Cython:为什么size_t比int快? [英] Cython: why is size_t faster than int?

问题描述

将某些Cython变量从 int 类型更改为 size_t 类型可以显着减少某些功能时间(约30％)，但是我不明白为什么.

Changing certain Cython variables from type int to type size_t can significantly reduce some functions times (~30%), but I do not understand why.

例如:

cimport numpy as cnp
import numpy as np

def sum_int(cnp.int64_t[::1] A):
    cdef unsigned long s = 0
    cdef int k
    for k in xrange(A.shape[0]):
        s += A[k]
    return s

def sum_size_t(cnp.int64_t[::1] A):
    cdef unsigned long s = 0
    cdef size_t k
    for k in xrange(A.shape[0]):
        s += A[k]
    return s

a = np.array(range(1000000))

计时结果:

In [17]: %timeit sum_int(a)   
1000 loops, best of 3: 652 µs per loop

In [18]: %timeit sum_size_t(a)
1000 loops, best of 3: 427 µs per loop

我是Cython的新手，对Fortran的了解比对C的了解还多.请帮帮我.导致这种性能差异的这两个变量类型之间的重要区别是什么?我不喜欢Cython是什么?

I am new to Cython, and know Fortran better than C. Help me out. What is the important difference between these two variable types that causes such a performance difference? What is it that I don't grok about Cython?

推荐答案

您可能必须逐行进行概要分析才能找出确切的位置，但是从生成的C文件中我有一件很突出的事情:检查int 版本是否可以换成负数，假定 size_t 没问题.

You'd likely have to do a line by line profiling to find out exactly, but one thing stands out to me from the produced C file: int version is checked for wraparound to negative numbers, size_t is assumed ok.

在int循环中:( t_3 是从 k 分配的，它们是同一类型)

In the int loop: (t_3 is assigned from k, they're the same type)

if (__pyx_t_3 < 0) {
  __pyx_t_3 += __pyx_v_A.shape[0];
  if (unlikely(__pyx_t_3 < 0)) __pyx_t_4 = 0;
} else if (unlikely(__pyx_t_3 >= __pyx_v_A.shape[0])) __pyx_t_4 = 0;

在size_t循环中:

In the size_t loop:

if (unlikely(__pyx_t_3 >= (size_t)__pyx_v_A.shape[0])) __pyx_t_4 = 0;

因此不需要环绕测试，因为 size_t 是未签名的，并且保证在索引内存中的项目时不会环绕.其余的几乎相同.

So no wraparound test is needed because size_t is unsigned and guaranteed not to wrap around when indexing items in memory. The rest is virtually the same.

更新:关于您的 unsigned int 结果-您的int和size_t大小是多少?它们大小不同是否有可能导致更改?在我的情况下，uint和size_t的C代码是相同的.(因为size_t是未签名的，并且在此系统上特别是unsigned int)

Update: regarding your unsigned int results - what's your size of int and size_t? Any chance they're different size, causing the change? In my case the C code for uint and size_t is identical. (since size_t is unsigned and specifically unsigned int on this system)

这篇关于Cython:为什么size_t比int快?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！