[MPlayer-cvslog] r25535 - trunk/configure

[Jindrich Makovicka]

On Fri, 28 Dec 2007 23:25:50 -0500
Rich Felker <dalias at aerifal.cx> wrote:

> On Fri, Dec 28, 2007 at 06:23:58PM +0100, eugeni wrote:
> > Author: eugeni
> > Date: Fri Dec 28 18:23:57 2007
> > New Revision: 25535
> > 
> > Log:
> > Add detection of *lrint* and round* functions to configure.
> > 
> > 
> > Modified:
> >    trunk/configure
> > 
> > Modified: trunk/configure
> > ==============================================================================
> > --- trunk/configure	(original)
> > +++ trunk/configure	Fri Dec 28 18:23:57 2007
> > @@ -2757,19 +2757,22 @@ if test "$_posix4" = yes ; then
> >  fi
> >  echores "$_posix4"
> >  
> > -echocheck "lrintf"
> > +for func in llrint lrint lrintf round roundf; do
> > +echocheck $func
> >  cat > $TMPC << EOF
> >  #include <math.h>
> > -int main(void) { long (*foo)(float); foo = lrintf;
> > (void)(*foo)(0.0); return 0; } +int main(void) { long
> > (*foo)(float); foo = $func; (void)(*foo)(0.0); return 0; } EOF
> > -_lrintf=no
> > -cc_check -D_GNU_SOURCE $_ld_lm && _lrintf=yes
> > -if test "$_lrintf" = yes ; then
> > -  _def_lrintf="#define HAVE_LRINTF 1"
> > +eval _$func=no
> > +cc_check -D_GNU_SOURCE $_ld_lm && eval _$func=yes
> 
> Could someone look into why this _GNU_SOURCE nonsense is here and, if
> there's a reason, find the proper fix? Using -D_GNU_SOURCE is
> obviously not correct or sane.

It enables C99 features and bunch of other stuff,
see /usr/include/features.h. -D_ISOC99_SOURCE could be used as well in
this case.

With newer GCCs, -std=c99 is preferred.

-- 
Jindrich Makovicka