首页 > 代码库 > Cocos2d之“引用计数”内存管理机制实现解析
Cocos2d之“引用计数”内存管理机制实现解析
一、引言
本文主要分析cocos2d游戏开发引擎的引用计数内存管理技术的实现原理。建议读者在阅读本文之前阅读笔者之前一篇介绍如何使用cocos2d内存管理技术的文章——《Cocos2d之Ref类与内存管理使用详解》。
二、相关概念
引用计数
引用计数是计算机编程语言的一种内存管理技术，是指将资源（对象、内存或者磁盘空间等）的被引用计数保存起来，当引用计数变为零时就将资源释放的过程。使用引用计数技术可以实现自动内存管理的目的。
当实例化一个类时，对象的引用计数为1，在其他对象需要持有这个对象时，就把该对象的引用计数加1，需要解除对该对象的持有关系时，就将该对象的引用计数碱1，直至对象的引用计数为0，对象的内存就被立即释放。
内存管理池
就是一个AutoreleasePool对象。
对象池
内存管理池中一个存储所有Ref对象的结构。
三、实现过程详解
Ref类
《Cocos2d之Ref类与内存管理使用详解》一问中已经详细介绍了Ref类的使用，这里就不再赘述了，直接看Ref的源码吧。
class CC_DLL Ref{public: /** * Retains the ownership. * * This increases the Ref‘s reference count. * * @see release, autorelease * @js NA */ void retain(); /** * Releases the ownership immediately. * * This decrements the Ref‘s reference count. * * If the reference count reaches 0 after the descrement, this Ref is * destructed. * * @see retain, autorelease * @js NA */ void release(); /** * Releases the ownership sometime soon automatically. * * This descrements the Ref‘s reference count at the end of current * autorelease pool block. * * If the reference count reaches 0 after the descrement, this Ref is * destructed. * * @returns The Ref itself. * * @see AutoreleasePool, retain, release * @js NA * @lua NA */ Ref* autorelease(); /** * Returns the Ref‘s current reference count. * * @returns The Ref‘s reference count. * @js NA */ unsigned int getReferenceCount() const;protected: /** * Constructor * * The Ref‘s reference count is 1 after construction. * @js NA */ Ref();public: /** * @js NA * @lua NA */ virtual ~Ref();protected: /// count of references unsigned int _referenceCount; friend class AutoreleasePool;#if CC_ENABLE_SCRIPT_BINDINGpublic: /// object id, ScriptSupport need public _ID unsigned int _ID; /// Lua reference id int _luaID; /// scriptObject, support for swift void* _scriptObject;#endif // Memory leak diagnostic data (only included when CC_USE_MEM_LEAK_DETECTION is defined and its value isn‘t zero)#if CC_USE_MEM_LEAK_DETECTIONpublic: static void printLeaks();#endif};
成员变量 _referenceCount 就是所谓的引用计数。下面看成员方法的具体实现。
Ref::Ref(): _referenceCount(1) // when the Ref is created, the reference count of it is 1{ ....}void Ref::retain(){ CCASSERT(_referenceCount > 0, "reference count should greater than 0"); ++_referenceCount;}void Ref::release(){ CCASSERT(_referenceCount > 0, "reference count should greater than 0"); --_referenceCount; if (_referenceCount == 0) { .... delete this; }}Ref* Ref::autorelease(){ PoolManager::getInstance()->getCurrentPool()->addObject(this); return this;}
从源码可以知道，该类在实例化的时候直接把引用计数 _referenceCount 置1，retain和release方法就是负责增减引用计数 _referenceCount 而已，autorelease将该对象交给内存管理池 AutoreleasePool 管理。
下面看一下如何利用这几个接口来创建一个Ref对象。
#define CREATE_FUNC(__TYPE__) static __TYPE__* create() { __TYPE__ *pRet = new __TYPE__(); if (pRet && pRet->init()) { pRet->autorelease(); return pRet; } else { delete pRet; pRet = NULL; return NULL; } }
这个宏定义了cocos2d里=框架中众多Ref子类（如Scene、Layer）的静态方法create(void)的实现。当创建Ref子类对象时，构造函数将 _referenceCount 置1，然后调用对象的autorelease函数把对象交由当前的内存池管理。因为对象一开始 _referenceCount 就为1，那如果对象创建后没有被持有也没有被调用release函数，会不会造成内存泄露呢？不会的，因为autorelease会在内存管理池销毁后将对象的 _referenceCount 减 1，以抵消创建对象时增加的引用计数，如果此时引用计数为0，那么对象就会被立即回收。
PoolManager类
Ref类的autorelease函数其实已经将内存管理的工作交给PoolManager完成了，而自己只是管理引用计数而已。下面看PoolManager类的声明。
class CC_DLL PoolManager{public: /** * @js NA * @lua NA */ CC_DEPRECATED_ATTRIBUTE static PoolManager* sharedPoolManager() { return getInstance(); } static PoolManager* getInstance(); /** * @js NA * @lua NA */ CC_DEPRECATED_ATTRIBUTE static void purgePoolManager() { destroyInstance(); } static void destroyInstance(); /** * Get current auto release pool, there is at least one auto release pool that created by engine. * You can create your own auto release pool at demand, which will be put into auto releae pool stack. */ AutoreleasePool *getCurrentPool() const; bool isObjectInPools(Ref* obj) const; /** * @js NA * @lua NA */ friend class AutoreleasePool; private: PoolManager(); ~PoolManager(); void push(AutoreleasePool *pool); void pop(); static PoolManager* s_singleInstance; std::vector<AutoreleasePool*> _releasePoolStack;};
从源码可以看出，PoolManager类被设置成单例模式。单例 s_singleInstance 持有一个管理 AutoreleasePool的 _releasePoolStack 栈。AutoreleasePool类是真正的内存管理池，其实现过程会在下文谈到。_releasePoolStack 栈中至少会有一个由引擎创建的AutoreleasePool对象。PoolManager类提供了获取当前内存管理池currentPool和向对象管理池查询对象的接口。注意，push和pop函数被限制为private，意味着开发者不能够之间向PoolManager单例中添加或者删除AutoreleasePool对象管理池。push和pop方法被AutoreleasePool使用的，也就是说AutoreleasePool创建后将自己添加到 _releasePoolStack 栈中。下面分析PoolManager的实现过程。接下来看PoolManager的实现。
PoolManager* PoolManager::s_singleInstance = nullptr;PoolManager* PoolManager::getInstance(){ if (s_singleInstance == nullptr) { s_singleInstance = new (std::nothrow) PoolManager(); // Add the first auto release pool new AutoreleasePool("cocos2d autorelease pool"); } return s_singleInstance;}void PoolManager::destroyInstance(){ delete s_singleInstance; s_singleInstance = nullptr;}PoolManager::PoolManager(){ _releasePoolStack.reserve(10);}PoolManager::~PoolManager(){ CCLOGINFO("deallocing PoolManager: %p", this); while (!_releasePoolStack.empty()) { AutoreleasePool* pool = _releasePoolStack.back(); delete pool; }}AutoreleasePool* PoolManager::getCurrentPool() const{ return _releasePoolStack.back();}bool PoolManager::isObjectInPools(Ref* obj) const{ for (const auto& pool : _releasePoolStack) { if (pool->contains(obj)) return true; } return false;}void PoolManager::push(AutoreleasePool *pool){ _releasePoolStack.push_back(pool);}void PoolManager::pop(){ CC_ASSERT(!_releasePoolStack.empty()); _releasePoolStack.pop_back();}
从源码不难看出，PoolManager持有一个存储所有所有对象管理池的栈 _releasePoolStack，PoolManger无非就是实现了对 releasePoolStack 栈的管理而已，内存管理的工作由栈中不同的AutoreleasePool对象完成。
AutoreleasePool类
AutoreleasePool对象就是内存管理池。下面看AutoreleasePool类的声明。
class CC_DLL AutoreleasePool{public: /** * @warn Don‘t create an auto release pool in heap, create it in stack. * @js NA * @lua NA */ AutoreleasePool(); /** * Create an autorelease pool with specific name. This name is useful for debugging. */ AutoreleasePool(const std::string &name); /** * @js NA * @lua NA */ ~AutoreleasePool(); /** * Add a given object to this pool. * * The same object may be added several times to the same pool; When the * pool is destructed, the object‘s Ref::release() method will be called * for each time it was added. * * @param object The object to add to the pool. * @js NA * @lua NA */ void addObject(Ref *object); /** * Clear the autorelease pool. * * Ref::release() will be called for each time the managed object is * added to the pool. * @js NA * @lua NA */ void clear(); /** * Checks whether the pool contains the specified object. */ bool contains(Ref* object) const; /** * Dump the objects that are put into autorelease pool. It is used for debugging. * * The result will look like: * Object pointer address object id reference count * */ void dump(); private: /** * The underlying array of object managed by the pool. * * Although Array retains the object once when an object is added, proper * Ref::release() is called outside the array to make sure that the pool * does not affect the managed object‘s reference count. So an object can * be destructed properly by calling Ref::release() even if the object * is in the pool. */ std::vector<Ref*> _managedObjectArray; std::string _name; };
从源码可以看出，AutoreleasePool内存管理池中只有一个Ref对象池 _managedObjectArray，AutoreleasePool类提供里往 _managedObjectArray 对象池添加对象、查询对象和清空对象池的接口。AutoreleasePool内存管理池自行维护 _managedObjectArray 对象池。
AutoreleasePool::AutoreleasePool(): _name(""){ _managedObjectArray.reserve(150); PoolManager::getInstance()->push(this);}AutoreleasePool::AutoreleasePool(const std::string &name): _name(name){ _managedObjectArray.reserve(150); PoolManager::getInstance()->push(this);}AutoreleasePool::~AutoreleasePool(){ CCLOGINFO("deallocing AutoreleasePool: %p", this); clear(); PoolManager::getInstance()->pop();}void AutoreleasePool::addObject(Ref* object){ _managedObjectArray.push_back(object);}void AutoreleasePool::clear(){ for (const auto &obj : _managedObjectArray) { obj->release(); } _managedObjectArray.clear();}bool AutoreleasePool::contains(Ref* object) const{ for (const auto& obj : _managedObjectArray) { if (obj == object) return true; } return false;}void AutoreleasePool::dump(){ CCLOG("autorelease pool: %s, number of managed object %d\n", _name.c_str(), static_cast<int>(_managedObjectArray.size())); CCLOG("%20s%20s%20s", "Object pointer", "Object id", "reference count"); for (const auto &obj : _managedObjectArray) { CC_UNUSED_PARAM(obj); CCLOG("%20p%20u\n", obj, obj->getReferenceCount()); }}
从AutoreleasePool的构造函数可知，AutoreleasePool内存管理池创建后会将自己添加到 PoolManager 的 _releasePoolStack 栈中，成为PoolManager的currentPool。AutoreleasePool的析构函数调用clear函数，clear函数会将调用对象池中每个对象的release函数，如果对象的引用计数为0则将对象释放，到这一步就完成了Ref类的autorelease函数了所有功能。
四、总结
看到这里，读者应该明白了cocos2d的引用计数内存管理技术是由AutoreleasePool、PoolManager和Ref三个类协助完成的。所有Ref的子类，都能够使用这套内存管理机制。AutoreleasePool（内存管理池）中有个对象池，存放了所有需要被管理的Ref对象。cocos2d引擎至少有一个默认的内存管理池，开发者也可以自定义和创建自己的内存管理池。PoolManager用一个栈结构管理所有的内存管理池。
Cocos2d之“引用计数”内存管理机制实现解析