Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934889AbXEUUJe (ORCPT ); Mon, 21 May 2007 16:09:34 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1759560AbXEUUI7 (ORCPT ); Mon, 21 May 2007 16:08:59 -0400 Received: from netops-testserver-3-out.sgi.com ([192.48.171.28]:50573 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1761217AbXEUUI5 (ORCPT ); Mon, 21 May 2007 16:08:57 -0400 Date: Mon, 21 May 2007 15:08:56 -0500 To: linux-kernel@vger.kernel.org Subject: hotplug cpu: PATCHes for 3 issues Message-ID: <4651FC58.mailxOW91VRGMT@eag09.americas.sgi.com> User-Agent: nail 11.4 8/29/04 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit From: cpw@sgi.com (Cliff Wickman) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1373 Lines: 36 In the 2.6.21 kernel there are still 3 hotplug issues that are cpuset- related, and that I find to still be problems. And for which I offer patches. These have been submitted before, and subsequently cleaned up per comments received. I'm resubmitting all 3 for consideration and further comment. 1) [PATCH 1/1] hotplug cpu: cpusets/sched_domain reconciliation 2) [PATCH 1/1] hotplug cpu: move tasks in empty cpusets to parent 3) [PATCH 1/1] hotplug cpu: migrate a task within its cpuset 1) Reconciles cpusets and sched_domains that get out of sync due to hotplug disabling and re-enabling of cpu's. Tasks can get into infinite hangs without this fix. kernel/cpuset.c kernel/sched.c 2) When a cpuset is emptied by disabling its cpus, move tasks to a parent cpuset. This is a correction of the current procedure, which moves such tasks to the wrong cpuset. kernel/cpuset.c 3) Causes a task running on a disabled cpu to migrate to a cpu within its cpuset. This behavior is particularly important for a NUMA system on which tasks have been explicitly placed. kernel/sched.c - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/