This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: egcs-19980425, system.h stuff part 3/6
- To: "Kaveh R. Ghazi" <ghazi at caip dot rutgers dot edu>
- Subject: Re: egcs-19980425, system.h stuff part 3/6
- From: Jeffrey A Law <law at cygnus dot com>
- Date: Tue, 05 May 1998 01:23:50 -0600
- cc: jfc at mit dot edu, egcs at cygnus dot com
- Reply-To: law at cygnus dot com
In message <199805041945.PAA09982@caip.rutgers.edu>you write:
> > From: John Carr <jfc@mit.edu>
> >
> > > * system.h: Wrap time.h and sys/file.h in autoconf checks.
> > > Provide default definitions for O_RDONLY and O_WRONLY here.
> >
> > What build systems supported by gcc need definitions of O_RDONLY and
> > O_WRONLY?
>
> None that I have access to or am aware of. However I suppose
> they must exist somewhere else gcc wouldn't have had support for it
> scattered amongst several files.
I see a half-dozen or so occurences of the O_RDONLY stuff throughout
the sources. O_WRONLY is less common.
I'd say let's go ahead and keep the code to provide default definitions
of O_RDONLY/O_WRONLY if none exist -- keeping that code can't hurt.
And burying it in system.h seems like the right place to work around
such lossage.
jeff