pub unsafe fn _mm256_cvtepi32_pd(a: __m128i) -> __m256d
Converts packed 32-bit integers in a to packed double-precision (64-bit) floating-point elements.
a
Intel’s documentation