Re: [PATCH] AES i586 optimized
From: Fruhwirth Clemens
Date: Thu May 13 2004 - 06:46:46 EST
On Thu, May 13, 2004 at 04:07:32AM -0700, Andrew Morton wrote:
> Fruhwirth Clemens <clemens-dated-1085310070.4b1f@xxxxxxxxxxxxx> wrote:
> > The following patch adds an i586 optimized implementation of AES aka
> > Rijndael. It's following an integration strategy similiar to recent
> > s390/z990 integration for DES/SHA1. aes-i586-glue.c, a glue layer for
> > CryptoAPI, and aes-i586-asm.S, the actual implementation, are added to
> > arch/i386/crypto, as well as a config section to crypto/Kconfig.
>
> Some benchmark figures would be useful. Cache-cold and cache-hot.
I posted this patch for the first time about 3/4 year ago. The first
response has been the same. Please have a look at
My first posting:
http://marc.theaimsgroup.com/?l=linux-kernel&m=106321924724484&w=2
My original claim:
http://www.kerneli.org/pipermail/cryptoapi-devel/2003-August/000607.html
Jari Ruusu's posting:
http://marc.theaimsgroup.com/?l=linux-kernel&m=106321682117697&w=2
I suspect the result to be similar for a P4 for example. I could do some P4
benchmarking.
> > The code has been in use for half a year by myself and the loop-aes project
> > for the last - I think - 2 years. I consider the implementation
> > production-stable. Andrew, please consider applying.
>
> James Morris <jmorris@xxxxxxxxxx> is the crypto guy....
Ack. I'll try to get his opinion on this.
Regards, Clemens
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/