[Bug c++/13160] New: ABI=1 implementation difference vs. g++ 3.2

[Bug c++/13160] New: ABI=1 implementation difference vs. g++ 3.2

[grigory at stl dot sarov dot ru]

For the simple class c1 listed below g++ 3.4 with default ABI (-fabi-version=1)
and g++ 3.2 give different size and alignment values. This is signifincant ABI
compatibility issue since -fabi-version=1 is stated as 3.2-compatible
implementation.

enum e0 { };
union c1 {
  enum e0 : 20;
};


[g++ 3.4 20031112]
Class c1
   size=3 align=1
   base size=3 base align=1
c1 (0x403cc8c0) 0


[g++ (GCC) 3.2 20020903 (Red Hat Linux 8.0 3.2-7)]
Class c1
   size=4 align=4
c1 (0x4031c3c0) 0


Note: the same difference exists in GCC 3.3.x series.

-- 
           Summary: ABI=1 implementation difference vs. g++ 3.2
           Product: gcc
           Version: 3.4
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: c++
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: grigory at stl dot sarov dot ru
                CC: gcc-bugs at gcc dot gnu dot org
 GCC build triplet: i686-redhat-linux
  GCC host triplet: i686-redhat-linux
GCC target triplet: i686-redhat-linux


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13160

[Bug c++/13160] ABI=1 implementation difference vs. g++ 3.2

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|normal                      |critical
           Keywords|                            |ABI
            Summary|ABI=1 implementation        |ABI=1 implementation
                   |difference vs. g++ 3.2      |difference vs. g++ 3.2
   Target Milestone|---                         |3.4.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13160

[Bug c++/13160] ABI=1 implementation difference vs. g++ 3.2

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |wrong-code


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13160

[Bug c++/13160] ABI=1 implementation difference vs. g++ 3.2

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|wrong-code                  |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13160

[Bug c++/13160] [3.3/3.4 Regression] ABI=1 implementation difference vs. g++ 3.2

[giovannibajo at libero dot it]

------- Additional Comments From giovannibajo at libero dot it  2004-01-11 23:53 -------
Confirmed on both 3.3 and 3.4. Critical regression I'd say.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |giovannibajo at libero dot
                   |                            |it
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
   Last reconfirmed|0000-00-00 00:00:00         |2004-01-11 23:53:36
               date|                            |
            Summary|ABI=1 implementation        |[3.3/3.4 Regression] ABI=1
                   |difference vs. g++ 3.2      |implementation difference
                   |                            |vs. g++ 3.2


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13160

[Bug c++/13160] [3.3/3.4 Regression] ABI=1 implementation difference vs. g++ 3.2

[giovannibajo at libero dot it]

------- Additional Comments From giovannibajo at libero dot it  2004-01-11 23:57 -------
Nathan, can you look into this regression? It seems pretty critical.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |nathan at gcc dot gnu dot
                   |                            |org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13160

[Bug c++/13160] [3.3/3.4 Regression] ABI=1 implementation difference vs. g++ 3.2

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-01-15 07:49 -------
This changed between 3.3 20020928 and 3.3 20021001 (this was done using Phil's 
regression hunter and using the testcase below and having a function which returns the 
sizeof of the union and looking at the asm).
Almost I think it is caused by this patch: <http://gcc.gnu.org/ml/gcc-patches/2002-09/
msg01814.html>, it is the only ABI patch that went in during that time.
Mark could you look at this because it might be caused by you patch

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mark at codesourcery dot com


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13160

[Bug c++/13160] [3.3/3.4/3.5 Regression] ABI=1 implementation difference vs. g++ 3.2

[mmitchel at gcc dot gnu dot org]

------- Additional Comments From mmitchel at gcc dot gnu dot org  2004-02-05 05:44 -------
This is not a C++ ABI change; it is a *C* ABI change, and an intentional one.

See:

  http://gcc.gnu.org/ml/gcc/2002-09/msg01109.html

and the follow-up thread for the rationale.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13160