notify通知的遗漏很容易理解,即threadA还没开始wait的时候,threadB已经notify了,这样,threadB通知是没有任何响应的,当threadB退出synchronized代码块后,threadA再开始wait,便会一直阻塞等待,直到被别的线程打断。
下面给出一段代码演示通知是如何遗漏的,如下:
public class MissedNotify extends Object { private Object proceedLock; public MissedNotify() { print("in MissedNotify()"); proceedLock = new Object(); } public void waitToProceed() throws InterruptedException { print("in waitToProceed() - entered"); synchronized ( proceedLock ) { print("in waitToProceed() - about to wait()"); proceedLock.wait(); print("in waitToProceed() - back from wait()"); } print("in waitToProceed() - leaving"); } public void proceed() { print("in proceed() - entered"); synchronized ( proceedLock ) { print("in proceed() - about to notifyAll()"); proceedLock.notifyAll(); print("in proceed() - back from notifyAll()"); } print("in proceed() - leaving"); } private static void print(String msg) { String name = Thread.currentThread().getName(); System.out.println(name + ": " + msg); } public static void main(String[] args) { final MissedNotify mn = new MissedNotify(); Runnable runA = new Runnable() { public void run() { try { //休眠1000ms,大于runB中的500ms, //是为了后调用waitToProceed,从而先notifyAll,后wait, //从而造成通知的遗漏 Thread.sleep(1000); mn.waitToProceed(); } catch ( InterruptedException x ) { x.printStackTrace(); } } }; Thread threadA = new Thread(runA, "threadA"); threadA.start(); Runnable runB = new Runnable() { public void run() { try { //休眠500ms,小于runA中的1000ms, //是为了先调用proceed,从而先notifyAll,后wait, //从而造成通知的遗漏 Thread.sleep(500); mn.proceed(); } catch ( InterruptedException x ) { x.printStackTrace(); } } }; Thread threadB = new Thread(runB, "threadB"); threadB.start(); try { Thread.sleep(10000); } catch ( InterruptedException x ) {} //试图打断wait阻塞 print("about to invoke interrupt() on threadA"); threadA.interrupt(); } }
执行结果如下:
分析:由于threadB在执行mn.proceed()之前只休眠了500ms,而threadA在执行mn.waitToProceed()之前休眠了1000ms,因此,threadB会先苏醒,继而执行mn.proceed(),获取到proceedLock的对象锁,继而执行其中的notifyAll(),当退出proceed()方法中的synchronized代码块时,threadA才有机会获取proceedLock的对象锁,继而执行其中的wait()方法,但此时notifyAll()方法已经执行完毕,threadA便漏掉了threadB的通知,便会阻塞下去。后面主线程休眠10秒后,尝试中断threadA线程,使其抛出InterruptedException。
为了修正MissedNotify,需要添加一个boolean指示变量,该变量只能在同步代码块内部访问和修改。修改后的代码如下:
public class MissedNotifyFix extends Object { private Object proceedLock; //该标志位用来指示线程是否需要等待 private boolean okToProceed; public MissedNotifyFix() { print("in MissedNotify()"); proceedLock = new Object(); //先设置为false okToProceed = false; } public void waitToProceed() throws InterruptedException { print("in waitToProceed() - entered"); synchronized ( proceedLock ) { print("in waitToProceed() - entered sync block"); //while循环判断,这里不用if的原因是为了防止早期通知 while ( okToProceed == false ) { print("in waitToProceed() - about to wait()"); proceedLock.wait(); print("in waitToProceed() - back from wait()"); } print("in waitToProceed() - leaving sync block"); } print("in waitToProceed() - leaving"); } public void proceed() { print("in proceed() - entered"); synchronized ( proceedLock ) { print("in proceed() - entered sync block"); //通知之前,将其设置为true,这样即使出现通知遗漏的情况,也不会使线程在wait出阻塞 okToProceed = true; print("in proceed() - changed okToProceed to true"); proceedLock.notifyAll(); print("in proceed() - just did notifyAll()"); print("in proceed() - leaving sync block"); } print("in proceed() - leaving"); } private static void print(String msg) { String name = Thread.currentThread().getName(); System.out.println(name + ": " + msg); } public static void main(String[] args) { final MissedNotifyFix mnf = new MissedNotifyFix(); Runnable runA = new Runnable() { public void run() { try { //休眠1000ms,大于runB中的500ms, //是为了后调用waitToProceed,从而先notifyAll,后wait, Thread.sleep(1000); mnf.waitToProceed(); } catch ( InterruptedException x ) { x.printStackTrace(); } } }; Thread threadA = new Thread(runA, "threadA"); threadA.start(); Runnable runB = new Runnable() { public void run() { try { //休眠500ms,小于runA中的1000ms, //是为了先调用proceed,从而先notifyAll,后wait, Thread.sleep(500); mnf.proceed(); } catch ( InterruptedException x ) { x.printStackTrace(); } } }; Thread threadB = new Thread(runB, "threadB"); threadB.start(); try { Thread.sleep(10000); } catch ( InterruptedException x ) {} print("about to invoke interrupt() on threadA"); threadA.interrupt(); } }
执行结果如下:
注意代码中加了注释的部分,在threadB进行通知之前,先将okToProceed置为true,这样如果threadA将通知遗漏,那么就不会进入while循环,也便不会执行wait方法,线程也就不会阻塞。如果通知没有被遗漏,wait方法返回后,okToProceed已经被置为true,下次while循环判断条件不成立,便会退出循环。
这样,通过标志位和wait、notifyAll的配合使用,便避免了通知遗漏而造成的阻塞问题。
总结:在使用线程的等待/通知机制时,一般都要配合一个boolean变量值(或者其他能够判断真假的条件),在notify之前改变该boolean变量的值,让wait返回后能够退出while循环(一般都要在wait方法外围加一层while循环,以防止早期通知),或在通知被遗漏后,不会被阻塞在wait方法处。这样便保证了程序的正确性。