From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jens Axboe Subject: Re: [PATCH] block: fix oops in blk_run_queue() Date: Wed, 25 May 2011 13:28:41 +0200 Message-ID: <4DDCE7E9.5090107@kernel.dk> References: <201105191944.p4JJiVVm029690@demeter1.kernel.org> <1306257531.10201.20.camel@mulgrave.site> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Return-path: Received: from 0122700014.0.fullrate.dk ([95.166.99.235]:56030 "EHLO kernel.dk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750783Ab1EYL2p (ORCPT ); Wed, 25 May 2011 07:28:45 -0400 In-Reply-To: <1306257531.10201.20.camel@mulgrave.site> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: James Bottomley Cc: linux-scsi@vger.kernel.org On 2011-05-24 19:18, James Bottomley wrote: > Bugzilla report is here: > > https://bugzilla.kernel.org/show_bug.cgi?id=35682 > > The problem occurs if blk_run_queue() or its variants is called after > blk_cleanup_queue() is called. The run checks should have a state guard > for QUEUE_FLAG_DEAD. Since a DEAD queue is always STOPPED, just fold it > into the blk_queue_stopped() test. > > Signed-off-by: James Bottomley > > --- > > Jens, can you look at this, please ... would you prefer a separate DEAD > test instead? Incidentally, this is in addition to the other DEAD test > you still haven't applied ... Seems a bit of a mixup. Would be cleaner to set STOPPED when it is marked dead, but I suspect that will kill erroring out IO when the device goes away unless we add some logic to ensure that happens without the drivers help. -- Jens Axboe