From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jarek Poplawski Subject: Re: deadlocks if use htb Date: Wed, 14 Jan 2009 13:57:06 +0000 Message-ID: <20090114135706.GE6117@ff.dom.local> References: <20081010090426.GA6054@ff.dom.local> <200901141417.58667.denys@visp.net.lb> <1231937404.14825.4.camel@laptop> <200901141505.46929.denys@visp.net.lb> <20090114131257.GC6117@ff.dom.local> <1231938929.14825.6.camel@laptop> <20090114132603.GD6117@ff.dom.local> <1231939946.14825.9.camel@laptop> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Denys Fedoryschenko , Chris Caputo , netdev@vger.kernel.org, linux-kernel@vger.kernel.org, Badalian Vyacheslav , Thomas Gleixner To: Peter Zijlstra Return-path: Received: from mail-ew0-f31.google.com ([209.85.219.31]:41351 "EHLO mail-ew0-f31.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751316AbZANN5O (ORCPT ); Wed, 14 Jan 2009 08:57:14 -0500 Content-Disposition: inline In-Reply-To: <1231939946.14825.9.camel@laptop> Sender: netdev-owner@vger.kernel.org List-ID: On Wed, Jan 14, 2009 at 02:32:26PM +0100, Peter Zijlstra wrote: > On Wed, 2009-01-14 at 13:26 +0000, Jarek Poplawski wrote: > > > OK, I hope Denys can apply more, but what about others? Without any > > patches the hole seems to be much bigger. > > OK, I read most of this thread on netdev, but didn't find a clear clue > on the specific hrtimer insertion race. > > Do you have any clear ideas or should I poke at the htb/hrtimer code a > little? > I didn't diagnose this entirely, and there were "a few" days since I wrote my opinion on this, so currently I forgot details, but I guess the most suspicious place is adding a hrtimer, when it's on the list, maybe even two times in a row. I guess it's not removed from the rbtree in some case. (I stopped searching after seeing you removed this list or something...) Jarek P.