Hi! Please benchmark this version. While its code is 25% larger now, it's faster for small messages, and overall it's more reliable now, lesser chances for occassional "blinding". I'm still not sure why you've got such small large-block hashing performance, on both Intel and AMD, please make sure you've enabled compiler optimizations.
avaneev
commented
2 years ago
Hi! Please benchmark this version. While its code is 25% larger now, it's faster for small messages, and overall it's more reliable now, lesser chances for occassional "blinding". I'm still not sure why you've got such small large-block hashing performance, on both Intel and AMD, please make sure you've enabled compiler optimizations.
https://github.com/avaneev/komihash