Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 26 Nov 2002 20:54:28 -0800
From:      Juli Mallett <jmallett@FreeBSD.org>
To:        Mike Barcroft <mike@FreeBSD.org>
Cc:        obrien@FreeBSD.org, freebsd-standards@FreeBSD.org, re@FreeBSD.org
Subject:   Re: Approval to fix non-GCC users of <stdarg.h>
Message-ID:  <20021126205428.A95993@FreeBSD.org>
In-Reply-To: <20021126231641.C67382@espresso.q9media.com>; from mike@FreeBSD.org on Tue, Nov 26, 2002 at 11:16:41PM -0500
References:  <20021126194219.A92609@FreeBSD.org> <20021126231641.C67382@espresso.q9media.com>

next in thread | previous in thread | raw e-mail | index | archive | help
* De: Mike Barcroft <mike@FreeBSD.org> [ Data: 2002-11-26 ]
	[ Subjecte: Re: Approval to fix non-GCC users of <stdarg.h> ]
> [re@ probably doesn't need to be CC'd on this thread.  I'm sure they
> have enough mail to read, but I'll include them anyway.]
> 
> Juli Mallett <jmallett@FreeBSD.org> writes:
> > The following patch fixes <stdarg.h> for use on i386 with
> > non-GCC consumers...  __va_size is defined inside an if GNUC
> > block, but not used solely within an if GNUC block.  This
> > allows me to compile a number of pieces of software I work
> > on with TenDRA 4.1.2 on -CURRENT again.
> > 
> > It only "affects" things that would break, behaviour stays
> > the same for modern GCC (unrelated block) and for old GCC
> > (definition still in scope).
> > 
> > I'd like to get approval to get this in for 5.0.
> 
> I sent David O'Brien a similar patch about a week ago.  IIRC, he's
> already seeking re@ approval.  There are some other issues which
> prevented me from completing the latest TenDRA snapshot build.

Oh, ok, cool.

> 1. The new #warning in <sys/syslimits.h> isn't #if __GNUC__ ... #endif
> protected.  Adding the protection doesn't fix it for some reason (ie.
> it's complaining about unknown CPP syntax in a conditional that
> doesn't apply.)  I haven't looked to see what the C rule is for this,
> but I suspect TenDRA is doing something wrong.

The "protect ancient compilers" trick of adding whitespace doesn't fix
it.  I tossed an email to BDE asking if we should just make it an error,
as a warning of that variety doesn't go a long way to preventing such
things, but *shrug*.

> 2. TenDRA assumes an older POSIX system and makes wrong assumptions
> about some limits it happens to duplicate in its version of
> <limits.h>.

This is a -Yansi / -Ysystem / -Yposix thing, and I haven't *seen* any
problems, but I don't know for sure if it is right or not.
-- 
Juli Mallett <jmallett@FreeBSD.org>
OpenDarwin, Mono, FreeBSD Developer.
ircd-hybrid Developer, EFnet addict.
FreeBSD on MIPS-Anything on FreeBSD.

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-standards" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20021126205428.A95993>