[FFmpeg-devel] [PATCH 1/2] avcodec/hevc_ps: Check log2_sao_offset_scale_*

James Almer jamrial at gmail.com
Wed Jan 24 05:47:18 EET 2018


On 1/24/2018 12:34 AM, Michael Niedermayer wrote:
> Fixes: 4868/clusterfuzz-testcase-minimized-6236542906400768
> Fixes: runtime error: shift exponent 126 is too large for 32-bit type 'int'
> 
> Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
> Signed-off-by: Michael Niedermayer <michael at niedermayer.cc>
> ---
>  libavcodec/hevc_ps.c | 11 +++++++++++
>  1 file changed, 11 insertions(+)
> 
> diff --git a/libavcodec/hevc_ps.c b/libavcodec/hevc_ps.c
> index 4787312cfa..746c96b17e 100644
> --- a/libavcodec/hevc_ps.c
> +++ b/libavcodec/hevc_ps.c
> @@ -1324,6 +1324,17 @@ static int pps_range_extensions(GetBitContext *gb, AVCodecContext *avctx,
>      pps->log2_sao_offset_scale_luma = get_ue_golomb_long(gb);
>      pps->log2_sao_offset_scale_chroma = get_ue_golomb_long(gb);
>  
> +    if (   pps->log2_sao_offset_scale_luma   > FFMAX(sps->bit_depth        - 10, 0)
> +        || pps->log2_sao_offset_scale_chroma > FFMAX(sps->bit_depth_chroma - 10, 0)
> +    ) {
> +        av_log(avctx, AV_LOG_ERROR,
> +                "log2 sao offset scales (%d %d) are invalid\n",
> +               pps->log2_sao_offset_scale_luma,
> +               pps->log2_sao_offset_scale_chroma
> +              );
> +        return AVERROR_INVALIDDATA;

Wouldn't it be better to just port the h264 and hevc decoder to use the
cbs API at this point? It correctly does a range check for every
sps/vps/pps/slice value already.

Otherwise you'll be adding a lot of range checks as oss-fuzz finds an
ubsan testcase for them.


More information about the ffmpeg-devel mailing list