nisse@lysator.liu.se (Niels Möller) writes:
I have it fixed in my working tree now (as well as _init at the end of _digest, and fixes for the carry handling in gost_compute_sum_and_hash (which I mailed Aleksey about and I hope he will comment on that), and an MIT-style (c) notice).
I've pushed the gost code to the repo now, with various smaller changes compared to your patch. Please have a look, and tell me if I've missed something. Some remaining things, in order of importance:
1. Documentation (manual and NEWS entry).
2. Support in nettle-benchmark.
3. I think it would be nice to include a program for generating the tables (could be based on the code in the first version of your patch). It's stil not clear to me if if/when alternative sboxes are used.
4. Maybe some reindentation and M-x whitespace-cleanup, for consistency.
BTW, with increasing number of hash functions, I think it would be nice to add some more structure to the manual, and at least split it into "recommended hash function" and "other/legacy hash functions". Not sure which category to put gost in, though.
Regards, /Niels