[MPlayer-cvslog] r30464 - trunk/configure

[Reimar Döffinger]

On Sun, Jan 31, 2010 at 11:53:39AM +0100, Diego Biurrun wrote:
> On Sat, Jan 30, 2010 at 10:46:27PM +0100, Reimar Döffinger wrote:
> > On Sat, Jan 30, 2010 at 10:17:38PM +0100, diego wrote:
> > > @@ -9176,6 +9177,8 @@ $def_yasm
> > >  #define HAVE_POLL_H 1
> > >  #define HAVE_PPC4XX 0
> > >  #define HAVE_SETMODE 0
> > > +#define HAVE_STRUCT_IPV6_MREQ 1
> > > +#define HAVE_STRUCT_SOCKADDR_IN6 1
> > 
> > Is that tested on MinGW?
> 
> No.  It could very well fail, we shall get reports.
> 
> I was too lazy to port the complete test and I must say I'm getting
> tired of porting over FFmpeg configure changes.  I'm reviving my
> patch to build FFmpeg in a separate directory as we speak..

I'm currently not 100% convinced it is such a big win, I expect
that way we will get e.g. issues because FFmpeg detects and treats
the CPU differently than the main MPlayer does.
Have you considered just having FFmpeg's configure in our tree and
just merging parts of its config.h/config.mak with ours?
No, it doesn't sound that good to me either, just thinking about different
solutions.