terrencewu

c++ 和cuda混合编程

装好CUDA 5.5 sdk后，默认会自动添加好系统环境变量。

因此不需要额外配置，不过为了保险起见，可以选择性地添加以下环境变量：

CUDA_BIN_PATH　　%CUDA_PATH%\bin
CUDA_LIB_PATH　　%CUDA_PATH%\lib\Win32
CUDA_SDK_BIN　　%CUDA_SDK_PATH%\bin\Win32
CUDA_SDK_LIB　　%CUDA_SDK_PATH%\common\lib\Win32
CUDA_SDK_PATH　　C:\cuda\cudasdk\common

这时可以打开CUDA自带的sample运行一下，运行能通过才可以继续下面的内容————cpp和cuda联调。

方法一：先建立cuda工程，再添加cpp文件

1.打开vs2010，新建一个cuda项目，名称CudaCpp。

2.cuda默认建立的工程是如下，实现了两个一维向量的并行相加。kernel函数和执行函数还有main函数全都写在了一个cu文件里。

3.接下来在工程里添加一个空的cpp文件。将原来cu文件里main函数里的内容剪切到cpp文件main函数里。

为了让cpp能够调用cu文件里面的函数，在addWithCuda函数前加上extern "C"关键字（注意C大写，为什么addKernel不用加呢？因为cpp里面直接调用的是addWithCuda）

4.在cpp里也要加上addWithCuda函数的完整前向声明。下图就是工程的完整结构

5.可以在cpp里的main函数return之间加入getchar()防止运行后一闪就退出，加上system("pause")或者直接ctrl+F5也行。

运行结果：

下面贴出CudaCpp项目代码。

kernel.cu

[plain] view plain copy

#include "cuda_runtime.h"
#include "device_launch_parameters.h"
#include
__global__ void addKernel(int *c, const int *a, const int *b)
{
int i = threadIdx.x;
c[i] = a[i] + b[i];
}
// Helper function for using CUDA to add vectors in parallel.
extern "C"
cudaError_t addWithCuda(int *c, const int *a, const int *b, unsigned int size)
{
int *dev_a = 0;
int *dev_b = 0;
int *dev_c = 0;
cudaError_t cudaStatus;
// Choose which GPU to run on, change this on a multi-GPU system.
cudaStatus = cudaSetDevice(0);
if (cudaStatus != cudaSuccess) {
fprintf(stderr, "cudaSetDevice failed! Do you have a CUDA-capable GPU installed?");
goto Error;
}
// Allocate GPU buffers for three vectors (two input, one output) .
cudaStatus = cudaMalloc((void**)&dev_c, size * sizeof(int));
if (cudaStatus != cudaSuccess) {
fprintf(stderr, "cudaMalloc failed!");
goto Error;
}
cudaStatus = cudaMalloc((void**)&dev_a, size * sizeof(int));
if (cudaStatus != cudaSuccess) {
fprintf(stderr, "cudaMalloc failed!");
goto Error;
}
cudaStatus = cudaMalloc((void**)&dev_b, size * sizeof(int));
if (cudaStatus != cudaSuccess) {
fprintf(stderr, "cudaMalloc failed!");
goto Error;
}
// Copy input vectors from host memory to GPU buffers.
cudaStatus = cudaMemcpy(dev_a, a, size * sizeof(int), cudaMemcpyHostToDevice);
if (cudaStatus != cudaSuccess) {
fprintf(stderr, "cudaMemcpy failed!");
goto Error;
}
cudaStatus = cudaMemcpy(dev_b, b, size * sizeof(int), cudaMemcpyHostToDevice);
if (cudaStatus != cudaSuccess) {
fprintf(stderr, "cudaMemcpy failed!");
goto Error;
}
// Launch a kernel on the GPU with one thread for each element.
addKernel<<<1, size>>>(dev_c, dev_a, dev_b);
// Check for any errors launching the kernel
cudaStatus = cudaGetLastError();
if (cudaStatus != cudaSuccess) {
fprintf(stderr, "addKernel launch failed: %s\n", cudaGetErrorString(cudaStatus));
goto Error;
}
// cudaDeviceSynchronize waits for the kernel to finish, and returns
// any errors encountered during the launch.
cudaStatus = cudaDeviceSynchronize();
if (cudaStatus != cudaSuccess) {
fprintf(stderr, "cudaDeviceSynchronize returned error code %d after launching addKernel!\n", cudaStatus);
goto Error;
}
// Copy output vector from GPU buffer to host memory.
cudaStatus = cudaMemcpy(c, dev_c, size * sizeof(int), cudaMemcpyDeviceToHost);
if (cudaStatus != cudaSuccess) {
fprintf(stderr, "cudaMemcpy failed!");
goto Error;
}
Error:
cudaFree(dev_c);
cudaFree(dev_a);
cudaFree(dev_b);
return cudaStatus;
}

#include "cuda_runtime.h"
#include "device_launch_parameters.h"

#include 

__global__ void addKernel(int *c, const int *a, const int *b)
{
    int i = threadIdx.x;
    c[i] = a[i] + b[i];
}
// Helper function for using CUDA to add vectors in parallel.
extern "C"
cudaError_t addWithCuda(int *c, const int *a, const int *b, unsigned int size)
{
    int *dev_a = 0;
    int *dev_b = 0;
    int *dev_c = 0;
    cudaError_t cudaStatus;

    // Choose which GPU to run on, change this on a multi-GPU system.
    cudaStatus = cudaSetDevice(0);
    if (cudaStatus != cudaSuccess) {
        fprintf(stderr, "cudaSetDevice failed!  Do you have a CUDA-capable GPU installed?");
        goto Error;
    }

    // Allocate GPU buffers for three vectors (two input, one output)    .
    cudaStatus = cudaMalloc((void**)&dev_c, size * sizeof(int));
    if (cudaStatus != cudaSuccess) {
        fprintf(stderr, "cudaMalloc failed!");
        goto Error;
    }

    cudaStatus = cudaMalloc((void**)&dev_a, size * sizeof(int));
    if (cudaStatus != cudaSuccess) {
        fprintf(stderr, "cudaMalloc failed!");
        goto Error;
    }

    cudaStatus = cudaMalloc((void**)&dev_b, size * sizeof(int));
    if (cudaStatus != cudaSuccess) {
        fprintf(stderr, "cudaMalloc failed!");
        goto Error;
    }

    // Copy input vectors from host memory to GPU buffers.
    cudaStatus = cudaMemcpy(dev_a, a, size * sizeof(int), cudaMemcpyHostToDevice);
    if (cudaStatus != cudaSuccess) {
        fprintf(stderr, "cudaMemcpy failed!");
        goto Error;
    }

    cudaStatus = cudaMemcpy(dev_b, b, size * sizeof(int), cudaMemcpyHostToDevice);
    if (cudaStatus != cudaSuccess) {
        fprintf(stderr, "cudaMemcpy failed!");
        goto Error;
    }

    // Launch a kernel on the GPU with one thread for each element.
    addKernel<<<1, size>>>(dev_c, dev_a, dev_b);

    // Check for any errors launching the kernel
    cudaStatus = cudaGetLastError();
    if (cudaStatus != cudaSuccess) {
        fprintf(stderr, "addKernel launch failed: %s\n", cudaGetErrorString(cudaStatus));
        goto Error;
    }
    
    // cudaDeviceSynchronize waits for the kernel to finish, and returns
    // any errors encountered during the launch.
    cudaStatus = cudaDeviceSynchronize();
    if (cudaStatus != cudaSuccess) {
        fprintf(stderr, "cudaDeviceSynchronize returned error code %d after launching addKernel!\n", cudaStatus);
        goto Error;
    }

    // Copy output vector from GPU buffer to host memory.
    cudaStatus = cudaMemcpy(c, dev_c, size * sizeof(int), cudaMemcpyDeviceToHost);
    if (cudaStatus != cudaSuccess) {
        fprintf(stderr, "cudaMemcpy failed!");
        goto Error;
    }

Error:
    cudaFree(dev_c);
    cudaFree(dev_a);
    cudaFree(dev_b);
    
    return cudaStatus;
}

main.cpp

[cpp] view plain copy

#include
#include "cuda_runtime.h"
#include "device_launch_parameters.h"
extern "C"
cudaError_t addWithCuda(int *c, const int *a, const int *b, unsigned int size);
int main()
{
const int arraySize = 5;
const int a[arraySize] = { 1, 2, 3, 4, 5 };
const int b[arraySize] = { 10, 20, 30, 40, 50 };
int c[arraySize] = { 0 };
// Add vectors in parallel.
cudaError_t cudaStatus = addWithCuda(c, a, b, arraySize);
if (cudaStatus != cudaSuccess) {
fprintf(stderr, "addWithCuda failed!");
return 1;
}
printf("{1,2,3,4,5} + {10,20,30,40,50} = {%d,%d,%d,%d,%d}\n",
c[0], c[1], c[2], c[3], c[4]);
printf("cuda工程中调用cpp成功！\n");
// cudaDeviceReset must be called before exiting in order for profiling and
// tracing tools such as Nsight and Visual Profiler to show complete traces.
cudaStatus = cudaDeviceReset();
if (cudaStatus != cudaSuccess) {
fprintf(stderr, "cudaDeviceReset failed!");
return 1;
}
getchar(); //here we want the console to hold for a while
return 0;
}

#include
#include "cuda_runtime.h"
#include "device_launch_parameters.h"
extern "C"
cudaError_t addWithCuda(int *c, const int *a, const int *b, unsigned int size);
int main()
{
const int arraySize = 5;
const int a[arraySize] = { 1, 2, 3, 4, 5 };
const int b[arraySize] = { 10, 20, 30, 40, 50 };
int c[arraySize] = { 0 };
// Add vectors in parallel.
cudaError_t cudaStatus = addWithCuda(c, a, b, arraySize);
if (cudaStatus != cudaSuccess) {
fprintf(stderr, "addWithCuda failed!");
return 1;
}
printf("{1,2,3,4,5} + {10,20,30,40,50} = {%d,%d,%d,%d,%d}\n",
c[0], c[1], c[2], c[3], c[4]);
printf("cuda工程中调用cpp成功！\n");
// cudaDeviceReset must be called before exiting in order for profiling and
// tracing tools such as Nsight and Visual Profiler to show complete traces.
cudaStatus = cudaDeviceReset();
if (cudaStatus != cudaSuccess) {
fprintf(stderr, "cudaDeviceReset failed!");
return 1;
}
getchar(); //here we want the console to hold for a while
return 0;
}

方法二：先建立cpp工程，再添加cu文件

方法一由于是cuda工程是自动建立的，所以比较简单，不需要多少额外的配置。而在cpp工程里面添加cu就要复杂一些。为了简单起见，这里采用console程序讲解，至于MFC或者Direct3D程序同理。

1.建立一个空的win32控制台工程，名称CppCuda。

2.然后右键工程-->添加一个cu文件

3.将方法一中cu和cpp文件的代码分别拷贝到这个工程里来（做了少许修改，extern "C"关键字和某些头文件不要忘了加），工程结构如图：

这个时候编译是通不过的，需要作一些配置。

4.关键的一步，右键工程-->生成自定义，将对话框中CUDA5.5前面的勾打上。

这时点击工程-->属性，会发现多了CUDA链接器这一项。

5.关键的一步，右键kernel.cu文件-->属性，在常规-->项类型里面选择CUDA C/C++（由于cu文件是由nvcc编译的，这里要修改编译链接属性）

6.工程-->属性-->链接器-->附加依赖项，加入cudart.lib

7.工具-->选项-->文本编辑器-->文件扩展名添加cu \cuh两个文件扩展名

8.至此配置成功。运行一下：

9.为了更加确信cuda中的函数确实被调用，在main.cpp里面调用cuda函数的地方加入了一个断点。

单步执行一下。

可以看到程序跳到了cu文件里去执行了，说明cpp调用cuda函数成功。

贴上代码（其实跟方式一基本一样，没怎么改），工程CppCuda