[HARDWARE] Just Courious......]]

stoney at sequent.com stoney at sequent.com
Mon Sep 20 13:42:22 EDT 1999


You will only be able to partially overlap the assignments to A and B
in the for loops because of the data dependencies.  You cannot feasibly
unroll the decryption loop

RC5-w/r/b 46/16/

  For i = r down to 1:

nor the 2r+2 subkeys loop.

  For i = 1 to 2(r+1) -1:



On Sun, 19 Sep 1999, Robert Norton wrote:

> Dave Avery wrote:
> 
> > <snip>
> > as far as I can tell you do require a programmable barrelshifter - the rotates are
> > data dependant , unlike DES . The amount to rotate is determined by the incoming
> > data/key
> 
> Thanks for the info, I didn't understand that the rotate was data dependent!  So it will
> take more gates to fully unroll the algorithm, dang it.  Might it still fit into a $50
> Xilinx part?  For me, the idea of 25Mkey/sec in a small, (cheap!,) box is very exciting.
> 
> 
> --
> To unsubscribe, send 'unsubscribe hardware' to majordomo at lists.distributed.net
> 

--
To unsubscribe, send 'unsubscribe hardware' to majordomo at lists.distributed.net



More information about the Hardware mailing list