CountDownLatch
CountDownLatch
基于AQS实现的同步器,允许一个或者多个线程通过await()
方法进入阻塞等待,直到一个或者多个线程执行countDown()
完成。CountDownLatch
在创建时需要传入一个count
值,一旦某个或者多个线程调用了await()
方法,那么需要等待count
值减为0,才能继续执行。
countDown()
方法每执行一次,count(state)值减1,直到减为0。一个线程可以多次调用countDown()
方法,每次调用都会造成count减1
CountDownLatch
在RocketMQ底层通信被大量使用,实现远程调用异步转同步。Netty Client
发送消息之前创建一个ResponseFuture
,ReponseFuture
中有一个CountDownLatch
属性,发送消息之后调用await()
,等待response,当接收到响应之后,调用对应ResponseFuture
中CountDownLatch#countDown
,唤醒阻塞线程。
内部类AQS实现
private static final class Sync extends AbstractQueuedSynchronizer { private static final long serialVersionUID = 4982264981922014374L; Sync(int count) { setState(count); } int getCount() { return getState(); } protected int tryAcquireShared(int acquires) { return (getState() == 0) ? 1 : -1; } protected boolean tryReleaseShared(int releases) { // Decrement count; signal when transition to zero for (;;) { int c = getState(); if (c == 0) return false; int nextc = c-1; if (compareAndSetState(c, nextc)) return nextc == 0; } } }
构造函数
public CountDownLatch(int count) { // count不能为负数 if (count < 0) throw new IllegalArgumentException("count < 0"); // 创建同步器,设置state为count this.sync = new Sync(count); }
await
public void await() throws InterruptedException { // AQS#acquireSharedInterruptibly -> Sync#tryAcquireShared(如果state=0 返回1,立即返回,线程继续向下执行,如果state != 0, 返回-1,线程进入同步队列,阻塞排队) sync.acquireSharedInterruptibly(1); } public final void acquireSharedInterruptibly(int arg) throws InterruptedException { if (Thread.interrupted()) throw new InterruptedException(); // 如果state != 0,tryAcquireShared()方法返回-1,说明需要等待其他线程执行countDown()方法,线程进入同步队列阻塞 // 如果state = 0,tryAcquireShared()方法返回1,线程立即返回,继续向下执行 if (tryAcquireShared(arg) < 0) doAcquireSharedInterruptibly(arg); } protected int tryAcquireShared(int acquires) { return (getState() == 0) ? 1 : -1; } private void doAcquireSharedInterruptibly(int arg) throws InterruptedException { // 进入同步队列阻塞 final Node node = addWaiter(Node.SHARED); boolean failed = true; try { // 自旋等待state = 0,等待其他线程执行完毕 for (;;) { final Node p = node.predecessor(); if (p == head) { // 如果state = 0,表明其他同步线程执行完毕,线程阻塞结束 int r = tryAcquireShared(arg); if (r >= 0) { // 更新头节点为自己,并向后唤醒其他阻塞的线程 setHeadAndPropagate(node, r); p.next = null; // help GC failed = false; return; } } if (shouldParkAfterFailedAcquire(p, node) && parkAndCheckInterrupt()) throw new InterruptedException(); } } finally { if (failed) cancelAcquire(node); } }
countDown
/** * count(state)值减1,当减为0时,由于await调用阻塞的线程将被唤醒继续执行 */ public void countDown() { sync.releaseShared(1); } public final boolean releaseShared(int arg) { if (tryReleaseShared(arg)) { // 将count值减-1,如果count值减1后等于0,返回true, // count值减1后等于0,唤醒在同步队列上等待的第一个线程,第一个线程会向后传播,唤醒后驱节点(doAcquireSharedInterruptibly) doReleaseShared(); return true; } return false; } /** * 自旋 + CAS完成更新 */ protected boolean tryReleaseShared(int releases) { // Decrement count; signal when transition to zero for (;;) { int c = getState(); if (c == 0) return false; int nextc = c-1; if (compareAndSetState(c, nextc)) return nextc == 0; } } /** * count(state)值减1后等于0,会调用该方法,该方法唤醒在同步队列上等待的第一个线程 */ private void doReleaseShared() { for (;;) { Node h = head; if (h != null && h != tail) { int ws = h.waitStatus; if (ws == Node.SIGNAL) { if (!compareAndSetWaitStatus(h, Node.SIGNAL, 0)) continue; // loop to recheck cases unparkSuccessor(h); } else if (ws == 0 && !compareAndSetWaitStatus(h, 0, Node.PROPAGATE)) continue; // loop on failed CAS } if (h == head) // loop if head changed break; } }
获取count
public long getCount() { return sync.getCount(); }
来源:https://www.cnblogs.com/QullLee/p/12247777.html