Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932090Ab1E0XkM (ORCPT ); Fri, 27 May 2011 19:40:12 -0400 Received: from cantor.suse.de ([195.135.220.2]:36404 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753884Ab1E0XkI (ORCPT ); Fri, 27 May 2011 19:40:08 -0400 Date: Sat, 28 May 2011 01:40:07 +0200 From: Michal Hocko To: David Rientjes Cc: Andrew Morton , KOSAKI Motohiro , LKML , Jack Steiner , Lee Schermerhorn , Christoph Lameter , Pekka Enberg , Paul Menage , Robin Holt , Linus Torvalds , linux-mm@kvack.org Subject: Re: [PATCH v2] cpusets: randomize node rotor used in cpuset_mem_spread_node() Message-ID: <20110527234007.GB4276@tiehlicka.suse.cz> References: <20110414065146.GA19685@tiehlicka.suse.cz> <20110414160145.0830.A69D9226@jp.fujitsu.com> <20110415161831.12F8.A69D9226@jp.fujitsu.com> <20110415082051.GB8828@tiehlicka.suse.cz> <20110526153319.b7e8c0b6.akpm@linux-foundation.org> <20110527124705.GB4067@tiehlicka.suse.cz> <20110527231708.GB3214@tiehlicka.suse.cz> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1332 Lines: 34 On Fri 27-05-11 16:27:34, David Rientjes wrote: > On Sat, 28 May 2011, Michal Hocko wrote: > > > > CONFIG_NODES_SHIFT is used for UMA machines that are using DISCONTIGMEM > > > usually because they have very large holes; such machines don't need > > > things like mempolicies but do need the data structures that abstract > > > ranges of memory in the physical address space. This build breakage > > > probably isn't restricted to only alpha, you could probably see it with at > > > least ia64 and mips as well. > > > > Hmmm. I just find strange that some UMA arch uses functions like > > {first,next}_online_node. > > > > They shouldn't, but they do use NUMA data structures like pg_data_t for > DISCONTIGMEM. The MAX_NUMNODES > 1 optimization in nodemask.h is to > prevent doing things like node_weight() on a nodemask when we know that > only one bit will ever be set, otherwise we could make it conditional on > CONFIG_NEED_MULTIPLE_NODES. Thanks for the explanation. -- Michal Hocko SUSE Labs SUSE LINUX s.r.o. Lihovarska 1060/12 190 00 Praha 9 Czech Republic -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/