The AI Engine matrix multiplication example demonstrates how to use AI Engine for scalar computation and how to use GMIO attributes for data movement. This example implements the standard matrix multiplication algorithm to multiply two matrixes on the AI Engine cores. You can change the matrix size and the number of cores used at compile-time. The default size is 1200 × 1200 (int32). The matrix can be further scaled to use the entire AI Engine's local data memory.
Note: This sample AI Engine illustration serves as a proof-of-concept included in
the default AMD Versal™
Linux image.