From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jens Axboe Subject: Re: What am I doing wrong? submit_bio() suddenly stops working... Date: Fri, 22 Oct 2010 09:19:37 +0200 Message-ID: <4CC13B09.6030104@kernel.dk> References: <4CBFE4E2.7050001@kernel.dk> <20101021165525.GB3127@thunk.org> <4CC07C67.9010502@kernel.dk> <20101022033424.GO3127@thunk.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit To: Ted Ts'o , "linux-kernel@vger.kernel.org" , "linux-ext4@vger.kernel.org" Return-path: Received: from 0122700014.0.fullrate.dk ([95.166.99.235]:33747 "EHLO kernel.dk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752671Ab0JVHTW (ORCPT ); Fri, 22 Oct 2010 03:19:22 -0400 In-Reply-To: <20101022033424.GO3127@thunk.org> Sender: linux-ext4-owner@vger.kernel.org List-ID: On 2010-10-22 05:34, Ted Ts'o wrote: > On Thu, Oct 21, 2010 at 07:46:15PM +0200, Jens Axboe wrote: >> By the sound of things, if I were you I'd turn on the mem and slab >> debugging to catch use-before-init and use-after-free. Mysterious hangs >> in the IO sub system are usually caused by such bugs. And the regular >> debugging aids, just to see if that produces anything of interest. > > It looks like it was a use-after-free bug in my code. I'm running a > full set of set of tests now, but so far, it's gotten a lot further > than it went before, so I think I've figured it out. > > I'm not sure why it caused the weird behaviour that it did (I got as > far as figuring out that somehow we lost the unplug timer, so after > the queue got plugged it never got unplugged), but I'm not going to > ask too many questions. :-) > > Maybe later on I'll try to figure out if there's any way to add some > kind of sanity checking so that screw ups in in the bio code's caller > cause a clearer failure (such as a BUG_ON), but that'll have to wait > for when I have some free time. In my experience, use-after-free bugs are best caught using the allocator poisoning along with lock checking. I guess that didn't trigger for you? Out of curiousity, what was the exact bug? -- Jens Axboe