Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers,
See Implementation
Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers,