我想计算 Cuda 中整个图像的平均值。为了测试二维数组的缩减是如何工作的，我在下面编写了这个内核。最终输出 o 应该是所有图像值的总和。输入 g 是一个二维数组，每个像素的值为 1。但是这个程序的结果是 0 作为总和。对我来说有点奇怪。
我在本教程中模仿一维数组的减少http://developer.download.nvidia.com/compute/cuda/1.1-Beta/x86_website/projects/reduction/doc/reduction.pdf我写了这个二维表格。我是 Cuda 的新手。欢迎对潜在错误和改进提出建议！
只需添加一条评论。我知道仅计算一维数组中的平均值是有意义的。但我想利用更多并测试更复杂的归约行为。可能不对。但只是一个测试。希望任何人都可以给我更多关于减少常见做法的建议。
#include <iostream>
#include <stdio.h>
#include <stdlib.h>
#include <time.h>
cudaEvent_t start, stop;
float elapsedTime;
__global__ void
reduce(float *g, float *o, const int dimx, const int dimy)
{
extern __shared__ float sdata[];
unsigned int tid_x = threadIdx.x;
unsigned int tid_y = threadIdx.y;
unsigned int i = blockDim.x * blockIdx.x + threadIdx.x;
unsigned int j = blockDim.y * blockIdx.y + threadIdx.y;
if (i >= dimx || j >= dimy)
return;
sdata[tid_x*blockDim.y + tid_y] = g[i*dimy + j];
__syncthreads();
for(unsigned int s_y = blockDim.y/2; s_y > 0; s_y >>= 1)
{
if (tid_y < s_y)
{
sdata[tid_x * dimy + tid_y] += sdata[tid_x * dimy + tid_y + s_y];
}
__syncthreads();
}
for(unsigned int s_x = blockDim.x/2; s_x > 0; s_x >>= 1 )
{
if(tid_x < s_x)
{
sdata[tid_x * dimy] += sdata[(tid_x + s_x) * dimy];
}
__syncthreads();
}
float sum;
if( tid_x == 0 && tid_y == 0)
{
sum = sdata[0];
atomicAdd (o, sum); // The result should be the sum of all pixel values. But the program produce 0
}
//if(tid_x==0 && tid__y == 0 )
//o[blockIdx.x] = sdata[0];
}
int
main()
{
int dimx = 320;
int dimy = 160;
int num_bytes = dimx*dimy*sizeof(float);
float *d_a, *h_a, // device and host pointers
*d_o=0, *h_o=0;
h_a = (float*)malloc(num_bytes);
h_o = (float*)malloc(sizeof(float));
srand(time(NULL));
for (int i=0; i < dimx; i++)
{
for (int j=0; j < dimy; j++)
{
h_a[i*dimy + j] = 1;
}
}
cudaMalloc( (void**)&d_a, num_bytes );
cudaMalloc( (void**)&d_o, sizeof(int) );
cudaMemcpy( d_a, h_a, num_bytes, cudaMemcpyHostToDevice);
cudaMemcpy( d_o, h_o, sizeof(int), cudaMemcpyHostToDevice);
dim3 grid, block;
block.x = 4;
block.y = 4;
grid.x = dimx / block.x;
grid.y = dimy / block.y;
cudaEventCreate(&start);
cudaEventRecord(start, 0);
int sizeofSharedMemory = dimx*dimy*sizeof(float);
reduce<<<grid, block, sizeofSharedMemory>>> (d_a, d_o, block.x, block.y);
cudaEventCreate(&stop);
cudaEventRecord(stop, 0);
cudaEventSynchronize(stop);
cudaEventElapsedTime(&elapsedTime, start, stop);
std::cout << "This kernel runs: " << elapsedTime << "ms" << std::endl;
std::cout << block.x << " " << block.y << std::endl;
std::cout << grid.x << " " << grid.y << std::endl;
std::cout << dimx << " " << dimy << " " << dimx*dimy << std::endl;
cudaMemcpy( h_a, d_a, num_bytes, cudaMemcpyDeviceToHost );
cudaMemcpy( h_o, d_o, sizeof(int), cudaMemcpyDeviceToHost );
std::cout << "The sum is:" << *h_o << std::endl;
free(h_a);
free(h_o);
cudaFree(d_a);
cudaFree(d_o);
}