Re: [PATCH 1/3] media: uapi: h264: update reference lists

From: Nicolas Dufresne
Date: Wed Jul 08 2020 - 20:59:39 EST


Le mercredi 08 juillet 2020 Ã 17:57 +0200, Jernej Åkrabec a Ãcrit :
> Hi!
>
> Dne sreda, 08. julij 2020 ob 15:28:52 CEST je Ezequiel Garcia napisal(a):
> > Hello Jernej,
> >
> > I'd like to post a new H264 uAPI cleanup series soon,
> > would you mind resending this, or otherwise do you
> > mind if I include this patch in the series?
>
> I don't mind at all. Currently my focus was elsewhere...
>
> > See below for a tiny comment.
> >
> > On Thu, 4 Jun 2020 at 15:55, Jernej Skrabec <jernej.skrabec@xxxxxxxx> wrote:
> > > When dealing with with interlaced frames, reference lists must tell if
> > > each particular reference is meant for top or bottom field. This info
> > > is currently not provided at all in the H264 related controls.
> > >
> > > Make reference lists hold a structure which will also hold flags along
> > > index into DPB array. Flags will tell if reference is meant for top or
> > > bottom field.
> > >
> > > Currently the only user of these lists is Cedrus which is just compile
> > > fixed here. Actual usage of newly introduced flags will come in
> > > following commit.
> > >
> > > Signed-off-by: Jernej Skrabec <jernej.skrabec@xxxxxxxx>
> > > ---
> > >
> > > .../media/v4l/ext-ctrls-codec.rst | 40 ++++++++++++++++++-
> > > .../staging/media/sunxi/cedrus/cedrus_h264.c | 6 +--
> > > include/media/h264-ctrls.h | 12 +++++-
> > > 3 files changed, 51 insertions(+), 7 deletions(-)
> > >
> > > diff --git a/Documentation/userspace-api/media/v4l/ext-ctrls-codec.rst
> > > b/Documentation/userspace-api/media/v4l/ext-ctrls-codec.rst index
> > > d0d506a444b1..6c36d298db20 100644
> > > --- a/Documentation/userspace-api/media/v4l/ext-ctrls-codec.rst
> > > +++ b/Documentation/userspace-api/media/v4l/ext-ctrls-codec.rst
> > > @@ -1843,10 +1843,10 @@ enum v4l2_mpeg_video_h264_hierarchical_coding_type
> > > ->
> > > * - __u32
> > >
> > > - ``slice_group_change_cycle``
> > > -
> > >
> > > - * - __u8
> > > + * - struct :c:type:`v4l2_h264_reference`
> > >
> > > - ``ref_pic_list0[32]``
> > > - Reference picture list after applying the per-slice modifications
> > >
> > > - * - __u8
> > > + * - struct :c:type:`v4l2_h264_reference`
> > >
> > > - ``ref_pic_list1[32]``
> > > - Reference picture list after applying the per-slice modifications
> > >
> > > * - __u32
> > >
> > > @@ -1926,6 +1926,42 @@ enum v4l2_mpeg_video_h264_hierarchical_coding_type
> > > -
> > >
> > > - ``chroma_offset[32][2]``
> > > -
> > >
> > > +``Picture Reference``
> > > +
> > > +.. c:type:: v4l2_h264_reference
> > > +
> > > +.. cssclass:: longtable
> > > +
> > > +.. flat-table:: struct v4l2_h264_reference
> > > + :header-rows: 0
> > > + :stub-columns: 0
> > > + :widths: 1 1 2
> > > +
> > > + * - __u16
> > > + - ``flags``
> > > + - See :ref:`Picture Reference Flags <h264_reference_flags>`
> > > + * - __u8
> > > + - ``index``
> > > + -
> > > +
> > > +.. _h264_reference_flags:
> > > +
> > > +``Picture Reference Flags``
> > > +
> > > +.. cssclass:: longtable
> > > +
> > > +.. flat-table::
> > > + :header-rows: 0
> > > + :stub-columns: 0
> > > + :widths: 1 1 2
> > > +
> > > + * - ``V4L2_H264_REFERENCE_FLAG_TOP_FIELD``
> > > + - 0x00000001
> > > + -
> > > + * - ``V4L2_H264_REFERENCE_FLAG_BOTTOM_FIELD``
> > > + - 0x00000002
> > > + -
> > > +
> > >
> > > ``V4L2_CID_MPEG_VIDEO_H264_DECODE_PARAMS (struct)``
> > >
> > > Specifies the decode parameters (as extracted from the bitstream)
> > > for the associated H264 slice data. This includes the necessary
> > >
> > > diff --git a/drivers/staging/media/sunxi/cedrus/cedrus_h264.c
> > > b/drivers/staging/media/sunxi/cedrus/cedrus_h264.c index
> > > 54ee2aa423e2..cce527bbdf86 100644
> > > --- a/drivers/staging/media/sunxi/cedrus/cedrus_h264.c
> > > +++ b/drivers/staging/media/sunxi/cedrus/cedrus_h264.c
> > > @@ -166,8 +166,8 @@ static void cedrus_write_frame_list(struct cedrus_ctx
> > > *ctx,>
> > > static void _cedrus_write_ref_list(struct cedrus_ctx *ctx,
> > >
> > > struct cedrus_run *run,
> > >
> > > - const u8 *ref_list, u8 num_ref,
> > > - enum cedrus_h264_sram_off sram)
> > > + const struct v4l2_h264_reference
> > > *ref_list, + u8 num_ref, enum
> > > cedrus_h264_sram_off sram)>
> > > {
> > >
> > > const struct v4l2_ctrl_h264_decode_params *decode =
> > > run->h264.decode_params; struct vb2_queue *cap_q;
> > >
> > > @@ -188,7 +188,7 @@ static void _cedrus_write_ref_list(struct cedrus_ctx
> > > *ctx,>
> > > int buf_idx;
> > > u8 dpb_idx;
> > >
> > > - dpb_idx = ref_list[i];
> > > + dpb_idx = ref_list[i].index;
> > >
> > > dpb = &decode->dpb[dpb_idx];
> > >
> > > if (!(dpb->flags & V4L2_H264_DPB_ENTRY_FLAG_ACTIVE))
> > >
> > > diff --git a/include/media/h264-ctrls.h b/include/media/h264-ctrls.h
> > > index 080fd1293c42..9b1cbc9bc38e 100644
> > > --- a/include/media/h264-ctrls.h
> > > +++ b/include/media/h264-ctrls.h
> > > @@ -140,6 +140,14 @@ struct v4l2_h264_pred_weight_table {
> > >
> > > #define V4L2_H264_SLICE_FLAG_DIRECT_SPATIAL_MV_PRED 0x04
> > > #define V4L2_H264_SLICE_FLAG_SP_FOR_SWITCH 0x08
> > >
> > > +#define V4L2_H264_REFERENCE_FLAG_TOP_FIELD 0x01
> > > +#define V4L2_H264_REFERENCE_FLAG_BOTTOM_FIELD 0x02
> > > +
> > > +struct v4l2_h264_reference {
> > > + __u8 flags;
> > > + __u8 index;
> > > +};
> > > +
> > >
> > > struct v4l2_ctrl_h264_slice_params {
> > >
> > > /* Size in bytes, including header */
> > > __u32 size;
> > >
> > > @@ -182,8 +190,8 @@ struct v4l2_ctrl_h264_slice_params {
> > >
> > > * Entries on each list are indices into
> > > * v4l2_ctrl_h264_decode_params.dpb[].
> > > */
> > >
> > > - __u8 ref_pic_list0[32];
> > > - __u8 ref_pic_list1[32];
> > > + struct v4l2_h264_reference ref_pic_list0[32];
> > > + struct v4l2_h264_reference ref_pic_list1[32];
> >
> > Could we use a macro for "32" here? Something like:
> >
> > #define V4L2_H264_REF_PIC_LIST_LEN (V4L2_H264_NUM_DPB_ENTRIES * 2).
> >
> > Does it make sense to add a comment as well?
> >
> > I was thinking something along these lines:
> >
> > """
> > Pictures in the DPB can be a frame, a complementary field pair or a
> > single field.
>
> To be honest, I don't know if user has a free choice to select same or
> different destination (capture) buffer for another field. I never tested it and
> I'm not sure how to test it with ffmpeg. HW deinterlacing cores on Allwinner
> SoCs support only interleaved fields as a input, that's why I never though
> about separate fields.

When setting the format, one will choose V4L2_FIELD_INTERLACED_TB/BT.
Most driver will support that, but if not, they can update field in the
structure. Interlaced being very common, I think it's fine to only
support that, but the drive must enforce this field value.

But I know the Xilinx ZynqMP will do ALTERNATE instead, which would be
each field get decoded in it's own buffer. That, to be honest, I
haven't though about. The tricky part is for H264, since it means 1
slice may populate two capture buffers and I don't know if we can
support this right now.

For now I'd say we should just ensure that whatever userspace ask, we
enforce V4L2_FIELD_INTERLACED_TB/BT appropriately. That is likely not
complete, just look at how interlacing is signal in the bitstream.

>
> Best regards,
> Jernej
>
> > Therefore, reference pictures lists need twice as much entries, so it
> > can reference
> > either field of a field pair.
> > """
> >
> > While it doesn't replace proper H264 specification reading,
> > it would add some clarity.
> >
> > Thanks,
> > Ezequiel
>
>
>