1 Overview

The primary goal of this Design is to demonstrate the capabilities of VCU hard block present in Zynq UltraScale+ EV devices. The TRD will serve as a platform to tune the performance parameters of VCU and arrive at optimal configurations for encoder and decoder blocks. 

This design supports the following video interfaces:

Sources:

• Test pattern generator (TPG) implemented in the PL.
• HDMI-Rx capture pipeline implemented in the PL.
• MIPI CSI-2 Rx capture pipeline implemented in the PL.
• File source (SD card, USB storage, SATA hard disk).
• Stream-In from network or internet.

Sinks:

• DP Tx display pipeline in the PS.
• HDMI-Tx display pipeline implemented in the PL.

VCU Codec:

• Video Encode/Decode capability using VCU hard block in PL
  • AVC/HEVC encoding
  • Encoder/decoder parameter configuration.
  • Demonstrate the multi-stream capability of VCU at 4k 60 Hz throughput

Streaming Interfaces:

• 1G Ethernet PS GEM 

Video format:

• NV12

The below figure shows the TRD block diagram.

• Demonstrate the multi-stream capability of VCU at 4k 60 Hz throughput.
  • Supports 2-4KP30 multi-stream feature with any 2 of HDMI-Rx, TPG, and MIPI as the input source and HDMI-Tx as display pipeline.
  • Supports 4-1080p60 multi-stream feature with 3 HDMI-Rx and 1 MIPI as the input source and HDMI-Tx as display pipeline.
  • Supports 8-1080p30 multi-stream feature with 7 HDMI-Rx and 1 MIPI as the input source and HDMI-Tx as display pipeline.

Other features:

• This design supports 8 channel memory based SCD(Scene Change Detection) IP. SCD can be enabled or disabled through configuration

Supported Resolution:

The table below provides the supported resolution from GUI and command line app in this design.

√ - Supported
NA – Not applicable
x – Not supported

The below table gives information about the features supported in this design. 

NOTE: DP will support a max resolution of 4KP30 

            TPG will not support 1080P30 resolution mode.

The below figure shows the VCU TRD design hardware block diagram.

The below figure shows the VCU TRD design software block diagram.

1.1 Board Setup

Refer below link for Board Setup

• Zynq UltraScale+ MPSoC VCU TRD 2019.1 Board Setup

1.2 Run Flow

The TRD package is released with the source code, Vivado project, Petalinux BSP, and SD card image that enables the user to run the demonstration. It also includes the binaries necessary to configure and boot the ZCU106 board. Prior to running the steps mentioned in this wiki page, download the TRD package and extract its contents to a directory referred to as ‘TRD_HOME' which is the home directory.

Refer below link to download all TRD contents.

• Zynq UltraScale+ MPSoC VCU TRD 2019.1 Download zip 

TRD package contents are placed in the following directory structure. The user needs to copy all the files from the $TRD_HOME/images/vcu_trd/ to FAT32 formatted SD card directory.

└── rdf0428-zcu106-vcu-trd-2019-1
    ├── apu
    │   ├── apps
    │   ├── vcu_petalinux_bsp
    │   ├── vcu_sdx
    │   └── ws_bypass
    ├── host_x86
    │   └── host_package
    ├── images
    │   ├── vcu_10g
    │   ├── vcu_audio
    │   ├── vcu_hdmirx
    │   ├── vcu_hdmitx
    │   ├── vcu_pcie
    │   ├── vcu_plddr_hdmi
    │   ├── vcu_sdirx
    │   ├── vcu_sdirxtx
    │   ├── vcu_sditx
    │   ├── vcu_sdx
    │   └── vcu_trd
    ├── pl
    │   ├── constrs
    │   ├── pre-built
    │   ├── scripts
    │   └── srcs
    └── README.txt

TRD package contents specific to VCU TRD design are placed in the following directory structure.

Example:

4kp60 HEVC_HIGH Display Pipeline execution

% vcu_gst_app /media/card/config/1-4kp60/Display/Single_4kp60_HEVC_HIGH.cfg

4kp60 HEVC_HIGH Record Pipeline execution

% vcu_gst_app /media/card/config/1-4kp60/Record/Single_4kp60_HEVC_HIGH.cfg

4kp60 HEVC_HIGH Stream-out Pipeline execution

% vcu_gst_app /media/card/config/1-4kp60/Stream-out/Single_4kp60_HEVC_HIGH.cfg

4kp60 HEVC_HIGH Stream-in Pipeline execution

% vcu_gst_app /media/card/config/1-4kp60/Stream-in/input.cfg

NOTE: Make sure HDMI-Rx should be configured to 4kp60 mode.

To measure the latency of the pipeline, run the below command. The latency data is huge, so dump it to a file.

% GST_DEBUG="GST_TRACER:7" GST_TRACERS="latency;scheduletime" ./vcu_gst_app ./input.cfg >& dump_log.txt

Refer below link for detailed run flow steps

• Zynq UltraScale+ MPSoC VCU TRD 2019.1 - Run Flow

1.3 Build Flow

Refer below link for detailed build flow steps

• Zynq UltraScale+ MPSoC VCU TRD 2019.1 - Build Flow

2 Other Information

2.1 Known Issues

• Frame drop observed in TPG pipeline
  

  • Frequency: Always
  • Workaround: None (Note: 56 fps in 4kp60/1080p60, and 28 fps in 4kp30 resolution)
    
• VCU QT: Green shade on DP when play 4kp30 HDMI/MIPI pipeline with SCD
  • Frequency: Always
  • Workaround: None
• For VCU related known issues please referAR# 72293: PetaLinux 2019.1 - Product Update Release Notes and Known Issues.

2.2 Limitations

• For playback in DP, video input resolution should match to DP's native resolution. This constraint is due to support of GUI. In GUI case if we allow video source other than native resolution(by setting fullscreen overlay) then graphics layer will disappear. To recover back GUI user need to kill and relaunch the GUI app. To avoid such condition TRD only supports video input resolution which is equal to DP's native resolution.
• For VCU related limitations please refer AR# 72293: PetaLinux 2019.1 - Product Update Release Notes and Known Issues and PG252 link.

2.3 Optimum VCU Encoder parameters for use-cases:

Video streaming:

• Video streaming use-case requires very stable bitrate graph for all pictures.
• It is good to avoid periodic large Intra pictures during the encoding session
• Low-latency rate control (hardware RC) is the preferred control-rate for video streaming, it tries to maintain equal amount frame sizes for all pictures.
• Good to avoid periodic Intra frames instead use low-delay-p (IPPPPP…)
• VBR is not a preferred mode of streaming.

Performance: AVC Encoder settings:

• It is preferred to use 8 or higher slices for better AVC encoder performance.
• AVC standard does not support Tile mode processing which results in the processing of MB rows sequentially for entropy coding.

Quality: Low bitrate AVC encoding:

• Enable profile=high and use qp-mode=auto for low-bitrate encoding use-cases.
• The high profile enables 8x8 transform which results in better video quality at low bitrates.

3 Appendix A - Input Configuration File (input.cfg)

The example configuration files are stored at /media/card/config/ folder.

Common Configuration:
It is the starting point of common configuration.
 
Num of Input:
Provide the number of input. It is ranging from 1 to 8.

Output:
Select the video interface.
Options: HDMI or DP

Out Type:
Options: display, record, and stream

Display Rate:
Pipeline frame rate.
Options: 30 FPS or 60 FPS for each stream.

Exit:
It indicates to the application that the configuration is over.

Input Configuration:
It is the starting point of the input configuration.

Input Num:
Starting N^th input configuration.
Options: 1-8

Input Type:
Input source type.
Options: TPG, HDMI, HDMI_2, HDMI_3, HDMI_4, HDMI_5, HDMI_6, HDMI_7, MIPI, File, Stream

Uri:
File path or Network URL. Applicable for file playback and stream-in pipeline only. Supported file formats for playback are ts, mp4, and mkv.
NOTE: See Mount Location for additional file paths.
Options: file:///run/media/sda/abc.ts (for file path), udp://192.168.25.89:5004/ (for Network streaming, Here 192.168.25.89 is IP address and 5004 is port no)

Raw:
To tell the pipeline is processed or pass-through.
Options: True, False

Width:
The width of the live source.
Options: 3840, 1920

Height:
The height of the live source.
Options: 2160, 1080

Enable SCD:
Enable or Disable Memory based SCD in the pipeline.
Options: True, False

Exit:
It indicates to the application that the configuration is over.

Encoder Configuration:
It is the starting point of encoder configuration.

Encoder Num:
Starting N^th encoder configuration.
Options: 1-8

Encoder Name:
Name of the encoder.
Options: AVC, HEVC

Profile:
Name of the profile.
Options: baseline, main or high for AVC. Main for HEVC.

Rate Control:
Rate control options.
Options: CBR, VBR, and low-latency.

Filler Data:
Filler Data NAL units for CBR rate control.
Options: True, False

QP:
QP control mode used by the VCU encoder.
Options: Uniform, Auto

L2 Cache:
Enable or Disable L2Cache buffer in encoding process.
Options: True, False

Latency Mode:
Encoder latency mode.
Options: normal, sub_frame

Low Bandwidth:
If enabled, decrease the vertical search range used for P-frame motion estimation to reduce the bandwidth.
Options: True, False

Gop Mode:
Group of Pictures mode.
Options: Basic, low_delay_p, low_delay_b

Bitrate:
Target bitrate in Kbps
Options: 1-60000

B Frames:
Number of B-frames between two consecutive P-frames
Options: 0-4

Slice:
The number of slices produced for each frame. Each slice contains one or more complete macroblock/CTU row(s). Slices are distributed over the frame as regularly as possible. If slice-size is defined as well more slices may be produced to fit the slice-size requirement.
Options:
4-22 4kp resolution with HEVC codec
4-32 4kp resolution with AVC codec
4-32 1080p resolution with HEVC codec
4-32 1080p resolution with AVC codec

GoP Length:
The distance between two consecutive I frames
Options: 1-1000

Format:
The format of input data.
Options: NV12

Preset:
Options: HEVC_HIGH, HEVC_MEDIUM, HEVC_LOW, AVC_HIGH, AVC_MEDIUM, AVC_LOW, Custom

Exit
It indicates to the application that the configuration is over.

Record Configuration:
It is the starting point of record configuration.

Record Num:
Starting N^th record configuration.
Options: 1-8

Out-File Name:
Record file path.
NOTE: See Mount Location for additional file paths.
Options: /run/media/sda/abc.ts

Duration:
Duration in minutes.
Options: 1-3

Exit
It indicates to the application that the configuration is over.

Streaming Configuration:
It is the starting point of streaming configuration.

Streaming Num:
Starting N^th Streaming configuration.
Options: 1-8

Host IP:
The host to send the packets to
Options: 192.168.25.89 or Windows PC IP

Port:
The port to send the packets to
Options: 5004, 5008, 5012, 5016, 5020, 5024, 5028, and 5032.

Exit
It indicates to the application that the configuration is over.

Trace Configuration:
It is the starting point of trace configuration.

FPS Info:
To display fps info on the console.
Options: True, False

APM Info:
To display APM counter number on the console.
Options: True, False

Pipeline Info:
To display pipeline info on console.
Options: True, False

Exit
It indicates to the application that the configuration is over.

4 Appendix B

• Kill the Qt GUI application running on target board by executing the below commands from the serial console.

$ killall -9 run_vcu.sh
$ killall -9 vcu_qt
$ killall -9 Xorg

• HDMI source can be locked to any resolution. Run the below command for all media nodes to print media device topology where "mediaX" represents different media nodes. In the topology, log look for the “v_hdmi_rx_ss” string to identify the HDMI input source media node.

$ xmedia-ctl -p -d /dev/mediaX

• To check the link status, resolution and video node of the HDMI input source, run below xmedia-ctl command where "mediaX" indicates media node for the HDMI input source.

$ xmedia-ctl -p -d /dev/mediaX

When HDMI source is connected to 4KP60 resolution, it shows:

root@zcu106_vcu_trd:/media/card# xmedia-ctl -p -d /dev/mediaX
Media controller API version 4.19.0

Media device information
------------------------
driver          xilinx-video
model           Xilinx Video Composite Device
serial          
bus info        
hw revision     0x0
driver version  4.19.0

Device topology
- entity 1: vcap_hdmi output 0 (1 pad, 1 link)
            type Node subtype V4L flags 0
            device node name /dev/video0 -----> Video node for HDMI Rx source
        pad0: Sink
                <- "a0080000.v_proc_ss":1 [ENABLED]

- entity 5: a0080000.v_proc_ss (2 pads, 2 links)
            type V4L2 subdev subtype Unknown flags 0
            device node name /dev/v4l-subdev21
        pad0: Sink
                [fmt:RBG888_1X24/3840x2160 field:none]
                <- "a0000000.v_hdmi_rx_ss":0 [ENABLED]
        pad1: Source
                [fmt:VYYUYY8_1X24/3840x2160 field:none]
                -> "vcap_hdmi output 0":0 [ENABLED]

- entity 8: a0000000.v_hdmi_rx_ss (1 pad, 1 link)
            type V4L2 subdev subtype Unknown flags 0
            device node name /dev/v4l-subdev22
        pad0: Source
                [fmt:RBG888_1X24/3840x2160 field:none colorspace:srgb]
                [dv.caps:BT.656/1120 min:0x0@25000000 max:4096x2160@297000000 stds:CEA-861,DMT,CVT,GTF caps:progressive,reduced-blanking,custom]
                [dv.detect:BT.656/1120 3840x2160p60 (4400x2250) stds:CEA-861 flags:CE-video] -----> Resolution and Frame-rate of HDMI Rx source
                -> "a0080000.v_proc_ss":0 [ENABLED]

NOTE: Check resolution and frame-rate of "dv.detect" under "v_hdmi_rx_ss" node.

When the HDMI source is not connected, it shows:

root@zcu106_vcu_trd:/media/card# xmedia-ctl -p -d /dev/mediaX
Media controller API version 4.19.0

Media device information
------------------------
driver          xilinx-video
model           Xilinx Video Composite Device
serial          
bus info        
hw revision     0x0
driver version  4.19.0

Device topology
- entity 1: vcap_hdmi output 0 (1 pad, 1 link)
            type Node subtype V4L flags 0
            device node name /dev/video0 -----> Video node for HDMI Rx source
        pad0: Sink
                <- "a0080000.v_proc_ss":1 [ENABLED]

- entity 5: a0080000.v_proc_ss (2 pads, 2 links)
            type V4L2 subdev subtype Unknown flags 0
            device node name /dev/v4l-subdev21
        pad0: Sink
                [fmt:RBG888_1X24/3840x2160 field:none]
                <- "a0000000.v_hdmi_rx_ss":0 [ENABLED]
        pad1: Source
                [fmt:VYYUYY8_1X24/3840x2160 field:none]
                -> "vcap_hdmi output 0":0 [ENABLED]

- entity 8: a0000000.v_hdmi_rx_ss (1 pad, 1 link)
            type V4L2 subdev subtype Unknown flags 0
            device node name /dev/v4l-subdev22
        pad0: Source
                [fmt:RBG888_1X24/3840x2160 field:none colorspace:srgb]
                [dv.caps:BT.656/1120 min:0x0@25000000 max:4096x2160@297000000 stds:CEA-861,DMT,CVT,GTF caps:progressive,reduced-blanking,custom]
                [dv.query:no-link] -----> HDMI Rx Link Status
                -> "a0080000.v_proc_ss":0 [ENABLED]

NOTE: Here "dv.query:no-link" under "v_hdmi_rx_ss" node shows HDMI-Rx source is not connected or HDMI-Rx source is not active(Try waking up the device by pressing a key on remote).

• Follow the below steps to switch the HDMI-Rx resolution from 1080p60 to 4kp60.
  • Check current HDMI Input Source Resolution (1080p60) by following the above-mentioned steps.
  • Run vcu_gst_app for current HDMI resolution (1080p60) by executing the following command.

$ vcu_gst_app /media/card/config/input.cfg

Below configurations needs to be set in input.cfg for HDMI-1080p60.

Common Configuration    : START
Num Of Input            : 1
Output                  : HDMI
Out Type                : Display
Frame Rate              : 60
Exit

Input Configuration     : START
Input Num               : 1
Input Type              : hdmi
Raw                     : TRUE
Width                   : 1920
Height                  : 1080
Exit

• • Change Resolution of HDMI Input Source from 1080p60 to 4kp60 by following below steps.
    • Set the HDMI source resolution to 4kp60 (Homepage → settings → display & Sound → Resolution → change to 4kp60).
    • Save the configuration to take place the change.
  • Verify the desired HDMI Input Source Resolution (4kp60) by following the above-mentioned steps.

• If HDMI Tx link-up issue is observed after Linux booting, use the following command:

$ modetest -D a0070000.v_mix -s 41:3840x2160-60@BG24 &

• Run the following gst-launch-1.0 command to display passthrough pipeline. Where "videoX" indicates a video node for the input source.

$ gst-launch-1.0 v4l2src device=/dev/videoX io-mode=4 ! video/x-raw, width=3840, height=2160, format=NV12, framerate=60/1 ! queue ! kmssink bus-id="a0070000.v_mix"

• Run the following gst-launch-1.0 command to display processed pipeline (capture → scd → encode → decode → display) on HDMI-Tx. Where "videoX" indicates a video node for the input source.

$ gst-launch-1.0 v4l2src device=/dev/videoX io-mode=4 ! video/x-raw, width=3840, height=2160, format=NV12, framerate=60/1 ! xilinxscd io-mode=5 ! omxh265enc qp-mode=auto gop-mode=basic gop-length=60 b-frames=0 target-bitrate=60000 num-slices=8 control-rate=constant prefetch-buffer=true low-bandwidth=false filler-data=true cpb-size=1000 initial-delay=500 ! video/x-h265, profile=main, alignment=au ! queue ! omxh265dec internal-entropy-buffers=5 low-latency=0 ! queue max-size-bytes=0 ! kmssink bus-id="a0070000.v_mix"

• Run the following gst-launch-1.0 command to record video using GStreamer pipeline. Where "videoX" indicates a video node for the input source.