From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753876AbaIPXtJ (ORCPT ); Tue, 16 Sep 2014 19:49:09 -0400 Received: from www.sr71.net ([198.145.64.142]:54518 "EHLO blackbird.sr71.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753032AbaIPXtH (ORCPT ); Tue, 16 Sep 2014 19:49:07 -0400 Message-ID: <5418CC6F.2040803@sr71.net> Date: Tue, 16 Sep 2014 16:49:03 -0700 From: Dave Hansen User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.6.0 MIME-Version: 1.0 To: Peter Zijlstra CC: linux-kernel@vger.kernel.org, mingo@kernel.org, hpa@linux.intel.com, ak@linux.intel.com, Alex Chiang , Borislav Petkov , Rusty Russell , Mike Travis , Greg Kroah-Hartman , Heiko Carstens Subject: Re: [PATCH] x86: Consider multiple nodes in a single socket to be "sane" References: <20140915222641.D640BD8A@viggo.jf.intel.com> <20140916032920.GH2840@worktop.localdomain> <5417F228.9060301@sr71.net> <20140916175851.GC2848@worktop.localdomain> In-Reply-To: <20140916175851.GC2848@worktop.localdomain> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 09/16/2014 10:58 AM, Peter Zijlstra wrote: > the 'problem' is that we currently have the static order of > the masks, if we were to flip the MC and NUMA masks we need a condition > to do that on and make sure everything is aware of that. > > CoD not being detectable sucks arse for sure :/ It's not like we can't detect it. We just can't detect it *explicitly*. I think when we see nodes inside a package now, we have to trust that they're OK. One other data point here. With an unpatched mainline, here's how the sched domains look with cluster-on-die enabled: # grep . /proc/sys/kernel/sched_domain/cpu9/domain?/name /proc/sys/kernel/sched_domain/cpu9/domain0/name:SMT // 2 threads /proc/sys/kernel/sched_domain/cpu9/domain1/name:MC // 18 threads cores /proc/sys/kernel/sched_domain/cpu9/domain2/name:NUMA // 36 threads /proc/sys/kernel/sched_domain/cpu9/domain3/name:NUMA // 72 threads and with cluster-on-die disabled: # grep . /proc/sys/kernel/sched_domain/cpu9/domain?/name /proc/sys/kernel/sched_domain/cpu9/domain0/name:SMT /proc/sys/kernel/sched_domain/cpu9/domain1/name:MC /proc/sys/kernel/sched_domain/cpu9/domain2/name:NUMA So, shockingly, the domains seem to be set up at at least conceptually OK in both cases. I think the domains in this case should _probably_ be conceptually: SMT -> COD_NUMA -> PKG -> SOCKET_NUMA We could probably rig up sched_init_numa() to mix topology levels between the ones that come out of sched_domain_topology and the NUMA levels, although that doesn't sound very appealing. Another option would be to: 1. Add a new "PKG" level and actually _build_ it with phys_proc_id 2. Make sure to tie the sysfs 'core_siblings' file to PKG 3. Leave the "MC" level as it is now, but define it as being the lowest- common-denominator of core grouping. In other words, the "MC" group will stop at a NUMA node or a socket boundary, whichever it sees first. 4. Chop the "COD_NUMA" level off in sched_init_numa() That seems a bit hackish though.