https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91889
Marek Polacek <mpolacek at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |jason at gcc dot gnu.org
--- Comment #5 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
I think this is not a bug, just a consequence of DR 2352.
The ICS in
void f(const int* const &);
converting "int *" to "const int *const &" no longer requires a qualification
conversion, and so its rank is cr_identity (before the DR it would be cr_exact
because of the qualification conversion).
The ICS in
void f(int*);
is and always was cr_identity too, because it's simply "int *" to "int *".
So now we have two standard conversion sequences with the same rank, and none
of the tiebreakers in [over.ics.rank] seem to apply.
