Python garbage collection mechanism

Reference count

Python in , Mainly garbage collection through reference counting

principle

Each object maintains a ob_ref Field , Used to record the number of times the object is currently referenced , Whenever a new reference points to the object , Its reference count ob_ref Add 1, Count every time a reference to this object fails ob_ref reduce 1, Once the reference count of an object is 0, The object is immediately recycled , The memory space occupied by the object will be freed . Its disadvantage is that it requires additional space to maintain reference counts , This problem is secondary to , But the main problem is that it can't solve the problem of objects “ Circular reference ”, therefore , There are also many languages, such as Java There is no garbage collection mechanism using this algorithm .

Causes the reference count to be incremented 1

1. Object created , for example a=2
2. Object is quoted ,b = a
3. Object as a parameter , Into a function
4. Object as an element , Stored in a container

Causes the reference count to decrease 1

1. The object alias is displayed del a
2. The object alias is given to the new object
3. An object leaves its scope
4. The container in which the object is located is destroyed or the object is deleted from the container

To put it bluntly, the number of object references increases or decreases

advantage

The citation counting method has its obvious advantages , If efficient 、 The implementation logic is simple 、 Real time , Once the reference count of an object is set to zero , Memory is released directly .

shortcoming

Increased the burden of space , Each object needs to allocate a separate space to count the reference count , At the same time, its maintenance will also increase the burden .
Circular reference cannot be implemented

Mark clear

Mark removal algorithm is a kind of garbage collection algorithm based on trace collection technology . It is divided into two stages
The first stage ： Marking stage ,GC Will mark all active objects
The second stage ： Put those unmarked objects 【 Inactive objects 】 To recycle .

Generational recycling

Generational recycling is a way of exchanging space for time ,Python The memory is divided into different sets according to the lifetime of the object , Each set is called a generation ,Python Divide memory into 3“ generation ”, They are the young generation （ The first 0 generation ）、 Middle age （ The first 1 generation ）、 Old age （ The first 2 generation ）, They correspond to 3 A linked list , Their garbage collection frequency and the survival time of objects increase and decrease . The newly created objects will be assigned to the younger generation , When the total number of young generation linked lists reaches the upper limit ,Python The garbage collection mechanism will be triggered , Recycle the objects that can be recycled , And objects that won't be recycled will be moved to the middle ages , And so on , The object of the old generation is the one who has the longest life , Even in the whole life cycle of the system . meanwhile , Generational recycling is based on marker removal technology . Generation recycling is also used as Python The auxiliary garbage collection technology of the company deals with those container objects

Reference resources ：
https://foofish.net/python-gc.html
https://zhuanlan.zhihu.com/p/83251959