Frank Denis 6669885aa2
Faster BLAKE3 implementation (#25574)
This is a rewrite of the BLAKE3 implementation, with vectorization.

On Apple Silicon, the new implementation is about twice as fast as the previous one.

With AVX2, it is more than 4 times faster.

With AVX512, it is more than 7.5x faster than the previous implementation (from 678 MB/s to 5086 MB/s).
2025-10-15 14:03:56 +02:00
..
2025-08-31 12:49:18 -07:00
2025-08-30 06:36:40 +02:00
2025-08-30 06:36:41 +02:00
2025-10-15 14:03:56 +02:00
2025-09-30 13:44:56 +01:00