[FFmpeg-devel] [RFC PATCH 2/4] avcodec/cbs_h2645: keep nal->nuh_layer_id > 0

["zhilizhao(赵志立)"]

> On Jan 5, 2023, at 19:18, James Almer <jamrial at gmail.com> wrote:
> 
> On 1/5/2023 1:14 PM, Zhao Zhili wrote:
>> From: Zhao Zhili <zhilizhao at tencent.com>
>> ---
>>  libavcodec/cbs_h2645.c | 3 ---
>>  1 file changed, 3 deletions(-)
>> diff --git a/libavcodec/cbs_h2645.c b/libavcodec/cbs_h2645.c
>> index 4ee06003c3..b12abede29 100644
>> --- a/libavcodec/cbs_h2645.c
>> +++ b/libavcodec/cbs_h2645.c
>> @@ -477,9 +477,6 @@ static int cbs_h2645_fragment_add_nals(CodedBitstreamContext *ctx,
>>          AVBufferRef *ref;
>>          size_t size = nal->size;
>>  -        if (nal->nuh_layer_id > 0)
>> -            continue;
> 
> Have you tested samples with NALUs with a layer > 0? How does CBS handle them in both reading and writing?

Yes, only with hevc alpha layer encoding. Both reading and writing work
fine. What is missing is parse VPS extension, SEI alpha channel info and
so on.

Patch 4/4 rewrite nuh_layer_id from non-zero to zero, and it works from
my test (for independent non-base layer).

> 
>> -
>>          // Remove trailing zeroes.
>>          while (size > 0 && nal->data[size - 1] == 0)
>>              --size;
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> 
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request at ffmpeg.org with subject "unsubscribe".