pub unsafe fn _mm256_abs_epi32(a: __m256i) -> __m256i
This is supported on x86-64 and target feature avx2 only.
Expand description

Computes the absolute values of packed 32-bit integers in a.

Intel’s documentation