[Bug fortran/30865] New: optional argument passed on to size(...,dim=)

[Bug fortran/30865] New: optional argument passed on to size(...,dim=)

[tkoenig at gcc dot gnu dot org]

The following is legal (see 
<7dpCh.29496$5j1.2977@bgtnsc04-news.ops.worldnet.att.net>), but we segfault
on execution:

$ cat size.f90
module foo
contains
  subroutine checkv(ires,a1,opt1)
    integer :: a1(:,:)
    integer, optional :: opt1

    ires = size (a1, dim=opt1)
  end subroutine checkv
end module foo

program main
  use foo
  implicit none
  integer :: a(2,3)
  integer :: ires

  call checkv (ires, a)
  print *,ires
  call checkv (ires, a, 1)
  print *,ires
end program main
$ gfortran size.f90
$ ./a.out
Segmentation fault


-- 
           Summary: optional argument passed on to size(...,dim=)
           Product: gcc
           Version: 4.3.0
            Status: UNCONFIRMED
          Keywords: wrong-code
          Severity: normal
          Priority: P3
         Component: fortran
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: tkoenig at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865

[Bug fortran/30865] optional argument passed on to size(...,dim=)

[pault at gcc dot gnu dot org]

------- Comment #1 from pault at gcc dot gnu dot org  2007-02-20 05:34 -------
Confirmed:

The first call to checkv produces
    checkv (&ires, &parm.10, 0B);

With prototype: checkv (ires, a1, opt1), the call to size1 is
*ires = _gfortran_size1 (&parm.6, *opt1);

with the obvious consequences.

Paul

As an aside, Thomas, it seems to me that the fix is relatively easy:
trans-intrinsic.c:2702

  actual = actual->next;
  if (actual->expr)
    {
      gfc_init_se (&argse, NULL);
      gfc_conv_expr_type (&argse, actual->expr, gfc_array_index_type);
      gfc_add_block_to_block (&se->pre, &argse.pre);
      args = gfc_chainon_list (args, argse.expr);
      fndecl = gfor_fndecl_size1;
    }
  else
    fndecl = gfor_fndecl_size0;

We need a tree conditional that effects this:
fndecl = argse.pre == 0B ? gfor_fndecl_size0 : gfor_fndecl_size1


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865

[Bug fortran/30865] optional argument passed on to size(...,dim=)

[pault at gcc dot gnu dot org]

-- 

pault at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2007-02-20 05:35:18
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865

[Bug fortran/30865] optional argument passed on to size(...,dim=)

[burnus at gcc dot gnu dot org]

------- Comment #2 from burnus at gcc dot gnu dot org  2007-02-20 08:57 -------
> The following is legal but we segfault on execution:

>  subroutine checkv(ires,a1,opt1)
>    integer :: a1(:,:)
>    integer, optional :: opt1
>    ires = size (a1, dim=opt1)

For those who wonder (as I did) why using an optional argument is legal:
It is only used as actual argument corresponding to an optional dummy argument.
(cf. 12.4.1.6 in the Fortran 2003 standard).


-- 

burnus at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |burnus at gcc dot gnu dot
                   |                            |org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865

[Bug fortran/30865] optional argument passed on to size(...,dim=)

[burnus at gcc dot gnu dot org]

------- Comment #3 from burnus at gcc dot gnu dot org  2007-02-20 13:16 -------
(In reply to comment #2)
> For those who wonder (as I did) why using an optional argument is legal:
> It is only used as actual argument corresponding to an optional dummy 
> argument. (cf. 12.4.1.6 in the Fortran 2003 standard).

And on the fact that the optional argument does not change the rank.

This PR is seemingly based on the following c.l.fortran thread:
http://groups.google.com/group/comp.lang.fortran/browse_thread/thread/1e7a4db9d998955a


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865

[Bug fortran/30865] optional argument passed on to size(...,dim=)

[pault at gcc dot gnu dot org]

------- Comment #4 from pault at gcc dot gnu dot org  2007-02-20 14:35 -------
Created an attachment (id=13073)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13073&action=view)
Fix for the problem

I will set this regtesting in a few minutes and will post the result.  Note,
however, that I can do nothing with this for a few weeks. Therefore if one of
you want to bring it forth to the commit stage, please be my guest; note that
the code could be tidied up and the testcase needs dejagnufying.

Paul


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865

[Bug fortran/30865] optional argument passed on to size(...,dim=)

[pault at gcc dot gnu dot org]

------- Comment #5 from pault at gcc dot gnu dot org  2007-02-20 14:39 -------
(In reply to comment #4)
> Created an attachment (id=13073)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13073&action=view) [edit]
> Fix for the problem
I should have remarked that the patch produces the following code:

    D.1015 = &parm.6;
    D.1016 = (void *) opt1 != 0B;
    *ires = D.1016 ? _gfortran_size1 (D.1015, *opt1) : _gfortran_size0
(D.1015);

Paul


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865

[Bug fortran/30865] optional argument passed on to size(...,dim=)

[pault at gcc dot gnu dot org]

------- Comment #6 from pault at gcc dot gnu dot org  2007-02-20 17:25 -------
(In reply to comment #5)
> (In reply to comment #4)
> > Created an attachment (id=13073)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13073&action=view) [edit]
> > Fix for the problem
> I should have remarked that the patch produces the following code:
..and it regests and bootstraps on Cygwin_NT/i686

Paul


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865

[Bug fortran/30865] optional argument passed on to size(...,dim=)

[kargl at gcc dot gnu dot org]

------- Comment #7 from kargl at gcc dot gnu dot org  2007-02-21 19:02 -------
(In reply to comment #4)
> Created an attachment (id=13073)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13073&action=view) [edit]
> Fix for the problem
> 

Paul, I tried to apply your patch, but it is rejected.  Is your
code up-to-date?  In particular, Sandra'a recent CALL_EXPR patch caused
a lot of code churn in trans-*.c files.

steve


-- 

kargl at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |kargl at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865

[Bug fortran/30865] optional argument passed on to size(...,dim=)

[tkoenig at gcc dot gnu dot org]

------- Comment #8 from tkoenig at gcc dot gnu dot org  2007-02-24 21:48 -------
(In reply to comment #7)

> Paul, I tried to apply your patch, but it is rejected.

I applied the patch with a recent trunk, which worked (except
I got a message that the patch file contained CR/LF).

I'm running a regression-test now.

Thomas


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865

[Bug fortran/30865] optional argument passed on to size(...,dim=)

[patchapp at dberlin dot org]

------- Comment #9 from patchapp at dberlin dot org  2007-02-25 02:46 -------
Subject: Bug number PR 30865

A patch for this bug has been added to the patch tracker.
The mailing list url for the patch is
http://gcc.gnu.org/ml/gcc-patches/2007-02/msg01956.html


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865

[Bug fortran/30865] optional argument passed on to size(...,dim=)

[tkoenig at gcc dot gnu dot org]

------- Comment #10 from tkoenig at gcc dot gnu dot org  2007-02-26 21:16 -------
Subject: Bug 30865

Author: tkoenig
Date: Mon Feb 26 21:16:00 2007
New Revision: 122342

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=122342
Log:
2007-02-26  Thomas Koenig  <Thomas.Koenig@online.de>
            Paul Thomas  <pault@gcc.gnu.org>

        PR fortran/30865
        * trans-intrinsic.c (gfc_conv_intrinsic_size):
        If dim is an optional argument, check for its
        presence and call size0 or size1, respectively.

2007-02-26  Thomas Koenig  <Thomas.Koenig@online.de>

        PR fortran/30865
        * size_optional_dim_1.f90:  New test.


Added:
    trunk/gcc/testsuite/gfortran.dg/size_optional_dim_1.f90
Modified:
    trunk/gcc/fortran/ChangeLog
    trunk/gcc/fortran/trans-intrinsic.c
    trunk/gcc/testsuite/ChangeLog


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865

[Bug fortran/30865] [4.1, 4.2 only] optional argument passed on to size(...,dim=)

[burnus at gcc dot gnu dot org]

-- 

burnus at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |tkoenig at gcc dot gnu dot
                   |dot org                     |org
             Status|NEW                         |ASSIGNED
            Summary|optional argument passed on |[4.1, 4.2 only] optional
                   |to size(...,dim=)           |argument passed on to
                   |                            |size(...,dim=)


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865

[Bug fortran/30865] [4.1, 4.2 only] optional argument passed on to size(...,dim=)

[burnus at gcc dot gnu dot org]

------- Comment #11 from burnus at gcc dot gnu dot org  2007-02-27 17:46 -------
The following additional patch needs to be applied when backporting:
http://gcc.gnu.org/ml/fortran/2007-02/msg00620.html


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865

[Bug fortran/30865] [4.1, 4.2 only] optional argument passed on to size(...,dim=)

[burnus at gcc dot gnu dot org]

------- Comment #12 from burnus at gcc dot gnu dot org  2007-02-28 11:28 -------
> The following additional patch needs to be applied when backporting:
> http://gcc.gnu.org/ml/fortran/2007-02/msg00620.html

This two-line patch is unrelated.
(I think one should nonetheless backport it to 4.2 [is not in 4.1].)

The following is the real patch:
http://gcc.gnu.org/ml/gcc-patches/2007-02/msg02200.html


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865

[Bug fortran/30865] [4.1, 4.2 only] optional argument passed on to size(...,dim=)

[burnus at gcc dot gnu dot org]

------- Comment #13 from burnus at gcc dot gnu dot org  2007-03-01 08:19 -------
Subject: Bug 30865

Author: burnus
Date: Thu Mar  1 08:19:09 2007
New Revision: 122423

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=122423
Log:
2007-03-01  Tobias Burnus  <burnus@net-b.de>

        PR fortran/30865
        * trans-intrinsic.c (gfc_conv_intrinsic_size): Compare pointers.


Modified:
    trunk/gcc/fortran/ChangeLog
    trunk/gcc/fortran/trans-intrinsic.c


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865

[Bug fortran/30865] [4.1, 4.2 only] optional argument passed on to size(...,dim=)

[pault at gcc dot gnu dot org]

------- Comment #14 from pault at gcc dot gnu dot org  2007-03-27 09:59 -------
Not a regression from 4.1, so closing.

Paul


-- 

pault at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30865