https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122035
--- Comment #11 from Krzysztof Pecka <krzysiu858519851 at gmail dot com> --- (In reply to Andrew Pinski from comment #10) > Note my self contained example I attached I had actually made the change to: > index = line.find_first_of(semicolonChar); > > This is why at first I didn't see the issue :). > Also I suspect it is not there is a null character after the stack variable > but rather GCC knows that strlen(&a); has to be 0; otherwise it is undefined > for reading past the bounds of the "array" of 1. Interesting. I honestly forgot about null character. Still for me, interesting to see those different behaviors with different optimization flags.
