https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102685
anlauf at gcc dot gnu.org changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |NEW
Priority|P3 |P4
Last reconfirmed| |2021-10-11
Ever confirmed|0 |1
CC| |anlauf at gcc dot gnu.org
--- Comment #2 from anlauf at gcc dot gnu.org ---
That's arguably a regression.
gcc-11 accepted the invalid testcase in comment#0 but generated wrong code.
The real reason is that the constructor is not dealth with properly,
and there are several related PRs I am too lazy to look up now.
The accepts-invalid for comment#1 is old and not a regression for that reason.
gcc-7 had that already, and maybe much longer.