美文网首页
CountDownLatch实现原理

CountDownLatch实现原理

作者: shark没有辣椒 | 来源:发表于2021-12-11 23:30 被阅读0次

1、创建计数器

当我们调用CountDownLatch countDownLatch=new CountDownLatch(4) 时候,此时会创建一个AQS的同步队列,并把创建CountDownLatch 传进来的计数器赋值给AQS队列的 state,所以state的值也代表CountDownLatch所剩余的计数次数;

public CountDownLatch(int count) {
        if (count < 0) throw new IllegalArgumentException("count < 0");
        this.sync = new Sync(count);//创建同步队列,并设置初始计数器值
    }

2、阻塞线程

当我们调用countDownLatch.wait()的时候,会创建一个节点,加入到AQS阻塞队列,并同时把当前线程(主线程)挂起。

public void await() throws InterruptedException {
        sync.acquireSharedInterruptibly(1);
    }

判断计数器是计数完毕,未完毕则把当前线程加入阻塞队列

public final void acquireSharedInterruptibly(int arg)
            throws InterruptedException {
        if (Thread.interrupted())
            throw new InterruptedException();
        //锁重入次数大于0 则新建节点加入阻塞队列,挂起当前线程
        if (tryAcquireShared(arg) < 0)
            doAcquireSharedInterruptibly(arg);
    }

构建阻塞队列的双向链表,挂起当前线程

private void doAcquireSharedInterruptibly(int arg)
        throws InterruptedException {
        //新建节点加入阻塞队列
        final Node node = addWaiter(Node.SHARED);
        boolean failed = true;
        try {
            for (;;) {
                //获得当前节点pre节点
                final Node p = node.predecessor();
                if (p == head) {
                    int r = tryAcquireShared(arg);//返回锁的state
                    if (r >= 0) {
                        setHeadAndPropagate(node, r);
                        p.next = null; // help GC
                        failed = false;
                        return;
                    }
                }
                //重组双向链表,清空无效节点,挂起当前线程
                if (shouldParkAfterFailedAcquire(p, node) &&
                    parkAndCheckInterrupt())
                    throw new InterruptedException();
            }
        } finally {
            if (failed)
                cancelAcquire(node);
        }
    }

3、计数器递减

当我们调用countDownLatch.down()方法的时候,会对计数器进行减1操作,AQS内部是通过释放锁的方式,对state进行减1操作,当state=0的时候证明计数器已经递减完毕,此时会将AQS阻塞队列里的节点线程全部唤醒。

public void countDown() {
        //递减锁重入次数,当state=0时唤醒所有阻塞线程
        sync.releaseShared(1);
    }
public final boolean releaseShared(int arg) {
        //递减锁的重入次数
        if (tryReleaseShared(arg)) {
            doReleaseShared();//唤醒队列所有阻塞的节点
            return true;
        }
        return false;
    }
 private void doReleaseShared() {
        //唤醒所有阻塞队列里面的线程
        for (;;) {
            Node h = head;
            if (h != null && h != tail) {
                int ws = h.waitStatus;
                if (ws == Node.SIGNAL) {//节点是否在等待唤醒状态
                    if (!compareAndSetWaitStatus(h, Node.SIGNAL, 0))//修改状态为初始
                        continue;
                    unparkSuccessor(h);//成功则唤醒线程
                }
                else if (ws == 0 &&
                         !compareAndSetWaitStatus(h, 0, Node.PROPAGATE))
                    continue;                // loop on failed CAS
            }
            if (h == head)                   // loop if head changed
                break;
        }
    }

本文参考自:https://zhuanlan.zhihu.com/p/95835099

相关文章

网友评论

      本文标题:CountDownLatch实现原理

      本文链接:https://www.haomeiwen.com/subject/zkwdfrtx.html