From: Borislav Petkov <bp@alien8.de>
To: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Cc: "Johannes Goetzfried"
<Johannes.Goetzfried@informatik.stud.uni-erlangen.de>,
"Herbert Xu" <herbert@gondor.apana.org.au>,
linux-kernel@vger.kernel.org, linux-crypto@vger.kernel.org,
"Tilo Müller" <tilo.mueller@informatik.uni-erlangen.de>
Subject: Re: [PATCH] crypto: twofish - add x86_64/avx assembler implementation
Date: Wed, 15 Aug 2012 11:28:04 +0200 [thread overview]
Message-ID: <20120815092804.GA14676@x1.osrc.amd.com> (raw)
In-Reply-To: <20120815114216.209814z4mq3hxqe8@www.81.fi>
On Wed, Aug 15, 2012 at 11:42:16AM +0300, Jussi Kivilinna wrote:
> I started thinking about the performance on AMD Bulldozer.
> vmovq/vmovd/vpextr*/vpinsr* between FPU and general purpose registers
> on AMD CPU is alot slower (latencies from 8 to 12 cycles) than on
> Intel sandy-bridge (where instructions have latency of 1 to 2). See:
> http://www.agner.org/optimize/instruction_tables.pdf
>
> It would be really good, if implementation could be tested on AMD CPU
> to determinate, if it causes performance regression. However I don't
> have access to machine with such CPU.
But I do. :)
And if you tell me exactly how to run the tests and on what kernel, I'll
try to do so.
HTH.
--
Regards/Gruss,
Boris.
WARNING: multiple messages have this Message-ID (diff)
From: Borislav Petkov <bp@alien8.de>
To: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Cc: "Johannes Goetzfried"
<Johannes.Goetzfried@informatik.stud.uni-erlangen.de>,
"Herbert Xu" <herbert@gondor.hengli.com.au>,
linux-kernel@vger.kernel.org, linux-crypto@vger.kernel.org,
"Tilo Müller" <tilo.mueller@informatik.uni-erlangen.de>
Subject: Re: [PATCH] crypto: twofish - add x86_64/avx assembler implementation
Date: Wed, 15 Aug 2012 11:28:04 +0200 [thread overview]
Message-ID: <20120815092804.GA14676@x1.osrc.amd.com> (raw)
In-Reply-To: <20120815114216.209814z4mq3hxqe8@www.81.fi>
On Wed, Aug 15, 2012 at 11:42:16AM +0300, Jussi Kivilinna wrote:
> I started thinking about the performance on AMD Bulldozer.
> vmovq/vmovd/vpextr*/vpinsr* between FPU and general purpose registers
> on AMD CPU is alot slower (latencies from 8 to 12 cycles) than on
> Intel sandy-bridge (where instructions have latency of 1 to 2). See:
> http://www.agner.org/optimize/instruction_tables.pdf
>
> It would be really good, if implementation could be tested on AMD CPU
> to determinate, if it causes performance regression. However I don't
> have access to machine with such CPU.
But I do. :)
And if you tell me exactly how to run the tests and on what kernel, I'll
try to do so.
HTH.
--
Regards/Gruss,
Boris.
next prev parent reply other threads:[~2012-08-15 9:28 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-05-27 14:49 [PATCH] crypto: twofish - add x86_64/avx assembler implementation Johannes Goetzfried
2012-05-27 14:49 ` Johannes Goetzfried
2012-05-28 6:25 ` Jussi Kivilinna
2012-05-28 6:25 ` Jussi Kivilinna
2012-05-28 13:52 ` Johannes Goetzfried
2012-08-15 8:42 ` Jussi Kivilinna
2012-08-15 8:42 ` Jussi Kivilinna
2012-08-15 9:28 ` Borislav Petkov [this message]
2012-08-15 9:28 ` Borislav Petkov
2012-08-15 11:00 ` Jussi Kivilinna
2012-08-15 11:00 ` Jussi Kivilinna
2012-08-15 12:52 ` Borislav Petkov
2012-08-15 12:52 ` Borislav Petkov
2012-08-15 13:48 ` Jussi Kivilinna
2012-08-15 13:48 ` Jussi Kivilinna
2012-08-15 14:03 ` Borislav Petkov
2012-08-15 14:03 ` Borislav Petkov
2012-08-15 14:22 ` Jussi Kivilinna
2012-08-15 14:22 ` Jussi Kivilinna
2012-08-15 15:33 ` Borislav Petkov
2012-08-15 15:33 ` Borislav Petkov
2012-08-15 17:34 ` Jussi Kivilinna
2012-08-15 17:34 ` Jussi Kivilinna
2012-08-16 13:29 ` Borislav Petkov
2012-08-16 13:29 ` Borislav Petkov
2012-08-16 14:26 ` Jussi Kivilinna
2012-08-16 14:26 ` Jussi Kivilinna
2012-08-17 7:37 ` Jussi Kivilinna
2012-08-17 7:37 ` Jussi Kivilinna
2012-08-20 17:32 ` Borislav Petkov
2012-08-20 17:32 ` Borislav Petkov
2012-08-22 4:35 ` Jussi Kivilinna
2012-08-22 4:35 ` Jussi Kivilinna
2012-08-22 13:31 ` Borislav Petkov
2012-08-22 13:31 ` Borislav Petkov
2012-08-22 19:20 ` Jussi Kivilinna
2012-08-22 19:20 ` Jussi Kivilinna
2012-08-23 0:05 ` Jason Garrett-Glaser
2012-08-23 0:05 ` Jason Garrett-Glaser
2012-08-23 8:33 ` Jussi Kivilinna
2012-08-23 8:33 ` Jussi Kivilinna
2012-08-23 14:36 ` Borislav Petkov
2012-08-23 14:36 ` Borislav Petkov
2012-08-28 9:17 ` Jussi Kivilinna
2012-08-28 9:17 ` Jussi Kivilinna
2012-08-28 16:25 ` Borislav Petkov
2012-08-28 16:25 ` Borislav Petkov
-- strict thread matches above, loose matches on Subject: below --
2012-05-28 13:54 Johannes Goetzfried
2012-06-12 10:05 ` Herbert Xu
2012-06-12 10:05 ` Herbert Xu
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=20120815092804.GA14676@x1.osrc.amd.com \
--to=bp@alien8.de \
--cc=Johannes.Goetzfried@informatik.stud.uni-erlangen.de \
--cc=herbert@gondor.apana.org.au \
--cc=jussi.kivilinna@mbnet.fi \
--cc=linux-crypto@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=tilo.mueller@informatik.uni-erlangen.de \
/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.