Palmer
Palmer
@tarcieri Its hard, but not impossible to write algorithms that are constant time in Rust. The tricky part is actually proving that. I have some ideas about how to do...
@genodeftest: I like your checklist and I think thats a good model for what we should have for a potential 1.0 release at some point. Right now, nothing would meet...
@genodeftest One of the big, giant challenges for rust-crypto is proving that it is fixed time for some subset of its algorithms. If that can't be done, its never really...
Its not currently possible, although I believe that constructing a new Salsa20 struct with a different offset would work. I'd certainly be happy to accept a patch to add this...
Thanks for the PR and sorry for taking so long to get to it. Hmm - I'd rather not pull in a dependency on another Blake2b implementation if ours is...
I think the Blake2b issue is fixed by #353. Does that remove the need to pull in an external blake2b dependency?
The other things I'm concerned about is adding in the shell scripts and the large test files - I like the strategy, but I'd like it to apply to other...
H.R.6977 appears to be available now, but I'm still not seeing S.3621
@crescentrose @jcdickinson I'm having the same issue with 1Password.