Example Design

The Xilinx® DPU targeted reference design (TRD) provides instructions on how to use the DPU with a Xilinx SoC platform to build and run deep neural network applications. The TRD includes two parts, the Vivado DPU TRD and the Vitis™ DPU TRD. The TRD uses the Vivado IP integrator flow for building the hardware design and the Xilinx Yocto PetaLinux flow for software design. The Zynq® UltraScale+™ MPSoC platform is used to create this TRD. The TRD can be accessed through this link: https://www.xilinx.com/products/intellectual-property/dpu.html#overview.

This chapter describes the architecture of the reference design and provides a functional description of its components. It is organized as follows:

• Vivado TRD Overview provides a high-level overview of the Zynq UltraScale+ MPSoC architecture, the reference design architecture, and a summary of key features.
• Hardware Design gives an overview of how to use the Xilinx Vivado Design Suite to generate the reference hardware design.
• Software Design describes the design flow of project creation in the PetaLinux environment.
• Demo execution describes how to run an application created by the TRD.

The architecture of the Vitis DPU TRD reference design is similar to the Vivado TRD. However, the Vitis DPU TRD is friendly and flexible to the software designer.

For the Vivado DPU TRD Flow, refer to the https://github.com/Xilinx/Vitis-AI/blob/master/dsa/DPU-TRD/prj/Vivado/README.md.

This tutorial contains information about:

• Setting up the ZCU102 evaluation board and run the TRD.
• Changing the configuration of DPU.

For the Vitis™ DPU TRD Flow, refer to https://github.com/Xilinx/Vitis-AI/blob/master/dsa/DPU-TRD/prj/Vitis/README.md.

This tutorial contains information about:

• Setting up the ZCU102 evaluation board and running the TRD.
• Changing the DPU configuration.
• Integrating the DPU in a custom platform in the Vitis environment.
• Run the DPU_TRD in Vitis GUI flow.