[FFmpeg-devel] [PATCH 1/4] avformat/avidec: Check for dv streams before using priv_data in parse ##dc/##wb

Andreas Rheinhardt andreas.rheinhardt at gmail.com
Fri Mar 19 02:19:28 EET 2021


Michael Niedermayer:
> On Fri, Mar 19, 2021 at 12:20:23AM +0100, Andreas Rheinhardt wrote:
>> Michael Niedermayer:
>>> Fixes: null pointer dereference
>>> Fixes: 31588/clusterfuzz-testcase-minimized-ffmpeg_dem_AVI_fuzzer-6165716135968768
>>>
>>> Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
>>> Signed-off-by: Michael Niedermayer <michael at niedermayer.cc>
>>> ---
>>>  libavformat/avidec.c | 2 +-
>>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/libavformat/avidec.c b/libavformat/avidec.c
>>> index fa0599501a..48370fe5ce 100644
>>> --- a/libavformat/avidec.c
>>> +++ b/libavformat/avidec.c
>>> @@ -1288,7 +1288,7 @@ start_sync:
>>>                  AVStream *st1   = s->streams[1];
>>>                  AVIStream *ast1 = st1->priv_data;
>>>                  // workaround for broken small-file-bug402.avi
>>> -                if (   d[2] == 'w' && d[3] == 'b'
>>> +                if (ast1 && d[2] == 'w' && d[3] == 'b'
>>>                     && n == 0
>>>                     && st ->codecpar->codec_type == AVMEDIA_TYPE_VIDEO
>>>                     && st1->codecpar->codec_type == AVMEDIA_TYPE_AUDIO
>>>
>> How is this possible? After all, dv streams also have an AVIStream as
> 
> The DV demuxer creates streams in dv_extract_audio_info() without a AVIStream
> 

That explains it. Thanks. Patch is fine by me, but I haven't looked at
it in detail. But neither dv nor avi set the AVFMTCTX_NOHEADER flag, so
adding streams later is an API violation.

> 
>> priv_data; and only the very first stream can ever be a dv stream due to
>> the check in line 605.
> 
> I assume they are created after that check
> 
> 



More information about the ffmpeg-devel mailing list