关于java多线程中的await和signalAll的使用问题

Question

package test.synchornize; import java.util.concurrent.locks.*; class runab implements Runnable{ boolean x = true; static Lock lock = new ReentrantLock(); static Condition con1 = lock.newCondition(); static Condition con2 = lock.newCondition(); runab(boolean x){  this.x = x; } public void run() {  lock.lock();  try {   System.out.println("Start---" + Thread.currentThread().getName());   if(x) {    x = false;    try {con1.await();con2.await();} catch (InterruptedException e) {}   }   System.out.println("End---" + Thread.currentThread().getName());  } finally {    con2.signalAll();   con1.signalAll();   lock.unlock();  } }} public class showLock {  public static void main(String[] args) {  runab rab = new runab(true);  runab rab1 = new runab(false);  Thread th = new Thread(rab);  Thread th1 = new Thread(rab1);  th.start();  try {   Thread.sleep(100);  } catch (InterruptedException e) {   // TODO Auto-generated catch block   e.printStackTrace();  }  th1.start();   } }   请问finally里面怎样写才能，让线程th继续下去呢？（即让程序最后一行输出 “End---Thread-0”）

Answer 1

呵，你这怎么写线程th都执行不下去啊，自己把自己锁死了。con1.await()和con2.await()是释放锁，线程进入阻塞状态并等待其他线程唤醒，finally根本就不会执行，那必然不会唤醒。你这con1和con2就不应该是对象内的局部变量，应该是线程间的共享变量，才能在自己等待的时候，被其他线程唤醒。