On 09/05/15 22:01, Anton Khirnov wrote:
> Fixes possible invalid reads, once one of those refs is freed, but the
> others remain.
> CC: [email protected]
> ---
>  libavcodec/h264_refs.c | 9 +++++++++
>  1 file changed, 9 insertions(+)
> 
> diff --git a/libavcodec/h264_refs.c b/libavcodec/h264_refs.c
> index 2ddbdb7..adc2213 100644
> --- a/libavcodec/h264_refs.c
> +++ b/libavcodec/h264_refs.c
> @@ -640,6 +640,15 @@ int ff_h264_execute_ref_pic_marking(H264Context *h, MMCO 
> *mmco, int mmco_count)
>              if (h->short_ref[0] == h->cur_pic_ptr)
>                  remove_short_at_index(h, 0);
>  
> +            /* make sure the current picture is not already assigned as a 
> long ref */
> +            if (h->cur_pic_ptr->long_ref) {
> +                for (j = 0; j < FF_ARRAY_ELEMS(h->long_ref); j++) {
> +                    if (h->long_ref[j] == h->cur_pic_ptr)
> +                        remove_long(h, j, 0);
> +                }
> +            }
> +
> +
>              if (h->long_ref[mmco[i].long_arg] != h->cur_pic_ptr) {
>                  remove_long(h, mmco[i].long_arg, 0);
>  
> 

Probably ok, Janne what do you think?

lu
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to