Early FPGA/SoC Design Verification with Simulink and the Vivado Simulator from AMD Xilinx

Overview

Versal^® Adaptive SoCs are software-programmable, heterogeneous compute platforms that combine processing cores, programmable logic, and AI Engines. Versal Adaptive SoCs can be used to accelerate machine learning inference and advanced signal processing workloads, such as beamforming, FFTs, and filters.

Versal Adaptive SoC designs include legacy RTL code for the programmable logic along with functions that have been newly implemented for the high-performance AI Engine array. Since designs for Versal architecture are partitioned across different engines, it is important to functionally verify all new design content in the context of the complete design.

An approach that enables fast, functional simulation of a hybrid system is to simulate using the MATLAB and Simulink enviroment in combination with Vitis™ Model Composer. The HDL Verifier product from MathWorks now supports cosimulation between the Vivado® Simulator and either MATLAB or Simulink.

In this webinar, engineers from AMD and MathWorks will demonstrate techniques for early functional verification of designs that use both the programmable logic and AI Engine array functions. 

Highlights

• Performing fast functional cosimulation of a design that uses both programmable logic and AI Engine implementation
• Performing cycle-approximate cosimulation of imported HDL code (using the Vivado Simulator)
• Using the HDL Coder Workflow Advisor to generate HDL code and automatically verify the code through cosimulation with the Vivado Simulator
• Using MATLAB and Simulink environment test benches to verify implementation on AMD Xilinx development boards with FPGA-in-the-loop testing
• Generating SystemVerilog test bench components automatically for use with Vivado IDE verification environments

About the Presenters

Rob Graessle is a staff product development engineer with AMD/Xilinx, focusing on Vitis Model Composer. He has more than 10 years of experience working in and for the DoD on radar and software-defined radio applications. His passion is taking projects from requirements to simulation to prototype, then experiencing the thrill of seeing designs come to life. Rob earned BS and MS degrees in Computational Science & Engineering from Miami University.

Eric Cigan is a principal product marketing manager at MathWorks for ASIC and FPGA design and verification. Prior to joining MathWorks, he held technical marketing roles at Mentor Graphics, MathStar, and AccelChip. Eric earned BS and MS degrees in mechanical engineering from the Massachusetts Institute of Technology. In his spare time, Eric curates a wide-ranging, ever-growing collection of development boards from AMD/Xilinx, Avnet, and other sources.