From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933735AbcATAtF (ORCPT ); Tue, 19 Jan 2016 19:49:05 -0500 Received: from LGEAMRELO13.lge.com ([156.147.23.53]:37604 "EHLO lgeamrelo13.lge.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933363AbcATAst (ORCPT ); Tue, 19 Jan 2016 19:48:49 -0500 X-Original-SENDERIP: 156.147.1.125 X-Original-MAILFROM: byungchul.park@lge.com X-Original-SENDERIP: 10.177.222.33 X-Original-MAILFROM: byungchul.park@lge.com Date: Wed, 20 Jan 2016 09:48:25 +0900 From: Byungchul Park To: Peter Zijlstra Cc: Dietmar Eggemann , perterz@infradead.org, Frederic Weisbecker , LKML , Chris Metcalf , Thomas Gleixner , Luiz Capitulino , Christoph Lameter , "Paul E . McKenney" , Mike Galbraith , Rik van Riel Subject: Re: [RFC PATCH 0/4] sched: Improve cpu load accounting with nohz Message-ID: <20160120004825.GB9882@X58A-UD3R> References: <1452700891-21807-1-git-send-email-fweisbec@gmail.com> <569810C4.7090900@arm.com> <20160114212704.GJ6357@twins.programming.kicks-ass.net> <56981FF2.6030700@arm.com> <20160115070749.GA1914@X58A-UD3R> <569924C4.8010903@arm.com> <20160119130457.GB6344@twins.programming.kicks-ass.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20160119130457.GB6344@twins.programming.kicks-ass.net> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Jan 19, 2016 at 02:04:57PM +0100, Peter Zijlstra wrote: > On Fri, Jan 15, 2016 at 04:56:36PM +0000, Dietmar Eggemann wrote: > > Couldn't we set tickless_load only in case: > > > > unsigned long tickless_load = (active && pending_updates > 1) ? > > this_rq->cpu_load[0] : 0; > > > > Even though update_cpu_load_nohz() can call with pending_updates=1 and > > active=1 but then we don't have to decay. > > decay_load_missed() has an early bail for !missed, which will be tickled > with pending_updates == 1. I think the way for decay_load_missed() to get an early bail for *!load*, which the Dietmar's proposal did, is also good. And the peterz's proposal avoiding an unnecessary "add" operation is also good. Whatever.. > > What I was thinking of doing however is: > > --- a/kernel/sched/fair.c > +++ b/kernel/sched/fair.c > @@ -4445,13 +4445,15 @@ static void __update_cpu_load(struct rq > > old_load = this_rq->cpu_load[i]; > old_load = decay_load_missed(old_load, pending_updates - 1, i); > - old_load -= decay_load_missed(tickless_load, pending_updates - 1, i); > - /* > - * old_load can never be a negative value because a decayed > - * tickless_load cannot be greater than the original > - * tickless_load. > - */ > - old_load += tickless_load; > + if (tickless_load) { And additionally, in this approach, why don't you do like, if (tickless_load || pending_updates - 1) > + old_load -= decay_load_missed(tickless_load, pending_updates - 1, i); > + /* > + * old_load can never be a negative value because a > + * decayed tickless_load cannot be greater than the > + * original tickless_load. > + */ > + old_load += tickless_load; > + } > new_load = this_load; > /* > * Round up the averaging division if load is increasing. This > > > Since regardless of the pending_updates, none of that makes sense if > !tickless_load. None of that makes sense if !(pending_updates - 1), too. In that case, it becomes, old_load -= tickless_load; old_load += tickless_load;