From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752921AbZDTQPw (ORCPT ); Mon, 20 Apr 2009 12:15:52 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752006AbZDTQPe (ORCPT ); Mon, 20 Apr 2009 12:15:34 -0400 Received: from e6.ny.us.ibm.com ([32.97.182.146]:36883 "EHLO e6.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752852AbZDTQPd (ORCPT ); Mon, 20 Apr 2009 12:15:33 -0400 Subject: Re: [PATCH V3] Fix Committed_AS underflow From: Dave Hansen To: Eric B Munson Cc: kosaki.motohiro@jp.fujitsu.com, linux-mm@kvack.org, linux-kernel@vger.kernel.org, mel@linux.vnet.ibm.com, cl@linux-foundation.org In-Reply-To: <1240218590-16714-1-git-send-email-ebmunson@us.ibm.com> References: <1240218590-16714-1-git-send-email-ebmunson@us.ibm.com> Content-Type: text/plain Date: Mon, 20 Apr 2009 09:15:20 -0700 Message-Id: <1240244120.32604.278.camel@nimitz> Mime-Version: 1.0 X-Mailer: Evolution 2.22.3.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 2009-04-20 at 10:09 +0100, Eric B Munson wrote: > 1. Change NR_CPUS to min(64, NR_CPUS) > This will limit the amount of possible skew on kernels compiled for very > large SMP machines. 64 is an arbitrary number selected to limit the worst > of the skew without using more cache lines. min(64, NR_CPUS) is used > instead of nr_online_cpus() because nr_online_cpus() requires a shared > cache line and a call to hweight to make the calculation. Its runtime > overhead and keeping this counter accurate showed up in profiles and it's > possible that nr_online_cpus() would also show. -- Dave