[−][src]Function safe_arch::shuffle_av_i32_all_m256

#[must_use]pub fn shuffle_av_i32_all_m256(a: m256, v: m256i) -> m256
This is supported with target feature avx2 only.

Shuffle f32 lanes in a using i32 values in v.

let a = m256::from_array([8.0, 9.0, 10.0, 11.0, 12.0, 13.0, 14.0, 15.0]);
let v = m256i::from([7, 6, 5, 5, 3, 2, 2, 0]);
let c: [f32; 8] = shuffle_av_i32_all_m256(a, v).to_array();
assert_eq!(c, [15.0, 14.0, 13.0, 13.0, 11.0, 10.0, 10.0, 8.0]);