Message-ID: <3A2A032A.3980F482@softhome.net>
Date: Sun, 03 Dec 2000 09:24:10 +0100
From: Laurynas Biveinis <lauras AT softhome DOT net>
X-Mailer: Mozilla 4.76 [en] (Win98; U)
X-Accept-Language: lt,en
MIME-Version: 1.0
To: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
CC: djgpp-workers AT delorie DOT com
Subject: Re: Patch: make GCC & DJGPP headers compatible
References: <Pine DOT SUN DOT 3 DOT 91 DOT 1001203085529 DOT 1124J-100000 AT is>
Content-Type: text/plain; charset=iso-8859-4
Content-Transfer-Encoding: 7bit
Reply-To: djgpp-workers AT delorie DOT com

Eli Zaretskii wrote:
> What happens if we change the definition of wchar_t or wint_t?  Will that
> break GCC, or require to rebuild it after changing djgpp.h?

GCC must be rebuilt, like in any other case of library changes.

> I thought that fixing fixincludes to DTRT with DJGPP should help solving
> these problems.  Doesn't it?

Fixincludes are meant to fix ANSI incompatibilities in system headers. Our
headers are all OK (except stdio.h with va_list - but I'll discuss it later).
So it can't make a big difference there, because it doesn't fix anything in
our headers. Another possibility is that fixincludes does not recognize logic
in DJGPP headers - the way they define size_t etc. is not very common.

> What I'm trying to understand is whether our headers or the headers
> installed by GCC's "make install" procedure take precedence when you
> compile a program.

GCC headers.

> Also, what is the updated list of headers installed by GCC in the current
> CVS?

For C:
<iso646.h>
<stddef.h>
<stdbool.h>
<stdarg.h>
<varargs.h>
and fixincluded our <stdio.h>.

I assume you don't need a list of C++ headers, do you?

Laurynas