# PP hw1

## Question 1.

void test1(float* __restrict a, float* __restrict b, float* __restrict c, int N) {
  __builtin_assume(N == 1024);
    a = (float *)__builtin_assume_aligned(a, 32);
    b = (float *)__builtin_assume_aligned(b, 32);
    c = (float *)__builtin_assume_aligned(c, 32);

  fasttime_t time1 = gettime();
  for (int i=0; i<I; i++) {
    for (int j=0; j<N; j++) {
      c[j] = a[j] + b[j];
    }
  }
  fasttime_t time2 = gettime();

  double elapsedf = tdiff(time1, time2);
  std::cout << "Elapsed execution time of the loop in test1():\n"
    << elapsedf << "sec (N: " << N << ", I: " << I << ")\n";
}

## Question 2.

Before optimized:

7.99886sec
9.14774sec
8.72205sec
9.1878sec
9.14816sec

After optimized:

3.70272sec
3.38494sec
3.11554sec
3.47381sec
3.37357sec 

Speed up: $9.14774/3.38494\sim2.70248217103$

After enable AVX2:

1.9841sec
1.89361sec 
1.95184sec
1.85509sec
1.8603sec

Speed up: $3.38494/1.89361\sim1.7875592123$

## Question 3.

After porting the patch of test2.cpp, the logic flow would become more direct, helping compiler easily use vectorization.