Macro safe_arch::shuffle_ai_i64_all_m256i

macro_rules! shuffle_ai_i64_all_m256i {
    ($a:expr, [$z:expr, $o:expr, $t:expr, $h:expr]) => { ... };
}

This is supported with target feature avx2 only.

Shuffle the f64 lanes in $a using an immediate control value.

let a = m256i::from([5_i64, 6, 7, 8]);
let b: [i64; 4] = shuffle_ai_i64_all_m256i!(a, [3, 2, 1, 0]).into();
assert_eq!(b, [8_i64, 7, 6, 5]);

• Intrinsic: _mm256_permute4x64_epi64
• Assembly: vpermq ymm, ymm, imm8