https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120704
--- Comment #4 from congli <congli at smail dot nju.edu.cn> --- By the C standard "The behavior is defined only if both the original pointer and the result pointer are pointing at elements of the same array or one past the end of that array", yes, "The only defined value for l is 0 or 1" is correct, since c is an int* (instead of an object pointer). So does GCC actually makes some UB-based optimizations based on the assumption that "&c + l" is definitely undefined after seeing l is neither 0 nor 1? However, when you change "k + l + 2147634" to "k + (l + 2147634)" in j(), the issue is still there. This time, (l + 2147634) is definitely 0.
