https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78659
kargl at gcc dot gnu.org changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |kargl at gcc dot gnu.org
--- Comment #5 from kargl at gcc dot gnu.org ---
(In reply to Jerry DeLisle from comment #4)
> I have the ICE resolved, but also note that the Fortran 95 standard has the
> constraint on namelist statements and F2003 does not.
>
> Constraint - namelist-group-object shall not be an array dummy argument with
> a nonconstant bound, a variable with nonconstant character length, an
> automatic object, a pointer, a variable of a type that has an ultimate
> component that is a pointer, or an allocatable array.
>
> We can got two ways here, forget about the F95 standard and accept it, or
> give an error with -std=f95. This is sort of troublsome because we have to
> maintain two or more behaviors if a user wants to check against F95. I
> wonder if we should get rid of the -std=f95 altogether and just keep
> -std=F2003 or similar.
>
> Any opinions?
I would prefer an error with -std=f95. This is what
gfc_notify_std() is used for. ISTR that list of things
in others places, but don't recall if we have a simply
function where you could do
if (check_list_of_stuff () && gfc_notify_std (GFC_F95,...))