【FutureTask】FutureTask原理-优快云博客

本文链接：https://blog.youkuaiyun.com/m0_64944491/article/details/142569357

前言

熟悉线程池的话，对 FutureTask 一定不会陌生，当我们提交任务时会拿到一个 Future 对象，它代表一个异步任务，而 FutureTask 则是它的实现类，示例如下：

ExecutorService executorService = Executors.newFixedThreadPool(1);
Future<?> future = executorService.submit(() -> {
    System.out.println("do something");
});
Object res = future.get();

Future 位于 java.util.concurrent 包下，是异步任务的顶层接口

public interface Future<V> {

    boolean cancel(boolean mayInterruptIfRunning);

    boolean isCancelled();

    boolean isDone();

    V get() throws InterruptedException, ExecutionException;

    V get(long timeout, TimeUnit unit)
        throws InterruptedException, ExecutionException, TimeoutException;
}

实现

在深入 FutureTask 原理之前，不妨先思考一下，如果让你来实现一个 FutureTask，你会怎么实现？

先不考虑线程安全，可以想到需要这些成员变量

Callable 变量，用于接收任务并做装饰，提交给线程执行
线程变量，用于记录当前等待任务的线程，当任务执行完后唤醒等待线程
执行状态，用于当前任务的执行状态记录

那就不难写出一个自己的实现类

public class MyFutureTask<V> implements Runnable, Future<V> {


    /**
     * 执行任务
     */
    private final Callable<V> callable;

    /**
     * 执行状态
     */
    private int state;

    /**
     * 执行结果
     */
    private V outcome;

    /**
     * 当前等待线程
     */
    private Thread runner;

    private static final int NEW          = 0;
    private static final int NORMAL       = 1;
    private static final int EXCEPTIONAL  = 2;
    private static final int CANCELLED    = 3;

    public MyFutureTask(Callable<V> callable) {
        this.callable = callable;
        this.state = NEW;
    }

    @Override
    public boolean cancel(boolean mayInterruptIfRunning) {
        if (state > NEW) {
            return false;
        }
        if (mayInterruptIfRunning) {
            runner.interrupt();
            state = CANCELLED;
        }
        return true;
    }

    @Override
    public boolean isCancelled() {
        return state == CANCELLED;
    }

    @Override
    public boolean isDone() {
        return state != NEW;
    }

    @Override
    public V get() {
        if (state == NEW) {
            for (; ; ) {
                if (state > NEW) break;
                runner = Thread.currentThread();
                LockSupport.park();
            }
        }
        if (state == NORMAL) {
            return outcome;
        } else {
            throw new RuntimeException("exception!");
        }
    }

    @Override
    public V get(long timeout, TimeUnit unit) throws InterruptedException, ExecutionException, TimeoutException {
        final long deadline = System.nanoTime() + timeout;
        if (state == NEW) {
            for (; ; ) {
                if (state > NEW) break;
                runner = Thread.currentThread();
                long nanos = deadline - System.nanoTime();
                LockSupport.parkNanos(nanos);
            }
        }
        if (state == NORMAL) {
            return outcome;
        } else {
            throw new RuntimeException("exception!");
        }
    }

    @Override
    public void run() {
        if (state != NEW) return;
        try {
            outcome = callable.call();
            state = NORMAL;
        } catch (Exception e) {
            e.printStackTrace();
            state = EXCEPTIONAL;
        } finally {
            LockSupport.unpark(runner);
        }
    }
}

原理

但是仔细想一下，这个 FutureTask 在我们使用的场景肯定是需要提交给不同线程的，也就是说必须满足线程安全

FutureTask#get 多线程场景，必须有一个类似队列的设计，任务完成后把所有等待线程逐个唤醒，同样也存在数据竞争
FutureTask#run 多线程场景，状态的修改存在数据竞争
还要支持可中断的等待

可见一旦引入多线程场景，事情就变得困难了，我们直接看看 FutureTask 源码是怎么做的

成员变量

    private volatile int state;
    private static final int NEW          = 0;
    private static final int COMPLETING   = 1;
    private static final int NORMAL       = 2;
    private static final int EXCEPTIONAL  = 3;
    private static final int CANCELLED    = 4;
    private static final int INTERRUPTING = 5;
    private static final int INTERRUPTED  = 6;

    /** The underlying callable; nulled out after running */
    private Callable<V> callable;
    /** The result to return or exception to throw from get() */
    private Object outcome; // non-volatile, protected by state reads/writes
    /** The thread running the callable; CASed during run() */
    private volatile Thread runner;
    /** Treiber stack of waiting threads */
    private volatile WaitNode waiters;

    static final class WaitNode {
        volatile Thread thread;
        volatile WaitNode next;
        WaitNode() { thread = Thread.currentThread(); }
    }

先看看成员变量，多线程场景存在内存可见性问题的成员变量都用了 volatile 修饰。此外，定义了一个链表的结构，每个数据节点是等待的线程，需要被唤醒，且被 volatile 修饰

CAS

    // Unsafe mechanics
    private static final sun.misc.Unsafe UNSAFE;
    private static final long stateOffset;
    private static final long runnerOffset;
    private static final long waitersOffset;
    static {
        try {
            UNSAFE = sun.misc.Unsafe.getUnsafe();
            Class<?> k = FutureTask.class;
            stateOffset = UNSAFE.objectFieldOffset
                (k.getDeclaredField("state"));
            runnerOffset = UNSAFE.objectFieldOffset
                (k.getDeclaredField("runner"));
            waitersOffset = UNSAFE.objectFieldOffset
                (k.getDeclaredField("waiters"));
        } catch (Exception e) {
            throw new Error(e);
        }
    }

在多线程场景，任务状态的设置、当前等待线程的设置、等待线程在链表中的设置这些都存在数据竞争，而这些场景是非常符合 CAS 的场景，比如多个线程设置任务状态，只能有一个成功，其他都会失败，在这个场景其他的线程都没必要阻塞，设置失败就是失败了

为了实现 CAS，它在这里引入了 Unsafe 类，是 Java 的魔法类之一，虽然我们平时写代码很少直接用它，但其实都间接地用到了它，比如用 Atomic 原子类

FutureTask#run

FutureTask#run 方法是由线程池中的线程执行

    public void run() {
        if (state != NEW ||
            !UNSAFE.compareAndSwapObject(this, runnerOffset,
                                         null, Thread.currentThread()))
            return;
        try {
            Callable<V> c = callable;
            if (c != null && state == NEW) {
                V result;
                boolean ran;
                try {
                    result = c.call();
                    ran = true;
                } catch (Throwable ex) {
                    result = null;
                    ran = false;
                    setException(ex);
                }
                if (ran)
                    set(result);
            }
        } finally {
            // runner must be non-null until state is settled to
            // prevent concurrent calls to run()
            runner = null;
            // state must be re-read after nulling runner to prevent
            // leaked interrupts
            int s = state;
            if (s >= INTERRUPTING)
                handlePossibleCancellationInterrupt(s);
        }
    }

这里我们逐行分析一下

如果执行 run 方法时，看到的 state 状态不是 NEW，或者当前线程 CAS 失败，直接返回。注意这里 CAS 的 expected 是 null，也就是说成员变量 runner 仅会被第一个 CAS 成功的线程赋值成功
对 callable 入参判空、再次判断 state 状态是 NEW
执行 callable 任务，对异常和正常两种情况做处理
异常和正常两种情况分别会通过 CAS 把状态设置为 EXCEPTIONAL 和 NORMAL
最终都会走到 finishCompletion 方法

    private void finishCompletion() {
        // assert state > COMPLETING;
        for (WaitNode q; (q = waiters) != null;) {
            // 每次取到下一个waitNode，就通过CAS把该值赋null
            if (UNSAFE.compareAndSwapObject(this, waitersOffset, q, null)) {
                for (;;) {
                    Thread t = q.thread;
                    if (t != null) {
                        q.thread = null;
                        // 唤醒该线程
                        LockSupport.unpark(t);
                    }
                    WaitNode next = q.next;
                    if (next == null)
                        break;
                    q.next = null; // unlink to help gc
                    q = next;
                }
                break;
            }
        }

        done();

        callable = null;        // to reduce footprint
    }

finishCompletion 方法有几个细节的地方

q.thread = null、q.next = null、callable = null 都是帮助 GC 回收
done 是钩子函数，由子类实现，默认是空实现

FutureTask#get

FutureTask#get 方法由外部提交任务的线程调用，同样做了并发和响应中断的设计，它支持可超时和无限等待两种方式，内部实现都是 awaitDone 方法

    private int awaitDone(boolean timed, long nanos)
        throws InterruptedException {
        final long deadline = timed ? System.nanoTime() + nanos : 0L;
        WaitNode q = null;
        boolean queued = false;
        for (;;) {
            // 响应中断
            if (Thread.interrupted()) {
                removeWaiter(q);
                throw new InterruptedException();
            }

            int s = state;
            // 任务已完成，返回最终的任务状态
            if (s > COMPLETING) {
                if (q != null)
                    q.thread = null;
                return s;
            }
            // 边界情况，直接让出CPU，等待下次循环
            else if (s == COMPLETING) // cannot time out yet
                Thread.yield();
            // waitNode初始化
            else if (q == null)
                q = new WaitNode();
            // 一直尝试CAS设置节点头
            else if (!queued)
                queued = UNSAFE.compareAndSwapObject(this, waitersOffset,
                                                     q.next = waiters, q);
            // 走到这里说明任务还在进行中，进入休眠，等待被唤醒
            else if (timed) {
                nanos = deadline - System.nanoTime();
                if (nanos <= 0L) {
                    removeWaiter(q);
                    return state;
                }
                // 休眠+超时
                LockSupport.parkNanos(this, nanos);
            }
            else
                // 休眠
                LockSupport.park(this);
        }
    }

FutureTask#cancel

FutureTask#cancel 是由外部提交任务的线程调用，取消当前的任务

    public boolean cancel(boolean mayInterruptIfRunning) {
        // 如果任务状态不是NEW或者CAS失败，返回失败
        if (!(state == NEW &&
              UNSAFE.compareAndSwapInt(this, stateOffset, NEW,
                  mayInterruptIfRunning ? INTERRUPTING : CANCELLED)))
            return false;
        try {    // in case call to interrupt throws exception
            // 如果入参为true，直接中断执行任务的线程
            if (mayInterruptIfRunning) {
                try {
                    Thread t = runner;
                    if (t != null)
                        t.interrupt();
                } finally { // final state
                    UNSAFE.putOrderedInt(this, stateOffset, INTERRUPTED);
                }
            }
        } finally {
            // 相当于run执行完成后的调用，唤醒当前get的线程
            finishCompletion();
        }
        return true;
    }