克雷格.兰丁&hagersten (CLH Lock)

CLH Lock摘要

CLH lock is Craig, Landin, and Hagersten (CLH) locks, CLH lock is a spin lock, can ensure no hunger, provide fairness first come first service.
The CLH lock is a scalable, high performance, fairness and spin lock based on the list, the application thread spin only on a local variable, it constantly polling the precursor state, if it is found that the pre release lock end spin.

CLH锁是自旋锁的一种，对它的研究是因为AQS源代码中使用了CLH锁的一个变种，为了更好的理解AQS中使用锁的思想，所以决定先好好理解CLH锁

Java代码实现

public class ClhSpinLock implements Lock{
    private final ThreadLocal<Node> prev;
    private final ThreadLocal<Node> node;
    private final AtomicReference<Node> tail = new AtomicReference<Node>(new Node());

    public ClhSpinLock() {
        this.node = new ThreadLocal<Node>() {
            protected Node initialValue() {
                return new Node();
            }
        };

        this.prev = new ThreadLocal<Node>() {
            protected Node initialValue() {
                return null;
            }
        };
    }

    /**
     * 1.初始状态 tail指向一个node(head)节点 
     * +------+ 
     * | head | <---- tail 
     * +------+
     * 
     * 2.lock-thread加入等待队列: tail指向新的Node，同时Prev指向tail之前指向的节点
     * +----------+
     * | Thread-A |
     * | := Node  | <---- tail
     * | := Prev  | -----> +------+
     * +----------+        | head |
     *                     +------+ 
     * 
     *             +----------+            +----------+
     *             | Thread-B |            | Thread-A |
     * tail ---->  | := Node  |     -->    | := Node  | 
     *             | := Prev  | ----|      | := Prev  | ----->  +------+
     *             +----------+            +----------+         | head |
     *                                                          +------+ 
     * 3.寻找当前node的prev-node然后开始自旋
     * 
     */
    public void lock() {
        final Node node = this.node.get();
        node.locked = true;
        Node pred = this.tail.getAndSet(node);
        this.prev.set(pred);
        // 自旋
        while (pred.locked);
    }

    public void unlock() {
        final Node node = this.node.get();
        node.locked = false;
        this.node.set(this.prev.get());
    }

    @Override
    public void lockInterruptibly() throws InterruptedException {
        // TODO Auto-generated method stub
        
    }

    @Override
    public boolean tryLock() {
        // TODO Auto-generated method stub
        return false;
    }

    @Override
    public boolean tryLock(long time, TimeUnit unit) throws InterruptedException {
        // TODO Auto-generated method stub
        return false;
    }

    @Override
    public Condition newCondition() {
        // TODO Auto-generated method stub
        return null;
    }
    
    private static class Node {
        private volatile boolean locked;
    }
}

简单的看一下CLH的算法定义

the list, the application thread spin only on a local variable, it constantly polling the precursor state, if it is found that the pre release lock end spin.

基于list，线程仅在一个局部变量上自旋，它不断轮询前一个节点状态，如果发现前一个节点释放锁结束.

所以在java中使用了ThreadLocal作为具体实现,AtomicReference为了消除多个线程并发对tail引用Node的影响，核心方法lock()中分为3个步骤去实现

1. 初始状态 tail指向一个node(head)节点

   private final AtomicReference<Node> tail = new AtomicReference<Node>(new Node());

2. thread加入等待队列: tail指向新的Node，同时Prev指向tail之前指向的节点，在java代码中使用了getAndSet即CAS操作使用

   Node pred = this.tail.getAndSet(node);
   this.prev.set(pred);

3. 寻找当前线程对应的node的前驱node然后开始自旋前驱node的status判断是否可以获取lock

   while (pred.locked);

同理unlock()方法，获取当前线程的node，设置lock status，将当前node指向前驱node(这样操作tail指向的就是前驱node等同于出队操作).至此CLH Lock的过程就结束了

测试ClhSpinLock

public class LockTest {
    static int count = 0;
    
    public static void testLock(Lock lock) {
        try {
            lock.lock();
            for (int i = 0; i < 10000000; i++) ++count;
        } finally {
            lock.unlock();
        }
    }
    
    public static void main(String[] args) throws InterruptedException, BrokenBarrierException {
        final ClhSpinLock clh = new ClhSpinLock();
        final CyclicBarrier cb = new CyclicBarrier(10, new Runnable() {
            @Override
            public void run() {
                System.out.println(count);
            }
        });
        
        for (int i = 0; i < 10; i++) {
            new Thread(new Runnable() {
                @Override
                public void run() {
                    testLock(clh);
                    // 这段代码是非lock比较使用
//                    for (int i = 0; i < 10000000; i++)
//                        count++;
                    try {
                        cb.await();
                    } catch (InterruptedException | BrokenBarrierException e) {
                        e.printStackTrace();
                    }
                }
            }).start();
        }
    }
}

测试代码使用了CyclicBarrier辅助当所有子线程完成后输出static变量count的值
结果发现输出的值和预期一样，CLH Lock完成了独占式锁的功能

小结

CLH Lock是一种比较简单的自旋锁算法之一，因为锁的CAS操作涉及到了硬件的锁定(锁总线或者是锁内存)所以性能和CPU架构也密不可分，该兴趣的同学可以继续深入研究包括MCS锁等。CLH Lock是独占式锁的一种，并且是不可重入的锁，这篇文章是对AQS锁源代码分析的预热篇