Convert the double-precision floating point value x to a signed 64-bit integer value in round-up (to positive infinity) mode.
x