This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: Today's source cannot build on NetBSD
- To: thorpej at zembu dot com
- Subject: Re: Today's source cannot build on NetBSD
- From: Richard Earnshaw <rearnsha at arm dot com>
- Date: Wed, 06 Dec 2000 09:14:52 +0000
- Cc: Krister Walfridsson <cato at df dot lth dot se>, "Kaveh R. Ghazi" <ghazi at caip dot rutgers dot edu>, gcc-bugs at gcc dot gnu dot org
- Cc: rearnsha at arm dot com
- Organization: ARM Ltd.
- Reply-To: rearnsha at arm dot com
> It basically does:
>
> #if __GNUC_PREREQ__(2, 96)
> #define _BSD_VA_LIST_ __builtin_va_list /* GCC built-in type */
> #else
> #define _BSD_VA_LIST_ char * /* va_list */
> #endif
Which is fine, and probably the correct thing for the master copy in
/usr/src/sys/arch/xxx/include/ansi.h.
> Anyway, if you do end up hacking fixincludes to fix this, make sure the
> rule can notice when it's been fixed already.
It won't matter. When/if gcc is integrated into the netbsd sources, you
won't run fixincludes there because you can fix your header files. When
gcc is used as a stand-alone package any files it needs to be fixed will
be squirrelled away in its own private install tree. Other compilers will
never see the changes. So if fixincludes fails to spot the above
workaround (and I don't think it does), it won't matter anyway, since it
will never be see by anything that doesn't require the fix anyway.
R.