From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753315Ab0DUIqA (ORCPT ); Wed, 21 Apr 2010 04:46:00 -0400 Received: from casper.infradead.org ([85.118.1.10]:51770 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753106Ab0DUIp6 (ORCPT ); Wed, 21 Apr 2010 04:45:58 -0400 Subject: Re: [Bug #15672] KVM bug, git bisected From: Peter Zijlstra To: Borislav Petkov Cc: "Rafael J. Wysocki" , Rik van Riel , Linux Kernel Mailing List , Kernel Testers List , Maciej Rutecki , Kent Overstreet , Ingo Molnar In-Reply-To: <20100421062015.GB3839@liondog.tnic> References: <4BCE188A.1060408@redhat.com> <4BCE72D6.6010809@redhat.com> <201004210702.02129.rjw@sisk.pl> <20100421062015.GB3839@liondog.tnic> Content-Type: text/plain; charset="UTF-8" Date: Wed, 21 Apr 2010 10:45:54 +0200 Message-ID: <1271839554.1776.56.camel@laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.28.3 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 2010-04-21 at 08:20 +0200, Borislav Petkov wrote: > From: "Rafael J. Wysocki" > Date: Wed, Apr 21, 2010 at 07:02:02AM +0200 > > > On Wednesday 21 April 2010, Rik van Riel wrote: > > > On 04/20/2010 05:11 PM, Rik van Riel wrote: > > > > On 04/19/2010 11:19 PM, Rafael J. Wysocki wrote: > > > >> This message has been generated automatically as a part of a summary > > > >> report > > > >> of recent regressions. > > > >> > > > >> The following bug entry is on the current list of known regressions > > > >> from 2.6.33. Please verify if it still should be listed and let the > > > >> tracking team > > > >> know (either way). > > > >> > > > >> > > > >> Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=15672 > > > >> Subject : KVM bug, git bisected > > > >> Submitter : Kent Overstreet > > > >> Date : 2010-03-27 12:43 (24 days old) > > > >> First-Bad-Commit: > > > >> http://kernel.org/git/linus/5beb49305251e5669852ed541e8e2f2f7696c53e > > > >> Message-ID :<4BADFD74.8060904@gmail.com> > > > >> References : http://marc.info/?l=linux-kernel&m=126969385121711&w=2 > > > > > > > > Should be fixed by commit ea90002b0fa7bdee86ec22eba1d951f30bf043a6 > > > > > > Never mind me - this is a harmless (but loud) overflow > > > of PREEMPT_BITS in the preempt count. > > > > OK, what am I supposed to do with this entry, then? Close? > > FWIW, I hit that warning too when chasing the anon_vma regression. It > seems on certain workloads (for me it was several kvm guests) we're > close to max preemption depth. > > Anyway, adding some more people to Cc. Right, so my proposed solution to this is to make those locks preemptible, but that's a large and unfinished patch-set. As it is, its only a warning, nothing really serious should happen, but the situation does suck. From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Zijlstra Subject: Re: [Bug #15672] KVM bug, git bisected Date: Wed, 21 Apr 2010 10:45:54 +0200 Message-ID: <1271839554.1776.56.camel@laptop> References: <4BCE188A.1060408@redhat.com> <4BCE72D6.6010809@redhat.com> <201004210702.02129.rjw@sisk.pl> <20100421062015.GB3839@liondog.tnic> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20100421062015.GB3839-f9CnO7I+Q6zU6FkGJEIX5A@public.gmane.org> Sender: kernel-testers-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-ID: Content-Type: text/plain; charset="us-ascii" To: Borislav Petkov Cc: "Rafael J. Wysocki" , Rik van Riel , Linux Kernel Mailing List , Kernel Testers List , Maciej Rutecki , Kent Overstreet , Ingo Molnar On Wed, 2010-04-21 at 08:20 +0200, Borislav Petkov wrote: > From: "Rafael J. Wysocki" > Date: Wed, Apr 21, 2010 at 07:02:02AM +0200 > > > On Wednesday 21 April 2010, Rik van Riel wrote: > > > On 04/20/2010 05:11 PM, Rik van Riel wrote: > > > > On 04/19/2010 11:19 PM, Rafael J. Wysocki wrote: > > > >> This message has been generated automatically as a part of a summary > > > >> report > > > >> of recent regressions. > > > >> > > > >> The following bug entry is on the current list of known regressions > > > >> from 2.6.33. Please verify if it still should be listed and let the > > > >> tracking team > > > >> know (either way). > > > >> > > > >> > > > >> Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=15672 > > > >> Subject : KVM bug, git bisected > > > >> Submitter : Kent Overstreet > > > >> Date : 2010-03-27 12:43 (24 days old) > > > >> First-Bad-Commit: > > > >> http://kernel.org/git/linus/5beb49305251e5669852ed541e8e2f2f7696c53e > > > >> Message-ID :<4BADFD74.8060904-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> > > > >> References : http://marc.info/?l=linux-kernel&m=126969385121711&w=2 > > > > > > > > Should be fixed by commit ea90002b0fa7bdee86ec22eba1d951f30bf043a6 > > > > > > Never mind me - this is a harmless (but loud) overflow > > > of PREEMPT_BITS in the preempt count. > > > > OK, what am I supposed to do with this entry, then? Close? > > FWIW, I hit that warning too when chasing the anon_vma regression. It > seems on certain workloads (for me it was several kvm guests) we're > close to max preemption depth. > > Anyway, adding some more people to Cc. Right, so my proposed solution to this is to make those locks preemptible, but that's a large and unfinished patch-set. As it is, its only a warning, nothing really serious should happen, but the situation does suck.