Subject: Re: difference in key calculations for ssh
To: Todd Vierling <tv@pobox.com>
From: Andrew Brown <atatat@atatdot.net>
List: current-users
Date: 05/25/2000 17:58:20
>: >My guess would be for the same reasons rc5des is significantly faster
>: >on Intel.  My understanding is that the Intel is much faster at certain
>: >instructions (shifts?) that are heavily used by the encryption code.
>: 
>: i dunno if it's shifts...i always figured it was intel being better at
>: integer arthmetic than sparcs.
>
>It's bitwise rotates, for one.  SPARCs have no such beasts.

that's something i always thought was missing from c: a bitwise rotate
primitive.  i mean...it's had a shift primitive for how long?

>It's also that there is a very highly optimised assembly gmp innards for x86
>in ssh.
>
>IIRC, SPARC in ssh uses the C versions of all the gmp stuff....

ah yes.  that will make a difference too.  hmmm.......

-- 
|-----< "CODE WARRIOR" >-----|
codewarrior@daemon.org             * "ah!  i see you have the internet
twofsonet@graffiti.com (Andrew Brown)                that goes *ping*!"
andrew@crossbar.com       * "information is power -- share the wealth."