2 ms·Last time I checked, the movs variants were microcoded slow path; you used SSE to get the fastest memcpy.by monocasa 10y agoLast time I checked, the movs variants were microcoded slow path; you used SSE to get the fastest memcpy.