[FFmpeg-cvslog] vp9: always consume the full input packet

[Hendrik Leppkes]

ffmpeg | branch: master | Hendrik Leppkes <h.leppkes at gmail.com> | Tue Oct 22 17:57:56 2013 +0200| [c2f3715cb3fce093137591e3005de3e831b1f9fd] | committer: Michael Niedermayer

vp9: always consume the full input packet

Reviewed-by: "Ronald S. Bultje" <rsbultje at gmail.com>
Signed-off-by: Michael Niedermayer <michaelni at gmx.at>

> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=c2f3715cb3fce093137591e3005de3e831b1f9fd
---

 libavcodec/vp9.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/libavcodec/vp9.c b/libavcodec/vp9.c
index 1d06460..cb14ec4 100644
--- a/libavcodec/vp9.c
+++ b/libavcodec/vp9.c
@@ -3495,14 +3495,14 @@ static int vp9_decode_packet(AVCodecContext *avctx, void *out_pic,
                 case_n(3, AV_RL24(idx));
                 case_n(4, AV_RL32(idx));
             }
-            return size;
+            return avpkt->size;
         }
     }
     // if we get here, there was no valid superframe index, i.e. this is just
     // one whole single frame - decode it as such from the complete input buf
     if ((res = vp9_decode_frame(avctx, out_pic, got_frame, data, size)) < 0)
         return res;
-    return size;
+    return avpkt->size;
 }
 
 static void vp9_decode_flush(AVCodecContext *ctx)