[Bug debug/53235] New: [4.8 Regression] 20120504 broke -fdebug-types-section

[Bug debug/53235] New: [4.8 Regression] 20120504 broke -fdebug-types-section

[jan.kratochvil at redhat dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

             Bug #: 53235
           Summary: [4.8 Regression] 20120504 broke -fdebug-types-section
    Classification: Unclassified
           Product: gcc
           Version: 4.7.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: debug
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: jan.kratochvil@redhat.com
            Target: x86_64-unknown-linux-gnu


PASS: gcc (GCC) 4.7.1 20120504 (prerelease)
FAIL: gcc (GCC) 4.8.0 20120504 (experimental)
(I do not have it here to verify but apparently 4.8.0 20120503 PASSed.)

There are 338 GDB testsuite regressions due to it, one for all:

 p acp->c1
-$9 = (A *) 0x0
-(gdb) PASS: gdb.cp/class2.exp: p acp->c1
+$9 = (struct A *) 0x0
+(gdb) FAIL: gdb.cp/class2.exp: p acp->c1

PASS:
 <2><1ad>: Abbrev Number: 4 (DW_TAG_member)
    <1ae>   DW_AT_name        : c1
    <1b3>   DW_AT_type        : <0x203>
    <1b7>   DW_AT_data_member_location: 24
 <1><203>: Abbrev Number: 19 (DW_TAG_pointer_type)
    <204>   DW_AT_byte_size   : 8
    <205>   DW_AT_type        : signature: 9f2338a31cdd67b0

FAIL:
 <2><1aa>: Abbrev Number: 4 (DW_TAG_member)
    <1ab>   DW_AT_name        : c1
    <1b0>   DW_AT_type        : <0x200>
    <1b4>   DW_AT_data_member_location: 24
 <1><200>: Abbrev Number: 9 (DW_TAG_pointer_type)
    <201>   DW_AT_byte_size   : 8
    <202>   DW_AT_type        : <0x21e>
 <1><21e>: Abbrev Number: 17 (DW_TAG_typedef)
    <21f>   DW_AT_type        : signature: 9f2338a31cdd67b0

<21e> seems to be excessive/redundant.

[Bug debug/53235] [4.8 Regression] 20120504 broke -fdebug-types-section

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

Richard Guenther <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Version|4.7.0                       |4.8.0
   Target Milestone|---                         |4.8.0

[Bug debug/53235] [4.8 Regression] 20120504 broke -fdebug-types-section

[jason at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

--- Comment #1 from Jason Merrill <jason at gcc dot gnu.org> 2012-05-07 14:21:30 UTC ---
<21e> is there as a local stub so that multiple references to the same
.debug_types type can use (smaller) local DIE references instead of all needing
to use DW_FORM_sig8.

The testcase is failing because of the change from "A*" to "struct A*"?  Is
that an important distinction?  If so, could you handle the case of a nameless
typedef specially on your end?

[Bug debug/53235] [4.8 Regression] 20120504 broke -fdebug-types-section

[jan.kratochvil at redhat dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

--- Comment #2 from Jan Kratochvil <jan.kratochvil at redhat dot com> 2012-05-07 14:32:59 UTC ---
(In reply to comment #1)
> The testcase is failing because of the change from "A*" to "struct A*"?

Yes.

> Is that an important distinction?

In other cases it even crashes GDB, originally filed as:
  http://sourceware.org/bugzilla/show_bug.cgi?id=14060
Or I see
  Internal error: non-aggregate type to value_struct_elt_for_reference
  Function "policy1::function" not defined.
  etc.

> If so, could you handle the case of a nameless typedef specially on your end?

Yes; I wanted to know first if this is an intended change.

FYI it does not seem to be DWARF compliant:
  The typedef entry has a DW_AT_name attribute whose value is a null-terminated
  string containing the name of the typedef as it appears in the source
program.

[Bug debug/53235] [4.8 Regression] 20120504 broke -fdebug-types-section

[jason at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

--- Comment #3 from Jason Merrill <jason at gcc dot gnu.org> 2012-05-07 14:47:27 UTC ---
(In reply to comment #2)
> Yes; I wanted to know first if this is an intended change.

It was.

> FYI it does not seem to be DWARF compliant:
>   The typedef entry has a DW_AT_name attribute whose value is a null-terminated
>   string containing the name of the typedef as it appears in the source
> program.

Yes, I agree, but it seemed in informal testing that GDB didn't mind; I
apologize for not running the testsuite.

Would

DW_TAG_structure_type
  DW_AT_declaration
  DW_AT_signature <sig8>

be better?  I suppose strictly speaking we need a DW_AT_name there, too...I'll
see about going in that direction instead.

[Bug debug/53235] [4.8 Regression] 20120504 broke -fdebug-types-section

[jan.kratochvil at redhat dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

--- Comment #4 from Jan Kratochvil <jan.kratochvil at redhat dot com> 2012-05-07 15:00:01 UTC ---
(In reply to comment #3)
> Would
> 
> DW_TAG_structure_type
>   DW_AT_declaration
>   DW_AT_signature <sig8>
> 
> be better?

I also think GDB read_structure_type should cope with it right.


> I suppose strictly speaking we need a DW_AT_name there, too...I'll
> see about going in that direction instead.

As GDB should not mind maybe we could propose DWARF extension
  DW_AT_specification -> DW_AT_specification or DW_AT_signature
in the paragraph:
  Structure, union and class entries containing the DW_AT_specification
  attribute do not need to duplicate information provided by the declaration
  entry referenced by the specification attribute. In particular, such entries
  do not need to contain an attribute for the name of the structure, class or
  union they represent if such information is already provided in the
  declaration.

[Bug debug/53235] [4.8 Regression] 20120504 broke -fdebug-types-section

[jason at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

--- Comment #5 from Jason Merrill <jason at gcc dot gnu.org> 2012-05-13 03:37:42 UTC ---
Author: jason
Date: Sun May 13 03:37:38 2012
New Revision: 187435

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=187435
Log:
    PR debug/53235
    * dwarf2out.c (build_local_stub): Prefer DW_AT_signature for
    comdat types.

Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/dwarf2out.c

[Bug debug/53235] [4.8 Regression] 20120504 broke -fdebug-types-section

[jason at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

Jason Merrill <jason at gcc dot gnu.org> changed:

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

--- Comment #6 from Jason Merrill <jason at gcc dot gnu.org> 2012-05-14 02:12:20 UTC ---
Should be fixed.

[Bug debug/53235] [4.8 Regression] 20120504 broke -fdebug-types-section

[jan.kratochvil at redhat dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

Jan Kratochvil <jan.kratochvil at redhat dot com> changed:

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

--- Comment #7 from Jan Kratochvil <jan.kratochvil at redhat dot com> 2012-05-19 07:59:05 UTC ---
Still getting for -fdebug-types-section regression against 2012-04-25 state:

Running ./gdb.cp/koenig.exp ...
FAIL: gdb.cp/koenig.exp: p foo(eo)
FAIL: gdb.cp/koenig.exp: p foo(eo, eo)
FAIL: gdb.cp/koenig.exp: p foo(eo, eo, 1)
FAIL: gdb.cp/koenig.exp: p foo(fo, eo)
FAIL: gdb.cp/koenig.exp: p foo(1 ,fo, eo)
FAIL: gdb.cp/koenig.exp: p foo(go, fo, eo)
FAIL: gdb.cp/koenig.exp: p foo(io)
FAIL: gdb.cp/koenig.exp: p foo(ix)
FAIL: gdb.cp/koenig.exp: p foo(ko,1)
Running ./gdb.cp/m-static.exp ...
FAIL: gdb.cp/m-static.exp: print svar
+FAIL: gdb.ada/array_bounds.exp: print table'first (GDB internal error)
+FAIL: gdb.ada/array_bounds.exp: print table'last (GDB internal error)
+FAIL: gdb.ada/arrayidx.exp: print e_one_two_three, indexes off (GDB internal
error)
+ERROR: Process no longer exists
+FAIL: gdb.ada/packed_tagged.exp: print x
+FAIL: gdb.ada/packed_tagged.exp: ptype x
+FAIL: gdb.ada/variant_record_packed_array.exp: print empty
+FAIL: gdb.ada/whatis_array_val.exp: print full (GDB internal error)
+FAIL: gdb.ada/whatis_array_val.exp: whatis $

But for example gdb.cp/class2.exp no longer regresses.

[Bug debug/53235] [4.8 Regression] 20120504 broke -fdebug-types-section

[jan.kratochvil at redhat dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

Jan Kratochvil <jan.kratochvil at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dje at google dot com

--- Comment #8 from Jan Kratochvil <jan.kratochvil at redhat dot com> 2012-05-24 03:36:08 UTC ---
-fdebug-types-section regresses static scope of types
http://sourceware.org/bugzilla/show_bug.cgi?id=14148

suggests the stub DW_TAG_struct_type should not have DW_AT_declaration as we
need definition of the type at its proper scope.

[Bug debug/53235] [4.8 Regression] 20120504 broke -fdebug-types-section

[jason at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

--- Comment #9 from Jason Merrill <jason at gcc dot gnu.org> 2012-07-07 14:13:38 UTC ---
(In reply to comment #8)
> -fdebug-types-section regresses static scope of types
> http://sourceware.org/bugzilla/show_bug.cgi?id=14148
> 
> suggests the stub DW_TAG_struct_type should not have DW_AT_declaration as we
> need definition of the type at its proper scope.

We have had declarations at the type's proper scope and definitions at CU scope
for a very long time, and moving the definition to a type unit should work
similarly.  But I'm currently not putting a name on the stub to save space;
would giving it a DW_AT_name make a difference?

[Bug debug/53235] [4.8 Regression] 20120504 broke -fdebug-types-section

[jan.kratochvil at redhat dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

--- Comment #10 from Jan Kratochvil <jan.kratochvil at redhat dot com> 2012-07-17 19:09:59 UTC ---
namespace E {
  class O {};
  void f (O o) {}
}
namespace F {
  class O {};
  void f (O fo) {}
}
E::O eo;
int main () {}
--------------------------------------------------------------------------------
-gdwarf-4 -fdebug-types-section
gdb ./a.out -ex 'ptype eo'
FAIL: gcc (GCC) 4.7.2 20120717 (prerelease)
FAIL: gcc (GCC) 4.8.0 20120717 (experimental)
type = class F::O {
    <no data fields>
}
--------------------------------------------------------------------------------
-gdwarf-4 -fno-debug-types-section
gdb ./a.out -ex 'ptype eo'
PASS: gcc (GCC) 4.7.2 20120717 (prerelease)
PASS: gcc (GCC) 4.8.0 20120717 (experimental)
type = class E::O {
    <no data fields>
}
--------------------------------------------------------------------------------
FAIL:
 <1><d4>: Abbrev Number: 13 (DW_TAG_variable)
    <d5>   DW_AT_name        : eo
    <da>   DW_AT_type        : <0xe8>
 <1><e8>: Abbrev Number: 14 (DW_TAG_class_type)
    <e9>   DW_AT_signature   : signature: 0x2ab8ef7c6fcf3265
Contents of the .debug_types section:
   Signature:     0x2ab8ef7c6fcf3265
 <1><25>: Abbrev Number: 2 (DW_TAG_namespace)
    <26>   DW_AT_name        : F

[Bug debug/53235] [4.8 Regression] 20120504 broke -fdebug-types-section

[jason at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

Jason Merrill <jason at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2012-07-19
         AssignedTo|unassigned at gcc dot       |jason at gcc dot gnu.org
                   |gnu.org                     |
     Ever Confirmed|0                           |1

[Bug debug/53235] [4.8 Regression] 20120504 broke -fdebug-types-section

[jason at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

--- Comment #11 from Jason Merrill <jason at gcc dot gnu.org> 2012-07-19 20:02:02 UTC ---
Author: jason
Date: Thu Jul 19 20:01:56 2012
New Revision: 189676

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=189676
Log:
    PR debug/53235
    * dwarf2out.c (generate_type_signature): Handle the case of DIE
    being nested, rather than its declaration..

Added:
    trunk/gcc/testsuite/g++.dg/debug/dwarf2/nested-4.C
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/dwarf2out.c

[Bug debug/53235] [4.8 Regression] 20120504 broke -fdebug-types-section

[jason at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

Jason Merrill <jason at gcc dot gnu.org> changed:

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

--- Comment #12 from Jason Merrill <jason at gcc dot gnu.org> 2012-07-19 20:09:35 UTC ---
Thanks for the testcase.  Fixed.

[Bug debug/53235] [4.8 Regression] 20120504 broke -fdebug-types-section

[ccoutant at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

--- Comment #13 from Cary Coutant <ccoutant at gcc dot gnu.org> 2012-09-06 03:34:27 UTC ---
Author: ccoutant
Date: Thu Sep  6 03:34:22 2012
New Revision: 191005

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=191005
Log:
2012-09-05  Cary Coutant  <ccoutant@gmail.com>

    Backport trunk patch to fix a problem where type signature does
    not include the type's context.

    2012-07-19  Jason Merrill  <jason@redhat.com>

    gcc/
        PR debug/53235
        * dwarf2out.c (get_die_parent): New.
        (generate_type_signature): Use it.

    gcc/testsuite/
        * g++.dg/debug/dwarf2/nested-4.C: New.

Added:
    branches/google/gcc-4_7/gcc/testsuite/g++.dg/debug/dwarf2/nested-4.C
Modified:
    branches/google/gcc-4_7/gcc/ChangeLog.google-4_7
    branches/google/gcc-4_7/gcc/dwarf2out.c
    branches/google/gcc-4_7/gcc/testsuite/ChangeLog.google-4_7

[Bug debug/53235] [4.8 Regression] 20120504 broke -fdebug-types-section

[mrs at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53235

--- Comment #14 from mrs at gcc dot gnu.org <mrs at gcc dot gnu.org> 2013-01-20 20:35:56 UTC ---
Author: mrs
Date: Sun Jan 20 20:35:48 2013
New Revision: 195326

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=195326
Log:
2013-01-20  Jack Howarth <howarth@bromo.med.uc.edu>

        PR debug/53235
        * g++.dg/debug/dwarf2/nested-4.C: XFAIL on darwin.

Modified:
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/testsuite/g++.dg/debug/dwarf2/nested-4.C