ViewVC Help

View File | Revision Log | Show Annotations | Download File | Root Listing

root/jsr166/jsr166/src/jsr166e/ConcurrentHashMapV8.java

Comparing jsr166/src/jsr166e/ConcurrentHashMapV8.java (file contents):
Revision 1.29 by dl, Mon Oct 3 11:20:47 2011 UTC vs.
Revision 1.106 by jsr166, Wed Jun 19 17:11:57 2013 UTC

#	Line 5 | Line 5
5		*/
6
7		package jsr166e;
8	<	import jsr166e.LongAdder;
8	>
9	>	import jsr166e.ForkJoinPool;
10	>
11	>	import java.io.ObjectStreamField;
12	>	import java.io.Serializable;
13	>	import java.lang.reflect.ParameterizedType;
14	>	import java.lang.reflect.Type;
15		import java.util.Arrays;
10	–	import java.util.Map;
11	–	import java.util.Set;
16		import java.util.Collection;
17	<	import java.util.AbstractMap;
18	<	import java.util.AbstractSet;
19	<	import java.util.AbstractCollection;
16	<	import java.util.Hashtable;
17	>	import java.util.Comparator;
18	>	import java.util.ConcurrentModificationException;
19	>	import java.util.Enumeration;
20		import java.util.HashMap;
21	+	import java.util.Hashtable;
22		import java.util.Iterator;
23	<	import java.util.Enumeration;
20	<	import java.util.ConcurrentModificationException;
23	>	import java.util.Map;
24		import java.util.NoSuchElementException;
25	+	import java.util.Set;
26		import java.util.concurrent.ConcurrentMap;
27	+	import java.util.concurrent.atomic.AtomicReference;
28	+	import java.util.concurrent.atomic.AtomicInteger;
29		import java.util.concurrent.locks.LockSupport;
30	<	import java.io.Serializable;
30	>	import java.util.concurrent.locks.ReentrantLock;
31
32		/**
33		* A hash table supporting full concurrency of retrievals and
#	Line 35 | Line 41 | import java.io.Serializable;
41		* interoperable with {@code Hashtable} in programs that rely on its
42		* thread safety but not on its synchronization details.
43		*
44	<	* <p> Retrieval operations (including {@code get}) generally do not
44	>	* <p>Retrieval operations (including {@code get}) generally do not
45		* block, so may overlap with update operations (including {@code put}
46		* and {@code remove}). Retrievals reflect the results of the most
47		* recently <em>completed</em> update operations holding upon their
48	<	* onset.  For aggregate operations such as {@code putAll} and {@code
49	<	* clear}, concurrent retrievals may reflect insertion or removal of
50	<	* only some entries.  Similarly, Iterators and Enumerations return
51	<	* elements reflecting the state of the hash table at some point at or
52	<	* since the creation of the iterator/enumeration.  They do
53	<	* <em>not</em> throw {@link ConcurrentModificationException}.
54	<	* However, iterators are designed to be used by only one thread at a
55	<	* time.  Bear in mind that the results of aggregate status methods
56	<	* including {@code size}, {@code isEmpty}, and {@code containsValue}
57	<	* are typically useful only when a map is not undergoing concurrent
58	<	* updates in other threads.  Otherwise the results of these methods
59	<	* reflect transient states that may be adequate for monitoring
60	<	* or estimation purposes, but not for program control.
48	>	* onset. (More formally, an update operation for a given key bears a
49	>	* <em>happens-before</em> relation with any (non-null) retrieval for
50	>	* that key reporting the updated value.)  For aggregate operations
51	>	* such as {@code putAll} and {@code clear}, concurrent retrievals may
52	>	* reflect insertion or removal of only some entries.  Similarly,
53	>	* Iterators and Enumerations return elements reflecting the state of
54	>	* the hash table at some point at or since the creation of the
55	>	* iterator/enumeration.  They do <em>not</em> throw {@link
56	>	* ConcurrentModificationException}.  However, iterators are designed
57	>	* to be used by only one thread at a time.  Bear in mind that the
58	>	* results of aggregate status methods including {@code size}, {@code
59	>	* isEmpty}, and {@code containsValue} are typically useful only when
60	>	* a map is not undergoing concurrent updates in other threads.
61	>	* Otherwise the results of these methods reflect transient states
62	>	* that may be adequate for monitoring or estimation purposes, but not
63	>	* for program control.
64		*
65	<	* <p> The table is dynamically expanded when there are too many
65	>	* <p>The table is dynamically expanded when there are too many
66		* collisions (i.e., keys that have distinct hash codes but fall into
67		* the same slot modulo the table size), with the expected average
68		* effect of maintaining roughly two bins per mapping (corresponding
#	Line 71 | Line 80 | import java.io.Serializable;
80		* versions of this class, constructors may optionally specify an
81		* expected {@code concurrencyLevel} as an additional hint for
82		* internal sizing.  Note that using many keys with exactly the same
83	<	* {@code hashCode{}} is a sure way to slow down performance of any
84	<	* hash table.
83	>	* {@code hashCode()} is a sure way to slow down performance of any
84	>	* hash table. To ameliorate impact, when keys are {@link Comparable},
85	>	* this class may use comparison order among keys to help break ties.
86	>	*
87	>	* <p>A {@link Set} projection of a ConcurrentHashMapV8 may be created
88	>	* (using {@link #newKeySet()} or {@link #newKeySet(int)}), or viewed
89	>	* (using {@link #keySet(Object)} when only keys are of interest, and the
90	>	* mapped values are (perhaps transiently) not used or all take the
91	>	* same mapping value.
92		*
93		* <p>This class and its views and iterators implement all of the
94		* <em>optional</em> methods of the {@link Map} and {@link Iterator}
95		* interfaces.
96		*
97	<	* <p> Like {@link Hashtable} but unlike {@link HashMap}, this class
97	>	* <p>Like {@link Hashtable} but unlike {@link HashMap}, this class
98		* does <em>not</em> allow {@code null} to be used as a key or value.
99		*
100	+	* <p>ConcurrentHashMapV8s support a set of sequential and parallel bulk
101	+	* operations that are designed
102	+	* to be safely, and often sensibly, applied even with maps that are
103	+	* being concurrently updated by other threads; for example, when
104	+	* computing a snapshot summary of the values in a shared registry.
105	+	* There are three kinds of operation, each with four forms, accepting
106	+	* functions with Keys, Values, Entries, and (Key, Value) arguments
107	+	* and/or return values. Because the elements of a ConcurrentHashMapV8
108	+	* are not ordered in any particular way, and may be processed in
109	+	* different orders in different parallel executions, the correctness
110	+	* of supplied functions should not depend on any ordering, or on any
111	+	* other objects or values that may transiently change while
112	+	* computation is in progress; and except for forEach actions, should
113	+	* ideally be side-effect-free. Bulk operations on {@link java.util.Map.Entry}
114	+	* objects do not support method {@code setValue}.
115	+	*
116	+	* <ul>
117	+	* <li> forEach: Perform a given action on each element.
118	+	* A variant form applies a given transformation on each element
119	+	* before performing the action.</li>
120	+	*
121	+	* <li> search: Return the first available non-null result of
122	+	* applying a given function on each element; skipping further
123	+	* search when a result is found.</li>
124	+	*
125	+	* <li> reduce: Accumulate each element.  The supplied reduction
126	+	* function cannot rely on ordering (more formally, it should be
127	+	* both associative and commutative).  There are five variants:
128	+	*
129	+	* <ul>
130	+	*
131	+	* <li> Plain reductions. (There is not a form of this method for
132	+	* (key, value) function arguments since there is no corresponding
133	+	* return type.)</li>
134	+	*
135	+	* <li> Mapped reductions that accumulate the results of a given
136	+	* function applied to each element.</li>
137	+	*
138	+	* <li> Reductions to scalar doubles, longs, and ints, using a
139	+	* given basis value.</li>
140	+	*
141	+	* </ul>
142	+	* </li>
143	+	* </ul>
144	+	*
145	+	* <p>These bulk operations accept a {@code parallelismThreshold}
146	+	* argument. Methods proceed sequentially if the current map size is
147	+	* estimated to be less than the given threshold. Using a value of
148	+	* {@code Long.MAX_VALUE} suppresses all parallelism.  Using a value
149	+	* of {@code 1} results in maximal parallelism by partitioning into
150	+	* enough subtasks to fully utilize the {@link
151	+	* ForkJoinPool#commonPool()} that is used for all parallel
152	+	* computations. Normally, you would initially choose one of these
153	+	* extreme values, and then measure performance of using in-between
154	+	* values that trade off overhead versus throughput.
155	+	*
156	+	* <p>The concurrency properties of bulk operations follow
157	+	* from those of ConcurrentHashMapV8: Any non-null result returned
158	+	* from {@code get(key)} and related access methods bears a
159	+	* happens-before relation with the associated insertion or
160	+	* update.  The result of any bulk operation reflects the
161	+	* composition of these per-element relations (but is not
162	+	* necessarily atomic with respect to the map as a whole unless it
163	+	* is somehow known to be quiescent).  Conversely, because keys
164	+	* and values in the map are never null, null serves as a reliable
165	+	* atomic indicator of the current lack of any result.  To
166	+	* maintain this property, null serves as an implicit basis for
167	+	* all non-scalar reduction operations. For the double, long, and
168	+	* int versions, the basis should be one that, when combined with
169	+	* any other value, returns that other value (more formally, it
170	+	* should be the identity element for the reduction). Most common
171	+	* reductions have these properties; for example, computing a sum
172	+	* with basis 0 or a minimum with basis MAX_VALUE.
173	+	*
174	+	* <p>Search and transformation functions provided as arguments
175	+	* should similarly return null to indicate the lack of any result
176	+	* (in which case it is not used). In the case of mapped
177	+	* reductions, this also enables transformations to serve as
178	+	* filters, returning null (or, in the case of primitive
179	+	* specializations, the identity basis) if the element should not
180	+	* be combined. You can create compound transformations and
181	+	* filterings by composing them yourself under this "null means
182	+	* there is nothing there now" rule before using them in search or
183	+	* reduce operations.
184	+	*
185	+	* <p>Methods accepting and/or returning Entry arguments maintain
186	+	* key-value associations. They may be useful for example when
187	+	* finding the key for the greatest value. Note that "plain" Entry
188	+	* arguments can be supplied using {@code new
189	+	* AbstractMap.SimpleEntry(k,v)}.
190	+	*
191	+	* <p>Bulk operations may complete abruptly, throwing an
192	+	* exception encountered in the application of a supplied
193	+	* function. Bear in mind when handling such exceptions that other
194	+	* concurrently executing functions could also have thrown
195	+	* exceptions, or would have done so if the first exception had
196	+	* not occurred.
197	+	*
198	+	* <p>Speedups for parallel compared to sequential forms are common
199	+	* but not guaranteed.  Parallel operations involving brief functions
200	+	* on small maps may execute more slowly than sequential forms if the
201	+	* underlying work to parallelize the computation is more expensive
202	+	* than the computation itself.  Similarly, parallelization may not
203	+	* lead to much actual parallelism if all processors are busy
204	+	* performing unrelated tasks.
205	+	*
206	+	* <p>All arguments to all task methods must be non-null.
207	+	*
208	+	* <p><em>jsr166e note: During transition, this class
209	+	* uses nested functional interfaces with different names but the
210	+	* same forms as those expected for JDK8.</em>
211	+	*
212		* <p>This class is a member of the
213		* <a href="{@docRoot}/../technotes/guides/collections/index.html">
214		* Java Collections Framework</a>.
215		*
88	–	* <p><em>jsr166e note: This class is a candidate replacement for
89	–	* java.util.concurrent.ConcurrentHashMap.<em>
90	–	*
216		* @since 1.5
217		* @author Doug Lea
218		* @param <K> the type of keys maintained by this map
219		* @param <V> the type of mapped values
220		*/
221	<	public class ConcurrentHashMapV8<K, V>
222	<	implements ConcurrentMap<K, V>, Serializable {
221	>	public class ConcurrentHashMapV8<K,V>
222	>	implements ConcurrentMap<K,V>, Serializable {
223		private static final long serialVersionUID = 7249069246763182397L;
224
225		/**
226	<	* A function computing a mapping from the given key to a value.
227	<	* This is a place-holder for an upcoming JDK8 interface.
226	>	* An object for traversing and partitioning elements of a source.
227	>	* This interface provides a subset of the functionality of JDK8
228	>	* java.util.Spliterator.
229		*/
230	<	public static interface MappingFunction<K, V> {
230	>	public static interface ConcurrentHashMapSpliterator<T> {
231		/**
232	<	* Returns a non-null value for the given key.
233	<	*
234	<	* @param key the (non-null) key
235	<	* @return a non-null value
232	>	* If possible, returns a new spliterator covering
233	>	* approximately one half of the elements, which will not be
234	>	* covered by this spliterator. Returns null if cannot be
235	>	* split.
236		*/
237	<	V map(K key);
112	<	}
113	<
114	<	/**
115	<	* A function computing a new mapping given a key and its current
116	<	* mapped value (or {@code null} if there is no current
117	<	* mapping). This is a place-holder for an upcoming JDK8
118	<	* interface.
119	<	*/
120	<	public static interface RemappingFunction<K, V> {
237	>	ConcurrentHashMapSpliterator<T> trySplit();
238		/**
239	<	* Returns a new value given a key and its current value.
240	<	*
124	<	* @param key the (non-null) key
125	<	* @param value the current value, or null if there is no mapping
126	<	* @return a non-null value
239	>	* Returns an estimate of the number of elements covered by
240	>	* this Spliterator.
241		*/
242	<	V remap(K key, V value);
243	<	}
242	>	long estimateSize();
243	>
244	>	/** Applies the action to each untraversed element */
245	>	void forEachRemaining(Action<? super T> action);
246	>	/** If an element remains, applies the action and returns true. */
247	>	boolean tryAdvance(Action<? super T> action);
248	>	}
249	>
250	>	// Sams
251	>	/** Interface describing a void action of one argument */
252	>	public interface Action<A> { void apply(A a); }
253	>	/** Interface describing a void action of two arguments */
254	>	public interface BiAction<A,B> { void apply(A a, B b); }
255	>	/** Interface describing a function of one argument */
256	>	public interface Fun<A,T> { T apply(A a); }
257	>	/** Interface describing a function of two arguments */
258	>	public interface BiFun<A,B,T> { T apply(A a, B b); }
259	>	/** Interface describing a function mapping its argument to a double */
260	>	public interface ObjectToDouble<A> { double apply(A a); }
261	>	/** Interface describing a function mapping its argument to a long */
262	>	public interface ObjectToLong<A> { long apply(A a); }
263	>	/** Interface describing a function mapping its argument to an int */
264	>	public interface ObjectToInt<A> {int apply(A a); }
265	>	/** Interface describing a function mapping two arguments to a double */
266	>	public interface ObjectByObjectToDouble<A,B> { double apply(A a, B b); }
267	>	/** Interface describing a function mapping two arguments to a long */
268	>	public interface ObjectByObjectToLong<A,B> { long apply(A a, B b); }
269	>	/** Interface describing a function mapping two arguments to an int */
270	>	public interface ObjectByObjectToInt<A,B> {int apply(A a, B b); }
271	>	/** Interface describing a function mapping two doubles to a double */
272	>	public interface DoubleByDoubleToDouble { double apply(double a, double b); }
273	>	/** Interface describing a function mapping two longs to a long */
274	>	public interface LongByLongToLong { long apply(long a, long b); }
275	>	/** Interface describing a function mapping two ints to an int */
276	>	public interface IntByIntToInt { int apply(int a, int b); }
277
278		/*
279		* Overview:
#	Line 138 | Line 285 | public class ConcurrentHashMapV8<K, V>
285		* the same or better than java.util.HashMap, and to support high
286		* initial insertion rates on an empty table by many threads.
287		*
288	<	* Each key-value mapping is held in a Node.  Because Node fields
289	<	* can contain special values, they are defined using plain Object
290	<	* types. Similarly in turn, all internal methods that use them
291	<	* work off Object types. And similarly, so do the internal
292	<	* methods of auxiliary iterator and view classes.  All public
293	<	* generic typed methods relay in/out of these internal methods,
294	<	* supplying null-checks and casts as needed. This also allows
295	<	* many of the public methods to be factored into a smaller number
296	<	* of internal methods (although sadly not so for the five
297	<	* sprawling variants of put-related operations).
288	>	* This map usually acts as a binned (bucketed) hash table.  Each
289	>	* key-value mapping is held in a Node.  Most nodes are instances
290	>	* of the basic Node class with hash, key, value, and next
291	>	* fields. However, various subclasses exist: TreeNodes are
292	>	* arranged in balanced trees, not lists.  TreeBins hold the roots
293	>	* of sets of TreeNodes. ForwardingNodes are placed at the heads
294	>	* of bins during resizing. ReservationNodes are used as
295	>	* placeholders while establishing values in computeIfAbsent and
296	>	* related methods.  The types TreeBin, ForwardingNode, and
297	>	* ReservationNode do not hold normal user keys, values, or
298	>	* hashes, and are readily distinguishable during search etc
299	>	* because they have negative hash fields and null key and value
300	>	* fields. (These special nodes are either uncommon or transient,
301	>	* so the impact of carrying around some unused fields is
302	>	* insignficant.)
303		*
304		* The table is lazily initialized to a power-of-two size upon the
305	<	* first insertion.  Each bin in the table contains a list of
306	<	* Nodes (most often, the list has only zero or one Node).  Table
307	<	* accesses require volatile/atomic reads, writes, and CASes.
308	<	* Because there is no other way to arrange this without adding
309	<	* further indirections, we use intrinsics (sun.misc.Unsafe)
310	<	* operations.  The lists of nodes within bins are always
311	<	* accurately traversable under volatile reads, so long as lookups
312	<	* check hash code and non-nullness of value before checking key
313	<	* equality.
314	<	*
315	<	* We use the top two bits of Node hash fields for control
164	<	* purposes -- they are available anyway because of addressing
165	<	* constraints.  As explained further below, these top bits are
166	<	* used as follows:
167	<	*  00 - Normal
168	<	*  01 - Locked
169	<	*  11 - Locked and may have a thread waiting for lock
170	<	*  10 - Node is a forwarding node
171	<	*
172	<	* The lower 30 bits of each Node's hash field contain a
173	<	* transformation (for better randomization -- method "spread") of
174	<	* the key's hash code, except for forwarding nodes, for which the
175	<	* lower bits are zero (and so always have hash field == MOVED).
305	>	* first insertion.  Each bin in the table normally contains a
306	>	* list of Nodes (most often, the list has only zero or one Node).
307	>	* Table accesses require volatile/atomic reads, writes, and
308	>	* CASes.  Because there is no other way to arrange this without
309	>	* adding further indirections, we use intrinsics
310	>	* (sun.misc.Unsafe) operations.
311	>	*
312	>	* We use the top (sign) bit of Node hash fields for control
313	>	* purposes -- it is available anyway because of addressing
314	>	* constraints.  Nodes with negative hash fields are specially
315	>	* handled or ignored in map methods.
316		*
317		* Insertion (via put or its variants) of the first node in an
318		* empty bin is performed by just CASing it to the bin.  This is
319	<	* by far the most common case for put operations.  Other update
320	<	* operations (insert, delete, and replace) require locks.  We do
321	<	* not want to waste the space required to associate a distinct
322	<	* lock object with each bin, so instead use the first node of a
323	<	* bin list itself as a lock. Blocking support for these locks
324	<	* relies on the builtin "synchronized" monitors.  However, we
325	<	* also need a tryLock construction, so we overlay these by using
186	<	* bits of the Node hash field for lock control (see above), and
187	<	* so normally use builtin monitors only for blocking and
188	<	* signalling using wait/notifyAll constructions. See
189	<	* Node.tryAwaitLock.
319	>	* by far the most common case for put operations under most
320	>	* key/hash distributions.  Other update operations (insert,
321	>	* delete, and replace) require locks.  We do not want to waste
322	>	* the space required to associate a distinct lock object with
323	>	* each bin, so instead use the first node of a bin list itself as
324	>	* a lock. Locking support for these locks relies on builtin
325	>	* "synchronized" monitors.
326		*
327		* Using the first node of a list as a lock does not by itself
328		* suffice though: When a node is locked, any update must first
329		* validate that it is still the first node after locking it, and
330		* retry if not. Because new nodes are always appended to lists,
331		* once a node is first in a bin, it remains first until deleted
332	<	* or the bin becomes invalidated (upon resizing).  However,
197	<	* operations that only conditionally update may inspect nodes
198	<	* until the point of update. This is a converse of sorts to the
199	<	* lazy locking technique described by Herlihy & Shavit.
332	>	* or the bin becomes invalidated (upon resizing).
333		*
334		* The main disadvantage of per-bin locks is that other update
335		* operations on other nodes in a bin list protected by the same
336		* lock can stall, for example when user equals() or mapping
337	<	* functions take a long time.  However, statistically, this is
338	<	* not a common enough problem to outweigh the time/space overhead
339	<	* of alternatives: Under random hash codes, the frequency of
207	<	* nodes in bins follows a Poisson distribution
337	>	* functions take a long time.  However, statistically, under
338	>	* random hash codes, this is not a common problem.  Ideally, the
339	>	* frequency of nodes in bins follows a Poisson distribution
340		* (http://en.wikipedia.org/wiki/Poisson_distribution) with a
341		* parameter of about 0.5 on average, given the resizing threshold
342		* of 0.75, although with a large variance because of resizing
343		* granularity. Ignoring variance, the expected occurrences of
344		* list size k are (exp(-0.5) * pow(0.5, k) / factorial(k)). The
345	<	* first few values are:
345	>	* first values are:
346		*
347	<	* 0:    0.607
348	<	* 1:    0.303
349	<	* 2:    0.076
350	<	* 3:    0.012
351	<	* more: 0.002
347	>	* 0:    0.60653066
348	>	* 1:    0.30326533
349	>	* 2:    0.07581633
350	>	* 3:    0.01263606
351	>	* 4:    0.00157952
352	>	* 5:    0.00015795
353	>	* 6:    0.00001316
354	>	* 7:    0.00000094
355	>	* 8:    0.00000006
356	>	* more: less than 1 in ten million
357		*
358		* Lock contention probability for two threads accessing distinct
359	<	* elements is roughly 1 / (8 * #elements).  Function "spread"
360	<	* performs hashCode randomization that improves the likelihood
361	<	* that these assumptions hold unless users define exactly the
362	<	* same value for too many hashCodes.
363	<	*
364	<	* The table is resized when occupancy exceeds an occupancy
365	<	* threshold (nominally, 0.75, but see below).  Only a single
366	<	* thread performs the resize (using field "sizeCtl", to arrange
367	<	* exclusion), but the table otherwise remains usable for reads
368	<	* and updates. Resizing proceeds by transferring bins, one by
369	<	* one, from the table to the next table.  Because we are using
370	<	* power-of-two expansion, the elements from each bin must either
371	<	* stay at same index, or move with a power of two offset. We
372	<	* eliminate unnecessary node creation by catching cases where old
373	<	* nodes can be reused because their next fields won't change.  On
374	<	* average, only about one-sixth of them need cloning when a table
375	<	* doubles. The nodes they replace will be garbage collectable as
376	<	* soon as they are no longer referenced by any reader thread that
377	<	* may be in the midst of concurrently traversing table.  Upon
378	<	* transfer, the old table bin contains only a special forwarding
379	<	* node (with hash field "MOVED") that contains the next table as
380	<	* its key. On encountering a forwarding node, access and update
381	<	* operations restart, using the new table.
382	<	*
383	<	* Each bin transfer requires its bin lock. However, unlike other
384	<	* cases, a transfer can skip a bin if it fails to acquire its
385	<	* lock, and revisit it later. Method rebuild maintains a buffer
386	<	* of TRANSFER_BUFFER_SIZE bins that have been skipped because of
387	<	* failure to acquire a lock, and blocks only if none are
388	<	* available (i.e., only very rarely).  The transfer operation
389	<	* must also ensure that all accessible bins in both the old and
390	<	* new table are usable by any traversal.  When there are no lock
391	<	* acquisition failures, this is arranged simply by proceeding
392	<	* from the last bin (table.length - 1) up towards the first.
393	<	* Upon seeing a forwarding node, traversals (see class
394	<	* InternalIterator) arrange to move to the new table without
395	<	* revisiting nodes.  However, when any node is skipped during a
396	<	* transfer, all earlier table bins may have become visible, so
397	<	* are initialized with a reverse-forwarding node back to the old
398	<	* table until the new ones are established. (This sometimes
399	<	* requires transiently locking a forwarding node, which is
400	<	* possible under the above encoding.) These more expensive
401	<	* mechanics trigger only when necessary.
359	>	* elements is roughly 1 / (8 * #elements) under random hashes.
360	>	*
361	>	* Actual hash code distributions encountered in practice
362	>	* sometimes deviate significantly from uniform randomness.  This
363	>	* includes the case when N > (1<<30), so some keys MUST collide.
364	>	* Similarly for dumb or hostile usages in which multiple keys are
365	>	* designed to have identical hash codes or ones that differs only
366	>	* in masked-out high bits. So we use a secondary strategy that
367	>	* applies when the number of nodes in a bin exceeds a
368	>	* threshold. These TreeBins use a balanced tree to hold nodes (a
369	>	* specialized form of red-black trees), bounding search time to
370	>	* O(log N).  Each search step in a TreeBin is at least twice as
371	>	* slow as in a regular list, but given that N cannot exceed
372	>	* (1<<64) (before running out of addresses) this bounds search
373	>	* steps, lock hold times, etc, to reasonable constants (roughly
374	>	* 100 nodes inspected per operation worst case) so long as keys
375	>	* are Comparable (which is very common -- String, Long, etc).
376	>	* TreeBin nodes (TreeNodes) also maintain the same "next"
377	>	* traversal pointers as regular nodes, so can be traversed in
378	>	* iterators in the same way.
379	>	*
380	>	* The table is resized when occupancy exceeds a percentage
381	>	* threshold (nominally, 0.75, but see below).  Any thread
382	>	* noticing an overfull bin may assist in resizing after the
383	>	* initiating thread allocates and sets up the replacement
384	>	* array. However, rather than stalling, these other threads may
385	>	* proceed with insertions etc.  The use of TreeBins shields us
386	>	* from the worst case effects of overfilling while resizes are in
387	>	* progress.  Resizing proceeds by transferring bins, one by one,
388	>	* from the table to the next table. To enable concurrency, the
389	>	* next table must be (incrementally) prefilled with place-holders
390	>	* serving as reverse forwarders to the old table.  Because we are
391	>	* using power-of-two expansion, the elements from each bin must
392	>	* either stay at same index, or move with a power of two
393	>	* offset. We eliminate unnecessary node creation by catching
394	>	* cases where old nodes can be reused because their next fields
395	>	* won't change.  On average, only about one-sixth of them need
396	>	* cloning when a table doubles. The nodes they replace will be
397	>	* garbage collectable as soon as they are no longer referenced by
398	>	* any reader thread that may be in the midst of concurrently
399	>	* traversing table.  Upon transfer, the old table bin contains
400	>	* only a special forwarding node (with hash field "MOVED") that
401	>	* contains the next table as its key. On encountering a
402	>	* forwarding node, access and update operations restart, using
403	>	* the new table.
404	>	*
405	>	* Each bin transfer requires its bin lock, which can stall
406	>	* waiting for locks while resizing. However, because other
407	>	* threads can join in and help resize rather than contend for
408	>	* locks, average aggregate waits become shorter as resizing
409	>	* progresses.  The transfer operation must also ensure that all
410	>	* accessible bins in both the old and new table are usable by any
411	>	* traversal.  This is arranged by proceeding from the last bin
412	>	* (table.length - 1) up towards the first.  Upon seeing a
413	>	* forwarding node, traversals (see class Traverser) arrange to
414	>	* move to the new table without revisiting nodes.  However, to
415	>	* ensure that no intervening nodes are skipped, bin splitting can
416	>	* only begin after the associated reverse-forwarders are in
417	>	* place.
418		*
419		* The traversal scheme also applies to partial traversals of
420	<	* ranges of bins (via an alternate InternalIterator constructor)
421	<	* to support partitioned aggregate operations (that are not
422	<	* otherwise implemented yet).  Also, read-only operations give up
423	<	* if ever forwarded to a null table, which provides support for
424	<	* shutdown-style clearing, which is also not currently
272	<	* implemented.
420	>	* ranges of bins (via an alternate Traverser constructor)
421	>	* to support partitioned aggregate operations.  Also, read-only
422	>	* operations give up if ever forwarded to a null table, which
423	>	* provides support for shutdown-style clearing, which is also not
424	>	* currently implemented.
425		*
426		* Lazy table initialization minimizes footprint until first use,
427		* and also avoids resizings when the first operation is from a
#	Line 277 | Line 429 | public class ConcurrentHashMapV8<K, V>
429		* These cases attempt to override the initial capacity settings,
430		* but harmlessly fail to take effect in cases of races.
431		*
432	<	* The element count is maintained using a LongAdder, which avoids
433	<	* contention on updates but can encounter cache thrashing if read
434	<	* too frequently during concurrent access. To avoid reading so
435	<	* often, resizing is attempted either when a bin lock is
436	<	* contended, or upon adding to a bin already holding two or more
437	<	* nodes (checked before adding in the xIfAbsent methods, after
438	<	* adding in others). Under uniform hash distributions, the
439	<	* probability of this occurring at threshold is around 13%,
440	<	* meaning that only about 1 in 8 puts check threshold (and after
441	<	* resizing, many fewer do so). But this approximation has high
442	<	* variance for small table sizes, so we check on any collision
443	<	* for sizes <= 64. The bulk putAll operation further reduces
444	<	* contention by only committing count updates upon these size
445	<	* checks.
432	>	* The element count is maintained using a specialization of
433	>	* LongAdder. We need to incorporate a specialization rather than
434	>	* just use a LongAdder in order to access implicit
435	>	* contention-sensing that leads to creation of multiple
436	>	* CounterCells.  The counter mechanics avoid contention on
437	>	* updates but can encounter cache thrashing if read too
438	>	* frequently during concurrent access. To avoid reading so often,
439	>	* resizing under contention is attempted only upon adding to a
440	>	* bin already holding two or more nodes. Under uniform hash
441	>	* distributions, the probability of this occurring at threshold
442	>	* is around 13%, meaning that only about 1 in 8 puts check
443	>	* threshold (and after resizing, many fewer do so).
444	>	*
445	>	* TreeBins use a special form of comparison for search and
446	>	* related operations (which is the main reason we cannot use
447	>	* existing collections such as TreeMaps). TreeBins contain
448	>	* Comparable elements, but may contain others, as well as
449	>	* elements that are Comparable but not necessarily Comparable
450	>	* for the same T, so we cannot invoke compareTo among them. To
451	>	* handle this, the tree is ordered primarily by hash value, then
452	>	* by Comparable.compareTo order if applicable.  On lookup at a
453	>	* node, if elements are not comparable or compare as 0 then both
454	>	* left and right children may need to be searched in the case of
455	>	* tied hash values. (This corresponds to the full list search
456	>	* that would be necessary if all elements were non-Comparable and
457	>	* had tied hashes.)  The red-black balancing code is updated from
458	>	* pre-jdk-collections
459	>	* (http://gee.cs.oswego.edu/dl/classes/collections/RBCell.java)
460	>	* based in turn on Cormen, Leiserson, and Rivest "Introduction to
461	>	* Algorithms" (CLR).
462	>	*
463	>	* TreeBins also require an additional locking mechanism.  While
464	>	* list traversal is always possible by readers even during
465	>	* updates, tree traversal is not, mainly beause of tree-rotations
466	>	* that may change the root node and/or its linkages.  TreeBins
467	>	* include a simple read-write lock mechanism parasitic on the
468	>	* main bin-synchronization strategy: Structural adjustments
469	>	* associated with an insertion or removal are already bin-locked
470	>	* (and so cannot conflict with other writers) but must wait for
471	>	* ongoing readers to finish. Since there can be only one such
472	>	* waiter, we use a simple scheme using a single "waiter" field to
473	>	* block writers.  However, readers need never block.  If the root
474	>	* lock is held, they proceed along the slow traversal path (via
475	>	* next-pointers) until the lock becomes available or the list is
476	>	* exhausted, whichever comes first. These cases are not fast, but
477	>	* maximize aggregate expected throughput.
478		*
479		* Maintaining API and serialization compatibility with previous
480		* versions of this class introduces several oddities. Mainly: We
#	Line 300 | Line 484 | public class ConcurrentHashMapV8<K, V>
484		* time that we can guarantee to honor it.) We also declare an
485		* unused "Segment" class that is instantiated in minimal form
486		* only when serializing.
487	+	*
488	+	* This file is organized to make things a little easier to follow
489	+	* while reading than they might otherwise: First the main static
490	+	* declarations and utilities, then fields, then main public
491	+	* methods (with a few factorings of multiple public methods into
492	+	* internal ones), then sizing methods, trees, traversers, and
493	+	* bulk operations.
494		*/
495
496		/* ---------------- Constants -------------- */
#	Line 341 | Line 532 | public class ConcurrentHashMapV8<K, V>
532		private static final float LOAD_FACTOR = 0.75f;
533
534		/**
535	<	* The buffer size for skipped bins during transfers. The
536	<	* value is arbitrary but should be large enough to avoid
537	<	* most locking stalls during resizes.
538	<	*/
539	<	private static final int TRANSFER_BUFFER_SIZE = 32;
540	<
350	<	/*
351	<	* Encodings for special uses of Node hash fields. See above for
352	<	* explanation.
535	>	* The bin count threshold for using a tree rather than list for a
536	>	* bin.  Bins are converted to trees when adding an element to a
537	>	* bin with at least this many nodes. The value must be greater
538	>	* than 2, and should be at least 8 to mesh with assumptions in
539	>	* tree removal about conversion back to plain bins upon
540	>	* shrinkage.
541		*/
542	<	static final int MOVED     = 0x80000000; // hash field for fowarding nodes
355	<	static final int LOCKED    = 0x40000000; // set/tested only as a bit
356	<	static final int WAITING   = 0xc0000000; // both bits set/tested together
357	<	static final int HASH_BITS = 0x3fffffff; // usable bits of normal node hash
358	<
359	<	/* ---------------- Fields -------------- */
542	>	static final int TREEIFY_THRESHOLD = 8;
543
544		/**
545	<	* The array of bins. Lazily initialized upon first insertion.
546	<	* Size is always a power of two. Accessed directly by iterators.
545	>	* The bin count threshold for untreeifying a (split) bin during a
546	>	* resize operation. Should be less than TREEIFY_THRESHOLD, and at
547	>	* most 6 to mesh with shrinkage detection under removal.
548		*/
549	<	transient volatile Node[] table;
549	>	static final int UNTREEIFY_THRESHOLD = 6;
550
551		/**
552	<	* The counter maintaining number of elements.
552	>	* The smallest table capacity for which bins may be treeified.
553	>	* (Otherwise the table is resized if too many nodes in a bin.)
554	>	* The value should be at least 4 * TREEIFY_THRESHOLD to avoid
555	>	* conflicts between resizing and treeification thresholds.
556		*/
557	<	private transient final LongAdder counter;
557	>	static final int MIN_TREEIFY_CAPACITY = 64;
558
559		/**
560	<	* Table initialization and resizing control.  When negative, the
561	<	* table is being initialized or resized. Otherwise, when table is
562	<	* null, holds the initial table size to use upon creation, or 0
563	<	* for default. After initialization, holds the next element count
564	<	* value upon which to resize the table.
560	>	* Minimum number of rebinnings per transfer step. Ranges are
561	>	* subdivided to allow multiple resizer threads.  This value
562	>	* serves as a lower bound to avoid resizers encountering
563	>	* excessive memory contention.  The value should be at least
564	>	* DEFAULT_CAPACITY.
565		*/
566	<	private transient volatile int sizeCtl;
566	>	private static final int MIN_TRANSFER_STRIDE = 16;
567
568	<	// views
569	<	private transient KeySet<K,V> keySet;
570	<	private transient Values<K,V> values;
571	<	private transient EntrySet<K,V> entrySet;
572	<
573	<	/** For serialization compatibility. Null unless serialized; see below */
574	<	private Segment<K,V>[] segments;
568	>	/*
569	>	* Encodings for Node hash fields. See above for explanation.
570	>	*/
571	>	static final int MOVED     = 0x8fffffff; // (-1) hash for forwarding nodes
572	>	static final int TREEBIN   = 0x80000000; // hash for heads of treea
573	>	static final int RESERVED  = 0x80000001; // hash for transient reservations
574	>	static final int HASH_BITS = 0x7fffffff; // usable bits of normal node hash
575	>
576	>	/** Number of CPUS, to place bounds on some sizings */
577	>	static final int NCPU = Runtime.getRuntime().availableProcessors();
578	>
579	>	/** For serialization compatibility. */
580	>	private static final ObjectStreamField[] serialPersistentFields = {
581	>	new ObjectStreamField("segments", Segment[].class),
582	>	new ObjectStreamField("segmentMask", Integer.TYPE),
583	>	new ObjectStreamField("segmentShift", Integer.TYPE)
584	>	};
585
586		/* ---------------- Nodes -------------- */
587
588		/**
589	<	* Key-value entry. Note that this is never exported out as a
590	<	* user-visible Map.Entry (see WriteThroughEntry and SnapshotEntry
591	<	* below). Nodes with a hash field of MOVED are special, and do
592	<	* not contain user keys or values.  Otherwise, keys are never
593	<	* null, and null val fields indicate that a node is in the
594	<	* process of being deleted or created. For purposes of read-only
595	<	* access, a key may be read before a val, but can only be used
596	<	* after checking val to be non-null.
597	<	*/
598	<	static final class Node {
599	<	volatile int hash;
600	<	final Object key;
404	<	volatile Object val;
405	<	volatile Node next;
589	>	* Key-value entry.  This class is never exported out as a
590	>	* user-mutable Map.Entry (i.e., one supporting setValue; see
591	>	* MapEntry below), but can be used for read-only traversals used
592	>	* in bulk tasks.  Subclasses of Node with a negativehash field
593	>	* are special, and contain null keys and values (but are never
594	>	* exported).  Otherwise, keys and vals are never null.
595	>	*/
596	>	static class Node<K,V> implements Map.Entry<K,V> {
597	>	final int hash;
598	>	final K key;
599	>	volatile V val;
600	>	Node<K,V> next;
601
602	<	Node(int hash, Object key, Object val, Node next) {
602	>	Node(int hash, K key, V val, Node<K,V> next) {
603		this.hash = hash;
604		this.key = key;
605		this.val = val;
606		this.next = next;
607		}
608
609	<	/** CompareAndSet the hash field */
610	<	final boolean casHash(int cmp, int val) {
611	<	return UNSAFE.compareAndSwapInt(this, hashOffset, cmp, val);
609	>	public final K getKey()       { return key; }
610	>	public final V getValue()     { return val; }
611	>	public final int hashCode()   { return key.hashCode() ^ val.hashCode(); }
612	>	public final String toString(){ return key + "=" + val; }
613	>	public final V setValue(V value) {
614	>	throw new UnsupportedOperationException();
615		}
616
617	<	/** The number of spins before blocking for a lock */
618	<	static final int MAX_SPINS =
619	<	Runtime.getRuntime().availableProcessors() > 1 ? 64 : 1;
620	<
621	<	/**
622	<	* Spins a while if LOCKED bit set and this node is the first
623	<	* of its bin, and then sets WAITING bits on hash field and
426	<	* blocks (once) if they are still set.  It is OK for this
427	<	* method to return even if lock is not available upon exit,
428	<	* which enables these simple single-wait mechanics.
429	<	*
430	<	* The corresponding signalling operation is performed within
431	<	* callers: Upon detecting that WAITING has been set when
432	<	* unlocking lock (via a failed CAS from non-waiting LOCKED
433	<	* state), unlockers acquire the sync lock and perform a
434	<	* notifyAll.
435	<	*/
436	<	final void tryAwaitLock(Node[] tab, int i) {
437	<	if (tab != null && i >= 0 && i < tab.length) { // bounds check
438	<	int spins = MAX_SPINS, h;
439	<	while (tabAt(tab, i) == this && ((h = hash) & LOCKED) != 0) {
440	<	if (spins >= 0) {
441	<	if (--spins == MAX_SPINS >>> 1)
442	<	Thread.yield();  // heuristically yield mid-way
443	<	}
444	<	else if (casHash(h, h | WAITING)) {
445	<	synchronized (this) {
446	<	if (tabAt(tab, i) == this &&
447	<	(hash & WAITING) == WAITING) {
448	<	try {
449	<	wait();
450	<	} catch (InterruptedException ie) {
451	<	Thread.currentThread().interrupt();
452	<	}
453	<	}
454	<	else
455	<	notifyAll(); // possibly won race vs signaller
456	<	}
457	<	break;
458	<	}
459	<	}
460	<	}
617	>	public final boolean equals(Object o) {
618	>	Object k, v, u; Map.Entry<?,?> e;
619	>	return ((o instanceof Map.Entry) &&
620	>	(k = (e = (Map.Entry<?,?>)o).getKey()) != null &&
621	>	(v = e.getValue()) != null &&
622	>	(k == key || k.equals(key)) &&
623	>	(v == (u = val) || v.equals(u)));
624		}
625
626	<	// Unsafe mechanics for casHash
627	<	private static final sun.misc.Unsafe UNSAFE;
628	<	private static final long hashOffset;
629	<
630	<	static {
631	<	try {
632	<	UNSAFE = getUnsafe();
633	<	Class<?> k = Node.class;
634	<	hashOffset = UNSAFE.objectFieldOffset
635	<	(k.getDeclaredField("hash"));
636	<	} catch (Exception e) {
637	<	throw new Error(e);
638	<	}
626	>	/**
627	>	* Virtualized support for map.get(); overridden in subclasses.
628	>	*/
629	>	Node<K,V> find(int h, Object k) {
630	>	Node<K,V> e = this;
631	>	if (k != null) {
632	>	do {
633	>	K ek;
634	>	if (e.hash == h &&
635	>	((ek = e.key) == k || (ek != null && k.equals(ek))))
636	>	return e;
637	>	} while ((e = e.next) != null);
638	>	}
639	>	return null;
640		}
641		}
642
643	<	/* ---------------- Table element access -------------- */
644	<
645	<	/*
646	<	* Volatile access methods are used for table elements as well as
647	<	* elements of in-progress next table while resizing.  Uses are
648	<	* null checked by callers, and implicitly bounds-checked, relying
649	<	* on the invariants that tab arrays have non-zero size, and all
650	<	* indices are masked with (tab.length - 1) which is never
651	<	* negative and always less than length. Note that, to be correct
652	<	* wrt arbitrary concurrency errors by users, bounds checks must
653	<	* operate on local variables, which accounts for some odd-looking
654	<	* inline assignments below.
643	>	/* ---------------- Static utilities -------------- */
644	>
645	>	/**
646	>	* Spreads (XORs) higher bits of hash to lower and also forces top
647	>	* bit to 0. Because the table uses power-of-two masking, sets of
648	>	* hashes that vary only in bits above the current mask will
649	>	* always collide. (Among known examples are sets of Float keys
650	>	* holding consecutive whole numbers in small tables.)  So we
651	>	* apply a transform that spreads the impact of higher bits
652	>	* downward. There is a tradeoff between speed, utility, and
653	>	* quality of bit-spreading. Because many common sets of hashes
654	>	* are already reasonably distributed (so don't benefit from
655	>	* spreading), and because we use trees to handle large sets of
656	>	* collisions in bins, we just XOR some shifted bits in the
657	>	* cheapest possible way to reduce systematic lossage, as well as
658	>	* to incorporate impact of the highest bits that would otherwise
659	>	* never be used in index calculations because of table bounds.
660		*/
661	<
662	<	static final Node tabAt(Node[] tab, int i) { // used by InternalIterator
494	<	return (Node)UNSAFE.getObjectVolatile(tab, ((long)i<<ASHIFT)+ABASE);
495	<	}
496	<
497	<	private static final boolean casTabAt(Node[] tab, int i, Node c, Node v) {
498	<	return UNSAFE.compareAndSwapObject(tab, ((long)i<<ASHIFT)+ABASE, c, v);
499	<	}
500	<
501	<	private static final void setTabAt(Node[] tab, int i, Node v) {
502	<	UNSAFE.putObjectVolatile(tab, ((long)i<<ASHIFT)+ABASE, v);
661	>	static final int spread(int h) {
662	>	return (h ^ (h >>> 16)) & HASH_BITS;
663		}
664
505	–	/* ---------------- Internal access and update methods -------------- */
506	–
665		/**
666	<	* Applies a supplemental hash function to a given hashCode, which
667	<	* defends against poor quality hash functions.  The result must
510	<	* be have the top 2 bits clear. For reasonable performance, this
511	<	* function must have good avalanche properties; i.e., that each
512	<	* bit of the argument affects each bit of the result. (Although
513	<	* we don't care about the unused top 2 bits.)
666	>	* Returns a power of two table size for the given desired capacity.
667	>	* See Hackers Delight, sec 3.2
668		*/
669	<	private static final int spread(int h) {
670	<	// Apply base step of MurmurHash; see http://code.google.com/p/smhasher/
671	<	// Despite two multiplies, this is often faster than others
672	<	// with comparable bit-spread properties.
673	<	h ^= h >>> 16;
674	<	h *= 0x85ebca6b;
675	<	h ^= h >>> 13;
676	<	h *= 0xc2b2ae35;
523	<	return ((h >>> 16) ^ h) & HASH_BITS; // mask out top bits
669	>	private static final int tableSizeFor(int c) {
670	>	int n = c - 1;
671	>	n |= n >>> 1;
672	>	n |= n >>> 2;
673	>	n |= n >>> 4;
674	>	n |= n >>> 8;
675	>	n |= n >>> 16;
676	>	return (n < 0) ? 1 : (n >= MAXIMUM_CAPACITY) ? MAXIMUM_CAPACITY : n + 1;
677		}
678
679	<	/** Implementation for get and containsKey */
680	<	private final Object internalGet(Object k) {
681	<	int h = spread(k.hashCode());
682	<	retry: for (Node[] tab = table; tab != null;) {
683	<	Node e; Object ek, ev; int eh;    // locals to read fields once
684	<	for (e = tabAt(tab, (tab.length - 1) & h); e != null; e = e.next) {
685	<	if ((eh = e.hash) == MOVED) {
686	<	tab = (Node[])e.key;      // restart with new table
687	<	continue retry;
679	>	/**
680	>	* Returns x's Class if it is of the form "class C implements
681	>	* Comparable<C>", else null.
682	>	*/
683	>	static Class<?> comparableClassFor(Object x) {
684	>	if (x instanceof Comparable) {
685	>	Class<?> c; Type[] ts, as; Type t; ParameterizedType p;
686	>	if ((c = x.getClass()) == String.class) // bypass checks
687	>	return c;
688	>	if ((ts = c.getGenericInterfaces()) != null) {
689	>	for (int i = 0; i < ts.length; ++i) {
690	>	if (((t = ts[i]) instanceof ParameterizedType) &&
691	>	((p = (ParameterizedType)t).getRawType() ==
692	>	Comparable.class) &&
693	>	(as = p.getActualTypeArguments()) != null &&
694	>	as.length == 1 && as[0] == c) // type arg is c
695	>	return c;
696		}
536	–	if ((eh & HASH_BITS) == h && (ev = e.val) != null &&
537	–	((ek = e.key) == k || k.equals(ek)))
538	–	return ev;
697		}
540	–	break;
698		}
699		return null;
700		}
701
702		/**
703	<	* Implementation for the four public remove/replace methods:
704	<	* Replaces node value with v, conditional upon match of cv if
548	<	* non-null.  If resulting value is null, delete.
703	>	* Returns k.compareTo(x) if x matches kc (k's screened comparable
704	>	* class), else 0.
705		*/
706	<	private final Object internalReplace(Object k, Object v, Object cv) {
707	<	int h = spread(k.hashCode());
708	<	Object oldVal = null;
709	<	for (Node[] tab = table;;) {
554	<	Node f; int i, fh;
555	<	if (tab == null ||
556	<	(f = tabAt(tab, i = (tab.length - 1) & h)) == null)
557	<	break;
558	<	else if ((fh = f.hash) == MOVED)
559	<	tab = (Node[])f.key;
560	<	else if ((fh & HASH_BITS) != h && f.next == null) // precheck
561	<	break;                          // rules out possible existence
562	<	else if ((fh & LOCKED) != 0) {
563	<	checkForResize();               // try resizing if can't get lock
564	<	f.tryAwaitLock(tab, i);
565	<	}
566	<	else if (f.casHash(fh, fh | LOCKED)) {
567	<	boolean validated = false;
568	<	boolean deleted = false;
569	<	try {
570	<	if (tabAt(tab, i) == f) {
571	<	validated = true;
572	<	for (Node e = f, pred = null;;) {
573	<	Object ek, ev;
574	<	if ((e.hash & HASH_BITS) == h &&
575	<	((ev = e.val) != null) &&
576	<	((ek = e.key) == k || k.equals(ek))) {
577	<	if (cv == null || cv == ev || cv.equals(ev)) {
578	<	oldVal = ev;
579	<	if ((e.val = v) == null) {
580	<	deleted = true;
581	<	Node en = e.next;
582	<	if (pred != null)
583	<	pred.next = en;
584	<	else
585	<	setTabAt(tab, i, en);
586	<	}
587	<	}
588	<	break;
589	<	}
590	<	pred = e;
591	<	if ((e = e.next) == null)
592	<	break;
593	<	}
594	<	}
595	<	} finally {
596	<	if (!f.casHash(fh | LOCKED, fh)) {
597	<	f.hash = fh;
598	<	synchronized(f) { f.notifyAll(); };
599	<	}
600	<	}
601	<	if (validated) {
602	<	if (deleted)
603	<	counter.add(-1L);
604	<	break;
605	<	}
606	<	}
607	<	}
608	<	return oldVal;
706	>	@SuppressWarnings({"rawtypes","unchecked"}) // for cast to Comparable
707	>	static int compareComparables(Class<?> kc, Object k, Object x) {
708	>	return (x == null || x.getClass() != kc ? 0 :
709	>	((Comparable)k).compareTo(x));
710		}
711
712	<	/*
612	<	* Internal versions of the five insertion methods, each a
613	<	* little more complicated than the last. All have
614	<	* the same basic structure as the first (internalPut):
615	<	*  1. If table uninitialized, create
616	<	*  2. If bin empty, try to CAS new node
617	<	*  3. If bin stale, use new table
618	<	*  4. Lock and validate; if valid, scan and add or update
619	<	*
620	<	* The others interweave other checks and/or alternative actions:
621	<	*  * Plain put checks for and performs resize after insertion.
622	<	*  * putIfAbsent prescans for mapping without lock (and fails to add
623	<	*    if present), which also makes pre-emptive resize checks worthwhile.
624	<	*  * computeIfAbsent extends form used in putIfAbsent with additional
625	<	*    mechanics to deal with, calls, potential exceptions and null
626	<	*    returns from function call.
627	<	*  * compute uses the same function-call mechanics, but without
628	<	*    the prescans
629	<	*  * putAll attempts to pre-allocate enough table space
630	<	*    and more lazily performs count updates and checks.
631	<	*
632	<	* Someday when details settle down a bit more, it might be worth
633	<	* some factoring to reduce sprawl.
634	<	*/
635	<
636	<	/** Implementation for put */
637	<	private final Object internalPut(Object k, Object v) {
638	<	int h = spread(k.hashCode());
639	<	boolean checkSize = false;
640	<	for (Node[] tab = table;;) {
641	<	int i; Node f; int fh;
642	<	if (tab == null)
643	<	tab = initTable();
644	<	else if ((f = tabAt(tab, i = (tab.length - 1) & h)) == null) {
645	<	if (casTabAt(tab, i, null, new Node(h, k, v, null)))
646	<	break;                   // no lock when adding to empty bin
647	<	}
648	<	else if ((fh = f.hash) == MOVED)
649	<	tab = (Node[])f.key;
650	<	else if ((fh & LOCKED) != 0) {
651	<	checkForResize();
652	<	f.tryAwaitLock(tab, i);
653	<	}
654	<	else if (f.casHash(fh, fh | LOCKED)) {
655	<	Object oldVal = null;
656	<	boolean validated = false;
657	<	try {                        // needed in case equals() throws
658	<	if (tabAt(tab, i) == f) {
659	<	validated = true;    // retry if 1st already deleted
660	<	for (Node e = f;;) {
661	<	Object ek, ev;
662	<	if ((e.hash & HASH_BITS) == h &&
663	<	(ev = e.val) != null &&
664	<	((ek = e.key) == k || k.equals(ek))) {
665	<	oldVal = ev;
666	<	e.val = v;
667	<	break;
668	<	}
669	<	Node last = e;
670	<	if ((e = e.next) == null) {
671	<	last.next = new Node(h, k, v, null);
672	<	if (last != f || tab.length <= 64)
673	<	checkSize = true;
674	<	break;
675	<	}
676	<	}
677	<	}
678	<	} finally {                  // unlock and signal if needed
679	<	if (!f.casHash(fh | LOCKED, fh)) {
680	<	f.hash = fh;
681	<	synchronized (f) { f.notifyAll(); };
682	<	}
683	<	}
684	<	if (validated) {
685	<	if (oldVal != null)
686	<	return oldVal;
687	<	break;
688	<	}
689	<	}
690	<	}
691	<	counter.add(1L);
692	<	if (checkSize)
693	<	checkForResize();
694	<	return null;
695	<	}
712	>	/* ---------------- Table element access -------------- */
713
714	<	/** Implementation for putIfAbsent */
715	<	private final Object internalPutIfAbsent(Object k, Object v) {
716	<	int h = spread(k.hashCode());
717	<	for (Node[] tab = table;;) {
718	<	int i; Node f; int fh; Object fk, fv;
719	<	if (tab == null)
720	<	tab = initTable();
721	<	else if ((f = tabAt(tab, i = (tab.length - 1) & h)) == null) {
722	<	if (casTabAt(tab, i, null, new Node(h, k, v, null)))
723	<	break;
724	<	}
725	<	else if ((fh = f.hash) == MOVED)
726	<	tab = (Node[])f.key;
727	<	else if ((fh & HASH_BITS) == h && (fv = f.val) != null &&
711	<	((fk = f.key) == k || k.equals(fk)))
712	<	return fv;
713	<	else {
714	<	Node g = f.next;
715	<	if (g != null) { // at least 2 nodes -- search and maybe resize
716	<	for (Node e = g;;) {
717	<	Object ek, ev;
718	<	if ((e.hash & HASH_BITS) == h && (ev = e.val) != null &&
719	<	((ek = e.key) == k || k.equals(ek)))
720	<	return ev;
721	<	if ((e = e.next) == null) {
722	<	checkForResize();
723	<	break;
724	<	}
725	<	}
726	<	}
727	<	if (((fh = f.hash) & LOCKED) != 0) {
728	<	checkForResize();
729	<	f.tryAwaitLock(tab, i);
730	<	}
731	<	else if (tabAt(tab, i) == f && f.casHash(fh, fh | LOCKED)) {
732	<	Object oldVal = null;
733	<	boolean validated = false;
734	<	try {
735	<	if (tabAt(tab, i) == f) {
736	<	validated = true;
737	<	for (Node e = f;;) {
738	<	Object ek, ev;
739	<	if ((e.hash & HASH_BITS) == h &&
740	<	(ev = e.val) != null &&
741	<	((ek = e.key) == k || k.equals(ek))) {
742	<	oldVal = ev;
743	<	break;
744	<	}
745	<	Node last = e;
746	<	if ((e = e.next) == null) {
747	<	last.next = new Node(h, k, v, null);
748	<	break;
749	<	}
750	<	}
751	<	}
752	<	} finally {
753	<	if (!f.casHash(fh | LOCKED, fh)) {
754	<	f.hash = fh;
755	<	synchronized(f) { f.notifyAll(); };
756	<	}
757	<	}
758	<	if (validated) {
759	<	if (oldVal != null)
760	<	return oldVal;
761	<	break;
762	<	}
763	<	}
764	<	}
765	<	}
766	<	counter.add(1L);
767	<	return null;
768	<	}
714	>	/*
715	>	* Volatile access methods are used for table elements as well as
716	>	* elements of in-progress next table while resizing.  All uses of
717	>	* the tab arguments must be null checked by callers.  All callers
718	>	* also paranoically precheck that tab's length is not zero (or an
719	>	* equivalent check), thus ensuring that any index argument taking
720	>	* the form of a hash value anded with (length - 1) is a valid
721	>	* index.  Note that, to be correct wrt arbitrary concurrency
722	>	* errors by users, these checks must operate on local variables,
723	>	* which accounts for some odd-looking inline assignments below.
724	>	* Note that calls to setTabAt always occur within locked regions,
725	>	* and so do not need full volatile semantics, but still require
726	>	* ordering to maintain concurrent readability.
727	>	*/
728
729	<	/** Implementation for computeIfAbsent */
730	<	private final Object internalComputeIfAbsent(K k,
731	<	MappingFunction<? super K, ?> mf) {
773	<	int h = spread(k.hashCode());
774	<	Object val = null;
775	<	for (Node[] tab = table;;) {
776	<	Node f; int i, fh; Object fk, fv;
777	<	if (tab == null)
778	<	tab = initTable();
779	<	else if ((f = tabAt(tab, i = (tab.length - 1) & h)) == null) {
780	<	Node node = new Node(fh = h | LOCKED, k, null, null);
781	<	boolean validated = false;
782	<	if (casTabAt(tab, i, null, node)) {
783	<	validated = true;
784	<	try {
785	<	if ((val = mf.map(k)) != null)
786	<	node.val = val;
787	<	} finally {
788	<	if (val == null)
789	<	setTabAt(tab, i, null);
790	<	if (!node.casHash(fh, h)) {
791	<	node.hash = h;
792	<	synchronized(node) { node.notifyAll(); };
793	<	}
794	<	}
795	<	}
796	<	if (validated)
797	<	break;
798	<	}
799	<	else if ((fh = f.hash) == MOVED)
800	<	tab = (Node[])f.key;
801	<	else if ((fh & HASH_BITS) == h && (fv = f.val) != null &&
802	<	((fk = f.key) == k || k.equals(fk)))
803	<	return fv;
804	<	else {
805	<	Node g = f.next;
806	<	if (g != null) {
807	<	for (Node e = g;;) {
808	<	Object ek, ev;
809	<	if ((e.hash & HASH_BITS) == h && (ev = e.val) != null &&
810	<	((ek = e.key) == k || k.equals(ek)))
811	<	return ev;
812	<	if ((e = e.next) == null) {
813	<	checkForResize();
814	<	break;
815	<	}
816	<	}
817	<	}
818	<	if (((fh = f.hash) & LOCKED) != 0) {
819	<	checkForResize();
820	<	f.tryAwaitLock(tab, i);
821	<	}
822	<	else if (tabAt(tab, i) == f && f.casHash(fh, fh | LOCKED)) {
823	<	boolean validated = false;
824	<	try {
825	<	if (tabAt(tab, i) == f) {
826	<	validated = true;
827	<	for (Node e = f;;) {
828	<	Object ek, ev;
829	<	if ((e.hash & HASH_BITS) == h &&
830	<	(ev = e.val) != null &&
831	<	((ek = e.key) == k || k.equals(ek))) {
832	<	val = ev;
833	<	break;
834	<	}
835	<	Node last = e;
836	<	if ((e = e.next) == null) {
837	<	if ((val = mf.map(k)) != null)
838	<	last.next = new Node(h, k, val, null);
839	<	break;
840	<	}
841	<	}
842	<	}
843	<	} finally {
844	<	if (!f.casHash(fh | LOCKED, fh)) {
845	<	f.hash = fh;
846	<	synchronized(f) { f.notifyAll(); };
847	<	}
848	<	}
849	<	if (validated)
850	<	break;
851	<	}
852	<	}
853	<	}
854	<	if (val == null)
855	<	throw new NullPointerException();
856	<	counter.add(1L);
857	<	return val;
729	>	@SuppressWarnings("unchecked")
730	>	static final <K,V> Node<K,V> tabAt(Node<K,V>[] tab, int i) {
731	>	return (Node<K,V>)U.getObjectVolatile(tab, ((long)i << ASHIFT) + ABASE);
732		}
733
734	<	/** Implementation for compute */
735	<	@SuppressWarnings("unchecked")
736	<	private final Object internalCompute(K k,
863	<	RemappingFunction<? super K, V> mf) {
864	<	int h = spread(k.hashCode());
865	<	Object val = null;
866	<	boolean added = false;
867	<	boolean checkSize = false;
868	<	for (Node[] tab = table;;) {
869	<	Node f; int i, fh;
870	<	if (tab == null)
871	<	tab = initTable();
872	<	else if ((f = tabAt(tab, i = (tab.length - 1) & h)) == null) {
873	<	Node node = new Node(fh = h | LOCKED, k, null, null);
874	<	boolean validated = false;
875	<	if (casTabAt(tab, i, null, node)) {
876	<	validated = true;
877	<	try {
878	<	if ((val = mf.remap(k, null)) != null) {
879	<	node.val = val;
880	<	added = true;
881	<	}
882	<	} finally {
883	<	if (!added)
884	<	setTabAt(tab, i, null);
885	<	if (!node.casHash(fh, h)) {
886	<	node.hash = h;
887	<	synchronized (node) { node.notifyAll(); };
888	<	}
889	<	}
890	<	}
891	<	if (validated)
892	<	break;
893	<	}
894	<	else if ((fh = f.hash) == MOVED)
895	<	tab = (Node[])f.key;
896	<	else if ((fh & LOCKED) != 0) {
897	<	checkForResize();
898	<	f.tryAwaitLock(tab, i);
899	<	}
900	<	else if (f.casHash(fh, fh | LOCKED)) {
901	<	boolean validated = false;
902	<	try {
903	<	if (tabAt(tab, i) == f) {
904	<	validated = true;
905	<	for (Node e = f;;) {
906	<	Object ek, ev;
907	<	if ((e.hash & HASH_BITS) == h &&
908	<	(ev = e.val) != null &&
909	<	((ek = e.key) == k || k.equals(ek))) {
910	<	val = mf.remap(k, (V)ev);
911	<	if (val != null)
912	<	e.val = val;
913	<	break;
914	<	}
915	<	Node last = e;
916	<	if ((e = e.next) == null) {
917	<	if ((val = mf.remap(k, null)) != null) {
918	<	last.next = new Node(h, k, val, null);
919	<	added = true;
920	<	if (last != f || tab.length <= 64)
921	<	checkSize = true;
922	<	}
923	<	break;
924	<	}
925	<	}
926	<	}
927	<	} finally {
928	<	if (!f.casHash(fh | LOCKED, fh)) {
929	<	f.hash = fh;
930	<	synchronized (f) { f.notifyAll(); };
931	<	}
932	<	}
933	<	if (validated)
934	<	break;
935	<	}
936	<	}
937	<	if (val == null)
938	<	throw new NullPointerException();
939	<	if (added) {
940	<	counter.add(1L);
941	<	if (checkSize)
942	<	checkForResize();
943	<	}
944	<	return val;
734	>	static final <K,V> boolean casTabAt(Node<K,V>[] tab, int i,
735	>	Node<K,V> c, Node<K,V> v) {
736	>	return U.compareAndSwapObject(tab, ((long)i << ASHIFT) + ABASE, c, v);
737		}
738
739	<	/** Implementation for putAll */
740	<	private final void internalPutAll(Map<?, ?> m) {
949	<	tryPresize(m.size());
950	<	long delta = 0L;     // number of uncommitted additions
951	<	boolean npe = false; // to throw exception on exit for nulls
952	<	try {                // to clean up counts on other exceptions
953	<	for (Map.Entry<?, ?> entry : m.entrySet()) {
954	<	Object k, v;
955	<	if (entry == null || (k = entry.getKey()) == null ||
956	<	(v = entry.getValue()) == null) {
957	<	npe = true;
958	<	break;
959	<	}
960	<	int h = spread(k.hashCode());
961	<	for (Node[] tab = table;;) {
962	<	int i; Node f; int fh;
963	<	if (tab == null)
964	<	tab = initTable();
965	<	else if ((f = tabAt(tab, i = (tab.length - 1) & h)) == null){
966	<	if (casTabAt(tab, i, null, new Node(h, k, v, null))) {
967	<	++delta;
968	<	break;
969	<	}
970	<	}
971	<	else if ((fh = f.hash) == MOVED)
972	<	tab = (Node[])f.key;
973	<	else if ((fh & LOCKED) != 0) {
974	<	counter.add(delta);
975	<	delta = 0L;
976	<	checkForResize();
977	<	f.tryAwaitLock(tab, i);
978	<	}
979	<	else if (f.casHash(fh, fh | LOCKED)) {
980	<	boolean validated = false;
981	<	boolean tooLong = false;
982	<	try {
983	<	if (tabAt(tab, i) == f) {
984	<	validated = true;
985	<	for (Node e = f;;) {
986	<	Object ek, ev;
987	<	if ((e.hash & HASH_BITS) == h &&
988	<	(ev = e.val) != null &&
989	<	((ek = e.key) == k || k.equals(ek))) {
990	<	e.val = v;
991	<	break;
992	<	}
993	<	Node last = e;
994	<	if ((e = e.next) == null) {
995	<	++delta;
996	<	last.next = new Node(h, k, v, null);
997	<	break;
998	<	}
999	<	tooLong = true;
1000	<	}
1001	<	}
1002	<	} finally {
1003	<	if (!f.casHash(fh | LOCKED, fh)) {
1004	<	f.hash = fh;
1005	<	synchronized(f) { f.notifyAll(); };
1006	<	}
1007	<	}
1008	<	if (validated) {
1009	<	if (tooLong) {
1010	<	counter.add(delta);
1011	<	delta = 0L;
1012	<	checkForResize();
1013	<	}
1014	<	break;
1015	<	}
1016	<	}
1017	<	}
1018	<	}
1019	<	} finally {
1020	<	if (delta != 0)
1021	<	counter.add(delta);
1022	<	}
1023	<	if (npe)
1024	<	throw new NullPointerException();
739	>	static final <K,V> void setTabAt(Node<K,V>[] tab, int i, Node<K,V> v) {
740	>	U.putOrderedObject(tab, ((long)i << ASHIFT) + ABASE, v);
741		}
742
743	<	/* ---------------- Table Initialization and Resizing -------------- */
743	>	/* ---------------- Fields -------------- */
744
745		/**
746	<	* Returns a power of two table size for the given desired capacity.
747	<	* See Hackers Delight, sec 3.2
746	>	* The array of bins. Lazily initialized upon first insertion.
747	>	* Size is always a power of two. Accessed directly by iterators.
748		*/
749	<	private static final int tableSizeFor(int c) {
1034	<	int n = c - 1;
1035	<	n |= n >>> 1;
1036	<	n |= n >>> 2;
1037	<	n |= n >>> 4;
1038	<	n |= n >>> 8;
1039	<	n |= n >>> 16;
1040	<	return (n < 0) ? 1 : (n >= MAXIMUM_CAPACITY) ? MAXIMUM_CAPACITY : n + 1;
1041	<	}
749	>	transient volatile Node<K,V>[] table;
750
751		/**
752	<	* Initializes table, using the size recorded in sizeCtl.
752	>	* The next table to use; non-null only while resizing.
753		*/
754	<	private final Node[] initTable() {
1047	<	Node[] tab; int sc;
1048	<	while ((tab = table) == null) {
1049	<	if ((sc = sizeCtl) < 0)
1050	<	Thread.yield(); // lost initialization race; just spin
1051	<	else if (UNSAFE.compareAndSwapInt(this, sizeCtlOffset, sc, -1)) {
1052	<	try {
1053	<	if ((tab = table) == null) {
1054	<	int n = (sc > 0) ? sc : DEFAULT_CAPACITY;
1055	<	tab = table = new Node[n];
1056	<	sc = n - (n >>> 2);
1057	<	}
1058	<	} finally {
1059	<	sizeCtl = sc;
1060	<	}
1061	<	break;
1062	<	}
1063	<	}
1064	<	return tab;
1065	<	}
754	>	private transient volatile Node<K,V>[] nextTable;
755
756		/**
757	<	* If table is too small and not already resizing, creates next
758	<	* table and transfers bins.  Rechecks occupancy after a transfer
759	<	* to see if another resize is already needed because resizings
760	<	* are lagging additions.
761	<	*/
1073	<	private final void checkForResize() {
1074	<	Node[] tab; int n, sc;
1075	<	while ((tab = table) != null &&
1076	<	(n = tab.length) < MAXIMUM_CAPACITY &&
1077	<	(sc = sizeCtl) >= 0 && counter.sum() >= (long)sc &&
1078	<	UNSAFE.compareAndSwapInt(this, sizeCtlOffset, sc, -1)) {
1079	<	try {
1080	<	if (tab == table) {
1081	<	table = rebuild(tab);
1082	<	sc = (n << 1) - (n >>> 1);
1083	<	}
1084	<	} finally {
1085	<	sizeCtl = sc;
1086	<	}
1087	<	}
1088	<	}
757	>	* Base counter value, used mainly when there is no contention,
758	>	* but also as a fallback during table initialization
759	>	* races. Updated via CAS.
760	>	*/
761	>	private transient volatile long baseCount;
762
763		/**
764	<	* Tries to presize table to accommodate the given number of elements.
765	<	*
766	<	* @param size number of elements (doesn't need to be perfectly accurate)
764	>	* Table initialization and resizing control.  When negative, the
765	>	* table is being initialized or resized: -1 for initialization,
766	>	* else -(1 + the number of active resizing threads).  Otherwise,
767	>	* when table is null, holds the initial table size to use upon
768	>	* creation, or 0 for default. After initialization, holds the
769	>	* next element count value upon which to resize the table.
770		*/
771	<	private final void tryPresize(int size) {
1096	<	int c = (size >= (MAXIMUM_CAPACITY >>> 1)) ? MAXIMUM_CAPACITY :
1097	<	tableSizeFor(size + (size >>> 1) + 1);
1098	<	int sc;
1099	<	while ((sc = sizeCtl) >= 0) {
1100	<	Node[] tab = table; int n;
1101	<	if (tab == null || (n = tab.length) == 0) {
1102	<	n = (sc > c)? sc : c;
1103	<	if (UNSAFE.compareAndSwapInt(this, sizeCtlOffset, sc, -1)) {
1104	<	try {
1105	<	if (table == tab) {
1106	<	table = new Node[n];
1107	<	sc = n - (n >>> 2);
1108	<	}
1109	<	} finally {
1110	<	sizeCtl = sc;
1111	<	}
1112	<	}
1113	<	}
1114	<	else if (c <= sc || n >= MAXIMUM_CAPACITY)
1115	<	break;
1116	<	else if (UNSAFE.compareAndSwapInt(this, sizeCtlOffset, sc, -1)) {
1117	<	try {
1118	<	if (table == tab) {
1119	<	table = rebuild(tab);
1120	<	sc = (n << 1) - (n >>> 1);
1121	<	}
1122	<	} finally {
1123	<	sizeCtl = sc;
1124	<	}
1125	<	}
1126	<	}
1127	<	}
771	>	private transient volatile int sizeCtl;
772
773	<	/*
774	<	* Moves and/or copies the nodes in each bin to new table. See
1131	<	* above for explanation.
1132	<	*
1133	<	* @return the new table
773	>	/**
774	>	* The next table index (plus one) to split while resizing.
775		*/
776	<	private static final Node[] rebuild(Node[] tab) {
1136	<	int n = tab.length;
1137	<	Node[] nextTab = new Node[n << 1];
1138	<	Node fwd = new Node(MOVED, nextTab, null, null);
1139	<	int[] buffer = null;       // holds bins to revisit; null until needed
1140	<	Node rev = null;           // reverse forwarder; null until needed
1141	<	int nbuffered = 0;         // the number of bins in buffer list
1142	<	int bufferIndex = 0;       // buffer index of current buffered bin
1143	<	int bin = n - 1;           // current non-buffered bin or -1 if none
1144	<
1145	<	for (int i = bin;;) {      // start upwards sweep
1146	<	int fh; Node f;
1147	<	if ((f = tabAt(tab, i)) == null) {
1148	<	if (bin >= 0) {    // no lock needed (or available)
1149	<	if (!casTabAt(tab, i, f, fwd))
1150	<	continue;
1151	<	}
1152	<	else {             // transiently use a locked forwarding node
1153	<	Node g =  new Node(MOVED|LOCKED, nextTab, null, null);
1154	<	if (!casTabAt(tab, i, f, g))
1155	<	continue;
1156	<	setTabAt(nextTab, i, null);
1157	<	setTabAt(nextTab, i + n, null);
1158	<	setTabAt(tab, i, fwd);
1159	<	if (!g.casHash(MOVED|LOCKED, MOVED)) {
1160	<	g.hash = MOVED;
1161	<	synchronized (g) { g.notifyAll(); }
1162	<	}
1163	<	}
1164	<	}
1165	<	else if (((fh = f.hash) & LOCKED) == 0 && f.casHash(fh, fh|LOCKED)) {
1166	<	boolean validated = false;
1167	<	try {              // split to lo and hi lists; copying as needed
1168	<	if (tabAt(tab, i) == f) {
1169	<	validated = true;
1170	<	Node e = f, lastRun = f;
1171	<	Node lo = null, hi = null;
1172	<	int runBit = e.hash & n;
1173	<	for (Node p = e.next; p != null; p = p.next) {
1174	<	int b = p.hash & n;
1175	<	if (b != runBit) {
1176	<	runBit = b;
1177	<	lastRun = p;
1178	<	}
1179	<	}
1180	<	if (runBit == 0)
1181	<	lo = lastRun;
1182	<	else
1183	<	hi = lastRun;
1184	<	for (Node p = e; p != lastRun; p = p.next) {
1185	<	int ph = p.hash & HASH_BITS;
1186	<	Object pk = p.key, pv = p.val;
1187	<	if ((ph & n) == 0)
1188	<	lo = new Node(ph, pk, pv, lo);
1189	<	else
1190	<	hi = new Node(ph, pk, pv, hi);
1191	<	}
1192	<	setTabAt(nextTab, i, lo);
1193	<	setTabAt(nextTab, i + n, hi);
1194	<	setTabAt(tab, i, fwd);
1195	<	}
1196	<	} finally {
1197	<	if (!f.casHash(fh | LOCKED, fh)) {
1198	<	f.hash = fh;
1199	<	synchronized (f) { f.notifyAll(); };
1200	<	}
1201	<	}
1202	<	if (!validated)
1203	<	continue;
1204	<	}
1205	<	else {
1206	<	if (buffer == null) // initialize buffer for revisits
1207	<	buffer = new int[TRANSFER_BUFFER_SIZE];
1208	<	if (bin < 0 && bufferIndex > 0) {
1209	<	int j = buffer[--bufferIndex];
1210	<	buffer[bufferIndex] = i;
1211	<	i = j;         // swap with another bin
1212	<	continue;
1213	<	}
1214	<	if (bin < 0 || nbuffered >= TRANSFER_BUFFER_SIZE) {
1215	<	f.tryAwaitLock(tab, i);
1216	<	continue;      // no other options -- block
1217	<	}
1218	<	if (rev == null)   // initialize reverse-forwarder
1219	<	rev = new Node(MOVED, tab, null, null);
1220	<	if (tabAt(tab, i) != f || (f.hash & LOCKED) == 0)
1221	<	continue;      // recheck before adding to list
1222	<	buffer[nbuffered++] = i;
1223	<	setTabAt(nextTab, i, rev);     // install place-holders
1224	<	setTabAt(nextTab, i + n, rev);
1225	<	}
1226	<
1227	<	if (bin > 0)
1228	<	i = --bin;
1229	<	else if (buffer != null && nbuffered > 0) {
1230	<	bin = -1;
1231	<	i = buffer[bufferIndex = --nbuffered];
1232	<	}
1233	<	else
1234	<	return nextTab;
1235	<	}
1236	<	}
776	>	private transient volatile int transferIndex;
777
778		/**
779	<	* Implementation for clear. Steps through each bin, removing all
1240	<	* nodes.
779	>	* The least available table index to split while resizing.
780		*/
781	<	private final void internalClear() {
1243	<	long delta = 0L; // negative number of deletions
1244	<	int i = 0;
1245	<	Node[] tab = table;
1246	<	while (tab != null && i < tab.length) {
1247	<	int fh;
1248	<	Node f = tabAt(tab, i);
1249	<	if (f == null)
1250	<	++i;
1251	<	else if ((fh = f.hash) == MOVED)
1252	<	tab = (Node[])f.key;
1253	<	else if ((fh & LOCKED) != 0) {
1254	<	counter.add(delta); // opportunistically update count
1255	<	delta = 0L;
1256	<	f.tryAwaitLock(tab, i);
1257	<	}
1258	<	else if (f.casHash(fh, fh | LOCKED)) {
1259	<	boolean validated = false;
1260	<	try {
1261	<	if (tabAt(tab, i) == f) {
1262	<	validated = true;
1263	<	for (Node e = f; e != null; e = e.next) {
1264	<	if (e.val != null) { // currently always true
1265	<	e.val = null;
1266	<	--delta;
1267	<	}
1268	<	}
1269	<	setTabAt(tab, i, null);
1270	<	}
1271	<	} finally {
1272	<	if (!f.casHash(fh | LOCKED, fh)) {
1273	<	f.hash = fh;
1274	<	synchronized(f) { f.notifyAll(); };
1275	<	}
1276	<	}
1277	<	if (validated)
1278	<	++i;
1279	<	}
1280	<	}
1281	<	if (delta != 0)
1282	<	counter.add(delta);
1283	<	}
1284	<
1285	<
1286	<	/* ----------------Table Traversal -------------- */
781	>	private transient volatile int transferOrigin;
782
783		/**
784	<	* Encapsulates traversal for methods such as containsValue; also
1290	<	* serves as a base class for other iterators.
1291	<	*
1292	<	* At each step, the iterator snapshots the key ("nextKey") and
1293	<	* value ("nextVal") of a valid node (i.e., one that, at point of
1294	<	* snapshot, has a nonnull user value). Because val fields can
1295	<	* change (including to null, indicating deletion), field nextVal
1296	<	* might not be accurate at point of use, but still maintains the
1297	<	* weak consistency property of holding a value that was once
1298	<	* valid.
1299	<	*
1300	<	* Internal traversals directly access these fields, as in:
1301	<	* {@code while (it.next != null) { process(it.nextKey); it.advance(); }}
1302	<	*
1303	<	* Exported iterators (subclasses of ViewIterator) extract key,
1304	<	* value, or key-value pairs as return values of Iterator.next(),
1305	<	* and encapsulate the it.next check as hasNext();
1306	<	*
1307	<	* The iterator visits once each still-valid node that was
1308	<	* reachable upon iterator construction. It might miss some that
1309	<	* were added to a bin after the bin was visited, which is OK wrt
1310	<	* consistency guarantees. Maintaining this property in the face
1311	<	* of possible ongoing resizes requires a fair amount of
1312	<	* bookkeeping state that is difficult to optimize away amidst
1313	<	* volatile accesses.  Even so, traversal maintains reasonable
1314	<	* throughput.
1315	<	*
1316	<	* Normally, iteration proceeds bin-by-bin traversing lists.
1317	<	* However, if the table has been resized, then all future steps
1318	<	* must traverse both the bin at the current index as well as at
1319	<	* (index + baseSize); and so on for further resizings. To
1320	<	* paranoically cope with potential sharing by users of iterators
1321	<	* across threads, iteration terminates if a bounds checks fails
1322	<	* for a table read.
1323	<	*
1324	<	* The range-based constructor enables creation of parallel
1325	<	* range-splitting traversals. (Not yet implemented.)
784	>	* Spinlock (locked via CAS) used when resizing and/or creating CounterCells.
785		*/
786	<	static class InternalIterator {
1328	<	Node next;           // the next entry to use
1329	<	Node last;           // the last entry used
1330	<	Object nextKey;      // cached key field of next
1331	<	Object nextVal;      // cached val field of next
1332	<	Node[] tab;          // current table; updated if resized
1333	<	int index;           // index of bin to use next
1334	<	int baseIndex;       // current index of initial table
1335	<	final int baseLimit; // index bound for initial table
1336	<	final int baseSize;  // initial table size
786	>	private transient volatile int cellsBusy;
787
788	<	/** Creates iterator for all entries in the table. */
789	<	InternalIterator(Node[] tab) {
790	<	this.tab = tab;
791	<	baseLimit = baseSize = (tab == null) ? 0 : tab.length;
1342	<	index = baseIndex = 0;
1343	<	next = null;
1344	<	advance();
1345	<	}
788	>	/**
789	>	* Table of counter cells. When non-null, size is a power of 2.
790	>	*/
791	>	private transient volatile CounterCell[] counterCells;
792
793	<	/** Creates iterator for the given range of the table */
794	<	InternalIterator(Node[] tab, int lo, int hi) {
795	<	this.tab = tab;
796	<	baseSize = (tab == null) ? 0 : tab.length;
1351	<	baseLimit = (hi <= baseSize) ? hi : baseSize;
1352	<	index = baseIndex = (lo >= 0) ? lo : 0;
1353	<	next = null;
1354	<	advance();
1355	<	}
793	>	// views
794	>	private transient KeySetView<K,V> keySet;
795	>	private transient ValuesView<K,V> values;
796	>	private transient EntrySetView<K,V> entrySet;
797
1357	–	/** Advances next. See above for explanation. */
1358	–	final void advance() {
1359	–	Node e = last = next;
1360	–	outer: do {
1361	–	if (e != null)                  // advance past used/skipped node
1362	–	e = e.next;
1363	–	while (e == null) {             // get to next non-null bin
1364	–	Node[] t; int b, i, n;      // checks must use locals
1365	–	if ((b = baseIndex) >= baseLimit || (i = index) < 0 ||
1366	–	(t = tab) == null || i >= (n = t.length))
1367	–	break outer;
1368	–	else if ((e = tabAt(t, i)) != null && e.hash == MOVED)
1369	–	tab = (Node[])e.key;    // restarts due to null val
1370	–	else                        // visit upper slots if present
1371	–	index = (i += baseSize) < n ? i : (baseIndex = b + 1);
1372	–	}
1373	–	nextKey = e.key;
1374	–	} while ((nextVal = e.val) == null);// skip deleted or special nodes
1375	–	next = e;
1376	–	}
1377	–	}
798
799		/* ---------------- Public operations -------------- */
800
801		/**
802	<	* Creates a new, empty map with the default initial table size (16),
802	>	* Creates a new, empty map with the default initial table size (16).
803		*/
804		public ConcurrentHashMapV8() {
1385	–	this.counter = new LongAdder();
805		}
806
807		/**
#	Line 1401 | Line 820 | public class ConcurrentHashMapV8<K, V>
820		int cap = ((initialCapacity >= (MAXIMUM_CAPACITY >>> 1)) ?
821		MAXIMUM_CAPACITY :
822		tableSizeFor(initialCapacity + (initialCapacity >>> 1) + 1));
1404	–	this.counter = new LongAdder();
823		this.sizeCtl = cap;
824		}
825
#	Line 1411 | Line 829 | public class ConcurrentHashMapV8<K, V>
829		* @param m the map
830		*/
831		public ConcurrentHashMapV8(Map<? extends K, ? extends V> m) {
1414	–	this.counter = new LongAdder();
832		this.sizeCtl = DEFAULT_CAPACITY;
833	<	internalPutAll(m);
833	>	putAll(m);
834		}
835
836		/**
#	Line 1454 | Line 871 | public class ConcurrentHashMapV8<K, V>
871		* nonpositive
872		*/
873		public ConcurrentHashMapV8(int initialCapacity,
874	<	float loadFactor, int concurrencyLevel) {
874	>	float loadFactor, int concurrencyLevel) {
875		if (!(loadFactor > 0.0f) || initialCapacity < 0 || concurrencyLevel <= 0)
876		throw new IllegalArgumentException();
877		if (initialCapacity < concurrencyLevel)   // Use at least as many bins
878		initialCapacity = concurrencyLevel;   // as estimated threads
879		long size = (long)(1.0 + (long)initialCapacity / loadFactor);
880	<	int cap =  ((size >= (long)MAXIMUM_CAPACITY) ?
881	<	MAXIMUM_CAPACITY: tableSizeFor((int)size));
1465	<	this.counter = new LongAdder();
880	>	int cap = (size >= (long)MAXIMUM_CAPACITY) ?
881	>	MAXIMUM_CAPACITY : tableSizeFor((int)size);
882		this.sizeCtl = cap;
883		}
884
885	<	/**
1470	<	* {@inheritDoc}
1471	<	*/
1472	<	public boolean isEmpty() {
1473	<	return counter.sum() <= 0L; // ignore transient negative values
1474	<	}
885	>	// Original (since JDK1.2) Map methods
886
887		/**
888		* {@inheritDoc}
889		*/
890		public int size() {
891	<	long n = counter.sum();
891	>	long n = sumCount();
892		return ((n < 0L) ? 0 :
893		(n > (long)Integer.MAX_VALUE) ? Integer.MAX_VALUE :
894		(int)n);
895		}
896
897	<	final long longSize() { // accurate version of size needed for views
898	<	long n = counter.sum();
899	<	return (n < 0L) ? 0L : n;
897	>	/**
898	>	* {@inheritDoc}
899	>	*/
900	>	public boolean isEmpty() {
901	>	return sumCount() <= 0L; // ignore transient negative values
902		}
903
904		/**
#	Line 1499 | Line 912 | public class ConcurrentHashMapV8<K, V>
912		*
913		* @throws NullPointerException if the specified key is null
914		*/
1502	–	@SuppressWarnings("unchecked")
915		public V get(Object key) {
916	<	if (key == null)
917	<	throw new NullPointerException();
918	<	return (V)internalGet(key);
916	>	Node<K,V>[] tab; Node<K,V> e, p; int n, eh; K ek;
917	>	int h = spread(key.hashCode());
918	>	if ((tab = table) != null && (n = tab.length) > 0 &&
919	>	(e = tabAt(tab, (n - 1) & h)) != null) {
920	>	if ((eh = e.hash) == h) {
921	>	if ((ek = e.key) == key || (ek != null && key.equals(ek)))
922	>	return e.val;
923	>	}
924	>	else if (eh < 0)
925	>	return (p = e.find(h, key)) != null ? p.val : null;
926	>	while ((e = e.next) != null) {
927	>	if (e.hash == h &&
928	>	((ek = e.key) == key || (ek != null && key.equals(ek))))
929	>	return e.val;
930	>	}
931	>	}
932	>	return null;
933		}
934
935		/**
936		* Tests if the specified object is a key in this table.
937		*
938	<	* @param  key   possible key
938	>	* @param  key possible key
939		* @return {@code true} if and only if the specified object
940		*         is a key in this table, as determined by the
941		*         {@code equals} method; {@code false} otherwise
942		* @throws NullPointerException if the specified key is null
943		*/
944		public boolean containsKey(Object key) {
945	<	if (key == null)
1520	<	throw new NullPointerException();
1521	<	return internalGet(key) != null;
945	>	return get(key) != null;
946		}
947
948		/**
#	Line 1534 | Line 958 | public class ConcurrentHashMapV8<K, V>
958		public boolean containsValue(Object value) {
959		if (value == null)
960		throw new NullPointerException();
961	<	Object v;
962	<	InternalIterator it = new InternalIterator(table);
963	<	while (it.next != null) {
964	<	if ((v = it.nextVal) == value || value.equals(v))
965	<	return true;
966	<	it.advance();
961	>	Node<K,V>[] t;
962	>	if ((t = table) != null) {
963	>	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
964	>	for (Node<K,V> p; (p = it.advance()) != null; ) {
965	>	V v;
966	>	if ((v = p.val) == value || (v != null && value.equals(v)))
967	>	return true;
968	>	}
969		}
970		return false;
971		}
972
973		/**
1548	–	* Legacy method testing if some key maps into the specified value
1549	–	* in this table.  This method is identical in functionality to
1550	–	* {@link #containsValue}, and exists solely to ensure
1551	–	* full compatibility with class {@link java.util.Hashtable},
1552	–	* which supported this method prior to introduction of the
1553	–	* Java Collections framework.
1554	–	*
1555	–	* @param  value a value to search for
1556	–	* @return {@code true} if and only if some key maps to the
1557	–	*         {@code value} argument in this table as
1558	–	*         determined by the {@code equals} method;
1559	–	*         {@code false} otherwise
1560	–	* @throws NullPointerException if the specified value is null
1561	–	*/
1562	–	public boolean contains(Object value) {
1563	–	return containsValue(value);
1564	–	}
1565	–
1566	–	/**
974		* Maps the specified key to the specified value in this table.
975		* Neither the key nor the value can be null.
976		*
977	<	* <p> The value can be retrieved by calling the {@code get} method
977	>	* <p>The value can be retrieved by calling the {@code get} method
978		* with a key that is equal to the original key.
979		*
980		* @param key key with which the specified value is to be associated
#	Line 1576 | Line 983 | public class ConcurrentHashMapV8<K, V>
983		*         {@code null} if there was no mapping for {@code key}
984		* @throws NullPointerException if the specified key or value is null
985		*/
1579	–	@SuppressWarnings("unchecked")
986		public V put(K key, V value) {
987	<	if (key == null || value == null)
1582	<	throw new NullPointerException();
1583	<	return (V)internalPut(key, value);
987	>	return putVal(key, value, false);
988		}
989
990	<	/**
991	<	* {@inheritDoc}
992	<	*
993	<	* @return the previous value associated with the specified key,
994	<	*         or {@code null} if there was no mapping for the key
995	<	* @throws NullPointerException if the specified key or value is null
996	<	*/
997	<	@SuppressWarnings("unchecked")
998	<	public V putIfAbsent(K key, V value) {
999	<	if (key == null || value == null)
1000	<	throw new NullPointerException();
1001	<	return (V)internalPutIfAbsent(key, value);
990	>	/** Implementation for put and putIfAbsent */
991	>	final V putVal(K key, V value, boolean onlyIfAbsent) {
992	>	if (key == null || value == null) throw new NullPointerException();
993	>	int hash = spread(key.hashCode());
994	>	int binCount = 0;
995	>	for (Node<K,V>[] tab = table;;) {
996	>	Node<K,V> f; int n, i, fh;
997	>	if (tab == null || (n = tab.length) == 0)
998	>	tab = initTable();
999	>	else if ((f = tabAt(tab, i = (n - 1) & hash)) == null) {
1000	>	if (casTabAt(tab, i, null,
1001	>	new Node<K,V>(hash, key, value, null)))
1002	>	break;                   // no lock when adding to empty bin
1003	>	}
1004	>	else if ((fh = f.hash) == MOVED)
1005	>	tab = helpTransfer(tab, f);
1006	>	else {
1007	>	V oldVal = null;
1008	>	synchronized (f) {
1009	>	if (tabAt(tab, i) == f) {
1010	>	if (fh >= 0) {
1011	>	binCount = 1;
1012	>	for (Node<K,V> e = f;; ++binCount) {
1013	>	K ek;
1014	>	if (e.hash == hash &&
1015	>	((ek = e.key) == key ||
1016	>	(ek != null && key.equals(ek)))) {
1017	>	oldVal = e.val;
1018	>	if (!onlyIfAbsent)
1019	>	e.val = value;
1020	>	break;
1021	>	}
1022	>	Node<K,V> pred = e;
1023	>	if ((e = e.next) == null) {
1024	>	pred.next = new Node<K,V>(hash, key,
1025	>	value, null);
1026	>	break;
1027	>	}
1028	>	}
1029	>	}
1030	>	else if (f instanceof TreeBin) {
1031	>	Node<K,V> p;
1032	>	binCount = 2;
1033	>	if ((p = ((TreeBin<K,V>)f).putTreeVal(hash, key,
1034	>	value)) != null) {
1035	>	oldVal = p.val;
1036	>	if (!onlyIfAbsent)
1037	>	p.val = value;
1038	>	}
1039	>	}
1040	>	}
1041	>	}
1042	>	if (binCount != 0) {
1043	>	if (binCount >= TREEIFY_THRESHOLD)
1044	>	treeifyBin(tab, i);
1045	>	if (oldVal != null)
1046	>	return oldVal;
1047	>	break;
1048	>	}
1049	>	}
1050	>	}
1051	>	addCount(1L, binCount);
1052	>	return null;
1053		}
1054
1055		/**
#	Line 1605 | Line 1060 | public class ConcurrentHashMapV8<K, V>
1060		* @param m mappings to be stored in this map
1061		*/
1062		public void putAll(Map<? extends K, ? extends V> m) {
1063	<	internalPutAll(m);
1064	<	}
1065	<
1611	<	/**
1612	<	* If the specified key is not already associated with a value,
1613	<	* computes its value using the given mappingFunction and
1614	<	* enters it into the map.  This is equivalent to
1615	<	* <pre> {@code
1616	<	* if (map.containsKey(key))
1617	<	*   return map.get(key);
1618	<	* value = mappingFunction.map(key);
1619	<	* map.put(key, value);
1620	<	* return value;}</pre>
1621	<	*
1622	<	* except that the action is performed atomically.  If the
1623	<	* function returns {@code null} (in which case a {@code
1624	<	* NullPointerException} is thrown), or the function itself throws
1625	<	* an (unchecked) exception, the exception is rethrown to its
1626	<	* caller, and no mapping is recorded.  Some attempted update
1627	<	* operations on this map by other threads may be blocked while
1628	<	* computation is in progress, so the computation should be short
1629	<	* and simple, and must not attempt to update any other mappings
1630	<	* of this Map. The most appropriate usage is to construct a new
1631	<	* object serving as an initial mapped value, or memoized result,
1632	<	* as in:
1633	<	*
1634	<	*  <pre> {@code
1635	<	* map.computeIfAbsent(key, new MappingFunction<K, V>() {
1636	<	*   public V map(K k) { return new Value(f(k)); }});}</pre>
1637	<	*
1638	<	* @param key key with which the specified value is to be associated
1639	<	* @param mappingFunction the function to compute a value
1640	<	* @return the current (existing or computed) value associated with
1641	<	*         the specified key.
1642	<	* @throws NullPointerException if the specified key, mappingFunction,
1643	<	*         or computed value is null
1644	<	* @throws IllegalStateException if the computation detectably
1645	<	*         attempts a recursive update to this map that would
1646	<	*         otherwise never complete
1647	<	* @throws RuntimeException or Error if the mappingFunction does so,
1648	<	*         in which case the mapping is left unestablished
1649	<	*/
1650	<	@SuppressWarnings("unchecked")
1651	<	public V computeIfAbsent(K key, MappingFunction<? super K, ? extends V> mappingFunction) {
1652	<	if (key == null || mappingFunction == null)
1653	<	throw new NullPointerException();
1654	<	return (V)internalComputeIfAbsent(key, mappingFunction);
1655	<	}
1656	<
1657	<	/**
1658	<	* Computes and enters a new mapping value given a key and
1659	<	* its current mapped value (or {@code null} if there is no current
1660	<	* mapping). This is equivalent to
1661	<	*  <pre> {@code
1662	<	*  map.put(key, remappingFunction.remap(key, map.get(key));
1663	<	* }</pre>
1664	<	*
1665	<	* except that the action is performed atomically.  If the
1666	<	* function returns {@code null} (in which case a {@code
1667	<	* NullPointerException} is thrown), or the function itself throws
1668	<	* an (unchecked) exception, the exception is rethrown to its
1669	<	* caller, and current mapping is left unchanged.  Some attempted
1670	<	* update operations on this map by other threads may be blocked
1671	<	* while computation is in progress, so the computation should be
1672	<	* short and simple, and must not attempt to update any other
1673	<	* mappings of this Map. For example, to either create or
1674	<	* append new messages to a value mapping:
1675	<	*
1676	<	* <pre> {@code
1677	<	* Map<Key, String> map = ...;
1678	<	* final String msg = ...;
1679	<	* map.compute(key, new RemappingFunction<Key, String>() {
1680	<	*   public String remap(Key k, String v) {
1681	<	*    return (v == null) ? msg : v + msg;});}}</pre>
1682	<	*
1683	<	* @param key key with which the specified value is to be associated
1684	<	* @param remappingFunction the function to compute a value
1685	<	* @return the new value associated with
1686	<	*         the specified key.
1687	<	* @throws NullPointerException if the specified key or remappingFunction
1688	<	*         or computed value is null
1689	<	* @throws IllegalStateException if the computation detectably
1690	<	*         attempts a recursive update to this map that would
1691	<	*         otherwise never complete
1692	<	* @throws RuntimeException or Error if the remappingFunction does so,
1693	<	*         in which case the mapping is unchanged
1694	<	*/
1695	<	@SuppressWarnings("unchecked")
1696	<	public V compute(K key, RemappingFunction<? super K, V> remappingFunction) {
1697	<	if (key == null || remappingFunction == null)
1698	<	throw new NullPointerException();
1699	<	return (V)internalCompute(key, remappingFunction);
1063	>	tryPresize(m.size());
1064	>	for (Map.Entry<? extends K, ? extends V> e : m.entrySet())
1065	>	putVal(e.getKey(), e.getValue(), false);
1066		}
1067
1068		/**
#	Line 1708 | Line 1074 | public class ConcurrentHashMapV8<K, V>
1074		*         {@code null} if there was no mapping for {@code key}
1075		* @throws NullPointerException if the specified key is null
1076		*/
1711	–	@SuppressWarnings("unchecked")
1077		public V remove(Object key) {
1078	<	if (key == null)
1714	<	throw new NullPointerException();
1715	<	return (V)internalReplace(key, null, null);
1716	<	}
1717	<
1718	<	/**
1719	<	* {@inheritDoc}
1720	<	*
1721	<	* @throws NullPointerException if the specified key is null
1722	<	*/
1723	<	public boolean remove(Object key, Object value) {
1724	<	if (key == null)
1725	<	throw new NullPointerException();
1726	<	if (value == null)
1727	<	return false;
1728	<	return internalReplace(key, null, value) != null;
1729	<	}
1730	<
1731	<	/**
1732	<	* {@inheritDoc}
1733	<	*
1734	<	* @throws NullPointerException if any of the arguments are null
1735	<	*/
1736	<	public boolean replace(K key, V oldValue, V newValue) {
1737	<	if (key == null || oldValue == null || newValue == null)
1738	<	throw new NullPointerException();
1739	<	return internalReplace(key, newValue, oldValue) != null;
1078	>	return replaceNode(key, null, null);
1079		}
1080
1081		/**
1082	<	* {@inheritDoc}
1083	<	*
1084	<	* @return the previous value associated with the specified key,
1746	<	*         or {@code null} if there was no mapping for the key
1747	<	* @throws NullPointerException if the specified key or value is null
1082	>	* Implementation for the four public remove/replace methods:
1083	>	* Replaces node value with v, conditional upon match of cv if
1084	>	* non-null.  If resulting value is null, delete.
1085		*/
1086	<	@SuppressWarnings("unchecked")
1087	<	public V replace(K key, V value) {
1088	<	if (key == null || value == null)
1089	<	throw new NullPointerException();
1090	<	return (V)internalReplace(key, value, null);
1086	>	final V replaceNode(Object key, V value, Object cv) {
1087	>	int hash = spread(key.hashCode());
1088	>	for (Node<K,V>[] tab = table;;) {
1089	>	Node<K,V> f; int n, i, fh;
1090	>	if (tab == null || (n = tab.length) == 0 ||
1091	>	(f = tabAt(tab, i = (n - 1) & hash)) == null)
1092	>	break;
1093	>	else if ((fh = f.hash) == MOVED)
1094	>	tab = helpTransfer(tab, f);
1095	>	else {
1096	>	V oldVal = null;
1097	>	boolean validated = false;
1098	>	synchronized (f) {
1099	>	if (tabAt(tab, i) == f) {
1100	>	if (fh >= 0) {
1101	>	validated = true;
1102	>	for (Node<K,V> e = f, pred = null;;) {
1103	>	K ek;
1104	>	if (e.hash == hash &&
1105	>	((ek = e.key) == key ||
1106	>	(ek != null && key.equals(ek)))) {
1107	>	V ev = e.val;
1108	>	if (cv == null || cv == ev ||
1109	>	(ev != null && cv.equals(ev))) {
1110	>	oldVal = ev;
1111	>	if (value != null)
1112	>	e.val = value;
1113	>	else if (pred != null)
1114	>	pred.next = e.next;
1115	>	else
1116	>	setTabAt(tab, i, e.next);
1117	>	}
1118	>	break;
1119	>	}
1120	>	pred = e;
1121	>	if ((e = e.next) == null)
1122	>	break;
1123	>	}
1124	>	}
1125	>	else if (f instanceof TreeBin) {
1126	>	validated = true;
1127	>	TreeBin<K,V> t = (TreeBin<K,V>)f;
1128	>	TreeNode<K,V> r, p;
1129	>	if ((r = t.root) != null &&
1130	>	(p = r.findTreeNode(hash, key, null)) != null) {
1131	>	V pv = p.val;
1132	>	if (cv == null || cv == pv ||
1133	>	(pv != null && cv.equals(pv))) {
1134	>	oldVal = pv;
1135	>	if (value != null)
1136	>	p.val = value;
1137	>	else if (t.removeTreeNode(p))
1138	>	setTabAt(tab, i, untreeify(t.first));
1139	>	}
1140	>	}
1141	>	}
1142	>	}
1143	>	}
1144	>	if (validated) {
1145	>	if (oldVal != null) {
1146	>	if (value == null)
1147	>	addCount(-1L, -1);
1148	>	return oldVal;
1149	>	}
1150	>	break;
1151	>	}
1152	>	}
1153	>	}
1154	>	return null;
1155		}
1156
1157		/**
1158		* Removes all of the mappings from this map.
1159		*/
1160		public void clear() {
1161	<	internalClear();
1161	>	long delta = 0L; // negative number of deletions
1162	>	int i = 0;
1163	>	Node<K,V>[] tab = table;
1164	>	while (tab != null && i < tab.length) {
1165	>	int fh;
1166	>	Node<K,V> f = tabAt(tab, i);
1167	>	if (f == null)
1168	>	++i;
1169	>	else if ((fh = f.hash) == MOVED) {
1170	>	tab = helpTransfer(tab, f);
1171	>	i = 0; // restart
1172	>	}
1173	>	else {
1174	>	synchronized (f) {
1175	>	if (tabAt(tab, i) == f) {
1176	>	Node<K,V> p = (fh >= 0 ? f :
1177	>	(f instanceof TreeBin) ?
1178	>	((TreeBin<K,V>)f).first : null);
1179	>	while (p != null) {
1180	>	--delta;
1181	>	p = p.next;
1182	>	}
1183	>	setTabAt(tab, i++, null);
1184	>	}
1185	>	}
1186	>	}
1187	>	}
1188	>	if (delta != 0L)
1189	>	addCount(delta, -1);
1190		}
1191
1192		/**
1193		* Returns a {@link Set} view of the keys contained in this map.
1194		* The set is backed by the map, so changes to the map are
1195	<	* reflected in the set, and vice-versa.  The set supports element
1195	>	* reflected in the set, and vice-versa. The set supports element
1196		* removal, which removes the corresponding mapping from this map,
1197		* via the {@code Iterator.remove}, {@code Set.remove},
1198		* {@code removeAll}, {@code retainAll}, and {@code clear}
#	Line 1775 | Line 1204 | public class ConcurrentHashMapV8<K, V>
1204		* and guarantees to traverse elements as they existed upon
1205		* construction of the iterator, and may (but is not guaranteed to)
1206		* reflect any modifications subsequent to construction.
1207	+	*
1208	+	* @return the set view
1209		*/
1210	<	public Set<K> keySet() {
1211	<	KeySet<K,V> ks = keySet;
1212	<	return (ks != null) ? ks : (keySet = new KeySet<K,V>(this));
1210	>	public KeySetView<K,V> keySet() {
1211	>	KeySetView<K,V> ks;
1212	>	return (ks = keySet) != null ? ks : (keySet = new KeySetView<K,V>(this, null));
1213		}
1214
1215		/**
#	Line 1796 | Line 1227 | public class ConcurrentHashMapV8<K, V>
1227		* and guarantees to traverse elements as they existed upon
1228		* construction of the iterator, and may (but is not guaranteed to)
1229		* reflect any modifications subsequent to construction.
1230	+	*
1231	+	* @return the collection view
1232		*/
1233		public Collection<V> values() {
1234	<	Values<K,V> vs = values;
1235	<	return (vs != null) ? vs : (values = new Values<K,V>(this));
1234	>	ValuesView<K,V> vs;
1235	>	return (vs = values) != null ? vs : (values = new ValuesView<K,V>(this));
1236		}
1237
1238		/**
#	Line 1809 | Line 1242 | public class ConcurrentHashMapV8<K, V>
1242		* removal, which removes the corresponding mapping from the map,
1243		* via the {@code Iterator.remove}, {@code Set.remove},
1244		* {@code removeAll}, {@code retainAll}, and {@code clear}
1245	<	* operations.  It does not support the {@code add} or
1813	<	* {@code addAll} operations.
1245	>	* operations.
1246		*
1247		* <p>The view's {@code iterator} is a "weakly consistent" iterator
1248		* that will never throw {@link ConcurrentModificationException},
1249		* and guarantees to traverse elements as they existed upon
1250		* construction of the iterator, and may (but is not guaranteed to)
1251		* reflect any modifications subsequent to construction.
1820	–	*/
1821	–	public Set<Map.Entry<K,V>> entrySet() {
1822	–	EntrySet<K,V> es = entrySet;
1823	–	return (es != null) ? es : (entrySet = new EntrySet<K,V>(this));
1824	–	}
1825	–
1826	–	/**
1827	–	* Returns an enumeration of the keys in this table.
1828	–	*
1829	–	* @return an enumeration of the keys in this table
1830	–	* @see #keySet()
1831	–	*/
1832	–	public Enumeration<K> keys() {
1833	–	return new KeyIterator<K,V>(this);
1834	–	}
1835	–
1836	–	/**
1837	–	* Returns an enumeration of the values in this table.
1252		*
1253	<	* @return an enumeration of the values in this table
1840	<	* @see #values()
1253	>	* @return the set view
1254		*/
1255	<	public Enumeration<V> elements() {
1256	<	return new ValueIterator<K,V>(this);
1255	>	public Set<Map.Entry<K,V>> entrySet() {
1256	>	EntrySetView<K,V> es;
1257	>	return (es = entrySet) != null ? es : (entrySet = new EntrySetView<K,V>(this));
1258		}
1259
1260		/**
#	Line 1852 | Line 1266 | public class ConcurrentHashMapV8<K, V>
1266		*/
1267		public int hashCode() {
1268		int h = 0;
1269	<	InternalIterator it = new InternalIterator(table);
1270	<	while (it.next != null) {
1271	<	h += it.nextKey.hashCode() ^ it.nextVal.hashCode();
1272	<	it.advance();
1269	>	Node<K,V>[] t;
1270	>	if ((t = table) != null) {
1271	>	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
1272	>	for (Node<K,V> p; (p = it.advance()) != null; )
1273	>	h += p.key.hashCode() ^ p.val.hashCode();
1274		}
1275		return h;
1276		}
#	Line 1872 | Line 1287 | public class ConcurrentHashMapV8<K, V>
1287		* @return a string representation of this map
1288		*/
1289		public String toString() {
1290	<	InternalIterator it = new InternalIterator(table);
1290	>	Node<K,V>[] t;
1291	>	int f = (t = table) == null ? 0 : t.length;
1292	>	Traverser<K,V> it = new Traverser<K,V>(t, f, 0, f);
1293		StringBuilder sb = new StringBuilder();
1294		sb.append('{');
1295	<	if (it.next != null) {
1295	>	Node<K,V> p;
1296	>	if ((p = it.advance()) != null) {
1297		for (;;) {
1298	<	Object k = it.nextKey, v = it.nextVal;
1298	>	K k = p.key;
1299	>	V v = p.val;
1300		sb.append(k == this ? "(this Map)" : k);
1301		sb.append('=');
1302		sb.append(v == this ? "(this Map)" : v);
1303	<	it.advance();
1885	<	if (it.next == null)
1303	>	if ((p = it.advance()) == null)
1304		break;
1305		sb.append(',').append(' ');
1306		}
#	Line 1905 | Line 1323 | public class ConcurrentHashMapV8<K, V>
1323		if (!(o instanceof Map))
1324		return false;
1325		Map<?,?> m = (Map<?,?>) o;
1326	<	InternalIterator it = new InternalIterator(table);
1327	<	while (it.next != null) {
1328	<	Object val = it.nextVal;
1329	<	Object v = m.get(it.nextKey);
1326	>	Node<K,V>[] t;
1327	>	int f = (t = table) == null ? 0 : t.length;
1328	>	Traverser<K,V> it = new Traverser<K,V>(t, f, 0, f);
1329	>	for (Node<K,V> p; (p = it.advance()) != null; ) {
1330	>	V val = p.val;
1331	>	Object v = m.get(p.key);
1332		if (v == null || (v != val && !v.equals(val)))
1333		return false;
1914	–	it.advance();
1334		}
1335		for (Map.Entry<?,?> e : m.entrySet()) {
1336		Object mk, mv, v;
1337		if ((mk = e.getKey()) == null ||
1338		(mv = e.getValue()) == null ||
1339	<	(v = internalGet(mk)) == null ||
1339	>	(v = get(mk)) == null ||
1340		(mv != v && !mv.equals(v)))
1341		return false;
1342		}
#	Line 1925 | Line 1344 | public class ConcurrentHashMapV8<K, V>
1344		return true;
1345		}
1346
1347	<	/* ----------------Iterators -------------- */
1347	>	/**
1348	>	* Stripped-down version of helper class used in previous version,
1349	>	* declared for the sake of serialization compatibility
1350	>	*/
1351	>	static class Segment<K,V> extends ReentrantLock implements Serializable {
1352	>	private static final long serialVersionUID = 2249069246763182397L;
1353	>	final float loadFactor;
1354	>	Segment(float lf) { this.loadFactor = lf; }
1355	>	}
1356	>
1357	>	/**
1358	>	* Saves the state of the {@code ConcurrentHashMapV8} instance to a
1359	>	* stream (i.e., serializes it).
1360	>	* @param s the stream
1361	>	* @serialData
1362	>	* the key (Object) and value (Object)
1363	>	* for each key-value mapping, followed by a null pair.
1364	>	* The key-value mappings are emitted in no particular order.
1365	>	*/
1366	>	private void writeObject(java.io.ObjectOutputStream s)
1367	>	throws java.io.IOException {
1368	>	// For serialization compatibility
1369	>	// Emulate segment calculation from previous version of this class
1370	>	int sshift = 0;
1371	>	int ssize = 1;
1372	>	while (ssize < DEFAULT_CONCURRENCY_LEVEL) {
1373	>	++sshift;
1374	>	ssize <<= 1;
1375	>	}
1376	>	int segmentShift = 32 - sshift;
1377	>	int segmentMask = ssize - 1;
1378	>	@SuppressWarnings("unchecked") Segment<K,V>[] segments = (Segment<K,V>[])
1379	>	new Segment<?,?>[DEFAULT_CONCURRENCY_LEVEL];
1380	>	for (int i = 0; i < segments.length; ++i)
1381	>	segments[i] = new Segment<K,V>(LOAD_FACTOR);
1382	>	s.putFields().put("segments", segments);
1383	>	s.putFields().put("segmentShift", segmentShift);
1384	>	s.putFields().put("segmentMask", segmentMask);
1385	>	s.writeFields();
1386	>
1387	>	Node<K,V>[] t;
1388	>	if ((t = table) != null) {
1389	>	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
1390	>	for (Node<K,V> p; (p = it.advance()) != null; ) {
1391	>	s.writeObject(p.key);
1392	>	s.writeObject(p.val);
1393	>	}
1394	>	}
1395	>	s.writeObject(null);
1396	>	s.writeObject(null);
1397	>	segments = null; // throw away
1398	>	}
1399	>
1400	>	/**
1401	>	* Reconstitutes the instance from a stream (that is, deserializes it).
1402	>	* @param s the stream
1403	>	*/
1404	>	private void readObject(java.io.ObjectInputStream s)
1405	>	throws java.io.IOException, ClassNotFoundException {
1406	>	/*
1407	>	* To improve performance in typical cases, we create nodes
1408	>	* while reading, then place in table once size is known.
1409	>	* However, we must also validate uniqueness and deal with
1410	>	* overpopulated bins while doing so, which requires
1411	>	* specialized versions of putVal mechanics.
1412	>	*/
1413	>	sizeCtl = -1; // force exclusion for table construction
1414	>	s.defaultReadObject();
1415	>	long size = 0L;
1416	>	Node<K,V> p = null;
1417	>	for (;;) {
1418	>	@SuppressWarnings("unchecked") K k = (K) s.readObject();
1419	>	@SuppressWarnings("unchecked") V v = (V) s.readObject();
1420	>	if (k != null && v != null) {
1421	>	p = new Node<K,V>(spread(k.hashCode()), k, v, p);
1422	>	++size;
1423	>	}
1424	>	else
1425	>	break;
1426	>	}
1427	>	if (size == 0L)
1428	>	sizeCtl = 0;
1429	>	else {
1430	>	int n;
1431	>	if (size >= (long)(MAXIMUM_CAPACITY >>> 1))
1432	>	n = MAXIMUM_CAPACITY;
1433	>	else {
1434	>	int sz = (int)size;
1435	>	n = tableSizeFor(sz + (sz >>> 1) + 1);
1436	>	}
1437	>	@SuppressWarnings({"rawtypes","unchecked"})
1438	>	Node<K,V>[] tab = (Node<K,V>[])new Node[n];
1439	>	int mask = n - 1;
1440	>	long added = 0L;
1441	>	while (p != null) {
1442	>	boolean insertAtFront;
1443	>	Node<K,V> next = p.next, first;
1444	>	int h = p.hash, j = h & mask;
1445	>	if ((first = tabAt(tab, j)) == null)
1446	>	insertAtFront = true;
1447	>	else {
1448	>	K k = p.key;
1449	>	if (first.hash < 0) {
1450	>	TreeBin<K,V> t = (TreeBin<K,V>)first;
1451	>	if (t.putTreeVal(h, k, p.val) == null)
1452	>	++added;
1453	>	insertAtFront = false;
1454	>	}
1455	>	else {
1456	>	int binCount = 0;
1457	>	insertAtFront = true;
1458	>	Node<K,V> q; K qk;
1459	>	for (q = first; q != null; q = q.next) {
1460	>	if (q.hash == h &&
1461	>	((qk = q.key) == k ||
1462	>	(qk != null && k.equals(qk)))) {
1463	>	insertAtFront = false;
1464	>	break;
1465	>	}
1466	>	++binCount;
1467	>	}
1468	>	if (insertAtFront && binCount >= TREEIFY_THRESHOLD) {
1469	>	insertAtFront = false;
1470	>	++added;
1471	>	p.next = first;
1472	>	TreeNode<K,V> hd = null, tl = null;
1473	>	for (q = p; q != null; q = q.next) {
1474	>	TreeNode<K,V> t = new TreeNode<K,V>
1475	>	(q.hash, q.key, q.val, null, null);
1476	>	if ((t.prev = tl) == null)
1477	>	hd = t;
1478	>	else
1479	>	tl.next = t;
1480	>	tl = t;
1481	>	}
1482	>	setTabAt(tab, j, new TreeBin<K,V>(hd));
1483	>	}
1484	>	}
1485	>	}
1486	>	if (insertAtFront) {
1487	>	++added;
1488	>	p.next = first;
1489	>	setTabAt(tab, j, p);
1490	>	}
1491	>	p = next;
1492	>	}
1493	>	table = tab;
1494	>	sizeCtl = n - (n >>> 2);
1495	>	baseCount = added;
1496	>	}
1497	>	}
1498	>
1499	>	// ConcurrentMap methods
1500	>
1501	>	/**
1502	>	* {@inheritDoc}
1503	>	*
1504	>	* @return the previous value associated with the specified key,
1505	>	*         or {@code null} if there was no mapping for the key
1506	>	* @throws NullPointerException if the specified key or value is null
1507	>	*/
1508	>	public V putIfAbsent(K key, V value) {
1509	>	return putVal(key, value, true);
1510	>	}
1511	>
1512	>	/**
1513	>	* {@inheritDoc}
1514	>	*
1515	>	* @throws NullPointerException if the specified key is null
1516	>	*/
1517	>	public boolean remove(Object key, Object value) {
1518	>	if (key == null)
1519	>	throw new NullPointerException();
1520	>	return value != null && replaceNode(key, null, value) != null;
1521	>	}
1522	>
1523	>	/**
1524	>	* {@inheritDoc}
1525	>	*
1526	>	* @throws NullPointerException if any of the arguments are null
1527	>	*/
1528	>	public boolean replace(K key, V oldValue, V newValue) {
1529	>	if (key == null || oldValue == null || newValue == null)
1530	>	throw new NullPointerException();
1531	>	return replaceNode(key, newValue, oldValue) != null;
1532	>	}
1533	>
1534	>	/**
1535	>	* {@inheritDoc}
1536	>	*
1537	>	* @return the previous value associated with the specified key,
1538	>	*         or {@code null} if there was no mapping for the key
1539	>	* @throws NullPointerException if the specified key or value is null
1540	>	*/
1541	>	public V replace(K key, V value) {
1542	>	if (key == null || value == null)
1543	>	throw new NullPointerException();
1544	>	return replaceNode(key, value, null);
1545	>	}
1546	>
1547	>	// Overrides of JDK8+ Map extension method defaults
1548	>
1549	>	/**
1550	>	* Returns the value to which the specified key is mapped, or the
1551	>	* given default value if this map contains no mapping for the
1552	>	* key.
1553	>	*
1554	>	* @param key the key whose associated value is to be returned
1555	>	* @param defaultValue the value to return if this map contains
1556	>	* no mapping for the given key
1557	>	* @return the mapping for the key, if present; else the default value
1558	>	* @throws NullPointerException if the specified key is null
1559	>	*/
1560	>	public V getOrDefault(Object key, V defaultValue) {
1561	>	V v;
1562	>	return (v = get(key)) == null ? defaultValue : v;
1563	>	}
1564	>
1565	>	public void forEach(BiAction<? super K, ? super V> action) {
1566	>	if (action == null) throw new NullPointerException();
1567	>	Node<K,V>[] t;
1568	>	if ((t = table) != null) {
1569	>	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
1570	>	for (Node<K,V> p; (p = it.advance()) != null; ) {
1571	>	action.apply(p.key, p.val);
1572	>	}
1573	>	}
1574	>	}
1575	>
1576	>	public void replaceAll(BiFun<? super K, ? super V, ? extends V> function) {
1577	>	if (function == null) throw new NullPointerException();
1578	>	Node<K,V>[] t;
1579	>	if ((t = table) != null) {
1580	>	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
1581	>	for (Node<K,V> p; (p = it.advance()) != null; ) {
1582	>	V oldValue = p.val;
1583	>	for (K key = p.key;;) {
1584	>	V newValue = function.apply(key, oldValue);
1585	>	if (newValue == null)
1586	>	throw new NullPointerException();
1587	>	if (replaceNode(key, newValue, oldValue) != null ||
1588	>	(oldValue = get(key)) == null)
1589	>	break;
1590	>	}
1591	>	}
1592	>	}
1593	>	}
1594	>
1595	>	/**
1596	>	* If the specified key is not already associated with a value,
1597	>	* attempts to compute its value using the given mapping function
1598	>	* and enters it into this map unless {@code null}.  The entire
1599	>	* method invocation is performed atomically, so the function is
1600	>	* applied at most once per key.  Some attempted update operations
1601	>	* on this map by other threads may be blocked while computation
1602	>	* is in progress, so the computation should be short and simple,
1603	>	* and must not attempt to update any other mappings of this map.
1604	>	*
1605	>	* @param key key with which the specified value is to be associated
1606	>	* @param mappingFunction the function to compute a value
1607	>	* @return the current (existing or computed) value associated with
1608	>	*         the specified key, or null if the computed value is null
1609	>	* @throws NullPointerException if the specified key or mappingFunction
1610	>	*         is null
1611	>	* @throws IllegalStateException if the computation detectably
1612	>	*         attempts a recursive update to this map that would
1613	>	*         otherwise never complete
1614	>	* @throws RuntimeException or Error if the mappingFunction does so,
1615	>	*         in which case the mapping is left unestablished
1616	>	*/
1617	>	public V computeIfAbsent(K key, Fun<? super K, ? extends V> mappingFunction) {
1618	>	if (key == null || mappingFunction == null)
1619	>	throw new NullPointerException();
1620	>	int h = spread(key.hashCode());
1621	>	V val = null;
1622	>	int binCount = 0;
1623	>	for (Node<K,V>[] tab = table;;) {
1624	>	Node<K,V> f; int n, i, fh;
1625	>	if (tab == null || (n = tab.length) == 0)
1626	>	tab = initTable();
1627	>	else if ((f = tabAt(tab, i = (n - 1) & h)) == null) {
1628	>	Node<K,V> r = new ReservationNode<K,V>();
1629	>	synchronized (r) {
1630	>	if (casTabAt(tab, i, null, r)) {
1631	>	binCount = 1;
1632	>	Node<K,V> node = null;
1633	>	try {
1634	>	if ((val = mappingFunction.apply(key)) != null)
1635	>	node = new Node<K,V>(h, key, val, null);
1636	>	} finally {
1637	>	setTabAt(tab, i, node);
1638	>	}
1639	>	}
1640	>	}
1641	>	if (binCount != 0)
1642	>	break;
1643	>	}
1644	>	else if ((fh = f.hash) == MOVED)
1645	>	tab = helpTransfer(tab, f);
1646	>	else {
1647	>	boolean added = false;
1648	>	synchronized (f) {
1649	>	if (tabAt(tab, i) == f) {
1650	>	if (fh >= 0) {
1651	>	binCount = 1;
1652	>	for (Node<K,V> e = f;; ++binCount) {
1653	>	K ek; V ev;
1654	>	if (e.hash == h &&
1655	>	((ek = e.key) == key ||
1656	>	(ek != null && key.equals(ek)))) {
1657	>	val = e.val;
1658	>	break;
1659	>	}
1660	>	Node<K,V> pred = e;
1661	>	if ((e = e.next) == null) {
1662	>	if ((val = mappingFunction.apply(key)) != null) {
1663	>	added = true;
1664	>	pred.next = new Node<K,V>(h, key, val, null);
1665	>	}
1666	>	break;
1667	>	}
1668	>	}
1669	>	}
1670	>	else if (f instanceof TreeBin) {
1671	>	binCount = 2;
1672	>	TreeBin<K,V> t = (TreeBin<K,V>)f;
1673	>	TreeNode<K,V> r, p;
1674	>	if ((r = t.root) != null &&
1675	>	(p = r.findTreeNode(h, key, null)) != null)
1676	>	val = p.val;
1677	>	else if ((val = mappingFunction.apply(key)) != null) {
1678	>	added = true;
1679	>	t.putTreeVal(h, key, val);
1680	>	}
1681	>	}
1682	>	}
1683	>	}
1684	>	if (binCount != 0) {
1685	>	if (binCount >= TREEIFY_THRESHOLD)
1686	>	treeifyBin(tab, i);
1687	>	if (!added)
1688	>	return val;
1689	>	break;
1690	>	}
1691	>	}
1692	>	}
1693	>	if (val != null)
1694	>	addCount(1L, binCount);
1695	>	return val;
1696	>	}
1697	>
1698	>	/**
1699	>	* If the value for the specified key is present, attempts to
1700	>	* compute a new mapping given the key and its current mapped
1701	>	* value.  The entire method invocation is performed atomically.
1702	>	* Some attempted update operations on this map by other threads
1703	>	* may be blocked while computation is in progress, so the
1704	>	* computation should be short and simple, and must not attempt to
1705	>	* update any other mappings of this map.
1706	>	*
1707	>	* @param key key with which a value may be associated
1708	>	* @param remappingFunction the function to compute a value
1709	>	* @return the new value associated with the specified key, or null if none
1710	>	* @throws NullPointerException if the specified key or remappingFunction
1711	>	*         is null
1712	>	* @throws IllegalStateException if the computation detectably
1713	>	*         attempts a recursive update to this map that would
1714	>	*         otherwise never complete
1715	>	* @throws RuntimeException or Error if the remappingFunction does so,
1716	>	*         in which case the mapping is unchanged
1717	>	*/
1718	>	public V computeIfPresent(K key, BiFun<? super K, ? super V, ? extends V> remappingFunction) {
1719	>	if (key == null || remappingFunction == null)
1720	>	throw new NullPointerException();
1721	>	int h = spread(key.hashCode());
1722	>	V val = null;
1723	>	int delta = 0;
1724	>	int binCount = 0;
1725	>	for (Node<K,V>[] tab = table;;) {
1726	>	Node<K,V> f; int n, i, fh;
1727	>	if (tab == null || (n = tab.length) == 0)
1728	>	tab = initTable();
1729	>	else if ((f = tabAt(tab, i = (n - 1) & h)) == null)
1730	>	break;
1731	>	else if ((fh = f.hash) == MOVED)
1732	>	tab = helpTransfer(tab, f);
1733	>	else {
1734	>	synchronized (f) {
1735	>	if (tabAt(tab, i) == f) {
1736	>	if (fh >= 0) {
1737	>	binCount = 1;
1738	>	for (Node<K,V> e = f, pred = null;; ++binCount) {
1739	>	K ek;
1740	>	if (e.hash == h &&
1741	>	((ek = e.key) == key ||
1742	>	(ek != null && key.equals(ek)))) {
1743	>	val = remappingFunction.apply(key, e.val);
1744	>	if (val != null)
1745	>	e.val = val;
1746	>	else {
1747	>	delta = -1;
1748	>	Node<K,V> en = e.next;
1749	>	if (pred != null)
1750	>	pred.next = en;
1751	>	else
1752	>	setTabAt(tab, i, en);
1753	>	}
1754	>	break;
1755	>	}
1756	>	pred = e;
1757	>	if ((e = e.next) == null)
1758	>	break;
1759	>	}
1760	>	}
1761	>	else if (f instanceof TreeBin) {
1762	>	binCount = 2;
1763	>	TreeBin<K,V> t = (TreeBin<K,V>)f;
1764	>	TreeNode<K,V> r, p;
1765	>	if ((r = t.root) != null &&
1766	>	(p = r.findTreeNode(h, key, null)) != null) {
1767	>	val = remappingFunction.apply(key, p.val);
1768	>	if (val != null)
1769	>	p.val = val;
1770	>	else {
1771	>	delta = -1;
1772	>	if (t.removeTreeNode(p))
1773	>	setTabAt(tab, i, untreeify(t.first));
1774	>	}
1775	>	}
1776	>	}
1777	>	}
1778	>	}
1779	>	if (binCount != 0)
1780	>	break;
1781	>	}
1782	>	}
1783	>	if (delta != 0)
1784	>	addCount((long)delta, binCount);
1785	>	return val;
1786	>	}
1787	>
1788	>	/**
1789	>	* Attempts to compute a mapping for the specified key and its
1790	>	* current mapped value (or {@code null} if there is no current
1791	>	* mapping). The entire method invocation is performed atomically.
1792	>	* Some attempted update operations on this map by other threads
1793	>	* may be blocked while computation is in progress, so the
1794	>	* computation should be short and simple, and must not attempt to
1795	>	* update any other mappings of this Map.
1796	>	*
1797	>	* @param key key with which the specified value is to be associated
1798	>	* @param remappingFunction the function to compute a value
1799	>	* @return the new value associated with the specified key, or null if none
1800	>	* @throws NullPointerException if the specified key or remappingFunction
1801	>	*         is null
1802	>	* @throws IllegalStateException if the computation detectably
1803	>	*         attempts a recursive update to this map that would
1804	>	*         otherwise never complete
1805	>	* @throws RuntimeException or Error if the remappingFunction does so,
1806	>	*         in which case the mapping is unchanged
1807	>	*/
1808	>	public V compute(K key,
1809	>	BiFun<? super K, ? super V, ? extends V> remappingFunction) {
1810	>	if (key == null || remappingFunction == null)
1811	>	throw new NullPointerException();
1812	>	int h = spread(key.hashCode());
1813	>	V val = null;
1814	>	int delta = 0;
1815	>	int binCount = 0;
1816	>	for (Node<K,V>[] tab = table;;) {
1817	>	Node<K,V> f; int n, i, fh;
1818	>	if (tab == null || (n = tab.length) == 0)
1819	>	tab = initTable();
1820	>	else if ((f = tabAt(tab, i = (n - 1) & h)) == null) {
1821	>	Node<K,V> r = new ReservationNode<K,V>();
1822	>	synchronized (r) {
1823	>	if (casTabAt(tab, i, null, r)) {
1824	>	binCount = 1;
1825	>	Node<K,V> node = null;
1826	>	try {
1827	>	if ((val = remappingFunction.apply(key, null)) != null) {
1828	>	delta = 1;
1829	>	node = new Node<K,V>(h, key, val, null);
1830	>	}
1831	>	} finally {
1832	>	setTabAt(tab, i, node);
1833	>	}
1834	>	}
1835	>	}
1836	>	if (binCount != 0)
1837	>	break;
1838	>	}
1839	>	else if ((fh = f.hash) == MOVED)
1840	>	tab = helpTransfer(tab, f);
1841	>	else {
1842	>	synchronized (f) {
1843	>	if (tabAt(tab, i) == f) {
1844	>	if (fh >= 0) {
1845	>	binCount = 1;
1846	>	for (Node<K,V> e = f, pred = null;; ++binCount) {
1847	>	K ek;
1848	>	if (e.hash == h &&
1849	>	((ek = e.key) == key ||
1850	>	(ek != null && key.equals(ek)))) {
1851	>	val = remappingFunction.apply(key, e.val);
1852	>	if (val != null)
1853	>	e.val = val;
1854	>	else {
1855	>	delta = -1;
1856	>	Node<K,V> en = e.next;
1857	>	if (pred != null)
1858	>	pred.next = en;
1859	>	else
1860	>	setTabAt(tab, i, en);
1861	>	}
1862	>	break;
1863	>	}
1864	>	pred = e;
1865	>	if ((e = e.next) == null) {
1866	>	val = remappingFunction.apply(key, null);
1867	>	if (val != null) {
1868	>	delta = 1;
1869	>	pred.next =
1870	>	new Node<K,V>(h, key, val, null);
1871	>	}
1872	>	break;
1873	>	}
1874	>	}
1875	>	}
1876	>	else if (f instanceof TreeBin) {
1877	>	binCount = 1;
1878	>	TreeBin<K,V> t = (TreeBin<K,V>)f;
1879	>	TreeNode<K,V> r, p;
1880	>	if ((r = t.root) != null)
1881	>	p = r.findTreeNode(h, key, null);
1882	>	else
1883	>	p = null;
1884	>	V pv = (p == null) ? null : p.val;
1885	>	val = remappingFunction.apply(key, pv);
1886	>	if (val != null) {
1887	>	if (p != null)
1888	>	p.val = val;
1889	>	else {
1890	>	delta = 1;
1891	>	t.putTreeVal(h, key, val);
1892	>	}
1893	>	}
1894	>	else if (p != null) {
1895	>	delta = -1;
1896	>	if (t.removeTreeNode(p))
1897	>	setTabAt(tab, i, untreeify(t.first));
1898	>	}
1899	>	}
1900	>	}
1901	>	}
1902	>	if (binCount != 0) {
1903	>	if (binCount >= TREEIFY_THRESHOLD)
1904	>	treeifyBin(tab, i);
1905	>	break;
1906	>	}
1907	>	}
1908	>	}
1909	>	if (delta != 0)
1910	>	addCount((long)delta, binCount);
1911	>	return val;
1912	>	}
1913
1914		/**
1915	<	* Base class for key, value, and entry iterators.  Adds a map
1916	<	* reference to InternalIterator to support Iterator.remove.
1915	>	* If the specified key is not already associated with a
1916	>	* (non-null) value, associates it with the given value.
1917	>	* Otherwise, replaces the value with the results of the given
1918	>	* remapping function, or removes if {@code null}. The entire
1919	>	* method invocation is performed atomically.  Some attempted
1920	>	* update operations on this map by other threads may be blocked
1921	>	* while computation is in progress, so the computation should be
1922	>	* short and simple, and must not attempt to update any other
1923	>	* mappings of this Map.
1924	>	*
1925	>	* @param key key with which the specified value is to be associated
1926	>	* @param value the value to use if absent
1927	>	* @param remappingFunction the function to recompute a value if present
1928	>	* @return the new value associated with the specified key, or null if none
1929	>	* @throws NullPointerException if the specified key or the
1930	>	*         remappingFunction is null
1931	>	* @throws RuntimeException or Error if the remappingFunction does so,
1932	>	*         in which case the mapping is unchanged
1933		*/
1934	<	static abstract class ViewIterator<K,V> extends InternalIterator {
1935	<	final ConcurrentHashMapV8<K, V> map;
1936	<	ViewIterator(ConcurrentHashMapV8<K, V> map) {
1937	<	super(map.table);
1934	>	public V merge(K key, V value, BiFun<? super V, ? super V, ? extends V> remappingFunction) {
1935	>	if (key == null || value == null || remappingFunction == null)
1936	>	throw new NullPointerException();
1937	>	int h = spread(key.hashCode());
1938	>	V val = null;
1939	>	int delta = 0;
1940	>	int binCount = 0;
1941	>	for (Node<K,V>[] tab = table;;) {
1942	>	Node<K,V> f; int n, i, fh;
1943	>	if (tab == null || (n = tab.length) == 0)
1944	>	tab = initTable();
1945	>	else if ((f = tabAt(tab, i = (n - 1) & h)) == null) {
1946	>	if (casTabAt(tab, i, null, new Node<K,V>(h, key, value, null))) {
1947	>	delta = 1;
1948	>	val = value;
1949	>	break;
1950	>	}
1951	>	}
1952	>	else if ((fh = f.hash) == MOVED)
1953	>	tab = helpTransfer(tab, f);
1954	>	else {
1955	>	synchronized (f) {
1956	>	if (tabAt(tab, i) == f) {
1957	>	if (fh >= 0) {
1958	>	binCount = 1;
1959	>	for (Node<K,V> e = f, pred = null;; ++binCount) {
1960	>	K ek;
1961	>	if (e.hash == h &&
1962	>	((ek = e.key) == key ||
1963	>	(ek != null && key.equals(ek)))) {
1964	>	val = remappingFunction.apply(e.val, value);
1965	>	if (val != null)
1966	>	e.val = val;
1967	>	else {
1968	>	delta = -1;
1969	>	Node<K,V> en = e.next;
1970	>	if (pred != null)
1971	>	pred.next = en;
1972	>	else
1973	>	setTabAt(tab, i, en);
1974	>	}
1975	>	break;
1976	>	}
1977	>	pred = e;
1978	>	if ((e = e.next) == null) {
1979	>	delta = 1;
1980	>	val = value;
1981	>	pred.next =
1982	>	new Node<K,V>(h, key, val, null);
1983	>	break;
1984	>	}
1985	>	}
1986	>	}
1987	>	else if (f instanceof TreeBin) {
1988	>	binCount = 2;
1989	>	TreeBin<K,V> t = (TreeBin<K,V>)f;
1990	>	TreeNode<K,V> r = t.root;
1991	>	TreeNode<K,V> p = (r == null) ? null :
1992	>	r.findTreeNode(h, key, null);
1993	>	val = (p == null) ? value :
1994	>	remappingFunction.apply(p.val, value);
1995	>	if (val != null) {
1996	>	if (p != null)
1997	>	p.val = val;
1998	>	else {
1999	>	delta = 1;
2000	>	t.putTreeVal(h, key, val);
2001	>	}
2002	>	}
2003	>	else if (p != null) {
2004	>	delta = -1;
2005	>	if (t.removeTreeNode(p))
2006	>	setTabAt(tab, i, untreeify(t.first));
2007	>	}
2008	>	}
2009	>	}
2010	>	}
2011	>	if (binCount != 0) {
2012	>	if (binCount >= TREEIFY_THRESHOLD)
2013	>	treeifyBin(tab, i);
2014	>	break;
2015	>	}
2016	>	}
2017	>	}
2018	>	if (delta != 0)
2019	>	addCount((long)delta, binCount);
2020	>	return val;
2021	>	}
2022	>
2023	>	// Hashtable legacy methods
2024	>
2025	>	/**
2026	>	* Legacy method testing if some key maps into the specified value
2027	>	* in this table.  This method is identical in functionality to
2028	>	* {@link #containsValue(Object)}, and exists solely to ensure
2029	>	* full compatibility with class {@link java.util.Hashtable},
2030	>	* which supported this method prior to introduction of the
2031	>	* Java Collections framework.
2032	>	*
2033	>	* @param  value a value to search for
2034	>	* @return {@code true} if and only if some key maps to the
2035	>	*         {@code value} argument in this table as
2036	>	*         determined by the {@code equals} method;
2037	>	*         {@code false} otherwise
2038	>	* @throws NullPointerException if the specified value is null
2039	>	*/
2040	>	@Deprecated public boolean contains(Object value) {
2041	>	return containsValue(value);
2042	>	}
2043	>
2044	>	/**
2045	>	* Returns an enumeration of the keys in this table.
2046	>	*
2047	>	* @return an enumeration of the keys in this table
2048	>	* @see #keySet()
2049	>	*/
2050	>	public Enumeration<K> keys() {
2051	>	Node<K,V>[] t;
2052	>	int f = (t = table) == null ? 0 : t.length;
2053	>	return new KeyIterator<K,V>(t, f, 0, f, this);
2054	>	}
2055	>
2056	>	/**
2057	>	* Returns an enumeration of the values in this table.
2058	>	*
2059	>	* @return an enumeration of the values in this table
2060	>	* @see #values()
2061	>	*/
2062	>	public Enumeration<V> elements() {
2063	>	Node<K,V>[] t;
2064	>	int f = (t = table) == null ? 0 : t.length;
2065	>	return new ValueIterator<K,V>(t, f, 0, f, this);
2066	>	}
2067	>
2068	>	// ConcurrentHashMapV8-only methods
2069	>
2070	>	/**
2071	>	* Returns the number of mappings. This method should be used
2072	>	* instead of {@link #size} because a ConcurrentHashMapV8 may
2073	>	* contain more mappings than can be represented as an int. The
2074	>	* value returned is an estimate; the actual count may differ if
2075	>	* there are concurrent insertions or removals.
2076	>	*
2077	>	* @return the number of mappings
2078	>	* @since 1.8
2079	>	*/
2080	>	public long mappingCount() {
2081	>	long n = sumCount();
2082	>	return (n < 0L) ? 0L : n; // ignore transient negative values
2083	>	}
2084	>
2085	>	/**
2086	>	* Creates a new {@link Set} backed by a ConcurrentHashMapV8
2087	>	* from the given type to {@code Boolean.TRUE}.
2088	>	*
2089	>	* @return the new set
2090	>	* @since 1.8
2091	>	*/
2092	>	public static <K> KeySetView<K,Boolean> newKeySet() {
2093	>	return new KeySetView<K,Boolean>
2094	>	(new ConcurrentHashMapV8<K,Boolean>(), Boolean.TRUE);
2095	>	}
2096	>
2097	>	/**
2098	>	* Creates a new {@link Set} backed by a ConcurrentHashMapV8
2099	>	* from the given type to {@code Boolean.TRUE}.
2100	>	*
2101	>	* @param initialCapacity The implementation performs internal
2102	>	* sizing to accommodate this many elements.
2103	>	* @throws IllegalArgumentException if the initial capacity of
2104	>	* elements is negative
2105	>	* @return the new set
2106	>	* @since 1.8
2107	>	*/
2108	>	public static <K> KeySetView<K,Boolean> newKeySet(int initialCapacity) {
2109	>	return new KeySetView<K,Boolean>
2110	>	(new ConcurrentHashMapV8<K,Boolean>(initialCapacity), Boolean.TRUE);
2111	>	}
2112	>
2113	>	/**
2114	>	* Returns a {@link Set} view of the keys in this map, using the
2115	>	* given common mapped value for any additions (i.e., {@link
2116	>	* Collection#add} and {@link Collection#addAll(Collection)}).
2117	>	* This is of course only appropriate if it is acceptable to use
2118	>	* the same value for all additions from this view.
2119	>	*
2120	>	* @param mappedValue the mapped value to use for any additions
2121	>	* @return the set view
2122	>	* @throws NullPointerException if the mappedValue is null
2123	>	*/
2124	>	public KeySetView<K,V> keySet(V mappedValue) {
2125	>	if (mappedValue == null)
2126	>	throw new NullPointerException();
2127	>	return new KeySetView<K,V>(this, mappedValue);
2128	>	}
2129	>
2130	>	/* ---------------- Special Nodes -------------- */
2131	>
2132	>	/**
2133	>	* A node inserted at head of bins during transfer operations.
2134	>	*/
2135	>	static final class ForwardingNode<K,V> extends Node<K,V> {
2136	>	final Node<K,V>[] nextTable;
2137	>	ForwardingNode(Node<K,V>[] tab) {
2138	>	super(MOVED, null, null, null);
2139	>	this.nextTable = tab;
2140	>	}
2141	>
2142	>	Node<K,V> find(int h, Object k) {
2143	>	Node<K,V> e; int n;
2144	>	Node<K,V>[] tab = nextTable;
2145	>	if (k != null && tab != null && (n = tab.length) > 0 &&
2146	>	(e = tabAt(tab, (n - 1) & h)) != null) {
2147	>	do {
2148	>	int eh; K ek;
2149	>	if ((eh = e.hash) == h &&
2150	>	((ek = e.key) == k || (ek != null && k.equals(ek))))
2151	>	return e;
2152	>	if (eh < 0)
2153	>	return e.find(h, k);
2154	>	} while ((e = e.next) != null);
2155	>	}
2156	>	return null;
2157	>	}
2158	>	}
2159	>
2160	>	/**
2161	>	* A place-holder node used in computeIfAbsent and compute
2162	>	*/
2163	>	static final class ReservationNode<K,V> extends Node<K,V> {
2164	>	ReservationNode() {
2165	>	super(RESERVED, null, null, null);
2166	>	}
2167	>
2168	>	Node<K,V> find(int h, Object k) {
2169	>	return null;
2170	>	}
2171	>	}
2172	>
2173	>	/* ---------------- Table Initialization and Resizing -------------- */
2174	>
2175	>	/**
2176	>	* Initializes table, using the size recorded in sizeCtl.
2177	>	*/
2178	>	private final Node<K,V>[] initTable() {
2179	>	Node<K,V>[] tab; int sc;
2180	>	while ((tab = table) == null || tab.length == 0) {
2181	>	if ((sc = sizeCtl) < 0)
2182	>	Thread.yield(); // lost initialization race; just spin
2183	>	else if (U.compareAndSwapInt(this, SIZECTL, sc, -1)) {
2184	>	try {
2185	>	if ((tab = table) == null || tab.length == 0) {
2186	>	int n = (sc > 0) ? sc : DEFAULT_CAPACITY;
2187	>	@SuppressWarnings({"rawtypes","unchecked"})
2188	>	Node<K,V>[] nt = (Node<K,V>[])new Node[n];
2189	>	table = tab = nt;
2190	>	sc = n - (n >>> 2);
2191	>	}
2192	>	} finally {
2193	>	sizeCtl = sc;
2194	>	}
2195	>	break;
2196	>	}
2197	>	}
2198	>	return tab;
2199	>	}
2200	>
2201	>	/**
2202	>	* Adds to count, and if table is too small and not already
2203	>	* resizing, initiates transfer. If already resizing, helps
2204	>	* perform transfer if work is available.  Rechecks occupancy
2205	>	* after a transfer to see if another resize is already needed
2206	>	* because resizings are lagging additions.
2207	>	*
2208	>	* @param x the count to add
2209	>	* @param check if <0, don't check resize, if <= 1 only check if uncontended
2210	>	*/
2211	>	private final void addCount(long x, int check) {
2212	>	CounterCell[] as; long b, s;
2213	>	if ((as = counterCells) != null ||
2214	>	!U.compareAndSwapLong(this, BASECOUNT, b = baseCount, s = b + x)) {
2215	>	CounterHashCode hc; CounterCell a; long v; int m;
2216	>	boolean uncontended = true;
2217	>	if ((hc = threadCounterHashCode.get()) == null ||
2218	>	as == null || (m = as.length - 1) < 0 ||
2219	>	(a = as[m & hc.code]) == null ||
2220	>	!(uncontended =
2221	>	U.compareAndSwapLong(a, CELLVALUE, v = a.value, v + x))) {
2222	>	fullAddCount(x, hc, uncontended);
2223	>	return;
2224	>	}
2225	>	if (check <= 1)
2226	>	return;
2227	>	s = sumCount();
2228	>	}
2229	>	if (check >= 0) {
2230	>	Node<K,V>[] tab, nt; int sc;
2231	>	while (s >= (long)(sc = sizeCtl) && (tab = table) != null &&
2232	>	tab.length < MAXIMUM_CAPACITY) {
2233	>	if (sc < 0) {
2234	>	if (sc == -1 || transferIndex <= transferOrigin ||
2235	>	(nt = nextTable) == null)
2236	>	break;
2237	>	if (U.compareAndSwapInt(this, SIZECTL, sc, sc - 1))
2238	>	transfer(tab, nt);
2239	>	}
2240	>	else if (U.compareAndSwapInt(this, SIZECTL, sc, -2))
2241	>	transfer(tab, null);
2242	>	s = sumCount();
2243	>	}
2244	>	}
2245	>	}
2246	>
2247	>	/**
2248	>	* Helps transfer if a resize is in progress.
2249	>	*/
2250	>	final Node<K,V>[] helpTransfer(Node<K,V>[] tab, Node<K,V> f) {
2251	>	Node<K,V>[] nextTab; int sc;
2252	>	if ((f instanceof ForwardingNode) &&
2253	>	(nextTab = ((ForwardingNode<K,V>)f).nextTable) != null) {
2254	>	if (nextTab == nextTable && tab == table &&
2255	>	transferIndex > transferOrigin && (sc = sizeCtl) < -1 &&
2256	>	U.compareAndSwapInt(this, SIZECTL, sc, sc - 1))
2257	>	transfer(tab, nextTab);
2258	>	return nextTab;
2259	>	}
2260	>	return table;
2261	>	}
2262	>
2263	>	/**
2264	>	* Tries to presize table to accommodate the given number of elements.
2265	>	*
2266	>	* @param size number of elements (doesn't need to be perfectly accurate)
2267	>	*/
2268	>	private final void tryPresize(int size) {
2269	>	int c = (size >= (MAXIMUM_CAPACITY >>> 1)) ? MAXIMUM_CAPACITY :
2270	>	tableSizeFor(size + (size >>> 1) + 1);
2271	>	int sc;
2272	>	while ((sc = sizeCtl) >= 0) {
2273	>	Node<K,V>[] tab = table; int n;
2274	>	if (tab == null || (n = tab.length) == 0) {
2275	>	n = (sc > c) ? sc : c;
2276	>	if (U.compareAndSwapInt(this, SIZECTL, sc, -1)) {
2277	>	try {
2278	>	if (table == tab) {
2279	>	@SuppressWarnings({"rawtypes","unchecked"})
2280	>	Node<K,V>[] nt = (Node<K,V>[])new Node[n];
2281	>	table = nt;
2282	>	sc = n - (n >>> 2);
2283	>	}
2284	>	} finally {
2285	>	sizeCtl = sc;
2286	>	}
2287	>	}
2288	>	}
2289	>	else if (c <= sc || n >= MAXIMUM_CAPACITY)
2290	>	break;
2291	>	else if (tab == table &&
2292	>	U.compareAndSwapInt(this, SIZECTL, sc, -2))
2293	>	transfer(tab, null);
2294	>	}
2295	>	}
2296	>
2297	>	/**
2298	>	* Moves and/or copies the nodes in each bin to new table. See
2299	>	* above for explanation.
2300	>	*/
2301	>	private final void transfer(Node<K,V>[] tab, Node<K,V>[] nextTab) {
2302	>	int n = tab.length, stride;
2303	>	if ((stride = (NCPU > 1) ? (n >>> 3) / NCPU : n) < MIN_TRANSFER_STRIDE)
2304	>	stride = MIN_TRANSFER_STRIDE; // subdivide range
2305	>	if (nextTab == null) {            // initiating
2306	>	try {
2307	>	@SuppressWarnings({"rawtypes","unchecked"})
2308	>	Node<K,V>[] nt = (Node<K,V>[])new Node[n << 1];
2309	>	nextTab = nt;
2310	>	} catch (Throwable ex) {      // try to cope with OOME
2311	>	sizeCtl = Integer.MAX_VALUE;
2312	>	return;
2313	>	}
2314	>	nextTable = nextTab;
2315	>	transferOrigin = n;
2316	>	transferIndex = n;
2317	>	ForwardingNode<K,V> rev = new ForwardingNode<K,V>(tab);
2318	>	for (int k = n; k > 0;) {    // progressively reveal ready slots
2319	>	int nextk = (k > stride) ? k - stride : 0;
2320	>	for (int m = nextk; m < k; ++m)
2321	>	nextTab[m] = rev;
2322	>	for (int m = n + nextk; m < n + k; ++m)
2323	>	nextTab[m] = rev;
2324	>	U.putOrderedInt(this, TRANSFERORIGIN, k = nextk);
2325	>	}
2326	>	}
2327	>	int nextn = nextTab.length;
2328	>	ForwardingNode<K,V> fwd = new ForwardingNode<K,V>(nextTab);
2329	>	boolean advance = true;
2330	>	for (int i = 0, bound = 0;;) {
2331	>	int nextIndex, nextBound, fh; Node<K,V> f;
2332	>	while (advance) {
2333	>	if (--i >= bound)
2334	>	advance = false;
2335	>	else if ((nextIndex = transferIndex) <= transferOrigin) {
2336	>	i = -1;
2337	>	advance = false;
2338	>	}
2339	>	else if (U.compareAndSwapInt
2340	>	(this, TRANSFERINDEX, nextIndex,
2341	>	nextBound = (nextIndex > stride ?
2342	>	nextIndex - stride : 0))) {
2343	>	bound = nextBound;
2344	>	i = nextIndex - 1;
2345	>	advance = false;
2346	>	}
2347	>	}
2348	>	if (i < 0 || i >= n || i + n >= nextn) {
2349	>	for (int sc;;) {
2350	>	if (U.compareAndSwapInt(this, SIZECTL, sc = sizeCtl, ++sc)) {
2351	>	if (sc == -1) {
2352	>	nextTable = null;
2353	>	table = nextTab;
2354	>	sizeCtl = (n << 1) - (n >>> 1);
2355	>	}
2356	>	return;
2357	>	}
2358	>	}
2359	>	}
2360	>	else if ((f = tabAt(tab, i)) == null) {
2361	>	if (casTabAt(tab, i, null, fwd)) {
2362	>	setTabAt(nextTab, i, null);
2363	>	setTabAt(nextTab, i + n, null);
2364	>	advance = true;
2365	>	}
2366	>	}
2367	>	else if ((fh = f.hash) == MOVED)
2368	>	advance = true; // already processed
2369	>	else {
2370	>	synchronized (f) {
2371	>	if (tabAt(tab, i) == f) {
2372	>	Node<K,V> ln, hn;
2373	>	if (fh >= 0) {
2374	>	int runBit = fh & n;
2375	>	Node<K,V> lastRun = f;
2376	>	for (Node<K,V> p = f.next; p != null; p = p.next) {
2377	>	int b = p.hash & n;
2378	>	if (b != runBit) {
2379	>	runBit = b;
2380	>	lastRun = p;
2381	>	}
2382	>	}
2383	>	if (runBit == 0) {
2384	>	ln = lastRun;
2385	>	hn = null;
2386	>	}
2387	>	else {
2388	>	hn = lastRun;
2389	>	ln = null;
2390	>	}
2391	>	for (Node<K,V> p = f; p != lastRun; p = p.next) {
2392	>	int ph = p.hash; K pk = p.key; V pv = p.val;
2393	>	if ((ph & n) == 0)
2394	>	ln = new Node<K,V>(ph, pk, pv, ln);
2395	>	else
2396	>	hn = new Node<K,V>(ph, pk, pv, hn);
2397	>	}
2398	>	}
2399	>	else if (f instanceof TreeBin) {
2400	>	TreeBin<K,V> t = (TreeBin<K,V>)f;
2401	>	TreeNode<K,V> lo = null, loTail = null;
2402	>	TreeNode<K,V> hi = null, hiTail = null;
2403	>	int lc = 0, hc = 0;
2404	>	for (Node<K,V> e = t.first; e != null; e = e.next) {
2405	>	int h = e.hash;
2406	>	TreeNode<K,V> p = new TreeNode<K,V>
2407	>	(h, e.key, e.val, null, null);
2408	>	if ((h & n) == 0) {
2409	>	if ((p.prev = loTail) == null)
2410	>	lo = p;
2411	>	else
2412	>	loTail.next = p;
2413	>	loTail = p;
2414	>	++lc;
2415	>	}
2416	>	else {
2417	>	if ((p.prev = hiTail) == null)
2418	>	hi = p;
2419	>	else
2420	>	hiTail.next = p;
2421	>	hiTail = p;
2422	>	++hc;
2423	>	}
2424	>	}
2425	>	ln = (lc <= UNTREEIFY_THRESHOLD) ? untreeify(lo) :
2426	>	(hc != 0) ? new TreeBin<K,V>(lo) : t;
2427	>	hn = (hc <= UNTREEIFY_THRESHOLD) ? untreeify(hi) :
2428	>	(lc != 0) ? new TreeBin<K,V>(hi) : t;
2429	>	}
2430	>	else
2431	>	ln = hn = null;
2432	>	setTabAt(nextTab, i, ln);
2433	>	setTabAt(nextTab, i + n, hn);
2434	>	setTabAt(tab, i, fwd);
2435	>	advance = true;
2436	>	}
2437	>	}
2438	>	}
2439	>	}
2440	>	}
2441	>
2442	>	/* ---------------- Conversion from/to TreeBins -------------- */
2443	>
2444	>	/**
2445	>	* Replaces all linked nodes in bin at given index unless table is
2446	>	* too small, in which case resizes instead.
2447	>	*/
2448	>	private final void treeifyBin(Node<K,V>[] tab, int index) {
2449	>	Node<K,V> b; int n, sc;
2450	>	if (tab != null) {
2451	>	if ((n = tab.length) < MIN_TREEIFY_CAPACITY) {
2452	>	if (tab == table && (sc = sizeCtl) >= 0 &&
2453	>	U.compareAndSwapInt(this, SIZECTL, sc, -2))
2454	>	transfer(tab, null);
2455	>	}
2456	>	else if ((b = tabAt(tab, index)) != null) {
2457	>	synchronized (b) {
2458	>	if (tabAt(tab, index) == b) {
2459	>	TreeNode<K,V> hd = null, tl = null;
2460	>	for (Node<K,V> e = b; e != null; e = e.next) {
2461	>	TreeNode<K,V> p =
2462	>	new TreeNode<K,V>(e.hash, e.key, e.val,
2463	>	null, null);
2464	>	if ((p.prev = tl) == null)
2465	>	hd = p;
2466	>	else
2467	>	tl.next = p;
2468	>	tl = p;
2469	>	}
2470	>	setTabAt(tab, index, new TreeBin<K,V>(hd));
2471	>	}
2472	>	}
2473	>	}
2474	>	}
2475	>	}
2476	>
2477	>	/**
2478	>	* Returns a list on non-TreeNodes replacing those in given list.
2479	>	*/
2480	>	static <K,V> Node<K,V> untreeify(Node<K,V> b) {
2481	>	Node<K,V> hd = null, tl = null;
2482	>	for (Node<K,V> q = b; q != null; q = q.next) {
2483	>	Node<K,V> p = new Node<K,V>(q.hash, q.key, q.val, null);
2484	>	if (tl == null)
2485	>	hd = p;
2486	>	else
2487	>	tl.next = p;
2488	>	tl = p;
2489	>	}
2490	>	return hd;
2491	>	}
2492	>
2493	>	/* ---------------- TreeNodes -------------- */
2494	>
2495	>	/**
2496	>	* Nodes for use in TreeBins
2497	>	*/
2498	>	static final class TreeNode<K,V> extends Node<K,V> {
2499	>	TreeNode<K,V> parent;  // red-black tree links
2500	>	TreeNode<K,V> left;
2501	>	TreeNode<K,V> right;
2502	>	TreeNode<K,V> prev;    // needed to unlink next upon deletion
2503	>	boolean red;
2504	>
2505	>	TreeNode(int hash, K key, V val, Node<K,V> next,
2506	>	TreeNode<K,V> parent) {
2507	>	super(hash, key, val, next);
2508	>	this.parent = parent;
2509	>	}
2510	>
2511	>	Node<K,V> find(int h, Object k) {
2512	>	return findTreeNode(h, k, null);
2513	>	}
2514	>
2515	>	/**
2516	>	* Returns the TreeNode (or null if not found) for the given key
2517	>	* starting at given root.
2518	>	*/
2519	>	final TreeNode<K,V> findTreeNode(int h, Object k, Class<?> kc) {
2520	>	if (k != null) {
2521	>	TreeNode<K,V> p = this;
2522	>	do  {
2523	>	int ph, dir; K pk; TreeNode<K,V> q;
2524	>	TreeNode<K,V> pl = p.left, pr = p.right;
2525	>	if ((ph = p.hash) > h)
2526	>	p = pl;
2527	>	else if (ph < h)
2528	>	p = pr;
2529	>	else if ((pk = p.key) == k || (pk != null && k.equals(pk)))
2530	>	return p;
2531	>	else if (pl == null && pr == null)
2532	>	break;
2533	>	else if ((kc != null ||
2534	>	(kc = comparableClassFor(k)) != null) &&
2535	>	(dir = compareComparables(kc, k, pk)) != 0)
2536	>	p = (dir < 0) ? pl : pr;
2537	>	else if (pl == null)
2538	>	p = pr;
2539	>	else if (pr == null ||
2540	>	(q = pr.findTreeNode(h, k, kc)) == null)
2541	>	p = pl;
2542	>	else
2543	>	return q;
2544	>	} while (p != null);
2545	>	}
2546	>	return null;
2547	>	}
2548	>	}
2549	>
2550	>	/* ---------------- TreeBins -------------- */
2551	>
2552	>	/**
2553	>	* TreeNodes used at the heads of bins. TreeBins do not hold user
2554	>	* keys or values, but instead point to list of TreeNodes and
2555	>	* their root. They also maintain a parasitic read-write lock
2556	>	* forcing writers (who hold bin lock) to wait for readers (who do
2557	>	* not) to complete before tree restructuring operations.
2558	>	*/
2559	>	static final class TreeBin<K,V> extends Node<K,V> {
2560	>	TreeNode<K,V> root;
2561	>	volatile TreeNode<K,V> first;
2562	>	volatile Thread waiter;
2563	>	volatile int lockState;
2564	>	// values for lockState
2565	>	static final int WRITER = 1; // set while holding write lock
2566	>	static final int WAITER = 2; // set when waiting for write lock
2567	>	static final int READER = 4; // increment value for setting read lock
2568	>
2569	>	/**
2570	>	* Creates bin with initial set of nodes headed by b.
2571	>	*/
2572	>	TreeBin(TreeNode<K,V> b) {
2573	>	super(TREEBIN, null, null, null);
2574	>	this.first = b;
2575	>	TreeNode<K,V> r = null;
2576	>	for (TreeNode<K,V> x = b, next; x != null; x = next) {
2577	>	next = (TreeNode<K,V>)x.next;
2578	>	x.left = x.right = null;
2579	>	if (r == null) {
2580	>	x.parent = null;
2581	>	x.red = false;
2582	>	r = x;
2583	>	}
2584	>	else {
2585	>	Object key = x.key;
2586	>	int hash = x.hash;
2587	>	Class<?> kc = null;
2588	>	for (TreeNode<K,V> p = r;;) {
2589	>	int dir, ph;
2590	>	if ((ph = p.hash) > hash)
2591	>	dir = -1;
2592	>	else if (ph < hash)
2593	>	dir = 1;
2594	>	else if ((kc != null ||
2595	>	(kc = comparableClassFor(key)) != null))
2596	>	dir = compareComparables(kc, key, p.key);
2597	>	else
2598	>	dir = 0;
2599	>	TreeNode<K,V> xp = p;
2600	>	if ((p = (dir <= 0) ? p.left : p.right) == null) {
2601	>	x.parent = xp;
2602	>	if (dir <= 0)
2603	>	xp.left = x;
2604	>	else
2605	>	xp.right = x;
2606	>	r = balanceInsertion(r, x);
2607	>	break;
2608	>	}
2609	>	}
2610	>	}
2611	>	}
2612	>	this.root = r;
2613	>	}
2614	>
2615	>	/**
2616	>	* Acquires write lock for tree restructuring.
2617	>	*/
2618	>	private final void lockRoot() {
2619	>	if (!U.compareAndSwapInt(this, LOCKSTATE, 0, WRITER))
2620	>	contendedLock(); // offload to separate method
2621	>	}
2622	>
2623	>	/**
2624	>	* Releases write lock for tree restructuring.
2625	>	*/
2626	>	private final void unlockRoot() {
2627	>	lockState = 0;
2628	>	}
2629	>
2630	>	/**
2631	>	* Possibly blocks awaiting root lock.
2632	>	*/
2633	>	private final void contendedLock() {
2634	>	boolean waiting = false;
2635	>	for (int s;;) {
2636	>	if (((s = lockState) & WRITER) == 0) {
2637	>	if (U.compareAndSwapInt(this, LOCKSTATE, s, WRITER)) {
2638	>	if (waiting)
2639	>	waiter = null;
2640	>	return;
2641	>	}
2642	>	}
2643	>	else if ((s | WAITER) == 0) {
2644	>	if (U.compareAndSwapInt(this, LOCKSTATE, s, s | WAITER)) {
2645	>	waiting = true;
2646	>	waiter = Thread.currentThread();
2647	>	}
2648	>	}
2649	>	else if (waiting)
2650	>	LockSupport.park(this);
2651	>	}
2652	>	}
2653	>
2654	>	/**
2655	>	* Returns matching node or null if none. Tries to search
2656	>	* using tree compareisons from root, but continues linear
2657	>	* search when lock not available.
2658	>	*/
2659	>	final Node<K,V> find(int h, Object k) {
2660	>	if (k != null) {
2661	>	for (Node<K,V> e = first; e != null; e = e.next) {
2662	>	int s; K ek;
2663	>	if (((s = lockState) & (WAITER|WRITER)) != 0) {
2664	>	if (e.hash == h &&
2665	>	((ek = e.key) == k || (ek != null && k.equals(ek))))
2666	>	return e;
2667	>	}
2668	>	else if (U.compareAndSwapInt(this, LOCKSTATE, s,
2669	>	s + READER)) {
2670	>	TreeNode<K,V> r, p;
2671	>	try {
2672	>	p = ((r = root) == null ? null :
2673	>	r.findTreeNode(h, k, null));
2674	>	} finally {
2675	>	Thread w;
2676	>	int ls;
2677	>	do {} while (!U.compareAndSwapInt
2678	>	(this, LOCKSTATE,
2679	>	ls = lockState, ls - READER));
2680	>	if (ls == (READER|WAITER) && (w = waiter) != null)
2681	>	LockSupport.unpark(w);
2682	>	}
2683	>	return p;
2684	>	}
2685	>	}
2686	>	}
2687	>	return null;
2688	>	}
2689	>
2690	>	/**
2691	>	* Finds or adds a node.
2692	>	* @return null if added
2693	>	*/
2694	>	final TreeNode<K,V> putTreeVal(int h, K k, V v) {
2695	>	Class<?> kc = null;
2696	>	for (TreeNode<K,V> p = root;;) {
2697	>	int dir, ph; K pk; TreeNode<K,V> q, pr;
2698	>	if (p == null) {
2699	>	first = root = new TreeNode<K,V>(h, k, v, null, null);
2700	>	break;
2701	>	}
2702	>	else if ((ph = p.hash) > h)
2703	>	dir = -1;
2704	>	else if (ph < h)
2705	>	dir = 1;
2706	>	else if ((pk = p.key) == k || (pk != null && k.equals(pk)))
2707	>	return p;
2708	>	else if ((kc == null &&
2709	>	(kc = comparableClassFor(k)) == null) ||
2710	>	(dir = compareComparables(kc, k, pk)) == 0) {
2711	>	if (p.left == null)
2712	>	dir = 1;
2713	>	else if ((pr = p.right) == null ||
2714	>	(q = pr.findTreeNode(h, k, kc)) == null)
2715	>	dir = -1;
2716	>	else
2717	>	return q;
2718	>	}
2719	>	TreeNode<K,V> xp = p;
2720	>	if ((p = (dir < 0) ? p.left : p.right) == null) {
2721	>	TreeNode<K,V> x, f = first;
2722	>	first = x = new TreeNode<K,V>(h, k, v, f, xp);
2723	>	if (f != null)
2724	>	f.prev = x;
2725	>	if (dir < 0)
2726	>	xp.left = x;
2727	>	else
2728	>	xp.right = x;
2729	>	if (!xp.red)
2730	>	x.red = true;
2731	>	else {
2732	>	lockRoot();
2733	>	try {
2734	>	root = balanceInsertion(root, x);
2735	>	} finally {
2736	>	unlockRoot();
2737	>	}
2738	>	}
2739	>	break;
2740	>	}
2741	>	}
2742	>	assert checkInvariants(root);
2743	>	return null;
2744	>	}
2745	>
2746	>	/**
2747	>	* Removes the given node, that must be present before this
2748	>	* call.  This is messier than typical red-black deletion code
2749	>	* because we cannot swap the contents of an interior node
2750	>	* with a leaf successor that is pinned by "next" pointers
2751	>	* that are accessible independently of lock. So instead we
2752	>	* swap the tree linkages.
2753	>	*
2754	>	* @return true if now too small, so should be untreeified
2755	>	*/
2756	>	final boolean removeTreeNode(TreeNode<K,V> p) {
2757	>	TreeNode<K,V> next = (TreeNode<K,V>)p.next;
2758	>	TreeNode<K,V> pred = p.prev;  // unlink traversal pointers
2759	>	TreeNode<K,V> r, rl;
2760	>	if (pred == null)
2761	>	first = next;
2762	>	else
2763	>	pred.next = next;
2764	>	if (next != null)
2765	>	next.prev = pred;
2766	>	if (first == null) {
2767	>	root = null;
2768	>	return true;
2769	>	}
2770	>	if ((r = root) == null || r.right == null || // too small
2771	>	(rl = r.left) == null || rl.left == null)
2772	>	return true;
2773	>	lockRoot();
2774	>	try {
2775	>	TreeNode<K,V> replacement;
2776	>	TreeNode<K,V> pl = p.left;
2777	>	TreeNode<K,V> pr = p.right;
2778	>	if (pl != null && pr != null) {
2779	>	TreeNode<K,V> s = pr, sl;
2780	>	while ((sl = s.left) != null) // find successor
2781	>	s = sl;
2782	>	boolean c = s.red; s.red = p.red; p.red = c; // swap colors
2783	>	TreeNode<K,V> sr = s.right;
2784	>	TreeNode<K,V> pp = p.parent;
2785	>	if (s == pr) { // p was s's direct parent
2786	>	p.parent = s;
2787	>	s.right = p;
2788	>	}
2789	>	else {
2790	>	TreeNode<K,V> sp = s.parent;
2791	>	if ((p.parent = sp) != null) {
2792	>	if (s == sp.left)
2793	>	sp.left = p;
2794	>	else
2795	>	sp.right = p;
2796	>	}
2797	>	if ((s.right = pr) != null)
2798	>	pr.parent = s;
2799	>	}
2800	>	p.left = null;
2801	>	if ((p.right = sr) != null)
2802	>	sr.parent = p;
2803	>	if ((s.left = pl) != null)
2804	>	pl.parent = s;
2805	>	if ((s.parent = pp) == null)
2806	>	r = s;
2807	>	else if (p == pp.left)
2808	>	pp.left = s;
2809	>	else
2810	>	pp.right = s;
2811	>	if (sr != null)
2812	>	replacement = sr;
2813	>	else
2814	>	replacement = p;
2815	>	}
2816	>	else if (pl != null)
2817	>	replacement = pl;
2818	>	else if (pr != null)
2819	>	replacement = pr;
2820	>	else
2821	>	replacement = p;
2822	>	if (replacement != p) {
2823	>	TreeNode<K,V> pp = replacement.parent = p.parent;
2824	>	if (pp == null)
2825	>	r = replacement;
2826	>	else if (p == pp.left)
2827	>	pp.left = replacement;
2828	>	else
2829	>	pp.right = replacement;
2830	>	p.left = p.right = p.parent = null;
2831	>	}
2832	>
2833	>	root = (p.red) ? r : balanceDeletion(r, replacement);
2834	>
2835	>	if (p == replacement) {  // detach pointers
2836	>	TreeNode<K,V> pp;
2837	>	if ((pp = p.parent) != null) {
2838	>	if (p == pp.left)
2839	>	pp.left = null;
2840	>	else if (p == pp.right)
2841	>	pp.right = null;
2842	>	p.parent = null;
2843	>	}
2844	>	}
2845	>	} finally {
2846	>	unlockRoot();
2847	>	}
2848	>	assert checkInvariants(root);
2849	>	return false;
2850	>	}
2851	>
2852	>	/* ------------------------------------------------------------ */
2853	>	// Red-black tree methods, all adapted from CLR
2854	>
2855	>	static <K,V> TreeNode<K,V> rotateLeft(TreeNode<K,V> root,
2856	>	TreeNode<K,V> p) {
2857	>	TreeNode<K,V> r, pp, rl;
2858	>	if (p != null && (r = p.right) != null) {
2859	>	if ((rl = p.right = r.left) != null)
2860	>	rl.parent = p;
2861	>	if ((pp = r.parent = p.parent) == null)
2862	>	(root = r).red = false;
2863	>	else if (pp.left == p)
2864	>	pp.left = r;
2865	>	else
2866	>	pp.right = r;
2867	>	r.left = p;
2868	>	p.parent = r;
2869	>	}
2870	>	return root;
2871	>	}
2872	>
2873	>	static <K,V> TreeNode<K,V> rotateRight(TreeNode<K,V> root,
2874	>	TreeNode<K,V> p) {
2875	>	TreeNode<K,V> l, pp, lr;
2876	>	if (p != null && (l = p.left) != null) {
2877	>	if ((lr = p.left = l.right) != null)
2878	>	lr.parent = p;
2879	>	if ((pp = l.parent = p.parent) == null)
2880	>	(root = l).red = false;
2881	>	else if (pp.right == p)
2882	>	pp.right = l;
2883	>	else
2884	>	pp.left = l;
2885	>	l.right = p;
2886	>	p.parent = l;
2887	>	}
2888	>	return root;
2889	>	}
2890	>
2891	>	static <K,V> TreeNode<K,V> balanceInsertion(TreeNode<K,V> root,
2892	>	TreeNode<K,V> x) {
2893	>	x.red = true;
2894	>	for (TreeNode<K,V> xp, xpp, xppl, xppr;;) {
2895	>	if ((xp = x.parent) == null) {
2896	>	x.red = false;
2897	>	return x;
2898	>	}
2899	>	else if (!xp.red || (xpp = xp.parent) == null)
2900	>	return root;
2901	>	if (xp == (xppl = xpp.left)) {
2902	>	if ((xppr = xpp.right) != null && xppr.red) {
2903	>	xppr.red = false;
2904	>	xp.red = false;
2905	>	xpp.red = true;
2906	>	x = xpp;
2907	>	}
2908	>	else {
2909	>	if (x == xp.right) {
2910	>	root = rotateLeft(root, x = xp);
2911	>	xpp = (xp = x.parent) == null ? null : xp.parent;
2912	>	}
2913	>	if (xp != null) {
2914	>	xp.red = false;
2915	>	if (xpp != null) {
2916	>	xpp.red = true;
2917	>	root = rotateRight(root, xpp);
2918	>	}
2919	>	}
2920	>	}
2921	>	}
2922	>	else {
2923	>	if (xppl != null && xppl.red) {
2924	>	xppl.red = false;
2925	>	xp.red = false;
2926	>	xpp.red = true;
2927	>	x = xpp;
2928	>	}
2929	>	else {
2930	>	if (x == xp.left) {
2931	>	root = rotateRight(root, x = xp);
2932	>	xpp = (xp = x.parent) == null ? null : xp.parent;
2933	>	}
2934	>	if (xp != null) {
2935	>	xp.red = false;
2936	>	if (xpp != null) {
2937	>	xpp.red = true;
2938	>	root = rotateLeft(root, xpp);
2939	>	}
2940	>	}
2941	>	}
2942	>	}
2943	>	}
2944	>	}
2945	>
2946	>	static <K,V> TreeNode<K,V> balanceDeletion(TreeNode<K,V> root,
2947	>	TreeNode<K,V> x) {
2948	>	for (TreeNode<K,V> xp, xpl, xpr;;)  {
2949	>	if (x == null || x == root)
2950	>	return root;
2951	>	else if ((xp = x.parent) == null) {
2952	>	x.red = false;
2953	>	return x;
2954	>	}
2955	>	else if (x.red) {
2956	>	x.red = false;
2957	>	return root;
2958	>	}
2959	>	else if ((xpl = xp.left) == x) {
2960	>	if ((xpr = xp.right) != null && xpr.red) {
2961	>	xpr.red = false;
2962	>	xp.red = true;
2963	>	root = rotateLeft(root, xp);
2964	>	xpr = (xp = x.parent) == null ? null : xp.right;
2965	>	}
2966	>	if (xpr == null)
2967	>	x = xp;
2968	>	else {
2969	>	TreeNode<K,V> sl = xpr.left, sr = xpr.right;
2970	>	if ((sr == null || !sr.red) &&
2971	>	(sl == null || !sl.red)) {
2972	>	xpr.red = true;
2973	>	x = xp;
2974	>	}
2975	>	else {
2976	>	if (sr == null || !sr.red) {
2977	>	if (sl != null)
2978	>	sl.red = false;
2979	>	xpr.red = true;
2980	>	root = rotateRight(root, xpr);
2981	>	xpr = (xp = x.parent) == null ?
2982	>	null : xp.right;
2983	>	}
2984	>	if (xpr != null) {
2985	>	xpr.red = (xp == null) ? false : xp.red;
2986	>	if ((sr = xpr.right) != null)
2987	>	sr.red = false;
2988	>	}
2989	>	if (xp != null) {
2990	>	xp.red = false;
2991	>	root = rotateLeft(root, xp);
2992	>	}
2993	>	x = root;
2994	>	}
2995	>	}
2996	>	}
2997	>	else { // symmetric
2998	>	if (xpl != null && xpl.red) {
2999	>	xpl.red = false;
3000	>	xp.red = true;
3001	>	root = rotateRight(root, xp);
3002	>	xpl = (xp = x.parent) == null ? null : xp.left;
3003	>	}
3004	>	if (xpl == null)
3005	>	x = xp;
3006	>	else {
3007	>	TreeNode<K,V> sl = xpl.left, sr = xpl.right;
3008	>	if ((sl == null || !sl.red) &&
3009	>	(sr == null || !sr.red)) {
3010	>	xpl.red = true;
3011	>	x = xp;
3012	>	}
3013	>	else {
3014	>	if (sl == null || !sl.red) {
3015	>	if (sr != null)
3016	>	sr.red = false;
3017	>	xpl.red = true;
3018	>	root = rotateLeft(root, xpl);
3019	>	xpl = (xp = x.parent) == null ?
3020	>	null : xp.left;
3021	>	}
3022	>	if (xpl != null) {
3023	>	xpl.red = (xp == null) ? false : xp.red;
3024	>	if ((sl = xpl.left) != null)
3025	>	sl.red = false;
3026	>	}
3027	>	if (xp != null) {
3028	>	xp.red = false;
3029	>	root = rotateRight(root, xp);
3030	>	}
3031	>	x = root;
3032	>	}
3033	>	}
3034	>	}
3035	>	}
3036	>	}
3037	>
3038	>	/**
3039	>	* Recursive invariant check
3040	>	*/
3041	>	static <K,V> boolean checkInvariants(TreeNode<K,V> t) {
3042	>	TreeNode<K,V> tp = t.parent, tl = t.left, tr = t.right,
3043	>	tb = t.prev, tn = (TreeNode<K,V>)t.next;
3044	>	if (tb != null && tb.next != t)
3045	>	return false;
3046	>	if (tn != null && tn.prev != t)
3047	>	return false;
3048	>	if (tp != null && t != tp.left && t != tp.right)
3049	>	return false;
3050	>	if (tl != null && (tl.parent != t || tl.hash > t.hash))
3051	>	return false;
3052	>	if (tr != null && (tr.parent != t || tr.hash < t.hash))
3053	>	return false;
3054	>	if (t.red && tl != null && tl.red && tr != null && tr.red)
3055	>	return false;
3056	>	if (tl != null && !checkInvariants(tl))
3057	>	return false;
3058	>	if (tr != null && !checkInvariants(tr))
3059	>	return false;
3060	>	return true;
3061	>	}
3062	>
3063	>	private static final sun.misc.Unsafe U;
3064	>	private static final long LOCKSTATE;
3065	>	static {
3066	>	try {
3067	>	U = getUnsafe();
3068	>	Class<?> k = TreeBin.class;
3069	>	LOCKSTATE = U.objectFieldOffset
3070	>	(k.getDeclaredField("lockState"));
3071	>	} catch (Exception e) {
3072	>	throw new Error(e);
3073	>	}
3074	>	}
3075	>	}
3076	>
3077	>	/* ----------------Table Traversal -------------- */
3078	>
3079	>	/**
3080	>	* Encapsulates traversal for methods such as containsValue; also
3081	>	* serves as a base class for other iterators and spliterators.
3082	>	*
3083	>	* Method advance visits once each still-valid node that was
3084	>	* reachable upon iterator construction. It might miss some that
3085	>	* were added to a bin after the bin was visited, which is OK wrt
3086	>	* consistency guarantees. Maintaining this property in the face
3087	>	* of possible ongoing resizes requires a fair amount of
3088	>	* bookkeeping state that is difficult to optimize away amidst
3089	>	* volatile accesses.  Even so, traversal maintains reasonable
3090	>	* throughput.
3091	>	*
3092	>	* Normally, iteration proceeds bin-by-bin traversing lists.
3093	>	* However, if the table has been resized, then all future steps
3094	>	* must traverse both the bin at the current index as well as at
3095	>	* (index + baseSize); and so on for further resizings. To
3096	>	* paranoically cope with potential sharing by users of iterators
3097	>	* across threads, iteration terminates if a bounds checks fails
3098	>	* for a table read.
3099	>	*/
3100	>	static class Traverser<K,V> {
3101	>	Node<K,V>[] tab;        // current table; updated if resized
3102	>	Node<K,V> next;         // the next entry to use
3103	>	int index;              // index of bin to use next
3104	>	int baseIndex;          // current index of initial table
3105	>	int baseLimit;          // index bound for initial table
3106	>	final int baseSize;     // initial table size
3107	>
3108	>	Traverser(Node<K,V>[] tab, int size, int index, int limit) {
3109	>	this.tab = tab;
3110	>	this.baseSize = size;
3111	>	this.baseIndex = this.index = index;
3112	>	this.baseLimit = limit;
3113	>	this.next = null;
3114	>	}
3115	>
3116	>	/**
3117	>	* Advances if possible, returning next valid node, or null if none.
3118	>	*/
3119	>	final Node<K,V> advance() {
3120	>	Node<K,V> e;
3121	>	if ((e = next) != null)
3122	>	e = e.next;
3123	>	for (;;) {
3124	>	Node<K,V>[] t; int i, n; K ek;  // must use locals in checks
3125	>	if (e != null)
3126	>	return next = e;
3127	>	if (baseIndex >= baseLimit || (t = tab) == null ||
3128	>	(n = t.length) <= (i = index) || i < 0)
3129	>	return next = null;
3130	>	if ((e = tabAt(t, index)) != null && e.hash < 0) {
3131	>	if (e instanceof ForwardingNode) {
3132	>	tab = ((ForwardingNode<K,V>)e).nextTable;
3133	>	e = null;
3134	>	continue;
3135	>	}
3136	>	else if (e instanceof TreeBin)
3137	>	e = ((TreeBin<K,V>)e).first;
3138	>	else
3139	>	e = null;
3140	>	}
3141	>	if ((index += baseSize) >= n)
3142	>	index = ++baseIndex;    // visit upper slots if present
3143	>	}
3144	>	}
3145	>	}
3146	>
3147	>	/**
3148	>	* Base of key, value, and entry Iterators. Adds fields to
3149	>	* Traverser to support iterator.remove.
3150	>	*/
3151	>	static class BaseIterator<K,V> extends Traverser<K,V> {
3152	>	final ConcurrentHashMapV8<K,V> map;
3153	>	Node<K,V> lastReturned;
3154	>	BaseIterator(Node<K,V>[] tab, int size, int index, int limit,
3155	>	ConcurrentHashMapV8<K,V> map) {
3156	>	super(tab, size, index, limit);
3157		this.map = map;
3158	+	advance();
3159		}
3160
3161	+	public final boolean hasNext() { return next != null; }
3162	+	public final boolean hasMoreElements() { return next != null; }
3163	+
3164		public final void remove() {
3165	<	if (last == null)
3165	>	Node<K,V> p;
3166	>	if ((p = lastReturned) == null)
3167		throw new IllegalStateException();
3168	<	map.remove(last.key);
3169	<	last = null;
3168	>	lastReturned = null;
3169	>	map.replaceNode(p.key, null, null);
3170		}
1947	–
1948	–	public final boolean hasNext()         { return next != null; }
1949	–	public final boolean hasMoreElements() { return next != null; }
3171		}
3172
3173	<	static final class KeyIterator<K,V> extends ViewIterator<K,V>
3173	>	static final class KeyIterator<K,V> extends BaseIterator<K,V>
3174		implements Iterator<K>, Enumeration<K> {
3175	<	KeyIterator(ConcurrentHashMapV8<K, V> map) { super(map); }
3175	>	KeyIterator(Node<K,V>[] tab, int index, int size, int limit,
3176	>	ConcurrentHashMapV8<K,V> map) {
3177	>	super(tab, index, size, limit, map);
3178	>	}
3179
1956	–	@SuppressWarnings("unchecked")
3180		public final K next() {
3181	<	if (next == null)
3181	>	Node<K,V> p;
3182	>	if ((p = next) == null)
3183		throw new NoSuchElementException();
3184	<	Object k = nextKey;
3184	>	K k = p.key;
3185	>	lastReturned = p;
3186		advance();
3187	<	return (K)k;
3187	>	return k;
3188		}
3189
3190		public final K nextElement() { return next(); }
3191		}
3192
3193	<	static final class ValueIterator<K,V> extends ViewIterator<K,V>
3193	>	static final class ValueIterator<K,V> extends BaseIterator<K,V>
3194		implements Iterator<V>, Enumeration<V> {
3195	<	ValueIterator(ConcurrentHashMapV8<K, V> map) { super(map); }
3195	>	ValueIterator(Node<K,V>[] tab, int index, int size, int limit,
3196	>	ConcurrentHashMapV8<K,V> map) {
3197	>	super(tab, index, size, limit, map);
3198	>	}
3199
1972	–	@SuppressWarnings("unchecked")
3200		public final V next() {
3201	<	if (next == null)
3201	>	Node<K,V> p;
3202	>	if ((p = next) == null)
3203		throw new NoSuchElementException();
3204	<	Object v = nextVal;
3204	>	V v = p.val;
3205	>	lastReturned = p;
3206		advance();
3207	<	return (V)v;
3207	>	return v;
3208		}
3209
3210		public final V nextElement() { return next(); }
3211		}
3212
3213	<	static final class EntryIterator<K,V> extends ViewIterator<K,V>
3213	>	static final class EntryIterator<K,V> extends BaseIterator<K,V>
3214		implements Iterator<Map.Entry<K,V>> {
3215	<	EntryIterator(ConcurrentHashMapV8<K, V> map) { super(map); }
3216	<
3217	<	@SuppressWarnings("unchecked")
1989	<	public final Map.Entry<K,V> next() {
1990	<	if (next == null)
1991	<	throw new NoSuchElementException();
1992	<	Object k = nextKey;
1993	<	Object v = nextVal;
1994	<	advance();
1995	<	return new WriteThroughEntry<K,V>((K)k, (V)v, map);
3215	>	EntryIterator(Node<K,V>[] tab, int index, int size, int limit,
3216	>	ConcurrentHashMapV8<K,V> map) {
3217	>	super(tab, index, size, limit, map);
3218		}
1997	–	}
3219
1999	–	static final class SnapshotEntryIterator<K,V> extends ViewIterator<K,V>
2000	–	implements Iterator<Map.Entry<K,V>> {
2001	–	SnapshotEntryIterator(ConcurrentHashMapV8<K, V> map) { super(map); }
2002	–
2003	–	@SuppressWarnings("unchecked")
3220		public final Map.Entry<K,V> next() {
3221	<	if (next == null)
3221	>	Node<K,V> p;
3222	>	if ((p = next) == null)
3223		throw new NoSuchElementException();
3224	<	Object k = nextKey;
3225	<	Object v = nextVal;
3224	>	K k = p.key;
3225	>	V v = p.val;
3226	>	lastReturned = p;
3227		advance();
3228	<	return new SnapshotEntry<K,V>((K)k, (V)v);
3228	>	return new MapEntry<K,V>(k, v, map);
3229		}
3230		}
3231
3232		/**
3233	<	* Base of writeThrough and Snapshot entry classes
3233	>	* Exported Entry for EntryIterator
3234		*/
3235	<	static abstract class MapEntry<K,V> implements Map.Entry<K, V> {
3235	>	static final class MapEntry<K,V> implements Map.Entry<K,V> {
3236		final K key; // non-null
3237		V val;       // non-null
3238	<	MapEntry(K key, V val)        { this.key = key; this.val = val; }
3239	<	public final K getKey()       { return key; }
3240	<	public final V getValue()     { return val; }
3241	<	public final int hashCode()   { return key.hashCode() ^ val.hashCode(); }
3242	<	public final String toString(){ return key + "=" + val; }
3238	>	final ConcurrentHashMapV8<K,V> map;
3239	>	MapEntry(K key, V val, ConcurrentHashMapV8<K,V> map) {
3240	>	this.key = key;
3241	>	this.val = val;
3242	>	this.map = map;
3243	>	}
3244	>	public K getKey()        { return key; }
3245	>	public V getValue()      { return val; }
3246	>	public int hashCode()    { return key.hashCode() ^ val.hashCode(); }
3247	>	public String toString() { return key + "=" + val; }
3248
3249	<	public final boolean equals(Object o) {
3249	>	public boolean equals(Object o) {
3250		Object k, v; Map.Entry<?,?> e;
3251		return ((o instanceof Map.Entry) &&
3252		(k = (e = (Map.Entry<?,?>)o).getKey()) != null &&
#	Line 2032 | Line 3255 | public class ConcurrentHashMapV8<K, V>
3255		(v == val || v.equals(val)));
3256		}
3257
2035	–	public abstract V setValue(V value);
2036	–	}
2037	–
2038	–	/**
2039	–	* Entry used by EntryIterator.next(), that relays setValue
2040	–	* changes to the underlying map.
2041	–	*/
2042	–	static final class WriteThroughEntry<K,V> extends MapEntry<K,V>
2043	–	implements Map.Entry<K, V> {
2044	–	final ConcurrentHashMapV8<K, V> map;
2045	–	WriteThroughEntry(K key, V val, ConcurrentHashMapV8<K, V> map) {
2046	–	super(key, val);
2047	–	this.map = map;
2048	–	}
2049	–
3258		/**
3259		* Sets our entry's value and writes through to the map. The
3260	<	* value to return is somewhat arbitrary here. Since a
3261	<	* WriteThroughEntry does not necessarily track asynchronous
3262	<	* changes, the most recent "previous" value could be
3263	<	* different from what we return (or could even have been
3264	<	* removed in which case the put will re-establish). We do not
2057	<	* and cannot guarantee more.
3260	>	* value to return is somewhat arbitrary here. Since we do not
3261	>	* necessarily track asynchronous changes, the most recent
3262	>	* "previous" value could be different from what we return (or
3263	>	* could even have been removed, in which case the put will
3264	>	* re-establish). We do not and cannot guarantee more.
3265		*/
3266	<	public final V setValue(V value) {
3266	>	public V setValue(V value) {
3267		if (value == null) throw new NullPointerException();
3268		V v = val;
3269		val = value;
#	Line 2065 | Line 3272 | public class ConcurrentHashMapV8<K, V>
3272		}
3273		}
3274
3275	+	static final class KeySpliterator<K,V> extends Traverser<K,V>
3276	+	implements ConcurrentHashMapSpliterator<K> {
3277	+	long est;               // size estimate
3278	+	KeySpliterator(Node<K,V>[] tab, int size, int index, int limit,
3279	+	long est) {
3280	+	super(tab, size, index, limit);
3281	+	this.est = est;
3282	+	}
3283	+
3284	+	public ConcurrentHashMapSpliterator<K> trySplit() {
3285	+	int i, f, h;
3286	+	return (h = ((i = baseIndex) + (f = baseLimit)) >>> 1) <= i ? null :
3287	+	new KeySpliterator<K,V>(tab, baseSize, baseLimit = h,
3288	+	f, est >>>= 1);
3289	+	}
3290	+
3291	+	public void forEachRemaining(Action<? super K> action) {
3292	+	if (action == null) throw new NullPointerException();
3293	+	for (Node<K,V> p; (p = advance()) != null;)
3294	+	action.apply(p.key);
3295	+	}
3296	+
3297	+	public boolean tryAdvance(Action<? super K> action) {
3298	+	if (action == null) throw new NullPointerException();
3299	+	Node<K,V> p;
3300	+	if ((p = advance()) == null)
3301	+	return false;
3302	+	action.apply(p.key);
3303	+	return true;
3304	+	}
3305	+
3306	+	public long estimateSize() { return est; }
3307	+
3308	+	}
3309	+
3310	+	static final class ValueSpliterator<K,V> extends Traverser<K,V>
3311	+	implements ConcurrentHashMapSpliterator<V> {
3312	+	long est;               // size estimate
3313	+	ValueSpliterator(Node<K,V>[] tab, int size, int index, int limit,
3314	+	long est) {
3315	+	super(tab, size, index, limit);
3316	+	this.est = est;
3317	+	}
3318	+
3319	+	public ConcurrentHashMapSpliterator<V> trySplit() {
3320	+	int i, f, h;
3321	+	return (h = ((i = baseIndex) + (f = baseLimit)) >>> 1) <= i ? null :
3322	+	new ValueSpliterator<K,V>(tab, baseSize, baseLimit = h,
3323	+	f, est >>>= 1);
3324	+	}
3325	+
3326	+	public void forEachRemaining(Action<? super V> action) {
3327	+	if (action == null) throw new NullPointerException();
3328	+	for (Node<K,V> p; (p = advance()) != null;)
3329	+	action.apply(p.val);
3330	+	}
3331	+
3332	+	public boolean tryAdvance(Action<? super V> action) {
3333	+	if (action == null) throw new NullPointerException();
3334	+	Node<K,V> p;
3335	+	if ((p = advance()) == null)
3336	+	return false;
3337	+	action.apply(p.val);
3338	+	return true;
3339	+	}
3340	+
3341	+	public long estimateSize() { return est; }
3342	+
3343	+	}
3344	+
3345	+	static final class EntrySpliterator<K,V> extends Traverser<K,V>
3346	+	implements ConcurrentHashMapSpliterator<Map.Entry<K,V>> {
3347	+	final ConcurrentHashMapV8<K,V> map; // To export MapEntry
3348	+	long est;               // size estimate
3349	+	EntrySpliterator(Node<K,V>[] tab, int size, int index, int limit,
3350	+	long est, ConcurrentHashMapV8<K,V> map) {
3351	+	super(tab, size, index, limit);
3352	+	this.map = map;
3353	+	this.est = est;
3354	+	}
3355	+
3356	+	public ConcurrentHashMapSpliterator<Map.Entry<K,V>> trySplit() {
3357	+	int i, f, h;
3358	+	return (h = ((i = baseIndex) + (f = baseLimit)) >>> 1) <= i ? null :
3359	+	new EntrySpliterator<K,V>(tab, baseSize, baseLimit = h,
3360	+	f, est >>>= 1, map);
3361	+	}
3362	+
3363	+	public void forEachRemaining(Action<? super Map.Entry<K,V>> action) {
3364	+	if (action == null) throw new NullPointerException();
3365	+	for (Node<K,V> p; (p = advance()) != null; )
3366	+	action.apply(new MapEntry<K,V>(p.key, p.val, map));
3367	+	}
3368	+
3369	+	public boolean tryAdvance(Action<? super Map.Entry<K,V>> action) {
3370	+	if (action == null) throw new NullPointerException();
3371	+	Node<K,V> p;
3372	+	if ((p = advance()) == null)
3373	+	return false;
3374	+	action.apply(new MapEntry<K,V>(p.key, p.val, map));
3375	+	return true;
3376	+	}
3377	+
3378	+	public long estimateSize() { return est; }
3379	+
3380	+	}
3381	+
3382	+	// Parallel bulk operations
3383	+
3384		/**
3385	<	* Internal version of entry, that doesn't write though changes
3385	>	* Computes initial batch value for bulk tasks. The returned value
3386	>	* is approximately exp2 of the number of times (minus one) to
3387	>	* split task by two before executing leaf action. This value is
3388	>	* faster to compute and more convenient to use as a guide to
3389	>	* splitting than is the depth, since it is used while dividing by
3390	>	* two anyway.
3391		*/
3392	<	static final class SnapshotEntry<K,V> extends MapEntry<K,V>
3393	<	implements Map.Entry<K, V> {
3394	<	SnapshotEntry(K key, V val) { super(key, val); }
3395	<	public final V setValue(V value) { // only locally update
3396	<	if (value == null) throw new NullPointerException();
3397	<	V v = val;
3398	<	val = value;
3399	<	return v;
3400	<	}
3392	>	final int batchFor(long b) {
3393	>	long n;
3394	>	if (b == Long.MAX_VALUE || (n = sumCount()) <= 1L || n < b)
3395	>	return 0;
3396	>	int sp = ForkJoinPool.getCommonPoolParallelism() << 2; // slack of 4
3397	>	return (b <= 0L || (n /= b) >= sp) ? sp : (int)n;
3398	>	}
3399	>
3400	>	/**
3401	>	* Performs the given action for each (key, value).
3402	>	*
3403	>	* @param parallelismThreshold the (estimated) number of elements
3404	>	* needed for this operation to be executed in parallel
3405	>	* @param action the action
3406	>	* @since 1.8
3407	>	*/
3408	>	public void forEach(long parallelismThreshold,
3409	>	BiAction<? super K,? super V> action) {
3410	>	if (action == null) throw new NullPointerException();
3411	>	new ForEachMappingTask<K,V>
3412	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3413	>	action).invoke();
3414	>	}
3415	>
3416	>	/**
3417	>	* Performs the given action for each non-null transformation
3418	>	* of each (key, value).
3419	>	*
3420	>	* @param parallelismThreshold the (estimated) number of elements
3421	>	* needed for this operation to be executed in parallel
3422	>	* @param transformer a function returning the transformation
3423	>	* for an element, or null if there is no transformation (in
3424	>	* which case the action is not applied)
3425	>	* @param action the action
3426	>	* @since 1.8
3427	>	*/
3428	>	public <U> void forEach(long parallelismThreshold,
3429	>	BiFun<? super K, ? super V, ? extends U> transformer,
3430	>	Action<? super U> action) {
3431	>	if (transformer == null || action == null)
3432	>	throw new NullPointerException();
3433	>	new ForEachTransformedMappingTask<K,V,U>
3434	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3435	>	transformer, action).invoke();
3436	>	}
3437	>
3438	>	/**
3439	>	* Returns a non-null result from applying the given search
3440	>	* function on each (key, value), or null if none.  Upon
3441	>	* success, further element processing is suppressed and the
3442	>	* results of any other parallel invocations of the search
3443	>	* function are ignored.
3444	>	*
3445	>	* @param parallelismThreshold the (estimated) number of elements
3446	>	* needed for this operation to be executed in parallel
3447	>	* @param searchFunction a function returning a non-null
3448	>	* result on success, else null
3449	>	* @return a non-null result from applying the given search
3450	>	* function on each (key, value), or null if none
3451	>	* @since 1.8
3452	>	*/
3453	>	public <U> U search(long parallelismThreshold,
3454	>	BiFun<? super K, ? super V, ? extends U> searchFunction) {
3455	>	if (searchFunction == null) throw new NullPointerException();
3456	>	return new SearchMappingsTask<K,V,U>
3457	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3458	>	searchFunction, new AtomicReference<U>()).invoke();
3459		}
3460
3461	+	/**
3462	+	* Returns the result of accumulating the given transformation
3463	+	* of all (key, value) pairs using the given reducer to
3464	+	* combine values, or null if none.
3465	+	*
3466	+	* @param parallelismThreshold the (estimated) number of elements
3467	+	* needed for this operation to be executed in parallel
3468	+	* @param transformer a function returning the transformation
3469	+	* for an element, or null if there is no transformation (in
3470	+	* which case it is not combined)
3471	+	* @param reducer a commutative associative combining function
3472	+	* @return the result of accumulating the given transformation
3473	+	* of all (key, value) pairs
3474	+	* @since 1.8
3475	+	*/
3476	+	public <U> U reduce(long parallelismThreshold,
3477	+	BiFun<? super K, ? super V, ? extends U> transformer,
3478	+	BiFun<? super U, ? super U, ? extends U> reducer) {
3479	+	if (transformer == null || reducer == null)
3480	+	throw new NullPointerException();
3481	+	return new MapReduceMappingsTask<K,V,U>
3482	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3483	+	null, transformer, reducer).invoke();
3484	+	}
3485	+
3486	+	/**
3487	+	* Returns the result of accumulating the given transformation
3488	+	* of all (key, value) pairs using the given reducer to
3489	+	* combine values, and the given basis as an identity value.
3490	+	*
3491	+	* @param parallelismThreshold the (estimated) number of elements
3492	+	* needed for this operation to be executed in parallel
3493	+	* @param transformer a function returning the transformation
3494	+	* for an element
3495	+	* @param basis the identity (initial default value) for the reduction
3496	+	* @param reducer a commutative associative combining function
3497	+	* @return the result of accumulating the given transformation
3498	+	* of all (key, value) pairs
3499	+	* @since 1.8
3500	+	*/
3501	+	public double reduceToDoubleIn(long parallelismThreshold,
3502	+	ObjectByObjectToDouble<? super K, ? super V> transformer,
3503	+	double basis,
3504	+	DoubleByDoubleToDouble reducer) {
3505	+	if (transformer == null || reducer == null)
3506	+	throw new NullPointerException();
3507	+	return new MapReduceMappingsToDoubleTask<K,V>
3508	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3509	+	null, transformer, basis, reducer).invoke();
3510	+	}
3511	+
3512	+	/**
3513	+	* Returns the result of accumulating the given transformation
3514	+	* of all (key, value) pairs using the given reducer to
3515	+	* combine values, and the given basis as an identity value.
3516	+	*
3517	+	* @param parallelismThreshold the (estimated) number of elements
3518	+	* needed for this operation to be executed in parallel
3519	+	* @param transformer a function returning the transformation
3520	+	* for an element
3521	+	* @param basis the identity (initial default value) for the reduction
3522	+	* @param reducer a commutative associative combining function
3523	+	* @return the result of accumulating the given transformation
3524	+	* of all (key, value) pairs
3525	+	* @since 1.8
3526	+	*/
3527	+	public long reduceToLong(long parallelismThreshold,
3528	+	ObjectByObjectToLong<? super K, ? super V> transformer,
3529	+	long basis,
3530	+	LongByLongToLong reducer) {
3531	+	if (transformer == null || reducer == null)
3532	+	throw new NullPointerException();
3533	+	return new MapReduceMappingsToLongTask<K,V>
3534	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3535	+	null, transformer, basis, reducer).invoke();
3536	+	}
3537	+
3538	+	/**
3539	+	* Returns the result of accumulating the given transformation
3540	+	* of all (key, value) pairs using the given reducer to
3541	+	* combine values, and the given basis as an identity value.
3542	+	*
3543	+	* @param parallelismThreshold the (estimated) number of elements
3544	+	* needed for this operation to be executed in parallel
3545	+	* @param transformer a function returning the transformation
3546	+	* for an element
3547	+	* @param basis the identity (initial default value) for the reduction
3548	+	* @param reducer a commutative associative combining function
3549	+	* @return the result of accumulating the given transformation
3550	+	* of all (key, value) pairs
3551	+	* @since 1.8
3552	+	*/
3553	+	public int reduceToInt(long parallelismThreshold,
3554	+	ObjectByObjectToInt<? super K, ? super V> transformer,
3555	+	int basis,
3556	+	IntByIntToInt reducer) {
3557	+	if (transformer == null || reducer == null)
3558	+	throw new NullPointerException();
3559	+	return new MapReduceMappingsToIntTask<K,V>
3560	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3561	+	null, transformer, basis, reducer).invoke();
3562	+	}
3563	+
3564	+	/**
3565	+	* Performs the given action for each key.
3566	+	*
3567	+	* @param parallelismThreshold the (estimated) number of elements
3568	+	* needed for this operation to be executed in parallel
3569	+	* @param action the action
3570	+	* @since 1.8
3571	+	*/
3572	+	public void forEachKey(long parallelismThreshold,
3573	+	Action<? super K> action) {
3574	+	if (action == null) throw new NullPointerException();
3575	+	new ForEachKeyTask<K,V>
3576	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3577	+	action).invoke();
3578	+	}
3579	+
3580	+	/**
3581	+	* Performs the given action for each non-null transformation
3582	+	* of each key.
3583	+	*
3584	+	* @param parallelismThreshold the (estimated) number of elements
3585	+	* needed for this operation to be executed in parallel
3586	+	* @param transformer a function returning the transformation
3587	+	* for an element, or null if there is no transformation (in
3588	+	* which case the action is not applied)
3589	+	* @param action the action
3590	+	* @since 1.8
3591	+	*/
3592	+	public <U> void forEachKey(long parallelismThreshold,
3593	+	Fun<? super K, ? extends U> transformer,
3594	+	Action<? super U> action) {
3595	+	if (transformer == null || action == null)
3596	+	throw new NullPointerException();
3597	+	new ForEachTransformedKeyTask<K,V,U>
3598	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3599	+	transformer, action).invoke();
3600	+	}
3601	+
3602	+	/**
3603	+	* Returns a non-null result from applying the given search
3604	+	* function on each key, or null if none. Upon success,
3605	+	* further element processing is suppressed and the results of
3606	+	* any other parallel invocations of the search function are
3607	+	* ignored.
3608	+	*
3609	+	* @param parallelismThreshold the (estimated) number of elements
3610	+	* needed for this operation to be executed in parallel
3611	+	* @param searchFunction a function returning a non-null
3612	+	* result on success, else null
3613	+	* @return a non-null result from applying the given search
3614	+	* function on each key, or null if none
3615	+	* @since 1.8
3616	+	*/
3617	+	public <U> U searchKeys(long parallelismThreshold,
3618	+	Fun<? super K, ? extends U> searchFunction) {
3619	+	if (searchFunction == null) throw new NullPointerException();
3620	+	return new SearchKeysTask<K,V,U>
3621	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3622	+	searchFunction, new AtomicReference<U>()).invoke();
3623	+	}
3624	+
3625	+	/**
3626	+	* Returns the result of accumulating all keys using the given
3627	+	* reducer to combine values, or null if none.
3628	+	*
3629	+	* @param parallelismThreshold the (estimated) number of elements
3630	+	* needed for this operation to be executed in parallel
3631	+	* @param reducer a commutative associative combining function
3632	+	* @return the result of accumulating all keys using the given
3633	+	* reducer to combine values, or null if none
3634	+	* @since 1.8
3635	+	*/
3636	+	public K reduceKeys(long parallelismThreshold,
3637	+	BiFun<? super K, ? super K, ? extends K> reducer) {
3638	+	if (reducer == null) throw new NullPointerException();
3639	+	return new ReduceKeysTask<K,V>
3640	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3641	+	null, reducer).invoke();
3642	+	}
3643	+
3644	+	/**
3645	+	* Returns the result of accumulating the given transformation
3646	+	* of all keys using the given reducer to combine values, or
3647	+	* null if none.
3648	+	*
3649	+	* @param parallelismThreshold the (estimated) number of elements
3650	+	* needed for this operation to be executed in parallel
3651	+	* @param transformer a function returning the transformation
3652	+	* for an element, or null if there is no transformation (in
3653	+	* which case it is not combined)
3654	+	* @param reducer a commutative associative combining function
3655	+	* @return the result of accumulating the given transformation
3656	+	* of all keys
3657	+	* @since 1.8
3658	+	*/
3659	+	public <U> U reduceKeys(long parallelismThreshold,
3660	+	Fun<? super K, ? extends U> transformer,
3661	+	BiFun<? super U, ? super U, ? extends U> reducer) {
3662	+	if (transformer == null || reducer == null)
3663	+	throw new NullPointerException();
3664	+	return new MapReduceKeysTask<K,V,U>
3665	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3666	+	null, transformer, reducer).invoke();
3667	+	}
3668	+
3669	+	/**
3670	+	* Returns the result of accumulating the given transformation
3671	+	* of all keys using the given reducer to combine values, and
3672	+	* the given basis as an identity value.
3673	+	*
3674	+	* @param parallelismThreshold the (estimated) number of elements
3675	+	* needed for this operation to be executed in parallel
3676	+	* @param transformer a function returning the transformation
3677	+	* for an element
3678	+	* @param basis the identity (initial default value) for the reduction
3679	+	* @param reducer a commutative associative combining function
3680	+	* @return the result of accumulating the given transformation
3681	+	* of all keys
3682	+	* @since 1.8
3683	+	*/
3684	+	public double reduceKeysToDouble(long parallelismThreshold,
3685	+	ObjectToDouble<? super K> transformer,
3686	+	double basis,
3687	+	DoubleByDoubleToDouble reducer) {
3688	+	if (transformer == null || reducer == null)
3689	+	throw new NullPointerException();
3690	+	return new MapReduceKeysToDoubleTask<K,V>
3691	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3692	+	null, transformer, basis, reducer).invoke();
3693	+	}
3694	+
3695	+	/**
3696	+	* Returns the result of accumulating the given transformation
3697	+	* of all keys using the given reducer to combine values, and
3698	+	* the given basis as an identity value.
3699	+	*
3700	+	* @param parallelismThreshold the (estimated) number of elements
3701	+	* needed for this operation to be executed in parallel
3702	+	* @param transformer a function returning the transformation
3703	+	* for an element
3704	+	* @param basis the identity (initial default value) for the reduction
3705	+	* @param reducer a commutative associative combining function
3706	+	* @return the result of accumulating the given transformation
3707	+	* of all keys
3708	+	* @since 1.8
3709	+	*/
3710	+	public long reduceKeysToLong(long parallelismThreshold,
3711	+	ObjectToLong<? super K> transformer,
3712	+	long basis,
3713	+	LongByLongToLong reducer) {
3714	+	if (transformer == null || reducer == null)
3715	+	throw new NullPointerException();
3716	+	return new MapReduceKeysToLongTask<K,V>
3717	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3718	+	null, transformer, basis, reducer).invoke();
3719	+	}
3720	+
3721	+	/**
3722	+	* Returns the result of accumulating the given transformation
3723	+	* of all keys using the given reducer to combine values, and
3724	+	* the given basis as an identity value.
3725	+	*
3726	+	* @param parallelismThreshold the (estimated) number of elements
3727	+	* needed for this operation to be executed in parallel
3728	+	* @param transformer a function returning the transformation
3729	+	* for an element
3730	+	* @param basis the identity (initial default value) for the reduction
3731	+	* @param reducer a commutative associative combining function
3732	+	* @return the result of accumulating the given transformation
3733	+	* of all keys
3734	+	* @since 1.8
3735	+	*/
3736	+	public int reduceKeysToInt(long parallelismThreshold,
3737	+	ObjectToInt<? super K> transformer,
3738	+	int basis,
3739	+	IntByIntToInt reducer) {
3740	+	if (transformer == null || reducer == null)
3741	+	throw new NullPointerException();
3742	+	return new MapReduceKeysToIntTask<K,V>
3743	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3744	+	null, transformer, basis, reducer).invoke();
3745	+	}
3746	+
3747	+	/**
3748	+	* Performs the given action for each value.
3749	+	*
3750	+	* @param parallelismThreshold the (estimated) number of elements
3751	+	* needed for this operation to be executed in parallel
3752	+	* @param action the action
3753	+	* @since 1.8
3754	+	*/
3755	+	public void forEachValue(long parallelismThreshold,
3756	+	Action<? super V> action) {
3757	+	if (action == null)
3758	+	throw new NullPointerException();
3759	+	new ForEachValueTask<K,V>
3760	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3761	+	action).invoke();
3762	+	}
3763	+
3764	+	/**
3765	+	* Performs the given action for each non-null transformation
3766	+	* of each value.
3767	+	*
3768	+	* @param parallelismThreshold the (estimated) number of elements
3769	+	* needed for this operation to be executed in parallel
3770	+	* @param transformer a function returning the transformation
3771	+	* for an element, or null if there is no transformation (in
3772	+	* which case the action is not applied)
3773	+	* @param action the action
3774	+	* @since 1.8
3775	+	*/
3776	+	public <U> void forEachValue(long parallelismThreshold,
3777	+	Fun<? super V, ? extends U> transformer,
3778	+	Action<? super U> action) {
3779	+	if (transformer == null || action == null)
3780	+	throw new NullPointerException();
3781	+	new ForEachTransformedValueTask<K,V,U>
3782	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3783	+	transformer, action).invoke();
3784	+	}
3785	+
3786	+	/**
3787	+	* Returns a non-null result from applying the given search
3788	+	* function on each value, or null if none.  Upon success,
3789	+	* further element processing is suppressed and the results of
3790	+	* any other parallel invocations of the search function are
3791	+	* ignored.
3792	+	*
3793	+	* @param parallelismThreshold the (estimated) number of elements
3794	+	* needed for this operation to be executed in parallel
3795	+	* @param searchFunction a function returning a non-null
3796	+	* result on success, else null
3797	+	* @return a non-null result from applying the given search
3798	+	* function on each value, or null if none
3799	+	* @since 1.8
3800	+	*/
3801	+	public <U> U searchValues(long parallelismThreshold,
3802	+	Fun<? super V, ? extends U> searchFunction) {
3803	+	if (searchFunction == null) throw new NullPointerException();
3804	+	return new SearchValuesTask<K,V,U>
3805	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3806	+	searchFunction, new AtomicReference<U>()).invoke();
3807	+	}
3808	+
3809	+	/**
3810	+	* Returns the result of accumulating all values using the
3811	+	* given reducer to combine values, or null if none.
3812	+	*
3813	+	* @param parallelismThreshold the (estimated) number of elements
3814	+	* needed for this operation to be executed in parallel
3815	+	* @param reducer a commutative associative combining function
3816	+	* @return the result of accumulating all values
3817	+	* @since 1.8
3818	+	*/
3819	+	public V reduceValues(long parallelismThreshold,
3820	+	BiFun<? super V, ? super V, ? extends V> reducer) {
3821	+	if (reducer == null) throw new NullPointerException();
3822	+	return new ReduceValuesTask<K,V>
3823	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3824	+	null, reducer).invoke();
3825	+	}
3826	+
3827	+	/**
3828	+	* Returns the result of accumulating the given transformation
3829	+	* of all values using the given reducer to combine values, or
3830	+	* null if none.
3831	+	*
3832	+	* @param parallelismThreshold the (estimated) number of elements
3833	+	* needed for this operation to be executed in parallel
3834	+	* @param transformer a function returning the transformation
3835	+	* for an element, or null if there is no transformation (in
3836	+	* which case it is not combined)
3837	+	* @param reducer a commutative associative combining function
3838	+	* @return the result of accumulating the given transformation
3839	+	* of all values
3840	+	* @since 1.8
3841	+	*/
3842	+	public <U> U reduceValues(long parallelismThreshold,
3843	+	Fun<? super V, ? extends U> transformer,
3844	+	BiFun<? super U, ? super U, ? extends U> reducer) {
3845	+	if (transformer == null || reducer == null)
3846	+	throw new NullPointerException();
3847	+	return new MapReduceValuesTask<K,V,U>
3848	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3849	+	null, transformer, reducer).invoke();
3850	+	}
3851	+
3852	+	/**
3853	+	* Returns the result of accumulating the given transformation
3854	+	* of all values using the given reducer to combine values,
3855	+	* and the given basis as an identity value.
3856	+	*
3857	+	* @param parallelismThreshold the (estimated) number of elements
3858	+	* needed for this operation to be executed in parallel
3859	+	* @param transformer a function returning the transformation
3860	+	* for an element
3861	+	* @param basis the identity (initial default value) for the reduction
3862	+	* @param reducer a commutative associative combining function
3863	+	* @return the result of accumulating the given transformation
3864	+	* of all values
3865	+	* @since 1.8
3866	+	*/
3867	+	public double reduceValuesToDouble(long parallelismThreshold,
3868	+	ObjectToDouble<? super V> transformer,
3869	+	double basis,
3870	+	DoubleByDoubleToDouble reducer) {
3871	+	if (transformer == null || reducer == null)
3872	+	throw new NullPointerException();
3873	+	return new MapReduceValuesToDoubleTask<K,V>
3874	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3875	+	null, transformer, basis, reducer).invoke();
3876	+	}
3877	+
3878	+	/**
3879	+	* Returns the result of accumulating the given transformation
3880	+	* of all values using the given reducer to combine values,
3881	+	* and the given basis as an identity value.
3882	+	*
3883	+	* @param parallelismThreshold the (estimated) number of elements
3884	+	* needed for this operation to be executed in parallel
3885	+	* @param transformer a function returning the transformation
3886	+	* for an element
3887	+	* @param basis the identity (initial default value) for the reduction
3888	+	* @param reducer a commutative associative combining function
3889	+	* @return the result of accumulating the given transformation
3890	+	* of all values
3891	+	* @since 1.8
3892	+	*/
3893	+	public long reduceValuesToLong(long parallelismThreshold,
3894	+	ObjectToLong<? super V> transformer,
3895	+	long basis,
3896	+	LongByLongToLong reducer) {
3897	+	if (transformer == null || reducer == null)
3898	+	throw new NullPointerException();
3899	+	return new MapReduceValuesToLongTask<K,V>
3900	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3901	+	null, transformer, basis, reducer).invoke();
3902	+	}
3903	+
3904	+	/**
3905	+	* Returns the result of accumulating the given transformation
3906	+	* of all values using the given reducer to combine values,
3907	+	* and the given basis as an identity value.
3908	+	*
3909	+	* @param parallelismThreshold the (estimated) number of elements
3910	+	* needed for this operation to be executed in parallel
3911	+	* @param transformer a function returning the transformation
3912	+	* for an element
3913	+	* @param basis the identity (initial default value) for the reduction
3914	+	* @param reducer a commutative associative combining function
3915	+	* @return the result of accumulating the given transformation
3916	+	* of all values
3917	+	* @since 1.8
3918	+	*/
3919	+	public int reduceValuesToInt(long parallelismThreshold,
3920	+	ObjectToInt<? super V> transformer,
3921	+	int basis,
3922	+	IntByIntToInt reducer) {
3923	+	if (transformer == null || reducer == null)
3924	+	throw new NullPointerException();
3925	+	return new MapReduceValuesToIntTask<K,V>
3926	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3927	+	null, transformer, basis, reducer).invoke();
3928	+	}
3929	+
3930	+	/**
3931	+	* Performs the given action for each entry.
3932	+	*
3933	+	* @param parallelismThreshold the (estimated) number of elements
3934	+	* needed for this operation to be executed in parallel
3935	+	* @param action the action
3936	+	* @since 1.8
3937	+	*/
3938	+	public void forEachEntry(long parallelismThreshold,
3939	+	Action<? super Map.Entry<K,V>> action) {
3940	+	if (action == null) throw new NullPointerException();
3941	+	new ForEachEntryTask<K,V>(null, batchFor(parallelismThreshold), 0, 0, table,
3942	+	action).invoke();
3943	+	}
3944	+
3945	+	/**
3946	+	* Performs the given action for each non-null transformation
3947	+	* of each entry.
3948	+	*
3949	+	* @param parallelismThreshold the (estimated) number of elements
3950	+	* needed for this operation to be executed in parallel
3951	+	* @param transformer a function returning the transformation
3952	+	* for an element, or null if there is no transformation (in
3953	+	* which case the action is not applied)
3954	+	* @param action the action
3955	+	* @since 1.8
3956	+	*/
3957	+	public <U> void forEachEntry(long parallelismThreshold,
3958	+	Fun<Map.Entry<K,V>, ? extends U> transformer,
3959	+	Action<? super U> action) {
3960	+	if (transformer == null || action == null)
3961	+	throw new NullPointerException();
3962	+	new ForEachTransformedEntryTask<K,V,U>
3963	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3964	+	transformer, action).invoke();
3965	+	}
3966	+
3967	+	/**
3968	+	* Returns a non-null result from applying the given search
3969	+	* function on each entry, or null if none.  Upon success,
3970	+	* further element processing is suppressed and the results of
3971	+	* any other parallel invocations of the search function are
3972	+	* ignored.
3973	+	*
3974	+	* @param parallelismThreshold the (estimated) number of elements
3975	+	* needed for this operation to be executed in parallel
3976	+	* @param searchFunction a function returning a non-null
3977	+	* result on success, else null
3978	+	* @return a non-null result from applying the given search
3979	+	* function on each entry, or null if none
3980	+	* @since 1.8
3981	+	*/
3982	+	public <U> U searchEntries(long parallelismThreshold,
3983	+	Fun<Map.Entry<K,V>, ? extends U> searchFunction) {
3984	+	if (searchFunction == null) throw new NullPointerException();
3985	+	return new SearchEntriesTask<K,V,U>
3986	+	(null, batchFor(parallelismThreshold), 0, 0, table,
3987	+	searchFunction, new AtomicReference<U>()).invoke();
3988	+	}
3989	+
3990	+	/**
3991	+	* Returns the result of accumulating all entries using the
3992	+	* given reducer to combine values, or null if none.
3993	+	*
3994	+	* @param parallelismThreshold the (estimated) number of elements
3995	+	* needed for this operation to be executed in parallel
3996	+	* @param reducer a commutative associative combining function
3997	+	* @return the result of accumulating all entries
3998	+	* @since 1.8
3999	+	*/
4000	+	public Map.Entry<K,V> reduceEntries(long parallelismThreshold,
4001	+	BiFun<Map.Entry<K,V>, Map.Entry<K,V>, ? extends Map.Entry<K,V>> reducer) {
4002	+	if (reducer == null) throw new NullPointerException();
4003	+	return new ReduceEntriesTask<K,V>
4004	+	(null, batchFor(parallelismThreshold), 0, 0, table,
4005	+	null, reducer).invoke();
4006	+	}
4007	+
4008	+	/**
4009	+	* Returns the result of accumulating the given transformation
4010	+	* of all entries using the given reducer to combine values,
4011	+	* or null if none.
4012	+	*
4013	+	* @param parallelismThreshold the (estimated) number of elements
4014	+	* needed for this operation to be executed in parallel
4015	+	* @param transformer a function returning the transformation
4016	+	* for an element, or null if there is no transformation (in
4017	+	* which case it is not combined)
4018	+	* @param reducer a commutative associative combining function
4019	+	* @return the result of accumulating the given transformation
4020	+	* of all entries
4021	+	* @since 1.8
4022	+	*/
4023	+	public <U> U reduceEntries(long parallelismThreshold,
4024	+	Fun<Map.Entry<K,V>, ? extends U> transformer,
4025	+	BiFun<? super U, ? super U, ? extends U> reducer) {
4026	+	if (transformer == null || reducer == null)
4027	+	throw new NullPointerException();
4028	+	return new MapReduceEntriesTask<K,V,U>
4029	+	(null, batchFor(parallelismThreshold), 0, 0, table,
4030	+	null, transformer, reducer).invoke();
4031	+	}
4032	+
4033	+	/**
4034	+	* Returns the result of accumulating the given transformation
4035	+	* of all entries using the given reducer to combine values,
4036	+	* and the given basis as an identity value.
4037	+	*
4038	+	* @param parallelismThreshold the (estimated) number of elements
4039	+	* needed for this operation to be executed in parallel
4040	+	* @param transformer a function returning the transformation
4041	+	* for an element
4042	+	* @param basis the identity (initial default value) for the reduction
4043	+	* @param reducer a commutative associative combining function
4044	+	* @return the result of accumulating the given transformation
4045	+	* of all entries
4046	+	* @since 1.8
4047	+	*/
4048	+	public double reduceEntriesToDouble(long parallelismThreshold,
4049	+	ObjectToDouble<Map.Entry<K,V>> transformer,
4050	+	double basis,
4051	+	DoubleByDoubleToDouble reducer) {
4052	+	if (transformer == null || reducer == null)
4053	+	throw new NullPointerException();
4054	+	return new MapReduceEntriesToDoubleTask<K,V>
4055	+	(null, batchFor(parallelismThreshold), 0, 0, table,
4056	+	null, transformer, basis, reducer).invoke();
4057	+	}
4058	+
4059	+	/**
4060	+	* Returns the result of accumulating the given transformation
4061	+	* of all entries using the given reducer to combine values,
4062	+	* and the given basis as an identity value.
4063	+	*
4064	+	* @param parallelismThreshold the (estimated) number of elements
4065	+	* needed for this operation to be executed in parallel
4066	+	* @param transformer a function returning the transformation
4067	+	* for an element
4068	+	* @param basis the identity (initial default value) for the reduction
4069	+	* @param reducer a commutative associative combining function
4070	+	* @return the result of accumulating the given transformation
4071	+	* of all entries
4072	+	* @since 1.8
4073	+	*/
4074	+	public long reduceEntriesToLong(long parallelismThreshold,
4075	+	ObjectToLong<Map.Entry<K,V>> transformer,
4076	+	long basis,
4077	+	LongByLongToLong reducer) {
4078	+	if (transformer == null || reducer == null)
4079	+	throw new NullPointerException();
4080	+	return new MapReduceEntriesToLongTask<K,V>
4081	+	(null, batchFor(parallelismThreshold), 0, 0, table,
4082	+	null, transformer, basis, reducer).invoke();
4083	+	}
4084	+
4085	+	/**
4086	+	* Returns the result of accumulating the given transformation
4087	+	* of all entries using the given reducer to combine values,
4088	+	* and the given basis as an identity value.
4089	+	*
4090	+	* @param parallelismThreshold the (estimated) number of elements
4091	+	* needed for this operation to be executed in parallel
4092	+	* @param transformer a function returning the transformation
4093	+	* for an element
4094	+	* @param basis the identity (initial default value) for the reduction
4095	+	* @param reducer a commutative associative combining function
4096	+	* @return the result of accumulating the given transformation
4097	+	* of all entries
4098	+	* @since 1.8
4099	+	*/
4100	+	public int reduceEntriesToInt(long parallelismThreshold,
4101	+	ObjectToInt<Map.Entry<K,V>> transformer,
4102	+	int basis,
4103	+	IntByIntToInt reducer) {
4104	+	if (transformer == null || reducer == null)
4105	+	throw new NullPointerException();
4106	+	return new MapReduceEntriesToIntTask<K,V>
4107	+	(null, batchFor(parallelismThreshold), 0, 0, table,
4108	+	null, transformer, basis, reducer).invoke();
4109	+	}
4110	+
4111	+
4112		/* ----------------Views -------------- */
4113
4114		/**
4115	<	* Base class for views. This is done mainly to allow adding
2086	<	* customized parallel traversals (not yet implemented.)
4115	>	* Base class for views.
4116		*/
4117	<	static abstract class MapView<K, V> {
4118	<	final ConcurrentHashMapV8<K, V> map;
4119	<	MapView(ConcurrentHashMapV8<K, V> map)  { this.map = map; }
4120	<	public final int size()                 { return map.size(); }
4121	<	public final boolean isEmpty()          { return map.isEmpty(); }
4122	<	public final void clear()               { map.clear(); }
4117	>	abstract static class CollectionView<K,V,E>
4118	>	implements Collection<E>, java.io.Serializable {
4119	>	private static final long serialVersionUID = 7249069246763182397L;
4120	>	final ConcurrentHashMapV8<K,V> map;
4121	>	CollectionView(ConcurrentHashMapV8<K,V> map)  { this.map = map; }
4122	>
4123	>	/**
4124	>	* Returns the map backing this view.
4125	>	*
4126	>	* @return the map backing this view
4127	>	*/
4128	>	public ConcurrentHashMapV8<K,V> getMap() { return map; }
4129	>
4130	>	/**
4131	>	* Removes all of the elements from this view, by removing all
4132	>	* the mappings from the map backing this view.
4133	>	*/
4134	>	public final void clear()      { map.clear(); }
4135	>	public final int size()        { return map.size(); }
4136	>	public final boolean isEmpty() { return map.isEmpty(); }
4137
4138		// implementations below rely on concrete classes supplying these
4139	<	abstract Iterator<?> iter();
4140	<	abstract public boolean contains(Object o);
4141	<	abstract public boolean remove(Object o);
4139	>	// abstract methods
4140	>	/**
4141	>	* Returns a "weakly consistent" iterator that will never
4142	>	* throw {@link ConcurrentModificationException}, and
4143	>	* guarantees to traverse elements as they existed upon
4144	>	* construction of the iterator, and may (but is not
4145	>	* guaranteed to) reflect any modifications subsequent to
4146	>	* construction.
4147	>	*/
4148	>	public abstract Iterator<E> iterator();
4149	>	public abstract boolean contains(Object o);
4150	>	public abstract boolean remove(Object o);
4151
4152		private static final String oomeMsg = "Required array size too large";
4153
4154		public final Object[] toArray() {
4155	<	long sz = map.longSize();
4156	<	if (sz > (long)(MAX_ARRAY_SIZE))
4155	>	long sz = map.mappingCount();
4156	>	if (sz > MAX_ARRAY_SIZE)
4157		throw new OutOfMemoryError(oomeMsg);
4158		int n = (int)sz;
4159		Object[] r = new Object[n];
4160		int i = 0;
4161	<	Iterator<?> it = iter();
2110	<	while (it.hasNext()) {
4161	>	for (E e : this) {
4162		if (i == n) {
4163		if (n >= MAX_ARRAY_SIZE)
4164		throw new OutOfMemoryError(oomeMsg);
#	Line 2117 | Line 4168 | public class ConcurrentHashMapV8<K, V>
4168		n += (n >>> 1) + 1;
4169		r = Arrays.copyOf(r, n);
4170		}
4171	<	r[i++] = it.next();
4171	>	r[i++] = e;
4172		}
4173		return (i == n) ? r : Arrays.copyOf(r, i);
4174		}
4175
4176		@SuppressWarnings("unchecked")
4177		public final <T> T[] toArray(T[] a) {
4178	<	long sz = map.longSize();
4179	<	if (sz > (long)(MAX_ARRAY_SIZE))
4178	>	long sz = map.mappingCount();
4179	>	if (sz > MAX_ARRAY_SIZE)
4180		throw new OutOfMemoryError(oomeMsg);
4181		int m = (int)sz;
4182		T[] r = (a.length >= m) ? a :
#	Line 2133 | Line 4184 | public class ConcurrentHashMapV8<K, V>
4184		.newInstance(a.getClass().getComponentType(), m);
4185		int n = r.length;
4186		int i = 0;
4187	<	Iterator<?> it = iter();
2137	<	while (it.hasNext()) {
4187	>	for (E e : this) {
4188		if (i == n) {
4189		if (n >= MAX_ARRAY_SIZE)
4190		throw new OutOfMemoryError(oomeMsg);
#	Line 2144 | Line 4194 | public class ConcurrentHashMapV8<K, V>
4194		n += (n >>> 1) + 1;
4195		r = Arrays.copyOf(r, n);
4196		}
4197	<	r[i++] = (T)it.next();
4197	>	r[i++] = (T)e;
4198		}
4199		if (a == r && i < n) {
4200		r[i] = null; // null-terminate
#	Line 2153 | Line 4203 | public class ConcurrentHashMapV8<K, V>
4203		return (i == n) ? r : Arrays.copyOf(r, i);
4204		}
4205
4206	<	public final int hashCode() {
4207	<	int h = 0;
4208	<	for (Iterator<?> it = iter(); it.hasNext();)
4209	<	h += it.next().hashCode();
4210	<	return h;
4211	<	}
4212	<
4206	>	/**
4207	>	* Returns a string representation of this collection.
4208	>	* The string representation consists of the string representations
4209	>	* of the collection's elements in the order they are returned by
4210	>	* its iterator, enclosed in square brackets ({@code "[]"}).
4211	>	* Adjacent elements are separated by the characters {@code ", "}
4212	>	* (comma and space).  Elements are converted to strings as by
4213	>	* {@link String#valueOf(Object)}.
4214	>	*
4215	>	* @return a string representation of this collection
4216	>	*/
4217		public final String toString() {
4218		StringBuilder sb = new StringBuilder();
4219		sb.append('[');
4220	<	Iterator<?> it = iter();
4220	>	Iterator<E> it = iterator();
4221		if (it.hasNext()) {
4222		for (;;) {
4223		Object e = it.next();
#	Line 2178 | Line 4232 | public class ConcurrentHashMapV8<K, V>
4232
4233		public final boolean containsAll(Collection<?> c) {
4234		if (c != this) {
4235	<	for (Iterator<?> it = c.iterator(); it.hasNext();) {
2182	<	Object e = it.next();
4235	>	for (Object e : c) {
4236		if (e == null || !contains(e))
4237		return false;
4238		}
#	Line 2187 | Line 4240 | public class ConcurrentHashMapV8<K, V>
4240		return true;
4241		}
4242
4243	<	public final boolean removeAll(Collection c) {
4243	>	public final boolean removeAll(Collection<?> c) {
4244		boolean modified = false;
4245	<	for (Iterator<?> it = iter(); it.hasNext();) {
4245	>	for (Iterator<E> it = iterator(); it.hasNext();) {
4246		if (c.contains(it.next())) {
4247		it.remove();
4248		modified = true;
#	Line 2200 | Line 4253 | public class ConcurrentHashMapV8<K, V>
4253
4254		public final boolean retainAll(Collection<?> c) {
4255		boolean modified = false;
4256	<	for (Iterator<?> it = iter(); it.hasNext();) {
4256	>	for (Iterator<E> it = iterator(); it.hasNext();) {
4257		if (!c.contains(it.next())) {
4258		it.remove();
4259		modified = true;
#	Line 2211 | Line 4264 | public class ConcurrentHashMapV8<K, V>
4264
4265		}
4266
4267	<	static final class KeySet<K,V> extends MapView<K,V> implements Set<K> {
4268	<	KeySet(ConcurrentHashMapV8<K, V> map)   { super(map); }
4269	<	public final boolean contains(Object o) { return map.containsKey(o); }
4270	<	public final boolean remove(Object o)   { return map.remove(o) != null; }
4267	>	/**
4268	>	* A view of a ConcurrentHashMapV8 as a {@link Set} of keys, in
4269	>	* which additions may optionally be enabled by mapping to a
4270	>	* common value.  This class cannot be directly instantiated.
4271	>	* See {@link #keySet() keySet()},
4272	>	* {@link #keySet(Object) keySet(V)},
4273	>	* {@link #newKeySet() newKeySet()},
4274	>	* {@link #newKeySet(int) newKeySet(int)}.
4275	>	*
4276	>	* @since 1.8
4277	>	*/
4278	>	public static class KeySetView<K,V> extends CollectionView<K,V,K>
4279	>	implements Set<K>, java.io.Serializable {
4280	>	private static final long serialVersionUID = 7249069246763182397L;
4281	>	private final V value;
4282	>	KeySetView(ConcurrentHashMapV8<K,V> map, V value) {  // non-public
4283	>	super(map);
4284	>	this.value = value;
4285	>	}
4286	>
4287	>	/**
4288	>	* Returns the default mapped value for additions,
4289	>	* or {@code null} if additions are not supported.
4290	>	*
4291	>	* @return the default mapped value for additions, or {@code null}
4292	>	* if not supported
4293	>	*/
4294	>	public V getMappedValue() { return value; }
4295
4296	<	public final Iterator<K> iterator() {
4297	<	return new KeyIterator<K,V>(map);
4296	>	/**
4297	>	* {@inheritDoc}
4298	>	* @throws NullPointerException if the specified key is null
4299	>	*/
4300	>	public boolean contains(Object o) { return map.containsKey(o); }
4301	>
4302	>	/**
4303	>	* Removes the key from this map view, by removing the key (and its
4304	>	* corresponding value) from the backing map.  This method does
4305	>	* nothing if the key is not in the map.
4306	>	*
4307	>	* @param  o the key to be removed from the backing map
4308	>	* @return {@code true} if the backing map contained the specified key
4309	>	* @throws NullPointerException if the specified key is null
4310	>	*/
4311	>	public boolean remove(Object o) { return map.remove(o) != null; }
4312	>
4313	>	/**
4314	>	* @return an iterator over the keys of the backing map
4315	>	*/
4316	>	public Iterator<K> iterator() {
4317	>	Node<K,V>[] t;
4318	>	ConcurrentHashMapV8<K,V> m = map;
4319	>	int f = (t = m.table) == null ? 0 : t.length;
4320	>	return new KeyIterator<K,V>(t, f, 0, f, m);
4321		}
4322	<	final Iterator<?> iter() {
4323	<	return new KeyIterator<K,V>(map);
4322	>
4323	>	/**
4324	>	* Adds the specified key to this set view by mapping the key to
4325	>	* the default mapped value in the backing map, if defined.
4326	>	*
4327	>	* @param e key to be added
4328	>	* @return {@code true} if this set changed as a result of the call
4329	>	* @throws NullPointerException if the specified key is null
4330	>	* @throws UnsupportedOperationException if no default mapped value
4331	>	* for additions was provided
4332	>	*/
4333	>	public boolean add(K e) {
4334	>	V v;
4335	>	if ((v = value) == null)
4336	>	throw new UnsupportedOperationException();
4337	>	return map.putVal(e, v, true) == null;
4338		}
4339	<	public final boolean add(K e) {
4340	<	throw new UnsupportedOperationException();
4339	>
4340	>	/**
4341	>	* Adds all of the elements in the specified collection to this set,
4342	>	* as if by calling {@link #add} on each one.
4343	>	*
4344	>	* @param c the elements to be inserted into this set
4345	>	* @return {@code true} if this set changed as a result of the call
4346	>	* @throws NullPointerException if the collection or any of its
4347	>	* elements are {@code null}
4348	>	* @throws UnsupportedOperationException if no default mapped value
4349	>	* for additions was provided
4350	>	*/
4351	>	public boolean addAll(Collection<? extends K> c) {
4352	>	boolean added = false;
4353	>	V v;
4354	>	if ((v = value) == null)
4355	>	throw new UnsupportedOperationException();
4356	>	for (K e : c) {
4357	>	if (map.putVal(e, v, true) == null)
4358	>	added = true;
4359	>	}
4360	>	return added;
4361		}
4362	<	public final boolean addAll(Collection<? extends K> c) {
4363	<	throw new UnsupportedOperationException();
4362	>
4363	>	public int hashCode() {
4364	>	int h = 0;
4365	>	for (K e : this)
4366	>	h += e.hashCode();
4367	>	return h;
4368		}
4369	+
4370		public boolean equals(Object o) {
4371		Set<?> c;
4372		return ((o instanceof Set) &&
4373		((c = (Set<?>)o) == this ||
4374		(containsAll(c) && c.containsAll(this))));
4375		}
4376	+
4377	+	public ConcurrentHashMapSpliterator<K> spliterator() {
4378	+	Node<K,V>[] t;
4379	+	ConcurrentHashMapV8<K,V> m = map;
4380	+	long n = m.sumCount();
4381	+	int f = (t = m.table) == null ? 0 : t.length;
4382	+	return new KeySpliterator<K,V>(t, f, 0, f, n < 0L ? 0L : n);
4383	+	}
4384	+
4385	+	public void forEach(Action<? super K> action) {
4386	+	if (action == null) throw new NullPointerException();
4387	+	Node<K,V>[] t;
4388	+	if ((t = map.table) != null) {
4389	+	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
4390	+	for (Node<K,V> p; (p = it.advance()) != null; )
4391	+	action.apply(p.key);
4392	+	}
4393	+	}
4394		}
4395
4396	<	static final class Values<K,V> extends MapView<K,V>
4397	<	implements Collection<V>  {
4398	<	Values(ConcurrentHashMapV8<K, V> map)   { super(map); }
4399	<	public final boolean contains(Object o) { return map.containsValue(o); }
4396	>	/**
4397	>	* A view of a ConcurrentHashMapV8 as a {@link Collection} of
4398	>	* values, in which additions are disabled. This class cannot be
4399	>	* directly instantiated. See {@link #values()}.
4400	>	*/
4401	>	static final class ValuesView<K,V> extends CollectionView<K,V,V>
4402	>	implements Collection<V>, java.io.Serializable {
4403	>	private static final long serialVersionUID = 2249069246763182397L;
4404	>	ValuesView(ConcurrentHashMapV8<K,V> map) { super(map); }
4405	>	public final boolean contains(Object o) {
4406	>	return map.containsValue(o);
4407	>	}
4408
4409		public final boolean remove(Object o) {
4410		if (o != null) {
4411	<	Iterator<V> it = new ValueIterator<K,V>(map);
2247	<	while (it.hasNext()) {
4411	>	for (Iterator<V> it = iterator(); it.hasNext();) {
4412		if (o.equals(it.next())) {
4413		it.remove();
4414		return true;
#	Line 2253 | Line 4417 | public class ConcurrentHashMapV8<K, V>
4417		}
4418		return false;
4419		}
4420	+
4421		public final Iterator<V> iterator() {
4422	<	return new ValueIterator<K,V>(map);
4423	<	}
4424	<	final Iterator<?> iter() {
4425	<	return new ValueIterator<K,V>(map);
4422	>	ConcurrentHashMapV8<K,V> m = map;
4423	>	Node<K,V>[] t;
4424	>	int f = (t = m.table) == null ? 0 : t.length;
4425	>	return new ValueIterator<K,V>(t, f, 0, f, m);
4426		}
4427	+
4428		public final boolean add(V e) {
4429		throw new UnsupportedOperationException();
4430		}
4431		public final boolean addAll(Collection<? extends V> c) {
4432		throw new UnsupportedOperationException();
4433		}
4434	+
4435	+	public ConcurrentHashMapSpliterator<V> spliterator() {
4436	+	Node<K,V>[] t;
4437	+	ConcurrentHashMapV8<K,V> m = map;
4438	+	long n = m.sumCount();
4439	+	int f = (t = m.table) == null ? 0 : t.length;
4440	+	return new ValueSpliterator<K,V>(t, f, 0, f, n < 0L ? 0L : n);
4441	+	}
4442	+
4443	+	public void forEach(Action<? super V> action) {
4444	+	if (action == null) throw new NullPointerException();
4445	+	Node<K,V>[] t;
4446	+	if ((t = map.table) != null) {
4447	+	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
4448	+	for (Node<K,V> p; (p = it.advance()) != null; )
4449	+	action.apply(p.val);
4450	+	}
4451	+	}
4452		}
4453
4454	<	static final class EntrySet<K,V>  extends MapView<K,V>
4455	<	implements Set<Map.Entry<K,V>> {
4456	<	EntrySet(ConcurrentHashMapV8<K, V> map) { super(map); }
4454	>	/**
4455	>	* A view of a ConcurrentHashMapV8 as a {@link Set} of (key, value)
4456	>	* entries.  This class cannot be directly instantiated. See
4457	>	* {@link #entrySet()}.
4458	>	*/
4459	>	static final class EntrySetView<K,V> extends CollectionView<K,V,Map.Entry<K,V>>
4460	>	implements Set<Map.Entry<K,V>>, java.io.Serializable {
4461	>	private static final long serialVersionUID = 2249069246763182397L;
4462	>	EntrySetView(ConcurrentHashMapV8<K,V> map) { super(map); }
4463
4464	<	public final boolean contains(Object o) {
4464	>	public boolean contains(Object o) {
4465		Object k, v, r; Map.Entry<?,?> e;
4466		return ((o instanceof Map.Entry) &&
4467		(k = (e = (Map.Entry<?,?>)o).getKey()) != null &&
#	Line 2280 | Line 4470 | public class ConcurrentHashMapV8<K, V>
4470		(v == r || v.equals(r)));
4471		}
4472
4473	<	public final boolean remove(Object o) {
4473	>	public boolean remove(Object o) {
4474		Object k, v; Map.Entry<?,?> e;
4475		return ((o instanceof Map.Entry) &&
4476		(k = (e = (Map.Entry<?,?>)o).getKey()) != null &&
#	Line 2288 | Line 4478 | public class ConcurrentHashMapV8<K, V>
4478		map.remove(k, v));
4479		}
4480
4481	<	public final Iterator<Map.Entry<K,V>> iterator() {
4482	<	return new EntryIterator<K,V>(map);
4481	>	/**
4482	>	* @return an iterator over the entries of the backing map
4483	>	*/
4484	>	public Iterator<Map.Entry<K,V>> iterator() {
4485	>	ConcurrentHashMapV8<K,V> m = map;
4486	>	Node<K,V>[] t;
4487	>	int f = (t = m.table) == null ? 0 : t.length;
4488	>	return new EntryIterator<K,V>(t, f, 0, f, m);
4489		}
4490	<	final Iterator<?> iter() {
4491	<	return new SnapshotEntryIterator<K,V>(map);
4490	>
4491	>	public boolean add(Entry<K,V> e) {
4492	>	return map.putVal(e.getKey(), e.getValue(), false) == null;
4493		}
4494	<	public final boolean add(Entry<K,V> e) {
4495	<	throw new UnsupportedOperationException();
4494	>
4495	>	public boolean addAll(Collection<? extends Entry<K,V>> c) {
4496	>	boolean added = false;
4497	>	for (Entry<K,V> e : c) {
4498	>	if (add(e))
4499	>	added = true;
4500	>	}
4501	>	return added;
4502		}
4503	<	public final boolean addAll(Collection<? extends Entry<K,V>> c) {
4504	<	throw new UnsupportedOperationException();
4503	>
4504	>	public final int hashCode() {
4505	>	int h = 0;
4506	>	Node<K,V>[] t;
4507	>	if ((t = map.table) != null) {
4508	>	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
4509	>	for (Node<K,V> p; (p = it.advance()) != null; ) {
4510	>	h += p.hashCode();
4511	>	}
4512	>	}
4513	>	return h;
4514		}
4515	<	public boolean equals(Object o) {
4515	>
4516	>	public final boolean equals(Object o) {
4517		Set<?> c;
4518		return ((o instanceof Set) &&
4519		((c = (Set<?>)o) == this ||
4520		(containsAll(c) && c.containsAll(this))));
4521		}
4522	+
4523	+	public ConcurrentHashMapSpliterator<Map.Entry<K,V>> spliterator() {
4524	+	Node<K,V>[] t;
4525	+	ConcurrentHashMapV8<K,V> m = map;
4526	+	long n = m.sumCount();
4527	+	int f = (t = m.table) == null ? 0 : t.length;
4528	+	return new EntrySpliterator<K,V>(t, f, 0, f, n < 0L ? 0L : n, m);
4529	+	}
4530	+
4531	+	public void forEach(Action<? super Map.Entry<K,V>> action) {
4532	+	if (action == null) throw new NullPointerException();
4533	+	Node<K,V>[] t;
4534	+	if ((t = map.table) != null) {
4535	+	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
4536	+	for (Node<K,V> p; (p = it.advance()) != null; )
4537	+	action.apply(new MapEntry<K,V>(p.key, p.val, map));
4538	+	}
4539	+	}
4540	+
4541		}
4542
4543	<	/* ---------------- Serialization Support -------------- */
4543	>	// -------------------------------------------------------
4544
4545		/**
4546	<	* Stripped-down version of helper class used in previous version,
4547	<	* declared for the sake of serialization compatibility
4546	>	* Base class for bulk tasks. Repeats some fields and code from
4547	>	* class Traverser, because we need to subclass CountedCompleter.
4548		*/
4549	<	static class Segment<K,V> implements Serializable {
4550	<	private static final long serialVersionUID = 2249069246763182397L;
4551	<	final float loadFactor;
4552	<	Segment(float lf) { this.loadFactor = lf; }
4549	>	abstract static class BulkTask<K,V,R> extends CountedCompleter<R> {
4550	>	Node<K,V>[] tab;        // same as Traverser
4551	>	Node<K,V> next;
4552	>	int index;
4553	>	int baseIndex;
4554	>	int baseLimit;
4555	>	final int baseSize;
4556	>	int batch;              // split control
4557	>
4558	>	BulkTask(BulkTask<K,V,?> par, int b, int i, int f, Node<K,V>[] t) {
4559	>	super(par);
4560	>	this.batch = b;
4561	>	this.index = this.baseIndex = i;
4562	>	if ((this.tab = t) == null)
4563	>	this.baseSize = this.baseLimit = 0;
4564	>	else if (par == null)
4565	>	this.baseSize = this.baseLimit = t.length;
4566	>	else {
4567	>	this.baseLimit = f;
4568	>	this.baseSize = par.baseSize;
4569	>	}
4570	>	}
4571	>
4572	>	/**
4573	>	* Same as Traverser version
4574	>	*/
4575	>	final Node<K,V> advance() {
4576	>	Node<K,V> e;
4577	>	if ((e = next) != null)
4578	>	e = e.next;
4579	>	for (;;) {
4580	>	Node<K,V>[] t; int i, n; K ek;  // must use locals in checks
4581	>	if (e != null)
4582	>	return next = e;
4583	>	if (baseIndex >= baseLimit || (t = tab) == null ||
4584	>	(n = t.length) <= (i = index) || i < 0)
4585	>	return next = null;
4586	>	if ((e = tabAt(t, index)) != null && e.hash < 0) {
4587	>	if (e instanceof ForwardingNode) {
4588	>	tab = ((ForwardingNode<K,V>)e).nextTable;
4589	>	e = null;
4590	>	continue;
4591	>	}
4592	>	else if (e instanceof TreeBin)
4593	>	e = ((TreeBin<K,V>)e).first;
4594	>	else
4595	>	e = null;
4596	>	}
4597	>	if ((index += baseSize) >= n)
4598	>	index = ++baseIndex;    // visit upper slots if present
4599	>	}
4600	>	}
4601	>	}
4602	>
4603	>	/*
4604	>	* Task classes. Coded in a regular but ugly format/style to
4605	>	* simplify checks that each variant differs in the right way from
4606	>	* others. The null screenings exist because compilers cannot tell
4607	>	* that we've already null-checked task arguments, so we force
4608	>	* simplest hoisted bypass to help avoid convoluted traps.
4609	>	*/
4610	>	@SuppressWarnings("serial")
4611	>	static final class ForEachKeyTask<K,V>
4612	>	extends BulkTask<K,V,Void> {
4613	>	final Action<? super K> action;
4614	>	ForEachKeyTask
4615	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4616	>	Action<? super K> action) {
4617	>	super(p, b, i, f, t);
4618	>	this.action = action;
4619	>	}
4620	>	public final void compute() {
4621	>	final Action<? super K> action;
4622	>	if ((action = this.action) != null) {
4623	>	for (int i = baseIndex, f, h; batch > 0 &&
4624	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4625	>	addToPendingCount(1);
4626	>	new ForEachKeyTask<K,V>
4627	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4628	>	action).fork();
4629	>	}
4630	>	for (Node<K,V> p; (p = advance()) != null;)
4631	>	action.apply(p.key);
4632	>	propagateCompletion();
4633	>	}
4634	>	}
4635	>	}
4636	>
4637	>	@SuppressWarnings("serial")
4638	>	static final class ForEachValueTask<K,V>
4639	>	extends BulkTask<K,V,Void> {
4640	>	final Action<? super V> action;
4641	>	ForEachValueTask
4642	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4643	>	Action<? super V> action) {
4644	>	super(p, b, i, f, t);
4645	>	this.action = action;
4646	>	}
4647	>	public final void compute() {
4648	>	final Action<? super V> action;
4649	>	if ((action = this.action) != null) {
4650	>	for (int i = baseIndex, f, h; batch > 0 &&
4651	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4652	>	addToPendingCount(1);
4653	>	new ForEachValueTask<K,V>
4654	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4655	>	action).fork();
4656	>	}
4657	>	for (Node<K,V> p; (p = advance()) != null;)
4658	>	action.apply(p.val);
4659	>	propagateCompletion();
4660	>	}
4661	>	}
4662	>	}
4663	>
4664	>	@SuppressWarnings("serial")
4665	>	static final class ForEachEntryTask<K,V>
4666	>	extends BulkTask<K,V,Void> {
4667	>	final Action<? super Entry<K,V>> action;
4668	>	ForEachEntryTask
4669	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4670	>	Action<? super Entry<K,V>> action) {
4671	>	super(p, b, i, f, t);
4672	>	this.action = action;
4673	>	}
4674	>	public final void compute() {
4675	>	final Action<? super Entry<K,V>> action;
4676	>	if ((action = this.action) != null) {
4677	>	for (int i = baseIndex, f, h; batch > 0 &&
4678	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4679	>	addToPendingCount(1);
4680	>	new ForEachEntryTask<K,V>
4681	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4682	>	action).fork();
4683	>	}
4684	>	for (Node<K,V> p; (p = advance()) != null; )
4685	>	action.apply(p);
4686	>	propagateCompletion();
4687	>	}
4688	>	}
4689	>	}
4690	>
4691	>	@SuppressWarnings("serial")
4692	>	static final class ForEachMappingTask<K,V>
4693	>	extends BulkTask<K,V,Void> {
4694	>	final BiAction<? super K, ? super V> action;
4695	>	ForEachMappingTask
4696	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4697	>	BiAction<? super K,? super V> action) {
4698	>	super(p, b, i, f, t);
4699	>	this.action = action;
4700	>	}
4701	>	public final void compute() {
4702	>	final BiAction<? super K, ? super V> action;
4703	>	if ((action = this.action) != null) {
4704	>	for (int i = baseIndex, f, h; batch > 0 &&
4705	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4706	>	addToPendingCount(1);
4707	>	new ForEachMappingTask<K,V>
4708	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4709	>	action).fork();
4710	>	}
4711	>	for (Node<K,V> p; (p = advance()) != null; )
4712	>	action.apply(p.key, p.val);
4713	>	propagateCompletion();
4714	>	}
4715	>	}
4716	>	}
4717	>
4718	>	@SuppressWarnings("serial")
4719	>	static final class ForEachTransformedKeyTask<K,V,U>
4720	>	extends BulkTask<K,V,Void> {
4721	>	final Fun<? super K, ? extends U> transformer;
4722	>	final Action<? super U> action;
4723	>	ForEachTransformedKeyTask
4724	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4725	>	Fun<? super K, ? extends U> transformer, Action<? super U> action) {
4726	>	super(p, b, i, f, t);
4727	>	this.transformer = transformer; this.action = action;
4728	>	}
4729	>	public final void compute() {
4730	>	final Fun<? super K, ? extends U> transformer;
4731	>	final Action<? super U> action;
4732	>	if ((transformer = this.transformer) != null &&
4733	>	(action = this.action) != null) {
4734	>	for (int i = baseIndex, f, h; batch > 0 &&
4735	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4736	>	addToPendingCount(1);
4737	>	new ForEachTransformedKeyTask<K,V,U>
4738	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4739	>	transformer, action).fork();
4740	>	}
4741	>	for (Node<K,V> p; (p = advance()) != null; ) {
4742	>	U u;
4743	>	if ((u = transformer.apply(p.key)) != null)
4744	>	action.apply(u);
4745	>	}
4746	>	propagateCompletion();
4747	>	}
4748	>	}
4749	>	}
4750	>
4751	>	@SuppressWarnings("serial")
4752	>	static final class ForEachTransformedValueTask<K,V,U>
4753	>	extends BulkTask<K,V,Void> {
4754	>	final Fun<? super V, ? extends U> transformer;
4755	>	final Action<? super U> action;
4756	>	ForEachTransformedValueTask
4757	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4758	>	Fun<? super V, ? extends U> transformer, Action<? super U> action) {
4759	>	super(p, b, i, f, t);
4760	>	this.transformer = transformer; this.action = action;
4761	>	}
4762	>	public final void compute() {
4763	>	final Fun<? super V, ? extends U> transformer;
4764	>	final Action<? super U> action;
4765	>	if ((transformer = this.transformer) != null &&
4766	>	(action = this.action) != null) {
4767	>	for (int i = baseIndex, f, h; batch > 0 &&
4768	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4769	>	addToPendingCount(1);
4770	>	new ForEachTransformedValueTask<K,V,U>
4771	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4772	>	transformer, action).fork();
4773	>	}
4774	>	for (Node<K,V> p; (p = advance()) != null; ) {
4775	>	U u;
4776	>	if ((u = transformer.apply(p.val)) != null)
4777	>	action.apply(u);
4778	>	}
4779	>	propagateCompletion();
4780	>	}
4781	>	}
4782	>	}
4783	>
4784	>	@SuppressWarnings("serial")
4785	>	static final class ForEachTransformedEntryTask<K,V,U>
4786	>	extends BulkTask<K,V,Void> {
4787	>	final Fun<Map.Entry<K,V>, ? extends U> transformer;
4788	>	final Action<? super U> action;
4789	>	ForEachTransformedEntryTask
4790	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4791	>	Fun<Map.Entry<K,V>, ? extends U> transformer, Action<? super U> action) {
4792	>	super(p, b, i, f, t);
4793	>	this.transformer = transformer; this.action = action;
4794	>	}
4795	>	public final void compute() {
4796	>	final Fun<Map.Entry<K,V>, ? extends U> transformer;
4797	>	final Action<? super U> action;
4798	>	if ((transformer = this.transformer) != null &&
4799	>	(action = this.action) != null) {
4800	>	for (int i = baseIndex, f, h; batch > 0 &&
4801	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4802	>	addToPendingCount(1);
4803	>	new ForEachTransformedEntryTask<K,V,U>
4804	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4805	>	transformer, action).fork();
4806	>	}
4807	>	for (Node<K,V> p; (p = advance()) != null; ) {
4808	>	U u;
4809	>	if ((u = transformer.apply(p)) != null)
4810	>	action.apply(u);
4811	>	}
4812	>	propagateCompletion();
4813	>	}
4814	>	}
4815	>	}
4816	>
4817	>	@SuppressWarnings("serial")
4818	>	static final class ForEachTransformedMappingTask<K,V,U>
4819	>	extends BulkTask<K,V,Void> {
4820	>	final BiFun<? super K, ? super V, ? extends U> transformer;
4821	>	final Action<? super U> action;
4822	>	ForEachTransformedMappingTask
4823	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4824	>	BiFun<? super K, ? super V, ? extends U> transformer,
4825	>	Action<? super U> action) {
4826	>	super(p, b, i, f, t);
4827	>	this.transformer = transformer; this.action = action;
4828	>	}
4829	>	public final void compute() {
4830	>	final BiFun<? super K, ? super V, ? extends U> transformer;
4831	>	final Action<? super U> action;
4832	>	if ((transformer = this.transformer) != null &&
4833	>	(action = this.action) != null) {
4834	>	for (int i = baseIndex, f, h; batch > 0 &&
4835	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4836	>	addToPendingCount(1);
4837	>	new ForEachTransformedMappingTask<K,V,U>
4838	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4839	>	transformer, action).fork();
4840	>	}
4841	>	for (Node<K,V> p; (p = advance()) != null; ) {
4842	>	U u;
4843	>	if ((u = transformer.apply(p.key, p.val)) != null)
4844	>	action.apply(u);
4845	>	}
4846	>	propagateCompletion();
4847	>	}
4848	>	}
4849	>	}
4850	>
4851	>	@SuppressWarnings("serial")
4852	>	static final class SearchKeysTask<K,V,U>
4853	>	extends BulkTask<K,V,U> {
4854	>	final Fun<? super K, ? extends U> searchFunction;
4855	>	final AtomicReference<U> result;
4856	>	SearchKeysTask
4857	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4858	>	Fun<? super K, ? extends U> searchFunction,
4859	>	AtomicReference<U> result) {
4860	>	super(p, b, i, f, t);
4861	>	this.searchFunction = searchFunction; this.result = result;
4862	>	}
4863	>	public final U getRawResult() { return result.get(); }
4864	>	public final void compute() {
4865	>	final Fun<? super K, ? extends U> searchFunction;
4866	>	final AtomicReference<U> result;
4867	>	if ((searchFunction = this.searchFunction) != null &&
4868	>	(result = this.result) != null) {
4869	>	for (int i = baseIndex, f, h; batch > 0 &&
4870	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4871	>	if (result.get() != null)
4872	>	return;
4873	>	addToPendingCount(1);
4874	>	new SearchKeysTask<K,V,U>
4875	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4876	>	searchFunction, result).fork();
4877	>	}
4878	>	while (result.get() == null) {
4879	>	U u;
4880	>	Node<K,V> p;
4881	>	if ((p = advance()) == null) {
4882	>	propagateCompletion();
4883	>	break;
4884	>	}
4885	>	if ((u = searchFunction.apply(p.key)) != null) {
4886	>	if (result.compareAndSet(null, u))
4887	>	quietlyCompleteRoot();
4888	>	break;
4889	>	}
4890	>	}
4891	>	}
4892	>	}
4893	>	}
4894	>
4895	>	@SuppressWarnings("serial")
4896	>	static final class SearchValuesTask<K,V,U>
4897	>	extends BulkTask<K,V,U> {
4898	>	final Fun<? super V, ? extends U> searchFunction;
4899	>	final AtomicReference<U> result;
4900	>	SearchValuesTask
4901	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4902	>	Fun<? super V, ? extends U> searchFunction,
4903	>	AtomicReference<U> result) {
4904	>	super(p, b, i, f, t);
4905	>	this.searchFunction = searchFunction; this.result = result;
4906	>	}
4907	>	public final U getRawResult() { return result.get(); }
4908	>	public final void compute() {
4909	>	final Fun<? super V, ? extends U> searchFunction;
4910	>	final AtomicReference<U> result;
4911	>	if ((searchFunction = this.searchFunction) != null &&
4912	>	(result = this.result) != null) {
4913	>	for (int i = baseIndex, f, h; batch > 0 &&
4914	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4915	>	if (result.get() != null)
4916	>	return;
4917	>	addToPendingCount(1);
4918	>	new SearchValuesTask<K,V,U>
4919	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4920	>	searchFunction, result).fork();
4921	>	}
4922	>	while (result.get() == null) {
4923	>	U u;
4924	>	Node<K,V> p;
4925	>	if ((p = advance()) == null) {
4926	>	propagateCompletion();
4927	>	break;
4928	>	}
4929	>	if ((u = searchFunction.apply(p.val)) != null) {
4930	>	if (result.compareAndSet(null, u))
4931	>	quietlyCompleteRoot();
4932	>	break;
4933	>	}
4934	>	}
4935	>	}
4936	>	}
4937	>	}
4938	>
4939	>	@SuppressWarnings("serial")
4940	>	static final class SearchEntriesTask<K,V,U>
4941	>	extends BulkTask<K,V,U> {
4942	>	final Fun<Entry<K,V>, ? extends U> searchFunction;
4943	>	final AtomicReference<U> result;
4944	>	SearchEntriesTask
4945	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4946	>	Fun<Entry<K,V>, ? extends U> searchFunction,
4947	>	AtomicReference<U> result) {
4948	>	super(p, b, i, f, t);
4949	>	this.searchFunction = searchFunction; this.result = result;
4950	>	}
4951	>	public final U getRawResult() { return result.get(); }
4952	>	public final void compute() {
4953	>	final Fun<Entry<K,V>, ? extends U> searchFunction;
4954	>	final AtomicReference<U> result;
4955	>	if ((searchFunction = this.searchFunction) != null &&
4956	>	(result = this.result) != null) {
4957	>	for (int i = baseIndex, f, h; batch > 0 &&
4958	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4959	>	if (result.get() != null)
4960	>	return;
4961	>	addToPendingCount(1);
4962	>	new SearchEntriesTask<K,V,U>
4963	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4964	>	searchFunction, result).fork();
4965	>	}
4966	>	while (result.get() == null) {
4967	>	U u;
4968	>	Node<K,V> p;
4969	>	if ((p = advance()) == null) {
4970	>	propagateCompletion();
4971	>	break;
4972	>	}
4973	>	if ((u = searchFunction.apply(p)) != null) {
4974	>	if (result.compareAndSet(null, u))
4975	>	quietlyCompleteRoot();
4976	>	return;
4977	>	}
4978	>	}
4979	>	}
4980	>	}
4981	>	}
4982	>
4983	>	@SuppressWarnings("serial")
4984	>	static final class SearchMappingsTask<K,V,U>
4985	>	extends BulkTask<K,V,U> {
4986	>	final BiFun<? super K, ? super V, ? extends U> searchFunction;
4987	>	final AtomicReference<U> result;
4988	>	SearchMappingsTask
4989	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4990	>	BiFun<? super K, ? super V, ? extends U> searchFunction,
4991	>	AtomicReference<U> result) {
4992	>	super(p, b, i, f, t);
4993	>	this.searchFunction = searchFunction; this.result = result;
4994	>	}
4995	>	public final U getRawResult() { return result.get(); }
4996	>	public final void compute() {
4997	>	final BiFun<? super K, ? super V, ? extends U> searchFunction;
4998	>	final AtomicReference<U> result;
4999	>	if ((searchFunction = this.searchFunction) != null &&
5000	>	(result = this.result) != null) {
5001	>	for (int i = baseIndex, f, h; batch > 0 &&
5002	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5003	>	if (result.get() != null)
5004	>	return;
5005	>	addToPendingCount(1);
5006	>	new SearchMappingsTask<K,V,U>
5007	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5008	>	searchFunction, result).fork();
5009	>	}
5010	>	while (result.get() == null) {
5011	>	U u;
5012	>	Node<K,V> p;
5013	>	if ((p = advance()) == null) {
5014	>	propagateCompletion();
5015	>	break;
5016	>	}
5017	>	if ((u = searchFunction.apply(p.key, p.val)) != null) {
5018	>	if (result.compareAndSet(null, u))
5019	>	quietlyCompleteRoot();
5020	>	break;
5021	>	}
5022	>	}
5023	>	}
5024	>	}
5025	>	}
5026	>
5027	>	@SuppressWarnings("serial")
5028	>	static final class ReduceKeysTask<K,V>
5029	>	extends BulkTask<K,V,K> {
5030	>	final BiFun<? super K, ? super K, ? extends K> reducer;
5031	>	K result;
5032	>	ReduceKeysTask<K,V> rights, nextRight;
5033	>	ReduceKeysTask
5034	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5035	>	ReduceKeysTask<K,V> nextRight,
5036	>	BiFun<? super K, ? super K, ? extends K> reducer) {
5037	>	super(p, b, i, f, t); this.nextRight = nextRight;
5038	>	this.reducer = reducer;
5039	>	}
5040	>	public final K getRawResult() { return result; }
5041	>	public final void compute() {
5042	>	final BiFun<? super K, ? super K, ? extends K> reducer;
5043	>	if ((reducer = this.reducer) != null) {
5044	>	for (int i = baseIndex, f, h; batch > 0 &&
5045	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5046	>	addToPendingCount(1);
5047	>	(rights = new ReduceKeysTask<K,V>
5048	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5049	>	rights, reducer)).fork();
5050	>	}
5051	>	K r = null;
5052	>	for (Node<K,V> p; (p = advance()) != null; ) {
5053	>	K u = p.key;
5054	>	r = (r == null) ? u : u == null ? r : reducer.apply(r, u);
5055	>	}
5056	>	result = r;
5057	>	CountedCompleter<?> c;
5058	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5059	>	@SuppressWarnings("unchecked") ReduceKeysTask<K,V>
5060	>	t = (ReduceKeysTask<K,V>)c,
5061	>	s = t.rights;
5062	>	while (s != null) {
5063	>	K tr, sr;
5064	>	if ((sr = s.result) != null)
5065	>	t.result = (((tr = t.result) == null) ? sr :
5066	>	reducer.apply(tr, sr));
5067	>	s = t.rights = s.nextRight;
5068	>	}
5069	>	}
5070	>	}
5071	>	}
5072	>	}
5073	>
5074	>	@SuppressWarnings("serial")
5075	>	static final class ReduceValuesTask<K,V>
5076	>	extends BulkTask<K,V,V> {
5077	>	final BiFun<? super V, ? super V, ? extends V> reducer;
5078	>	V result;
5079	>	ReduceValuesTask<K,V> rights, nextRight;
5080	>	ReduceValuesTask
5081	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5082	>	ReduceValuesTask<K,V> nextRight,
5083	>	BiFun<? super V, ? super V, ? extends V> reducer) {
5084	>	super(p, b, i, f, t); this.nextRight = nextRight;
5085	>	this.reducer = reducer;
5086	>	}
5087	>	public final V getRawResult() { return result; }
5088	>	public final void compute() {
5089	>	final BiFun<? super V, ? super V, ? extends V> reducer;
5090	>	if ((reducer = this.reducer) != null) {
5091	>	for (int i = baseIndex, f, h; batch > 0 &&
5092	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5093	>	addToPendingCount(1);
5094	>	(rights = new ReduceValuesTask<K,V>
5095	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5096	>	rights, reducer)).fork();
5097	>	}
5098	>	V r = null;
5099	>	for (Node<K,V> p; (p = advance()) != null; ) {
5100	>	V v = p.val;
5101	>	r = (r == null) ? v : reducer.apply(r, v);
5102	>	}
5103	>	result = r;
5104	>	CountedCompleter<?> c;
5105	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5106	>	@SuppressWarnings("unchecked") ReduceValuesTask<K,V>
5107	>	t = (ReduceValuesTask<K,V>)c,
5108	>	s = t.rights;
5109	>	while (s != null) {
5110	>	V tr, sr;
5111	>	if ((sr = s.result) != null)
5112	>	t.result = (((tr = t.result) == null) ? sr :
5113	>	reducer.apply(tr, sr));
5114	>	s = t.rights = s.nextRight;
5115	>	}
5116	>	}
5117	>	}
5118	>	}
5119	>	}
5120	>
5121	>	@SuppressWarnings("serial")
5122	>	static final class ReduceEntriesTask<K,V>
5123	>	extends BulkTask<K,V,Map.Entry<K,V>> {
5124	>	final BiFun<Map.Entry<K,V>, Map.Entry<K,V>, ? extends Map.Entry<K,V>> reducer;
5125	>	Map.Entry<K,V> result;
5126	>	ReduceEntriesTask<K,V> rights, nextRight;
5127	>	ReduceEntriesTask
5128	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5129	>	ReduceEntriesTask<K,V> nextRight,
5130	>	BiFun<Entry<K,V>, Map.Entry<K,V>, ? extends Map.Entry<K,V>> reducer) {
5131	>	super(p, b, i, f, t); this.nextRight = nextRight;
5132	>	this.reducer = reducer;
5133	>	}
5134	>	public final Map.Entry<K,V> getRawResult() { return result; }
5135	>	public final void compute() {
5136	>	final BiFun<Map.Entry<K,V>, Map.Entry<K,V>, ? extends Map.Entry<K,V>> reducer;
5137	>	if ((reducer = this.reducer) != null) {
5138	>	for (int i = baseIndex, f, h; batch > 0 &&
5139	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5140	>	addToPendingCount(1);
5141	>	(rights = new ReduceEntriesTask<K,V>
5142	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5143	>	rights, reducer)).fork();
5144	>	}
5145	>	Map.Entry<K,V> r = null;
5146	>	for (Node<K,V> p; (p = advance()) != null; )
5147	>	r = (r == null) ? p : reducer.apply(r, p);
5148	>	result = r;
5149	>	CountedCompleter<?> c;
5150	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5151	>	@SuppressWarnings("unchecked") ReduceEntriesTask<K,V>
5152	>	t = (ReduceEntriesTask<K,V>)c,
5153	>	s = t.rights;
5154	>	while (s != null) {
5155	>	Map.Entry<K,V> tr, sr;
5156	>	if ((sr = s.result) != null)
5157	>	t.result = (((tr = t.result) == null) ? sr :
5158	>	reducer.apply(tr, sr));
5159	>	s = t.rights = s.nextRight;
5160	>	}
5161	>	}
5162	>	}
5163	>	}
5164	>	}
5165	>
5166	>	@SuppressWarnings("serial")
5167	>	static final class MapReduceKeysTask<K,V,U>
5168	>	extends BulkTask<K,V,U> {
5169	>	final Fun<? super K, ? extends U> transformer;
5170	>	final BiFun<? super U, ? super U, ? extends U> reducer;
5171	>	U result;
5172	>	MapReduceKeysTask<K,V,U> rights, nextRight;
5173	>	MapReduceKeysTask
5174	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5175	>	MapReduceKeysTask<K,V,U> nextRight,
5176	>	Fun<? super K, ? extends U> transformer,
5177	>	BiFun<? super U, ? super U, ? extends U> reducer) {
5178	>	super(p, b, i, f, t); this.nextRight = nextRight;
5179	>	this.transformer = transformer;
5180	>	this.reducer = reducer;
5181	>	}
5182	>	public final U getRawResult() { return result; }
5183	>	public final void compute() {
5184	>	final Fun<? super K, ? extends U> transformer;
5185	>	final BiFun<? super U, ? super U, ? extends U> reducer;
5186	>	if ((transformer = this.transformer) != null &&
5187	>	(reducer = this.reducer) != null) {
5188	>	for (int i = baseIndex, f, h; batch > 0 &&
5189	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5190	>	addToPendingCount(1);
5191	>	(rights = new MapReduceKeysTask<K,V,U>
5192	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5193	>	rights, transformer, reducer)).fork();
5194	>	}
5195	>	U r = null;
5196	>	for (Node<K,V> p; (p = advance()) != null; ) {
5197	>	U u;
5198	>	if ((u = transformer.apply(p.key)) != null)
5199	>	r = (r == null) ? u : reducer.apply(r, u);
5200	>	}
5201	>	result = r;
5202	>	CountedCompleter<?> c;
5203	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5204	>	@SuppressWarnings("unchecked") MapReduceKeysTask<K,V,U>
5205	>	t = (MapReduceKeysTask<K,V,U>)c,
5206	>	s = t.rights;
5207	>	while (s != null) {
5208	>	U tr, sr;
5209	>	if ((sr = s.result) != null)
5210	>	t.result = (((tr = t.result) == null) ? sr :
5211	>	reducer.apply(tr, sr));
5212	>	s = t.rights = s.nextRight;
5213	>	}
5214	>	}
5215	>	}
5216	>	}
5217	>	}
5218	>
5219	>	@SuppressWarnings("serial")
5220	>	static final class MapReduceValuesTask<K,V,U>
5221	>	extends BulkTask<K,V,U> {
5222	>	final Fun<? super V, ? extends U> transformer;
5223	>	final BiFun<? super U, ? super U, ? extends U> reducer;
5224	>	U result;
5225	>	MapReduceValuesTask<K,V,U> rights, nextRight;
5226	>	MapReduceValuesTask
5227	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5228	>	MapReduceValuesTask<K,V,U> nextRight,
5229	>	Fun<? super V, ? extends U> transformer,
5230	>	BiFun<? super U, ? super U, ? extends U> reducer) {
5231	>	super(p, b, i, f, t); this.nextRight = nextRight;
5232	>	this.transformer = transformer;
5233	>	this.reducer = reducer;
5234	>	}
5235	>	public final U getRawResult() { return result; }
5236	>	public final void compute() {
5237	>	final Fun<? super V, ? extends U> transformer;
5238	>	final BiFun<? super U, ? super U, ? extends U> reducer;
5239	>	if ((transformer = this.transformer) != null &&
5240	>	(reducer = this.reducer) != null) {
5241	>	for (int i = baseIndex, f, h; batch > 0 &&
5242	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5243	>	addToPendingCount(1);
5244	>	(rights = new MapReduceValuesTask<K,V,U>
5245	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5246	>	rights, transformer, reducer)).fork();
5247	>	}
5248	>	U r = null;
5249	>	for (Node<K,V> p; (p = advance()) != null; ) {
5250	>	U u;
5251	>	if ((u = transformer.apply(p.val)) != null)
5252	>	r = (r == null) ? u : reducer.apply(r, u);
5253	>	}
5254	>	result = r;
5255	>	CountedCompleter<?> c;
5256	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5257	>	@SuppressWarnings("unchecked") MapReduceValuesTask<K,V,U>
5258	>	t = (MapReduceValuesTask<K,V,U>)c,
5259	>	s = t.rights;
5260	>	while (s != null) {
5261	>	U tr, sr;
5262	>	if ((sr = s.result) != null)
5263	>	t.result = (((tr = t.result) == null) ? sr :
5264	>	reducer.apply(tr, sr));
5265	>	s = t.rights = s.nextRight;
5266	>	}
5267	>	}
5268	>	}
5269	>	}
5270	>	}
5271	>
5272	>	@SuppressWarnings("serial")
5273	>	static final class MapReduceEntriesTask<K,V,U>
5274	>	extends BulkTask<K,V,U> {
5275	>	final Fun<Map.Entry<K,V>, ? extends U> transformer;
5276	>	final BiFun<? super U, ? super U, ? extends U> reducer;
5277	>	U result;
5278	>	MapReduceEntriesTask<K,V,U> rights, nextRight;
5279	>	MapReduceEntriesTask
5280	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5281	>	MapReduceEntriesTask<K,V,U> nextRight,
5282	>	Fun<Map.Entry<K,V>, ? extends U> transformer,
5283	>	BiFun<? super U, ? super U, ? extends U> reducer) {
5284	>	super(p, b, i, f, t); this.nextRight = nextRight;
5285	>	this.transformer = transformer;
5286	>	this.reducer = reducer;
5287	>	}
5288	>	public final U getRawResult() { return result; }
5289	>	public final void compute() {
5290	>	final Fun<Map.Entry<K,V>, ? extends U> transformer;
5291	>	final BiFun<? super U, ? super U, ? extends U> reducer;
5292	>	if ((transformer = this.transformer) != null &&
5293	>	(reducer = this.reducer) != null) {
5294	>	for (int i = baseIndex, f, h; batch > 0 &&
5295	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5296	>	addToPendingCount(1);
5297	>	(rights = new MapReduceEntriesTask<K,V,U>
5298	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5299	>	rights, transformer, reducer)).fork();
5300	>	}
5301	>	U r = null;
5302	>	for (Node<K,V> p; (p = advance()) != null; ) {
5303	>	U u;
5304	>	if ((u = transformer.apply(p)) != null)
5305	>	r = (r == null) ? u : reducer.apply(r, u);
5306	>	}
5307	>	result = r;
5308	>	CountedCompleter<?> c;
5309	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5310	>	@SuppressWarnings("unchecked") MapReduceEntriesTask<K,V,U>
5311	>	t = (MapReduceEntriesTask<K,V,U>)c,
5312	>	s = t.rights;
5313	>	while (s != null) {
5314	>	U tr, sr;
5315	>	if ((sr = s.result) != null)
5316	>	t.result = (((tr = t.result) == null) ? sr :
5317	>	reducer.apply(tr, sr));
5318	>	s = t.rights = s.nextRight;
5319	>	}
5320	>	}
5321	>	}
5322	>	}
5323	>	}
5324	>
5325	>	@SuppressWarnings("serial")
5326	>	static final class MapReduceMappingsTask<K,V,U>
5327	>	extends BulkTask<K,V,U> {
5328	>	final BiFun<? super K, ? super V, ? extends U> transformer;
5329	>	final BiFun<? super U, ? super U, ? extends U> reducer;
5330	>	U result;
5331	>	MapReduceMappingsTask<K,V,U> rights, nextRight;
5332	>	MapReduceMappingsTask
5333	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5334	>	MapReduceMappingsTask<K,V,U> nextRight,
5335	>	BiFun<? super K, ? super V, ? extends U> transformer,
5336	>	BiFun<? super U, ? super U, ? extends U> reducer) {
5337	>	super(p, b, i, f, t); this.nextRight = nextRight;
5338	>	this.transformer = transformer;
5339	>	this.reducer = reducer;
5340	>	}
5341	>	public final U getRawResult() { return result; }
5342	>	public final void compute() {
5343	>	final BiFun<? super K, ? super V, ? extends U> transformer;
5344	>	final BiFun<? super U, ? super U, ? extends U> reducer;
5345	>	if ((transformer = this.transformer) != null &&
5346	>	(reducer = this.reducer) != null) {
5347	>	for (int i = baseIndex, f, h; batch > 0 &&
5348	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5349	>	addToPendingCount(1);
5350	>	(rights = new MapReduceMappingsTask<K,V,U>
5351	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5352	>	rights, transformer, reducer)).fork();
5353	>	}
5354	>	U r = null;
5355	>	for (Node<K,V> p; (p = advance()) != null; ) {
5356	>	U u;
5357	>	if ((u = transformer.apply(p.key, p.val)) != null)
5358	>	r = (r == null) ? u : reducer.apply(r, u);
5359	>	}
5360	>	result = r;
5361	>	CountedCompleter<?> c;
5362	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5363	>	@SuppressWarnings("unchecked") MapReduceMappingsTask<K,V,U>
5364	>	t = (MapReduceMappingsTask<K,V,U>)c,
5365	>	s = t.rights;
5366	>	while (s != null) {
5367	>	U tr, sr;
5368	>	if ((sr = s.result) != null)
5369	>	t.result = (((tr = t.result) == null) ? sr :
5370	>	reducer.apply(tr, sr));
5371	>	s = t.rights = s.nextRight;
5372	>	}
5373	>	}
5374	>	}
5375	>	}
5376	>	}
5377	>
5378	>	@SuppressWarnings("serial")
5379	>	static final class MapReduceKeysToDoubleTask<K,V>
5380	>	extends BulkTask<K,V,Double> {
5381	>	final ObjectToDouble<? super K> transformer;
5382	>	final DoubleByDoubleToDouble reducer;
5383	>	final double basis;
5384	>	double result;
5385	>	MapReduceKeysToDoubleTask<K,V> rights, nextRight;
5386	>	MapReduceKeysToDoubleTask
5387	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5388	>	MapReduceKeysToDoubleTask<K,V> nextRight,
5389	>	ObjectToDouble<? super K> transformer,
5390	>	double basis,
5391	>	DoubleByDoubleToDouble reducer) {
5392	>	super(p, b, i, f, t); this.nextRight = nextRight;
5393	>	this.transformer = transformer;
5394	>	this.basis = basis; this.reducer = reducer;
5395	>	}
5396	>	public final Double getRawResult() { return result; }
5397	>	public final void compute() {
5398	>	final ObjectToDouble<? super K> transformer;
5399	>	final DoubleByDoubleToDouble reducer;
5400	>	if ((transformer = this.transformer) != null &&
5401	>	(reducer = this.reducer) != null) {
5402	>	double r = this.basis;
5403	>	for (int i = baseIndex, f, h; batch > 0 &&
5404	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5405	>	addToPendingCount(1);
5406	>	(rights = new MapReduceKeysToDoubleTask<K,V>
5407	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5408	>	rights, transformer, r, reducer)).fork();
5409	>	}
5410	>	for (Node<K,V> p; (p = advance()) != null; )
5411	>	r = reducer.apply(r, transformer.apply(p.key));
5412	>	result = r;
5413	>	CountedCompleter<?> c;
5414	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5415	>	@SuppressWarnings("unchecked") MapReduceKeysToDoubleTask<K,V>
5416	>	t = (MapReduceKeysToDoubleTask<K,V>)c,
5417	>	s = t.rights;
5418	>	while (s != null) {
5419	>	t.result = reducer.apply(t.result, s.result);
5420	>	s = t.rights = s.nextRight;
5421	>	}
5422	>	}
5423	>	}
5424	>	}
5425	>	}
5426	>
5427	>	@SuppressWarnings("serial")
5428	>	static final class MapReduceValuesToDoubleTask<K,V>
5429	>	extends BulkTask<K,V,Double> {
5430	>	final ObjectToDouble<? super V> transformer;
5431	>	final DoubleByDoubleToDouble reducer;
5432	>	final double basis;
5433	>	double result;
5434	>	MapReduceValuesToDoubleTask<K,V> rights, nextRight;
5435	>	MapReduceValuesToDoubleTask
5436	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5437	>	MapReduceValuesToDoubleTask<K,V> nextRight,
5438	>	ObjectToDouble<? super V> transformer,
5439	>	double basis,
5440	>	DoubleByDoubleToDouble reducer) {
5441	>	super(p, b, i, f, t); this.nextRight = nextRight;
5442	>	this.transformer = transformer;
5443	>	this.basis = basis; this.reducer = reducer;
5444	>	}
5445	>	public final Double getRawResult() { return result; }
5446	>	public final void compute() {
5447	>	final ObjectToDouble<? super V> transformer;
5448	>	final DoubleByDoubleToDouble reducer;
5449	>	if ((transformer = this.transformer) != null &&
5450	>	(reducer = this.reducer) != null) {
5451	>	double r = this.basis;
5452	>	for (int i = baseIndex, f, h; batch > 0 &&
5453	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5454	>	addToPendingCount(1);
5455	>	(rights = new MapReduceValuesToDoubleTask<K,V>
5456	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5457	>	rights, transformer, r, reducer)).fork();
5458	>	}
5459	>	for (Node<K,V> p; (p = advance()) != null; )
5460	>	r = reducer.apply(r, transformer.apply(p.val));
5461	>	result = r;
5462	>	CountedCompleter<?> c;
5463	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5464	>	@SuppressWarnings("unchecked") MapReduceValuesToDoubleTask<K,V>
5465	>	t = (MapReduceValuesToDoubleTask<K,V>)c,
5466	>	s = t.rights;
5467	>	while (s != null) {
5468	>	t.result = reducer.apply(t.result, s.result);
5469	>	s = t.rights = s.nextRight;
5470	>	}
5471	>	}
5472	>	}
5473	>	}
5474	>	}
5475	>
5476	>	@SuppressWarnings("serial")
5477	>	static final class MapReduceEntriesToDoubleTask<K,V>
5478	>	extends BulkTask<K,V,Double> {
5479	>	final ObjectToDouble<Map.Entry<K,V>> transformer;
5480	>	final DoubleByDoubleToDouble reducer;
5481	>	final double basis;
5482	>	double result;
5483	>	MapReduceEntriesToDoubleTask<K,V> rights, nextRight;
5484	>	MapReduceEntriesToDoubleTask
5485	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5486	>	MapReduceEntriesToDoubleTask<K,V> nextRight,
5487	>	ObjectToDouble<Map.Entry<K,V>> transformer,
5488	>	double basis,
5489	>	DoubleByDoubleToDouble reducer) {
5490	>	super(p, b, i, f, t); this.nextRight = nextRight;
5491	>	this.transformer = transformer;
5492	>	this.basis = basis; this.reducer = reducer;
5493	>	}
5494	>	public final Double getRawResult() { return result; }
5495	>	public final void compute() {
5496	>	final ObjectToDouble<Map.Entry<K,V>> transformer;
5497	>	final DoubleByDoubleToDouble reducer;
5498	>	if ((transformer = this.transformer) != null &&
5499	>	(reducer = this.reducer) != null) {
5500	>	double r = this.basis;
5501	>	for (int i = baseIndex, f, h; batch > 0 &&
5502	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5503	>	addToPendingCount(1);
5504	>	(rights = new MapReduceEntriesToDoubleTask<K,V>
5505	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5506	>	rights, transformer, r, reducer)).fork();
5507	>	}
5508	>	for (Node<K,V> p; (p = advance()) != null; )
5509	>	r = reducer.apply(r, transformer.apply(p));
5510	>	result = r;
5511	>	CountedCompleter<?> c;
5512	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5513	>	@SuppressWarnings("unchecked") MapReduceEntriesToDoubleTask<K,V>
5514	>	t = (MapReduceEntriesToDoubleTask<K,V>)c,
5515	>	s = t.rights;
5516	>	while (s != null) {
5517	>	t.result = reducer.apply(t.result, s.result);
5518	>	s = t.rights = s.nextRight;
5519	>	}
5520	>	}
5521	>	}
5522	>	}
5523	>	}
5524	>
5525	>	@SuppressWarnings("serial")
5526	>	static final class MapReduceMappingsToDoubleTask<K,V>
5527	>	extends BulkTask<K,V,Double> {
5528	>	final ObjectByObjectToDouble<? super K, ? super V> transformer;
5529	>	final DoubleByDoubleToDouble reducer;
5530	>	final double basis;
5531	>	double result;
5532	>	MapReduceMappingsToDoubleTask<K,V> rights, nextRight;
5533	>	MapReduceMappingsToDoubleTask
5534	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5535	>	MapReduceMappingsToDoubleTask<K,V> nextRight,
5536	>	ObjectByObjectToDouble<? super K, ? super V> transformer,
5537	>	double basis,
5538	>	DoubleByDoubleToDouble reducer) {
5539	>	super(p, b, i, f, t); this.nextRight = nextRight;
5540	>	this.transformer = transformer;
5541	>	this.basis = basis; this.reducer = reducer;
5542	>	}
5543	>	public final Double getRawResult() { return result; }
5544	>	public final void compute() {
5545	>	final ObjectByObjectToDouble<? super K, ? super V> transformer;
5546	>	final DoubleByDoubleToDouble reducer;
5547	>	if ((transformer = this.transformer) != null &&
5548	>	(reducer = this.reducer) != null) {
5549	>	double r = this.basis;
5550	>	for (int i = baseIndex, f, h; batch > 0 &&
5551	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5552	>	addToPendingCount(1);
5553	>	(rights = new MapReduceMappingsToDoubleTask<K,V>
5554	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5555	>	rights, transformer, r, reducer)).fork();
5556	>	}
5557	>	for (Node<K,V> p; (p = advance()) != null; )
5558	>	r = reducer.apply(r, transformer.apply(p.key, p.val));
5559	>	result = r;
5560	>	CountedCompleter<?> c;
5561	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5562	>	@SuppressWarnings("unchecked") MapReduceMappingsToDoubleTask<K,V>
5563	>	t = (MapReduceMappingsToDoubleTask<K,V>)c,
5564	>	s = t.rights;
5565	>	while (s != null) {
5566	>	t.result = reducer.apply(t.result, s.result);
5567	>	s = t.rights = s.nextRight;
5568	>	}
5569	>	}
5570	>	}
5571	>	}
5572	>	}
5573	>
5574	>	@SuppressWarnings("serial")
5575	>	static final class MapReduceKeysToLongTask<K,V>
5576	>	extends BulkTask<K,V,Long> {
5577	>	final ObjectToLong<? super K> transformer;
5578	>	final LongByLongToLong reducer;
5579	>	final long basis;
5580	>	long result;
5581	>	MapReduceKeysToLongTask<K,V> rights, nextRight;
5582	>	MapReduceKeysToLongTask
5583	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5584	>	MapReduceKeysToLongTask<K,V> nextRight,
5585	>	ObjectToLong<? super K> transformer,
5586	>	long basis,
5587	>	LongByLongToLong reducer) {
5588	>	super(p, b, i, f, t); this.nextRight = nextRight;
5589	>	this.transformer = transformer;
5590	>	this.basis = basis; this.reducer = reducer;
5591	>	}
5592	>	public final Long getRawResult() { return result; }
5593	>	public final void compute() {
5594	>	final ObjectToLong<? super K> transformer;
5595	>	final LongByLongToLong reducer;
5596	>	if ((transformer = this.transformer) != null &&
5597	>	(reducer = this.reducer) != null) {
5598	>	long r = this.basis;
5599	>	for (int i = baseIndex, f, h; batch > 0 &&
5600	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5601	>	addToPendingCount(1);
5602	>	(rights = new MapReduceKeysToLongTask<K,V>
5603	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5604	>	rights, transformer, r, reducer)).fork();
5605	>	}
5606	>	for (Node<K,V> p; (p = advance()) != null; )
5607	>	r = reducer.apply(r, transformer.apply(p.key));
5608	>	result = r;
5609	>	CountedCompleter<?> c;
5610	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5611	>	@SuppressWarnings("unchecked") MapReduceKeysToLongTask<K,V>
5612	>	t = (MapReduceKeysToLongTask<K,V>)c,
5613	>	s = t.rights;
5614	>	while (s != null) {
5615	>	t.result = reducer.apply(t.result, s.result);
5616	>	s = t.rights = s.nextRight;
5617	>	}
5618	>	}
5619	>	}
5620	>	}
5621	>	}
5622	>
5623	>	@SuppressWarnings("serial")
5624	>	static final class MapReduceValuesToLongTask<K,V>
5625	>	extends BulkTask<K,V,Long> {
5626	>	final ObjectToLong<? super V> transformer;
5627	>	final LongByLongToLong reducer;
5628	>	final long basis;
5629	>	long result;
5630	>	MapReduceValuesToLongTask<K,V> rights, nextRight;
5631	>	MapReduceValuesToLongTask
5632	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5633	>	MapReduceValuesToLongTask<K,V> nextRight,
5634	>	ObjectToLong<? super V> transformer,
5635	>	long basis,
5636	>	LongByLongToLong reducer) {
5637	>	super(p, b, i, f, t); this.nextRight = nextRight;
5638	>	this.transformer = transformer;
5639	>	this.basis = basis; this.reducer = reducer;
5640	>	}
5641	>	public final Long getRawResult() { return result; }
5642	>	public final void compute() {
5643	>	final ObjectToLong<? super V> transformer;
5644	>	final LongByLongToLong reducer;
5645	>	if ((transformer = this.transformer) != null &&
5646	>	(reducer = this.reducer) != null) {
5647	>	long r = this.basis;
5648	>	for (int i = baseIndex, f, h; batch > 0 &&
5649	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5650	>	addToPendingCount(1);
5651	>	(rights = new MapReduceValuesToLongTask<K,V>
5652	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5653	>	rights, transformer, r, reducer)).fork();
5654	>	}
5655	>	for (Node<K,V> p; (p = advance()) != null; )
5656	>	r = reducer.apply(r, transformer.apply(p.val));
5657	>	result = r;
5658	>	CountedCompleter<?> c;
5659	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5660	>	@SuppressWarnings("unchecked") MapReduceValuesToLongTask<K,V>
5661	>	t = (MapReduceValuesToLongTask<K,V>)c,
5662	>	s = t.rights;
5663	>	while (s != null) {
5664	>	t.result = reducer.apply(t.result, s.result);
5665	>	s = t.rights = s.nextRight;
5666	>	}
5667	>	}
5668	>	}
5669	>	}
5670	>	}
5671	>
5672	>	@SuppressWarnings("serial")
5673	>	static final class MapReduceEntriesToLongTask<K,V>
5674	>	extends BulkTask<K,V,Long> {
5675	>	final ObjectToLong<Map.Entry<K,V>> transformer;
5676	>	final LongByLongToLong reducer;
5677	>	final long basis;
5678	>	long result;
5679	>	MapReduceEntriesToLongTask<K,V> rights, nextRight;
5680	>	MapReduceEntriesToLongTask
5681	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5682	>	MapReduceEntriesToLongTask<K,V> nextRight,
5683	>	ObjectToLong<Map.Entry<K,V>> transformer,
5684	>	long basis,
5685	>	LongByLongToLong reducer) {
5686	>	super(p, b, i, f, t); this.nextRight = nextRight;
5687	>	this.transformer = transformer;
5688	>	this.basis = basis; this.reducer = reducer;
5689	>	}
5690	>	public final Long getRawResult() { return result; }
5691	>	public final void compute() {
5692	>	final ObjectToLong<Map.Entry<K,V>> transformer;
5693	>	final LongByLongToLong reducer;
5694	>	if ((transformer = this.transformer) != null &&
5695	>	(reducer = this.reducer) != null) {
5696	>	long r = this.basis;
5697	>	for (int i = baseIndex, f, h; batch > 0 &&
5698	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5699	>	addToPendingCount(1);
5700	>	(rights = new MapReduceEntriesToLongTask<K,V>
5701	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5702	>	rights, transformer, r, reducer)).fork();
5703	>	}
5704	>	for (Node<K,V> p; (p = advance()) != null; )
5705	>	r = reducer.apply(r, transformer.apply(p));
5706	>	result = r;
5707	>	CountedCompleter<?> c;
5708	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5709	>	@SuppressWarnings("unchecked") MapReduceEntriesToLongTask<K,V>
5710	>	t = (MapReduceEntriesToLongTask<K,V>)c,
5711	>	s = t.rights;
5712	>	while (s != null) {
5713	>	t.result = reducer.apply(t.result, s.result);
5714	>	s = t.rights = s.nextRight;
5715	>	}
5716	>	}
5717	>	}
5718	>	}
5719	>	}
5720	>
5721	>	@SuppressWarnings("serial")
5722	>	static final class MapReduceMappingsToLongTask<K,V>
5723	>	extends BulkTask<K,V,Long> {
5724	>	final ObjectByObjectToLong<? super K, ? super V> transformer;
5725	>	final LongByLongToLong reducer;
5726	>	final long basis;
5727	>	long result;
5728	>	MapReduceMappingsToLongTask<K,V> rights, nextRight;
5729	>	MapReduceMappingsToLongTask
5730	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5731	>	MapReduceMappingsToLongTask<K,V> nextRight,
5732	>	ObjectByObjectToLong<? super K, ? super V> transformer,
5733	>	long basis,
5734	>	LongByLongToLong reducer) {
5735	>	super(p, b, i, f, t); this.nextRight = nextRight;
5736	>	this.transformer = transformer;
5737	>	this.basis = basis; this.reducer = reducer;
5738	>	}
5739	>	public final Long getRawResult() { return result; }
5740	>	public final void compute() {
5741	>	final ObjectByObjectToLong<? super K, ? super V> transformer;
5742	>	final LongByLongToLong reducer;
5743	>	if ((transformer = this.transformer) != null &&
5744	>	(reducer = this.reducer) != null) {
5745	>	long r = this.basis;
5746	>	for (int i = baseIndex, f, h; batch > 0 &&
5747	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5748	>	addToPendingCount(1);
5749	>	(rights = new MapReduceMappingsToLongTask<K,V>
5750	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5751	>	rights, transformer, r, reducer)).fork();
5752	>	}
5753	>	for (Node<K,V> p; (p = advance()) != null; )
5754	>	r = reducer.apply(r, transformer.apply(p.key, p.val));
5755	>	result = r;
5756	>	CountedCompleter<?> c;
5757	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5758	>	@SuppressWarnings("unchecked") MapReduceMappingsToLongTask<K,V>
5759	>	t = (MapReduceMappingsToLongTask<K,V>)c,
5760	>	s = t.rights;
5761	>	while (s != null) {
5762	>	t.result = reducer.apply(t.result, s.result);
5763	>	s = t.rights = s.nextRight;
5764	>	}
5765	>	}
5766	>	}
5767	>	}
5768	>	}
5769	>
5770	>	@SuppressWarnings("serial")
5771	>	static final class MapReduceKeysToIntTask<K,V>
5772	>	extends BulkTask<K,V,Integer> {
5773	>	final ObjectToInt<? super K> transformer;
5774	>	final IntByIntToInt reducer;
5775	>	final int basis;
5776	>	int result;
5777	>	MapReduceKeysToIntTask<K,V> rights, nextRight;
5778	>	MapReduceKeysToIntTask
5779	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5780	>	MapReduceKeysToIntTask<K,V> nextRight,
5781	>	ObjectToInt<? super K> transformer,
5782	>	int basis,
5783	>	IntByIntToInt reducer) {
5784	>	super(p, b, i, f, t); this.nextRight = nextRight;
5785	>	this.transformer = transformer;
5786	>	this.basis = basis; this.reducer = reducer;
5787	>	}
5788	>	public final Integer getRawResult() { return result; }
5789	>	public final void compute() {
5790	>	final ObjectToInt<? super K> transformer;
5791	>	final IntByIntToInt reducer;
5792	>	if ((transformer = this.transformer) != null &&
5793	>	(reducer = this.reducer) != null) {
5794	>	int r = this.basis;
5795	>	for (int i = baseIndex, f, h; batch > 0 &&
5796	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5797	>	addToPendingCount(1);
5798	>	(rights = new MapReduceKeysToIntTask<K,V>
5799	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5800	>	rights, transformer, r, reducer)).fork();
5801	>	}
5802	>	for (Node<K,V> p; (p = advance()) != null; )
5803	>	r = reducer.apply(r, transformer.apply(p.key));
5804	>	result = r;
5805	>	CountedCompleter<?> c;
5806	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5807	>	@SuppressWarnings("unchecked") MapReduceKeysToIntTask<K,V>
5808	>	t = (MapReduceKeysToIntTask<K,V>)c,
5809	>	s = t.rights;
5810	>	while (s != null) {
5811	>	t.result = reducer.apply(t.result, s.result);
5812	>	s = t.rights = s.nextRight;
5813	>	}
5814	>	}
5815	>	}
5816	>	}
5817	>	}
5818	>
5819	>	@SuppressWarnings("serial")
5820	>	static final class MapReduceValuesToIntTask<K,V>
5821	>	extends BulkTask<K,V,Integer> {
5822	>	final ObjectToInt<? super V> transformer;
5823	>	final IntByIntToInt reducer;
5824	>	final int basis;
5825	>	int result;
5826	>	MapReduceValuesToIntTask<K,V> rights, nextRight;
5827	>	MapReduceValuesToIntTask
5828	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5829	>	MapReduceValuesToIntTask<K,V> nextRight,
5830	>	ObjectToInt<? super V> transformer,
5831	>	int basis,
5832	>	IntByIntToInt reducer) {
5833	>	super(p, b, i, f, t); this.nextRight = nextRight;
5834	>	this.transformer = transformer;
5835	>	this.basis = basis; this.reducer = reducer;
5836	>	}
5837	>	public final Integer getRawResult() { return result; }
5838	>	public final void compute() {
5839	>	final ObjectToInt<? super V> transformer;
5840	>	final IntByIntToInt reducer;
5841	>	if ((transformer = this.transformer) != null &&
5842	>	(reducer = this.reducer) != null) {
5843	>	int r = this.basis;
5844	>	for (int i = baseIndex, f, h; batch > 0 &&
5845	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5846	>	addToPendingCount(1);
5847	>	(rights = new MapReduceValuesToIntTask<K,V>
5848	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5849	>	rights, transformer, r, reducer)).fork();
5850	>	}
5851	>	for (Node<K,V> p; (p = advance()) != null; )
5852	>	r = reducer.apply(r, transformer.apply(p.val));
5853	>	result = r;
5854	>	CountedCompleter<?> c;
5855	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5856	>	@SuppressWarnings("unchecked") MapReduceValuesToIntTask<K,V>
5857	>	t = (MapReduceValuesToIntTask<K,V>)c,
5858	>	s = t.rights;
5859	>	while (s != null) {
5860	>	t.result = reducer.apply(t.result, s.result);
5861	>	s = t.rights = s.nextRight;
5862	>	}
5863	>	}
5864	>	}
5865	>	}
5866	>	}
5867	>
5868	>	@SuppressWarnings("serial")
5869	>	static final class MapReduceEntriesToIntTask<K,V>
5870	>	extends BulkTask<K,V,Integer> {
5871	>	final ObjectToInt<Map.Entry<K,V>> transformer;
5872	>	final IntByIntToInt reducer;
5873	>	final int basis;
5874	>	int result;
5875	>	MapReduceEntriesToIntTask<K,V> rights, nextRight;
5876	>	MapReduceEntriesToIntTask
5877	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5878	>	MapReduceEntriesToIntTask<K,V> nextRight,
5879	>	ObjectToInt<Map.Entry<K,V>> transformer,
5880	>	int basis,
5881	>	IntByIntToInt reducer) {
5882	>	super(p, b, i, f, t); this.nextRight = nextRight;
5883	>	this.transformer = transformer;
5884	>	this.basis = basis; this.reducer = reducer;
5885	>	}
5886	>	public final Integer getRawResult() { return result; }
5887	>	public final void compute() {
5888	>	final ObjectToInt<Map.Entry<K,V>> transformer;
5889	>	final IntByIntToInt reducer;
5890	>	if ((transformer = this.transformer) != null &&
5891	>	(reducer = this.reducer) != null) {
5892	>	int r = this.basis;
5893	>	for (int i = baseIndex, f, h; batch > 0 &&
5894	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5895	>	addToPendingCount(1);
5896	>	(rights = new MapReduceEntriesToIntTask<K,V>
5897	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5898	>	rights, transformer, r, reducer)).fork();
5899	>	}
5900	>	for (Node<K,V> p; (p = advance()) != null; )
5901	>	r = reducer.apply(r, transformer.apply(p));
5902	>	result = r;
5903	>	CountedCompleter<?> c;
5904	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5905	>	@SuppressWarnings("unchecked") MapReduceEntriesToIntTask<K,V>
5906	>	t = (MapReduceEntriesToIntTask<K,V>)c,
5907	>	s = t.rights;
5908	>	while (s != null) {
5909	>	t.result = reducer.apply(t.result, s.result);
5910	>	s = t.rights = s.nextRight;
5911	>	}
5912	>	}
5913	>	}
5914	>	}
5915	>	}
5916	>
5917	>	@SuppressWarnings("serial")
5918	>	static final class MapReduceMappingsToIntTask<K,V>
5919	>	extends BulkTask<K,V,Integer> {
5920	>	final ObjectByObjectToInt<? super K, ? super V> transformer;
5921	>	final IntByIntToInt reducer;
5922	>	final int basis;
5923	>	int result;
5924	>	MapReduceMappingsToIntTask<K,V> rights, nextRight;
5925	>	MapReduceMappingsToIntTask
5926	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5927	>	MapReduceMappingsToIntTask<K,V> nextRight,
5928	>	ObjectByObjectToInt<? super K, ? super V> transformer,
5929	>	int basis,
5930	>	IntByIntToInt reducer) {
5931	>	super(p, b, i, f, t); this.nextRight = nextRight;
5932	>	this.transformer = transformer;
5933	>	this.basis = basis; this.reducer = reducer;
5934	>	}
5935	>	public final Integer getRawResult() { return result; }
5936	>	public final void compute() {
5937	>	final ObjectByObjectToInt<? super K, ? super V> transformer;
5938	>	final IntByIntToInt reducer;
5939	>	if ((transformer = this.transformer) != null &&
5940	>	(reducer = this.reducer) != null) {
5941	>	int r = this.basis;
5942	>	for (int i = baseIndex, f, h; batch > 0 &&
5943	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5944	>	addToPendingCount(1);
5945	>	(rights = new MapReduceMappingsToIntTask<K,V>
5946	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5947	>	rights, transformer, r, reducer)).fork();
5948	>	}
5949	>	for (Node<K,V> p; (p = advance()) != null; )
5950	>	r = reducer.apply(r, transformer.apply(p.key, p.val));
5951	>	result = r;
5952	>	CountedCompleter<?> c;
5953	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5954	>	@SuppressWarnings("unchecked") MapReduceMappingsToIntTask<K,V>
5955	>	t = (MapReduceMappingsToIntTask<K,V>)c,
5956	>	s = t.rights;
5957	>	while (s != null) {
5958	>	t.result = reducer.apply(t.result, s.result);
5959	>	s = t.rights = s.nextRight;
5960	>	}
5961	>	}
5962	>	}
5963	>	}
5964	>	}
5965	>
5966	>	/* ---------------- Counters -------------- */
5967	>
5968	>	// Adapted from LongAdder and Striped64.
5969	>	// See their internal docs for explanation.
5970	>
5971	>	// A padded cell for distributing counts
5972	>	static final class CounterCell {
5973	>	volatile long p0, p1, p2, p3, p4, p5, p6;
5974	>	volatile long value;
5975	>	volatile long q0, q1, q2, q3, q4, q5, q6;
5976	>	CounterCell(long x) { value = x; }
5977		}
5978
5979		/**
5980	<	* Saves the state of the {@code ConcurrentHashMapV8} instance to a
5981	<	* stream (i.e., serializes it).
5982	<	* @param s the stream
2327	<	* @serialData
2328	<	* the key (Object) and value (Object)
2329	<	* for each key-value mapping, followed by a null pair.
2330	<	* The key-value mappings are emitted in no particular order.
5980	>	* Holder for the thread-local hash code determining which
5981	>	* CounterCell to use. The code is initialized via the
5982	>	* counterHashCodeGenerator, but may be moved upon collisions.
5983		*/
5984	<	@SuppressWarnings("unchecked")
5985	<	private void writeObject(java.io.ObjectOutputStream s)
2334	<	throws java.io.IOException {
2335	<	if (segments == null) { // for serialization compatibility
2336	<	segments = (Segment<K,V>[])
2337	<	new Segment<?,?>[DEFAULT_CONCURRENCY_LEVEL];
2338	<	for (int i = 0; i < segments.length; ++i)
2339	<	segments[i] = new Segment<K,V>(LOAD_FACTOR);
2340	<	}
2341	<	s.defaultWriteObject();
2342	<	InternalIterator it = new InternalIterator(table);
2343	<	while (it.next != null) {
2344	<	s.writeObject(it.nextKey);
2345	<	s.writeObject(it.nextVal);
2346	<	it.advance();
2347	<	}
2348	<	s.writeObject(null);
2349	<	s.writeObject(null);
2350	<	segments = null; // throw away
5984	>	static final class CounterHashCode {
5985	>	int code;
5986		}
5987
5988		/**
5989	<	* Reconstitutes the instance from a stream (that is, deserializes it).
2355	<	* @param s the stream
5989	>	* Generates initial value for per-thread CounterHashCodes.
5990		*/
5991	<	@SuppressWarnings("unchecked")
2358	<	private void readObject(java.io.ObjectInputStream s)
2359	<	throws java.io.IOException, ClassNotFoundException {
2360	<	s.defaultReadObject();
2361	<	this.segments = null; // unneeded
2362	<	// initialize transient final field
2363	<	UNSAFE.putObjectVolatile(this, counterOffset, new LongAdder());
5991	>	static final AtomicInteger counterHashCodeGenerator = new AtomicInteger();
5992
5993	<	// Create all nodes, then place in table once size is known
5994	<	long size = 0L;
5995	<	Node p = null;
5996	<	for (;;) {
5997	<	K k = (K) s.readObject();
5998	<	V v = (V) s.readObject();
5999	<	if (k != null && v != null) {
6000	<	p = new Node(spread(k.hashCode()), k, v, p);
6001	<	++size;
5993	>	/**
5994	>	* Increment for counterHashCodeGenerator. See class ThreadLocal
5995	>	* for explanation.
5996	>	*/
5997	>	static final int SEED_INCREMENT = 0x61c88647;
5998	>
5999	>	/**
6000	>	* Per-thread counter hash codes. Shared across all instances.
6001	>	*/
6002	>	static final ThreadLocal<CounterHashCode> threadCounterHashCode =
6003	>	new ThreadLocal<CounterHashCode>();
6004	>
6005	>
6006	>	final long sumCount() {
6007	>	CounterCell[] as = counterCells; CounterCell a;
6008	>	long sum = baseCount;
6009	>	if (as != null) {
6010	>	for (int i = 0; i < as.length; ++i) {
6011	>	if ((a = as[i]) != null)
6012	>	sum += a.value;
6013		}
2375	–	else
2376	–	break;
6014		}
6015	<	if (p != null) {
6016	<	boolean init = false;
6017	<	int n;
6018	<	if (size >= (long)(MAXIMUM_CAPACITY >>> 1))
6019	<	n = MAXIMUM_CAPACITY;
6020	<	else {
6021	<	int sz = (int)size;
6022	<	n = tableSizeFor(sz + (sz >>> 1) + 1);
6023	<	}
6024	<	int sc = sizeCtl;
6025	<	if (n > sc &&
6026	<	UNSAFE.compareAndSwapInt(this, sizeCtlOffset, sc, -1)) {
6027	<	try {
6028	<	if (table == null) {
6029	<	init = true;
6030	<	Node[] tab = new Node[n];
6031	<	int mask = n - 1;
6032	<	while (p != null) {
6033	<	int j = p.hash & mask;
6034	<	Node next = p.next;
6035	<	p.next = tabAt(tab, j);
6036	<	setTabAt(tab, j, p);
6037	<	p = next;
6015	>	return sum;
6016	>	}
6017	>
6018	>	// See LongAdder version for explanation
6019	>	private final void fullAddCount(long x, CounterHashCode hc,
6020	>	boolean wasUncontended) {
6021	>	int h;
6022	>	if (hc == null) {
6023	>	hc = new CounterHashCode();
6024	>	int s = counterHashCodeGenerator.addAndGet(SEED_INCREMENT);
6025	>	h = hc.code = (s == 0) ? 1 : s; // Avoid zero
6026	>	threadCounterHashCode.set(hc);
6027	>	}
6028	>	else
6029	>	h = hc.code;
6030	>	boolean collide = false;                // True if last slot nonempty
6031	>	for (;;) {
6032	>	CounterCell[] as; CounterCell a; int n; long v;
6033	>	if ((as = counterCells) != null && (n = as.length) > 0) {
6034	>	if ((a = as[(n - 1) & h]) == null) {
6035	>	if (cellsBusy == 0) {            // Try to attach new Cell
6036	>	CounterCell r = new CounterCell(x); // Optimistic create
6037	>	if (cellsBusy == 0 &&
6038	>	U.compareAndSwapInt(this, CELLSBUSY, 0, 1)) {
6039	>	boolean created = false;
6040	>	try {               // Recheck under lock
6041	>	CounterCell[] rs; int m, j;
6042	>	if ((rs = counterCells) != null &&
6043	>	(m = rs.length) > 0 &&
6044	>	rs[j = (m - 1) & h] == null) {
6045	>	rs[j] = r;
6046	>	created = true;
6047	>	}
6048	>	} finally {
6049	>	cellsBusy = 0;
6050	>	}
6051	>	if (created)
6052	>	break;
6053	>	continue;           // Slot is now non-empty
6054		}
2402	–	table = tab;
2403	–	counter.add(size);
2404	–	sc = n - (n >>> 2);
6055		}
6056	<	} finally {
2407	<	sizeCtl = sc;
6056	>	collide = false;
6057		}
6058	<	}
6059	<	if (!init) { // Can only happen if unsafely published.
6060	<	while (p != null) {
6061	<	internalPut(p.key, p.val);
6062	<	p = p.next;
6058	>	else if (!wasUncontended)       // CAS already known to fail
6059	>	wasUncontended = true;      // Continue after rehash
6060	>	else if (U.compareAndSwapLong(a, CELLVALUE, v = a.value, v + x))
6061	>	break;
6062	>	else if (counterCells != as || n >= NCPU)
6063	>	collide = false;            // At max size or stale
6064	>	else if (!collide)
6065	>	collide = true;
6066	>	else if (cellsBusy == 0 &&
6067	>	U.compareAndSwapInt(this, CELLSBUSY, 0, 1)) {
6068	>	try {
6069	>	if (counterCells == as) {// Expand table unless stale
6070	>	CounterCell[] rs = new CounterCell[n << 1];
6071	>	for (int i = 0; i < n; ++i)
6072	>	rs[i] = as[i];
6073	>	counterCells = rs;
6074	>	}
6075	>	} finally {
6076	>	cellsBusy = 0;
6077	>	}
6078	>	collide = false;
6079	>	continue;                   // Retry with expanded table
6080	>	}
6081	>	h ^= h << 13;                   // Rehash
6082	>	h ^= h >>> 17;
6083	>	h ^= h << 5;
6084	>	}
6085	>	else if (cellsBusy == 0 && counterCells == as &&
6086	>	U.compareAndSwapInt(this, CELLSBUSY, 0, 1)) {
6087	>	boolean init = false;
6088	>	try {                           // Initialize table
6089	>	if (counterCells == as) {
6090	>	CounterCell[] rs = new CounterCell[2];
6091	>	rs[h & 1] = new CounterCell(x);
6092	>	counterCells = rs;
6093	>	init = true;
6094	>	}
6095	>	} finally {
6096	>	cellsBusy = 0;
6097		}
6098	+	if (init)
6099	+	break;
6100		}
6101	+	else if (U.compareAndSwapLong(this, BASECOUNT, v = baseCount, v + x))
6102	+	break;                          // Fall back on using base
6103		}
6104	+	hc.code = h;                            // Record index for next time
6105		}
6106
6107		// Unsafe mechanics
6108	<	private static final sun.misc.Unsafe UNSAFE;
6109	<	private static final long counterOffset;
6110	<	private static final long sizeCtlOffset;
6108	>	private static final sun.misc.Unsafe U;
6109	>	private static final long SIZECTL;
6110	>	private static final long TRANSFERINDEX;
6111	>	private static final long TRANSFERORIGIN;
6112	>	private static final long BASECOUNT;
6113	>	private static final long CELLSBUSY;
6114	>	private static final long CELLVALUE;
6115		private static final long ABASE;
6116		private static final int ASHIFT;
6117
6118		static {
2427	–	int ss;
6119		try {
6120	<	UNSAFE = getUnsafe();
6120	>	U = getUnsafe();
6121		Class<?> k = ConcurrentHashMapV8.class;
6122	<	counterOffset = UNSAFE.objectFieldOffset
2432	<	(k.getDeclaredField("counter"));
2433	<	sizeCtlOffset = UNSAFE.objectFieldOffset
6122	>	SIZECTL = U.objectFieldOffset
6123		(k.getDeclaredField("sizeCtl"));
6124	<	Class<?> sc = Node[].class;
6125	<	ABASE = UNSAFE.arrayBaseOffset(sc);
6126	<	ss = UNSAFE.arrayIndexScale(sc);
6124	>	TRANSFERINDEX = U.objectFieldOffset
6125	>	(k.getDeclaredField("transferIndex"));
6126	>	TRANSFERORIGIN = U.objectFieldOffset
6127	>	(k.getDeclaredField("transferOrigin"));
6128	>	BASECOUNT = U.objectFieldOffset
6129	>	(k.getDeclaredField("baseCount"));
6130	>	CELLSBUSY = U.objectFieldOffset
6131	>	(k.getDeclaredField("cellsBusy"));
6132	>	Class<?> ck = CounterCell.class;
6133	>	CELLVALUE = U.objectFieldOffset
6134	>	(ck.getDeclaredField("value"));
6135	>	Class<?> ak = Node[].class;
6136	>	ABASE = U.arrayBaseOffset(ak);
6137	>	int scale = U.arrayIndexScale(ak);
6138	>	if ((scale & (scale - 1)) != 0)
6139	>	throw new Error("data type scale not a power of two");
6140	>	ASHIFT = 31 - Integer.numberOfLeadingZeros(scale);
6141		} catch (Exception e) {
6142		throw new Error(e);
6143		}
2441	–	if ((ss & (ss-1)) != 0)
2442	–	throw new Error("data type scale not a power of two");
2443	–	ASHIFT = 31 - Integer.numberOfLeadingZeros(ss);
6144		}
6145
6146		/**
#	Line 2453 | Line 6153 | public class ConcurrentHashMapV8<K, V>
6153		private static sun.misc.Unsafe getUnsafe() {
6154		try {
6155		return sun.misc.Unsafe.getUnsafe();
6156	<	} catch (SecurityException se) {
6157	<	try {
6158	<	return java.security.AccessController.doPrivileged
6159	<	(new java.security
6160	<	.PrivilegedExceptionAction<sun.misc.Unsafe>() {
6161	<	public sun.misc.Unsafe run() throws Exception {
6162	<	java.lang.reflect.Field f = sun.misc
6163	<	.Unsafe.class.getDeclaredField("theUnsafe");
6164	<	f.setAccessible(true);
6165	<	return (sun.misc.Unsafe) f.get(null);
6166	<	}});
6167	<	} catch (java.security.PrivilegedActionException e) {
6168	<	throw new RuntimeException("Could not initialize intrinsics",
6169	<	e.getCause());
6170	<	}
6156	>	} catch (SecurityException tryReflectionInstead) {}
6157	>	try {
6158	>	return java.security.AccessController.doPrivileged
6159	>	(new java.security.PrivilegedExceptionAction<sun.misc.Unsafe>() {
6160	>	public sun.misc.Unsafe run() throws Exception {
6161	>	Class<sun.misc.Unsafe> k = sun.misc.Unsafe.class;
6162	>	for (java.lang.reflect.Field f : k.getDeclaredFields()) {
6163	>	f.setAccessible(true);
6164	>	Object x = f.get(null);
6165	>	if (k.isInstance(x))
6166	>	return k.cast(x);
6167	>	}
6168	>	throw new NoSuchFieldError("the Unsafe");
6169	>	}});
6170	>	} catch (java.security.PrivilegedActionException e) {
6171	>	throw new RuntimeException("Could not initialize intrinsics",
6172	>	e.getCause());
6173		}
6174		}
2473	–
6175		}

Diff Legend

–	Removed lines
+	Added lines
<	Changed lines
>	Changed lines