Biham's paper on 64-bit DES implementations

Steven K. Kasow (kasow@rabi.phys.columbia.edu)
Thu, 24 Apr 1997 02:43:02 -0400 (EDT)


Matt Curtin <cmcurtin@research.megasoft.com> wrote:

>The Alpha (and other 64-bit processors) have an especially interesting
>ability, in taking a different approach altogether from the 32-bit
>processors, which could yield significant speed increases. Biham
>discussed 64-bit implementations of DES in software that are
>super-fast...

I've read Biham's paper on this stuff- I assume he was
referring to the idea of viewing a 64-bit CPU as a SIMD machine that
will do 64 one-bit operations at the same time. This seems to me to be
a really cool idea, and was wondering if there is any interest in
writing DESCHALL clients that used this. And it seems to my mind that
the kind of code resulting from that approach would seem to me to be
relatively easily optimized for a RISC chip by a decent compiler.

I suppose I'm volunteering to do some development work- while
I'm not exceedingly familiar with DES implementations in general, and
I don't have a vast amount of free time (I'm a physics grad student),
I am tempted to try and implement that routine in C and see what kind
of speeds I can achieve. I have access to various flavors of Sparcs
(sparc 20s and ultra sparcs) as well as an SGI Irix and Power PCs. Any
interest out there?

cheers,

Steven

--
"There are three stages in the killing of an astrophysicist."
	--Misner, C. W., K. S. Thorne, and J. A. Wheeler, 1973,
		_Gravitation_, W.H. Freeman and Company, New York.