From: Chris Mason Subject: Re: konqueror deadlocks on 2.6.22 Date: Tue, 22 Jan 2008 09:25:49 -0500 Message-ID: <200801220925.50314.chris.mason@oracle.com> References: <200801192114.41427.a1426z@gawab.com> <20080122101014.GD5722@elte.hu> <200801221623.42989.a1426z@gawab.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Ingo Molnar , Oliver Pinter (=?iso-8859-1?q?Pint=E9r?= =?iso-8859-1?q?_Oliv=E9r?=) , linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org To: Al Boldi Return-path: Received: from agminet01.oracle.com ([141.146.126.228]:34219 "EHLO agminet01.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750781AbYAVO1v convert rfc822-to-8bit (ORCPT ); Tue, 22 Jan 2008 09:27:51 -0500 In-Reply-To: <200801221623.42989.a1426z@gawab.com> Content-Disposition: inline Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Tuesday 22 January 2008, Al Boldi wrote: > Ingo Molnar wrote: > > * Oliver Pinter (Pint=E9r Oliv=E9r) wrote: > > > and then please update to CFS-v24.1 > > > http://people.redhat.com/~mingo/cfs-scheduler/sched-cfs-v2.6.22.1= 5-v24. > > >1 .patch > > > > > > > Yes with CFSv20.4, as in the log. > > > > > > > > It also hangs on 2.6.23.13 > > > > my feeling is that this is some sort of timing dependent race in > > konqueror/kde/qt that is exposed when a different scheduler is put = in. > > > > If it disappears with CFS-v24.1 it is probably just because the tim= ings > > will change again. Would be nice to debug this on the konqueror sid= e and > > analyze why it fails and how. You can probably tune the timings by > > enabling SCHED_DEBUG and tweaking /proc/sys/kernel/*sched* values -= in > > particular sched_latency and the granularity settings. Setting wake= up > > granularity to 0 might be one of the things that could make a > > difference. > > Thanks Ingo, but Mike suggested that data=3Dwriteback may make a diff= erence, > which it does indeed. > > So the bug seems to be related to data=3Dordered, although I haven't = gotten > any feedback from the ext3 gurus yet. > > Seems rather critical though, as data=3Dwriteback is a dangerous mode= to run. Running fsync in data=3Dordered means that all of the dirty blocks on t= he FS=20 will get written before fsync returns. Your original stack trace shows= =20 everyone either performing writeback for a log commit or waiting for th= e log=20 commit to return. They key task in your trace is kjournald, stuck in get_request_wait. I= t could=20 be a block layer bug, not giving him requests quickly enough, or it cou= ld be=20 the scheduler not giving him back the cpu fast enough. At any rate, that's where to concentrate the debugging. You should be = able to=20 simulate this by running a few instances of the below loop and looking = for=20 stalls: while(true) ; do time dd if=3D/dev/zero of=3Dfoo bs=3D50M count=3D4 oflags=3Dsync = =20 done - To unsubscribe from this list: send the line "unsubscribe linux-fsdevel= " in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html