From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: Re: ICH6-M libata disk timeouts in 2.6.19 (except with small queue depth) Date: Sun, 21 Jan 2007 03:21:48 +0900 Message-ID: <45B25DBC.4020102@gmail.com> References: <798.1169174977@mdt.dhcp.pit.laurelnetworks.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from ug-out-1314.google.com ([66.249.92.169]:48951 "EHLO ug-out-1314.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S965263AbXATSV4 (ORCPT ); Sat, 20 Jan 2007 13:21:56 -0500 Received: by ug-out-1314.google.com with SMTP id 44so720274uga for ; Sat, 20 Jan 2007 10:21:55 -0800 (PST) In-Reply-To: <798.1169174977@mdt.dhcp.pit.laurelnetworks.com> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Mike Accetta Cc: linux-ide@vger.kernel.org Mike Accetta wrote: > The distinguishing factor appears to be the queue depth (4 works, 5 > and various values up to and including 31 fail) not the kernel version. > I am going to try running with the queue depth clamped at 4 to see if > this consistently masks the problem. I may also try some more experiments > if I have the time, like instrumenting what command was issued right > before the group that all time out or increasing the SCSI timeout, > in order to get some more insight into what is going on at the time of > the failure. Please report the result of 'hdparm -I /dev/sdX' where sdX is the problematic disk. Quite a few drives have problem with NCQ and we have to blacklist them. Thanks. -- tejun