[FFmpeg-devel] [PATCH V2] libavcodec/qsvdec.c: using queue count to unref frame

[Xiang, Haihao]

On Fri, 2021-03-12 at 10:44 +0800, wenbin.chen at intel.com wrote:
> From: "Chen,Wenbin" <wenbin.chen at intel.com>
> 
> MSDK vc1 and av1 sometimes output frame into the same suface, but
> ffmpeg-qsv assume the surface will be used only once, so it will
> unref the frame when it receives the output surface. Now change
> it to unref frame according to queue count.
> 
> Signed-off-by Wenbin Chen <wenbin.chen at intel.com>
> ---
>  libavcodec/qsvdec.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/libavcodec/qsvdec.c b/libavcodec/qsvdec.c
> index 5f2e641373..3abd7bd4a2 100644
> --- a/libavcodec/qsvdec.c
> +++ b/libavcodec/qsvdec.c
> @@ -513,7 +513,7 @@ static int qsv_decode(AVCodecContext *avctx, QSVContext
> *q,
>              return AVERROR_BUG;
>          }
>  
> -        out_frame->queued = 1;
> +        out_frame->queued += 1;
>          av_fifo_generic_write(q->async_fifo, &out_frame, sizeof(out_frame),
> NULL);
>          av_fifo_generic_write(q->async_fifo,
> &sync,      sizeof(sync),      NULL);
>      } else {
> @@ -526,7 +526,7 @@ static int qsv_decode(AVCodecContext *avctx, QSVContext
> *q,
>  
>          av_fifo_generic_read(q->async_fifo, &out_frame, sizeof(out_frame),
> NULL);
>          av_fifo_generic_read(q->async_fifo,
> &sync,      sizeof(sync),      NULL);
> -        out_frame->queued = 0;
> +        out_frame->queued -= 1;
>  
>          if (avctx->pix_fmt != AV_PIX_FMT_QSV) {
>              do {

LGTM, thanks!