Function _mm256_mask_blend_epi8

#[target_feature(enable = "avx512bw", enable = "avx512vl")]
pub const fn _mm256_mask_blend_epi8(k: __mmask32, a: __m256i, b: __m256i) -> __m256i

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

Intel's documentation