使用synchronized(lock)时，锁对象最好是静态的还是非静态的? [英] When using synchronized(lock), the lock object is prefered to be static or non-static?

问题描述

当使用这样的东西时:

private final Object lock = new Object()

静态和非静态有什么区别吗?

Is there any difference between static and non-static one?

非静态对象是否可以锁定静态方法，反之亦然?

Can non-static object lock static method or vice versa?

推荐答案

如果您使用的是非静态锁，则该对象的每个实例都会有一个不同的锁对象，这可能是更细粒度的等效项调用:

If you're using a non-static lock, every instance of the object would have a different lock object, and it would be a potentially more fine grained equivalent of calling:

synchronized(this) {

}

也就是说:您只是锁定来自同一对象内的其他访问.使用静态锁，类的每个实例都共享该锁对象.所以在任何给定时间只有一个线程可以访问同步块.

That is to say: you're only locking against other accesses from within the same object. With a static lock, every instance of the class shares that lock object. So only one thread can access the synchronized block at any given time.

所以这取决于你在做什么.在前一种情况下，分配锁对象没有意义，除非您有多个这些锁保护较小的数据子集.在后一种情况下，您基本上(再次，更细粒度)这样做:

So it depends on what you're doing. In the former case, it doesn't make sense to bother allocating a lock object unless you have multiple of these locks protecting a smaller subset of the data. In the latter case, you're basically (again, more fine grained) doing this:

synchronized(MyObject.class) {

}

也就是说，无论您有相同的对象还是不同的对象进行访问，您都会锁定所有访问.

That is, you're locking against ALL accesses regardless of whether you have the same object or different ones doing the accessing.

这取决于您要完成的任务、您锁定的对象以及这些锁的使用方式.如果您要保护每个实例状态，您可能需要一个每个实例(非静态)锁定对象.如果您要保护全局状态，您需要一个在所有人之间共享的静态锁对象.

It'll depend on what you're trying to accomplish, and what you're locking against, and how those locks are being used. If you're protecting per-instance state, you probably want a per instance (non-static) lock object. If you're protecting global state, you want a static lock object that's shared amongst everyone.

这篇关于使用synchronized(lock)时，锁对象最好是静态的还是非静态的?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！