Re: [PATCH v4 1/2] media: uapi: Add H264 low-level decoder API compound controls.

From: Maxime Ripard
Date: Tue Mar 05 2019 - 06:16:33 EST


On Fri, Feb 22, 2019 at 04:46:17PM +0900, Tomasz Figa wrote:
> Hi Maxime,
>
> On Wed, Feb 20, 2019 at 11:17 PM Maxime Ripard
> <maxime.ripard@xxxxxxxxxxx> wrote:
> >
> > From: Pawel Osciak <posciak@xxxxxxxxxxxx>
> >
> > Stateless video codecs will require both the H264 metadata and slices in
> > order to be able to decode frames.
> >
> > This introduces the definitions for a new pixel format for H264 slices that
> > have been parsed, as well as the structures used to pass the metadata from
> > the userspace to the kernel.
> >
> > Co-Developped-by: Maxime Ripard <maxime.ripard@xxxxxxxxxxx>
> > Signed-off-by: Pawel Osciak <posciak@xxxxxxxxxxxx>
> > Signed-off-by: Guenter Roeck <groeck@xxxxxxxxxxxx>
> > Signed-off-by: Maxime Ripard <maxime.ripard@xxxxxxxxxxx>
>
> Thanks for the patch. Some comments inline.
>
> [snip]
> > +``V4L2_CID_MPEG_VIDEO_H264_SLICE_PARAMS (struct)``
> > + Specifies the slice parameters (as extracted from the bitstream)
> > + for the associated H264 slice data. This includes the necessary
> > + parameters for configuring a stateless hardware decoding pipeline
> > + for H264. The bitstream parameters are defined according to
> > + :ref:`h264`. Unless there's a specific comment, refer to the
> > + specification for the documentation of these fields, section 7.4.3
> > + "Slice Header Semantics".
>
> Note that this is expected to be an array, with entries for all the
> slices included in the bitstream buffer.
>
> > +
> > + .. note::
> > +
> > + This compound control is not yet part of the public kernel API and
> > + it is expected to change.
> > +
> > +.. c:type:: v4l2_ctrl_h264_slice_param
> > +
> > +.. cssclass:: longtable
> > +
> > +.. flat-table:: struct v4l2_ctrl_h264_slice_param
> > + :header-rows: 0
> > + :stub-columns: 0
> > + :widths: 1 1 2
> > +
> > + * - __u32
> > + - ``size``
> > + -
> > + * - __u32
> > + - ``header_bit_size``
> > + -
> > + * - __u16
> > + - ``first_mb_in_slice``
> > + -
> > + * - __u8
> > + - ``slice_type``
> > + -
> > + * - __u8
> > + - ``pic_parameter_set_id``
> > + -
> > + * - __u8
> > + - ``colour_plane_id``
> > + -
> > + * - __u8
> > + - ``redundant_pic_cnt``
> > + -
> > + * - __u16
> > + - ``frame_num``
> > + -
> > + * - __u16
> > + - ``idr_pic_id``
> > + -
> > + * - __u16
> > + - ``pic_order_cnt_lsb``
> > + -
> > + * - __s32
> > + - ``delta_pic_order_cnt_bottom``
> > + -
> > + * - __s32
> > + - ``delta_pic_order_cnt0``
> > + -
> > + * - __s32
> > + - ``delta_pic_order_cnt1``
> > + -
> > + * - struct :c:type:`v4l2_h264_pred_weight_table`
> > + - ``pred_weight_table``
> > + -
> > + * - __u32
> > + - ``dec_ref_pic_marking_bit_size``
> > + -
> > + * - __u32
> > + - ``pic_order_cnt_bit_size``
> > + -
> > + * - __u8
> > + - ``cabac_init_idc``
> > + -
> > + * - __s8
> > + - ``slice_qp_delta``
> > + -
> > + * - __s8
> > + - ``slice_qs_delta``
> > + -
> > + * - __u8
> > + - ``disable_deblocking_filter_idc``
> > + -
> > + * - __s8
> > + - ``slice_alpha_c0_offset_div2``
> > + -
> > + * - __s8
> > + - ``slice_beta_offset_div2``
> > + -
> > + * - __u8
> > + - ``num_ref_idx_l0_active_minus1``
> > + -
> > + * - __u8
> > + - ``num_ref_idx_l1_active_minus1``
> > + -
> > + * - __u32
> > + - ``slice_group_change_cycle``
> > + -
> > + * - __u8
> > + - ``ref_pic_list0[32]``
> > + -
> > + * - __u8
> > + - ``ref_pic_list1[32]``
> > + -
>
> Should we explicitly document that these are the lists after applying
> the per-slice modifications, as opposed to the original order from
> v4l2_ctrl_h264_decode_param?
>
> [snip]
> > + * .. _V4L2-PIX-FMT-H264-SLICE:
> > +
> > + - ``V4L2_PIX_FMT_H264_SLICE``
> > + - 'S264'
> > + - H264 parsed slice data, as extracted from the H264 bitstream.
> > + This format is adapted for stateless video decoders that
> > + implement an H264 pipeline (using the :ref:`codec` and
> > + :ref:`media-request-api`). Metadata associated with the frame
> > + to decode are required to be passed through the
> > + ``V4L2_CID_MPEG_VIDEO_H264_SPS``,
> > + ``V4L2_CID_MPEG_VIDEO_H264_PPS``,
> > + ``V4L2_CID_MPEG_VIDEO_H264_SLICE_PARAMS`` and
> > + ``V4L2_CID_MPEG_VIDEO_H264_DECODE_PARAMS`` controls and
> > + scaling matrices can optionally be specified through the
> > + ``V4L2_CID_MPEG_VIDEO_H264_SCALING_MATRIX`` control. See the
> > + :ref:`associated Codec Control IDs <v4l2-mpeg-h264>`.
> > + Exactly one output and one capture buffer must be provided for
> > + use with this pixel format. The output buffer must contain the
> > + appropriate number of macroblocks to decode a full
> > + corresponding frame to the matching capture buffer.
>
> What does it mean that a control can be optionally specified? A
> control always has a value, so how do we decide that it was specified
> or not? Should we have another control (or flag) that selects whether
> to use the control? How is it better than just having the control
> initialized with the default scaling matrix and always using it?

Ok, I'll change it.

> > diff --git a/include/uapi/linux/videodev2.h b/include/uapi/linux/videodev2.h
> > index 9a920f071ff9..6443ae53597f 100644
> > --- a/include/uapi/linux/videodev2.h
> > +++ b/include/uapi/linux/videodev2.h
> > @@ -653,6 +653,7 @@ struct v4l2_pix_format {
> > #define V4L2_PIX_FMT_H264 v4l2_fourcc('H', '2', '6', '4') /* H264 with start codes */
> > #define V4L2_PIX_FMT_H264_NO_SC v4l2_fourcc('A', 'V', 'C', '1') /* H264 without start codes */
> > #define V4L2_PIX_FMT_H264_MVC v4l2_fourcc('M', '2', '6', '4') /* H264 MVC */
> > +#define V4L2_PIX_FMT_H264_SLICE v4l2_fourcc('S', '2', '6', '4') /* H264 parsed slices */
>
> Are we okay with adding here already, without going through staging first?

This is what we did for MPEG-2 already (the format is public but the
controls are not), so I'm not sure this is causing any issue.

Thanks!
Maxime

--
Maxime Ripard, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

Attachment: signature.asc
Description: PGP signature