From: simon@himi.org (Simon Fowler)
To: linux-kernel@vger.kernel.org
Subject: Re: [CFT] faster athlon/duron memory copy implementation
Date: Fri, 25 Oct 2002 04:43:28 +1000 [thread overview]
Message-ID: <20021024184328.GA5667@himi.org> (raw)
In-Reply-To: <3DB82ABF.8030706@colorfullife.com>
[-- Attachment #1: Type: text/plain, Size: 3000 bytes --]
On Thu, Oct 24, 2002 at 07:15:43PM +0200, Manfred Spraul wrote:
> AMD recommends to perform memory copies with backward read operations
> instead of prefetch.
>
> http://208.15.46.63/events/gdc2002.htm
>
> Attached is a test app that compares several memory copy implementations.
> Could you run it and report the results to me, together with cpu,
> chipset and memory type?
>
> Please run 2 or 3 times.
>
simon@caccini:~/hacking$ cat /proc/cpuinfo
processor : 0
vendor_id : AuthenticAMD
cpu family : 6
model : 1
model name : AMD-K7(tm) Processor
stepping : 2
cpu MHz : 553.880
cache size : 512 KB
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 1
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca cmov pat mmx syscall mmxext 3dnowext 3dnow
bogomips : 1104.28
simon@caccini:~/hacking$ ./athlon; ./athlon; ./athlon
Athlon test program $Id: fast.c,v 1.6 2000/09/23 09:05:45 arjan Exp $
copy_page() tests
copy_page function 'warm up run' took 12855 cycles per page
copy_page function '2.4 non MMX' took 17267 cycles per page
copy_page function '2.4 MMX fallback' took 14930 cycles per page
copy_page function '2.4 MMX version' took 10642 cycles per page
copy_page function 'faster_copy' took 10591 cycles per page
copy_page function 'even_faster' took 13035 cycles per page
copy_page function 'no_prefetch' took 11657 cycles per page
Athlon test program $Id: fast.c,v 1.6 2000/09/23 09:05:45 arjan Exp $
copy_page() tests
copy_page function 'warm up run' took 12871 cycles per page
copy_page function '2.4 non MMX' took 18482 cycles per page
copy_page function '2.4 MMX fallback' took 15013 cycles per page
copy_page function '2.4 MMX version' took 10679 cycles per page
copy_page function 'faster_copy' took 12268 cycles per page
copy_page function 'even_faster' took 10789 cycles per page
copy_page function 'no_prefetch' took 11691 cycles per page
Athlon test program $Id: fast.c,v 1.6 2000/09/23 09:05:45 arjan Exp $
copy_page() tests
copy_page function 'warm up run' took 13110 cycles per page
copy_page function '2.4 non MMX' took 14958 cycles per page
copy_page function '2.4 MMX fallback' took 14952 cycles per page
copy_page function '2.4 MMX version' took 12864 cycles per page
copy_page function 'faster_copy' took 10581 cycles per page
copy_page function 'even_faster' took 10629 cycles per page
copy_page function 'no_prefetch' took 11607 cycles per page
Simon
--
PGP public key Id 0x144A991C, or http://himi.org/stuff/himi.asc
(crappy) Homepage: http://himi.org
doe #237 (see http://www.lemuria.org/DeCSS)
My DeCSS mirror: ftp://himi.org/pub/mirrors/css/
[-- Attachment #2: Type: application/pgp-signature, Size: 232 bytes --]
next prev parent reply other threads:[~2002-10-24 18:37 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-10-24 17:15 [CFT] faster athlon/duron memory copy implementation Manfred Spraul
2002-10-24 17:37 ` Robert Love
2002-10-24 18:05 ` Zach Brown
2002-10-24 17:41 ` Andreas Steinmetz
2002-10-24 17:48 ` Matthias Welk
2002-10-24 19:01 ` erich
2002-10-24 19:11 ` Arjan van de Ven
2002-10-24 19:38 ` Manfred Spraul
2002-10-25 0:59 ` Panagiotis Papadakos
2002-10-24 17:53 ` Roger Luethi
2002-10-24 18:10 ` Daniel Egger
2002-10-24 19:15 ` Florin Iucha
2002-10-24 19:28 ` Manfred Spraul
2002-10-24 19:38 ` Dave Jones
2002-10-24 19:43 ` Ken Witherow
2002-10-25 13:08 ` Daniel Egger
2002-10-24 18:17 ` Eric Lammerts
2002-10-24 18:26 ` David Rees
2002-10-24 18:35 ` Josh McKinney
2002-10-24 18:36 ` Dave Jones
2002-10-24 18:43 ` Simon Fowler [this message]
2002-10-24 18:50 ` Simon Fowler
2002-10-24 18:56 ` Dave Jones
2002-10-24 18:48 ` Ernst Herzberg
2002-10-24 20:09 ` Ed Sweetman
2002-10-24 20:13 ` Robert Love
2002-10-24 20:31 ` Ed Sweetman
2002-10-24 20:49 ` Dave Jones
2002-10-24 20:26 ` Dave Jones
2002-10-25 9:19 ` Måns Rullgård
2002-10-24 19:11 ` Marcus Libäck
2002-10-24 19:19 ` Brian Gerst
2002-10-24 19:31 ` Matthias Schniedermeyer
2002-10-24 19:33 ` Pascal Schmidt
2002-10-24 19:39 ` Olaf Dietsche
2002-10-24 20:27 ` Mike Civil
2002-10-24 20:44 ` Willy TARREAU
2002-10-24 21:46 ` Josh Fryman
2002-10-24 22:18 ` Tim Schmielau
2002-10-24 23:09 ` Hirokazu Takahashi
2002-10-24 23:37 ` Ryan Cumming
2002-10-25 0:10 ` Matthias Andree
2002-10-25 8:35 ` venom
2002-10-25 13:31 ` Denis Vlasenko
2002-10-26 12:11 ` Jurjen Oskam
-- strict thread matches above, loose matches on Subject: below --
2002-10-24 18:27 Shawn Starr
2002-10-24 20:51 Dieter Nützel
2002-10-24 21:01 ` Dieter Nützel
2002-10-24 21:16 ` Willy TARREAU
2002-10-24 22:01 Harm Verhagen
2002-10-25 16:29 Jorge Bernal "Koke"
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=20021024184328.GA5667@himi.org \
--to=simon@himi.org \
--cc=linux-kernel@vger.kernel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.