From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753329AbaGJMoN (ORCPT ); Thu, 10 Jul 2014 08:44:13 -0400 Received: from cantor2.suse.de ([195.135.220.15]:35484 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752904AbaGJMoM (ORCPT ); Thu, 10 Jul 2014 08:44:12 -0400 Date: Thu, 10 Jul 2014 13:44:07 +0100 From: Mel Gorman To: Johannes Weiner Cc: Andrew Morton , Linux Kernel , Linux-MM , Linux-FSDevel Subject: Re: [PATCH 5/6] mm: page_alloc: Abort fair zone allocation policy when remotes nodes are encountered Message-ID: <20140710124407.GH10819@suse.de> References: <1404893588-21371-1-git-send-email-mgorman@suse.de> <1404893588-21371-6-git-send-email-mgorman@suse.de> <20140710121419.GM29639@cmpxchg.org> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline In-Reply-To: <20140710121419.GM29639@cmpxchg.org> 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 On Thu, Jul 10, 2014 at 08:14:19AM -0400, Johannes Weiner wrote: > On Wed, Jul 09, 2014 at 09:13:07AM +0100, Mel Gorman wrote: > > The purpose of numa_zonelist_order=zone is to preserve lower zones > > for use with 32-bit devices. If locality is preferred then the > > numa_zonelist_order=node policy should be used. Unfortunately, the fair > > zone allocation policy overrides this by skipping zones on remote nodes > > until the lower one is found. While this makes sense from a page aging > > and performance perspective, it breaks the expected zonelist policy. This > > patch restores the expected behaviour for zone-list ordering. > > > > Signed-off-by: Mel Gorman > > 32-bit NUMA? :-) I'm tempted to just say "it can go on fire" but realistically speaking they should be configured to use node ordering. I was very tempted to always force node ordering but I didn't have good data on how often lowmem allocations are required on NUMA machines. > Anyway, this change also cuts down the fair pass > overhead on bigger NUMA machines, so I'm all for it. > > Acked-by: Johannes Weiner Thanks for the reviews! -- Mel Gorman SUSE Labs