前言
线程池实现原理-1
addWorker实现
在看addWorker方法之前，我们先看一个例子，了解一下retry的使用
- break retry 跳到retry处，且不再进入循环
- continue retry 跳到retry处，且再次进入循环
public static void main(String[] args) {
breakRetry();
continueRetry();
}
private static void breakRetry() {
int i = 0;
retry:
for (; ; ) {
System.out.println("start");
for (; ; ) {
i ;
if (i == 4)
break retry;
}
}
//start 进入外层循环
//4
System.out.println(i);
}
private static void continueRetry() {
int i = 0;
retry:
for(;;) {
System.out.println("start");
for(;;) {
i ;
if (i == 3)
continue retry;
System.out.println("end");
if (i == 4)
break retry;
}
}
//start 第一次进入外层循环
//end i=1输出
//end i=2输出
//start 再次进入外层循环
//end i=4输出
//4 最后输出
System.out.println(i);
}这里说一下Runnable 参数的含义
- firstTask != null 说明任务被添加了，我们需要启动一个线程去执行它
- fistTask == null 说明我只想启动一个线程去消费阻塞队列中的任务
// core为ture表示是核心线程，否则非核心线程
private boolean addWorker(Runnable firstTask, boolean core) {
retry:
for (;;) {
int c = ctl.get();
int rs = runStateOf(c);
// Check if queue empty only if necessary.
/**
* 将条件改为如下形式，方便理解
* rs >= SHUTDOWN && (rs != SHUTDOWN || fistTask != null || workQueue.isEmpty)
* 1.如果当前线程池的状态>SHUTDOWN，addWorker返回false，添加任务失败
* 2.如果当前线程池的状态=SHUTDOWN，分为如下2种情况
* (1)workQueue为空，fistTask == null 和fistTask != null的任务都不能
* (2)workQueue不为空，可以添加fistTask != null的任务
*/
if (rs >= SHUTDOWN &&
! (rs == SHUTDOWN &&
firstTask == null &&
! workQueue.isEmpty()))
return false;
for (;;) {
int wc = workerCountOf(c);
if (wc >= CAPACITY ||
// 1.是核心线程 >= corePoolSize
// 2.非核心线程 >= maximumPoolSize
wc >= (core ? corePoolSize : maximumPoolSize))
return false;
// 成功将线程数 1，跳到retry处，并且不再进入死循环
if (compareAndIncrementWorkerCount(c))
break retry;
// 否则重新读取ctl
c = ctl.get(); // Re-read ctl
// 线程状态发生改变，跳到retry处，并且进入死循环
if (runStateOf(c) != rs)
continue retry;
// else CAS failed due to workerCount change; retry inner loop
}
}
// 线程是否启动的标志位
boolean workerStarted = false;
// 线程封装成Worker对象，是否添加到线程池中的标志位
boolean workerAdded = false;
Worker w = null;
try {
w = new Worker(firstTask);
final Thread t = w.thread;
if (t != null) {
final ReentrantLock mainLock = this.mainLock;
mainLock.lock();
try {
// Recheck while holding lock.
// Back out on ThreadFactory failure or if
// shut down before lock acquired.
int rs = runStateOf(ctl.get());
// 1.rs < SHUTDOWN 即 rs = RUNNING
// 2.rs == SHUTDOWN && firstTask == null
if (rs < SHUTDOWN ||
(rs == SHUTDOWN && firstTask == null)) {
if (t.isAlive()) // precheck that t is startable
throw new IllegalThreadStateException();
workers.add(w);
int s = workers.size();
// 刷新了largestPoolSize
if (s > largestPoolSize)
largestPoolSize = s;
workerAdded = true;
}
} finally {
mainLock.unlock();
}
if (workerAdded) {
// 留心一下这里，后面会从这里开始讲起
t.start();
workerStarted = true;
}
}
} finally {
if (! workerStarted)
addWorkerFailed(w);
}
return workerStarted;
}仔细理解一下这段代码，其实就能理解，当线程池处于RUNNING 接受新任务，并且处理进入队列的任务，处于SHUTDOWN 不接受新任务，处理进入队列的任务，剩余状态都不会处理任务，上面代码中的注释有详细解释
代码语言：javascript复制if (rs >= SHUTDOWN &&
! (rs == SHUTDOWN &&
firstTask == null &&
! workQueue.isEmpty()))
return false;线程池在执行任务的时候，会把任务对象包装成一个Worker对象，Worker对象是ThreadPoolExecutor的一个内部类，继承了AbstractQueuedSynchronizer，实现了一个独占锁，status值为0表示未锁定状态，status值为1表示锁定状态，实现了Runnable接口，在执行run方法的时候，它执行完初始化的firstTask后，还会从workQueue中取出任务执行，这样就不用新建一个线程执行任务，而是在一个线程中执行了好几个任务
Worker内部类
代码语言：javascript复制// 省略了一部分对锁的操作，简单的对AQS的一个实现
private final class Worker
extends AbstractQueuedSynchronizer
implements Runnable
{
/** Thread this worker is running in. Null if factory fails. */
final Thread thread;
/** Initial task to run. Possibly null. */
Runnable firstTask;
/** Per-thread task counter */
volatile long completedTasks;
/**
* Creates with given first task and thread from ThreadFactory.
*/
Worker(Runnable firstTask) {
setState(-1); // inhibit interrupts until runWorker
this.firstTask = firstTask;
this.thread = getThreadFactory().newThread(this);
}
/** Delegates main run loop to outer runWorker */
public void run() {
runWorker(this);
}
}runWorker实现
当t.start()被执行后，run方法会执行runWorker方法，来看runWorker方法
代码语言：javascript复制final void runWorker(Worker w) {
Thread wt = Thread.currentThread();
Runnable task = w.firstTask;
w.firstTask = null;
// 允许中断
w.unlock(); // allow interrupts
// 标识线程是不是异常终止的
boolean completedAbruptly = true;
try {
// 先执行初始化的fistTask，执行完成后还会无限循环获取workQueue里的任务来执行
while (task != null || (task = getTask()) != null) {
w.lock();
// If pool is stopping, ensure thread is interrupted;
// if not, ensure thread is not interrupted. This
// requires a recheck in second case to deal with
// shutdownNow race while clearing interrupt
// 配合shutdownNow 方法
if ((runStateAtLeast(ctl.get(), STOP) ||
(Thread.interrupted() &&
runStateAtLeast(ctl.get(), STOP))) &&
!wt.isInterrupted())
wt.interrupt();
try {
// 线程开始开始执行之前执行此方法
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 {
// 线程执行后执行
afterExecute(task, thrown);
}
} finally {
// 运行过的置为null
task = null;
w.completedTasks ;
w.unlock();
}
}
completedAbruptly = false;
} finally {
processWorkerExit(w, completedAbruptly);
}
}这个方法需要注意的就是
- getTask()从阻塞队列中获取任务，如果队列中没有任务会被阻塞，并不会占用CPU资源
- 可以根据业务需要自定义beforeExecute和afterExecute方法
getTask实现
代码语言：javascript复制private Runnable getTask() {
// 标记获取头结点并且移除头结点的方法是否超时
boolean timedOut = false; // Did the last poll() time out?
for (;;) {
int c = ctl.get();
int rs = runStateOf(c);
// Check if queue empty only if necessary.
// 1.rs >= STOP
// 2.rs == SHUTDOWN && workQueue.isEmpty()
if (rs >= SHUTDOWN && (rs >= STOP || workQueue.isEmpty())) {
// 用CAS将线程池中的数量-1，直到成功才会退出
decrementWorkerCount();
return null;
}
int wc = workerCountOf(c);
// Are workers subject to culling?
// 1.核心线程允许被销毁
// 2.核心线程数 > corePoolSize
boolean timed = allowCoreThreadTimeOut || wc > corePoolSize;
// 1.timeOut为true，表示超时获取，workQueue没有任务，说明线程应该被销毁，但是还是要 && timed
// 2.wc > maximumPoolSize肯定要删除线程了
// 3.workQueue为空可以销毁线程，此时有可能所有线程都被销毁了
// 4.workQueue不为空，只有wc > 1才能被删除
if ((wc > maximumPoolSize || (timed && timedOut))
&& (wc > 1 || workQueue.isEmpty())) {
if (compareAndDecrementWorkerCount(c))
return null;
continue;
}
try {
// timed为true，超过keepAliveTime还是没有任务，返回null
// timed为false，则一直阻塞等待任务
Runnable r = timed ?
workQueue.poll(keepAliveTime, TimeUnit.NANOSECONDS) :
workQueue.take();
if (r != null)
return r;
timedOut = true;
} catch (InterruptedException retry) {
timedOut = false;
}
}
}processWorkerExit实现
线程执行完毕执行的方法
代码语言：javascript复制// processWorkerExit在runWorker结束之后被调用
private void processWorkerExit(Worker w, boolean completedAbruptly) {
// 如果是异常终止，或者被中断，减少workerCount
if (completedAbruptly) // If abrupt, then workerCount wasn't adjusted
decrementWorkerCount();
final ReentrantLock mainLock = this.mainLock;
mainLock.lock();
try {
completedTaskCount = w.completedTasks;
workers.remove(w);
} finally {
mainLock.unlock();
}
// Transitions to TERMINATED state if either (SHUTDOWN and pool
// and queue empty) or (STOP and pool empty)
tryTerminate();
int c = ctl.get();
// 状态为RUNNING或者SHUTDOWN
if (runStateLessThan(c, STOP)) {
if (!completedAbruptly) {
int min = allowCoreThreadTimeOut ? 0 : corePoolSize;
if (min == 0 && ! workQueue.isEmpty())
min = 1;
// 目前核心线程已经够用了，不用再创建
if (workerCountOf(c) >= min)
return; // replacement not needed
}
// 增加一个消费的线程
addWorker(null, false);
}
}shutdown实现
代码语言：javascript复制public void shutdown() {
final ReentrantLock mainLock = this.mainLock;
mainLock.lock();
try {
// 检查能否操作线程
checkShutdownAccess();
// 确保状态 >= SHUTDOWN
advanceRunState(SHUTDOWN);
// 中断所有的空闲线程
interruptIdleWorkers();
// ScheduledThreadPoolExecutor会重写这个方法，做一些其他的运算
onShutdown(); // hook for ScheduledThreadPoolExecutor
} finally {
mainLock.unlock();
}
tryTerminate();
}// 中断空闲线程
private void interruptIdleWorkers() {
interruptIdleWorkers(false);
}// onlyOne为true则只中断一个空闲线程，否则全部中断
private void interruptIdleWorkers(boolean onlyOne) {
final ReentrantLock mainLock = this.mainLock;
mainLock.lock();
try {
for (Worker w : workers) {
Thread t = w.thread;
// 遍历Worker并执行中断操作，w.tryLock()保证了正在执行的Worker不会被中断
// 因为正在运行的Worker再次获取锁会失败
if (!t.isInterrupted() && w.tryLock()) {
try {
t.interrupt();
} catch (SecurityException ignore) {
} finally {
w.unlock();
}
}
if (onlyOne)
break;
}
} finally {
mainLock.unlock();
}
}这里需要注意的是不会中断正在运行的线程，因为正在运行的线程w.tryLock()会返回false
shutdownNow实现
代码语言：javascript复制public List<Runnable> shutdownNow() {
List<Runnable> tasks;
final ReentrantLock mainLock = this.mainLock;
mainLock.lock();
try {
checkShutdownAccess();
// 确保状态 >= STOP
advanceRunState(STOP);
// 中断所有线程
interruptWorkers();
// 获取所有没有执行完成的task
// 即将阻塞队列中的任务放到tasks中
tasks = drainQueue();
} finally {
mainLock.unlock();
}
tryTerminate();
return tasks;
}private void interruptWorkers() {
final ReentrantLock mainLock = this.mainLock;
mainLock.lock();
try {
for (Worker w : workers)
w.interruptIfStarted();
} finally {
mainLock.unlock();
}
}// 这个是Worker内部类的方法
void interruptIfStarted() {
Thread t;
// state的初始值为-1，运行到runWorker才允许中断
if (getState() >= 0 && (t = thread) != null && !t.isInterrupted()) {
try {
t.interrupt();
} catch (SecurityException ignore) {
}
}
}shutdownNow会中断所有的线程，因为和shutdown相比在中断之前，不用获取锁
tryTerminate实现
代码语言：javascript复制// 将状态转换到TERMINATED
final void tryTerminate() {
for (;;) {
int c = ctl.get();
// 以下几种状态不能转换为TERMINATED
// 1.RUNNING状态
// 2.TIDYING或TERMINATED
// 3.SHUTDOWN状态，但是workQueue不为空
if (isRunning(c) ||
runStateAtLeast(c, TIDYING) ||
(runStateOf(c) == SHUTDOWN && ! workQueue.isEmpty()))
return;
if (workerCountOf(c) != 0) { // Eligible to terminate
interruptIdleWorkers(ONLY_ONE);
return;
}
final ReentrantLock mainLock = this.mainLock;
mainLock.lock();
try {
if (ctl.compareAndSet(c, ctlOf(TIDYING, 0))) {
try {
// 让子类去实现，做一些操作
terminated();
} finally {
ctl.set(ctlOf(TERMINATED, 0));
termination.signalAll();
}
return;
}
} finally {
mainLock.unlock();
}
// else retry on failed CAS
}
}从上面可看出状态转换的条件
- SHUTDOWN想转化为TIDYING，需要workQueue为空，同时workerCount为0
- STOP转化为TIDYING，需要workerCount为0
