Re: [FFmpeg-devel] [PATCH] avcodec/vda: define av_vda_default_init2 when CONFIG_H264_VDA_HWACCEL equ 0

2016-10-17 Thread Xidorn Quan
This isn't maintained by me. You should probaby ping Sebastien Zwickert <
dilar...@free.fr>.

But this change looks trivial, so I think it is fine.

- Xidorn

On Mon, Oct 17, 2016 at 7:01 PM, Steven Liu  wrote:

> ping Xidorn Quan
>
> 2016-10-15 6:01 GMT+08:00 wm4 :
>
>> On Thu, 13 Oct 2016 20:57:07 +0800
>> Steven Liu  wrote:
>>
>> > ping
>> >
>> > 2016-10-12 17:36 GMT+08:00 Steven Liu :
>> >
>> > > on OSX:
>> > > ../configure --disable-everything --enable-demuxer=hls make
>> > > error message: Undefined symbols for architecture x86_64:
>> > > "_av_vda_default_init2", referenced from:_videotoolbox_init in
>> > > ffmpeg_videotoolbox.o
>> > > so add av_vda_default_init2 when CONFIG_H264_VDA_HWACCEL=0
>> > >
>> > > Signed-off-by: Steven Liu 
>> > > ---
>> > >  libavcodec/vda.c |5 +
>> > >  1 files changed, 5 insertions(+), 0 deletions(-)
>> > >
>> > > diff --git a/libavcodec/vda.c b/libavcodec/vda.c
>> > > index 4670140..819ae03 100644
>> > > --- a/libavcodec/vda.c
>> > > +++ b/libavcodec/vda.c
>> > > @@ -73,6 +73,11 @@ int av_vda_default_init(AVCodecContext *avctx)
>> > >  return AVERROR(ENOSYS);
>> > >  }
>> > >
>> > > +int av_vda_default_init2(AVCodecContext *avctx, AVVDAContext
>> *vdactx)
>> > > +{
>> > > +return AVERROR(ENOSYS);
>> > > +}
>> > > +
>> > >  void av_vda_default_free(AVCodecContext *ctx)
>> > >  {
>> > >  }
>> > > --
>> > > 1.7.1
>> > >
>> > >
>> > >
>> > > ___
>> > > ffmpeg-devel mailing list
>> > > ffmpeg-devel@ffmpeg.org
>> > > http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>> > >
>> > ___
>> > ffmpeg-devel mailing list
>> > ffmpeg-devel@ffmpeg.org
>> > http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>>
>> Patch ok, but I can't push for a while (and I'm not VDA maintainer).
>> ___
>> ffmpeg-devel mailing list
>> ffmpeg-devel@ffmpeg.org
>> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>>
>
>
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel


Re: [FFmpeg-devel] vda: fix h64 decoding, vda wants the entire buffer

2015-04-01 Thread Xidorn Quan
On Wed, Apr 1, 2015 at 10:10 PM, Hendrik Leppkes 
wrote:

> On Wed, Apr 1, 2015 at 12:40 PM, Rainer Hochecker 
> wrote:
> > vda wants the entire buffer with all info. it has no other means to
> deliver
> > picture parameters.
> > according to documentation start_frame can pass the entire buffer
>
>
> This doesn't make much sense. If I follow that argument, then vda
> would never have worked, at all. How is it that it does work for many
> other people, presumably including the author of this code?


I haven't looked into the code for a long time. But what I can tell is, VDA
only worked
for really limited cases when I played with it. For example, it required
all extra data,
also it only recognized 'AVC1' as fourcc, 'H264' would make it failed.

I'd recommend you to try to use VT instead of VDA. In new OS X systems, VDA
is
just a simple wrapper of VT. VT is probably more powerful (though I haven't
ever
tried it myself.)

- Xidorn
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel