将 Thrust 设备迭代器转换为原始指针

Question

我正在考虑以下简单代码，其中我将thrust::host_vector<int>::iterator h_temp_iterator = h_temp.begin();和转换thrust::device_vector<int>::iterator d_temp_iterator = d_temp.begin();为原始指针。

为此，我分别将&(h_temp_iterator[0])和传递&(d_temp_iterator[0])给函数和内核。前者（CPU 情况）可以编译，后者（GPU 情况）则不能。这两种情况原则上应该是对称的，所以我不明白错误消息的原因是：

Error   1   error : no suitable conversion function from "thrust::device_ptr<int>" to "int *" exists    

配置是：

1. Windows 7, Visual Studio 2010, CUDA 7.5, 编译架构3.5。
2. Windows 10, Visual Studio 2013, CUDA 8.0, 编译架构5.2。

代码

#include <thrust\host_vector.h>
#include <thrust\device_vector.h>

__global__ void testKernel(int *a, const int N)
{
    int i = threadIdx.x;

    if (i >= N) return;

    a[i] = 2;
}

void testFunction(int *a, const int N)
{
    for (int i = 0; i < N; i++) a[i] = 2;
}

int main()
{
    const int N = 10;

    thrust::host_vector<int> h_temp(N);
    thrust::device_vector<int> d_temp(N);

    thrust::host_vector<int>::iterator h_temp_iterator = h_temp.begin();
    thrust::device_vector<int>::iterator d_temp_iterator = d_temp.begin();

    testFunction(&(h_temp_iterator[0]), N);
    testKernel<<<1, N>>>(&(d_temp_iterator[0]), N);

    for (int i = 0; i < N; i++) printf("%i %i\n", i, h_temp[i]);

    return 0;
}

Answer 1

根据talonmies的评论，解决方案是通过

thrust::raw_pointer_cast(&d_temp_iterator[0])

并不是

&d_temp_iterator[0]

下面是完整的工作代码

#include <thrust\host_vector.h>
#include <thrust\device_vector.h>

__global__ void testKernel(int *a, const int N)
{
    int i = threadIdx.x;

    if (i >= N) return;

    a[i] = 2;

    printf("GPU %i %i\n", i, a[i]);
}

void testFunction(int *a, const int N)
{
    for (int i = 0; i < N; i++) {
        a[i] = 2;
        printf("CPU %i %i\n", i, a[i]);
    }
}

int main()
{
    const int N = 10;

    thrust::host_vector<int> h_temp(N);
    thrust::device_vector<int> d_temp(N);

    thrust::host_vector<int>::iterator h_temp_iterator = h_temp.begin();
    thrust::device_vector<int>::iterator d_temp_iterator = d_temp.begin();

    int *temp = thrust::raw_pointer_cast(&d_temp_iterator[0]);

    testFunction(&(h_temp_iterator[0]), N);
    testKernel<<<1, N>>>(temp, N);

    return 0;
}