From mboxrd@z Thu Jan 1 00:00:00 1970 From: James Bottomley Subject: Re: Fw: oops in kernel 2.6.9 Date: 20 Oct 2004 08:32:38 -0500 Sender: linux-scsi-owner@vger.kernel.org Message-ID: <1098279164.2002.7.camel@mulgrave> References: <20041019175146.15889e11.akpm@osdl.org> <20041020064255.GB8755@suse.de> <20041019234857.7ddf5bc7.akpm@osdl.org> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from stat16.steeleye.com ([209.192.50.48]:49337 "EHLO hancock.sc.steeleye.com") by vger.kernel.org with ESMTP id S270009AbUJTNc5 (ORCPT ); Wed, 20 Oct 2004 09:32:57 -0400 In-Reply-To: <20041019234857.7ddf5bc7.akpm@osdl.org> List-Id: linux-scsi@vger.kernel.org To: Andrew Morton Cc: Jens Axboe , nickpiggin@yahoo.com.au, SCSI Mailing List On Wed, 2004-10-20 at 01:48, Andrew Morton wrote: > Jens Axboe wrote: > > > > On Tue, Oct 19 2004, Andrew Morton wrote: > > > > > > oy. > > > > Looks like a SCSI issue, it probably killed the queue before the io > > scheduler was entered again. > > Better add linux-scsi to cc then... The SCSI half of the fix for this is in bk latest (that was adding a refcount to the device as soon as the command is issued). I'm not sure about the usb-storage piece (not exciting the error handler by ignoring commands when the device is unplugged). James