Re: [PATCH 2/2] media: uapi: h264: clarify pic_order_cnt_bit_size field

[Hans Verkuil <hverkuil@xs4all.nl>]

On 9/26/19 4:09 PM, Philipp Zabel wrote:
> Hi Hans,
> 
> On Thu, 2019-09-26 at 15:23 +0200, Hans Verkuil wrote:
>> On 9/5/19 3:12 PM, Philipp Zabel wrote:
>>> Since pic_order_cnt_bit_size is not a syntax element itself, explicitly
>>> state that it is the total size in bits of the pic_order_cnt_lsb,
>>> delta_pic_order_cnt_bottom, delta_pic_order_cnt[0], and
>>> delta_pic_order_cnt[1] syntax elements contained in the slice.
>>>
>>> Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
>>> ---
>>>  Documentation/media/uapi/v4l/ext-ctrls-codec.rst | 3 ++-
>>>  1 file changed, 2 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/Documentation/media/uapi/v4l/ext-ctrls-codec.rst b/Documentation/media/uapi/v4l/ext-ctrls-codec.rst
>>> index c281bc7ed1b3..08b49b2bbfa8 100644
>>> --- a/Documentation/media/uapi/v4l/ext-ctrls-codec.rst
>>> +++ b/Documentation/media/uapi/v4l/ext-ctrls-codec.rst
>>> @@ -1799,7 +1799,8 @@ enum v4l2_mpeg_video_h264_hierarchical_coding_type -
>>>        - Size in bits of the dec_ref_pic_marking() syntax element.
>>>      * - __u32
>>>        - ``pic_order_cnt_bit_size``
>>> -      -
>>> +      - Size in bits of the pic_order_cnt_lsb, delta_pic_order_cnt_bottom,
>>> +        delta_pic_order_cnt[0], and delta_pic_order_cnt[1] syntax elements.
>>
>> It's a little bit ambiguous: this field contains the size in bits of all these
>> fields combined, right? Not the size in bits of each field separately.
> 
> Yes.
> 
>> I.e. if these fields are all 8 bits, then pic_order_cnt_bit_size is 4*8 and
>> not just 8.
> 
> The size of pic_order_cnt_lsb is determined by another field's value
> (log2_max_pic_order_cnt_lsb_minus4), and the other three are
> exponential-Golomb coded, so each can be of different length (including
> zero).
> 
>> I think this should be rephrased.
> 
> Ok, how about:
> 
>  "Combined size in bits of the picture order count related syntax
>   elements: pic_order_cnt_lsb, delta_pic_order_cnt_bottom,
>   delta_pic_order_cnt[0], and delta_pic_order_cnt[1]."

That's better.

> 
> Actually, there's either pic_order_cnt_lsb + (optionally)
> delta_pic_order_cnt_bottom present, or delta_pic_order_cnt[0] +
> (optionally) delta_pic_order_cnt[1], never all four. Describing that in
> the uapi docs seemed overly verbose, though.

I agree.

Regards,

	Hans

 The relevant part in the
> slice_header() syntax spec looks like this:
> 
>   if (pic_order_cnt_type == 0) {
>     pic_order_cnt_lsb
>     if (bottom_field_pic_order_in_frame_present_flag && !field_pic_flag)
>       delta_pic_order_cnt_bottom
>   }
>   if (pic_order_cnt_type == 1 && !delta_pic_order_always_zero_flag) {
>     delta_pic_order_cnt[0]
>     if (bottom_field_pic_order_in_frame_present_flag && !field_pic_flag)
>       delta_pic_order_cnt[1]
>   }
> 
> pic_order_cnt_bit_size is supposed to be the size in bits of this whole
> block. pic_order_cnt_type and log2_max_pic_order_cnt_lsb_minus4 are from
> the SPS header, bottom_field_pic_order_in_frame_present_flag is from the
> PPS header, and field_pic_flag is contained earlier in the slice header.
> 
> regards
> Philipp
>