linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Mel Gorman <mel@csn.ul.ie>
To: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: akpm@linux-foundation.org, ak@suse.de, clameter@sgi.com,
	kamezawa.hiroyu@jp.fujitsu.com, linux-mm@kvack.org,
	rientjes@google.com, eric.whitney@hp.com
Subject: Re: [PATCH 0/6] Use two zonelists per node instead of multiple zonelists v11r3
Date: Fri, 29 Feb 2008 14:12:51 +0000	[thread overview]
Message-ID: <20080229141250.GA6045@csn.ul.ie> (raw)
In-Reply-To: <20080227214708.6858.53458.sendpatchset@localhost>

On (27/02/08 16:47), Lee Schermerhorn didst pronounce:
> From: Mel Gorman <mel@csn.ul.ie>
> [PATCH 0/6] Use two zonelists per node instead of multiple zonelists v11r3
> 
> This is a rebase of the two-zonelist patchset to 2.6.25-rc2-mm1.
> 
> Mel, still on vacation last I checked,  asked me to repost these
> as I'd already rebased them and I've been testing them continually
> on each -mm tree for months, hoping to see them in -mm for wider
> testing.
> 

Thanks a lot, Lee. I tested this patchset against a slightly patched
2.6.25-rc2-mm1 (compile-failure fix and a memoryless-related bug that is
fixed in git-x86#testing).

> These are the range of performance losses/gains when running against
> 2.6.24-rc4-mm1. The set and these machines are a mix of i386, x86_64 and
> ppc64 both NUMA and non-NUMA.
> 

Against 2.6.25-rc5-mm1, the results are
				loss	to	gain
Total CPU time on Kernbench:	-0.23%		 1.04%
Elapsed time on Kernbench:	-0.69%		 3.86%
page_test from aim9:		-3.74%		 5.72%
brk_test from aim9:		-7.37%		10.98%
fork_test from aim9:		-3.52%		 3.17%
exec_test from aim9:		-2.78%		 2.34%
TBench:				-1.93%		 2.96%

Hackbench was similarly variable but most machines showed little or no
difference. As before, whether the changes are a performance win/loss
depends on the machine but the majority of results showed little
difference.

> 			     loss   to  gain
> Total CPU time on Kernbench: -0.86% to  1.13%
> Elapsed   time on Kernbench: -0.79% to  0.76%
> page_test from aim9:         -4.37% to  0.79%
> brk_test  from aim9:         -0.71% to  4.07%
> fork_test from aim9:         -1.84% to  4.60%
> exec_test from aim9:         -0.71% to  1.08%
> 

-- 
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: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

      parent reply	other threads:[~2008-02-29 14:12 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-02-27 21:47 Lee Schermerhorn, Mel Gorman
2008-02-27 21:47 ` [PATCH 1/6] Use zonelists instead of zones when direct reclaiming pages Lee Schermerhorn, Mel Gorman
2008-02-27 21:47 ` [PATCH 2/6] Introduce node_zonelist() for accessing the zonelist for a GFP mask Lee Schermerhorn, Mel Gorman
2008-02-27 21:47 ` [PATCH 3/6] Remember what the preferred zone is for zone_statistics Lee Schermerhorn, Mel Gorman
2008-02-27 22:00   ` Christoph Lameter
2008-02-28 17:45     ` Lee Schermerhorn
2008-02-29 14:19     ` Mel Gorman
2008-02-29  2:30   ` KAMEZAWA Hiroyuki
2008-02-29 14:32     ` Mel Gorman
2008-02-27 21:47 ` [PATCH 4/6] Use two zonelist that are filtered by GFP mask Lee Schermerhorn, Mel Gorman
2008-02-28 21:32   ` Andrew Morton
2008-02-28 21:53     ` Lee Schermerhorn
2008-02-29  2:37       ` KAMEZAWA Hiroyuki
2008-02-29 14:50     ` Mel Gorman
2008-02-29 15:48       ` Lee Schermerhorn
2008-02-29 21:07         ` Christoph Lameter
2008-03-04 18:01         ` Mel Gorman
2008-03-05 16:06           ` [PATCH] 2.6.25-rc3-mm1 - Mempolicy: make dequeue_huge_page_vma() obey MPOL_BIND nodemask Lee Schermerhorn
2008-03-05 18:03             ` Nishanth Aravamudan
2008-03-05 19:02               ` Lee Schermerhorn
2008-03-06  1:04                 ` Nishanth Aravamudan
2008-03-06 15:38                   ` Lee Schermerhorn
2008-03-06 21:24                   ` [PATCH] Mempolicy: make dequeue_huge_page_vma() obey MPOL_BIND nodemask rework Lee Schermerhorn
2008-03-07 17:35                     ` Nishanth Aravamudan
2008-03-07 18:31                       ` Lee Schermerhorn
2008-03-08  0:27                         ` Nishanth Aravamudan
2008-03-06  0:39             ` [PATCH] 2.6.25-rc3-mm1 - Mempolicy: make dequeue_huge_page_vma() obey MPOL_BIND nodemask Andrew Morton
2008-03-06 15:17               ` Lee Schermerhorn
2008-03-06 18:41     ` [PATCH 4/6] Use two zonelist that are filtered by GFP mask Mel Gorman
2008-02-27 21:47 ` [PATCH 5/6] Have zonelist contains structs with both a zone pointer and zone_idx Lee Schermerhorn, Mel Gorman
2008-02-29  7:49   ` KOSAKI Motohiro
2008-02-27 21:47 ` [PATCH 6/6] Filter based on a nodemask as well as a gfp_mask Lee Schermerhorn, Mel Gorman
2008-02-29  2:59   ` KAMEZAWA Hiroyuki
2008-03-07 11:56     ` Mel Gorman
2008-02-29  8:48   ` KOSAKI Motohiro
2008-02-27 21:53 ` [PATCH 0/6] Use two zonelists per node instead of multiple zonelists v11r3 Lee Schermerhorn
2008-02-29 14:12 ` Mel Gorman [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20080229141250.GA6045@csn.ul.ie \
    --to=mel@csn.ul.ie \
    --cc=ak@suse.de \
    --cc=akpm@linux-foundation.org \
    --cc=clameter@sgi.com \
    --cc=eric.whitney@hp.com \
    --cc=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=lee.schermerhorn@hp.com \
    --cc=linux-mm@kvack.org \
    --cc=rientjes@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox