https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85872
--- Comment #2 from jakub.leszczak at nokia dot com --- > The dependence on TEST is beacause we limit the compile-time it takes to emit > those warnings. But compilation fails when TEST=1 that is when there is more code to compile. So if there is time limit for this warning then I would expect it failing when TEST=0 and passing when TEST=1 not the other way around. Or do I not understand what you meant?
