CUDA之矩阵乘法——globalmemory

CUDA 矩阵乘法

使用global memory

报错

• 错误 17 error : no instance of overloaded function “cudaMalloc” matches the argument list E:\Niki\MVDR_BTR\MVDR_BTR\MatrixMulOnDevice.cu 16
  修正：把CUDAcudaMalloc(&Nd, size);改成cudaMalloc((void**)&Nd, size);

• 错误 17 error : argument of type “float” is incompatible with parameter of type “void *” E:\Niki\MVDR_BTR\MVDR_BTR\MatrixMulOnDevice.cu 17
  修正：把float Md, *Nd, *Pd; 改成float Md, Nd, Pd;

编译通过的代码

#include "cuda_runtime.h"
#include "device_launch_parameters.h"

#include <stdio.h>

__global__ void MatrixMulKernel(float* Md, float* Nd, float* Pd, int Width);

void MatrixMulOnDevice(float* M, float* N, float* P, int Width)
{