From owner-freebsd-bugs  Thu Oct  7 17:20: 5 1999
Delivered-To: freebsd-bugs@freebsd.org
Received: from freefall.freebsd.org (freefall.FreeBSD.ORG [204.216.27.21])
	by hub.freebsd.org (Postfix) with ESMTP id AF02114BB7
	for <freebsd-bugs@FreeBSD.org>; Thu,  7 Oct 1999 17:20:02 -0700 (PDT)
	(envelope-from gnats@FreeBSD.org)
Received: (from gnats@localhost)
	by freefall.freebsd.org (8.9.3/8.9.2) id RAA77164;
	Thu, 7 Oct 1999 17:20:02 -0700 (PDT)
	(envelope-from gnats@FreeBSD.org)
Date: Thu, 7 Oct 1999 17:20:02 -0700 (PDT)
Message-Id: <199910080020.RAA77164@freefall.freebsd.org>
To: freebsd-bugs@FreeBSD.org
Cc: 
From: Jacques Vidrine <n@nectar.com>
Subject: Re: bin/13383 sys/netinet/in.h violates C++ spec. 
Reply-To: Jacques Vidrine <n@nectar.com>
Sender: owner-freebsd-bugs@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.org

The following reply was made to PR bin/13383; it has been noted by GNATS.

From: Jacques Vidrine <n@nectar.com>
To: gibbs@caspian.plutotech.com
Cc: freebsd-gnats-submit@freebsd.org
Subject: Re: bin/13383 sys/netinet/in.h violates C++ spec. 
Date: Thu, 07 Oct 1999 19:18:55 -0500

 On Thu, 07 Oct 1999 13:18:10 -0600, "Justin T. Gibbs" <gibbs@caspian.plutotech.com> wrote:
 [snip]
 > This certainly will show the error, my complicated C++ program shows the
 > error, but your example does not:
 > foo.cc:
 > #include <sys/types.h>
 > #include "foo.h"
 > 
 > int
 > main()
 > {
 >         return 1;
 > }
 > 
 > foo.h:
 > #include <sys/cdefs.h>
 > 
 > __BEGIN_DECLS   /* extern C protection has no effect */
 > struct in_addr {
 >         u_int32_t s_addr;
 > };
 > 
 > struct ip_opts {
 >         struct  in_addr ip_dst;         /* first hop, 0 w/o src rt */
 >         char    ip_opts[40];            /* actually variable in size */
 > };
 > __END_DECLS 
 
 Ah ha!
 
 Change "foo.h" to <foo.h> and use ``g++ -I$PWD [...]'' -- the
 compilation will complete successfully and without warnings.
 
 GCC is treating ``system'' headers specially.  Seems reasonable to me.
 
 As far as I am concerned, there are two alternatives we can follow:
 
 1) Do nothing, all versions of gcc and egcs seem to handle the situation
    in a reasonable and useful manner (aka if it isn't broke, don't fix it).
 2) Adopt the change that OSF/1 and OpenBSD have (see patch at the end of
    this message).
 
 I'm going to research the C++ standards I can get my hands on this
 weekend to determine if the current behavior is conformant or not.
 If it is not conformant, I will do (2).
 
 --- in.h.orig   Thu Oct  7 19:17:26 1999
 +++ in.h        Thu Oct  7 19:18:28 1999
 @@ -289,7 +289,11 @@
   */
  struct ip_opts {
         struct  in_addr ip_dst;         /* first hop, 0 w/o src rt */
 +#if defined(__cplusplus)
 +       char    Ip_opts[40];            /* the member name must be different */
 +#else
         char    ip_opts[40];            /* actually variable in size */
 +#endif
  };
  /*
 
 Jacques Vidrine / n@nectar.com / nectar@FreeBSD.org
 


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