[Bug fortran/95879] [10/11/12 Regression] ICE in gfc_resolve_formal_arglist, at fortran/resolve.c:313

["anlauf at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95879

anlauf at gcc dot gnu.org changed:

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

--- Comment #6 from anlauf at gcc dot gnu.org ---
(In reply to sandra from comment #5)
> The ICE from the z1.90 example is gone on mainline now.  I don't know if the
> error on y1.f90 is correct or not without further research.

y1.f90 is invalid code.  However, gfortran's checking looks suspicious to me.

Rewriting the testcase slightly:

module m
  implicit none
contains
  integer function f(x) bind(c)
    use iso_c_binding, only: c_funloc
    real, intent(in) :: x
    c_funloc(f) = x
    f = 0
  end
end

This would give:

pr95879-y1.f90:7:13:

    7 |     c_funloc(f) = x
      |             1
Error: Function result 'f' at (1) is invalid as X argument to C_FUNLOC

I think this is wrong.

F2018: 18.2.3.5  C_FUNLOC (X)

Argument. X shall be a procedure; if it is a procedure pointer it shall be
associated. It shall not be a coindexed object.

Deactivating the related check in gfc_check_c_funloc, I get:

pr95879-y1.f90:7:4:

    7 |     c_funloc(f) = x
      |    1
Error: The function result on the lhs of the assignment at (1) must have the
pointer attribute.

That looks like a more reasonable error.