This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: AIX C++ fixinc
- From: Bruce Korb <bkorb at veritas dot com>
- To: David Edelsohn <dje at watson dot ibm dot com>
- Cc: Bruce Korb <bkorb at gnu dot org>, gcc-patches at gcc dot gnu dot org
- Date: Mon, 30 Dec 2002 08:42:03 -0800
- Subject: Re: AIX C++ fixinc
- References: <200212300313.WAA27242@makai.watson.ibm.com>
- Reply-to: bkorb at veritas dot com
David Edelsohn wrote:
>
> The following patch fixes some C++-unfriendly elements of AIX's
> system header files. The extern "C" wrapping allows AIX to bootstrap
> while the new C++ parser is fixed.
>
> Is this okay to commit along with the regenrated fixincl.x?
aix_class looks fine, as long as you are reasonably sure the
named files won't show up on any other platform. ("ODM" may
be unique to AIX, but is "sys/gswio.h"?) Otherwise,
you might consider a straight-out substitution on the "class"
field names:
select = "([ *])(class[;[])";
c_fix = format;
c_fix_arg = "%1____%2";
I don't see any issues with aix_extern_c. It will ensure that every
assert.h on every platform is wrapped with the ``extern "C"'' stuff.
The fix is bypassed on Solaris, Linux and HP/UX. However, please
cd $top_builddir/gcc/fixinc
make check
and verify that the highlighted "failures" are what you would
expect, and then update gcc/fixinc/test/base in the obvious ways, too.
Summary: approved with your judgement call on aix_class & please
update the baseline test files, too. They will be "obvious".