#include <stdio.h>

// __global__ indicates this function runs on the GPU and is called by the CPU

__global__ void cudaHello() {

    printf("Hello from the GPU!\n");

}

int main() {

    // Launch the kernel with 1 block and 1 thread

    cudaHello<<<1, 1>>>();

    // Wait for the GPU to finish before exiting the program

    cudaDeviceSynchronize();

    printf("Hello from the CPU!\n");

    return 0;

}

on console

Running in FUNCTIONAL mode...

Compiling...

Executing...

Hello from the GPU!

Hello from the CPU!

Exit status: 0