From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756040AbYIDMMT (ORCPT ); Thu, 4 Sep 2008 08:12:19 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751927AbYIDMML (ORCPT ); Thu, 4 Sep 2008 08:12:11 -0400 Received: from mga11.intel.com ([192.55.52.93]:43390 "EHLO mga11.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751584AbYIDMMK (ORCPT ); Thu, 4 Sep 2008 08:12:10 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.32,320,1217833200"; d="scan'208";a="376682069" Subject: Re: oltp ~10% regression with 2.6.27-rc5 on stoakley machine From: Lin Ming To: Peter Zijlstra Cc: linux-kernel , "Zhang, Yanmin" , mingo , Gregory Haskins In-Reply-To: <1220526360.8609.213.camel@twins> References: <1220518266.9590.22.camel@minggr> <1220519034.8609.206.camel@twins> <1220525579.12161.8.camel@minggr> <1220526360.8609.213.camel@twins> Content-Type: text/plain Date: Thu, 04 Sep 2008 20:12:08 +0800 Message-Id: <1220530328.12161.29.camel@minggr> Mime-Version: 1.0 X-Mailer: Evolution 2.12.1 (2.12.1-3.fc8) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org > > > Thats bizarre... that just indicates the better clock, which should give > > > better (read fairer) scheduling hurts your workload. > > > > > > Is there anything I can run to see if we can fix the scheduler perhaps? > > > > I observed schedstats of sysbench, there's more > > "nr_failed_migrations_hot" > > > > 2.6.27-rc4: se.nr_failed_migrations_hot 11 > > 2.6.27-rc5: se.nr_failed_migrations_hot 95 > > > > task migration failed because of task_hot, the system is un-balanced? > > Ah, that makes sense, a more accurate clock could indeed make more tasks > hot. > > Can you try fiddling with: /proc/sys/kernel/sched_migration_cost ? sched_migration_cost regression ---------------------- ------------- 50000 ~6% 0 ~8% 500000 (default) ~10% 5000000 ~14% -1 ~19% > > Also, we used to have some auto-tuning in there, which dissapeared some > time ago, gregory brought it back to live recently, perhaps he likes to > share? :-) >