https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94355
--- Comment #12 from Jonathan Wakely <redi at gcc dot gnu.org> --- (In reply to Jonathan Wakely from comment #11) > Ideally it would have complained when we copied the pointer value in the > trivial copy constructor and didn't zero out the original. That shallow copy > was the actual bug, the double-free is a symptom of it. I've created PR 107534 for that.
