请教CUDA启动返回错误码4问题

加速计算 CUDA
1

使用1080 GPU，VS2015
简化源码，如下

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

global void TestKernel()
{
/*************** thread shared para setting *************************/
for (int j = 0; j < 600; ++j) {
printf(“%d”,j);
}
}

int main()
{
cudaError_t cudaStatus = cudaSuccess;

TestKernel << < 20 , 256 >> >();

/***** Check for any errors launching the kernel*******/
cudaStatus = cudaGetLastError();
if (cudaStatus != cudaSuccess) {
fprintf(stderr, “\n\nTestKernel launch failed: %s\n”, cudaGetErrorString(cudaStatus));
return cudaStatus;
}

cudaStatus = cudaDeviceSynchronize();
if (cudaStatus != cudaSuccess) {
fprintf(stderr, “\n\ncudaDeviceSynchronize returned error code %d after launching TestKernel!\n”, cudaStatus);
return cudaStatus;
}

printf(“end\n”);
getchar();
return cudaStatus;
}

执行后报错返回错误码4，请问是什么问题？