Working with I/O Traffic Generators - 2023.2 English

Vitis Unified Software Platform Documentation: Application Acceleration Development (UG1393)

Document ID
UG1393
Release Date
2023-12-13
Version
2023.2 English

Some user applications such as video streaming and Ethernet-based applications make use of I/O ports on the platform to stream data into and out of the platform. For these applications, performing software and hardware emulation of the design, or running AI Engine simulation, requires a mechanism to mimic the hardware behavior of the I/O port, and to simulate data traffic running through the ports. I/O traffic generators let you model traffic through the I/O ports during software and hardware emulation in the AMD Vitis™ application acceleration development flow, during the AI Engine simulation flows (x86sim, AI Enginesim), or during logic simulation in the AMD Vivado™ Design Suite.

Important: Software emulation supports only AXI4-Stream I/O emulation and is not supported on AMD Zynq™ 7000 devices. Hardware emulation supports both AXI4-Stream and AXI4 memory map interface I/O emulation.

As described in AXI4-Stream I/O Model for Streaming Traffic, traffic generators can be written in Python, MATLAB, C/C++, or RTL (Verilog/SV) modules. They are launched in an external process which communicates with Vitis Emulation process or AI Engine simulation process using Inter Process Communication (IPC). The IPC connections are established using IPC AXI4-Stream master/slave modules as described in Running Traffic Generators in Python/C++/MATLAB.

Traffic generators are designed to pass data into your system, or receive data from your system. The APIs are provided to handle data transfer for standard datatypes, or for more complex datatypes. The API you use to create the traffic generator is determined by the datatype your system requires. For instance, standard datatypes are covered by simple API such as send_data, or receive_data as described in Python API for AI Engine Graphs or Python API for PL Kernels. More complex datatypes require API like those described in General Purpose Python API or Advanced Python Traffic Generator API.

The following are additional details on ways to integrate traffic generators in Python/C/C++/Verilog with subsequent PL kernels or the AI Engine kernels based on your application.