From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754681Ab3AZBtB (ORCPT ); Fri, 25 Jan 2013 20:49:01 -0500 Received: from mga09.intel.com ([134.134.136.24]:41584 "EHLO mga09.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753242Ab3AZBs7 (ORCPT ); Fri, 25 Jan 2013 20:48:59 -0500 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.84,542,1355126400"; d="scan'208";a="276820836" From: Alex Shi To: mingo@redhat.com, peterz@infradead.org, tglx@linutronix.de, akpm@linux-foundation.org, arjan@linux.intel.com, bp@alien8.de, pjt@google.com, namhyung@kernel.org, efault@gmx.de Cc: vincent.guittot@linaro.org, gregkh@linuxfoundation.org, preeti@linux.vnet.ibm.com, viresh.kumar@linaro.org, linux-kernel@vger.kernel.org, alex.shi@intel.com Subject: [RFC patch v2] sched: use runnable load avg in cfs balance instead of instant load Date: Sat, 26 Jan 2013 09:49:17 +0800 Message-Id: <1359164964-29052-1-git-send-email-alex.shi@intel.com> X-Mailer: git-send-email 1.7.12 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org This patchset can be used, but causes burst waking benchmark aim9 drop 5~7% on my 2 sockets machine. The reason is too light runnable load in early stage of waked tasks causes imbalance in balancing. So, it is immature and just a reference for guys who want to go gurther. V2 change: 1, attached the 1~3 patches, which were sent in power awareness scheduling 2, remove CONFIG_FAIR_GROUP_SCHED mask in patch 5th. Thanks Ingo's comments and testing provided by Fengguang's kbuild system. Now it is indepent patchset bases on Linus' tree. Thanks Alex [RFC patch v2 1/7] sched: give initial value for runnable avg of [RFC patch v2 2/7] sched: set initial load avg of new forked task [RFC patch v2 3/7] Revert "sched: Introduce temporary [RFC patch v2 4/7] sched: update cpu load after task_tick. [RFC patch v2 5/7] sched: compute runnable load avg in cpu_load and [RFC patch v2 6/7] sched: consider runnable load average in [RFC patch v2 7/7] sched: consider runnable load average in