源码解读之FutureTask如何实现最大等待时间

预备知识:Java 线程挂起的常用方式有以下几种

  1. Thread.sleep(long millis):这个方法可以让线程挂起一段时间,并释放 CPU 时间片,等待一段时间后自动恢复执行。这种方式可以用来实现简单的定时器功能,但如果不恰当使用会影响系统性能。
  2. Object.wait()Object.notify()Object.notifyAll():这是一种通过等待某个条件的发生来挂起线程的方式。wait() 方法会让线程等待,直到其他线程调用了 notify()notifyAll() 方法来通知它。这种方式需要使用 synchronized 或者 ReentrantLock 等同步机制来保证线程之间的协作和通信。
  3. LockSupport.park()LockSupport.unpark(Thread thread):这两个方法可以让线程挂起和恢复。park() 方法会使当前线程挂起,直到其他线程调用了 unpark(Thread thread) 方法来唤醒它。这种方式比较灵活,可以根据需要控制线程的挂起和恢复。

先上结论:

1.futureTask.get时通过LockSupport.park()挂起线程

2.在Thread.run() 方法中 调用 setException(ex)或set(result),然后调用LockSupport.unpark(t)唤醒线程。

一:示例-引入主题

public class FutureTaskDemo {
    public static void main(String[] args) {
        FutureTask<String> futureTask = new FutureTask<>(new Callable() {
            @Override
            public Object call() throws Exception {
                System.out.println("异步线程执行");
                Thread.sleep(3000);//模拟线程执行任务需要3秒
                return "ok";
            }
        });
        Thread t1 = new Thread(futureTask, "线程一");
        t1.start();

        try {
            //关键代码
            String s = futureTask.get(2, TimeUnit.SECONDS); //最大等待线程2秒
        } catch (InterruptedException e) {
            e.printStackTrace();
        } catch (ExecutionException e) {
            e.printStackTrace();
        } catch (TimeoutException e) {
            e.printStackTrace();
        }
    }
}

二:进入futureTask.get(2, TimeUnit.SECONDS);

  public V get(long timeout, TimeUnit unit)
        throws InterruptedException, ExecutionException, TimeoutException {
        if (unit == null)
            throw new NullPointerException();
        int s = state;
        if (s <= COMPLETING &&
            (s = awaitDone(true, unit.toNanos(timeout))) <= COMPLETING) //重点awaitDone,即完成了最大等待,依然没有结果就抛出异常逻辑
            throw new TimeoutException();
        return report(s);
    }

​ awaitDone返回线程任务执行状态,即小于等于COMPLETING(任务正在运行,等待完成)抛出异常TimeoutException

三:进入(awaitDone(true, unit.toNanos(timeout)))原理分析

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

            int s = state;
            if (s > COMPLETING) {
                if (q != null)
                    q.thread = null;
                return s;
            }
            else if (s == COMPLETING) // cannot time out yet
                Thread.yield();
            else if (q == null)
                q = new WaitNode();
            else if (!queued)
                queued = UNSAFE.compareAndSwapObject(this, waitersOffset,
                                                     q.next = waiters, q);
            else if (timed) {
                nanos = deadline - System.nanoTime();
                if (nanos <= 0L) {
                    removeWaiter(q);
                    return state;
                }
                LockSupport.parkNanos(this, nanos);
            }
            else
                LockSupport.park(this);
        }
    }

3.1 总体解读awaitDone

利用自旋(for (;😉)的方式 ,检查state(任务状态)与waitNode(维护等待的线程),

第一步:首先检查if (Thread.interrupted()) 线程是否被打断(LockSupport.parkNanos挂起的线程被打断不抛出异常),

第二步:判断任务状态与waitNode是否入队+确定最大等待时间

​ 若已完成(if (s > COMPLETING))返回任务状态

​ 若已完成(if (s == COMPLETING))-->表示正在完成,但尚未完成。则让出 CPU,进入就绪状态,等待其他线程的执行

​ 若if (q == null)==>创建等待等待节点

​ 若if (!queued)==>表示上一步创建的节点没有和当前线程绑定,故绑定

​ 最后else if (timed)与else,判断最大等待时间

static final class WaitNode {
        volatile Thread thread;
        volatile WaitNode next;
        WaitNode() { thread = Thread.currentThread(); }
    }
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;
state可能转换的过程 
    1.NEW -> COMPLETING -> NORMAL (成功完成)
    2.NEW -> COMPLETING -> EXCEPTIONAL (异常)
    3.NEW -> CANCELLED (任务被取消)
    4.NEW -> INTERRUPTING -> INTERRUPTED(任务被打断)

3.2 关键代码

LockSupport.park(this, nanos) ==内部实现==> UNSAFE.park(false, nanos)();

​ 即让当前线程堵塞直至指定的时间(nanos),该方法同Thread.sleep()一样不会释放持有的对象锁,但不同的是Thread.sleep会被打断(interrupted)并抛出异常,而LockSupport.park被打断不会抛出异常,故在自旋时(for (;😉)需判断if (Thread.interrupted())线程是否被打断(手动抛出异常)。

四:线程运行时state的变化轨迹

4.1:新建时利用构造器设置state=NEW

 public FutureTask(Runnable runnable, V result) {
     this.callable = Executors.callable(runnable, result);
     this.state = NEW;   // 赋值状态
 }

4.2: 线程运行时state可能变化轨迹

public void run() {
        ..........防止多次运行stat()方法..............
        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); //异常轨迹---> 见下分析
                }
                if (ran)
                    set(result); // 正常轨迹--->见下分析
            }
        } finally {
            runner = null;
    		//----最后结束---防止线程被打断
            int s = state;
            if (s >= INTERRUPTING)
                handlePossibleCancellationInterrupt(s);
        }
    }

异常轨迹setException(ex)

protected void setException(Throwable t) {
    if (UNSAFE.compareAndSwapInt(this, stateOffset, NEW, COMPLETING)) {
        outcome = t;
        UNSAFE.putOrderedInt(this, stateOffset, EXCEPTIONAL); // final state
        finishCompletion();
        //轨迹变化 2.NEW -> COMPLETING -> EXCEPTIONAL (异常)
    }
    //否则1: 3.NEW -> CANCELLED (任务被取消)
    //否则2: 4.NEW -> INTERRUPTING -> INTERRUPTED(任务被打断)
}

正常轨迹 set(result);

 1.NEW -> COMPLETING -> NORMAL (成功完成)
posted @ 2023-03-28 14:04  爱我-中华  阅读(151)  评论(0编辑  收藏  举报