From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755411Ab0HPRi2 (ORCPT ); Mon, 16 Aug 2010 13:38:28 -0400 Received: from mga03.intel.com ([143.182.124.21]:39848 "EHLO mga03.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752725Ab0HPRi1 (ORCPT ); Mon, 16 Aug 2010 13:38:27 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.55,377,1278313200"; d="scan'208";a="313082967" Subject: Re: [patch 2/3] sched: fix minimum power returned by update_cpu_power() From: Suresh Siddha Reply-To: Suresh Siddha To: Peter Zijlstra Cc: "mingo@elte.hu" , "linux-kernel@vger.kernel.org" , "chris@frostnet.net" , "debian00@aliceadsl.fr" , "hpa@zytor.com" , "jonathan.protzenko@gmail.com" , "mans@mansr.com" , "psastudio@mail.ru" , "rjw@sisk.pl" , "stephan.eicher@web.de" , "sxxe@gmx.de" , "thomas@archlinux.org" , "venki@google.com" , "wonghow@gmail.com" In-Reply-To: <1281945037.1926.954.camel@laptop> References: <20100813190539.410550989@sbsiddha-MOBL3.sc.intel.com> <20100813193911.909546779@sbsiddha-MOBL3.sc.intel.com> <1281945037.1926.954.camel@laptop> Content-Type: text/plain Organization: Intel Corp Date: Mon, 16 Aug 2010 10:37:59 -0700 Message-Id: <1281980279.2676.24.camel@sbsiddha-MOBL3.sc.intel.com> Mime-Version: 1.0 X-Mailer: Evolution 2.26.3 (2.26.3-1.fc11) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 2010-08-16 at 00:50 -0700, Peter Zijlstra wrote: > On Fri, 2010-08-13 at 12:45 -0700, Suresh Siddha wrote: > > plain text document attachment (fix_update_cpu_power.patch) > > Default cpu_power needs to be multiples of SCHED_LOAD_SCALE and not '1'. > > Fix it. > > > > Signed-off-by: Suresh Siddha > > --- > > kernel/sched_fair.c | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > Index: tree/kernel/sched_fair.c > > =================================================================== > > --- tree.orig/kernel/sched_fair.c > > +++ tree/kernel/sched_fair.c > > @@ -2309,7 +2309,7 @@ static void update_cpu_power(struct sche > > power >>= SCHED_LOAD_SHIFT; > > > > if (!power) > > - power = 1; > > + power = SCHED_LOAD_SCALE; > > > > smt_power freq_power rt_power > power = SCHED_LOAD_SCALE * ---------- * ---------- * ---------- > LOAD_SCALE LOAD_SCALE LOAD_SCALE > > Which, in the above case ends up being 0, so how does resetting it back > to LOAD_SCALE make sense? hmm, true, but I thought I saw some load balancing code which was depending on SCHED_LOAD_SCALE value. Ignore this for now. Will get back to you if this indeed is a problem. thanks, suresh