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 1.35 by dl, Wed Jul 7 19:52:32 2010 UTC vs.
Revision 1.47 by jsr166, Tue Sep 7 07:10:53 2010 UTC

#	Line 85 | Line 85 | public class ForkJoinWorkerThread extend
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 stolen) the most recent task it stole
89	<	* from some other worker. Plus, it records (in field joining) the
90	<	* task it is currently actively joining. Method joinTask uses
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
#	Line 97 | Line 97 | public class ForkJoinWorkerThread extend
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 maintain per-task bookkeeping.  This
101	<	* requires a linear scan of workers array to locate stealers,
102	<	* which isolates cost to when it is needed, rather than adding to
103	<	* per-task overhead.  (2) It is "shallow", ignoring nesting and
104	<	* potentially cyclic mutual steals.  (3) It is intentionally
105	<	* racy: field joining is updated only while actively joining,
106	<	* which means that we could miss links in the chain during
107	<	* long-lived tasks, GC stalls etc.  (4) We fall back to
108	<	* suspending the worker and if necessary replacing it with a
109	<	* spare (see ForkJoinPool.tryAwaitJoin).
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 on
116	<	* an uncomfortable amount of "Unsafe" mechanics. To maintain
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	<	* scanWhileJoining.)  Because they are protected by volatile base
123	<	* reads, reads of the queue array and its slots by other threads
124	<	* do not need volatile load semantics, but writes (in push)
125	<	* require store order and CASes (in pop and deq) require
126	<	* (volatile) CAS semantics.  (Michael, Saraswat, and Vechev's
127	<	* algorithm has similar properties, but without support for
128	<	* nulling slots.)  Since these combinations aren't supported
129	<	* using ordinary volatiles, the only way to accomplish these
130	<	* efficiently is to use direct Unsafe calls. (Using external
131	<	* AtomicIntegers and AtomicReferenceArrays for the indices and
132	<	* array is significantly slower because of memory locality and
133	<	* indirection effects.)
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.)
134		*
135		* Further, performance on most platforms is very sensitive to
136		* placement and sizing of the (resizable) queue array.  Even
#	Line 151 | Line 155 | public class ForkJoinWorkerThread extend
155		private static final Random seedGenerator = new Random();
156
157		/**
158	<	* The timeout value for suspending spares. Spare workers that
159	<	* remain unsignalled for more than this time may be trimmed
160	<	* (killed and removed from pool).  Since our goal is to avoid
161	<	* long-term thread buildup, the exact value of timeout does not
162	<	* matter too much so long as it avoids most false-alarm timeouts
159	<	* under GC stalls or momentarily high system load.
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 long SPARE_KEEPALIVE_NANOS =
162	<	5L * 1000L * 1000L * 1000L; // 5 secs
164	>	private static final int MAX_HELP_DEPTH = 8;
165
166		/**
167		* Capacity of work-stealing queue array upon initialization.
#	Line 170 | Line 172 | public class ForkJoinWorkerThread extend
172
173		/**
174		* Maximum work-stealing queue array size.  Must be less than or
175	<	* equal to 1 << 28 to ensure lack of index wraparound. (This
176	<	* is less than usual bounds, because we need leftshift by 3
177	<	* to be in int range).
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 = 1 << 28;
179	>	private static final int MAXIMUM_QUEUE_CAPACITY;
180
181		/**
182		* The pool this thread works in. Accessed directly by ForkJoinTask.
#	Line 182 | Line 184 | public class ForkJoinWorkerThread extend
184		final ForkJoinPool pool;
185
186		/**
185	–	* The task most recently stolen from another worker
186	–	*/
187	–	private volatile ForkJoinTask<?> stolen;
188	–
189	–	/**
190	–	* The task currently being joined, set only when actively
191	–	* trying to helpStealer.
192	–	*/
193	–	private volatile ForkJoinTask<?> joining;
194	–
195	–	/**
187		* The work-stealing queue array. Size must be a power of two.
188		* Initialized in onStart, to improve memory locality.
189		*/
#	Line 214 | Line 205 | public class ForkJoinWorkerThread extend
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.
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.
224		*/
225	<	private volatile int runState;
225	>	volatile int runState;
226
227		private static final int TERMINATING = 0x01;
228		private static final int TERMINATED  = 0x02;
#	Line 227 | Line 230 | public class ForkJoinWorkerThread extend
230		private static final int TRIMMED     = 0x08; // killed while suspended
231
232		/**
233	<	* Number of LockSupport.park calls to block this thread for
234	<	* suspension or event waits. Used for internal instrumention;
232	<	* currently not exported but included because volatile write upon
233	<	* park also provides a workaround for a JVM bug.
234	<	*/
235	<	volatile int parkCount;
236	<
237	<	/**
238	<	* Number of steals, transferred and reset in pool callbacks pool
239	<	* when idle Accessed directly by pool.
233	>	* Number of steals. Directly accessed (and reset) by
234	>	* pool.tryAccumulateStealCount when idle.
235		*/
236		int stealCount;
237
#	Line 254 | Line 249 | public class ForkJoinWorkerThread extend
249
250		/**
251		* True if use local fifo, not default lifo, for local polling.
252	<	* Shadows value from ForkJoinPool, which resets it if changed
258	<	* pool-wide.
252	>	* Shadows value from ForkJoinPool.
253		*/
254		private final boolean locallyFifo;
255	<
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
#	Line 273 | Line 267 | public class ForkJoinWorkerThread extend
267		int lastEventCount;
268
269		/**
270	<	* Encoded index and event count of next event waiter. Used only
271	<	* by ForkJoinPool for managing event waiters.
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;
300	+
301	+	/**
302		* Creates a ForkJoinWorkerThread operating in the given pool.
303		*
304		* @param pool the pool this thread works in
#	Line 287 | Line 307 | public class ForkJoinWorkerThread extend
307		protected ForkJoinWorkerThread(ForkJoinPool pool) {
308		this.pool = pool;
309		this.locallyFifo = pool.locallyFifo;
310	+	setDaemon(true);
311		// To avoid exposing construction details to subclasses,
312		// remaining initialization is in start() and onStart()
313		}
314
315		/**
316	<	* Performs additional initialization and starts this thread
316	>	* Performs additional initialization and starts this thread.
317		*/
318		final void start(int poolIndex, UncaughtExceptionHandler ueh) {
319		this.poolIndex = poolIndex;
320		if (ueh != null)
321		setUncaughtExceptionHandler(ueh);
301	–	setDaemon(true);
322		start();
323		}
324
#	Line 329 | Line 349 | public class ForkJoinWorkerThread extend
349		/**
350		* Initializes internal state after construction but before
351		* processing any tasks. If you override this method, you must
352	<	* invoke super.onStart() at the beginning of the method.
352	>	* invoke @code{super.onStart()} at the beginning of the method.
353		* Initialization requires care: Most fields must have legal
354		* default values, to ensure that attempted accesses from other
355		* threads work correctly even before this thread starts
#	Line 357 | Line 377 | public class ForkJoinWorkerThread extend
377		*/
378		protected void onTermination(Throwable exception) {
379		try {
380	<	stolen = null;
381	<	joining = null;
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	<	pool.workerTerminated(this);
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;
#	Line 391 | Line 416 | public class ForkJoinWorkerThread extend
416		// helpers for run()
417
418		/**
419	<	* Find and execute tasks and check status while running
419	>	* Finds and executes tasks, and checks status while running.
420		*/
421		private void mainLoop() {
422	<	boolean ran = false;      // true if ran task in last loop iter
398	<	boolean prevRan = false;  // true if ran on last or previous step
422	>	boolean ran = false; // true if ran a task on last step
423		ForkJoinPool p = pool;
424		for (;;) {
425	<	p.preStep(this, prevRan);
425	>	p.preStep(this, ran);
426		if (runState != 0)
427	<	return;
428	<	ForkJoinTask<?> t; // try to get and run stolen or submitted task
405	<	if ((t = scan()) != null || (t = pollSubmission()) != null) {
406	<	t.tryExec();
407	<	if (base != sp)
408	<	runLocalTasks();
409	<	stolen = null;
410	<	prevRan = ran = true;
411	<	}
412	<	else {
413	<	prevRan = ran;
414	<	ran = false;
415	<	}
427	>	break;
428	>	ran = tryExecSteal() || tryExecSubmission();
429		}
430		}
431
432		/**
433	<	* Runs local tasks until queue is empty or shut down.  Call only
434	<	* while active.
433	>	* Tries to steal a task and execute it.
434	>	*
435	>	* @return true if ran a task
436		*/
437	<	private void runLocalTasks() {
438	<	while (runState == 0) {
439	<	ForkJoinTask<?> t = locallyFifo? locallyDeqTask() : popTask();
440	<	if (t != null)
441	<	t.tryExec();
442	<	else if (base == sp)
443	<	break;
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 take it
450	>	* If a submission exists, try to activate and run it.
451		*
452	<	* @return a task, if available
452	>	* @return true if ran a task
453		*/
454	<	private ForkJoinTask<?> pollSubmission() {
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	<	if (active || (active = p.tryIncrementActiveCount())) {
461	<	ForkJoinTask<?> t = p.pollSubmission();
462	<	return t != null ? t : scan(); // if missed, rescan
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 null;
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		/*
#	Line 454 | Line 496 | public class ForkJoinWorkerThread extend
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 protected by other volatile reads and are
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
#	Line 478 | Line 520 | public class ForkJoinWorkerThread extend
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) {
523	>	ForkJoinTask<?> t) {
524		UNSAFE.putObjectVolatile(q, (i << qShift) + qBase, t);
525		}
526
#	Line 511 | Line 553 | public class ForkJoinWorkerThread extend
553		ForkJoinTask<?> t;
554		ForkJoinTask<?>[] q;
555		int b, i;
556	<	if ((b = base) != sp &&
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)) {
#	Line 523 | Line 565 | public class ForkJoinWorkerThread extend
565
566		/**
567		* Tries to take a task from the base of own queue. Assumes active
568	<	* status.  Called only by current thread.
568	>	* status.  Called only by this thread.
569		*
570		* @return a task, or null if none
571		*/
#	Line 546 | Line 588 | public class ForkJoinWorkerThread extend
588
589		/**
590		* Returns a popped task, or null if empty. Assumes active status.
591	<	* Called only by current thread. (Note: a specialization of this
550	<	* code appears in popWhileJoining.)
591	>	* Called only by this thread.
592		*/
593	<	final ForkJoinTask<?> popTask() {
594	<	int s;
595	<	ForkJoinTask<?>[] q;
596	<	if (base != (s = sp) && (q = queue) != null) {
597	<	int i = (q.length - 1) & --s;
598	<	ForkJoinTask<?> t = q[i];
599	<	if (t != null && UNSAFE.compareAndSwapObject
600	<	(q, (i << qShift) + qBase, t, null)) {
601	<	sp = s;
602	<	return t;
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	>	sp = s; // putOrderedInt may encourage more timely write
605	>	// UNSAFE.putOrderedInt(this, spOffset, s);
606	>	return t;
607	>	}
608		}
609		}
610		return null;
#	Line 566 | Line 612 | public class ForkJoinWorkerThread extend
612
613		/**
614		* Specialized version of popTask to pop only if topmost element
615	<	* is the given task. Called only by current thread while
570	<	* active.
615	>	* is the given task. Called only by this thread while active.
616		*
617		* @param t the task. Caller must ensure non-null.
618		*/
619		final boolean unpushTask(ForkJoinTask<?> t) {
620		int s;
621	<	ForkJoinTask<?>[] q;
622	<	if (base != (s = sp) && (q = queue) != null &&
621	>	ForkJoinTask<?>[] q = queue;
622	>	if ((s = sp) != base && q != null &&
623		UNSAFE.compareAndSwapObject
624		(q, (((q.length - 1) & --s) << qShift) + qBase, t, null)) {
625	<	sp = s;
625	>	sp = s; // putOrderedInt may encourage more timely write
626	>	// UNSAFE.putOrderedInt(this, spOffset, s);
627		return true;
628		}
629		return false;
630		}
631
632		/**
633	<	* Returns next task or null if empty or contended
633	>	* Returns next task, or null if empty or contended.
634		*/
635		final ForkJoinTask<?> peekTask() {
636		ForkJoinTask<?>[] q = queue;
#	Line 626 | Line 672 | public class ForkJoinWorkerThread extend
672		* Computes next value for random victim probe in scan().  Scans
673		* don't require a very high quality generator, but also not a
674		* crummy one.  Marsaglia xor-shift is cheap and works well enough.
675	<	* Note: This is manually inlined in scan()
675	>	* Note: This is manually inlined in scan().
676		*/
677		private static final int xorShift(int r) {
678		r ^= r << 13;
#	Line 665 | Line 711 | public class ForkJoinWorkerThread extend
711		for (;;) {
712		ForkJoinWorkerThread v = ws[k & mask];
713		r ^= r << 13; r ^= r >>> 17; r ^= r << 5; // inline xorshift
714	<	if (v != null && v.base != v.sp) {
715	<	if (canSteal ||       // ensure active status
716	<	(canSteal = active = p.tryIncrementActiveCount())) {
717	<	int b = v.base;   // inline specialized deqTask
718	<	ForkJoinTask<?>[] q;
719	<	if (b != v.sp && (q = v.queue) != null) {
720	<	ForkJoinTask<?> t;
721	<	int i = (q.length - 1) & b;
722	<	long u = (i << qShift) + qBase; // raw offset
723	<	if ((t = q[i]) != null && v.base == b &&
724	<	UNSAFE.compareAndSwapObject(q, u, t, null)) {
725	<	stolen = t;
726	<	v.base = b + 1;
727	<	seed = r;
728	<	++stealCount;
729	<	return t;
730	<	}
714	>	ForkJoinTask<?>[] q; ForkJoinTask<?> t; int b, a;
715	>	if (v != null && (b = v.base) != v.sp &&
716	>	(q = v.queue) != null) {
717	>	int i = (q.length - 1) & b;
718	>	long u = (i << qShift) + qBase; // raw offset
719	>	int pid = poolIndex;
720	>	if ((t = q[i]) != null) {
721	>	if (!canSteal &&  // inline p.tryIncrementActiveCount
722	>	UNSAFE.compareAndSwapInt(p, poolRunStateOffset,
723	>	a = p.runState, a + 1))
724	>	canSteal = active = true;
725	>	if (canSteal && v.base == b++ &&
726	>	UNSAFE.compareAndSwapObject(q, u, t, null)) {
727	>	v.base = b;
728	>	v.stealHint = pid;
729	>	UNSAFE.putOrderedObject(this,
730	>	currentStealOffset, t);
731	>	seed = r;
732	>	++stealCount;
733	>	return t;
734		}
735		}
736		j = -n;
#	Line 701 | Line 750 | public class ForkJoinWorkerThread extend
750		// Run State management
751
752		// status check methods used mainly by ForkJoinPool
753	+	final boolean isRunning()     { return runState == 0; }
754		final boolean isTerminating() { return (runState & TERMINATING) != 0; }
755		final boolean isTerminated()  { return (runState & TERMINATED) != 0; }
756		final boolean isSuspended()   { return (runState & SUSPENDED) != 0; }
757		final boolean isTrimmed()     { return (runState & TRIMMED) != 0; }
758
759		/**
760	<	* Sets state to TERMINATING, also resuming if suspended.
760	>	* Sets state to TERMINATING. Does NOT unpark or interrupt
761	>	* to wake up if currently blocked. Callers must do so if desired.
762		*/
763		final void shutdown() {
764		for (;;) {
765		int s = runState;
766	+	if ((s & (TERMINATING|TERMINATED)) != 0)
767	+	break;
768		if ((s & SUSPENDED) != 0) { // kill and wakeup if suspended
769		if (UNSAFE.compareAndSwapInt(this, runStateOffset, s,
770		(s & ~SUSPENDED) |
771	<	(TRIMMED|TERMINATING))) {
719	<	LockSupport.unpark(this);
771	>	(TRIMMED|TERMINATING)))
772		break;
721	–	}
773		}
774		else if (UNSAFE.compareAndSwapInt(this, runStateOffset, s,
775		s | TERMINATING))
#	Line 727 | Line 778 | public class ForkJoinWorkerThread extend
778		}
779
780		/**
781	<	* Sets state to TERMINATED. Called only by this thread.
781	>	* Sets state to TERMINATED. Called only by onTermination().
782		*/
783		private void setTerminated() {
784		int s;
#	Line 737 | Line 788 | public class ForkJoinWorkerThread extend
788		}
789
790		/**
740	–	* Instrumented version of park used by ForkJoinPool.awaitEvent
741	–	*/
742	–	final void doPark() {
743	–	++parkCount;
744	–	LockSupport.park(this);
745	–	}
746	–
747	–	/**
791		* If suspended, tries to set status to unsuspended.
792	<	* Caller must unpark to actually resume
792	>	* Does NOT wake up if blocked.
793		*
794		* @return true if successful
795		*/
796		final boolean tryUnsuspend() {
797	<	int s = runState;
798	<	if ((s & SUSPENDED) != 0)
799	<	return UNSAFE.compareAndSwapInt(this, runStateOffset, s,
800	<	s & ~SUSPENDED);
797	>	int s;
798	>	while (((s = runState) & SUSPENDED) != 0) {
799	>	if (UNSAFE.compareAndSwapInt(this, runStateOffset, s,
800	>	s & ~SUSPENDED))
801	>	return true;
802	>	}
803		return false;
804		}
805
806		/**
807	<	* Sets suspended status and blocks as spare until resumed,
808	<	* shutdown, or timed out.
764	<	*
765	<	* @return false if trimmed
807	>	* Sets suspended status and blocks as spare until resumed
808	>	* or shutdown.
809		*/
810	<	final boolean suspendAsSpare() {
811	<	for (;;) {               // set suspended unless terminating
810	>	final void suspendAsSpare() {
811	>	for (;;) {                  // set suspended unless terminating
812		int s = runState;
813		if ((s & TERMINATING) != 0) { // must kill
814		if (UNSAFE.compareAndSwapInt(this, runStateOffset, s,
815		s | (TRIMMED | TERMINATING)))
816	<	return false;
816	>	return;
817		}
818		else if (UNSAFE.compareAndSwapInt(this, runStateOffset, s,
819		s | SUSPENDED))
820		break;
821		}
822	<	boolean timed;
823	<	long nanos;
781	<	long startTime;
782	<	if (poolIndex < pool.parallelism) {
783	<	timed = false;
784	<	nanos = 0L;
785	<	startTime = 0L;
786	<	}
787	<	else {
788	<	timed = true;
789	<	nanos = SPARE_KEEPALIVE_NANOS;
790	<	startTime = System.nanoTime();
791	<	}
792	<	pool.accumulateStealCount(this);
793	<	lastEventCount = 0;      // reset upon resume
794	<	interrupted();           // clear/ignore interrupts
822	>	ForkJoinPool p = pool;
823	>	p.pushSpare(this);
824		while ((runState & SUSPENDED) != 0) {
825	<	++parkCount;
826	<	if (!timed)
825	>	if (p.tryAccumulateStealCount(this)) {
826	>	interrupted();          // clear/ignore interrupts
827	>	if ((runState & SUSPENDED) == 0)
828	>	break;
829		LockSupport.park(this);
799	–	else if ((nanos -= (System.nanoTime() - startTime)) > 0)
800	–	LockSupport.parkNanos(this, nanos);
801	–	else { // try to trim on timeout
802	–	int s = runState;
803	–	if (UNSAFE.compareAndSwapInt(this, runStateOffset, s,
804	–	(s & ~SUSPENDED) |
805	–	(TRIMMED|TERMINATING)))
806	–	return false;
830		}
831		}
809	–	return true;
832		}
833
834		// Misc support methods for ForkJoinPool
#	Line 816 | Line 838 | public class ForkJoinWorkerThread extend
838		* used by ForkJoinTask.
839		*/
840		final int getQueueSize() {
841	<	return -base + sp;
841	>	int n; // external calls must read base first
842	>	return (n = -base + sp) <= 0 ? 0 : n;
843		}
844
845		/**
#	Line 824 | Line 847 | public class ForkJoinWorkerThread extend
847		* thread.
848		*/
849		final void cancelTasks() {
850	+	ForkJoinTask<?> cj = currentJoin; // try to cancel ongoing tasks
851	+	if (cj != null) {
852	+	currentJoin = null;
853	+	cj.cancelIgnoringExceptions();
854	+	try {
855	+	this.interrupt(); // awaken wait
856	+	} catch (SecurityException ignore) {
857	+	}
858	+	}
859	+	ForkJoinTask<?> cs = currentSteal;
860	+	if (cs != null) {
861	+	currentSteal = null;
862	+	cs.cancelIgnoringExceptions();
863	+	}
864		while (base != sp) {
865		ForkJoinTask<?> t = deqTask();
866		if (t != null)
#	Line 851 | Line 888 | public class ForkJoinWorkerThread extend
888		// Support methods for ForkJoinTask
889
890		/**
891	+	* Gets and removes a local task.
892	+	*
893	+	* @return a task, if available
894	+	*/
895	+	final ForkJoinTask<?> pollLocalTask() {
896	+	ForkJoinPool p = pool;
897	+	while (sp != base) {
898	+	int a; // inline p.tryIncrementActiveCount
899	+	if (active ||
900	+	(active = UNSAFE.compareAndSwapInt(p, poolRunStateOffset,
901	+	a = p.runState, a + 1)))
902	+	return locallyFifo ? locallyDeqTask() : popTask();
903	+	}
904	+	return null;
905	+	}
906	+
907	+	/**
908	+	* Gets and removes a local or stolen task.
909	+	*
910	+	* @return a task, if available
911	+	*/
912	+	final ForkJoinTask<?> pollTask() {
913	+	ForkJoinTask<?> t = pollLocalTask();
914	+	if (t == null) {
915	+	t = scan();
916	+	// cannot retain/track/help steal
917	+	UNSAFE.putOrderedObject(this, currentStealOffset, null);
918	+	}
919	+	return t;
920	+	}
921	+
922	+	/**
923		* Possibly runs some tasks and/or blocks, until task is done.
924		*
925		* @param joinMe the task to join
926		*/
927		final void joinTask(ForkJoinTask<?> joinMe) {
928	<	ForkJoinTask<?> prevJoining = joining;
929	<	joining = joinMe;
930	<	while (joinMe.status >= 0) {
931	<	int s = sp;
932	<	if (s == base) {
933	<	nonlocalJoinTask(joinMe);
934	<	break;
935	<	}
936	<	// process local task
937	<	ForkJoinTask<?> t;
938	<	ForkJoinTask<?>[] q = queue;
928	>	// currentJoin only written by this thread; only need ordered store
929	>	ForkJoinTask<?> prevJoin = currentJoin;
930	>	UNSAFE.putOrderedObject(this, currentJoinOffset, joinMe);
931	>	if (sp != base)
932	>	localHelpJoinTask(joinMe);
933	>	if (joinMe.status >= 0)
934	>	pool.awaitJoin(joinMe, this);
935	>	UNSAFE.putOrderedObject(this, currentJoinOffset, prevJoin);
936	>	}
937	>
938	>	/**
939	>	* Run tasks in local queue until given task is done.
940	>	*
941	>	* @param joinMe the task to join
942	>	*/
943	>	private void localHelpJoinTask(ForkJoinTask<?> joinMe) {
944	>	int s;
945	>	ForkJoinTask<?>[] q;
946	>	while (joinMe.status >= 0 && (s = sp) != base && (q = queue) != null) {
947		int i = (q.length - 1) & --s;
948		long u = (i << qShift) + qBase; // raw offset
949	<	if ((t = q[i]) != null &&
950	<	UNSAFE.compareAndSwapObject(q, u, t, null)) {
949	>	ForkJoinTask<?> t = q[i];
950	>	if (t == null)  // lost to a stealer
951	>	break;
952	>	if (UNSAFE.compareAndSwapObject(q, u, t, null)) {
953		/*
954	<	* This recheck (and similarly in nonlocalJoinTask)
954	>	* This recheck (and similarly in helpJoinTask)
955		* handles cases where joinMe is independently
956		* cancelled or forced even though there is other work
957		* available. Back out of the pop by putting t back
958	<	* into slot before we commit by setting sp.
958	>	* into slot before we commit by writing sp.
959		*/
960		if (joinMe.status < 0) {
961		UNSAFE.putObjectVolatile(q, u, t);
962		break;
963		}
964		sp = s;
965	<	t.tryExec();
965	>	// UNSAFE.putOrderedInt(this, spOffset, s);
966	>	t.quietlyExec();
967		}
968		}
889	–	joining = prevJoining;
969		}
970
971		/**
972	<	* Tries to locate and help perform tasks for a stealer of the
973	<	* given task (or in turn one of its stealers), blocking (via
974	<	* pool.tryAwaitJoin) upon failure to find work.  Traces
975	<	* stolen->joining links looking for a thread working on
976	<	* a descendant of the given task and with a non-empty queue to
977	<	* steal back and execute tasks from. Inhibits mutual steal chains
978	<	* and scans on outer joins upon nesting to avoid unbounded
979	<	* growth.  Restarts search upon encountering inconsistencies.
980	<	* Tries to block if two passes agree that there are no remaining
981	<	* targets.
972	>	* Unless terminating, tries to locate and help perform tasks for
973	>	* a stealer of the given task, or in turn one of its stealers.
974	>	* Traces currentSteal->currentJoin links looking for a thread
975	>	* working on a descendant of the given task and with a non-empty
976	>	* queue to steal back and execute tasks from.
977	>	*
978	>	* The implementation is very branchy to cope with potential
979	>	* inconsistencies or loops encountering chains that are stale,
980	>	* unknown, or of length greater than MAX_HELP_DEPTH links.  All
981	>	* of these cases are dealt with by just returning back to the
982	>	* caller, who is expected to retry if other join mechanisms also
983	>	* don't work out.
984		*
985		* @param joinMe the task to join
986		*/
987	<	private void nonlocalJoinTask(ForkJoinTask<?> joinMe) {
988	<	ForkJoinPool p = pool;
989	<	int scans = p.parallelism;       // give up if too many retries
990	<	ForkJoinTask<?> bottom = null;   // target seen when can't descend
991	<	restart: while (joinMe.status >= 0) {
992	<	ForkJoinTask<?> target = null;
993	<	ForkJoinTask<?> next = joinMe;
994	<	while (scans >= 0 && next != null) {
995	<	--scans;
996	<	target = next;
997	<	next = null;
998	<	ForkJoinWorkerThread v = null;
999	<	ForkJoinWorkerThread[] ws = p.workers;
1000	<	int n = ws.length;
1001	<	for (int j = 0; j < n; ++j) {
1002	<	ForkJoinWorkerThread w = ws[j];
1003	<	if (w != null && w.stolen == target) {
1004	<	v = w;
1005	<	break;
1006	<	}
1007	<	}
1008	<	if (v != null && v != this) {
1009	<	ForkJoinTask<?> prevStolen = stolen;
1010	<	int b;
1011	<	ForkJoinTask<?>[] q;
1012	<	while ((b = v.base) != v.sp && (q = v.queue) != null) {
1013	<	int i = (q.length - 1) & b;
1014	<	long u = (i << qShift) + qBase;
934	<	ForkJoinTask<?> t = q[i];
935	<	if (target.status < 0)
936	<	continue restart;
937	<	if (t != null && v.base == b &&
938	<	UNSAFE.compareAndSwapObject(q, u, t, null)) {
939	<	if (joinMe.status < 0) {
940	<	UNSAFE.putObjectVolatile(q, u, t);
941	<	return; // back out
987	>	final void helpJoinTask(ForkJoinTask<?> joinMe) {
988	>	ForkJoinWorkerThread[] ws;
989	>	int n;
990	>	if (joinMe.status < 0)                // already done
991	>	return;
992	>	if ((runState & TERMINATING) != 0) {  // cancel if shutting down
993	>	joinMe.cancelIgnoringExceptions();
994	>	return;
995	>	}
996	>	if ((ws = pool.workers) == null || (n = ws.length) <= 1)
997	>	return;                           // need at least 2 workers
998	>
999	>	ForkJoinTask<?> task = joinMe;        // base of chain
1000	>	ForkJoinWorkerThread thread = this;   // thread with stolen task
1001	>	for (int d = 0; d < MAX_HELP_DEPTH; ++d) { // chain length
1002	>	// Try to find v, the stealer of task, by first using hint
1003	>	ForkJoinWorkerThread v = ws[thread.stealHint & (n - 1)];
1004	>	if (v == null || v.currentSteal != task) {
1005	>	for (int j = 0; ; ++j) {      // search array
1006	>	if (j < n) {
1007	>	ForkJoinTask<?> vs;
1008	>	if ((v = ws[j]) != null &&
1009	>	(vs = v.currentSteal) != null) {
1010	>	if (joinMe.status < 0 || task.status < 0)
1011	>	return;       // stale or done
1012	>	if (vs == task) {
1013	>	thread.stealHint = j;
1014	>	break;        // save hint for next time
1015		}
943	–	stolen = t;
944	–	v.base = b + 1;
945	–	t.tryExec();
946	–	stolen = prevStolen;
1016		}
948	–	if (joinMe.status < 0)
949	–	return;
1017		}
1018	<	next = v.joining;
1018	>	else
1019	>	return;               // no stealer
1020		}
1021	<	if (target.status < 0)
1022	<	continue restart;  // inconsistent
1023	<	if (joinMe.status < 0)
1021	>	}
1022	>	for (;;) { // Try to help v, using specialized form of deqTask
1023	>	if (joinMe.status < 0)
1024		return;
1025	+	int b = v.base;
1026	+	ForkJoinTask<?>[] q = v.queue;
1027	+	if (b == v.sp || q == null)
1028	+	break;
1029	+	int i = (q.length - 1) & b;
1030	+	long u = (i << qShift) + qBase;
1031	+	ForkJoinTask<?> t = q[i];
1032	+	int pid = poolIndex;
1033	+	ForkJoinTask<?> ps = currentSteal;
1034	+	if (task.status < 0)
1035	+	return;                   // stale or done
1036	+	if (t != null && v.base == b++ &&
1037	+	UNSAFE.compareAndSwapObject(q, u, t, null)) {
1038	+	if (joinMe.status < 0) {
1039	+	UNSAFE.putObjectVolatile(q, u, t);
1040	+	return;               // back out on cancel
1041	+	}
1042	+	v.base = b;
1043	+	v.stealHint = pid;
1044	+	UNSAFE.putOrderedObject(this, currentStealOffset, t);
1045	+	t.quietlyExec();
1046	+	UNSAFE.putOrderedObject(this, currentStealOffset, ps);
1047	+	}
1048		}
1049	<
1050	<	if (bottom != target)
1051	<	bottom = target;    // recheck landing spot
1052	<	else if (p.tryAwaitJoin(joinMe) < 0)
1053	<	return;             // successfully blocked
1054	<	Thread.yield();         // tame spin in case too many active
1049	>	// Try to descend to find v's stealer
1050	>	ForkJoinTask<?> next = v.currentJoin;
1051	>	if (task.status < 0 || next == null || next == task ||
1052	>	joinMe.status < 0)
1053	>	return;
1054	>	task = next;
1055	>	thread = v;
1056		}
1057		}
1058
1059		/**
1060	+	* Implements ForJoinTask.getSurplusQueuedTaskCount().
1061		* Returns an estimate of the number of tasks, offset by a
1062		* function of number of idle workers.
1063		*
#	Line 1016 | Line 1109 | public class ForkJoinWorkerThread extend
1109		}
1110
1111		/**
1019	–	* Gets and removes a local task.
1020	–	*
1021	–	* @return a task, if available
1022	–	*/
1023	–	final ForkJoinTask<?> pollLocalTask() {
1024	–	while (sp != base) {
1025	–	if (active || (active = pool.tryIncrementActiveCount()))
1026	–	return locallyFifo? locallyDeqTask() : popTask();
1027	–	}
1028	–	return null;
1029	–	}
1030	–
1031	–	/**
1032	–	* Gets and removes a local or stolen task.
1033	–	*
1034	–	* @return a task, if available
1035	–	*/
1036	–	final ForkJoinTask<?> pollTask() {
1037	–	ForkJoinTask<?> t;
1038	–	return (t = pollLocalTask()) != null ? t : scan();
1039	–	}
1040	–
1041	–	/**
1112		* Runs tasks until {@code pool.isQuiescent()}.
1113		*/
1114		final void helpQuiescePool() {
1115	+	ForkJoinTask<?> ps = currentSteal; // to restore below
1116		for (;;) {
1117		ForkJoinTask<?> t = pollLocalTask();
1118	<	if (t != null || (t = scan()) != null) {
1119	<	t.tryExec();
1049	<	stolen = null;
1050	<	}
1118	>	if (t != null || (t = scan()) != null)
1119	>	t.quietlyExec();
1120		else {
1121		ForkJoinPool p = pool;
1122	+	int a; // to inline CASes
1123		if (active) {
1124	+	if (!UNSAFE.compareAndSwapInt
1125	+	(p, poolRunStateOffset, a = p.runState, a - 1))
1126	+	continue;   // retry later
1127		active = false; // inactivate
1128	<	do {} while (!p.tryDecrementActiveCount());
1128	>	UNSAFE.putOrderedObject(this, currentStealOffset, ps);
1129		}
1130		if (p.isQuiescent()) {
1131		active = true; // re-activate
1132	<	do {} while (!p.tryIncrementActiveCount());
1132	>	do {} while (!UNSAFE.compareAndSwapInt
1133	>	(p, poolRunStateOffset, a = p.runState, a+1));
1134		return;
1135		}
1136		}
#	Line 1066 | Line 1140 | public class ForkJoinWorkerThread extend
1140		// Unsafe mechanics
1141
1142		private static final sun.misc.Unsafe UNSAFE = getUnsafe();
1143	+	private static final long spOffset =
1144	+	objectFieldOffset("sp", ForkJoinWorkerThread.class);
1145		private static final long runStateOffset =
1146		objectFieldOffset("runState", ForkJoinWorkerThread.class);
1147	+	private static final long currentJoinOffset =
1148	+	objectFieldOffset("currentJoin", ForkJoinWorkerThread.class);
1149	+	private static final long currentStealOffset =
1150	+	objectFieldOffset("currentSteal", ForkJoinWorkerThread.class);
1151		private static final long qBase =
1152		UNSAFE.arrayBaseOffset(ForkJoinTask[].class);
1153	+	private static final long poolRunStateOffset = // to inline CAS
1154	+	objectFieldOffset("runState", ForkJoinPool.class);
1155	+
1156		private static final int qShift;
1157
1158		static {
#	Line 1077 | Line 1160 | public class ForkJoinWorkerThread extend
1160		if ((s & (s-1)) != 0)
1161		throw new Error("data type scale not a power of two");
1162		qShift = 31 - Integer.numberOfLeadingZeros(s);
1163	+	MAXIMUM_QUEUE_CAPACITY = 1 << (31 - qShift);
1164		}
1165
1166		private static long objectFieldOffset(String field, Class<?> klazz) {

Diff Legend

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