Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752143AbcDKGSh (ORCPT ); Mon, 11 Apr 2016 02:18:37 -0400 Received: from mga01.intel.com ([192.55.52.88]:10158 "EHLO mga01.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750760AbcDKGSg (ORCPT ); Mon, 11 Apr 2016 02:18:36 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.24,462,1455004800"; d="scan'208";a="929706648" From: Yuyang Du To: peterz@infradead.org, mingo@kernel.org, linux-kernel@vger.kernel.org Cc: bsegall@google.com, pjt@google.com, morten.rasmussen@arm.com, vincent.guittot@linaro.org, dietmar.eggemann@arm.com, juri.lelli@arm.com, Yuyang Du Subject: [PATCH 0/4] Optimize sched avg computation and implement flat util hierarchy Date: Mon, 11 Apr 2016 06:36:01 +0800 Message-Id: <1460327765-18024-1-git-send-email-yuyang.du@intel.com> X-Mailer: git-send-email 1.7.9.5 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 911 Lines: 26 This patch series takes an effort to implement a flat util hierarchy, in order to reflect group task's migration immediately. Simply doing so may increase overhead in computing sched averages, or lose accuracy if only updating util of root cfs_rq. So, to not add overhead or to even reduce it, we must optimize __update_load_avg() greatly. Hopefully, these patches can achieve it. --- Yuyang Du (4): sched/fair: Optimize sum computation with a lookup table sched/fair: Drop out incomplete current period when sched averages accrue sched/fair: Modify accumulated sums for load/util averages sched/fair: Implement flat hierarchical structure for util_avg include/linux/sched.h | 2 +- kernel/sched/debug.c | 11 +- kernel/sched/fair.c | 313 ++++++++++++++++++++++++-------------------------- kernel/sched/sched.h | 5 +- 4 files changed, 163 insertions(+), 168 deletions(-) -- 2.1.4