Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sources.redhat.com/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sources.redhat.com/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com
Reply-To: <mhcox AT bluezoosoftware DOT com>
From: "Michael H. Cox" <mhcox AT bluezoosoftware DOT com>
To: <cygwin AT cygwin DOT com>
Subject: RE: gcc-3.2-3:  bootstrap build fails (HAVE_DECL_GETOPT not in config.h?)
Date: Fri, 22 Nov 2002 01:26:01 -0700
Message-ID: <NDBBJEBGAAGAIKPGOOFIIEDIDMAA.mhcox@bluezoosoftware.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106
Importance: Normal
In-Reply-To: <20021120000137.GB27204@redhat.com>



> -----Original Message-----
> From: Christopher Faylor [mailto:cgf AT redhat DOT com]
> Sent: Tuesday, November 19, 2002 5:02 PM
> To: cygwin AT cygwin DOT com
> Subject: Re: gcc-3.2-3: bootstrap build fails (HAVE_DECL_GETOPT not in
> config.h?)
>
>
> On Wed, Nov 20, 2002 at 10:14:59AM +1100, Nigel Stewart & Fiona
> Smith wrote:
> >Michael,
> >
> >I have encountered the same problem on my setup.
> >
> >Regards,
>
> I have not encountered the same problem in my setup.

You may have something wrong with your setup, i.e. you may have other
changes besides what's in gcc-3.2-3.
I looked at the gcc/include/getopt.h in CVS at gcc.gnu.org and they have the
following change (which is
basically what I came up with also):

===================================================================
RCS file: /cvs/gcc/gcc/include/getopt.h,v
retrieving revision 1.6
retrieving revision 1.6.30.1
diff -u -r1.6 -r1.6.30.1
--- gcc/include/getopt.h	2001/03/14 19:44:38	1.6
+++ gcc/include/getopt.h	2002/11/09 18:12:56	1.6.30.1
@@ -111,7 +111,7 @@
    errors, only prototype getopt for the GNU C library.  */
 extern int getopt (int argc, char *const *argv, const char *shortopts);
 #else /* not __GNU_LIBRARY__ */
-# if !defined (HAVE_DECL_GETOPT)
+# if !defined (HAVE_DECL_GETOPT) && !defined (__cplusplus)
 extern int getopt ();
 # endif
 #endif /* __GNU_LIBRARY__ */
===================================================================

Although there still maybe a more general problem with the way GCC's cpp
finds header files.
The compile of pure.cc is finding the <unistd.h> in /usr/include which
includes <getopt.h>, but
instead of getting the getopt.h in /usr/include, it's grabbing the one in
./gcc/include.  I
vaguely remember something in the ISO C standard regarding C pre-processing
that states all
angle-bracket includes strictly search for files in the "standard
directories" before searching
in the "" include directories.  Therefore the ./gcc/include/getopt.h
shouldn't come into play.

In addition to the change shown above, to make the file more portable, line
102

	#if defined (__STDC__) && __STDC__

should be changed to

	#if (defined (__STDC__) && __STDC__) || defined (__cplusplus)

since ISO C++ Appendix C.1.9 Clause 16.8.1 says that "Whether __STDC__ is
defined and if so, what its
value is, are implementation-defined".  G++ defines it, but doesn't have to.
If getopt.h was compiled
with a C++ compiler that chose not to define/set __STDC__, the build would
again break.

Mike
>
> HTH,
> cgf
>


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/