countDownLatch使用说明-转发

转：
https://www.jianshu.com/p/e233bb37d2e6

1.背景：

countDownLatch是在java1.5被引入，跟它一起被引入的工具类还有CyclicBarrier、Semaphore、concurrentHashMap和BlockingQueue。
存在于java.util.cucurrent包下。

2.概念

countDownLatch这个类使一个线程等待其他线程各自执行完毕后再执行。
是通过一个计数器来实现的，计数器的初始值是线程的数量。每当一个线程执行完毕后，计数器的值就-1，当计数器的值为0时，表示所有线程都执行完毕，然后在闭锁上等待的线程就可以恢复工作了。

3.源码

countDownLatch类中只提供了一个构造器：
//参数count为计数值
public CountDownLatch(int count) { };
类中有三个方法是最重要的：
//调用await()方法的线程会被挂起，它会等待直到count值为0才继续执行
public void await() throws InterruptedException { };
//和await()类似，只不过等待一定的时间后count值还没变为0的话就会继续执行
public boolean await(long timeout, TimeUnit unit) throws InterruptedException { };
//将count值减1
public void countDown() { };

4.示例

普通示例：

这是例子在一般的书籍上都有，比较初级的例子，实战意义不大。

public class CountDownLatchTest {

public static void main(String[] args) {
    final CountDownLatch latch = new CountDownLatch(2);
    System.out.println("主线程开始执行…… ……");
    //第一个子线程执行
    ExecutorService es1 = Executors.newSingleThreadExecutor();
    es1.execute(new Runnable() {
        @Override
        public void run() {
            try {
                Thread.sleep(3000);
                System.out.println("子线程："+Thread.currentThread().getName()+"执行");
            } catch (InterruptedException e) {
                e.printStackTrace();
            }
            latch.countDown();
        }
    });
    es1.shutdown();

    //第二个子线程执行
    ExecutorService es2 = Executors.newSingleThreadExecutor();
    es2.execute(new Runnable() {
        @Override
        public void run() {
            try {
                Thread.sleep(3000);
            } catch (InterruptedException e) {
                e.printStackTrace();
            }
            System.out.println("子线程："+Thread.currentThread().getName()+"执行");
            latch.countDown();
        }
    });
    es2.shutdown();
    System.out.println("等待两个线程执行完毕…… ……");
    try {
        latch.await();
    } catch (InterruptedException e) {
        e.printStackTrace();
    }
    System.out.println("两个子线程都执行完毕，继续执行主线程");
}

}
结果集：

主线程开始执行…… ……
等待两个线程执行完毕…… ……
子线程：pool-1-thread-1执行
子线程：pool-2-thread-1执行
两个子线程都执行完毕，继续执行主线程

模拟并发示例：

public class Parallellimit {
    public static void main(String[] args) {
        ExecutorService pool = Executors.newCachedThreadPool();
        CountDownLatch cdl = new CountDownLatch(100);
        for (int i = 0; i < 100; i++) {
            CountRunnable runnable = new CountRunnable(cdl);
            pool.execute(runnable);
        }
    }
}

 class CountRunnable implements Runnable {
    private CountDownLatch countDownLatch;
    public CountRunnable(CountDownLatch countDownLatch) {
        this.countDownLatch = countDownLatch;
    }
    @Override
    public void run() {
        try {
            synchronized (countDownLatch) {
                /*** 每次减少一个容量*/
                countDownLatch.countDown();
                System.out.println("thread counts = " + (countDownLatch.getCount()));
            }
            countDownLatch.await();
            System.out.println("concurrency counts = " + (100 - countDownLatch.getCount()));
        } catch (InterruptedException e) {
            e.printStackTrace();
        }
    }
}

修改后的代码：

package juc;

import java.util.concurrent.CountDownLatch;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;

public class CountDownLatchParallellimit {
    public static void main(String[] args) throws InterruptedException {
        ExecutorService pool = Executors.newCachedThreadPool();
        CountDownLatch cdl = new CountDownLatch(100);
        for (int i = 0; i < 100; i++) {
            CountRunnable runnable = new CountRunnable(cdl);
            pool.execute(runnable);
        }
        cdl.await();
        System.out.println("concurrency counts = " + (100 - cdl.getCount()));
        pool.shutdown();

    }

    static class CountRunnable implements Runnable {
        private CountDownLatch countDownLatch;
        public CountRunnable(CountDownLatch countDownLatch) {
            this.countDownLatch = countDownLatch;
        }
        public void run() {
                synchronized (countDownLatch) {
                    /*** 每次减少一个容量*/
                    countDownLatch.countDown();
                    System.out.println("thread counts = " + (countDownLatch.getCount()));
                }
        }
    }
}

说明：
上面代码里的synchronized 用法有点奇怪，CountDownLatch 是线程安全的采用CAS算法，在countDown()后面打印getCount()感觉就是一个悖论。countDown()是线程体完成之后执行的，可是在后面又打印语句。

加synchronized 能让线程顺序打印出来。

自问自答环节

countDown()方法添加在哪里？
CountDownLatch的countDown()方法必须要放在线程内部业务处理完成之后最后一段添加。

如果业务发生了异常怎么办？
捕获异常在finally执行countDown()，否则会导致主线程阻塞。

package juc;

import java.util.concurrent.CountDownLatch;

public class CountDownLatchStudy {

    public static void main(String[] args) {
        final  int SIZE = 10;

        final CountDownLatch  cdl = new CountDownLatch(SIZE);
        for(int i=0;i < SIZE; i++){
            new Thread(new Runnable() {
                public void run() {
                    System.out.println(Thread.currentThread().getName());
                    try {
                        Thread.sleep(100);
                        int a = 1/0;
                    } catch (InterruptedException e) {
                        e.printStackTrace();
                    }finally {
                        //执行完毕后再调用
                        //一定要捕获异常处理
                        cdl.countDown();
                    }
                    //System.out.println(Thread.currentThread().getName()+" : "+cdl.getCount());
                }
            }).start();
        }
        System.out.println(Thread.currentThread().getName()+" : "+cdl.getCount());
        try {
            cdl.await();
        } catch (InterruptedException e) {
            e.printStackTrace();
        }
        System.out.println("done");
    }
}

如果有的线程countDown()没有完成，getCount不为0,，主线程会怎么样？
主线程会一直阻塞。

*CountDownLatch和CyclicBarrier区别：
1.countDownLatch是一个计数器，线程完成一个记录一个，计数器递减，只能只用一次
2.CyclicBarrier的计数器更像一个阀门，需要所有线程都到达，然后继续执行，计数器递增，提供reset功能，可以多次使用

countDownLatch类中只提供了一个构造器：
https://www.jianshu.com/p/e233bb37d2e6

https://www.cnblogs.com/Lee_xy_z/p/10470181.html