[Bug regression/57551] New: [4.9 Regression]: g++.dg/ext/visibility/anon6.C scan-assembler 1BIiE1cE

[Bug regression/57551] New: [4.9 Regression]: g++.dg/ext/visibility/anon6.C scan-assembler 1BIiE1cE

[hp at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57551

            Bug ID: 57551
           Summary: [4.9 Regression]: g++.dg/ext/visibility/anon6.C
                    scan-assembler 1BIiE1cE
           Product: gcc
           Version: unknown
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: regression
          Assignee: unassigned at gcc dot gnu.org
          Reporter: hp at gcc dot gnu.org
                CC: hubicka at ucw dot cz

With revision 199692 this test passed.
>From revision 199702 and on, this test has failed as follows:

Running /tmp/hpautotest-gcc0/gcc/gcc/testsuite/g++.dg/dg.exp ...
...
FAIL: g++.dg/ext/visibility/anon6.C -std=c++98  scan-assembler 1BIiE1cE
FAIL: g++.dg/ext/visibility/anon6.C -std=c++11  scan-assembler 1BIiE1cE

With the message in the logfile being:

Executing on host:
/tmp/hpautotest-gcc0/cris-elf/gccobj/gcc/testsuite/g++/../../xg++
-B/tmp/hpautotest-gcc0/cris-elf/gccobj/gcc/testsuite/g++/../../
/tmp/hpautotest-gcc0/gcc/gcc/testsuite/g++.dg/ext/visibility/anon6.C 
-fno-diagnostics-show-caret -fdiagnostics-color=never  -nostdinc++
-I/tmp/hpautotest-gcc0/cris-elf/gccobj/cris-elf/libstdc++-v3/include/cris-elf
-I/tmp/hpautotest-gcc0/cris-elf/gccobj/cris-elf/libstdc++-v3/include
-I/tmp/hpautotest-gcc0/gcc/libstdc++-v3/libsupc++
-I/tmp/hpautotest-gcc0/gcc/libstdc++-v3/include/backward
-I/tmp/hpautotest-gcc0/gcc/libstdc++-v3/testsuite/util -fmessage-length=0
-std=c++98 -pedantic-errors -Wno-long-long -ffat-lto-objects -ffat-lto-objects
-ffat-lto-objects  -S   -isystem
/tmp/hpautotest-gcc0/cris-elf/gccobj/cris-elf/./newlib/targ-include -isystem
/tmp/hpautotest-gcc0/gcc/newlib/libc/include  -o anon6.s    (timeout = 300)
PASS: g++.dg/ext/visibility/anon6.C -std=c++98 (test for excess errors)
FAIL: g++.dg/ext/visibility/anon6.C -std=c++98  scan-assembler 1BIiE1cE
PASS: g++.dg/ext/visibility/anon6.C -std=c++98  scan-assembler-not
globl.*1BIiE1cE
PASS: g++.dg/ext/visibility/anon6.C -std=c++98  scan-assembler-not 1CIiE1cE

(similarly for -std=c++11)

The generated assembly file is empty.
Messages on gcc-patches indicate that the failure is universal; at a glance it
seems the suspect commit 199695 made the whole code go away as unused and
unreachable.  Tweak or remove test-case?

Author of suspect patch in revision range CC:ed.

[Bug regression/57551] [4.9 Regression]: g++.dg/ext/visibility/anon6.C scan-assembler 1BIiE1cE

[hubicka at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57551

Jan Hubicka <hubicka at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2013-06-07
                 CC|                            |hubicka at gcc dot gnu.org
           Assignee|unassigned at gcc dot gnu.org      |hubicka at gcc dot gnu.org
     Ever confirmed|0                           |1

--- Comment #1 from Jan Hubicka <hubicka at gcc dot gnu.org> ---
mine.

[Bug regression/57551] [4.9 Regression]: g++.dg/ext/visibility/anon6.C scan-assembler 1BIiE1cE

[hubicka at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57551

Jan Hubicka <hubicka at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jason at redhat dot com

--- Comment #2 from Jan Hubicka <hubicka at gcc dot gnu.org> ---
You are right, I added attribute used to the variable and forgot about it. The
testcase is bogus by expecting unused local static to not be optmized away.

Looking closer, I am however confused. While the attribute fixes the testcase,
i think something wrong is still going on hre.

We get the following symbol out of FE:
_ZN12_GLOBAL__N_11BIiE1cE/1 (const int {anonymous}::B<int>::c) @0x7ffff75141a0
  Type: variable definition analyzed
  Visibility: force_output comdat
  Aux: @0x1  References:
  Referring:
  Availability: not-ready
  Varpool flags: initialized read-only const-value-known

i.e. it is COMDAT but not public. It is the force_output flag that keeps
variable alive and that is there because of keying. The variable is output as
static object:
jh@gcc10:~/trunk/build3/gcc$ more anon6.s
        .file   "anon6.C"
        .section        .rodata
        .align 4
        .type   _ZN12_GLOBAL__N_11BIiE1cE, @object
        .size   _ZN12_GLOBAL__N_11BIiE1cE, 4
_ZN12_GLOBAL__N_11BIiE1cE:
        .zero   4
        .ident  "GCC: (GNU) 4.9.0 20130602 (experimental)"
        .section        .note.GNU-stack,"",@progbits

i.e. no comdat visibility. This flag stays through the compilation and seems to
be ignored by output routines.  

And now adding the attribute used:

Index: /home/jh/trunk/gcc/testsuite/g++.dg/ext/visibility/anon6.C
===================================================================
--- /home/jh/trunk/gcc/testsuite/g++.dg/ext/visibility/anon6.C  (revision
199698)
+++ /home/jh/trunk/gcc/testsuite/g++.dg/ext/visibility/anon6.C  (working copy)
@@ -18,7 +18,7 @@ namespace
   template <typename T>
   class B
   {
-    static const T c = 0;
+    __attribute__ ((__used__)) static const T c = 0;
   };

   template <typename T> const T B<T>::c;

changes object to:
_ZN12_GLOBAL__N_11BIiE1cE/1 (const int {anonymous}::B<int>::c) @0x7ffff75141a0
  Type: variable definition analyzed
  Visibility: force_output forced_by_abi public weak comdat
comdat_group:_ZN12_GLOBAL__N_11BIiE1cE one_only
  Aux: @0x7ffff7402850  References:
  Referring:
  Availability: not-ready
  Varpool flags: initialized read-only const-value-known

which changes assembly to:
jh@gcc10:~/trunk/build/gcc$ ./xgcc -B ./ -O2
/home/jh/trunk/gcc/testsuite/g++.dg/ext/visibility/anon6.C -S -std=c++98
-pedantic-errors -Wno-long-long -S -fdump-ipa-all-details -fdump-ipa-cgraph
jh@gcc10:~/trunk/build/gcc$ more anon6.s 
        .file   "anon6.C"
        .weak   _ZN12_GLOBAL__N_11BIiE1cE
        .section       
.rodata._ZN12_GLOBAL__N_11BIiE1cE,"aG",@progbits,_ZN12_GLOBAL__N_11BIiE1cE,comdat
        .align 4
        .type   _ZN12_GLOBAL__N_11BIiE1cE, @gnu_unique_object
        .size   _ZN12_GLOBAL__N_11BIiE1cE, 4
_ZN12_GLOBAL__N_11BIiE1cE:
        .zero   4
        .ident  "GCC: (GNU) 4.9.0 20130605 (experimental)"
        .section        .note.GNU-stack,"",@progbits

Here we now disagree with GCC 4.5 that still outputs this as static.

Jason, why C++ FE behaves this way now?

clang also optimizes out the variable in testcase and adding attribute used
makes the var to be static.

I think the testcase should be fixed by adding the used flag and by checking
that weaks/comdats are not involved in the assembly and C++ FE should be fixed.
or do I miss something?

Honza

[Bug regression/57551] [4.9 Regression]: g++.dg/ext/visibility/anon6.C scan-assembler 1BIiE1cE

[jason at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57551

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jason at gcc dot gnu.org
           Assignee|hubicka at gcc dot gnu.org         |jason at gcc dot gnu.org
   Target Milestone|---                         |4.9.0

--- Comment #3 from Jason Merrill <jason at gcc dot gnu.org> ---
Sounds right to me.

[Bug regression/57551] [4.9 Regression]: g++.dg/ext/visibility/anon6.C scan-assembler 1BIiE1cE

[hubicka at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57551

--- Comment #4 from Jan Hubicka <hubicka at gcc dot gnu.org> ---
OK, C++ FE properly brings the decl local as part of constrain_visibility,
however later in pt.c it is made public again in 
mark_decl_instantiated as explicit instantiation.

My wild try is the following:
Index: pt.c
===================================================================
--- pt.c        (revision 199698)
+++ pt.c        (working copy)
@@ -17402,6 +17402,13 @@ mark_decl_instantiated (tree result, int
   if (TREE_ASM_WRITTEN (result))
     return;

+  /* For anonymous namespace we don't need to do anything.  */
+  if (decl_anon_ns_mem_p (result))
+    {
+      gcc_assert (!TREE_PUBLIC (result));
+      return;
+    }
+
   if (TREE_CODE (result) != FUNCTION_DECL)
     /* The TREE_PUBLIC flag for function declarations will have been
        set correctly by tsubst.  */

[Bug regression/57551] [4.9 Regression]: g++.dg/ext/visibility/anon6.C scan-assembler 1BIiE1cE

[hubicka at ucw dot cz]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57551

--- Comment #6 from Jan Hubicka <hubicka at ucw dot cz> ---
Good to be assigned.  I posted the patch 
Jason, can you please look at
http://gcc.gnu.org/ml/gcc-patches/2013-06/msg00433.html

Honza

[Bug regression/57551] [4.9 Regression]: g++.dg/ext/visibility/anon6.C scan-assembler 1BIiE1cE

[jason at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57551

--- Comment #7 from Jason Merrill <jason at gcc dot gnu.org> ---
(In reply to Jan Hubicka from comment #6)
> Jason, can you please look at
> http://gcc.gnu.org/ml/gcc-patches/2013-06/msg00433.html

I approved it the same day you posted it. :)

[Bug regression/57551] [4.9 Regression]: g++.dg/ext/visibility/anon6.C scan-assembler 1BIiE1cE

[hp at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57551

Hans-Peter Nilsson <hp at gcc dot gnu.org> changed:

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

--- Comment #8 from Hans-Peter Nilsson <hp at gcc dot gnu.org> ---
Intended fix committed as noticed by my autotester, so I'm closing this.

Looks like the svn commit <-> bugzilla interface is broken or severely delayed;
the log seems ok and should match, AFAICT:

Log:
        PR c++/57551
        * g++.dg/ext/visibility/anon6.C: Update testcase.
        * cp/pt.c (mark_decl_instantiated): Do not export explicit
instantiations
        of anonymous namespace templates.

[Bug regression/57551] [4.9 Regression]: g++.dg/ext/visibility/anon6.C scan-assembler 1BIiE1cE

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57551

--- Comment #9 from Paolo Carlini <paolo.carlini at oracle dot com> ---
It's broken since the Bugzilla update, unfortunately.

[Bug regression/57551] [4.9 Regression]: g++.dg/ext/visibility/anon6.C scan-assembler 1BIiE1cE

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57551

--- Comment #10 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Paolo Carlini from comment #9)
> It's broken since the Bugzilla update, unfortunately.

No broken since updating the server to a new machine.  Updating Bugzilla was
not the issue but updating the subversion on the server was.

[Bug regression/57551] [4.9 Regression]: g++.dg/ext/visibility/anon6.C scan-assembler 1BIiE1cE

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57551

--- Comment #11 from Paolo Carlini <paolo.carlini at oracle dot com> ---
Right sorry, the problem is the new subversion not sending commit messages at
all to Bugzilla.

[Bug regression/57551] [4.9 Regression]: g++.dg/ext/visibility/anon6.C scan-assembler 1BIiE1cE

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57551

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |fche at redhat dot com,
                   |                            |jakub at gcc dot gnu.org

--- Comment #12 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
In particular, the problem is that we used to have a hacked up version of
svnmailer, which created the special emails for bugzilla, now the server has a
new upstream version of svnmailer and these hacks weren't ported to it.