pgi编译器 linux

// example.cu
#include <stdio.h>

__global__ void add(int *a, int *b, int *c) {
    int tid = threadIdx.x;
    c[tid] = a[tid] + b[tid];
}

int main() {
    int n = 10;
    int a[n], b[n], c[n];
    int *dev_a, *dev_b, *dev_c;

    // 初始化数据
    for (int i = 0; i < n; i++) {
        a[i] = i;
        b[i] = i * 2;
    }

    // 分配GPU内存
    cudaMalloc((void**)&dev_a, n * sizeof(int));
    cudaMalloc((void**)&dev_b, n * sizeof(int));
    cudaMalloc((void**)&dev_c, n * sizeof(int));

    // 将数据复制到GPU
    cudaMemcpy(dev_a, a, n * sizeof(int), cudaMemcpyHostToDevice);
    cudaMemcpy(dev_b, b, n * sizeof(int), cudaMemcpyHostToDevice);

    // 调用内核函数
    add<<<1, n>>>(dev_a, dev_b, dev_c);

    // 将结果复制回CPU
    cudaMemcpy(c, dev_c, n * sizeof(int), cudaMemcpyDeviceToHost);

    // 打印结果
    for (int i = 0; i < n; i++) {
        printf("%d + %d = %d
", a[i], b[i], c[i]);
    }

    // 释放GPU内存
    cudaFree(dev_a);
    cudaFree(dev_b);
    cudaFree(dev_c);

    return 0;
}

pgc++ -o example example.cu

./example