面试的时候可能会让你写一段死锁的代码,其实如果对死锁理解深刻,写出来并不难。
其中一个典型场景,就是一个线程持有A锁,然后请求获取B锁。另外一个线程正好相反,持有B锁,等待获取A锁。
模拟死锁的线程代码
public class DeadLockThread extends Thread {
private final Lock lock1;
private final Lock lock2;
public DeadLockThread(Lock lock1, Lock lock2) {
this.lock1 = lock1;
this.lock2 = lock2;
}
@Override
public void run() {
synchronized (lock1) {
try {
Thread.sleep(5000);
System.out.println(Thread.currentThread().getName() + "先拿" + lock1.getName() + ",然后尝试获取锁" + lock2.getName());
synchronized (lock2) {
System.out.println(Thread.currentThread().getName() + "获取到锁" + lock2.getName());
}
} catch (InterruptedException e) {
e.printStackTrace();
}
}
}
}
测试代码:
public class DeadLockTest {
public static void main(String[] args) {
Lock lock1 = new Lock("[锁1]");
Lock lock2 = new Lock("[锁2]");
Thread thread1 = new DeadLockThread(lock1, lock2);
thread1.setName("<线程1>");
Thread thread2 = new DeadLockThread(lock2, lock1);
thread2.setName("<线程2>");
thread1.start();
thread2.start();
}
}
运行结果
根据这个具体案例,该如何避免死锁呢?
第一种最简单的方案,就是顺序获取锁。
public class DeadLockTest {
public static void main(String[] args) {
Lock lock1 = new Lock("[锁1]");
Lock lock2 = new Lock("[锁2]");
Thread thread1 = new DeadLockThread(lock1, lock2);
thread1.setName("<线程1>");
Thread thread2 = new DeadLockThread(lock1, lock2);
thread2.setName("<线程2>");
thread1.start();
thread2.start();
}
}
或者线程依次执行
或者检测到死锁,强制中断其中一个线程。