java - 并发ConcurrentHashMap的问题

Question

这个问题的 说明有点长，

step1 并发map中的put

step2

step3

step4

step5

step6

step7

step8

step9

上面中steop3 表示重复尝试获取锁达到一定的次数之后就 调用lock阻塞当前线程

我想问的问题是关键在 step7 这个if中的tryAcquire 表示的是尝试获取锁，如果获取不成功就返回false，然后将创建一个Node 封装 当前线程添加到一个队列中
（1）其中if的第二个条件表示什么意思？ ，好像不仅仅是获取一个节点那么简单的

（2）selfInterrupt 表示自我中断，中断不是说让当前线程放弃正在执行的任务吗（并不是立即放弃，是下次cpu检查的时候查看中断位是吗）

而我的理解是：这里的需求是需要将当前线程阻塞，因为当前线程始终获取不到锁，而这里的实现是使用了中断，中断可以阻塞线程吗？（平时使用中断不都是中断正在等待或者阻塞的线程吗？）

（3）其实最本质想问的是，ReentrantLock实现中 ，当一个线程多次尝试获取不到锁之后是如何阻塞当前线程的？ 阻塞之后又是如何唤醒的

求各位开导呐

Answer 1

Thank you for the invitation, but I really can’t answer because first of all I haven’t studied the source code of ConcurrentHashMap myself, and secondly it is almost impossible to ask such details in the interview. Unless you need to write a class library like guava yourself, I don't think it makes much sense to study these.

Answer 2

Thank you for the invitation, but I'm sorry, my knowledge of concurrency-related learning is still very limited. Let me help you invite people who may be able to answer it

Answer 3

I’ll answer as many as I can
Question (1) involves the issue of fair locks. If you can’t get it, insert the Node into the queue. Focus on java.util.concurrent.locks.AbstractQueuedSynchronizer< /code>java.util.concurrent.locks.AbstractQueuedSynchronizer
问题（3）线程将自己挂起，这部分的实现是依靠底层的，参照LockSupportProblem (3) The thread hangs itself. The implementation of this part depends on the underlying layer. Please refer to LockSupport

public static void park(Object blocker) {
    Thread t = Thread.currentThread();
    setBlocker(t, blocker);
    unsafe.park(false, 0L);
    setBlocker(t, null);
}

Wake up using unparkmethod