[FFmpeg-devel] [PATCH] H264 parser fix
Wed Oct 6 19:50:23 CEST 2010
On Wed, Oct 06, 2010 at 10:32:17AM -0700, John Stebbins wrote:
> On 09/28/2010 11:46 AM, John Stebbins wrote:
> > On 09/28/2010 11:20 AM, Michael Niedermayer wrote:
> >> On Tue, Sep 28, 2010 at 09:41:03AM -0700, John Stebbins wrote:
> >> [...]
> >>>> also the factorization should be a seperate patch
> >>>> [...]
> >>> Any movement on this issue? Still seeing it in svn 25249. Avatar BD m2ts is a good example of the problem.
> >> A variant of this patch has been commited.
> >> Is there an issue that is fixed by a prior patch of this thread?
> > Sorry, I didn't test the prior patch. I assumed it hadn't been committed since I was seeing what appeared to be the
> > same problem. Perhaps it's not the same issue. I do not see these errors with an earlier version of ffmpeg, svn 22950
> > which is about 5 months old.
> > If you would like, I can back out to a version prior to the variant that was committed and try the prior patch. Can you
> > tell me at what revision that was committed?
> I did some more investigation of this and found the source of the problem. The time_offset_length from hrd_parameters
> is not being set to the correct default value when hrd_parameters are not present. According to the specification:
> "When the time_offset_length syntax element is not present, it shall be inferred to be equal to 24."
> The proper number of bits are not being read in decode_picture_timing() in the SEI. Since it is now out of sync with
> the bitstream, it starts misinterpreting the remaining data.
> Patch attached.
> h264_ps.c | 1 +
> 1 file changed, 1 insertion(+)
> fea8210b0b86c01e203a8816387417595bf9f5c0 time_offset_length_default.patch
Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB
The worst form of inequality is to try to make unequal things equal.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Size: 198 bytes
Desc: Digital signature
More information about the ffmpeg-devel