This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Improve vxWorks support on x86
- To: Hans-Peter Nilsson <hans-peter dot nilsson at axis dot com>
- Subject: Re: Improve vxWorks support on x86
- From: Jeffrey A Law <law at cygnus dot com>
- Date: Tue, 26 Jan 1999 06:51:50 -0700
- cc: egcs-patches at cygnus dot com, mrs at wrs dot com
- Reply-To: law at cygnus dot com
In message <199901261348.OAA22230@ignucius.axis.se>you write:
> > From: Jeffrey A Law <law@cygnus.com>
> > Date: Fri Jan 22 08:36:57 CET 1999
> >
> > In message <199901211606.IAA06808@kankakee.wrs.com>you write:
> > > 1999-01-21 Mike Stump <mrs@wrs.com>
> > >
> > > * i386/vxi386.h: Improve vxWorks support.
> > Mostly OK. However, you have several ANSI namespace violations that need
> > to be addressed.
> >
> > We should not be defining anything that does not start with __ when -ansi
> > is in effect.
> >
> > It is not OK to use -Di386 in CPP_PREDEFINES. __i386__ is OK, but not -D
> i386.
>
> CPP_PREDEFINES has magic attached to it so the driver (gcc, g++)
> takes care of the ANSI (really: ISO) stuff, so that -Dfoo never
> appears without "__" when -ansi.
> So it *is* ok and recommended to have -Di386 and there was
> nothing wrong with Mike's CPP_PREDEFINES in the first place, or
> any port that has this, AFAICT. Correct-me.
Not everything calls cpp via the gcc driver. imake for example.
jeff