Function _mm512_maddubs_epi16

#[target_feature(enable = "avx512bw")]
pub fn _mm512_maddubs_epi16(a: __m512i, b: __m512i) -> __m512i

Vertically multiply each unsigned 8-bit integer from a with the corresponding signed 8-bit integer from b, producing intermediate signed 16-bit integers. Horizontally add adjacent pairs of intermediate signed 16-bit integers, and pack the saturated results in dst.

Intel's documentation