[Bug fortran/111218] New: Conflict in BIND(C) INTERFACEs in two Modules leads to ICE.

[Bug fortran/111218] New: Conflict in BIND(C) INTERFACEs in two Modules leads to ICE.

[toon at moene dot org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111218

            Bug ID: 111218
           Summary: Conflict in BIND(C) INTERFACEs in two Modules leads to
                    ICE.
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: fortran
          Assignee: unassigned at gcc dot gnu.org
          Reporter: toon at moene dot org
  Target Milestone: ---

The following program:

MODULE FIELD_2RM_UTIL_MODULE

INTERFACE

SUBROUTINE SET_ABOR1_EXCEPTION_HANDLER()
BIND(C,name="set_abor1_exception_handler")
END SUBROUTINE SET_ABOR1_EXCEPTION_HANDLER

END INTERFACE

END MODULE
MODULE FIELD_3RM_UTIL_MODULE

INTERFACE

SUBROUTINE SET_ABOR1_EXCEPTION_HANDLER()
BIND(C,name="set_abor1_exception_handler")
END SUBROUTINE SET_ABOR1_EXCEPTION_HANDLER

END INTERFACE

END MODULE
MODULE FIELD_UTIL_MODULE


USE FIELD_2RM_UTIL_MODULE
USE FIELD_3RM_UTIL_MODULE

IMPLICIT NONE

END MODULE

leads to the following internal compiler error:

/home/toon/compilers/install/gcc/bin/gfortran -c -g a.f90

in gfc_format_decoder, at fortran/error.cc:1078
0x75917d gfc_format_decoder
        /home/toon/compilers/gcc/gcc/fortran/error.cc:1078
0x2153e1f pp_format(pretty_printer*, text_info*)
        /home/toon/compilers/gcc/gcc/pretty-print.cc:1475
0x21315be diagnostic_report_diagnostic(diagnostic_context*, diagnostic_info*)
        /home/toon/compilers/gcc/gcc/diagnostic.cc:1606
0x9b628e gfc_report_diagnostic
        /home/toon/compilers/gcc/gcc/fortran/error.cc:890
0x9b628e gfc_error_opt
        /home/toon/compilers/gcc/gcc/fortran/error.cc:1460
0x9b7470 gfc_error(char const*, ...)
        /home/toon/compilers/gcc/gcc/fortran/error.cc:1489
0xa6205b ambiguous_symbol
        /home/toon/compilers/gcc/gcc/fortran/symbol.cc:3167
0xa6ce9e gfc_find_sym_tree(char const*, gfc_namespace*, int, gfc_symtree**)
        /home/toon/compilers/gcc/gcc/fortran/symbol.cc:3240
0xa6cec1 gfc_find_symbol(char const*, gfc_namespace*, int, gfc_symbol**)
        /home/toon/compilers/gcc/gcc/fortran/symbol.cc:3291
0xb27a05 check_against_globals
        /home/toon/compilers/gcc/gcc/fortran/frontend-passes.cc:5842
0xa630e2 do_traverse_symtree
        /home/toon/compilers/gcc/gcc/fortran/symbol.cc:4190
0xb30231 gfc_check_externals(gfc_namespace*)
        /home/toon/compilers/gcc/gcc/fortran/frontend-passes.cc:5888
0xa293d8 gfc_parse_file()
        /home/toon/compilers/gcc/gcc/fortran/parse.cc:7195
0xa7aecf gfc_be_parse_file
        /home/toon/compilers/gcc/gcc/fortran/f95-lang.cc:229
Please submit a full bug report, with preprocessed source (by using
-freport-bug).
Please include the complete backtrace with any bug report.
See <https://gcc.gnu.org/bugs/> for instructions.

(tested with gcc version 14.0.0 20230828 (experimental) [master
r14-3528-gc3669bb677b] (GCC)

[Bug fortran/111218] Conflict in BIND(C) INTERFACEs in two Modules leads to ICE.

[kargl at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111218

kargl at gcc dot gnu.org changed:

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

--- Comment #1 from kargl at gcc dot gnu.org ---

> (tested with gcc version 14.0.0 20230828 (experimental) [master
> r14-3528-gc3669bb677b] (GCC)

No ICE with a 14.0.0 20230824 gfortran

[Bug fortran/111218] Conflict in BIND(C) INTERFACEs in two Modules leads to ICE.

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111218

anlauf at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|ice-on-invalid-code         |ice-on-valid-code
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2023-08-29
     Ever confirmed|0                           |1

--- Comment #2 from anlauf at gcc dot gnu.org ---
I get an ICE for all gcc >= 11 here, and I think the code is actually valid.
(Other compilers, like NAG, Intel, NVidia accept it).

Might be a regression, as my installs of 7 <= gcc <= 10 all pass.

[Bug fortran/111218] Conflict in BIND(C) INTERFACEs in two Modules leads to ICE.

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111218

--- Comment #3 from anlauf at gcc dot gnu.org ---
A workaround is to add a 'private' statement to any of the first two modules.

[Bug fortran/111218] Conflict in BIND(C) INTERFACEs in two Modules leads to ICE.

[rimvydas.jas at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111218

Rimvydas (RJ) <rimvydas.jas at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |rimvydas.jas at gmail dot com

--- Comment #4 from Rimvydas (RJ) <rimvydas.jas at gmail dot com> ---
Another testcase resulting in identical ICE.

$ cat test.f90
module foo1
  interface !bail ! generic works ok
    subroutine bail() bind(c,name="bail_")
    end subroutine
  end interface
end

module foo2
  interface !bail
    subroutine sfx_bail(k)
      integer :: k
    end subroutine
  end interface
end

module foo
  use foo1
  use foo2 , bail => sfx_bail
end

subroutine unice_dummy ! avoid ice
end

$ ./gcc/f951 test.f90
test.f90:21:22:

   21 | subroutine unice_dummy
      |                      1
Error: Name 'bail' at (1) is an ambiguous reference to 'bail' from module
'foo1'

Using extra dummy unit at the end of file helps to avoid the ICE (avoids ICE on
original case too), but notice the wrong locus.  Issue here happens because
ambiguous_symbol() is called after parsing was already done, resulting in
nullification of gfc_current_locus global in gfc_advance_line () at
gcc/fortran/scanner.cc:651
651           gfc_current_locus.nextc = NULL;
So the "%C" cannot be used in gcc/fortran/symbol.cc:3167
      gfc_error ("Name %qs at %C is an ambiguous reference to %qs "
Note: there is an assertion in gcc/fortran/error.cc:1078
      gcc_assert (loc->nextc - loc->lb->line >= 0);


Question is weather bind(c) rename attribute should somehow affect interface
generic status in any way?
Testcase does pass with bind(c) commented out or with explicitly marked
interface as generic in foo1.

[Bug fortran/111218] Conflict in BIND(C) INTERFACEs in two Modules leads to ICE.

[kargl at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111218

--- Comment #5 from kargl at gcc dot gnu.org ---
Interesting bug.  If one puts a break point ...

0x75917d gfc_format_decoder
        /home/toon/compilers/gcc/gcc/fortran/error.cc:1078
0x2153e1f pp_format(pretty_printer*, text_info*)
        /home/toon/compilers/gcc/gcc/pretty-print.cc:1475
0x21315be diagnostic_report_diagnostic(diagnostic_context*, diagnostic_info*)
        /home/toon/compilers/gcc/gcc/diagnostic.cc:1606
0x9b628e gfc_report_diagnostic
        /home/toon/compilers/gcc/gcc/fortran/error.cc:890
0x9b628e gfc_error_opt
        /home/toon/compilers/gcc/gcc/fortran/error.cc:1460
0x9b7470 gfc_error(char const*, ...)
        /home/toon/compilers/gcc/gcc/fortran/error.cc:1489
0xa6205b ambiguous_symbol
        /home/toon/compilers/gcc/gcc/fortran/symbol.cc:3167

here, one ends up at 

        gfc_error ("Name %qs at %C is an ambiguous reference to %qs from"
                   "module %qs", name, st->n.sym->name, st->n.sym->module);

%C means one should use gfc_current_locus to print the offending line.
If one looks are gfc_current_locus, one sees gfc_current_locus.lb = 0x0,
an infamous NULL pointer dereference then occurs.  Ouch.  This patch 
fixes the issue:

diff --git a/gcc/fortran/symbol.cc b/gcc/fortran/symbol.cc
index a6078bc608a..5b726f8b715 100644
--- a/gcc/fortran/symbol.cc
+++ b/gcc/fortran/symbol.cc
@@ -3164,8 +3164,16 @@ ambiguous_symbol (const char *name, gfc_symtree *st)
     return;

   if (st->n.sym->module)
-    gfc_error ("Name %qs at %C is an ambiguous reference to %qs "
-              "from module %qs", name, st->n.sym->name, st->n.sym->module);
+    {
+      if (gfc_current_locus.lb)
+       gfc_error ("Name %qs at %C is an ambiguous reference to %qs from"
+                  "module %qs", name, st->n.sym->name, st->n.sym->module);
+      else
+       gfc_error ("Name %qs is an ambiguous reference to %qs, which is "
+                  "available through USE association from module %qs at %L",
+                  name, st->n.sym->name, st->n.sym->module,
+                  &st->n.sym->declared_at);
+    }
   else
     gfc_error ("Name %qs at %C is an ambiguous reference to %qs "
               "from current program unit", name, st->n.sym->name);

[Bug fortran/111218] Conflict in BIND(C) INTERFACEs in two Modules leads to ICE.

[kargl at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111218

kargl at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P4

--- Comment #6 from kargl at gcc dot gnu.org ---

>    if (st->n.sym->module)
> -    gfc_error ("Name %qs at %C is an ambiguous reference to %qs "
> -	       "from module %qs", name, st->n.sym->name, st->n.sym->module);
> +    {
> +      if (gfc_current_locus.lb)
> +	gfc_error ("Name %qs at %C is an ambiguous reference to %qs from"

A space is missing after "... from".  Should be "... from "

> +		   "module %qs", name, st->n.sym->name, st->n.sym->module);
> +      else
> +	gfc_error ("Name %qs is an ambiguous reference to %qs, which is "
> +		   "available through USE association from module %qs at %L",
> +		   name, st->n.sym->name, st->n.sym->module,
> +		   &st->n.sym->declared_at);
> +    }
>    else
>      gfc_error ("Name %qs at %C is an ambiguous reference to %qs "
>  	       "from current program unit", name, st->n.sym->name);

The original code is too painful to read (aka uppercase and no formatting).
The claim is that the code is valid.  Maybe it is as the conflicting 
names in the modules are not referenced in the program unit.  If OP
references SET_ABOR1_EXCEPTION_HANDLER, good luck.

[Bug fortran/111218] Conflict in BIND(C) INTERFACEs in two Modules leads to ICE.

[kargl at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111218

--- Comment #7 from kargl at gcc dot gnu.org ---
An alternative patch that allows the original code to compile is

diff --git a/gcc/fortran/symbol.cc b/gcc/fortran/symbol.cc
index a6078bc608a..34eb3a65e8f 100644
--- a/gcc/fortran/symbol.cc
+++ b/gcc/fortran/symbol.cc
@@ -3160,7 +3160,7 @@ static void
 ambiguous_symbol (const char *name, gfc_symtree *st)
 {

-  if (st->n.sym->error)
+  if (st->n.sym->error || (st->n.sym->module && !gfc_current_locus.lb))
     return;

   if (st->n.sym->module)

This one simply returns.  If one actually tries to reference the
conflicting entity an error message is generated.