Skip to content

v6.1.2: Avoid AVX-512BW on Skylake-X
Compare
Choose a tag to compare
@ashvardanian ashvardanian released this 24 Nov 19:19
· 39 commits to main since this release
v6.1.2
37ea5a2

Previous AVX-512 implementation of complex products used an extra ZMM register for swap_adjacent_vec. Moreover, they used the vpshufb instruction available only with the Ice Lake capability and newer. The replacement uses the _mm512_permute_ps and its double-precision variant.

Assets 2
Loading