pub unsafe fn _mm_loaddup_pd(mem_addr: *const f64) -> __m128d
Loads a double-precision (64-bit) floating-point element from memory into both elements of return vector.
Intel’s documentation