CUDA学习笔记05：矩阵相乘

参考资料

CUDA编程模型系列三(矩阵乘)_哔哩哔哩_bilibili

极力推荐的良心课程~

代码片段

#include 
 
const int matrix_size = 32;
const int BLOCK_SIZE = 16;
 
__global__ void matrix_multi(const int* a1, const int *b1, int *c1, const int size)
{
    int y = blockDim.y * blockIdx.y + threadIdx.y;
    int x = blockDim.x * blockIdx.x + threadIdx.x;
 
    int t_sum = 0;
    if(x < size && y < size) {
        for(int step = 0; step < size; step++) {
            t_sum += a1[y * size + step] * b1[step * size + x];
        }
        c1[y * size + x] = t_sum;
        printf("x: %d  y: %d  sum: %d \n", x, y, t_sum);
    }
}
 
int main()
{
    int* a1, *b1, *c1;
    int msize = matrix_size * matrix_size * sizeof(int);
 
    // 分配内存空间
    cudaMallocHost((void**)&a1, msize);
    cudaMallocHost((void**)&b1, msize);
    cudaMallocHost((void**)&c1, msize);
 
    // 初始化
    for(int i = 0; i < matrix_size; i++) {
        for(int j = 0; j < matrix_size; j++) {
            a1[i * matrix_size + j] = 1;
            b1[i * matrix_size + j] = 1;
        }
    }
 
    printf("Check result a1: %d \n", a1[1]);
    printf("Check result b1: %d \n", b1[1]);
 
    // 分配设备空间
    int *x1, *y1, *z1;
    cudaMalloc((void**)&x1, msize);
    cudaMalloc((void**)&y1, msize);
    cudaMalloc((void**)&z1, msize); 
 
    // Host => Device
    cudaMemcpy(x1, a1, msize, cudaMemcpyHostToDevice);
    cudaMemcpy(y1, b1, msize, cudaMemcpyHostToDevice);
    // printf("Check result: %d \n", x1[1]);
    // printf("Check result: %d \n", y1[1]);
 
    int grid_rows = (matrix_size + BLOCK_SIZE - 1) / BLOCK_SIZE;
    int grid_cols = (matrix_size + BLOCK_SIZE - 1) / BLOCK_SIZE;
 
    dim3 gridDim(grid_cols, grid_rows);
    dim3 blockDim(BLOCK_SIZE, BLOCK_SIZE);
 
    // 计算
    matrix_multi<<>>(x1, y1, z1, matrix_size);
 
    // Device => Host
    cudaMemcpy(c1, z1, msize, cudaMemcpyDeviceToHost);
 
    printf("Check result c1: %d \n", c1[1]);
 
    // 释放空间
    cudaFreeHost(a1);
    cudaFreeHost(b1);
    cudaFreeHost(c1);
    cudaFree(x1);
    cudaFree(y1);
    cudaFree(z1);
    return 0;
}