[Bug c++/12574] New: another unit-at-a-time bug?

[Bug c++/12574] New: another unit-at-a-time bug?

[gcc-bugzilla at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12574

           Summary: another unit-at-a-time bug?
           Product: gcc
           Version: 3.4
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: snyder at fnal dot gov
                CC: gcc-bugs at gcc dot gnu dot org
 GCC build triplet: i686-pc-linux-gnu
  GCC host triplet: i686-pc-linux-gnu
GCC target triplet: i686-pc-linux-gnu


Here's another case where gcc 3.4 fails to emit a required function definition.
>From a quick look, it looks like all the existing unit-at-a-time PRs have
been closed, so i hope this isn't a duplicate.

When the source below is compiled at -O2 or above, there's a reference
to the function `A::foo', but the definition isn't emitted:

$ cc1plus -quiet -O2 x.cc
$ grep foo x.s
        movl    $_ZN1A3fooEv, (%eax)
$

Environment:
System: Linux karma 2.4.19-emp_2419p5a829i #1 Tue Sep 3 17:42:17 EST 2002 i686 i686 i386 GNU/Linux
Architecture: i686

	
host: i686-pc-linux-gnu
build: i686-pc-linux-gnu
target: i686-pc-linux-gnu
configured with: ../gcc/configure --prefix=/usr/local/gcc --enable-threads=posix --enable-long-long --enable-languages=c,c++,f77

How-To-Repeat:

Compile this with -O2 or above:

------------------------------------------------------------------------
struct A
{
  static int foo() { return 0; }
};


template <class TYPE>
class C
{
public:
  C ();
  int (*f_)(void);
};


template <class TYPE>
C<TYPE>::C ()
  : f_ (A::foo)
{
}


void open ()
{
  C<int> th;
}
------------------------------------------------------------------------

[Bug c++/12574] [3.4 Regression] [unti-at-a-time] not emitting function referenced from a weak symbol

[pinskia at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12574


pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|normal                      |critical
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
           Keywords|                            |wrong-code
   Last reconfirmed|0000-00-00 00:00:00         |2003-10-11 02:28:16
               date|                            |
            Summary|another unit-at-a-time bug? |[3.4 Regression] [unti-at-a-
                   |                            |time] not emitting function
                   |                            |referenced from a weak
                   |                            |symbol
   Target Milestone|---                         |3.4


------- Additional Comments From pinskia at gcc dot gnu dot org  2003-10-11 02:28 -------
I can confirm this on the mainline (20031010).

[Bug c++/12574] [3.4 Regression] [unti-at-a-time] not emitting function referenced from a weak symbol

[pinskia at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12574



------- Additional Comments From pinskia at gcc dot gnu dot org  2003-10-11 02:29 -------
Most likely related to bug 12324.

[Bug c++/12574] [3.4 Regression] [unit-at-a-time] not emitting function referenced from a weak symbol

[pinskia at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12574



------- Additional Comments From pinskia at gcc dot gnu dot org  2003-10-11 17:56 -------
Might be fixed by (not commited yet):
Sat Oct 11 15:39:11 CEST 2003  Jan Hubicka  <jh@suse.cz>
	* cgraphunit.c (decide_is_function_needed):  Fix test dealing
	with functions implicitly made inline.

	* decl.c (start_cleanup_fn):  Set DECL_DECLARED_INLINE_P to deffer
	the expansion.

[Bug c++/12574] [3.4 Regression] [unit-at-a-time] not emitting function referenced from a weak symbol

[pinskia at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12574


pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |jh at suse dot cz
                   |dot org                     |
             Status|NEW                         |ASSIGNED


------- Additional Comments From pinskia at gcc dot gnu dot org  2003-10-14 05:16 -------
Jan I noticed your post at <http://gcc.gnu.org/ml/gcc-patches/2003-10/msg00907.html> 
that should fix this bug but you said you will be installing this but you have not yet, why?

[Bug c++/12574] [3.4 Regression] [unit-at-a-time] not emitting function referenced from a weak symbol

[jh at suse dot cz]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12574



------- Additional Comments From jh at suse dot cz  2003-10-14 08:16 -------
Subject: Re:  [3.4 Regression] [unit-at-a-time] not emitting function referenced from a weak symbol

> PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.
> 
> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12574
> 
> 
> pinskia at gcc dot gnu dot org changed:
> 
>            What    |Removed                     |Added
> ----------------------------------------------------------------------------
>          AssignedTo|unassigned at gcc dot gnu   |jh at suse dot cz
>                    |dot org                     |
>              Status|NEW                         |ASSIGNED
> 
> 
> ------- Additional Comments From pinskia at gcc dot gnu dot org  2003-10-14 05:16 -------
> Jan I noticed your post at <http://gcc.gnu.org/ml/gcc-patches/2003-10/msg00907.html> 
> that should fix this bug but you said you will be installing this but you have not yet, why?

No, this is yet another bug of missing MARK_USED somewhere in C++
frontend, unfortunately.  Will this ever take end?

Thanks for pointing this out.  Please if you notice something related to
unit-at-a-time, simply assign the bug to me.  For some reason I missed
this bug even when I search the database for unit-at-a-time related bugs
from time to time.

Honza
> 
> 
> 
> ------- You are receiving this mail because: -------
> You are the assignee for the bug, or are watching the assignee.

[Bug c++/12574] [3.4 Regression] [unit-at-a-time] not emitting function referenced from a weak symbol

[pinskia at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12574


pinskia at gcc dot gnu dot org changed:

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


------- Additional Comments From pinskia at gcc dot gnu dot org  2003-10-15 23:14 -------
Fixed by:
+2003-10-15  Jan Hubicka  <jh@suse.cz>
+
+	pr c++/12574
+        * decl2.c (cxx_callgraph_analyze_expr): deal with baselink.
+