From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754643AbZDWATf (ORCPT ); Wed, 22 Apr 2009 20:19:35 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752591AbZDWATN (ORCPT ); Wed, 22 Apr 2009 20:19:13 -0400 Received: from gir.skynet.ie ([193.1.99.77]:38719 "EHLO gir.skynet.ie" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752079AbZDWATM (ORCPT ); Wed, 22 Apr 2009 20:19:12 -0400 Date: Thu, 23 Apr 2009 01:19:08 +0100 From: Mel Gorman To: David Rientjes Cc: Linux Memory Management List , KOSAKI Motohiro , Christoph Lameter , Nick Piggin , Linux Kernel Mailing List , Lin Ming , Zhang Yanmin , Peter Zijlstra , Pekka Enberg , Andrew Morton Subject: Re: [PATCH 16/22] Do not setup zonelist cache when there is only one node Message-ID: <20090423001907.GB26643@csn.ul.ie> References: <1240408407-21848-1-git-send-email-mel@csn.ul.ie> <1240408407-21848-17-git-send-email-mel@csn.ul.ie> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.17+20080114 (2008-01-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Apr 22, 2009 at 01:24:26PM -0700, David Rientjes wrote: > On Wed, 22 Apr 2009, Mel Gorman wrote: > > > There is a zonelist cache which is used to track zones that are not in > > the allowed cpuset or found to be recently full. This is to reduce cache > > footprint on large machines. On smaller machines, it just incurs cost > > for no gain. This patch only uses the zonelist cache when there are NUMA > > nodes. > > > > Signed-off-by: Mel Gorman > > Reviewed-by: Christoph Lameter > > --- > > mm/page_alloc.c | 7 +++++-- > > 1 files changed, 5 insertions(+), 2 deletions(-) > > > > diff --git a/mm/page_alloc.c b/mm/page_alloc.c > > index 7f45de1..e59bb80 100644 > > --- a/mm/page_alloc.c > > +++ b/mm/page_alloc.c > > @@ -1467,8 +1467,11 @@ this_zone_full: > > if (NUMA_BUILD) > > zlc_mark_zone_full(zonelist, z); > > If zonelist caching is never used for UMA machines, why should they ever > call zlc_mark_zone_full()? It will always dereference > zonelist->zlcache_ptr and immediately return without doing anything. > > Wouldn't it better to just add > > if (num_online_nodes() == 1) > continue; > num_online_nodes() is actually a really heavy function. It calls hweight on a bitmap which is probably why it's not happening already. There is a nr_online_nodes later in the patchset though. With nr_online_nodes, it's a good idea to avoid a function call so I've taken note to do that patch in pass 2. Thanks > right before this call to zlc_mark_zone_full()? This should compile out > the remainder of the loop for !CONFIG_NUMA kernels anyway. > > > try_next_zone: > > - if (NUMA_BUILD && !did_zlc_setup) { > > - /* we do zlc_setup after the first zone is tried */ > > + if (NUMA_BUILD && !did_zlc_setup && num_online_nodes() > 1) { > > + /* > > + * we do zlc_setup after the first zone is tried but only > > + * if there are multiple nodes make it worthwhile > > + */ > > allowednodes = zlc_setup(zonelist, alloc_flags); > > zlc_active = 1; > > did_zlc_setup = 1; > > -- > > 1.5.6.5 > -- Mel Gorman Part-time Phd Student Linux Technology Center University of Limerick IBM Dublin Software Lab From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail191.messagelabs.com (mail191.messagelabs.com [216.82.242.19]) by kanga.kvack.org (Postfix) with ESMTP id 622AC6B010E for ; Wed, 22 Apr 2009 20:18:53 -0400 (EDT) Date: Thu, 23 Apr 2009 01:19:08 +0100 From: Mel Gorman Subject: Re: [PATCH 16/22] Do not setup zonelist cache when there is only one node Message-ID: <20090423001907.GB26643@csn.ul.ie> References: <1240408407-21848-1-git-send-email-mel@csn.ul.ie> <1240408407-21848-17-git-send-email-mel@csn.ul.ie> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline In-Reply-To: Sender: owner-linux-mm@kvack.org To: David Rientjes Cc: Linux Memory Management List , KOSAKI Motohiro , Christoph Lameter , Nick Piggin , Linux Kernel Mailing List , Lin Ming , Zhang Yanmin , Peter Zijlstra , Pekka Enberg , Andrew Morton List-ID: On Wed, Apr 22, 2009 at 01:24:26PM -0700, David Rientjes wrote: > On Wed, 22 Apr 2009, Mel Gorman wrote: > > > There is a zonelist cache which is used to track zones that are not in > > the allowed cpuset or found to be recently full. This is to reduce cache > > footprint on large machines. On smaller machines, it just incurs cost > > for no gain. This patch only uses the zonelist cache when there are NUMA > > nodes. > > > > Signed-off-by: Mel Gorman > > Reviewed-by: Christoph Lameter > > --- > > mm/page_alloc.c | 7 +++++-- > > 1 files changed, 5 insertions(+), 2 deletions(-) > > > > diff --git a/mm/page_alloc.c b/mm/page_alloc.c > > index 7f45de1..e59bb80 100644 > > --- a/mm/page_alloc.c > > +++ b/mm/page_alloc.c > > @@ -1467,8 +1467,11 @@ this_zone_full: > > if (NUMA_BUILD) > > zlc_mark_zone_full(zonelist, z); > > If zonelist caching is never used for UMA machines, why should they ever > call zlc_mark_zone_full()? It will always dereference > zonelist->zlcache_ptr and immediately return without doing anything. > > Wouldn't it better to just add > > if (num_online_nodes() == 1) > continue; > num_online_nodes() is actually a really heavy function. It calls hweight on a bitmap which is probably why it's not happening already. There is a nr_online_nodes later in the patchset though. With nr_online_nodes, it's a good idea to avoid a function call so I've taken note to do that patch in pass 2. Thanks > right before this call to zlc_mark_zone_full()? This should compile out > the remainder of the loop for !CONFIG_NUMA kernels anyway. > > > try_next_zone: > > - if (NUMA_BUILD && !did_zlc_setup) { > > - /* we do zlc_setup after the first zone is tried */ > > + if (NUMA_BUILD && !did_zlc_setup && num_online_nodes() > 1) { > > + /* > > + * we do zlc_setup after the first zone is tried but only > > + * if there are multiple nodes make it worthwhile > > + */ > > allowednodes = zlc_setup(zonelist, alloc_flags); > > zlc_active = 1; > > did_zlc_setup = 1; > > -- > > 1.5.6.5 > -- Mel Gorman Part-time Phd Student Linux Technology Center University of Limerick IBM Dublin Software Lab -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org