Function _mm512_mask_blend_epi8 

pub fn _mm512_mask_blend_epi8(k: __mmask64, a: __m512i, b: __m512i) -> __m512i

Available on x86-64 and target feature avx512bw only.

Blend packed 8-bit integers from a and b using control mask k, and store the results in dst.

Intel’s documentation