[RC5] Hello / kernel optimization

bwilson at fers.com bwilson at fers.com
Wed Jul 11 08:49:35 EDT 2001


Sounds like you're just the kind of person we want to talk to.  I've forwarded

your message to some of our coders. 

The cores are available for download at http://www.distributed.net/source/ - 
have a look and give it your best shot.  The actual algorithms used in RC5 
encryption can be found starting at 
http://www.rsasecurity.com/rsalabs/challenges/secretkey/index.html 

Have a look at the information, and if you come up with an optimization to the

core for P4, send it to bovine at distributed.net, daa at distributed.net and 
cyp at distributed.net (as mentioned on the /source page) 

(pardon any awkward line breaks... I'm frustratingly locked into Lotus Notes) 

__
Bruce Wilson, Manager, FERS Business Services
bwilson at fers.com, 312.245.1750, http://www.fers.com/
PGP KeyID: 5430B995, http://www.toomuchblue.com/

Domain names are like women (or men): The ones you want are
already taken, but you can always get one from a strange country. 



	rc5 at lists.distributed.net 
	
	07/10/2001 22:02 
		        
		        To:        rc5 at llamas.net 
		        cc:        (bcc: Bruce Wilson/Chicago/MP/RSMi) 
		        Subject:        [RC5] Hello / kernel optimization



Hello, 

I just subscribed to the list, so my apologies if this is any breach of
etiquitte. 

I have heard that the RC5 kernel for Pentium 4 is, well, slow.  I consider
myself pretty handy with an assembler so I'd like to write a new kernel.  My
problem is that I don't know much about RC5. 

What's the quickest way to ramp up on what the RC5 algorithm actually does, at
a
high level of abstraction?  In lieu of not finding any "How to Understand RC5"
documentation, I've looked at the existing kernel sources and I don't find
them
transparent.  I figure the experts are here, so this is the right place to
ask. 

My particular concerns are: 

  - What are the operations to be performed for each data element?  (i.e. what
do the rounds do and how do they break down?) 

  - Are there any interactions between data elements that make parallelization
tricky?  (Is this what the "mangle-increment" stuff is all about?) 

Thanks! 

---
Kyle Markley
--
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest 




--
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest



More information about the rc5 mailing list