From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S936187AbXGWVKg (ORCPT ); Mon, 23 Jul 2007 17:10:36 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1759749AbXGWVKZ (ORCPT ); Mon, 23 Jul 2007 17:10:25 -0400 Received: from mga03.intel.com ([143.182.124.21]:21260 "EHLO mga03.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757576AbXGWVKY (ORCPT ); Mon, 23 Jul 2007 17:10:24 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.16,572,1175497200"; d="scan'208";a="253756357" Subject: Re: [RFC] scheduler: improve SMP fairness in CFS From: "Li, Tong N" To: Andi Kleen Cc: mingo@elte.hu, linux-kernel@vger.kernel.org In-Reply-To: References: Content-Type: text/plain Content-Transfer-Encoding: 7bit Date: Mon, 23 Jul 2007 14:10:15 -0700 Message-Id: <1185225015.3593.15.camel@tongli.jf.intel.com> Mime-Version: 1.0 X-Mailer: Evolution 2.10.2 (2.10.2-3.fc7) X-OriginalArrivalTime: 23 Jul 2007 21:10:16.0313 (UTC) FILETIME=[DE01AE90:01C7CD6D] Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org I benchmarked an early version of this code (against 2.6.21) with SPECjbb, SPEComp, kernbench, etc. on an 8-processor system, and didn't see any slowdown compared to the stock scheduler. I'll generate the data again with this version of the code. On the other hand, if locking does become a problem for certain systems/workloads, increasing sysctl_base_round_slice can reduce the locking frequency and alleviate the problem, at the cost of being relatively less fair across the CPUs. tong On Mon, 2007-07-23 at 22:00 +0200, Andi Kleen wrote: > Tong Li writes: > > + > > + read_lock_irqsave(&dwrr_highest_lock, flags); > > That lock looks like it would bounce between CPUs like crazy. > Did you try any benchmarks on a larger system? > > -Andi