问题描述
我试图了解以下两个代码块之间的区别
I am trying to understand the difference between the two following code blocks
AtomicBoolean ab = new AtomicBoolean(false); using the following to get and set state. . ab.get(); ab.set(X); vs. private boolean ab = false; private final Object myboollock = new O**ect(); public void setAB(boolean state) { synchronized(myboollock) { ab = state; } } public boolean getAB() { synchronized(myboollock) { return ab; } }
我需要线程保护一个布尔值,仅此而已,并且有以前用过后来的方法,但是想开始用Atomic对象,(如果)它们是安全的?,
I need to thread protect a boolean, that is all, and have in the past used the later method, but would like to start to use Atomic objects, (if ) they are safe?,
推荐答案
这两个代码片段有一些细微的差别,但从外观上看是相似的:如果你调用 set 方法,随后调用 get 的其他线程将可以看到更改.
There are a few subtle differences but seen from the outside the two code snippets behave similarly: if you call the set method, the change will be visible to other threads calling get subsequently.
主要区别在于:
- 性能:根据争用的级别,您可能会使用 synchronized 或 AtomicBoolean 获得更好的性能
- 原子性:如果在某个阶段您想做的不仅仅是设置布尔值,synchronized 块将允许您以原子方式添加指令,但 AtomicBoolean 不会
- performance: depending on the level of contention, you may get better performance with synchronized or AtomicBoolean
- atomicity: if at some stage you want to do more than just setting the boolean value, a synchronized block will allow you to add instructions atomically but AtomicBoolean won't