float ldexpf(float x, int exp);
ldexpf multiplies a floating-point number by an integral power of 2.
ldexpf returns x * 2^exp. If the result overflows, errno is set to ERANGE and ldexpf returns HUGE_VALF.
If x is ∞ or NaN, ldexpf returns x. If the result overflows, ldexpf returns ∞.