From mboxrd@z Thu Jan 1 00:00:00 1970 From: Michael Tokarev Subject: Re: WDC WD1600BEVS-2 on NCQ blacklist Date: Sun, 09 Dec 2007 18:17:26 +0300 Message-ID: <475C0706.7090709@msgid.tls.msk.ru> References: <475BC8D7.5070407@harddisk.is-a-geek.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from hobbit.corpit.ru ([81.13.94.6]:20117 "EHLO hobbit.corpit.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751136AbXLIPR2 (ORCPT ); Sun, 9 Dec 2007 10:17:28 -0500 In-Reply-To: <475BC8D7.5070407@harddisk.is-a-geek.org> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Marco Schuster Cc: jgarzik@pobox.com, linux-ide@vger.kernel.org Marco Schuster wrote: > Hi all, > > I use a WDC WD1600BEVS-2 SATA drive (160GB). > The controller is a standard Intel ICH8 chipset. > During high disk load (fsck, large file copies) the system or the > process (mostly both) lock up and resume some secs later. In dmesg for > every of those breaks this appears: > > ata3.00: exception Emask 0x2 SAct 0x8001 SErr 0x0 action 0x2 frozen > ata3.00: spurious completions during NCQ issue=0x0 SAct=0x8001 FIS=004040a1:00004000 [] > Someone other than me also has this problem with this disk, see > http://marc.info/?l=linux-kernel&m=118789093418142&w=2 for details. > xingu on freenode/#debian suggested me to do echo 1 > > /sys/block/sdX/device/queue_depth, which solved these problems for one > session and to email you to put this drive/controller combo on the NCQ > blacklist. Marco, please see http://marc.info/?l=linux-ide&m=119707355016060&w=2 before adding more entries to the blacklist... /mjt