From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932525AbdDZR7U (ORCPT ); Wed, 26 Apr 2017 13:59:20 -0400 Received: from mx1.redhat.com ([209.132.183.28]:53126 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932497AbdDZR7L (ORCPT ); Wed, 26 Apr 2017 13:59:11 -0400 DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com 4E0D31FEE70 Authentication-Results: ext-mx10.extmail.prod.ext.phx2.redhat.com; dmarc=none (p=none dis=none) header.from=redhat.com Authentication-Results: ext-mx10.extmail.prod.ext.phx2.redhat.com; spf=pass smtp.mailfrom=lvenanci@redhat.com DKIM-Filter: OpenDKIM Filter v2.11.0 mx1.redhat.com 4E0D31FEE70 Reply-To: lvenanci@redhat.com Subject: Re: [PATCH 0/4] sched/topology: fix overlap group capacity and balance cpu References: <1492717903-5195-1-git-send-email-lvenanci@redhat.com> <20170426163106.zwtlrvkc2wuezn6g@hirez.programming.kicks-ass.net> To: Peter Zijlstra Cc: lwang@redhat.com, riel@redhat.com, Mike Galbraith , Thomas Gleixner , Ingo Molnar , linux-kernel@vger.kernel.org From: Lauro Venancio Organization: Red Hat Message-ID: <1b6b08ca-7a13-5cf8-06c7-421258f79655@redhat.com> Date: Wed, 26 Apr 2017 14:59:09 -0300 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.4.0 MIME-Version: 1.0 In-Reply-To: <20170426163106.zwtlrvkc2wuezn6g@hirez.programming.kicks-ass.net> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.39]); Wed, 26 Apr 2017 17:59:11 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 04/26/2017 01:31 PM, Peter Zijlstra wrote: > > Please have a look here: > > git://git.kernel.org/pub/scm/linux/kernel/git/peterz/queue.git sched/wip > > I think that has them all. Yes. I agree. > > I'm still thinking on how to best create a test to assert on the SGC > condition, but I don't seem to get much further than brute-force > matching all group masks. At least we now display (albeit I don't like > the format, suggestions?) the stuff. I cannot think in anything other than brute-force, too. Maybe, to make the debug messages clear, you could print "balance_cpu: %d" instead of "capacity(%d): " and assert (sgc id == balance cpu). You would need to restart the sgc id count on every topology level.