Report question

Question: Why doesn't a compiler optimize floating-point *2 into an exponent increment?