From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Gleixner Subject: Re: Deadlocks due to per-process plugging Date: Sun, 15 Jul 2012 10:59:34 +0200 (CEST) Message-ID: References: <20120711133735.GA8122@quack.suse.cz> <20120711201601.GB9779@quack.suse.cz> <20120713123318.GB20361@quack.suse.cz> <20120713144622.GB28715@quack.suse.cz> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Cc: Jeff Moyer , LKML , linux-fsdevel@vger.kernel.org, Tejun Heo , Jens Axboe , mgalbraith@suse.com To: Jan Kara Return-path: Received: from www.linutronix.de ([62.245.132.108]:49584 "EHLO Galois.linutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751615Ab2GOI7k (ORCPT ); Sun, 15 Jul 2012 04:59:40 -0400 In-Reply-To: <20120713144622.GB28715@quack.suse.cz> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Fri, 13 Jul 2012, Jan Kara wrote: > On Fri 13-07-12 16:25:05, Thomas Gleixner wrote: > > So the patch below should allow the unplug to take place when blocked > > on mutexes etc. > Thanks for the patch! Mike will give it some testing. I just found out that this patch will explode nicely when the unplug code runs into a contended lock. Then we try to block on that lock and make the rtmutex code unhappy as we are already blocked on something else. So no, it's not a solution to the problem. Sigh. Can you figure out on which lock the stuck thread which did not unplug due to tsk_is_pi_blocked was blocked? Thanks, tglx