線程執行序列

Question

在下面的代碼，我試圖找出線程在synchronized塊處理執行：

public class ThreadExp { 
    public static void main(String[] args) throws InterruptedException { 
     MyRunnable r = new MyRunnable(); 
     Thread t1 = new Thread(r, "Thread1"); 
     Thread t2 = new Thread(r, "Thread2"); 
     t1.start(); 
     t2.start(); 
    } 

} 

class MyRunnable implements Runnable { 

    @Override 
    public void run() { 
     callFn(); 
    } 

    private void callFn() { 
     System.out.println(Thread.currentThread().getName() + ": entered callSync()"); 
     synchronized (Thread.currentThread()) { 
      System.out.println(Thread.currentThread().getName() + ": inside sync"); 
      try { 
       Thread.currentThread().sleep(5000); 
      } catch (InterruptedException e) { 
       e.printStackTrace(); 
      } 
      System.out.println(Thread.currentThread().getName() + " leaving sync"); 
     } 
     System.out.println(Thread.currentThread().getName() + ": leaving callSync()"); 
    } 
} 

實際輸出：

Thread1: entered callFn() 
    Thread2: entered callFn() 
    Thread1: inside sync block 
    Thread2: inside sync block 
// 5000 millisec sleep 
    Thread2 leaving sync block 
    Thread1 leaving sync block 
    Thread2: leaving callFn() 
    Thread1: leaving callFn() 

而我預期的東西如：

Thread1: entered callFn() 
Thread2: entered callFn() 
Thread1: inside sync block 
// 5000 millisec sleep 
Thread1 leaving sync block 
Thread1: leaving callFn() 
Thread2: inside sync block 
Thread2 leaving sync block 
Thread2: leaving callFn() 

總的來說，我認爲Thread1會獲得鎖，然後進入睡眠。只有在線程1完成後，線程2才能進入同步塊。

Answer 1

正在同步線程本身上：

synchronized (Thread.currentThread()) 

所以每個線程都有自己的鎖，他們可以同時執行​​塊。

如果您希望​​塊一次只能由一個線程運行，則需要使用相同的鎖。

在您的例子，它可能僅僅是this，因爲Runnable實例同樣爲你的兩個線程：

synchronized(this) {...} 

然後輸出如下：當前

Thread1: entered callSync() 
Thread1: inside sync 
Thread2: entered callSync() 
Thread1 leaving sync 
Thread1: leaving callSync() 
Thread2: inside sync 
Thread2 leaving sync 
Thread2: leaving callSync()