谈论高并发（二十二）解决java.util.concurrent各种组件（四） 深入了解AQS（二）

http://blog.csdn.net/ITer_ZC/article/category/2592507/2

上一页介绍AQS其基本设计思路以及两个内部类Node和ConditionObject实现 聊聊高并发（二十一）解析java.util.concurrent各个组件（三） 深入理解AQS（一） 这篇说一说AQS的主要方法的实现。AQS和CLHLock的最大差别是，CLHLock是自旋锁，而AQS使用Unsafe的park操作让线程进入等待（堵塞）。

线程增加同步队列，和CLHLock一样，从队尾入队列，使用CAS+轮询的方式实现无锁化。

入队列后设置节点的prev和next引用，形成双向链表的结构

private Node enq(final Node node) {        for (;;) {            Node t = tail;            if (t == null) { // Must initialize                if (compareAndSetHead(new Node()))                    tail = head;            } else {                node.prev = t;                if (compareAndSetTail(t, node)) {                    t.next = node;                    return t;                }            }        }    }

线程指定独享还是共享方式增加队列，先尝试增加一次，假设失败再用enq()轮询地尝试，比方addWaiter(Node.EXCLUSIVE), addWaiter(Node.SHARED)

private Node addWaiter(Node mode) {        Node node = new Node(Thread.currentThread(), mode);        // Try the fast path of enq; backup to full enq on failure        Node pred = tail;        if (pred != null) {            node.prev = pred;            if (compareAndSetTail(pred, node)) {                pred.next = node;                return node;            }        }        enq(node);        return node;    }

唤醒后继节点，最典型的情况就是在线程释放锁后，会唤醒后继节点。会从节点的next開始，找到一个后继节点，假设next是null。就从队尾開始往head找，直到找到最靠近当前节点的兴许节点。 waitStatus <= 0的隐含意思是线程没有被取消。 然后用LockSupport唤醒这个找到的后继节点的线程。

这种方法类似于CLHLock里面释放锁时，通知兴许节点来获取锁。AQS使用了堵塞的方式，所以这种方法的兴许方法是acquireXXX方法，它负责将兴许节点唤醒，兴许节点再依据状态去推断是否获得锁

private void unparkSuccessor(Node node) {        /*         * If status is negative (i.e., possibly needing signal) try         * to clear in anticipation of signalling.  It is OK if this         * fails or if status is changed by waiting thread.         */        int ws = node.waitStatus;        if (ws < 0)            compareAndSetWaitStatus(node, ws, 0);        /*         * Thread to unpark is held in successor, which is normally         * just the next node.  But if cancelled or apparently null,         * traverse backwards from tail to find the actual         * non-cancelled successor.         */        Node s = node.next;        if (s == null || s.waitStatus > 0) {            s = null;            for (Node t = tail; t != null && t != node; t = t.prev)                if (t.waitStatus <= 0)                    s = t;        }        if (s != null)            LockSupport.unpark(s.thread);    }

共享模式下的释放操作。从队首開始向队尾扩散，假设节点的waitStatu是SIGNAL,就唤醒后继节点。假设waitStatus是0，就设置标记成PROPAGATE

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;        }    }

取消获取操作，要把节点从同步队列中去除。通过链表操作将它的前置节点的next指向它的后继节点集合。假设该节点是在队尾，直接删除就可以，否则要通知后继节点去获取锁

private void cancelAcquire(Node node) {        // Ignore if node doesn't exist        if (node == null)            return;        node.thread = null;        // Skip cancelled predecessors        Node pred = node.prev;        while (pred.waitStatus > 0)            node.prev = pred = pred.prev;        // predNext is the apparent node to unsplice. CASes below will        // fail if not, in which case, we lost race vs another cancel        // or signal, so no further action is necessary.        Node predNext = pred.next;        // Can use unconditional write instead of CAS here.        // After this atomic step, other Nodes can skip past us.        // Before, we are free of interference from other threads.        node.waitStatus = Node.CANCELLED;        // If we are the tail, remove ourselves.        if (node == tail && compareAndSetTail(node, pred)) {            compareAndSetNext(pred, predNext, null);        } else {            // If successor needs signal, try to set pred's next-link            // so it will get one. Otherwise wake it up to propagate.            int ws;            if (pred != head &&                ((ws = pred.waitStatus) == Node.SIGNAL ||                 (ws <= 0 && compareAndSetWaitStatus(pred, ws, Node.SIGNAL))) &&                pred.thread != null) {                Node next = node.next;                if (next != null && next.waitStatus <= 0)                    compareAndSetNext(pred, predNext, next);            } else {                unparkSuccessor(node);            }            node.next = node; // help GC        }    }

独占模式而且不可中断地获取队列锁的操作。这种方法在ConditionObject.await()中被使用。当线程被Unsafe.unpark唤醒后，须要调用acquireQueued来获取锁，从而结束await(). accquireQueued()方法要么获得锁，要么被tryAcquire方法抛出的异常打断，假设抛出异常，最后在finally里面取消获取

值得注意的是仅仅有节点的前驱节点是head的时候，才干获得锁。

这里隐含了一个意思，就是head指向当前获得锁的节点。当程序进入if(p == head and tryAcquire(arg))这个分支时。表示线程获得了锁或者被中断。将自己设置为head，将next设置为null.

shouldParkAfterFailedAcquired()方法的目的是将节点的前驱节点的waitStatus设置为SIGNAL，表示会通知兴许节点，这样兴许节点才干放心去park。而不用操心被丢失唤醒的通知。

parkAndCheckInterupt()方法会真正运行堵塞，并返回中断状态，这种方法有两种情况返回。一种是park被unpark唤醒。这时候中断状态为false。还有一种情况是park被中断了，因为这个accquireQueued方法是不可中断的版本号。所以即使线程被中断了，也仅仅是设置了中断标志为true,没有跑出中断异常。

在支持中断的获取版本号里。这时会抛出中断异常。

这种方法能够理解为Lock的lock里没有获取锁的分支。在CLHLock自旋锁的实现里。是对前驱节点的状态自旋，而AQS是堵塞。所以这里是在同步队列里面进入了堵塞状态。等待被前驱节点释放锁时唤醒。

释放锁时会依据状态调用unparkSuccessor()方法来唤醒兴许节点，这样就会在这种方法里面把堵塞的线程唤醒并获得锁。

队列锁的优点是线程都在多个共享状态上自旋或堵塞。所以unparkSuccessor()方法仅仅会唤醒它后继没有取消的节点。

而取消仅仅有两种情况，中断或者超时

final boolean acquireQueued(final Node node, int arg) {        boolean failed = true;        try {            boolean interrupted = false;            for (;;) {                final Node p = node.predecessor();                if (p == head && tryAcquire(arg)) {                    setHead(node);                    p.next = null; // help GC                    failed = false;                    return interrupted;                }                if (shouldParkAfterFailedAcquire(p, node) &&                    parkAndCheckInterrupt())                    interrupted = true;            }        } finally {            if (failed)                cancelAcquire(node);        }    }

独占模式支持中断的获取队列锁操作。能够看到和不支持中断版本号的差别，这里假设parkAndCheckInterrupt()方法返回时显示被中断了，就抛出中断异常

private void doAcquireInterruptibly(int arg)        throws InterruptedException {        final Node node = addWaiter(Node.EXCLUSIVE);        boolean failed = true;        try {            for (;;) {                final Node p = node.predecessor();                if (p == head && tryAcquire(arg)) {                    setHead(node);                    p.next = null; // help GC                    failed = false;                    return;                }                if (shouldParkAfterFailedAcquire(p, node) &&                    parkAndCheckInterrupt())                    throw new InterruptedException();            }        } finally {            if (failed)                cancelAcquire(node);        }    }

独占模式限时获取队列锁操作, 这个获取的总体逻辑和前面的类似，差别是它支持限时操作，假设等待时间大于spinForTimeoutThreshold，就使用堵塞的方式等待，否则用自旋等待。使用了LockSupport.parkNanos（）方法来实现限时地等待，并支持中断

这里隐含的一个含义是parkNanos方法退出有3种方式，

1. 限时到了自己主动退出，这时候会超时

2. 没有到限时被唤醒了。这时候是不超时的

3. 被中断

private boolean doAcquireNanos(int arg, long nanosTimeout)        throws InterruptedException {        long lastTime = System.nanoTime();        final Node node = addWaiter(Node.EXCLUSIVE);        boolean failed = true;        try {            for (;;) {                final Node p = node.predecessor();                if (p == head && tryAcquire(arg)) {                    setHead(node);                    p.next = null; // help GC                    failed = false;                    return true;                }                if (nanosTimeout <= 0)                    return false;                if (shouldParkAfterFailedAcquire(p, node) &&                    nanosTimeout > spinForTimeoutThreshold)                    LockSupport.parkNanos(this, nanosTimeout);                long now = System.nanoTime();                nanosTimeout -= now - lastTime;                lastTime = now;                if (Thread.interrupted())                    throw new InterruptedException();            }        } finally {            if (failed)                cancelAcquire(node);        }    }

共享模式获得队列锁操作。获得操作也是从head的下一个节点開始，和独占模式仅仅unparkSuccessor一个节点不同，共享模式下，等head的兴许节点被唤醒了，它要扩散这样的共享的获取，使用setHeadAndPropagate操作，把自己设置为head，而且把释放的状态往下传递。这里採用了链式唤醒的方法，1个节点负责唤醒1个兴许节点，直到不能唤醒。当后继节点是共享模式isShared，就调用doReleaseShared来唤醒后继节点

doReleaseShared会从head開始往后检查状态，假设节点是SIGNAL状态，就唤醒它的后继节点。

假设是0就标记为PROPAGATE, 等它释放锁的时候会再次唤醒后继节点。

这里有个隐含的意思:

1. 增加同步队列并堵塞的节点，它的前驱节点仅仅会是SIGNAL。表示前驱节点释放锁时。后继节点会被唤醒。shouldParkAfterFailedAcquire()方法保证了这点，假设前驱节点不是SIGNAL,它会把它改动成SIGNAL。

这里不是SIGNAL就有可能是PROPAGATE

2. 造成前驱节点是PROPAGATE的情况是前驱节点获得锁时。会唤醒一次后继节点。但这时候后继节点还没有增加到同步队列，所以临时把节点状态设置为PROPAGATE,当后继节点增加同步队列后，会把PROPAGATE设置为SIGNAL,这样前驱节点释放锁时会再次doReleaseShared，这时候它的状态已经是SIGNAL了，就能够唤醒兴许节点了

private void doAcquireShared(int arg) {        final Node node = addWaiter(Node.SHARED);        boolean failed = true;        try {            boolean interrupted = false;            for (;;) {                final Node p = node.predecessor();                if (p == head) {                    int r = tryAcquireShared(arg);                    if (r >= 0) {                        setHeadAndPropagate(node, r);                        p.next = null; // help GC                        if (interrupted)                            selfInterrupt();                        failed = false;                        return;                    }                }                if (shouldParkAfterFailedAcquire(p, node) &&                    parkAndCheckInterrupt())                    interrupted = true;            }        } finally {            if (failed)                cancelAcquire(node);        }    }private void setHeadAndPropagate(Node node, int propagate) {        Node h = head; // Record old head for check below        setHead(node);        /*         * Try to signal next queued node if:         *   Propagation was indicated by caller,         *     or was recorded (as h.waitStatus) by a previous operation         *     (note: this uses sign-check of waitStatus because         *      PROPAGATE status may transition to SIGNAL.)         * and         *   The next node is waiting in shared mode,         *     or we don't know, because it appears null         *         * The conservatism in both of these checks may cause         * unnecessary wake-ups, but only when there are multiple         * racing acquires/releases, so most need signals now or soon         * anyway.         */        if (propagate > 0 || h == null || h.waitStatus < 0) {            Node s = node.next;            if (s == null || s.isShared())                doReleaseShared();        }    }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;        }    }

tryXXXX 方法，这几个方法是给子类重写的。用来扩展响应的同步器操作

protected boolean tryAcquire(int arg) {        throw new UnsupportedOperationException();    }protected boolean tryRelease(int arg) {        throw new UnsupportedOperationException();    }protected int tryAcquireShared(int arg) {        throw new UnsupportedOperationException();    }protected boolean tryReleaseShared(int arg) {        throw new UnsupportedOperationException();    }

独占模式获取操作的顶层方法，假设没有tryAcquired，或者没有获得队列锁，就中断

public final void acquire(int arg) {        if (!tryAcquire(arg) &&            acquireQueued(addWaiter(Node.EXCLUSIVE), arg))            selfInterrupt();    }

独占模式释放操作的顶层方法，假设tryRelease()成功，那么就唤醒后继节点去获取锁

public final boolean release(int arg) {        if (tryRelease(arg)) {            Node h = head;            if (h != null && h.waitStatus != 0)                unparkSuccessor(h);            return true;        }        return false;    }