From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Al Boldi" Subject: IOWAIT block layer problem Date: Tue, 5 Jul 2005 06:43:33 +0300 Message-ID: <200507050343.GAA26158@raad.intranet> References: <42C9C56D.7040701@tomt.net> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1256 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from [213.184.187.60] ([213.184.187.60]:25861 "EHLO raad.intranet") by vger.kernel.org with ESMTP id S261755AbVGEDop convert rfc822-to-8bit (ORCPT ); Mon, 4 Jul 2005 23:44:45 -0400 In-Reply-To: <42C9C56D.7040701@tomt.net> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: =?windows-1256?Q?'Andr=E9_Tomt'?= Cc: 'Bartlomiej Zolnierkiewicz' , 'Ondrej Zary' , 'Linus Torvalds' , linux-ide@vger.kernel.org, linux-kernel@vger.kernel.org Andr=E9 Tomt wrote: { >>>>On 7/4/05, Al Boldi 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% IOWAIT The "hdparm doesn't get as high scores as in 2.4" is a old discussed to death "problem" on LKML. So far nobody has been able to show it affects anything but that pretty useless quasi-benchmark. >>>No errors though, only sluggish system. Really sluggish or just "benchmark-sluggish"? If the former, try selecting a different IO elevator/sheduler. > When you compare 2.4.31 with 2.6.12 don't you see this problem on you= r=20 > machine? > If you have a fast system the slowdown won't show, but your IOWAIT=20 > will be higher anyway! Nothing wrong with 73% iowait Beware 2.4 didn't export that statistic at all to userspace, so 0% iowa= it gets reported from most 2.6-ready reporting tools on 2.4. } 1. hdparm only gives you an indicator. 2. Really sluggish, as in unresponsive to commands. 3. Tried all IO elevators/shedulers to no avail. 4. In 2.4 IOWAIT is added to sys, as in 8% IOWAIT.