https://gcc.gnu.org/bugzilla/show_bug.cgi?id=125805
Drea Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |NEW
Ever confirmed|0 |1
Last reconfirmed| |2026-06-16
Known to fail| |4.9.4
--- Comment #1 from Drea Pinski <pinskia at gcc dot gnu.org> ---
Confirmed. Definitely not a regression. GCC thinks the size (internally) of an
incomplete type might be 0 and that a1 and a2 could have the same address.