Macro safe_arch::insert_m128i_to_m256i_slow_avx

macro_rules! insert_m128i_to_m256i_slow_avx {
    ($a:expr, $b:expr, $imm:expr) => { ... };
}

This is supported with target feature avx only.

Slowly inserts an m128i to m256i.

This is a "historical artifact" that was potentially useful if you have AVX but not AVX2. If you plan on having AVX2 available please use insert_m128i_to_m256i, it will do the same task with better performance.

let a = m256i::from([0_i32; 8]);
let b: [i32; 8] =
  insert_m128i_to_m256i_slow_avx!(a, m128i::from([1, 2, 3, 4]), 1).into();
assert_eq!(b, [0, 0, 0, 0, 1, 2, 3, 4]);