From: Christoph Lameter <clameter@engr.sgi.com>
To: Andrew Morton <akpm@osdl.org>
Cc: peterc@gelato.unsw.edu.au, torvalds@osdl.org, hugh@veritas.com,
piggin@cyberone.com.au, linux-mm@kvack.org
Subject: Re: pagefault scalability patches
Date: Thu, 18 Aug 2005 09:09:10 -0700 (PDT) [thread overview]
Message-ID: <Pine.LNX.4.62.0508180904440.25799@schroedinger.engr.sgi.com> (raw)
In-Reply-To: <20050817174720.47ac351f.akpm@osdl.org>
On Wed, 17 Aug 2005, Andrew Morton wrote:
> > Patched:
> >
> > Gb Rep Threads User System Wall flt/cpu/s fault/wsec
> > 16 3 1 0.859s 64.994s 65.084s 47768.542 47771.664
>
> Versus:
>
> > Unpatched:
> >
> > Gb Rep Threads User System Wall flt/cpu/s fault/wsec
> > 16 3 1 0.757s 62.772s 63.052s 49515.393 49522.112
>
> It got slower?
For that sample yes. There is a certain unpredictability coming with NUMA
systems. Memory placement affects the tests. This in the margin of error.
Another test shows just the opposite:
unpatched:
Gb Rep Threads User System Wall flt/cpu/s fault/wsec
16 3 1 0.735s 64.074s 64.083s 48537.993 48519.120
16 3 2 0.773s 94.774s 49.046s 32923.047 63588.898
16 3 4 0.717s 87.110s 29.092s 35816.846 105117.121
16 3 8 0.677s 136.768s 21.069s 22886.951 145008.228
16 3 16 0.757s 288.464s 23.045s 10876.524 134128.797
16 3 32 13.612s 297.150s 23.034s 10122.600 134723.354
16 3 64 60.201s 318.414s 27.048s 8308.505 114470.017
16 3 128 279.422s 322.942s 41.063s 5222.299 75562.812
16 3 256 280.823s 146.732s 28.073s 7357.466 109486.455
16 3 512 282.124s 77.636s 24.023s 8743.940 129787.460
patched:
Gb Rep Threads User System Wall flt/cpu/s fault/wsec
16 3 1 0.702s 62.858s 63.056s 49491.809 49489.633
16 3 2 0.734s 72.348s 38.004s 43043.199 82674.132
16 3 4 0.718s 76.552s 25.012s 40710.056 125186.047
16 3 8 0.782s 58.417s 12.020s 53137.972 257740.814
16 3 16 1.534s 93.568s 9.092s 33077.207 316995.454
16 3 32 3.297s 173.145s 9.078s 17828.534 321373.156
16 3 64 9.001s 445.874s 11.064s 6915.569 270213.663
16 3 128 27.157s 1500.321s 16.060s 2059.426 189481.849
16 3 256 25.647s 762.183s 8.083s 3992.895 355973.645
16 3 512 26.167s 407.595s 5.008s 7252.183 619054.581
--
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>
next prev parent reply other threads:[~2005-08-18 16:09 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-08-17 22:17 Andrew Morton
2005-08-17 22:19 ` Christoph Lameter
2005-08-17 22:36 ` Linus Torvalds
2005-08-17 22:51 ` Christoph Lameter
2005-08-17 23:01 ` Linus Torvalds
2005-08-17 23:12 ` Christoph Lameter
2005-08-17 23:23 ` Linus Torvalds
2005-08-17 23:31 ` Christoph Lameter
2005-08-17 23:30 ` Andrew Morton
2005-08-17 23:33 ` Christoph Lameter
2005-08-17 23:44 ` Andrew Morton
2005-08-17 23:52 ` Peter Chubb
2005-08-17 23:58 ` Christoph Lameter
2005-08-18 0:47 ` Andrew Morton
2005-08-18 16:09 ` Christoph Lameter [this message]
2005-08-22 2:13 ` Benjamin Herrenschmidt
2005-08-18 0:43 ` Andrew Morton
2005-08-18 16:04 ` Christoph Lameter
2005-08-18 20:16 ` Hugh Dickins
2005-08-19 1:22 ` [PATCH] use mm_counter macros for nr_pte since its also under ptl Christoph Lameter
2005-08-19 3:17 ` Andrew Morton
2005-08-19 3:51 ` Christoph Lameter
2005-08-19 1:33 ` pagefault scalability patches Christoph Lameter
2005-08-19 3:53 ` [RFC] Concept for delayed counter updates in mm_struct Christoph Lameter
2005-08-19 4:29 ` Andrew Morton
2005-08-19 4:34 ` Andi Kleen
2005-08-19 4:49 ` Linus Torvalds
2005-08-19 16:06 ` Christoph Lameter
2005-08-20 7:33 ` [PATCH] mm_struct counter deltas in task_struct Christoph Lameter
2005-08-20 7:35 ` [PATCH] Use deltas to replace atomic inc Christoph Lameter
2005-08-20 7:58 ` Andrew Morton
2005-08-22 3:32 ` Christoph Lameter
2005-08-22 3:48 ` Linus Torvalds
2005-08-22 4:06 ` Christoph Lameter
2005-08-22 4:18 ` Linus Torvalds
2005-08-22 13:23 ` Christoph Lameter
2005-08-22 14:22 ` Hugh Dickins
2005-08-22 15:24 ` Christoph Lameter
2005-08-22 15:43 ` Andi Kleen
2005-08-22 16:24 ` Christoph Lameter
2005-08-22 20:30 ` [PATCH] mm_struct counter deltas V2 Christoph Lameter
2005-08-22 20:31 ` [PATCH] Use deltas to replace atomic inc V2 Christoph Lameter
2005-08-22 2:09 ` pagefault scalability patches Benjamin Herrenschmidt
2005-08-18 2:00 ` Nick Piggin
2005-08-18 8:38 ` Nick Piggin
2005-08-18 16:17 ` Christoph Lameter
2005-08-22 2:04 ` Benjamin Herrenschmidt
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=Pine.LNX.4.62.0508180904440.25799@schroedinger.engr.sgi.com \
--to=clameter@engr.sgi.com \
--cc=akpm@osdl.org \
--cc=hugh@veritas.com \
--cc=linux-mm@kvack.org \
--cc=peterc@gelato.unsw.edu.au \
--cc=piggin@cyberone.com.au \
--cc=torvalds@osdl.org \
/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