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