[FFmpeg-devel] [PATCH 3/3] avcodec/cbs_av1: abort when the written inferred value is not the expected one

[James Almer]

On 6/14/2020 11:11 AM, Carl Eugen Hoyos wrote:
> Am Do., 11. Juni 2020 um 18:14 Uhr schrieb James Almer <jamrial at gmail.com>:
>>
>> If this happens, it's a sign of parsing issues earlier in the process, or
>> misuse by the calling module.
>>
>> Prevents creating invalid bitstreams.
>>
>> Signed-off-by: James Almer <jamrial at gmail.com>
>> ---
>>  libavcodec/cbs_av1.c | 3 ++-
>>  1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/libavcodec/cbs_av1.c b/libavcodec/cbs_av1.c
>> index fc228086c2..456bd9b1d5 100644
>> --- a/libavcodec/cbs_av1.c
>> +++ b/libavcodec/cbs_av1.c
>> @@ -711,10 +711,11 @@ static size_t cbs_av1_get_payload_bytes_left(GetBitContext *gbc)
>>
>>  #define infer(name, value) do { \
>>          if (current->name != (value)) { \
>> -            av_log(ctx->log_ctx, AV_LOG_WARNING, "Warning: " \
>> +            av_log(ctx->log_ctx, AV_LOG_ERROR, \
>>                     "%s does not match inferred value: " \
>>                     "%"PRId64", but should be %"PRId64".\n", \
>>                     #name, (int64_t)current->name, (int64_t)(value)); \
> 
>> +            return AVERROR_BUG; \
> 
> For this return value, please explain in the commit message
> why this can never happen and in the mail why you are not
> using an assert().

I could, i suggested it. But nobody commented if they preferred it or
not, so i went with a normal error code.
I could also just use the invalid data code, same as the other cases
here. I'm fine with any of those options.