Prepare the Kernels - 2022.2 English

AI Engine Kernel and Graph Programming Guide (UG1079)

Document ID
Release Date
2022.2 English

Kernels are computation functions that form the fundamental building blocks of the data flow graph specifications. Kernels are declared as ordinary C/C++ functions that return void and can use special data types as arguments (discussed in Window and Streaming Data API). Each kernel must be defined in its own source file. This organization is recommended for reusability and faster compilation. Furthermore, the kernel source files should include all relevant header files to allow for independent compilation.

Note: For the AI Engine kernel to use AI Engine API, include the following files in the kernel source code:
  • #include "aie_api/aie.hpp"
  • #include "aie_api/aie_adf.hpp"

It is recommended that a header file (kernels.h in this documentation) should declare the function prototypes for all kernels used in a graph. An example is as follows.


void simple(input_window<cint16> * in, output_window<cint16> * out);


In the example, the #ifndef and #endif are present to ensure that the include file is only included once, which is good C/C++ practice.

The is the implementation file for the simple function. The kernel implementation declares two complex int 16 variables c1 and c2. It then reads the complex int16 input data into c1 using the window_readincr API, adds the real and imaginary values and assigns it back to c2.real. It also subtracts the imaginary value from the real value and assign it to c2.imag. Finally c2 is written back to the output window using the window_writeincr API.

/* A simple kernel */
#include <adf.h>
#include "include.h"

void simple(input_window<cint16> * in, output_window<cint16> * out) {
  cint16 c1, c2;

  for (unsigned i=0; i<NUM_SAMPLES; i++) {
    c1 = window_readincr(in);
    c2.real = c1.real+c1.imag;
    c2.imag = c1.real-c1.imag;
    window_writeincr(out, c2);