linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@osdl.org>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
	bob.picco@hp.com, iwamoto@valinux.co.jp, christoph@lameter.com,
	wfg@mail.ustc.edu.cn, npiggin@suse.de, torvalds@osdl.org,
	riel@redhat.com, marcelo.tosatti@cyclades.com
Subject: Re: [PATCH 00/34] mm: Page Replacement Policy Framework
Date: Wed, 22 Mar 2006 14:51:32 -0800	[thread overview]
Message-ID: <20060322145132.0886f742.akpm@osdl.org> (raw)
In-Reply-To: <20060322223107.12658.14997.sendpatchset@twins.localnet>

Peter Zijlstra <a.p.zijlstra@chello.nl> wrote:
>
> 
> This patch-set introduces a page replacement policy framework and 4 new 
> experimental policies.

Holy cow.

> The page replacement algorithm determines which pages to swap out.
> The current algorithm has some problems that are increasingly noticable, even
> on desktop workloads.

Rather than replacing the whole lot four times I'd really prefer to see
precise descriptions of these problems, see if we can improve the situation
incrementally rather than wholesale slash-n-burn...

Once we've done that work to the best of our ability, *then* we're in a
position to evaluate the performance benefits of this new work.  Because
there's not much point in comparing known-to-have-unaddressed-problems old
code with fancy new code.

> Measurements:
> 
> (Walltime, so lower is better)
> 
> cyclic-anon ; Cyclic access pattern with anonymous memory.
>               (http://programming.kicks-ass.net/benchmarks/cyclic-anon.c)
> 
> 2.6.16-rc6              14:28
> 2.6.16-rc6-useonce      15:11
> 2.6.16-rc6-clockpro     10:51
> 2.6.16-rc6-cart          8:55
> 2.6.16-rc6-random     1:09:50
> 
> cyclic-file ; Cyclic access pattern with file backed memory.
>               (http://programming.kicks-ass.net/benchmarks/cyclic-file.c)
> 
> 2.6.16-rc6              11:24
> 2.6.16-rc6-clockpro      8:14
> 2.6.16-rc6-cart          8:09
> 
> webtrace ; Replay of an IO trace from the Umass trace repository
>            (http://programming.kicks-ass.net/benchmarks/spc/)
> 
> 2.6.16-rc6               8:27
> 2.6.16-rc6-useonce       8:24
> 2.6.16-rc6-clockpro     10:23
> 2.6.16-rc6-cart         15:30
> 2.6.16-rc6-random       15:52
> 
> mdb-bench ; Low frequency benchmark.
>             (http://linux-mm.org/PageReplacementTesting)
> 
> 2.6.16-rc6            4:20:44
> 2.6.16-rc6 (mlock)    3:52:15
> 2.6.16-rc6-useonce    4:20:59
> 2.6.16-rc6-clockpro   3:56:17
> 2.6.16-rc6-cart       4:11:54
> 2.6.16-rc6-random     5:21:30

2.6.16-rc6 seems to do OK.  I assume the cyclic patterns exploit the lru
worst case thing?  Has consideration been given to tweaking the existing
code, detect the situation and work avoid the problem?

--
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:[~2006-03-22 22:51 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-03-22 22:31 Peter Zijlstra
2006-03-22 22:31 ` [PATCH 01/34] mm: kill-page-activate.patch Peter Zijlstra
2006-03-22 22:32 ` [PATCH 02/34] mm: page-replace-kconfig-makefile.patch Peter Zijlstra
2006-03-22 23:03   ` Jeff Garzik
2006-03-22 22:32 ` [PATCH 03/34] mm: page-replace-insert.patch Peter Zijlstra
2006-03-22 22:32 ` [PATCH 04/34] mm: page-replace-use_once.patch Peter Zijlstra
2006-03-22 22:32 ` [PATCH 05/34] mm: page-replace-generic-pagevec.patch Peter Zijlstra
2006-03-22 22:32 ` [PATCH 06/34] mm: page-replace-activate.patch Peter Zijlstra
2006-03-22 22:32 ` [PATCH 07/34] mm: page-replace-move-macros.patch Peter Zijlstra
2006-03-22 22:33 ` [PATCH 08/34] mm: page-replace-move-scan_control.patch Peter Zijlstra
2006-03-22 22:33 ` [PATCH 09/34] mm: page-replace-move-isolate_lru_pages.patch Peter Zijlstra
2006-03-22 22:33 ` [PATCH 10/34] mm: page-replace-reinsert.patch Peter Zijlstra
2006-03-22 22:33 ` [PATCH 11/34] mm: page-replace-should_reclaim_mapped.patch Peter Zijlstra
2006-03-22 22:33 ` [PATCH 12/34] mm: page-replace-shrink.patch Peter Zijlstra
2006-03-22 22:33 ` [PATCH 13/34] mm: page-replace-mark-accessed.patch Peter Zijlstra
2006-03-22 22:34 ` [PATCH 14/34] mm: page-replace-remove-mm_inline.patch Peter Zijlstra
2006-03-22 22:34 ` [PATCH 15/34] mm: page-replace-rotate.patch Peter Zijlstra
2006-03-22 22:34 ` [PATCH 16/34] mm: page-replace-init.patch Peter Zijlstra
2006-03-22 22:34 ` [PATCH 17/34] mm: page-replace-info.patch Peter Zijlstra
2006-03-22 22:34 ` [PATCH 18/34] mm: page-replace-counts.patch Peter Zijlstra
2006-03-22 22:34 ` [PATCH 19/34] mm: page-replace-data.patch Peter Zijlstra
2006-03-22 22:35 ` [PATCH 20/34] mm: page-replace-pg_flags.patch Peter Zijlstra
2006-03-22 22:35 ` [PATCH 21/34] mm: page-replace-nonresident.patch Peter Zijlstra
2006-03-22 22:35 ` [PATCH 22/34] mm: page-replace-shrink-new.patch Peter Zijlstra
2006-03-22 22:35 ` [PATCH 23/34] mm: page-replace-documentation.patch Peter Zijlstra
2006-03-22 22:35 ` [PATCH 24/34] mm: sum_cpu_var.patch Peter Zijlstra
2006-03-22 22:35 ` [PATCH 25/34] mm: kswapd-writeout-wait.patch Peter Zijlstra
2006-03-22 22:36 ` [PATCH 26/34] mm: clockpro-nonresident.patch Peter Zijlstra
2006-03-22 22:36 ` [PATCH 27/34] mm: clockpro-ignore_token.patch Peter Zijlstra
2006-03-22 22:36 ` [PATCH 28/34] mm: clockpro-PG_reclaim2.patch Peter Zijlstra
2006-03-22 22:36 ` [PATCH 29/34] mm: clockpro-clockpro.patch Peter Zijlstra
2006-03-22 22:36 ` [PATCH 30/34] mm: cart-nonresident.patch Peter Zijlstra
2006-03-22 22:36 ` [PATCH 31/34] mm: cart-PG_reclaim3.patch Peter Zijlstra
2006-03-22 22:37 ` [PATCH 32/34] mm: cart-cart.patch Peter Zijlstra
2006-03-22 22:37 ` [PATCH 33/34] mm: cart-r.patch Peter Zijlstra
2006-03-22 22:37 ` [PATCH 34/34] mm: random.patch Peter Zijlstra
2006-03-22 22:51 ` Andrew Morton [this message]
2006-03-23  2:21   ` [PATCH 00/34] mm: Page Replacement Policy Framework Nick Piggin
2006-03-23 21:13     ` Marcelo Tosatti
2006-03-23  4:01   ` Rik van Riel
2006-03-23 20:53   ` Marcelo Tosatti
2006-03-23 18:15     ` Linus Torvalds
2006-03-23 18:26       ` Rik van Riel
2006-03-23 18:48       ` Diego Calleja
2006-03-23 19:03       ` Peter Zijlstra
2006-03-23 22:30       ` Marcelo Tosatti
2006-03-23 20:49         ` Linus Torvalds
2006-03-23 20:59           ` Rik van Riel
2006-03-24 15:06       ` Helge Hafting
2006-03-28 23:05       ` Elladan

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=20060322145132.0886f742.akpm@osdl.org \
    --to=akpm@osdl.org \
    --cc=a.p.zijlstra@chello.nl \
    --cc=bob.picco@hp.com \
    --cc=christoph@lameter.com \
    --cc=iwamoto@valinux.co.jp \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=marcelo.tosatti@cyclades.com \
    --cc=npiggin@suse.de \
    --cc=riel@redhat.com \
    --cc=torvalds@osdl.org \
    --cc=wfg@mail.ustc.edu.cn \
    /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