From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751417AbWDNXAy (ORCPT ); Fri, 14 Apr 2006 19:00:54 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751434AbWDNXAy (ORCPT ); Fri, 14 Apr 2006 19:00:54 -0400 Received: from ishtar.tlinx.org ([64.81.245.74]:38109 "EHLO ishtar.tlinx.org") by vger.kernel.org with ESMTP id S1751417AbWDNXAy (ORCPT ); Fri, 14 Apr 2006 19:00:54 -0400 Message-ID: <444029A2.3060702@tlinx.org> Date: Fri, 14 Apr 2006 16:00:50 -0700 From: Linda Walsh User-Agent: Thunderbird 1.5 (Windows/20051201) MIME-Version: 1.0 To: Andreas Schnaiter CC: linux-kernel@vger.kernel.org Subject: Re: Linux 2.6.16 - SATA read performance drop ~50% on Intel 82801GB/GR/GH References: <200604120136.28681.schnaiter@gmx.net> In-Reply-To: <200604120136.28681.schnaiter@gmx.net> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Andreas Schnaiter wrote: > 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. > > Linux 2.6.15.7 > --- > # time dd if=/benchfile of=/dev/null bs=1M > 8192+0 records in, 8192+0 records out > 8589934592 bytes (8.6 GB) copied, 130.547 seconds, 65.8 MB/s > real 2m10.670s user 0m0.023s sys 0m14.238s > > Linux 2.6.16.2 > --- > # time dd if=/benchfile of=/dev/null bs=1M > 8192+0 records in, 8192+0 records out > 8589934592 bytes (8.6 GB) copied, 302.452 seconds, 28.4 MB/s > real 5m3.100s user 0m0.021s sys 0m40.521s --- Slightly echoing Jeff G's question, but rephrasing for read, could you try a read on the actual device? I.e. without destroying your partition, you could try a direct read from the device: time dd if=/dev/sda of=/dev/null bs=1M count=8192. BTW, how much memory do you have on the system? Not that I have know much about block-i/o, but it might narrow things down. Have you tried the tests in single-user or run-level 1 to help rule out other noise? Linda