AI Engines are an array of very-long
instruction word (VLIW) processors with single instruction multiple data (SIMD) vector
units. An AI Engine kernel is a C/C++ program which
is written using specialized APIs that target the VLIW vector processor. The AI Engine kernel code is compiled using the v++ -mode aie command that is used by the AMD Vitis™
IDE to produce an ELF file that is run on the AI Engine processors.
An AI Engine component consists of an adaptive data flow (ADF) graph application written in C++, containing nodes and edges where nodes represent compute kernel functions, and edges represent data connections. Kernels in the application can be compiled to run on the AI Engines or in the PL region of the device as described in the AI Engine Kernel and Graph Programming Guide (UG1079).
AI Engine components can be built, analyzed, and debugged as a standalone component, and then be added to an System project as part of an embedded system design, or for application acceleration in a data center. The following sections detail building and analyzing the standalone AI Engine component.