[Bug c++/13225] New: GDB fails to find non-overloaded C++ function

[Bug c++/13225] New: GDB fails to find non-overloaded C++ function

[jimb@red-bean.com]

http://sourceware.org/bugzilla/show_bug.cgi?id=13225

             Bug #: 13225
           Summary: GDB fails to find non-overloaded C++ function
           Product: gdb
           Version: unknown
            Status: NEW
          Severity: normal
          Priority: P2
         Component: c++
        AssignedTo: unassigned@sourceware.org
        ReportedBy: jimb@red-bean.com
    Classification: Unclassified


When I try to call JS_CompileFunction in the following test program, GDB says:

Cannot resolve function JS_CompileFunction to any overloaded instance

However, if I cast one of the zeros to "const char **", GDB can find the
function.

$ cat archer-bug.cpp
#include <stdlib.h>

typedef unsigned uintN;

struct JSContext;
struct JSFunction;
struct JSObject;

JSFunction *
JS_CompileFunction(JSContext *cx, JSObject *obj, const char *name,
                   uintN nargs, const char **argnames,
                   const char *bytes, size_t length,
                   const char *filename, uintN lineno) { }


int main(int argc, char **argv) {
  JSContext *cx = NULL;
  JSObject *global = NULL;
  JS_CompileFunction(cx, global, "f", 0, 0, "1", 1, "foo.js", 1);
  return 0;
}
$ g++ -g archer-bug.cpp -o archer-bug$ ~/gdb/bin/gdb archer-bug
GNU gdb (GDB) 7.3.50.20110925-cvs
Copyright (C) 2011 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-unknown-linux-gnu".
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>...
Reading symbols from /home/jimb/play/archer-bug...done.
(gdb) start
Temporary breakpoint 1 at 0x4005a1: file archer-bug.cpp, line 17.
Starting program: /home/jimb/play/archer-bug 

Temporary breakpoint 1, main (argc=1, argv=0x7fffffffe738) at archer-bug.cpp:17
17      JSContext *cx = NULL;
(gdb) next
18      JSObject *global = NULL;
(gdb) 
19      JS_CompileFunction(cx, global, "f", 0, 0, "1", 1, "foo.js", 1);
(gdb) 
20      return 0;
(gdb) p JS_CompileFunction(cx, global, "f", 0, 0, "1", 1, "foo.js", 1)
Cannot resolve function JS_CompileFunction to any overloaded instance
(gdb) p JS_CompileFunction(cx, global, "f", 0, (const char **) 0, "1", 1,
"foo.js", 1)
$1 = NULL
(gdb)

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

[Bug c++/13225] GDB fails to find non-overloaded C++ function

[keiths at redhat dot com]

http://sourceware.org/bugzilla/show_bug.cgi?id=13225

Keith Seitz <keiths at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
                 CC|                            |keiths at redhat dot com
         AssignedTo|unassigned at sourceware    |keiths at redhat dot com
                   |dot org                     |

--- Comment #1 from Keith Seitz <keiths at redhat dot com> 2011-10-11 18:53:23 UTC ---
There are actually two problems mentioned here. First, the fact that the parser
is seeing the NULL ("0") char const** parameter as an INT, but the evaluator
will not allow that conversion. I don't see why we couldn't permit that. I can
imagine scenarios where the user might want to pass an address this way. While
a real source file would require the cast, there is no reason why we cannot be
a little more permissive in the debugger.

Second, there is a bug alluded to by the OP: casting the first "0" to some
other type will get gdb to "work." This is happening because when the
function's overload suitability is calculated, the code is currently early
returning on the very first non-STANDARD rank that it sees, instead of
returning the worst one that it sees.

Patch/tests pending.

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

[Bug c++/13225] GDB fails to find non-overloaded C++ function

[keiths at redhat dot com]

http://sourceware.org/bugzilla/show_bug.cgi?id=13225

--- Comment #2 from Keith Seitz <keiths at redhat dot com> 2011-10-11 19:24:27 UTC ---
Patch submitted:
http://sourceware.org/ml/gdb-patches/2011-10/msg00327.html

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

[Bug c++/13225] GDB fails to find non-overloaded C++ function

[jimb@red-bean.com]

http://sourceware.org/bugzilla/show_bug.cgi?id=13225

--- Comment #3 from Jim Blandy <jimb@red-bean.com> 2011-10-11 19:30:21 UTC ---
Thanks for the patch, Keith!

You say:

While a real source file would require the cast, there is no reason why we
cannot be a little more permissive in the debugger.

I don't think that's correct. The transcript (with its missing newline after
the g++ command...) shows that the expression I try to evaluate in GDB is
identical to the expression appearing in the source code, which the compiler
accepts without complaint. The C++ language does not require that cast.

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

[Bug c++/13225] GDB fails to find non-overloaded C++ function

[cvs-commit at gcc dot gnu.org]

http://sourceware.org/bugzilla/show_bug.cgi?id=13225

--- Comment #4 from cvs-commit at gcc dot gnu.org <cvs-commit at gcc dot gnu.org> 2011-10-14 20:22:26 UTC ---
CVSROOT:    /cvs/src
Module name:    src
Changes by:    kseitz@sourceware.org    2011-10-14 20:22:17

Modified files:
    gdb            : ChangeLog eval.c gdbtypes.h gdbtypes.c 
                     valarith.c valops.c value.h 

Log message:
    PR c++/13225
    * eval.c (evaluate_subexp_standard): Do not construct
    an array of types; pass the value array directly to
    find_overload_match.
    * gdbtypes.h (NULL_POINTER_CONVERSION_BADNESS): Declare.
    (rank_function): Take an array of values instead of types.
    (rank_one_type): Add struct value * parameter.
    * gdbtypes.c (NULL_POINTER_CONVERSION_BADNESS): Define.
    (rank_function): For each argument, pass the argument's
    value to rank_one_type.
    (rank_one_type): Add VALUE parameter.
    If the parameter type is a pointer and the argument type
    is an integer, return NULL_POINTER_CONVERSION_BADNESS if
    VALUE is zero.
    Update all calls to rank_one_type, passing NULL for new
    VALUE parameter.
    * valarith.c (value_user_defined_cpp_op): Do not construct
    an array of types; pass the value array directly to
    find_overload_match.
    * valops.c (find_overload_method_list): Take an array of
    values instead of types.
    Save the type of OBJP for later use.
    Update calls to find_oload_champ, and find_oload_champ_namespace.
    (find_oload_champ_namespace): Take an array of values instead
    of types.
    (find_oload_champ_namespace_loop): Likewise.
    (find_oload_champ): Likewise.
    (classify_oload_match): Inspect all arguments
    until INCOMPATIBLE is found. Return the worst badness found
    otherwise.
    (compare_parameters): Update call to rank_one_type.
    * value.h (find_overload_match): Take an array of values instead
    of types.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/src/gdb/ChangeLog.diff?cvsroot=src&r1=1.13437&r2=1.13438
http://sourceware.org/cgi-bin/cvsweb.cgi/src/gdb/eval.c.diff?cvsroot=src&r1=1.154&r2=1.155
http://sourceware.org/cgi-bin/cvsweb.cgi/src/gdb/gdbtypes.h.diff?cvsroot=src&r1=1.156&r2=1.157
http://sourceware.org/cgi-bin/cvsweb.cgi/src/gdb/gdbtypes.c.diff?cvsroot=src&r1=1.219&r2=1.220
http://sourceware.org/cgi-bin/cvsweb.cgi/src/gdb/valarith.c.diff?cvsroot=src&r1=1.99&r2=1.100
http://sourceware.org/cgi-bin/cvsweb.cgi/src/gdb/valops.c.diff?cvsroot=src&r1=1.287&r2=1.288
http://sourceware.org/cgi-bin/cvsweb.cgi/src/gdb/value.h.diff?cvsroot=src&r1=1.187&r2=1.188

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

[Bug c++/13225] GDB fails to find non-overloaded C++ function

[cvs-commit at gcc dot gnu.org]

http://sourceware.org/bugzilla/show_bug.cgi?id=13225

--- Comment #5 from cvs-commit at gcc dot gnu.org <cvs-commit at gcc dot gnu.org> 2011-10-14 20:23:01 UTC ---
CVSROOT:    /cvs/src
Module name:    src
Changes by:    kseitz@sourceware.org    2011-10-14 20:22:50

Modified files:
    gdb/testsuite  : ChangeLog 
    gdb/testsuite/gdb.cp: converts.exp converts.cc 

Log message:
    PR c++/13225
    * gdb.cp/converts.cc (foo3_1): New function.
    (foo3_2): New functions.
    * gdb.cp/converts.exp: Add tests for int to pointer conversion
    and null pointer conversions of integer constant zero.
    Add test to check if all arguments are checked for incompatible
    conversion BADNESS.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/src/gdb/testsuite/ChangeLog.diff?cvsroot=src&r1=1.2897&r2=1.2898
http://sourceware.org/cgi-bin/cvsweb.cgi/src/gdb/testsuite/gdb.cp/converts.exp.diff?cvsroot=src&r1=1.3&r2=1.4
http://sourceware.org/cgi-bin/cvsweb.cgi/src/gdb/testsuite/gdb.cp/converts.cc.diff?cvsroot=src&r1=1.2&r2=1.3

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

[Bug c++/13225] GDB fails to find non-overloaded C++ function

[keiths at redhat dot com]

http://sourceware.org/bugzilla/show_bug.cgi?id=13225

Keith Seitz <keiths at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED
   Target Milestone|---                         |7.4

--- Comment #6 from Keith Seitz <keiths at redhat dot com> 2011-10-14 20:30:39 UTC ---
Final patch committed.

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

[Bug c++/13225] GDB fails to find non-overloaded C++ function

[pedro at codesourcery dot com]

http://sourceware.org/bugzilla/show_bug.cgi?id=13225

Pedro Alves <pedro at codesourcery dot com> changed:

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

--- Comment #7 from Pedro Alves <pedro at codesourcery dot com> 2011-10-28 15:33:08 UTC ---
*** Bug 13356 has been marked as a duplicate of this bug. ***

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.