#include <stdio.h>
#include <stdlib.h>
#include <ctime>
#include <iostream>
#include <ostream>
#include <arm_neon.h>
#define SIZE 1024
int main(){
	void matmul_optimized(float **A,float **B,float **C,int n);
	int i,j,n;
	float **A=new float*[n];
	float **B=new float*[n];
	float **C=new float*[n];
	for(i=0;i<n;i++){
		A[i]=new float[n];
		B[i]=new float[n];
		C[i]=new float[n];
	}
	for(i=0;i<n;i++){
		for(j=0;j<n;j++){
			A[i][j]=rand()%100;
			B[i][j]=rand()%100;
		}
	}
	clock_t start=clock();
	matmul_optimized(A,B,C,SIZE);
	clock_t end=clock();
    std::cout<<"优化稠密矩阵乘法时间："<<double(end-start)/CLOCKS_PER_SEC<<"秒"<<std::endl; 
	for(i=0;i<n;i++){
		delete[] A[i]; 
		delete[] B[i];
		delete[] C[i];
	}
	delete[] A;
	delete[] B;
	delete[] C;
}
void matmul_optimized(float **A,float **B,float **C,int n){
	int i,j,k;
	for(i=0;i<n;i++){
		for(j=0;j<n;j++){
			float32x4_t vecC=vdupq_n_f32(0.0);
			for(k=0;k<n;k+=4){
				float32x4_t vecA=vld1q_f32(&A[i][j]);	
				float32x4_t vecB=vld1q_f32(&B[i][j]);
				vecC=vmlap_f32(vecC,vecA,vecB);
				C[i][j]=vgetq_lane_f32(vecC,0)+vgetq_lane_f32(vecC,1)+vgetq_lane_f32(vecC,2)+vgetq_lane_f32(vecC,3);
			}
		}
	}
}