From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754026AbaCDPck (ORCPT ); Tue, 4 Mar 2014 10:32:40 -0500 Received: from mail-pb0-f43.google.com ([209.85.160.43]:38699 "EHLO mail-pb0-f43.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753758AbaCDPci (ORCPT ); Tue, 4 Mar 2014 10:32:38 -0500 Message-ID: <5315F203.80602@linaro.org> Date: Tue, 04 Mar 2014 23:32:19 +0800 From: Alex Shi User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.0 MIME-Version: 1.0 To: Fengguang Wu CC: LKML , lkp@01.org Subject: Re: [sched/balance] 18e6296b85e: -96.9% interrupts.RES References: <20140304130559.GA28781@localhost> In-Reply-To: <20140304130559.GA28781@localhost> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Thanks a lot for the testing result! I know the problem I have. :) On 03/04/2014 09:05 PM, Fengguang Wu wrote: > Hi Alex, > > FYI, we noticed the below changes on > > https://github.com/alexshi/power-scheduling.git single-balance > commit 18e6296b85ea72c81aa485653d442ad88296d475 ("sched/balance: central balance prototype") > > test case: snb-drag/crypto/tcrypt/2s-301-319 > > 598143917dc5ecf 18e6296b85ea72c81aa485653 > --------------- ------------------------- > 206307 ~ 0% -96.9% 6337 ~ 4% TOTAL interrupts.RES > 506552 ~ 1% -24.3% 383290 ~ 3% TOTAL cpuidle.C6-SNB.usage > 18.27 ~ 9% +1063.5% 212.53 ~25% TOTAL time.system_time > 1 ~30% +1262.5% 21 ~31% TOTAL time.percent_of_cpu_this_job_got > 821 ~12% -22.3% 638 ~19% TOTAL vmstat.system.in > > > > time.system_time > > 400 ++-------------------------------------------------------------------+ > | | > 350 ++ O O O | > 300 ++O O O | > O O O O O O | > 250 ++ O O | > | O O | > 200 ++ | > | O | > 150 ++ O | > 100 ++ | > | | > 50 ++ | > *.*..*.*..*.*..*.*..*.*..*.*..*.*..*.*.*..*.*..*.*..*.*..*.*..*.*..*.* > 0 ++-------------------------------------------------------------------+ > > > Thanks, > Fengguang