animetosho/md5-optimisation

The fastest MD5 implementation using x86 assembly

3 mentions on Hacker News by 2 people: 2 stories, 1 comment · first 2020-12-10 · latest 2023-06-20

Mentions per year 3 mentions · busiest year 2023 (2)

20082026

All Stories Comments Newest Top Oldest

…useful. In fact, it speeds up a single instance (i.e. not taking advantage of SIMD) of MD5 by 20%: https://github.com/animetosho/md5-optimisation#x86-avx512-vl...
comment by Nyan · 2023-06-20 · in The least interesting part about AVX-512 is the 512 bits vector width