[FFmpeg-devel] [PATCH] H.264/AVCHD interlaced fixes
Mon Feb 9 16:24:32 CET 2009
On Mon, Feb 09, 2009 at 02:47:53PM +0100, Ivan Schreter wrote:
> Michael Niedermayer wrote:
> > On Mon, Feb 09, 2009 at 12:52:46AM +0100, Ivan Schreter wrote:
> > [...]
> >> @@ -7340,6 +7344,8 @@
> >> int context_count = 0;
> >> h->max_contexts = avctx->thread_count;
> >> + h->sei_recovery_frame_cnt = -1;
> >> +
> > i suspect that this will misbehave if CODEC_FLAG2_CHUNKS is set
> The docs for the flag is quite brief. I suppose, the application will
> say that it won't give whole frame/field buffer at once, but send it in
> chunks? If that's the case, yes, recovery frame count will be overwritten.
yes though slices may not be split across chunks
> First I thought initializing it in frame_start() should do the trick.
> However, this is called only _after_ SEI recovery point is processed.
> There seems to be no good point where to initialize it.
> So where the hell can I put the initialization for this case?
hmm, my first thought was to add the var to AVFrame, that would also
output it which we may want one day but i think we dont have the AVFrame
the other way you might try is reset it when a pic is done/output
Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB
Let us carefully observe those good qualities wherein our enemies excel us
and endeavor to excel them, by avoiding what is faulty, and imitating what
is excellent in them. -- Plutarch
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Size: 189 bytes
Desc: Digital signature
More information about the ffmpeg-devel