NVIDIA OpenCL实现C++ API问题

[ErrorErro]

NVIDIA CUDA Toolkit 6.0中没有提供C+ API（cl.hpp），所以我从khronos下载了一个OpenCL 1.1的cl.hpp
然后，如下代码

#include <iostream>
#include <CL\cl.hpp>
#include <vector>
#include <string>


using namespace std;

int main()
{	
std::vector<cl::Platform> platformList;
	cl::Platform::get(&platformList);
	for (int i = 0; i < platformList.size(); ++i)
	{
		std::string cppname;
		std::string cppversion;
		std::string cppverndor;
		std::string cppprofile;
	 	platformList[i].getInfo(CL_PLATFORM_NAME, &cppname);
	 	platformList[i].getInfo(CL_PLATFORM_VERSION, &cppversion);
	  	platformList[i].getInfo(CL_PLATFORM_VENDOR, &cppverndor);
    	platformList[i].getInfo(CL_PLATFORM_PROFILE, &cppprofile);
		cout << "OpenCL平台：" << cppname << "\n"
			<< "OpenCL版本：" << cppversion << "\n"
			<< "OpenCL开发商：" << cppverndor << "\n"
			<< "OpenCL简档：" << cppprofile << "\n" << "\n" << endl;
	}
	cin.get();
	return 0;
}

编译错误：错误 1 error C2664: “cl_int clEnqueueNativeKernel(cl_command_queue,void (__stdcall *)(void *),void *,size_t,cl_uint,const cl_mem *,const void **,cl_uint,const cl_event *,cl_event *)”: 无法将参数 2 从“void (__cdecl *)(void *)”转换为“void (__stdcall *)(void *)” c:\program files\nvidia gpu computing toolkit\cuda\v6.0\include\cl\cl.hpp 2996 1 OpenCLExample
错误指向的是cl.hpp中的一个函数，这是么情况？NVIDIA 的OpenCL事先到底支持不支持C++ API？如何解决