error: a host function call cannot be configured CUDA Linux

写一个入门的vec　add。

出了这么个错误。之前是没有的。现在有了，想想找了，结果这类错误好像还不多。

一定又是个低级的错误。记录下来，让那些和我一样(对，就是你，看这个帖子的你)下次写程序时小心些。

__global__　　　没写。

error: a host function call cannot be configured

# include <stdio.h>
# include <stdlib.h>
# include <string.h>
# include <time.h>
# include <math.h>
# include <cuda.h>
#include <cuda_runtime.h>

void sumArraysOnHost(float *A, float *B, float *C, const int N)
{
	for ( int idx = 0; idx < N; idx++ )
	{
		C[idx] = A[idx] + B[idx];
	}
}

void sumArrayOnDevice(float *A, float *B, float *C) //error version
//__global__ void sumArrayOnDevice(float *A, float *B, float *C)　// right version

{
	C[blockIdx.x] = A[blockIdx.x] + B[blockIdx.x];
}

void initialData(float *ip, int size)
{
//	time_t t;
//	srand((unsigned int) time(&t));

	for (int i = 0; i< size; i++ )
	{
		//ip[i] = (float)(rand() & 0XFF)/10.0f;
		ip[i] = (float)(rand() & 0XFF)/10.0f;
	}
}


int main(int argc, char ** argv)
{
	int nElem = 12;
	size_t nBytes = nElem * sizeof(float);

	float *h_A, *h_B, *h_C, *h_C_host;
	h_A = (float *)malloc(nBytes);
	h_B = (float *)malloc(nBytes);
	h_C = (float *)malloc(nBytes);
	h_C_host = (float *)malloc(nBytes);

	initialData(h_A, nElem);	
	initialData(h_B, nElem);
    initialData(h_C,nElem);
    initialData(h_C_host,nElem);

	sumArraysOnHost(h_A,h_B,h_C,nElem);

	float *d_A, *d_B, *d_C;
	cudaMalloc((float **)&d_A, nBytes);
	cudaMalloc((float **)&d_B, nBytes);
	cudaMalloc((float **)&d_C, nBytes);

	cudaMemcpy(d_A,h_A,nBytes,cudaMemcpyHostToDevice);
	cudaMemcpy(d_B,h_B,nBytes,cudaMemcpyHostToDevice);

	sumArrayOnDevice<<<12,1>>>(d_A,d_B,d_C);
	
	cudaMemcpy(h_C_host,d_C,nBytes,cudaMemcpyDeviceToHost);

	for (int i=0;i<nElem;i++)
	{
		printf("A[%d] = %5.2f   B[%d]= %5.2f C[%d]= %5.2f C_dev[%d]=%5.2f \n",\
		i,h_A[i],i,h_B[i],i,h_C[i],i,h_C_host[i]);
	}

	free(h_A);
	free(h_B);
	free(h_C);
	free(h_C_host);

	cudaFree(d_A);
	cudaFree(d_B);
	cudaFree(d_C);

	return(0);
}