This is the mail archive of the
mailing list for the GCC project.
Re: Patch to add __builtin_printf -> fputs transformations
- To: bkorb at cruzio dot com, jsm28 at cam dot ac dot uk
- Subject: Re: Patch to add __builtin_printf -> fputs transformations
- From: "Kaveh R. Ghazi" <ghazi at caip dot rutgers dot edu>
- Date: Tue, 19 Sep 2000 22:48:46 -0400 (EDT)
- Cc: gcc-patches at gcc dot gnu dot org
> From: Bruce Korb <firstname.lastname@example.org>
> "Joseph S. Myers" wrote:
> > In general, when wrapping <stdio.h> you need to be very careful about not
> > interfering with glibc's __need_FILE. That is, any wrapper applied to
> > glibc's <stdio.h> must _not_ have multiple-include protection, since if it
> > does, one header includes <stdio.h> with __need_FILE defined and then if
> > the program has a later direct include of <stdio.h> that doesn't work if a
> > wrapper prevents reinclusion.
> Well, then never mind. If glibc has forced stdio.h to not be wrapable,
> then you cannot do that for glibc's stdio.h. In fact, it is probably
> necessary to verify that fixincludes cannot wrap the thing. Yuck.
BTW, stdio-opt.h is itself self-exclusionary, so at worst you'll have
an extra nanosecond of cpp time, but cc1 won't see the code twice.
Kaveh R. Ghazi Engagement Manager / Project Services
email@example.com Qwest Internet Solutions