2025最新版使用VSCode和CMake图形化编译调试Cuda C++程序(保姆级教学)

VSCode +CMake 简直就是C++程序开发者的福音，无论是进行Qt开发，音视频开发，后端程序开发，Cuda开发，CMake都是神一样的存在。

首先确保你已经安装好了Cuda Toolkit，比如我的是Cuda 12.3。

确保你已经配置了环境变量：

VSCode下载好插件：

CMakeLists.txt如下：

cpp 复制代码

cmake_minimum_required(VERSION 3.18)
project(CudaTest LANGUAGES CXX CUDA)
# 设置CUDA标准
set(CMAKE_CUDA_STANDARD 17)
set(CMAKE_CUDA_STANDARD_REQUIRED ON)
#设置CUDA架构
set(CMAKE_CUDA_ARCHITECTURES "75")
# 添加CUDA头文件
include_directories("$ENV{CUDA_PATH}/include")
# 添加可执行文件
add_executable(cuda_test  cuda_kernel.cu)

main.cu如下：

cpp 复制代码

#include <stdio.h>
#include <cuda_runtime.h>

// CUDA核函数：执行向量加法
__global__ void vectorAdd(const float *a, const float *b, float *c, int n) {
    int idx = blockIdx.x * blockDim.x + threadIdx.x;
    if (idx < n) {
        c[idx] = a[idx] + b[idx];
    }
}
// 验证GPU计算结果
void verifyResult(float *a, float *b, float *c, int n) {
    for (int i = 0; i < n; i++) {
        if (fabs(a[i] + b[i] - c[i]) > 1e-5) {
            printf("计算结果错误！\n");
            return;
        }
    }
    printf("计算结果正确！\n");
}
int main() {
    const int n = 1000;
    const int size = n * sizeof(float);
    float *h_a = (float*)malloc(size);
    float *h_b = (float*)malloc(size);
    float *h_c = (float*)malloc(size);
    for (int i = 0; i < n; i++) {
        h_a[i] = rand() / (float)RAND_MAX;
        h_b[i] = rand() / (float)RAND_MAX;
    }
    float *d_a, *d_b, *d_c;
    cudaMalloc(&d_a, size);
    cudaMalloc(&d_b, size);
    cudaMalloc(&d_c, size);
    cudaMemcpy(d_a, h_a, size, cudaMemcpyHostToDevice);
    cudaMemcpy(d_b, h_b, size, cudaMemcpyHostToDevice);
    int threadsPerBlock = 256;
    int blocksPerGrid = (n + threadsPerBlock - 1) / threadsPerBlock;
    vectorAdd<<<blocksPerGrid, threadsPerBlock>>>(d_a, d_b, d_c, n);
    cudaMemcpy(h_c, d_c, size, cudaMemcpyDeviceToHost);
    verifyResult(h_a, h_b, h_c, n);
    cudaFree(d_a);
    cudaFree(d_b);
    cudaFree(d_c);
    free(h_a);
    free(h_b);
    free(h_c);
    return 0;
}