Compiling and Running the Graph from the Command Line - 2025.2 English - UG1603

v++ -c --mode aie --target hw --config ./config.cfg project.cpp

The program is named project.cpp. The AI Engine compiler reads the input graph, compiles it to the AI Engine-ML array, produces reports, and generates output files in the Work directory.

The input graph is first partitioned into groups of kernels to be mapped to the same core.

You can view the output of the mapper as a tabular report in the file project_mapping_analysis_report.txt. This file reports the mapping of nodes to processing cores and data buffers to memory banks. Inter-processor communication is appropriately double-banked as ping-pong buffers.

The AI Engine compiler allocates the necessary locks, memory buffers, DMA channels, descriptors. The compiler also generates routing information to map the graph onto the AI Engine-ML array.

The compiler synthesizes a main program for each core. The program schedules all the kernels on the cores, and implements the necessary locking mechanism and data copy among buffers. The C/C++ program for each core is compiled to produce loadable ELF files. The AI Engine compiler also generates control APIs to control the graph initialization, execution, and termination from the main application and a simulator configuration script scsim_config.json. These are all stored within the Work directory under various sub-folders. See Compiling an AI Engine Graph Application in AI Engine Tools and Flows User Guide (UG1076) for more details.

To open the AI Engine compiler summary, use the following command:

vitis -a ./Work/graph.aiecompile_summary

aiesimulator –-pkg-dir=./Work

This command starts the SystemC-based simulator with the control program being the main application. The graph APIs which are used in the control program configure the AI Engine-ML array. Configuration includes setting up static routing, programming the DMAs, loading the ELF files onto the individual cores, and initiating AI Engine-ML array execution.

At the end of the simulation, the output data is produced in the directory aiesimulator_output and it should match the reference data.

You can load the graph at device boot time in hardware, or through the host application. Building and Running the System in the Data Center Acceleration using Vitis (UG1700) describes deploying the graph in hardware and the flow associated with it.