詳解Java 線程中斷
一、前言
大傢肯定都使用過 Java 線程開發(Thread / Runnable),啟動一個線程的做法通常是:
new Thread(new Runnable( @Override public void run() { // todo sth... } )).start();
然而線程退出,大傢是如何做的呢?一般做法可能不外乎以下兩種:
- 設置一個標志位:true / false 來退出;
- 強制退出:thread.stop;(我相信,現在應該沒人會使用這種方式瞭,因為JDK也很早就廢棄瞭該方法)
可能還會有人提出,我可以用中斷來退出線程! 我隻能說:Too Young Too Simple!中斷並不會使得線程結束而退出,中斷(interrupt)隻是喚醒被阻塞的線程而已。
本篇,我們就來好好的聊聊:線程中斷,以及如何正確的使用線程中斷,和正確的線程退出。
二、為何 Thread.stop 被廢棄
This method is inherently unsafe. Stopping a thread with Thread.stop causes it to unlock all of the monitors that it has locked (as a natural consequence of the unchecked ThreadDeath exception propagating up the stack). If any of the objects previously protected by these monitors were in an inconsistent state, the damaged objects become visible to other threads, potentially resulting in arbitrary behavior. Many uses of stop should be replaced by code that simply modifies some variable to indicate that the target thread should stop running. The target thread should check this variable regularly, and return from its run method in an orderly fashion if the variable indicates that it is to stop running. If the target thread waits for long periods (on a condition variable, for example), the interrupt method should be used to interrupt the wait.
以上是官方 JDK 中的源碼註釋說明,其含義如下:
**Thread.stop 方法天生就不安全。**使用該方法來停止線程,將會導致其它因為監視器鎖『監視器我們在 synchronized 中就講過,是 Java 的內置鎖』而被鎖住的線程全部都解鎖!(本質的後果是:沒有檢查的 ThreadDeath 異常會在棧中傳播,因而使得監視器鎖解鎖)。如果任何一個被監視器鎖給鎖住的對象處於一個不一致的狀態,那麼其被解鎖後將會被其它線程可見,潛在的結果是產生任何後果。**我們應該使用一個變量來代替使用 stop 方法,告訴目標線程退出『這裡就是我們開頭所說的第一種方法,設置一個標志位』。**目標線程應該周期性的檢查這個變量,並根據這個變量來正確的退出 run 方法。如果目標線程處於阻塞/休眠狀態(如:使用 wait、sleep、yield 方法後,線程讓出瞭 CPU 使用權,進而阻塞/休眠),此時,該標志位變量將不會起作用,那麼,應該使用 interrupt 方法來中斷目標線程的阻塞/休眠狀態,將其喚醒!
對於 ThreadDeath 對象,官方還有補充:
- 線程可以在幾乎任何地方拋出 ThreadDeath 異常。由於這一點,所有的同步方法和(代碼)塊將必須被考慮得事無巨細。
- 線程在清理第一個 ThreadDeath 異常的時候(在 catch 或 finally 語句中),可能會拋出第二個。清理工作將不得不重復直到到其成功。保障這一點的代碼將會很復雜。
所以,我們也別想著去 try-catch ThreadDeath Exception!
同樣,被廢棄的還有 Thread.resume 和 Thread.suspend。這倆方法有造成死鎖的危險:
- 使用suspend時,並不會釋放鎖;
- 如果存在某種情況要先獲取該鎖,再進行resume,那麼就造成死鎖瞭;
取代這兩方法的正確方式是:Object.wait 和 Object.notify :
因為 Object.wait 進入阻塞時,會釋放鎖。
三、線程中斷的含義
Thread 中有三個與中斷相關的方法:
- 成員方法 interrupt():設置線程中斷標志為 true ;
- 成員方法 isInterrupted():獲取線程的中斷狀態,默認為 false,調用 interrupt() 後,該方法返回 true;
- 靜態方法 Thread.interrupted():獲取線程的中斷狀態,並且清除中斷狀態(設置為 false);
註:如果線程中斷後,連續兩次調用 Thread.interrupted(),第一次是 true & 清除狀態,第二次結果是 false。
3.1、初步瞭解
我們先來通過一個例子來初步瞭解 thread.interrupt :
public class InterruptDemo implements Runnable { @Override public void run() { while (true) { System.out.println("Thread running..."); } } public static void main(String[] args) throws InterruptedException { Thread thread = new Thread(new InterruptDemo(), "InterruptDemo"); System.out.println("start thread"); thread.start(); Thread.sleep(50); System.out.println("interrupt thread"); thread.interrupt(); Thread.sleep(50); System.out.println("thread's status = " + thread.isInterrupted()); } }
輸出結果:
start thread Thread running... Thread running... ...... interrupt thread Thread running... Thread running... ...... thread's status = true Thread running... ......
我們可以看到,即便我們調用瞭 thread.interrupt 方法,線程也並沒有退出,仍舊繼續運行。因此,這個例子證明瞭一點:我們並不能通過”我們所認為的”中斷來試圖”結束”正在運行的線程。
3.2、中斷即喚醒阻塞/休眠的線程
同樣,我們再來看一個例子:
public class InterruptDemo implements Runnable { @Override public void run() { while (true) { System.out.println("Thread will sleep 10s ------------------------- running"); long timestamp = System.currentTimeMillis(); try { Thread.sleep(10000); } catch (InterruptedException e) { System.out.println("thread interrupted..."); } timestamp = System.currentTimeMillis() - timestamp; System.out.println("Thread run, total sleep = " + timestamp + "(ms)"); } } public static void main(String[] args) throws InterruptedException { Thread thread = new Thread(new InterruptDemo(), "InterruptDemo"); System.out.println("start thread"); thread.start(); Thread.sleep(3000); System.out.println("interrupt thread"); thread.interrupt(); System.out.println("main exit"); } }
輸出結果:
start thread Thread will sleep 10s ------------------------- running interrupt thread main exit thread interrupted... Thread run, total sleep = 3002(ms) Thread will sleep 10s ------------------------- running Thread run, total sleep = 10002(ms) Thread will sleep 10s ------------------------- running
我們可以看到,線程啟動後,進入睡眠(10s),3秒後被中斷喚醒,執行完一個 while 後再次進入第二次睡眠(10s),然後周而復始。
3.3、一般標志位法退出線程
public class InterruptDemo implements Runnable { private static final AtomicBoolean running = new AtomicBoolean(true); @Override public void run() { while (running.get()) { long timestamp = System.currentTimeMillis(); timestamp = System.currentTimeMillis() - timestamp; System.out.println("Thread run, total sleep = " + timestamp + "(ms)"); } System.out.println("Thread exit"); } public static void main(String[] args) throws InterruptedException { Thread thread = new Thread(new InterruptDemo(), "InterruptDemo"); System.out.println("start thread"); thread.start(); Thread.sleep(100); System.out.println("interrupt thread"); thread.interrupt(); running.set(false); System.out.println("main exit"); } }
輸出結果:
start thread ....... Thread run, total sleep = 0(ms) interrupt thread Thread run, total sleep = 0(ms) Thread run, total sleep = 0(ms) Thread run, total sleep = 0(ms) main exit Thread exit
我們通過使用一個 AtomicBoolean 變量來當作標志位,使得我們的線程能正常退出。 我們也可以判斷線程是否被中斷而選擇性的退出。
3.4、線程中斷退出
public class InterruptDemo implements Runnable { @Override public void run() { while (!Thread.currentThread().isInterrupted()) { long timestamp = System.currentTimeMillis(); timestamp = System.currentTimeMillis() - timestamp; System.out.println("Thread run, total sleep = " + timestamp + "(ms)"); } System.out.println("Thread exit"); } public static void main(String[] args) throws InterruptedException { Thread thread = new Thread(new InterruptDemo(), "InterruptDemo"); System.out.println("start thread"); thread.start(); Thread.sleep(100); System.out.println("interrupt thread"); thread.interrupt(); System.out.println("main exit"); } }
輸出結果:
start thread ....... Thread run, total sleep = 0(ms) interrupt thread Thread run, total sleep = 0(ms) Thread run, total sleep = 0(ms) Thread run, total sleep = 0(ms) main exit Thread exit
3.5、標志位 + 線程中斷結合
public class InterruptDemo implements Runnable { private static final AtomicBoolean running = new AtomicBoolean(true); @Override public void run() { while (running.get()) { System.out.println("Thread will sleep 10s ------------------------- running"); long timestamp = System.currentTimeMillis(); try { Thread.sleep(10000); } catch (InterruptedException e) { System.out.println("Interrupted... Todo other things then exit......"); running.set(false); continue; } timestamp = System.currentTimeMillis() - timestamp; System.out.println("Thread run, total sleep = " + timestamp + "(ms)"); } System.out.println("Thread exit"); } public static void main(String[] args) throws InterruptedException { Thread thread = new Thread(new InterruptDemo(), "InterruptDemo"); System.out.println("start thread"); thread.start(); Thread.sleep(3000); System.out.println("interrupt thread"); thread.interrupt(); System.out.println("main exit"); } }
輸出結果:
start thread Thread will sleep 10s ------------------------- running interrupt thread main exit Interrupted... Todo other things then exit...... Thread exit
四、總結
本文我們分析瞭線程的中斷,並讓大傢瞭解瞭中斷的含義:隻是告訴該線程,你被『中斷』瞭,至於你想幹嘛,還是由你自己來決定。同時,我們也簡單分析瞭幾個廢棄的方法的原因。希望大傢學習瞭本文之後,能正確且合理的設計,線程如何安全的退出。
五、附錄
- Object.wait:阻塞當前線程,釋放持有的鎖;
- Object.notify:喚醒當前對象上被阻塞的線程,使其進入就緒狀態;
- Object.notifyAll:喚醒所有線程;
- Thread.sleep:指定當前線程休眠一定時間,讓出CPU,但不會釋放同步資源鎖;
- Thread.yield:讓出CPU使用權,讓自己和其它線程來爭奪使用CPU的機會,因此,使用此方法後,並不能保證該線程又再次拿到CPU而恢復運行(使用此方法後,優先級高的線程拿到CPU的概率較大,但優先級低的線程也有概率拿到CPU而執行),同理不會釋放同步資源鎖;
以上就是詳解Java 線程中斷的詳細內容,更多關於Java 線程中斷的資料請關註WalkonNet其它相關文章!
推薦閱讀:
- 一篇文章掌握Java Thread的類及其常見方法
- 詳解Java停止線程的四種方法
- Java線程的停止實現原理詳解
- java 兩階段終止線程的正確做法
- Java線程中的常見方法(start方法和run方法)