欧美bbbwbbbw肥妇,免费乱码人妻系列日韩,一级黄片

java多線程-讀寫鎖原理

 更新時(shí)間:2016年09月02日 16:34:08   作者:DemonWang  
本文主要介紹java多線程的知識(shí),這里整理了相關(guān)資料及簡(jiǎn)單示例代碼,有興趣的小伙伴可以參考下

Java5 在 java.util.concurrent 包中已經(jīng)包含了讀寫鎖。盡管如此,我們還是應(yīng)該了解其實(shí)現(xiàn)背后的原理。

  1. 讀/寫鎖的 Java 實(shí)現(xiàn)(Read / Write Lock Java Implementation)
  2. 讀/寫鎖的重入(Read / Write Lock Reentrance)
  3. 讀鎖重入(Read Reentrance)
  4. 寫鎖重入(Write Reentrance)
  5. 讀鎖升級(jí)到寫鎖(Read to Write Reentrance)
  6. 寫鎖降級(jí)到讀鎖(Write to Read Reentrance)
  7. 可重入的 ReadWriteLock 的完整實(shí)現(xiàn)(Fully Reentrant ReadWriteLock)
  8. 在 finally 中調(diào)用 unlock() (Calling unlock() from a finally-clause)

讀/寫鎖的 Java 實(shí)現(xiàn)

先讓我們對(duì)讀寫訪問資源的條件做個(gè)概述:

讀取 沒有線程正在做寫操作,且沒有線程在請(qǐng)求寫操作。

寫入 沒有線程正在做讀寫操作。

如果某個(gè)線程想要讀取資源,只要沒有線程正在對(duì)該資源進(jìn)行寫操作且沒有線程請(qǐng)求對(duì)該資源的寫操作即可。我們假設(shè)對(duì)寫操作的請(qǐng)求比對(duì)讀操作的請(qǐng)求更重要,就要提升寫請(qǐng)求的優(yōu)先級(jí)。此外,如果讀操作發(fā)生的比較頻繁,我們又沒有提升寫操作的優(yōu)先級(jí),那么就會(huì)產(chǎn)生“饑餓”現(xiàn)象。請(qǐng)求寫操作的線程會(huì)一直阻塞,直到所有的讀線程都從 ReadWriteLock 上解鎖了。如果一直保證新線程的讀操作權(quán)限,那么等待寫操作的線程就會(huì)一直阻塞下去,結(jié)果就是發(fā)生“饑餓”。因此,只有當(dāng)沒有線程正在鎖住 ReadWriteLock 進(jìn)行寫操作,且沒有線程請(qǐng)求該鎖準(zhǔn)備執(zhí)行寫操作時(shí),才能保證讀操作繼續(xù)。

當(dāng)其它線程沒有對(duì)共享資源進(jìn)行讀操作或者寫操作時(shí),某個(gè)線程就有可能獲得該共享資源的寫鎖,進(jìn)而對(duì)共享資源進(jìn)行寫操作。有多少線程請(qǐng)求了寫鎖以及以何種順序請(qǐng)求寫鎖并不重要,除非你想保證寫鎖請(qǐng)求的公平性。

按照上面的敘述,簡(jiǎn)單的實(shí)現(xiàn)出一個(gè)讀/寫鎖,代碼如下

public class ReadWriteLock{
 private int readers = 0;
 private int writers = 0;
 private int writeRequests = 0;

 public synchronized void lockRead() 
  throws InterruptedException{
  while(writers > 0 || writeRequests > 0){
   wait();
  }
  readers++;
 }

 public synchronized void unlockRead(){
  readers--;
  notifyAll();
 }

 public synchronized void lockWrite() 
  throws InterruptedException{
  writeRequests++;

  while(readers > 0 || writers > 0){
   wait();
  }
  writeRequests--;
  writers++;
 }

 public synchronized void unlockWrite() 
  throws InterruptedException{
  writers--;
  notifyAll();
 }
}

ReadWriteLock 類中,讀鎖和寫鎖各有一個(gè)獲取鎖和釋放鎖的方法。

讀鎖的實(shí)現(xiàn)在 lockRead()中,只要沒有線程擁有寫鎖(writers==0),且沒有線程在請(qǐng)求寫鎖(writeRequests ==0),所有想獲得讀鎖的線程都能成功獲取。

寫鎖的實(shí)現(xiàn)在 lockWrite()中,當(dāng)一個(gè)線程想獲得寫鎖的時(shí)候,首先會(huì)把寫鎖請(qǐng)求數(shù)加 1(writeRequests++),然后再去判斷是否能夠真能獲得寫鎖,當(dāng)沒有線程持有讀鎖(readers==0 ),且沒有線程持有寫鎖(writers==0)時(shí)就能獲得寫鎖。有多少線程在請(qǐng)求寫鎖并無(wú)關(guān)系。

需要注意的是,在兩個(gè)釋放鎖的方法(unlockRead,unlockWrite)中,都調(diào)用了 notifyAll 方法,而不是 notify。要解釋這個(gè)原因,我們可以想象下面一種情形:

如果有線程在等待獲取讀鎖,同時(shí)又有線程在等待獲取寫鎖。如果這時(shí)其中一個(gè)等待讀鎖的線程被 notify 方法喚醒,但因?yàn)榇藭r(shí)仍有請(qǐng)求寫鎖的線程存在(writeRequests>0),所以被喚醒的線程會(huì)再次進(jìn)入阻塞狀態(tài)。然而,等待寫鎖的線程一個(gè)也沒被喚醒,就像什么也沒發(fā)生過一樣(譯者注:信號(hào)丟失現(xiàn)象)。如果用的是 notifyAll 方法,所有的線程都會(huì)被喚醒,然后判斷能否獲得其請(qǐng)求的鎖。

用 notifyAll 還有一個(gè)好處。如果有多個(gè)讀線程在等待讀鎖且沒有線程在等待寫鎖時(shí),調(diào)用 unlockWrite()后,所有等待讀鎖的線程都能立馬成功獲取讀鎖 —— 而不是一次只允許一個(gè)。

讀/寫鎖的重入

上面實(shí)現(xiàn)的讀/寫鎖(ReadWriteLock) 是不可重入的,當(dāng)一個(gè)已經(jīng)持有寫鎖的線程再次請(qǐng)求寫鎖時(shí),就會(huì)被阻塞。原因是已經(jīng)有一個(gè)寫線程了——就是它自己。此外,考慮下面的例子:

  1. Thread 1 獲得了讀鎖。
  2. Thread 2 請(qǐng)求寫鎖,但因?yàn)?Thread 1 持有了讀鎖,所以寫鎖請(qǐng)求被阻塞。
  3. Thread 1 再想請(qǐng)求一次讀鎖,但因?yàn)?Thread 2 處于請(qǐng)求寫鎖的狀態(tài),所以想再次獲取讀鎖也會(huì)被阻塞。 上面這種情形使用前面的 ReadWriteLock 就會(huì)被鎖定——一種類似于死鎖的情形。不會(huì)再有線程能夠成功獲取讀鎖或?qū)戞i了。

為了讓 ReadWriteLock 可重入,需要對(duì)它做一些改進(jìn)。下面會(huì)分別處理讀鎖的重入和寫鎖的重入。

讀鎖重入

為了讓 ReadWriteLock 的讀鎖可重入,我們要先為讀鎖重入建立規(guī)則:

要保證某個(gè)線程中的讀鎖可重入,要么滿足獲取讀鎖的條件(沒有寫或?qū)懻?qǐng)求),要么已經(jīng)持有讀鎖(不管是否有寫請(qǐng)求)。 要確定一個(gè)線程是否已經(jīng)持有讀鎖,可以用一個(gè) map 來存儲(chǔ)已經(jīng)持有讀鎖的線程以及對(duì)應(yīng)線程獲取讀鎖的次數(shù),當(dāng)需要判斷某個(gè)線程能否獲得讀鎖時(shí),就利用 map 中存儲(chǔ)的數(shù)據(jù)進(jìn)行判斷。下面是方法 lockRead 和 unlockRead 修改后的的代碼:

public class ReadWriteLock{
 private Map<Thread, Integer> readingThreads =
  new HashMap<Thread, Integer>();

 private int writers = 0;
 private int writeRequests = 0;

 public synchronized void lockRead() 
  throws InterruptedException{
  Thread callingThread = Thread.currentThread();
  while(! canGrantReadAccess(callingThread)){
   wait();                 
  }

  readingThreads.put(callingThread,
   (getAccessCount(callingThread) + 1));
 }

 public synchronized void unlockRead(){
  Thread callingThread = Thread.currentThread();
  int accessCount = getAccessCount(callingThread);
  if(accessCount == 1) { 
   readingThreads.remove(callingThread); 
  } else {
   readingThreads.put(callingThread, (accessCount -1)); 
  }
  notifyAll();
 }

 private boolean canGrantReadAccess(Thread callingThread){
  if(writers > 0) return false;
  if(isReader(callingThread) return true;
  if(writeRequests > 0) return false;
  return true;
 }

 private int getReadAccessCount(Thread callingThread){
  Integer accessCount = readingThreads.get(callingThread);
  if(accessCount == null) return 0;
  return accessCount.intValue();
 }

 private boolean isReader(Thread callingThread){
  return readingThreads.get(callingThread) != null;
 }
}

代碼中我們可以看到,只有在沒有線程擁有寫鎖的情況下才允許讀鎖的重入。此外,重入的讀鎖比寫鎖優(yōu)先級(jí)高。

寫鎖重入

僅當(dāng)一個(gè)線程已經(jīng)持有寫鎖,才允許寫鎖重入(再次獲得寫鎖)。下面是方法 lockWrite 和 unlockWrite 修改后的的代碼。

public class ReadWriteLock{
 private Map<Thread, Integer> readingThreads =
  new HashMap<Thread, Integer>();

 private int writeAccesses = 0;
 private int writeRequests = 0;
 private Thread writingThread = null;

 public synchronized void lockWrite() 
  throws InterruptedException{
  writeRequests++;
  Thread callingThread = Thread.currentThread();
  while(!canGrantWriteAccess(callingThread)){
   wait();
  }
  writeRequests--;
  writeAccesses++;
  writingThread = callingThread;
 }

 public synchronized void unlockWrite() 
  throws InterruptedException{
  writeAccesses--;
  if(writeAccesses == 0){
   writingThread = null;
  }
  notifyAll();
 }

 private boolean canGrantWriteAccess(Thread callingThread){
  if(hasReaders()) return false;
  if(writingThread == null) return true;
  if(!isWriter(callingThread)) return false;
  return true;
 }

 private boolean hasReaders(){
  return readingThreads.size() > 0;
 }

 private boolean isWriter(Thread callingThread){
  return writingThread == callingThread;
 }
}

注意在確定當(dāng)前線程是否能夠獲取寫鎖的時(shí)候,是如何處理的。

讀鎖升級(jí)到寫鎖

有時(shí),我們希望一個(gè)擁有讀鎖的線程,也能獲得寫鎖。想要允許這樣的操作,要求這個(gè)線程是唯一一個(gè)擁有讀鎖的線程。writeLock()需要做點(diǎn)改動(dòng)來達(dá)到這個(gè)目的:

public class ReadWriteLock{
 private Map<Thread, Integer> readingThreads =
  new HashMap<Thread, Integer>();

 private int writeAccesses = 0;
 private int writeRequests = 0;
 private Thread writingThread = null;

 public synchronized void lockWrite() 
  throws InterruptedException{
  writeRequests++;
  Thread callingThread = Thread.currentThread();
  while(!canGrantWriteAccess(callingThread)){
   wait();
  }
  writeRequests--;
  writeAccesses++;
  writingThread = callingThread;
 }

 public synchronized void unlockWrite() throws InterruptedException{
  writeAccesses--;
  if(writeAccesses == 0){
   writingThread = null;
  }
  notifyAll();
 }

 private boolean canGrantWriteAccess(Thread callingThread){
  if(isOnlyReader(callingThread)) return true;
  if(hasReaders()) return false;
  if(writingThread == null) return true;
  if(!isWriter(callingThread)) return false;
  return true;
 }

 private boolean hasReaders(){
  return readingThreads.size() > 0;
 }

 private boolean isWriter(Thread callingThread){
  return writingThread == callingThread;
 }

 private boolean isOnlyReader(Thread thread){
  return readers == 1 && readingThreads.get(callingThread) != null;
 }
}

現(xiàn)在 ReadWriteLock 類就可以從讀鎖升級(jí)到寫鎖了。

寫鎖降級(jí)到讀鎖

有時(shí)擁有寫鎖的線程也希望得到讀鎖。如果一個(gè)線程擁有了寫鎖,那么自然其它線程是不可能擁有讀鎖或?qū)戞i了。所以對(duì)于一個(gè)擁有寫鎖的線程,再獲得讀鎖,是不會(huì)有什么危險(xiǎn)的。我們僅僅需要對(duì)上面 canGrantReadAccess 方法進(jìn)行簡(jiǎn)單地修改:

public class ReadWriteLock{
 private boolean canGrantReadAccess(Thread callingThread){
  if(isWriter(callingThread)) return true;
  if(writingThread != null) return false;
  if(isReader(callingThread) return true;
  if(writeRequests > 0) return false;
  return true;
 }
}

可重入的 ReadWriteLock 的完整實(shí)現(xiàn)

下面是完整的 ReadWriteLock 實(shí)現(xiàn)。為了便于代碼的閱讀與理解,簡(jiǎn)單對(duì)上面的代碼做了重構(gòu)。重構(gòu)后的代碼如下。

public class ReadWriteLock{
 private Map<Thread, Integer> readingThreads =
  new HashMap<Thread, Integer>();

 private int writeAccesses = 0;
 private int writeRequests = 0;
 private Thread writingThread = null;

 public synchronized void lockRead() 
  throws InterruptedException{
  Thread callingThread = Thread.currentThread();
  while(! canGrantReadAccess(callingThread)){
   wait();
  }

  readingThreads.put(callingThread,
   (getReadAccessCount(callingThread) + 1));
 }

 private boolean canGrantReadAccess(Thread callingThread){
  if(isWriter(callingThread)) return true;
  if(hasWriter()) return false;
  if(isReader(callingThread)) return true;
  if(hasWriteRequests()) return false;
  return true;
 }

 public synchronized void unlockRead(){
  Thread callingThread = Thread.currentThread();
  if(!isReader(callingThread)){
   throw new IllegalMonitorStateException(
    "Calling Thread does not" +
    " hold a read lock on this ReadWriteLock");
  }
  int accessCount = getReadAccessCount(callingThread);
  if(accessCount == 1){ 
   readingThreads.remove(callingThread); 
  } else { 
   readingThreads.put(callingThread, (accessCount -1));
  }
  notifyAll();
 }

 public synchronized void lockWrite() 
  throws InterruptedException{
  writeRequests++;
  Thread callingThread = Thread.currentThread();
  while(!canGrantWriteAccess(callingThread)){
   wait();
  }
  writeRequests--;
  writeAccesses++;
  writingThread = callingThread;
 }

 public synchronized void unlockWrite() 
  throws InterruptedException{
  if(!isWriter(Thread.currentThread()){
  throw new IllegalMonitorStateException(
   "Calling Thread does not" +
   " hold the write lock on this ReadWriteLock");
  }
  writeAccesses--;
  if(writeAccesses == 0){
   writingThread = null;
  }
  notifyAll();
 }

 private boolean canGrantWriteAccess(Thread callingThread){
  if(isOnlyReader(callingThread)) return true;
  if(hasReaders()) return false;
  if(writingThread == null) return true;
  if(!isWriter(callingThread)) return false;
  return true;
 }

 private int getReadAccessCount(Thread callingThread){
  Integer accessCount = readingThreads.get(callingThread);
  if(accessCount == null) return 0;
  return accessCount.intValue();
 }

 private boolean hasReaders(){
  return readingThreads.size() > 0;
 }

 private boolean isReader(Thread callingThread){
  return readingThreads.get(callingThread) != null;
 }

 private boolean isOnlyReader(Thread callingThread){
  return readingThreads.size() == 1 &&
   readingThreads.get(callingThread) != null;
 }

 private boolean hasWriter(){
  return writingThread != null;
 }

 private boolean isWriter(Thread callingThread){
  return writingThread == callingThread;
 }

 private boolean hasWriteRequests(){
  return this.writeRequests > 0;
 }
}

在 finally 中調(diào)用 unlock()

在利用 ReadWriteLock 來保護(hù)臨界區(qū)時(shí),如果臨界區(qū)可能拋出異常,在 finally 塊中調(diào)用 readUnlock()和 writeUnlock()就顯得很重要了。這樣做是為了保證 ReadWriteLock 能被成功解鎖,然后其它線程可以請(qǐng)求到該鎖。這里有個(gè)例子:

lock.lockWrite();
try{
 //do critical section code, which may throw exception
} finally {
 lock.unlockWrite();
}

上面這樣的代碼結(jié)構(gòu)能夠保證臨界區(qū)中拋出異常時(shí) ReadWriteLock 也會(huì)被釋放。如果 unlockWrite 方法不是在 finally 塊中調(diào)用的,當(dāng)臨界區(qū)拋出了異常時(shí),ReadWriteLock 會(huì)一直保持在寫鎖定狀態(tài),就會(huì)導(dǎo)致所有調(diào)用 lockRead()或 lockWrite()的線程一直阻塞。唯一能夠重新解鎖 ReadWriteLock 的因素可能就是 ReadWriteLock 是可重入的,當(dāng)拋出異常時(shí),這個(gè)線程后續(xù)還可以成功獲取這把鎖,然后執(zhí)行臨界區(qū)以及再次調(diào)用 unlockWrite(),這就會(huì)再次釋放 ReadWriteLock。但是如果該線程后續(xù)不再獲取這把鎖了呢?所以,在 finally 中調(diào)用 unlockWrite 對(duì)寫出健壯代碼是很重要的。

 以上就是對(duì)java  多線程的資料整理,后續(xù)繼續(xù)補(bǔ)充相關(guān)資料,謝謝大家對(duì)本站的支持!

相關(guān)文章

最新評(píng)論