Convert the double-precision floating point value x to an unsigned 64-bit integer value in round-to-nearest-even mode.
x