From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754755Ab0IMSYN (ORCPT ); Mon, 13 Sep 2010 14:24:13 -0400 Received: from mx3.mail.elte.hu ([157.181.1.138]:59247 "EHLO mx3.mail.elte.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752515Ab0IMSYK (ORCPT ); Mon, 13 Sep 2010 14:24:10 -0400 Date: Mon, 13 Sep 2010 20:23:55 +0200 From: Ingo Molnar To: Mathieu Desnoyers Cc: Linus Torvalds , Peter Zijlstra , LKML , Andrew Morton , Steven Rostedt , Thomas Gleixner , Tony Lindgren , Mike Galbraith Subject: [PATCH] sched: Improve latencies under load by decreasing minimum scheduling granularity Message-ID: <20100913182355.GC20171@elte.hu> References: <20100912203712.GD32327@Krystal> <1284382387.2275.265.camel@laptop> <1284383758.2275.283.camel@laptop> <20100913135621.GA13442@Krystal> <1284387398.2275.311.camel@laptop> <20100913161641.GA28707@Krystal> <20100913174533.GA15653@Krystal> <20100913180348.GA20171@elte.hu> <20100913181925.GA15107@Krystal> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100913181925.GA15107@Krystal> User-Agent: Mutt/1.5.20 (2009-08-17) X-ELTE-SpamScore: -2.0 X-ELTE-SpamLevel: X-ELTE-SpamCheck: no X-ELTE-SpamVersion: ELTE 2.0 X-ELTE-SpamCheck-Details: score=-2.0 required=5.9 tests=BAYES_00 autolearn=no SpamAssassin version=3.2.5 -2.0 BAYES_00 BODY: Bayesian spam probability is 0 to 1% [score: 0.0000] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Mathieu Desnoyers wrote: > * Ingo Molnar (mingo@elte.hu) wrote: > > > > * Mathieu Desnoyers wrote: > > > > > * Linus Torvalds (torvalds@linux-foundation.org) wrote: > > > > On Mon, Sep 13, 2010 at 9:16 AM, Mathieu Desnoyers > > > > wrote: > > > > > > > > > > OK, the long IRC discussions we just had convinced me that the current scheme > > > > > takes things into account by adapting the granularity dynamically, but also got > > > > > me to notice that check_preempt seems to compare vruntime with wall time, which > > > > > is utterly incorrect. So maybe all my patch was doing was to expose this bug: > > > > > > > > Do you have latency numbers for this patch? > > > > > > Sure, see below, > > > > > > In addition to this patch, [...] > > > > Note, which is a NOP for your latency workload. > > > > > [...] I also used Peter's approach of reducing the minimum granularity > > > > Ok, that's the very first patch i sent yesterday morning - so we also > > have my numbers that it reduces latencies. > > > > To move things along i'll apply it with your Reported-by and Acked-by > > line, ok? > > > > We can also work on the other, more complex things after that, but first > > lets make some progress on the latency front ... > > Yep, that's fine with me. > > Thanks! You are welcome! Linus, Mathieu, you can test the granularity reduction patch via: git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git sched/urgent Patch also attached below. Note, i'd like to keep this separate from the check_preempt() change - which only affects reniced tasks and isnt essential to these tests. (we want such things to be in separate commits, for bisectability) Thanks, Ingo ------------------> Ingo Molnar (1): sched: Improve latencies under load by decreasing minimum scheduling granularity kernel/sched_fair.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 9b5b4f8..a171138 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -54,13 +54,13 @@ enum sched_tunable_scaling sysctl_sched_tunable_scaling * Minimal preemption granularity for CPU-bound tasks: * (default: 2 msec * (1 + ilog(ncpus)), units: nanoseconds) */ -unsigned int sysctl_sched_min_granularity = 2000000ULL; -unsigned int normalized_sysctl_sched_min_granularity = 2000000ULL; +unsigned int sysctl_sched_min_granularity = 750000ULL; +unsigned int normalized_sysctl_sched_min_granularity = 750000ULL; /* * is kept at sysctl_sched_latency / sysctl_sched_min_granularity */ -static unsigned int sched_nr_latency = 3; +static unsigned int sched_nr_latency = 8; /* * After fork, child runs first. If set to 0 (default) then