From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753020Ab1I1FHG (ORCPT ); Wed, 28 Sep 2011 01:07:06 -0400 Received: from e5.ny.us.ibm.com ([32.97.182.145]:47852 "EHLO e5.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752664Ab1I1FHE (ORCPT ); Wed, 28 Sep 2011 01:07:04 -0400 Date: Wed, 28 Sep 2011 10:36:58 +0530 From: Srivatsa Vaddagiri To: Venki Pallipadi Cc: Peter Zijlstra , Paul Turner , Ingo Molnar , Vaidyanathan Srinivasan , Kamalesh Babulal , linux-kernel@vger.kernel.org Subject: Re: [PATCH v1] sched: fix nohz idle load balancer issues Message-ID: <20110928050344.GD8593@linux.vnet.ibm.com> Reply-To: Srivatsa Vaddagiri References: <20110926115049.GA22604@linux.vnet.ibm.com> <20110928043427.GI4357@linux.vnet.ibm.com> <20110928043921.GC8593@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline In-Reply-To: <20110928043921.GC8593@linux.vnet.ibm.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Srivatsa Vaddagiri [2011-09-28 10:09:21]: > * Srivatsa Vaddagiri [2011-09-28 10:04:27]: > > > unless we want to go for more sophisticated like a sorted list of > > rq->next_balance in a rb-tree > > Given that this list will need to be global (and not per-cpu) and it will need > to be updated reqularly (even when cpus are busy and as they update their > rq->next_balance), that will probably not fly! On third thoughts, this list needs to be updated only in idle cpu context (idle cpus going tickless will be inserted in the list (indexed at rq->next_balance and cpu exiting idle will need to be removed). So the global list may not hurt so much ..will see how that comes up in code. - vatsa