请问为什么我的程序结果都是0呢？

[over_sky]

我的程序目的就是分别计算一个数与其他数之间的距离，然后将符合一定条件的距离点的位置存储起来。输入数据目前是300*3.但是输出的数据都是0，应该是我的核函数没有正确执行，但是我自己找不到错误在哪？还有就是使用atomicAdd和__syncthreads()需要头文件吗？

#include "cuda_runtime.h"
#include "device_launch_parameters.h"
#include <iostream>
#include <fstream>
#include <thrust/device_vector.h>
#include <thrust/host_vector.h>
#include <device_functions.h>

using namespace std;
const int lineNum = 300;
const int threadsPerBlock = 64;
const int blocksPerGrid = (lineNum + threadsPerBlock - 1) % threadsPerBlock;
const int neighborMaxNum = 300;
const float R = 0.03;

__global__ void calDistance(int num, float standard_x, float standard_y, float standard_z, float *x_dev, float *y_dev, float *z_dev, int* result, int line, int width, float R)
{
	int tid = blockIdx.x * blockDim.x + threadIdx.x;
	int count = 0;
	if (tid < line)
	{
		float dist = pow(x_dev[tid] - standard_x, 2) + pow(y_dev[tid] - standard_y, 2) + pow(z_dev[tid] - standard_z, 2);
		if ((dist < pow(R, 2)) && (count<width))
		{
			result[num*neighborMaxNum+count] = tid;
			atomicAdd(&count, 1);
			//cout << count << endl;
			//tid += blockDim.x*blockIdx.x;
			//__syncthreads();
		}
	}
}

int main()
{
	cout << "begin" << endl;
	ifstream input("test.txt");
	float *x_Host=new float[lineNum];
	float *y_Host=new float[lineNum];
	float *z_Host=new float[lineNum];
	//int *result_Host=[lineNum][neighborMaxNum] = {};
	/*int **result_Host;
	result_Host = new int *[lineNum];
	for (int i = 0; i < lineNum; i++)
	{
		result_Host[i] = new int[neighborMaxNum];
	}
	for (int m = 0; m < lineNum; m++)
		for (int n = 0; n < neighborMaxNum; n++)
		{
		result_Host[m][n] = -1;
		}*/
	int *result_Host = new int[lineNum*neighborMaxNum];
	for (int i = 0; i < lineNum*neighborMaxNum; i++)
	{
		result_Host[i] = -1;
	}
	ofstream output("result2.txt");
	for (int j = 0; j < lineNum; j++)
	{
		for (int k = 0; k < neighborMaxNum; k++)
		{
			output << result_Host[j*neighborMaxNum + k] << " ";
		}
		output << endl;
	}
	output.close();
	float *x_dev, *y_dev, *z_dev;
	int *result_device;
	//thrust::host_vector<thrust::host_vector> result_host(lineNum);
	//thrust::device_vector<thrust::device_vector> result_device(lineNum);

	cudaMalloc((void**)&x_dev, lineNum*sizeof(float));
	cudaMalloc((void**)&y_dev, lineNum*sizeof(float));
	cudaMalloc((void**)&z_dev, lineNum*sizeof(float));
	cudaMalloc((void**)&result_device, sizeof(int)*lineNum*neighborMaxNum);

	for (int i = 0; i<lineNum; i++)
	{
		input >> x_Host[i] >> y_Host[i] >> z_Host[i];
	}
	input.close();

	cudaMemcpy(x_dev, x_Host, lineNum*sizeof(float), cudaMemcpyHostToDevice);
	cudaMemcpy(y_dev, y_Host, lineNum*sizeof(float), cudaMemcpyHostToDevice);
	cudaMemcpy(z_dev, z_Host, lineNum*sizeof(float), cudaMemcpyHostToDevice);
	float time;
	cudaEvent_t start, stop;
	cudaEventCreate(&start);
	cudaEventCreate(&stop);
	cudaEventRecord(start, 0);
	for (int i = 0; i<lineNum; i++)
	{
		calDistance <<<blocksPerGrid, threadsPerBlock >>>(i, x_dev[i], y_dev[i], z_dev[i], x_dev, y_dev, z_dev, result_device, lineNum, neighborMaxNum, R);
		cudaThreadSynchronize();
		//cudaError_t error = cudaGetLastError();
		//cout << cudaGetErrorString(error) << endl;
	}
	cout << "kernel" << endl;
	cudaEventRecord(stop, 0);
	cudaEventSynchronize(stop);
	cudaEventElapsedTime(&time, start, stop);
	cout << "execute time " << time << endl;
	//cudaMemcpy2D(result_Host, neighborMaxNum*sizeof(int), result_device, neighborMaxNum*sizeof(int), neighborMaxNum*sizeof(int), lineNum, cudaMemcpyDeviceToHost);
	cudaMemcpy(result_Host, result_device, lineNum*neighborMaxNum*sizeof(int), cudaMemcpyDeviceToHost);
	cout << "here1" << endl;
	cudaFree(x_dev);
	cudaFree(y_dev);
	cudaFree(z_dev);
	cudaFree(result_device);
	//cudaFree(result_device);

	//for (int j = 0; j<lineNum; j++)
	//	for (int k = 0; k<neighborMaxNum; k++)
	//	{
	//	if (result_Host[j][k] != 0)
	//	{
	//		cout << result_Host[j][k];
	//	}
	//	}
	ofstream output2("result.txt");
	for (int j = 0; j < lineNum; j++)
	{
		for (int k = 0; k < neighborMaxNum; k++)
		{
			output2 << result_Host[j*neighborMaxNum+k] << " ";
		}
		output2 << endl;
	}
	output2.close();
	delete[] x_Host;
	delete[] y_Host;
	delete[] z_Host;
	//for (int i = 0; i < lineNum; i++)
	//{
	//	delete[] result_Host[i];
	//}
	delete[] result_Host;
	cout << "finished" << endl;
	return 0;

}


//#include<iostream>
//#include<fstream>
//
//using namespace std;
//const int lineNum = 1007428;
//int main()
//{
//	cout << "begin" << endl;
//	ifstream input("model_test.txt");
//	float x_Host[lineNum] = {};
//	float y_Host[lineNum] = {};
//	float z_Host[lineNum] = {};
//	//thrust::host_vector<thrust::host_vector> result_host(lineNum);
//	//thrust::device_vector<thrust::device_vector> result_device(lineNum);
//
//
//	for (int i = 0; i < lineNum; i++)
//	{
//		input >> x_Host[i] >> y_Host[i] >> z_Host[i];
//	}
//	input.close();
//	return 0;
//}