In the case that a larger unsigned increment type was added to a signed int, a compiler optimization did not account for possible wraparound resulting in a negative value and erroneously assumed the signed int variable was never negative.