contents

• related file
• memory layout
• example
  • 0
  • 1
  • 0.1
  • 1.1
  • -0.1
• free_list

related file

• cpython/Objects/floatobject.c
• cpython/Include/floatobject.h
• cpython/Objects/clinic/floatobject.c.h

memory layout

PyFloatObject is simply a wrapper around the C type double, which takes 8 bytes to represent a floating-point number

you can refer to IEEE 754/IEEE-754标准与浮点数运算 for more detail

example

0

The binary representation of 0.0 in IEEE 754 format is 64 zero bits

1.0

0.1

1.1

The difference between 1.1 and 0.1 is the last few exponent bits

-0.1

The difference between -0.1 and 0.1 is the first sign bit

free_list

#ifndef PyFloat_MAXFREELIST
#define PyFloat_MAXFREELIST    100
#endif
static int numfree = 0;
static PyFloatObject *free_list = NULL;

free_list is a singly linked list that stores at most PyFloat_MAXFREELIST PyFloatObjects

The linked list is linked via the ob_type field

>>> f = 0.0
>>> id(f)
4551393664
>>> f2 = 1.0
>>> id(f2)
4551393616
del f
del f2

f3 comes from the head of the free_list

>>> f3 = 3.0
>>> id(f3)
4551393616