Returns a 32-bit integer corresponding to the more significant 32 bits of a double-precision floating-point number.
