pub fn mul_u16_keep_high_m256i(a: m256i, b: m256i) -> m256i
Available with target feature avx2 only.
Expand description

Multiply the u16 lanes and keep the high half of each 32-bit output.

let a = m256i::from([5_u16, 6, 2, 5, 4, 3, 1, 0, 12000, 13, 56, 21, 8, 7, 6, 5]);
let b = m256i::from([12000_u16, 13000, 2000, 800, 0, 1, 2, 3, 8, 7, 6, 5, 234, 654, 123, 978]);
let c: [u16; 16] = mul_u16_keep_high_m256i(a, b).into();
assert_eq!(c, [0_u16, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0]);