[FFmpeg-devel] [PATCH 1/4] avcodec/av1dec: Check for unset obu instead of crashing

James Almer jamrial at gmail.com
Mon Oct 26 02:00:53 EET 2020


On 10/25/2020 8:00 PM, Michael Niedermayer wrote:
> Fixes: NULL pointer dereference
> Fixes: 26550/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_AV1_fuzzer-5417762807349248
> 
> 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/av1dec.c | 7 ++++++-
>  1 file changed, 6 insertions(+), 1 deletion(-)
> 
> diff --git a/libavcodec/av1dec.c b/libavcodec/av1dec.c
> index a0a279d65b..bb4295aa6f 100644
> --- a/libavcodec/av1dec.c
> +++ b/libavcodec/av1dec.c
> @@ -673,7 +673,12 @@ static int av1_decode_frame(AVCodecContext *avctx, void *frame,
>      for (int i = 0; i < s->current_obu.nb_units; i++) {
>          CodedBitstreamUnit *unit = &s->current_obu.units[i];
>          AV1RawOBU *obu = unit->content;
> -        const AV1RawOBUHeader *header = &obu->header;
> +        const AV1RawOBUHeader *header;
> +
> +        if (!obu)

This looks like a "regression" since ae7686a64f, where ENOSYS (an error
code that's not propagated to the caller) on unsupported OBU types is
now returned before cbs_av1 even gets a chance to allocate the unit's
content.

Should be ok.

> +            continue;
> +
> +        header = &obu->header;
>          av_log(avctx, AV_LOG_DEBUG, "Obu idx:%d, obu type:%d.\n", i, unit->type);
>  
>          switch (unit->type) {
> 



More information about the ffmpeg-devel mailing list