Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1765336AbZAOSxl (ORCPT ); Thu, 15 Jan 2009 13:53:41 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755942AbZAOSxd (ORCPT ); Thu, 15 Jan 2009 13:53:33 -0500 Received: from mx04.mailboxcop.com ([206.125.223.74]:44978 "EHLO mx04.mailboxcop.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755791AbZAOSxc (ORCPT ); Thu, 15 Jan 2009 13:53:32 -0500 Message-ID: <496F8631.3070902@jaysonking.com> Date: Thu, 15 Jan 2009 12:53:37 -0600 From: Jayson King User-Agent: Thunderbird 2.0.0.19 (X11/20090105) MIME-Version: 1.0 To: Peter Zijlstra CC: Lin Ming , Mike Galbraith , "Zhang, Yanmin" , LKML Subject: Re: Performance regression of specjbb2005/aim7 with 2.6.29-rc1 References: <1231837036.11429.27.camel@ymzhang> <1231839464.14655.12.camel@marge.simson.net> <1231986636.10565.159.camel@minggr.sh.intel.com> <1232036235.8870.75.camel@laptop> In-Reply-To: <1232036235.8870.75.camel@laptop> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Canit-CHI2: 0.04 X-Bayes-Prob: 0.0001 (Score 0, tokens from: @@RPTN, outgoing) X-Spam-Score: 0.10 () [Hold at 5.20] RDNS_NONE,2(1.2),6947(-1.2) X-CanItPRO-Stream: outgoing (inherits from default) X-Canit-Stats-ID: 341642990 - 59b5111924c9 X-Antispam-Training-Forget: http://mailboxcop.com/canit/b.php?i=341642990&m=59b5111924c9&c=f X-Antispam-Training-Nonspam: http://mailboxcop.com/canit/b.php?i=341642990&m=59b5111924c9&c=n X-Antispam-Training-Spam: http://mailboxcop.com/canit/b.php?i=341642990&m=59b5111924c9&c=s Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1582 Lines: 55 Peter Zijlstra wrote: > On Thu, 2009-01-15 at 10:30 +0800, Lin Ming wrote: > >> Below patch fixes aim7 regression and specjbb2005 regression becomes >> less than 1.5% on 8-core stokley. > > Ingo, please apply the below. > > Ming, would you also provide a S-o-b line? > > --- > Subject: sched_slice fixlet > From: Lin Ming > Date: Thu Jan 15 17:10:02 CET 2009 > > Mike's change: 0a582440f -- sched: fix sched_slice()) > Broke group scheduling by forgetting to reload cfs_rq on each loop. > > Signed-off-by: Peter Zijlstra > --- > kernel/sched_fair.c | 5 ++++- > 1 file changed, 4 insertions(+), 1 deletion(-) > > Index: linux-2.6/kernel/sched_fair.c > =================================================================== > --- linux-2.6.orig/kernel/sched_fair.c > +++ linux-2.6/kernel/sched_fair.c > @@ -429,7 +429,10 @@ static u64 sched_slice(struct cfs_rq *cf > u64 slice = __sched_period(cfs_rq->nr_running + !se->on_rq); > > for_each_sched_entity(se) { > - struct load_weight *load = &cfs_rq->load; > + struct load_weight *load; > + > + cfs_rq = cfs_rq_of(se); > + load = &cfs_rq->load; > > if (unlikely(!se->on_rq)) { > struct load_weight lw = cfs_rq->load; > > > That still works for me. You may add: Tested-by: Jayson King -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/