From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756737AbaEIPHB (ORCPT ); Fri, 9 May 2014 11:07:01 -0400 Received: from mx1.redhat.com ([209.132.183.28]:9429 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755219AbaEIPHA (ORCPT ); Fri, 9 May 2014 11:07:00 -0400 Message-ID: <536CEE5E.9010609@redhat.com> Date: Fri, 09 May 2014 11:03:58 -0400 From: Rik van Riel User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.2.0 MIME-Version: 1.0 To: Peter Zijlstra CC: linux-kernel@vger.kernel.org, mingo@kernel.org, mgorman@suse.de, chegu_vinod@hp.com Subject: Re: [PATCH 2/4] sched,numa: weigh nearby nodes for task placement on complex NUMA topologies References: <1399569811-14362-1-git-send-email-riel@redhat.com> <1399569811-14362-3-git-send-email-riel@redhat.com> <20140509101317.GU30445@twins.programming.kicks-ass.net> In-Reply-To: <20140509101317.GU30445@twins.programming.kicks-ass.net> 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 On 05/09/2014 06:13 AM, Peter Zijlstra wrote: > On Thu, May 08, 2014 at 01:23:29PM -0400, riel@redhat.com wrote: >> This patch does nothing on machines with simple NUMA topologies. > > Was this: > >> + /* >> + * No need to calculate a score if the system has a simple NUMA >> + * topology, with no node distances between "local" and "far away". >> + */ >> + if (max_distance == LOCAL_DISTANCE) >> + return 0; > > Supposed to make that true? > > It doesn't. That test is a !numa test, not a fully connected test. > Look at patch 1/4. I only set max_distance to !LOCAL_DISTANCE if the system has multiple different distances in the SLIT table. I guess that needs to be cleaned up :)