CUDA 二维矩阵操作

二维矩阵

代码如下：

#include <stdio.h>

#include <stdlib.h>

#include"cuda_runtime.h"

#include"device_launch_parameters.h"

#include"iostream"

using namespace std;

/************************************************************************/

/* Example:矩阵加法                                                              */

const int N =10;

__global__ void add(float m[]
, float n[]
);

void main()

{
float(*a)
= new float

;
float(*b)
= new float

;
float(*dev_a)
, (*dev_b)
;
cout << "a:" << endl;
for (int i = 0; i < N; i++)
{
for (int  j = 0; j < N; j++)
{
a[i][j] = i + j;
cout << a[i][j] << "  ";
}
cout << endl;
}
cudaMalloc((void **)&dev_a, N * N * sizeof(float));
cudaMalloc((void **)&dev_b, N * N * sizeof(float));
cudaMemcpy(dev_a, a, N * N * sizeof(float), cudaMemcpyHostToDevice);

// delete[]a;
dim3 blocks(N, N);
dim3 threads(1);
add << <blocks, threads >> >(dev_a, dev_b);
cudaMemcpy(b, dev_b, N * N * sizeof(float), cudaMemcpyDeviceToHost);
cudaFree(dev_a);
cudaFree(dev_b);
cout << "b:" << endl;
for (int i = 0; i < N; i++)
{
for (int j = 0; j < N; j++)
{
cout << b[i][j] << "  ";
}
cout << endl;
}
delete[]b;

}

__global__ void add(float m[]
, float n[]
)

{
int x = threadIdx.x + blockIdx.x * blockDim.x;
int y = threadIdx.y + blockIdx.y * blockDim.y;
{
n[x][y] = 200*m[x][y];
}

}