[meta-ti] [PATCH] omapfbplay: import from OE classic

[Denys Dmytriyenko]

On Thu, May 10, 2012 at 02:23:35PM +0000, Monk, Roger wrote:
> On Thu, May 10, 2012 at 15:20:31, Koen Kooi wrote:
> > Subject: Re: [meta-ti] [PATCH] omapfbplay: import from OE classic
> >
> >
> > Op 10 mei 2012, om 16:13 heeft Monk, Roger het volgende geschreven:
> >
> > > On Thu, May 10, 2012 at 15:09:36, Koen Kooi wrote:
> > >> Subject: Re: [meta-ti] [PATCH] omapfbplay: import from OE classic
> > >>
> > >>
> > >> Op 10 mei 2012, om 15:42 heeft Monk, Roger het volgende geschreven:
> > >>
> > >>> Hi Koen,
> > >>>
> > >>> Think we've still got some API incompatibility here.  I needed to
> > >> roll
> > >>> back to an earlier omapfbplay commit
> > >>> (34293052c5a2ae328eac6903512e6b4ce19b5639) and patch CODEC_TYPE :-
> > >>>
> > >>> -        if (afc->streams[i]->codec->codec_type == CODEC_TYPE_VIDEO
> > >> && !st)
> > >>> +        if (afc->streams[i]->codec->codec_type ==
> > >>> + AVMEDIA_TYPE_VIDEO && !st)
> > >>>
> > >>> to get this working on beagleboard-xm, so I suspect there may have
> > >> been an API change after Aug11, or
> > >> d03f042ecac0a231bedeadc7a6c67f0102d5505a was rebased against 0.8
> > >> branch?  I guess we should ask Mans...
> > >>
> > >> We still use libav 0.7.4 in meta-oe, I suspect Mans updated to
> > >> 0.8.x or even newer. I only build tested it, so I'm open for fixes
> > >> that make it run :) Upgrading libav might be an option as well, but
> > >> I'm not sure of that will get accepted for the meta-oe release branch.
> > >
> > > See attached for now - probably ok as reasonable known working
> > baseline before we upgrade other bits?
> >
> > The patch looks good to me. Once the beaglebone a6 release it out I'll
> > have time to look at libav. I'll need it for a customer project in
> > that timeframe as well. Thanks for updating the patch!
> 
> Np! - you gave me great head-start! - I'll re-send via git send-email.

BTW, do we want to have PROVIDES and PREFERRED_PROVIDER for all those 
different variants of omapfbplay?

-- 
Denys