[Bug testsuite/44798] inconsistent interfaces

[Bug testsuite/44798] inconsistent interfaces

[zeccav at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44798

--- Comment #1 from Vittorio Zecca <zeccav at gmail dot com> ---
The problem is still present in 4.8.2
If nobody is interested in fixing the test case I trust it is just as good
to close this issue.

[Bug testsuite/44798] inconsistent interfaces

[dominiq at lps dot ens.fr]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44798

Dominique d'Humieres <dominiq at lps dot ens.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2013-10-19
                 CC|                            |janus at gcc dot gnu.org
     Ever confirmed|0                           |1

--- Comment #2 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
> INTRINSIC_ASSOCIATED.f90 lines 38 and 124 are inconsistent. 
> I believe the same array bounds should be specified.

This has been fixed on trunk by revision 197922:

...
subroutine sub1 (a, ap)
   integer, pointer :: ap(:, :)
   integer, target :: a(10, 10)    <--- 38

   ap => a
end
...
subroutine associated_2 ()
   integer, pointer :: xp(:, :)
   integer, target  :: x(10, 10)
   integer, target  :: y(100, 100)
   interface
      subroutine sub1 (a, ap)
         integer, pointer :: ap(:, :)
         integer, target  :: a(10, 10) <--- 124
      end
   endinterface

   xp => y
   if (.not. associated (xp)) call abort ()
   call sub1 (x, xp)
   if (associated (xp, y)) call abort ()
   if (.not. associated (xp, x)) call abort ()
end

but the fix has not been backported. The patch was

---
trunk/gcc/testsuite/gfortran.fortran-torture/execute/intrinsic_associated.f90  
 2013/04/12 14:19:10    197921
+++
trunk/gcc/testsuite/gfortran.fortran-torture/execute/intrinsic_associated.f90  
 2013/04/12 14:21:39    197922
@@ -121,7 +121,7 @@
    interface
       subroutine sub1 (a, ap)
          integer, pointer :: ap(:, :)
-         integer, target  :: a(10, 1)
+         integer, target  :: a(10, 10)
       end
    endinterface

[Bug testsuite/44798] inconsistent interfaces

[zeccav at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44798

--- Comment #3 from Vittorio Zecca <zeccav at gmail dot com> ---
It would be even better if gfortran detected the inconsistent interfaces,
as NAG nagfor does, as in the following:

nagfor -S gfortran.fortran-torture/execute/intrinsic_associated.f90
NAG Fortran Compiler Release 5.3.2(968)
Error: gfortran.fortran-torture/execute/intrinsic_associated.f90:
Inconsistent INTERFACE block for procedure SUB1 from ASSOCIATED_2
       Argument A (no. 1) has a different array size
[NAG Fortran Compiler error termination, 2 errors]

[Bug testsuite/44798] inconsistent interfaces

[zeccav at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=44798

Vittorio Zecca <zeccav at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |FIXED

--- Comment #4 from Vittorio Zecca <zeccav at gmail dot com> ---
This issue has been resolved by adjusting the array bounds.