From: Andi Kleen <ak@suse.de>
To: linux-crypto@vger.kernel.org
Subject: {twofish,aes}-{x86_64,i586} versus C implementations
Date: Mon, 20 Aug 2007 02:34:25 +0200 [thread overview]
Message-ID: <200708200234.25620.ak@suse.de> (raw)
Hallo,
Currently there are two twofish and two aes implementions on x86.
Worse when both are enabled as modules a modprobe aes
will get the C version which seems to be slower on K8
and about the same speed on Core2 on my tests.
Is there a specific reason why anybody would prefer the C functions
over the assembler functions?
Possible reasons I could think of:
- If the assembler functions are optimized for a specific CPU the compiler
might be able to do a better job on other CPUs.
Is there evidence for this? I suspect it's not true.
- They are not trusted and might be buggy. I assume they have
been validated against the C versions with a wide range of input
data, correct?
If none of these reasons are valid it might make sense to disable
the C versions for x86 and only offer the assembler versions.
Then modprobe aes|twofish would DTRT automatically.
Comments?
-Andi
next reply other threads:[~2007-08-20 0:34 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-08-20 0:34 Andi Kleen [this message]
2007-08-20 1:01 ` {twofish,aes}-{x86_64,i586} versus C implementations Herbert Xu
2007-08-20 10:16 ` Andi Kleen
2007-08-20 9:45 ` Sebastian Siewior
2007-08-20 10:47 ` Andi Kleen
2007-08-20 10:08 ` Sebastian Siewior
2007-08-20 11:12 ` Andi Kleen
2007-08-20 11:27 ` Sebastian Siewior
2007-08-20 12:06 ` Herbert Xu
2007-08-20 13:06 ` Andi Kleen
2007-08-20 13:07 ` Herbert Xu
2007-09-02 22:42 ` Sebastian Siewior
2007-09-04 13:58 ` Andi Kleen
2007-09-19 12:29 ` Herbert Xu
2007-09-19 21:46 ` Sebastian Siewior
2007-09-20 0:20 ` Herbert Xu
2007-09-20 21:09 ` Sebastian Siewior
2007-09-30 12:23 ` Sebastian Siewior
2007-09-30 12:42 ` Sebastian Siewior
2007-10-03 7:35 ` Herbert Xu
2007-10-04 8:35 ` Sebastian Siewior
2007-10-03 19:23 ` [PATCH] [crypto] load the DES module by an alias Sebastian Siewior
2007-10-05 8:48 ` Herbert Xu
2007-10-04 7:37 ` [PATCH] [crypto] load the AES " Sebastian Siewior
2007-10-05 8:52 ` Herbert Xu
2007-10-04 7:37 ` [PATCH] [crypto] load the SHA1[1|256] " Sebastian Siewior
2007-10-05 8:57 ` Herbert Xu
2007-10-05 13:50 ` Sebastian Siewior
2007-10-05 13:12 ` [PATCH] [crypto] load the SHA1[1|256] module by an alias (v2) Sebastian Siewior
2007-10-05 15:10 ` Herbert Xu
2007-10-06 22:02 ` Sebastian Siewior
2007-10-08 3:21 ` Herbert Xu
2007-10-07 21:42 ` Sebastian Siewior
2007-10-08 3:20 ` Herbert Xu
2007-10-08 12:35 ` Sebastian Siewior
2007-10-08 4:12 ` Herbert Xu
2007-10-05 14:20 ` [PATCH] [crypto] load the SHA1[1|256] module by an alias Herbert Xu
2007-10-06 21:54 ` Sebastian Siewior
2007-10-08 11:25 ` Jan Glauber
2007-10-08 11:30 ` Sebastian Siewior
2007-10-04 8:48 ` {twofish,aes}-{x86_64,i586} versus C implementations Herbert Xu
2007-10-04 9:31 ` Andi Kleen
2007-10-04 10:00 ` Sebastian Siewior
2007-10-04 10:00 ` Herbert Xu
2007-10-04 9:39 ` Sebastian Siewior
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=200708200234.25620.ak@suse.de \
--to=ak@suse.de \
--cc=linux-crypto@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox