On: Ara: A 1 GHz+ Scalable and Energy-Efficient RISC-V Vector Processor

[brandmeyer]

Can you be more specific?

[snovv_crash → brandmeyer]

I've used it for bitwise hamming distance - xor+popcnt - having a hardware instruction on x86 made the entire pipeline over 6x faster.

[stochastic_monk → snovv_crash]

I’ve used popcounts for full vectorization for b-bit minwise hashing, comparing bloom filters efficiently, and bitvector traversal. It’s a standard primitive in a lot of low-level code.