A system-level view of program execution can be helpful in identifying problems during program execution including functional correctness and performance related challenges. While system level hardware emulation helps during the development phase of the system, AMD also offers functional and performance debug flows in hardware. The AI Engine architecture has support for generation, collection, and streaming of profile related data, and events as trace data during hardware execution.
Tip: To support the testing of AI Engine graphs on VCK190, AMD provides a ready-to-use test harness. This test
harness allows a simple and natural transition from the AI Engine simulation environment to testing on
hardware. Using the test harness, AI Engine
graphs in simulation can run on hardware within minutes with only a few minor
modifications. See the https://xilinx.github.io/AI-Engine-Test-Harness/
AI Engine for more
details.
Warning: This chapter covers performance
analysis on hardware. This cannot be used in hardware emulation.