[Bug c++/7385] g++ problems with enum and __attribute__

[Bug c++/7385] g++ problems with enum and __attribute__

[pinskia at physics dot uc dot edu]

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

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=7385


pinskia at physics dot uc dot edu changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |pinskia at physics dot uc
                   |                            |dot edu
   Last reconfirmed|2003-07-14 04:06:26         |2003-07-17 02:53:28
               date|                            |


------- Additional Comments From pinskia at physics dot uc dot edu  2003-07-17 02:53 -------
Mark are you still working on a fix?

[Bug c++/7385] g++ problems with enum and __attribute__

[mark at codesourcery dot com]

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

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=7385



------- Additional Comments From mark at codesourcery dot com  2003-07-17 07:02 -------
Subject: Re:  g++ problems with enum and __attribute__

On Wed, 2003-07-16 at 19:53, pinskia at physics dot uc dot edu wrote:
> PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.
> 
> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=7385
> 
> 
> pinskia at physics dot uc dot edu changed:
> 
>            What    |Removed                     |Added
> ----------------------------------------------------------------------------
>                  CC|                            |pinskia at physics dot uc
>                    |                            |dot edu
>    Last reconfirmed|2003-07-14 04:06:26         |2003-07-17 02:53:28
>                date|                            |
> 
> 
> ------- Additional Comments From pinskia at physics dot uc dot edu  2003-07-17 02:53 -------
> Mark are you still working on a fix?

No, I gave up.

It is probably relatively more tractable in the new paser, but even
there it is probably not easy.  I'm not sure what the new attribute
syntax document says about this case.

[Bug c++/7385] g++ problems with enum and __attribute__

[pinskia at physics dot uc dot edu]

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

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=7385


pinskia at physics dot uc dot edu changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|mmitchel at gcc dot gnu dot |unassigned at gcc dot gnu
                   |org                         |dot org
             Status|ASSIGNED                    |NEW


------- Additional Comments From pinskia at physics dot uc dot edu  2003-07-17 12:48 -------
Mark is no longer working on it.

[Bug c++/7385] g++ problems with enum and __attribute__

[pinskia at physics dot uc dot edu]

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

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=7385



------- Additional Comments From pinskia at physics dot uc dot edu  2003-07-22 18:29 -------
The mainline (20030722) now  gives:
pr7385.cc:3: warning: `packed' attribute ignored
pr7385.cc:4: error: non-local function `void g1(<anonymous enum>)' uses 
    anonymous type
pr7385.cc:6: warning: `packed' attribute ignored
pr7385.cc:7: error: non-local function `void g2(<anonymous enum>)' uses 
    anonymous type

[Bug c++/7385] [3.3/3.4 Regression] g++ problems with enum and __attribute__

[pinskia at physics dot uc dot edu]

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

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=7385


pinskia at physics dot uc dot edu changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|normal                      |critical
           Priority|P3                          |P1
   Last reconfirmed|2003-07-27 06:58:45         |2003-07-29 18:00:01
               date|                            |
            Summary|g++ problems with enum and  |[3.3/3.4 Regression] g++
                   |__attribute__               |problems with enum and
                   |                            |__attribute__
   Target Milestone|---                         |3.3.2


------- Additional Comments From pinskia at physics dot uc dot edu  2003-07-29 18:00 -------
This is actually a regression from 3.0.4 so some patch made this fail.
>From Phil's regression hunter: Search converges between 2002-04-28-trunk (#75) and 
2002-05-05-trunk (#76). 

The only reason I say this was because Apple's 3.1 based compiler is based on 3.1 
(20020420) which does not have this regression in it.

[Bug c++/7385] [3.3/3.4 Regression] g++ problems with enum and __attribute__

[janis187 at us dot ibm dot com]

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

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=7385


janis187 at us dot ibm dot com changed:

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


------- Additional Comments From janis187 at us dot ibm dot com  2003-08-05 20:25 -------
The regression in PR 7385 was introduced or exposed by this patch:

--- gcc/gcc/cp/ChangeLog ---

2002-04-30  Jason Merrill  <jason@redhat.com>

        * decl.c (grokdeclarator): Don't override TYPE_NAME of an
        anonymous class with a typedef if there are attributes.

The regression hunt took place on i686-pc-linux-gnu using the submitter's
test case.   Like many other recent hunts, it started with the dates
identified by Andrew Pinski using Phil's regression hunter.

[Bug c++/7385] [3.3/3.4 Regression] g++ problems with enum and __attribute__

[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=7385



------- Additional Comments From pinskia at gcc dot gnu dot org  2003-08-09 02:02 -------
The error message has changed again:
pr7385.cc:3: warning: `packed' attribute ignored
pr7385.cc:4: error: non-local function `void g1(e1)' uses anonymous type
pr7385.cc:3: error: `typedef enum <anonymous> e1' does not refer to the 
   unqualified type, so it is not used for linkage
pr7385.cc:6: warning: `packed' attribute ignored
pr7385.cc:7: error: non-local function `void g2(e2)' uses anonymous type
pr7385.cc:6: error: `typedef enum <anonymous> e2' does not refer to the 
   unqualified type, so it is not used for linkage

[Bug c++/7385] [3.3/3.4 Regression] g++ problems with enum and __attribute__

[jason 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=7385


jason at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
   Last reconfirmed|2003-08-24 17:25:54         |2003-08-29 20:40:31
               date|                            |

[Bug c++/7385] [3.3/3.4 Regression] g++ problems with enum and __attribute__

[jason 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=7385


jason at gcc dot gnu dot org changed:

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


------- Additional Comments From jason at gcc dot gnu dot org  2003-08-29 20:48 -------
This observed behavior was the intent of the patch.  The gcc trunk gives this
additional error for your testcase:

error: `typedef enum <anonymous> e1' does not refer to the unqualified 
   type, so it is not used for linkage

You want to write

  enum e1 { v11=1, v12=2 } __attribute__ ((packed));

instead, so that the attribute applies to the enum rather than the typedef.  If
you want to use "e1" without "enum" in C, you can then write

  typedef enum e1 e1;