Re: [f-cpu] RC5, F-CPU and srotl

[cedric <cedric.bail@free.fr>]

> > >From my point of view, the dnet project is a specific part of the
> > application domain of F-CPU. Can i point that crypto is used in
> > communication (like in SSH) and storage (encrypted filesystems) ?
> > This means that if we find something interesting when examining dnet,
> > it can also enhance the performance of a normal computer that works
> > in a secured environment, such as an internet server. It is maybe
> > worth noting it and may explain why i don't focus exclusively on
> > RC5 cracking (though i know it's a good entry point and a "cool way"
> > to make others think about F-CPU).
> Crypto stuff may be "cool", but something that fits into the L1 cache
> doesn't qualify as a real application, IMHO :)  Besides that, many crypto
> algorithms can be done better (and faster) in dedicated hardware.
Ok, DNETC-RC5 is a real application, it's why I decided to implement it. When 
I will finished to port in, I will perhaps look to cryptographic algorithm 
(like AES), but that an other discussion.

What I want to do by the port of the dnetc client is to see where we can have 
problem with our instruction set on high parellisable algorithm. The 
conclusion of this discussion is that we will have a "real" simd rotl, but 
not because of this algorithm, but because we want a "clean" CPU.

> What's a 64-bit CPU good for, compared to a 32-bit one? It's not
> necessarily faster, but it can handle large working sets. Therefore
> I'd rather focus on scientific applications, databases and so on.
Yes, but we need to start from a point, if you have some idee of which 
algorithm or part of program to port, say it here, and if some one have time 
he can start to port them.

> > The case of dnet is looking different but i can't involve myself
> > in this now. The case of the SIMD shift has been proved and there
> > does not seem to be any difficulty in implementing it with F-CPU,
> > we simply disagree on a "small" implementation detail :-)
> Huh? No difficulty?
I see your last post, so you find a solution, no ? ;-)

A+
   Cedric

*************************************************************
To unsubscribe, send an e-mail to majordomo@seul.org with
unsubscribe f-cpu       in the body. http://f-cpu.seul.org/