On Saturday 08 February 2014 06:16:10 Luca Barbato wrote:
> It was done only in check_mvset(), while mv_scale() is called also by
> dist_scale().
>
> Sample-Id: 00001579-google
> Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
> CC: [email protected]
> ---
> libavcodec/hevc_mvs.c | 8 ++++++--
> 1 file changed, 6 insertions(+), 2 deletions(-)
>
> diff --git a/libavcodec/hevc_mvs.c b/libavcodec/hevc_mvs.c
> index 49d5ff2..2fe4dbb 100644
> --- a/libavcodec/hevc_mvs.c
> +++ b/libavcodec/hevc_mvs.c
> @@ -568,8 +568,12 @@ static av_always_inline void dist_scale(HEVCContext *s,
> Mv *mv, int ref_pic_elist = refPicList[elist].list[TAB_MVF(x,
> y).ref_idx[elist]]; int ref_pic_curr =
> refPicList[ref_idx_curr].list[ref_idx];
>
> - if (ref_pic_elist != ref_pic_curr)
> - mv_scale(mv, mv, s->poc - ref_pic_elist, s->poc - ref_pic_curr);
> + if (ref_pic_elist != ref_pic_curr) {
> + int poc_diff = s->poc - ref_pic_elist;
> + if (!poc_diff)
> + poc_diff = 1;
> + mv_scale(mv, mv, poc_diff, s->poc - ref_pic_curr);
> + }
> }
>
> static int mv_mp_mode_mx(HEVCContext *s, int x, int y, int pred_flag_index,
Looks OK, but it seems like it'd be simpler to do the check directly in
mv_scale (and maybe also print an error since this should not happen in a
valid bitstream).
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
