public class SyncBlock {
  static class DataWrap {
    int i;
  }
  
  static class SyncBlockThread extends Thread {
    private DataWrap date;
    
    public SyncBlockThread(DataWrap dataWrap) {
      this.date = dataWrap;
    }
    
    @Override
    public void run() {
      
      for (int i = 0; i < 10; i++) {
        synchronized (date) {
          date.i++;
          try {
            sleep(1);
          } catch (InterruptedException e) {
            e.printStackTrace();
          }
          System.out.println(getName() + " " + date.i);
        }
      }
    }
  }
  
  public static void main(String[] args) {
    //多線程實(shí)現(xiàn)變量i依次加一輸出
    DataWrap dataWrap = new DataWrap();
    
    new SyncBlockThread(dataWrap).start();
    new SyncBlockThread(dataWrap).start();
    new SyncBlockThread(dataWrap).start();
  }

}

示例中希望按照順序依次輸出整數(shù)。

通常同步代碼塊是需要鎖定的對(duì)象，一般是需要并發(fā)訪問的共享資源，任何線程在修改指定資源之前都首先對(duì)該資源加鎖，在加鎖期間其它線程無法修改該資源。從而保證了線程的安全性。另外線程在調(diào)用sleep或者yield時(shí)并不會(huì)讓出資源鎖。

2、同步方法

public class SyncMethod {
  
  static class DataWrap{
    int i;
    
    public synchronized void valueGrow(){
      i++;
      
      try {
        Thread.sleep(1);
      } catch (InterruptedException e) {
        // TODO Auto-generated catch block
        e.printStackTrace();
      }
      System.out.println(Thread.currentThread().getName() + " " + i);
    }
  }
  
  static class SyncMethodThread extends Thread {
    DataWrap dataWrap;
    
    public SyncMethodThread(DataWrap dataWrap){
      this.dataWrap = dataWrap;
    }
    
    @Override
    public void run() {
      for (int i = 0; i < 10; i++) {
        dataWrap.valueGrow();  
      }
    }
  }
  
  public static void main(String[] args) {
    //實(shí)現(xiàn)順序增長(zhǎng)并輸出Datawrap中的i
    
    DataWrap dataWrap = new DataWrap();
    
    new SyncMethodThread(dataWrap).start();
    new SyncMethodThread(dataWrap).start();
    new SyncMethodThread(dataWrap).start();
  }

}

同步方法是使用synchronized關(guān)鍵字修飾的某個(gè)方法，同步方法鎖定的就是該對(duì)象本身，所以當(dāng)一個(gè)線程調(diào)用了某個(gè)對(duì)象的同步方法后，如有其它線程調(diào)用該對(duì)象的其它同步方法，也依然要等待釋放該對(duì)象的鎖，因?yàn)樵搶?duì)象已被鎖定。

3、同步鎖

通過定義同步鎖對(duì)象實(shí)現(xiàn)同步，這種情況下，同步鎖使用Lock對(duì)象充當(dāng)。

import java.util.concurrent.locks.Lock;
import java.util.concurrent.locks.ReentrantLock;

public class SyncLock {

  static class DataWrap{
    Lock lock = new ReentrantLock();
    int i;
    
    public void valueGrow(){
      lock.lock();
      try {
        i++;
        
        try {
          Thread.sleep(1);
        } catch (InterruptedException e) {
          // TODO Auto-generated catch block
          e.printStackTrace();
        }
        System.out.println(Thread.currentThread().getName() + " " + i);  
      } finally {
        lock.unlock();
      }
      
    }
  }
  
  static class SyncLockThread extends Thread {
    DataWrap dataWrap;
    
    public SyncLockThread(DataWrap dataWrap){
      this.dataWrap = dataWrap;
    }
    
    @Override
    public void run() {
      for (int i = 0; i < 10; i++) {
        dataWrap.valueGrow();  
      }
    }
  }
  
  public static void main(String[] args) {
    //實(shí)現(xiàn)順序增長(zhǎng)并輸出Datawrap中的i
    
    DataWrap dataWrap = new DataWrap();
        
    new SyncLockThread(dataWrap).start();
    new SyncLockThread(dataWrap).start();
    new SyncLockThread(dataWrap).start();
  }

}

使用鎖對(duì)象實(shí)現(xiàn)線程同步會(huì)更靈活一些，某些鎖還具有一些特定的功能，其中比較常用的ReadWriteLock讀寫鎖，ReentrantLock可重入鎖。

您可能感興趣的文章: