java并发编程学习之ConcurrentHashMap(JDK1.7)

之前HashMap中提过，并发的时候，可能造成死循环，且线程不安全，那么在多线程中可以用ConcurrentHashMap来避免这一情况。

Segment

ConcurrentHashMap是由多个Segment组成的，Segment继承了ReentrantLock，每次加锁都是对某个Segment，不会影响其他Segment，达到了锁分离（也叫分段锁）的作用。
每个Segment又包含了HashEntry数组，HashEntry是一个链表。如下图所示：

初始化

initialCapacity：初始容量大小，默认16。
loadFactor：扩容因子，table扩容使用，Segments不扩容。默认0.75，当Segment容量大于initialCapacity*loadFactor时，开始扩容
concurrencyLevel：并发数，默认16，直接影响segmentShift和segmentMask的值，以及Segment的初始化数量。Segment初始化的数量，为最接近且大于的办等于2的N次方的值，比如concurrencyLevel=16，Segment数量为16，concurrencyLevel=17，Segment数量为32。segmentShift的值是这样的，比如Segment是32，相对于2的5次方，那么他的值就是32-5，为27，后面无符号右移27位，也就是取高5位的时候，就是0到31的值，此时Segment的下标也是0到31，取模后对应着每个Segment。segmentMask就是2的n次方-1，这边n是5，用于取模。之前在hashmap的indexFor方法有提过。
初始化的时候，还要初始化第一个Segment，以及Segment中table数组的大小，这边大小是大于等于initialCapacity除以Segment数组的个数，平均分配，最小是2，且是2的N次方。比如initialCapacity是32，concurrencyLevel是16的时候，那么Segment的个数也是16，32除以16，等于2，如果initialCapacity是33，Segment是16，33除以16，取4。

public ConcurrentHashMap() {
    this(DEFAULT_INITIAL_CAPACITY, DEFAULT_LOAD_FACTOR, DEFAULT_CONCURRENCY_LEVEL);
}
public ConcurrentHashMap(int initialCapacity,
                             float loadFactor, int concurrencyLevel) {
    if (!(loadFactor > 0) || initialCapacity < 0 || concurrencyLevel <= 0)
        throw new IllegalArgumentException();
    if (concurrencyLevel > MAX_SEGMENTS)
        concurrencyLevel = MAX_SEGMENTS;
    // Find power-of-two sizes best matching arguments
    int sshift = 0;
    int ssize = 1;
    while (ssize < concurrencyLevel) {
        ++sshift;
        ssize <<= 1;
    }
    this.segmentShift = 32 - sshift;//用于高位，判断落在哪个Segment
    this.segmentMask = ssize - 1;//用于取模。之前在hashmap的indexFor方法有提过。2的n次方-1
    if (initialCapacity > MAXIMUM_CAPACITY)
        initialCapacity = MAXIMUM_CAPACITY;
    int c = initialCapacity / ssize;
    if (c * ssize < initialCapacity)
        ++c;
    int cap = MIN_SEGMENT_TABLE_CAPACITY;
    while (cap < c)
        cap <<= 1;
    // create segments and segments[0]
    Segment<K,V> s0 =
        new Segment<K,V>(loadFactor, (int)(cap * loadFactor),
                         (HashEntry<K,V>[])new HashEntry[cap]);//初始化第一个位置的Segment
    Segment<K,V>[] ss = (Segment<K,V>[])new Segment[ssize];//初始化Segments
    UNSAFE.putOrderedObject(ss, SBASE, s0); // ordered write of segments[0]
    this.segments = ss;
}

详细流程

put方法

public V put(K key, V value) {
    Segment<K,V> s;
    if (value == null)
        throw new NullPointerException();
    int hash = hash(key);
    //无符号右移后取模，落在哪个Segment上面
    int j = (hash >>> segmentShift) & segmentMask;
    if ((s = (Segment<K,V>)UNSAFE.getObject          // nonvolatile; recheck
         (segments, (j << SSHIFT) + SBASE)) == null) //  in ensureSegment
        s = ensureSegment(j);
    return s.put(key, hash, value, false);
}

ensureSegment方法
确定落在哪个Segment上，如果为空，就初始化，因为之前就初始化第一个Segment

private Segment<K,V> ensureSegment(int k) {
    final Segment<K,V>[] ss = this.segments;
    long u = (k << SSHIFT) + SBASE; // raw offset
    Segment<K,V> seg;
    if ((seg = (Segment<K,V>)UNSAFE.getObjectVolatile(ss, u)) == null) {
        //使用segment[0]的table长度和loadFactor来初始化
        Segment<K,V> proto = ss[0]; // use segment 0 as prototype
        int cap = proto.table.length;
        float lf = proto.loadFactor;
        int threshold = (int)(cap * lf);
        HashEntry<K,V>[] tab = (HashEntry<K,V>[])new HashEntry[cap];
        if ((seg = (Segment<K,V>)UNSAFE.getObjectVolatile(ss, u))
            == null) { // recheck
            Segment<K,V> s = new Segment<K,V>(lf, threshold, tab);
            while ((seg = (Segment<K,V>)UNSAFE.getObjectVolatile(ss, u))//cas操作，只能一个设值成功，如果其他成功了，就赋值，并返回
                   == null) {
                if (UNSAFE.compareAndSwapObject(ss, u, null, seg = s))
                    break;
            }
        }
    }
    return seg;
}

put方法

final V put(K key, int hash, V value, boolean onlyIfAbsent) {
    HashEntry<K,V> node = tryLock() ? null :
        scanAndLockForPut(key, hash, value);//获取Segment的锁
    V oldValue;
    try {
        HashEntry<K,V>[] tab = table;
        int index = (tab.length - 1) & hash;//上面是获取Segment取高位的hash，这边是tabel的hash，
        HashEntry<K,V> first = entryAt(tab, index);//取到hash位置的数组的表头
        for (HashEntry<K,V> e = first;;) {//从头结点遍历
            if (e != null) {
                K k;
                if ((k = e.key) == key ||
                    (e.hash == hash && key.equals(k))) {//key相同，或者hash值一样
                    oldValue = e.value;
                    if (!onlyIfAbsent) {//是否替换
                        e.value = value;
                        ++modCount;
                    }
                    break;
                }
                e = e.next;
            }
            else {
                if (node != null)//不为空，设置为表头
                    node.setNext(first);
                else
                    node = new HashEntry<K,V>(hash, key, value, first);/初始化后放表头
                int c = count + 1;
                if (c > threshold && tab.length < MAXIMUM_CAPACITY)
                    rehash(node);// 扩容
                else
                    setEntryAt(tab, index, node);//把新的节点放在tab的index上面
                ++modCount;
                count = c;
                oldValue = null;
                break;
            }
        }
    } finally {
        unlock();//释放锁
    }
    return oldValue;
}

scanAndLockForPut方法
尝试获取锁，没获取到先初始化node

private HashEntry<K,V> scanAndLockForPut(K key, int hash, V value) {
    HashEntry<K,V> first = entryForHash(this, hash);//获取hash后的头结点，有存在null的情况
    HashEntry<K,V> e = first;
    HashEntry<K,V> node = null;
    int retries = -1; // negative while locating node
    while (!tryLock()) {//这个put方法先尝试获取，获取不到，这边while循环尝试获取
        HashEntry<K,V> f; // to recheck first below
        if (retries < 0) {
            if (e == null) {//结点为空的时候
                if (node == null) // speculatively create node
                    node = new HashEntry<K,V>(hash, key, value, null);//初始化node
                retries = 0;
            }
            else if (key.equals(e.key))//头结点不为空的时候
                retries = 0;
            else
                e = e.next;
        }
        else if (++retries > MAX_SCAN_RETRIES) {//超过重试次数，直接进入阻塞队列等待锁
            lock();
            break;
        }
        else if ((retries & 1) == 0 &&
                 (f = entryForHash(this, hash)) != first) {//不等于first，就是已经有其他节点进入
            e = first = f; // re-traverse if entry changed
            retries = -1;
        }
    }
    return node;
}

rehash方法，扩容，对table扩容

private void rehash(HashEntry<K,V> node) {
    HashEntry<K,V>[] oldTable = table;
    int oldCapacity = oldTable.length;
    int newCapacity = oldCapacity << 1;//左移，之前的2倍
    threshold = (int)(newCapacity * loadFactor);
    HashEntry<K,V>[] newTable =
        (HashEntry<K,V>[]) new HashEntry[newCapacity];
    int sizeMask = newCapacity - 1;
    for (int i = 0; i < oldCapacity ; i++) {
        HashEntry<K,V> e = oldTable[i];
        if (e != null) {
            HashEntry<K,V> next = e.next;
            int idx = e.hash & sizeMask;
            if (next == null)   //  为空，没有后面的节点，直接给新数组
                newTable[idx] = e;
            else { // Reuse consecutive sequence at same slot
                HashEntry<K,V> lastRun = e;
                int lastIdx = idx;
                //因为数组是2倍的扩容，所以重新hash后，要么落在跟之前索引一样的位置，要么就是加上oldCapacity 的值，
                //比如容量是2，扩容4，现在hash是2，4，6，10，14那么后面3个都是除4余2，可以直接复制
                for (HashEntry<K,V> last = next;
                     last != null;
                     last = last.next) {
                    int k = last.hash & sizeMask;
                    if (k != lastIdx) {//hash不一样，重新
                        lastIdx = k;
                        lastRun = last;
                    }
                }
                //执行上面，就是lastRun是6，10，14
                newTable[lastIdx] = lastRun;//上面
                // Clone remaining nodes克隆的时候，碰到lastrun，直接根据所以给值，但是前面有可能的索引跟lastrun一样，比如2
                for (HashEntry<K,V> p = e; p != lastRun; p = p.next) {
                    V v = p.value;
                    int h = p.hash;
                    int k = h & sizeMask;
                    HashEntry<K,V> n = newTable[k];
                    newTable[k] = new HashEntry<K,V>(h, p.key, v, n);
                }
            }
        }
    }
    int nodeIndex = node.hash & sizeMask; // add the new node
    node.setNext(newTable[nodeIndex]);//加入到头结点
    newTable[nodeIndex] = node;
    table = newTable;
}

get方法

public V get(Object key) {
    Segment<K,V> s; // manually integrate access methods to reduce overhead
    HashEntry<K,V>[] tab;
    int h = hash(key);
    long u = (((h >>> segmentShift) & segmentMask) << SSHIFT) + SBASE;
    if ((s = (Segment<K,V>)UNSAFE.getObjectVolatile(segments, u)) != null &&//找到Segment，逻辑同put
        (tab = s.table) != null) {
        for (HashEntry<K,V> e = (HashEntry<K,V>) UNSAFE.getObjectVolatile
                 (tab, ((long)(((tab.length - 1) & h)) << TSHIFT) + TBASE);//找到table，逻辑同put
             e != null; e = e.next) {//遍历table
            K k;
            if ((k = e.key) == key || (e.hash == h && key.equals(k)))
                return e.value;
        }
    }
    return null;
}