CUDA之矩阵乘法——非方阵计算

说明

A矩阵为M * N，B矩阵为N * M

代码

#include "device_functions.h"
#include "cuda_runtime.h"
#include "device_launch_parameters.h"
#include <stdio.h>
#include <iostream>

typedef struct {
    int width;
    int height;
    int stride;
    float* elements;
} Matrix;

#define BLOCK_SIZE  2
#define N           4
#define M           8

__device__ float GetElement(const Matrix A, int row, int col) {
    return A.elements[row * A.stride + col];
}

__device__ void SetElement(Matrix A, int row, int col, float value) {
    A.elements[row * A.stride + col] = value;
}

__device__ Matrix GetSubMatrix(Matrix A,