I've been playing with adding anti-spam hash-cash stamping support to
Mozilla/Thuderbird, and it requires calling sha1 many, many times in a
loop. I have a hacked-together version that works, but I'm struggling
with the right way to call sha1. The stamping is extremely CPU intensive
with respect to computing hashes, so I'd like the code to be near
optimal. This is important because the spammers will have every
incentive for their stamp code to be near optimal, and the whole system
works best if the good-guy stamping is about as fast as the bad-guy
stamping.
Anyway, the question is: what's the best way to call sha1 many times in
a loop from Mozilla C++?
It seems like the best candidates are the HASH_Begin() etc.
functions in
http://lxr.mozilla.org/seamonkey/source/security/nss/lib/cryptohi/sechash.h
But with my very primitive understanding of XPCOM and my makefile.in, I
cannot get that to compile. I saw the new nsICryptoHash.idl, but I think
that's too new for the mozilla 1.7.8 I'm targetting, and it also seems
to create more overhead than makes sense in this case, forcing a re-init
of the context for every hash.
Anyway, any help is much appreciated.
Thanks,
Nick
_______________________________________________
mozilla-crypto mailing list
[email protected]
http://mail.mozilla.org/listinfo/mozilla-crypto
