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 1.59 by dl, Sun Nov 21 13:55:04 2010 UTC vs.
Revision 1.72 by dl, Mon Nov 19 18:12:42 2012 UTC

#	Line 1 | Line 1
1		/*
2		* Written by Doug Lea with assistance from members of JCP JSR-166
3		* Expert Group and released to the public domain, as explained at
4	<	* http://creativecommons.org/licenses/publicdomain
4	>	* http://creativecommons.org/publicdomain/zero/1.0/
5		*/
6
7		package jsr166y;
8
9	<	import java.util.Random;
10	<	import java.util.Collection;
11	<	import java.util.concurrent.locks.LockSupport;
12	<	import java.util.concurrent.RejectedExecutionException;
9	>	import java.util.concurrent.atomic.AtomicInteger;
10
11		/**
12		* A thread managed by a {@link ForkJoinPool}, which executes
#	Line 27 | Line 24 | import java.util.concurrent.RejectedExec
24		*/
25		public class ForkJoinWorkerThread extends Thread {
26		/*
30	–	* Overview:
31	–	*
27		* ForkJoinWorkerThreads are managed by ForkJoinPools and perform
28	<	* ForkJoinTasks. This class includes bookkeeping in support of
29	<	* worker activation, suspension, and lifecycle control described
35	<	* in more detail in the internal documentation of class
36	<	* ForkJoinPool. And as described further below, this class also
37	<	* includes special-cased support for some ForkJoinTask
38	<	* methods. But the main mechanics involve work-stealing:
39	<	*
40	<	* Work-stealing queues are special forms of Deques that support
41	<	* only three of the four possible end-operations -- push, pop,
42	<	* and deq (aka steal), under the further constraints that push
43	<	* and pop are called only from the owning thread, while deq may
44	<	* be called from other threads.  (If you are unfamiliar with
45	<	* them, you probably want to read Herlihy and Shavit's book "The
46	<	* Art of Multiprocessor programming", chapter 16 describing these
47	<	* in more detail before proceeding.)  The main work-stealing
48	<	* queue design is roughly similar to those in the papers "Dynamic
49	<	* Circular Work-Stealing Deque" by Chase and Lev, SPAA 2005
50	<	* (http://research.sun.com/scalable/pubs/index.html) and
51	<	* "Idempotent work stealing" by Michael, Saraswat, and Vechev,
52	<	* PPoPP 2009 (http://portal.acm.org/citation.cfm?id=1504186).
53	<	* The main differences ultimately stem from gc requirements that
54	<	* we null out taken slots as soon as we can, to maintain as small
55	<	* a footprint as possible even in programs generating huge
56	<	* numbers of tasks. To accomplish this, we shift the CAS
57	<	* arbitrating pop vs deq (steal) from being on the indices
58	<	* ("base" and "sp") to the slots themselves (mainly via method
59	<	* "casSlotNull()"). So, both a successful pop and deq mainly
60	<	* entail a CAS of a slot from non-null to null.  Because we rely
61	<	* on CASes of references, we do not need tag bits on base or sp.
62	<	* They are simple ints as used in any circular array-based queue
63	<	* (see for example ArrayDeque).  Updates to the indices must
64	<	* still be ordered in a way that guarantees that sp == base means
65	<	* the queue is empty, but otherwise may err on the side of
66	<	* possibly making the queue appear nonempty when a push, pop, or
67	<	* deq have not fully committed. Note that this means that the deq
68	<	* operation, considered individually, is not wait-free. One thief
69	<	* cannot successfully continue until another in-progress one (or,
70	<	* if previously empty, a push) completes.  However, in the
71	<	* aggregate, we ensure at least probabilistic non-blockingness.
72	<	* If an attempted steal fails, a thief always chooses a different
73	<	* random victim target to try next. So, in order for one thief to
74	<	* progress, it suffices for any in-progress deq or new push on
75	<	* any empty queue to complete. One reason this works well here is
76	<	* that apparently-nonempty often means soon-to-be-stealable,
77	<	* which gives threads a chance to set activation status if
78	<	* necessary before stealing.
79	<	*
80	<	* This approach also enables support for "async mode" where local
81	<	* task processing is in FIFO, not LIFO order; simply by using a
82	<	* version of deq rather than pop when locallyFifo is true (as set
83	<	* by the ForkJoinPool).  This allows use in message-passing
84	<	* frameworks in which tasks are never joined.
85	<	*
86	<	* When a worker would otherwise be blocked waiting to join a
87	<	* task, it first tries a form of linear helping: Each worker
88	<	* records (in field currentSteal) the most recent task it stole
89	<	* from some other worker. Plus, it records (in field currentJoin)
90	<	* the task it is currently actively joining. Method joinTask uses
91	<	* these markers to try to find a worker to help (i.e., steal back
92	<	* a task from and execute it) that could hasten completion of the
93	<	* actively joined task. In essence, the joiner executes a task
94	<	* that would be on its own local deque had the to-be-joined task
95	<	* not been stolen. This may be seen as a conservative variant of
96	<	* the approach in Wagner & Calder "Leapfrogging: a portable
97	<	* technique for implementing efficient futures" SIGPLAN Notices,
98	<	* 1993 (http://portal.acm.org/citation.cfm?id=155354). It differs
99	<	* in that: (1) We only maintain dependency links across workers
100	<	* upon steals, rather than use per-task bookkeeping.  This may
101	<	* require a linear scan of workers array to locate stealers, but
102	<	* usually doesn't because stealers leave hints (that may become
103	<	* stale/wrong) of where to locate them. This isolates cost to
104	<	* when it is needed, rather than adding to per-task overhead.
105	<	* (2) It is "shallow", ignoring nesting and potentially cyclic
106	<	* mutual steals.  (3) It is intentionally racy: field currentJoin
107	<	* is updated only while actively joining, which means that we
108	<	* miss links in the chain during long-lived tasks, GC stalls etc
109	<	* (which is OK since blocking in such cases is usually a good
110	<	* idea).  (4) We bound the number of attempts to find work (see
111	<	* MAX_HELP_DEPTH) and fall back to suspending the worker and if
112	<	* necessary replacing it with a spare (see
113	<	* ForkJoinPool.awaitJoin).
114	<	*
115	<	* Efficient implementation of these algorithms currently relies
116	<	* on an uncomfortable amount of "Unsafe" mechanics. To maintain
117	<	* correct orderings, reads and writes of variable base require
118	<	* volatile ordering.  Variable sp does not require volatile
119	<	* writes but still needs store-ordering, which we accomplish by
120	<	* pre-incrementing sp before filling the slot with an ordered
121	<	* store.  (Pre-incrementing also enables backouts used in
122	<	* joinTask.)  Because they are protected by volatile base reads,
123	<	* reads of the queue array and its slots by other threads do not
124	<	* need volatile load semantics, but writes (in push) require
125	<	* store order and CASes (in pop and deq) require (volatile) CAS
126	<	* semantics.  (Michael, Saraswat, and Vechev's algorithm has
127	<	* similar properties, but without support for nulling slots.)
128	<	* Since these combinations aren't supported using ordinary
129	<	* volatiles, the only way to accomplish these efficiently is to
130	<	* use direct Unsafe calls. (Using external AtomicIntegers and
131	<	* AtomicReferenceArrays for the indices and array is
132	<	* significantly slower because of memory locality and indirection
133	<	* effects.)
28	>	* ForkJoinTasks. For explanation, see the internal documentation
29	>	* of class ForkJoinPool.
30		*
31	<	* Further, performance on most platforms is very sensitive to
32	<	* placement and sizing of the (resizable) queue array.  Even
33	<	* though these queues don't usually become all that big, the
34	<	* initial size must be large enough to counteract cache
35	<	* contention effects across multiple queues (especially in the
36	<	* presence of GC cardmarking). Also, to improve thread-locality,
37	<	* queues are initialized after starting.  All together, these
142	<	* low-level implementation choices produce as much as a factor of
143	<	* 4 performance improvement compared to naive implementations,
144	<	* and enable the processing of billions of tasks per second,
145	<	* sometimes at the expense of ugliness.
146	<	*/
147	<
148	<	/**
149	<	* Generator for initial random seeds for random victim
150	<	* selection. This is used only to create initial seeds. Random
151	<	* steals use a cheaper xorshift generator per steal attempt. We
152	<	* expect only rare contention on seedGenerator, so just use a
153	<	* plain Random.
154	<	*/
155	<	private static final Random seedGenerator = new Random();
156	<
157	<	/**
158	<	* The maximum stolen->joining link depth allowed in helpJoinTask.
159	<	* Depths for legitimate chains are unbounded, but we use a fixed
160	<	* constant to avoid (otherwise unchecked) cycles and bound
161	<	* staleness of traversal parameters at the expense of sometimes
162	<	* blocking when we could be helping.
163	<	*/
164	<	private static final int MAX_HELP_DEPTH = 8;
165	<
166	<	/**
167	<	* Capacity of work-stealing queue array upon initialization.
168	<	* Must be a power of two. Initial size must be at least 4, but is
169	<	* padded to minimize cache effects.
170	<	*/
171	<	private static final int INITIAL_QUEUE_CAPACITY = 1 << 13;
172	<
173	<	/**
174	<	* Maximum work-stealing queue array size.  Must be less than or
175	<	* equal to 1 << (31 - width of array entry) to ensure lack of
176	<	* index wraparound. The value is set in the static block
177	<	* at the end of this file after obtaining width.
178	<	*/
179	<	private static final int MAXIMUM_QUEUE_CAPACITY;
180	<
181	<	/**
182	<	* The pool this thread works in. Accessed directly by ForkJoinTask.
183	<	*/
184	<	final ForkJoinPool pool;
185	<
186	<	/**
187	<	* The work-stealing queue array. Size must be a power of two.
188	<	* Initialized in onStart, to improve memory locality.
189	<	*/
190	<	private ForkJoinTask<?>[] queue;
191	<
192	<	/**
193	<	* Index (mod queue.length) of least valid queue slot, which is
194	<	* always the next position to steal from if nonempty.
195	<	*/
196	<	private volatile int base;
197	<
198	<	/**
199	<	* Index (mod queue.length) of next queue slot to push to or pop
200	<	* from. It is written only by owner thread, and accessed by other
201	<	* threads only after reading (volatile) base.  Both sp and base
202	<	* are allowed to wrap around on overflow, but (sp - base) still
203	<	* estimates size.
204	<	*/
205	<	private int sp;
206	<
207	<	/**
208	<	* The index of most recent stealer, used as a hint to avoid
209	<	* traversal in method helpJoinTask. This is only a hint because a
210	<	* worker might have had multiple steals and this only holds one
211	<	* of them (usually the most current). Declared non-volatile,
212	<	* relying on other prevailing sync to keep reasonably current.
213	<	*/
214	<	private int stealHint;
215	<
216	<	/**
217	<	* Run state of this worker. In addition to the usual run levels,
218	<	* tracks if this worker is suspended as a spare, and if it was
219	<	* killed (trimmed) while suspended. However, "active" status is
220	<	* maintained separately and modified only in conjunction with
221	<	* CASes of the pool's runState (which are currently sadly
222	<	* manually inlined for performance.)  Accessed directly by pool
223	<	* to simplify checks for normal (zero) status.
31	>	* This class just maintains links to its pool and WorkQueue.  The
32	>	* pool field is set upon construction, but the workQueue field is
33	>	* not set until the thread has started (unless forced early by a
34	>	* subclass constructor call to poolIndex()).  This provides
35	>	* better memory placement (because this thread allocates queue
36	>	* and bookkeeping fields) but because the field is non-final, we
37	>	* require that it never be accessed except by the owning thread.
38		*/
225	–	volatile int runState;
39
40	<	private static final int TERMINATING = 0x01;
41	<	private static final int TERMINATED  = 0x02;
229	<	private static final int SUSPENDED   = 0x04; // inactive spare
230	<	private static final int TRIMMED     = 0x08; // killed while suspended
40	>	final ForkJoinPool pool;                // the pool this thread works in
41	>	ForkJoinPool.WorkQueue workQueue;       // Work-stealing mechanics
42
43		/**
44	<	* Number of steals. Directly accessed (and reset) by
234	<	* pool.tryAccumulateStealCount when idle.
44	>	* Sequence number for creating worker Names
45		*/
46	<	int stealCount;
237	<
238	<	/**
239	<	* Seed for random number generator for choosing steal victims.
240	<	* Uses Marsaglia xorshift. Must be initialized as nonzero.
241	<	*/
242	<	private int seed;
243	<
244	<	/**
245	<	* Activity status. When true, this worker is considered active.
246	<	* Accessed directly by pool.  Must be false upon construction.
247	<	*/
248	<	boolean active;
249	<
250	<	/**
251	<	* True if use local fifo, not default lifo, for local polling.
252	<	* Shadows value from ForkJoinPool.
253	<	*/
254	<	private final boolean locallyFifo;
255	<
256	<	/**
257	<	* Index of this worker in pool array. Set once by pool before
258	<	* running, and accessed directly by pool to locate this worker in
259	<	* its workers array.
260	<	*/
261	<	int poolIndex;
262	<
263	<	/**
264	<	* The last pool event waited for. Accessed only by pool in
265	<	* callback methods invoked within this thread.
266	<	*/
267	<	int lastEventCount;
268	<
269	<	/**
270	<	* Encoded index and event count of next event waiter. Accessed
271	<	* only by ForkJoinPool for managing event waiters.
272	<	*/
273	<	volatile long nextWaiter;
274	<
275	<	/**
276	<	* Number of times this thread suspended as spare. Accessed only
277	<	* by pool.
278	<	*/
279	<	int spareCount;
280	<
281	<	/**
282	<	* Encoded index and count of next spare waiter. Accessed only
283	<	* by ForkJoinPool for managing spares.
284	<	*/
285	<	volatile int nextSpare;
286	<
287	<	/**
288	<	* The task currently being joined, set only when actively trying
289	<	* to help other stealers in helpJoinTask. Written only by this
290	<	* thread, but read by others.
291	<	*/
292	<	private volatile ForkJoinTask<?> currentJoin;
293	<
294	<	/**
295	<	* The task most recently stolen from another worker (or
296	<	* submission queue).  Written only by this thread, but read by
297	<	* others.
298	<	*/
299	<	private volatile ForkJoinTask<?> currentSteal;
46	>	private static final AtomicInteger threadNumber = new AtomicInteger();
47
48		/**
49		* Creates a ForkJoinWorkerThread operating in the given pool.
#	Line 305 | Line 52 | public class ForkJoinWorkerThread extend
52		* @throws NullPointerException if pool is null
53		*/
54		protected ForkJoinWorkerThread(ForkJoinPool pool) {
55	<	this.pool = pool;
309	<	this.locallyFifo = pool.locallyFifo;
55	>	super(pool.workerNamePrefix.concat(Integer.toString(threadNumber.incrementAndGet())));
56		setDaemon(true);
57	<	// To avoid exposing construction details to subclasses,
58	<	// remaining initialization is in start() and onStart()
313	<	}
314	<
315	<	/**
316	<	* Performs additional initialization and starts this thread.
317	<	*/
318	<	final void start(int poolIndex, UncaughtExceptionHandler ueh) {
319	<	this.poolIndex = poolIndex;
57	>	this.pool = pool;
58	>	Thread.UncaughtExceptionHandler ueh = pool.ueh;
59		if (ueh != null)
60		setUncaughtExceptionHandler(ueh);
322	–	start();
61		}
62
325	–	// Public/protected methods
326	–
63		/**
64		* Returns the pool hosting this thread.
65		*
#	Line 343 | Line 79 | public class ForkJoinWorkerThread extend
79		* @return the index number
80		*/
81		public int getPoolIndex() {
82	<	return poolIndex;
82	>	// force early registration if called before started
83	>	ForkJoinPool.WorkQueue q;
84	>	if ((q = workQueue) == null) {
85	>	pool.registerWorker(this);
86	>	q = workQueue;
87	>	}
88	>	return q.poolIndex;
89		}
90
91		/**
#	Line 356 | Line 98 | public class ForkJoinWorkerThread extend
98		* processing tasks.
99		*/
100		protected void onStart() {
359	–	int rs = seedGenerator.nextInt();
360	–	seed = rs == 0? 1 : rs; // seed must be nonzero
361	–
362	–	// Allocate name string and arrays in this thread
363	–	String pid = Integer.toString(pool.getPoolNumber());
364	–	String wid = Integer.toString(poolIndex);
365	–	setName("ForkJoinPool-" + pid + "-worker-" + wid);
366	–
367	–	queue = new ForkJoinTask<?>[INITIAL_QUEUE_CAPACITY];
101		}
102
103		/**
#	Line 376 | Line 109 | public class ForkJoinWorkerThread extend
109		* to an unrecoverable error, or {@code null} if completed normally
110		*/
111		protected void onTermination(Throwable exception) {
379	–	try {
380	–	ForkJoinPool p = pool;
381	–	if (active) {
382	–	int a; // inline p.tryDecrementActiveCount
383	–	active = false;
384	–	do {} while (!UNSAFE.compareAndSwapInt
385	–	(p, poolRunStateOffset, a = p.runState, a - 1));
386	–	}
387	–	cancelTasks();
388	–	setTerminated();
389	–	p.workerTerminated(this);
390	–	} catch (Throwable ex) {        // Shouldn't ever happen
391	–	if (exception == null)      // but if so, at least rethrown
392	–	exception = ex;
393	–	} finally {
394	–	if (exception != null)
395	–	UNSAFE.throwException(exception);
396	–	}
112		}
113
114		/**
#	Line 404 | Line 119 | public class ForkJoinWorkerThread extend
119		public void run() {
120		Throwable exception = null;
121		try {
122	+	pool.registerWorker(this);
123		onStart();
124	<	mainLoop();
124	>	pool.runWorker(workQueue);
125		} catch (Throwable ex) {
126		exception = ex;
127		} finally {
412	–	onTermination(exception);
413	–	}
414	–	}
415	–
416	–	// helpers for run()
417	–
418	–	/**
419	–	* Finds and executes tasks, and checks status while running.
420	–	*/
421	–	private void mainLoop() {
422	–	boolean ran = false; // true if ran a task on last step
423	–	ForkJoinPool p = pool;
424	–	for (;;) {
425	–	p.preStep(this, ran);
426	–	if (runState != 0)
427	–	break;
428	–	ran = tryExecSteal() || tryExecSubmission();
429	–	}
430	–	}
431	–
432	–	/**
433	–	* Tries to steal a task and execute it.
434	–	*
435	–	* @return true if ran a task
436	–	*/
437	–	private boolean tryExecSteal() {
438	–	ForkJoinTask<?> t;
439	–	if ((t = scan()) != null) {
440	–	t.quietlyExec();
441	–	UNSAFE.putOrderedObject(this, currentStealOffset, null);
442	–	if (sp != base)
443	–	execLocalTasks();
444	–	return true;
445	–	}
446	–	return false;
447	–	}
448	–
449	–	/**
450	–	* If a submission exists, try to activate and run it.
451	–	*
452	–	* @return true if ran a task
453	–	*/
454	–	private boolean tryExecSubmission() {
455	–	ForkJoinPool p = pool;
456	–	// This loop is needed in case attempt to activate fails, in
457	–	// which case we only retry if there still appears to be a
458	–	// submission.
459	–	while (p.hasQueuedSubmissions()) {
460	–	ForkJoinTask<?> t; int a;
461	–	if (active || // inline p.tryIncrementActiveCount
462	–	(active = UNSAFE.compareAndSwapInt(p, poolRunStateOffset,
463	–	a = p.runState, a + 1))) {
464	–	if ((t = p.pollSubmission()) != null) {
465	–	UNSAFE.putOrderedObject(this, currentStealOffset, t);
466	–	t.quietlyExec();
467	–	UNSAFE.putOrderedObject(this, currentStealOffset, null);
468	–	if (sp != base)
469	–	execLocalTasks();
470	–	return true;
471	–	}
472	–	}
473	–	}
474	–	return false;
475	–	}
476	–
477	–	/**
478	–	* Runs local tasks until queue is empty or shut down.  Call only
479	–	* while active.
480	–	*/
481	–	private void execLocalTasks() {
482	–	while (runState == 0) {
483	–	ForkJoinTask<?> t = locallyFifo ? locallyDeqTask() : popTask();
484	–	if (t != null)
485	–	t.quietlyExec();
486	–	else if (sp == base)
487	–	break;
488	–	}
489	–	}
490	–
491	–	/*
492	–	* Intrinsics-based atomic writes for queue slots. These are
493	–	* basically the same as methods in AtomicReferenceArray, but
494	–	* specialized for (1) ForkJoinTask elements (2) requirement that
495	–	* nullness and bounds checks have already been performed by
496	–	* callers and (3) effective offsets are known not to overflow
497	–	* from int to long (because of MAXIMUM_QUEUE_CAPACITY). We don't
498	–	* need corresponding version for reads: plain array reads are OK
499	–	* because they are protected by other volatile reads and are
500	–	* confirmed by CASes.
501	–	*
502	–	* Most uses don't actually call these methods, but instead contain
503	–	* inlined forms that enable more predictable optimization.  We
504	–	* don't define the version of write used in pushTask at all, but
505	–	* instead inline there a store-fenced array slot write.
506	–	*/
507	–
508	–	/**
509	–	* CASes slot i of array q from t to null. Caller must ensure q is
510	–	* non-null and index is in range.
511	–	*/
512	–	private static final boolean casSlotNull(ForkJoinTask<?>[] q, int i,
513	–	ForkJoinTask<?> t) {
514	–	return UNSAFE.compareAndSwapObject(q, (i << qShift) + qBase, t, null);
515	–	}
516	–
517	–	/**
518	–	* Performs a volatile write of the given task at given slot of
519	–	* array q.  Caller must ensure q is non-null and index is in
520	–	* range. This method is used only during resets and backouts.
521	–	*/
522	–	private static final void writeSlot(ForkJoinTask<?>[] q, int i,
523	–	ForkJoinTask<?> t) {
524	–	UNSAFE.putObjectVolatile(q, (i << qShift) + qBase, t);
525	–	}
526	–
527	–	// queue methods
528	–
529	–	/**
530	–	* Pushes a task. Call only from this thread.
531	–	*
532	–	* @param t the task. Caller must ensure non-null.
533	–	*/
534	–	final void pushTask(ForkJoinTask<?> t) {
535	–	ForkJoinTask<?>[] q = queue;
536	–	int mask = q.length - 1; // implicit assert q != null
537	–	int s = sp++;            // ok to increment sp before slot write
538	–	UNSAFE.putOrderedObject(q, ((s & mask) << qShift) + qBase, t);
539	–	if ((s -= base) == 0)
540	–	pool.signalWork();   // was empty
541	–	else if (s == mask)
542	–	growQueue();         // is full
543	–	}
544	–
545	–	/**
546	–	* Tries to take a task from the base of the queue, failing if
547	–	* empty or contended. Note: Specializations of this code appear
548	–	* in locallyDeqTask and elsewhere.
549	–	*
550	–	* @return a task, or null if none or contended
551	–	*/
552	–	final ForkJoinTask<?> deqTask() {
553	–	ForkJoinTask<?> t;
554	–	ForkJoinTask<?>[] q;
555	–	int b, i;
556	–	if (sp != (b = base) &&
557	–	(q = queue) != null && // must read q after b
558	–	(t = q[i = (q.length - 1) & b]) != null && base == b &&
559	–	UNSAFE.compareAndSwapObject(q, (i << qShift) + qBase, t, null)) {
560	–	base = b + 1;
561	–	return t;
562	–	}
563	–	return null;
564	–	}
565	–
566	–	/**
567	–	* Tries to take a task from the base of own queue. Assumes active
568	–	* status.  Called only by this thread.
569	–	*
570	–	* @return a task, or null if none
571	–	*/
572	–	final ForkJoinTask<?> locallyDeqTask() {
573	–	ForkJoinTask<?>[] q = queue;
574	–	if (q != null) {
575	–	ForkJoinTask<?> t;
576	–	int b, i;
577	–	while (sp != (b = base)) {
578	–	if ((t = q[i = (q.length - 1) & b]) != null && base == b &&
579	–	UNSAFE.compareAndSwapObject(q, (i << qShift) + qBase,
580	–	t, null)) {
581	–	base = b + 1;
582	–	return t;
583	–	}
584	–	}
585	–	}
586	–	return null;
587	–	}
588	–
589	–	/**
590	–	* Returns a popped task, or null if empty. Assumes active status.
591	–	* Called only by this thread.
592	–	*/
593	–	private ForkJoinTask<?> popTask() {
594	–	ForkJoinTask<?>[] q = queue;
595	–	if (q != null) {
596	–	int s;
597	–	while ((s = sp) != base) {
598	–	int i = (q.length - 1) & --s;
599	–	long u = (i << qShift) + qBase; // raw offset
600	–	ForkJoinTask<?> t = q[i];
601	–	if (t == null)   // lost to stealer
602	–	break;
603	–	if (UNSAFE.compareAndSwapObject(q, u, t, null)) {
604	–	/*
605	–	* Note: here and in related methods, as a
606	–	* performance (not correctness) issue, we'd like
607	–	* to encourage compiler not to arbitrarily
608	–	* postpone setting sp after successful CAS.
609	–	* Currently there is no intrinsic for arranging
610	–	* this, but using Unsafe putOrderedInt may be a
611	–	* preferable strategy on some compilers even
612	–	* though its main effect is a pre-, not post-
613	–	* fence. To simplify possible changes, the option
614	–	* is left in comments next to the associated
615	–	* assignments.
616	–	*/
617	–	sp = s; // putOrderedInt may encourage more timely write
618	–	// UNSAFE.putOrderedInt(this, spOffset, s);
619	–	return t;
620	–	}
621	–	}
622	–	}
623	–	return null;
624	–	}
625	–
626	–	/**
627	–	* Specialized version of popTask to pop only if topmost element
628	–	* is the given task. Called only by this thread while active.
629	–	*
630	–	* @param t the task. Caller must ensure non-null.
631	–	*/
632	–	final boolean unpushTask(ForkJoinTask<?> t) {
633	–	int s;
634	–	ForkJoinTask<?>[] q = queue;
635	–	if ((s = sp) != base && q != null &&
636	–	UNSAFE.compareAndSwapObject
637	–	(q, (((q.length - 1) & --s) << qShift) + qBase, t, null)) {
638	–	sp = s; // putOrderedInt may encourage more timely write
639	–	// UNSAFE.putOrderedInt(this, spOffset, s);
640	–	return true;
641	–	}
642	–	return false;
643	–	}
644	–
645	–	/**
646	–	* Returns next task, or null if empty or contended.
647	–	*/
648	–	final ForkJoinTask<?> peekTask() {
649	–	ForkJoinTask<?>[] q = queue;
650	–	if (q == null)
651	–	return null;
652	–	int mask = q.length - 1;
653	–	int i = locallyFifo ? base : (sp - 1);
654	–	return q[i & mask];
655	–	}
656	–
657	–	/**
658	–	* Doubles queue array size. Transfers elements by emulating
659	–	* steals (deqs) from old array and placing, oldest first, into
660	–	* new array.
661	–	*/
662	–	private void growQueue() {
663	–	ForkJoinTask<?>[] oldQ = queue;
664	–	int oldSize = oldQ.length;
665	–	int newSize = oldSize << 1;
666	–	if (newSize > MAXIMUM_QUEUE_CAPACITY)
667	–	throw new RejectedExecutionException("Queue capacity exceeded");
668	–	ForkJoinTask<?>[] newQ = queue = new ForkJoinTask<?>[newSize];
669	–
670	–	int b = base;
671	–	int bf = b + oldSize;
672	–	int oldMask = oldSize - 1;
673	–	int newMask = newSize - 1;
674	–	do {
675	–	int oldIndex = b & oldMask;
676	–	ForkJoinTask<?> t = oldQ[oldIndex];
677	–	if (t != null && !casSlotNull(oldQ, oldIndex, t))
678	–	t = null;
679	–	writeSlot(newQ, b & newMask, t);
680	–	} while (++b != bf);
681	–	pool.signalWork();
682	–	}
683	–
684	–	/**
685	–	* Computes next value for random victim probe in scan().  Scans
686	–	* don't require a very high quality generator, but also not a
687	–	* crummy one.  Marsaglia xor-shift is cheap and works well enough.
688	–	* Note: This is manually inlined in scan().
689	–	*/
690	–	private static final int xorShift(int r) {
691	–	r ^= r << 13;
692	–	r ^= r >>> 17;
693	–	return r ^ (r << 5);
694	–	}
695	–
696	–	/**
697	–	* Tries to steal a task from another worker. Starts at a random
698	–	* index of workers array, and probes workers until finding one
699	–	* with non-empty queue or finding that all are empty.  It
700	–	* randomly selects the first n probes. If these are empty, it
701	–	* resorts to a circular sweep, which is necessary to accurately
702	–	* set active status. (The circular sweep uses steps of
703	–	* approximately half the array size plus 1, to avoid bias
704	–	* stemming from leftmost packing of the array in ForkJoinPool.)
705	–	*
706	–	* This method must be both fast and quiet -- usually avoiding
707	–	* memory accesses that could disrupt cache sharing etc other than
708	–	* those needed to check for and take tasks (or to activate if not
709	–	* already active). This accounts for, among other things,
710	–	* updating random seed in place without storing it until exit.
711	–	*
712	–	* @return a task, or null if none found
713	–	*/
714	–	private ForkJoinTask<?> scan() {
715	–	ForkJoinPool p = pool;
716	–	ForkJoinWorkerThread[] ws;        // worker array
717	–	int n;                            // upper bound of #workers
718	–	if ((ws = p.workers) != null && (n = ws.length) > 1) {
719	–	boolean canSteal = active;    // shadow active status
720	–	int r = seed;                 // extract seed once
721	–	int mask = n - 1;
722	–	int j = -n;                   // loop counter
723	–	int k = r;                    // worker index, random if j < 0
724	–	for (;;) {
725	–	ForkJoinWorkerThread v = ws[k & mask];
726	–	r ^= r << 13; r ^= r >>> 17; r ^= r << 5; // inline xorshift
727	–	ForkJoinTask<?>[] q; ForkJoinTask<?> t; int b, a;
728	–	if (v != null && (b = v.base) != v.sp &&
729	–	(q = v.queue) != null) {
730	–	int i = (q.length - 1) & b;
731	–	long u = (i << qShift) + qBase; // raw offset
732	–	int pid = poolIndex;
733	–	if ((t = q[i]) != null) {
734	–	if (!canSteal &&  // inline p.tryIncrementActiveCount
735	–	UNSAFE.compareAndSwapInt(p, poolRunStateOffset,
736	–	a = p.runState, a + 1))
737	–	canSteal = active = true;
738	–	if (canSteal && v.base == b++ &&
739	–	UNSAFE.compareAndSwapObject(q, u, t, null)) {
740	–	v.base = b;
741	–	v.stealHint = pid;
742	–	UNSAFE.putOrderedObject(this,
743	–	currentStealOffset, t);
744	–	seed = r;
745	–	++stealCount;
746	–	return t;
747	–	}
748	–	}
749	–	j = -n;
750	–	k = r;                // restart on contention
751	–	}
752	–	else if (++j <= 0)
753	–	k = r;
754	–	else if (j <= n)
755	–	k += (n >>> 1) | 1;
756	–	else
757	–	break;
758	–	}
759	–	}
760	–	return null;
761	–	}
762	–
763	–	// Run State management
764	–
765	–	// status check methods used mainly by ForkJoinPool
766	–	final boolean isRunning()    { return runState == 0; }
767	–	final boolean isTerminated() { return (runState & TERMINATED) != 0; }
768	–	final boolean isSuspended()  { return (runState & SUSPENDED) != 0; }
769	–	final boolean isTrimmed()    { return (runState & TRIMMED) != 0; }
770	–
771	–	final boolean isTerminating() {
772	–	if ((runState & TERMINATING) != 0)
773	–	return true;
774	–	if (pool.isAtLeastTerminating()) { // propagate pool state
775	–	shutdown();
776	–	return true;
777	–	}
778	–	return false;
779	–	}
780	–
781	–	/**
782	–	* Sets state to TERMINATING. Does NOT unpark or interrupt
783	–	* to wake up if currently blocked. Callers must do so if desired.
784	–	*/
785	–	final void shutdown() {
786	–	for (;;) {
787	–	int s = runState;
788	–	if ((s & (TERMINATING|TERMINATED)) != 0)
789	–	break;
790	–	if ((s & SUSPENDED) != 0) { // kill and wakeup if suspended
791	–	if (UNSAFE.compareAndSwapInt(this, runStateOffset, s,
792	–	(s & ~SUSPENDED) |
793	–	(TRIMMED|TERMINATING)))
794	–	break;
795	–	}
796	–	else if (UNSAFE.compareAndSwapInt(this, runStateOffset, s,
797	–	s | TERMINATING))
798	–	break;
799	–	}
800	–	}
801	–
802	–	/**
803	–	* Sets state to TERMINATED. Called only by onTermination().
804	–	*/
805	–	private void setTerminated() {
806	–	int s;
807	–	do {} while (!UNSAFE.compareAndSwapInt(this, runStateOffset,
808	–	s = runState,
809	–	s | (TERMINATING|TERMINATED)));
810	–	}
811	–
812	–	/**
813	–	* If suspended, tries to set status to unsuspended.
814	–	* Does NOT wake up if blocked.
815	–	*
816	–	* @return true if successful
817	–	*/
818	–	final boolean tryUnsuspend() {
819	–	int s;
820	–	while (((s = runState) & SUSPENDED) != 0) {
821	–	if (UNSAFE.compareAndSwapInt(this, runStateOffset, s,
822	–	s & ~SUSPENDED))
823	–	return true;
824	–	}
825	–	return false;
826	–	}
827	–
828	–	/**
829	–	* Sets suspended status and blocks as spare until resumed
830	–	* or shutdown.
831	–	*/
832	–	final void suspendAsSpare() {
833	–	for (;;) {                  // set suspended unless terminating
834	–	int s = runState;
835	–	if ((s & TERMINATING) != 0) { // must kill
836	–	if (UNSAFE.compareAndSwapInt(this, runStateOffset, s,
837	–	s | (TRIMMED | TERMINATING)))
838	–	return;
839	–	}
840	–	else if (UNSAFE.compareAndSwapInt(this, runStateOffset, s,
841	–	s | SUSPENDED))
842	–	break;
843	–	}
844	–	ForkJoinPool p = pool;
845	–	p.pushSpare(this);
846	–	while ((runState & SUSPENDED) != 0) {
847	–	if (p.tryAccumulateStealCount(this)) {
848	–	interrupted();          // clear/ignore interrupts
849	–	if ((runState & SUSPENDED) == 0)
850	–	break;
851	–	LockSupport.park(this);
852	–	}
853	–	}
854	–	}
855	–
856	–	// Misc support methods for ForkJoinPool
857	–
858	–	/**
859	–	* Returns an estimate of the number of tasks in the queue.  Also
860	–	* used by ForkJoinTask.
861	–	*/
862	–	final int getQueueSize() {
863	–	int n; // external calls must read base first
864	–	return (n = -base + sp) <= 0 ? 0 : n;
865	–	}
866	–
867	–	/**
868	–	* Removes and cancels all tasks in queue.  Can be called from any
869	–	* thread.
870	–	*/
871	–	final void cancelTasks() {
872	–	ForkJoinTask<?> cj = currentJoin; // try to cancel ongoing tasks
873	–	if (cj != null && cj.status >= 0) {
874	–	cj.cancelIgnoringExceptions();
875	–	try {
876	–	this.interrupt(); // awaken wait
877	–	} catch (SecurityException ignore) {
878	–	}
879	–	}
880	–	ForkJoinTask<?> cs = currentSteal;
881	–	if (cs != null && cs.status >= 0)
882	–	cs.cancelIgnoringExceptions();
883	–	while (base != sp) {
884	–	ForkJoinTask<?> t = deqTask();
885	–	if (t != null)
886	–	t.cancelIgnoringExceptions();
887	–	}
888	–	}
889	–
890	–	/**
891	–	* Drains tasks to given collection c.
892	–	*
893	–	* @return the number of tasks drained
894	–	*/
895	–	final int drainTasksTo(Collection<? super ForkJoinTask<?>> c) {
896	–	int n = 0;
897	–	while (base != sp) {
898	–	ForkJoinTask<?> t = deqTask();
899	–	if (t != null) {
900	–	c.add(t);
901	–	++n;
902	–	}
903	–	}
904	–	return n;
905	–	}
906	–
907	–	// Support methods for ForkJoinTask
908	–
909	–	/**
910	–	* Gets and removes a local task.
911	–	*
912	–	* @return a task, if available
913	–	*/
914	–	final ForkJoinTask<?> pollLocalTask() {
915	–	ForkJoinPool p = pool;
916	–	while (sp != base) {
917	–	int a; // inline p.tryIncrementActiveCount
918	–	if (active ||
919	–	(active = UNSAFE.compareAndSwapInt(p, poolRunStateOffset,
920	–	a = p.runState, a + 1)))
921	–	return locallyFifo ? locallyDeqTask() : popTask();
922	–	}
923	–	return null;
924	–	}
925	–
926	–	/**
927	–	* Gets and removes a local or stolen task.
928	–	*
929	–	* @return a task, if available
930	–	*/
931	–	final ForkJoinTask<?> pollTask() {
932	–	ForkJoinTask<?> t = pollLocalTask();
933	–	if (t == null) {
934	–	t = scan();
935	–	// cannot retain/track/help steal
936	–	UNSAFE.putOrderedObject(this, currentStealOffset, null);
937	–	}
938	–	return t;
939	–	}
940	–
941	–	/**
942	–	* Possibly runs some tasks and/or blocks, until task is done.
943	–	*
944	–	* @param joinMe the task to join
945	–	* @param timed true if use timed wait
946	–	* @param nanos wait time if timed
947	–	*/
948	–	final void joinTask(ForkJoinTask<?> joinMe, boolean timed, long nanos) {
949	–	// currentJoin only written by this thread; only need ordered store
950	–	ForkJoinTask<?> prevJoin = currentJoin;
951	–	UNSAFE.putOrderedObject(this, currentJoinOffset, joinMe);
952	–	pool.awaitJoin(joinMe, this, timed, nanos);
953	–	UNSAFE.putOrderedObject(this, currentJoinOffset, prevJoin);
954	–	}
955	–
956	–	/**
957	–	* Tries to locate and help perform tasks for a stealer of the
958	–	* given task, or in turn one of its stealers.  Traces
959	–	* currentSteal->currentJoin links looking for a thread working on
960	–	* a descendant of the given task and with a non-empty queue to
961	–	* steal back and execute tasks from.
962	–	*
963	–	* The implementation is very branchy to cope with potential
964	–	* inconsistencies or loops encountering chains that are stale,
965	–	* unknown, or of length greater than MAX_HELP_DEPTH links.  All
966	–	* of these cases are dealt with by just returning back to the
967	–	* caller, who is expected to retry if other join mechanisms also
968	–	* don't work out.
969	–	*
970	–	* @param joinMe the task to join
971	–	* @param running if false, then must update pool count upon
972	–	*  running a task
973	–	* @return value of running on exit
974	–	*/
975	–	final boolean helpJoinTask(ForkJoinTask<?> joinMe, boolean running) {
976	–	/*
977	–	* Initial checks to (1) abort if terminating; (2) clean out
978	–	* old cancelled tasks from local queue; (3) if joinMe is next
979	–	* task, run it; (4) omit scan if local queue nonempty (since
980	–	* it may contain non-descendents of joinMe).
981	–	*/
982	–	ForkJoinPool p = pool;
983	–	for (;;) {
984	–	ForkJoinTask<?>[] q;
985	–	int s;
986	–	if (joinMe.status < 0)
987	–	return running;
988	–	else if ((runState & TERMINATING) != 0)
989	–	joinMe.cancelIgnoringExceptions();
990	–	else if ((s = sp) == base || (q = queue) == null)
991	–	break;                            // queue empty
992	–	else {
993	–	int i = (q.length - 1) & --s;
994	–	long u = (i << qShift) + qBase;   // raw offset
995	–	ForkJoinTask<?> t = q[i];
996	–	if (t == null)
997	–	break;                        // lost to a stealer
998	–	else if (t != joinMe && t.status >= 0)
999	–	return running;               // cannot safely help
1000	–	else if ((running ||
1001	–	(running = p.tryIncrementRunningCount())) &&
1002	–	UNSAFE.compareAndSwapObject(q, u, t, null)) {
1003	–	sp = s; // putOrderedInt may encourage more timely write
1004	–	// UNSAFE.putOrderedInt(this, spOffset, s);
1005	–	if (t.status >= 0)
1006	–	t.quietlyExec();
1007	–	}
1008	–	}
1009	–	}
1010	–
1011	–	int n;                                    // worker array size
1012	–	ForkJoinWorkerThread[] ws = p.workers;
1013	–	if (ws != null && (n = ws.length) > 1) {  // need at least 2 workers
1014	–	ForkJoinTask<?> task = joinMe;        // base of chain
1015	–	ForkJoinWorkerThread thread = this;   // thread with stolen task
1016	–
1017	–	outer:for (int d = 0; d < MAX_HELP_DEPTH; ++d) { // chain length
1018	–	// Try to find v, the stealer of task, by first using hint
1019	–	ForkJoinWorkerThread v = ws[thread.stealHint & (n - 1)];
1020	–	if (v == null || v.currentSteal != task) {
1021	–	for (int j = 0; ; ++j) {      // search array
1022	–	if (j < n) {
1023	–	ForkJoinTask<?> vs;
1024	–	if ((v = ws[j]) != null &&
1025	–	(vs = v.currentSteal) != null) {
1026	–	if (joinMe.status < 0)
1027	–	break outer;
1028	–	if (vs == task) {
1029	–	if (task.status < 0)
1030	–	break outer; // stale
1031	–	thread.stealHint = j;
1032	–	break;        // save hint for next time
1033	–	}
1034	–	}
1035	–	}
1036	–	else
1037	–	break outer;          // no stealer
1038	–	}
1039	–	}
1040	–
1041	–	// Try to help v, using specialized form of deqTask
1042	–	for (;;) {
1043	–	if (joinMe.status < 0)
1044	–	break outer;
1045	–	int b = v.base;
1046	–	ForkJoinTask<?>[] q = v.queue;
1047	–	if (b == v.sp || q == null)
1048	–	break;                    // empty
1049	–	int i = (q.length - 1) & b;
1050	–	long u = (i << qShift) + qBase;
1051	–	ForkJoinTask<?> t = q[i];
1052	–	if (task.status < 0)
1053	–	break outer;              // stale
1054	–	if (t != null &&
1055	–	(running ||
1056	–	(running = p.tryIncrementRunningCount())) &&
1057	–	v.base == b++ &&
1058	–	UNSAFE.compareAndSwapObject(q, u, t, null)) {
1059	–	if (t != joinMe && joinMe.status < 0) {
1060	–	UNSAFE.putObjectVolatile(q, u, t);
1061	–	break outer;          // joinMe cancelled; back out
1062	–	}
1063	–	v.base = b;
1064	–	if (t.status >= 0) {
1065	–	ForkJoinTask<?> ps = currentSteal;
1066	–	int pid = poolIndex;
1067	–	v.stealHint = pid;
1068	–	UNSAFE.putOrderedObject(this,
1069	–	currentStealOffset, t);
1070	–	t.quietlyExec();
1071	–	UNSAFE.putOrderedObject(this,
1072	–	currentStealOffset, ps);
1073	–	}
1074	–	}
1075	–	}
1076	–
1077	–	// Try to descend to find v's stealer
1078	–	ForkJoinTask<?> next = v.currentJoin;
1079	–	if (task.status < 0 || next == null || next == task)
1080	–	break;                       // stale, dead-end, or cyclic
1081	–	if ((runState & TERMINATING) != 0)
1082	–	joinMe.cancelIgnoringExceptions();
1083	–	if (joinMe.status < 0)
1084	–	break;
1085	–	task = next;
1086	–	thread = v;
1087	–	}
1088	–	}
1089	–	return running;
1090	–	}
1091	–
1092	–	/**
1093	–	* Implements ForkJoinTask.getSurplusQueuedTaskCount().
1094	–	* Returns an estimate of the number of tasks, offset by a
1095	–	* function of number of idle workers.
1096	–	*
1097	–	* This method provides a cheap heuristic guide for task
1098	–	* partitioning when programmers, frameworks, tools, or languages
1099	–	* have little or no idea about task granularity.  In essence by
1100	–	* offering this method, we ask users only about tradeoffs in
1101	–	* overhead vs expected throughput and its variance, rather than
1102	–	* how finely to partition tasks.
1103	–	*
1104	–	* In a steady state strict (tree-structured) computation, each
1105	–	* thread makes available for stealing enough tasks for other
1106	–	* threads to remain active. Inductively, if all threads play by
1107	–	* the same rules, each thread should make available only a
1108	–	* constant number of tasks.
1109	–	*
1110	–	* The minimum useful constant is just 1. But using a value of 1
1111	–	* would require immediate replenishment upon each steal to
1112	–	* maintain enough tasks, which is infeasible.  Further,
1113	–	* partitionings/granularities of offered tasks should minimize
1114	–	* steal rates, which in general means that threads nearer the top
1115	–	* of computation tree should generate more than those nearer the
1116	–	* bottom. In perfect steady state, each thread is at
1117	–	* approximately the same level of computation tree. However,
1118	–	* producing extra tasks amortizes the uncertainty of progress and
1119	–	* diffusion assumptions.
1120	–	*
1121	–	* So, users will want to use values larger, but not much larger
1122	–	* than 1 to both smooth over transient shortages and hedge
1123	–	* against uneven progress; as traded off against the cost of
1124	–	* extra task overhead. We leave the user to pick a threshold
1125	–	* value to compare with the results of this call to guide
1126	–	* decisions, but recommend values such as 3.
1127	–	*
1128	–	* When all threads are active, it is on average OK to estimate
1129	–	* surplus strictly locally. In steady-state, if one thread is
1130	–	* maintaining say 2 surplus tasks, then so are others. So we can
1131	–	* just use estimated queue length (although note that (sp - base)
1132	–	* can be an overestimate because of stealers lagging increments
1133	–	* of base).  However, this strategy alone leads to serious
1134	–	* mis-estimates in some non-steady-state conditions (ramp-up,
1135	–	* ramp-down, other stalls). We can detect many of these by
1136	–	* further considering the number of "idle" threads, that are
1137	–	* known to have zero queued tasks, so compensate by a factor of
1138	–	* (#idle/#active) threads.
1139	–	*/
1140	–	final int getEstimatedSurplusTaskCount() {
1141	–	return sp - base - pool.idlePerActive();
1142	–	}
1143	–
1144	–	/**
1145	–	* Runs tasks until {@code pool.isQuiescent()}.
1146	–	*/
1147	–	final void helpQuiescePool() {
1148	–	ForkJoinTask<?> ps = currentSteal; // to restore below
1149	–	for (;;) {
1150	–	ForkJoinTask<?> t = pollLocalTask();
1151	–	if (t != null || (t = scan()) != null)
1152	–	t.quietlyExec();
1153	–	else {
1154	–	ForkJoinPool p = pool;
1155	–	int a; // to inline CASes
1156	–	if (active) {
1157	–	if (!UNSAFE.compareAndSwapInt
1158	–	(p, poolRunStateOffset, a = p.runState, a - 1))
1159	–	continue;   // retry later
1160	–	active = false; // inactivate
1161	–	UNSAFE.putOrderedObject(this, currentStealOffset, ps);
1162	–	}
1163	–	if (p.isQuiescent()) {
1164	–	active = true; // re-activate
1165	–	do {} while (!UNSAFE.compareAndSwapInt
1166	–	(p, poolRunStateOffset, a = p.runState, a+1));
1167	–	return;
1168	–	}
1169	–	}
1170	–	}
1171	–	}
1172	–
1173	–	// Unsafe mechanics
1174	–
1175	–	private static final sun.misc.Unsafe UNSAFE = getUnsafe();
1176	–	private static final long spOffset =
1177	–	objectFieldOffset("sp", ForkJoinWorkerThread.class);
1178	–	private static final long runStateOffset =
1179	–	objectFieldOffset("runState", ForkJoinWorkerThread.class);
1180	–	private static final long currentJoinOffset =
1181	–	objectFieldOffset("currentJoin", ForkJoinWorkerThread.class);
1182	–	private static final long currentStealOffset =
1183	–	objectFieldOffset("currentSteal", ForkJoinWorkerThread.class);
1184	–	private static final long qBase =
1185	–	UNSAFE.arrayBaseOffset(ForkJoinTask[].class);
1186	–	private static final long poolRunStateOffset = // to inline CAS
1187	–	objectFieldOffset("runState", ForkJoinPool.class);
1188	–
1189	–	private static final int qShift;
1190	–
1191	–	static {
1192	–	int s = UNSAFE.arrayIndexScale(ForkJoinTask[].class);
1193	–	if ((s & (s-1)) != 0)
1194	–	throw new Error("data type scale not a power of two");
1195	–	qShift = 31 - Integer.numberOfLeadingZeros(s);
1196	–	MAXIMUM_QUEUE_CAPACITY = 1 << (31 - qShift);
1197	–	}
1198	–
1199	–	private static long objectFieldOffset(String field, Class<?> klazz) {
1200	–	try {
1201	–	return UNSAFE.objectFieldOffset(klazz.getDeclaredField(field));
1202	–	} catch (NoSuchFieldException e) {
1203	–	// Convert Exception to corresponding Error
1204	–	NoSuchFieldError error = new NoSuchFieldError(field);
1205	–	error.initCause(e);
1206	–	throw error;
1207	–	}
1208	–	}
1209	–
1210	–	/**
1211	–	* Returns a sun.misc.Unsafe.  Suitable for use in a 3rd party package.
1212	–	* Replace with a simple call to Unsafe.getUnsafe when integrating
1213	–	* into a jdk.
1214	–	*
1215	–	* @return a sun.misc.Unsafe
1216	–	*/
1217	–	private static sun.misc.Unsafe getUnsafe() {
1218	–	try {
1219	–	return sun.misc.Unsafe.getUnsafe();
1220	–	} catch (SecurityException se) {
128		try {
129	<	return java.security.AccessController.doPrivileged
130	<	(new java.security
131	<	.PrivilegedExceptionAction<sun.misc.Unsafe>() {
132	<	public sun.misc.Unsafe run() throws Exception {
133	<	java.lang.reflect.Field f = sun.misc
134	<	.Unsafe.class.getDeclaredField("theUnsafe");
1228	<	f.setAccessible(true);
1229	<	return (sun.misc.Unsafe) f.get(null);
1230	<	}});
1231	<	} catch (java.security.PrivilegedActionException e) {
1232	<	throw new RuntimeException("Could not initialize intrinsics",
1233	<	e.getCause());
129	>	onTermination(exception);
130	>	} catch (Throwable ex) {
131	>	if (exception == null)
132	>	exception = ex;
133	>	} finally {
134	>	pool.deregisterWorker(this, exception);
135		}
136		}
137		}

Diff Legend

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