pub unsafe fn _mm256_add_ps(a: __m256, b: __m256) -> __m256
Adds packed single-precision (32-bit) floating-point elements in a and b.
a
b
Intel’s documentation