Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S964850AbcDMBy2 (ORCPT ); Tue, 12 Apr 2016 21:54:28 -0400 Received: from mga09.intel.com ([134.134.136.24]:2189 "EHLO mga09.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758563AbcDMBy2 (ORCPT ); Tue, 12 Apr 2016 21:54:28 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.24,477,1455004800"; d="scan'208";a="931202261" Date: Wed, 13 Apr 2016 02:12:01 +0800 From: Yuyang Du To: Peter Zijlstra Cc: Dietmar Eggemann , mingo@kernel.org, linux-kernel@vger.kernel.org, bsegall@google.com, pjt@google.com, morten.rasmussen@arm.com, vincent.guittot@linaro.org, juri.lelli@arm.com Subject: Re: [PATCH 1/4] sched/fair: Optimize sum computation with a lookup table Message-ID: <20160412181200.GK8697@intel.com> References: <1460327765-18024-1-git-send-email-yuyang.du@intel.com> <1460327765-18024-2-git-send-email-yuyang.du@intel.com> <570BD7DF.6070400@arm.com> <20160411191712.GG8697@intel.com> <20160412141952.GF1087@worktop> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20160412141952.GF1087@worktop> 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 Content-Length: 1787 Lines: 51 On Tue, Apr 12, 2016 at 04:19:52PM +0200, Peter Zijlstra wrote: > On Tue, Apr 12, 2016 at 03:17:12AM +0800, Yuyang Du wrote: > > On Mon, Apr 11, 2016 at 05:59:11PM +0100, Dietmar Eggemann wrote: > > > On 10/04/16 23:36, Yuyang Du wrote: > > > > __compute_runnable_contrib() uses a loop to compute sum, whereas a > > > > table loopup can do it faster in a constant time. > > > > > > > > The following python script can be used to generate the constants: > > > > > > > > print " #: yN_inv yN_sum" > > > > print "-----------------------" > > > > y = (0.5)**(1/32.0) > > > > x = 2**32 > > > > xx = 1024 > > > > for i in range(0, 32): > > > > if i == 0: > > > > x = x-1 > > > > xx = xx*y > > > > else: > > > > x = x*y > > > > xx = int(xx*y + 1024*y) > > > > print "%2d: %#x %8d" % (i, int(x), int(xx)) > > > > > > > > print " #: sum_N32" > > > > print "------------" > > > > xxx = xx > > > > for i in range(0, 11): > > > > if i == 0: > > > > xxx = xx > > > > else: > > > > xxx = xxx/2 + xx > > > > print "%2d: %8d" % (i, xxx) > > > > > > > > > > IMHO, it would be nice to add this to the existing tool from the patch > > > header of commit 5b51f2f80b3b > > > ("sched: Make __update_entity_runnable_avg() fast") simply because people > > > already use this one to tweak their pelt tables. Maybe something like > > > > I'd prefer not, and recommend switching from the C program for this > > kind of job. :) > > I much prefer C because I don't speak snake or any of the other popular > languages -- mostly because I simply don't use them enough to remember > how they work. > > Also, if we're going to edit that program, maybe change it such that at > the end it prints the numbers in a copy/paste-able C form, just for the > lazy amongst us :-) Sure thing, :)