[FFmpeg-devel] [PATCH 4/6] avcodec/evc_parse: Check log2_sub_gop_length

[James Almer]

On 6/18/2023 7:27 PM, James Almer wrote:
> On 6/18/2023 6:50 PM, Michael Niedermayer wrote:
>> Fixes: 1.70141e+38 is outside the range of representable values of 
>> type 'int'
>> Fixes: 
>> 59883/clusterfuzz-testcase-minimized-ffmpeg_BSF_EVC_FRAME_MERGE_fuzzer-5557887217565696
>>
>> 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/evc_parse.c | 2 ++
>>   1 file changed, 2 insertions(+)
>>
>> diff --git a/libavcodec/evc_parse.c b/libavcodec/evc_parse.c
>> index 44be5c5291..822b236423 100644
>> --- a/libavcodec/evc_parse.c
>> +++ b/libavcodec/evc_parse.c
>> @@ -277,6 +277,8 @@ EVCParserSPS *ff_evc_parse_sps(EVCParserContext 
>> *ctx, const uint8_t *bs, int bs_
>>       if (!sps->sps_pocs_flag || !sps->sps_rpl_flag) {
>>           sps->log2_sub_gop_length = get_ue_golomb(&gb);
>> +        if (sps->log2_sub_gop_length > 5U)
>> +            return NULL;
>>           if (sps->log2_sub_gop_length == 0)
>>               sps->log2_ref_pic_gap_length = get_ue_golomb(&gb);
>>       }
> 
> LGTM, but please let me apply it as part of my evc patchset to prevent 
> conflicts.

Actually, this is leaving the SPS allocated in the array, which should 
be freed if we're going to start erroring out on failed range checks.
I'll amend it before applying it.