From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751722AbeFDR7c (ORCPT ); Mon, 4 Jun 2018 13:59:32 -0400 Received: from shelob.surriel.com ([96.67.55.147]:53672 "EHLO shelob.surriel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751058AbeFDR7b (ORCPT ); Mon, 4 Jun 2018 13:59:31 -0400 Message-ID: <1528135170.7898.113.camel@surriel.com> Subject: Re: [PATCH 11/19] sched/numa: Restrict migrating in parallel to the same node. From: Rik van Riel To: Srikar Dronamraju , Ingo Molnar , Peter Zijlstra Cc: LKML , Mel Gorman , Thomas Gleixner Date: Mon, 04 Jun 2018 13:59:30 -0400 In-Reply-To: <1528106428-19992-12-git-send-email-srikar@linux.vnet.ibm.com> References: <1528106428-19992-1-git-send-email-srikar@linux.vnet.ibm.com> <1528106428-19992-12-git-send-email-srikar@linux.vnet.ibm.com> Content-Type: multipart/signed; micalg="pgp-sha256"; protocol="application/pgp-signature"; boundary="=-M9gQjdaWU+P8s3O/lyZE" 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 --=-M9gQjdaWU+P8s3O/lyZE Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Mon, 2018-06-04 at 15:30 +0530, Srikar Dronamraju wrote: > Since task migration under numa balancing can happen in parallel, > more > than one task might choose to move to the same node at the same time. > This can cause load imbalances at the node level. >=20 > The problem is more likely if there are more cores per node or more > nodes in system. >=20 > Use a per-node variable to indicate if task migration > to the node under numa balance is currently active. > This per-node variable will not track swapping of tasks. > The commit does cause some performance regression but is needed from > a fairness/correctness perspective. Does it help any "real workloads", even simple things like SpecJBB2005? If this patch only causes regressions, and does not help any workloads, I would argue that it is not in fact needed. --=20 All Rights Reversed. --=-M9gQjdaWU+P8s3O/lyZE Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part Content-Transfer-Encoding: 7bit -----BEGIN PGP SIGNATURE----- iQEzBAABCAAdFiEEKR73pCCtJ5Xj3yADznnekoTE3oMFAlsVfgIACgkQznnekoTE 3oPMvwgAkzjioIrsHYjus5SI4uFsOza1t5Nub0dXNfyxikjymRWDzCNeDgLOiuY2 m1wMnplkuwMqrrWtJcIEzQDIuHGTc/w7W63WKt4jQDT7oWYY5/gkHqbamQvxDryK WkZFS4WXrO25zycqlBCkvuVo4vkhckS03mJ0G/QZLvVW3pLMTV+i+rpnVH9NToXY sH/wCdZ7BxgNp1bFjT4g68Rg64NHSmCwCozuAMXxnneJLmvoo+uSbpA62xKp5TE0 3NN1rFE+ytRPijXPbr/MgnN6ozh2YxpmsyxBJ4HF4OPmF/i6vOUpxNd0YT45TY4g xHn+cOxVFUPjynValYUUq0CQNT0Y7g== =Y12M -----END PGP SIGNATURE----- --=-M9gQjdaWU+P8s3O/lyZE--