Date: Wed, 14 Jun 2000 18:45:05 +0300 (IDT)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: djgpp-workers AT delorie DOT com
Subject: Re: Patch: sentinels for typedefs in headers
In-Reply-To: <Pine.A41.4.05.10006141622240.32182-100000@ieva06.lanet.lv>
Message-ID: <Pine.SUN.3.91.1000614184408.2198B-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk


On Wed, 14 Jun 2000, Andris Pavenis wrote:

> gcc/ginclude/stdarg.h uses builtin gcc functions for va_list, va_arg, etc.
> So unless there is a bug in gcc they should correpond the real stack
> layout. If we provide our own definitions (like DJGPP's stdarg.c) we
> should carefully verify that none of gcc options that influences data
> alignment in stack (-mpreferred-stack-boundary) doesn't break it.
> 
> I think it would be better to use gcc builtin definitions if possible.

I agree: in the case of va_list, we should track the GCC definitions, 
since it's the compiler who generates the code for that.