分析下1.8的ConcurrentHashMap是怎么进行实现的,怎么保证数据安全的,跟1.7的差别在哪,做了哪些优化,还是从构造方法开始看吧
public ConcurrentHashMap(int initialCapacity, float loadFactor) {
// 默认并发粒度为1
this(initialCapacity, loadFactor, 1);
}
public ConcurrentHashMap(int initialCapacity,
float loadFactor, int concurrencyLevel) {
if (!(loadFactor > 0.0f) || initialCapacity < 0 || concurrencyLevel <= 0)
throw new IllegalArgumentException();
// 并发粒度如果大于容量,那么扩大容量
if (initialCapacity < concurrencyLevel) // Use at least as many bins
initialCapacity = concurrencyLevel; // as estimated threads
// 向上取整
long size = (long)(1.0 + (long)initialCapacity / loadFactor);
//最大值判断,然后获取对应的数组的长度还是2的n次幂
int cap = (size >= (long)MAXIMUM_CAPACITY) ?
MAXIMUM_CAPACITY : tableSizeFor((int)size);
this.sizeCtl = cap;
}
构造参数简单看下得到的是并发粒度默认的是1,初始化的容量含义跟之前的HashMap存在差别,之前是在*加载因子,现在直接作为实际的容量,去除以加载因子,得到需要初始化的数组长度,接下来还是看下put方法逻辑吧
put 放入数据
final V putVal(K key, V value, boolean onlyIfAbsent) {
// key 和value都不允许为空
if (key == null || value == null) throw new NullPointerException();
int hash = spread(key.hashCode());
int binCount = 0;
for (Node<K,V>[] tab = table;;) {
Node<K,V> f; int n, i, fh; K fk; V fv;
if (tab == null || (n = tab.length) == 0)
// tab为空的时候初始化table,单独分析
tab = initTable();
else if ((f = tabAt(tab, i = (n - 1) & hash)) == null) {
// 为空的时候cas插入
if (casTabAt(tab, i, null, new Node<K,V>(hash, key, value)))
break; // no lock when adding to empty bin
}
else if ((fh = f.hash) == MOVED)
// 如果正在转移数据,帮助转移,单独分析
tab = helpTransfer(tab, f);
else if (onlyIfAbsent // check first node without acquiring lock
&& fh == hash
&& ((fk = f.key) == key || (fk != null && key.equals(fk)))
&& (fv = f.val) != null)
// 这个是在只有不存在才插入的时候找到相同的直接返回
return fv;
else {
V oldVal = null;
synchronized (f)
// 加锁进行处理
if (tabAt(tab, i) == f) {
// 确认是锁的头节点
if (fh >= 0) {
// fh是当前节点hash值
binCount = 1;
for (Node<K,V> e = f;; ++binCount) {
K ek;
// 找到相同的替换返回
if (e.hash == hash &&
((ek = e.key) == key ||
(ek != null && key.equals(ek)))) {
oldVal = e.val;
if (!onlyIfAbsent)
e.val = value;
break;
}
// 这里是插入到尾部
Node<K,V> pred = e;
if ((e = e.next) == null) {
pred.next = new Node<K,V>(hash, key, value);
break;
}
}
}
// 如果是树节点
else if (f instanceof TreeBin) {
Node<K,V> p;
binCount = 2;
// 采用的树节点的插入,同hashMap
if ((p = ((TreeBin<K,V>)f).putTreeVal(hash, key,
value)) != null) {
oldVal = p.val;
if (!onlyIfAbsent)
p.val = value;
}
}
// 占位符节点
else if (f instanceof ReservationNode)
throw new IllegalStateException("Recursive update");
}
}
if (binCount != 0) {
if (binCount >= TREEIFY_THRESHOLD)
// 树化,同hashMap不单独分析
treeifyBin(tab, i);
if (oldVal != null)
return oldVal;
break;
}
}
}
// 增加计数,单独分析下
addCount(1L, binCount);
return null;
}
简单总结下,跟1.7的差别是不使用分段锁了,如果没有hash冲突直接cas替换下,不成功的话,再锁头节点进行安全操作,在synchronized里面操作,然后又了帮助转移数据操作,增加计数扩容方面,下面单独分析下这几块
初始化数组分析initTable
private final Node<K,V>[] initTable() {
Node<K,V>[] tab; int sc;
while ((tab = table) == null || tab.length == 0) {
if ((sc = sizeCtl) < 0)
Thread.yield(); // lost initialization race; just spin
// cas成功才能进去
else if (U.compareAndSetInt(this, SIZECTL, sc, -1)) {
try {
if ((tab = table) == null || tab.length == 0) {
// 数组胃为空的时候看下sc是不是正常的,否则就是默认值
int n = (sc > 0) ? sc : DEFAULT_CAPACITY;
@SuppressWarnings("unchecked")
Node<K,V>[] nt = (Node<K,V>[])new Node<?,?>[n];
table = tab = nt;
// 前面说了,当时给的容量直接除扩容因子得到的值赋值给其的
// 这里又等于对n*0.75的操作
sc = n - (n >>> 2);
}
} finally {
sizeCtl = sc;
}
break;
}
}
return tab;
}
可以看到是在初始化一个数组,数组长度为之前的sc,然后sc重新赋值为原值的0.75 倍,扩容因子没参与,可以后续看下sizeCtl 的作用,计算方式差异,数值使用方式的差异,都是小问题,下面看下,什么时候会触发扩容?怎么转移数据的,
addCount
private final void addCount(long x, int check) {
CounterCell[] cs; long b, s;
//counterCells 记录每个节点对应数量的,或者就是设置baseCount失败
if ((cs = counterCells) != null ||
!U.compareAndSetLong(this, BASECOUNT, b = baseCount, s = b + x)) {
CounterCell c; long v; int m;
boolean uncontended = true;
// 如果CounterCell 为null或者cs长度为0
//或者线程对应的位置为空或者就是给cs里面的cell设置值失败
if (cs == null || (m = cs.length - 1) < 0 ||
(c = cs[ThreadLocalRandom.getProbe() & m]) == null ||
!(uncontended =
U.compareAndSetLong(c, CELLVALUE, v = c.value, v + x))) {
// 走到这个进行设置,可以单独分析
fullAddCount(x, uncontended);
return;
}
if (check <= 1)
return;
// 计算下数量
s = sumCount();
}
if (check >= 0) {
Node<K,V>[] tab, nt; int n, sc;
while (s >= (long)(sc = sizeCtl) && (tab = table) != null &&
(n = tab.length) < MAXIMUM_CAPACITY) {
int rs = resizeStamp(n);
if (sc < 0) {
if ((sc >>> RESIZE_STAMP_SHIFT) != rs || sc == rs + 1 ||
sc == rs + MAX_RESIZERS || (nt = nextTable) == null ||
transferIndex <= 0)
break;
if (U.compareAndSetInt(this, SIZECTL, sc, sc + 1))
transfer(tab, nt);
}
else if (U.compareAndSetInt(this, SIZECTL, sc,
(rs << RESIZE_STAMP_SHIFT) + 2))
transfer(tab, null);
s = sumCount();
}
}
}
这里主要就是在增加对应的数量,map里面存放了多少元素,采用cas设置,cas设置不到再用的CountCell数组,然后这里是check是大于0还是小于0,代表的是增加元素还是删除元素,下面增加元素的时候,会判断是不是需要扩容,不过不是所有都能走到这里,比如上面走到**fullAddCount **的时候,后面增加完元素数量就不会出发扩容了,直接返回了,扩容触发条件比HashMap更加复杂
fullAddCount
private final void fullAddCount(long x, boolean wasUncontended) {
int h;
// 线程的标志
if ((h = ThreadLocalRandom.getProbe()) == 0) {
ThreadLocalRandom.localInit(); // force initialization
h = ThreadLocalRandom.getProbe();
wasUncontended = true;
}
boolean collide = false; // True if last slot nonempty
for (;;) {
CounterCell[] cs; CounterCell c; int n; long v;
// cs不为空
if ((cs = counterCells) != null && (n = cs.length) > 0) {
// 线程对应的为空
if ((c = cs[(n - 1) & h]) == null) {
if (cellsBusy == 0) { // Try to attach new Cell
CounterCell r = new CounterCell(x); // Optimistic create
//cas赋值,等于对cellsBusy加锁
if (cellsBusy == 0 &&
U.compareAndSetInt(this, CELLSBUSY, 0, 1)) {
// 成功之后
boolean created = false;
try { // Recheck under lock
CounterCell[] rs; int m, j;
// counterCells不为空,长度大于0,但是线程对应的为空
if ((rs = counterCells) != null &&
(m = rs.length) > 0 &&
rs[j = (m - 1) & h] == null) {
// 直接尽性赋值
rs[j] = r;
created = true;
}
} finally {
cellsBusy = 0;
}
if (created)
break;
continue; // Slot is now non-empty
}
}
collide = false;
}
else if (!wasUncontended) // CAS already known to fail
wasUncontended = true; // Continue after rehash
//直接给cell cas赋值成功
else if (U.compareAndSetLong(c, CELLVALUE, v = c.value, v + x))
break;
// 不相等了(被人改了) 或者大与等于cpu的个数
else if (counterCells != cs || n >= NCPU)
collide = false; // At max size or stale
else if (!collide)
collide = true;
// cas获取busy的锁
else if (cellsBusy == 0 &&
U.compareAndSetInt(this, CELLSBUSY, 0, 1)) {
try {
// 如果相等,扩容了原来的二倍,然后把源数据拷贝过去了
if (counterCells == cs) // Expand table unless stale
counterCells = Arrays.copyOf(cs, n << 1);
} finally {
cellsBusy = 0;
}
collide = false;
continue; // Retry with expanded table
}
h = ThreadLocalRandom.advanceProbe(h);
}
else if (cellsBusy == 0 && counterCells == cs &&
U.compareAndSetInt(this, CELLSBUSY, 0, 1)) {
boolean init = false;
try { // Initialize table
// 初始化为2
if (counterCells == cs) {
CounterCell[] rs = new CounterCell[2];
rs[h & 1] = new CounterCell(x);
counterCells = rs;
init = true;
}
} finally {
cellsBusy = 0;
}
if (init)
break;
}
// 直接给baseCount加上了
else if (U.compareAndSetLong(this, BASECOUNT, v = baseCount, v + x))
break; // Fall back on using base
}
}
这里是在设置增加值冲突的时候走到这里,采用了一个CountCell的数组cas增加,baseCount竞争不到时候,走CountCell,提高并发粒度
转移数据transfer(tab, nextTab);
sizeCtl :默认为0,用来控制table的初始化和扩容操作,具体应用在后续会体现出来。
- -1 代表table正在初始化
- -N 表示有N-1个线程正在进行扩容操作
- 其余情况:
1、如果table未初始化,表示table需要初始化的大小。
2、如果table初始化完成,表示table的容量,默认是table大小的0.75倍
private final void transfer(Node<K,V>[] tab, Node<K,V>[] nextTab) {
int n = tab.length, stride;
if ((stride = (NCPU > 1) ? (n >>> 3) / NCPU : n) < MIN_TRANSFER_STRIDE)
stride = MIN_TRANSFER_STRIDE; // subdivide range
if (nextTab == null) { // initiating
try {
@SuppressWarnings("unchecked")
// 二倍大小的新的数组
Node<K,V>[] nt = (Node<K,V>[])new Node<?,?>[n << 1];
nextTab = nt;
} catch (Throwable ex) { // try to cope with OOME
sizeCtl = Integer.MAX_VALUE;
return;
}
nextTable = nextTab;
// 这个数字从数组大小开始的
transferIndex = n;
}
int nextn = nextTab.length;
// fwd节点,证明在转移的占位节点
ForwardingNode<K,V> fwd = new ForwardingNode<K,V>(nextTab);
boolean advance = true;
boolean finishing = false; // to ensure sweep before committing nextTab
for (int i = 0, bound = 0;;) {
Node<K,V> f; int fh;
while (advance) {
int nextIndex, nextBound;
if (--i >= bound || finishing)
advance = false;
else if ((nextIndex = transferIndex) <= 0) {
i = -1;
advance = false;
}
//这里是在分配节点,从后向前数一定的步长
else if (U.compareAndSetInt
(this, TRANSFERINDEX, nextIndex,
nextBound = (nextIndex > stride ?
nextIndex - stride : 0))) {
bound = nextBound;
i = nextIndex - 1;
advance = false;
}
}
if (i < 0 || i >= n || i + n >= nextn) {
int sc;
if (finishing) {
nextTable = null;
table = nextTab;
sizeCtl = (n << 1) - (n >>> 1);
return;
}
if (U.compareAndSetInt(this, SIZECTL, sc = sizeCtl, sc - 1)) {
if ((sc - 2) != resizeStamp(n) << RESIZE_STAMP_SHIFT)
return;
finishing = advance = true;
i = n; // recheck before commit
}
}
else if ((f = tabAt(tab, i)) == null)
advance = casTabAt(tab, i, null, fwd);
else if ((fh = f.hash) == MOVED)
advance = true; // already processed
else {
// 锁住头节点
synchronized (f) {
if (tabAt(tab, i) == f) {
Node<K,V> ln, hn;
if (fh >= 0) {
int runBit = fh & n;
Node<K,V> lastRun = f;
for (Node<K,V> p = f.next; p != null; p = p.next) {
int b = p.hash & n;
// 需要移动下标的,跟现在下标不同
if (b != runBit) {
runBit = b;
lastRun = p;
}
}
if (runBit == 0) {
ln = lastRun;
hn = null;
}
else {
hn = lastRun;
ln = null;
}
// 遍历链表,组成俩新的链表,然后插入到数组上
for (Node<K,V> p = f; p != lastRun; p = p.next) {
int ph = p.hash; K pk = p.key; V pv = p.val;
if ((ph & n) == 0)
ln = new Node<K,V>(ph, pk, pv, ln);
else
hn = new Node<K,V>(ph, pk, pv, hn);
}
setTabAt(nextTab, i, ln);
setTabAt(nextTab, i + n, hn);
setTabAt(tab, i, fwd);
advance = true;
}
// 是树的时候
else if (f instanceof TreeBin) {
TreeBin<K,V> t = (TreeBin<K,V>)f;
TreeNode<K,V> lo = null, loTail = null;
TreeNode<K,V> hi = null, hiTail = null;
int lc = 0, hc = 0;
// 原理同hashMap
for (Node<K,V> e = t.first; e != null; e = e.next) {
int h = e.hash;
TreeNode<K,V> p = new TreeNode<K,V>
(h, e.key, e.val, null, null);
if ((h & n) == 0) {
if ((p.prev = loTail) == null)
lo = p;
else
loTail.next = p;
loTail = p;
++lc;
}
else {
if ((p.prev = hiTail) == null)
hi = p;
else
hiTail.next = p;
hiTail = p;
++hc;
}
}
ln = (lc <= UNTREEIFY_THRESHOLD) ? untreeify(lo) :
(hc != 0) ? new TreeBin<K,V>(lo) : t;
hn = (hc <= UNTREEIFY_THRESHOLD) ? untreeify(hi) :
(lc != 0) ? new TreeBin<K,V>(hi) : t;
setTabAt(nextTab, i, ln);
setTabAt(nextTab, i + n, hn);
setTabAt(tab, i, fwd);
advance = true;
}
}
}
}
}
}
转移数据的时候会锁定头节点然后进行数据转移,树节点的TreeBIn不同于HashMap的TreeNode,这里是一整颗树,不是一个单独的树节点,为了防止红黑树的重新平衡导致的更换跟节点导致的数据加锁无效,
,helpTransfer 帮助转移数据
final Node<K,V>[] helpTransfer(Node<K,V>[] tab, Node<K,V> f) {
Node<K,V>[] nextTab; int sc;
// tab存在并且是
if (tab != null && (f instanceof ForwardingNode) &&
(nextTab = ((ForwardingNode<K,V>)f).nextTable) != null) {
int rs = resizeStamp(tab.length);
// 小于0的时候,正在扩容
while (nextTab == nextTable && table == tab &&
(sc = sizeCtl) < 0) {
// 在判断是不是符合帮助扩容的条件
if ((sc >>> RESIZE_STAMP_SHIFT) != rs || sc == rs + 1 ||
sc == rs + MAX_RESIZERS || transferIndex <= 0)
break;
// 每个线程对其扩容的时候sc的低16位记录正在扩容的线程个数
if (U.compareAndSetInt(this, SIZECTL, sc, sc + 1)) {
transfer(tab, nextTab);
break;
}
}
return nextTab;
}
return table;
}
总结
对于1.8的concurrentHashMap比较复杂,有些点没分析透彻,不过先看最主要的流程吧,就是并发粒度比较高,等于数组的大小,存放数据如果数组对应位置没有值,直接cas放上去,如果存在值,需要锁对应数组节点,然后对其进行插入,结构同HashMap的链表和红黑树,不过这里成树的时候就是存的树对象,而不是存的一个树节点(防止红黑树插入之后旋转更换根节点),扩容机制类似,不过是不是HashMap那样超过那么多一定触发扩容,还有点就是扩容,扩容的时候如果有新的数据进来,会帮助一起扩容,按照步长,大家一起瓜分节点,扩容的时候也类似HashMap1.8的,分为高低段,使用&得到的下标判断,组起来之后一起放到对应数组位置,对于get就是直接获取,因为volitale修饰的保证数据可见性,存储的数据数量统计也做了并发处理,直接cas失败的时候增加了一个CountCell数组来计数,想得到总数的时候就把这些都加起来。