From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756327Ab1LBBKa (ORCPT ); Thu, 1 Dec 2011 20:10:30 -0500 Received: from mga02.intel.com ([134.134.136.20]:34883 "EHLO mga02.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755224Ab1LBBKU (ORCPT ); Thu, 1 Dec 2011 20:10:20 -0500 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.67,352,1309762800"; d="scan'208";a="81983579" Message-Id: <20111202010731.344451602@sbsiddha-desk.sc.intel.com> User-Agent: quilt/0.48-1 Date: Thu, 01 Dec 2011 17:07:31 -0800 From: Suresh Siddha To: Peter Zijlstra , Ingo Molnar , Venki Pallipadi , Srivatsa Vaddagiri , Mike Galbraith Cc: linux-kernel , Tim Chen , alex.shi@intel.com Subject: [patch v3 0/6] nohz idle load balancing patches Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org This is the -v3 of the load balancing patches that makes nohz idle load balancing more scalabale. While the last 2 in the series are not directly related to nohz, I clubbed them as part of this series as I noticed them while I was working in this area and they helped the context switch intensive workloads improve on the 8-socket system. First RFC version of this can be found at https://lkml.org/lkml/2011/11/1/368 Second version of this can be found at https://lkml.org/lkml/2011/11/18/518 changes from v2: a. Make the changelogs more descriptive. b. nr_busy_cpus in the sched_group_power is updated when entering idle irrespective of whether the tick is stopped or not. c. kick the idle load balancer for the groups sharing package resources and having more than one cpu busy. d. Incroporate Mike's patch for enhancing ttwu_feature instead of disabling the feature by default. thanks, suresh