[Bug fortran/60718] Test case gfortran.dg/select_type_4.f90 fails on ARM

[mikael at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60718

Mikael Morin <mikael at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mikael at gcc dot gnu.org

--- Comment #7 from Mikael Morin <mikael at gcc dot gnu.org> ---
(In reply to Bernd Edlinger from comment #6)
> hmm,
> 
> gfc_build_class_symbol creates different types,
> if a class has the allocatable, target or pointer attribute.
> 
[...]
> 
> But the class(xxx), pointer is passed to class(xxx), target when
> used as a procedure parameter. this makes one type alias another.
> 
> maybe the test case is wrong, passing one flavor of object to
> another?
> 
This is not a problem; a pointer has the target attribute.
In fortran pointers are not first class citizens, so a pointer variable is the
pointer's target.

I would say that gfc_build_class_symbol should look at the attr->target instead
of attr->pointer, but I don't know how much fallout would come from that.