From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S936977AbYCSXFx (ORCPT ); Wed, 19 Mar 2008 19:05:53 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S936013AbYCSVb0 (ORCPT ); Wed, 19 Mar 2008 17:31:26 -0400 Received: from fallback.mail.elte.hu ([157.181.151.13]:45627 "EHLO fallback.mail.elte.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933750AbYCSVbZ (ORCPT ); Wed, 19 Mar 2008 17:31:25 -0400 Date: Wed, 19 Mar 2008 04:49:53 +0100 From: Ingo Molnar To: Linus Torvalds Cc: linux-kernel@vger.kernel.org Subject: [git pull] scheduler fixes Message-ID: <20080319034952.GB21347@elte.hu> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.17 (2007-11-01) X-ELTE-VirusStatus: clean X-ELTE-SpamScore: -1.5 X-ELTE-SpamLevel: X-ELTE-SpamCheck: no X-ELTE-SpamVersion: ELTE 2.0 X-ELTE-SpamCheck-Details: score=-1.5 required=5.9 tests=BAYES_00 autolearn=no SpamAssassin version=3.2.3 -1.5 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 Linus, please pull the latest scheduler fixes git tree from: git://git.kernel.org/pub/scm/linux/kernel/git/mingo/linux-2.6-sched-devel.git for-linus [ Note: it touches net/core/sock.c to tweak two socket wakeups there. ] this tree improves/fixes lat_ctx/lat_pipe/bw_pipe latency/performance on SMP systems: .25-rc6 .25-rc6-sched/for-linus ------------------------------------------------------------- lat_ctx 2 : 1.93 us 0.69 us lat_pipe : 4.96 us 2.49 us bw_pipe : 2841.11 MB/sec 3436.61 MB/sec it also improves the ramp-up phase of sysbench on a quad: .25-rc6 .25-rc6-sched/for-linus ------------------------------------------------------------- 1: ( 836.64 per sec.) (1216.90 per sec.) 2: (2407.02 per sec.) (2361.12 per sec.) 4: (4468.59 per sec.) (4510.14 per sec.) ( similar improvements were measured on an 8-way and on various dual boxes as well - a 16way box was largely unchanged. ) these changes also improved some other workloads as well: mysql, apache and others - and it did not impact some others that we handled well before (such as kbuild, hackbench and others). Thanks, Ingo ------------------> Ingo Molnar (8): sched: clean up wakeup balancing, move wake_affine() sched: clean up wakeup balancing, rename variables sched: clean up wakeup balancing, code flow sched, net: socket wakeups are sync sched: improve affine wakeups sched: wakeup-buddy tasks are cache-hot sched: retune wake granularity sched: tune multi-core idle balancing include/linux/sched.h | 3 + include/linux/topology.h | 1 - kernel/sched.c | 11 +++- kernel/sched_debug.c | 1 + kernel/sched_fair.c | 191 ++++++++++++++++++++++++++++------------------ net/core/sock.c | 4 +- 6 files changed, 134 insertions(+), 77 deletions(-)