Java 死鎖

什麼是死鎖？

當一個線程永遠地持有一個鎖，而且其餘線程都嘗試去得到這個鎖時，那麼它們將永遠被阻塞，當線程A持有鎖1想獲取鎖2，當線程B持有鎖2想獲取鎖1 這種狀況下就會產生2個線程一直在阻塞等待其餘線程釋放鎖1 鎖2 就會形成死鎖。

下面咱們來看一下死鎖的示例：

package cn.com.example;

/**
 * Created by Jack on 2017/2/10.
 */
public class DeadLock {

    private static Object o1 = new Object(), o2 = new Object();

    public static void main(String[] args) {
        Thread t1 = new Thread(new DeadLockThread(1, o1, o2));
        Thread t2 = new Thread(new DeadLockThread(2, o1, o2));

        t1.start();
        t2.start();
    }
}

class DeadLockThread implements Runnable {
    private int i;

    private Object o1;

    private Object o2;

    public DeadLockThread(int i, Object o1, Object o2) {
        this.i = i;
        this.o1 = o1;
        this.o2 = o2;
    }

    @Override
    public void run() {
        if (i == 1) {
            synchronized (o1) {
                try {
                    Thread.sleep(500);
                } catch (InterruptedException e) {
                    e.printStackTrace();
                }
                System.out.println("i = 1 synchronized o1");
                synchronized (o2) {
                    System.out.println("i = 1 synchronized o2");
                }
            }
        } else if (i == 2) {
            synchronized (o2) {
                try {
                    Thread.sleep(500);
                } catch (InterruptedException e) {
                    e.printStackTrace();
                }
                System.out.println("i = 2 synchronized o2");
                synchronized (o1) {
                    System.out.println("i = 2 synchronized o1");
                }
            }
        }
    }
}

輸出：

i = 1 synchronized o1
i = 2 synchronized o2

後面什麼都不輸出了， 爲何要 Thread.sleep(500);  是怕 某個線程先運行 搶佔了 o1 o2 的鎖。