Function safe_arch::sub_i32_m256i

#[must_use]pub fn sub_i32_m256i(a: m256i, b: m256i) -> m256i

This is supported with target feature avx2 only.

Lanewise a - b with lanes as i32.

let a = m256i::from([5_i32; 8]);
let b = m256i::from([10_i32; 8]);
let c: [i32; 8] = sub_i32_m256i(a, b).into();
assert_eq!(c, [-5_i32; 8]);

• Intrinsic: _mm256_sub_epi32
• Assembly: vpsubd ymm, ymm, ymm