ThreadLocal類分析
首先試想一個場景:
多個線程都要訪問數據庫,先要獲得一個Connection,然後執行一些操作。為了線程安全,如果用synchronized鎖定一個Connection對象,那麽任何時候,都只有一個線程能通過Connection對象操作數據庫。這樣的話,程序的效率太低。反過來,如果每次需要Connection對象就去new一個的話,就會同時存在數量龐大的數據庫連接,你受得了,數據庫受不了。於是就有人提出折中方案:為每個線程只生成一個Connection對象,這樣別的線程訪問不到這個對象,線程安全問題解決;而且無論線程有多少地方需要數據庫連接,都是在復用這個Connection對象,數據庫的壓力會小很多。
其實不僅僅是數據庫,其它的場景比如說,SimpleDateFormat。我們處理日期的時候,經常要用到這個類,但是這個類不是線程安全的,在多線程下是會出問題的。這時候,采用上述折中方案是比較合理的。
那麽如何實現這種折中方案呢?我們先動手試一試唄!!!
要確保某類型的變量,每個線程只有一份。因為每個線程的ID是唯一的,這是JVM保證的,所有我們可以定義一個Map:線程ID作為key,我們要用的變量作為value。
稍微對這個Map進行簡單的封裝,當做一個類來用:
package threadlocal; import java.util.HashMap; import java.util.Map;public class ThreadLocalVar<T> { Map<Long, T> threadVarMap = new HashMap<Long, T>(); public T get() { return threadVarMap.get(Thread.currentThread().getId()); } public void set(T value) { threadVarMap.put(Thread.currentThread().getId(), value); } }
接下來,就把這個類扔到多線程環境裏面練一練
package threadlocal; public class MyTest { ThreadLocalVar<Long> longLocal = new ThreadLocalVar<Long>(); ThreadLocalVar<String> stringLocal = new ThreadLocalVar<String>(); public void set() { longLocal.set(Thread.currentThread().getId()); stringLocal.set(Thread.currentThread().getName()); } public long getLong() { return longLocal.get(); } public String getString() { return stringLocal.get(); } public static void main(String[] args) throws InterruptedException { final MyTest test = new MyTest(); test.set(); System.out.println(test.getLong()); System.out.println(test.getString()); for (int i=0; i<3; i++) { Thread thread1 = new Thread(){ public void run() { test.set(); System.out.println(test.getLong()); System.out.println(test.getString()); }; }; thread1.start(); thread1.join(); } System.out.println(test.getLong()); System.out.println(test.getString()); } }
這個程序很簡單,看一遍就能明白具體邏輯。雖然都是調用的同一個對象test的getLong和getString方法,但是不同的線程獲取到的值不一樣。
運行結果:
1 main 9 Thread-0 10 Thread-1 11 Thread-2 1 mainView Code
哈哈,我們就是使用了奇淫巧技,把一個對象簡單的get和set操作,轉到了對Map的get和set操作。如果光看MyTest這個類,再看結果,還是挺迷惑的吧。
這個時候就有人說了,Java的ThreadLocal機制,不是這麽實現的。對,也不對。JDK之前的老版本其實就是這麽實現來著,不過後來改了。為什麽改,且聽我慢慢道來。
先上一個真正的ThreadLocal版本的test程序:
package threadlocal; public class Test { ThreadLocal<Long> longLocal = new ThreadLocal<Long>(); ThreadLocal<String> stringLocal = new ThreadLocal<String>(); public void set() { longLocal.set(Thread.currentThread().getId()); stringLocal.set(Thread.currentThread().getName()); } public long getLong() { return longLocal.get(); } public String getString() { return stringLocal.get(); } public static void main(String[] args) throws InterruptedException { final Test test = new Test(); test.set(); System.out.println(test.getLong()); System.out.println(test.getString()); for (int i=0; i<3; i++) { Thread thread1 = new Thread(){ public void run() { test.set(); System.out.println(test.getLong()); System.out.println(test.getString()); }; }; thread1.start(); thread1.join(); } System.out.println(test.getLong()); System.out.println(test.getString()); } }
和我們之前的test程序唯一的區別,就是使用了Java自帶的ThreadLocal類,那就進去看一看。
/** * Returns the value in the current thread‘s copy of this * thread-local variable. If the variable has no value for the * current thread, it is first initialized to the value returned * by an invocation of the {@link #initialValue} method. * * @return the current thread‘s value of this thread-local */ public T get() { Thread t = Thread.currentThread(); // 其實還是通過Map的數據結構 ThreadLocalMap map = getMap(t); if (map != null) { ThreadLocalMap.Entry e = map.getEntry(this); if (e != null) { @SuppressWarnings("unchecked") T result = (T)e.value; return result; } } return setInitialValue(); }
這是ThreadLocal的get方法,最終還是Map操作,但是這個Map以及Map裏面的Entry都是為ThreadLocal專門定制的,後面再說。看看getMap方法的邏輯
/** * Get the map associated with a ThreadLocal. Overridden in * InheritableThreadLocal. * * @param t the current thread * @return the map */ ThreadLocalMap getMap(Thread t) { return t.threadLocals; }
/* ThreadLocal values pertaining to this thread. This map is maintained * by the ThreadLocal class. */ // 定義在Thread類裏面 ThreadLocal.ThreadLocalMap threadLocals = null;
從這裏能看出2點:
1、ThreadLocalMap這個Map是ThreadLocal的內部類
2、這個Map的持有者是Thread類,就是說每個線程都直接持有自己的Map
第2點跟我們之前的實現思路截然不同,我們定義的ThreadLocalVar類不被任何線程直接持有,只是獨立的第三方,保持各個線程的數據。
後面再詳細分析這裏為什麽要這麽實現。
先來看看ThreadLocal的內部類ThreadLocalMap的內部類Entry(別繞暈了)
static class Entry extends WeakReference<ThreadLocal<?>> { /** The value associated with this ThreadLocal. */ Object value; Entry(ThreadLocal<?> k, Object v) { super(k); value = v; } }
Entry繼承自弱引用,說明持有key的弱引用,而且key是ThreadLocal類型(跟之前的實現方式也截然不同)。
為了說明ThreadLocal的實現機制和類直接的關系,從網上盜一張圖,圖中實線是強引用,虛線是弱引用。
每個線程持有Map有什麽好處?
1、線程消失,Map跟著消失,釋放了內存
2、保存數據的Map數量變多了,但是每個Map裏面Entry數量變少了。之前的實現裏面,每個Map裏面的Entry數量是線程的個數,現在是ThreadLocal的個數。熟悉Map數據結構的人都知道,這樣對Map的操作性能會提升。
至於為什麽要用弱引用,先來看看Entry類的註釋
/** * The entries in this hash map extend WeakReference, using * its main ref field as the key (which is always a * ThreadLocal object). Note that null keys (i.e. entry.get() * == null) mean that the key is no longer referenced, so the * entry can be expunged from table. Such entries are referred to * as "stale entries" in the code that follows. */
簡單來說,就是當ThreadLocal類型的key不再被引用時(值為null),對應的Entry能夠被刪除。
具體的實現就是,get操作會調用expungeStaleEntry,set操作會調用replaceStaleEntry,它們的效果就是遇到的key為null的Entry都會被刪除,那麽Entry內的value也就沒有強引用鏈,自然會被回收,防止內存泄露。這部分,請讀者仔細閱讀源碼。
經這麽一分析,是不是豁然開朗。
下面在看看ThreadLocal在一些框架裏面的應用:
1、Hibernate處理session,看看一個類ThreadLocalSessionContext
private static final ThreadLocal<Map> CONTEXT_TL = new ThreadLocal<Map>(); protected static Map sessionMap() { return CONTEXT_TL.get(); } @SuppressWarnings({"unchecked"}) private static void doBind(org.hibernate.Session session, SessionFactory factory) { Map sessionMap = sessionMap(); if ( sessionMap == null ) { sessionMap = new HashMap(); CONTEXT_TL.set( sessionMap ); } sessionMap.put( factory, session ); }
2、Spring處理事務,看看一個類TransactionSynchronizationManager
private static final ThreadLocal<Map<Object, Object>> resources = new NamedThreadLocal<Map<Object, Object>>("Transactional resources"); private static final ThreadLocal<Set<TransactionSynchronization>> synchronizations = new NamedThreadLocal<Set<TransactionSynchronization>>("Transaction synchronizations"); private static final ThreadLocal<String> currentTransactionName = new NamedThreadLocal<String>("Current transaction name"); private static final ThreadLocal<Boolean> currentTransactionReadOnly = new NamedThreadLocal<Boolean>("Current transaction read-only status"); private static final ThreadLocal<Integer> currentTransactionIsolationLevel = new NamedThreadLocal<Integer>("Current transaction isolation level"); private static final ThreadLocal<Boolean> actualTransactionActive = new NamedThreadLocal<Boolean>("Actual transaction active"); public static void bindResource(Object key, Object value) throws IllegalStateException { Object actualKey = TransactionSynchronizationUtils.unwrapResourceIfNecessary(key); Assert.notNull(value, "Value must not be null"); // 處理ThreadLocal Map<Object, Object> map = resources.get(); // set ThreadLocal Map if none found if (map == null) { map = new HashMap<Object, Object>(); // 處理ThreadLocal resources.set(map); } Object oldValue = map.put(actualKey, value); // Transparently suppress a ResourceHolder that was marked as void... if (oldValue instanceof ResourceHolder && ((ResourceHolder) oldValue).isVoid()) { oldValue = null; } if (oldValue != null) { throw new IllegalStateException("Already value [" + oldValue + "] for key [" + actualKey + "] bound to thread [" + Thread.currentThread().getName() + "]"); } if (logger.isTraceEnabled()) { logger.trace("Bound value [" + value + "] for key [" + actualKey + "] to thread [" + Thread.currentThread().getName() + "]"); } }
ThreadLocal類分析