pub unsafe fn _mm_cvtps_pd(a: __m128) -> __m128d
Converts packed single-precision (32-bit) floating-point elements in a to packed double-precision (64-bit) floating-point elements.
a
Intel’s documentation