java多线程-8-线程池

问题

• Worker怎么维持？
  • run方法中有个while循环，getTask时阻塞在阻塞队列的消费端
• Worker怎么失效？
  • getTask中，超过核心池的线程会返回null，然后run中的while被打破，run结束了，自然就被回收了

ThreadPoolExecutor

构造函数

public ThreadPoolExecutor(int corePoolSize,
                              int maximumPoolSize,
                              long keepAliveTime,
                              TimeUnit unit,
                              BlockingQueue workQueue,
                              ThreadFactory threadFactory,
                              RejectedExecutionHandler handler) {
    if (corePoolSize < 0 || maximumPoolSize <= 0 || maximumPoolSize < corePoolSize || keepAliveTime < 0)
        throw new IllegalArgumentException();
    if (workQueue == null || threadFactory == null || handler == null)
        throw new NullPointerException();
    this.acc = System.getSecurityManager() == null ? null : AccessController.getContext();
    this.corePoolSize = corePoolSize;
    this.maximumPoolSize = maximumPoolSize;
    this.workQueue = workQueue;
    this.keepAliveTime = unit.toNanos(keepAliveTime);
    this.threadFactory = threadFactory;
    this.handler = handler;
}

状态参数

private final AtomicInteger ctl = new AtomicInteger(ctlOf(RUNNING, 0));
private static final int COUNT_BITS = Integer.SIZE - 3;
private static final int CAPACITY   = (1 << COUNT_BITS) - 1; // 29个1

// runState is stored in the high-order bits
private static final int RUNNING    = -1 << COUNT_BITS; // 111 + 29个0
private static final int SHUTDOWN   =  0 << COUNT_BITS; // 000 + 29个0
private static final int STOP       =  1 << COUNT_BITS; // 001 + 29个0
private static final int TIDYING    =  2 << COUNT_BITS; // 010 + 29个0
private static final int TERMINATED =  3 << COUNT_BITS; // 011 + 29个0

// Packing and unpacking ctl
private static int runStateOf(int c)     { return c & ~CAPACITY; } // 取c前3位
private static int workerCountOf(int c)  { return c & CAPACITY; }  // 取c后29位
private static int ctlOf(int rs, int wc) { return rs | wc; }

• ctl初始值：111 + 29个0

execute

public void execute(Runnable command) {
    if (command == null)
        throw new NullPointerException();
    int c = ctl.get();
    if (workerCountOf(c) < corePoolSize) { // 第1次时workerCountOf(c)为0
        if (addWorker(command, true))
            return;
        c = ctl.get();
    }
    if (isRunning(c) && workQueue.offer(command)) {
        int recheck = ctl.get();
        if (!isRunning(recheck) && remove(command))
            reject(command);
        else if (workerCountOf(recheck) == 0)
            addWorker(null, false);
    }
    else if (!addWorker(command, false))
        reject(command);
}

// Worker是个AQS，也是个Runnable
private boolean addWorker(Runnable firstTask, boolean core) {
    retry:
    for (;;) {
        int c = ctl.get();
        int rs = runStateOf(c);

        // 大于SHUTDOWN || 等于SHUTDOWN && firstTask不为空 || 等于SHUTDOWN && workQueue为空，返回false
        // 含义是：仅在等于SHUTDOWN时，会接着消耗workQueue
        if (rs >= SHUTDOWN && ! (rs == SHUTDOWN && firstTask == null && ! workQueue.isEmpty()))
            return false;

        for (;;) {
            int wc = workerCountOf(c);
            if (wc >= CAPACITY || wc >= (core ? corePoolSize : maximumPoolSize))
                return false; // 超CAPACITY || 超size，返回false
            if (compareAndIncrementWorkerCount(c)) // 111 + 29个0 -> 111 + 28个0 + 1
                break retry; // +1成功，直接break
            c = ctl.get();  // Re-read ctl
            if (runStateOf(c) != rs)
                continue retry;
            // else CAS failed due to workerCount change; retry inner loop
        }
    }

    boolean workerStarted = false;
    boolean workerAdded = false;
    Worker w = null;
    try {
        w = new Worker(firstTask); // Worker extends AQS implements Runnable
        final Thread t = w.thread;
        if (t != null) {
            final ReentrantLock mainLock = this.mainLock; // 非公平ReentrantLock
            mainLock.lock();
            try {
                // Recheck while holding lock.
                // Back out on ThreadFactory failure or if shut down before lock acquired.
                int rs = runStateOf(ctl.get());

                if (rs < SHUTDOWN || (rs == SHUTDOWN && firstTask == null)) {
                    if (t.isAlive()) // precheck that t is startable
                        throw new IllegalThreadStateException();
                    workers.add(w); // workers：HashSet
                    int s = workers.size();
                    if (s > largestPoolSize) // int largestPoolSize
                        largestPoolSize = s;
                    workerAdded = true;
                }
            } finally {
                mainLock.unlock();
            }
            if (workerAdded) {
                t.start();
                workerStarted = true;
            }
        }
    } finally {
        if (! workerStarted)
            addWorkerFailed(w);
    }
    return workerStarted;
}

Worker(Runnable firstTask) {
    setState(-1); // inhibit interrupts until runWorker
    this.firstTask = firstTask;
    this.thread = getThreadFactory().newThread(this);
}

• 最后Worker构造函数这里可以看出thread的参数其实是Worker自己，而真实任务则是firstTask属性
• 那么，上面t.start()那里，其实执行的是Worker的run方法

run

public void run() {
    runWorker(this); // this是Worker
}

final void runWorker(Worker w) {
    Thread wt = Thread.currentThread();
    Runnable task = w.firstTask;
    w.firstTask = null;
    w.unlock(); // allow interrupts
    boolean completedAbruptly = true;
    try {
        while (task != null || (task = getTask()) != null) { // firstTask不为空，或者workQueue不为空，轮询执行任务
            w.lock();
            if ((runStateAtLeast(ctl.get(), STOP) || (Thread.interrupted() && runStateAtLeast(ctl.get(), STOP))) && !wt.isInterrupted())
                wt.interrupt();
            try {
                beforeExecute(wt, task);
                Throwable thrown = null;
                try {
                    task.run();
                } catch (RuntimeException x) {
                    thrown = x; throw x;
                } catch (Error x) {
                    thrown = x; throw x;
                } catch (Throwable x) {
                    thrown = x; throw new Error(x);
                } finally {
                    afterExecute(task, thrown);
                }
            } finally {
                task = null;
                w.completedTasks++;
                w.unlock();
            }
        }
        completedAbruptly = false;
    } finally {
        processWorkerExit(w, completedAbruptly);
    }
}

• execute -> addWorker(ctl+1) -> t.start -> run -> runWorker -> task -> getTask -> workQueue.take处阻塞
• corePoolSize未满前，新来的任务都去新增Worker，因为之前的Worker都阻塞在workQueue
• corePoolSize满时，新来的任务进workQueue；之前阻塞在take的Worker去消耗任务
• workQueue也满了，启用maximumPoolSize
• 超出corePoolSize的Worker，在消耗完workQueue后，会进入processWorkerExit，把自己移出workers，即自失效

ThreadFactory

// default
Executors.defaultThreadFactory()

static class DefaultThreadFactory implements ThreadFactory {
    private static final AtomicInteger poolNumber = new AtomicInteger(1);
    private final ThreadGroup group;
    private final AtomicInteger threadNumber = new AtomicInteger(1);
    private final String namePrefix;

    DefaultThreadFactory() {
        SecurityManager s = System.getSecurityManager();
        group = (s != null) ? s.getThreadGroup() : Thread.currentThread().getThreadGroup();
        namePrefix = "pool-" + poolNumber.getAndIncrement() + "-thread-";
    }

    public Thread newThread(Runnable r) {
        Thread t = new Thread(group, r, namePrefix + threadNumber.getAndIncrement(), 0);
        if (t.isDaemon())
            t.setDaemon(false);
        if (t.getPriority() != Thread.NORM_PRIORITY)
            t.setPriority(Thread.NORM_PRIORITY);
        return t;
    }
}

• new个Thread
• 设置成非daemon
• 设置成普通优先级
• 线程池中的线程时由线程工厂创建并返回的

reject策略

• 发生在maximumPoolSize依旧被耗光时
• 内置4种，默认AbortPolicy
• AbortPolicy - 简单粗暴抛异常

public static class AbortPolicy implements RejectedExecutionHandler {
    public AbortPolicy() {}

    public void rejectedExecution(Runnable r, ThreadPoolExecutor e) {
        throw new RejectedExecutionException("Task " + r.toString() + " rejected from " + e.toString());
    }
}

• DiscardPolicy - 丢弃任务，啥也不干

public static class DiscardPolicy implements RejectedExecutionHandler {
    public DiscardPolicy() {}

    public void rejectedExecution(Runnable r, ThreadPoolExecutor e) {
    }
}

• DiscardOldestPolicy - 把workQueue中最老的command扔掉，把新的执行execute

public static class DiscardOldestPolicy implements RejectedExecutionHandler {
    public DiscardOldestPolicy() {}

    public void rejectedExecution(Runnable r, ThreadPoolExecutor e) {
        if (!e.isShutdown()) {
            e.getQueue().poll();
            e.execute(r);
        }
    }
}

• CallerRunsPolicy - 调用线程池方法的线程同步执行run方法

public static class CallerRunsPolicy implements RejectedExecutionHandler {
    public CallerRunsPolicy() {}

    public void rejectedExecution(Runnable r, ThreadPoolExecutor e) {
        if (!e.isShutdown()) {
            r.run();
        }
    }
}

线程池状态

• RUNNING
  • 创建后的初始状态
  • 可接收新任务
  • 可处理workQueue任务
• SHUTDOWN
  • RUNNING -> shutdown() -> SHUTDOWN
  • 不接收新任务
  • 可处理workQueue任务
• STOP
  • RUNNING | SHUTDOWN -> shutdownNow() -> STOP
  • 不接收新任务
  • 不处理workQueue任务
  • 中断正在处理的任务
• TIDYING
  • SHUTDOWN -> workQueue 和 执行中变空 -> TIDYING
  • STOP -> 执行中变空 -> TIDYING
  • 标识：workerCountOf(ctl) == 0
  • 会执行terminated()方法做后续处理，默认为空
• TERMINATED
  • TIDYING -> terminated() -> TERMINATED
  • 线程池彻底终止