Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 13 Mar 2003 13:15:08 +0100
From:      Andre Albsmeier <andre.albsmeier@siemens.com>
To:        Andre Albsmeier <andre.albsmeier@siemens.com>, Eric Anholt <eta@lclark.edu>, Garance A Drosihn <drosih@rpi.edu>, ports@FreeBSD.ORG, stable@FreeBSD.ORG
Subject:   Re: HEADSUP: XFree86 4.3.0 update
Message-ID:  <20030313131508.A1456@curry.mchp.siemens.de>
In-Reply-To: <20030313120807.GA90326@falcon.midgard.homeip.net>; from ertr1013@student.uu.se on Thu, Mar 13, 2003 at 01:08:08PM %2B0100
References:  <1047429424.8471.40.camel@leguin> <p05200f25ba9537ca5fb3@[128.113.24.47]> <1047515632.614.163.camel@leguin> <20030313114036.A81385@curry.mchp.siemens.de> <20030313120807.GA90326@falcon.midgard.homeip.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 13-Mar-2003 at 13:08:08 +0100, Erik Trulsson wrote:
> On Thu, Mar 13, 2003 at 11:40:36AM +0100, Andre Albsmeier wrote:
> > On Wed, 12-Mar-2003 at 16:33:52 -0800, Eric Anholt wrote:
> > > On Wed, 2003-03-12 at 11:21, Garance A Drosihn wrote:
> > > > At 4:37 PM -0800 3/11/03, Eric Anholt wrote:
> > > > >I've committed the update of XFree86 to 4.3.0 to ports.  I think
> > > > >I've cleaned up after my mess at this point, but there may still
> > > > >be issues.  Please report to me if you have any problems with the
> > > > >new ports or any issues with XFree86 that you didn't have in 4.2.0.
> > > > 
> > > > A minor observation:
> > > > 
> > > > As I watch this all compiling, I notice a lot of compiles
> > > > are generating:
> > > > 
> > > > /usr/include/sys/cdefs.h:273: warning: `_POSIX_C_SOURCE' is not defined
> > > > /usr/include/sys/cdefs.h:279: warning: `_POSIX_C_SOURCE' is not defined
> > > 
> > > These ones annoy me, too (-current).
> > > 
> > > /usr/include/sys/time.h: In function `timespec2bintime':
> > > /usr/include/sys/time.h:125: warning: ISO C89 forbids long long integer constants
> > > /usr/include/sys/time.h: In function `timeval2bintime':
> > > /usr/include/sys/time.h:142: warning: ISO C89 forbids long long integer constants
> > > 
> > > If anyone proposes the right way to clean up these messages, I would
> > > love to hear.  XFree86 uses -ansi -pedantic in its compiles, and I'm
> > > wondering if we could just drop that and make things look prettier, at
> > > least for this second message.
> 
> Dropping -pedantic should get rid of the warnings about 'long long' and
> shouldn't hurt anything. (People developing XFree86 will probably want
> to keep it, but for those of us who are just compiling it from ports
> there is not much point in using -pedantic.)
> Personally I got annoyed enough by warnings about 'long long' that I
> modified the source for gcc such that -pedantic no longer warns about
> 'long long'.  This is almost certainly not the correct solution, though
> it stops the warnings.
> The correct solution is probably to fix the header files so they don't
> use 'long long' when compiled in strict ANSI C mode.
> 
> 
> > 
> > I am using this patch now (which might be complete nonsense but it
> > stops the warnings):
> 
> I have been using an identical patch for some time now (after being
> overly annoyed by the warnings) and I believe it is the correct way of
> getting rid of the warnings about _POSIX_C_SOURCE being undefined.
> Since this patch
> a) is perfectly legal C
> b) does not change the semantics of the header file
> c) does get rid of the warnings
> d) prevents the code from using the value of a preprocessor macro
>    without defining it first which, while legal, is hardly good
>    practice (which is why the warning is there in the first place) 

I have already mailed Garrett Wollman <wollman@freebsd.org> about
this. It was his patch in

http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/sys/cdefs.h.diff?r1=1.58&r2=1.59

that changed lines like

#if defined(_POSIX_C_SOURCE) && _POSIX_C_SOURCE < 199309

to 

#if _POSIX_C_SOURCE == 1

So we had the "#if defined" already in there a while ago...

> I wouldn't call the patch complete nonsense but rather call it a good idea.

Well, I just wanted to protect myself from flames since I have no
ideas what styles/standards/... I am violating with it :-)

	-Andre

> 
> 
> > 
> > --- sys/cdefs.h.ORI	Wed Sep 18 16:52:23 2002
> > +++ sys/cdefs.h	Thu Mar 13 10:20:54 2003
> > @@ -269,6 +269,8 @@
> >   * Our macros begin with two underscores to avoid namespace screwage.
> >   */
> >  
> > +#ifdef _POSIX_C_SOURCE
> > +
> >  /* Deal with IEEE Std. 1003.1-1990, in which _POSIX_C_SOURCE == 1. */
> >  #if _POSIX_C_SOURCE == 1
> >  #undef _POSIX_C_SOURCE		/* Probably illegal, but beyond caring now. */
> > @@ -280,6 +282,8 @@
> >  #undef _POSIX_C_SOURCE
> >  #define	_POSIX_C_SOURCE		199209
> >  #endif
> > +
> > +#endif
> >  
> >  /* Deal with various X/Open Portability Guides and Single UNIX Spec. */
> >  #ifdef _XOPEN_SOURCE
> 
> 
> 
> 
> -- 
> <Insert your favourite quote here.>
> Erik Trulsson
> ertr1013@student.uu.se

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




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030313131508.A1456>