https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84342
--- Comment #7 from Tamar Christina <tnfchris at gcc dot gnu.org> --- Jeff: ccfsm is indeed the problem here, as it does indeed decide quite late to remove a branch. Though we've tried emitting a nop instead of nothing just to test and this fixes the branch but causes another failure. So it looks like there might be multiple things wrong here. Haven't taken a look at the new failure yet. Alexandre: I assume you mean TARGET_RESET_LOCATION_VIEW? So for this we should just return a 1 for the branch?
