In this assembly code it cannot overflow because N is a 32-bit integer and the multiplication gives a 64-bit result, which is converted to 32-bit only after shi

In this assembly code it cannot overflow because N is a 32-bit integer and the multiplication gives a 64-bit result, which is converted to 32-bit only after shifting. I can't figure out why it doesn't use the simpler formula (other than the optimizer being bad).