This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: boolstrap failure: /usr/include/stdlib.h:45: redefinition of `rune_t'


 > From: "Kaveh R. Ghazi" <ghazi@caip.rutgers.edu>
 > 
 >  > From: Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
 >  > 
 >  > On Thu, 3 Feb 2000, Gerald Pfeifer wrote:
 >  > > Kaveh, I am afraid your change
 >  > > 
 >  > >   2000-02-01  Kaveh R. Ghazi  <ghazi@caip.rutgers.edu>
 >  > > 
 >  > >         * tsystem.h: New file.
 >  > > 
 >  > >         * Makefile.in (libgcc2.a, stmp-multilib, crtbegin.o, crtend.o,
 >  > >         s-crtS): Depend on tsystem.h.
 >  > > 
 >  > >         * crtstuff.c: Include tsystem.h.
 >  > >         * frame.c: Likewise.
 >  > >         * libgcc2.c: Likewise.
 >  > > 
 >  > > broke bootstrap on FreeBSD 3.3:
 >  > 
 >  > FYI, this is fixed by
 >  >   http://gcc.gnu.org/ml/gcc-patches/2000-02/msg00098.html
 >  > which has not been reviewed/installed yet.
 >  > 
 >  > Gerald
 > 
 > 	I think the bug is definitely in the freebsd headers.  I don't
 > have access to freebsd to test that patch or help fix it though.
 > 
 > 		--Kaveh


Hmm, you say this error is triggered by including stddef.h & stdlib.h,
I'm pretty sure we were doing this prior to my tsystem.h patch.

I think you mentioned that freebsd treats it as a warning if it knows
they are system headers?  If so, then perhaps the problem is that gcc
is not properly determining that gcc/include is a system header dir.
Maybe some recent change (like the cccp->cpplib cutover) caused it?

(I still think the freebsd headers should be fixed though.)

		--Kaveh
--
Kaveh R. Ghazi			Engagement Manager / Project Services
ghazi@caip.rutgers.edu		Qwest Internet Solutions

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]