为什么必须在g ++中启用优化才能进行简单的数组访问？ [英] Why do I have to turn on optimization in g++ for simple array access?

查看：72 发布时间：2020/10/6 23:21:24 c++ c performance compiler-optimization

本文介绍了为什么必须在g ++中启用优化才能进行简单的数组访问？的处理方法，对大家解决问题具有一定的参考价值，需要的朋友们下面随着小编来一起学习吧！

问题描述

我用C ++中的 double s的 std :: vector 编写了一个简单的高斯消去算法（gcc / Linux）。现在，我看到运行时取决于编译器的优化级别（ -O3 的运行速度提高了5倍）。我编写了一个小型测试程序，并收到了类似的结果。问题不在于向量的分配，也不在于向量的大小调整等。

I have written a simple Gaussian elimination algorithm using a std::vector of doubles in C++ (gcc / Linux). Now I have seen that the runtime depends on the optimization level of the compiler (up to 5-times faster with -O3). I wrote a small test program and received similar results. The problem is not the allocation of the vector nor any resizing etc.

简单的事实是：

v[i] = x + y / z;

（或类似的东西）在没有优化的情况下要慢得多。我认为问题在于索引运算符。如果没有编译器优化，则 std :: vector 会比原始的 double * v 慢，但是当我打开优化时，性能是相等的，令我惊讶的是，甚至访问原始 double * v 的速度也更快。

(or something like that) is much slower without optimization. I think the problem is the index operator. Without compiler optimization, the std::vector is slower than a raw double *v, but when I turn on optimization, the performance is equal and, to my surprise, even the access to the raw double *v is faster.

对此行为有解释吗？我确实不是一名专业开发人员，但是我认为编译器应该能够将上述语句转换为硬件指令，而不是直接转换为硬件指令。为什么需要启动优化，更重要的是，优化的缺点是什么？（如果没有，我想知道为什么优化不是标准的。）

Is there an explanation for this behaviour? I'm really not a professional developer, but I thought the compiler should be able to transfer statements like the above one rather directly to hardware instructions. Why is there a need to turn on an optimization and, more importantly, what is the disadvantage of the optimization? (If there is none, I wonder why the optimization is not the standard.)

这是我的矢量测试代码：

Here is my vector test code:

const long int count = 100000;
const double pi = 3.1416;
void C_array (long int size)
{
  long int start = time(0);
  double *x = (double*) malloc (size * sizeof(double));
  for (long int n = 0; n < count; n++)
    for (long int i = 0; i < size; i++)
      x[i] = i;
      //x[i] = pi * (i-n);
  printf ("C array   : %li s\n", time(0) - start);
  free (x);
}

void CPP_vector (long int size)
{
  long int start = time(0);
  std::vector<double> x(size);
  for (long int n = 0; n < count; n++)
    for (long int i = 0; i < size; i++)
      x[i] = i;
      //x[i] = pi * (i-n);
  printf ("C++ vector: %li s\n", time(0) - start);
}

int main ()
{
  printf ("Size of vector: ");
  long int size;
  scanf ("%li", &size);

  C_array (size);
  CPP_vector (size);
  return 0;
}

我收到了一些奇怪的结果。一个标准的g ++编译器会生成一个8 s（C数组）或18 s（ std :: vector ）的运行时，其向量大小为20000。如果使用更复杂的行在 // .. 后面，运行时间为8/15 s（是，更快）。如果我打开 -O3 ，则对于40,000个矢量大小，运行时间为5/5 s。

I received some weird results. A standard g++ compile produces a runtime 8 s (C array) or 18 s (std::vector) for a vector size of 20 000. If I use the more complex line behind the //.., the runtime is 8 / 15 s (yes, faster). If I turn on -O3 then, the runtime is 5 / 5 s for a 40,000 vector size.

为什么必须在g ++中启用优化才能进行简单的数组访问？ [英] Why do I have to turn on optimization in g++ for simple array access?

问题描述

推荐答案

相关文章

C/C++开发最新文章

热门教程

热门工具

登录关闭

为什么必须在g ++中启用优化才能进行简单的数组访问？ [英] Why do I have to turn on optimization in g++ for simple array access?

问题描述

推荐答案

相关文章

C/C++开发最新文章

热门教程

热门工具

登录 关闭

登录关闭