core::arch::x86_64

Function _mm256_maddubs_epi16

1.27.0 · source
pub unsafe fn _mm256_maddubs_epi16(a: __m256i, b: __m256i) -> __m256i
Available on (x86 or x86-64) and target feature avx2 and x86-64 only.
Expand description

Vertically multiplies 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

Intel’s documentation