pub unsafe fn _mm256_blend_epi16( a: __m256i, b: __m256i, const IMM8: i32) -> __m256i
avx2
Blends packed 16-bit integers from a and b using control mask IMM8.
a
b
IMM8
Intel’s documentation