pub fn _kadd_mask32(a: __mmask32, b: __mmask32) -> __mmask32
avx512bw
Add 32-bit masks in a and b, and store the result in k.
Intel’s documentation