[FFmpeg-devel] Google Summer of Code participation

[Michael Niedermayer]

On Tue, Apr 07, 2009 at 10:54:52PM +0200, Thilo Borgmann wrote:
>
>
> Michael Niedermayer schrieb:
>> they break vp6 decoding
>>
>>
>>   
> I think we are talking about revision 8?
> The vp6 decoder defines vp56_decode_frame as their decoding function, which 
> looks good in the source file and compiles fine. I do not see anything 
> wrong about vp56 or vp6 in the output of make test.
> What exactly does not work?

the patch broke fate, someone else fixed it already


>>> diff --git a/ffmpeg.c b/ffmpeg.c
>>> index cb15120..a35f282 100644
>>> --- a/ffmpeg.c
>>> +++ b/ffmpeg.c
>>> @@ -1224,8 +1224,8 @@ static int output_packet(AVInputStream *ist, int 
>>> ist_index,
>>>                  data_size= samples_size;
>>>                      /* XXX: could avoid copy if PCM 16 bits with same
>>>                         endianness as CPU */
>>> -                ret = avcodec_decode_audio2(ist->st->codec, samples, 
>>> &data_size,
>>> -                                           ptr, len);
>>> +                ret = avcodec_decode_audio3(ist->st->codec, samples, 
>>> &data_size,
>>> +                                           pkt);
>>>                  if (ret < 0)
>>>                      goto fail_decode;
>>>                  ptr += ret;
>>>     
>>
>> this looks wrong
>>   
> Hmm. It compiles and works fine here.
> What exactly is looking wrong? In context to the broken vp6 decoding?

ptr is changed, so is len, pkt is not, you replace ptr/len by pkt

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

It is dangerous to be right in matters on which the established authorities
are wrong. -- Voltaire
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20090407/8ab8b561/attachment.pgp>