多线程等待所有子线程执行完使用总结——CountDownLatch使用和源码初步分析

Posted 2023-02-17

问题背景

我们在日常开发和学习过程中，经常会使用到多线程的场景，其中我们经常会碰到，我们代码需要等待某个或者多个线程执行完再开始执行，上一篇文章中（参考 https://blog.51cto.com/baorant24/6059489 ），我们介绍了object的wait()和notify（），以及线程的join()方法来实现，本文将介绍一种新的方案，CountDownLatch类的使用。

问题分析

大家在日常开发和学习过程中，或多或少都使用过CountDownLatch，知道CountDownLatch的一般用法，CountDownLatch类是用在同步，允许一个或多个线程去等待直到另外的线程完成了一组操作。它通过count进行初始化，await方法会阻塞直到当前的count为0,之后所有的线程将被释放。了解CountDownLatch的一般作用后，我们先来一起看看CountDownLatch的源码相关分析。 重点源码梳理如下：

public class CountDownLatch 
    // 内部类Sync，继承自AbstractQueuedSynchronizer，并且重写了tryAcquireShared（）和tryReleaseShared方法。
    private static final class Sync extends AbstractQueuedSynchronizer 
        private static final long serialVersionUID = 4982264981922014374L;

        Sync(int count) 
            setState(count);
        

        int getCount() 
            return getState();
        

        protected int tryAcquireShared(int acquires) 
            return (getState() == 0) ? 1 : -1;
        

        protected boolean tryReleaseShared(int releases) 
            // Decrement count; signal when transition to zero
            for (;;) 
                int c = getState();
                if (c == 0)
                    return false;
                int nextc = c - 1;
                if (compareAndSetState(c, nextc))
                    return nextc == 0;
            
        
    

    private final Sync sync;

    // 构造方法传入count计数值，然后构造内部的sync对象
    public CountDownLatch(int count) 
        if (count < 0) throw new IllegalArgumentException("count < 0");
        this.sync = new Sync(count);
    
    
    
    public void await() throws InterruptedException 
        sync.acquireSharedInterruptibly(1);
    
    ...

    public void countDown() 
        sync.releaseShared(1);
    
    
    // 获取计数值
    public long getCount() 
        return sync.getCount();
    
    ...

核心代码分析： （1）await()方法分析 java.util.concurrent.CountDownLatch#await()

    public void await() throws InterruptedException 
       sync.acquireSharedInterruptibly(1);
   

java.util.concurrent.locks.AbstractQueuedSynchronizer#acquireSharedInterruptibly

    public final void acquireSharedInterruptibly(int arg)
           throws InterruptedException 
       if (Thread.interrupted())
           throw new InterruptedException();
       // 前面提到CountDownLatch类中Sync内部类对tryAcquireShared进行了重写
       if (tryAcquireShared(arg) < 0)
           doAcquireSharedInterruptibly(arg);
   

java.util.concurrent.CountDownLatch.Sync#tryAcquireShared

              
        // 判断state值是否为0,0的话返回1，不然返回-1。
        protected int tryAcquireShared(int acquires) 
           return (getState() == 0) ? 1 : -1;
       

所有上面 java.util.concurrent.locks.AbstractQueuedSynchronizer#acquireSharedInterruptibly方法会走doAcquireSharedInterruptibly(arg)，继续分析该方法。 java.util.concurrent.locks.AbstractQueuedSynchronizer#doAcquireSharedInterruptibly

    private void doAcquireSharedInterruptibly(int arg)
       throws InterruptedException 
       final Node node = addWaiter(Node.SHARED);
       try 
           for (;;) 
               final Node p = node.predecessor();
               if (p == head) 
                   int r = tryAcquireShared(arg);
                   // 根据前面的分析知道，这块r在state不等于0时，返回-1，所有state未减少到0时，这块回事for循环自旋一直等待。
                   if (r >= 0) 
                       setHeadAndPropagate(node, r);
                       p.next = null; // help GC
                       return;
                   
               
               if (shouldParkAfterFailedAcquire(p, node) &&
                   parkAndCheckInterrupt())
                   throw new InterruptedException();
           
        catch (Throwable t) 
           cancelAcquire(node);
           throw t;
       
   

（2）countDown()方法分析 java.util.concurrent.CountDownLatch#countDown

    public void countDown() 
       sync.releaseShared(1);
   

java.util.concurrent.locks.AbstractQueuedSynchronizer#releaseShared

   public final boolean releaseShared(int arg) 
       // 同样，前面提到，tryReleaseShared方法在CountDownLatch类中Sync内部类中进行了重写
       if (tryReleaseShared(arg)) 
           doReleaseShared();
           return true;
       
       return false;
   

java.util.concurrent.CountDownLatch.Sync#tryReleaseShared

        protected boolean tryReleaseShared(int releases) 
           // Decrement count; signal when transition to zero
           for (;;) 
               int c = getState();
               if (c == 0)
                   return false;
               int nextc = c - 1;
               // 通过compareAndSet保证state-1的同步正确性。
               if (compareAndSetState(c, nextc))
   	    // 当改到0后return TRUE，不然 return FALSE。
                   return nextc == 0;
           
       

问题解决

大概梳理了CountDownLatch的源码后，我们实践一个demo具体再看下，代码如下：

import android.os.Bundle
import android.util.Log
import androidx.appcompat.app.AppCompatActivity
import java.util.concurrent.CountDownLatch

class TestCountDownLatchActivity : AppCompatActivity() 
    private var mCountDownLatch: CountDownLatch? = null

    override fun onCreate(savedInstanceState: Bundle?) 
        super.onCreate(savedInstanceState)
        setContentView(R.layout.activity_test_count_down_latch)

        mCountDownLatch = CountDownLatch(3);

        Thread 
            try 
                Log.d("TAG", "Thread Main A start:" + mCountDownLatch?.count);
                mCountDownLatch?.await();
                Log.d("TAG", "Thread Main A end:" + mCountDownLatch?.count);
             catch (e: InterruptedException) 
                e.printStackTrace();
            
        .start()

        Thread 
            try 
                Log.d("TAG", "Thread Main B start:" + mCountDownLatch?.count);
                mCountDownLatch?.await();
                Log.d("TAG", "Thread Main B end:" + mCountDownLatch?.count);
             catch (e: InterruptedException) 
                e.printStackTrace();
            
        .start()

        Thread 
            Log.d("TAG", "Thread A start:" + mCountDownLatch?.count);
            mCountDownLatch?.countDown();
            Log.d("TAG", "Thread A end:" + mCountDownLatch?.count);
        .start()

        Thread 
            Log.d("TAG", "Thread B start:" + mCountDownLatch?.count);
            mCountDownLatch?.countDown();
            Log.d("TAG", "Thread B end:" + mCountDownLatch?.count);
        .start()

        Thread 
            Log.d("TAG", "Thread C start:" + mCountDownLatch?.count);
            mCountDownLatch?.countDown();
            Log.d("TAG", "Thread C end:" + mCountDownLatch?.count);
        .start()
    

运行结果如下： 结果分析： 线程Main A和线程Main B在执行过程中调用mCountDownLatch?.await()方法，会等待mCountDownLatch的计数值减少到0后再继续执行。其余三个线程Thread A、Thread B、 Thread C执行过程将CountDownLatch分别减1，所以这三个线程都执行完后，线程Main A和线程Main B才会继续执行。

问题总结

我们在日常开发和学习过程中，经常会使用到多线程的场景，其中经常会碰到，我们代码需要等待某个或者多个线程执行完再开始执行，上一篇文章中，我们介绍了object的wait()和notify（），以及线程的join()方法来实现，本文介绍了一种新的方案，CountDownLatch类的使用和源码的初步分析，有兴趣的同学可以进一步深入研究。