Java线程池：简明易懂的源码分析

痛定思痛。 2023-05-22 10:30 37阅读 0赞

## Java线程池：简明易懂的源码分析 ##

> Java线程池的一些基础知识，可以参考博客
> 
> [Java线程池，你五分钟讲完，而我和面试官聊了半小时][Java]
> 
> 本文将从源码角度分析线程池原理，加深对线程池原理的理解，简单背几个原理知识，其实很难得到面试官的青睐，了解源码知识，可以由内而外得征服面试官。

### Java线程池状态转换 ###

使用Integer类型(32bit) `ctl`中**高3位**记录Java线程池的状态,**低29位**记录线程数量

**RUNNING**:`111`、**SHUTDOWN**:`000`、**STOP**:`001`

**TIDYING**:`010`、**TERMINATED**:`011`.后面均又29位零

private final AtomicInteger ctl = new AtomicInteger(ctlOf(RUNNING, 0));
    private static final int COUNT_BITS = Integer.SIZE - 3; // 29
    //线程池最大容量2^29-1
    private static final int CAPACITY   = (1 << COUNT_BITS) - 1;
    
    private static final int RUNNING    = -1 << COUNT_BITS;
    private static final int SHUTDOWN   =  0 << COUNT_BITS;
    private static final int STOP       =  1 << COUNT_BITS;
    private static final int TIDYING    =  2 << COUNT_BITS;
    private static final int TERMINATED =  3 << COUNT_BITS;
    //取得线程池运行状态
    private static int runStateOf(int c)     {  return c & ~CAPACITY; }
    //取得线程池线程数量
    private static int workerCountOf(int c)  {  return c & CAPACITY; }
    private static int ctlOf(int rs, int wc) {  return rs | wc; }

![format_png][]

> 下面从线程池的submit方法作为出发点，从源码角度分析，线程池的实现原理

### submit方法 ###

`submit`方法将提交的任务`task`,包装成`RunnableFuture`对象,既实现了`Runnable`接口，又实现了`Future`接口。之后会调用`execute`方法

public Future<?> submit(Runnable task) { 
        if (task == null) throw new NullPointerException();
        RunnableFuture<Void> ftask = newTaskFor(task, null);
        execute(ftask);
        return ftask;
    }

### execute方法 ###

当提交一个新任务，线程池的处理流程如下：

*  判断线程池中核心线程数是否达到`corePoolSize`,若否，则创建一个新核心线程执行任务
 *  若核心线程数已达阈值,判断`workQueue`是否已满，若未满，则将新任务添加进阻塞队列
 *  若满，再判断，线程池中线程数是否达到阈值`maximumPoolSize`,若否，则新建一个非核心线程执行任务。若达到阈值，则执行**线程池饱和策略**。

**execute方法，各种参数**

<table> 
 <thead> 
  <tr> 
   <th>参数类型</th> 
   <th>用途</th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <td><code>addWorker(firstTask, true)</code></td> 
   <td>创建核心线程，执行提交的任务</td> 
  </tr> 
  <tr> 
   <td><code>addWorker(firstTask, false)</code></td> 
   <td>创建非核心线程，执行提交的任务</td> 
  </tr> 
  <tr> 
   <td><code>addWorker(null, false)</code></td> 
   <td>创建非核心线程，执行工作队列中任务</td> 
  </tr> 
  <tr> 
   <td><code>addWorker(null, true)</code></td> 
   <td>创建核心线程，执行工作队列中任务</td> 
  </tr> 
 </tbody> 
</table>

public void execute(Runnable command) { 
        if (command == null)
            throw new NullPointerException();
            
        int c = ctl.get();
        //线程数小于corePoolSize
        if (workerCountOf(c) < corePoolSize) { 
            if (addWorker(command, true))
                return;
            c = ctl.get();
        }
        //线程数大于等于corePoolSize,尝试添加进workQueue
        if (isRunning(c) && workQueue.offer(command)) { 
            //再次检查
            int recheck = ctl.get();
            //如果状态不为Running，则从队列中移除任务
            if (!isRunning(recheck) && remove(command))
                reject(command);
            //如果线程池数量为0，则添加一个非核心线程执行任务
            else if (workerCountOf(recheck) == 0)
                addWorker(null, false);
        }
        //队列满，尝试新建非核心线程，执行任务
        else if (!addWorker(command, false))
            //否则执行拒绝策略
            reject(command);
    }

### addWorker方法 ###

首先通过自旋操作，将线程总数加1.之后用独占锁锁住，构建一个Worker对象，将Worker对象添加进workers(`HashSet<Worker>`),释放锁，启动线程。

private boolean addWorker(Runnable firstTask, boolean core) { 
        retry:
        for (;;) { 
            int c = ctl.get();
            int rs = runStateOf(c);
            //以下情况直接返回
    	    //线程池处于STOP、TYDING、TERMINATED状态
            //线程池处于SHUTDOWN状态，firstTask!=null
            //线程池处于SHUTDOWN状态，firstTask==null, workQueue为空
            if (rs >= SHUTDOWN &&
                ! (rs == SHUTDOWN &&
                   firstTask == null &&
                   ! workQueue.isEmpty()))
                return false;
    		//循环CAS操作，增加元素个数
            for (;;) { 
                int wc = workerCountOf(c);
                if (wc >= CAPACITY ||
                    wc >= (core ? corePoolSize : maximumPoolSize))
                    return false;
                if (compareAndIncrementWorkerCount(c))
                    break retry;
                //CAS失败，查看线程池状态是否改变，变化则跳到最外层循环
                c = ctl.get(); 
                if (runStateOf(c) != rs)
                    continue retry;
            }
        }
        boolean workerStarted = false;
        boolean workerAdded = false;
        Worker w = null;
        try { 
            w = new Worker(firstTask);
            final Thread t = w.thread;
            if (t != null) { 
                final ReentrantLock mainLock = this.mainLock;
                //独占锁进行同步
                mainLock.lock();
                try { 
                    int rs = runStateOf(ctl.get());
    			   //重新check一下线程池状态
                    if (rs < SHUTDOWN ||
                        (rs == SHUTDOWN && firstTask == null)) { 
                        if (t.isAlive()) 
                            throw new IllegalThreadStateException();
                        //添加任务
                        workers.add(w);
                        int s = workers.size();
                        if (s > largestPoolSize)
                            largestPoolSize = s;
                        workerAdded = true;
                    }
                } finally { 
                    mainLock.unlock();
                }
                //添加成功，执行任务
                if (workerAdded) { 
                    t.start();
                    workerStarted = true;
                }
            }
        } finally { 
            if (!workerStarted)
                addWorkerFailed(w);
        }
        return workerStarted;
    }

### Worker类( 实现了 AQS) ###

Worker对象，包括`thread`和`firstTask`成员变量。`thread`成员变量的构造函数的参数是自身的Worker对象。

所以**调用thread成员变量的run方法，其实就是调用本身Worker对象的run方法**。

本身Worker对象的run方法，会调用`runWorker`方法，参数为本身Worker对象

private final class Worker extends AbstractQueuedSynchronizer
            implements Runnable
    { 
        
        final Thread thread;
        Runnable firstTask;
        volatile long completedTasks;
    
        Worker(Runnable firstTask) { 
            this.firstTask = firstTask;
            this.thread = getThreadFactory().newThread(this);
        }
    
        public void run() { 
            runWorker(this);
        }
    }

### runWorker方法 ###

final void runWorker(Worker w) { 
        Thread wt = Thread.currentThread();
        Runnable task = w.firstTask;
        w.firstTask = null;
        w.unlock(); //允许中断
        boolean completedAbruptly = true;
        try { 
            //循环过程，执行完第一个任务后，一直从队列中拿取任务
            while (task != null || (task = getTask()) != null) { 
                w.lock();
                //处于SHUTDOWN状态，并不会中断正在运行的任务
                if ((runStateAtLeast(ctl.get(), STOP) ||
                     (Thread.interrupted() &&
                      runStateAtLeast(ctl.get(), STOP))) &&
                    !wt.isInterrupted())
                    wt.interrupt();
                try { 
                    //执行任务前干一些事
                    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 { 
                        //执行任务后干一些事
                        afterExecute(task, thrown);
                    }
                } finally { 
                    task = null;
                    w.completedTasks++;
                    w.unlock();
                }
            }
            completedAbruptly = false;
        } finally { 
            processWorkerExit(w, completedAbruptly);
        }
    }

### getTask方法 ###

从工作队列中，拿取任务，如果满足某几个条件(线程超时与否、线程池状态、工作队列是否为空)，直接返回null.

private Runnable getTask() { 
        boolean timedOut = false; 
        for (;;) { 
            int c = ctl.get();
            int rs = runStateOf(c);
    
            //线程池状态为SHUTDOWN，并且workQueue为空
            //线程池处于STOP、TYDING、TERMINATED状态
            //以上两种情况,返回null
            if (rs >= SHUTDOWN && (rs >= STOP || workQueue.isEmpty())) { 
                //线程总数减一
                decrementWorkerCount();
                return null;
            }
            int wc = workerCountOf(c);
            //allowCoreThreadTimeOut为true,运行核心线程受超时机制影响
            boolean timed = allowCoreThreadTimeOut || wc > corePoolSize;
    		
            //空闲线程超时，直接返回null
            if ((wc > maximumPoolSize || (timed && timedOut))
                && (wc > 1 || workQueue.isEmpty())) { 
                if (compareAndDecrementWorkerCount(c))
                    return null;
                continue;
            }
    
            try { 
                //从队列中取任务
                //如果空闲线程超时，workQueue.poll方法返回null
                Runnable r = timed ?
                    workQueue.poll(keepAliveTime, TimeUnit.NANOSECONDS) :
                    workQueue.take();
                if (r != null)
                    return r;
                timedOut = true;
            } catch (InterruptedException retry) { 
                timedOut = false;
            }
        }
    }

### 参考文章 ###

[Java线程池源码分析（一)][Java 1]

[Java 线程池 ThreadPoolExecutor 源码分析][Java _ ThreadPoolExecutor]  
[submit与execute方法的区别][submit_execute]

[Java]: https://blog.csdn.net/zycxnanwang/article/details/105563351
[format_png]: /images/20230521/1f8245444ff04c2abc7ca70ca2330020.png
[Java 1]: https://blog.csdn.net/mayongzhan_csdn/article/details/80790966
[Java _ ThreadPoolExecutor]: https://blog.csdn.net/cleverGump/article/details/50688008
[submit_execute]: https://www.cnblogs.com/handsomeye/p/6225033.html