[Bug fortran/31219] New: ICE on valid code with gfortran

[Bug fortran/31219] New: ICE on valid code with gfortran

[jv244 at cam dot ac dot uk]

A recent gfortran ICEs on the following code:

INTEGER :: J
CHARACTER(LEN=8) :: str
J=3
write(str,'(2A4)') (/(F(I,J),I=1,2)/)
IF (str.NE." ODD EVE") CALL ABORT()
CONTAINS
 FUNCTION F(K,J) RESULT(I)
  INTEGER :: K,J
  CHARACTER(LEN=J) :: I
  IF (MODULO(K,2).EQ.0) THEN
     I="EVEN"
  ELSE
     I="ODD"
  ENDIF
 END FUNCTION
END


-- 
           Summary: ICE on valid code with gfortran
           Product: gcc
           Version: 4.3.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: fortran
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: jv244 at cam dot ac dot uk


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31219

[Bug fortran/31219] ICE on valid code with gfortran

[fxcoudert at gcc dot gnu dot org]

------- Comment #1 from fxcoudert at gcc dot gnu dot org  2007-03-16 15:23 -------
Reduced testcase:

$ cat a.f90
  call gee( [f(1)] )
contains
 function f(k) result(i)
  integer :: k
  character(len=k) :: i
 end function
end
$ ./bin/gfortran a.f90
a.f90:1: internal compiler error: Segmentation fault


The backtrace is:

Program received signal SIGSEGV, Segmentation fault.
get_frame_type (info=0x0) at /utmp/coudert/gfortran/trunk/gcc/tree-nested.c:199
199       tree type = info->frame_type;
(gdb) where
#0  get_frame_type (info=0x0)
    at /utmp/coudert/gfortran/trunk/gcc/tree-nested.c:199
#1  0x00000000006b8a04 in get_chain_decl (info=0xdcc3f0)
    at /utmp/coudert/gfortran/trunk/gcc/tree-nested.c:305
#2  0x00000000006b9af4 in get_nonlocal_debug_decl (info=0xdcc3f0, 
    decl=0x2a984ca000) at /utmp/coudert/gfortran/trunk/gcc/tree-nested.c:852
#3  0x00000000006bc1d3 in convert_nonlocal_reference (tp=0x2a984cecb0, 
    walk_subtrees=0x7fbfffea7c, data=Variable "data" is not available.
)
    at /utmp/coudert/gfortran/trunk/gcc/tree-nested.c:923
#4  0x00000000007b076b in walk_tree (tp=0x2a984cecb0, 
    func=0x6bc100 <convert_nonlocal_reference>, data=0x7fbfffec90, pset=0x0)
    at /utmp/coudert/gfortran/trunk/gcc/tree.c:8000
#5  0x00000000007b087f in walk_tree (tp=0x2a984cd448, 
    func=0x6bc100 <convert_nonlocal_reference>, data=0x7fbfffec90, pset=0x0)
    at /utmp/coudert/gfortran/trunk/gcc/tree.c:8235


-- 

fxcoudert at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
           Keywords|                            |ice-on-valid-code
      Known to fail|                            |4.1.3 4.2.0 4.3.0
   Last reconfirmed|0000-00-00 00:00:00         |2007-03-16 15:23:52
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31219

[Bug fortran/31219] ICE on valid code with gfortran

[fxcoudert at gcc dot gnu dot org]

------- Comment #2 from fxcoudert at gcc dot gnu dot org  2007-03-16 15:25 -------
Forgot to say: the original tree dump is

f (__result, .__result, k)
{
  (void) 0;


MAIN__ ()
{
  static void f (char[1:] &, int4, int4 &);

  _gfortran_set_std (68, 127, 0);
  {
    void * D.1377;
    void * D.1374;
    struct array1_unknown atmp.3;
    void * D.1367;
    struct array1_unknown atmp.0;

    atmp.0.dtype = (SAVE_EXPR <(<unnamed type>) (..__result + -1) + 1> << 6) +
49;
    atmp.0.dim[0].stride = 1;
    atmp.0.dim[0].lbound = 0;
    atmp.0.dim[0].ubound = 0;
    D.1367 = _gfortran_internal_malloc64 ((int8) (SAVE_EXPR <(<unnamed type>)
(..__result + -1) + 1>));
    atmp.0.data = D.1367;
    atmp.0.offset = 0;
    {
      char[1:D.1371] * pstr.2;
      int4 D.1371;
      int4 * D.1370;
      static int4 C.1369 = 1;

      D.1370 = &C.1369;
      D.1371 = *D.1370;
      pstr.2 = (char[1:D.1371] *) _gfortran_internal_malloc (D.1371);
      f (pstr.2, D.1371, D.1370);
      __builtin_memcpy (&(*(char[0:][1:..__result] *) atmp.0.data)[0], pstr.2,
D.1371);
      _gfortran_internal_free ((void *) pstr.2);
    }
    atmp.3.dtype = (SAVE_EXPR <(<unnamed type>) (..__result + -1) + 1> << 6) +
49;
    atmp.3.dim[0].stride = 1;
    atmp.3.dim[0].lbound = 0;
    atmp.3.dim[0].ubound = 0;
    D.1374 = _gfortran_internal_malloc64 ((int8) (SAVE_EXPR <(<unnamed type>)
(..__result + -1) + 1>));
    atmp.3.data = D.1374;
    atmp.3.offset = 0;
    {
      int8 S.4;

      S.4 = 0;
      while (1)
        {
          if (S.4 > 0) goto L.1;
          (*(char[0:][1:..__result] *) atmp.3.data)[S.4] =
(*(char[0:][1:..__result] *) atmp.0.data)[S.4];
          S.4 = S.4 + 1;
        }
      L.1:;
    }
    D.1377 = _gfortran_internal_pack (&atmp.3);
    gee (D.1377, ..__result);
    if (D.1377 != (char[0:][1:..__result] *) atmp.3.data)
      {
        _gfortran_internal_unpack (&atmp.3, D.1377);
        _gfortran_internal_free (D.1377);
      }
    _gfortran_internal_free (atmp.0.data);
    _gfortran_internal_free (atmp.3.data);
  }
}


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31219

[Bug fortran/31219] ICE on array of character function results

[pault at gcc dot gnu dot org]

------- Comment #3 from pault at gcc dot gnu dot org  2007-03-24 12:31 -------
Subject: Bug 31219

Author: pault
Date: Sat Mar 24 12:30:58 2007
New Revision: 123183

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=123183
Log:
2007-03-24  Paul Thomas  <pault@gcc.gnu.org>

        PR fortran/31215
        * trans-expr.c (gfc_apply_interface_mapping_to_expr): Return
        int result that is non-zero if the expression is the function
        result.  Only the characteristics of the result expression
        can be used in a procedure interface, so simplify LEN in situ
        using its character length.

        PR fortran/31219
        PR fortran/31200
        * trans-expr.c (gfc_conv_function_call): Do not use
        gfc_conv_expr_reference for actual pointer function with formal
        target because a temporary is created that does not transfer
        the reference correctly.  Do not indirect formal pointer
        functions since it is the function reference that is needed.

2007-03-24  Paul Thomas  <pault@gcc.gnu.org>

        PR fortran/31219
        * gfortran.dg/pointer_function_actual_1.f90: New test.

        PR fortran/31200
        * gfortran.dg/pointer_function_actual_2.f90: New test.

        PR fortran/31215
        * gfortran.dg/result_in_spec_1.f90: New test.

Added:
    trunk/gcc/testsuite/gfortran.dg/pointer_function_actual_1.f90
    trunk/gcc/testsuite/gfortran.dg/pointer_function_actual_2.f90
    trunk/gcc/testsuite/gfortran.dg/result_in_spec_1.f90
Modified:
    trunk/gcc/fortran/ChangeLog
    trunk/gcc/fortran/trans-expr.c
    trunk/gcc/testsuite/ChangeLog


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31219

[Bug fortran/31219] ICE on array of character function results

[pault at gcc dot gnu dot org]

------- Comment #4 from pault at gcc dot gnu dot org  2007-03-24 13:56 -------
Fixed on trunk

Paul


-- 

pault at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31219

[Bug fortran/31219] ICE on array of character function results

[pault at gcc dot gnu dot org]

------- Comment #5 from pault at gcc dot gnu dot org  2007-03-24 14:00 -------
Sorry, this was not fixed - it was PR31209 that was fixed.

Paul


-- 

pault at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31219

[Bug fortran/31219] ICE on array of character function results

[pault at gcc dot gnu dot org]

------- Comment #6 from pault at gcc dot gnu dot org  2007-03-24 14:17 -------
Subject: Bug 31219

Author: pault
Date: Sat Mar 24 14:17:34 2007
New Revision: 123184

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=123184
Log:
2007-03-24  Paul Thomas  <pault@gcc.gnu.org>

        PR fortran/31215
        * trans-expr.c (gfc_apply_interface_mapping_to_expr): Return
        int result that is non-zero if the expression is the function
        result.  Only the characteristics of the result expression
        can be used in a procedure interface, so simplify LEN in situ
        using its character length.

        PR fortran/31219
        PR fortran/31200
        * trans-expr.c (gfc_conv_function_call): Do not use
        gfc_conv_expr_reference for actual pointer function with formal
        target because a temporary is created that does not transfer
        the reference correctly.  Do not indirect formal pointer
        functions since it is the function reference that is needed.

2007-03-24  Paul Thomas  <pault@gcc.gnu.org>

        PR fortran/31219
        * gfortran.dg/pointer_function_actual_1.f90: New test.

        PR fortran/31200
        * gfortran.dg/pointer_function_actual_2.f90: New test.

        PR fortran/31215
        * gfortran.dg/result_in_spec_1.f90: New test.

Modified:
    trunk/gcc/fortran/ChangeLog
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/testsuite/gfortran.dg/pointer_function_actual_1.f90


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31219

[Bug fortran/31219] ICE on array of character function results

[pault at gcc dot gnu dot org]

------- Comment #7 from pault at gcc dot gnu dot org  2007-05-22 21:56 -------
The patch below does the job but needs some sorting out.  In particular, the
function call generated by the calculation of the character length needs to be
supressed, in case the function has side effects.

It's regtesting now but I am pretty sure that it is OK.

Paul

Index: gcc/fortran/trans-array.c
===================================================================
*** gcc/fortran/trans-array.c   (revision 124953)
--- gcc/fortran/trans-array.c   (working copy)
*************** get_array_ctor_var_strlen (gfc_expr * ex
*** 1368,1374 ****
     Returns TRUE if all elements are character constants.  */

  bool
! get_array_ctor_strlen (gfc_constructor * c, tree * len)
  {
    bool is_const;

--- 1368,1374 ----
     Returns TRUE if all elements are character constants.  */

  bool
! get_array_ctor_strlen (stmtblock_t * block, gfc_constructor * c, tree * len)
  {
    bool is_const;

*************** get_array_ctor_strlen (gfc_constructor *
*** 1384,1390 ****
          break;

        case EXPR_ARRAY:
!         if (!get_array_ctor_strlen (c->expr->value.constructor, len))
            is_const = false;
          break;

--- 1384,1390 ----
          break;

        case EXPR_ARRAY:
!         if (!get_array_ctor_strlen (block, c->expr->value.constructor, len))
            is_const = false;
          break;

*************** get_array_ctor_strlen (gfc_constructor *
*** 1396,1411 ****
        default:
          is_const = false;

-         /* Hope that whatever we have possesses a constant character
-            length!  */
          if (!(*len && INTEGER_CST_P (*len)) && c->expr->ts.cl)
            {
!             gfc_conv_const_charlen (c->expr->ts.cl);
!             *len = c->expr->ts.cl->backend_decl;
            }
!         /* TODO: For now we just ignore anything we don't know how to
!            handle, and hope we can figure it out a different way.  */
!         break;
        }
      }

--- 1396,1418 ----
        default:
          is_const = false;

          if (!(*len && INTEGER_CST_P (*len)) && c->expr->ts.cl)
            {
!             gfc_se argse;
!             gfc_ss *ss;
!             ss = gfc_walk_expr (c->expr);
!             gfc_init_se (&argse, NULL);
!             if (ss == gfc_ss_terminator)
!               gfc_conv_expr (&argse, c->expr);
!             else
!               gfc_conv_expr_descriptor (&argse, c->expr, ss);
!             *len = gfc_evaluate_now (argse.string_length, &argse.pre);
!             gfc_add_block_to_block (block, &argse.pre);
!             gfc_add_block_to_block (block, &argse.post);
!             c->expr->ts.cl->backend_decl = *len;
            }
! 
!           break;
        }
      }

*************** gfc_trans_array_constructor (gfc_loopinf
*** 1595,1604 ****
    c = ss->expr->value.constructor;
    if (ss->expr->ts.type == BT_CHARACTER)
      {
!       bool const_string = get_array_ctor_strlen (c, &ss->string_length);
        if (!ss->string_length)
        gfc_todo_error ("complex character array constructors");

        type = gfc_get_character_type_len (ss->expr->ts.kind,
ss->string_length);
        if (const_string)
        type = build_pointer_type (type);
--- 1602,1614 ----
    c = ss->expr->value.constructor;
    if (ss->expr->ts.type == BT_CHARACTER)
      {
!       bool const_string = get_array_ctor_strlen (&loop->pre, c,
&ss->string_length);
        if (!ss->string_length)
        gfc_todo_error ("complex character array constructors");

+       ss->expr->ts.cl->backend_decl = ss->string_length;
+ 
+ 
        type = gfc_get_character_type_len (ss->expr->ts.kind,
ss->string_length);
        if (const_string)
        type = build_pointer_type (type);
Index: gcc/fortran/trans.h
===================================================================
*** gcc/fortran/trans.h (revision 124953)
--- gcc/fortran/trans.h (working copy)
*************** extern GTY(()) tree gfc_static_ctors;
*** 434,440 ****
  void gfc_generate_constructors (void);

  /* Get the string length of an array constructor.  */
! bool get_array_ctor_strlen (gfc_constructor *, tree *);

  /* Generate a runtime error check.  */
  void gfc_trans_runtime_check (tree, const char *, stmtblock_t *, locus *);
--- 434,440 ----
  void gfc_generate_constructors (void);

  /* Get the string length of an array constructor.  */
! bool get_array_ctor_strlen (stmtblock_t *, gfc_constructor *, tree *);

  /* Generate a runtime error check.  */
  void gfc_trans_runtime_check (tree, const char *, stmtblock_t *, locus *);
Index: gcc/fortran/trans-intrinsic.c
===================================================================
*** gcc/fortran/trans-intrinsic.c       (revision 124953)
--- gcc/fortran/trans-intrinsic.c       (working copy)
*************** gfc_conv_intrinsic_len (gfc_se * se, gfc
*** 2537,2543 ****
        /* Obtain the string length from the function used by
           trans-array.c(gfc_trans_array_constructor).  */
        len = NULL_TREE;
!       get_array_ctor_strlen (arg->value.constructor, &len);
        break;

      case EXPR_VARIABLE:
--- 2537,2543 ----
        /* Obtain the string length from the function used by
           trans-array.c(gfc_trans_array_constructor).  */
        len = NULL_TREE;
!       get_array_ctor_strlen (&se->pre, arg->value.constructor, &len);
        break;

      case EXPR_VARIABLE:


-- 

pault at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |pault at gcc dot gnu dot org
                   |dot org                     |
             Status|REOPENED                    |ASSIGNED
   Last reconfirmed|2007-03-16 15:23:52         |2007-05-22 21:56:32
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31219

[Bug fortran/31219] ICE on array of character function results

[patchapp at dberlin dot org]

------- Comment #8 from patchapp at dberlin dot org  2007-05-23 11:25 -------
Subject: Bug number PR31219

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/2007-05/msg01540.html


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31219

[Bug fortran/31219] ICE on array of character function results

[pault at gcc dot gnu dot org]

------- Comment #9 from pault at gcc dot gnu dot org  2007-05-26 11:25 -------
Subject: Bug 31219

Author: pault
Date: Sat May 26 11:25:36 2007
New Revision: 125088

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=125088
Log:
2007-05-26 Paul Thomas <pault@gcc.gnu.org>

        PR fortran/31219
        * trans.h : Add no_function_call bitfield to gfc_se structure.
        Add stmtblock_t argument to prototype of get_array_ctor_strlen.
        * trans-array.c (get_array_ctor_all_strlen): New function.
        (get_array_ctor_strlen): Add new stmtblock_t argument and call
        new function for character elements that are not constants,
        arrays or variables.
        (gfc_conv_array_parameter): Call get_array_ctor_strlen to get
        good string length.
        * trans-intrinsic (gfc_conv_intrinsic_len): Add new argument
        to call of get_array_ctor_strlen.


2007-05-26 Paul Thomas <pault@gcc.gnu.org>

        PR fortran/31219
        * gfortran.dg/array_constructor_17.f90: New test.


Added:
    trunk/gcc/testsuite/gfortran.dg/array_constructor_17.f90
Modified:
    trunk/gcc/fortran/ChangeLog
    trunk/gcc/fortran/trans-array.c
    trunk/gcc/fortran/trans-intrinsic.c
    trunk/gcc/fortran/trans.h
    trunk/gcc/testsuite/ChangeLog


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31219

[Bug fortran/31219] ICE on array of character function results

[burnus at gcc dot gnu dot org]

------- Comment #10 from burnus at gcc dot gnu dot org  2007-05-26 11:59 -------
Fixed in 4.3.0; no regression, not serious wrong-code PR -> no backporting to
4.2


-- 

burnus at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED
   Target Milestone|---                         |4.3.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31219