CUDA MATRIX MULTIPLICATION PROGRAM

#include<stdio.h>
#include<stdlib.h>
#define SIZE 3
int matrixA[SIZE][SIZE],matrixB[SIZE][SIZE],matrixC[SIZE][SIZE];

__global__ void matrix_mult(int *dA,int *dB,int *dC,int dim)
{
	//	int e1,e2;
	int i=threadIdx.x;
	int j=threadIdx.y;
	int product=0,k;
	for(k=0;k<dim;k++)
	{
		product=product+dA[i*dim+k]*dB[k*dim+j];
	}
	dC[i*dim+j]=product;

}

void read_matrix_from_file(FILE *fp,int m,int n,int matrix[SIZE][SIZE])
{
int i,j;
	for(i=0;i<m;i++)
	{
		for(j=0;j<n;j++)
		{
			fscanf(fp,"%d",&matrix[i][j]);
		}
	}
}

void print_matrix(int matrix[SIZE][SIZE],int m,int n)
{
int i,j;
	for(i=0;i<n;i++)
	{
		for(j=0;j<m;j++)
		{
			printf("%d  ",matrix[i][j]);
		}
		printf("\n");
	}
}
int main(int argc,char *argv[])
{
	FILE *fp1=NULL,*fp2=NULL;
	int m,n,o,p,size;
	int *dA=NULL,*dB=NULL,*dC=NULL;
	enum cudaError error;
	if(argc!=7)
	{
		printf("\n(7 Parameters)./a.out matrixfile1 m n matrixfile2 o p");
		exit(0);
	}
	else
	{
		fp1=fopen(argv[1],"r+");
		fp2=fopen(argv[4],"r+");
		if(fp1==NULL||fp2==NULL)
		{
			printf("\nError in opening the file");
			exit(0);
		}
		else
		{
			m=atoi(argv[2]);
			n=atoi(argv[3]);
			o=atoi(argv[5]);
			p=atoi(argv[6]);
			size=sizeof(int)*m*m;
			// Reading the matrix from file
			read_matrix_from_file(fp1,m,n,matrixA);
			read_matrix_from_file(fp2,o,p,matrixB);
			//Print the matrix on the console
			printf("\n\nMatrixA is:\n");
			print_matrix(matrixA,m,n);
			printf("\n\nMatrixB is:\n");
			print_matrix(matrixB,o,p);
			//The first parameter of the cudaMalloc() function is the address of a pointer variable that must point to the allocated object after allocation.

			error=cudaMalloc((void**)&dA,size);
			if(error) printf("\nError in allocation");
			error=cudaMalloc((void**)&dB,size);
			if(error)printf("\nError in allocation");
			cudaMalloc((void**)&dC,size);
			if(error)printf("\nError in allocation");
			error=cudaMemcpy(dA,matrixA,size,cudaMemcpyHostToDevice);
			if(error)printf("\nError in copying data from host to device");
			error=cudaMemcpy(dB,matrixB,size,cudaMemcpyHostToDevice);
			if(error)printf("\nError in copying data from host to device");

			dim3 dimBlock(m,m);
			dim3 dimGrid(1,1);
			matrix_mult<<<dimGrid,dimBlock>>>(dA,dB,dC,m);

			error=cudaMemcpy(matrixC,dC,size,cudaMemcpyDeviceToHost);
			if(error)printf("\nError in copying data from device to host");
			cudaFree(dA);
			cudaFree(dB);
			cudaFree(dC);
			printf("\n\nThe Product matrix is:(C=A*B):\n");
			print_matrix(matrixC,m,m);

		}
	}
}