...

The Xilinx Certified Ubuntu 20.04 LTS for Xilinx Devices image is an official Ubuntu image with certified hardware support for select Xilinx evaluation boards. This page details how to boot and use the official desktop environment image released by Canonical for Xilinx ZCU102, ZCU104, and ZCU106 evaluation boards as well as the Kria KV260 Starter Kit. You can read more and download the certified Ubuntu hardware here.

Table of Contents

Table of Contents

Hardware and Software Requirements

...

• ZCU102: https://github.com/Xilinx/Vitis-AI/tree/v1.3.2/dsa/DPU-TRD/prj/Vitis#51-board-setup

• ZCU104: https://xilinx-wiki.atlassian.net/wiki/spaces/A/pages/1268973650/Zynq+UltraScale+MPSoC+ZCU104+VCU+HDMI+ROI+2020.2#1.5-Board-Setup

• ZCU106: https://xilinx-wiki.atlassian.net/wiki/spaces/A/pages/1186693157/Zynq+UltraScale+MPSoC+ZCU106+VCU+HDMI+ROI+TRD+2020.2#1.5-Board-Setup  

• Kria KV260 Starter Kit: Getting Started with Kria KV260 Starter Kit, Step 3: Connecting Everything

For general information about how the boot process works for the ZCU10x image, please refer to Booting Certified Ubuntu 20.04 LTS for Xilinx Devices. For information on how the boot process works for the Kria KV260 Starter Kit, please refer to the Kria K26 SOM wiki page.

...

$ sudo snap install xlnx-config --classic

After installing the xlnx-config snap, you can use it to set up the remainder of the system by executing the following command in a terminal:

$ xlnx-config.sysinit

...

 --channel=1.x

The heart of the ZCU10x demonstration experience is the Xilinx Deep Learning Processing Unit (DPU). In addition to setting up the system, the xlnx-config snap provides the DPU via dpu.xclbin to other snaps that depend on it. The xlnx-config snap also provides infrastructure which allows users to package up other hardware platforms to enable other custom development and demonstration workflows.  For more information about creating and installing custom platforms via xlnx-config, see the Usage Example section on the xlnx-config developer page.

Install the Custom Xilinx gstreamer

By default, the Ubuntu 20.04 LTS root filesystem includes the open source upstream version of gstreamer that is packaged with Ubuntu 20.04 LTS. In order to take advantage of the unique features provided by Xilinx multimedia IP, xlnx-config.sysinit installs a customized Xilinx-specific version of gstreamer with specialized extensions and plugins. 

Download a Vitis AI Model (ZCU104 and ZCU106 Only)

During xlnx-config.sysinit, the densebox_640_360 model is downloaded from the Xilinx Model Zoo to facilitate running of the VCU ROI TRD Demo on ZCU10x platforms.  

...

Each evaluation board boot image contains a default bitstream & device tree combination that is configured during the first boot.  Each ZCU10x platform includes the Xilinx Deep Learning Processing Unit (DPU) in the Zynq UltraScale+ Programmable Logic (PL). These platforms also include a software stack  (XRT + Vitis AI) required to take advantage of it.  The Kria KV260 Starter Kit includes a series of Vitis platforms to support various accelerated applications (AAs). All of the out-of-the-box images are based on Xilinx 2020.2 tools, IP, drivers, and firmware. The default platforms for the three boards are: 

...

Evaluation Board

...

Reference Hardware Platform

...

ZCU102

...

Vitis AI v1.3 MPSoC DPU TRD

...

ZCU104

...

2020.2 VCU HDMI ROI TRD

...

ZCU106

...

2020.2 VCU HDMI ROI TRD

...

Kria KV260 Starter Kit

...

Kria KV260 Vitis Platforms

Running the ZCU10x DPU Demos

Each of the Out-of-the-Box demo designs includes the Xilinx DPU for machine learning inference.  The ZCU104 & ZCU106 designs include a 4096x1 configuration. The ZCU102 includes 4096x3 configuration. Each DPU configuration has an identical feature set and they are compatible with the pre-built models in the Xilinx Model Zoo that target the ZCU10x boards.

Running the Vitis AI Library Samples Snap

The xlnx-vai-lib-samples snap includes a pre-built subset of the Vitis AI Library sample applications available as source in the Vitis AI Github repository.  After installing the snap, users can experiment with twenty-seven different Vitis AI samples and over 90 different pre-built models. For users on the ZCU10x platforms, this is all you need to do. Users on the Kria KV260 Starter Kit will require a compatible bitstream. Details on downloading and deploying this bitstream are coming soon.

For more information regarding using the xlnx-vai-lib-samples snap, please visit the xlnx-vail-lib-samples developer page.

Building the Vitis AI Library Sample Applications from Source

For users that would like to build the Vitis AI Library sample applications from source on the target, please visit the Building Vitis-AI Sample Applications After installing the xlnx-config snap, you can use it to set up the remainder of the system by executing the following command in a terminal:

$ xlnx-config.sysinit

xlnx-config.sysinit takes care of the following items, depending on the system it's run on:

Install the Custom Xilinx gstreamer

By default, the Ubuntu 20.04 LTS root filesystem includes the open source upstream version of gstreamer that is packaged with Ubuntu 20.04 LTS. In order to take advantage of the unique features provided by Xilinx multimedia IP, xlnx-config.sysinit installs a customized Xilinx-specific version of gstreamer with specialized extensions and plugins. 

Download a Vitis AI Model (ZCU104 and ZCU106 Only)

During xlnx-config.sysinit, the densebox_640_360 model is downloaded from the Xilinx Model Zoo to facilitate running of the VCU ROI TRD Demo on ZCU10x platforms.  

Anchor
ubuntu-demo-overview ubuntu-demo-overview

Overview of the Out-of-the-Box Demos

Each ZCU10x evaluation board boot image contains a default bitstream & device tree combination that is configured during the first boot.  Each ZCU10x platform includes the Xilinx Deep Learning Processing Unit (DPU) in the Zynq UltraScale+ Programmable Logic (PL). These platforms also include a software stack  (XRT + Vitis AI) required to take advantage of it.  The Kria KV260 Starter Kit includes a series of Vitis platforms to support various accelerated applications (AAs). All of the out-of-the-box images are based on Xilinx 2020.2 (2020.2.2 for Kria) tools, IP, drivers, and firmware.

The default platforms for the three boards are: 

Running the ZCU10x DPU Demos

Each of the Out-of-the-Box demo designs includes the Xilinx DPU for machine learning inference.  The ZCU104 & ZCU106 designs include a 4096x1 configuration. The ZCU102 includes 4096x3 configuration. Each DPU configuration has an identical feature set and they are compatible with the pre-built models in the Xilinx Model Zoo that target the ZCU10x boards.

Running the Vitis AI Library Samples Snap

The xlnx-vai-lib-samples snap includes a pre-built subset of the Vitis AI Library sample applications available as source in the Vitis AI Github repository.  After installing the snap, users can experiment with twenty-seven different Vitis AI samples and over 90 different pre-built models. For users on the ZCU10x platforms, this is all you need to do. Users on the Kria KV260 Starter Kit will require a compatible bitstream. The xlnx-nlp-smartvision snap includes a bitstream that is compatible with the xlnx-vai-lib-samples application.

For more information regarding using the xlnx-vai-lib-samples snap, please visit the xlnx-vail-lib-samples developer page.

Building the Vitis AI Library Sample Applications from Source

For users that would like to build the Vitis AI Library sample applications from source on the target, please visit the Building Vitis-AI Sample Applications on Certified Ubuntu 20.04 LTS for Xilinx Devices page for detailed steps.

Using the ZCU104 & ZCU106  VCU Region of Interest (ROI) TRD Demo

The primary goal of the VCU ROI design is to demonstrate the use of the Xilinx Deep learning Processor Unit (DPU) block for extracting the Region of Interest (ROI) data from input video frames.  This information is then used to perform ROI-based encoding using the Video Codec Unit (VCU) encoder hard block present in Zynq UltraScale+ EV devices.

For instruction for running the VCU ROI TRD Demo on the supported platforms, please visit the Running the VCU ROI Demo on Certified Ubuntu 20.04 LTS for Xilinx Devices page for detailed steps.

Using the ZCU104 & ZCU106  VCU Region of Interest (ROI) TRD Demo

The primary goal of the VCU ROI design is to demonstrate the use of the Xilinx Deep learning Processor Unit (DPU) block for extracting the Region of Interest (ROI) data from input video frames.  This information is then used to perform ROI-based encoding using the Video Codec Unit (VCU) encoder hard block present in Zynq UltraScale+ EV devices.

For instruction for running the VCU ROI TRD Demo on the supported platforms, please visit the Running the VCU ROI Demo on

Anchor
kv260_demos kv260_demos

Running the Kria KV260 Demos

The NLP SmartVision accelerated application is available as an Ubuntu snap. For more information, see the NLP SmartVision Snap wiki page.

Creating a Custom Hardware Platform

For the ZCU10x boards, users can replace the out-of-box hardware platform with customized versions. The xlnx-config snap helps facilitate switching between platforms.

For more information on how to package your custom boot assets into a Platform Assets Container (PAC) that xlnx-config can consume,, please refer to the xlnx-config Usage Example for ZCU10x.

Getting Help

Xilinx recommends posting on the Xilinx Embedded Linux Forum for any questions related to the Certified Ubuntu 20.04 LTS for on Xilinx Devices page for detailed stepsrelease.

Running the Kria KV260 Demos

The NLP SmartVision application is available as an Ubuntu snap. For more information, see the NLP SmartVision Snap wiki page.

Creating a Custom Hardware Platform

COMING SOON!

Getting Help

Xilinx recommends posting on the Xilinx Embedded Linux Forum for any questions related to the Certified Ubuntu 20.04 LTS on Xilinx Devices release.

...

Resources & References

This section outlines how to begin extending your usage beyond the initial Getting Started instructions.

For even more tips and tricks, see the Tips & Tricks page.

General Unsupported Usage Exceptions

The following configurations and use models are not supported:

• Portrait mode is unsupported

• Wake on USB (eg, pressing a key on a keyboard or clicking a button on a mouse) is not supported

...

This appends options to the kernel command line - the latest value is the one actually used. For this example, we're overriding the default CMA value of 1000M with cma=256M: 

...

...

Resources & References

This section outlines how to begin extending your usage beyond the initial Getting Started instructions.

For even more tips and tricks, see the Tips & Tricks page.

General Unsupported Usage Exceptions

The following configurations and use models are not supported:

• Portrait mode is unsupported

• Wake on USB (eg, pressing a key on a keyboard or clicking a button on a mouse) is not supported

Anchor
bootargs bootargs

Changing the Kernel bootargs Used By U-Boot

If your application or configuration requires specific arguments you can configure U-Boot to pass these arguments automatically.

Example bootargs lines

To update only the CMA value (256M in this case, 1000M is the default):

sudo sh -c 'echo "LINUX_KERNEL_CMDLINE=\"cma=256M\"" > /etc/default/u-boot-xlnx'

To update only the root filesystem location:

sudo sh -c 'echo "LINUX_KERNEL_CMDLINE=\"root=\\/dev\\/sda1\"" > /etc/default/u-boot-xlnx'

To update multiple arguments (CMA and root filesystem location):

sudo sh -c 'echo "LINUX_KERNEL_CMDLINE=\"cma=256M\ root=\\/dev\\/sda1\"" > /etc/default/u-

...

boot-xlnx'

After updating the Linux kernel boot arguments, be sure to update the U-Boot configuration so that these arguments are passed to the Linux kernel on the next reboot

ubuntu@zynqmp:~$ sudo dpkg-reconfigure u-boot-zynqmp -boot-zynqmp 

Remember to reboot the system for the changes to take place.

...

Installing & Using the Xilinx PPA

...

Anchor
ubuntu-kv260-special-notes ubuntu-kv260-special-notes

Special Considerations for the Kria KV260 Vision AI Starter Kit

COMING SOON!

Anchor
DP_Adapter DP_Adapter

DisplayPort to HDMI Adapters 

...

Related Links

• Booting Certified Ubuntu 20.04 LTS for Xilinx Devices

• Building Vitis-AI Sample Applications on Certified Ubuntu 20.04 LTS for Xilinx Devices

• https://xilinx-wiki.atlassian.net/wiki/pages/resumedraft.action?draftId=2105376824

• Tips & Tricks for Certified Ubuntu 20.04 LTS on Xilinx Devices

• Rebuilding the Certified Ubuntu for Xilinx Devices 20.04 LTS Kernel from Source

• Using the Certified Ubuntu on Xilinx Devices image with a SATA hard disk

• Snaps for Certified Ubuntu on Xilinx Devices

  • Snaps - xlnx-config Snap for Certified Ubuntu on Xilinx Devices

  • Snaps - xlnx-vai-lib-samples Snap for Certified Ubuntu on Xilinx Devices

  • Snaps - xlnx-nlp-smartvision Snap for Certified Ubuntu on Xilinx Devices