pub fn _mm512_kxor(a: __mmask16, b: __mmask16) -> __mmask16
avx512f
Compute the bitwise XOR of 16-bit masks a and b, and store the result in k.
Intel’s documentation