https://gcc.gnu.org/bugzilla/show_bug.cgi?id=124107
--- Comment #6 from Liam Powell <liam at liampwll dot com> --- The new testcase is *technically* still using a dangling reference according to the exact wording of the RM. I just want to show that this has nothing to do with using Unchecked_Access and clobbering the memory. The subpool still remains attached to the pool after finalization without that.
