From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Zijlstra Subject: Re: [Announce] 2.6.29.3-rt12 Date: Wed, 13 May 2009 12:24:48 +0200 Message-ID: <1242210288.26820.18.camel@twins> References: <5a7b8b7b0905121716y646a2f68qad9181362caa22f6@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Cc: Hugo Vincent , LKML , rt-users , Ingo Molnar , Steven Rostedt , Carsten Emde , Clark Williams , Frank Rowand , Robin Gareus , Gregory Haskins , Philippe Reynes , Fernando Lopez-Lezcano , Will Schmidt , Darren Hart , Jan Blunck , Sven-Thorsten Dietrich To: Thomas Gleixner Return-path: Received: from casper.infradead.org ([85.118.1.10]:46634 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758135AbZEMKZL (ORCPT ); Wed, 13 May 2009 06:25:11 -0400 In-Reply-To: Sender: linux-rt-users-owner@vger.kernel.org List-ID: On Wed, 2009-05-13 at 12:19 +0200, Thomas Gleixner wrote: > On Wed, 13 May 2009, Hugo Vincent wrote: > > This patch resulted in a broken build for me, until I did this: > > > > --- a/patch-2.6.29.3-rt12 > > +++ b/patch-2.6.29.3-rt12 > > @@ -121754,7 +121754,7 @@ Index: linux-2.6-tip/kernel/sched.c > > } > > > > /* > > -@@ -2801,6 +3177,15 @@ static void update_cpu_load(struct rq *t > > +@@ -2801,6 +3177,17 @@ static void update_cpu_load(struct rq *t > > new_load += scale-1; > > this_rq->cpu_load[i] = (old_load*(scale-1) + new_load) >> i; > > } > > @@ -121765,8 +121765,10 @@ Index: linux-2.6-tip/kernel/sched.c > > + this_rq->calc_load_update += LOAD_FREQ; > > + calc_load_account_active(this_rq); > > + > > ++#ifdef CONFIG_SMP > > + if (this_rq->cpu == calc_load_cpu) > > + calc_global_load(); > > ++#endif > > } > > Darn. the patch is halfways right. the #ifdef needs to be only around > the if() line. Otherwise your /proc/loadavg values will never be > updated. I'll push out -rt13 with the proper fix.. if (cpu_of(this_rq) == calc_load_cpu) to reduce if-deffery