The part about data dependencies across loop iterations is fascinating to me, becuase it's mostly invisible even when you look at the generated assembly. There'

The part about data dependencies across loop iterations is fascinating to me, becuase it's mostly invisible even when you look at the generated assembly. There's a related optimization that comes up in implementations of ChaCha/BLAKE, where we permute columns around in a kind of weird order, because it breaks a data dependency for an operation that's about to happen: https://github.com/sneves/blake2-avx2/pull/4#issuecomment-502507027 https://github.com/sneves/blake2-avx2/pull/4#issuecomment-50...