#include <stdio.h>
#include <stdlib.h>
#include <time.h>
#include <arm_neon.h>

#define SIZE 1024

// 基础向量加法
void vector_add(float* A, float* B, float* C, int size) {
    for (int i = 0; i < size; i++) {
        C[i] = A[i] + B[i];
    }
}

// NEON 优化向量加法
void vector_add_optimized(float* A, float* B, float* C, int size) {
    int i = 0;
    for (; i <= size - 4; i += 4) {
        float32x4_t vecA = vld1q_f32(&A[i]);
        float32x4_t vecB = vld1q_f32(&B[i]);
        float32x4_t vecC = vaddq_f32(vecA, vecB);
        vst1q_f32(&C[i], vecC);
    }
    for (; i < size; i++) { // 处理剩余元素
        C[i] = A[i] + B[i];
    }
}

int main() {
    float* A = (float*)malloc(SIZE * sizeof(float));
    float* B = (float*)malloc(SIZE * sizeof(float));
    float* C = (float*)malloc(SIZE * sizeof(float));

    srand(time(NULL));
    for (int i = 0; i < SIZE; i++) {
        A[i] = rand() % 100;
        B[i] = rand() % 100;
    }

    clock_t start, end;

    // 基础向量加法
    start = clock();
    vector_add(A, B, C, SIZE);
    end = clock();
    printf("基础向量加法时间: %f 秒\n", (double)(end - start) / CLOCKS_PER_SEC);

    // NEON 优化向量加法
    start = clock();
    vector_add_optimized(A, B, C, SIZE);
    end = clock();
    printf("NEON 优化向量加法时间: %f 秒\n", (double)(end - start) / CLOCKS_PER_SEC);

    free(A);
    free(B);
    free(C);

    return 0;
}