【并发编程】ThreadLocal 源码分析

女爷i 2022-03-17 00:45 326阅读 0赞

# 前言 #

Github：[https://github.com/yihonglei/jdk-source-code-reading][https_github.com_yihonglei_jdk-source-code-reading]（java-concurrent）

# 一 ThreadLocal 原理 #

ThreadLocal 称为线程本地变量或线程本地存储，ThreadLocal 的作用就是让每个线程绑定自己的局部变量，

用于存储每个线程的私有数据，在线程的生命周期内起作用，减少同一个线程内多个函数或者组件之间一些

公共变量传递的复杂性。

## 1、创建一个 ThreadLocal 对象 ##

ThreadLocal myThreadLocal = new ThreadLocal();

实例化了一个 ThreadLocal 对象。每个线程仅需要实例化一次即可。

虽然不同的线程执行同一段代码时，访问同一个 ThreadLocal 变量，但是每个线程只能看到私有的 ThreadLocal 实例。

所以不同的线程在给 ThreadLocal 对象设置不同的值时，他们也不能看到彼此的修改。

## 2、访问 ThreadLocal 对象 ##

一旦创建了一个 ThreadLocal 对象，你就可以通过以下方式来存储此对象的值。

myThreadLocal.set("A thread local value");

也可以直接读取一个 ThreadLocal 对象的值。

String threadLocalValue = (String) myThreadLocal.get();

get() 方法会返回一个 Object 对象，而 set() 方法则依赖一个 Object 对象参数。

## 3、ThreadLocal 泛型 ##

为了使 get() 方法返回值不用做强制类型转换，通常可以创建一个泛型化的 ThreadLocal 对象。

ThreadLocal<String> myThreadLocal1 = new ThreadLocal();

现在你可以存储一个字符串到 ThreadLocal 实例里，此外，当你从此 ThreadLocal 实例中获取值的时候，

就不必要做强制类型转换。

myThreadLocal1.set("Hello ThreadLocal");
    
    String threadLocalValues = myThreadLocal.get();

## **4、初始化 ThreadLocal** ##

由于 ThreadLocal 对象的 set() 方法设置的值只对当前线程可见，那有什么方法可以为 ThreadLocal 对象设置的值对

所有线程都可见。为此，我们可以通过 ThreadLocal 子类的实现，并覆写 initialValue() 方法，就可以为 ThreadLocal 对象

指定一个初始化值。

ThreadLocal myThreadLocal = new ThreadLocal<String>() {
       @Override protected String initialValue() {
           return "This is the initial value";
       }
    };

此时，在 set() 方法调用前，当调用 get() 方法的时候，所有线程都可以看到同一个初始化值。

# 二 ThreadLocal 实战 #

## 1、ThreadLocal 类的 get 方法和 null 值 ##

package com.jpeony.concurrent.threadlocal;
    
    /**
     * ThreadLocal类的get方法和null值
     *
     * @author yihonglei
     */
    public class RunSimple {
        private static ThreadLocal THREAD_LOCAL = new ThreadLocal();
    
        public static void main(String[] args) {
            if (THREAD_LOCAL.get() == null) {
                System.out.println("未放入值");
                THREAD_LOCAL.set("放入值");
            }
            // 获取值
            System.out.println("print：" + THREAD_LOCAL.get());
        }
    }

当第一次调用 get 方法时，由于没有放入过值，返回 null，通过 set 方法放入值，之后通过 get 方法可以获取当前线程的私有数据。

**运行结果**

![20200728194947467.png][]

## 2、ThreadLocal 隔离实验 ##

**创建一个数据工具类：**

package com.jpeony.concurrent.threadlocal;
    
    /**
     * @author yihonglei
     */
    public class DataTools {
        public static ThreadLocal<String> THREAD_LOCAL = new ThreadLocal<>();
    }

**创建线程A：**

package com.jpeony.concurrent.threadlocal;
    
    /**
     * @author yihonglei
     */
    public class ThreadA extends Thread {
        @Override
        public void run() {
            try {
                for (int i = 0; i < 100; i++) {
                    DataTools.THREAD_LOCAL.set("A_Thread_" + System.currentTimeMillis());
                    // 获取当前线程私有数据
                    System.out.println("线程A获取的值:" + DataTools.THREAD_LOCAL.get());
                    Thread.sleep(200);
                }
            } catch (InterruptedException ex) {
                ex.printStackTrace();
            }
        }
    }

**创建线程B：**

package com.jpeony.concurrent.threadlocal;
    
    /**
     * @author yihonglei
     */
    public class ThreadB extends Thread {
        @Override
        public void run() {
            try {
                for (int i = 0; i < 100; i++) {
                    DataTools.THREAD_LOCAL.set("B_Thread_" + System.currentTimeMillis());
                    // 获取当前线程私有数据
                    System.out.println("线程Ｂ获取的值：" + DataTools.THREAD_LOCAL.get());
                    Thread.sleep(200);
                }
            } catch (InterruptedException ex) {
                ex.printStackTrace();
            }
        }
    }

**测试类：**

package com.jpeony.concurrent.threadlocal;
    
    /**
     * @author yihonglei
     */
    public class RunTest {
        public static void main(String[] args) {
            try {
                ThreadA threadA = new ThreadA();
                threadA.start();
    
                Thread.sleep(3000);
    
                ThreadB threadB = new ThreadB();
                threadB.start();
            } catch (InterruptedException ex) {
                ex.printStackTrace();
            }
        }
    }

**运行结果：**

![2020072819534572.png][]

从程序运行结果，线程获取属于自己的私有数据，证明了 ThreadLocal 实现线程私有数据保证数据的隔离性。

## 3、如何让 get 方法第一次获取的值不为 null ##

写一个类继承与 ThreadLocal，重写 initialValue() 方法，设置 get 的默认值。

package com.jpeony.concurrent.threadlocal;
    
    /**
     * 重写ThreadLocal初始化方法，自定义初始值
     *
     * @author yihonglei
     */
    public class ThreadLocalExt extends ThreadLocal {
    
        @Override
        protected Object initialValue() {
            //return super.initialValue();
            return "设置默认值，第一次获取不在为null";
        }
    
    }

# 三 ThreadLocal 源码分析 #

## 1、ThreadLocal 源码结构 ##

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3lobF9qeHk_size_16_color_FFFFFF_t_70][]

### 1.1 主要结构说明 ###

SuppliedThreadLocal：jdk8的增强，支持 lambda 表达式；

ThreadLocalMap：ThreadLocal 内部数据结构，类似 Map，但是 key 是对应 ThreadLocal 对象的弱引用；

initilaValue()：ThreadLocal 初始值；

set(T)：往 ThreadLocalMap 放值；

get()：从 ThreadLocalMap 取值；

### 1.2 ThreadLocal 如何实现线程私有的？ ###

在 Thread 里面，有一个局部变量 ThreadLocal.ThreadLocalMap threadLocals = null；

通过这个局部变量维护线程自己的一份数据，这个局部变量值持续到线程结束。

至于这个 ThreadLocalMap 是个什么，下面分析它在 ThreadLocal 中的地位和数据结构。

![20200729103001957.png][]

Thread 的局部变量 threadLocals 是私有的，数据结构是 ThreadLocal 的 ThreadLocalMap，如果一个线程想拥有

多个 ThreadLocal，则 new 多个 ThreadLocal 对象，根据 ThreadLocal 对象计算 key，都存在线程的

ThreadLoca.ThreadLocalMap 数据结构里面，多个 ThreadLocal 共享 ThreadLocalMap 数据结构。

![20200729135419254.png][]

## 2、set 过程源码分析 ##

**2.1 set(T value)**

将应用数据 set 到 ThreadLocal 中。

public void set(T value) {
        // 获取当前线程
        Thread t = Thread.currentThread();
        // 获取ThreadLocalMap
        ThreadLocalMap map = getMap(t);
        // 判断是否已经存在ThreadLocalMap
        if (map != null)
            // 添加值到ThreadLocalMap
            map.set(this, value);
        else
            // 创建ThreadLocalMap并添加值
            createMap(t, value);
    }

**2.2 getMap(Thread t)**

ThreadLocalMap getMap(Thread t) {
        return t.threadLocals;
    }

threadLocals 为 ThreadLocal 的内部类成员变量 ThreadLocalMap，第一次添加值获取的是 null，这个时候还没有初始化。

ThreadLocal.ThreadLocalMap threadLocals = null;

**2.3 createMap(Thread t, T firstValue)**

创建 ThreadLocal 的内部数据结构 ThreadLocalMap。

void createMap(Thread t, T firstValue) {
        // 初始化并set值
        t.threadLocals = new ThreadLocalMap(this, firstValue);
    }

先看下 ThreadLocalMap 是啥？

**2.4 ThreadLocalMap 是啥？**

ThreadLocalMap 是 ThreadLocal 静态内部类，专为 ThreadLocal 定制的高效存储实现，基于弱引用的垃圾清理机制。

弱引用就是垃圾回收的时候，不管内存空间是否足够，直接回收掉，避免占用内存。ThreadLocal 跟线程生命周期一致，

所以，当线程结束，即可回收，比较适合弱引用清理方式。

关于jvm对象回收判定：[https://jpeony.blog.csdn.net/article/details/80950336][https_jpeony.blog.csdn.net_article_details_80950336]

ThreadLocalMap 源码结构：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3lobF9qeHk_size_16_color_FFFFFF_t_70 1][]

ThreadLocalMap 存储结构：

/**
     * 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.
     * 
     * key是对ThreadLocal的弱引用，value是一个Object，存储要放入ThreadLocal的值。
     * 当没有引用的时候，Entry自动从table中剔除。
     */
    static class Entry extends WeakReference<ThreadLocal<?>> {
        /** The value associated with this ThreadLocal. */
        // 存储要放入ThreadLocal的值
        Object value;
    
        Entry(ThreadLocal<?> k, Object v) {
            super(k);
            value = v;
        }
    }

![20200729150258365.png][]

结构类似 Map，Entry 继承 WeakReference（弱引用），key 跟普通 Map 的 key 有区别，这个 key 存放的是对 ThreadLocal 的弱引用，

当对 key 的引用不存在的时候，自动从 Entry 数组移出，被 GC 自动回收，弱引用的回收，GC 是不判断内存是否还够的，

Object 对象存储要放入 ThreadLocal 的值。debug 可以看下这个 key 长的样子。

HashMap：数组+链表（解决Hash冲突），k-v 都是强引用。

ThreadLocalMap：数组（通过线性探测解决Hash冲突，即如果当前数组位置有值，则往下查找，往空的地方存入值），

k 是弱引用，v 是强引用。

![20200325173922730.png][]

ThreadLocalMap 成员变量：

/**
      * The initial capacity -- MUST be a power of two.
      * 初始化容量大小，必须是2的幂次方
      */
      private static final int INITIAL_CAPACITY = 16;
      
     /**
      * The table, resized as necessary.
      * table.length MUST always be a power of two.
      * table是一个Entry数组
      */
      private Entry[] table;
      
     /**
      * The number of entries in the table.
      * table的元素个数（Entry数组的元素个数）
      */
      private int size = 0;
      
     /**
      * The next size value at which to resize.
      * 重新分配表大小的阀值，默认值是0
      */
      private int threshold; // Default to 0

ThreadLocalMap 主要方法：

/**
     * Set the resize threshold to maintain at worst a 2/3 load factor.
     * 设置阀值上限，不能超过最坏2/3负载因子，因为超过之后性能极速下降
     */
    private void setThreshold(int len) {
        threshold = len * 2 / 3;
    }
    
    /**
     * Increment i modulo len.
     * 环形意义上的下一个索引
     */
    private static int nextIndex(int i, int len) {
        return ((i + 1 < len) ? i + 1 : 0);
    }
    
    /**
     * Decrement i modulo len.
     * 环形意义上的上一个索引
     */
    private static int prevIndex(int i, int len) {
        return ((i - 1 >= 0) ? i - 1 : len - 1);
    }

ThreadLocalMap 使用线性探测法来解决散列冲突，所以实际上 Entry\[\] 数组在程序逻辑上是作为一个环形存在的。

图片源于网络。

![format_png][]

ThreadLocalMap 维护了 Entry 环形数组，数组中元素 Entry 的逻辑上的 key 为某个 ThreadLocal 对象（实际上是指向该 ThreadLocal 对

象的弱引用），value 为代码中该线程往该 ThreadLoacl 变量实际塞入的值。

ThreadLocalMap 构造函数：

/**
     * Construct a new map initially containing (firstKey, firstValue).
     * ThreadLocalMaps are constructed lazily, so we only create
     * one when we have at least one entry to put in it.
     * 构造一个包含firstKey和firstValue的map。
     * ThreadLocalMap是惰性构造的，所以只有当至少要往里面放一个元素的时候才会构建它。
     */
     hreadLocalMap(ThreadLocal<?> firstKey, Object firstValue) {
        // 初始化table，设置Entry数组初始容量为16
        table = new Entry[INITIAL_CAPACITY];
        // firstKey的threadLocalHashCode与初始大小(16-1)取模算entry的下标值
        int i = firstKey.threadLocalHashCode & (INITIAL_CAPACITY - 1);
        // 初始化节点
        table[i] = new Entry(firstKey, firstValue);
        size = 1;
        // 设置扩容的阀值，当达到这个值的时候，Entry数据要进行扩容处理
        setThreshold(INITIAL_CAPACITY);
     }

ThreadLocalMap 的 set 方法：

/**
     * Set the value associated with key.
     *
     * @param key the thread local object
     * @param value the value to be set
     */
    private void set(ThreadLocal<?> key, Object value) {
    
        // We don't use a fast path as with get() because it is at
        // least as common to use set() to create new entries as
        // it is to replace existing ones, in which case, a fast
        // path would fail more often than not.
    
        Entry[] tab = table;
        int len = tab.length;
        int i = key.threadLocalHashCode & (len-1);
    
        // 线性探测
        for (Entry e = tab[i];
             e != null;
             e = tab[i = nextIndex(i, len)]) {
            // 找到对应的entry
            ThreadLocal<?> k = e.get();
            // 获取对应的value
            if (k == key) {
                e.value = value;
                return;
            }
    
            // 替换失效的entry
            if (k == null) {
                replaceStaleEntry(key, value, i);
                return;
            }
        }
        // 在table中没有找到对应的key，会基于k-v构建节点，并赋值到Entry对应的位置
        tab[i] = new Entry(key, value);
        int sz = ++size;
        // 从数组中移出空闲的entry
        if (!cleanSomeSlots(i, sz) && sz >= threshold)
            rehash();
    }
    /**
     * Replace a stale entry encountered during a set operation
     * with an entry for the specified key.  The value passed in
     * the value parameter is stored in the entry, whether or not
     * an entry already exists for the specified key.
     *
     * As a side effect, this method expunges all stale entries in the
     * "run" containing the stale entry.  (A run is a sequence of entries
     * between two null slots.)
     *
     * @param  key the key
     * @param  value the value to be associated with key
     * @param  staleSlot index of the first stale entry encountered while
     *         searching for key.
     */
     rivate void replaceStaleEntry(ThreadLocal<?> key, Object value,
                                   int staleSlot) {
        Entry[] tab = table;
        int len = tab.length;
        Entry e;
    
        // Back up to check for prior stale entry in current run.
        // We clean out whole runs at a time to avoid continual
        // incremental rehashing due to garbage collector freeing
        // up refs in bunches (i.e., whenever the collector runs).
        // 向前遍历Entry数组，查找最近一个无效的slot
        int slotToExpunge = staleSlot;
        for (int i = prevIndex(staleSlot, len);
             (e = tab[i]) != null;
             i = prevIndex(i, len))
            if (e.get() == null)
                slotToExpunge = i;
    
        // Find either the key or trailing null slot of run, whichever
        // occurs first
        // 向后遍历
        for (int i = nextIndex(staleSlot, len);
             (e = tab[i]) != null;
             i = nextIndex(i, len)) {
            // 找到key
            ThreadLocal<?> k = e.get();
    
            // If we find key, then we need to swap it
            // with the stale entry to maintain hash table order.
            // The newly stale slot, or any other stale slot
            // encountered above it, can then be sent to expungeStaleEntry
            // to remove or rehash all of the other entries in run.
            // 找到了key，把新值替换旧值
            if (k == key) {
                e.value = value;
    
                tab[i] = tab[staleSlot];
                tab[staleSlot] = e;
    
                // Start expunge at preceding stale entry if it exists
                // 如果找到了无效的slot，则这个作为清理的起点，否则，当前的值作为清理的起点
                if (slotToExpunge == staleSlot)
                    slotToExpunge = i;
                // 从slotToExpunge开始做一次连续段的清理，再做一次启发式清理
                cleanSomeSlots(expungeStaleEntry(slotToExpunge), len);
                return;
            }
    
            // If we didn't find stale entry on backward scan, the
            // first stale entry seen while scanning for key is the
            // first still present in the run.
            // 如果当前的slot已经无效，并且向前扫描过程中没有无效slot，则更新slotToExpunge为当前位置
            if (k == null && slotToExpunge == staleSlot)
                slotToExpunge = i;
        }
    
        // If key not found, put new entry in stale slot
        // 如果key在table中不存在，则在原地放一个即可
        tab[staleSlot].value = null;
        tab[staleSlot] = new Entry(key, value);
    
        // If there are any other stale entries in run, expunge them
        // 在探测过程中如果发现任何无效slot，则做一次清理（连续段清理+启发式清理）
        if (slotToExpunge != staleSlot)
            cleanSomeSlots(expungeStaleEntry(slotToExpunge), len);
     }
    /**
     * Heuristically scan some cells looking for stale entries.
     * This is invoked when either a new element is added, or
     * another stale one has been expunged. It performs a
     * logarithmic number of scans, as a balance between no
     * scanning (fast but retains garbage) and a number of scans
     * proportional to number of elements, that would find all
     * garbage but would cause some insertions to take O(n) time.
     *
     * @param i a position known NOT to hold a stale entry. The
     * scan starts at the element after i.
     *
     * @param n scan control: {@code log2(n)} cells are scanned,
     * unless a stale entry is found, in which case
     * {@code log2(table.length)-1} additional cells are scanned.
     * When called from insertions, this parameter is the number
     * of elements, but when from replaceStaleEntry, it is the
     * table length. (Note: all this could be changed to be either
     * more or less aggressive by weighting n instead of just
     * using straight log n. But this version is simple, fast, and
     * seems to work well.)
     *
     * @return true if any stale entries have been removed.
     */
    private boolean cleanSomeSlots(int i, int n) {
        boolean removed = false;
        Entry[] tab = table;
        int len = tab.length;
        do {
            // i是刚构建的元素下标，所以不会是一个无效slot，所以从下一个开始判断
            i = nextIndex(i, len);
            Entry e = tab[i];
            if (e != null && e.get() == null) {
                // 扩大扫描控制因子
                n = len;
                removed = true;
                // 清理一个连续段
                i = expungeStaleEntry(i);
            }
        } while ( (n >>>= 1) != 0);
        return removed;
    }
    
    /**
     * Expunge a stale entry by rehashing any possibly colliding entries
     * lying between staleSlot and the next null slot.  This also expunges
     * any other stale entries encountered before the trailing null.  See
     * Knuth, Section 6.4
     *
     * @param staleSlot index of slot known to have null key
     * @return the index of the next null slot after staleSlot
     * (all between staleSlot and this slot will have been checked
     * for expunging).
     */
    private int  expungeStaleEntry(int staleSlot) {
        Entry[] tab = table;
        int len = tab.length;
    
        // expunge entry at staleSlot
        tab[staleSlot].value = null;
        tab[staleSlot] = null;
        size--;
    
        // Rehash until we encounter null
        Entry e;
        int i;
        for (i = nextIndex(staleSlot, len);
             (e = tab[i]) != null;
             i = nextIndex(i, len)) {
            ThreadLocal<?> k = e.get();
            if (k == null) {
                e.value = null;
                tab[i] = null;
                size--;
            } else {
                int h = k.threadLocalHashCode & (len - 1);
                if (h != i) {
                    tab[i] = null;
    
                    // Unlike Knuth 6.4 Algorithm R, we must scan until
                    // null because multiple entries could have been stale.
                    while (tab[h] != null)
                        h = nextIndex(h, len);
                    tab[h] = e;
                }
            }
        }
        return i;
    }
    /**
     * Re-pack and/or re-size the table. First scan the entire
     * table removing stale entries. If this doesn't sufficiently
     * shrink the size of the table, double the table size.
     */
    private void rehash() {
        expungeStaleEntries();
    
        // Use lower threshold for doubling to avoid hysteresis
        /*
         * 因为做了一次清理，所以size很可能会变小。
         * ThreadLocalMap这里的实现是调低阈值来判断是否需要扩容，
         * threshold默认为len*2/3，所以这里的threshold - threshold / 4相当于len/2
         */
        if (size >= threshold - threshold / 4)
            resize();
    }
    
    /**
     * Expunge all stale entries in the table.
     */
    private void expungeStaleEntries() {
        Entry[] tab = table;
        int len = tab.length;
        for (int j = 0; j < len; j++) {
            Entry e = tab[j];
            if (e != null && e.get() == null)
                expungeStaleEntry(j);
        }
    }
    
    /**
     * Double the capacity of the table.
     * 扩容，因为需要保证table的容量len为2的幂，所以扩容即扩大2倍
     */
    private void resize() {
        Entry[] oldTab = table;
        int oldLen = oldTab.length;
        int newLen = oldLen * 2;
        Entry[] newTab = new Entry[newLen];
        int count = 0;
    
        for (int j = 0; j < oldLen; ++j) {
            Entry e = oldTab[j];
            if (e != null) {
                ThreadLocal<?> k = e.get();
                if (k == null) {
                    e.value = null; // Help the GC
                } else {
                    // 线性探测来存放Entry
                    int h = k.threadLocalHashCode & (newLen - 1);
                    while (newTab[h] != null)
                        h = nextIndex(h, newLen);
                    newTab[h] = e;
                    count++;
                }
            }
        }
    
        setThreshold(newLen);
        size = count;
        table = newTab;
    }

ThreadLocalMap 的 getEntry 方法：

/**
     * Get the entry associated with key.  This method
     * itself handles only the fast path: a direct hit of existing
     * key. It otherwise relays to getEntryAfterMiss.  This is
     * designed to maximize performance for direct hits, in part
     * by making this method readily inlinable.
     *
     * @param  key the thread local object
     * @return the entry associated with key, or null if no such
     */
    private Entry getEntry(ThreadLocal<?> key) {
        // 根据key这个ThreadLocal的ID来获取索引，也即哈希值
        int i = key.threadLocalHashCode & (table.length - 1);
        Entry e = table[i];
        // 对应的entry存在且未失效且弱引用指向的ThreadLocal就是key，则命中返回
        if (e != null && e.get() == key)
            return e;
        else
            // 因为用的是线性探测，所以往后找还是有可能能够找到目标Entry的。
            return getEntryAfterMiss(key, i, e);
    }
    
    /**
     * Version of getEntry method for use when key is not found in
     * its direct hash slot.
     *
     * @param  key the thread local object
     * @param  i the table index for key's hash code
     * @param  e the entry at table[i]
     * @return the entry associated with key, or null if no such
     */
    private Entry getEntryAfterMiss(ThreadLocal<?> key, int i, Entry e) {
        Entry[] tab = table;
        int len = tab.length;
    
        // 基于线性探测法不断向后探测直到遇到空entry
        while (e != null) {
            ThreadLocal<?> k = e.get();
            // 找到目标
            if (k == key)
                return e;
            if (k == null)
                // 该entry对应的ThreadLocal已经被回收，调用expungeStaleEntry来清理无效的entry
                expungeStaleEntry(i);
            else
                // 环形意义下往后面走
                i = nextIndex(i, len);
            e = tab[i];
        }
        return null;
    }
    
    /**
     * Expunge a stale entry by rehashing any possibly colliding entries
     * lying between staleSlot and the next null slot.  This also expunges
     * any other stale entries encountered before the trailing null.  See
     * Knuth, Section 6.4
     *
     * @param staleSlot index of slot known to have null key
     * @return the index of the next null slot after staleSlot
     * (all between staleSlot and this slot will have been checked
     * for expunging).
     */
    private int expungeStaleEntry(int staleSlot) {
        Entry[] tab = table;
        int len = tab.length;
    
        // expunge entry at staleSlot
        // 因为entry对应的ThreadLocal已经被回收，value设为null，显式断开强引用
        tab[staleSlot].value = null;
        tab[staleSlot] = null;
        size--;
    
        // Rehash until we encounter null
        Entry e;
        int i;
        for (i = nextIndex(staleSlot, len);
             (e = tab[i]) != null;
             i = nextIndex(i, len)) {
            ThreadLocal<?> k = e.get();
            if (k == null) {
                e.value = null;
                tab[i] = null;
                size--;
            } else {
                int h = k.threadLocalHashCode & (len - 1);
                if (h != i) {
                    tab[i] = null;
    
                    // Unlike Knuth 6.4 Algorithm R, we must scan until
                    // null because multiple entries could have been stale.
                    while (tab[h] != null)
                        h = nextIndex(h, len);
                    tab[h] = e;
                }
            }
        }
        return i;
    }

## 3、get 过程源码分析 ##

/**
     * Get the entry associated with key.  This method
     * itself handles only the fast path: a direct hit of existing
     * key. It otherwise relays to getEntryAfterMiss.  This is
     * designed to maximize performance for direct hits, in part
     * by making this method readily inlinable.
     *
     * @param  key the thread local object
     * @return the entry associated with key, or null if no such
     */
    private Entry getEntry(ThreadLocal<?> key) {
    	// 根据key，即ThreadLocal的哈希值计算索引
        int i = key.threadLocalHashCode & (table.length - 1);
        Entry e = table[i];
        // 如果对应的Entry存在，并且ThreadLocal的弱引用指向key，则key命中返回
        if (e != null && e.get() == key)
            return e;
        else
        	// 如果直接查找不到，根据当前位置往后继续找
            return getEntryAfterMiss(key, i, e);
    }
    
    /**
     * Version of getEntry method for use when key is not found in
     * its direct hash slot.
     *
     * @param  key the thread local object
     * @param  i the table index for key's hash code
     * @param  e the entry at table[i]
     * @return the entry associated with key, or null if no such
     */
    private Entry getEntryAfterMiss(ThreadLocal<?> key, int i, Entry e) {
        Entry[] tab = table;
        int len = tab.length;
    
        while (e != null) {
            ThreadLocal<?> k = e.get();
            // 如果找到，则返回
            if (k == key)
                return e;
            // 如果Entry对应的ThreadLocal被回收，通过expungeStaleEntry清理无效的Entry
            if (k == null)
                expungeStaleEntry(i);
            else
            	// 获取环形意义上的下一个索引
                i = nextIndex(i, len);
            e = tab[i];
        }
        return null;
    }
    
    /**
     * Expunge a stale entry by rehashing any possibly colliding entries
     * lying between staleSlot and the next null slot.  This also expunges
     * any other stale entries encountered before the trailing null.  See
     * Knuth, Section 6.4
     *
     * @param staleSlot index of slot known to have null key
     * @return the index of the next null slot after staleSlot
     * (all between staleSlot and this slot will have been checked
     * for expunging).
     */
    private int expungeStaleEntry(int staleSlot) {
        Entry[] tab = table;
        int len = tab.length;
    
        // expunge entry at staleSlot
        // 设置要清理的位置value为空
        tab[staleSlot].value = null;
        // 整个Entry置为空，方便GC回收
        tab[staleSlot] = null;
        size--;
    
        // Rehash until we encounter null
        Entry e;
        int i;
        for (i = nextIndex(staleSlot, len);
             (e = tab[i]) != null;
             i = nextIndex(i, len)) {
            ThreadLocal<?> k = e.get();
            if (k == null) {
                e.value = null;
                tab[i] = null;
                size--;
            } else {
                int h = k.threadLocalHashCode & (len - 1);
                if (h != i) {
                    tab[i] = null;
    
                    // Unlike Knuth 6.4 Algorithm R, we must scan until
                    // null because multiple entries could have been stale.
                    while (tab[h] != null)
                        h = nextIndex(h, len);
                    tab[h] = e;
                }
            }
        }
        return i;
    }

**InheritableThreadLocal 是什么？**

**子类线程获取父线程的 ThreadLocal 值。**

package com.jpeony.concurrent.threadlocal;
    
    /**
     * InheritableThreadLocal获取父线程往ThreadLocal里面塞入的值。
     *
     * @author yihonglei
     */
    public class InheritableThreadLocalSimple {
        public static ThreadLocal<String> THREAD_LOCAL = new InheritableThreadLocal<>();
    
        public static void main(String[] args) {
            // 往main-thread塞入值
            THREAD_LOCAL.set("parentThread");
            // 启动新线程
            Thread thread = new MyThread();
            thread.start();
            System.out.println("main = " + THREAD_LOCAL.get());
        }
    
        static class MyThread extends Thread {
            @Override
            public void run() {
                System.out.println("MyThread = " + THREAD_LOCAL.get());
            }
        }
    }

**运行结果**

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3lobF9qeHk_size_16_color_FFFFFF_t_70 2][]

如果缓存 ThreadLocal，则 MyThread 输出的将是 null。

**实现本质**

如果只是跟踪源码，并没有发现跟 ThreadLocal 有什么不同，为啥用 ThreadLocal 输出是null，而 InheritableThreadLocal 确

能在子线程里面拿到父线程塞入的值。

关键点在于线程初始化的时候捣鬼了。

java.lang.Thread\#init  
    --》ThreadLocal.createInheritedMap(parent.inheritableThreadLocals);  
        --》java.lang.ThreadLocal.ThreadLocalMap\#ThreadLocalMap(java.lang.ThreadLocal.ThreadLocalMap)

new Thread()，即子线程在初始化的时候，将父线程的 ThreadLocal 复制了一份，这是精髓代码。

Thread\#init 触发 ThreadLocal 复制的部分源码如下：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3lobF9qeHk_size_16_color_FFFFFF_t_70 3][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3lobF9qeHk_size_16_color_FFFFFF_t_70 4][]

# 四 ThreadLocal 存在哪些问题 #

## 1、内存溢出问题(OOM) ##

ThreadLocal 是 Thread 里面维护的一个局部变量，线程不退出，局部变量的值就一直存在。正常情况下单个线程执行完就退出，

那么就会做清理工作，但是当使用多线程的时候，如果我们在 ThreadLocal 存的是大对象，并发量高的时候，每个线程都持有大

对象，很容易造成内存溢出，ThreadLocal 是以“时间换空间”。

# 五 ThreadLocal总结 #

1、ThreadLocal.ThreadLocalMap 是 Thread 的一个局部变量；

2、ThreadLocalMap 是类似 Map 的结构，最大的区别在于 key，可以看为是 ThreadLocal 对象实例的 HashCode 值作为 Key

来存储 value，采用弱引用内存回收机制。

3、 ThreadLocal 实现了线程与线程之间的隔离，因为是线程的局部变量，自己有一份同时，也实现了每个 ThreadLocal 之间的

隔离，因为 key 是 ThreadLocal 对象实例的 HashCode 值，是不一样的。

4、线程退出时，局部变量自动回收内存。

5、ThreaLocal 是以空间换时间的做法，就是用内存空间存起来，维持到整个线程结束，但是，线程之间就不需要排队访问，

在自己的线程里面随心所欲。

[https_github.com_yihonglei_jdk-source-code-reading]: https://github.com/yihonglei/jdk-source-code-reading
[20200728194947467.png]: /images/20220317/5edb8f40615d423c84801352a18306f3.png
[2020072819534572.png]: /images/20220317/e381718d5e9c4bbea3a7726fd70385a1.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3lobF9qeHk_size_16_color_FFFFFF_t_70]: /images/20220317/ce137d80b47e4592bbffcdb98bafd515.png
[20200729103001957.png]: /images/20220317/0b560bea76414fd482cdded1417eaa33.png
[20200729135419254.png]: /images/20220317/8bacde33be8545c8a8e30152725e421a.png
[https_jpeony.blog.csdn.net_article_details_80950336]: https://jpeony.blog.csdn.net/article/details/80950336
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3lobF9qeHk_size_16_color_FFFFFF_t_70 1]: /images/20220317/faf0d275b2ca451f94261c3feee93250.png
[20200729150258365.png]: /images/20220317/490a13a274fc474a89ae9a2f1f117f97.png
[20200325173922730.png]: /images/20220317/8cb2358230cb47c5a715ae42f5374bc4.png
[format_png]: /images/20220317/fc1f16c82a39480d87a05eec0b7ff815.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3lobF9qeHk_size_16_color_FFFFFF_t_70 2]: /images/20220317/172b105a5a804f799afd5b314061ebaf.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3lobF9qeHk_size_16_color_FFFFFF_t_70 3]: /images/20220317/763ecfdb9b36490cacd15b7db4c90d9b.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3lobF9qeHk_size_16_color_FFFFFF_t_70 4]: /images/20220317/b3c62447398747649621eb044d26ffc9.png