Re: [tor-dev] New paper by Goldberg, Stebila, and Ustaoglu with proposed circuit handshake

On Thu, May 12, 2011 at 9:51 AM, Nick Mathewson nickm@freehaven.net wrote:

It's likely we'll want to use the fast reference implementation on 32-bit intel (It's assembly, right?), and donna on 64-bit platforms. We're going to need to find an answer for 32-bit PPC and ARM platforms, though.  Any suggestions there?

The 32-bit C version will function for these platforms and they're only clients, right? So if they're slower it's dramatically less important. I'm sure that I can also tune up the 32-bit version should there be anyone using it.

Cheers

AGL