ViewVC Help

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

root/jsr166/jsr166/src/jsr166y/ForkJoinWorkerThread.java

Comparing jsr166/src/jsr166y/ForkJoinWorkerThread.java (file contents):
Revision 1.1 by dl, Tue Jan 6 14:30:31 2009 UTC vs.
Revision 1.54 by dl, Wed Nov 17 12:06:46 2010 UTC

#	Line 5 | Line 5
5		*/
6
7		package jsr166y;
8	<	import java.util.*;
9	<	import java.util.concurrent.*;
10	<	import java.util.concurrent.atomic.*;
11	<	import java.util.concurrent.locks.*;
12	<	import sun.misc.Unsafe;
13	<	import java.lang.reflect.*;
8	>
9	>	import java.util.Random;
10	>	import java.util.Collection;
11	>	import java.util.concurrent.locks.LockSupport;
12	>	import java.util.concurrent.RejectedExecutionException;
13
14		/**
15	<	* A thread that is internally managed by a ForkJoinPool to execute
16	<	* ForkJoinTasks. This class additionally provides public
17	<	* <tt>static</tt> methods accessing some basic scheduling and
18	<	* execution mechanics for the <em>current</em>
19	<	* ForkJoinWorkerThread. These methods may be invoked only from within
20	<	* other ForkJoinTask computations. Attempts to invoke in other
21	<	* contexts result in exceptions or errors including
22	<	* ClassCastException.  These methods enable construction of
24	<	* special-purpose task classes, as well as specialized idioms
25	<	* occasionally useful in ForkJoinTask processing.
26	<	*
27	<	* <p>The form of supported static methods reflects the fact that
28	<	* worker threads may access and process tasks obtained in any of
29	<	* three ways. In preference order: <em>Local</em> tasks are processed
30	<	* in LIFO (newest first) order. <em>Stolen</em> tasks are obtained
31	<	* from other threads in FIFO (oldest first) order, only if there are
32	<	* no local tasks to run.  <em>Submissions</em> form a FIFO queue
33	<	* common to the entire pool, and are started only if no other
34	<	* work is available.
15	>	* A thread managed by a {@link ForkJoinPool}.  This class is
16	>	* subclassable solely for the sake of adding functionality -- there
17	>	* are no overridable methods dealing with scheduling or execution.
18	>	* However, you can override initialization and termination methods
19	>	* surrounding the main task processing loop.  If you do create such a
20	>	* subclass, you will also need to supply a custom {@link
21	>	* ForkJoinPool.ForkJoinWorkerThreadFactory} to use it in a {@code
22	>	* ForkJoinPool}.
23		*
24	<	* <p> This class is subclassable solely for the sake of adding
25	<	* functionality -- there are no overridable methods dealing with
38	<	* scheduling or execution. However, you can override initialization
39	<	* and termination cleanup methods surrounding the main task
40	<	* processing loop.  If you do create such a subclass, you will also
41	<	* need to supply a custom ForkJoinWorkerThreadFactory to use it in a
42	<	* ForkJoinPool.
24	>	* @since 1.7
25	>	* @author Doug Lea
26		*/
27		public class ForkJoinWorkerThread extends Thread {
28		/*
29	<	* Algorithm overview:
29	>	* Overview:
30	>	*
31	>	* ForkJoinWorkerThreads are managed by ForkJoinPools and perform
32	>	* ForkJoinTasks. This class includes bookkeeping in support of
33	>	* worker activation, suspension, and lifecycle control described
34	>	* in more detail in the internal documentation of class
35	>	* ForkJoinPool. And as described further below, this class also
36	>	* includes special-cased support for some ForkJoinTask
37	>	* methods. But the main mechanics involve work-stealing:
38		*
39	<	* 1. Work-Stealing: Work-stealing queues are special forms of
40	<	* Deques that support only three of the four possible
41	<	* end-operations -- push, pop, and deq (aka steal), and only do
42	<	* so under the constraints that push and pop are called only from
43	<	* the owning thread, while deq may be called from other threads.
44	<	* (If you are unfamiliar with them, you probably want to read
45	<	* Herlihy and Shavit's book "The Art of Multiprocessor
46	<	* programming", chapter 16 describing these in more detail before
47	<	* proceeding.)  The main work-stealing queue design is roughly
48	<	* similar to "Dynamic Circular Work-Stealing Deque" by David
49	<	* Chase and Yossi Lev, SPAA 2005
50	<	* (http://research.sun.com/scalable/pubs/index.html).  The main
51	<	* difference ultimately stems from gc requirements that we null
52	<	* out taken slots as soon as we can, to maintain as small a
53	<	* footprint as possible even in programs generating huge numbers
54	<	* of tasks. To accomplish this, we shift the CAS arbitrating pop
55	<	* vs deq (steal) from being on the indices ("base" and "sp") to
56	<	* the slots themselves (mainly via method "casSlotNull()"). So,
57	<	* both a successful pop and deq mainly entail CAS'ing a nonnull
58	<	* slot to null.  Because we rely on CASes of references, we do
59	<	* not need tag bits on base or sp.  They are simple ints as used
60	<	* in any circular array-based queue (see for example ArrayDeque).
61	<	* Updates to the indices must still be ordered in a way that
62	<	* guarantees that (sp - base) > 0 means the queue is empty, but
63	<	* otherwise may err on the side of possibly making the queue
64	<	* appear nonempty when a push, pop, or deq have not fully
65	<	* committed. Note that this means that the deq operation,
66	<	* considered individually, is not wait-free. One thief cannot
67	<	* successfully continue until another in-progress one (or, if
68	<	* previously empty, a push) completes.  However, in the
69	<	* aggregate, we ensure at least probablistic non-blockingness. If
70	<	* an attempted steal fails, a thief always chooses a different
39	>	* Work-stealing queues are special forms of Deques that support
40	>	* only three of the four possible end-operations -- push, pop,
41	>	* and deq (aka steal), under the further constraints that push
42	>	* and pop are called only from the owning thread, while deq may
43	>	* be called from other threads.  (If you are unfamiliar with
44	>	* them, you probably want to read Herlihy and Shavit's book "The
45	>	* Art of Multiprocessor programming", chapter 16 describing these
46	>	* in more detail before proceeding.)  The main work-stealing
47	>	* queue design is roughly similar to those in the papers "Dynamic
48	>	* Circular Work-Stealing Deque" by Chase and Lev, SPAA 2005
49	>	* (http://research.sun.com/scalable/pubs/index.html) and
50	>	* "Idempotent work stealing" by Michael, Saraswat, and Vechev,
51	>	* PPoPP 2009 (http://portal.acm.org/citation.cfm?id=1504186).
52	>	* The main differences ultimately stem from gc requirements that
53	>	* we null out taken slots as soon as we can, to maintain as small
54	>	* a footprint as possible even in programs generating huge
55	>	* numbers of tasks. To accomplish this, we shift the CAS
56	>	* arbitrating pop vs deq (steal) from being on the indices
57	>	* ("base" and "sp") to the slots themselves (mainly via method
58	>	* "casSlotNull()"). So, both a successful pop and deq mainly
59	>	* entail a CAS of a slot from non-null to null.  Because we rely
60	>	* on CASes of references, we do not need tag bits on base or sp.
61	>	* They are simple ints as used in any circular array-based queue
62	>	* (see for example ArrayDeque).  Updates to the indices must
63	>	* still be ordered in a way that guarantees that sp == base means
64	>	* the queue is empty, but otherwise may err on the side of
65	>	* possibly making the queue appear nonempty when a push, pop, or
66	>	* deq have not fully committed. Note that this means that the deq
67	>	* operation, considered individually, is not wait-free. One thief
68	>	* cannot successfully continue until another in-progress one (or,
69	>	* if previously empty, a push) completes.  However, in the
70	>	* aggregate, we ensure at least probabilistic non-blockingness.
71	>	* If an attempted steal fails, a thief always chooses a different
72		* random victim target to try next. So, in order for one thief to
73		* progress, it suffices for any in-progress deq or new push on
74		* any empty queue to complete. One reason this works well here is
75		* that apparently-nonempty often means soon-to-be-stealable,
76	<	* which gives threads a chance to activate if necessary before
77	<	* stealing (see below).
76	>	* which gives threads a chance to set activation status if
77	>	* necessary before stealing.
78		*
79	<	* Efficient implementation of this approach currently relies on
80	<	* an uncomfortable amount of "Unsafe" mechanics. To maintain
79	>	* This approach also enables support for "async mode" where local
80	>	* task processing is in FIFO, not LIFO order; simply by using a
81	>	* version of deq rather than pop when locallyFifo is true (as set
82	>	* by the ForkJoinPool).  This allows use in message-passing
83	>	* frameworks in which tasks are never joined.
84	>	*
85	>	* When a worker would otherwise be blocked waiting to join a
86	>	* task, it first tries a form of linear helping: Each worker
87	>	* records (in field currentSteal) the most recent task it stole
88	>	* from some other worker. Plus, it records (in field currentJoin)
89	>	* the task it is currently actively joining. Method joinTask uses
90	>	* these markers to try to find a worker to help (i.e., steal back
91	>	* a task from and execute it) that could hasten completion of the
92	>	* actively joined task. In essence, the joiner executes a task
93	>	* that would be on its own local deque had the to-be-joined task
94	>	* not been stolen. This may be seen as a conservative variant of
95	>	* the approach in Wagner & Calder "Leapfrogging: a portable
96	>	* technique for implementing efficient futures" SIGPLAN Notices,
97	>	* 1993 (http://portal.acm.org/citation.cfm?id=155354). It differs
98	>	* in that: (1) We only maintain dependency links across workers
99	>	* upon steals, rather than use per-task bookkeeping.  This may
100	>	* require a linear scan of workers array to locate stealers, but
101	>	* usually doesn't because stealers leave hints (that may become
102	>	* stale/wrong) of where to locate them. This isolates cost to
103	>	* when it is needed, rather than adding to per-task overhead.
104	>	* (2) It is "shallow", ignoring nesting and potentially cyclic
105	>	* mutual steals.  (3) It is intentionally racy: field currentJoin
106	>	* is updated only while actively joining, which means that we
107	>	* miss links in the chain during long-lived tasks, GC stalls etc
108	>	* (which is OK since blocking in such cases is usually a good
109	>	* idea).  (4) We bound the number of attempts to find work (see
110	>	* MAX_HELP_DEPTH) and fall back to suspending the worker and if
111	>	* necessary replacing it with a spare (see
112	>	* ForkJoinPool.awaitJoin).
113	>	*
114	>	* Efficient implementation of these algorithms currently relies
115	>	* on an uncomfortable amount of "Unsafe" mechanics. To maintain
116		* correct orderings, reads and writes of variable base require
117	<	* volatile ordering.  Variable sp does not require volatile write
118	<	* but needs cheaper store-ordering on writes.  Because they are
119	<	* protected by volatile base reads, reads of the queue array and
120	<	* its slots do not need volatile load semantics, but writes (in
121	<	* push) require store order and CASes (in pop and deq) require
122	<	* (volatile) CAS semantics. Since these combinations aren't
123	<	* supported using ordinary volatiles, the only way to accomplish
124	<	* these effciently is to use direct Unsafe calls. (Using external
125	<	* AtomicIntegers and AtomicReferenceArrays for the indices and
126	<	* array is significantly slower because of memory locality and
127	<	* indirection effects.) Further, performance on most platforms is
128	<	* very sensitive to placement and sizing of the (resizable) queue
129	<	* array.  Even though these queues don't usually become all that
130	<	* big, the initial size must be large enough to counteract cache
117	>	* volatile ordering.  Variable sp does not require volatile
118	>	* writes but still needs store-ordering, which we accomplish by
119	>	* pre-incrementing sp before filling the slot with an ordered
120	>	* store.  (Pre-incrementing also enables backouts used in
121	>	* joinTask.)  Because they are protected by volatile base reads,
122	>	* reads of the queue array and its slots by other threads do not
123	>	* need volatile load semantics, but writes (in push) require
124	>	* store order and CASes (in pop and deq) require (volatile) CAS
125	>	* semantics.  (Michael, Saraswat, and Vechev's algorithm has
126	>	* similar properties, but without support for nulling slots.)
127	>	* Since these combinations aren't supported using ordinary
128	>	* volatiles, the only way to accomplish these efficiently is to
129	>	* use direct Unsafe calls. (Using external AtomicIntegers and
130	>	* AtomicReferenceArrays for the indices and array is
131	>	* significantly slower because of memory locality and indirection
132	>	* effects.)
133	>	*
134	>	* Further, performance on most platforms is very sensitive to
135	>	* placement and sizing of the (resizable) queue array.  Even
136	>	* though these queues don't usually become all that big, the
137	>	* initial size must be large enough to counteract cache
138		* contention effects across multiple queues (especially in the
139		* presence of GC cardmarking). Also, to improve thread-locality,
140	<	* queues are currently initialized immediately after the thread
141	<	* gets the initial signal to start processing tasks.  However,
142	<	* all queue-related methods except pushTask are written in a way
143	<	* that allows them to instead be lazily allocated and/or disposed
144	<	* of when empty. All together, these low-level implementation
145	<	* choices produce as much as a factor of 4 performance
146	<	* improvement compared to naive implementations, and enable the
147	<	* processing of billions of tasks per second, sometimes at the
148	<	* expense of ugliness.
149	<	*
150	<	* 2. Run control: The primary run control is based on a global
151	<	* counter (activeCount) held by the pool. It uses an algorithm
152	<	* similar to that in Herlihy and Shavit section 17.6 to cause
153	<	* threads to eventually block when all threads declare they are
154	<	* inactive. (See variable "scans".)  For this to work, threads
155	<	* must be declared active when executing tasks, and before
156	<	* stealing a task. They must be inactive before blocking on the
157	<	* Pool Barrier (awaiting a new submission or other Pool
158	<	* event). In between, there is some free play which we take
159	<	* advantage of to avoid contention and rapid flickering of the
160	<	* global activeCount: If inactive, we activate only if a victim
161	<	* queue appears to be nonempty (see above).  Similarly, a thread
128	<	* tries to inactivate only after a full scan of other threads.
129	<	* The net effect is that contention on activeCount is rarely a
130	<	* measurable performance issue. (There are also a few other cases
131	<	* where we scan for work rather than retry/block upon
132	<	* contention.)
133	<	*
134	<	* 3. Selection control. We maintain policy of always choosing to
135	<	* run local tasks rather than stealing, and always trying to
136	<	* steal tasks before trying to run a new submission. All steals
137	<	* are currently performed in randomly-chosen deq-order. It may be
138	<	* worthwhile to bias these with locality / anti-locality
139	<	* information, but doing this well probably requires more
140	<	* lower-level information from JVMs than currently provided.
140	>	* queues are initialized after starting.  All together, these
141	>	* low-level implementation choices produce as much as a factor of
142	>	* 4 performance improvement compared to naive implementations,
143	>	* and enable the processing of billions of tasks per second,
144	>	* sometimes at the expense of ugliness.
145	>	*/
146	>
147	>	/**
148	>	* Generator for initial random seeds for random victim
149	>	* selection. This is used only to create initial seeds. Random
150	>	* steals use a cheaper xorshift generator per steal attempt. We
151	>	* expect only rare contention on seedGenerator, so just use a
152	>	* plain Random.
153	>	*/
154	>	private static final Random seedGenerator = new Random();
155	>
156	>	/**
157	>	* The maximum stolen->joining link depth allowed in helpJoinTask.
158	>	* Depths for legitimate chains are unbounded, but we use a fixed
159	>	* constant to avoid (otherwise unchecked) cycles and bound
160	>	* staleness of traversal parameters at the expense of sometimes
161	>	* blocking when we could be helping.
162		*/
163	+	private static final int MAX_HELP_DEPTH = 8;
164
165		/**
166		* Capacity of work-stealing queue array upon initialization.
167	<	* Must be a power of two. Initial size must be at least 2, but is
167	>	* Must be a power of two. Initial size must be at least 4, but is
168		* padded to minimize cache effects.
169		*/
170		private static final int INITIAL_QUEUE_CAPACITY = 1 << 13;
171
172		/**
173		* Maximum work-stealing queue array size.  Must be less than or
174	<	* equal to 1 << 30 to ensure lack of index wraparound.
174	>	* equal to 1 << (31 - width of array entry) to ensure lack of
175	>	* index wraparound. The value is set in the static block
176	>	* at the end of this file after obtaining width.
177		*/
178	<	private static final int MAXIMUM_QUEUE_CAPACITY = 1 << 30;
178	>	private static final int MAXIMUM_QUEUE_CAPACITY;
179
180		/**
181	<	* Generator of seeds for per-thread random numbers.
181	>	* The pool this thread works in. Accessed directly by ForkJoinTask.
182		*/
183	<	private static final Random randomSeedGenerator = new Random();
183	>	final ForkJoinPool pool;
184
185		/**
186		* The work-stealing queue array. Size must be a power of two.
187	+	* Initialized in onStart, to improve memory locality.
188		*/
189		private ForkJoinTask<?>[] queue;
190
191		/**
192	+	* Index (mod queue.length) of least valid queue slot, which is
193	+	* always the next position to steal from if nonempty.
194	+	*/
195	+	private volatile int base;
196	+
197	+	/**
198		* Index (mod queue.length) of next queue slot to push to or pop
199	<	* from. It is written only by owner thread, via ordered store.
200	<	* Both sp and base are allowed to wrap around on overflow, but
201	<	* (sp - base) still estimates size.
199	>	* from. It is written only by owner thread, and accessed by other
200	>	* threads only after reading (volatile) base.  Both sp and base
201	>	* are allowed to wrap around on overflow, but (sp - base) still
202	>	* estimates size.
203		*/
204	<	private volatile int sp;
204	>	private int sp;
205
206		/**
207	<	* Index (mod queue.length) of least valid queue slot, which is
208	<	* always the next position to steal from if nonempty.
207	>	* The index of most recent stealer, used as a hint to avoid
208	>	* traversal in method helpJoinTask. This is only a hint because a
209	>	* worker might have had multiple steals and this only holds one
210	>	* of them (usually the most current). Declared non-volatile,
211	>	* relying on other prevailing sync to keep reasonably current.
212		*/
213	<	private volatile int base;
213	>	private int stealHint;
214
215		/**
216	<	* The pool this thread works in.
216	>	* Run state of this worker. In addition to the usual run levels,
217	>	* tracks if this worker is suspended as a spare, and if it was
218	>	* killed (trimmed) while suspended. However, "active" status is
219	>	* maintained separately and modified only in conjunction with
220	>	* CASes of the pool's runState (which are currently sadly
221	>	* manually inlined for performance.)  Accessed directly by pool
222	>	* to simplify checks for normal (zero) status.
223		*/
224	<	final ForkJoinPool pool;
224	>	volatile int runState;
225	>
226	>	private static final int TERMINATING = 0x01;
227	>	private static final int TERMINATED  = 0x02;
228	>	private static final int SUSPENDED   = 0x04; // inactive spare
229	>	private static final int TRIMMED     = 0x08; // killed while suspended
230	>
231	>	/**
232	>	* Number of steals. Directly accessed (and reset) by
233	>	* pool.tryAccumulateStealCount when idle.
234	>	*/
235	>	int stealCount;
236	>
237	>	/**
238	>	* Seed for random number generator for choosing steal victims.
239	>	* Uses Marsaglia xorshift. Must be initialized as nonzero.
240	>	*/
241	>	private int seed;
242	>
243	>	/**
244	>	* Activity status. When true, this worker is considered active.
245	>	* Accessed directly by pool.  Must be false upon construction.
246	>	*/
247	>	boolean active;
248	>
249	>	/**
250	>	* True if use local fifo, not default lifo, for local polling.
251	>	* Shadows value from ForkJoinPool.
252	>	*/
253	>	private final boolean locallyFifo;
254
255		/**
256		* Index of this worker in pool array. Set once by pool before
257	<	* running, and accessed directly by pool during cleanup etc
257	>	* running, and accessed directly by pool to locate this worker in
258	>	* its workers array.
259		*/
260		int poolIndex;
261
262		/**
263	<	* Run state of this worker. Supports simple versions of the usual
264	<	* shutdown/shutdownNow control.
263	>	* The last pool event waited for. Accessed only by pool in
264	>	* callback methods invoked within this thread.
265		*/
266	<	private volatile int runState;
196	<
197	<	// Runstate values. Order matters
198	<	private static final int RUNNING     = 0;
199	<	private static final int SHUTDOWN    = 1;
200	<	private static final int TERMINATING = 2;
201	<	private static final int TERMINATED  = 3;
266	>	int lastEventCount;
267
268		/**
269	<	* Activity status. When true, this worker is considered active.
270	<	* Must be false upon construction. It must be true when executing
206	<	* tasks, and BEFORE stealing a task. It must be false before
207	<	* blocking on the Pool Barrier.
269	>	* Encoded index and event count of next event waiter. Accessed
270	>	* only by ForkJoinPool for managing event waiters.
271		*/
272	<	private boolean active;
272	>	volatile long nextWaiter;
273
274		/**
275	<	* Number of steals, transferred to pool when idle
275	>	* Number of times this thread suspended as spare. Accessed only
276	>	* by pool.
277		*/
278	<	private int stealCount;
278	>	int spareCount;
279
280		/**
281	<	* Seed for random number generator for choosing steal victims
281	>	* Encoded index and count of next spare waiter. Accessed only
282	>	* by ForkJoinPool for managing spares.
283		*/
284	<	private int randomVictimSeed;
284	>	volatile int nextSpare;
285
286		/**
287	<	* Seed for embedded Jurandom
287	>	* The task currently being joined, set only when actively trying
288	>	* to help other stealers in helpJoinTask. Written only by this
289	>	* thread, but read by others.
290		*/
291	<	private long juRandomSeed;
291	>	private volatile ForkJoinTask<?> currentJoin;
292
293		/**
294	<	* The last barrier event waited for
294	>	* The task most recently stolen from another worker (or
295	>	* submission queue).  Written only by this thread, but read by
296	>	* others.
297		*/
298	<	private long eventCount;
298	>	private volatile ForkJoinTask<?> currentSteal;
299
300		/**
301		* Creates a ForkJoinWorkerThread operating in the given pool.
302	+	*
303		* @param pool the pool this thread works in
304		* @throws NullPointerException if pool is null
305		*/
306		protected ForkJoinWorkerThread(ForkJoinPool pool) {
237	–	if (pool == null) throw new NullPointerException();
307		this.pool = pool;
308	<	// remaining initialization deferred to onStart
309	<	}
310	<
311	<	//  Access methods used by Pool
243	<
244	<	/**
245	<	* Get and clear steal count for accumulation by pool.  Called
246	<	* only when known to be idle (in pool.sync and termination).
247	<	*/
248	<	final int getAndClearStealCount() {
249	<	int sc = stealCount;
250	<	stealCount = 0;
251	<	return sc;
308	>	this.locallyFifo = pool.locallyFifo;
309	>	setDaemon(true);
310	>	// To avoid exposing construction details to subclasses,
311	>	// remaining initialization is in start() and onStart()
312		}
313
314		/**
315	<	* Returns estimate of the number of tasks in the queue, without
256	<	* correcting for transient negative values
315	>	* Performs additional initialization and starts this thread.
316		*/
317	<	final int getRawQueueSize() {
318	<	return sp - base;
317	>	final void start(int poolIndex, UncaughtExceptionHandler ueh) {
318	>	this.poolIndex = poolIndex;
319	>	if (ueh != null)
320	>	setUncaughtExceptionHandler(ueh);
321	>	start();
322		}
323
324	<	// Intrinsics-based support for queue operations.
263	<	// Currently these three (setSp, setSlot, casSlotNull) are
264	<	// usually manually inlined to improve performance
324	>	// Public/protected methods
325
326		/**
327	<	* Sets sp in store-order.
327	>	* Returns the pool hosting this thread.
328	>	*
329	>	* @return the pool
330		*/
331	<	private void setSp(int s) {
332	<	_unsafe.putOrderedInt(this, spOffset, s);
331	>	public ForkJoinPool getPool() {
332	>	return pool;
333		}
334
335		/**
336	<	* Add in store-order the given task at given slot of q to
337	<	* null. Caller must ensure q is nonnull and index is in range.
336	>	* Returns the index number of this thread in its pool.  The
337	>	* returned value ranges from zero to the maximum number of
338	>	* threads (minus one) that have ever been created in the pool.
339	>	* This method may be useful for applications that track status or
340	>	* collect results per-worker rather than per-task.
341	>	*
342	>	* @return the index number
343		*/
344	<	private static void setSlot(ForkJoinTask<?>[] q, int i,
345	<	ForkJoinTask<?> t){
279	<	_unsafe.putOrderedObject(q, (i << qShift) + qBase, t);
344	>	public int getPoolIndex() {
345	>	return poolIndex;
346		}
347
348		/**
349	<	* CAS given slot of q to null. Caller must ensure q is nonnull
350	<	* and index is in range.
349	>	* Initializes internal state after construction but before
350	>	* processing any tasks. If you override this method, you must
351	>	* invoke @code{super.onStart()} at the beginning of the method.
352	>	* Initialization requires care: Most fields must have legal
353	>	* default values, to ensure that attempted accesses from other
354	>	* threads work correctly even before this thread starts
355	>	* processing tasks.
356		*/
357	<	private static boolean casSlotNull(ForkJoinTask<?>[] q, int i,
358	<	ForkJoinTask<?> t) {
359	<	return _unsafe.compareAndSwapObject(q, (i << qShift) + qBase, t, null);
360	<	}
357	>	protected void onStart() {
358	>	int rs = seedGenerator.nextInt();
359	>	seed = rs == 0? 1 : rs; // seed must be nonzero
360	>
361	>	// Allocate name string and arrays in this thread
362	>	String pid = Integer.toString(pool.getPoolNumber());
363	>	String wid = Integer.toString(poolIndex);
364	>	setName("ForkJoinPool-" + pid + "-worker-" + wid);
365
366	<	// Main queue methods
366	>	queue = new ForkJoinTask<?>[INITIAL_QUEUE_CAPACITY];
367	>	}
368
369		/**
370	<	* Pushes a task. Called only by current thread.
371	<	* @param t the task. Caller must ensure nonnull
370	>	* Performs cleanup associated with termination of this worker
371	>	* thread.  If you override this method, you must invoke
372	>	* {@code super.onTermination} at the end of the overridden method.
373	>	*
374	>	* @param exception the exception causing this thread to abort due
375	>	* to an unrecoverable error, or {@code null} if completed normally
376		*/
377	<	final void pushTask(ForkJoinTask<?> t) {
378	<	ForkJoinTask<?>[] q = queue;
379	<	int mask = q.length - 1;
380	<	int s = sp;
381	<	_unsafe.putOrderedObject(q, ((s & mask) << qShift) + qBase, t);
382	<	_unsafe.putOrderedInt(this, spOffset, ++s);
383	<	if ((s -= base) == 1)
384	<	pool.signalNonEmptyWorkerQueue();
385	<	else if (s >= mask)
386	<	growQueue();
377	>	protected void onTermination(Throwable exception) {
378	>	try {
379	>	ForkJoinPool p = pool;
380	>	if (active) {
381	>	int a; // inline p.tryDecrementActiveCount
382	>	active = false;
383	>	do {} while (!UNSAFE.compareAndSwapInt
384	>	(p, poolRunStateOffset, a = p.runState, a - 1));
385	>	}
386	>	cancelTasks();
387	>	setTerminated();
388	>	p.workerTerminated(this);
389	>	} catch (Throwable ex) {        // Shouldn't ever happen
390	>	if (exception == null)      // but if so, at least rethrown
391	>	exception = ex;
392	>	} finally {
393	>	if (exception != null)
394	>	UNSAFE.throwException(exception);
395	>	}
396		}
397
398		/**
399	<	* Tries to take a task from the base of the queue, failing if
400	<	* either empty or contended.
401	<	* @return a task, or null if none or contended.
399	>	* This method is required to be public, but should never be
400	>	* called explicitly. It performs the main run loop to execute
401	>	* ForkJoinTasks.
402		*/
403	<	private ForkJoinTask<?> deqTask() {
404	<	ForkJoinTask<?>[] q;
405	<	ForkJoinTask<?> t;
406	<	int i;
407	<	int b;
408	<	if (sp != (b = base) &&
409	<	(q = queue) != null && // must read q after b
410	<	(t = q[i = (q.length - 1) & b]) != null &&
411	<	_unsafe.compareAndSwapObject(q, (i << qShift) + qBase, t, null)) {
323	<	base = b + 1;
324	<	return t;
403	>	public void run() {
404	>	Throwable exception = null;
405	>	try {
406	>	onStart();
407	>	mainLoop();
408	>	} catch (Throwable ex) {
409	>	exception = ex;
410	>	} finally {
411	>	onTermination(exception);
412		}
326	–	return null;
413		}
414
415	+	// helpers for run()
416	+
417		/**
418	<	* Returns a popped task, or null if empty.  Called only by
331	<	* current thread.
418	>	* Finds and executes tasks, and checks status while running.
419		*/
420	<	final ForkJoinTask<?> popTask() {
421	<	ForkJoinTask<?> t;
422	<	int i;
423	<	ForkJoinTask<?>[] q = queue;
424	<	int mask = q.length - 1;
425	<	int s = sp;
426	<	if (s != base &&
427	<	(t = q[i = (s - 1) & mask]) != null &&
341	<	_unsafe.compareAndSwapObject(q, (i << qShift) + qBase, t, null)) {
342	<	_unsafe.putOrderedInt(this, spOffset, s - 1);
343	<	return t;
420	>	private void mainLoop() {
421	>	boolean ran = false; // true if ran a task on last step
422	>	ForkJoinPool p = pool;
423	>	for (;;) {
424	>	p.preStep(this, ran);
425	>	if (runState != 0)
426	>	break;
427	>	ran = tryExecSteal() || tryExecSubmission();
428		}
345	–	return null;
429		}
430
431		/**
432	<	* Specialized version of popTask to pop only if
433	<	* topmost element is the given task. Called only
434	<	* by current thread.
352	<	* @param t the task. Caller must ensure nonnull
432	>	* Tries to steal a task and execute it.
433	>	*
434	>	* @return true if ran a task
435		*/
436	<	final boolean unpushTask(ForkJoinTask<?> t) {
437	<	ForkJoinTask<?>[] q = queue;
438	<	int mask = q.length - 1;
439	<	int s = sp - 1;
440	<	if (_unsafe.compareAndSwapObject(q, ((s & mask) << qShift) + qBase,
441	<	t, null)) {
442	<	_unsafe.putOrderedInt(this, spOffset, s);
436	>	private boolean tryExecSteal() {
437	>	ForkJoinTask<?> t;
438	>	if ((t = scan()) != null) {
439	>	t.quietlyExec();
440	>	UNSAFE.putOrderedObject(this, currentStealOffset, null);
441	>	if (sp != base)
442	>	execLocalTasks();
443		return true;
444		}
445		return false;
446		}
447
448		/**
449	<	* Returns next task to pop.
449	>	* If a submission exists, try to activate and run it.
450	>	*
451	>	* @return true if ran a task
452		*/
453	<	private ForkJoinTask<?> peekTask() {
454	<	ForkJoinTask<?>[] q = queue;
455	<	return q == null? null : q[(sp - 1) & (q.length - 1)];
453	>	private boolean tryExecSubmission() {
454	>	ForkJoinPool p = pool;
455	>	// This loop is needed in case attempt to activate fails, in
456	>	// which case we only retry if there still appears to be a
457	>	// submission.
458	>	while (p.hasQueuedSubmissions()) {
459	>	ForkJoinTask<?> t; int a;
460	>	if (active || // inline p.tryIncrementActiveCount
461	>	(active = UNSAFE.compareAndSwapInt(p, poolRunStateOffset,
462	>	a = p.runState, a + 1))) {
463	>	if ((t = p.pollSubmission()) != null) {
464	>	UNSAFE.putOrderedObject(this, currentStealOffset, t);
465	>	t.quietlyExec();
466	>	UNSAFE.putOrderedObject(this, currentStealOffset, null);
467	>	if (sp != base)
468	>	execLocalTasks();
469	>	return true;
470	>	}
471	>	}
472	>	}
473	>	return false;
474		}
475
476		/**
477	<	* Doubles queue array size. Transfers elements by emulating
478	<	* steals (deqs) from old array and placing, oldest first, into
377	<	* new array.
477	>	* Runs local tasks until queue is empty or shut down.  Call only
478	>	* while active.
479		*/
480	<	private void growQueue() {
481	<	ForkJoinTask<?>[] oldQ = queue;
482	<	int oldSize = oldQ.length;
483	<	int newSize = oldSize << 1;
484	<	if (newSize > MAXIMUM_QUEUE_CAPACITY)
485	<	throw new RejectedExecutionException("Queue capacity exceeded");
486	<	ForkJoinTask<?>[] newQ = queue = new ForkJoinTask<?>[newSize];
487	<
387	<	int b = base;
388	<	int bf = b + oldSize;
389	<	int oldMask = oldSize - 1;
390	<	int newMask = newSize - 1;
391	<	do {
392	<	int oldIndex = b & oldMask;
393	<	ForkJoinTask<?> t = oldQ[oldIndex];
394	<	if (t != null && !casSlotNull(oldQ, oldIndex, t))
395	<	t = null;
396	<	setSlot(newQ, b & newMask, t);
397	<	} while (++b != bf);
398	<	pool.signalIdleWorkers(false);
480	>	private void execLocalTasks() {
481	>	while (runState == 0) {
482	>	ForkJoinTask<?> t = locallyFifo ? locallyDeqTask() : popTask();
483	>	if (t != null)
484	>	t.quietlyExec();
485	>	else if (sp == base)
486	>	break;
487	>	}
488		}
489
490	<	// Runstate management
491	<
492	<	final boolean isShutdown()    { return runState >= SHUTDOWN;  }
493	<	final boolean isTerminating() { return runState >= TERMINATING;  }
494	<	final boolean isTerminated()  { return runState == TERMINATED; }
495	<	final boolean shutdown()      { return transitionRunStateTo(SHUTDOWN); }
496	<	final boolean shutdownNow()   { return transitionRunStateTo(TERMINATING); }
497	<
498	<	/**
499	<	* Transition to at least the given state. Return true if not
500	<	* already at least given state.
490	>	/*
491	>	* Intrinsics-based atomic writes for queue slots. These are
492	>	* basically the same as methods in AtomicReferenceArray, but
493	>	* specialized for (1) ForkJoinTask elements (2) requirement that
494	>	* nullness and bounds checks have already been performed by
495	>	* callers and (3) effective offsets are known not to overflow
496	>	* from int to long (because of MAXIMUM_QUEUE_CAPACITY). We don't
497	>	* need corresponding version for reads: plain array reads are OK
498	>	* because they are protected by other volatile reads and are
499	>	* confirmed by CASes.
500	>	*
501	>	* Most uses don't actually call these methods, but instead contain
502	>	* inlined forms that enable more predictable optimization.  We
503	>	* don't define the version of write used in pushTask at all, but
504	>	* instead inline there a store-fenced array slot write.
505		*/
413	–	private boolean transitionRunStateTo(int state) {
414	–	for (;;) {
415	–	int s = runState;
416	–	if (s >= state)
417	–	return false;
418	–	if (_unsafe.compareAndSwapInt(this, runStateOffset, s, state))
419	–	return true;
420	–	}
421	–	}
506
507		/**
508	<	* Ensure status is active and if necessary adjust pool active count
508	>	* CASes slot i of array q from t to null. Caller must ensure q is
509	>	* non-null and index is in range.
510		*/
511	<	final void activate() {
512	<	if (!active) {
513	<	active = true;
429	<	pool.incrementActiveCount();
430	<	}
511	>	private static final boolean casSlotNull(ForkJoinTask<?>[] q, int i,
512	>	ForkJoinTask<?> t) {
513	>	return UNSAFE.compareAndSwapObject(q, (i << qShift) + qBase, t, null);
514		}
515
516		/**
517	<	* Ensure status is inactive and if necessary adjust pool active count
517	>	* Performs a volatile write of the given task at given slot of
518	>	* array q.  Caller must ensure q is non-null and index is in
519	>	* range. This method is used only during resets and backouts.
520		*/
521	<	final void inactivate() {
522	<	if (active) {
523	<	active = false;
439	<	pool.decrementActiveCount();
440	<	}
521	>	private static final void writeSlot(ForkJoinTask<?>[] q, int i,
522	>	ForkJoinTask<?> t) {
523	>	UNSAFE.putObjectVolatile(q, (i << qShift) + qBase, t);
524		}
525
526	<	// Lifecycle methods
526	>	// queue methods
527
528		/**
529	<	* Initializes internal state after construction but before
530	<	* processing any tasks. If you override this method, you must
531	<	* invoke super.onStart() at the beginning of the method.
449	<	* Initialization requires care: Most fields must have legal
450	<	* default values, to ensure that attempted accesses from other
451	<	* threads work correctly even before this thread starts
452	<	* processing tasks.
529	>	* Pushes a task. Call only from this thread.
530	>	*
531	>	* @param t the task. Caller must ensure non-null.
532		*/
533	<	protected void onStart() {
534	<	juRandomSeed = randomSeedGenerator.nextLong();
535	<	do;while((randomVictimSeed = nextRandomInt()) == 0); // must be nonzero
536	<	if (queue == null)
537	<	queue = new ForkJoinTask<?>[INITIAL_QUEUE_CAPACITY];
538	<
539	<	// Heuristically allow one initial thread to warm up; others wait
540	<	if (poolIndex < pool.getParallelism() - 1) {
541	<	eventCount = pool.sync(this, 0);
463	<	activate();
464	<	}
533	>	final void pushTask(ForkJoinTask<?> t) {
534	>	ForkJoinTask<?>[] q = queue;
535	>	int mask = q.length - 1; // implicit assert q != null
536	>	int s = sp++;            // ok to increment sp before slot write
537	>	UNSAFE.putOrderedObject(q, ((s & mask) << qShift) + qBase, t);
538	>	if ((s -= base) == 0)
539	>	pool.signalWork();   // was empty
540	>	else if (s == mask)
541	>	growQueue();         // is full
542		}
543
544		/**
545	<	* Perform cleanup associated with termination of this worker
546	<	* thread.  If you override this method, you must invoke
547	<	* super.onTermination at the end of the overridden method.
545	>	* Tries to take a task from the base of the queue, failing if
546	>	* empty or contended. Note: Specializations of this code appear
547	>	* in locallyDeqTask and elsewhere.
548		*
549	<	* @param exception the exception causing this thread to abort due
473	<	* to an unrecoverable error, or null if completed normally.
549	>	* @return a task, or null if none or contended
550		*/
551	<	protected void onTermination(Throwable exception) {
552	<	try {
553	<	clearLocalTasks();
554	<	inactivate();
555	<	cancelTasks();
556	<	} finally {
557	<	terminate(exception);
551	>	final ForkJoinTask<?> deqTask() {
552	>	ForkJoinTask<?> t;
553	>	ForkJoinTask<?>[] q;
554	>	int b, i;
555	>	if (sp != (b = base) &&
556	>	(q = queue) != null && // must read q after b
557	>	(t = q[i = (q.length - 1) & b]) != null && base == b &&
558	>	UNSAFE.compareAndSwapObject(q, (i << qShift) + qBase, t, null)) {
559	>	base = b + 1;
560	>	return t;
561		}
562	+	return null;
563		}
564
565		/**
566	<	* Notify pool of termination and, if exception is nonnull,
567	<	* rethrow it to trigger this thread's uncaughtExceptionHandler
566	>	* Tries to take a task from the base of own queue. Assumes active
567	>	* status.  Called only by this thread.
568	>	*
569	>	* @return a task, or null if none
570		*/
571	<	private void terminate(Throwable exception) {
572	<	transitionRunStateTo(TERMINATED);
573	<	try {
574	<	pool.workerTerminated(this);
575	<	} finally {
576	<	if (exception != null)
577	<	ForkJoinTask.rethrowException(exception);
571	>	final ForkJoinTask<?> locallyDeqTask() {
572	>	ForkJoinTask<?>[] q = queue;
573	>	if (q != null) {
574	>	ForkJoinTask<?> t;
575	>	int b, i;
576	>	while (sp != (b = base)) {
577	>	if ((t = q[i = (q.length - 1) & b]) != null && base == b &&
578	>	UNSAFE.compareAndSwapObject(q, (i << qShift) + qBase,
579	>	t, null)) {
580	>	base = b + 1;
581	>	return t;
582	>	}
583	>	}
584		}
585	+	return null;
586		}
587
588		/**
589	<	* Run local tasks on exit from main.
589	>	* Returns a popped task, or null if empty. Assumes active status.
590	>	* Called only by this thread.
591		*/
592	<	private void clearLocalTasks() {
593	<	while (base != sp && !pool.isTerminating()) {
594	<	ForkJoinTask<?> t = popTask();
595	<	if (t != null) {
596	<	activate(); // ensure active status
597	<	t.quietlyExec();
592	>	private ForkJoinTask<?> popTask() {
593	>	ForkJoinTask<?>[] q = queue;
594	>	if (q != null) {
595	>	int s;
596	>	while ((s = sp) != base) {
597	>	int i = (q.length - 1) & --s;
598	>	long u = (i << qShift) + qBase; // raw offset
599	>	ForkJoinTask<?> t = q[i];
600	>	if (t == null)   // lost to stealer
601	>	break;
602	>	if (UNSAFE.compareAndSwapObject(q, u, t, null)) {
603	>	sp = s; // putOrderedInt may encourage more timely write
604	>	// UNSAFE.putOrderedInt(this, spOffset, s);
605	>	return t;
606	>	}
607		}
608		}
609	+	return null;
610		}
611
612		/**
613	<	* Removes and cancels all tasks in queue.  Can be called from any
614	<	* thread.
613	>	* Specialized version of popTask to pop only if topmost element
614	>	* is the given task. Called only by this thread while active.
615	>	*
616	>	* @param t the task. Caller must ensure non-null.
617		*/
618	<	final void cancelTasks() {
619	<	while (base != sp) {
620	<	ForkJoinTask<?> t = deqTask();
621	<	if (t != null)
622	<	t.cancelIgnoreExceptions();
618	>	final boolean unpushTask(ForkJoinTask<?> t) {
619	>	int s;
620	>	ForkJoinTask<?>[] q = queue;
621	>	if ((s = sp) != base && q != null &&
622	>	UNSAFE.compareAndSwapObject
623	>	(q, (((q.length - 1) & --s) << qShift) + qBase, t, null)) {
624	>	sp = s; // putOrderedInt may encourage more timely write
625	>	// UNSAFE.putOrderedInt(this, spOffset, s);
626	>	return true;
627		}
628	+	return false;
629		}
630
631		/**
632	<	* This method is required to be public, but should never be
526	<	* called explicitly. It performs the main run loop to execute
527	<	* ForkJoinTasks.
632	>	* Returns next task, or null if empty or contended.
633		*/
634	<	public void run() {
635	<	Throwable exception = null;
636	<	try {
637	<	onStart();
638	<	while (!isShutdown())
639	<	step();
640	<	} catch (Throwable ex) {
536	<	exception = ex;
537	<	} finally {
538	<	onTermination(exception);
539	<	}
634	>	final ForkJoinTask<?> peekTask() {
635	>	ForkJoinTask<?>[] q = queue;
636	>	if (q == null)
637	>	return null;
638	>	int mask = q.length - 1;
639	>	int i = locallyFifo ? base : (sp - 1);
640	>	return q[i & mask];
641		}
642
643		/**
644	<	* Main top-level action.
644	>	* Doubles queue array size. Transfers elements by emulating
645	>	* steals (deqs) from old array and placing, oldest first, into
646	>	* new array.
647		*/
648	<	private void step() {
649	<	ForkJoinTask<?> t = sp != base? popTask() : null;
650	<	if (t != null || (t = scan(null, true)) != null) {
651	<	activate();
652	<	t.quietlyExec();
653	<	}
654	<	else {
552	<	inactivate();
553	<	eventCount = pool.sync(this, eventCount);
554	<	}
555	<	}
648	>	private void growQueue() {
649	>	ForkJoinTask<?>[] oldQ = queue;
650	>	int oldSize = oldQ.length;
651	>	int newSize = oldSize << 1;
652	>	if (newSize > MAXIMUM_QUEUE_CAPACITY)
653	>	throw new RejectedExecutionException("Queue capacity exceeded");
654	>	ForkJoinTask<?>[] newQ = queue = new ForkJoinTask<?>[newSize];
655
656	<	// scanning for and stealing tasks
656	>	int b = base;
657	>	int bf = b + oldSize;
658	>	int oldMask = oldSize - 1;
659	>	int newMask = newSize - 1;
660	>	do {
661	>	int oldIndex = b & oldMask;
662	>	ForkJoinTask<?> t = oldQ[oldIndex];
663	>	if (t != null && !casSlotNull(oldQ, oldIndex, t))
664	>	t = null;
665	>	writeSlot(newQ, b & newMask, t);
666	>	} while (++b != bf);
667	>	pool.signalWork();
668	>	}
669
670		/**
671	<	* Computes next value for random victim probe. Scans don't
672	<	* require a very high quality generator, but also not a crummy
673	<	* one. Marsaglia xor-shift is cheap and works well.
674	<	*
564	<	* This is currently unused, and manually inlined
671	>	* Computes next value for random victim probe in scan().  Scans
672	>	* don't require a very high quality generator, but also not a
673	>	* crummy one.  Marsaglia xor-shift is cheap and works well enough.
674	>	* Note: This is manually inlined in scan().
675		*/
676	<	private static int xorShift(int r) {
677	<	r ^= r << 1;
678	<	r ^= r >>> 3;
679	<	r ^= r << 10;
570	<	return r;
676	>	private static final int xorShift(int r) {
677	>	r ^= r << 13;
678	>	r ^= r >>> 17;
679	>	return r ^ (r << 5);
680		}
681
682		/**
683	<	* Tries to steal a task from another worker and/or, if enabled,
684	<	* submission queue. Starts at a random index of workers array,
685	<	* and probes workers until finding one with non-empty queue or
686	<	* finding that all are empty.  It randomly selects the first n-1
687	<	* probes. If these are empty, it resorts to full circular
688	<	* traversal, which is necessary to accurately set active status
689	<	* by caller. Also restarts if pool barrier has tripped since last
690	<	* scan, which forces refresh of workers array, in case barrier
582	<	* was associated with resize.
683	>	* Tries to steal a task from another worker. Starts at a random
684	>	* index of workers array, and probes workers until finding one
685	>	* with non-empty queue or finding that all are empty.  It
686	>	* randomly selects the first n probes. If these are empty, it
687	>	* resorts to a circular sweep, which is necessary to accurately
688	>	* set active status. (The circular sweep uses steps of
689	>	* approximately half the array size plus 1, to avoid bias
690	>	* stemming from leftmost packing of the array in ForkJoinPool.)
691		*
692		* This method must be both fast and quiet -- usually avoiding
693		* memory accesses that could disrupt cache sharing etc other than
694	<	* those needed to check for and take tasks. This accounts for,
695	<	* among other things, updating random seed in place without
696	<	* storing it until exit. (Note that we only need to store it if
589	<	* we found a task; otherwise it doesn't matter if we start at the
590	<	* same place next time.)
694	>	* those needed to check for and take tasks (or to activate if not
695	>	* already active). This accounts for, among other things,
696	>	* updating random seed in place without storing it until exit.
697		*
592	–	* @param joinMe if non null; exit early if done
593	–	* @param checkSubmissions true if OK to take submissions
698		* @return a task, or null if none found
699		*/
700	<	private ForkJoinTask<?> scan(ForkJoinTask<?> joinMe,
597	<	boolean checkSubmissions) {
700	>	private ForkJoinTask<?> scan() {
701		ForkJoinPool p = pool;
702	<	if (p == null)                    // Never null, but avoids
703	<	return null;                  //   implicit nullchecks below
704	<	int r = randomVictimSeed;         // extract once to keep scan quiet
705	<	restart:                          // outer loop refreshes ws array
706	<	while (joinMe == null || joinMe.status >= 0) {
707	<	int mask;
708	<	ForkJoinWorkerThread[] ws = p.workers;
709	<	if (ws != null && (mask = ws.length - 1) > 0) {
710	<	int probes = -mask;       // use random index while negative
711	<	int idx = r;
712	<	for (;;) {
713	<	ForkJoinWorkerThread v;
714	<	// inlined xorshift to update seed
715	<	r ^= r << 1;  r ^= r >>> 3; r ^= r << 10;
716	<	if ((v = ws[mask & idx]) != null && v.sp != v.base) {
717	<	ForkJoinTask<?> t;
718	<	activate();
719	<	if ((joinMe == null || joinMe.status >= 0) &&
720	<	(t = v.deqTask()) != null) {
721	<	randomVictimSeed = r;
702	>	ForkJoinWorkerThread[] ws;        // worker array
703	>	int n;                            // upper bound of #workers
704	>	if ((ws = p.workers) != null && (n = ws.length) > 1) {
705	>	boolean canSteal = active;    // shadow active status
706	>	int r = seed;                 // extract seed once
707	>	int mask = n - 1;
708	>	int j = -n;                   // loop counter
709	>	int k = r;                    // worker index, random if j < 0
710	>	for (;;) {
711	>	ForkJoinWorkerThread v = ws[k & mask];
712	>	r ^= r << 13; r ^= r >>> 17; r ^= r << 5; // inline xorshift
713	>	ForkJoinTask<?>[] q; ForkJoinTask<?> t; int b, a;
714	>	if (v != null && (b = v.base) != v.sp &&
715	>	(q = v.queue) != null) {
716	>	int i = (q.length - 1) & b;
717	>	long u = (i << qShift) + qBase; // raw offset
718	>	int pid = poolIndex;
719	>	if ((t = q[i]) != null) {
720	>	if (!canSteal &&  // inline p.tryIncrementActiveCount
721	>	UNSAFE.compareAndSwapInt(p, poolRunStateOffset,
722	>	a = p.runState, a + 1))
723	>	canSteal = active = true;
724	>	if (canSteal && v.base == b++ &&
725	>	UNSAFE.compareAndSwapObject(q, u, t, null)) {
726	>	v.base = b;
727	>	v.stealHint = pid;
728	>	UNSAFE.putOrderedObject(this,
729	>	currentStealOffset, t);
730	>	seed = r;
731		++stealCount;
732		return t;
733		}
622	–	continue restart; // restart on contention
734		}
735	<	if ((probes >> 1) <= mask) // n-1 random then circular
736	<	idx = (probes++ < 0)? r : (idx + 1);
626	<	else
627	<	break;
735	>	j = -n;
736	>	k = r;                // restart on contention
737		}
738	<	}
739	<	if (checkSubmissions && p.hasQueuedSubmissions()) {
740	<	activate();
741	<	ForkJoinTask<?> t = p.pollSubmission();
742	<	if (t != null)
634	<	return t;
635	<	}
636	<	else {
637	<	long ec = eventCount;     // restart on pool event
638	<	if ((eventCount = p.getEventCount()) == ec)
738	>	else if (++j <= 0)
739	>	k = r;
740	>	else if (j <= n)
741	>	k += (n >>> 1) | 1;
742	>	else
743		break;
744		}
745		}
746		return null;
747		}
748
749	<	/**
750	<	* Callback from pool.sync to rescan before blocking.  If a
751	<	* task is found, it is pushed so it can be executed upon return.
752	<	* @return true if found and pushed a task
753	<	*/
754	<	final boolean prescan() {
755	<	ForkJoinTask<?> t = scan(null, true);
756	<	if (t != null) {
757	<	pushTask(t);
749	>	// Run State management
750	>
751	>	// status check methods used mainly by ForkJoinPool
752	>	final boolean isRunning()    { return runState == 0; }
753	>	final boolean isTerminated() { return (runState & TERMINATED) != 0; }
754	>	final boolean isSuspended()  { return (runState & SUSPENDED) != 0; }
755	>	final boolean isTrimmed()    { return (runState & TRIMMED) != 0; }
756	>
757	>	final boolean isTerminating() {
758	>	if ((runState & TERMINATING) != 0)
759	>	return true;
760	>	if (pool.isAtLeastTerminating()) { // propagate pool state
761	>	shutdown();
762		return true;
763		}
764	<	else {
657	<	inactivate();
658	<	return false;
659	<	}
764	>	return false;
765		}
766
767		/**
768	<	* Implements ForkJoinTask.helpJoin
768	>	* Sets state to TERMINATING. Does NOT unpark or interrupt
769	>	* to wake up if currently blocked. Callers must do so if desired.
770		*/
771	<	final int helpJoinTask(ForkJoinTask<?> joinMe) {
772	<	ForkJoinTask<?> t = null;
773	<	int s;
774	<	while ((s = joinMe.status) >= 0) {
775	<	if (t == null) {
776	<	if ((t = scan(joinMe, false)) == null)  // block if no work
777	<	return joinMe.awaitDone(this, false);
778	<	// else recheck status before exec
779	<	}
780	<	else {
675	<	t.quietlyExec();
676	<	t = null;
771	>	final void shutdown() {
772	>	for (;;) {
773	>	int s = runState;
774	>	if ((s & (TERMINATING|TERMINATED)) != 0)
775	>	break;
776	>	if ((s & SUSPENDED) != 0) { // kill and wakeup if suspended
777	>	if (UNSAFE.compareAndSwapInt(this, runStateOffset, s,
778	>	(s & ~SUSPENDED) |
779	>	(TRIMMED|TERMINATING)))
780	>	break;
781		}
782	+	else if (UNSAFE.compareAndSwapInt(this, runStateOffset, s,
783	+	s | TERMINATING))
784	+	break;
785		}
679	–	if (t != null) // unsteal
680	–	pushTask(t);
681	–	return s;
682	–	}
683	–
684	–	// Support for public static and/or ForkJoinTask methods
685	–
686	–	/**
687	–	* Returns an estimate of the number of tasks in the queue.
688	–	*/
689	–	final int getQueueSize() {
690	–	int b = base;
691	–	int n = sp - b;
692	–	return n <= 0? 0 : n; // suppress momentarily negative values
693	–	}
694	–
695	–	/**
696	–	* Runs one popped task, if available
697	–	* @return true if ran a task
698	–	*/
699	–	private boolean runLocalTask() {
700	–	ForkJoinTask<?> t = popTask();
701	–	if (t == null)
702	–	return false;
703	–	t.quietlyExec();
704	–	return true;
786		}
787
788		/**
789	<	* Pops or steals a task
709	<	* @return task, or null if none available
789	>	* Sets state to TERMINATED. Called only by onTermination().
790		*/
791	<	private ForkJoinTask<?> getLocalOrStolenTask() {
792	<	ForkJoinTask<?> t = popTask();
793	<	return t != null? t : scan(null, false);
791	>	private void setTerminated() {
792	>	int s;
793	>	do {} while (!UNSAFE.compareAndSwapInt(this, runStateOffset,
794	>	s = runState,
795	>	s | (TERMINATING|TERMINATED)));
796		}
797
798		/**
799	<	* Runs a popped or stolen task, if available
800	<	* @return true if ran a task
799	>	* If suspended, tries to set status to unsuspended.
800	>	* Does NOT wake up if blocked.
801	>	*
802	>	* @return true if successful
803		*/
804	<	private boolean runLocalOrStolenTask() {
805	<	ForkJoinTask<?> t = getLocalOrStolenTask();
806	<	if (t == null)
807	<	return false;
808	<	t.quietlyExec();
809	<	return true;
804	>	final boolean tryUnsuspend() {
805	>	int s;
806	>	while (((s = runState) & SUSPENDED) != 0) {
807	>	if (UNSAFE.compareAndSwapInt(this, runStateOffset, s,
808	>	s & ~SUSPENDED))
809	>	return true;
810	>	}
811	>	return false;
812		}
813
814		/**
815	<	* Runs tasks until pool isQuiescent
815	>	* Sets suspended status and blocks as spare until resumed
816	>	* or shutdown.
817		*/
818	<	final void helpQuiescePool() {
819	<	activate();
820	<	for (;;) {
821	<	if (!runLocalOrStolenTask()) {
822	<	inactivate();
823	<	if (pool.isQuiescent()) {
824	<	activate(); // re-activate on exit
818	>	final void suspendAsSpare() {
819	>	for (;;) {                  // set suspended unless terminating
820	>	int s = runState;
821	>	if ((s & TERMINATING) != 0) { // must kill
822	>	if (UNSAFE.compareAndSwapInt(this, runStateOffset, s,
823	>	s | (TRIMMED | TERMINATING)))
824	>	return;
825	>	}
826	>	else if (UNSAFE.compareAndSwapInt(this, runStateOffset, s,
827	>	s | SUSPENDED))
828	>	break;
829	>	}
830	>	ForkJoinPool p = pool;
831	>	p.pushSpare(this);
832	>	while ((runState & SUSPENDED) != 0) {
833	>	if (p.tryAccumulateStealCount(this)) {
834	>	interrupted();          // clear/ignore interrupts
835	>	if ((runState & SUSPENDED) == 0)
836		break;
837	<	}
837	>	LockSupport.park(this);
838		}
839		}
840		}
841
842	+	// Misc support methods for ForkJoinPool
843	+
844		/**
845	<	* Returns an estimate of the number of tasks, offset by a
846	<	* function of number of idle workers.
845	>	* Returns an estimate of the number of tasks in the queue.  Also
846	>	* used by ForkJoinTask.
847		*/
848	<	final int getEstimatedSurplusTaskCount() {
849	<	return (sp - base) - (pool.getIdleThreadCount() >>> 1);
848	>	final int getQueueSize() {
849	>	int n; // external calls must read base first
850	>	return (n = -base + sp) <= 0 ? 0 : n;
851		}
852
752	–	// Public methods on current thread
753	–
853		/**
854	<	* Returns the pool hosting the current task execution.
855	<	* @return the pool
854	>	* Removes and cancels all tasks in queue.  Can be called from any
855	>	* thread.
856		*/
857	<	public static ForkJoinPool getPool() {
858	<	return ((ForkJoinWorkerThread)(Thread.currentThread())).pool;
857	>	final void cancelTasks() {
858	>	ForkJoinTask<?> cj = currentJoin; // try to cancel ongoing tasks
859	>	if (cj != null) {
860	>	currentJoin = null;
861	>	cj.cancelIgnoringExceptions();
862	>	try {
863	>	this.interrupt(); // awaken wait
864	>	} catch (SecurityException ignore) {
865	>	}
866	>	}
867	>	ForkJoinTask<?> cs = currentSteal;
868	>	if (cs != null) {
869	>	currentSteal = null;
870	>	cs.cancelIgnoringExceptions();
871	>	}
872	>	while (base != sp) {
873	>	ForkJoinTask<?> t = deqTask();
874	>	if (t != null)
875	>	t.cancelIgnoringExceptions();
876	>	}
877		}
878
879		/**
880	<	* Returns the index number of the current worker thread in its
881	<	* pool.  The returned value ranges from zero to the maximum
882	<	* number of threads (minus one) that have ever been created in
766	<	* the pool.  This method may be useful for applications that
767	<	* track status or collect results per-worker rather than
768	<	* per-task.
769	<	* @return the index number.
880	>	* Drains tasks to given collection c.
881	>	*
882	>	* @return the number of tasks drained
883		*/
884	<	public static int getPoolIndex() {
885	<	return ((ForkJoinWorkerThread)(Thread.currentThread())).poolIndex;
884	>	final int drainTasksTo(Collection<? super ForkJoinTask<?>> c) {
885	>	int n = 0;
886	>	while (base != sp) {
887	>	ForkJoinTask<?> t = deqTask();
888	>	if (t != null) {
889	>	c.add(t);
890	>	++n;
891	>	}
892	>	}
893	>	return n;
894		}
895
896	+	// Support methods for ForkJoinTask
897	+
898		/**
899	<	* Returns an estimate of the number of tasks waiting to be run by
900	<	* the current worker thread. This value may be useful for
901	<	* heuristic decisions about whether to fork other tasks.
779	<	* @return the number of tasks
899	>	* Gets and removes a local task.
900	>	*
901	>	* @return a task, if available
902		*/
903	<	public static int getLocalQueueSize() {
904	<	return ((ForkJoinWorkerThread)(Thread.currentThread())).
905	<	getQueueSize();
903	>	final ForkJoinTask<?> pollLocalTask() {
904	>	ForkJoinPool p = pool;
905	>	while (sp != base) {
906	>	int a; // inline p.tryIncrementActiveCount
907	>	if (active ||
908	>	(active = UNSAFE.compareAndSwapInt(p, poolRunStateOffset,
909	>	a = p.runState, a + 1)))
910	>	return locallyFifo ? locallyDeqTask() : popTask();
911	>	}
912	>	return null;
913		}
914
915		/**
916	<	* Returns, but does not remove or execute, the next task locally
917	<	* queued for execution by the current worker thread. There is no
918	<	* guarantee that this task will be the next one actually returned
790	<	* or executed from other polling or execution methods.
791	<	* @return the next task or null if none
916	>	* Gets and removes a local or stolen task.
917	>	*
918	>	* @return a task, if available
919		*/
920	<	public static ForkJoinTask<?> peekLocalTask() {
921	<	return ((ForkJoinWorkerThread)(Thread.currentThread())).peekTask();
920	>	final ForkJoinTask<?> pollTask() {
921	>	ForkJoinTask<?> t = pollLocalTask();
922	>	if (t == null) {
923	>	t = scan();
924	>	// cannot retain/track/help steal
925	>	UNSAFE.putOrderedObject(this, currentStealOffset, null);
926	>	}
927	>	return t;
928		}
929
930		/**
931	<	* Removes and returns, without executing, the next task queued
932	<	* for execution in the current worker thread's local queue.
933	<	* @return the next task to execute, or null if none
934	<	*/
935	<	public static ForkJoinTask<?> pollLocalTask() {
936	<	return ((ForkJoinWorkerThread)(Thread.currentThread())).popTask();
931	>	* Possibly runs some tasks and/or blocks, until task is done.
932	>	*
933	>	* @param joinMe the task to join
934	>	* @param timed true if use timed wait
935	>	* @param nanos wait time if timed
936	>	*/
937	>	final void joinTask(ForkJoinTask<?> joinMe, boolean timed, long nanos) {
938	>	// currentJoin only written by this thread; only need ordered store
939	>	ForkJoinTask<?> prevJoin = currentJoin;
940	>	UNSAFE.putOrderedObject(this, currentJoinOffset, joinMe);
941	>	if (isTerminating())                // cancel if shutting down
942	>	joinMe.cancelIgnoringExceptions();
943	>	else
944	>	pool.awaitJoin(joinMe, this, timed, nanos);
945	>	UNSAFE.putOrderedObject(this, currentJoinOffset, prevJoin);
946		}
947
948		/**
949	<	* Execute the next task locally queued by the current worker, if
950	<	* one is available.
951	<	* @return true if a task was run; a false return indicates
952	<	* that no task was available.
949	>	* Run tasks in local queue until given task is done.
950	>	* Not currently used because it complicates semantics.
951	>	*
952	>	* @param joinMe the task to join
953		*/
954	<	public static boolean executeLocalTask() {
955	<	return ((ForkJoinWorkerThread)(Thread.currentThread())).
956	<	runLocalTask();
954	>	private void localHelpJoinTask(ForkJoinTask<?> joinMe) {
955	>	int s;
956	>	ForkJoinTask<?>[] q;
957	>	while (joinMe.status >= 0 && (s = sp) != base && (q = queue) != null) {
958	>	int i = (q.length - 1) & --s;
959	>	long u = (i << qShift) + qBase; // raw offset
960	>	ForkJoinTask<?> t = q[i];
961	>	if (t == null)  // lost to a stealer
962	>	break;
963	>	if (UNSAFE.compareAndSwapObject(q, u, t, null)) {
964	>	/*
965	>	* This recheck (and similarly in helpJoinTask)
966	>	* handles cases where joinMe is independently
967	>	* cancelled or forced even though there is other work
968	>	* available. Back out of the pop by putting t back
969	>	* into slot before we commit by writing sp.
970	>	*/
971	>	if (joinMe.status < 0) {
972	>	UNSAFE.putObjectVolatile(q, u, t);
973	>	break;
974	>	}
975	>	sp = s;
976	>	// UNSAFE.putOrderedInt(this, spOffset, s);
977	>	t.quietlyExec();
978	>	}
979	>	}
980		}
981
982		/**
983	<	* Removes and returns, without executing, the next task queued
984	<	* for execution in the current worker thread's local queue or if
985	<	* none, a task stolen from another worker, if one is available.
986	<	* A null return does not necessarily imply that all tasks are
987	<	* completed, only that there are currently none available.
988	<	* @return the next task to execute, or null if none
983	>	* Tries to locate and help perform tasks for a stealer of the
984	>	* given task, or in turn one of its stealers.  Traces
985	>	* currentSteal->currentJoin links looking for a thread working on
986	>	* a descendant of the given task and with a non-empty queue to
987	>	* steal back and execute tasks from.
988	>	*
989	>	* The implementation is very branchy to cope with potential
990	>	* inconsistencies or loops encountering chains that are stale,
991	>	* unknown, or of length greater than MAX_HELP_DEPTH links.  All
992	>	* of these cases are dealt with by just returning back to the
993	>	* caller, who is expected to retry if other join mechanisms also
994	>	* don't work out.
995	>	*
996	>	* @param joinMe the task to join
997		*/
998	<	public static ForkJoinTask<?> pollTask() {
999	<	return ((ForkJoinWorkerThread)(Thread.currentThread())).
1000	<	getLocalOrStolenTask();
998	>	final void helpJoinTask(ForkJoinTask<?> joinMe) {
999	>	ForkJoinWorkerThread[] ws;
1000	>	int n;
1001	>	if (joinMe.status < 0)                // already done
1002	>	return;
1003	>	if ((ws = pool.workers) == null || (n = ws.length) <= 1)
1004	>	return;                           // need at least 2 workers
1005	>
1006	>	ForkJoinTask<?> task = joinMe;        // base of chain
1007	>	ForkJoinWorkerThread thread = this;   // thread with stolen task
1008	>	for (int d = 0; d < MAX_HELP_DEPTH; ++d) { // chain length
1009	>	// Try to find v, the stealer of task, by first using hint
1010	>	ForkJoinWorkerThread v = ws[thread.stealHint & (n - 1)];
1011	>	if (v == null || v.currentSteal != task) {
1012	>	for (int j = 0; ; ++j) {      // search array
1013	>	if (j < n) {
1014	>	ForkJoinTask<?> vs;
1015	>	if ((v = ws[j]) != null &&
1016	>	(vs = v.currentSteal) != null) {
1017	>	if (joinMe.status < 0 || task.status < 0)
1018	>	return;       // stale or done
1019	>	if (vs == task) {
1020	>	thread.stealHint = j;
1021	>	break;        // save hint for next time
1022	>	}
1023	>	}
1024	>	}
1025	>	else
1026	>	return;               // no stealer
1027	>	}
1028	>	}
1029	>	for (;;) { // Try to help v, using specialized form of deqTask
1030	>	if (joinMe.status < 0)
1031	>	return;
1032	>	int b = v.base;
1033	>	ForkJoinTask<?>[] q = v.queue;
1034	>	if (b == v.sp || q == null)
1035	>	break;
1036	>	int i = (q.length - 1) & b;
1037	>	long u = (i << qShift) + qBase;
1038	>	ForkJoinTask<?> t = q[i];
1039	>	int pid = poolIndex;
1040	>	ForkJoinTask<?> ps = currentSteal;
1041	>	if (task.status < 0)
1042	>	return;                   // stale or done
1043	>	if (t != null && v.base == b++ &&
1044	>	UNSAFE.compareAndSwapObject(q, u, t, null)) {
1045	>	if (joinMe.status < 0) {
1046	>	UNSAFE.putObjectVolatile(q, u, t);
1047	>	return;               // back out on cancel
1048	>	}
1049	>	v.base = b;
1050	>	v.stealHint = pid;
1051	>	UNSAFE.putOrderedObject(this, currentStealOffset, t);
1052	>	t.quietlyExec();
1053	>	UNSAFE.putOrderedObject(this, currentStealOffset, ps);
1054	>	}
1055	>	}
1056	>	// Try to descend to find v's stealer
1057	>	ForkJoinTask<?> next = v.currentJoin;
1058	>	if (task.status < 0 || next == null || next == task ||
1059	>	joinMe.status < 0)
1060	>	return;
1061	>	task = next;
1062	>	thread = v;
1063	>	}
1064		}
1065
1066		/**
1067	<	* Helps this program complete by processing a local or stolen
1068	<	* task, if one is available.  This method may be useful when
1069	<	* several tasks are forked, and only one of them must be joined,
834	<	* as in:
1067	>	* Implements ForkJoinTask.getSurplusQueuedTaskCount().
1068	>	* Returns an estimate of the number of tasks, offset by a
1069	>	* function of number of idle workers.
1070		*
1071	<	* <pre>
1072	<	*   while (!t1.isDone() &amp;&amp; !t2.isDone())
1073	<	*     ForkJoinWorkerThread.executeTask();
1074	<	* </pre>
1071	>	* This method provides a cheap heuristic guide for task
1072	>	* partitioning when programmers, frameworks, tools, or languages
1073	>	* have little or no idea about task granularity.  In essence by
1074	>	* offering this method, we ask users only about tradeoffs in
1075	>	* overhead vs expected throughput and its variance, rather than
1076	>	* how finely to partition tasks.
1077		*
1078	<	* @return true if a task was run; a false return indicates
1079	<	* that no task was available.
1078	>	* In a steady state strict (tree-structured) computation, each
1079	>	* thread makes available for stealing enough tasks for other
1080	>	* threads to remain active. Inductively, if all threads play by
1081	>	* the same rules, each thread should make available only a
1082	>	* constant number of tasks.
1083	>	*
1084	>	* The minimum useful constant is just 1. But using a value of 1
1085	>	* would require immediate replenishment upon each steal to
1086	>	* maintain enough tasks, which is infeasible.  Further,
1087	>	* partitionings/granularities of offered tasks should minimize
1088	>	* steal rates, which in general means that threads nearer the top
1089	>	* of computation tree should generate more than those nearer the
1090	>	* bottom. In perfect steady state, each thread is at
1091	>	* approximately the same level of computation tree. However,
1092	>	* producing extra tasks amortizes the uncertainty of progress and
1093	>	* diffusion assumptions.
1094	>	*
1095	>	* So, users will want to use values larger, but not much larger
1096	>	* than 1 to both smooth over transient shortages and hedge
1097	>	* against uneven progress; as traded off against the cost of
1098	>	* extra task overhead. We leave the user to pick a threshold
1099	>	* value to compare with the results of this call to guide
1100	>	* decisions, but recommend values such as 3.
1101	>	*
1102	>	* When all threads are active, it is on average OK to estimate
1103	>	* surplus strictly locally. In steady-state, if one thread is
1104	>	* maintaining say 2 surplus tasks, then so are others. So we can
1105	>	* just use estimated queue length (although note that (sp - base)
1106	>	* can be an overestimate because of stealers lagging increments
1107	>	* of base).  However, this strategy alone leads to serious
1108	>	* mis-estimates in some non-steady-state conditions (ramp-up,
1109	>	* ramp-down, other stalls). We can detect many of these by
1110	>	* further considering the number of "idle" threads, that are
1111	>	* known to have zero queued tasks, so compensate by a factor of
1112	>	* (#idle/#active) threads.
1113		*/
1114	<	public static boolean executeTask() {
1115	<	return ((ForkJoinWorkerThread)(Thread.currentThread())).
846	<	runLocalOrStolenTask();
847	<	}
848	<
849	<	// Per-worker exported random numbers
850	<
851	<	// Same constants as java.util.Random
852	<	final static long JURandomMultiplier = 0x5DEECE66DL;
853	<	final static long JURandomAddend = 0xBL;
854	<	final static long JURandomMask = (1L << 48) - 1;
855	<
856	<	private final int nextJURandom(int bits) {
857	<	long next = (juRandomSeed * JURandomMultiplier + JURandomAddend) &
858	<	JURandomMask;
859	<	juRandomSeed = next;
860	<	return (int)(next >>> (48 - bits));
1114	>	final int getEstimatedSurplusTaskCount() {
1115	>	return sp - base - pool.idlePerActive();
1116		}
1117
1118	<	private final int nextJURandomInt(int n) {
1119	<	if (n <= 0)
1120	<	throw new IllegalArgumentException("n must be positive");
1121	<	int bits = nextJURandom(31);
1122	<	if ((n & -n) == n)
868	<	return (int)((n * (long)bits) >> 31);
869	<
1118	>	/**
1119	>	* Runs tasks until {@code pool.isQuiescent()}.
1120	>	*/
1121	>	final void helpQuiescePool() {
1122	>	ForkJoinTask<?> ps = currentSteal; // to restore below
1123		for (;;) {
1124	<	int val = bits % n;
1125	<	if (bits - val + (n-1) >= 0)
1126	<	return val;
1127	<	bits = nextJURandom(31);
1128	<	}
1129	<	}
1130	<
1131	<	private final long nextJURandomLong() {
1132	<	return ((long)(nextJURandom(32)) << 32) + nextJURandom(32);
1133	<	}
1134	<
1135	<	private final long nextJURandomLong(long n) {
1136	<	if (n <= 0)
1137	<	throw new IllegalArgumentException("n must be positive");
1138	<	long offset = 0;
1139	<	while (n >= Integer.MAX_VALUE) { // randomly pick half range
1140	<	int bits = nextJURandom(2); // 2nd bit for odd vs even split
1141	<	long half = n >>> 1;
1142	<	long nextn = ((bits & 2) == 0)? half : n - half;
1143	<	if ((bits & 1) == 0)
891	<	offset += n - nextn;
892	<	n = nextn;
1124	>	ForkJoinTask<?> t = pollLocalTask();
1125	>	if (t != null || (t = scan()) != null)
1126	>	t.quietlyExec();
1127	>	else {
1128	>	ForkJoinPool p = pool;
1129	>	int a; // to inline CASes
1130	>	if (active) {
1131	>	if (!UNSAFE.compareAndSwapInt
1132	>	(p, poolRunStateOffset, a = p.runState, a - 1))
1133	>	continue;   // retry later
1134	>	active = false; // inactivate
1135	>	UNSAFE.putOrderedObject(this, currentStealOffset, ps);
1136	>	}
1137	>	if (p.isQuiescent()) {
1138	>	active = true; // re-activate
1139	>	do {} while (!UNSAFE.compareAndSwapInt
1140	>	(p, poolRunStateOffset, a = p.runState, a+1));
1141	>	return;
1142	>	}
1143	>	}
1144		}
894	–	return offset + nextJURandomInt((int)n);
1145		}
1146
1147	<	private final double nextJURandomDouble() {
898	<	return (((long)(nextJURandom(26)) << 27) + nextJURandom(27))
899	<	/ (double)(1L << 53);
900	<	}
1147	>	// Unsafe mechanics
1148
1149	<	/**
1150	<	* Returns a random integer using a per-worker random
1151	<	* number generator with the same properties as
1152	<	* {@link java.util.Random#nextInt}
1153	<	* @return the next pseudorandom, uniformly distributed {@code int}
1154	<	*         value from this worker's random number generator's sequence
1155	<	*/
1156	<	public static int nextRandomInt() {
1157	<	return ((ForkJoinWorkerThread)(Thread.currentThread())).
1158	<	nextJURandom(32);
1159	<	}
1149	>	private static final sun.misc.Unsafe UNSAFE = getUnsafe();
1150	>	private static final long spOffset =
1151	>	objectFieldOffset("sp", ForkJoinWorkerThread.class);
1152	>	private static final long runStateOffset =
1153	>	objectFieldOffset("runState", ForkJoinWorkerThread.class);
1154	>	private static final long currentJoinOffset =
1155	>	objectFieldOffset("currentJoin", ForkJoinWorkerThread.class);
1156	>	private static final long currentStealOffset =
1157	>	objectFieldOffset("currentSteal", ForkJoinWorkerThread.class);
1158	>	private static final long qBase =
1159	>	UNSAFE.arrayBaseOffset(ForkJoinTask[].class);
1160	>	private static final long poolRunStateOffset = // to inline CAS
1161	>	objectFieldOffset("runState", ForkJoinPool.class);
1162
1163	<	/**
915	<	* Returns a random integer using a per-worker random
916	<	* number generator with the same properties as
917	<	* {@link java.util.Random#nextInt(int)}
918	<	* @param n the bound on the random number to be returned.  Must be
919	<	*        positive.
920	<	* @return the next pseudorandom, uniformly distributed {@code int}
921	<	*         value between {@code 0} (inclusive) and {@code n} (exclusive)
922	<	*         from this worker's random number generator's sequence
923	<	* @throws IllegalArgumentException if n is not positive
924	<	*/
925	<	public static int nextRandomInt(int n) {
926	<	return ((ForkJoinWorkerThread)(Thread.currentThread())).
927	<	nextJURandomInt(n);
928	<	}
1163	>	private static final int qShift;
1164
1165	<	/**
1166	<	* Returns a random long using a per-worker random
1167	<	* number generator with the same properties as
1168	<	* {@link java.util.Random#nextLong}
1169	<	* @return the next pseudorandom, uniformly distributed {@code long}
1170	<	*         value from this worker's random number generator's sequence
936	<	*/
937	<	public static long nextRandomLong() {
938	<	return ((ForkJoinWorkerThread)(Thread.currentThread())).
939	<	nextJURandomLong();
1165	>	static {
1166	>	int s = UNSAFE.arrayIndexScale(ForkJoinTask[].class);
1167	>	if ((s & (s-1)) != 0)
1168	>	throw new Error("data type scale not a power of two");
1169	>	qShift = 31 - Integer.numberOfLeadingZeros(s);
1170	>	MAXIMUM_QUEUE_CAPACITY = 1 << (31 - qShift);
1171		}
1172
1173	<	/**
1174	<	* Returns a random integer using a per-worker random
1175	<	* number generator with the same properties as
1176	<	* {@link java.util.Random#nextInt(int)}
1177	<	* @param n the bound on the random number to be returned.  Must be
1178	<	*        positive.
1179	<	* @return the next pseudorandom, uniformly distributed {@code int}
1180	<	*         value between {@code 0} (inclusive) and {@code n} (exclusive)
1181	<	*         from this worker's random number generator's sequence
951	<	* @throws IllegalArgumentException if n is not positive
952	<	*/
953	<	public static long nextRandomLong(long n) {
954	<	return ((ForkJoinWorkerThread)(Thread.currentThread())).
955	<	nextJURandomLong(n);
1173	>	private static long objectFieldOffset(String field, Class<?> klazz) {
1174	>	try {
1175	>	return UNSAFE.objectFieldOffset(klazz.getDeclaredField(field));
1176	>	} catch (NoSuchFieldException e) {
1177	>	// Convert Exception to corresponding Error
1178	>	NoSuchFieldError error = new NoSuchFieldError(field);
1179	>	error.initCause(e);
1180	>	throw error;
1181	>	}
1182		}
1183
1184		/**
1185	<	* Returns a random double using a per-worker random
1186	<	* number generator with the same properties as
1187	<	* {@link java.util.Random#nextDouble}
1188	<	* @return the next pseudorandom, uniformly distributed {@code double}
1189	<	*         value between {@code 0.0} and {@code 1.0} from this
964	<	*         worker's random number generator's sequence
1185	>	* Returns a sun.misc.Unsafe.  Suitable for use in a 3rd party package.
1186	>	* Replace with a simple call to Unsafe.getUnsafe when integrating
1187	>	* into a jdk.
1188	>	*
1189	>	* @return a sun.misc.Unsafe
1190		*/
1191	<	public static double nextRandomDouble() {
967	<	return ((ForkJoinWorkerThread)(Thread.currentThread())).
968	<	nextJURandomDouble();
969	<	}
970	<
971	<	// Temporary Unsafe mechanics for preliminary release
972	<
973	<	static final Unsafe _unsafe;
974	<	static final long baseOffset;
975	<	static final long spOffset;
976	<	static final long qBase;
977	<	static final int qShift;
978	<	static final long runStateOffset;
979	<	static {
1191	>	private static sun.misc.Unsafe getUnsafe() {
1192		try {
1193	<	if (ForkJoinWorkerThread.class.getClassLoader() != null) {
1194	<	Field f = Unsafe.class.getDeclaredField("theUnsafe");
1195	<	f.setAccessible(true);
1196	<	_unsafe = (Unsafe)f.get(null);
1197	<	}
1198	<	else
1199	<	_unsafe = Unsafe.getUnsafe();
1200	<	baseOffset = _unsafe.objectFieldOffset
1201	<	(ForkJoinWorkerThread.class.getDeclaredField("base"));
1202	<	spOffset = _unsafe.objectFieldOffset
1203	<	(ForkJoinWorkerThread.class.getDeclaredField("sp"));
1204	<	runStateOffset = _unsafe.objectFieldOffset
1205	<	(ForkJoinWorkerThread.class.getDeclaredField("runState"));
1206	<	qBase = _unsafe.arrayBaseOffset(ForkJoinTask[].class);
1207	<	int s = _unsafe.arrayIndexScale(ForkJoinTask[].class);
1208	<	if ((s & (s-1)) != 0)
997	<	throw new Error("data type scale not a power of two");
998	<	qShift = 31 - Integer.numberOfLeadingZeros(s);
999	<	} catch (Exception e) {
1000	<	throw new RuntimeException("Could not initialize intrinsics", e);
1193	>	return sun.misc.Unsafe.getUnsafe();
1194	>	} catch (SecurityException se) {
1195	>	try {
1196	>	return java.security.AccessController.doPrivileged
1197	>	(new java.security
1198	>	.PrivilegedExceptionAction<sun.misc.Unsafe>() {
1199	>	public sun.misc.Unsafe run() throws Exception {
1200	>	java.lang.reflect.Field f = sun.misc
1201	>	.Unsafe.class.getDeclaredField("theUnsafe");
1202	>	f.setAccessible(true);
1203	>	return (sun.misc.Unsafe) f.get(null);
1204	>	}});
1205	>	} catch (java.security.PrivilegedActionException e) {
1206	>	throw new RuntimeException("Could not initialize intrinsics",
1207	>	e.getCause());
1208	>	}
1209		}
1210		}
1211		}

Diff Legend

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