[FFmpeg-devel] [PATCH 2/2] avcodec/libdav1d: honor the requested skip_frame level

James Almer jamrial at gmail.com
Fri Jan 7 19:21:54 EET 2022


On 1/7/2022 1:14 PM, Hendrik Leppkes wrote:
> On Fri, Jan 7, 2022 at 3:22 PM James Almer <jamrial at gmail.com> wrote:
>>
>> This supports dropping non-intra, non-key, or all frames.
>>
>> Signed-off-by: James Almer <jamrial at gmail.com>
>> ---
>>   libavcodec/libdav1d.c | 9 +++++++++
>>   1 file changed, 9 insertions(+)
>>
>> diff --git a/libavcodec/libdav1d.c b/libavcodec/libdav1d.c
>> index 8e45f533cb..9252a2572d 100644
>> --- a/libavcodec/libdav1d.c
>> +++ b/libavcodec/libdav1d.c
>> @@ -299,6 +299,7 @@ static int libdav1d_receive_frame(AVCodecContext *c, AVFrame *frame)
>>   #endif
>>       int res;
>>
>> +redo:
>>       if (!data->sz) {
>>           AVPacket *const pkt = dav1d->pkt;
>>
>> @@ -363,6 +364,14 @@ static int libdav1d_receive_frame(AVCodecContext *c, AVFrame *frame)
>>
>>       av_assert0(p->data[0] && p->allocator_data);
>>
>> +    if ((c->skip_frame >= AVDISCARD_NONINTRA && (p->frame_hdr->frame_type != DAV1D_FRAME_TYPE_KEY &&
>> +                                                 p->frame_hdr->frame_type != DAV1D_FRAME_TYPE_INTRA)) ||
>> +        (c->skip_frame >= AVDISCARD_NONKEY   &&  p->frame_hdr->frame_type != DAV1D_FRAME_TYPE_KEY) ||
>> +         c->skip_frame >= AVDISCARD_ALL) {
>> +        dav1d_picture_unref(p);
>> +        goto redo;
>> +    }
> 
> Is skipping still meaningful if it doesn't actually speed up decoding?

Outputting only I frames, but that's also possible using a filter, i guess.

In any case, no alternative for libdav1d unless the actual library adds 
support for this, so I'll drop this patch.

> 
> - Hendrik
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> 
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request at ffmpeg.org with subject "unsubscribe".



More information about the ffmpeg-devel mailing list