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