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 1.12 by jsr166, Tue Aug 30 18:31:54 2011 UTC vs.
Revision 1.116 by dl, Wed Sep 11 14:53:38 2013 UTC

#	Line 5 | Line 5
5		*/
6
7		package jsr166e;
8	<	import jsr166e.LongAdder;
9	<	import java.util.Map;
10	<	import java.util.Set;
11	<	import java.util.Collection;
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.AbstractMap;
16	<	import java.util.AbstractSet;
17	<	import java.util.AbstractCollection;
18	<	import java.util.Hashtable;
16	>	import java.util.Arrays;
17	>	import java.util.Collection;
18	>	import java.util.Comparator;
19	>	import java.util.ConcurrentModificationException;
20	>	import java.util.Enumeration;
21		import java.util.HashMap;
22	+	import java.util.Hashtable;
23		import java.util.Iterator;
24	<	import java.util.Enumeration;
19	<	import java.util.ConcurrentModificationException;
24	>	import java.util.Map;
25		import java.util.NoSuchElementException;
26	+	import java.util.Set;
27		import java.util.concurrent.ConcurrentMap;
28	<	import java.io.Serializable;
28	>	import java.util.concurrent.atomic.AtomicReference;
29	>	import java.util.concurrent.atomic.AtomicInteger;
30	>	import java.util.concurrent.locks.LockSupport;
31	>	import java.util.concurrent.locks.ReentrantLock;
32
33		/**
34		* A hash table supporting full concurrency of retrievals and
#	Line 33 | Line 42 | import java.io.Serializable;
42		* interoperable with {@code Hashtable} in programs that rely on its
43		* thread safety but not on its synchronization details.
44		*
45	<	* <p> Retrieval operations (including {@code get}) generally do not
45	>	* <p>Retrieval operations (including {@code get}) generally do not
46		* block, so may overlap with update operations (including {@code put}
47		* and {@code remove}). Retrievals reflect the results of the most
48		* recently <em>completed</em> update operations holding upon their
49	<	* onset.  For aggregate operations such as {@code putAll} and {@code
50	<	* clear}, concurrent retrievals may reflect insertion or removal of
51	<	* only some entries.  Similarly, Iterators and Enumerations return
52	<	* elements reflecting the state of the hash table at some point at or
53	<	* since the creation of the iterator/enumeration.  They do
54	<	* <em>not</em> throw {@link ConcurrentModificationException}.
55	<	* However, iterators are designed to be used by only one thread at a
56	<	* time.  Bear in mind that the results of aggregate status methods
57	<	* including {@code size}, {@code isEmpty}, and {@code containsValue}
58	<	* are typically useful only when a map is not undergoing concurrent
59	<	* updates in other threads.  Otherwise the results of these methods
60	<	* reflect transient states that may be adequate for monitoring
61	<	* purposes, but not for program control.
62	<	*
63	<	* <p> Resizing this or any other kind of hash table is a relatively
64	<	* slow operation, so, when possible, it is a good idea to provide
65	<	* estimates of expected table sizes in constructors. Also, for
66	<	* compatibility with previous versions of this class, constructors
67	<	* may optionally specify an expected {@code concurrencyLevel} as an
68	<	* additional hint for internal sizing.
49	>	* onset. (More formally, an update operation for a given key bears a
50	>	* <em>happens-before</em> relation with any (non-null) retrieval for
51	>	* that key reporting the updated value.)  For aggregate operations
52	>	* such as {@code putAll} and {@code clear}, concurrent retrievals may
53	>	* reflect insertion or removal of only some entries.  Similarly,
54	>	* Iterators and Enumerations return elements reflecting the state of
55	>	* the hash table at some point at or since the creation of the
56	>	* iterator/enumeration.  They do <em>not</em> throw {@link
57	>	* ConcurrentModificationException}.  However, iterators are designed
58	>	* to be used by only one thread at a time.  Bear in mind that the
59	>	* results of aggregate status methods including {@code size}, {@code
60	>	* isEmpty}, and {@code containsValue} are typically useful only when
61	>	* a map is not undergoing concurrent updates in other threads.
62	>	* Otherwise the results of these methods reflect transient states
63	>	* that may be adequate for monitoring or estimation purposes, but not
64	>	* for program control.
65	>	*
66	>	* <p>The table is dynamically expanded when there are too many
67	>	* collisions (i.e., keys that have distinct hash codes but fall into
68	>	* the same slot modulo the table size), with the expected average
69	>	* effect of maintaining roughly two bins per mapping (corresponding
70	>	* to a 0.75 load factor threshold for resizing). There may be much
71	>	* variance around this average as mappings are added and removed, but
72	>	* overall, this maintains a commonly accepted time/space tradeoff for
73	>	* hash tables.  However, resizing this or any other kind of hash
74	>	* table may be a relatively slow operation. When possible, it is a
75	>	* good idea to provide a size estimate as an optional {@code
76	>	* initialCapacity} constructor argument. An additional optional
77	>	* {@code loadFactor} constructor argument provides a further means of
78	>	* customizing initial table capacity by specifying the table density
79	>	* to be used in calculating the amount of space to allocate for the
80	>	* given number of elements.  Also, for compatibility with previous
81	>	* versions of this class, constructors may optionally specify an
82	>	* expected {@code concurrencyLevel} as an additional hint for
83	>	* internal sizing.  Note that using many keys with exactly the same
84	>	* {@code hashCode()} is a sure way to slow down performance of any
85	>	* hash table. To ameliorate impact, when keys are {@link Comparable},
86	>	* this class may use comparison order among keys to help break ties.
87	>	*
88	>	* <p>A {@link Set} projection of a ConcurrentHashMapV8 may be created
89	>	* (using {@link #newKeySet()} or {@link #newKeySet(int)}), or viewed
90	>	* (using {@link #keySet(Object)} when only keys are of interest, and the
91	>	* mapped values are (perhaps transiently) not used or all take the
92	>	* same mapping value.
93		*
94		* <p>This class and its views and iterators implement all of the
95		* <em>optional</em> methods of the {@link Map} and {@link Iterator}
96		* interfaces.
97		*
98	<	* <p> Like {@link Hashtable} but unlike {@link HashMap}, this class
98	>	* <p>Like {@link Hashtable} but unlike {@link HashMap}, this class
99		* does <em>not</em> allow {@code null} to be used as a key or value.
100		*
101	+	* <p>ConcurrentHashMapV8s support a set of sequential and parallel bulk
102	+	* operations that are designed
103	+	* to be safely, and often sensibly, applied even with maps that are
104	+	* being concurrently updated by other threads; for example, when
105	+	* computing a snapshot summary of the values in a shared registry.
106	+	* There are three kinds of operation, each with four forms, accepting
107	+	* functions with Keys, Values, Entries, and (Key, Value) arguments
108	+	* and/or return values. Because the elements of a ConcurrentHashMapV8
109	+	* are not ordered in any particular way, and may be processed in
110	+	* different orders in different parallel executions, the correctness
111	+	* of supplied functions should not depend on any ordering, or on any
112	+	* other objects or values that may transiently change while
113	+	* computation is in progress; and except for forEach actions, should
114	+	* ideally be side-effect-free. Bulk operations on {@link java.util.Map.Entry}
115	+	* objects do not support method {@code setValue}.
116	+	*
117	+	* <ul>
118	+	* <li> forEach: Perform a given action on each element.
119	+	* A variant form applies a given transformation on each element
120	+	* before performing the action.</li>
121	+	*
122	+	* <li> search: Return the first available non-null result of
123	+	* applying a given function on each element; skipping further
124	+	* search when a result is found.</li>
125	+	*
126	+	* <li> reduce: Accumulate each element.  The supplied reduction
127	+	* function cannot rely on ordering (more formally, it should be
128	+	* both associative and commutative).  There are five variants:
129	+	*
130	+	* <ul>
131	+	*
132	+	* <li> Plain reductions. (There is not a form of this method for
133	+	* (key, value) function arguments since there is no corresponding
134	+	* return type.)</li>
135	+	*
136	+	* <li> Mapped reductions that accumulate the results of a given
137	+	* function applied to each element.</li>
138	+	*
139	+	* <li> Reductions to scalar doubles, longs, and ints, using a
140	+	* given basis value.</li>
141	+	*
142	+	* </ul>
143	+	* </li>
144	+	* </ul>
145	+	*
146	+	* <p>These bulk operations accept a {@code parallelismThreshold}
147	+	* argument. Methods proceed sequentially if the current map size is
148	+	* estimated to be less than the given threshold. Using a value of
149	+	* {@code Long.MAX_VALUE} suppresses all parallelism.  Using a value
150	+	* of {@code 1} results in maximal parallelism by partitioning into
151	+	* enough subtasks to fully utilize the {@link
152	+	* ForkJoinPool#commonPool()} that is used for all parallel
153	+	* computations. Normally, you would initially choose one of these
154	+	* extreme values, and then measure performance of using in-between
155	+	* values that trade off overhead versus throughput.
156	+	*
157	+	* <p>The concurrency properties of bulk operations follow
158	+	* from those of ConcurrentHashMapV8: Any non-null result returned
159	+	* from {@code get(key)} and related access methods bears a
160	+	* happens-before relation with the associated insertion or
161	+	* update.  The result of any bulk operation reflects the
162	+	* composition of these per-element relations (but is not
163	+	* necessarily atomic with respect to the map as a whole unless it
164	+	* is somehow known to be quiescent).  Conversely, because keys
165	+	* and values in the map are never null, null serves as a reliable
166	+	* atomic indicator of the current lack of any result.  To
167	+	* maintain this property, null serves as an implicit basis for
168	+	* all non-scalar reduction operations. For the double, long, and
169	+	* int versions, the basis should be one that, when combined with
170	+	* any other value, returns that other value (more formally, it
171	+	* should be the identity element for the reduction). Most common
172	+	* reductions have these properties; for example, computing a sum
173	+	* with basis 0 or a minimum with basis MAX_VALUE.
174	+	*
175	+	* <p>Search and transformation functions provided as arguments
176	+	* should similarly return null to indicate the lack of any result
177	+	* (in which case it is not used). In the case of mapped
178	+	* reductions, this also enables transformations to serve as
179	+	* filters, returning null (or, in the case of primitive
180	+	* specializations, the identity basis) if the element should not
181	+	* be combined. You can create compound transformations and
182	+	* filterings by composing them yourself under this "null means
183	+	* there is nothing there now" rule before using them in search or
184	+	* reduce operations.
185	+	*
186	+	* <p>Methods accepting and/or returning Entry arguments maintain
187	+	* key-value associations. They may be useful for example when
188	+	* finding the key for the greatest value. Note that "plain" Entry
189	+	* arguments can be supplied using {@code new
190	+	* AbstractMap.SimpleEntry(k,v)}.
191	+	*
192	+	* <p>Bulk operations may complete abruptly, throwing an
193	+	* exception encountered in the application of a supplied
194	+	* function. Bear in mind when handling such exceptions that other
195	+	* concurrently executing functions could also have thrown
196	+	* exceptions, or would have done so if the first exception had
197	+	* not occurred.
198	+	*
199	+	* <p>Speedups for parallel compared to sequential forms are common
200	+	* but not guaranteed.  Parallel operations involving brief functions
201	+	* on small maps may execute more slowly than sequential forms if the
202	+	* underlying work to parallelize the computation is more expensive
203	+	* than the computation itself.  Similarly, parallelization may not
204	+	* lead to much actual parallelism if all processors are busy
205	+	* performing unrelated tasks.
206	+	*
207	+	* <p>All arguments to all task methods must be non-null.
208	+	*
209	+	* <p><em>jsr166e note: During transition, this class
210	+	* uses nested functional interfaces with different names but the
211	+	* same forms as those expected for JDK8.</em>
212	+	*
213		* <p>This class is a member of the
214		* <a href="{@docRoot}/../technotes/guides/collections/index.html">
215		* Java Collections Framework</a>.
216		*
72	–	* <p><em>jsr166e note: This class is a candidate replacement for
73	–	* java.util.concurrent.ConcurrentHashMap.<em>
74	–	*
217		* @since 1.5
218		* @author Doug Lea
219		* @param <K> the type of keys maintained by this map
220		* @param <V> the type of mapped values
221		*/
222	<	public class ConcurrentHashMapV8<K, V>
223	<	implements ConcurrentMap<K, V>, Serializable {
222	>	public class ConcurrentHashMapV8<K,V> extends AbstractMap<K,V>
223	>	implements ConcurrentMap<K,V>, Serializable {
224		private static final long serialVersionUID = 7249069246763182397L;
225
226		/**
227	<	* A function computing a mapping from the given key to a value,
228	<	* or {@code null} if there is no mapping. This is a place-holder
229	<	* for an upcoming JDK8 interface.
230	<	*/
231	<	public static interface MappingFunction<K, V> {
232	<	/**
233	<	* Returns a value for the given key, or null if there is no
234	<	* mapping. If this function throws an (unchecked) exception,
235	<	* the exception is rethrown to its caller, and no mapping is
236	<	* recorded.  Because this function is invoked within
95	<	* atomicity control, the computation should be short and
96	<	* simple. The most common usage is to construct a new object
97	<	* serving as an initial mapped value.
98	<	*
99	<	* @param key the (non-null) key
100	<	* @return a value, or null if none
227	>	* An object for traversing and partitioning elements of a source.
228	>	* This interface provides a subset of the functionality of JDK8
229	>	* java.util.Spliterator.
230	>	*/
231	>	public static interface ConcurrentHashMapSpliterator<T> {
232	>	/**
233	>	* If possible, returns a new spliterator covering
234	>	* approximately one half of the elements, which will not be
235	>	* covered by this spliterator. Returns null if cannot be
236	>	* split.
237		*/
238	<	V map(K key);
239	<	}
238	>	ConcurrentHashMapSpliterator<T> trySplit();
239	>	/**
240	>	* Returns an estimate of the number of elements covered by
241	>	* this Spliterator.
242	>	*/
243	>	long estimateSize();
244	>
245	>	/** Applies the action to each untraversed element */
246	>	void forEachRemaining(Action<? super T> action);
247	>	/** If an element remains, applies the action and returns true. */
248	>	boolean tryAdvance(Action<? super T> action);
249	>	}
250	>
251	>	// Sams
252	>	/** Interface describing a void action of one argument */
253	>	public interface Action<A> { void apply(A a); }
254	>	/** Interface describing a void action of two arguments */
255	>	public interface BiAction<A,B> { void apply(A a, B b); }
256	>	/** Interface describing a function of one argument */
257	>	public interface Fun<A,T> { T apply(A a); }
258	>	/** Interface describing a function of two arguments */
259	>	public interface BiFun<A,B,T> { T apply(A a, B b); }
260	>	/** Interface describing a function mapping its argument to a double */
261	>	public interface ObjectToDouble<A> { double apply(A a); }
262	>	/** Interface describing a function mapping its argument to a long */
263	>	public interface ObjectToLong<A> { long apply(A a); }
264	>	/** Interface describing a function mapping its argument to an int */
265	>	public interface ObjectToInt<A> {int apply(A a); }
266	>	/** Interface describing a function mapping two arguments to a double */
267	>	public interface ObjectByObjectToDouble<A,B> { double apply(A a, B b); }
268	>	/** Interface describing a function mapping two arguments to a long */
269	>	public interface ObjectByObjectToLong<A,B> { long apply(A a, B b); }
270	>	/** Interface describing a function mapping two arguments to an int */
271	>	public interface ObjectByObjectToInt<A,B> {int apply(A a, B b); }
272	>	/** Interface describing a function mapping two doubles to a double */
273	>	public interface DoubleByDoubleToDouble { double apply(double a, double b); }
274	>	/** Interface describing a function mapping two longs to a long */
275	>	public interface LongByLongToLong { long apply(long a, long b); }
276	>	/** Interface describing a function mapping two ints to an int */
277	>	public interface IntByIntToInt { int apply(int a, int b); }
278	>
279
280		/*
281		* Overview:
#	Line 108 | Line 283 | public class ConcurrentHashMapV8<K, V>
283		* The primary design goal of this hash table is to maintain
284		* concurrent readability (typically method get(), but also
285		* iterators and related methods) while minimizing update
286	<	* contention.
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. All public generic-typed methods relay
292	<	* in/out of these internal methods, supplying casts as needed.
286	>	* contention. Secondary goals are to keep space consumption about
287	>	* the same or better than java.util.HashMap, and to support high
288	>	* initial insertion rates on an empty table by many threads.
289	>	*
290	>	* This map usually acts as a binned (bucketed) hash table.  Each
291	>	* key-value mapping is held in a Node.  Most nodes are instances
292	>	* of the basic Node class with hash, key, value, and next
293	>	* fields. However, various subclasses exist: TreeNodes are
294	>	* arranged in balanced trees, not lists.  TreeBins hold the roots
295	>	* of sets of TreeNodes. ForwardingNodes are placed at the heads
296	>	* of bins during resizing. ReservationNodes are used as
297	>	* placeholders while establishing values in computeIfAbsent and
298	>	* related methods.  The types TreeBin, ForwardingNode, and
299	>	* ReservationNode do not hold normal user keys, values, or
300	>	* hashes, and are readily distinguishable during search etc
301	>	* because they have negative hash fields and null key and value
302	>	* fields. (These special nodes are either uncommon or transient,
303	>	* so the impact of carrying around some unused fields is
304	>	* insignificant.)
305		*
306		* The table is lazily initialized to a power-of-two size upon the
307	<	* first insertion.  Each bin in the table contains a (typically
308	<	* short) list of Nodes.  Table accesses require volatile/atomic
309	<	* reads, writes, and CASes.  Because there is no other way to
310	<	* arrange this without adding further indirections, we use
311	<	* intrinsics (sun.misc.Unsafe) operations.  The lists of nodes
312	<	* within bins are always accurately traversable under volatile
313	<	* reads, so long as lookups check hash code and non-nullness of
314	<	* key and value before checking key equality. (All valid hash
315	<	* codes are nonnegative. Negative values are reserved for special
316	<	* forwarding nodes; see below.)
317	<	*
318	<	* A bin may be locked during update (insert, delete, and replace)
319	<	* operations.  We do not want to waste the space required to
320	<	* associate a distinct lock object with each bin, so instead use
321	<	* the first node of a bin list itself as a lock, using builtin
322	<	* "synchronized" locks. These save space and we can live with
323	<	* only plain block-structured lock/unlock operations. Using the
324	<	* first node of a list as a lock does not by itself suffice
325	<	* though: When a node is locked, any update must first validate
326	<	* that it is still the first node, and retry if not. (Because new
327	<	* nodes are always appended to lists, once a node is first in a
328	<	* bin, it remains first until deleted or the bin becomes
329	<	* invalidated.)  However, update operations can and sometimes do
330	<	* still traverse the bin until the point of update, which helps
331	<	* reduce cache misses on retries.  This is a converse of sorts to
332	<	* the lazy locking technique described by Herlihy & Shavit. If
333	<	* there is no existing node during a put operation, then one can
334	<	* be CAS'ed in (without need for lock except in computeIfAbsent);
335	<	* the CAS serves as validation. This is on average the most
336	<	* common case for put operations -- under random hash codes, the
337	<	* distribution of nodes in bins follows a Poisson distribution
338	<	* (see http://en.wikipedia.org/wiki/Poisson_distribution) with a
339	<	* parameter of 0.5 on average under the default loadFactor of
340	<	* 0.75.  The expected number of locks covering different elements
341	<	* (i.e., bins with 2 or more nodes) is approximately 10% at
342	<	* steady state under default settings.  Lock contention
343	<	* probability for two threads accessing arbitrary distinct
344	<	* elements is, roughly, 1 / (8 * #elements).
345	<	*
346	<	* The table is resized when occupancy exceeds a threshold.  Only
347	<	* a single thread performs the resize (using field "resizing", to
348	<	* arrange exclusion), but the table otherwise remains usable for
349	<	* both reads and updates. Resizing proceeds by transferring bins,
350	<	* one by one, from the table to the next table.  Upon transfer,
351	<	* the old table bin contains only a special forwarding node (with
352	<	* negative hash code ("MOVED")) that contains the next table as
307	>	* first insertion.  Each bin in the table normally contains a
308	>	* list of Nodes (most often, the list has only zero or one Node).
309	>	* Table accesses require volatile/atomic reads, writes, and
310	>	* CASes.  Because there is no other way to arrange this without
311	>	* adding further indirections, we use intrinsics
312	>	* (sun.misc.Unsafe) operations.
313	>	*
314	>	* We use the top (sign) bit of Node hash fields for control
315	>	* purposes -- it is available anyway because of addressing
316	>	* constraints.  Nodes with negative hash fields are specially
317	>	* handled or ignored in map methods.
318	>	*
319	>	* Insertion (via put or its variants) of the first node in an
320	>	* empty bin is performed by just CASing it to the bin.  This is
321	>	* by far the most common case for put operations under most
322	>	* key/hash distributions.  Other update operations (insert,
323	>	* delete, and replace) require locks.  We do not want to waste
324	>	* the space required to associate a distinct lock object with
325	>	* each bin, so instead use the first node of a bin list itself as
326	>	* a lock. Locking support for these locks relies on builtin
327	>	* "synchronized" monitors.
328	>	*
329	>	* Using the first node of a list as a lock does not by itself
330	>	* suffice though: When a node is locked, any update must first
331	>	* validate that it is still the first node after locking it, and
332	>	* retry if not. Because new nodes are always appended to lists,
333	>	* once a node is first in a bin, it remains first until deleted
334	>	* or the bin becomes invalidated (upon resizing).
335	>	*
336	>	* The main disadvantage of per-bin locks is that other update
337	>	* operations on other nodes in a bin list protected by the same
338	>	* lock can stall, for example when user equals() or mapping
339	>	* functions take a long time.  However, statistically, under
340	>	* random hash codes, this is not a common problem.  Ideally, the
341	>	* frequency of nodes in bins follows a Poisson distribution
342	>	* (http://en.wikipedia.org/wiki/Poisson_distribution) with a
343	>	* parameter of about 0.5 on average, given the resizing threshold
344	>	* of 0.75, although with a large variance because of resizing
345	>	* granularity. Ignoring variance, the expected occurrences of
346	>	* list size k are (exp(-0.5) * pow(0.5, k) / factorial(k)). The
347	>	* first values are:
348	>	*
349	>	* 0:    0.60653066
350	>	* 1:    0.30326533
351	>	* 2:    0.07581633
352	>	* 3:    0.01263606
353	>	* 4:    0.00157952
354	>	* 5:    0.00015795
355	>	* 6:    0.00001316
356	>	* 7:    0.00000094
357	>	* 8:    0.00000006
358	>	* more: less than 1 in ten million
359	>	*
360	>	* Lock contention probability for two threads accessing distinct
361	>	* elements is roughly 1 / (8 * #elements) under random hashes.
362	>	*
363	>	* Actual hash code distributions encountered in practice
364	>	* sometimes deviate significantly from uniform randomness.  This
365	>	* includes the case when N > (1<<30), so some keys MUST collide.
366	>	* Similarly for dumb or hostile usages in which multiple keys are
367	>	* designed to have identical hash codes or ones that differs only
368	>	* in masked-out high bits. So we use a secondary strategy that
369	>	* applies when the number of nodes in a bin exceeds a
370	>	* threshold. These TreeBins use a balanced tree to hold nodes (a
371	>	* specialized form of red-black trees), bounding search time to
372	>	* O(log N).  Each search step in a TreeBin is at least twice as
373	>	* slow as in a regular list, but given that N cannot exceed
374	>	* (1<<64) (before running out of addresses) this bounds search
375	>	* steps, lock hold times, etc, to reasonable constants (roughly
376	>	* 100 nodes inspected per operation worst case) so long as keys
377	>	* are Comparable (which is very common -- String, Long, etc).
378	>	* TreeBin nodes (TreeNodes) also maintain the same "next"
379	>	* traversal pointers as regular nodes, so can be traversed in
380	>	* iterators in the same way.
381	>	*
382	>	* The table is resized when occupancy exceeds a percentage
383	>	* threshold (nominally, 0.75, but see below).  Any thread
384	>	* noticing an overfull bin may assist in resizing after the
385	>	* initiating thread allocates and sets up the replacement array.
386	>	* However, rather than stalling, these other threads may proceed
387	>	* with insertions etc.  The use of TreeBins shields us from the
388	>	* worst case effects of overfilling while resizes are in
389	>	* progress.  Resizing proceeds by transferring bins, one by one,
390	>	* from the table to the next table. However, threads claim small
391	>	* blocks of indices to transfer (via field transferIndex) before
392	>	* doing so, reducing contention.  Because we are using
393	>	* power-of-two expansion, the elements from each bin must either
394	>	* stay at same index, or move with a power of two offset. We
395	>	* eliminate unnecessary node creation by catching cases where old
396	>	* nodes can be reused because their next fields won't change.  On
397	>	* average, only about one-sixth of them need cloning when a table
398	>	* doubles. The nodes they replace will be garbage collectable as
399	>	* soon as they are no longer referenced by any reader thread that
400	>	* may be in the midst of concurrently traversing table.  Upon
401	>	* transfer, the old table bin contains only a special forwarding
402	>	* node (with hash field "MOVED") that contains the next table as
403		* its key. On encountering a forwarding node, access and update
404	<	* operations restart, using the new table. To ensure concurrent
405	<	* readability of traversals, transfers must proceed from the last
406	<	* bin (table.length - 1) up towards the first.  Any traversal
407	<	* starting from the first bin can then arrange to move to the new
408	<	* table for the rest of the traversal without revisiting nodes.
409	<	* This constrains bin transfers to a particular order, and so can
410	<	* block indefinitely waiting for the next lock, and other threads
411	<	* cannot help with the transfer. However, expected stalls are
412	<	* infrequent enough to not warrant the additional overhead and
413	<	* complexity of access and iteration schemes that could admit
414	<	* out-of-order or concurrent bin transfers.
415	<	*
416	<	* A similar traversal scheme (not yet implemented) can apply to
417	<	* partial traversals during partitioned aggregate operations.
418	<	* Also, read-only operations give up if ever forwarded to a null
419	<	* table, which provides support for shutdown-style clearing,
420	<	* which is also not currently implemented.
421	<	*
422	<	* The element count is maintained using a LongAdder, which avoids
423	<	* contention on updates but can encounter cache thrashing if read
424	<	* too frequently during concurrent updates. To avoid reading so
425	<	* often, resizing is normally attempted only upon adding to a bin
426	<	* already holding two or more nodes. Under the default threshold
427	<	* (0.75), and uniform hash distributions, the probability of this
428	<	* occurring at threshold is around 13%, meaning that only about 1
429	<	* in 8 puts check threshold (and after resizing, many fewer do
430	<	* so). But this approximation has high variance for small table
431	<	* sizes, so we check on any collision for sizes <= 64.  Further,
432	<	* to increase the probability that a resize occurs soon enough, we
433	<	* offset the threshold (see THRESHOLD_OFFSET) by the expected
434	<	* number of puts between checks. This is currently set to 8, in
435	<	* accord with the default load factor. In practice, this is
436	<	* rarely overridden, and in any case is close enough to other
437	<	* plausible values not to waste dynamic probability computation
438	<	* for more precision.
404	>	* operations restart, using the new table.
405	>	*
406	>	* Each bin transfer requires its bin lock, which can stall
407	>	* waiting for locks while resizing. However, because other
408	>	* threads can join in and help resize rather than contend for
409	>	* locks, average aggregate waits become shorter as resizing
410	>	* progresses.  The transfer operation must also ensure that all
411	>	* accessible bins in both the old and new table are usable by any
412	>	* traversal.  This is arranged in part by proceeding from the
413	>	* last bin (table.length - 1) up towards the first.  Upon seeing
414	>	* a forwarding node, traversals (see class Traverser) arrange to
415	>	* move to the new table without revisiting nodes.  To ensure that
416	>	* no intervening nodes are skipped even when moved out of order,
417	>	* a stack (see class TableStack) is created on first encounter of
418	>	* a forwarding node during a traversal, to maintain its place if
419	>	* later processing the current table. The need for these
420	>	* save/restore mechanics is relatively rare, but when one
421	>	* forwarding node is encountered, typically many more will be.
422	>	* So Traversers use a simple caching scheme to avoid creating so
423	>	* many new TableStack nodes. (Thanks to Peter Levart for
424	>	* suggesting use of a stack here.)
425	>	*
426	>	* The traversal scheme also applies to partial traversals of
427	>	* ranges of bins (via an alternate Traverser constructor)
428	>	* to support partitioned aggregate operations.  Also, read-only
429	>	* operations give up if ever forwarded to a null table, which
430	>	* provides support for shutdown-style clearing, which is also not
431	>	* currently implemented.
432	>	*
433	>	* Lazy table initialization minimizes footprint until first use,
434	>	* and also avoids resizings when the first operation is from a
435	>	* putAll, constructor with map argument, or deserialization.
436	>	* These cases attempt to override the initial capacity settings,
437	>	* but harmlessly fail to take effect in cases of races.
438	>	*
439	>	* The element count is maintained using a specialization of
440	>	* LongAdder. We need to incorporate a specialization rather than
441	>	* just use a LongAdder in order to access implicit
442	>	* contention-sensing that leads to creation of multiple
443	>	* CounterCells.  The counter mechanics avoid contention on
444	>	* updates but can encounter cache thrashing if read too
445	>	* frequently during concurrent access. To avoid reading so often,
446	>	* resizing under contention is attempted only upon adding to a
447	>	* bin already holding two or more nodes. Under uniform hash
448	>	* distributions, the probability of this occurring at threshold
449	>	* is around 13%, meaning that only about 1 in 8 puts check
450	>	* threshold (and after resizing, many fewer do so).
451	>	*
452	>	* TreeBins use a special form of comparison for search and
453	>	* related operations (which is the main reason we cannot use
454	>	* existing collections such as TreeMaps). TreeBins contain
455	>	* Comparable elements, but may contain others, as well as
456	>	* elements that are Comparable but not necessarily Comparable for
457	>	* the same T, so we cannot invoke compareTo among them. To handle
458	>	* this, the tree is ordered primarily by hash value, then by
459	>	* Comparable.compareTo order if applicable.  On lookup at a node,
460	>	* if elements are not comparable or compare as 0 then both left
461	>	* and right children may need to be searched in the case of tied
462	>	* hash values. (This corresponds to the full list search that
463	>	* would be necessary if all elements were non-Comparable and had
464	>	* tied hashes.) On insertion, to keep a total ordering (or as
465	>	* close as is required here) across rebalancings, we compare
466	>	* classes and identityHashCodes as tie-breakers. The red-black
467	>	* balancing code is updated from pre-jdk-collections
468	>	* (http://gee.cs.oswego.edu/dl/classes/collections/RBCell.java)
469	>	* based in turn on Cormen, Leiserson, and Rivest "Introduction to
470	>	* Algorithms" (CLR).
471	>	*
472	>	* TreeBins also require an additional locking mechanism.  While
473	>	* list traversal is always possible by readers even during
474	>	* updates, tree traversal is not, mainly because of tree-rotations
475	>	* that may change the root node and/or its linkages.  TreeBins
476	>	* include a simple read-write lock mechanism parasitic on the
477	>	* main bin-synchronization strategy: Structural adjustments
478	>	* associated with an insertion or removal are already bin-locked
479	>	* (and so cannot conflict with other writers) but must wait for
480	>	* ongoing readers to finish. Since there can be only one such
481	>	* waiter, we use a simple scheme using a single "waiter" field to
482	>	* block writers.  However, readers need never block.  If the root
483	>	* lock is held, they proceed along the slow traversal path (via
484	>	* next-pointers) until the lock becomes available or the list is
485	>	* exhausted, whichever comes first. These cases are not fast, but
486	>	* maximize aggregate expected throughput.
487	>	*
488	>	* Maintaining API and serialization compatibility with previous
489	>	* versions of this class introduces several oddities. Mainly: We
490	>	* leave untouched but unused constructor arguments refering to
491	>	* concurrencyLevel. We accept a loadFactor constructor argument,
492	>	* but apply it only to initial table capacity (which is the only
493	>	* time that we can guarantee to honor it.) We also declare an
494	>	* unused "Segment" class that is instantiated in minimal form
495	>	* only when serializing.
496	>	*
497	>	* Also, solely for compatibility with previous versions of this
498	>	* class, it extends AbstractMap, even though all of its methods
499	>	* are overridden, so it is just useless baggage.
500	>	*
501	>	* This file is organized to make things a little easier to follow
502	>	* while reading than they might otherwise: First the main static
503	>	* declarations and utilities, then fields, then main public
504	>	* methods (with a few factorings of multiple public methods into
505	>	* internal ones), then sizing methods, trees, traversers, and
506	>	* bulk operations.
507		*/
508
509		/* ---------------- Constants -------------- */
510
511		/**
512	<	* The smallest allowed table capacity.  Must be a power of 2, at
513	<	* least 2.
512	>	* The largest possible table capacity.  This value must be
513	>	* exactly 1<<30 to stay within Java array allocation and indexing
514	>	* bounds for power of two table sizes, and is further required
515	>	* because the top two bits of 32bit hash fields are used for
516	>	* control purposes.
517		*/
518	<	static final int MINIMUM_CAPACITY = 2;
518	>	private static final int MAXIMUM_CAPACITY = 1 << 30;
519
520		/**
521	<	* The largest allowed table capacity.  Must be a power of 2, at
522	<	* most 1<<30.
521	>	* The default initial table capacity.  Must be a power of 2
522	>	* (i.e., at least 1) and at most MAXIMUM_CAPACITY.
523		*/
524	<	static final int MAXIMUM_CAPACITY = 1 << 30;
524	>	private static final int DEFAULT_CAPACITY = 16;
525
526		/**
527	<	* The default initial table capacity.  Must be a power of 2, at
528	<	* least MINIMUM_CAPACITY and at most MAXIMUM_CAPACITY.
527	>	* The largest possible (non-power of two) array size.
528	>	* Needed by toArray and related methods.
529		*/
530	<	static final int DEFAULT_CAPACITY = 16;
530	>	static final int MAX_ARRAY_SIZE = Integer.MAX_VALUE - 8;
531
532		/**
533	<	* The default load factor for this table, used when not otherwise
534	<	* specified in a constructor.
533	>	* The default concurrency level for this table. Unused but
534	>	* defined for compatibility with previous versions of this class.
535		*/
536	<	static final float DEFAULT_LOAD_FACTOR = 0.75f;
536	>	private static final int DEFAULT_CONCURRENCY_LEVEL = 16;
537
538		/**
539	<	* The default concurrency level for this table. Unused, but
540	<	* defined for compatibility with previous versions of this class.
539	>	* The load factor for this table. Overrides of this value in
540	>	* constructors affect only the initial table capacity.  The
541	>	* actual floating point value isn't normally used -- it is
542	>	* simpler to use expressions such as {@code n - (n >>> 2)} for
543	>	* the associated resizing threshold.
544		*/
545	<	static final int DEFAULT_CONCURRENCY_LEVEL = 16;
545	>	private static final float LOAD_FACTOR = 0.75f;
546
547		/**
548	<	* The count value to offset thresholds to compensate for checking
549	<	* for resizing only when inserting into bins with two or more
550	<	* elements. See above for explanation.
548	>	* The bin count threshold for using a tree rather than list for a
549	>	* bin.  Bins are converted to trees when adding an element to a
550	>	* bin with at least this many nodes. The value must be greater
551	>	* than 2, and should be at least 8 to mesh with assumptions in
552	>	* tree removal about conversion back to plain bins upon
553	>	* shrinkage.
554		*/
555	<	static final int THRESHOLD_OFFSET = 8;
555	>	static final int TREEIFY_THRESHOLD = 8;
556
557		/**
558	<	* Special node hash value indicating to use table in node.key
559	<	* Must be negative.
558	>	* The bin count threshold for untreeifying a (split) bin during a
559	>	* resize operation. Should be less than TREEIFY_THRESHOLD, and at
560	>	* most 6 to mesh with shrinkage detection under removal.
561		*/
562	<	static final int MOVED = -1;
248	<
249	<	/* ---------------- Fields -------------- */
562	>	static final int UNTREEIFY_THRESHOLD = 6;
563
564		/**
565	<	* The array of bins. Lazily initialized upon first insertion.
566	<	* Size is always a power of two. Accessed directly by inner
567	<	* classes.
565	>	* The smallest table capacity for which bins may be treeified.
566	>	* (Otherwise the table is resized if too many nodes in a bin.)
567	>	* The value should be at least 4 * TREEIFY_THRESHOLD to avoid
568	>	* conflicts between resizing and treeification thresholds.
569		*/
570	<	transient volatile Node[] table;
257	<
258	<	/** The counter maintaining number of elements. */
259	<	private transient final LongAdder counter;
260	<	/** Nonzero when table is being initialized or resized. Updated via CAS. */
261	<	private transient volatile int resizing;
262	<	/** The target load factor for the table. */
263	<	private transient float loadFactor;
264	<	/** The next element count value upon which to resize the table. */
265	<	private transient int threshold;
266	<	/** The initial capacity of the table. */
267	<	private transient int initCap;
268	<
269	<	// views
270	<	transient Set<K> keySet;
271	<	transient Set<Map.Entry<K,V>> entrySet;
272	<	transient Collection<V> values;
273	<
274	<	/** For serialization compatibility. Null unless serialized; see below */
275	<	Segment<K,V>[] segments;
570	>	static final int MIN_TREEIFY_CAPACITY = 64;
571
572		/**
573	<	* Applies a supplemental hash function to a given hashCode, which
574	<	* defends against poor quality hash functions.  The result must
575	<	* be non-negative, and for reasonable performance must have good
576	<	* avalanche properties; i.e., that each bit of the argument
577	<	* affects each bit (except sign bit) of the result.
573	>	* Minimum number of rebinnings per transfer step. Ranges are
574	>	* subdivided to allow multiple resizer threads.  This value
575	>	* serves as a lower bound to avoid resizers encountering
576	>	* excessive memory contention.  The value should be at least
577	>	* DEFAULT_CAPACITY.
578		*/
579	<	private static final int spread(int h) {
285	<	// Apply base step of MurmurHash; see http://code.google.com/p/smhasher/
286	<	h ^= h >>> 16;
287	<	h *= 0x85ebca6b;
288	<	h ^= h >>> 13;
289	<	h *= 0xc2b2ae35;
290	<	return (h >>> 16) ^ (h & 0x7fffffff); // mask out sign bit
291	<	}
579	>	private static final int MIN_TRANSFER_STRIDE = 16;
580
581	<	/**
582	<	* Key-value entry. Note that this is never exported out as a
295	<	* user-visible Map.Entry.
581	>	/*
582	>	* Encodings for Node hash fields. See above for explanation.
583		*/
584	<	static final class Node {
584	>	static final int MOVED     = -1; // hash for forwarding nodes
585	>	static final int TREEBIN   = -2; // hash for roots of trees
586	>	static final int RESERVED  = -3; // hash for transient reservations
587	>	static final int HASH_BITS = 0x7fffffff; // usable bits of normal node hash
588	>
589	>	/** Number of CPUS, to place bounds on some sizings */
590	>	static final int NCPU = Runtime.getRuntime().availableProcessors();
591	>
592	>	/** For serialization compatibility. */
593	>	private static final ObjectStreamField[] serialPersistentFields = {
594	>	new ObjectStreamField("segments", Segment[].class),
595	>	new ObjectStreamField("segmentMask", Integer.TYPE),
596	>	new ObjectStreamField("segmentShift", Integer.TYPE)
597	>	};
598	>
599	>	/* ---------------- Nodes -------------- */
600	>
601	>	/**
602	>	* Key-value entry.  This class is never exported out as a
603	>	* user-mutable Map.Entry (i.e., one supporting setValue; see
604	>	* MapEntry below), but can be used for read-only traversals used
605	>	* in bulk tasks.  Subclasses of Node with a negative hash field
606	>	* are special, and contain null keys and values (but are never
607	>	* exported).  Otherwise, keys and vals are never null.
608	>	*/
609	>	static class Node<K,V> implements Map.Entry<K,V> {
610		final int hash;
611	<	final Object key;
612	<	volatile Object val;
613	<	volatile Node next;
611	>	final K key;
612	>	volatile V val;
613	>	volatile Node<K,V> next;
614
615	<	Node(int hash, Object key, Object val, Node next) {
615	>	Node(int hash, K key, V val, Node<K,V> next) {
616		this.hash = hash;
617		this.key = key;
618		this.val = val;
619		this.next = next;
620		}
621	+
622	+	public final K getKey()       { return key; }
623	+	public final V getValue()     { return val; }
624	+	public final int hashCode()   { return key.hashCode() ^ val.hashCode(); }
625	+	public final String toString(){ return key + "=" + val; }
626	+	public final V setValue(V value) {
627	+	throw new UnsupportedOperationException();
628	+	}
629	+
630	+	public final boolean equals(Object o) {
631	+	Object k, v, u; Map.Entry<?,?> e;
632	+	return ((o instanceof Map.Entry) &&
633	+	(k = (e = (Map.Entry<?,?>)o).getKey()) != null &&
634	+	(v = e.getValue()) != null &&
635	+	(k == key || k.equals(key)) &&
636	+	(v == (u = val) || v.equals(u)));
637	+	}
638	+
639	+	/**
640	+	* Virtualized support for map.get(); overridden in subclasses.
641	+	*/
642	+	Node<K,V> find(int h, Object k) {
643	+	Node<K,V> e = this;
644	+	if (k != null) {
645	+	do {
646	+	K ek;
647	+	if (e.hash == h &&
648	+	((ek = e.key) == k || (ek != null && k.equals(ek))))
649	+	return e;
650	+	} while ((e = e.next) != null);
651	+	}
652	+	return null;
653	+	}
654		}
655
656	+	/* ---------------- Static utilities -------------- */
657	+
658	+	/**
659	+	* Spreads (XORs) higher bits of hash to lower and also forces top
660	+	* bit to 0. Because the table uses power-of-two masking, sets of
661	+	* hashes that vary only in bits above the current mask will
662	+	* always collide. (Among known examples are sets of Float keys
663	+	* holding consecutive whole numbers in small tables.)  So we
664	+	* apply a transform that spreads the impact of higher bits
665	+	* downward. There is a tradeoff between speed, utility, and
666	+	* quality of bit-spreading. Because many common sets of hashes
667	+	* are already reasonably distributed (so don't benefit from
668	+	* spreading), and because we use trees to handle large sets of
669	+	* collisions in bins, we just XOR some shifted bits in the
670	+	* cheapest possible way to reduce systematic lossage, as well as
671	+	* to incorporate impact of the highest bits that would otherwise
672	+	* never be used in index calculations because of table bounds.
673	+	*/
674	+	static final int spread(int h) {
675	+	return (h ^ (h >>> 16)) & HASH_BITS;
676	+	}
677	+
678	+	/**
679	+	* Returns a power of two table size for the given desired capacity.
680	+	* See Hackers Delight, sec 3.2
681	+	*/
682	+	private static final int tableSizeFor(int c) {
683	+	int n = c - 1;
684	+	n |= n >>> 1;
685	+	n |= n >>> 2;
686	+	n |= n >>> 4;
687	+	n |= n >>> 8;
688	+	n |= n >>> 16;
689	+	return (n < 0) ? 1 : (n >= MAXIMUM_CAPACITY) ? MAXIMUM_CAPACITY : n + 1;
690	+	}
691	+
692	+	/**
693	+	* Returns x's Class if it is of the form "class C implements
694	+	* Comparable<C>", else null.
695	+	*/
696	+	static Class<?> comparableClassFor(Object x) {
697	+	if (x instanceof Comparable) {
698	+	Class<?> c; Type[] ts, as; Type t; ParameterizedType p;
699	+	if ((c = x.getClass()) == String.class) // bypass checks
700	+	return c;
701	+	if ((ts = c.getGenericInterfaces()) != null) {
702	+	for (int i = 0; i < ts.length; ++i) {
703	+	if (((t = ts[i]) instanceof ParameterizedType) &&
704	+	((p = (ParameterizedType)t).getRawType() ==
705	+	Comparable.class) &&
706	+	(as = p.getActualTypeArguments()) != null &&
707	+	as.length == 1 && as[0] == c) // type arg is c
708	+	return c;
709	+	}
710	+	}
711	+	}
712	+	return null;
713	+	}
714	+
715	+	/**
716	+	* Returns k.compareTo(x) if x matches kc (k's screened comparable
717	+	* class), else 0.
718	+	*/
719	+	@SuppressWarnings({"rawtypes","unchecked"}) // for cast to Comparable
720	+	static int compareComparables(Class<?> kc, Object k, Object x) {
721	+	return (x == null || x.getClass() != kc ? 0 :
722	+	((Comparable)k).compareTo(x));
723	+	}
724	+
725	+	/* ---------------- Table element access -------------- */
726	+
727		/*
728		* Volatile access methods are used for table elements as well as
729	<	* elements of in-progress next table while resizing.  Uses in
730	<	* access and update methods are null checked by callers, and
731	<	* implicitly bounds-checked, relying on the invariants that tab
732	<	* arrays have non-zero size, and all indices are masked with
733	<	* (tab.length - 1) which is never negative and always less than
734	<	* length. The "relaxed" non-volatile forms are used only during
735	<	* table initialization. The only other usage is in
736	<	* HashIterator.advance, which performs explicit checks.
729	>	* elements of in-progress next table while resizing.  All uses of
730	>	* the tab arguments must be null checked by callers.  All callers
731	>	* also paranoically precheck that tab's length is not zero (or an
732	>	* equivalent check), thus ensuring that any index argument taking
733	>	* the form of a hash value anded with (length - 1) is a valid
734	>	* index.  Note that, to be correct wrt arbitrary concurrency
735	>	* errors by users, these checks must operate on local variables,
736	>	* which accounts for some odd-looking inline assignments below.
737	>	* Note that calls to setTabAt always occur within locked regions,
738	>	* and so in principle require only release ordering, not need
739	>	* full volatile semantics, but are currently coded as volatile
740	>	* writes to be conservative.
741	>	*/
742	>
743	>	@SuppressWarnings("unchecked")
744	>	static final <K,V> Node<K,V> tabAt(Node<K,V>[] tab, int i) {
745	>	return (Node<K,V>)U.getObjectVolatile(tab, ((long)i << ASHIFT) + ABASE);
746	>	}
747	>
748	>	static final <K,V> boolean casTabAt(Node<K,V>[] tab, int i,
749	>	Node<K,V> c, Node<K,V> v) {
750	>	return U.compareAndSwapObject(tab, ((long)i << ASHIFT) + ABASE, c, v);
751	>	}
752	>
753	>	static final <K,V> void setTabAt(Node<K,V>[] tab, int i, Node<K,V> v) {
754	>	U.putObjectVolatile(tab, ((long)i << ASHIFT) + ABASE, v);
755	>	}
756	>
757	>	/* ---------------- Fields -------------- */
758	>
759	>	/**
760	>	* The array of bins. Lazily initialized upon first insertion.
761	>	* Size is always a power of two. Accessed directly by iterators.
762	>	*/
763	>	transient volatile Node<K,V>[] table;
764	>
765	>	/**
766	>	* The next table to use; non-null only while resizing.
767	>	*/
768	>	private transient volatile Node<K,V>[] nextTable;
769	>
770	>	/**
771	>	* Base counter value, used mainly when there is no contention,
772	>	* but also as a fallback during table initialization
773	>	* races. Updated via CAS.
774	>	*/
775	>	private transient volatile long baseCount;
776	>
777	>	/**
778	>	* Table initialization and resizing control.  When negative, the
779	>	* table is being initialized or resized: -1 for initialization,
780	>	* else -(1 + the number of active resizing threads).  Otherwise,
781	>	* when table is null, holds the initial table size to use upon
782	>	* creation, or 0 for default. After initialization, holds the
783	>	* next element count value upon which to resize the table.
784	>	*/
785	>	private transient volatile int sizeCtl;
786	>
787	>	/**
788	>	* The next table index (plus one) to split while resizing.
789	>	*/
790	>	private transient volatile int transferIndex;
791	>
792	>	/**
793	>	* Spinlock (locked via CAS) used when resizing and/or creating CounterCells.
794	>	*/
795	>	private transient volatile int cellsBusy;
796	>
797	>	/**
798	>	* Table of counter cells. When non-null, size is a power of 2.
799		*/
800	+	private transient volatile CounterCell[] counterCells;
801	+
802	+	// views
803	+	private transient KeySetView<K,V> keySet;
804	+	private transient ValuesView<K,V> values;
805	+	private transient EntrySetView<K,V> entrySet;
806	+
807
808	<	static final Node tabAt(Node[] tab, int i) { // used in HashIterator
809	<	return (Node)UNSAFE.getObjectVolatile(tab, ((long)i<<ASHIFT)+ABASE);
808	>	/* ---------------- Public operations -------------- */
809	>
810	>	/**
811	>	* Creates a new, empty map with the default initial table size (16).
812	>	*/
813	>	public ConcurrentHashMapV8() {
814		}
815
816	<	private static final boolean casTabAt(Node[] tab, int i, Node c, Node v) {
817	<	return UNSAFE.compareAndSwapObject(tab, ((long)i<<ASHIFT)+ABASE, c, v);
816	>	/**
817	>	* Creates a new, empty map with an initial table size
818	>	* accommodating the specified number of elements without the need
819	>	* to dynamically resize.
820	>	*
821	>	* @param initialCapacity The implementation performs internal
822	>	* sizing to accommodate this many elements.
823	>	* @throws IllegalArgumentException if the initial capacity of
824	>	* elements is negative
825	>	*/
826	>	public ConcurrentHashMapV8(int initialCapacity) {
827	>	if (initialCapacity < 0)
828	>	throw new IllegalArgumentException();
829	>	int cap = ((initialCapacity >= (MAXIMUM_CAPACITY >>> 1)) ?
830	>	MAXIMUM_CAPACITY :
831	>	tableSizeFor(initialCapacity + (initialCapacity >>> 1) + 1));
832	>	this.sizeCtl = cap;
833		}
834
835	<	private static final void setTabAt(Node[] tab, int i, Node v) {
836	<	UNSAFE.putObjectVolatile(tab, ((long)i<<ASHIFT)+ABASE, v);
835	>	/**
836	>	* Creates a new map with the same mappings as the given map.
837	>	*
838	>	* @param m the map
839	>	*/
840	>	public ConcurrentHashMapV8(Map<? extends K, ? extends V> m) {
841	>	this.sizeCtl = DEFAULT_CAPACITY;
842	>	putAll(m);
843		}
844
845	<	private static final Node relaxedTabAt(Node[] tab, int i) {
846	<	return (Node)UNSAFE.getObject(tab, ((long)i<<ASHIFT)+ABASE);
845	>	/**
846	>	* Creates a new, empty map with an initial table size based on
847	>	* the given number of elements ({@code initialCapacity}) and
848	>	* initial table density ({@code loadFactor}).
849	>	*
850	>	* @param initialCapacity the initial capacity. The implementation
851	>	* performs internal sizing to accommodate this many elements,
852	>	* given the specified load factor.
853	>	* @param loadFactor the load factor (table density) for
854	>	* establishing the initial table size
855	>	* @throws IllegalArgumentException if the initial capacity of
856	>	* elements is negative or the load factor is nonpositive
857	>	*
858	>	* @since 1.6
859	>	*/
860	>	public ConcurrentHashMapV8(int initialCapacity, float loadFactor) {
861	>	this(initialCapacity, loadFactor, 1);
862		}
863
864	<	private static final void relaxedSetTabAt(Node[] tab, int i, Node v) {
865	<	UNSAFE.putObject(tab, ((long)i<<ASHIFT)+ABASE, v);
864	>	/**
865	>	* Creates a new, empty map with an initial table size based on
866	>	* the given number of elements ({@code initialCapacity}), table
867	>	* density ({@code loadFactor}), and number of concurrently
868	>	* updating threads ({@code concurrencyLevel}).
869	>	*
870	>	* @param initialCapacity the initial capacity. The implementation
871	>	* performs internal sizing to accommodate this many elements,
872	>	* given the specified load factor.
873	>	* @param loadFactor the load factor (table density) for
874	>	* establishing the initial table size
875	>	* @param concurrencyLevel the estimated number of concurrently
876	>	* updating threads. The implementation may use this value as
877	>	* a sizing hint.
878	>	* @throws IllegalArgumentException if the initial capacity is
879	>	* negative or the load factor or concurrencyLevel are
880	>	* nonpositive
881	>	*/
882	>	public ConcurrentHashMapV8(int initialCapacity,
883	>	float loadFactor, int concurrencyLevel) {
884	>	if (!(loadFactor > 0.0f) || initialCapacity < 0 || concurrencyLevel <= 0)
885	>	throw new IllegalArgumentException();
886	>	if (initialCapacity < concurrencyLevel)   // Use at least as many bins
887	>	initialCapacity = concurrencyLevel;   // as estimated threads
888	>	long size = (long)(1.0 + (long)initialCapacity / loadFactor);
889	>	int cap = (size >= (long)MAXIMUM_CAPACITY) ?
890	>	MAXIMUM_CAPACITY : tableSizeFor((int)size);
891	>	this.sizeCtl = cap;
892		}
893
894	<	/* ---------------- Access and update operations -------------- */
894	>	// Original (since JDK1.2) Map methods
895
896	<	/** Implementation for get and containsKey */
897	<	private final Object internalGet(Object k) {
898	<	int h = spread(k.hashCode());
899	<	Node[] tab = table;
900	<	retry: while (tab != null) {
901	<	Node e = tabAt(tab, (tab.length - 1) & h);
902	<	while (e != null) {
903	<	int eh = e.hash;
904	<	if (eh == h) {
905	<	Object ek = e.key, ev = e.val;
906	<	if (ev != null && ek != null && (k == ek || k.equals(ek)))
907	<	return ev;
908	<	}
909	<	else if (eh < 0) { // bin was moved during resize
910	<	tab = (Node[])e.key;
911	<	continue retry;
912	<	}
913	<	e = e.next;
896	>	/**
897	>	* {@inheritDoc}
898	>	*/
899	>	public int size() {
900	>	long n = sumCount();
901	>	return ((n < 0L) ? 0 :
902	>	(n > (long)Integer.MAX_VALUE) ? Integer.MAX_VALUE :
903	>	(int)n);
904	>	}
905	>
906	>	/**
907	>	* {@inheritDoc}
908	>	*/
909	>	public boolean isEmpty() {
910	>	return sumCount() <= 0L; // ignore transient negative values
911	>	}
912	>
913	>	/**
914	>	* Returns the value to which the specified key is mapped,
915	>	* or {@code null} if this map contains no mapping for the key.
916	>	*
917	>	* <p>More formally, if this map contains a mapping from a key
918	>	* {@code k} to a value {@code v} such that {@code key.equals(k)},
919	>	* then this method returns {@code v}; otherwise it returns
920	>	* {@code null}.  (There can be at most one such mapping.)
921	>	*
922	>	* @throws NullPointerException if the specified key is null
923	>	*/
924	>	public V get(Object key) {
925	>	Node<K,V>[] tab; Node<K,V> e, p; int n, eh; K ek;
926	>	int h = spread(key.hashCode());
927	>	if ((tab = table) != null && (n = tab.length) > 0 &&
928	>	(e = tabAt(tab, (n - 1) & h)) != null) {
929	>	if ((eh = e.hash) == h) {
930	>	if ((ek = e.key) == key || (ek != null && key.equals(ek)))
931	>	return e.val;
932	>	}
933	>	else if (eh < 0)
934	>	return (p = e.find(h, key)) != null ? p.val : null;
935	>	while ((e = e.next) != null) {
936	>	if (e.hash == h &&
937	>	((ek = e.key) == key || (ek != null && key.equals(ek))))
938	>	return e.val;
939		}
364	–	break;
940		}
941		return null;
942		}
943
944	+	/**
945	+	* Tests if the specified object is a key in this table.
946	+	*
947	+	* @param  key possible key
948	+	* @return {@code true} if and only if the specified object
949	+	*         is a key in this table, as determined by the
950	+	*         {@code equals} method; {@code false} otherwise
951	+	* @throws NullPointerException if the specified key is null
952	+	*/
953	+	public boolean containsKey(Object key) {
954	+	return get(key) != null;
955	+	}
956	+
957	+	/**
958	+	* Returns {@code true} if this map maps one or more keys to the
959	+	* specified value. Note: This method may require a full traversal
960	+	* of the map, and is much slower than method {@code containsKey}.
961	+	*
962	+	* @param value value whose presence in this map is to be tested
963	+	* @return {@code true} if this map maps one or more keys to the
964	+	*         specified value
965	+	* @throws NullPointerException if the specified value is null
966	+	*/
967	+	public boolean containsValue(Object value) {
968	+	if (value == null)
969	+	throw new NullPointerException();
970	+	Node<K,V>[] t;
971	+	if ((t = table) != null) {
972	+	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
973	+	for (Node<K,V> p; (p = it.advance()) != null; ) {
974	+	V v;
975	+	if ((v = p.val) == value || (v != null && value.equals(v)))
976	+	return true;
977	+	}
978	+	}
979	+	return false;
980	+	}
981	+
982	+	/**
983	+	* Maps the specified key to the specified value in this table.
984	+	* Neither the key nor the value can be null.
985	+	*
986	+	* <p>The value can be retrieved by calling the {@code get} method
987	+	* with a key that is equal to the original key.
988	+	*
989	+	* @param key key with which the specified value is to be associated
990	+	* @param value value to be associated with the specified key
991	+	* @return the previous value associated with {@code key}, or
992	+	*         {@code null} if there was no mapping for {@code key}
993	+	* @throws NullPointerException if the specified key or value is null
994	+	*/
995	+	public V put(K key, V value) {
996	+	return putVal(key, value, false);
997	+	}
998
999		/** Implementation for put and putIfAbsent */
1000	<	private final Object internalPut(Object k, Object v, boolean replace) {
1001	<	int h = spread(k.hashCode());
1002	<	Object oldVal = null;  // the previous value or null if none
1003	<	Node[] tab = table;
1004	<	for (;;) {
1005	<	Node e; int i;
1006	<	if (tab == null)
1007	<	tab = grow(0);
1008	<	else if ((e = tabAt(tab, i = (tab.length - 1) & h)) == null) {
1009	<	if (casTabAt(tab, i, null, new Node(h, k, v, null)))
1010	<	break;
1000	>	final V putVal(K key, V value, boolean onlyIfAbsent) {
1001	>	if (key == null || value == null) throw new NullPointerException();
1002	>	int hash = spread(key.hashCode());
1003	>	int binCount = 0;
1004	>	for (Node<K,V>[] tab = table;;) {
1005	>	Node<K,V> f; int n, i, fh;
1006	>	if (tab == null || (n = tab.length) == 0)
1007	>	tab = initTable();
1008	>	else if ((f = tabAt(tab, i = (n - 1) & hash)) == null) {
1009	>	if (casTabAt(tab, i, null,
1010	>	new Node<K,V>(hash, key, value, null)))
1011	>	break;                   // no lock when adding to empty bin
1012		}
1013	<	else if (e.hash < 0)
1014	<	tab = (Node[])e.key;
1013	>	else if ((fh = f.hash) == MOVED)
1014	>	tab = helpTransfer(tab, f);
1015		else {
1016	<	boolean validated = false;
1017	<	boolean checkSize = false;
1018	<	synchronized (e) {
1019	<	if (tabAt(tab, i) == e) {
1020	<	validated = true;
1021	<	for (Node first = e;;) {
1022	<	Object ek, ev;
1023	<	if (e.hash == h &&
1024	<	(ek = e.key) != null &&
1025	<	(ev = e.val) != null &&
1026	<	(k == ek || k.equals(ek))) {
1027	<	oldVal = ev;
1028	<	if (replace)
1029	<	e.val = v;
1030	<	break;
1016	>	V oldVal = null;
1017	>	synchronized (f) {
1018	>	if (tabAt(tab, i) == f) {
1019	>	if (fh >= 0) {
1020	>	binCount = 1;
1021	>	for (Node<K,V> e = f;; ++binCount) {
1022	>	K ek;
1023	>	if (e.hash == hash &&
1024	>	((ek = e.key) == key ||
1025	>	(ek != null && key.equals(ek)))) {
1026	>	oldVal = e.val;
1027	>	if (!onlyIfAbsent)
1028	>	e.val = value;
1029	>	break;
1030	>	}
1031	>	Node<K,V> pred = e;
1032	>	if ((e = e.next) == null) {
1033	>	pred.next = new Node<K,V>(hash, key,
1034	>	value, null);
1035	>	break;
1036	>	}
1037		}
1038	<	Node last = e;
1039	<	if ((e = e.next) == null) {
1040	<	last.next = new Node(h, k, v, null);
1041	<	if (last != first || tab.length <= 64)
1042	<	checkSize = true;
1043	<	break;
1038	>	}
1039	>	else if (f instanceof TreeBin) {
1040	>	Node<K,V> p;
1041	>	binCount = 2;
1042	>	if ((p = ((TreeBin<K,V>)f).putTreeVal(hash, key,
1043	>	value)) != null) {
1044	>	oldVal = p.val;
1045	>	if (!onlyIfAbsent)
1046	>	p.val = value;
1047		}
1048		}
1049		}
1050		}
1051	<	if (validated) {
1052	<	if (checkSize && tab.length < MAXIMUM_CAPACITY &&
1053	<	resizing == 0 && counter.sum() >= threshold)
1054	<	grow(0);
1051	>	if (binCount != 0) {
1052	>	if (binCount >= TREEIFY_THRESHOLD)
1053	>	treeifyBin(tab, i);
1054	>	if (oldVal != null)
1055	>	return oldVal;
1056		break;
1057		}
1058		}
1059		}
1060	<	if (oldVal == null)
1061	<	counter.increment();
1062	<	return oldVal;
1060	>	addCount(1L, binCount);
1061	>	return null;
1062	>	}
1063	>
1064	>	/**
1065	>	* Copies all of the mappings from the specified map to this one.
1066	>	* These mappings replace any mappings that this map had for any of the
1067	>	* keys currently in the specified map.
1068	>	*
1069	>	* @param m mappings to be stored in this map
1070	>	*/
1071	>	public void putAll(Map<? extends K, ? extends V> m) {
1072	>	tryPresize(m.size());
1073	>	for (Map.Entry<? extends K, ? extends V> e : m.entrySet())
1074	>	putVal(e.getKey(), e.getValue(), false);
1075	>	}
1076	>
1077	>	/**
1078	>	* Removes the key (and its corresponding value) from this map.
1079	>	* This method does nothing if the key is not in the map.
1080	>	*
1081	>	* @param  key the key that needs to be removed
1082	>	* @return the previous value associated with {@code key}, or
1083	>	*         {@code null} if there was no mapping for {@code key}
1084	>	* @throws NullPointerException if the specified key is null
1085	>	*/
1086	>	public V remove(Object key) {
1087	>	return replaceNode(key, null, null);
1088		}
1089
1090		/**
#	Line 427 | Line 1092 | public class ConcurrentHashMapV8<K, V>
1092		* Replaces node value with v, conditional upon match of cv if
1093		* non-null.  If resulting value is null, delete.
1094		*/
1095	<	private final Object internalReplace(Object k, Object v, Object cv) {
1096	<	int h = spread(k.hashCode());
1097	<	Object oldVal = null;
1098	<	Node e; int i;
1099	<	Node[] tab = table;
1100	<	while (tab != null &&
1101	<	(e = tabAt(tab, i = (tab.length - 1) & h)) != null) {
1102	<	if (e.hash < 0)
1103	<	tab = (Node[])e.key;
1095	>	final V replaceNode(Object key, V value, Object cv) {
1096	>	int hash = spread(key.hashCode());
1097	>	for (Node<K,V>[] tab = table;;) {
1098	>	Node<K,V> f; int n, i, fh;
1099	>	if (tab == null || (n = tab.length) == 0 ||
1100	>	(f = tabAt(tab, i = (n - 1) & hash)) == null)
1101	>	break;
1102	>	else if ((fh = f.hash) == MOVED)
1103	>	tab = helpTransfer(tab, f);
1104		else {
1105	+	V oldVal = null;
1106		boolean validated = false;
1107	<	boolean deleted = false;
1108	<	synchronized (e) {
1109	<	if (tabAt(tab, i) == e) {
1110	<	validated = true;
1111	<	Node pred = null;
1112	<	do {
1113	<	Object ek, ev;
1114	<	if (e.hash == h &&
1115	<	(ek = e.key) != null &&
1116	<	(ev = e.val) != null &&
1117	<	(k == ek || k.equals(ek))) {
1118	<	if (cv == null || cv == ev || cv.equals(ev)) {
1119	<	oldVal = ev;
1120	<	if ((e.val = v) == null) {
1121	<	deleted = true;
1122	<	Node en = e.next;
1123	<	if (pred != null)
458	<	pred.next = en;
1107	>	synchronized (f) {
1108	>	if (tabAt(tab, i) == f) {
1109	>	if (fh >= 0) {
1110	>	validated = true;
1111	>	for (Node<K,V> e = f, pred = null;;) {
1112	>	K ek;
1113	>	if (e.hash == hash &&
1114	>	((ek = e.key) == key ||
1115	>	(ek != null && key.equals(ek)))) {
1116	>	V ev = e.val;
1117	>	if (cv == null || cv == ev ||
1118	>	(ev != null && cv.equals(ev))) {
1119	>	oldVal = ev;
1120	>	if (value != null)
1121	>	e.val = value;
1122	>	else if (pred != null)
1123	>	pred.next = e.next;
1124		else
1125	<	setTabAt(tab, i, en);
1125	>	setTabAt(tab, i, e.next);
1126		}
1127	+	break;
1128		}
1129	<	break;
1129	>	pred = e;
1130	>	if ((e = e.next) == null)
1131	>	break;
1132		}
1133	<	pred = e;
1134	<	} while ((e = e.next) != null);
1133	>	}
1134	>	else if (f instanceof TreeBin) {
1135	>	validated = true;
1136	>	TreeBin<K,V> t = (TreeBin<K,V>)f;
1137	>	TreeNode<K,V> r, p;
1138	>	if ((r = t.root) != null &&
1139	>	(p = r.findTreeNode(hash, key, null)) != null) {
1140	>	V pv = p.val;
1141	>	if (cv == null || cv == pv ||
1142	>	(pv != null && cv.equals(pv))) {
1143	>	oldVal = pv;
1144	>	if (value != null)
1145	>	p.val = value;
1146	>	else if (t.removeTreeNode(p))
1147	>	setTabAt(tab, i, untreeify(t.first));
1148	>	}
1149	>	}
1150	>	}
1151		}
1152		}
1153		if (validated) {
1154	<	if (deleted)
1155	<	counter.decrement();
1154	>	if (oldVal != null) {
1155	>	if (value == null)
1156	>	addCount(-1L, -1);
1157	>	return oldVal;
1158	>	}
1159		break;
1160		}
1161		}
1162		}
1163	<	return oldVal;
1163	>	return null;
1164		}
1165
1166	<	/** Implementation for computeIfAbsent and compute */
1167	<	@SuppressWarnings("unchecked")
1168	<	private final V internalCompute(K k,
1169	<	MappingFunction<? super K, ? extends V> f,
1170	<	boolean replace) {
1171	<	int h = spread(k.hashCode());
1172	<	V val = null;
1173	<	boolean added = false;
1174	<	Node[] tab = table;
1166	>	/**
1167	>	* Removes all of the mappings from this map.
1168	>	*/
1169	>	public void clear() {
1170	>	long delta = 0L; // negative number of deletions
1171	>	int i = 0;
1172	>	Node<K,V>[] tab = table;
1173	>	while (tab != null && i < tab.length) {
1174	>	int fh;
1175	>	Node<K,V> f = tabAt(tab, i);
1176	>	if (f == null)
1177	>	++i;
1178	>	else if ((fh = f.hash) == MOVED) {
1179	>	tab = helpTransfer(tab, f);
1180	>	i = 0; // restart
1181	>	}
1182	>	else {
1183	>	synchronized (f) {
1184	>	if (tabAt(tab, i) == f) {
1185	>	Node<K,V> p = (fh >= 0 ? f :
1186	>	(f instanceof TreeBin) ?
1187	>	((TreeBin<K,V>)f).first : null);
1188	>	while (p != null) {
1189	>	--delta;
1190	>	p = p.next;
1191	>	}
1192	>	setTabAt(tab, i++, null);
1193	>	}
1194	>	}
1195	>	}
1196	>	}
1197	>	if (delta != 0L)
1198	>	addCount(delta, -1);
1199	>	}
1200	>
1201	>	/**
1202	>	* Returns a {@link Set} view of the keys contained in this map.
1203	>	* The set is backed by the map, so changes to the map are
1204	>	* reflected in the set, and vice-versa. The set supports element
1205	>	* removal, which removes the corresponding mapping from this map,
1206	>	* via the {@code Iterator.remove}, {@code Set.remove},
1207	>	* {@code removeAll}, {@code retainAll}, and {@code clear}
1208	>	* operations.  It does not support the {@code add} or
1209	>	* {@code addAll} operations.
1210	>	*
1211	>	* <p>The view's {@code iterator} is a "weakly consistent" iterator
1212	>	* that will never throw {@link ConcurrentModificationException},
1213	>	* and guarantees to traverse elements as they existed upon
1214	>	* construction of the iterator, and may (but is not guaranteed to)
1215	>	* reflect any modifications subsequent to construction.
1216	>	*
1217	>	* @return the set view
1218	>	*/
1219	>	public KeySetView<K,V> keySet() {
1220	>	KeySetView<K,V> ks;
1221	>	return (ks = keySet) != null ? ks : (keySet = new KeySetView<K,V>(this, null));
1222	>	}
1223	>
1224	>	/**
1225	>	* Returns a {@link Collection} view of the values contained in this map.
1226	>	* The collection is backed by the map, so changes to the map are
1227	>	* reflected in the collection, and vice-versa.  The collection
1228	>	* supports element removal, which removes the corresponding
1229	>	* mapping from this map, via the {@code Iterator.remove},
1230	>	* {@code Collection.remove}, {@code removeAll},
1231	>	* {@code retainAll}, and {@code clear} operations.  It does not
1232	>	* support the {@code add} or {@code addAll} operations.
1233	>	*
1234	>	* <p>The view's {@code iterator} is a "weakly consistent" iterator
1235	>	* that will never throw {@link ConcurrentModificationException},
1236	>	* and guarantees to traverse elements as they existed upon
1237	>	* construction of the iterator, and may (but is not guaranteed to)
1238	>	* reflect any modifications subsequent to construction.
1239	>	*
1240	>	* @return the collection view
1241	>	*/
1242	>	public Collection<V> values() {
1243	>	ValuesView<K,V> vs;
1244	>	return (vs = values) != null ? vs : (values = new ValuesView<K,V>(this));
1245	>	}
1246	>
1247	>	/**
1248	>	* Returns a {@link Set} view of the mappings contained in this map.
1249	>	* The set is backed by the map, so changes to the map are
1250	>	* reflected in the set, and vice-versa.  The set supports element
1251	>	* removal, which removes the corresponding mapping from the map,
1252	>	* via the {@code Iterator.remove}, {@code Set.remove},
1253	>	* {@code removeAll}, {@code retainAll}, and {@code clear}
1254	>	* operations.
1255	>	*
1256	>	* <p>The view's {@code iterator} is a "weakly consistent" iterator
1257	>	* that will never throw {@link ConcurrentModificationException},
1258	>	* and guarantees to traverse elements as they existed upon
1259	>	* construction of the iterator, and may (but is not guaranteed to)
1260	>	* reflect any modifications subsequent to construction.
1261	>	*
1262	>	* @return the set view
1263	>	*/
1264	>	public Set<Map.Entry<K,V>> entrySet() {
1265	>	EntrySetView<K,V> es;
1266	>	return (es = entrySet) != null ? es : (entrySet = new EntrySetView<K,V>(this));
1267	>	}
1268	>
1269	>	/**
1270	>	* Returns the hash code value for this {@link Map}, i.e.,
1271	>	* the sum of, for each key-value pair in the map,
1272	>	* {@code key.hashCode() ^ value.hashCode()}.
1273	>	*
1274	>	* @return the hash code value for this map
1275	>	*/
1276	>	public int hashCode() {
1277	>	int h = 0;
1278	>	Node<K,V>[] t;
1279	>	if ((t = table) != null) {
1280	>	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
1281	>	for (Node<K,V> p; (p = it.advance()) != null; )
1282	>	h += p.key.hashCode() ^ p.val.hashCode();
1283	>	}
1284	>	return h;
1285	>	}
1286	>
1287	>	/**
1288	>	* Returns a string representation of this map.  The string
1289	>	* representation consists of a list of key-value mappings (in no
1290	>	* particular order) enclosed in braces ("{@code {}}").  Adjacent
1291	>	* mappings are separated by the characters {@code ", "} (comma
1292	>	* and space).  Each key-value mapping is rendered as the key
1293	>	* followed by an equals sign ("{@code =}") followed by the
1294	>	* associated value.
1295	>	*
1296	>	* @return a string representation of this map
1297	>	*/
1298	>	public String toString() {
1299	>	Node<K,V>[] t;
1300	>	int f = (t = table) == null ? 0 : t.length;
1301	>	Traverser<K,V> it = new Traverser<K,V>(t, f, 0, f);
1302	>	StringBuilder sb = new StringBuilder();
1303	>	sb.append('{');
1304	>	Node<K,V> p;
1305	>	if ((p = it.advance()) != null) {
1306	>	for (;;) {
1307	>	K k = p.key;
1308	>	V v = p.val;
1309	>	sb.append(k == this ? "(this Map)" : k);
1310	>	sb.append('=');
1311	>	sb.append(v == this ? "(this Map)" : v);
1312	>	if ((p = it.advance()) == null)
1313	>	break;
1314	>	sb.append(',').append(' ');
1315	>	}
1316	>	}
1317	>	return sb.append('}').toString();
1318	>	}
1319	>
1320	>	/**
1321	>	* Compares the specified object with this map for equality.
1322	>	* Returns {@code true} if the given object is a map with the same
1323	>	* mappings as this map.  This operation may return misleading
1324	>	* results if either map is concurrently modified during execution
1325	>	* of this method.
1326	>	*
1327	>	* @param o object to be compared for equality with this map
1328	>	* @return {@code true} if the specified object is equal to this map
1329	>	*/
1330	>	public boolean equals(Object o) {
1331	>	if (o != this) {
1332	>	if (!(o instanceof Map))
1333	>	return false;
1334	>	Map<?,?> m = (Map<?,?>) o;
1335	>	Node<K,V>[] t;
1336	>	int f = (t = table) == null ? 0 : t.length;
1337	>	Traverser<K,V> it = new Traverser<K,V>(t, f, 0, f);
1338	>	for (Node<K,V> p; (p = it.advance()) != null; ) {
1339	>	V val = p.val;
1340	>	Object v = m.get(p.key);
1341	>	if (v == null || (v != val && !v.equals(val)))
1342	>	return false;
1343	>	}
1344	>	for (Map.Entry<?,?> e : m.entrySet()) {
1345	>	Object mk, mv, v;
1346	>	if ((mk = e.getKey()) == null ||
1347	>	(mv = e.getValue()) == null ||
1348	>	(v = get(mk)) == null ||
1349	>	(mv != v && !mv.equals(v)))
1350	>	return false;
1351	>	}
1352	>	}
1353	>	return true;
1354	>	}
1355	>
1356	>	/**
1357	>	* Stripped-down version of helper class used in previous version,
1358	>	* declared for the sake of serialization compatibility
1359	>	*/
1360	>	static class Segment<K,V> extends ReentrantLock implements Serializable {
1361	>	private static final long serialVersionUID = 2249069246763182397L;
1362	>	final float loadFactor;
1363	>	Segment(float lf) { this.loadFactor = lf; }
1364	>	}
1365	>
1366	>	/**
1367	>	* Saves the state of the {@code ConcurrentHashMapV8} instance to a
1368	>	* stream (i.e., serializes it).
1369	>	* @param s the stream
1370	>	* @throws java.io.IOException if an I/O error occurs
1371	>	* @serialData
1372	>	* the key (Object) and value (Object)
1373	>	* for each key-value mapping, followed by a null pair.
1374	>	* The key-value mappings are emitted in no particular order.
1375	>	*/
1376	>	private void writeObject(java.io.ObjectOutputStream s)
1377	>	throws java.io.IOException {
1378	>	// For serialization compatibility
1379	>	// Emulate segment calculation from previous version of this class
1380	>	int sshift = 0;
1381	>	int ssize = 1;
1382	>	while (ssize < DEFAULT_CONCURRENCY_LEVEL) {
1383	>	++sshift;
1384	>	ssize <<= 1;
1385	>	}
1386	>	int segmentShift = 32 - sshift;
1387	>	int segmentMask = ssize - 1;
1388	>	@SuppressWarnings("unchecked") Segment<K,V>[] segments = (Segment<K,V>[])
1389	>	new Segment<?,?>[DEFAULT_CONCURRENCY_LEVEL];
1390	>	for (int i = 0; i < segments.length; ++i)
1391	>	segments[i] = new Segment<K,V>(LOAD_FACTOR);
1392	>	s.putFields().put("segments", segments);
1393	>	s.putFields().put("segmentShift", segmentShift);
1394	>	s.putFields().put("segmentMask", segmentMask);
1395	>	s.writeFields();
1396	>
1397	>	Node<K,V>[] t;
1398	>	if ((t = table) != null) {
1399	>	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
1400	>	for (Node<K,V> p; (p = it.advance()) != null; ) {
1401	>	s.writeObject(p.key);
1402	>	s.writeObject(p.val);
1403	>	}
1404	>	}
1405	>	s.writeObject(null);
1406	>	s.writeObject(null);
1407	>	segments = null; // throw away
1408	>	}
1409	>
1410	>	/**
1411	>	* Reconstitutes the instance from a stream (that is, deserializes it).
1412	>	* @param s the stream
1413	>	* @throws ClassNotFoundException if the class of a serialized object
1414	>	*         could not be found
1415	>	* @throws java.io.IOException if an I/O error occurs
1416	>	*/
1417	>	private void readObject(java.io.ObjectInputStream s)
1418	>	throws java.io.IOException, ClassNotFoundException {
1419	>	/*
1420	>	* To improve performance in typical cases, we create nodes
1421	>	* while reading, then place in table once size is known.
1422	>	* However, we must also validate uniqueness and deal with
1423	>	* overpopulated bins while doing so, which requires
1424	>	* specialized versions of putVal mechanics.
1425	>	*/
1426	>	sizeCtl = -1; // force exclusion for table construction
1427	>	s.defaultReadObject();
1428	>	long size = 0L;
1429	>	Node<K,V> p = null;
1430		for (;;) {
1431	<	Node e; int i;
1432	<	if (tab == null)
1433	<	tab = grow(0);
1434	<	else if ((e = tabAt(tab, i = (tab.length - 1) & h)) == null) {
1435	<	Node node = new Node(h, k, null, null);
1436	<	boolean validated = false;
1437	<	synchronized (node) {
1438	<	if (casTabAt(tab, i, null, node)) {
1439	<	validated = true;
1431	>	@SuppressWarnings("unchecked") K k = (K) s.readObject();
1432	>	@SuppressWarnings("unchecked") V v = (V) s.readObject();
1433	>	if (k != null && v != null) {
1434	>	p = new Node<K,V>(spread(k.hashCode()), k, v, p);
1435	>	++size;
1436	>	}
1437	>	else
1438	>	break;
1439	>	}
1440	>	if (size == 0L)
1441	>	sizeCtl = 0;
1442	>	else {
1443	>	int n;
1444	>	if (size >= (long)(MAXIMUM_CAPACITY >>> 1))
1445	>	n = MAXIMUM_CAPACITY;
1446	>	else {
1447	>	int sz = (int)size;
1448	>	n = tableSizeFor(sz + (sz >>> 1) + 1);
1449	>	}
1450	>	@SuppressWarnings("unchecked")
1451	>	Node<K,V>[] tab = (Node<K,V>[])new Node<?,?>[n];
1452	>	int mask = n - 1;
1453	>	long added = 0L;
1454	>	while (p != null) {
1455	>	boolean insertAtFront;
1456	>	Node<K,V> next = p.next, first;
1457	>	int h = p.hash, j = h & mask;
1458	>	if ((first = tabAt(tab, j)) == null)
1459	>	insertAtFront = true;
1460	>	else {
1461	>	K k = p.key;
1462	>	if (first.hash < 0) {
1463	>	TreeBin<K,V> t = (TreeBin<K,V>)first;
1464	>	if (t.putTreeVal(h, k, p.val) == null)
1465	>	++added;
1466	>	insertAtFront = false;
1467	>	}
1468	>	else {
1469	>	int binCount = 0;
1470	>	insertAtFront = true;
1471	>	Node<K,V> q; K qk;
1472	>	for (q = first; q != null; q = q.next) {
1473	>	if (q.hash == h &&
1474	>	((qk = q.key) == k ||
1475	>	(qk != null && k.equals(qk)))) {
1476	>	insertAtFront = false;
1477	>	break;
1478	>	}
1479	>	++binCount;
1480	>	}
1481	>	if (insertAtFront && binCount >= TREEIFY_THRESHOLD) {
1482	>	insertAtFront = false;
1483	>	++added;
1484	>	p.next = first;
1485	>	TreeNode<K,V> hd = null, tl = null;
1486	>	for (q = p; q != null; q = q.next) {
1487	>	TreeNode<K,V> t = new TreeNode<K,V>
1488	>	(q.hash, q.key, q.val, null, null);
1489	>	if ((t.prev = tl) == null)
1490	>	hd = t;
1491	>	else
1492	>	tl.next = t;
1493	>	tl = t;
1494	>	}
1495	>	setTabAt(tab, j, new TreeBin<K,V>(hd));
1496	>	}
1497	>	}
1498	>	}
1499	>	if (insertAtFront) {
1500	>	++added;
1501	>	p.next = first;
1502	>	setTabAt(tab, j, p);
1503	>	}
1504	>	p = next;
1505	>	}
1506	>	table = tab;
1507	>	sizeCtl = n - (n >>> 2);
1508	>	baseCount = added;
1509	>	}
1510	>	}
1511	>
1512	>	// ConcurrentMap methods
1513	>
1514	>	/**
1515	>	* {@inheritDoc}
1516	>	*
1517	>	* @return the previous value associated with the specified key,
1518	>	*         or {@code null} if there was no mapping for the key
1519	>	* @throws NullPointerException if the specified key or value is null
1520	>	*/
1521	>	public V putIfAbsent(K key, V value) {
1522	>	return putVal(key, value, true);
1523	>	}
1524	>
1525	>	/**
1526	>	* {@inheritDoc}
1527	>	*
1528	>	* @throws NullPointerException if the specified key is null
1529	>	*/
1530	>	public boolean remove(Object key, Object value) {
1531	>	if (key == null)
1532	>	throw new NullPointerException();
1533	>	return value != null && replaceNode(key, null, value) != null;
1534	>	}
1535	>
1536	>	/**
1537	>	* {@inheritDoc}
1538	>	*
1539	>	* @throws NullPointerException if any of the arguments are null
1540	>	*/
1541	>	public boolean replace(K key, V oldValue, V newValue) {
1542	>	if (key == null || oldValue == null || newValue == null)
1543	>	throw new NullPointerException();
1544	>	return replaceNode(key, newValue, oldValue) != null;
1545	>	}
1546	>
1547	>	/**
1548	>	* {@inheritDoc}
1549	>	*
1550	>	* @return the previous value associated with the specified key,
1551	>	*         or {@code null} if there was no mapping for the key
1552	>	* @throws NullPointerException if the specified key or value is null
1553	>	*/
1554	>	public V replace(K key, V value) {
1555	>	if (key == null || value == null)
1556	>	throw new NullPointerException();
1557	>	return replaceNode(key, value, null);
1558	>	}
1559	>
1560	>	// Overrides of JDK8+ Map extension method defaults
1561	>
1562	>	/**
1563	>	* Returns the value to which the specified key is mapped, or the
1564	>	* given default value if this map contains no mapping for the
1565	>	* key.
1566	>	*
1567	>	* @param key the key whose associated value is to be returned
1568	>	* @param defaultValue the value to return if this map contains
1569	>	* no mapping for the given key
1570	>	* @return the mapping for the key, if present; else the default value
1571	>	* @throws NullPointerException if the specified key is null
1572	>	*/
1573	>	public V getOrDefault(Object key, V defaultValue) {
1574	>	V v;
1575	>	return (v = get(key)) == null ? defaultValue : v;
1576	>	}
1577	>
1578	>	public void forEach(BiAction<? super K, ? super V> action) {
1579	>	if (action == null) throw new NullPointerException();
1580	>	Node<K,V>[] t;
1581	>	if ((t = table) != null) {
1582	>	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
1583	>	for (Node<K,V> p; (p = it.advance()) != null; ) {
1584	>	action.apply(p.key, p.val);
1585	>	}
1586	>	}
1587	>	}
1588	>
1589	>	public void replaceAll(BiFun<? super K, ? super V, ? extends V> function) {
1590	>	if (function == null) throw new NullPointerException();
1591	>	Node<K,V>[] t;
1592	>	if ((t = table) != null) {
1593	>	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
1594	>	for (Node<K,V> p; (p = it.advance()) != null; ) {
1595	>	V oldValue = p.val;
1596	>	for (K key = p.key;;) {
1597	>	V newValue = function.apply(key, oldValue);
1598	>	if (newValue == null)
1599	>	throw new NullPointerException();
1600	>	if (replaceNode(key, newValue, oldValue) != null ||
1601	>	(oldValue = get(key)) == null)
1602	>	break;
1603	>	}
1604	>	}
1605	>	}
1606	>	}
1607	>
1608	>	/**
1609	>	* If the specified key is not already associated with a value,
1610	>	* attempts to compute its value using the given mapping function
1611	>	* and enters it into this map unless {@code null}.  The entire
1612	>	* method invocation is performed atomically, so the function is
1613	>	* applied at most once per key.  Some attempted update operations
1614	>	* on this map by other threads may be blocked while computation
1615	>	* is in progress, so the computation should be short and simple,
1616	>	* and must not attempt to update any other mappings of this map.
1617	>	*
1618	>	* @param key key with which the specified value is to be associated
1619	>	* @param mappingFunction the function to compute a value
1620	>	* @return the current (existing or computed) value associated with
1621	>	*         the specified key, or null if the computed value is null
1622	>	* @throws NullPointerException if the specified key or mappingFunction
1623	>	*         is null
1624	>	* @throws IllegalStateException if the computation detectably
1625	>	*         attempts a recursive update to this map that would
1626	>	*         otherwise never complete
1627	>	* @throws RuntimeException or Error if the mappingFunction does so,
1628	>	*         in which case the mapping is left unestablished
1629	>	*/
1630	>	public V computeIfAbsent(K key, Fun<? super K, ? extends V> mappingFunction) {
1631	>	if (key == null || mappingFunction == null)
1632	>	throw new NullPointerException();
1633	>	int h = spread(key.hashCode());
1634	>	V val = null;
1635	>	int binCount = 0;
1636	>	for (Node<K,V>[] tab = table;;) {
1637	>	Node<K,V> f; int n, i, fh;
1638	>	if (tab == null || (n = tab.length) == 0)
1639	>	tab = initTable();
1640	>	else if ((f = tabAt(tab, i = (n - 1) & h)) == null) {
1641	>	Node<K,V> r = new ReservationNode<K,V>();
1642	>	synchronized (r) {
1643	>	if (casTabAt(tab, i, null, r)) {
1644	>	binCount = 1;
1645	>	Node<K,V> node = null;
1646		try {
1647	<	val = f.map(k);
1648	<	if (val != null) {
1649	<	node.val = val;
1647	>	if ((val = mappingFunction.apply(key)) != null)
1648	>	node = new Node<K,V>(h, key, val, null);
1649	>	} finally {
1650	>	setTabAt(tab, i, node);
1651	>	}
1652	>	}
1653	>	}
1654	>	if (binCount != 0)
1655	>	break;
1656	>	}
1657	>	else if ((fh = f.hash) == MOVED)
1658	>	tab = helpTransfer(tab, f);
1659	>	else {
1660	>	boolean added = false;
1661	>	synchronized (f) {
1662	>	if (tabAt(tab, i) == f) {
1663	>	if (fh >= 0) {
1664	>	binCount = 1;
1665	>	for (Node<K,V> e = f;; ++binCount) {
1666	>	K ek; V ev;
1667	>	if (e.hash == h &&
1668	>	((ek = e.key) == key ||
1669	>	(ek != null && key.equals(ek)))) {
1670	>	val = e.val;
1671	>	break;
1672	>	}
1673	>	Node<K,V> pred = e;
1674	>	if ((e = e.next) == null) {
1675	>	if ((val = mappingFunction.apply(key)) != null) {
1676	>	added = true;
1677	>	pred.next = new Node<K,V>(h, key, val, null);
1678	>	}
1679	>	break;
1680	>	}
1681	>	}
1682	>	}
1683	>	else if (f instanceof TreeBin) {
1684	>	binCount = 2;
1685	>	TreeBin<K,V> t = (TreeBin<K,V>)f;
1686	>	TreeNode<K,V> r, p;
1687	>	if ((r = t.root) != null &&
1688	>	(p = r.findTreeNode(h, key, null)) != null)
1689	>	val = p.val;
1690	>	else if ((val = mappingFunction.apply(key)) != null) {
1691		added = true;
1692	+	t.putTreeVal(h, key, val);
1693	+	}
1694	+	}
1695	+	}
1696	+	}
1697	+	if (binCount != 0) {
1698	+	if (binCount >= TREEIFY_THRESHOLD)
1699	+	treeifyBin(tab, i);
1700	+	if (!added)
1701	+	return val;
1702	+	break;
1703	+	}
1704	+	}
1705	+	}
1706	+	if (val != null)
1707	+	addCount(1L, binCount);
1708	+	return val;
1709	+	}
1710	+
1711	+	/**
1712	+	* If the value for the specified key is present, attempts to
1713	+	* compute a new mapping given the key and its current mapped
1714	+	* value.  The entire method invocation is performed atomically.
1715	+	* Some attempted update operations on this map by other threads
1716	+	* may be blocked while computation is in progress, so the
1717	+	* computation should be short and simple, and must not attempt to
1718	+	* update any other mappings of this map.
1719	+	*
1720	+	* @param key key with which a value may be associated
1721	+	* @param remappingFunction the function to compute a value
1722	+	* @return the new value associated with the specified key, or null if none
1723	+	* @throws NullPointerException if the specified key or remappingFunction
1724	+	*         is null
1725	+	* @throws IllegalStateException if the computation detectably
1726	+	*         attempts a recursive update to this map that would
1727	+	*         otherwise never complete
1728	+	* @throws RuntimeException or Error if the remappingFunction does so,
1729	+	*         in which case the mapping is unchanged
1730	+	*/
1731	+	public V computeIfPresent(K key, BiFun<? super K, ? super V, ? extends V> remappingFunction) {
1732	+	if (key == null || remappingFunction == null)
1733	+	throw new NullPointerException();
1734	+	int h = spread(key.hashCode());
1735	+	V val = null;
1736	+	int delta = 0;
1737	+	int binCount = 0;
1738	+	for (Node<K,V>[] tab = table;;) {
1739	+	Node<K,V> f; int n, i, fh;
1740	+	if (tab == null || (n = tab.length) == 0)
1741	+	tab = initTable();
1742	+	else if ((f = tabAt(tab, i = (n - 1) & h)) == null)
1743	+	break;
1744	+	else if ((fh = f.hash) == MOVED)
1745	+	tab = helpTransfer(tab, f);
1746	+	else {
1747	+	synchronized (f) {
1748	+	if (tabAt(tab, i) == f) {
1749	+	if (fh >= 0) {
1750	+	binCount = 1;
1751	+	for (Node<K,V> e = f, pred = null;; ++binCount) {
1752	+	K ek;
1753	+	if (e.hash == h &&
1754	+	((ek = e.key) == key ||
1755	+	(ek != null && key.equals(ek)))) {
1756	+	val = remappingFunction.apply(key, e.val);
1757	+	if (val != null)
1758	+	e.val = val;
1759	+	else {
1760	+	delta = -1;
1761	+	Node<K,V> en = e.next;
1762	+	if (pred != null)
1763	+	pred.next = en;
1764	+	else
1765	+	setTabAt(tab, i, en);
1766	+	}
1767	+	break;
1768	+	}
1769	+	pred = e;
1770	+	if ((e = e.next) == null)
1771	+	break;
1772	+	}
1773	+	}
1774	+	else if (f instanceof TreeBin) {
1775	+	binCount = 2;
1776	+	TreeBin<K,V> t = (TreeBin<K,V>)f;
1777	+	TreeNode<K,V> r, p;
1778	+	if ((r = t.root) != null &&
1779	+	(p = r.findTreeNode(h, key, null)) != null) {
1780	+	val = remappingFunction.apply(key, p.val);
1781	+	if (val != null)
1782	+	p.val = val;
1783	+	else {
1784	+	delta = -1;
1785	+	if (t.removeTreeNode(p))
1786	+	setTabAt(tab, i, untreeify(t.first));
1787	+	}
1788	+	}
1789	+	}
1790	+	}
1791	+	}
1792	+	if (binCount != 0)
1793	+	break;
1794	+	}
1795	+	}
1796	+	if (delta != 0)
1797	+	addCount((long)delta, binCount);
1798	+	return val;
1799	+	}
1800	+
1801	+	/**
1802	+	* Attempts to compute a mapping for the specified key and its
1803	+	* current mapped value (or {@code null} if there is no current
1804	+	* mapping). The entire method invocation is performed atomically.
1805	+	* Some attempted update operations on this map by other threads
1806	+	* may be blocked while computation is in progress, so the
1807	+	* computation should be short and simple, and must not attempt to
1808	+	* update any other mappings of this Map.
1809	+	*
1810	+	* @param key key with which the specified value is to be associated
1811	+	* @param remappingFunction the function to compute a value
1812	+	* @return the new value associated with the specified key, or null if none
1813	+	* @throws NullPointerException if the specified key or remappingFunction
1814	+	*         is null
1815	+	* @throws IllegalStateException if the computation detectably
1816	+	*         attempts a recursive update to this map that would
1817	+	*         otherwise never complete
1818	+	* @throws RuntimeException or Error if the remappingFunction does so,
1819	+	*         in which case the mapping is unchanged
1820	+	*/
1821	+	public V compute(K key,
1822	+	BiFun<? super K, ? super V, ? extends V> remappingFunction) {
1823	+	if (key == null || remappingFunction == null)
1824	+	throw new NullPointerException();
1825	+	int h = spread(key.hashCode());
1826	+	V val = null;
1827	+	int delta = 0;
1828	+	int binCount = 0;
1829	+	for (Node<K,V>[] tab = table;;) {
1830	+	Node<K,V> f; int n, i, fh;
1831	+	if (tab == null || (n = tab.length) == 0)
1832	+	tab = initTable();
1833	+	else if ((f = tabAt(tab, i = (n - 1) & h)) == null) {
1834	+	Node<K,V> r = new ReservationNode<K,V>();
1835	+	synchronized (r) {
1836	+	if (casTabAt(tab, i, null, r)) {
1837	+	binCount = 1;
1838	+	Node<K,V> node = null;
1839	+	try {
1840	+	if ((val = remappingFunction.apply(key, null)) != null) {
1841	+	delta = 1;
1842	+	node = new Node<K,V>(h, key, val, null);
1843		}
1844		} finally {
1845	<	if (!added)
506	<	setTabAt(tab, i, null);
1845	>	setTabAt(tab, i, node);
1846		}
1847		}
1848		}
1849	<	if (validated)
1849	>	if (binCount != 0)
1850		break;
1851		}
1852	<	else if (e.hash < 0)
1853	<	tab = (Node[])e.key;
515	<	else if (Thread.holdsLock(e))
516	<	throw new IllegalStateException("Recursive map computation");
1852	>	else if ((fh = f.hash) == MOVED)
1853	>	tab = helpTransfer(tab, f);
1854		else {
1855	<	boolean validated = false;
1856	<	boolean checkSize = false;
1857	<	synchronized (e) {
1858	<	if (tabAt(tab, i) == e) {
1859	<	validated = true;
1860	<	for (Node first = e;;) {
1861	<	Object ek, ev, fv;
1862	<	if (e.hash == h &&
1863	<	(ek = e.key) != null &&
1864	<	(ev = e.val) != null &&
1865	<	(k == ek || k.equals(ek))) {
1866	<	if (replace && (fv = f.map(k)) != null)
1867	<	ev = e.val = fv;
1868	<	val = (V)ev;
1869	<	break;
1855	>	synchronized (f) {
1856	>	if (tabAt(tab, i) == f) {
1857	>	if (fh >= 0) {
1858	>	binCount = 1;
1859	>	for (Node<K,V> e = f, pred = null;; ++binCount) {
1860	>	K ek;
1861	>	if (e.hash == h &&
1862	>	((ek = e.key) == key ||
1863	>	(ek != null && key.equals(ek)))) {
1864	>	val = remappingFunction.apply(key, e.val);
1865	>	if (val != null)
1866	>	e.val = val;
1867	>	else {
1868	>	delta = -1;
1869	>	Node<K,V> en = e.next;
1870	>	if (pred != null)
1871	>	pred.next = en;
1872	>	else
1873	>	setTabAt(tab, i, en);
1874	>	}
1875	>	break;
1876	>	}
1877	>	pred = e;
1878	>	if ((e = e.next) == null) {
1879	>	val = remappingFunction.apply(key, null);
1880	>	if (val != null) {
1881	>	delta = 1;
1882	>	pred.next =
1883	>	new Node<K,V>(h, key, val, null);
1884	>	}
1885	>	break;
1886	>	}
1887		}
1888	<	Node last = e;
1889	<	if ((e = e.next) == null) {
1890	<	if ((val = f.map(k)) != null) {
1891	<	last.next = new Node(h, k, val, null);
1892	<	added = true;
1893	<	if (last != first || tab.length <= 64)
1894	<	checkSize = true;
1888	>	}
1889	>	else if (f instanceof TreeBin) {
1890	>	binCount = 1;
1891	>	TreeBin<K,V> t = (TreeBin<K,V>)f;
1892	>	TreeNode<K,V> r, p;
1893	>	if ((r = t.root) != null)
1894	>	p = r.findTreeNode(h, key, null);
1895	>	else
1896	>	p = null;
1897	>	V pv = (p == null) ? null : p.val;
1898	>	val = remappingFunction.apply(key, pv);
1899	>	if (val != null) {
1900	>	if (p != null)
1901	>	p.val = val;
1902	>	else {
1903	>	delta = 1;
1904	>	t.putTreeVal(h, key, val);
1905		}
1906	<	break;
1906	>	}
1907	>	else if (p != null) {
1908	>	delta = -1;
1909	>	if (t.removeTreeNode(p))
1910	>	setTabAt(tab, i, untreeify(t.first));
1911		}
1912		}
1913		}
1914		}
1915	<	if (validated) {
1916	<	if (checkSize && tab.length < MAXIMUM_CAPACITY &&
1917	<	resizing == 0 && counter.sum() >= threshold)
550	<	grow(0);
1915	>	if (binCount != 0) {
1916	>	if (binCount >= TREEIFY_THRESHOLD)
1917	>	treeifyBin(tab, i);
1918		break;
1919		}
1920		}
1921		}
1922	<	if (added)
1923	<	counter.increment();
1922	>	if (delta != 0)
1923	>	addCount((long)delta, binCount);
1924		return val;
1925		}
1926
1927	<	/*
1928	<	* Reclassifies nodes in each bin to new table.  Because we are
1929	<	* using power-of-two expansion, the elements from each bin must
1930	<	* either stay at same index, or move with a power of two
1931	<	* offset. We eliminate unnecessary node creation by catching
1932	<	* cases where old nodes can be reused because their next fields
1933	<	* won't change.  Statistically, at the default threshold, only
1934	<	* about one-sixth of them need cloning when a table doubles. The
1935	<	* nodes they replace will be garbage collectable as soon as they
1936	<	* are no longer referenced by any reader thread that may be in
1937	<	* the midst of concurrently traversing table.
1938	<	*
1939	<	* Transfers are done from the bottom up to preserve iterator
1940	<	* traversability. On each step, the old bin is locked,
1941	<	* moved/copied, and then replaced with a forwarding node.
1942	<	*/
1943	<	private static final void transfer(Node[] tab, Node[] nextTab) {
1944	<	int n = tab.length;
1945	<	int mask = nextTab.length - 1;
1946	<	Node fwd = new Node(MOVED, nextTab, null, null);
1947	<	for (int i = n - 1; i >= 0; --i) {
1948	<	for (Node e;;) {
1949	<	if ((e = tabAt(tab, i)) == null) {
1950	<	if (casTabAt(tab, i, e, fwd))
1951	<	break;
1927	>	/**
1928	>	* If the specified key is not already associated with a
1929	>	* (non-null) value, associates it with the given value.
1930	>	* Otherwise, replaces the value with the results of the given
1931	>	* remapping function, or removes if {@code null}. The entire
1932	>	* method invocation is performed atomically.  Some attempted
1933	>	* update operations on this map by other threads may be blocked
1934	>	* while computation is in progress, so the computation should be
1935	>	* short and simple, and must not attempt to update any other
1936	>	* mappings of this Map.
1937	>	*
1938	>	* @param key key with which the specified value is to be associated
1939	>	* @param value the value to use if absent
1940	>	* @param remappingFunction the function to recompute a value if present
1941	>	* @return the new value associated with the specified key, or null if none
1942	>	* @throws NullPointerException if the specified key or the
1943	>	*         remappingFunction is null
1944	>	* @throws RuntimeException or Error if the remappingFunction does so,
1945	>	*         in which case the mapping is unchanged
1946	>	*/
1947	>	public V merge(K key, V value, BiFun<? super V, ? super V, ? extends V> remappingFunction) {
1948	>	if (key == null || value == null || remappingFunction == null)
1949	>	throw new NullPointerException();
1950	>	int h = spread(key.hashCode());
1951	>	V val = null;
1952	>	int delta = 0;
1953	>	int binCount = 0;
1954	>	for (Node<K,V>[] tab = table;;) {
1955	>	Node<K,V> f; int n, i, fh;
1956	>	if (tab == null || (n = tab.length) == 0)
1957	>	tab = initTable();
1958	>	else if ((f = tabAt(tab, i = (n - 1) & h)) == null) {
1959	>	if (casTabAt(tab, i, null, new Node<K,V>(h, key, value, null))) {
1960	>	delta = 1;
1961	>	val = value;
1962	>	break;
1963		}
1964	<	else {
1965	<	int idx = e.hash & mask;
1966	<	boolean validated = false;
1967	<	synchronized (e) {
1968	<	if (tabAt(tab, i) == e) {
1969	<	validated = true;
1970	<	Node lastRun = e;
1971	<	for (Node p = e.next; p != null; p = p.next) {
1972	<	int j = p.hash & mask;
1973	<	if (j != idx) {
1974	<	idx = j;
1975	<	lastRun = p;
1964	>	}
1965	>	else if ((fh = f.hash) == MOVED)
1966	>	tab = helpTransfer(tab, f);
1967	>	else {
1968	>	synchronized (f) {
1969	>	if (tabAt(tab, i) == f) {
1970	>	if (fh >= 0) {
1971	>	binCount = 1;
1972	>	for (Node<K,V> e = f, pred = null;; ++binCount) {
1973	>	K ek;
1974	>	if (e.hash == h &&
1975	>	((ek = e.key) == key ||
1976	>	(ek != null && key.equals(ek)))) {
1977	>	val = remappingFunction.apply(e.val, value);
1978	>	if (val != null)
1979	>	e.val = val;
1980	>	else {
1981	>	delta = -1;
1982	>	Node<K,V> en = e.next;
1983	>	if (pred != null)
1984	>	pred.next = en;
1985	>	else
1986	>	setTabAt(tab, i, en);
1987	>	}
1988	>	break;
1989	>	}
1990	>	pred = e;
1991	>	if ((e = e.next) == null) {
1992	>	delta = 1;
1993	>	val = value;
1994	>	pred.next =
1995	>	new Node<K,V>(h, key, val, null);
1996	>	break;
1997		}
1998		}
1999	<	relaxedSetTabAt(nextTab, idx, lastRun);
2000	<	for (Node p = e; p != lastRun; p = p.next) {
2001	<	int h = p.hash;
2002	<	int j = h & mask;
2003	<	Node r = relaxedTabAt(nextTab, j);
2004	<	relaxedSetTabAt(nextTab, j,
2005	<	new Node(h, p.key, p.val, r));
1999	>	}
2000	>	else if (f instanceof TreeBin) {
2001	>	binCount = 2;
2002	>	TreeBin<K,V> t = (TreeBin<K,V>)f;
2003	>	TreeNode<K,V> r = t.root;
2004	>	TreeNode<K,V> p = (r == null) ? null :
2005	>	r.findTreeNode(h, key, null);
2006	>	val = (p == null) ? value :
2007	>	remappingFunction.apply(p.val, value);
2008	>	if (val != null) {
2009	>	if (p != null)
2010	>	p.val = val;
2011	>	else {
2012	>	delta = 1;
2013	>	t.putTreeVal(h, key, val);
2014	>	}
2015	>	}
2016	>	else if (p != null) {
2017	>	delta = -1;
2018	>	if (t.removeTreeNode(p))
2019	>	setTabAt(tab, i, untreeify(t.first));
2020		}
608	–	setTabAt(tab, i, fwd);
2021		}
2022		}
2023	<	if (validated)
2024	<	break;
2023	>	}
2024	>	if (binCount != 0) {
2025	>	if (binCount >= TREEIFY_THRESHOLD)
2026	>	treeifyBin(tab, i);
2027	>	break;
2028		}
2029		}
2030		}
2031	+	if (delta != 0)
2032	+	addCount((long)delta, binCount);
2033	+	return val;
2034		}
2035
2036	+	// Hashtable legacy methods
2037	+
2038		/**
2039	<	* If not already resizing, initializes or creates next table and
2040	<	* transfers bins. Rechecks occupancy after a transfer to see if
2041	<	* another resize is already needed because resizings are lagging
2042	<	* additions.
2043	<	*
2044	<	* @param sizeHint overridden capacity target (nonzero only from putAll)
2045	<	* @return current table
2046	<	*/
2047	<	private final Node[] grow(int sizeHint) {
2048	<	if (resizing == 0 &&
2049	<	UNSAFE.compareAndSwapInt(this, resizingOffset, 0, 1)) {
2050	<	try {
2039	>	* Legacy method testing if some key maps into the specified value
2040	>	* in this table.  This method is identical in functionality to
2041	>	* {@link #containsValue(Object)}, and exists solely to ensure
2042	>	* full compatibility with class {@link java.util.Hashtable},
2043	>	* which supported this method prior to introduction of the
2044	>	* Java Collections framework.
2045	>	*
2046	>	* @param  value a value to search for
2047	>	* @return {@code true} if and only if some key maps to the
2048	>	*         {@code value} argument in this table as
2049	>	*         determined by the {@code equals} method;
2050	>	*         {@code false} otherwise
2051	>	* @throws NullPointerException if the specified value is null
2052	>	*/
2053	>	@Deprecated public boolean contains(Object value) {
2054	>	return containsValue(value);
2055	>	}
2056	>
2057	>	/**
2058	>	* Returns an enumeration of the keys in this table.
2059	>	*
2060	>	* @return an enumeration of the keys in this table
2061	>	* @see #keySet()
2062	>	*/
2063	>	public Enumeration<K> keys() {
2064	>	Node<K,V>[] t;
2065	>	int f = (t = table) == null ? 0 : t.length;
2066	>	return new KeyIterator<K,V>(t, f, 0, f, this);
2067	>	}
2068	>
2069	>	/**
2070	>	* Returns an enumeration of the values in this table.
2071	>	*
2072	>	* @return an enumeration of the values in this table
2073	>	* @see #values()
2074	>	*/
2075	>	public Enumeration<V> elements() {
2076	>	Node<K,V>[] t;
2077	>	int f = (t = table) == null ? 0 : t.length;
2078	>	return new ValueIterator<K,V>(t, f, 0, f, this);
2079	>	}
2080	>
2081	>	// ConcurrentHashMapV8-only methods
2082	>
2083	>	/**
2084	>	* Returns the number of mappings. This method should be used
2085	>	* instead of {@link #size} because a ConcurrentHashMapV8 may
2086	>	* contain more mappings than can be represented as an int. The
2087	>	* value returned is an estimate; the actual count may differ if
2088	>	* there are concurrent insertions or removals.
2089	>	*
2090	>	* @return the number of mappings
2091	>	* @since 1.8
2092	>	*/
2093	>	public long mappingCount() {
2094	>	long n = sumCount();
2095	>	return (n < 0L) ? 0L : n; // ignore transient negative values
2096	>	}
2097	>
2098	>	/**
2099	>	* Creates a new {@link Set} backed by a ConcurrentHashMapV8
2100	>	* from the given type to {@code Boolean.TRUE}.
2101	>	*
2102	>	* @return the new set
2103	>	* @since 1.8
2104	>	*/
2105	>	public static <K> KeySetView<K,Boolean> newKeySet() {
2106	>	return new KeySetView<K,Boolean>
2107	>	(new ConcurrentHashMapV8<K,Boolean>(), Boolean.TRUE);
2108	>	}
2109	>
2110	>	/**
2111	>	* Creates a new {@link Set} backed by a ConcurrentHashMapV8
2112	>	* from the given type to {@code Boolean.TRUE}.
2113	>	*
2114	>	* @param initialCapacity The implementation performs internal
2115	>	* sizing to accommodate this many elements.
2116	>	* @return the new set
2117	>	* @throws IllegalArgumentException if the initial capacity of
2118	>	* elements is negative
2119	>	* @since 1.8
2120	>	*/
2121	>	public static <K> KeySetView<K,Boolean> newKeySet(int initialCapacity) {
2122	>	return new KeySetView<K,Boolean>
2123	>	(new ConcurrentHashMapV8<K,Boolean>(initialCapacity), Boolean.TRUE);
2124	>	}
2125	>
2126	>	/**
2127	>	* Returns a {@link Set} view of the keys in this map, using the
2128	>	* given common mapped value for any additions (i.e., {@link
2129	>	* Collection#add} and {@link Collection#addAll(Collection)}).
2130	>	* This is of course only appropriate if it is acceptable to use
2131	>	* the same value for all additions from this view.
2132	>	*
2133	>	* @param mappedValue the mapped value to use for any additions
2134	>	* @return the set view
2135	>	* @throws NullPointerException if the mappedValue is null
2136	>	*/
2137	>	public KeySetView<K,V> keySet(V mappedValue) {
2138	>	if (mappedValue == null)
2139	>	throw new NullPointerException();
2140	>	return new KeySetView<K,V>(this, mappedValue);
2141	>	}
2142	>
2143	>	/* ---------------- Special Nodes -------------- */
2144	>
2145	>	/**
2146	>	* A node inserted at head of bins during transfer operations.
2147	>	*/
2148	>	static final class ForwardingNode<K,V> extends Node<K,V> {
2149	>	final Node<K,V>[] nextTable;
2150	>	ForwardingNode(Node<K,V>[] tab) {
2151	>	super(MOVED, null, null, null);
2152	>	this.nextTable = tab;
2153	>	}
2154	>
2155	>	Node<K,V> find(int h, Object k) {
2156	>	// loop to avoid arbitrarily deep recursion on forwarding nodes
2157	>	outer: for (Node<K,V>[] tab = nextTable;;) {
2158	>	Node<K,V> e; int n;
2159	>	if (k == null || tab == null || (n = tab.length) == 0 ||
2160	>	(e = tabAt(tab, (n - 1) & h)) == null)
2161	>	return null;
2162		for (;;) {
2163	<	int cap, n;
2164	<	Node[] tab = table;
2165	<	if (tab == null) {
2166	<	int c = initCap;
2167	<	if (c < sizeHint)
2168	<	c = sizeHint;
2169	<	if (c == DEFAULT_CAPACITY)
2170	<	cap = c;
640	<	else if (c >= MAXIMUM_CAPACITY)
641	<	cap = MAXIMUM_CAPACITY;
642	<	else {
643	<	cap = MINIMUM_CAPACITY;
644	<	while (cap < c)
645	<	cap <<= 1;
2163	>	int eh; K ek;
2164	>	if ((eh = e.hash) == h &&
2165	>	((ek = e.key) == k || (ek != null && k.equals(ek))))
2166	>	return e;
2167	>	if (eh < 0) {
2168	>	if (e instanceof ForwardingNode) {
2169	>	tab = ((ForwardingNode<K,V>)e).nextTable;
2170	>	continue outer;
2171		}
2172	+	else
2173	+	return e.find(h, k);
2174		}
2175	<	else if ((n = tab.length) < MAXIMUM_CAPACITY &&
2176	<	(sizeHint <= 0 || n < sizeHint))
2177	<	cap = n << 1;
2178	<	else
2179	<	break;
2180	<	threshold = (int)(cap * loadFactor) - THRESHOLD_OFFSET;
2181	<	Node[] nextTab = new Node[cap];
2182	<	if (tab != null)
2183	<	transfer(tab, nextTab);
2184	<	table = nextTab;
2185	<	if (tab == null || cap >= MAXIMUM_CAPACITY ||
2186	<	((sizeHint > 0) ? cap >= sizeHint :
2187	<	counter.sum() < threshold))
2175	>	if ((e = e.next) == null)
2176	>	return null;
2177	>	}
2178	>	}
2179	>	}
2180	>	}
2181	>
2182	>	/**
2183	>	* A place-holder node used in computeIfAbsent and compute
2184	>	*/
2185	>	static final class ReservationNode<K,V> extends Node<K,V> {
2186	>	ReservationNode() {
2187	>	super(RESERVED, null, null, null);
2188	>	}
2189	>
2190	>	Node<K,V> find(int h, Object k) {
2191	>	return null;
2192	>	}
2193	>	}
2194	>
2195	>	/* ---------------- Table Initialization and Resizing -------------- */
2196	>
2197	>	/**
2198	>	* Initializes table, using the size recorded in sizeCtl.
2199	>	*/
2200	>	private final Node<K,V>[] initTable() {
2201	>	Node<K,V>[] tab; int sc;
2202	>	while ((tab = table) == null || tab.length == 0) {
2203	>	if ((sc = sizeCtl) < 0)
2204	>	Thread.yield(); // lost initialization race; just spin
2205	>	else if (U.compareAndSwapInt(this, SIZECTL, sc, -1)) {
2206	>	try {
2207	>	if ((tab = table) == null || tab.length == 0) {
2208	>	int n = (sc > 0) ? sc : DEFAULT_CAPACITY;
2209	>	@SuppressWarnings("unchecked")
2210	>	Node<K,V>[] nt = (Node<K,V>[])new Node<?,?>[n];
2211	>	table = tab = nt;
2212	>	sc = n - (n >>> 2);
2213	>	}
2214	>	} finally {
2215	>	sizeCtl = sc;
2216	>	}
2217	>	break;
2218	>	}
2219	>	}
2220	>	return tab;
2221	>	}
2222	>
2223	>	/**
2224	>	* Adds to count, and if table is too small and not already
2225	>	* resizing, initiates transfer. If already resizing, helps
2226	>	* perform transfer if work is available.  Rechecks occupancy
2227	>	* after a transfer to see if another resize is already needed
2228	>	* because resizings are lagging additions.
2229	>	*
2230	>	* @param x the count to add
2231	>	* @param check if <0, don't check resize, if <= 1 only check if uncontended
2232	>	*/
2233	>	private final void addCount(long x, int check) {
2234	>	CounterCell[] as; long b, s;
2235	>	if ((as = counterCells) != null ||
2236	>	!U.compareAndSwapLong(this, BASECOUNT, b = baseCount, s = b + x)) {
2237	>	CounterHashCode hc; CounterCell a; long v; int m;
2238	>	boolean uncontended = true;
2239	>	if ((hc = threadCounterHashCode.get()) == null ||
2240	>	as == null || (m = as.length - 1) < 0 ||
2241	>	(a = as[m & hc.code]) == null ||
2242	>	!(uncontended =
2243	>	U.compareAndSwapLong(a, CELLVALUE, v = a.value, v + x))) {
2244	>	fullAddCount(x, hc, uncontended);
2245	>	return;
2246	>	}
2247	>	if (check <= 1)
2248	>	return;
2249	>	s = sumCount();
2250	>	}
2251	>	if (check >= 0) {
2252	>	Node<K,V>[] tab, nt; int sc;
2253	>	while (s >= (long)(sc = sizeCtl) && (tab = table) != null &&
2254	>	tab.length < MAXIMUM_CAPACITY) {
2255	>	if (sc < 0) {
2256	>	if (sc == -1 || transferIndex <= 0 ||
2257	>	(nt = nextTable) == null)
2258		break;
2259	+	if (U.compareAndSwapInt(this, SIZECTL, sc, sc - 1))
2260	+	transfer(tab, nt);
2261		}
2262	<	} finally {
2263	<	resizing = 0;
2262	>	else if (U.compareAndSwapInt(this, SIZECTL, sc, -2))
2263	>	transfer(tab, null);
2264	>	s = sumCount();
2265		}
2266		}
2267	<	else if (table == null)
2268	<	Thread.yield(); // lost initialization race; just spin
2267	>	}
2268	>
2269	>	/**
2270	>	* Helps transfer if a resize is in progress.
2271	>	*/
2272	>	final Node<K,V>[] helpTransfer(Node<K,V>[] tab, Node<K,V> f) {
2273	>	Node<K,V>[] nextTab; int sc;
2274	>	if ((f instanceof ForwardingNode) &&
2275	>	(nextTab = ((ForwardingNode<K,V>)f).nextTable) != null) {
2276	>	while (transferIndex > 0 && nextTab == nextTable &&
2277	>	(sc = sizeCtl) < -1) {
2278	>	if (U.compareAndSwapInt(this, SIZECTL, sc, sc - 1)) {
2279	>	transfer(tab, nextTab);
2280	>	break;
2281	>	}
2282	>	}
2283	>	return nextTab;
2284	>	}
2285		return table;
2286		}
2287
2288		/**
2289	<	* Implementation for putAll and constructor with Map
2290	<	* argument. Tries to first override initial capacity or grow
2291	<	* based on map size to pre-allocate table space.
2289	>	* Tries to presize table to accommodate the given number of elements.
2290	>	*
2291	>	* @param size number of elements (doesn't need to be perfectly accurate)
2292		*/
2293	<	private final void internalPutAll(Map<? extends K, ? extends V> m) {
2294	<	int s = m.size();
2295	<	grow((s >= (MAXIMUM_CAPACITY >>> 1)) ? s : s + (s >>> 1));
2296	<	for (Map.Entry<? extends K, ? extends V> e : m.entrySet()) {
2297	<	Object k = e.getKey();
2298	<	Object v = e.getValue();
2299	<	if (k == null || v == null)
2300	<	throw new NullPointerException();
2301	<	internalPut(k, v, true);
2293	>	private final void tryPresize(int size) {
2294	>	int c = (size >= (MAXIMUM_CAPACITY >>> 1)) ? MAXIMUM_CAPACITY :
2295	>	tableSizeFor(size + (size >>> 1) + 1);
2296	>	int sc;
2297	>	while ((sc = sizeCtl) >= 0) {
2298	>	Node<K,V>[] tab = table; int n;
2299	>	if (tab == null || (n = tab.length) == 0) {
2300	>	n = (sc > c) ? sc : c;
2301	>	if (U.compareAndSwapInt(this, SIZECTL, sc, -1)) {
2302	>	try {
2303	>	if (table == tab) {
2304	>	@SuppressWarnings("unchecked")
2305	>	Node<K,V>[] nt = (Node<K,V>[])new Node<?,?>[n];
2306	>	table = nt;
2307	>	sc = n - (n >>> 2);
2308	>	}
2309	>	} finally {
2310	>	sizeCtl = sc;
2311	>	}
2312	>	}
2313	>	}
2314	>	else if (c <= sc || n >= MAXIMUM_CAPACITY)
2315	>	break;
2316	>	else if (tab == table &&
2317	>	U.compareAndSwapInt(this, SIZECTL, sc, -2))
2318	>	transfer(tab, null);
2319		}
2320		}
2321
2322		/**
2323	<	* Implementation for clear. Steps through each bin, removing all nodes.
2323	>	* Moves and/or copies the nodes in each bin to new table. See
2324	>	* above for explanation.
2325		*/
2326	<	private final void internalClear() {
2327	<	long deletions = 0L;
2328	<	int i = 0;
2329	<	Node[] tab = table;
2330	<	while (tab != null && i < tab.length) {
2331	<	Node e = tabAt(tab, i);
2332	<	if (e == null)
2333	<	++i;
2334	<	else if (e.hash < 0)
2335	<	tab = (Node[])e.key;
2326	>	private final void transfer(Node<K,V>[] tab, Node<K,V>[] nextTab) {
2327	>	int n = tab.length, stride;
2328	>	if ((stride = (NCPU > 1) ? (n >>> 3) / NCPU : n) < MIN_TRANSFER_STRIDE)
2329	>	stride = MIN_TRANSFER_STRIDE; // subdivide range
2330	>	if (nextTab == null) {            // initiating
2331	>	try {
2332	>	@SuppressWarnings("unchecked")
2333	>	Node<K,V>[] nt = (Node<K,V>[])new Node<?,?>[n << 1];
2334	>	nextTab = nt;
2335	>	} catch (Throwable ex) {      // try to cope with OOME
2336	>	sizeCtl = Integer.MAX_VALUE;
2337	>	return;
2338	>	}
2339	>	nextTable = nextTab;
2340	>	transferIndex = n;
2341	>	}
2342	>	int nextn = nextTab.length;
2343	>	ForwardingNode<K,V> fwd = new ForwardingNode<K,V>(nextTab);
2344	>	boolean advance = true;
2345	>	boolean finishing = false; // to ensure sweep before committing nextTab
2346	>	for (int i = 0, bound = 0;;) {
2347	>	Node<K,V> f; int fh;
2348	>	while (advance) {
2349	>	int nextIndex, nextBound;
2350	>	if (--i >= bound || finishing)
2351	>	advance = false;
2352	>	else if ((nextIndex = transferIndex) <= 0) {
2353	>	i = -1;
2354	>	advance = false;
2355	>	}
2356	>	else if (U.compareAndSwapInt
2357	>	(this, TRANSFERINDEX, nextIndex,
2358	>	nextBound = (nextIndex > stride ?
2359	>	nextIndex - stride : 0))) {
2360	>	bound = nextBound;
2361	>	i = nextIndex - 1;
2362	>	advance = false;
2363	>	}
2364	>	}
2365	>	if (i < 0 || i >= n || i + n >= nextn) {
2366	>	int sc;
2367	>	if (finishing) {
2368	>	nextTable = null;
2369	>	table = nextTab;
2370	>	sizeCtl = (n << 1) - (n >>> 1);
2371	>	return;
2372	>	}
2373	>	if (U.compareAndSwapInt(this, SIZECTL, sc = sizeCtl, ++sc)) {
2374	>	if (sc != -1)
2375	>	return;
2376	>	finishing = advance = true;
2377	>	i = n; // recheck before commit
2378	>	}
2379	>	}
2380	>	else if ((f = tabAt(tab, i)) == null)
2381	>	advance = casTabAt(tab, i, null, fwd);
2382	>	else if ((fh = f.hash) == MOVED)
2383	>	advance = true; // already processed
2384		else {
2385	<	boolean validated = false;
2386	<	synchronized (e) {
2387	<	if (tabAt(tab, i) == e) {
2388	<	validated = true;
2389	<	do {
2390	<	if (e.val != null) {
2391	<	e.val = null;
2392	<	++deletions;
2385	>	synchronized (f) {
2386	>	if (tabAt(tab, i) == f) {
2387	>	Node<K,V> ln, hn;
2388	>	if (fh >= 0) {
2389	>	int runBit = fh & n;
2390	>	Node<K,V> lastRun = f;
2391	>	for (Node<K,V> p = f.next; p != null; p = p.next) {
2392	>	int b = p.hash & n;
2393	>	if (b != runBit) {
2394	>	runBit = b;
2395	>	lastRun = p;
2396	>	}
2397	>	}
2398	>	if (runBit == 0) {
2399	>	ln = lastRun;
2400	>	hn = null;
2401	>	}
2402	>	else {
2403	>	hn = lastRun;
2404	>	ln = null;
2405		}
2406	<	} while ((e = e.next) != null);
2407	<	setTabAt(tab, i, null);
2406	>	for (Node<K,V> p = f; p != lastRun; p = p.next) {
2407	>	int ph = p.hash; K pk = p.key; V pv = p.val;
2408	>	if ((ph & n) == 0)
2409	>	ln = new Node<K,V>(ph, pk, pv, ln);
2410	>	else
2411	>	hn = new Node<K,V>(ph, pk, pv, hn);
2412	>	}
2413	>	setTabAt(nextTab, i, ln);
2414	>	setTabAt(nextTab, i + n, hn);
2415	>	setTabAt(tab, i, fwd);
2416	>	advance = true;
2417	>	}
2418	>	else if (f instanceof TreeBin) {
2419	>	TreeBin<K,V> t = (TreeBin<K,V>)f;
2420	>	TreeNode<K,V> lo = null, loTail = null;
2421	>	TreeNode<K,V> hi = null, hiTail = null;
2422	>	int lc = 0, hc = 0;
2423	>	for (Node<K,V> e = t.first; e != null; e = e.next) {
2424	>	int h = e.hash;
2425	>	TreeNode<K,V> p = new TreeNode<K,V>
2426	>	(h, e.key, e.val, null, null);
2427	>	if ((h & n) == 0) {
2428	>	if ((p.prev = loTail) == null)
2429	>	lo = p;
2430	>	else
2431	>	loTail.next = p;
2432	>	loTail = p;
2433	>	++lc;
2434	>	}
2435	>	else {
2436	>	if ((p.prev = hiTail) == null)
2437	>	hi = p;
2438	>	else
2439	>	hiTail.next = p;
2440	>	hiTail = p;
2441	>	++hc;
2442	>	}
2443	>	}
2444	>	ln = (lc <= UNTREEIFY_THRESHOLD) ? untreeify(lo) :
2445	>	(hc != 0) ? new TreeBin<K,V>(lo) : t;
2446	>	hn = (hc <= UNTREEIFY_THRESHOLD) ? untreeify(hi) :
2447	>	(lc != 0) ? new TreeBin<K,V>(hi) : t;
2448	>	setTabAt(nextTab, i, ln);
2449	>	setTabAt(nextTab, i + n, hn);
2450	>	setTabAt(tab, i, fwd);
2451	>	advance = true;
2452	>	}
2453		}
2454		}
2455	<	if (validated) {
2456	<	++i;
2457	<	if (deletions > THRESHOLD_OFFSET) { // bound lag in counts
2458	<	counter.add(-deletions);
2459	<	deletions = 0L;
2455	>	}
2456	>	}
2457	>	}
2458	>
2459	>	/* ---------------- Conversion from/to TreeBins -------------- */
2460	>
2461	>	/**
2462	>	* Replaces all linked nodes in bin at given index unless table is
2463	>	* too small, in which case resizes instead.
2464	>	*/
2465	>	private final void treeifyBin(Node<K,V>[] tab, int index) {
2466	>	Node<K,V> b; int n, sc;
2467	>	if (tab != null) {
2468	>	if ((n = tab.length) < MIN_TREEIFY_CAPACITY) {
2469	>	if (tab == table && (sc = sizeCtl) >= 0 &&
2470	>	U.compareAndSwapInt(this, SIZECTL, sc, -2))
2471	>	transfer(tab, null);
2472	>	}
2473	>	else if ((b = tabAt(tab, index)) != null && b.hash >= 0) {
2474	>	synchronized (b) {
2475	>	if (tabAt(tab, index) == b) {
2476	>	TreeNode<K,V> hd = null, tl = null;
2477	>	for (Node<K,V> e = b; e != null; e = e.next) {
2478	>	TreeNode<K,V> p =
2479	>	new TreeNode<K,V>(e.hash, e.key, e.val,
2480	>	null, null);
2481	>	if ((p.prev = tl) == null)
2482	>	hd = p;
2483	>	else
2484	>	tl.next = p;
2485	>	tl = p;
2486	>	}
2487	>	setTabAt(tab, index, new TreeBin<K,V>(hd));
2488		}
2489		}
2490		}
2491		}
725	–	if (deletions != 0L)
726	–	counter.add(-deletions);
2492		}
2493
2494		/**
2495	<	* Base class for key, value, and entry iterators, plus internal
731	<	* implementations of public traversal-based methods, to avoid
732	<	* duplicating traversal code.
2495	>	* Returns a list on non-TreeNodes replacing those in given list.
2496		*/
2497	<	class HashIterator {
2498	<	private Node next;          // the next entry to return
2499	<	private Node[] tab;         // current table; updated if resized
2500	<	private Node lastReturned;  // the last entry returned, for remove
2501	<	private Object nextVal;     // cached value of next
2502	<	private int index;          // index of bin to use next
2503	<	private int baseIndex;      // current index of initial table
2504	<	private final int baseSize; // initial table size
2497	>	static <K,V> Node<K,V> untreeify(Node<K,V> b) {
2498	>	Node<K,V> hd = null, tl = null;
2499	>	for (Node<K,V> q = b; q != null; q = q.next) {
2500	>	Node<K,V> p = new Node<K,V>(q.hash, q.key, q.val, null);
2501	>	if (tl == null)
2502	>	hd = p;
2503	>	else
2504	>	tl.next = p;
2505	>	tl = p;
2506	>	}
2507	>	return hd;
2508	>	}
2509
2510	<	HashIterator() {
2511	<	Node[] t = tab = table;
2512	<	if (t == null)
2513	<	baseSize = 0;
2514	<	else {
2515	<	baseSize = t.length;
2516	<	advance(null);
2517	<	}
2510	>	/* ---------------- TreeNodes -------------- */
2511	>
2512	>	/**
2513	>	* Nodes for use in TreeBins
2514	>	*/
2515	>	static final class TreeNode<K,V> extends Node<K,V> {
2516	>	TreeNode<K,V> parent;  // red-black tree links
2517	>	TreeNode<K,V> left;
2518	>	TreeNode<K,V> right;
2519	>	TreeNode<K,V> prev;    // needed to unlink next upon deletion
2520	>	boolean red;
2521	>
2522	>	TreeNode(int hash, K key, V val, Node<K,V> next,
2523	>	TreeNode<K,V> parent) {
2524	>	super(hash, key, val, next);
2525	>	this.parent = parent;
2526		}
2527
2528	<	public final boolean hasNext()         { return next != null; }
2529	<	public final boolean hasMoreElements() { return next != null; }
2528	>	Node<K,V> find(int h, Object k) {
2529	>	return findTreeNode(h, k, null);
2530	>	}
2531
2532		/**
2533	<	* Advances next.  Normally, iteration proceeds bin-by-bin
2534	<	* traversing lists.  However, if the table has been resized,
759	<	* then all future steps must traverse both the bin at the
760	<	* current index as well as at (index + baseSize); and so on
761	<	* for further resizings. To paranoically cope with potential
762	<	* (improper) sharing of iterators across threads, table reads
763	<	* are bounds-checked.
2533	>	* Returns the TreeNode (or null if not found) for the given key
2534	>	* starting at given root.
2535		*/
2536	<	final void advance(Node e) {
2537	<	for (;;) {
2538	<	Node[] t; int i; // for bounds checks
2539	<	if (e != null) {
2540	<	Object ek = e.key, ev = e.val;
2541	<	if (ev != null && ek != null) {
2542	<	nextVal = ev;
2543	<	next = e;
2544	<	break;
2545	<	}
2546	<	e = e.next;
2547	<	}
2548	<	else if (baseIndex < baseSize && (t = tab) != null &&
2549	<	t.length > (i = index) && i >= 0) {
2550	<	if ((e = tabAt(t, i)) != null && e.hash < 0) {
2551	<	tab = (Node[])e.key;
2552	<	e = null;
2553	<	}
2554	<	else if (i + baseSize < t.length)
2555	<	index += baseSize;    // visit forwarded upper slots
2536	>	final TreeNode<K,V> findTreeNode(int h, Object k, Class<?> kc) {
2537	>	if (k != null) {
2538	>	TreeNode<K,V> p = this;
2539	>	do  {
2540	>	int ph, dir; K pk; TreeNode<K,V> q;
2541	>	TreeNode<K,V> pl = p.left, pr = p.right;
2542	>	if ((ph = p.hash) > h)
2543	>	p = pl;
2544	>	else if (ph < h)
2545	>	p = pr;
2546	>	else if ((pk = p.key) == k || (pk != null && k.equals(pk)))
2547	>	return p;
2548	>	else if (pl == null)
2549	>	p = pr;
2550	>	else if (pr == null)
2551	>	p = pl;
2552	>	else if ((kc != null ||
2553	>	(kc = comparableClassFor(k)) != null) &&
2554	>	(dir = compareComparables(kc, k, pk)) != 0)
2555	>	p = (dir < 0) ? pl : pr;
2556	>	else if ((q = pr.findTreeNode(h, k, kc)) != null)
2557	>	return q;
2558		else
2559	<	index = ++baseIndex;
2559	>	p = pl;
2560	>	} while (p != null);
2561	>	}
2562	>	return null;
2563	>	}
2564	>	}
2565	>
2566	>	/* ---------------- TreeBins -------------- */
2567	>
2568	>	/**
2569	>	* TreeNodes used at the heads of bins. TreeBins do not hold user
2570	>	* keys or values, but instead point to list of TreeNodes and
2571	>	* their root. They also maintain a parasitic read-write lock
2572	>	* forcing writers (who hold bin lock) to wait for readers (who do
2573	>	* not) to complete before tree restructuring operations.
2574	>	*/
2575	>	static final class TreeBin<K,V> extends Node<K,V> {
2576	>	TreeNode<K,V> root;
2577	>	volatile TreeNode<K,V> first;
2578	>	volatile Thread waiter;
2579	>	volatile int lockState;
2580	>	// values for lockState
2581	>	static final int WRITER = 1; // set while holding write lock
2582	>	static final int WAITER = 2; // set when waiting for write lock
2583	>	static final int READER = 4; // increment value for setting read lock
2584	>
2585	>	/**
2586	>	* Tie-breaking utility for ordering insertions when equal
2587	>	* hashCodes and non-comparable. We don't require a total
2588	>	* order, just a consistent insertion rule to maintain
2589	>	* equivalence across rebalancings. Tie-breaking further than
2590	>	* necessary simplifies testing a bit.
2591	>	*/
2592	>	static int tieBreakOrder(Object a, Object b) {
2593	>	int d;
2594	>	if (a == null || b == null ||
2595	>	(d = a.getClass().getName().
2596	>	compareTo(b.getClass().getName())) == 0)
2597	>	d = (System.identityHashCode(a) <= System.identityHashCode(b) ?
2598	>	-1 : 1);
2599	>	return d;
2600	>	}
2601	>
2602	>	/**
2603	>	* Creates bin with initial set of nodes headed by b.
2604	>	*/
2605	>	TreeBin(TreeNode<K,V> b) {
2606	>	super(TREEBIN, null, null, null);
2607	>	this.first = b;
2608	>	TreeNode<K,V> r = null;
2609	>	for (TreeNode<K,V> x = b, next; x != null; x = next) {
2610	>	next = (TreeNode<K,V>)x.next;
2611	>	x.left = x.right = null;
2612	>	if (r == null) {
2613	>	x.parent = null;
2614	>	x.red = false;
2615	>	r = x;
2616		}
2617		else {
2618	<	next = null;
2619	<	break;
2618	>	K k = x.key;
2619	>	int h = x.hash;
2620	>	Class<?> kc = null;
2621	>	for (TreeNode<K,V> p = r;;) {
2622	>	int dir, ph;
2623	>	K pk = p.key;
2624	>	if ((ph = p.hash) > h)
2625	>	dir = -1;
2626	>	else if (ph < h)
2627	>	dir = 1;
2628	>	else if ((kc == null &&
2629	>	(kc = comparableClassFor(k)) == null) ||
2630	>	(dir = compareComparables(kc, k, pk)) == 0)
2631	>	dir = tieBreakOrder(k, pk);
2632	>	TreeNode<K,V> xp = p;
2633	>	if ((p = (dir <= 0) ? p.left : p.right) == null) {
2634	>	x.parent = xp;
2635	>	if (dir <= 0)
2636	>	xp.left = x;
2637	>	else
2638	>	xp.right = x;
2639	>	r = balanceInsertion(r, x);
2640	>	break;
2641	>	}
2642	>	}
2643		}
2644		}
2645	+	this.root = r;
2646	+	assert checkInvariants(root);
2647		}
2648
2649	<	final Object nextKey() {
2650	<	Node e = next;
2651	<	if (e == null)
2652	<	throw new NoSuchElementException();
2653	<	Object k = e.key;
2654	<	advance((lastReturned = e).next);
801	<	return k;
2649	>	/**
2650	>	* Acquires write lock for tree restructuring.
2651	>	*/
2652	>	private final void lockRoot() {
2653	>	if (!U.compareAndSwapInt(this, LOCKSTATE, 0, WRITER))
2654	>	contendedLock(); // offload to separate method
2655		}
2656
2657	<	final Object nextValue() {
2658	<	Node e = next;
2659	<	if (e == null)
2660	<	throw new NoSuchElementException();
2661	<	Object v = nextVal;
809	<	advance((lastReturned = e).next);
810	<	return v;
2657	>	/**
2658	>	* Releases write lock for tree restructuring.
2659	>	*/
2660	>	private final void unlockRoot() {
2661	>	lockState = 0;
2662		}
2663
2664	<	final WriteThroughEntry nextEntry() {
2665	<	Node e = next;
2666	<	if (e == null)
2667	<	throw new NoSuchElementException();
2668	<	WriteThroughEntry entry =
2669	<	new WriteThroughEntry(e.key, nextVal);
2670	<	advance((lastReturned = e).next);
2671	<	return entry;
2664	>	/**
2665	>	* Possibly blocks awaiting root lock.
2666	>	*/
2667	>	private final void contendedLock() {
2668	>	boolean waiting = false;
2669	>	for (int s;;) {
2670	>	if (((s = lockState) & WRITER) == 0) {
2671	>	if (U.compareAndSwapInt(this, LOCKSTATE, s, WRITER)) {
2672	>	if (waiting)
2673	>	waiter = null;
2674	>	return;
2675	>	}
2676	>	}
2677	>	else if ((s & WAITER) == 0) {
2678	>	if (U.compareAndSwapInt(this, LOCKSTATE, s, s | WAITER)) {
2679	>	waiting = true;
2680	>	waiter = Thread.currentThread();
2681	>	}
2682	>	}
2683	>	else if (waiting)
2684	>	LockSupport.park(this);
2685	>	}
2686		}
2687
2688	<	public final void remove() {
2689	<	if (lastReturned == null)
2690	<	throw new IllegalStateException();
2691	<	ConcurrentHashMapV8.this.remove(lastReturned.key);
2692	<	lastReturned = null;
2688	>	/**
2689	>	* Returns matching node or null if none. Tries to search
2690	>	* using tree comparisons from root, but continues linear
2691	>	* search when lock not available.
2692	>	*/
2693	>	final Node<K,V> find(int h, Object k) {
2694	>	if (k != null) {
2695	>	for (Node<K,V> e = first; e != null; e = e.next) {
2696	>	int s; K ek;
2697	>	if (((s = lockState) & (WAITER|WRITER)) != 0) {
2698	>	if (e.hash == h &&
2699	>	((ek = e.key) == k || (ek != null && k.equals(ek))))
2700	>	return e;
2701	>	}
2702	>	else if (U.compareAndSwapInt(this, LOCKSTATE, s,
2703	>	s + READER)) {
2704	>	TreeNode<K,V> r, p;
2705	>	try {
2706	>	p = ((r = root) == null ? null :
2707	>	r.findTreeNode(h, k, null));
2708	>	} finally {
2709	>	Thread w;
2710	>	int ls;
2711	>	do {} while (!U.compareAndSwapInt
2712	>	(this, LOCKSTATE,
2713	>	ls = lockState, ls - READER));
2714	>	if (ls == (READER|WAITER) && (w = waiter) != null)
2715	>	LockSupport.unpark(w);
2716	>	}
2717	>	return p;
2718	>	}
2719	>	}
2720	>	}
2721	>	return null;
2722		}
2723
2724	<	/** Helper for serialization */
2725	<	final void writeEntries(java.io.ObjectOutputStream s)
2726	<	throws java.io.IOException {
2727	<	Node e;
2728	<	while ((e = next) != null) {
2729	<	s.writeObject(e.key);
2730	<	s.writeObject(nextVal);
2731	<	advance(e.next);
2724	>	/**
2725	>	* Finds or adds a node.
2726	>	* @return null if added
2727	>	*/
2728	>	final TreeNode<K,V> putTreeVal(int h, K k, V v) {
2729	>	Class<?> kc = null;
2730	>	boolean searched = false;
2731	>	for (TreeNode<K,V> p = root;;) {
2732	>	int dir, ph; K pk;
2733	>	if (p == null) {
2734	>	first = root = new TreeNode<K,V>(h, k, v, null, null);
2735	>	break;
2736	>	}
2737	>	else if ((ph = p.hash) > h)
2738	>	dir = -1;
2739	>	else if (ph < h)
2740	>	dir = 1;
2741	>	else if ((pk = p.key) == k || (pk != null && k.equals(pk)))
2742	>	return p;
2743	>	else if ((kc == null &&
2744	>	(kc = comparableClassFor(k)) == null) ||
2745	>	(dir = compareComparables(kc, k, pk)) == 0) {
2746	>	if (!searched) {
2747	>	TreeNode<K,V> q, ch;
2748	>	searched = true;
2749	>	if (((ch = p.left) != null &&
2750	>	(q = ch.findTreeNode(h, k, kc)) != null) ||
2751	>	((ch = p.right) != null &&
2752	>	(q = ch.findTreeNode(h, k, kc)) != null))
2753	>	return q;
2754	>	}
2755	>	dir = tieBreakOrder(k, pk);
2756	>	}
2757	>
2758	>	TreeNode<K,V> xp = p;
2759	>	if ((p = (dir <= 0) ? p.left : p.right) == null) {
2760	>	TreeNode<K,V> x, f = first;
2761	>	first = x = new TreeNode<K,V>(h, k, v, f, xp);
2762	>	if (f != null)
2763	>	f.prev = x;
2764	>	if (dir <= 0)
2765	>	xp.left = x;
2766	>	else
2767	>	xp.right = x;
2768	>	if (!xp.red)
2769	>	x.red = true;
2770	>	else {
2771	>	lockRoot();
2772	>	try {
2773	>	root = balanceInsertion(root, x);
2774	>	} finally {
2775	>	unlockRoot();
2776	>	}
2777	>	}
2778	>	break;
2779	>	}
2780		}
2781	+	assert checkInvariants(root);
2782	+	return null;
2783		}
2784
2785	<	/** Helper for containsValue */
2786	<	final boolean containsVal(Object value) {
2787	<	if (value != null) {
2788	<	Node e;
2789	<	while ((e = next) != null) {
2790	<	Object v = nextVal;
2791	<	if (value == v || value.equals(v))
2792	<	return true;
2793	<	advance(e.next);
2785	>	/**
2786	>	* Removes the given node, that must be present before this
2787	>	* call.  This is messier than typical red-black deletion code
2788	>	* because we cannot swap the contents of an interior node
2789	>	* with a leaf successor that is pinned by "next" pointers
2790	>	* that are accessible independently of lock. So instead we
2791	>	* swap the tree linkages.
2792	>	*
2793	>	* @return true if now too small, so should be untreeified
2794	>	*/
2795	>	final boolean removeTreeNode(TreeNode<K,V> p) {
2796	>	TreeNode<K,V> next = (TreeNode<K,V>)p.next;
2797	>	TreeNode<K,V> pred = p.prev;  // unlink traversal pointers
2798	>	TreeNode<K,V> r, rl;
2799	>	if (pred == null)
2800	>	first = next;
2801	>	else
2802	>	pred.next = next;
2803	>	if (next != null)
2804	>	next.prev = pred;
2805	>	if (first == null) {
2806	>	root = null;
2807	>	return true;
2808	>	}
2809	>	if ((r = root) == null || r.right == null || // too small
2810	>	(rl = r.left) == null || rl.left == null)
2811	>	return true;
2812	>	lockRoot();
2813	>	try {
2814	>	TreeNode<K,V> replacement;
2815	>	TreeNode<K,V> pl = p.left;
2816	>	TreeNode<K,V> pr = p.right;
2817	>	if (pl != null && pr != null) {
2818	>	TreeNode<K,V> s = pr, sl;
2819	>	while ((sl = s.left) != null) // find successor
2820	>	s = sl;
2821	>	boolean c = s.red; s.red = p.red; p.red = c; // swap colors
2822	>	TreeNode<K,V> sr = s.right;
2823	>	TreeNode<K,V> pp = p.parent;
2824	>	if (s == pr) { // p was s's direct parent
2825	>	p.parent = s;
2826	>	s.right = p;
2827	>	}
2828	>	else {
2829	>	TreeNode<K,V> sp = s.parent;
2830	>	if ((p.parent = sp) != null) {
2831	>	if (s == sp.left)
2832	>	sp.left = p;
2833	>	else
2834	>	sp.right = p;
2835	>	}
2836	>	if ((s.right = pr) != null)
2837	>	pr.parent = s;
2838	>	}
2839	>	p.left = null;
2840	>	if ((p.right = sr) != null)
2841	>	sr.parent = p;
2842	>	if ((s.left = pl) != null)
2843	>	pl.parent = s;
2844	>	if ((s.parent = pp) == null)
2845	>	r = s;
2846	>	else if (p == pp.left)
2847	>	pp.left = s;
2848	>	else
2849	>	pp.right = s;
2850	>	if (sr != null)
2851	>	replacement = sr;
2852	>	else
2853	>	replacement = p;
2854	>	}
2855	>	else if (pl != null)
2856	>	replacement = pl;
2857	>	else if (pr != null)
2858	>	replacement = pr;
2859	>	else
2860	>	replacement = p;
2861	>	if (replacement != p) {
2862	>	TreeNode<K,V> pp = replacement.parent = p.parent;
2863	>	if (pp == null)
2864	>	r = replacement;
2865	>	else if (p == pp.left)
2866	>	pp.left = replacement;
2867	>	else
2868	>	pp.right = replacement;
2869	>	p.left = p.right = p.parent = null;
2870		}
2871	+
2872	+	root = (p.red) ? r : balanceDeletion(r, replacement);
2873	+
2874	+	if (p == replacement) {  // detach pointers
2875	+	TreeNode<K,V> pp;
2876	+	if ((pp = p.parent) != null) {
2877	+	if (p == pp.left)
2878	+	pp.left = null;
2879	+	else if (p == pp.right)
2880	+	pp.right = null;
2881	+	p.parent = null;
2882	+	}
2883	+	}
2884	+	} finally {
2885	+	unlockRoot();
2886		}
2887	+	assert checkInvariants(root);
2888		return false;
2889		}
2890
2891	<	/** Helper for Map.hashCode */
2892	<	final int mapHashCode() {
2893	<	int h = 0;
2894	<	Node e;
2895	<	while ((e = next) != null) {
2896	<	h += e.key.hashCode() ^ nextVal.hashCode();
2897	<	advance(e.next);
2891	>	/* ------------------------------------------------------------ */
2892	>	// Red-black tree methods, all adapted from CLR
2893	>
2894	>	static <K,V> TreeNode<K,V> rotateLeft(TreeNode<K,V> root,
2895	>	TreeNode<K,V> p) {
2896	>	TreeNode<K,V> r, pp, rl;
2897	>	if (p != null && (r = p.right) != null) {
2898	>	if ((rl = p.right = r.left) != null)
2899	>	rl.parent = p;
2900	>	if ((pp = r.parent = p.parent) == null)
2901	>	(root = r).red = false;
2902	>	else if (pp.left == p)
2903	>	pp.left = r;
2904	>	else
2905	>	pp.right = r;
2906	>	r.left = p;
2907	>	p.parent = r;
2908		}
2909	<	return h;
2909	>	return root;
2910		}
2911
2912	<	/** Helper for Map.toString */
2913	<	final String mapToString() {
2914	<	Node e = next;
2915	<	if (e == null)
2916	<	return "{}";
2917	<	StringBuilder sb = new StringBuilder();
2918	<	sb.append('{');
2919	<	for (;;) {
2920	<	sb.append(e.key   == this ? "(this Map)" : e.key);
2921	<	sb.append('=');
876	<	sb.append(nextVal == this ? "(this Map)" : nextVal);
877	<	advance(e.next);
878	<	if ((e = next) != null)
879	<	sb.append(',').append(' ');
2912	>	static <K,V> TreeNode<K,V> rotateRight(TreeNode<K,V> root,
2913	>	TreeNode<K,V> p) {
2914	>	TreeNode<K,V> l, pp, lr;
2915	>	if (p != null && (l = p.left) != null) {
2916	>	if ((lr = p.left = l.right) != null)
2917	>	lr.parent = p;
2918	>	if ((pp = l.parent = p.parent) == null)
2919	>	(root = l).red = false;
2920	>	else if (pp.right == p)
2921	>	pp.right = l;
2922		else
2923	<	return sb.append('}').toString();
2923	>	pp.left = l;
2924	>	l.right = p;
2925	>	p.parent = l;
2926	>	}
2927	>	return root;
2928	>	}
2929	>
2930	>	static <K,V> TreeNode<K,V> balanceInsertion(TreeNode<K,V> root,
2931	>	TreeNode<K,V> x) {
2932	>	x.red = true;
2933	>	for (TreeNode<K,V> xp, xpp, xppl, xppr;;) {
2934	>	if ((xp = x.parent) == null) {
2935	>	x.red = false;
2936	>	return x;
2937	>	}
2938	>	else if (!xp.red || (xpp = xp.parent) == null)
2939	>	return root;
2940	>	if (xp == (xppl = xpp.left)) {
2941	>	if ((xppr = xpp.right) != null && xppr.red) {
2942	>	xppr.red = false;
2943	>	xp.red = false;
2944	>	xpp.red = true;
2945	>	x = xpp;
2946	>	}
2947	>	else {
2948	>	if (x == xp.right) {
2949	>	root = rotateLeft(root, x = xp);
2950	>	xpp = (xp = x.parent) == null ? null : xp.parent;
2951	>	}
2952	>	if (xp != null) {
2953	>	xp.red = false;
2954	>	if (xpp != null) {
2955	>	xpp.red = true;
2956	>	root = rotateRight(root, xpp);
2957	>	}
2958	>	}
2959	>	}
2960	>	}
2961	>	else {
2962	>	if (xppl != null && xppl.red) {
2963	>	xppl.red = false;
2964	>	xp.red = false;
2965	>	xpp.red = true;
2966	>	x = xpp;
2967	>	}
2968	>	else {
2969	>	if (x == xp.left) {
2970	>	root = rotateRight(root, x = xp);
2971	>	xpp = (xp = x.parent) == null ? null : xp.parent;
2972	>	}
2973	>	if (xp != null) {
2974	>	xp.red = false;
2975	>	if (xpp != null) {
2976	>	xpp.red = true;
2977	>	root = rotateLeft(root, xpp);
2978	>	}
2979	>	}
2980	>	}
2981	>	}
2982	>	}
2983	>	}
2984	>
2985	>	static <K,V> TreeNode<K,V> balanceDeletion(TreeNode<K,V> root,
2986	>	TreeNode<K,V> x) {
2987	>	for (TreeNode<K,V> xp, xpl, xpr;;)  {
2988	>	if (x == null || x == root)
2989	>	return root;
2990	>	else if ((xp = x.parent) == null) {
2991	>	x.red = false;
2992	>	return x;
2993	>	}
2994	>	else if (x.red) {
2995	>	x.red = false;
2996	>	return root;
2997	>	}
2998	>	else if ((xpl = xp.left) == x) {
2999	>	if ((xpr = xp.right) != null && xpr.red) {
3000	>	xpr.red = false;
3001	>	xp.red = true;
3002	>	root = rotateLeft(root, xp);
3003	>	xpr = (xp = x.parent) == null ? null : xp.right;
3004	>	}
3005	>	if (xpr == null)
3006	>	x = xp;
3007	>	else {
3008	>	TreeNode<K,V> sl = xpr.left, sr = xpr.right;
3009	>	if ((sr == null || !sr.red) &&
3010	>	(sl == null || !sl.red)) {
3011	>	xpr.red = true;
3012	>	x = xp;
3013	>	}
3014	>	else {
3015	>	if (sr == null || !sr.red) {
3016	>	if (sl != null)
3017	>	sl.red = false;
3018	>	xpr.red = true;
3019	>	root = rotateRight(root, xpr);
3020	>	xpr = (xp = x.parent) == null ?
3021	>	null : xp.right;
3022	>	}
3023	>	if (xpr != null) {
3024	>	xpr.red = (xp == null) ? false : xp.red;
3025	>	if ((sr = xpr.right) != null)
3026	>	sr.red = false;
3027	>	}
3028	>	if (xp != null) {
3029	>	xp.red = false;
3030	>	root = rotateLeft(root, xp);
3031	>	}
3032	>	x = root;
3033	>	}
3034	>	}
3035	>	}
3036	>	else { // symmetric
3037	>	if (xpl != null && xpl.red) {
3038	>	xpl.red = false;
3039	>	xp.red = true;
3040	>	root = rotateRight(root, xp);
3041	>	xpl = (xp = x.parent) == null ? null : xp.left;
3042	>	}
3043	>	if (xpl == null)
3044	>	x = xp;
3045	>	else {
3046	>	TreeNode<K,V> sl = xpl.left, sr = xpl.right;
3047	>	if ((sl == null || !sl.red) &&
3048	>	(sr == null || !sr.red)) {
3049	>	xpl.red = true;
3050	>	x = xp;
3051	>	}
3052	>	else {
3053	>	if (sl == null || !sl.red) {
3054	>	if (sr != null)
3055	>	sr.red = false;
3056	>	xpl.red = true;
3057	>	root = rotateLeft(root, xpl);
3058	>	xpl = (xp = x.parent) == null ?
3059	>	null : xp.left;
3060	>	}
3061	>	if (xpl != null) {
3062	>	xpl.red = (xp == null) ? false : xp.red;
3063	>	if ((sl = xpl.left) != null)
3064	>	sl.red = false;
3065	>	}
3066	>	if (xp != null) {
3067	>	xp.red = false;
3068	>	root = rotateRight(root, xp);
3069	>	}
3070	>	x = root;
3071	>	}
3072	>	}
3073	>	}
3074	>	}
3075	>	}
3076	>
3077	>	/**
3078	>	* Recursive invariant check
3079	>	*/
3080	>	static <K,V> boolean checkInvariants(TreeNode<K,V> t) {
3081	>	TreeNode<K,V> tp = t.parent, tl = t.left, tr = t.right,
3082	>	tb = t.prev, tn = (TreeNode<K,V>)t.next;
3083	>	if (tb != null && tb.next != t)
3084	>	return false;
3085	>	if (tn != null && tn.prev != t)
3086	>	return false;
3087	>	if (tp != null && t != tp.left && t != tp.right)
3088	>	return false;
3089	>	if (tl != null && (tl.parent != t || tl.hash > t.hash))
3090	>	return false;
3091	>	if (tr != null && (tr.parent != t || tr.hash < t.hash))
3092	>	return false;
3093	>	if (t.red && tl != null && tl.red && tr != null && tr.red)
3094	>	return false;
3095	>	if (tl != null && !checkInvariants(tl))
3096	>	return false;
3097	>	if (tr != null && !checkInvariants(tr))
3098	>	return false;
3099	>	return true;
3100	>	}
3101	>
3102	>	private static final sun.misc.Unsafe U;
3103	>	private static final long LOCKSTATE;
3104	>	static {
3105	>	try {
3106	>	U = getUnsafe();
3107	>	Class<?> k = TreeBin.class;
3108	>	LOCKSTATE = U.objectFieldOffset
3109	>	(k.getDeclaredField("lockState"));
3110	>	} catch (Exception e) {
3111	>	throw new Error(e);
3112		}
3113		}
3114		}
3115
3116	<	/* ---------------- Public operations -------------- */
3116	>	/* ----------------Table Traversal -------------- */
3117
3118		/**
3119	<	* Creates a new, empty map with the specified initial
3120	<	* capacity, load factor and concurrency level.
3121	<	*
3122	<	* @param initialCapacity the initial capacity. The implementation
3123	<	* performs internal sizing to accommodate this many elements.
3124	<	* @param loadFactor  the load factor threshold, used to control resizing.
3125	<	* Resizing may be performed when the average number of elements per
3126	<	* bin exceeds this threshold.
3127	<	* @param concurrencyLevel the estimated number of concurrently
898	<	* updating threads. The implementation may use this value as
899	<	* a sizing hint.
900	<	* @throws IllegalArgumentException if the initial capacity is
901	<	* negative or the load factor or concurrencyLevel are
902	<	* nonpositive.
903	<	*/
904	<	public ConcurrentHashMapV8(int initialCapacity,
905	<	float loadFactor, int concurrencyLevel) {
906	<	if (!(loadFactor > 0) || initialCapacity < 0 || concurrencyLevel <= 0)
907	<	throw new IllegalArgumentException();
908	<	this.initCap = initialCapacity;
909	<	this.loadFactor = loadFactor;
910	<	this.counter = new LongAdder();
3119	>	* Records the table, its length, and current traversal index for a
3120	>	* traverser that must process a region of a forwarded table before
3121	>	* proceeding with current table.
3122	>	*/
3123	>	static final class TableStack<K,V> {
3124	>	int length;
3125	>	int index;
3126	>	Node<K,V>[] tab;
3127	>	TableStack<K,V> next;
3128		}
3129
3130		/**
3131	<	* Creates a new, empty map with the specified initial capacity
3132	<	* and load factor and with the default concurrencyLevel (16).
3133	<	*
3134	<	* @param initialCapacity The implementation performs internal
3135	<	* sizing to accommodate this many elements.
3136	<	* @param loadFactor  the load factor threshold, used to control resizing.
3137	<	* Resizing may be performed when the average number of elements per
3138	<	* bin exceeds this threshold.
3139	<	* @throws IllegalArgumentException if the initial capacity of
3140	<	* elements is negative or the load factor is nonpositive
3141	<	*
3142	<	* @since 1.6
3143	<	*/
3144	<	public ConcurrentHashMapV8(int initialCapacity, float loadFactor) {
3145	<	this(initialCapacity, loadFactor, DEFAULT_CONCURRENCY_LEVEL);
3131	>	* Encapsulates traversal for methods such as containsValue; also
3132	>	* serves as a base class for other iterators and spliterators.
3133	>	*
3134	>	* Method advance visits once each still-valid node that was
3135	>	* reachable upon iterator construction. It might miss some that
3136	>	* were added to a bin after the bin was visited, which is OK wrt
3137	>	* consistency guarantees. Maintaining this property in the face
3138	>	* of possible ongoing resizes requires a fair amount of
3139	>	* bookkeeping state that is difficult to optimize away amidst
3140	>	* volatile accesses.  Even so, traversal maintains reasonable
3141	>	* throughput.
3142	>	*
3143	>	* Normally, iteration proceeds bin-by-bin traversing lists.
3144	>	* However, if the table has been resized, then all future steps
3145	>	* must traverse both the bin at the current index as well as at
3146	>	* (index + baseSize); and so on for further resizings. To
3147	>	* paranoically cope with potential sharing by users of iterators
3148	>	* across threads, iteration terminates if a bounds checks fails
3149	>	* for a table read.
3150	>	*/
3151	>	static class Traverser<K,V> {
3152	>	Node<K,V>[] tab;        // current table; updated if resized
3153	>	Node<K,V> next;         // the next entry to use
3154	>	TableStack<K,V> stack, spare; // to save/restore on ForwardingNodes
3155	>	int index;              // index of bin to use next
3156	>	int baseIndex;          // current index of initial table
3157	>	int baseLimit;          // index bound for initial table
3158	>	final int baseSize;     // initial table size
3159	>
3160	>	Traverser(Node<K,V>[] tab, int size, int index, int limit) {
3161	>	this.tab = tab;
3162	>	this.baseSize = size;
3163	>	this.baseIndex = this.index = index;
3164	>	this.baseLimit = limit;
3165	>	this.next = null;
3166	>	}
3167	>
3168	>	/**
3169	>	* Advances if possible, returning next valid node, or null if none.
3170	>	*/
3171	>	final Node<K,V> advance() {
3172	>	Node<K,V> e;
3173	>	if ((e = next) != null)
3174	>	e = e.next;
3175	>	for (;;) {
3176	>	Node<K,V>[] t; int i, n;  // must use locals in checks
3177	>	if (e != null)
3178	>	return next = e;
3179	>	if (baseIndex >= baseLimit || (t = tab) == null ||
3180	>	(n = t.length) <= (i = index) || i < 0)
3181	>	return next = null;
3182	>	if ((e = tabAt(t, i)) != null && e.hash < 0) {
3183	>	if (e instanceof ForwardingNode) {
3184	>	tab = ((ForwardingNode<K,V>)e).nextTable;
3185	>	e = null;
3186	>	pushState(t, i, n);
3187	>	continue;
3188	>	}
3189	>	else if (e instanceof TreeBin)
3190	>	e = ((TreeBin<K,V>)e).first;
3191	>	else
3192	>	e = null;
3193	>	}
3194	>	if (stack != null)
3195	>	recoverState(n);
3196	>	else if ((index = i + baseSize) >= n)
3197	>	index = ++baseIndex; // visit upper slots if present
3198	>	}
3199	>	}
3200	>
3201	>	/**
3202	>	* Saves traversal state upon encountering a forwarding node.
3203	>	*/
3204	>	private void pushState(Node<K,V>[] t, int i, int n) {
3205	>	TableStack<K,V> s = spare;  // reuse if possible
3206	>	if (s != null)
3207	>	spare = s.next;
3208	>	else
3209	>	s = new TableStack<K,V>();
3210	>	s.tab = t;
3211	>	s.length = n;
3212	>	s.index = i;
3213	>	s.next = stack;
3214	>	stack = s;
3215	>	}
3216	>
3217	>	/**
3218	>	* Possibly pops traversal state.
3219	>	*
3220	>	* @param n length of current table
3221	>	*/
3222	>	private void recoverState(int n) {
3223	>	TableStack<K,V> s; int len;
3224	>	while ((s = stack) != null && (index += (len = s.length)) >= n) {
3225	>	n = len;
3226	>	index = s.index;
3227	>	tab = s.tab;
3228	>	s.tab = null;
3229	>	TableStack<K,V> next = s.next;
3230	>	s.next = spare; // save for reuse
3231	>	stack = next;
3232	>	spare = s;
3233	>	}
3234	>	if (s == null && (index += baseSize) >= n)
3235	>	index = ++baseIndex;
3236	>	}
3237		}
3238
3239		/**
3240	<	* Creates a new, empty map with the specified initial capacity,
3241	<	* and with default load factor (0.75) and concurrencyLevel (16).
934	<	*
935	<	* @param initialCapacity the initial capacity. The implementation
936	<	* performs internal sizing to accommodate this many elements.
937	<	* @throws IllegalArgumentException if the initial capacity of
938	<	* elements is negative.
3240	>	* Base of key, value, and entry Iterators. Adds fields to
3241	>	* Traverser to support iterator.remove.
3242		*/
3243	<	public ConcurrentHashMapV8(int initialCapacity) {
3244	<	this(initialCapacity, DEFAULT_LOAD_FACTOR, DEFAULT_CONCURRENCY_LEVEL);
3243	>	static class BaseIterator<K,V> extends Traverser<K,V> {
3244	>	final ConcurrentHashMapV8<K,V> map;
3245	>	Node<K,V> lastReturned;
3246	>	BaseIterator(Node<K,V>[] tab, int size, int index, int limit,
3247	>	ConcurrentHashMapV8<K,V> map) {
3248	>	super(tab, size, index, limit);
3249	>	this.map = map;
3250	>	advance();
3251	>	}
3252	>
3253	>	public final boolean hasNext() { return next != null; }
3254	>	public final boolean hasMoreElements() { return next != null; }
3255	>
3256	>	public final void remove() {
3257	>	Node<K,V> p;
3258	>	if ((p = lastReturned) == null)
3259	>	throw new IllegalStateException();
3260	>	lastReturned = null;
3261	>	map.replaceNode(p.key, null, null);
3262	>	}
3263	>	}
3264	>
3265	>	static final class KeyIterator<K,V> extends BaseIterator<K,V>
3266	>	implements Iterator<K>, Enumeration<K> {
3267	>	KeyIterator(Node<K,V>[] tab, int index, int size, int limit,
3268	>	ConcurrentHashMapV8<K,V> map) {
3269	>	super(tab, index, size, limit, map);
3270	>	}
3271	>
3272	>	public final K next() {
3273	>	Node<K,V> p;
3274	>	if ((p = next) == null)
3275	>	throw new NoSuchElementException();
3276	>	K k = p.key;
3277	>	lastReturned = p;
3278	>	advance();
3279	>	return k;
3280	>	}
3281	>
3282	>	public final K nextElement() { return next(); }
3283	>	}
3284	>
3285	>	static final class ValueIterator<K,V> extends BaseIterator<K,V>
3286	>	implements Iterator<V>, Enumeration<V> {
3287	>	ValueIterator(Node<K,V>[] tab, int index, int size, int limit,
3288	>	ConcurrentHashMapV8<K,V> map) {
3289	>	super(tab, index, size, limit, map);
3290	>	}
3291	>
3292	>	public final V next() {
3293	>	Node<K,V> p;
3294	>	if ((p = next) == null)
3295	>	throw new NoSuchElementException();
3296	>	V v = p.val;
3297	>	lastReturned = p;
3298	>	advance();
3299	>	return v;
3300	>	}
3301	>
3302	>	public final V nextElement() { return next(); }
3303	>	}
3304	>
3305	>	static final class EntryIterator<K,V> extends BaseIterator<K,V>
3306	>	implements Iterator<Map.Entry<K,V>> {
3307	>	EntryIterator(Node<K,V>[] tab, int index, int size, int limit,
3308	>	ConcurrentHashMapV8<K,V> map) {
3309	>	super(tab, index, size, limit, map);
3310	>	}
3311	>
3312	>	public final Map.Entry<K,V> next() {
3313	>	Node<K,V> p;
3314	>	if ((p = next) == null)
3315	>	throw new NoSuchElementException();
3316	>	K k = p.key;
3317	>	V v = p.val;
3318	>	lastReturned = p;
3319	>	advance();
3320	>	return new MapEntry<K,V>(k, v, map);
3321	>	}
3322		}
3323
3324		/**
3325	<	* Creates a new, empty map with a default initial capacity (16),
946	<	* load factor (0.75) and concurrencyLevel (16).
3325	>	* Exported Entry for EntryIterator
3326		*/
3327	<	public ConcurrentHashMapV8() {
3328	<	this(DEFAULT_CAPACITY, DEFAULT_LOAD_FACTOR, DEFAULT_CONCURRENCY_LEVEL);
3327	>	static final class MapEntry<K,V> implements Map.Entry<K,V> {
3328	>	final K key; // non-null
3329	>	V val;       // non-null
3330	>	final ConcurrentHashMapV8<K,V> map;
3331	>	MapEntry(K key, V val, ConcurrentHashMapV8<K,V> map) {
3332	>	this.key = key;
3333	>	this.val = val;
3334	>	this.map = map;
3335	>	}
3336	>	public K getKey()        { return key; }
3337	>	public V getValue()      { return val; }
3338	>	public int hashCode()    { return key.hashCode() ^ val.hashCode(); }
3339	>	public String toString() { return key + "=" + val; }
3340	>
3341	>	public boolean equals(Object o) {
3342	>	Object k, v; Map.Entry<?,?> e;
3343	>	return ((o instanceof Map.Entry) &&
3344	>	(k = (e = (Map.Entry<?,?>)o).getKey()) != null &&
3345	>	(v = e.getValue()) != null &&
3346	>	(k == key || k.equals(key)) &&
3347	>	(v == val || v.equals(val)));
3348	>	}
3349	>
3350	>	/**
3351	>	* Sets our entry's value and writes through to the map. The
3352	>	* value to return is somewhat arbitrary here. Since we do not
3353	>	* necessarily track asynchronous changes, the most recent
3354	>	* "previous" value could be different from what we return (or
3355	>	* could even have been removed, in which case the put will
3356	>	* re-establish). We do not and cannot guarantee more.
3357	>	*/
3358	>	public V setValue(V value) {
3359	>	if (value == null) throw new NullPointerException();
3360	>	V v = val;
3361	>	val = value;
3362	>	map.put(key, value);
3363	>	return v;
3364	>	}
3365		}
3366
3367	+	static final class KeySpliterator<K,V> extends Traverser<K,V>
3368	+	implements ConcurrentHashMapSpliterator<K> {
3369	+	long est;               // size estimate
3370	+	KeySpliterator(Node<K,V>[] tab, int size, int index, int limit,
3371	+	long est) {
3372	+	super(tab, size, index, limit);
3373	+	this.est = est;
3374	+	}
3375	+
3376	+	public ConcurrentHashMapSpliterator<K> trySplit() {
3377	+	int i, f, h;
3378	+	return (h = ((i = baseIndex) + (f = baseLimit)) >>> 1) <= i ? null :
3379	+	new KeySpliterator<K,V>(tab, baseSize, baseLimit = h,
3380	+	f, est >>>= 1);
3381	+	}
3382	+
3383	+	public void forEachRemaining(Action<? super K> action) {
3384	+	if (action == null) throw new NullPointerException();
3385	+	for (Node<K,V> p; (p = advance()) != null;)
3386	+	action.apply(p.key);
3387	+	}
3388	+
3389	+	public boolean tryAdvance(Action<? super K> action) {
3390	+	if (action == null) throw new NullPointerException();
3391	+	Node<K,V> p;
3392	+	if ((p = advance()) == null)
3393	+	return false;
3394	+	action.apply(p.key);
3395	+	return true;
3396	+	}
3397	+
3398	+	public long estimateSize() { return est; }
3399	+
3400	+	}
3401	+
3402	+	static final class ValueSpliterator<K,V> extends Traverser<K,V>
3403	+	implements ConcurrentHashMapSpliterator<V> {
3404	+	long est;               // size estimate
3405	+	ValueSpliterator(Node<K,V>[] tab, int size, int index, int limit,
3406	+	long est) {
3407	+	super(tab, size, index, limit);
3408	+	this.est = est;
3409	+	}
3410	+
3411	+	public ConcurrentHashMapSpliterator<V> trySplit() {
3412	+	int i, f, h;
3413	+	return (h = ((i = baseIndex) + (f = baseLimit)) >>> 1) <= i ? null :
3414	+	new ValueSpliterator<K,V>(tab, baseSize, baseLimit = h,
3415	+	f, est >>>= 1);
3416	+	}
3417	+
3418	+	public void forEachRemaining(Action<? super V> action) {
3419	+	if (action == null) throw new NullPointerException();
3420	+	for (Node<K,V> p; (p = advance()) != null;)
3421	+	action.apply(p.val);
3422	+	}
3423	+
3424	+	public boolean tryAdvance(Action<? super V> action) {
3425	+	if (action == null) throw new NullPointerException();
3426	+	Node<K,V> p;
3427	+	if ((p = advance()) == null)
3428	+	return false;
3429	+	action.apply(p.val);
3430	+	return true;
3431	+	}
3432	+
3433	+	public long estimateSize() { return est; }
3434	+
3435	+	}
3436	+
3437	+	static final class EntrySpliterator<K,V> extends Traverser<K,V>
3438	+	implements ConcurrentHashMapSpliterator<Map.Entry<K,V>> {
3439	+	final ConcurrentHashMapV8<K,V> map; // To export MapEntry
3440	+	long est;               // size estimate
3441	+	EntrySpliterator(Node<K,V>[] tab, int size, int index, int limit,
3442	+	long est, ConcurrentHashMapV8<K,V> map) {
3443	+	super(tab, size, index, limit);
3444	+	this.map = map;
3445	+	this.est = est;
3446	+	}
3447	+
3448	+	public ConcurrentHashMapSpliterator<Map.Entry<K,V>> trySplit() {
3449	+	int i, f, h;
3450	+	return (h = ((i = baseIndex) + (f = baseLimit)) >>> 1) <= i ? null :
3451	+	new EntrySpliterator<K,V>(tab, baseSize, baseLimit = h,
3452	+	f, est >>>= 1, map);
3453	+	}
3454	+
3455	+	public void forEachRemaining(Action<? super Map.Entry<K,V>> action) {
3456	+	if (action == null) throw new NullPointerException();
3457	+	for (Node<K,V> p; (p = advance()) != null; )
3458	+	action.apply(new MapEntry<K,V>(p.key, p.val, map));
3459	+	}
3460	+
3461	+	public boolean tryAdvance(Action<? super Map.Entry<K,V>> action) {
3462	+	if (action == null) throw new NullPointerException();
3463	+	Node<K,V> p;
3464	+	if ((p = advance()) == null)
3465	+	return false;
3466	+	action.apply(new MapEntry<K,V>(p.key, p.val, map));
3467	+	return true;
3468	+	}
3469	+
3470	+	public long estimateSize() { return est; }
3471	+
3472	+	}
3473	+
3474	+	// Parallel bulk operations
3475	+
3476		/**
3477	<	* Creates a new map with the same mappings as the given map.
3478	<	* The map is created with a capacity of 1.5 times the number
3479	<	* of mappings in the given map or 16 (whichever is greater),
3480	<	* and a default load factor (0.75) and concurrencyLevel (16).
3481	<	*
3482	<	* @param m the map
3477	>	* Computes initial batch value for bulk tasks. The returned value
3478	>	* is approximately exp2 of the number of times (minus one) to
3479	>	* split task by two before executing leaf action. This value is
3480	>	* faster to compute and more convenient to use as a guide to
3481	>	* splitting than is the depth, since it is used while dividing by
3482	>	* two anyway.
3483		*/
3484	<	public ConcurrentHashMapV8(Map<? extends K, ? extends V> m) {
3485	<	this(DEFAULT_CAPACITY, DEFAULT_LOAD_FACTOR, DEFAULT_CONCURRENCY_LEVEL);
3486	<	if (m == null)
3487	<	throw new NullPointerException();
3488	<	internalPutAll(m);
3484	>	final int batchFor(long b) {
3485	>	long n;
3486	>	if (b == Long.MAX_VALUE || (n = sumCount()) <= 1L || n < b)
3487	>	return 0;
3488	>	int sp = ForkJoinPool.getCommonPoolParallelism() << 2; // slack of 4
3489	>	return (b <= 0L || (n /= b) >= sp) ? sp : (int)n;
3490		}
3491
3492		/**
3493	<	* Returns {@code true} if this map contains no key-value mappings.
3493	>	* Performs the given action for each (key, value).
3494		*
3495	<	* @return {@code true} if this map contains no key-value mappings
3495	>	* @param parallelismThreshold the (estimated) number of elements
3496	>	* needed for this operation to be executed in parallel
3497	>	* @param action the action
3498	>	* @since 1.8
3499		*/
3500	<	public boolean isEmpty() {
3501	<	return counter.sum() <= 0L; // ignore transient negative values
3500	>	public void forEach(long parallelismThreshold,
3501	>	BiAction<? super K,? super V> action) {
3502	>	if (action == null) throw new NullPointerException();
3503	>	new ForEachMappingTask<K,V>
3504	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3505	>	action).invoke();
3506		}
3507
3508		/**
3509	<	* Returns the number of key-value mappings in this map.  If the
3510	<	* map contains more than {@code Integer.MAX_VALUE} elements, returns
979	<	* {@code Integer.MAX_VALUE}.
3509	>	* Performs the given action for each non-null transformation
3510	>	* of each (key, value).
3511		*
3512	<	* @return the number of key-value mappings in this map
3512	>	* @param parallelismThreshold the (estimated) number of elements
3513	>	* needed for this operation to be executed in parallel
3514	>	* @param transformer a function returning the transformation
3515	>	* for an element, or null if there is no transformation (in
3516	>	* which case the action is not applied)
3517	>	* @param action the action
3518	>	* @since 1.8
3519		*/
3520	<	public int size() {
3521	<	long n = counter.sum();
3522	<	return ((n >>> 31) == 0) ? (int)n : (n < 0L) ? 0 : Integer.MAX_VALUE;
3520	>	public <U> void forEach(long parallelismThreshold,
3521	>	BiFun<? super K, ? super V, ? extends U> transformer,
3522	>	Action<? super U> action) {
3523	>	if (transformer == null || action == null)
3524	>	throw new NullPointerException();
3525	>	new ForEachTransformedMappingTask<K,V,U>
3526	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3527	>	transformer, action).invoke();
3528		}
3529
3530		/**
3531	<	* Returns the value to which the specified key is mapped,
3532	<	* or {@code null} if this map contains no mapping for the key.
3533	<	*
3534	<	* <p>More formally, if this map contains a mapping from a key
3535	<	* {@code k} to a value {@code v} such that {@code key.equals(k)},
3536	<	* then this method returns {@code v}; otherwise it returns
3537	<	* {@code null}.  (There can be at most one such mapping.)
3538	<	*
3539	<	* @throws NullPointerException if the specified key is null
3540	<	*/
3541	<	@SuppressWarnings("unchecked")
3542	<	public V get(Object key) {
3543	<	if (key == null)
3531	>	* Returns a non-null result from applying the given search
3532	>	* function on each (key, value), or null if none.  Upon
3533	>	* success, further element processing is suppressed and the
3534	>	* results of any other parallel invocations of the search
3535	>	* function are ignored.
3536	>	*
3537	>	* @param parallelismThreshold the (estimated) number of elements
3538	>	* needed for this operation to be executed in parallel
3539	>	* @param searchFunction a function returning a non-null
3540	>	* result on success, else null
3541	>	* @return a non-null result from applying the given search
3542	>	* function on each (key, value), or null if none
3543	>	* @since 1.8
3544	>	*/
3545	>	public <U> U search(long parallelismThreshold,
3546	>	BiFun<? super K, ? super V, ? extends U> searchFunction) {
3547	>	if (searchFunction == null) throw new NullPointerException();
3548	>	return new SearchMappingsTask<K,V,U>
3549	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3550	>	searchFunction, new AtomicReference<U>()).invoke();
3551	>	}
3552	>
3553	>	/**
3554	>	* Returns the result of accumulating the given transformation
3555	>	* of all (key, value) pairs using the given reducer to
3556	>	* combine values, or null if none.
3557	>	*
3558	>	* @param parallelismThreshold the (estimated) number of elements
3559	>	* needed for this operation to be executed in parallel
3560	>	* @param transformer a function returning the transformation
3561	>	* for an element, or null if there is no transformation (in
3562	>	* which case it is not combined)
3563	>	* @param reducer a commutative associative combining function
3564	>	* @return the result of accumulating the given transformation
3565	>	* of all (key, value) pairs
3566	>	* @since 1.8
3567	>	*/
3568	>	public <U> U reduce(long parallelismThreshold,
3569	>	BiFun<? super K, ? super V, ? extends U> transformer,
3570	>	BiFun<? super U, ? super U, ? extends U> reducer) {
3571	>	if (transformer == null || reducer == null)
3572		throw new NullPointerException();
3573	<	return (V)internalGet(key);
3573	>	return new MapReduceMappingsTask<K,V,U>
3574	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3575	>	null, transformer, reducer).invoke();
3576		}
3577
3578		/**
3579	<	* Tests if the specified object is a key in this table.
3580	<	*
3581	<	* @param  key   possible key
3582	<	* @return {@code true} if and only if the specified object
3583	<	*         is a key in this table, as determined by the
3584	<	*         {@code equals} method; {@code false} otherwise.
3585	<	* @throws NullPointerException if the specified key is null
3586	<	*/
3587	<	public boolean containsKey(Object key) {
3588	<	if (key == null)
3579	>	* Returns the result of accumulating the given transformation
3580	>	* of all (key, value) pairs using the given reducer to
3581	>	* combine values, and the given basis as an identity value.
3582	>	*
3583	>	* @param parallelismThreshold the (estimated) number of elements
3584	>	* needed for this operation to be executed in parallel
3585	>	* @param transformer a function returning the transformation
3586	>	* for an element
3587	>	* @param basis the identity (initial default value) for the reduction
3588	>	* @param reducer a commutative associative combining function
3589	>	* @return the result of accumulating the given transformation
3590	>	* of all (key, value) pairs
3591	>	* @since 1.8
3592	>	*/
3593	>	public double reduceToDouble(long parallelismThreshold,
3594	>	ObjectByObjectToDouble<? super K, ? super V> transformer,
3595	>	double basis,
3596	>	DoubleByDoubleToDouble reducer) {
3597	>	if (transformer == null || reducer == null)
3598		throw new NullPointerException();
3599	<	return internalGet(key) != null;
3599	>	return new MapReduceMappingsToDoubleTask<K,V>
3600	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3601	>	null, transformer, basis, reducer).invoke();
3602		}
3603
3604		/**
3605	<	* Returns {@code true} if this map maps one or more keys to the
3606	<	* specified value. Note: This method requires a full internal
3607	<	* traversal of the hash table, and so is much slower than
3608	<	* method {@code containsKey}.
3609	<	*
3610	<	* @param value value whose presence in this map is to be tested
3611	<	* @return {@code true} if this map maps one or more keys to the
3612	<	*         specified value
3613	<	* @throws NullPointerException if the specified value is null
3614	<	*/
3615	<	public boolean containsValue(Object value) {
3616	<	if (value == null)
3605	>	* Returns the result of accumulating the given transformation
3606	>	* of all (key, value) pairs using the given reducer to
3607	>	* combine values, and the given basis as an identity value.
3608	>	*
3609	>	* @param parallelismThreshold the (estimated) number of elements
3610	>	* needed for this operation to be executed in parallel
3611	>	* @param transformer a function returning the transformation
3612	>	* for an element
3613	>	* @param basis the identity (initial default value) for the reduction
3614	>	* @param reducer a commutative associative combining function
3615	>	* @return the result of accumulating the given transformation
3616	>	* of all (key, value) pairs
3617	>	* @since 1.8
3618	>	*/
3619	>	public long reduceToLong(long parallelismThreshold,
3620	>	ObjectByObjectToLong<? super K, ? super V> transformer,
3621	>	long basis,
3622	>	LongByLongToLong reducer) {
3623	>	if (transformer == null || reducer == null)
3624		throw new NullPointerException();
3625	<	return new HashIterator().containsVal(value);
3625	>	return new MapReduceMappingsToLongTask<K,V>
3626	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3627	>	null, transformer, basis, reducer).invoke();
3628		}
3629
3630		/**
3631	<	* Legacy method testing if some key maps into the specified value
3632	<	* in this table.  This method is identical in functionality to
3633	<	* {@link #containsValue}, and exists solely to ensure
3634	<	* full compatibility with class {@link java.util.Hashtable},
3635	<	* which supported this method prior to introduction of the
3636	<	* Java Collections framework.
3637	<	*
3638	<	* @param  value a value to search for
3639	<	* @return {@code true} if and only if some key maps to the
3640	<	*         {@code value} argument in this table as
3641	<	*         determined by the {@code equals} method;
3642	<	*         {@code false} otherwise
3643	<	* @throws NullPointerException if the specified value is null
3644	<	*/
3645	<	public boolean contains(Object value) {
3646	<	return containsValue(value);
3631	>	* Returns the result of accumulating the given transformation
3632	>	* of all (key, value) pairs using the given reducer to
3633	>	* combine values, and the given basis as an identity value.
3634	>	*
3635	>	* @param parallelismThreshold the (estimated) number of elements
3636	>	* needed for this operation to be executed in parallel
3637	>	* @param transformer a function returning the transformation
3638	>	* for an element
3639	>	* @param basis the identity (initial default value) for the reduction
3640	>	* @param reducer a commutative associative combining function
3641	>	* @return the result of accumulating the given transformation
3642	>	* of all (key, value) pairs
3643	>	* @since 1.8
3644	>	*/
3645	>	public int reduceToInt(long parallelismThreshold,
3646	>	ObjectByObjectToInt<? super K, ? super V> transformer,
3647	>	int basis,
3648	>	IntByIntToInt reducer) {
3649	>	if (transformer == null || reducer == null)
3650	>	throw new NullPointerException();
3651	>	return new MapReduceMappingsToIntTask<K,V>
3652	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3653	>	null, transformer, basis, reducer).invoke();
3654		}
3655
3656		/**
3657	<	* Maps the specified key to the specified value in this table.
1059	<	* Neither the key nor the value can be null.
3657	>	* Performs the given action for each key.
3658		*
3659	<	* <p> The value can be retrieved by calling the {@code get} method
3660	<	* with a key that is equal to the original key.
3661	<	*
3662	<	* @param key key with which the specified value is to be associated
1065	<	* @param value value to be associated with the specified key
1066	<	* @return the previous value associated with {@code key}, or
1067	<	*         {@code null} if there was no mapping for {@code key}
1068	<	* @throws NullPointerException if the specified key or value is null
3659	>	* @param parallelismThreshold the (estimated) number of elements
3660	>	* needed for this operation to be executed in parallel
3661	>	* @param action the action
3662	>	* @since 1.8
3663		*/
3664	<	@SuppressWarnings("unchecked")
3665	<	public V put(K key, V value) {
3666	<	if (key == null || value == null)
3667	<	throw new NullPointerException();
3668	<	return (V)internalPut(key, value, true);
3664	>	public void forEachKey(long parallelismThreshold,
3665	>	Action<? super K> action) {
3666	>	if (action == null) throw new NullPointerException();
3667	>	new ForEachKeyTask<K,V>
3668	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3669	>	action).invoke();
3670		}
3671
3672		/**
3673	<	* {@inheritDoc}
3673	>	* Performs the given action for each non-null transformation
3674	>	* of each key.
3675		*
3676	<	* @return the previous value associated with the specified key,
3677	<	*         or {@code null} if there was no mapping for the key
3678	<	* @throws NullPointerException if the specified key or value is null
3676	>	* @param parallelismThreshold the (estimated) number of elements
3677	>	* needed for this operation to be executed in parallel
3678	>	* @param transformer a function returning the transformation
3679	>	* for an element, or null if there is no transformation (in
3680	>	* which case the action is not applied)
3681	>	* @param action the action
3682	>	* @since 1.8
3683		*/
3684	<	@SuppressWarnings("unchecked")
3685	<	public V putIfAbsent(K key, V value) {
3686	<	if (key == null || value == null)
3684	>	public <U> void forEachKey(long parallelismThreshold,
3685	>	Fun<? super K, ? extends U> transformer,
3686	>	Action<? super U> action) {
3687	>	if (transformer == null || action == null)
3688	>	throw new NullPointerException();
3689	>	new ForEachTransformedKeyTask<K,V,U>
3690	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3691	>	transformer, action).invoke();
3692	>	}
3693	>
3694	>	/**
3695	>	* Returns a non-null result from applying the given search
3696	>	* function on each key, or null if none. Upon success,
3697	>	* further element processing is suppressed and the results of
3698	>	* any other parallel invocations of the search function are
3699	>	* ignored.
3700	>	*
3701	>	* @param parallelismThreshold the (estimated) number of elements
3702	>	* needed for this operation to be executed in parallel
3703	>	* @param searchFunction a function returning a non-null
3704	>	* result on success, else null
3705	>	* @return a non-null result from applying the given search
3706	>	* function on each key, or null if none
3707	>	* @since 1.8
3708	>	*/
3709	>	public <U> U searchKeys(long parallelismThreshold,
3710	>	Fun<? super K, ? extends U> searchFunction) {
3711	>	if (searchFunction == null) throw new NullPointerException();
3712	>	return new SearchKeysTask<K,V,U>
3713	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3714	>	searchFunction, new AtomicReference<U>()).invoke();
3715	>	}
3716	>
3717	>	/**
3718	>	* Returns the result of accumulating all keys using the given
3719	>	* reducer to combine values, or null if none.
3720	>	*
3721	>	* @param parallelismThreshold the (estimated) number of elements
3722	>	* needed for this operation to be executed in parallel
3723	>	* @param reducer a commutative associative combining function
3724	>	* @return the result of accumulating all keys using the given
3725	>	* reducer to combine values, or null if none
3726	>	* @since 1.8
3727	>	*/
3728	>	public K reduceKeys(long parallelismThreshold,
3729	>	BiFun<? super K, ? super K, ? extends K> reducer) {
3730	>	if (reducer == null) throw new NullPointerException();
3731	>	return new ReduceKeysTask<K,V>
3732	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3733	>	null, reducer).invoke();
3734	>	}
3735	>
3736	>	/**
3737	>	* Returns the result of accumulating the given transformation
3738	>	* of all keys using the given reducer to combine values, or
3739	>	* null if none.
3740	>	*
3741	>	* @param parallelismThreshold the (estimated) number of elements
3742	>	* needed for this operation to be executed in parallel
3743	>	* @param transformer a function returning the transformation
3744	>	* for an element, or null if there is no transformation (in
3745	>	* which case it is not combined)
3746	>	* @param reducer a commutative associative combining function
3747	>	* @return the result of accumulating the given transformation
3748	>	* of all keys
3749	>	* @since 1.8
3750	>	*/
3751	>	public <U> U reduceKeys(long parallelismThreshold,
3752	>	Fun<? super K, ? extends U> transformer,
3753	>	BiFun<? super U, ? super U, ? extends U> reducer) {
3754	>	if (transformer == null || reducer == null)
3755		throw new NullPointerException();
3756	<	return (V)internalPut(key, value, false);
3756	>	return new MapReduceKeysTask<K,V,U>
3757	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3758	>	null, transformer, reducer).invoke();
3759		}
3760
3761		/**
3762	<	* Copies all of the mappings from the specified map to this one.
3763	<	* These mappings replace any mappings that this map had for any of the
3764	<	* keys currently in the specified map.
3765	<	*
3766	<	* @param m mappings to be stored in this map
3767	<	*/
3768	<	public void putAll(Map<? extends K, ? extends V> m) {
3769	<	if (m == null)
3762	>	* Returns the result of accumulating the given transformation
3763	>	* of all keys using the given reducer to combine values, and
3764	>	* the given basis as an identity value.
3765	>	*
3766	>	* @param parallelismThreshold the (estimated) number of elements
3767	>	* needed for this operation to be executed in parallel
3768	>	* @param transformer a function returning the transformation
3769	>	* for an element
3770	>	* @param basis the identity (initial default value) for the reduction
3771	>	* @param reducer a commutative associative combining function
3772	>	* @return the result of accumulating the given transformation
3773	>	* of all keys
3774	>	* @since 1.8
3775	>	*/
3776	>	public double reduceKeysToDouble(long parallelismThreshold,
3777	>	ObjectToDouble<? super K> transformer,
3778	>	double basis,
3779	>	DoubleByDoubleToDouble reducer) {
3780	>	if (transformer == null || reducer == null)
3781		throw new NullPointerException();
3782	<	internalPutAll(m);
3782	>	return new MapReduceKeysToDoubleTask<K,V>
3783	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3784	>	null, transformer, basis, reducer).invoke();
3785		}
3786
3787		/**
3788	<	* If the specified key is not already associated with a value,
3789	<	* computes its value using the given mappingFunction, and if
3790	<	* non-null, enters it into the map.  This is equivalent to
3791	<	*
3792	<	* <pre>
3793	<	*   if (map.containsKey(key))
3794	<	*       return map.get(key);
3795	<	*   value = mappingFunction.map(key);
3796	<	*   if (value != null)
3797	<	*      map.put(key, value);
3798	<	*   return value;
3799	<	* </pre>
3800	<	*
3801	<	* except that the action is performed atomically.  Some attempted
3802	<	* update operations on this map by other threads may be blocked
3803	<	* while computation is in progress, so the computation should be
3804	<	* short and simple, and must not attempt to update any other
3805	<	* mappings of this Map. The most appropriate usage is to
3806	<	* construct a new object serving as an initial mapped value, or
1124	<	* memoized result, as in:
1125	<	* <pre>{@code
1126	<	* map.computeIfAbsent(key, new MappingFunction<K, V>() {
1127	<	*   public V map(K k) { return new Value(f(k)); }};
1128	<	* }</pre>
1129	<	*
1130	<	* @param key key with which the specified value is to be associated
1131	<	* @param mappingFunction the function to compute a value
1132	<	* @return the current (existing or computed) value associated with
1133	<	*         the specified key, or {@code null} if the computation
1134	<	*         returned {@code null}.
1135	<	* @throws NullPointerException if the specified key or mappingFunction
1136	<	*         is null,
1137	<	* @throws IllegalStateException if the computation detectably
1138	<	*         attempts a recursive update to this map that would
1139	<	*         otherwise never complete.
1140	<	* @throws RuntimeException or Error if the mappingFunction does so,
1141	<	*         in which case the mapping is left unestablished.
1142	<	*/
1143	<	public V computeIfAbsent(K key, MappingFunction<? super K, ? extends V> mappingFunction) {
1144	<	if (key == null || mappingFunction == null)
3788	>	* Returns the result of accumulating the given transformation
3789	>	* of all keys using the given reducer to combine values, and
3790	>	* the given basis as an identity value.
3791	>	*
3792	>	* @param parallelismThreshold the (estimated) number of elements
3793	>	* needed for this operation to be executed in parallel
3794	>	* @param transformer a function returning the transformation
3795	>	* for an element
3796	>	* @param basis the identity (initial default value) for the reduction
3797	>	* @param reducer a commutative associative combining function
3798	>	* @return the result of accumulating the given transformation
3799	>	* of all keys
3800	>	* @since 1.8
3801	>	*/
3802	>	public long reduceKeysToLong(long parallelismThreshold,
3803	>	ObjectToLong<? super K> transformer,
3804	>	long basis,
3805	>	LongByLongToLong reducer) {
3806	>	if (transformer == null || reducer == null)
3807		throw new NullPointerException();
3808	<	return internalCompute(key, mappingFunction, false);
3808	>	return new MapReduceKeysToLongTask<K,V>
3809	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3810	>	null, transformer, basis, reducer).invoke();
3811		}
3812
3813		/**
3814	<	* Computes the value associated with the given key using the given
3815	<	* mappingFunction, and if non-null, enters it into the map.  This
3816	<	* is equivalent to
3817	<	*
3818	<	* <pre>
3819	<	*   value = mappingFunction.map(key);
3820	<	*   if (value != null)
3821	<	*      map.put(key, value);
3822	<	*   else
3823	<	*      value = map.get(key);
3824	<	*   return value;
3825	<	* </pre>
3826	<	*
3827	<	* except that the action is performed atomically.  Some attempted
3828	<	* update operations on this map by other threads may be blocked
3829	<	* while computation is in progress, so the computation should be
3830	<	* short and simple, and must not attempt to update any other
3831	<	* mappings of this Map.
3832	<	*
1169	<	* @param key key with which the specified value is to be associated
1170	<	* @param mappingFunction the function to compute a value
1171	<	* @return the current value associated with
1172	<	*         the specified key, or {@code null} if the computation
1173	<	*         returned {@code null} and the value was not otherwise present.
1174	<	* @throws NullPointerException if the specified key or mappingFunction
1175	<	*         is null,
1176	<	* @throws IllegalStateException if the computation detectably
1177	<	*         attempts a recursive update to this map that would
1178	<	*         otherwise never complete.
1179	<	* @throws RuntimeException or Error if the mappingFunction does so,
1180	<	*         in which case the mapping is unchanged.
1181	<	*/
1182	<	public V compute(K key, MappingFunction<? super K, ? extends V> mappingFunction) {
1183	<	if (key == null || mappingFunction == null)
3814	>	* Returns the result of accumulating the given transformation
3815	>	* of all keys using the given reducer to combine values, and
3816	>	* the given basis as an identity value.
3817	>	*
3818	>	* @param parallelismThreshold the (estimated) number of elements
3819	>	* needed for this operation to be executed in parallel
3820	>	* @param transformer a function returning the transformation
3821	>	* for an element
3822	>	* @param basis the identity (initial default value) for the reduction
3823	>	* @param reducer a commutative associative combining function
3824	>	* @return the result of accumulating the given transformation
3825	>	* of all keys
3826	>	* @since 1.8
3827	>	*/
3828	>	public int reduceKeysToInt(long parallelismThreshold,
3829	>	ObjectToInt<? super K> transformer,
3830	>	int basis,
3831	>	IntByIntToInt reducer) {
3832	>	if (transformer == null || reducer == null)
3833		throw new NullPointerException();
3834	<	return internalCompute(key, mappingFunction, true);
3834	>	return new MapReduceKeysToIntTask<K,V>
3835	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3836	>	null, transformer, basis, reducer).invoke();
3837		}
3838
3839		/**
3840	<	* Removes the key (and its corresponding value) from this map.
1190	<	* This method does nothing if the key is not in the map.
3840	>	* Performs the given action for each value.
3841		*
3842	<	* @param  key the key that needs to be removed
3843	<	* @return the previous value associated with {@code key}, or
3844	<	*         {@code null} if there was no mapping for {@code key}
3845	<	* @throws NullPointerException if the specified key is null
3842	>	* @param parallelismThreshold the (estimated) number of elements
3843	>	* needed for this operation to be executed in parallel
3844	>	* @param action the action
3845	>	* @since 1.8
3846		*/
3847	<	@SuppressWarnings("unchecked")
3848	<	public V remove(Object key) {
3849	<	if (key == null)
3847	>	public void forEachValue(long parallelismThreshold,
3848	>	Action<? super V> action) {
3849	>	if (action == null)
3850		throw new NullPointerException();
3851	<	return (V)internalReplace(key, null, null);
3851	>	new ForEachValueTask<K,V>
3852	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3853	>	action).invoke();
3854		}
3855
3856		/**
3857	<	* {@inheritDoc}
3858	<	*
3859	<	* @throws NullPointerException if the specified key is null
3860	<	*/
3861	<	public boolean remove(Object key, Object value) {
3862	<	if (key == null)
3857	>	* Performs the given action for each non-null transformation
3858	>	* of each value.
3859	>	*
3860	>	* @param parallelismThreshold the (estimated) number of elements
3861	>	* needed for this operation to be executed in parallel
3862	>	* @param transformer a function returning the transformation
3863	>	* for an element, or null if there is no transformation (in
3864	>	* which case the action is not applied)
3865	>	* @param action the action
3866	>	* @since 1.8
3867	>	*/
3868	>	public <U> void forEachValue(long parallelismThreshold,
3869	>	Fun<? super V, ? extends U> transformer,
3870	>	Action<? super U> action) {
3871	>	if (transformer == null || action == null)
3872		throw new NullPointerException();
3873	<	if (value == null)
3874	<	return false;
3875	<	return internalReplace(key, null, value) != null;
3873	>	new ForEachTransformedValueTask<K,V,U>
3874	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3875	>	transformer, action).invoke();
3876	>	}
3877	>
3878	>	/**
3879	>	* Returns a non-null result from applying the given search
3880	>	* function on each value, or null if none.  Upon success,
3881	>	* further element processing is suppressed and the results of
3882	>	* any other parallel invocations of the search function are
3883	>	* ignored.
3884	>	*
3885	>	* @param parallelismThreshold the (estimated) number of elements
3886	>	* needed for this operation to be executed in parallel
3887	>	* @param searchFunction a function returning a non-null
3888	>	* result on success, else null
3889	>	* @return a non-null result from applying the given search
3890	>	* function on each value, or null if none
3891	>	* @since 1.8
3892	>	*/
3893	>	public <U> U searchValues(long parallelismThreshold,
3894	>	Fun<? super V, ? extends U> searchFunction) {
3895	>	if (searchFunction == null) throw new NullPointerException();
3896	>	return new SearchValuesTask<K,V,U>
3897	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3898	>	searchFunction, new AtomicReference<U>()).invoke();
3899	>	}
3900	>
3901	>	/**
3902	>	* Returns the result of accumulating all values using the
3903	>	* given reducer to combine values, or null if none.
3904	>	*
3905	>	* @param parallelismThreshold the (estimated) number of elements
3906	>	* needed for this operation to be executed in parallel
3907	>	* @param reducer a commutative associative combining function
3908	>	* @return the result of accumulating all values
3909	>	* @since 1.8
3910	>	*/
3911	>	public V reduceValues(long parallelismThreshold,
3912	>	BiFun<? super V, ? super V, ? extends V> reducer) {
3913	>	if (reducer == null) throw new NullPointerException();
3914	>	return new ReduceValuesTask<K,V>
3915	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3916	>	null, reducer).invoke();
3917	>	}
3918	>
3919	>	/**
3920	>	* Returns the result of accumulating the given transformation
3921	>	* of all values using the given reducer to combine values, or
3922	>	* null if none.
3923	>	*
3924	>	* @param parallelismThreshold the (estimated) number of elements
3925	>	* needed for this operation to be executed in parallel
3926	>	* @param transformer a function returning the transformation
3927	>	* for an element, or null if there is no transformation (in
3928	>	* which case it is not combined)
3929	>	* @param reducer a commutative associative combining function
3930	>	* @return the result of accumulating the given transformation
3931	>	* of all values
3932	>	* @since 1.8
3933	>	*/
3934	>	public <U> U reduceValues(long parallelismThreshold,
3935	>	Fun<? super V, ? extends U> transformer,
3936	>	BiFun<? super U, ? super U, ? extends U> reducer) {
3937	>	if (transformer == null || reducer == null)
3938	>	throw new NullPointerException();
3939	>	return new MapReduceValuesTask<K,V,U>
3940	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3941	>	null, transformer, reducer).invoke();
3942		}
3943
3944		/**
3945	<	* {@inheritDoc}
3946	<	*
3947	<	* @throws NullPointerException if any of the arguments are null
3948	<	*/
3949	<	public boolean replace(K key, V oldValue, V newValue) {
3950	<	if (key == null || oldValue == null || newValue == null)
3945	>	* Returns the result of accumulating the given transformation
3946	>	* of all values using the given reducer to combine values,
3947	>	* and the given basis as an identity value.
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
3953	>	* @param basis the identity (initial default value) for the reduction
3954	>	* @param reducer a commutative associative combining function
3955	>	* @return the result of accumulating the given transformation
3956	>	* of all values
3957	>	* @since 1.8
3958	>	*/
3959	>	public double reduceValuesToDouble(long parallelismThreshold,
3960	>	ObjectToDouble<? super V> transformer,
3961	>	double basis,
3962	>	DoubleByDoubleToDouble reducer) {
3963	>	if (transformer == null || reducer == null)
3964		throw new NullPointerException();
3965	<	return internalReplace(key, newValue, oldValue) != null;
3965	>	return new MapReduceValuesToDoubleTask<K,V>
3966	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3967	>	null, transformer, basis, reducer).invoke();
3968		}
3969
3970		/**
3971	<	* {@inheritDoc}
3972	<	*
3973	<	* @return the previous value associated with the specified key,
3974	<	*         or {@code null} if there was no mapping for the key
3975	<	* @throws NullPointerException if the specified key or value is null
3976	<	*/
3977	<	@SuppressWarnings("unchecked")
3978	<	public V replace(K key, V value) {
3979	<	if (key == null || value == null)
3971	>	* Returns the result of accumulating the given transformation
3972	>	* of all values using the given reducer to combine values,
3973	>	* and the given basis as an identity value.
3974	>	*
3975	>	* @param parallelismThreshold the (estimated) number of elements
3976	>	* needed for this operation to be executed in parallel
3977	>	* @param transformer a function returning the transformation
3978	>	* for an element
3979	>	* @param basis the identity (initial default value) for the reduction
3980	>	* @param reducer a commutative associative combining function
3981	>	* @return the result of accumulating the given transformation
3982	>	* of all values
3983	>	* @since 1.8
3984	>	*/
3985	>	public long reduceValuesToLong(long parallelismThreshold,
3986	>	ObjectToLong<? super V> transformer,
3987	>	long basis,
3988	>	LongByLongToLong reducer) {
3989	>	if (transformer == null || reducer == null)
3990		throw new NullPointerException();
3991	<	return (V)internalReplace(key, value, null);
3991	>	return new MapReduceValuesToLongTask<K,V>
3992	>	(null, batchFor(parallelismThreshold), 0, 0, table,
3993	>	null, transformer, basis, reducer).invoke();
3994		}
3995
3996		/**
3997	<	* Removes all of the mappings from this map.
3998	<	*/
3999	<	public void clear() {
4000	<	internalClear();
3997	>	* Returns the result of accumulating the given transformation
3998	>	* of all values using the given reducer to combine values,
3999	>	* and the given basis as an identity value.
4000	>	*
4001	>	* @param parallelismThreshold the (estimated) number of elements
4002	>	* needed for this operation to be executed in parallel
4003	>	* @param transformer a function returning the transformation
4004	>	* for an element
4005	>	* @param basis the identity (initial default value) for the reduction
4006	>	* @param reducer a commutative associative combining function
4007	>	* @return the result of accumulating the given transformation
4008	>	* of all values
4009	>	* @since 1.8
4010	>	*/
4011	>	public int reduceValuesToInt(long parallelismThreshold,
4012	>	ObjectToInt<? super V> transformer,
4013	>	int basis,
4014	>	IntByIntToInt reducer) {
4015	>	if (transformer == null || reducer == null)
4016	>	throw new NullPointerException();
4017	>	return new MapReduceValuesToIntTask<K,V>
4018	>	(null, batchFor(parallelismThreshold), 0, 0, table,
4019	>	null, transformer, basis, reducer).invoke();
4020		}
4021
4022		/**
4023	<	* Returns a {@link Set} view of the keys contained in this map.
1251	<	* The set is backed by the map, so changes to the map are
1252	<	* reflected in the set, and vice-versa.  The set supports element
1253	<	* removal, which removes the corresponding mapping from this map,
1254	<	* via the {@code Iterator.remove}, {@code Set.remove},
1255	<	* {@code removeAll}, {@code retainAll}, and {@code clear}
1256	<	* operations.  It does not support the {@code add} or
1257	<	* {@code addAll} operations.
4023	>	* Performs the given action for each entry.
4024		*
4025	<	* <p>The view's {@code iterator} is a "weakly consistent" iterator
4026	<	* that will never throw {@link ConcurrentModificationException},
4027	<	* and guarantees to traverse elements as they existed upon
4028	<	* construction of the iterator, and may (but is not guaranteed to)
1263	<	* reflect any modifications subsequent to construction.
4025	>	* @param parallelismThreshold the (estimated) number of elements
4026	>	* needed for this operation to be executed in parallel
4027	>	* @param action the action
4028	>	* @since 1.8
4029		*/
4030	<	public Set<K> keySet() {
4031	<	Set<K> ks = keySet;
4032	<	return (ks != null) ? ks : (keySet = new KeySet());
4030	>	public void forEachEntry(long parallelismThreshold,
4031	>	Action<? super Map.Entry<K,V>> action) {
4032	>	if (action == null) throw new NullPointerException();
4033	>	new ForEachEntryTask<K,V>(null, batchFor(parallelismThreshold), 0, 0, table,
4034	>	action).invoke();
4035		}
4036
4037		/**
4038	<	* Returns a {@link Collection} view of the values contained in this map.
4039	<	* The collection is backed by the map, so changes to the map are
1273	<	* reflected in the collection, and vice-versa.  The collection
1274	<	* supports element removal, which removes the corresponding
1275	<	* mapping from this map, via the {@code Iterator.remove},
1276	<	* {@code Collection.remove}, {@code removeAll},
1277	<	* {@code retainAll}, and {@code clear} operations.  It does not
1278	<	* support the {@code add} or {@code addAll} operations.
4038	>	* Performs the given action for each non-null transformation
4039	>	* of each entry.
4040		*
4041	<	* <p>The view's {@code iterator} is a "weakly consistent" iterator
4042	<	* that will never throw {@link ConcurrentModificationException},
4043	<	* and guarantees to traverse elements as they existed upon
4044	<	* construction of the iterator, and may (but is not guaranteed to)
4045	<	* reflect any modifications subsequent to construction.
4041	>	* @param parallelismThreshold the (estimated) number of elements
4042	>	* needed for this operation to be executed in parallel
4043	>	* @param transformer a function returning the transformation
4044	>	* for an element, or null if there is no transformation (in
4045	>	* which case the action is not applied)
4046	>	* @param action the action
4047	>	* @since 1.8
4048		*/
4049	<	public Collection<V> values() {
4050	<	Collection<V> vs = values;
4051	<	return (vs != null) ? vs : (values = new Values());
4049	>	public <U> void forEachEntry(long parallelismThreshold,
4050	>	Fun<Map.Entry<K,V>, ? extends U> transformer,
4051	>	Action<? super U> action) {
4052	>	if (transformer == null || action == null)
4053	>	throw new NullPointerException();
4054	>	new ForEachTransformedEntryTask<K,V,U>
4055	>	(null, batchFor(parallelismThreshold), 0, 0, table,
4056	>	transformer, action).invoke();
4057	>	}
4058	>
4059	>	/**
4060	>	* Returns a non-null result from applying the given search
4061	>	* function on each entry, or null if none.  Upon success,
4062	>	* further element processing is suppressed and the results of
4063	>	* any other parallel invocations of the search function are
4064	>	* ignored.
4065	>	*
4066	>	* @param parallelismThreshold the (estimated) number of elements
4067	>	* needed for this operation to be executed in parallel
4068	>	* @param searchFunction a function returning a non-null
4069	>	* result on success, else null
4070	>	* @return a non-null result from applying the given search
4071	>	* function on each entry, or null if none
4072	>	* @since 1.8
4073	>	*/
4074	>	public <U> U searchEntries(long parallelismThreshold,
4075	>	Fun<Map.Entry<K,V>, ? extends U> searchFunction) {
4076	>	if (searchFunction == null) throw new NullPointerException();
4077	>	return new SearchEntriesTask<K,V,U>
4078	>	(null, batchFor(parallelismThreshold), 0, 0, table,
4079	>	searchFunction, new AtomicReference<U>()).invoke();
4080	>	}
4081	>
4082	>	/**
4083	>	* Returns the result of accumulating all entries using the
4084	>	* given reducer to combine values, or null if none.
4085	>	*
4086	>	* @param parallelismThreshold the (estimated) number of elements
4087	>	* needed for this operation to be executed in parallel
4088	>	* @param reducer a commutative associative combining function
4089	>	* @return the result of accumulating all entries
4090	>	* @since 1.8
4091	>	*/
4092	>	public Map.Entry<K,V> reduceEntries(long parallelismThreshold,
4093	>	BiFun<Map.Entry<K,V>, Map.Entry<K,V>, ? extends Map.Entry<K,V>> reducer) {
4094	>	if (reducer == null) throw new NullPointerException();
4095	>	return new ReduceEntriesTask<K,V>
4096	>	(null, batchFor(parallelismThreshold), 0, 0, table,
4097	>	null, reducer).invoke();
4098	>	}
4099	>
4100	>	/**
4101	>	* Returns the result of accumulating the given transformation
4102	>	* of all entries using the given reducer to combine values,
4103	>	* or null if none.
4104	>	*
4105	>	* @param parallelismThreshold the (estimated) number of elements
4106	>	* needed for this operation to be executed in parallel
4107	>	* @param transformer a function returning the transformation
4108	>	* for an element, or null if there is no transformation (in
4109	>	* which case it is not combined)
4110	>	* @param reducer a commutative associative combining function
4111	>	* @return the result of accumulating the given transformation
4112	>	* of all entries
4113	>	* @since 1.8
4114	>	*/
4115	>	public <U> U reduceEntries(long parallelismThreshold,
4116	>	Fun<Map.Entry<K,V>, ? extends U> transformer,
4117	>	BiFun<? super U, ? super U, ? extends U> reducer) {
4118	>	if (transformer == null || reducer == null)
4119	>	throw new NullPointerException();
4120	>	return new MapReduceEntriesTask<K,V,U>
4121	>	(null, batchFor(parallelismThreshold), 0, 0, table,
4122	>	null, transformer, reducer).invoke();
4123		}
4124
4125		/**
4126	<	* Returns a {@link Set} view of the mappings contained in this map.
4127	<	* The set is backed by the map, so changes to the map are
4128	<	* reflected in the set, and vice-versa.  The set supports element
4129	<	* removal, which removes the corresponding mapping from the map,
4130	<	* via the {@code Iterator.remove}, {@code Set.remove},
4131	<	* {@code removeAll}, {@code retainAll}, and {@code clear}
4132	<	* operations.  It does not support the {@code add} or
4133	<	* {@code addAll} operations.
4134	<	*
4135	<	* <p>The view's {@code iterator} is a "weakly consistent" iterator
4136	<	* that will never throw {@link ConcurrentModificationException},
4137	<	* and guarantees to traverse elements as they existed upon
4138	<	* construction of the iterator, and may (but is not guaranteed to)
4139	<	* reflect any modifications subsequent to construction.
4140	<	*/
4141	<	public Set<Map.Entry<K,V>> entrySet() {
4142	<	Set<Map.Entry<K,V>> es = entrySet;
4143	<	return (es != null) ? es : (entrySet = new EntrySet());
4126	>	* Returns the result of accumulating the given transformation
4127	>	* of all entries using the given reducer to combine values,
4128	>	* and the given basis as an identity value.
4129	>	*
4130	>	* @param parallelismThreshold the (estimated) number of elements
4131	>	* needed for this operation to be executed in parallel
4132	>	* @param transformer a function returning the transformation
4133	>	* for an element
4134	>	* @param basis the identity (initial default value) for the reduction
4135	>	* @param reducer a commutative associative combining function
4136	>	* @return the result of accumulating the given transformation
4137	>	* of all entries
4138	>	* @since 1.8
4139	>	*/
4140	>	public double reduceEntriesToDouble(long parallelismThreshold,
4141	>	ObjectToDouble<Map.Entry<K,V>> transformer,
4142	>	double basis,
4143	>	DoubleByDoubleToDouble reducer) {
4144	>	if (transformer == null || reducer == null)
4145	>	throw new NullPointerException();
4146	>	return new MapReduceEntriesToDoubleTask<K,V>
4147	>	(null, batchFor(parallelismThreshold), 0, 0, table,
4148	>	null, transformer, basis, reducer).invoke();
4149		}
4150
4151		/**
4152	<	* Returns an enumeration of the keys in this table.
4153	<	*
4154	<	* @return an enumeration of the keys in this table
4155	<	* @see #keySet()
4156	<	*/
4157	<	public Enumeration<K> keys() {
4158	<	return new KeyIterator();
4152	>	* Returns the result of accumulating the given transformation
4153	>	* of all entries using the given reducer to combine values,
4154	>	* and the given basis as an identity value.
4155	>	*
4156	>	* @param parallelismThreshold the (estimated) number of elements
4157	>	* needed for this operation to be executed in parallel
4158	>	* @param transformer a function returning the transformation
4159	>	* for an element
4160	>	* @param basis the identity (initial default value) for the reduction
4161	>	* @param reducer a commutative associative combining function
4162	>	* @return the result of accumulating the given transformation
4163	>	* of all entries
4164	>	* @since 1.8
4165	>	*/
4166	>	public long reduceEntriesToLong(long parallelismThreshold,
4167	>	ObjectToLong<Map.Entry<K,V>> transformer,
4168	>	long basis,
4169	>	LongByLongToLong reducer) {
4170	>	if (transformer == null || reducer == null)
4171	>	throw new NullPointerException();
4172	>	return new MapReduceEntriesToLongTask<K,V>
4173	>	(null, batchFor(parallelismThreshold), 0, 0, table,
4174	>	null, transformer, basis, reducer).invoke();
4175		}
4176
4177		/**
4178	<	* Returns an enumeration of the values in this table.
4179	<	*
4180	<	* @return an enumeration of the values in this table
4181	<	* @see #values()
4182	<	*/
4183	<	public Enumeration<V> elements() {
4184	<	return new ValueIterator();
4178	>	* Returns the result of accumulating the given transformation
4179	>	* of all entries using the given reducer to combine values,
4180	>	* and the given basis as an identity value.
4181	>	*
4182	>	* @param parallelismThreshold the (estimated) number of elements
4183	>	* needed for this operation to be executed in parallel
4184	>	* @param transformer a function returning the transformation
4185	>	* for an element
4186	>	* @param basis the identity (initial default value) for the reduction
4187	>	* @param reducer a commutative associative combining function
4188	>	* @return the result of accumulating the given transformation
4189	>	* of all entries
4190	>	* @since 1.8
4191	>	*/
4192	>	public int reduceEntriesToInt(long parallelismThreshold,
4193	>	ObjectToInt<Map.Entry<K,V>> transformer,
4194	>	int basis,
4195	>	IntByIntToInt reducer) {
4196	>	if (transformer == null || reducer == null)
4197	>	throw new NullPointerException();
4198	>	return new MapReduceEntriesToIntTask<K,V>
4199	>	(null, batchFor(parallelismThreshold), 0, 0, table,
4200	>	null, transformer, basis, reducer).invoke();
4201		}
4202
4203	+
4204	+	/* ----------------Views -------------- */
4205	+
4206		/**
4207	<	* Returns the hash code value for this {@link Map}, i.e.,
1334	<	* the sum of, for each key-value pair in the map,
1335	<	* {@code key.hashCode() ^ value.hashCode()}.
1336	<	*
1337	<	* @return the hash code value for this map
4207	>	* Base class for views.
4208		*/
4209	<	public int hashCode() {
4210	<	return new HashIterator().mapHashCode();
4209	>	abstract static class CollectionView<K,V,E>
4210	>	implements Collection<E>, java.io.Serializable {
4211	>	private static final long serialVersionUID = 7249069246763182397L;
4212	>	final ConcurrentHashMapV8<K,V> map;
4213	>	CollectionView(ConcurrentHashMapV8<K,V> map)  { this.map = map; }
4214	>
4215	>	/**
4216	>	* Returns the map backing this view.
4217	>	*
4218	>	* @return the map backing this view
4219	>	*/
4220	>	public ConcurrentHashMapV8<K,V> getMap() { return map; }
4221	>
4222	>	/**
4223	>	* Removes all of the elements from this view, by removing all
4224	>	* the mappings from the map backing this view.
4225	>	*/
4226	>	public final void clear()      { map.clear(); }
4227	>	public final int size()        { return map.size(); }
4228	>	public final boolean isEmpty() { return map.isEmpty(); }
4229	>
4230	>	// implementations below rely on concrete classes supplying these
4231	>	// abstract methods
4232	>	/**
4233	>	* Returns a "weakly consistent" iterator that will never
4234	>	* throw {@link ConcurrentModificationException}, and
4235	>	* guarantees to traverse elements as they existed upon
4236	>	* construction of the iterator, and may (but is not
4237	>	* guaranteed to) reflect any modifications subsequent to
4238	>	* construction.
4239	>	*/
4240	>	public abstract Iterator<E> iterator();
4241	>	public abstract boolean contains(Object o);
4242	>	public abstract boolean remove(Object o);
4243	>
4244	>	private static final String oomeMsg = "Required array size too large";
4245	>
4246	>	public final Object[] toArray() {
4247	>	long sz = map.mappingCount();
4248	>	if (sz > MAX_ARRAY_SIZE)
4249	>	throw new OutOfMemoryError(oomeMsg);
4250	>	int n = (int)sz;
4251	>	Object[] r = new Object[n];
4252	>	int i = 0;
4253	>	for (E e : this) {
4254	>	if (i == n) {
4255	>	if (n >= MAX_ARRAY_SIZE)
4256	>	throw new OutOfMemoryError(oomeMsg);
4257	>	if (n >= MAX_ARRAY_SIZE - (MAX_ARRAY_SIZE >>> 1) - 1)
4258	>	n = MAX_ARRAY_SIZE;
4259	>	else
4260	>	n += (n >>> 1) + 1;
4261	>	r = Arrays.copyOf(r, n);
4262	>	}
4263	>	r[i++] = e;
4264	>	}
4265	>	return (i == n) ? r : Arrays.copyOf(r, i);
4266	>	}
4267	>
4268	>	@SuppressWarnings("unchecked")
4269	>	public final <T> T[] toArray(T[] a) {
4270	>	long sz = map.mappingCount();
4271	>	if (sz > MAX_ARRAY_SIZE)
4272	>	throw new OutOfMemoryError(oomeMsg);
4273	>	int m = (int)sz;
4274	>	T[] r = (a.length >= m) ? a :
4275	>	(T[])java.lang.reflect.Array
4276	>	.newInstance(a.getClass().getComponentType(), m);
4277	>	int n = r.length;
4278	>	int i = 0;
4279	>	for (E e : this) {
4280	>	if (i == n) {
4281	>	if (n >= MAX_ARRAY_SIZE)
4282	>	throw new OutOfMemoryError(oomeMsg);
4283	>	if (n >= MAX_ARRAY_SIZE - (MAX_ARRAY_SIZE >>> 1) - 1)
4284	>	n = MAX_ARRAY_SIZE;
4285	>	else
4286	>	n += (n >>> 1) + 1;
4287	>	r = Arrays.copyOf(r, n);
4288	>	}
4289	>	r[i++] = (T)e;
4290	>	}
4291	>	if (a == r && i < n) {
4292	>	r[i] = null; // null-terminate
4293	>	return r;
4294	>	}
4295	>	return (i == n) ? r : Arrays.copyOf(r, i);
4296	>	}
4297	>
4298	>	/**
4299	>	* Returns a string representation of this collection.
4300	>	* The string representation consists of the string representations
4301	>	* of the collection's elements in the order they are returned by
4302	>	* its iterator, enclosed in square brackets ({@code "[]"}).
4303	>	* Adjacent elements are separated by the characters {@code ", "}
4304	>	* (comma and space).  Elements are converted to strings as by
4305	>	* {@link String#valueOf(Object)}.
4306	>	*
4307	>	* @return a string representation of this collection
4308	>	*/
4309	>	public final String toString() {
4310	>	StringBuilder sb = new StringBuilder();
4311	>	sb.append('[');
4312	>	Iterator<E> it = iterator();
4313	>	if (it.hasNext()) {
4314	>	for (;;) {
4315	>	Object e = it.next();
4316	>	sb.append(e == this ? "(this Collection)" : e);
4317	>	if (!it.hasNext())
4318	>	break;
4319	>	sb.append(',').append(' ');
4320	>	}
4321	>	}
4322	>	return sb.append(']').toString();
4323	>	}
4324	>
4325	>	public final boolean containsAll(Collection<?> c) {
4326	>	if (c != this) {
4327	>	for (Object e : c) {
4328	>	if (e == null || !contains(e))
4329	>	return false;
4330	>	}
4331	>	}
4332	>	return true;
4333	>	}
4334	>
4335	>	public final boolean removeAll(Collection<?> c) {
4336	>	boolean modified = false;
4337	>	for (Iterator<E> it = iterator(); it.hasNext();) {
4338	>	if (c.contains(it.next())) {
4339	>	it.remove();
4340	>	modified = true;
4341	>	}
4342	>	}
4343	>	return modified;
4344	>	}
4345	>
4346	>	public final boolean retainAll(Collection<?> c) {
4347	>	boolean modified = false;
4348	>	for (Iterator<E> it = iterator(); it.hasNext();) {
4349	>	if (!c.contains(it.next())) {
4350	>	it.remove();
4351	>	modified = true;
4352	>	}
4353	>	}
4354	>	return modified;
4355	>	}
4356	>
4357		}
4358
4359		/**
4360	<	* Returns a string representation of this map.  The string
4361	<	* representation consists of a list of key-value mappings (in no
4362	<	* particular order) enclosed in braces ("{@code {}}").  Adjacent
4363	<	* mappings are separated by the characters {@code ", "} (comma
4364	<	* and space).  Each key-value mapping is rendered as the key
4365	<	* followed by an equals sign ("{@code =}") followed by the
4366	<	* associated value.
4360	>	* A view of a ConcurrentHashMapV8 as a {@link Set} of keys, in
4361	>	* which additions may optionally be enabled by mapping to a
4362	>	* common value.  This class cannot be directly instantiated.
4363	>	* See {@link #keySet() keySet()},
4364	>	* {@link #keySet(Object) keySet(V)},
4365	>	* {@link #newKeySet() newKeySet()},
4366	>	* {@link #newKeySet(int) newKeySet(int)}.
4367		*
4368	<	* @return a string representation of this map
4368	>	* @since 1.8
4369		*/
4370	<	public String toString() {
4371	<	return new HashIterator().mapToString();
4370	>	public static class KeySetView<K,V> extends CollectionView<K,V,K>
4371	>	implements Set<K>, java.io.Serializable {
4372	>	private static final long serialVersionUID = 7249069246763182397L;
4373	>	private final V value;
4374	>	KeySetView(ConcurrentHashMapV8<K,V> map, V value) {  // non-public
4375	>	super(map);
4376	>	this.value = value;
4377	>	}
4378	>
4379	>	/**
4380	>	* Returns the default mapped value for additions,
4381	>	* or {@code null} if additions are not supported.
4382	>	*
4383	>	* @return the default mapped value for additions, or {@code null}
4384	>	* if not supported
4385	>	*/
4386	>	public V getMappedValue() { return value; }
4387	>
4388	>	/**
4389	>	* {@inheritDoc}
4390	>	* @throws NullPointerException if the specified key is null
4391	>	*/
4392	>	public boolean contains(Object o) { return map.containsKey(o); }
4393	>
4394	>	/**
4395	>	* Removes the key from this map view, by removing the key (and its
4396	>	* corresponding value) from the backing map.  This method does
4397	>	* nothing if the key is not in the map.
4398	>	*
4399	>	* @param  o the key to be removed from the backing map
4400	>	* @return {@code true} if the backing map contained the specified key
4401	>	* @throws NullPointerException if the specified key is null
4402	>	*/
4403	>	public boolean remove(Object o) { return map.remove(o) != null; }
4404	>
4405	>	/**
4406	>	* @return an iterator over the keys of the backing map
4407	>	*/
4408	>	public Iterator<K> iterator() {
4409	>	Node<K,V>[] t;
4410	>	ConcurrentHashMapV8<K,V> m = map;
4411	>	int f = (t = m.table) == null ? 0 : t.length;
4412	>	return new KeyIterator<K,V>(t, f, 0, f, m);
4413	>	}
4414	>
4415	>	/**
4416	>	* Adds the specified key to this set view by mapping the key to
4417	>	* the default mapped value in the backing map, if defined.
4418	>	*
4419	>	* @param e key to be added
4420	>	* @return {@code true} if this set changed as a result of the call
4421	>	* @throws NullPointerException if the specified key is null
4422	>	* @throws UnsupportedOperationException if no default mapped value
4423	>	* for additions was provided
4424	>	*/
4425	>	public boolean add(K e) {
4426	>	V v;
4427	>	if ((v = value) == null)
4428	>	throw new UnsupportedOperationException();
4429	>	return map.putVal(e, v, true) == null;
4430	>	}
4431	>
4432	>	/**
4433	>	* Adds all of the elements in the specified collection to this set,
4434	>	* as if by calling {@link #add} on each one.
4435	>	*
4436	>	* @param c the elements to be inserted into this set
4437	>	* @return {@code true} if this set changed as a result of the call
4438	>	* @throws NullPointerException if the collection or any of its
4439	>	* elements are {@code null}
4440	>	* @throws UnsupportedOperationException if no default mapped value
4441	>	* for additions was provided
4442	>	*/
4443	>	public boolean addAll(Collection<? extends K> c) {
4444	>	boolean added = false;
4445	>	V v;
4446	>	if ((v = value) == null)
4447	>	throw new UnsupportedOperationException();
4448	>	for (K e : c) {
4449	>	if (map.putVal(e, v, true) == null)
4450	>	added = true;
4451	>	}
4452	>	return added;
4453	>	}
4454	>
4455	>	public int hashCode() {
4456	>	int h = 0;
4457	>	for (K e : this)
4458	>	h += e.hashCode();
4459	>	return h;
4460	>	}
4461	>
4462	>	public boolean equals(Object o) {
4463	>	Set<?> c;
4464	>	return ((o instanceof Set) &&
4465	>	((c = (Set<?>)o) == this ||
4466	>	(containsAll(c) && c.containsAll(this))));
4467	>	}
4468	>
4469	>	public ConcurrentHashMapSpliterator<K> spliterator() {
4470	>	Node<K,V>[] t;
4471	>	ConcurrentHashMapV8<K,V> m = map;
4472	>	long n = m.sumCount();
4473	>	int f = (t = m.table) == null ? 0 : t.length;
4474	>	return new KeySpliterator<K,V>(t, f, 0, f, n < 0L ? 0L : n);
4475	>	}
4476	>
4477	>	public void forEach(Action<? super K> action) {
4478	>	if (action == null) throw new NullPointerException();
4479	>	Node<K,V>[] t;
4480	>	if ((t = map.table) != null) {
4481	>	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
4482	>	for (Node<K,V> p; (p = it.advance()) != null; )
4483	>	action.apply(p.key);
4484	>	}
4485	>	}
4486		}
4487
4488		/**
4489	<	* Compares the specified object with this map for equality.
4490	<	* Returns {@code true} if the given object is a map with the same
4491	<	* mappings as this map.  This operation may return misleading
1362	<	* results if either map is concurrently modified during execution
1363	<	* of this method.
1364	<	*
1365	<	* @param o object to be compared for equality with this map
1366	<	* @return {@code true} if the specified object is equal to this map
4489	>	* A view of a ConcurrentHashMapV8 as a {@link Collection} of
4490	>	* values, in which additions are disabled. This class cannot be
4491	>	* directly instantiated. See {@link #values()}.
4492		*/
4493	<	public boolean equals(Object o) {
4494	<	if (o == this)
4495	<	return true;
4496	<	if (!(o instanceof Map))
4497	<	return false;
4498	<	Map<?,?> m = (Map<?,?>) o;
4499	<	try {
4500	<	for (Map.Entry<K,V> e : this.entrySet())
4501	<	if (! e.getValue().equals(m.get(e.getKey())))
4502	<	return false;
4503	<	for (Map.Entry<?,?> e : m.entrySet()) {
4504	<	Object k = e.getKey();
4505	<	Object v = e.getValue();
4506	<	if (k == null || v == null || !v.equals(get(k)))
4507	<	return false;
4493	>	static final class ValuesView<K,V> extends CollectionView<K,V,V>
4494	>	implements Collection<V>, java.io.Serializable {
4495	>	private static final long serialVersionUID = 2249069246763182397L;
4496	>	ValuesView(ConcurrentHashMapV8<K,V> map) { super(map); }
4497	>	public final boolean contains(Object o) {
4498	>	return map.containsValue(o);
4499	>	}
4500	>
4501	>	public final boolean remove(Object o) {
4502	>	if (o != null) {
4503	>	for (Iterator<V> it = iterator(); it.hasNext();) {
4504	>	if (o.equals(it.next())) {
4505	>	it.remove();
4506	>	return true;
4507	>	}
4508	>	}
4509		}
1384	–	return true;
1385	–	} catch (ClassCastException unused) {
1386	–	return false;
1387	–	} catch (NullPointerException unused) {
4510		return false;
4511		}
4512	+
4513	+	public final Iterator<V> iterator() {
4514	+	ConcurrentHashMapV8<K,V> m = map;
4515	+	Node<K,V>[] t;
4516	+	int f = (t = m.table) == null ? 0 : t.length;
4517	+	return new ValueIterator<K,V>(t, f, 0, f, m);
4518	+	}
4519	+
4520	+	public final boolean add(V e) {
4521	+	throw new UnsupportedOperationException();
4522	+	}
4523	+	public final boolean addAll(Collection<? extends V> c) {
4524	+	throw new UnsupportedOperationException();
4525	+	}
4526	+
4527	+	public ConcurrentHashMapSpliterator<V> spliterator() {
4528	+	Node<K,V>[] t;
4529	+	ConcurrentHashMapV8<K,V> m = map;
4530	+	long n = m.sumCount();
4531	+	int f = (t = m.table) == null ? 0 : t.length;
4532	+	return new ValueSpliterator<K,V>(t, f, 0, f, n < 0L ? 0L : n);
4533	+	}
4534	+
4535	+	public void forEach(Action<? super V> action) {
4536	+	if (action == null) throw new NullPointerException();
4537	+	Node<K,V>[] t;
4538	+	if ((t = map.table) != null) {
4539	+	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
4540	+	for (Node<K,V> p; (p = it.advance()) != null; )
4541	+	action.apply(p.val);
4542	+	}
4543	+	}
4544		}
4545
4546		/**
4547	<	* Custom Entry class used by EntryIterator.next(), that relays
4548	<	* setValue changes to the underlying map.
4547	>	* A view of a ConcurrentHashMapV8 as a {@link Set} of (key, value)
4548	>	* entries.  This class cannot be directly instantiated. See
4549	>	* {@link #entrySet()}.
4550		*/
4551	<	final class WriteThroughEntry extends AbstractMap.SimpleEntry<K,V> {
4552	<	@SuppressWarnings("unchecked")
4553	<	WriteThroughEntry(Object k, Object v) {
4554	<	super((K)k, (V)v);
4551	>	static final class EntrySetView<K,V> extends CollectionView<K,V,Map.Entry<K,V>>
4552	>	implements Set<Map.Entry<K,V>>, java.io.Serializable {
4553	>	private static final long serialVersionUID = 2249069246763182397L;
4554	>	EntrySetView(ConcurrentHashMapV8<K,V> map) { super(map); }
4555	>
4556	>	public boolean contains(Object o) {
4557	>	Object k, v, r; Map.Entry<?,?> e;
4558	>	return ((o instanceof Map.Entry) &&
4559	>	(k = (e = (Map.Entry<?,?>)o).getKey()) != null &&
4560	>	(r = map.get(k)) != null &&
4561	>	(v = e.getValue()) != null &&
4562	>	(v == r || v.equals(r)));
4563	>	}
4564	>
4565	>	public boolean remove(Object o) {
4566	>	Object k, v; Map.Entry<?,?> e;
4567	>	return ((o instanceof Map.Entry) &&
4568	>	(k = (e = (Map.Entry<?,?>)o).getKey()) != null &&
4569	>	(v = e.getValue()) != null &&
4570	>	map.remove(k, v));
4571		}
4572
4573		/**
4574	<	* Sets our entry's value and writes through to the map. The
1404	<	* value to return is somewhat arbitrary here. Since a
1405	<	* WriteThroughEntry does not necessarily track asynchronous
1406	<	* changes, the most recent "previous" value could be
1407	<	* different from what we return (or could even have been
1408	<	* removed in which case the put will re-establish). We do not
1409	<	* and cannot guarantee more.
4574	>	* @return an iterator over the entries of the backing map
4575		*/
4576	<	public V setValue(V value) {
4577	<	if (value == null) throw new NullPointerException();
4578	<	V v = super.setValue(value);
4579	<	ConcurrentHashMapV8.this.put(getKey(), value);
4580	<	return v;
4576	>	public Iterator<Map.Entry<K,V>> iterator() {
4577	>	ConcurrentHashMapV8<K,V> m = map;
4578	>	Node<K,V>[] t;
4579	>	int f = (t = m.table) == null ? 0 : t.length;
4580	>	return new EntryIterator<K,V>(t, f, 0, f, m);
4581		}
4582	+
4583	+	public boolean add(Entry<K,V> e) {
4584	+	return map.putVal(e.getKey(), e.getValue(), false) == null;
4585	+	}
4586	+
4587	+	public boolean addAll(Collection<? extends Entry<K,V>> c) {
4588	+	boolean added = false;
4589	+	for (Entry<K,V> e : c) {
4590	+	if (add(e))
4591	+	added = true;
4592	+	}
4593	+	return added;
4594	+	}
4595	+
4596	+	public final int hashCode() {
4597	+	int h = 0;
4598	+	Node<K,V>[] t;
4599	+	if ((t = map.table) != null) {
4600	+	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
4601	+	for (Node<K,V> p; (p = it.advance()) != null; ) {
4602	+	h += p.hashCode();
4603	+	}
4604	+	}
4605	+	return h;
4606	+	}
4607	+
4608	+	public final boolean equals(Object o) {
4609	+	Set<?> c;
4610	+	return ((o instanceof Set) &&
4611	+	((c = (Set<?>)o) == this ||
4612	+	(containsAll(c) && c.containsAll(this))));
4613	+	}
4614	+
4615	+	public ConcurrentHashMapSpliterator<Map.Entry<K,V>> spliterator() {
4616	+	Node<K,V>[] t;
4617	+	ConcurrentHashMapV8<K,V> m = map;
4618	+	long n = m.sumCount();
4619	+	int f = (t = m.table) == null ? 0 : t.length;
4620	+	return new EntrySpliterator<K,V>(t, f, 0, f, n < 0L ? 0L : n, m);
4621	+	}
4622	+
4623	+	public void forEach(Action<? super Map.Entry<K,V>> action) {
4624	+	if (action == null) throw new NullPointerException();
4625	+	Node<K,V>[] t;
4626	+	if ((t = map.table) != null) {
4627	+	Traverser<K,V> it = new Traverser<K,V>(t, t.length, 0, t.length);
4628	+	for (Node<K,V> p; (p = it.advance()) != null; )
4629	+	action.apply(new MapEntry<K,V>(p.key, p.val, map));
4630	+	}
4631	+	}
4632	+
4633		}
4634
4635	<	final class KeyIterator extends HashIterator
4636	<	implements Iterator<K>, Enumeration<K> {
4637	<	@SuppressWarnings("unchecked")
4638	<	public final K next()        { return (K)super.nextKey(); }
4639	<	@SuppressWarnings("unchecked")
4640	<	public final K nextElement() { return (K)super.nextKey(); }
4635	>	// -------------------------------------------------------
4636	>
4637	>	/**
4638	>	* Base class for bulk tasks. Repeats some fields and code from
4639	>	* class Traverser, because we need to subclass CountedCompleter.
4640	>	*/
4641	>	abstract static class BulkTask<K,V,R> extends CountedCompleter<R> {
4642	>	Node<K,V>[] tab;        // same as Traverser
4643	>	Node<K,V> next;
4644	>	int index;
4645	>	int baseIndex;
4646	>	int baseLimit;
4647	>	final int baseSize;
4648	>	int batch;              // split control
4649	>
4650	>	BulkTask(BulkTask<K,V,?> par, int b, int i, int f, Node<K,V>[] t) {
4651	>	super(par);
4652	>	this.batch = b;
4653	>	this.index = this.baseIndex = i;
4654	>	if ((this.tab = t) == null)
4655	>	this.baseSize = this.baseLimit = 0;
4656	>	else if (par == null)
4657	>	this.baseSize = this.baseLimit = t.length;
4658	>	else {
4659	>	this.baseLimit = f;
4660	>	this.baseSize = par.baseSize;
4661	>	}
4662	>	}
4663	>
4664	>	/**
4665	>	* Same as Traverser version
4666	>	*/
4667	>	final Node<K,V> advance() {
4668	>	Node<K,V> e;
4669	>	if ((e = next) != null)
4670	>	e = e.next;
4671	>	for (;;) {
4672	>	Node<K,V>[] t; int i, n; K ek;  // must use locals in checks
4673	>	if (e != null)
4674	>	return next = e;
4675	>	if (baseIndex >= baseLimit || (t = tab) == null ||
4676	>	(n = t.length) <= (i = index) || i < 0)
4677	>	return next = null;
4678	>	if ((e = tabAt(t, index)) != null && e.hash < 0) {
4679	>	if (e instanceof ForwardingNode) {
4680	>	tab = ((ForwardingNode<K,V>)e).nextTable;
4681	>	e = null;
4682	>	continue;
4683	>	}
4684	>	else if (e instanceof TreeBin)
4685	>	e = ((TreeBin<K,V>)e).first;
4686	>	else
4687	>	e = null;
4688	>	}
4689	>	if ((index += baseSize) >= n)
4690	>	index = ++baseIndex;    // visit upper slots if present
4691	>	}
4692	>	}
4693		}
4694
4695	<	final class ValueIterator extends HashIterator
4696	<	implements Iterator<V>, Enumeration<V> {
4697	<	@SuppressWarnings("unchecked")
4698	<	public final V next()        { return (V)super.nextValue(); }
4699	<	@SuppressWarnings("unchecked")
4700	<	public final V nextElement() { return (V)super.nextValue(); }
4695	>	/*
4696	>	* Task classes. Coded in a regular but ugly format/style to
4697	>	* simplify checks that each variant differs in the right way from
4698	>	* others. The null screenings exist because compilers cannot tell
4699	>	* that we've already null-checked task arguments, so we force
4700	>	* simplest hoisted bypass to help avoid convoluted traps.
4701	>	*/
4702	>	@SuppressWarnings("serial")
4703	>	static final class ForEachKeyTask<K,V>
4704	>	extends BulkTask<K,V,Void> {
4705	>	final Action<? super K> action;
4706	>	ForEachKeyTask
4707	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4708	>	Action<? super K> action) {
4709	>	super(p, b, i, f, t);
4710	>	this.action = action;
4711	>	}
4712	>	public final void compute() {
4713	>	final Action<? super K> action;
4714	>	if ((action = this.action) != null) {
4715	>	for (int i = baseIndex, f, h; batch > 0 &&
4716	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4717	>	addToPendingCount(1);
4718	>	new ForEachKeyTask<K,V>
4719	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4720	>	action).fork();
4721	>	}
4722	>	for (Node<K,V> p; (p = advance()) != null;)
4723	>	action.apply(p.key);
4724	>	propagateCompletion();
4725	>	}
4726	>	}
4727		}
4728
4729	<	final class EntryIterator extends HashIterator
4730	<	implements Iterator<Entry<K,V>> {
4731	<	public final Map.Entry<K,V> next() { return super.nextEntry(); }
4729	>	@SuppressWarnings("serial")
4730	>	static final class ForEachValueTask<K,V>
4731	>	extends BulkTask<K,V,Void> {
4732	>	final Action<? super V> action;
4733	>	ForEachValueTask
4734	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4735	>	Action<? super V> action) {
4736	>	super(p, b, i, f, t);
4737	>	this.action = action;
4738	>	}
4739	>	public final void compute() {
4740	>	final Action<? super V> action;
4741	>	if ((action = this.action) != null) {
4742	>	for (int i = baseIndex, f, h; batch > 0 &&
4743	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4744	>	addToPendingCount(1);
4745	>	new ForEachValueTask<K,V>
4746	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4747	>	action).fork();
4748	>	}
4749	>	for (Node<K,V> p; (p = advance()) != null;)
4750	>	action.apply(p.val);
4751	>	propagateCompletion();
4752	>	}
4753	>	}
4754		}
4755
4756	<	final class KeySet extends AbstractSet<K> {
4757	<	public int size() {
4758	<	return ConcurrentHashMapV8.this.size();
4756	>	@SuppressWarnings("serial")
4757	>	static final class ForEachEntryTask<K,V>
4758	>	extends BulkTask<K,V,Void> {
4759	>	final Action<? super Entry<K,V>> action;
4760	>	ForEachEntryTask
4761	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4762	>	Action<? super Entry<K,V>> action) {
4763	>	super(p, b, i, f, t);
4764	>	this.action = action;
4765	>	}
4766	>	public final void compute() {
4767	>	final Action<? super Entry<K,V>> action;
4768	>	if ((action = this.action) != null) {
4769	>	for (int i = baseIndex, f, h; batch > 0 &&
4770	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4771	>	addToPendingCount(1);
4772	>	new ForEachEntryTask<K,V>
4773	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4774	>	action).fork();
4775	>	}
4776	>	for (Node<K,V> p; (p = advance()) != null; )
4777	>	action.apply(p);
4778	>	propagateCompletion();
4779	>	}
4780		}
4781	<	public boolean isEmpty() {
4782	<	return ConcurrentHashMapV8.this.isEmpty();
4781	>	}
4782	>
4783	>	@SuppressWarnings("serial")
4784	>	static final class ForEachMappingTask<K,V>
4785	>	extends BulkTask<K,V,Void> {
4786	>	final BiAction<? super K, ? super V> action;
4787	>	ForEachMappingTask
4788	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4789	>	BiAction<? super K,? super V> action) {
4790	>	super(p, b, i, f, t);
4791	>	this.action = action;
4792	>	}
4793	>	public final void compute() {
4794	>	final BiAction<? super K, ? super V> action;
4795	>	if ((action = this.action) != null) {
4796	>	for (int i = baseIndex, f, h; batch > 0 &&
4797	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4798	>	addToPendingCount(1);
4799	>	new ForEachMappingTask<K,V>
4800	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4801	>	action).fork();
4802	>	}
4803	>	for (Node<K,V> p; (p = advance()) != null; )
4804	>	action.apply(p.key, p.val);
4805	>	propagateCompletion();
4806	>	}
4807		}
4808	<	public void clear() {
4809	<	ConcurrentHashMapV8.this.clear();
4808	>	}
4809	>
4810	>	@SuppressWarnings("serial")
4811	>	static final class ForEachTransformedKeyTask<K,V,U>
4812	>	extends BulkTask<K,V,Void> {
4813	>	final Fun<? super K, ? extends U> transformer;
4814	>	final Action<? super U> action;
4815	>	ForEachTransformedKeyTask
4816	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4817	>	Fun<? super K, ? extends U> transformer, Action<? super U> action) {
4818	>	super(p, b, i, f, t);
4819	>	this.transformer = transformer; this.action = action;
4820	>	}
4821	>	public final void compute() {
4822	>	final Fun<? super K, ? extends U> transformer;
4823	>	final Action<? super U> action;
4824	>	if ((transformer = this.transformer) != null &&
4825	>	(action = this.action) != null) {
4826	>	for (int i = baseIndex, f, h; batch > 0 &&
4827	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4828	>	addToPendingCount(1);
4829	>	new ForEachTransformedKeyTask<K,V,U>
4830	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4831	>	transformer, action).fork();
4832	>	}
4833	>	for (Node<K,V> p; (p = advance()) != null; ) {
4834	>	U u;
4835	>	if ((u = transformer.apply(p.key)) != null)
4836	>	action.apply(u);
4837	>	}
4838	>	propagateCompletion();
4839	>	}
4840		}
4841	<	public Iterator<K> iterator() {
4842	<	return new KeyIterator();
4841	>	}
4842	>
4843	>	@SuppressWarnings("serial")
4844	>	static final class ForEachTransformedValueTask<K,V,U>
4845	>	extends BulkTask<K,V,Void> {
4846	>	final Fun<? super V, ? extends U> transformer;
4847	>	final Action<? super U> action;
4848	>	ForEachTransformedValueTask
4849	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4850	>	Fun<? super V, ? extends U> transformer, Action<? super U> action) {
4851	>	super(p, b, i, f, t);
4852	>	this.transformer = transformer; this.action = action;
4853	>	}
4854	>	public final void compute() {
4855	>	final Fun<? super V, ? extends U> transformer;
4856	>	final Action<? super U> action;
4857	>	if ((transformer = this.transformer) != null &&
4858	>	(action = this.action) != null) {
4859	>	for (int i = baseIndex, f, h; batch > 0 &&
4860	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4861	>	addToPendingCount(1);
4862	>	new ForEachTransformedValueTask<K,V,U>
4863	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4864	>	transformer, action).fork();
4865	>	}
4866	>	for (Node<K,V> p; (p = advance()) != null; ) {
4867	>	U u;
4868	>	if ((u = transformer.apply(p.val)) != null)
4869	>	action.apply(u);
4870	>	}
4871	>	propagateCompletion();
4872	>	}
4873		}
4874	<	public boolean contains(Object o) {
4875	<	return ConcurrentHashMapV8.this.containsKey(o);
4874	>	}
4875	>
4876	>	@SuppressWarnings("serial")
4877	>	static final class ForEachTransformedEntryTask<K,V,U>
4878	>	extends BulkTask<K,V,Void> {
4879	>	final Fun<Map.Entry<K,V>, ? extends U> transformer;
4880	>	final Action<? super U> action;
4881	>	ForEachTransformedEntryTask
4882	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4883	>	Fun<Map.Entry<K,V>, ? extends U> transformer, Action<? super U> action) {
4884	>	super(p, b, i, f, t);
4885	>	this.transformer = transformer; this.action = action;
4886	>	}
4887	>	public final void compute() {
4888	>	final Fun<Map.Entry<K,V>, ? extends U> transformer;
4889	>	final Action<? super U> action;
4890	>	if ((transformer = this.transformer) != null &&
4891	>	(action = this.action) != null) {
4892	>	for (int i = baseIndex, f, h; batch > 0 &&
4893	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4894	>	addToPendingCount(1);
4895	>	new ForEachTransformedEntryTask<K,V,U>
4896	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4897	>	transformer, action).fork();
4898	>	}
4899	>	for (Node<K,V> p; (p = advance()) != null; ) {
4900	>	U u;
4901	>	if ((u = transformer.apply(p)) != null)
4902	>	action.apply(u);
4903	>	}
4904	>	propagateCompletion();
4905	>	}
4906		}
4907	<	public boolean remove(Object o) {
4908	<	return ConcurrentHashMapV8.this.remove(o) != null;
4907	>	}
4908	>
4909	>	@SuppressWarnings("serial")
4910	>	static final class ForEachTransformedMappingTask<K,V,U>
4911	>	extends BulkTask<K,V,Void> {
4912	>	final BiFun<? super K, ? super V, ? extends U> transformer;
4913	>	final Action<? super U> action;
4914	>	ForEachTransformedMappingTask
4915	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4916	>	BiFun<? super K, ? super V, ? extends U> transformer,
4917	>	Action<? super U> action) {
4918	>	super(p, b, i, f, t);
4919	>	this.transformer = transformer; this.action = action;
4920	>	}
4921	>	public final void compute() {
4922	>	final BiFun<? super K, ? super V, ? extends U> transformer;
4923	>	final Action<? super U> action;
4924	>	if ((transformer = this.transformer) != null &&
4925	>	(action = this.action) != null) {
4926	>	for (int i = baseIndex, f, h; batch > 0 &&
4927	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4928	>	addToPendingCount(1);
4929	>	new ForEachTransformedMappingTask<K,V,U>
4930	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4931	>	transformer, action).fork();
4932	>	}
4933	>	for (Node<K,V> p; (p = advance()) != null; ) {
4934	>	U u;
4935	>	if ((u = transformer.apply(p.key, p.val)) != null)
4936	>	action.apply(u);
4937	>	}
4938	>	propagateCompletion();
4939	>	}
4940	>	}
4941	>	}
4942	>
4943	>	@SuppressWarnings("serial")
4944	>	static final class SearchKeysTask<K,V,U>
4945	>	extends BulkTask<K,V,U> {
4946	>	final Fun<? super K, ? extends U> searchFunction;
4947	>	final AtomicReference<U> result;
4948	>	SearchKeysTask
4949	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4950	>	Fun<? super K, ? extends U> searchFunction,
4951	>	AtomicReference<U> result) {
4952	>	super(p, b, i, f, t);
4953	>	this.searchFunction = searchFunction; this.result = result;
4954	>	}
4955	>	public final U getRawResult() { return result.get(); }
4956	>	public final void compute() {
4957	>	final Fun<? super K, ? extends U> searchFunction;
4958	>	final AtomicReference<U> result;
4959	>	if ((searchFunction = this.searchFunction) != null &&
4960	>	(result = this.result) != null) {
4961	>	for (int i = baseIndex, f, h; batch > 0 &&
4962	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
4963	>	if (result.get() != null)
4964	>	return;
4965	>	addToPendingCount(1);
4966	>	new SearchKeysTask<K,V,U>
4967	>	(this, batch >>>= 1, baseLimit = h, f, tab,
4968	>	searchFunction, result).fork();
4969	>	}
4970	>	while (result.get() == null) {
4971	>	U u;
4972	>	Node<K,V> p;
4973	>	if ((p = advance()) == null) {
4974	>	propagateCompletion();
4975	>	break;
4976	>	}
4977	>	if ((u = searchFunction.apply(p.key)) != null) {
4978	>	if (result.compareAndSet(null, u))
4979	>	quietlyCompleteRoot();
4980	>	break;
4981	>	}
4982	>	}
4983	>	}
4984	>	}
4985	>	}
4986	>
4987	>	@SuppressWarnings("serial")
4988	>	static final class SearchValuesTask<K,V,U>
4989	>	extends BulkTask<K,V,U> {
4990	>	final Fun<? super V, ? extends U> searchFunction;
4991	>	final AtomicReference<U> result;
4992	>	SearchValuesTask
4993	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
4994	>	Fun<? super V, ? extends U> searchFunction,
4995	>	AtomicReference<U> result) {
4996	>	super(p, b, i, f, t);
4997	>	this.searchFunction = searchFunction; this.result = result;
4998	>	}
4999	>	public final U getRawResult() { return result.get(); }
5000	>	public final void compute() {
5001	>	final Fun<? super V, ? extends U> searchFunction;
5002	>	final AtomicReference<U> result;
5003	>	if ((searchFunction = this.searchFunction) != null &&
5004	>	(result = this.result) != null) {
5005	>	for (int i = baseIndex, f, h; batch > 0 &&
5006	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5007	>	if (result.get() != null)
5008	>	return;
5009	>	addToPendingCount(1);
5010	>	new SearchValuesTask<K,V,U>
5011	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5012	>	searchFunction, result).fork();
5013	>	}
5014	>	while (result.get() == null) {
5015	>	U u;
5016	>	Node<K,V> p;
5017	>	if ((p = advance()) == null) {
5018	>	propagateCompletion();
5019	>	break;
5020	>	}
5021	>	if ((u = searchFunction.apply(p.val)) != null) {
5022	>	if (result.compareAndSet(null, u))
5023	>	quietlyCompleteRoot();
5024	>	break;
5025	>	}
5026	>	}
5027	>	}
5028		}
5029		}
5030
5031	<	final class Values extends AbstractCollection<V> {
5032	<	public int size() {
5033	<	return ConcurrentHashMapV8.this.size();
5031	>	@SuppressWarnings("serial")
5032	>	static final class SearchEntriesTask<K,V,U>
5033	>	extends BulkTask<K,V,U> {
5034	>	final Fun<Entry<K,V>, ? extends U> searchFunction;
5035	>	final AtomicReference<U> result;
5036	>	SearchEntriesTask
5037	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5038	>	Fun<Entry<K,V>, ? extends U> searchFunction,
5039	>	AtomicReference<U> result) {
5040	>	super(p, b, i, f, t);
5041	>	this.searchFunction = searchFunction; this.result = result;
5042	>	}
5043	>	public final U getRawResult() { return result.get(); }
5044	>	public final void compute() {
5045	>	final Fun<Entry<K,V>, ? extends U> searchFunction;
5046	>	final AtomicReference<U> result;
5047	>	if ((searchFunction = this.searchFunction) != null &&
5048	>	(result = this.result) != null) {
5049	>	for (int i = baseIndex, f, h; batch > 0 &&
5050	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5051	>	if (result.get() != null)
5052	>	return;
5053	>	addToPendingCount(1);
5054	>	new SearchEntriesTask<K,V,U>
5055	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5056	>	searchFunction, result).fork();
5057	>	}
5058	>	while (result.get() == null) {
5059	>	U u;
5060	>	Node<K,V> p;
5061	>	if ((p = advance()) == null) {
5062	>	propagateCompletion();
5063	>	break;
5064	>	}
5065	>	if ((u = searchFunction.apply(p)) != null) {
5066	>	if (result.compareAndSet(null, u))
5067	>	quietlyCompleteRoot();
5068	>	return;
5069	>	}
5070	>	}
5071	>	}
5072		}
5073	<	public boolean isEmpty() {
5074	<	return ConcurrentHashMapV8.this.isEmpty();
5073	>	}
5074	>
5075	>	@SuppressWarnings("serial")
5076	>	static final class SearchMappingsTask<K,V,U>
5077	>	extends BulkTask<K,V,U> {
5078	>	final BiFun<? super K, ? super V, ? extends U> searchFunction;
5079	>	final AtomicReference<U> result;
5080	>	SearchMappingsTask
5081	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5082	>	BiFun<? super K, ? super V, ? extends U> searchFunction,
5083	>	AtomicReference<U> result) {
5084	>	super(p, b, i, f, t);
5085	>	this.searchFunction = searchFunction; this.result = result;
5086	>	}
5087	>	public final U getRawResult() { return result.get(); }
5088	>	public final void compute() {
5089	>	final BiFun<? super K, ? super V, ? extends U> searchFunction;
5090	>	final AtomicReference<U> result;
5091	>	if ((searchFunction = this.searchFunction) != null &&
5092	>	(result = this.result) != null) {
5093	>	for (int i = baseIndex, f, h; batch > 0 &&
5094	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5095	>	if (result.get() != null)
5096	>	return;
5097	>	addToPendingCount(1);
5098	>	new SearchMappingsTask<K,V,U>
5099	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5100	>	searchFunction, result).fork();
5101	>	}
5102	>	while (result.get() == null) {
5103	>	U u;
5104	>	Node<K,V> p;
5105	>	if ((p = advance()) == null) {
5106	>	propagateCompletion();
5107	>	break;
5108	>	}
5109	>	if ((u = searchFunction.apply(p.key, p.val)) != null) {
5110	>	if (result.compareAndSet(null, u))
5111	>	quietlyCompleteRoot();
5112	>	break;
5113	>	}
5114	>	}
5115	>	}
5116		}
5117	<	public void clear() {
5118	<	ConcurrentHashMapV8.this.clear();
5117	>	}
5118	>
5119	>	@SuppressWarnings("serial")
5120	>	static final class ReduceKeysTask<K,V>
5121	>	extends BulkTask<K,V,K> {
5122	>	final BiFun<? super K, ? super K, ? extends K> reducer;
5123	>	K result;
5124	>	ReduceKeysTask<K,V> rights, nextRight;
5125	>	ReduceKeysTask
5126	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5127	>	ReduceKeysTask<K,V> nextRight,
5128	>	BiFun<? super K, ? super K, ? extends K> reducer) {
5129	>	super(p, b, i, f, t); this.nextRight = nextRight;
5130	>	this.reducer = reducer;
5131	>	}
5132	>	public final K getRawResult() { return result; }
5133	>	public final void compute() {
5134	>	final BiFun<? super K, ? super K, ? extends K> reducer;
5135	>	if ((reducer = this.reducer) != null) {
5136	>	for (int i = baseIndex, f, h; batch > 0 &&
5137	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5138	>	addToPendingCount(1);
5139	>	(rights = new ReduceKeysTask<K,V>
5140	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5141	>	rights, reducer)).fork();
5142	>	}
5143	>	K r = null;
5144	>	for (Node<K,V> p; (p = advance()) != null; ) {
5145	>	K u = p.key;
5146	>	r = (r == null) ? u : u == null ? r : reducer.apply(r, u);
5147	>	}
5148	>	result = r;
5149	>	CountedCompleter<?> c;
5150	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5151	>	@SuppressWarnings("unchecked") ReduceKeysTask<K,V>
5152	>	t = (ReduceKeysTask<K,V>)c,
5153	>	s = t.rights;
5154	>	while (s != null) {
5155	>	K 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	<	public Iterator<V> iterator() {
5165	<	return new ValueIterator();
5164	>	}
5165	>
5166	>	@SuppressWarnings("serial")
5167	>	static final class ReduceValuesTask<K,V>
5168	>	extends BulkTask<K,V,V> {
5169	>	final BiFun<? super V, ? super V, ? extends V> reducer;
5170	>	V result;
5171	>	ReduceValuesTask<K,V> rights, nextRight;
5172	>	ReduceValuesTask
5173	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5174	>	ReduceValuesTask<K,V> nextRight,
5175	>	BiFun<? super V, ? super V, ? extends V> reducer) {
5176	>	super(p, b, i, f, t); this.nextRight = nextRight;
5177	>	this.reducer = reducer;
5178	>	}
5179	>	public final V getRawResult() { return result; }
5180	>	public final void compute() {
5181	>	final BiFun<? super V, ? super V, ? extends V> reducer;
5182	>	if ((reducer = this.reducer) != null) {
5183	>	for (int i = baseIndex, f, h; batch > 0 &&
5184	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5185	>	addToPendingCount(1);
5186	>	(rights = new ReduceValuesTask<K,V>
5187	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5188	>	rights, reducer)).fork();
5189	>	}
5190	>	V r = null;
5191	>	for (Node<K,V> p; (p = advance()) != null; ) {
5192	>	V v = p.val;
5193	>	r = (r == null) ? v : reducer.apply(r, v);
5194	>	}
5195	>	result = r;
5196	>	CountedCompleter<?> c;
5197	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5198	>	@SuppressWarnings("unchecked") ReduceValuesTask<K,V>
5199	>	t = (ReduceValuesTask<K,V>)c,
5200	>	s = t.rights;
5201	>	while (s != null) {
5202	>	V tr, sr;
5203	>	if ((sr = s.result) != null)
5204	>	t.result = (((tr = t.result) == null) ? sr :
5205	>	reducer.apply(tr, sr));
5206	>	s = t.rights = s.nextRight;
5207	>	}
5208	>	}
5209	>	}
5210		}
5211	<	public boolean contains(Object o) {
5212	<	return ConcurrentHashMapV8.this.containsValue(o);
5211	>	}
5212	>
5213	>	@SuppressWarnings("serial")
5214	>	static final class ReduceEntriesTask<K,V>
5215	>	extends BulkTask<K,V,Map.Entry<K,V>> {
5216	>	final BiFun<Map.Entry<K,V>, Map.Entry<K,V>, ? extends Map.Entry<K,V>> reducer;
5217	>	Map.Entry<K,V> result;
5218	>	ReduceEntriesTask<K,V> rights, nextRight;
5219	>	ReduceEntriesTask
5220	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5221	>	ReduceEntriesTask<K,V> nextRight,
5222	>	BiFun<Entry<K,V>, Map.Entry<K,V>, ? extends Map.Entry<K,V>> reducer) {
5223	>	super(p, b, i, f, t); this.nextRight = nextRight;
5224	>	this.reducer = reducer;
5225	>	}
5226	>	public final Map.Entry<K,V> getRawResult() { return result; }
5227	>	public final void compute() {
5228	>	final BiFun<Map.Entry<K,V>, Map.Entry<K,V>, ? extends Map.Entry<K,V>> reducer;
5229	>	if ((reducer = this.reducer) != null) {
5230	>	for (int i = baseIndex, f, h; batch > 0 &&
5231	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5232	>	addToPendingCount(1);
5233	>	(rights = new ReduceEntriesTask<K,V>
5234	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5235	>	rights, reducer)).fork();
5236	>	}
5237	>	Map.Entry<K,V> r = null;
5238	>	for (Node<K,V> p; (p = advance()) != null; )
5239	>	r = (r == null) ? p : reducer.apply(r, p);
5240	>	result = r;
5241	>	CountedCompleter<?> c;
5242	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5243	>	@SuppressWarnings("unchecked") ReduceEntriesTask<K,V>
5244	>	t = (ReduceEntriesTask<K,V>)c,
5245	>	s = t.rights;
5246	>	while (s != null) {
5247	>	Map.Entry<K,V> tr, sr;
5248	>	if ((sr = s.result) != null)
5249	>	t.result = (((tr = t.result) == null) ? sr :
5250	>	reducer.apply(tr, sr));
5251	>	s = t.rights = s.nextRight;
5252	>	}
5253	>	}
5254	>	}
5255		}
5256		}
5257
5258	<	final class EntrySet extends AbstractSet<Map.Entry<K,V>> {
5259	<	public int size() {
5260	<	return ConcurrentHashMapV8.this.size();
5258	>	@SuppressWarnings("serial")
5259	>	static final class MapReduceKeysTask<K,V,U>
5260	>	extends BulkTask<K,V,U> {
5261	>	final Fun<? super K, ? extends U> transformer;
5262	>	final BiFun<? super U, ? super U, ? extends U> reducer;
5263	>	U result;
5264	>	MapReduceKeysTask<K,V,U> rights, nextRight;
5265	>	MapReduceKeysTask
5266	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5267	>	MapReduceKeysTask<K,V,U> nextRight,
5268	>	Fun<? super K, ? extends U> transformer,
5269	>	BiFun<? super U, ? super U, ? extends U> reducer) {
5270	>	super(p, b, i, f, t); this.nextRight = nextRight;
5271	>	this.transformer = transformer;
5272	>	this.reducer = reducer;
5273	>	}
5274	>	public final U getRawResult() { return result; }
5275	>	public final void compute() {
5276	>	final Fun<? super K, ? extends U> transformer;
5277	>	final BiFun<? super U, ? super U, ? extends U> reducer;
5278	>	if ((transformer = this.transformer) != null &&
5279	>	(reducer = this.reducer) != null) {
5280	>	for (int i = baseIndex, f, h; batch > 0 &&
5281	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5282	>	addToPendingCount(1);
5283	>	(rights = new MapReduceKeysTask<K,V,U>
5284	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5285	>	rights, transformer, reducer)).fork();
5286	>	}
5287	>	U r = null;
5288	>	for (Node<K,V> p; (p = advance()) != null; ) {
5289	>	U u;
5290	>	if ((u = transformer.apply(p.key)) != null)
5291	>	r = (r == null) ? u : reducer.apply(r, u);
5292	>	}
5293	>	result = r;
5294	>	CountedCompleter<?> c;
5295	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5296	>	@SuppressWarnings("unchecked") MapReduceKeysTask<K,V,U>
5297	>	t = (MapReduceKeysTask<K,V,U>)c,
5298	>	s = t.rights;
5299	>	while (s != null) {
5300	>	U tr, sr;
5301	>	if ((sr = s.result) != null)
5302	>	t.result = (((tr = t.result) == null) ? sr :
5303	>	reducer.apply(tr, sr));
5304	>	s = t.rights = s.nextRight;
5305	>	}
5306	>	}
5307	>	}
5308		}
5309	<	public boolean isEmpty() {
5310	<	return ConcurrentHashMapV8.this.isEmpty();
5309	>	}
5310	>
5311	>	@SuppressWarnings("serial")
5312	>	static final class MapReduceValuesTask<K,V,U>
5313	>	extends BulkTask<K,V,U> {
5314	>	final Fun<? super V, ? extends U> transformer;
5315	>	final BiFun<? super U, ? super U, ? extends U> reducer;
5316	>	U result;
5317	>	MapReduceValuesTask<K,V,U> rights, nextRight;
5318	>	MapReduceValuesTask
5319	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5320	>	MapReduceValuesTask<K,V,U> nextRight,
5321	>	Fun<? super V, ? extends U> transformer,
5322	>	BiFun<? super U, ? super U, ? extends U> reducer) {
5323	>	super(p, b, i, f, t); this.nextRight = nextRight;
5324	>	this.transformer = transformer;
5325	>	this.reducer = reducer;
5326	>	}
5327	>	public final U getRawResult() { return result; }
5328	>	public final void compute() {
5329	>	final Fun<? super V, ? extends U> transformer;
5330	>	final BiFun<? super U, ? super U, ? extends U> reducer;
5331	>	if ((transformer = this.transformer) != null &&
5332	>	(reducer = this.reducer) != null) {
5333	>	for (int i = baseIndex, f, h; batch > 0 &&
5334	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5335	>	addToPendingCount(1);
5336	>	(rights = new MapReduceValuesTask<K,V,U>
5337	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5338	>	rights, transformer, reducer)).fork();
5339	>	}
5340	>	U r = null;
5341	>	for (Node<K,V> p; (p = advance()) != null; ) {
5342	>	U u;
5343	>	if ((u = transformer.apply(p.val)) != null)
5344	>	r = (r == null) ? u : reducer.apply(r, u);
5345	>	}
5346	>	result = r;
5347	>	CountedCompleter<?> c;
5348	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5349	>	@SuppressWarnings("unchecked") MapReduceValuesTask<K,V,U>
5350	>	t = (MapReduceValuesTask<K,V,U>)c,
5351	>	s = t.rights;
5352	>	while (s != null) {
5353	>	U tr, sr;
5354	>	if ((sr = s.result) != null)
5355	>	t.result = (((tr = t.result) == null) ? sr :
5356	>	reducer.apply(tr, sr));
5357	>	s = t.rights = s.nextRight;
5358	>	}
5359	>	}
5360	>	}
5361		}
5362	<	public void clear() {
5363	<	ConcurrentHashMapV8.this.clear();
5362	>	}
5363	>
5364	>	@SuppressWarnings("serial")
5365	>	static final class MapReduceEntriesTask<K,V,U>
5366	>	extends BulkTask<K,V,U> {
5367	>	final Fun<Map.Entry<K,V>, ? extends U> transformer;
5368	>	final BiFun<? super U, ? super U, ? extends U> reducer;
5369	>	U result;
5370	>	MapReduceEntriesTask<K,V,U> rights, nextRight;
5371	>	MapReduceEntriesTask
5372	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5373	>	MapReduceEntriesTask<K,V,U> nextRight,
5374	>	Fun<Map.Entry<K,V>, ? extends U> transformer,
5375	>	BiFun<? super U, ? super U, ? extends U> reducer) {
5376	>	super(p, b, i, f, t); this.nextRight = nextRight;
5377	>	this.transformer = transformer;
5378	>	this.reducer = reducer;
5379	>	}
5380	>	public final U getRawResult() { return result; }
5381	>	public final void compute() {
5382	>	final Fun<Map.Entry<K,V>, ? extends U> transformer;
5383	>	final BiFun<? super U, ? super U, ? extends U> reducer;
5384	>	if ((transformer = this.transformer) != null &&
5385	>	(reducer = this.reducer) != null) {
5386	>	for (int i = baseIndex, f, h; batch > 0 &&
5387	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5388	>	addToPendingCount(1);
5389	>	(rights = new MapReduceEntriesTask<K,V,U>
5390	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5391	>	rights, transformer, reducer)).fork();
5392	>	}
5393	>	U r = null;
5394	>	for (Node<K,V> p; (p = advance()) != null; ) {
5395	>	U u;
5396	>	if ((u = transformer.apply(p)) != null)
5397	>	r = (r == null) ? u : reducer.apply(r, u);
5398	>	}
5399	>	result = r;
5400	>	CountedCompleter<?> c;
5401	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5402	>	@SuppressWarnings("unchecked") MapReduceEntriesTask<K,V,U>
5403	>	t = (MapReduceEntriesTask<K,V,U>)c,
5404	>	s = t.rights;
5405	>	while (s != null) {
5406	>	U tr, sr;
5407	>	if ((sr = s.result) != null)
5408	>	t.result = (((tr = t.result) == null) ? sr :
5409	>	reducer.apply(tr, sr));
5410	>	s = t.rights = s.nextRight;
5411	>	}
5412	>	}
5413	>	}
5414		}
5415	<	public Iterator<Map.Entry<K,V>> iterator() {
5416	<	return new EntryIterator();
5415	>	}
5416	>
5417	>	@SuppressWarnings("serial")
5418	>	static final class MapReduceMappingsTask<K,V,U>
5419	>	extends BulkTask<K,V,U> {
5420	>	final BiFun<? super K, ? super V, ? extends U> transformer;
5421	>	final BiFun<? super U, ? super U, ? extends U> reducer;
5422	>	U result;
5423	>	MapReduceMappingsTask<K,V,U> rights, nextRight;
5424	>	MapReduceMappingsTask
5425	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5426	>	MapReduceMappingsTask<K,V,U> nextRight,
5427	>	BiFun<? super K, ? super V, ? extends U> transformer,
5428	>	BiFun<? super U, ? super U, ? extends U> reducer) {
5429	>	super(p, b, i, f, t); this.nextRight = nextRight;
5430	>	this.transformer = transformer;
5431	>	this.reducer = reducer;
5432	>	}
5433	>	public final U getRawResult() { return result; }
5434	>	public final void compute() {
5435	>	final BiFun<? super K, ? super V, ? extends U> transformer;
5436	>	final BiFun<? super U, ? super U, ? extends U> reducer;
5437	>	if ((transformer = this.transformer) != null &&
5438	>	(reducer = this.reducer) != null) {
5439	>	for (int i = baseIndex, f, h; batch > 0 &&
5440	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5441	>	addToPendingCount(1);
5442	>	(rights = new MapReduceMappingsTask<K,V,U>
5443	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5444	>	rights, transformer, reducer)).fork();
5445	>	}
5446	>	U r = null;
5447	>	for (Node<K,V> p; (p = advance()) != null; ) {
5448	>	U u;
5449	>	if ((u = transformer.apply(p.key, p.val)) != null)
5450	>	r = (r == null) ? u : reducer.apply(r, u);
5451	>	}
5452	>	result = r;
5453	>	CountedCompleter<?> c;
5454	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5455	>	@SuppressWarnings("unchecked") MapReduceMappingsTask<K,V,U>
5456	>	t = (MapReduceMappingsTask<K,V,U>)c,
5457	>	s = t.rights;
5458	>	while (s != null) {
5459	>	U tr, sr;
5460	>	if ((sr = s.result) != null)
5461	>	t.result = (((tr = t.result) == null) ? sr :
5462	>	reducer.apply(tr, sr));
5463	>	s = t.rights = s.nextRight;
5464	>	}
5465	>	}
5466	>	}
5467		}
5468	<	public boolean contains(Object o) {
5469	<	if (!(o instanceof Map.Entry))
5470	<	return false;
5471	<	Map.Entry<?,?> e = (Map.Entry<?,?>)o;
5472	<	V v = ConcurrentHashMapV8.this.get(e.getKey());
5473	<	return v != null && v.equals(e.getValue());
5468	>	}
5469	>
5470	>	@SuppressWarnings("serial")
5471	>	static final class MapReduceKeysToDoubleTask<K,V>
5472	>	extends BulkTask<K,V,Double> {
5473	>	final ObjectToDouble<? super K> transformer;
5474	>	final DoubleByDoubleToDouble reducer;
5475	>	final double basis;
5476	>	double result;
5477	>	MapReduceKeysToDoubleTask<K,V> rights, nextRight;
5478	>	MapReduceKeysToDoubleTask
5479	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5480	>	MapReduceKeysToDoubleTask<K,V> nextRight,
5481	>	ObjectToDouble<? super K> transformer,
5482	>	double basis,
5483	>	DoubleByDoubleToDouble reducer) {
5484	>	super(p, b, i, f, t); this.nextRight = nextRight;
5485	>	this.transformer = transformer;
5486	>	this.basis = basis; this.reducer = reducer;
5487	>	}
5488	>	public final Double getRawResult() { return result; }
5489	>	public final void compute() {
5490	>	final ObjectToDouble<? super K> transformer;
5491	>	final DoubleByDoubleToDouble reducer;
5492	>	if ((transformer = this.transformer) != null &&
5493	>	(reducer = this.reducer) != null) {
5494	>	double r = this.basis;
5495	>	for (int i = baseIndex, f, h; batch > 0 &&
5496	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5497	>	addToPendingCount(1);
5498	>	(rights = new MapReduceKeysToDoubleTask<K,V>
5499	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5500	>	rights, transformer, r, reducer)).fork();
5501	>	}
5502	>	for (Node<K,V> p; (p = advance()) != null; )
5503	>	r = reducer.apply(r, transformer.apply(p.key));
5504	>	result = r;
5505	>	CountedCompleter<?> c;
5506	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5507	>	@SuppressWarnings("unchecked") MapReduceKeysToDoubleTask<K,V>
5508	>	t = (MapReduceKeysToDoubleTask<K,V>)c,
5509	>	s = t.rights;
5510	>	while (s != null) {
5511	>	t.result = reducer.apply(t.result, s.result);
5512	>	s = t.rights = s.nextRight;
5513	>	}
5514	>	}
5515	>	}
5516		}
5517	<	public boolean remove(Object o) {
5518	<	if (!(o instanceof Map.Entry))
5519	<	return false;
5520	<	Map.Entry<?,?> e = (Map.Entry<?,?>)o;
5521	<	return ConcurrentHashMapV8.this.remove(e.getKey(), e.getValue());
5517	>	}
5518	>
5519	>	@SuppressWarnings("serial")
5520	>	static final class MapReduceValuesToDoubleTask<K,V>
5521	>	extends BulkTask<K,V,Double> {
5522	>	final ObjectToDouble<? super V> transformer;
5523	>	final DoubleByDoubleToDouble reducer;
5524	>	final double basis;
5525	>	double result;
5526	>	MapReduceValuesToDoubleTask<K,V> rights, nextRight;
5527	>	MapReduceValuesToDoubleTask
5528	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5529	>	MapReduceValuesToDoubleTask<K,V> nextRight,
5530	>	ObjectToDouble<? super V> transformer,
5531	>	double basis,
5532	>	DoubleByDoubleToDouble reducer) {
5533	>	super(p, b, i, f, t); this.nextRight = nextRight;
5534	>	this.transformer = transformer;
5535	>	this.basis = basis; this.reducer = reducer;
5536	>	}
5537	>	public final Double getRawResult() { return result; }
5538	>	public final void compute() {
5539	>	final ObjectToDouble<? super V> transformer;
5540	>	final DoubleByDoubleToDouble reducer;
5541	>	if ((transformer = this.transformer) != null &&
5542	>	(reducer = this.reducer) != null) {
5543	>	double r = this.basis;
5544	>	for (int i = baseIndex, f, h; batch > 0 &&
5545	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5546	>	addToPendingCount(1);
5547	>	(rights = new MapReduceValuesToDoubleTask<K,V>
5548	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5549	>	rights, transformer, r, reducer)).fork();
5550	>	}
5551	>	for (Node<K,V> p; (p = advance()) != null; )
5552	>	r = reducer.apply(r, transformer.apply(p.val));
5553	>	result = r;
5554	>	CountedCompleter<?> c;
5555	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5556	>	@SuppressWarnings("unchecked") MapReduceValuesToDoubleTask<K,V>
5557	>	t = (MapReduceValuesToDoubleTask<K,V>)c,
5558	>	s = t.rights;
5559	>	while (s != null) {
5560	>	t.result = reducer.apply(t.result, s.result);
5561	>	s = t.rights = s.nextRight;
5562	>	}
5563	>	}
5564	>	}
5565	>	}
5566	>	}
5567	>
5568	>	@SuppressWarnings("serial")
5569	>	static final class MapReduceEntriesToDoubleTask<K,V>
5570	>	extends BulkTask<K,V,Double> {
5571	>	final ObjectToDouble<Map.Entry<K,V>> transformer;
5572	>	final DoubleByDoubleToDouble reducer;
5573	>	final double basis;
5574	>	double result;
5575	>	MapReduceEntriesToDoubleTask<K,V> rights, nextRight;
5576	>	MapReduceEntriesToDoubleTask
5577	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5578	>	MapReduceEntriesToDoubleTask<K,V> nextRight,
5579	>	ObjectToDouble<Map.Entry<K,V>> transformer,
5580	>	double basis,
5581	>	DoubleByDoubleToDouble reducer) {
5582	>	super(p, b, i, f, t); this.nextRight = nextRight;
5583	>	this.transformer = transformer;
5584	>	this.basis = basis; this.reducer = reducer;
5585	>	}
5586	>	public final Double getRawResult() { return result; }
5587	>	public final void compute() {
5588	>	final ObjectToDouble<Map.Entry<K,V>> transformer;
5589	>	final DoubleByDoubleToDouble reducer;
5590	>	if ((transformer = this.transformer) != null &&
5591	>	(reducer = this.reducer) != null) {
5592	>	double r = this.basis;
5593	>	for (int i = baseIndex, f, h; batch > 0 &&
5594	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5595	>	addToPendingCount(1);
5596	>	(rights = new MapReduceEntriesToDoubleTask<K,V>
5597	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5598	>	rights, transformer, r, reducer)).fork();
5599	>	}
5600	>	for (Node<K,V> p; (p = advance()) != null; )
5601	>	r = reducer.apply(r, transformer.apply(p));
5602	>	result = r;
5603	>	CountedCompleter<?> c;
5604	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5605	>	@SuppressWarnings("unchecked") MapReduceEntriesToDoubleTask<K,V>
5606	>	t = (MapReduceEntriesToDoubleTask<K,V>)c,
5607	>	s = t.rights;
5608	>	while (s != null) {
5609	>	t.result = reducer.apply(t.result, s.result);
5610	>	s = t.rights = s.nextRight;
5611	>	}
5612	>	}
5613	>	}
5614	>	}
5615	>	}
5616	>
5617	>	@SuppressWarnings("serial")
5618	>	static final class MapReduceMappingsToDoubleTask<K,V>
5619	>	extends BulkTask<K,V,Double> {
5620	>	final ObjectByObjectToDouble<? super K, ? super V> transformer;
5621	>	final DoubleByDoubleToDouble reducer;
5622	>	final double basis;
5623	>	double result;
5624	>	MapReduceMappingsToDoubleTask<K,V> rights, nextRight;
5625	>	MapReduceMappingsToDoubleTask
5626	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5627	>	MapReduceMappingsToDoubleTask<K,V> nextRight,
5628	>	ObjectByObjectToDouble<? super K, ? super V> transformer,
5629	>	double basis,
5630	>	DoubleByDoubleToDouble reducer) {
5631	>	super(p, b, i, f, t); this.nextRight = nextRight;
5632	>	this.transformer = transformer;
5633	>	this.basis = basis; this.reducer = reducer;
5634	>	}
5635	>	public final Double getRawResult() { return result; }
5636	>	public final void compute() {
5637	>	final ObjectByObjectToDouble<? super K, ? super V> transformer;
5638	>	final DoubleByDoubleToDouble reducer;
5639	>	if ((transformer = this.transformer) != null &&
5640	>	(reducer = this.reducer) != null) {
5641	>	double r = this.basis;
5642	>	for (int i = baseIndex, f, h; batch > 0 &&
5643	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5644	>	addToPendingCount(1);
5645	>	(rights = new MapReduceMappingsToDoubleTask<K,V>
5646	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5647	>	rights, transformer, r, reducer)).fork();
5648	>	}
5649	>	for (Node<K,V> p; (p = advance()) != null; )
5650	>	r = reducer.apply(r, transformer.apply(p.key, p.val));
5651	>	result = r;
5652	>	CountedCompleter<?> c;
5653	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5654	>	@SuppressWarnings("unchecked") MapReduceMappingsToDoubleTask<K,V>
5655	>	t = (MapReduceMappingsToDoubleTask<K,V>)c,
5656	>	s = t.rights;
5657	>	while (s != null) {
5658	>	t.result = reducer.apply(t.result, s.result);
5659	>	s = t.rights = s.nextRight;
5660	>	}
5661	>	}
5662	>	}
5663		}
5664		}
5665
5666	<	/* ---------------- Serialization Support -------------- */
5666	>	@SuppressWarnings("serial")
5667	>	static final class MapReduceKeysToLongTask<K,V>
5668	>	extends BulkTask<K,V,Long> {
5669	>	final ObjectToLong<? super K> transformer;
5670	>	final LongByLongToLong reducer;
5671	>	final long basis;
5672	>	long result;
5673	>	MapReduceKeysToLongTask<K,V> rights, nextRight;
5674	>	MapReduceKeysToLongTask
5675	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5676	>	MapReduceKeysToLongTask<K,V> nextRight,
5677	>	ObjectToLong<? super K> transformer,
5678	>	long basis,
5679	>	LongByLongToLong reducer) {
5680	>	super(p, b, i, f, t); this.nextRight = nextRight;
5681	>	this.transformer = transformer;
5682	>	this.basis = basis; this.reducer = reducer;
5683	>	}
5684	>	public final Long getRawResult() { return result; }
5685	>	public final void compute() {
5686	>	final ObjectToLong<? super K> transformer;
5687	>	final LongByLongToLong reducer;
5688	>	if ((transformer = this.transformer) != null &&
5689	>	(reducer = this.reducer) != null) {
5690	>	long r = this.basis;
5691	>	for (int i = baseIndex, f, h; batch > 0 &&
5692	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5693	>	addToPendingCount(1);
5694	>	(rights = new MapReduceKeysToLongTask<K,V>
5695	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5696	>	rights, transformer, r, reducer)).fork();
5697	>	}
5698	>	for (Node<K,V> p; (p = advance()) != null; )
5699	>	r = reducer.apply(r, transformer.apply(p.key));
5700	>	result = r;
5701	>	CountedCompleter<?> c;
5702	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5703	>	@SuppressWarnings("unchecked") MapReduceKeysToLongTask<K,V>
5704	>	t = (MapReduceKeysToLongTask<K,V>)c,
5705	>	s = t.rights;
5706	>	while (s != null) {
5707	>	t.result = reducer.apply(t.result, s.result);
5708	>	s = t.rights = s.nextRight;
5709	>	}
5710	>	}
5711	>	}
5712	>	}
5713	>	}
5714	>
5715	>	@SuppressWarnings("serial")
5716	>	static final class MapReduceValuesToLongTask<K,V>
5717	>	extends BulkTask<K,V,Long> {
5718	>	final ObjectToLong<? super V> transformer;
5719	>	final LongByLongToLong reducer;
5720	>	final long basis;
5721	>	long result;
5722	>	MapReduceValuesToLongTask<K,V> rights, nextRight;
5723	>	MapReduceValuesToLongTask
5724	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5725	>	MapReduceValuesToLongTask<K,V> nextRight,
5726	>	ObjectToLong<? super V> transformer,
5727	>	long basis,
5728	>	LongByLongToLong reducer) {
5729	>	super(p, b, i, f, t); this.nextRight = nextRight;
5730	>	this.transformer = transformer;
5731	>	this.basis = basis; this.reducer = reducer;
5732	>	}
5733	>	public final Long getRawResult() { return result; }
5734	>	public final void compute() {
5735	>	final ObjectToLong<? super V> transformer;
5736	>	final LongByLongToLong reducer;
5737	>	if ((transformer = this.transformer) != null &&
5738	>	(reducer = this.reducer) != null) {
5739	>	long r = this.basis;
5740	>	for (int i = baseIndex, f, h; batch > 0 &&
5741	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5742	>	addToPendingCount(1);
5743	>	(rights = new MapReduceValuesToLongTask<K,V>
5744	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5745	>	rights, transformer, r, reducer)).fork();
5746	>	}
5747	>	for (Node<K,V> p; (p = advance()) != null; )
5748	>	r = reducer.apply(r, transformer.apply(p.val));
5749	>	result = r;
5750	>	CountedCompleter<?> c;
5751	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5752	>	@SuppressWarnings("unchecked") MapReduceValuesToLongTask<K,V>
5753	>	t = (MapReduceValuesToLongTask<K,V>)c,
5754	>	s = t.rights;
5755	>	while (s != null) {
5756	>	t.result = reducer.apply(t.result, s.result);
5757	>	s = t.rights = s.nextRight;
5758	>	}
5759	>	}
5760	>	}
5761	>	}
5762	>	}
5763	>
5764	>	@SuppressWarnings("serial")
5765	>	static final class MapReduceEntriesToLongTask<K,V>
5766	>	extends BulkTask<K,V,Long> {
5767	>	final ObjectToLong<Map.Entry<K,V>> transformer;
5768	>	final LongByLongToLong reducer;
5769	>	final long basis;
5770	>	long result;
5771	>	MapReduceEntriesToLongTask<K,V> rights, nextRight;
5772	>	MapReduceEntriesToLongTask
5773	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5774	>	MapReduceEntriesToLongTask<K,V> nextRight,
5775	>	ObjectToLong<Map.Entry<K,V>> transformer,
5776	>	long basis,
5777	>	LongByLongToLong reducer) {
5778	>	super(p, b, i, f, t); this.nextRight = nextRight;
5779	>	this.transformer = transformer;
5780	>	this.basis = basis; this.reducer = reducer;
5781	>	}
5782	>	public final Long getRawResult() { return result; }
5783	>	public final void compute() {
5784	>	final ObjectToLong<Map.Entry<K,V>> transformer;
5785	>	final LongByLongToLong reducer;
5786	>	if ((transformer = this.transformer) != null &&
5787	>	(reducer = this.reducer) != null) {
5788	>	long r = this.basis;
5789	>	for (int i = baseIndex, f, h; batch > 0 &&
5790	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5791	>	addToPendingCount(1);
5792	>	(rights = new MapReduceEntriesToLongTask<K,V>
5793	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5794	>	rights, transformer, r, reducer)).fork();
5795	>	}
5796	>	for (Node<K,V> p; (p = advance()) != null; )
5797	>	r = reducer.apply(r, transformer.apply(p));
5798	>	result = r;
5799	>	CountedCompleter<?> c;
5800	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5801	>	@SuppressWarnings("unchecked") MapReduceEntriesToLongTask<K,V>
5802	>	t = (MapReduceEntriesToLongTask<K,V>)c,
5803	>	s = t.rights;
5804	>	while (s != null) {
5805	>	t.result = reducer.apply(t.result, s.result);
5806	>	s = t.rights = s.nextRight;
5807	>	}
5808	>	}
5809	>	}
5810	>	}
5811	>	}
5812	>
5813	>	@SuppressWarnings("serial")
5814	>	static final class MapReduceMappingsToLongTask<K,V>
5815	>	extends BulkTask<K,V,Long> {
5816	>	final ObjectByObjectToLong<? super K, ? super V> transformer;
5817	>	final LongByLongToLong reducer;
5818	>	final long basis;
5819	>	long result;
5820	>	MapReduceMappingsToLongTask<K,V> rights, nextRight;
5821	>	MapReduceMappingsToLongTask
5822	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5823	>	MapReduceMappingsToLongTask<K,V> nextRight,
5824	>	ObjectByObjectToLong<? super K, ? super V> transformer,
5825	>	long basis,
5826	>	LongByLongToLong reducer) {
5827	>	super(p, b, i, f, t); this.nextRight = nextRight;
5828	>	this.transformer = transformer;
5829	>	this.basis = basis; this.reducer = reducer;
5830	>	}
5831	>	public final Long getRawResult() { return result; }
5832	>	public final void compute() {
5833	>	final ObjectByObjectToLong<? super K, ? super V> transformer;
5834	>	final LongByLongToLong reducer;
5835	>	if ((transformer = this.transformer) != null &&
5836	>	(reducer = this.reducer) != null) {
5837	>	long r = this.basis;
5838	>	for (int i = baseIndex, f, h; batch > 0 &&
5839	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5840	>	addToPendingCount(1);
5841	>	(rights = new MapReduceMappingsToLongTask<K,V>
5842	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5843	>	rights, transformer, r, reducer)).fork();
5844	>	}
5845	>	for (Node<K,V> p; (p = advance()) != null; )
5846	>	r = reducer.apply(r, transformer.apply(p.key, p.val));
5847	>	result = r;
5848	>	CountedCompleter<?> c;
5849	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5850	>	@SuppressWarnings("unchecked") MapReduceMappingsToLongTask<K,V>
5851	>	t = (MapReduceMappingsToLongTask<K,V>)c,
5852	>	s = t.rights;
5853	>	while (s != null) {
5854	>	t.result = reducer.apply(t.result, s.result);
5855	>	s = t.rights = s.nextRight;
5856	>	}
5857	>	}
5858	>	}
5859	>	}
5860	>	}
5861	>
5862	>	@SuppressWarnings("serial")
5863	>	static final class MapReduceKeysToIntTask<K,V>
5864	>	extends BulkTask<K,V,Integer> {
5865	>	final ObjectToInt<? super K> transformer;
5866	>	final IntByIntToInt reducer;
5867	>	final int basis;
5868	>	int result;
5869	>	MapReduceKeysToIntTask<K,V> rights, nextRight;
5870	>	MapReduceKeysToIntTask
5871	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5872	>	MapReduceKeysToIntTask<K,V> nextRight,
5873	>	ObjectToInt<? super K> transformer,
5874	>	int basis,
5875	>	IntByIntToInt reducer) {
5876	>	super(p, b, i, f, t); this.nextRight = nextRight;
5877	>	this.transformer = transformer;
5878	>	this.basis = basis; this.reducer = reducer;
5879	>	}
5880	>	public final Integer getRawResult() { return result; }
5881	>	public final void compute() {
5882	>	final ObjectToInt<? super K> transformer;
5883	>	final IntByIntToInt reducer;
5884	>	if ((transformer = this.transformer) != null &&
5885	>	(reducer = this.reducer) != null) {
5886	>	int r = this.basis;
5887	>	for (int i = baseIndex, f, h; batch > 0 &&
5888	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5889	>	addToPendingCount(1);
5890	>	(rights = new MapReduceKeysToIntTask<K,V>
5891	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5892	>	rights, transformer, r, reducer)).fork();
5893	>	}
5894	>	for (Node<K,V> p; (p = advance()) != null; )
5895	>	r = reducer.apply(r, transformer.apply(p.key));
5896	>	result = r;
5897	>	CountedCompleter<?> c;
5898	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5899	>	@SuppressWarnings("unchecked") MapReduceKeysToIntTask<K,V>
5900	>	t = (MapReduceKeysToIntTask<K,V>)c,
5901	>	s = t.rights;
5902	>	while (s != null) {
5903	>	t.result = reducer.apply(t.result, s.result);
5904	>	s = t.rights = s.nextRight;
5905	>	}
5906	>	}
5907	>	}
5908	>	}
5909	>	}
5910	>
5911	>	@SuppressWarnings("serial")
5912	>	static final class MapReduceValuesToIntTask<K,V>
5913	>	extends BulkTask<K,V,Integer> {
5914	>	final ObjectToInt<? super V> transformer;
5915	>	final IntByIntToInt reducer;
5916	>	final int basis;
5917	>	int result;
5918	>	MapReduceValuesToIntTask<K,V> rights, nextRight;
5919	>	MapReduceValuesToIntTask
5920	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5921	>	MapReduceValuesToIntTask<K,V> nextRight,
5922	>	ObjectToInt<? super V> transformer,
5923	>	int basis,
5924	>	IntByIntToInt reducer) {
5925	>	super(p, b, i, f, t); this.nextRight = nextRight;
5926	>	this.transformer = transformer;
5927	>	this.basis = basis; this.reducer = reducer;
5928	>	}
5929	>	public final Integer getRawResult() { return result; }
5930	>	public final void compute() {
5931	>	final ObjectToInt<? super V> transformer;
5932	>	final IntByIntToInt reducer;
5933	>	if ((transformer = this.transformer) != null &&
5934	>	(reducer = this.reducer) != null) {
5935	>	int r = this.basis;
5936	>	for (int i = baseIndex, f, h; batch > 0 &&
5937	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5938	>	addToPendingCount(1);
5939	>	(rights = new MapReduceValuesToIntTask<K,V>
5940	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5941	>	rights, transformer, r, reducer)).fork();
5942	>	}
5943	>	for (Node<K,V> p; (p = advance()) != null; )
5944	>	r = reducer.apply(r, transformer.apply(p.val));
5945	>	result = r;
5946	>	CountedCompleter<?> c;
5947	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5948	>	@SuppressWarnings("unchecked") MapReduceValuesToIntTask<K,V>
5949	>	t = (MapReduceValuesToIntTask<K,V>)c,
5950	>	s = t.rights;
5951	>	while (s != null) {
5952	>	t.result = reducer.apply(t.result, s.result);
5953	>	s = t.rights = s.nextRight;
5954	>	}
5955	>	}
5956	>	}
5957	>	}
5958	>	}
5959	>
5960	>	@SuppressWarnings("serial")
5961	>	static final class MapReduceEntriesToIntTask<K,V>
5962	>	extends BulkTask<K,V,Integer> {
5963	>	final ObjectToInt<Map.Entry<K,V>> transformer;
5964	>	final IntByIntToInt reducer;
5965	>	final int basis;
5966	>	int result;
5967	>	MapReduceEntriesToIntTask<K,V> rights, nextRight;
5968	>	MapReduceEntriesToIntTask
5969	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
5970	>	MapReduceEntriesToIntTask<K,V> nextRight,
5971	>	ObjectToInt<Map.Entry<K,V>> transformer,
5972	>	int basis,
5973	>	IntByIntToInt reducer) {
5974	>	super(p, b, i, f, t); this.nextRight = nextRight;
5975	>	this.transformer = transformer;
5976	>	this.basis = basis; this.reducer = reducer;
5977	>	}
5978	>	public final Integer getRawResult() { return result; }
5979	>	public final void compute() {
5980	>	final ObjectToInt<Map.Entry<K,V>> transformer;
5981	>	final IntByIntToInt reducer;
5982	>	if ((transformer = this.transformer) != null &&
5983	>	(reducer = this.reducer) != null) {
5984	>	int r = this.basis;
5985	>	for (int i = baseIndex, f, h; batch > 0 &&
5986	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
5987	>	addToPendingCount(1);
5988	>	(rights = new MapReduceEntriesToIntTask<K,V>
5989	>	(this, batch >>>= 1, baseLimit = h, f, tab,
5990	>	rights, transformer, r, reducer)).fork();
5991	>	}
5992	>	for (Node<K,V> p; (p = advance()) != null; )
5993	>	r = reducer.apply(r, transformer.apply(p));
5994	>	result = r;
5995	>	CountedCompleter<?> c;
5996	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
5997	>	@SuppressWarnings("unchecked") MapReduceEntriesToIntTask<K,V>
5998	>	t = (MapReduceEntriesToIntTask<K,V>)c,
5999	>	s = t.rights;
6000	>	while (s != null) {
6001	>	t.result = reducer.apply(t.result, s.result);
6002	>	s = t.rights = s.nextRight;
6003	>	}
6004	>	}
6005	>	}
6006	>	}
6007	>	}
6008	>
6009	>	@SuppressWarnings("serial")
6010	>	static final class MapReduceMappingsToIntTask<K,V>
6011	>	extends BulkTask<K,V,Integer> {
6012	>	final ObjectByObjectToInt<? super K, ? super V> transformer;
6013	>	final IntByIntToInt reducer;
6014	>	final int basis;
6015	>	int result;
6016	>	MapReduceMappingsToIntTask<K,V> rights, nextRight;
6017	>	MapReduceMappingsToIntTask
6018	>	(BulkTask<K,V,?> p, int b, int i, int f, Node<K,V>[] t,
6019	>	MapReduceMappingsToIntTask<K,V> nextRight,
6020	>	ObjectByObjectToInt<? super K, ? super V> transformer,
6021	>	int basis,
6022	>	IntByIntToInt reducer) {
6023	>	super(p, b, i, f, t); this.nextRight = nextRight;
6024	>	this.transformer = transformer;
6025	>	this.basis = basis; this.reducer = reducer;
6026	>	}
6027	>	public final Integer getRawResult() { return result; }
6028	>	public final void compute() {
6029	>	final ObjectByObjectToInt<? super K, ? super V> transformer;
6030	>	final IntByIntToInt reducer;
6031	>	if ((transformer = this.transformer) != null &&
6032	>	(reducer = this.reducer) != null) {
6033	>	int r = this.basis;
6034	>	for (int i = baseIndex, f, h; batch > 0 &&
6035	>	(h = ((f = baseLimit) + i) >>> 1) > i;) {
6036	>	addToPendingCount(1);
6037	>	(rights = new MapReduceMappingsToIntTask<K,V>
6038	>	(this, batch >>>= 1, baseLimit = h, f, tab,
6039	>	rights, transformer, r, reducer)).fork();
6040	>	}
6041	>	for (Node<K,V> p; (p = advance()) != null; )
6042	>	r = reducer.apply(r, transformer.apply(p.key, p.val));
6043	>	result = r;
6044	>	CountedCompleter<?> c;
6045	>	for (c = firstComplete(); c != null; c = c.nextComplete()) {
6046	>	@SuppressWarnings("unchecked") MapReduceMappingsToIntTask<K,V>
6047	>	t = (MapReduceMappingsToIntTask<K,V>)c,
6048	>	s = t.rights;
6049	>	while (s != null) {
6050	>	t.result = reducer.apply(t.result, s.result);
6051	>	s = t.rights = s.nextRight;
6052	>	}
6053	>	}
6054	>	}
6055	>	}
6056	>	}
6057	>
6058	>	/* ---------------- Counters -------------- */
6059	>
6060	>	// Adapted from LongAdder and Striped64.
6061	>	// See their internal docs for explanation.
6062	>
6063	>	// A padded cell for distributing counts
6064	>	static final class CounterCell {
6065	>	volatile long p0, p1, p2, p3, p4, p5, p6;
6066	>	volatile long value;
6067	>	volatile long q0, q1, q2, q3, q4, q5, q6;
6068	>	CounterCell(long x) { value = x; }
6069	>	}
6070
6071		/**
6072	<	* Helper class used in previous version, declared for the sake of
6073	<	* serialization compatibility
6072	>	* Holder for the thread-local hash code determining which
6073	>	* CounterCell to use. The code is initialized via the
6074	>	* counterHashCodeGenerator, but may be moved upon collisions.
6075		*/
6076	<	static class Segment<K,V> extends java.util.concurrent.locks.ReentrantLock
6077	<	implements Serializable {
1515	<	private static final long serialVersionUID = 2249069246763182397L;
1516	<	final float loadFactor;
1517	<	Segment(float lf) { this.loadFactor = lf; }
6076	>	static final class CounterHashCode {
6077	>	int code;
6078		}
6079
6080		/**
6081	<	* Saves the state of the {@code ConcurrentHashMapV8} instance to a
1522	<	* stream (i.e., serializes it).
1523	<	* @param s the stream
1524	<	* @serialData
1525	<	* the key (Object) and value (Object)
1526	<	* for each key-value mapping, followed by a null pair.
1527	<	* The key-value mappings are emitted in no particular order.
6081	>	* Generates initial value for per-thread CounterHashCodes.
6082		*/
6083	<	@SuppressWarnings("unchecked")
1530	<	private void writeObject(java.io.ObjectOutputStream s)
1531	<	throws java.io.IOException {
1532	<	if (segments == null) { // for serialization compatibility
1533	<	segments = (Segment<K,V>[])
1534	<	new Segment<?,?>[DEFAULT_CONCURRENCY_LEVEL];
1535	<	for (int i = 0; i < segments.length; ++i)
1536	<	segments[i] = new Segment<K,V>(loadFactor);
1537	<	}
1538	<	s.defaultWriteObject();
1539	<	new HashIterator().writeEntries(s);
1540	<	s.writeObject(null);
1541	<	s.writeObject(null);
1542	<	segments = null; // throw away
1543	<	}
6083	>	static final AtomicInteger counterHashCodeGenerator = new AtomicInteger();
6084
6085		/**
6086	<	* Reconstitutes the instance from a stream (that is, deserializes it).
6087	<	* @param s the stream
6086	>	* Increment for counterHashCodeGenerator. See class ThreadLocal
6087	>	* for explanation.
6088		*/
6089	<	@SuppressWarnings("unchecked")
6090	<	private void readObject(java.io.ObjectInputStream s)
6091	<	throws java.io.IOException, ClassNotFoundException {
6092	<	s.defaultReadObject();
6093	<	// find load factor in a segment, if one exists
6094	<	if (segments != null && segments.length != 0)
6095	<	this.loadFactor = segments[0].loadFactor;
6096	<	else
1557	<	this.loadFactor = DEFAULT_LOAD_FACTOR;
1558	<	this.initCap = DEFAULT_CAPACITY;
1559	<	LongAdder ct = new LongAdder(); // force final field write
1560	<	UNSAFE.putObjectVolatile(this, counterOffset, ct);
1561	<	this.segments = null; // unneeded
6089	>	static final int SEED_INCREMENT = 0x61c88647;
6090	>
6091	>	/**
6092	>	* Per-thread counter hash codes. Shared across all instances.
6093	>	*/
6094	>	static final ThreadLocal<CounterHashCode> threadCounterHashCode =
6095	>	new ThreadLocal<CounterHashCode>();
6096	>
6097
6098	<	// Read the keys and values, and put the mappings in the table
6098	>	final long sumCount() {
6099	>	CounterCell[] as = counterCells; CounterCell a;
6100	>	long sum = baseCount;
6101	>	if (as != null) {
6102	>	for (int i = 0; i < as.length; ++i) {
6103	>	if ((a = as[i]) != null)
6104	>	sum += a.value;
6105	>	}
6106	>	}
6107	>	return sum;
6108	>	}
6109	>
6110	>	// See LongAdder version for explanation
6111	>	private final void fullAddCount(long x, CounterHashCode hc,
6112	>	boolean wasUncontended) {
6113	>	int h;
6114	>	if (hc == null) {
6115	>	hc = new CounterHashCode();
6116	>	int s = counterHashCodeGenerator.addAndGet(SEED_INCREMENT);
6117	>	h = hc.code = (s == 0) ? 1 : s; // Avoid zero
6118	>	threadCounterHashCode.set(hc);
6119	>	}
6120	>	else
6121	>	h = hc.code;
6122	>	boolean collide = false;                // True if last slot nonempty
6123		for (;;) {
6124	<	K key = (K) s.readObject();
6125	<	V value = (V) s.readObject();
6126	<	if (key == null)
6127	<	break;
6128	<	put(key, value);
6124	>	CounterCell[] as; CounterCell a; int n; long v;
6125	>	if ((as = counterCells) != null && (n = as.length) > 0) {
6126	>	if ((a = as[(n - 1) & h]) == null) {
6127	>	if (cellsBusy == 0) {            // Try to attach new Cell
6128	>	CounterCell r = new CounterCell(x); // Optimistic create
6129	>	if (cellsBusy == 0 &&
6130	>	U.compareAndSwapInt(this, CELLSBUSY, 0, 1)) {
6131	>	boolean created = false;
6132	>	try {               // Recheck under lock
6133	>	CounterCell[] rs; int m, j;
6134	>	if ((rs = counterCells) != null &&
6135	>	(m = rs.length) > 0 &&
6136	>	rs[j = (m - 1) & h] == null) {
6137	>	rs[j] = r;
6138	>	created = true;
6139	>	}
6140	>	} finally {
6141	>	cellsBusy = 0;
6142	>	}
6143	>	if (created)
6144	>	break;
6145	>	continue;           // Slot is now non-empty
6146	>	}
6147	>	}
6148	>	collide = false;
6149	>	}
6150	>	else if (!wasUncontended)       // CAS already known to fail
6151	>	wasUncontended = true;      // Continue after rehash
6152	>	else if (U.compareAndSwapLong(a, CELLVALUE, v = a.value, v + x))
6153	>	break;
6154	>	else if (counterCells != as || n >= NCPU)
6155	>	collide = false;            // At max size or stale
6156	>	else if (!collide)
6157	>	collide = true;
6158	>	else if (cellsBusy == 0 &&
6159	>	U.compareAndSwapInt(this, CELLSBUSY, 0, 1)) {
6160	>	try {
6161	>	if (counterCells == as) {// Expand table unless stale
6162	>	CounterCell[] rs = new CounterCell[n << 1];
6163	>	for (int i = 0; i < n; ++i)
6164	>	rs[i] = as[i];
6165	>	counterCells = rs;
6166	>	}
6167	>	} finally {
6168	>	cellsBusy = 0;
6169	>	}
6170	>	collide = false;
6171	>	continue;                   // Retry with expanded table
6172	>	}
6173	>	h ^= h << 13;                   // Rehash
6174	>	h ^= h >>> 17;
6175	>	h ^= h << 5;
6176	>	}
6177	>	else if (cellsBusy == 0 && counterCells == as &&
6178	>	U.compareAndSwapInt(this, CELLSBUSY, 0, 1)) {
6179	>	boolean init = false;
6180	>	try {                           // Initialize table
6181	>	if (counterCells == as) {
6182	>	CounterCell[] rs = new CounterCell[2];
6183	>	rs[h & 1] = new CounterCell(x);
6184	>	counterCells = rs;
6185	>	init = true;
6186	>	}
6187	>	} finally {
6188	>	cellsBusy = 0;
6189	>	}
6190	>	if (init)
6191	>	break;
6192	>	}
6193	>	else if (U.compareAndSwapLong(this, BASECOUNT, v = baseCount, v + x))
6194	>	break;                          // Fall back on using base
6195		}
6196	+	hc.code = h;                            // Record index for next time
6197		}
6198
6199		// Unsafe mechanics
6200	<	private static final sun.misc.Unsafe UNSAFE;
6201	<	private static final long counterOffset;
6202	<	private static final long resizingOffset;
6200	>	private static final sun.misc.Unsafe U;
6201	>	private static final long SIZECTL;
6202	>	private static final long TRANSFERINDEX;
6203	>	private static final long BASECOUNT;
6204	>	private static final long CELLSBUSY;
6205	>	private static final long CELLVALUE;
6206		private static final long ABASE;
6207		private static final int ASHIFT;
6208
6209		static {
1581	–	int ss;
6210		try {
6211	<	UNSAFE = getUnsafe();
6211	>	U = getUnsafe();
6212		Class<?> k = ConcurrentHashMapV8.class;
6213	<	counterOffset = UNSAFE.objectFieldOffset
6214	<	(k.getDeclaredField("counter"));
6215	<	resizingOffset = UNSAFE.objectFieldOffset
6216	<	(k.getDeclaredField("resizing"));
6217	<	Class<?> sc = Node[].class;
6218	<	ABASE = UNSAFE.arrayBaseOffset(sc);
6219	<	ss = UNSAFE.arrayIndexScale(sc);
6213	>	SIZECTL = U.objectFieldOffset
6214	>	(k.getDeclaredField("sizeCtl"));
6215	>	TRANSFERINDEX = U.objectFieldOffset
6216	>	(k.getDeclaredField("transferIndex"));
6217	>	BASECOUNT = U.objectFieldOffset
6218	>	(k.getDeclaredField("baseCount"));
6219	>	CELLSBUSY = U.objectFieldOffset
6220	>	(k.getDeclaredField("cellsBusy"));
6221	>	Class<?> ck = CounterCell.class;
6222	>	CELLVALUE = U.objectFieldOffset
6223	>	(ck.getDeclaredField("value"));
6224	>	Class<?> ak = Node[].class;
6225	>	ABASE = U.arrayBaseOffset(ak);
6226	>	int scale = U.arrayIndexScale(ak);
6227	>	if ((scale & (scale - 1)) != 0)
6228	>	throw new Error("data type scale not a power of two");
6229	>	ASHIFT = 31 - Integer.numberOfLeadingZeros(scale);
6230		} catch (Exception e) {
6231		throw new Error(e);
6232		}
1595	–	if ((ss & (ss-1)) != 0)
1596	–	throw new Error("data type scale not a power of two");
1597	–	ASHIFT = 31 - Integer.numberOfLeadingZeros(ss);
6233		}
6234
6235		/**
#	Line 1607 | Line 6242 | public class ConcurrentHashMapV8<K, V>
6242		private static sun.misc.Unsafe getUnsafe() {
6243		try {
6244		return sun.misc.Unsafe.getUnsafe();
6245	<	} catch (SecurityException se) {
6246	<	try {
6247	<	return java.security.AccessController.doPrivileged
6248	<	(new java.security
6249	<	.PrivilegedExceptionAction<sun.misc.Unsafe>() {
6250	<	public sun.misc.Unsafe run() throws Exception {
6251	<	java.lang.reflect.Field f = sun.misc
6252	<	.Unsafe.class.getDeclaredField("theUnsafe");
6253	<	f.setAccessible(true);
6254	<	return (sun.misc.Unsafe) f.get(null);
6255	<	}});
6256	<	} catch (java.security.PrivilegedActionException e) {
6257	<	throw new RuntimeException("Could not initialize intrinsics",
6258	<	e.getCause());
6259	<	}
6245	>	} catch (SecurityException tryReflectionInstead) {}
6246	>	try {
6247	>	return java.security.AccessController.doPrivileged
6248	>	(new java.security.PrivilegedExceptionAction<sun.misc.Unsafe>() {
6249	>	public sun.misc.Unsafe run() throws Exception {
6250	>	Class<sun.misc.Unsafe> k = sun.misc.Unsafe.class;
6251	>	for (java.lang.reflect.Field f : k.getDeclaredFields()) {
6252	>	f.setAccessible(true);
6253	>	Object x = f.get(null);
6254	>	if (k.isInstance(x))
6255	>	return k.cast(x);
6256	>	}
6257	>	throw new NoSuchFieldError("the Unsafe");
6258	>	}});
6259	>	} catch (java.security.PrivilegedActionException e) {
6260	>	throw new RuntimeException("Could not initialize intrinsics",
6261	>	e.getCause());
6262		}
6263		}
1627	–
6264		}

Diff Legend

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