From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S935948AbXGXJne (ORCPT ); Tue, 24 Jul 2007 05:43:34 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1759321AbXGXJnU (ORCPT ); Tue, 24 Jul 2007 05:43:20 -0400 Received: from one.firstfloor.org ([213.235.205.2]:50384 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1760241AbXGXJnT (ORCPT ); Tue, 24 Jul 2007 05:43:19 -0400 Date: Tue, 24 Jul 2007 11:43:14 +0200 From: Andi Kleen To: Chris Friesen Cc: "Li, Tong N" , Andi Kleen , mingo@elte.hu, linux-kernel@vger.kernel.org Subject: Re: [RFC] scheduler: improve SMP fairness in CFS Message-ID: <20070724094313.GB26724@one.firstfloor.org> References: <1185225015.3593.15.camel@tongli.jf.intel.com> <46A51CB8.1090400@nortel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <46A51CB8.1090400@nortel.com> User-Agent: Mutt/1.4.2.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Jul 23, 2007 at 03:25:12PM -0600, Chris Friesen wrote: > Li, Tong N wrote: > >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. > > If locking does become a problem, it may make sense to switch instead to > a lockless algorithm of some sort... That usually has the same issue with cache lines. It would be better to analyze first if such global synchronization is really needed. I hope not. -Andi