From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753384AbeFDOhe (ORCPT ); Mon, 4 Jun 2018 10:37:34 -0400 Received: from shelob.surriel.com ([96.67.55.147]:50934 "EHLO shelob.surriel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751354AbeFDOhc (ORCPT ); Mon, 4 Jun 2018 10:37:32 -0400 Message-ID: <1528123050.7898.99.camel@surriel.com> Subject: Re: [PATCH 04/19] sched/numa: Set preferred_node based on best_cpu From: Rik van Riel To: Srikar Dronamraju , Peter Zijlstra Cc: Ingo Molnar , LKML , Mel Gorman , Thomas Gleixner Date: Mon, 04 Jun 2018 10:37:30 -0400 In-Reply-To: <20180604125939.GB38574@linux.vnet.ibm.com> References: <1528106428-19992-1-git-send-email-srikar@linux.vnet.ibm.com> <1528106428-19992-5-git-send-email-srikar@linux.vnet.ibm.com> <20180604122336.GS12217@hirez.programming.kicks-ass.net> <20180604125939.GB38574@linux.vnet.ibm.com> Content-Type: multipart/signed; micalg="pgp-sha256"; protocol="application/pgp-signature"; boundary="=-yx+VoeBPYZ/jkcI2Rzj0" X-Mailer: Evolution 3.26.6 (3.26.6-1.fc27) Mime-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --=-yx+VoeBPYZ/jkcI2Rzj0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Mon, 2018-06-04 at 05:59 -0700, Srikar Dronamraju wrote: > * Peter Zijlstra [2018-06-04 14:23:36]: >=20 > > OK, the above matches the description, but I'm puzzled by the > > remainder: > >=20 > > >=20 > > > - if (ng->active_nodes > 1 && > > > numa_is_active_node(env.dst_nid, ng)) > > > - sched_setnuma(p, env.dst_nid); > > > + if (nid !=3D p->numa_preferred_nid) > > > + sched_setnuma(p, nid); > > > } > >=20 > > That seems to entirely loose the active_node thing, or are you > > saying > > best_cpu already includes that? (Changelog could use a little help > > there > > I suppose) >=20 > I think checking for active_nodes before calling sched_setnuma was a > mistake. >=20 > Before this change, we may be retaining numa_preferred_nid to be the > source node while we select another node with better numa affinity to > run on.=20 Sometimes workloads are so large they get spread around to multiple NUMA nodes. In that case, you do NOT want all the tasks of that workload (numa group) to try squeezing onto the same load, only to have the load balancer randomly move tasks off of that node again later. How do you keep that from happening? --=20 All Rights Reversed. --=-yx+VoeBPYZ/jkcI2Rzj0 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part Content-Transfer-Encoding: 7bit -----BEGIN PGP SIGNATURE----- iQEzBAABCAAdFiEEKR73pCCtJ5Xj3yADznnekoTE3oMFAlsVTqoACgkQznnekoTE 3oMv6gf/e+XjU8bS5G/mkPOc9AlMIK8KPBAMWH56QdxipZk7O8kbBOXSgMCNOcBf l862+/CK6bTBky/Ol2ztzCVbvK0vqLHljZantJ3da5rjqq70zR45+FKmHmS67AjE MRhUnyR0y0AU/QLJ/q5qK1F01zUColATBstisDl/qDWAarOI7Jl+fYqSKMzvjlV1 IlYbGXRybYzUj5xuCGnflGrYu1LoLvPVqFvNTdR50I1MGeGKXw9ZrW5ESo7yI9wT w4FsCJqh+pcgvyksSrYsFzQpF3M+/2emQ+FPL6NzhVeFlR9MAbD2T4q1yvjwUZar 8MScwgjxLj8W58d6vAFDW37MFYW9fQ== =36/e -----END PGP SIGNATURE----- --=-yx+VoeBPYZ/jkcI2Rzj0--