From mboxrd@z Thu Jan 1 00:00:00 1970 From: James Bottomley Date: Wed, 20 Jul 2011 16:45:16 +0000 Subject: Re: [PATCH SCSI] sr: check_events() ignore GET_EVENT when TUR says Message-Id: <1311180316.8570.1.camel@mulgrave> List-Id: References: <20110630130348.GQ3386@htj.dyndns.org> <20110720163308.GC9924@mtj.dyndns.org> In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Kay Sievers Cc: Tejun Heo , linux-scsi@vger.kernel.org, maggu2810@googlemail.com, linux-hotplug@vger.kernel.org On Wed, 2011-07-20 at 18:42 +0200, Kay Sievers wrote: > On Wed, Jul 20, 2011 at 18:33, Tejun Heo wrote: > > On Thu, Jun 30, 2011 at 03:03:48PM +0200, Tejun Heo wrote: > >> From: Kay Sievers > >> > >> Some broken devices indicates that media has changed on every > >> GET_EVENT_STATUS_NOTIFICATION. This translates into MEDIA_CHANGE > >> uevent on every open() which lets udev run into a loop. > >> > >> Verify GET_EVENT result against TUR and if it generates spurious > >> events for several times in a row, ignore the GET_EVENT events, and > >> trust only the TUR status. > > > James, ping. > > This is a rather urgent fix, that needs to go in, and also to -stable. > > Systems run into endless loops with ever-growing event queues. Nothing > really works any more if we hit that condition, unless the (broken) > device is disconnected. Sorry ... I have it marked for inclusion in my internal queue, I'm just having difficulty finding time to do a dump from queue and test. I should be able to get all the fixes sorted on the 'plane home tomorrow. James