[Bug fortran/101536] New: ICE in gfc_conv_expr_descriptor, at fortran/trans-array.c:7324

[Bug fortran/101536] New: ICE in gfc_conv_expr_descriptor, at fortran/trans-array.c:7324

[gscfq@t-online.de]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101536

            Bug ID: 101536
           Summary: ICE in gfc_conv_expr_descriptor, at
                    fortran/trans-array.c:7324
           Product: gcc
           Version: 12.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: fortran
          Assignee: unassigned at gcc dot gnu.org
          Reporter: gscfq@t-online.de
  Target Milestone: ---

An invalid argument affects versions down to at least r5 :


$ cat z1.f90
program p
   type t
   end type
contains
   integer function f(x)
      class(t), allocatable :: x(:)
      f = size(x(1))
   end
end


$ cat z2.f90
program p
   type t
   end type
contains
   integer function f(x)
      class(t) :: x(:)
      f = size(x(1))
   end
end


$ gfortran-12-20210718 -c z1.f90
z1.f90:7:20:

    7 |       f = size(x(1))
      |                    1
internal compiler error: in gfc_conv_expr_descriptor, at
fortran/trans-array.c:7324
0x8093ab gfc_conv_expr_descriptor(gfc_se*, gfc_expr*)
        ../../gcc/fortran/trans-array.c:7324
0x84c9a2 gfc_conv_intrinsic_size
        ../../gcc/fortran/trans-intrinsic.c:8062
0x86a6b7 gfc_conv_intrinsic_function(gfc_se*, gfc_expr*)
        ../../gcc/fortran/trans-intrinsic.c:10702
0x839faa gfc_conv_expr(gfc_se*, gfc_expr*)
        ../../gcc/fortran/trans-expr.c:9018
0x846081 gfc_trans_assignment_1
        ../../gcc/fortran/trans-expr.c:11311
0x7f57f7 trans_code
        ../../gcc/fortran/trans.c:1918
0x8287d5 gfc_generate_function_code(gfc_namespace*)
        ../../gcc/fortran/trans-decl.c:6898
0x8284dc gfc_generate_contained_functions
        ../../gcc/fortran/trans-decl.c:5892
0x8284dc gfc_generate_function_code(gfc_namespace*)
        ../../gcc/fortran/trans-decl.c:6830
0x7a0026 translate_all_program_units
        ../../gcc/fortran/parse.c:6464
0x7a0026 gfc_parse_file()
        ../../gcc/fortran/parse.c:6733
0x7ecf1f gfc_be_parse_file
        ../../gcc/fortran/f95-lang.c:216

[Bug fortran/101536] ICE in gfc_conv_expr_descriptor, at fortran/trans-array.c:7324

[gscfq@t-online.de]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101536

G. Steinmetz <gscfq@t-online.de> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |ice-on-invalid-code

--- Comment #1 from G. Steinmetz <gscfq@t-online.de> ---

Detected with "type" instead of "class" :


$ cat z0.f90
program p
   type t
   end type
contains
   integer function f(x)
      type(t), allocatable :: x(:)
      f = size(x(1))
   end
end


$ gfortran-12-20210718 -c z0.f90
z0.f90:7:15:

    7 |       f = size(x(1))
      |               1
Error: 'array' argument of 'size' intrinsic at (1) must be an array

[Bug fortran/101536] ICE in gfc_conv_expr_descriptor, at fortran/trans-array.c:7324

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101536

anlauf at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2021-07-20

--- Comment #2 from anlauf at gcc dot gnu.org ---
Confirmed.

I am testing the following:

diff --git a/gcc/fortran/check.c b/gcc/fortran/check.c
index 27bf3a7eafe..6d2d9fe4007 100644
--- a/gcc/fortran/check.c
+++ b/gcc/fortran/check.c
@@ -735,6 +735,10 @@ array_check (gfc_expr *e, int n)
        && CLASS_DATA (e)->attr.dimension
        && CLASS_DATA (e)->as->rank)
     {
+      if (e->ref && e->ref->type == REF_ARRAY
+         && e->ref->u.ar.type == AR_ELEMENT)
+       goto error;
+
       gfc_add_class_array_ref (e);
       return true;
     }
@@ -742,6 +746,7 @@ array_check (gfc_expr *e, int n)
   if (e->rank != 0 && e->ts.type != BT_PROCEDURE)
     return true;

+error:
   gfc_error ("%qs argument of %qs intrinsic at %L must be an array",
             gfc_current_intrinsic_arg[n]->name, gfc_current_intrinsic,
             &e->where);

[Bug fortran/101536] ICE in gfc_conv_expr_descriptor, at fortran/trans-array.c:7324

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101536

anlauf at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Assignee|unassigned at gcc dot gnu.org      |anlauf at gcc dot gnu.org
             Status|NEW                         |ASSIGNED
                 CC|                            |anlauf at gcc dot gnu.org

--- Comment #3 from anlauf at gcc dot gnu.org ---
Patch: https://gcc.gnu.org/pipermail/fortran/2021-July/056263.html

[Bug fortran/101536] ICE in gfc_conv_expr_descriptor, at fortran/trans-array.c:7324

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101536

--- Comment #4 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Harald Anlauf <anlauf@gcc.gnu.org>:

https://gcc.gnu.org/g:e314cfc371d8b2405a1d81e51b90f9fb24b9061f

commit r12-2500-ge314cfc371d8b2405a1d81e51b90f9fb24b9061f
Author: Harald Anlauf <anlauf@gmx.de>
Date:   Fri Jul 23 21:00:10 2021 +0200

    Fortran: extend check for array arguments and reject CLASS array elements.

    gcc/fortran/ChangeLog:

            PR fortran/101536
            * check.c (array_check): Adjust check for the case of CLASS
            arrays.

    gcc/testsuite/ChangeLog:

            PR fortran/101536
            * gfortran.dg/pr101536.f90: New test.

[Bug fortran/101536] ICE in gfc_conv_expr_descriptor, at fortran/trans-array.c:7324

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101536

--- Comment #5 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-11 branch has been updated by Harald Anlauf
<anlauf@gcc.gnu.org>:

https://gcc.gnu.org/g:35c7067f7b31682e4d5da7ec83759ab1a17cead9

commit r11-8813-g35c7067f7b31682e4d5da7ec83759ab1a17cead9
Author: Harald Anlauf <anlauf@gmx.de>
Date:   Fri Jul 23 21:00:10 2021 +0200

    Fortran: extend check for array arguments and reject CLASS array elements.

    gcc/fortran/ChangeLog:

            PR fortran/101536
            * check.c (array_check): Adjust check for the case of CLASS
            arrays.

    gcc/testsuite/ChangeLog:

            PR fortran/101536
            * gfortran.dg/pr101536.f90: New test.

    (cherry picked from commit e314cfc371d8b2405a1d81e51b90f9fb24b9061f)

[Bug fortran/101536] ICE in gfc_conv_expr_descriptor, at fortran/trans-array.c:7324

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101536

--- Comment #6 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-10 branch has been updated by Harald Anlauf
<anlauf@gcc.gnu.org>:

https://gcc.gnu.org/g:95cda605a8f7e6b54c428b7d2c37b4f2ebe4a8d8

commit r10-10014-g95cda605a8f7e6b54c428b7d2c37b4f2ebe4a8d8
Author: Harald Anlauf <anlauf@gmx.de>
Date:   Fri Jul 23 21:00:10 2021 +0200

    Fortran: extend check for array arguments and reject CLASS array elements.

    gcc/fortran/ChangeLog:

            PR fortran/101536
            * check.c (array_check): Adjust check for the case of CLASS
            arrays.

    gcc/testsuite/ChangeLog:

            PR fortran/101536
            * gfortran.dg/pr101536.f90: New test.

    (cherry picked from commit e314cfc371d8b2405a1d81e51b90f9fb24b9061f)

[Bug fortran/101536] ICE in gfc_conv_expr_descriptor, at fortran/trans-array.c:7324

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101536

anlauf at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
   Target Milestone|---                         |10.4
         Resolution|---                         |FIXED

--- Comment #7 from anlauf at gcc dot gnu.org ---
Fixed on mainline for gcc-12, and on 11- and 10-branch.  Not a regression,
so not backporting further.  Closing.

Thanks for the report!