ViewVC Help

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

root/jsr166/jsr166/src/jsr166y/Phaser.java

Comparing jsr166/src/jsr166y/Phaser.java (file contents):
Revision 1.9 by jsr166, Mon Jan 5 09:11:26 2009 UTC vs.
Revision 1.44 by dl, Tue Aug 25 16:32:28 2009 UTC

#	Line 7 | Line 7
7		package jsr166y;
8
9		import java.util.concurrent.*;
10	<	import java.util.concurrent.atomic.*;
10	>
11	>	import java.util.concurrent.atomic.AtomicReference;
12		import java.util.concurrent.locks.LockSupport;
12	–	import sun.misc.Unsafe;
13	–	import java.lang.reflect.*;
13
14		/**
15	<	* A reusable synchronization barrier, similar in functionality to a
15	>	* A reusable synchronization barrier, similar in functionality to
16		* {@link java.util.concurrent.CyclicBarrier CyclicBarrier} and
17		* {@link java.util.concurrent.CountDownLatch CountDownLatch}
18		* but supporting more flexible usage.
19		*
20	<	* <ul>
21	<	*
22	<	* <li> The number of parties synchronizing on a phaser may vary over
23	<	* time.  A task may register to be a party at any time, and may
24	<	* deregister upon arriving at the barrier.  As is the case with most
25	<	* basic synchronization constructs, registration and deregistration
26	<	* affect only internal counts; they do not establish any further
27	<	* internal bookkeeping, so tasks cannot query whether they are
28	<	* registered. (However, you can introduce such bookkeeping by
29	<	* subclassing this class.)
30	<	*
31	<	* <li> Each generation has an associated phase value, starting at
32	<	* zero, and advancing when all parties reach the barrier (wrapping
33	<	* around to zero after reaching {@code Integer.MAX_VALUE}).
34	<	*
35	<	* <li> Like a CyclicBarrier, a Phaser may be repeatedly awaited.
36	<	* Method {@code arriveAndAwaitAdvance} has effect analogous to
37	<	* {@code CyclicBarrier.await}.  However, Phasers separate two
38	<	* aspects of coordination, that may also be invoked independently:
20	>	* <p> <b>Registration.</b> Unlike the case for other barriers, the
21	>	* number of parties <em>registered</em> to synchronize on a phaser
22	>	* may vary over time.  Tasks may be registered at any time (using
23	>	* methods {@link #register}, {@link #bulkRegister}, or forms of
24	>	* constructors establishing initial numbers of parties), and
25	>	* optionally deregistered upon any arrival (using {@link
26	>	* #arriveAndDeregister}).  As is the case with most basic
27	>	* synchronization constructs, registration and deregistration affect
28	>	* only internal counts; they do not establish any further internal
29	>	* bookkeeping, so tasks cannot query whether they are registered.
30	>	* (However, you can introduce such bookkeeping by subclassing this
31	>	* class.)
32	>	*
33	>	* <p> <b>Synchronization.</b> Like a {@code CyclicBarrier}, a {@code
34	>	* Phaser} may be repeatedly awaited.  Method {@link
35	>	* #arriveAndAwaitAdvance} has effect analogous to {@link
36	>	* java.util.concurrent.CyclicBarrier#await CyclicBarrier.await}. Each
37	>	* generation of a {@code Phaser} has an associated phase number. The
38	>	* phase number starts at zero, and advances when all parties arrive
39	>	* at the barrier, wrapping around to zero after reaching {@code
40	>	* Integer.MAX_VALUE}. The use of phase numbers enables independent
41	>	* control of actions upon arrival at a barrier and upon awaiting
42	>	* others, via two kinds of methods that may be invoked by any
43	>	* registered party:
44		*
45		* <ul>
46		*
47	<	*   <li> Arriving at a barrier. Methods {@code arrive} and
48	<	*       {@code arriveAndDeregister} do not block, but return
49	<	*       the phase value current upon entry to the method.
50	<	*
51	<	*   <li> Awaiting others. Method {@code awaitAdvance} requires an
52	<	*       argument indicating the entry phase, and returns when the
53	<	*       barrier advances to a new phase.
54	<	* </ul>
47	>	*   <li> <b>Arrival.</b> Methods {@link #arrive} and
48	>	*       {@link #arriveAndDeregister} record arrival at a
49	>	*       barrier. These methods do not block, but return an associated
50	>	*       <em>arrival phase number</em>; that is, the phase number of
51	>	*       the barrier to which the arrival applied. When the final
52	>	*       party for a given phase arrives, an optional barrier action
53	>	*       is performed and the phase advances.  Barrier actions,
54	>	*       performed by the party triggering a phase advance, are
55	>	*       arranged by overriding method {@link #onAdvance(int, int)},
56	>	*       which also controls termination. Overriding this method is
57	>	*       similar to, but more flexible than, providing a barrier
58	>	*       action to a {@code CyclicBarrier}.
59	>	*
60	>	*   <li> <b>Waiting.</b> Method {@link #awaitAdvance} requires an
61	>	*       argument indicating an arrival phase number, and returns when
62	>	*       the barrier advances to (or is already at) a different phase.
63	>	*       Unlike similar constructions using {@code CyclicBarrier},
64	>	*       method {@code awaitAdvance} continues to wait even if the
65	>	*       waiting thread is interrupted. Interruptible and timeout
66	>	*       versions are also available, but exceptions encountered while
67	>	*       tasks wait interruptibly or with timeout do not change the
68	>	*       state of the barrier. If necessary, you can perform any
69	>	*       associated recovery within handlers of those exceptions,
70	>	*       often after invoking {@code forceTermination}.  Phasers may
71	>	*       also be used by tasks executing in a {@link ForkJoinPool},
72	>	*       which will ensure sufficient parallelism to execute tasks
73	>	*       when others are blocked waiting for a phase to advance.
74		*
75	+	* </ul>
76		*
77	<	* <li> Barrier actions, performed by the task triggering a phase
78	<	* advance while others may be waiting, are arranged by overriding
79	<	* method {@code onAdvance}, that also controls termination.
80	<	* Overriding this method may be used to similar but more flexible
81	<	* effect as providing a barrier action to a CyclicBarrier.
82	<	*
83	<	* <li> Phasers may enter a <em>termination</em> state in which all
60	<	* await actions immediately return, indicating (via a negative phase
61	<	* value) that execution is complete.  Termination is triggered by
62	<	* executing the overridable {@code onAdvance} method that is invoked
63	<	* each time the barrier is about to be tripped. When a Phaser is
64	<	* controlling an action with a fixed number of iterations, it is
77	>	* <p> <b>Termination.</b> A {@code Phaser} may enter a
78	>	* <em>termination</em> state in which all synchronization methods
79	>	* immediately return without updating phaser state or waiting for
80	>	* advance, and indicating (via a negative phase value) that execution
81	>	* is complete.  Termination is triggered when an invocation of {@code
82	>	* onAdvance} returns {@code true}.  As illustrated below, when
83	>	* phasers control actions with a fixed number of iterations, it is
84		* often convenient to override this method to cause termination when
85	<	* the current phase number reaches a threshold. Method
86	<	* {@code forceTermination} is also available to abruptly release
87	<	* waiting threads and allow them to terminate.
85	>	* the current phase number reaches a threshold. Method {@link
86	>	* #forceTermination} is also available to abruptly release waiting
87	>	* threads and allow them to terminate.
88		*
89	<	* <li> Phasers may be tiered to reduce contention. Phasers with large
89	>	* <p> <b>Tiering.</b> Phasers may be <em>tiered</em> (i.e., arranged
90	>	* in tree structures) to reduce contention. Phasers with large
91		* numbers of parties that would otherwise experience heavy
92	<	* synchronization contention costs may instead be arranged in trees.
93	<	* This will typically greatly increase throughput even though it
94	<	* incurs somewhat greater per-operation overhead.
95	<	*
96	<	* <li> By default, {@code awaitAdvance} continues to wait even if
97	<	* the waiting thread is interrupted. And unlike the case in
98	<	* CyclicBarriers, exceptions encountered while tasks wait
99	<	* interruptibly or with timeout do not change the state of the
100	<	* barrier. If necessary, you can perform any associated recovery
101	<	* within handlers of those exceptions, often after invoking
102	<	* {@code forceTermination}.
103	<	*
104	<	* </ul>
92	>	* synchronization contention costs may instead be set up so that
93	>	* groups of sub-phasers share a common parent.  This may greatly
94	>	* increase throughput even though it incurs greater per-operation
95	>	* overhead.
96	>	*
97	>	* <p><b>Monitoring.</b> While synchronization methods may be invoked
98	>	* only by registered parties, the current state of a phaser may be
99	>	* monitored by any caller.  At any given moment there are {@link
100	>	* #getRegisteredParties} parties in total, of which {@link
101	>	* #getArrivedParties} have arrived at the current phase ({@link
102	>	* #getPhase}).  When the remaining ({@link #getUnarrivedParties})
103	>	* parties arrive, the phase advances.  The values returned by these
104	>	* methods may reflect transient states and so are not in general
105	>	* useful for synchronization control.  Method {@link #toString}
106	>	* returns snapshots of these state queries in a form convenient for
107	>	* informal monitoring.
108		*
109		* <p><b>Sample usages:</b>
110		*
111	<	* <p>A Phaser may be used instead of a {@code CountDownLatch} to control
112	<	* a one-shot action serving a variable number of parties. The typical
113	<	* idiom is for the method setting this up to first register, then
114	<	* start the actions, then deregister, as in:
115	<	*
116	<	* <pre>
117	<	*  void runTasks(List&lt;Runnable&gt; list) {
118	<	*    final Phaser phaser = new Phaser(1); // "1" to register self
119	<	*    for (Runnable r : list) {
120	<	*      phaser.register();
121	<	*      new Thread() {
122	<	*        public void run() {
123	<	*          phaser.arriveAndAwaitAdvance(); // await all creation
124	<	*          r.run();
125	<	*          phaser.arriveAndDeregister();   // signal completion
126	<	*        }
127	<	*      }.start();
111	>	* <p>A {@code Phaser} may be used instead of a {@code CountDownLatch}
112	>	* to control a one-shot action serving a variable number of
113	>	* parties. The typical idiom is for the method setting this up to
114	>	* first register, then start the actions, then deregister, as in:
115	>	*
116	>	*  <pre> {@code
117	>	* void runTasks(List<Runnable> tasks) {
118	>	*   final Phaser phaser = new Phaser(1); // "1" to register self
119	>	*   // create and start threads
120	>	*   for (Runnable task : tasks) {
121	>	*     phaser.register();
122	>	*     new Thread() {
123	>	*       public void run() {
124	>	*         phaser.arriveAndAwaitAdvance(); // await all creation
125	>	*         task.run();
126	>	*       }
127	>	*     }.start();
128		*   }
129		*
130	<	*   doSomethingOnBehalfOfWorkers();
131	<	*   phaser.arrive(); // allow threads to start
132	<	*   int p = phaser.arriveAndDeregister(); // deregister self  ...
110	<	*   p = phaser.awaitAdvance(p); // ... and await arrival
111	<	*   otherActions(); // do other things while tasks execute
112	<	*   phaser.awaitAdvance(p); // await final completion
113	<	* }
114	<	* </pre>
130	>	*   // allow threads to start and deregister self
131	>	*   phaser.arriveAndDeregister();
132	>	* }}</pre>
133		*
134		* <p>One way to cause a set of threads to repeatedly perform actions
135		* for a given number of iterations is to override {@code onAdvance}:
136		*
137	<	* <pre>
138	<	*  void startTasks(List&lt;Runnable&gt; list, final int iterations) {
139	<	*    final Phaser phaser = new Phaser() {
140	<	*       public boolean onAdvance(int phase, int registeredParties) {
141	<	*         return phase &gt;= iterations || registeredParties == 0;
137	>	*  <pre> {@code
138	>	* void startTasks(List<Runnable> tasks, final int iterations) {
139	>	*   final Phaser phaser = new Phaser() {
140	>	*     protected boolean onAdvance(int phase, int registeredParties) {
141	>	*       return phase >= iterations || registeredParties == 0;
142	>	*     }
143	>	*   };
144	>	*   phaser.register();
145	>	*   for (Runnable task : tasks) {
146	>	*     phaser.register();
147	>	*     new Thread() {
148	>	*       public void run() {
149	>	*         do {
150	>	*           task.run();
151	>	*           phaser.arriveAndAwaitAdvance();
152	>	*         } while(!phaser.isTerminated();
153		*       }
154	<	*    };
126	<	*    phaser.register();
127	<	*    for (Runnable r : list) {
128	<	*      phaser.register();
129	<	*      new Thread() {
130	<	*        public void run() {
131	<	*           do {
132	<	*             r.run();
133	<	*             phaser.arriveAndAwaitAdvance();
134	<	*           } while(!phaser.isTerminated();
135	<	*        }
136	<	*      }.start();
154	>	*     }.start();
155		*   }
156		*   phaser.arriveAndDeregister(); // deregister self, don't wait
157	<	* }
158	<	* </pre>
157	>	* }}</pre>
158	>	*
159	>	* If the main task must later await termination, it
160	>	* may re-register and then execute a similar loop:
161	>	* <pre> {@code
162	>	*   // ...
163	>	*   phaser.register();
164	>	*   while (!phaser.isTerminated())
165	>	*     phaser.arriveAndAwaitAdvance();
166	>	* }</pre>
167	>	*
168	>	* Related constructions may be used to await particular phase numbers
169	>	* in contexts where you are sure that the phase will never wrap around
170	>	* {@code Integer.MAX_VALUE}. For example:
171	>	*
172	>	* <pre> {@code
173	>	*   void awaitPhase(Phaser phaser, int phase) {
174	>	*     int p = phaser.register(); // assumes caller not already registered
175	>	*     while (p < phase) {
176	>	*       if (phaser.isTerminated())
177	>	*         // ... deal with unexpected termination
178	>	*       else
179	>	*         p = phaser.arriveAndAwaitAdvance();
180	>	*     }
181	>	*     phaser.arriveAndDeregister();
182	>	*   }
183	>	* }</pre>
184	>	*
185		*
186	<	* <p> To create a set of tasks using a tree of Phasers,
186	>	* <p>To create a set of tasks using a tree of phasers,
187		* you could use code of the following form, assuming a
188	<	* Task class with a constructor accepting a Phaser that
188	>	* Task class with a constructor accepting a phaser that
189		* it registers for upon construction:
190	<	* <pre>
191	<	*  void build(Task[] actions, int lo, int hi, Phaser b) {
192	<	*    int step = (hi - lo) / TASKS_PER_PHASER;
193	<	*    if (step &gt; 1) {
194	<	*       int i = lo;
195	<	*       while (i &lt; hi) {
196	<	*         int r = Math.min(i + step, hi);
197	<	*         build(actions, i, r, new Phaser(b));
198	<	*         i = r;
199	<	*       }
200	<	*    }
201	<	*    else {
202	<	*      for (int i = lo; i &lt; hi; ++i)
203	<	*        actions[i] = new Task(b);
204	<	*        // assumes new Task(b) performs b.register()
161	<	*    }
162	<	*  }
163	<	*  // .. initially called, for n tasks via
164	<	*  build(new Task[n], 0, n, new Phaser());
165	<	* </pre>
190	>	*  <pre> {@code
191	>	* void build(Task[] actions, int lo, int hi, Phaser ph) {
192	>	*   if (hi - lo > TASKS_PER_PHASER) {
193	>	*     for (int i = lo; i < hi; i += TASKS_PER_PHASER) {
194	>	*       int j = Math.min(i + TASKS_PER_PHASER, hi);
195	>	*       build(actions, i, j, new Phaser(ph));
196	>	*     }
197	>	*   } else {
198	>	*     for (int i = lo; i < hi; ++i)
199	>	*       actions[i] = new Task(ph);
200	>	*       // assumes new Task(ph) performs ph.register()
201	>	*   }
202	>	* }
203	>	* // .. initially called, for n tasks via
204	>	* build(new Task[n], 0, n, new Phaser());}</pre>
205		*
206		* The best value of {@code TASKS_PER_PHASER} depends mainly on
207		* expected barrier synchronization rates. A value as low as four may
#	Line 173 | Line 212 | import java.lang.reflect.*;
212		*
213		* <p><b>Implementation notes</b>: This implementation restricts the
214		* maximum number of parties to 65535. Attempts to register additional
215	<	* parties result in IllegalStateExceptions. However, you can and
215	>	* parties result in {@code IllegalStateException}. However, you can and
216		* should create tiered phasers to accommodate arbitrarily large sets
217		* of participants.
218	+	*
219	+	* @since 1.7
220	+	* @author Doug Lea
221		*/
222		public class Phaser {
223		/*
#	Line 200 | Line 242 | public class Phaser {
242		* and encoding simple, and keeping race windows short.
243		*
244		* Note: there are some cheats in arrive() that rely on unarrived
245	<	* being lowest 16 bits.
245	>	* count being lowest 16 bits.
246		*/
247		private volatile long state;
248
249	<	private static final int ushortBits = 16;
250	<	private static final int ushortMask =  (1 << ushortBits) - 1;
209	<	private static final int phaseMask = 0x7fffffff;
249	>	private static final int ushortMask = 0xffff;
250	>	private static final int phaseMask  = 0x7fffffff;
251
252		private static int unarrivedOf(long s) {
253	<	return (int)(s & ushortMask);
253	>	return (int) (s & ushortMask);
254		}
255
256		private static int partiesOf(long s) {
257	<	return (int)(s & (ushortMask << 16)) >>> 16;
257	>	return ((int) s) >>> 16;
258		}
259
260		private static int phaseOf(long s) {
261	<	return (int)(s >>> 32);
261	>	return (int) (s >>> 32);
262		}
263
264		private static int arrivedOf(long s) {
#	Line 225 | Line 266 | public class Phaser {
266		}
267
268		private static long stateFor(int phase, int parties, int unarrived) {
269	<	return (((long)phase) << 32) | ((parties << 16) | unarrived);
269	>	return ((((long) phase) << 32) | (((long) parties) << 16) |
270	>	(long) unarrived);
271		}
272
273		private static long trippedStateFor(int phase, int parties) {
274	<	return (((long)phase) << 32) | ((parties << 16) | parties);
274	>	long lp = (long) parties;
275	>	return (((long) phase) << 32) | (lp << 16) | lp;
276		}
277
278	<	private static IllegalStateException badBounds(int parties, int unarrived) {
279	<	return new IllegalStateException
280	<	("Attempt to set " + unarrived +
281	<	" unarrived of " + parties + " parties");
278	>	/**
279	>	* Returns message string for bad bounds exceptions.
280	>	*/
281	>	private static String badBounds(int parties, int unarrived) {
282	>	return ("Attempt to set " + unarrived +
283	>	" unarrived of " + parties + " parties");
284		}
285
286		/**
#	Line 244 | Line 289 | public class Phaser {
289		private final Phaser parent;
290
291		/**
292	<	* The root of Phaser tree. Equals this if not in a tree.  Used to
292	>	* The root of phaser tree. Equals this if not in a tree.  Used to
293		* support faster state push-down.
294		*/
295		private final Phaser root;
#	Line 252 | Line 297 | public class Phaser {
297		// Wait queues
298
299		/**
300	<	* Heads of Treiber stacks waiting for nonFJ threads. To eliminate
300	>	* Heads of Treiber stacks for waiting threads. To eliminate
301		* contention while releasing some threads while adding others, we
302		* use two of them, alternating across even and odd phases.
303		*/
#	Line 260 | Line 305 | public class Phaser {
305		private final AtomicReference<QNode> oddQ  = new AtomicReference<QNode>();
306
307		private AtomicReference<QNode> queueFor(int phase) {
308	<	return (phase & 1) == 0? evenQ : oddQ;
308	>	return ((phase & 1) == 0) ? evenQ : oddQ;
309		}
310
311		/**
#	Line 268 | Line 313 | public class Phaser {
313		* root if necessary.
314		*/
315		private long getReconciledState() {
316	<	return parent == null? state : reconcileState();
316	>	return (parent == null) ? state : reconcileState();
317		}
318
319		/**
#	Line 295 | Line 340 | public class Phaser {
340		}
341
342		/**
343	<	* Creates a new Phaser without any initially registered parties,
344	<	* initial phase number 0, and no parent.
343	>	* Creates a new phaser without any initially registered parties,
344	>	* initial phase number 0, and no parent. Any thread using this
345	>	* phaser will need to first register for it.
346		*/
347		public Phaser() {
348		this(null);
349		}
350
351		/**
352	<	* Creates a new Phaser with the given numbers of registered
352	>	* Creates a new phaser with the given numbers of registered
353		* unarrived parties, initial phase number 0, and no parent.
354	<	* @param parties the number of parties required to trip barrier.
354	>	*
355	>	* @param parties the number of parties required to trip barrier
356		* @throws IllegalArgumentException if parties less than zero
357	<	* or greater than the maximum number of parties supported.
357	>	* or greater than the maximum number of parties supported
358		*/
359		public Phaser(int parties) {
360		this(null, parties);
361		}
362
363		/**
364	<	* Creates a new Phaser with the given parent, without any
364	>	* Creates a new phaser with the given parent, without any
365		* initially registered parties. If parent is non-null this phaser
366		* is registered with the parent and its initial phase number is
367		* the same as that of parent phaser.
368	<	* @param parent the parent phaser.
368	>	*
369	>	* @param parent the parent phaser
370		*/
371		public Phaser(Phaser parent) {
372		int phase = 0;
#	Line 333 | Line 381 | public class Phaser {
381		}
382
383		/**
384	<	* Creates a new Phaser with the given parent and numbers of
385	<	* registered unarrived parties. If parent is non-null this phaser
384	>	* Creates a new phaser with the given parent and numbers of
385	>	* registered unarrived parties. If parent is non-null, this phaser
386		* is registered with the parent and its initial phase number is
387		* the same as that of parent phaser.
388	<	* @param parent the parent phaser.
389	<	* @param parties the number of parties required to trip barrier.
388	>	*
389	>	* @param parent the parent phaser
390	>	* @param parties the number of parties required to trip barrier
391		* @throws IllegalArgumentException if parties less than zero
392	<	* or greater than the maximum number of parties supported.
392	>	* or greater than the maximum number of parties supported
393		*/
394		public Phaser(Phaser parent, int parties) {
395		if (parties < 0 || parties > ushortMask)
#	Line 358 | Line 407 | public class Phaser {
407
408		/**
409		* Adds a new unarrived party to this phaser.
410	<	* @return the current barrier phase number upon registration
410	>	*
411	>	* @return the arrival phase number to which this registration applied
412		* @throws IllegalStateException if attempting to register more
413	<	* than the maximum supported number of parties.
413	>	* than the maximum supported number of parties
414		*/
415		public int register() {
416		return doRegister(1);
#	Line 368 | Line 418 | public class Phaser {
418
419		/**
420		* Adds the given number of new unarrived parties to this phaser.
421	<	* @param parties the number of parties required to trip barrier.
422	<	* @return the current barrier phase number upon registration
421	>	*
422	>	* @param parties the number of parties required to trip barrier
423	>	* @return the arrival phase number to which this registration applied
424		* @throws IllegalStateException if attempting to register more
425	<	* than the maximum supported number of parties.
425	>	* than the maximum supported number of parties
426		*/
427		public int bulkRegister(int parties) {
428		if (parties < 0)
#	Line 394 | Line 445 | public class Phaser {
445		if (phase < 0)
446		break;
447		if (parties > ushortMask || unarrived > ushortMask)
448	<	throw badBounds(parties, unarrived);
448	>	throw new IllegalStateException(badBounds(parties, unarrived));
449		if (phase == phaseOf(root.state) &&
450		casState(s, stateFor(phase, parties, unarrived)))
451		break;
#	Line 404 | Line 455 | public class Phaser {
455
456		/**
457		* Arrives at the barrier, but does not wait for others.  (You can
458	<	* in turn wait for others via {@link #awaitAdvance}).
458	>	* in turn wait for others via {@link #awaitAdvance}).  It is an
459	>	* unenforced usage error for an unregistered party to invoke this
460	>	* method.
461		*
462	<	* @return the barrier phase number upon entry to this method, or a
410	<	* negative value if terminated;
462	>	* @return the arrival phase number, or a negative value if terminated
463		* @throws IllegalStateException if not terminated and the number
464	<	* of unarrived parties would become negative.
464	>	* of unarrived parties would become negative
465		*/
466		public int arrive() {
467		int phase;
468		for (;;) {
469		long s = state;
470		phase = phaseOf(s);
471	+	if (phase < 0)
472	+	break;
473		int parties = partiesOf(s);
474		int unarrived = unarrivedOf(s) - 1;
475		if (unarrived > 0) {        // Not the last arrival
#	Line 427 | Line 481 | public class Phaser {
481		if (par == null) {      // directly trip
482		if (casState
483		(s,
484	<	trippedStateFor(onAdvance(phase, parties)? -1 :
484	>	trippedStateFor(onAdvance(phase, parties) ? -1 :
485		((phase + 1) & phaseMask), parties))) {
486		releaseWaiters(phase);
487		break;
#	Line 441 | Line 495 | public class Phaser {
495		}
496		}
497		}
444	–	else if (phase < 0) // Don't throw exception if terminated
445	–	break;
498		else if (phase != phaseOf(root.state)) // or if unreconciled
499		reconcileState();
500		else
501	<	throw badBounds(parties, unarrived);
501	>	throw new IllegalStateException(badBounds(parties, unarrived));
502		}
503		return phase;
504		}
505
506		/**
507	<	* Arrives at the barrier, and deregisters from it, without
508	<	* waiting for others. Deregistration reduces number of parties
507	>	* Arrives at the barrier and deregisters from it without waiting
508	>	* for others. Deregistration reduces the number of parties
509		* required to trip the barrier in future phases.  If this phaser
510		* has a parent, and deregistration causes this phaser to have
511	<	* zero parties, this phaser is also deregistered from its parent.
511	>	* zero parties, this phaser also arrives at and is deregistered
512	>	* from its parent.  It is an unenforced usage error for an
513	>	* unregistered party to invoke this method.
514		*
515	<	* @return the current barrier phase number upon entry to
462	<	* this method, or a negative value if terminated;
515	>	* @return the arrival phase number, or a negative value if terminated
516		* @throws IllegalStateException if not terminated and the number
517	<	* of registered or unarrived parties would become negative.
517	>	* of registered or unarrived parties would become negative
518		*/
519		public int arriveAndDeregister() {
520		// similar code to arrive, but too different to merge
#	Line 470 | Line 523 | public class Phaser {
523		for (;;) {
524		long s = state;
525		phase = phaseOf(s);
526	+	if (phase < 0)
527	+	break;
528		int parties = partiesOf(s) - 1;
529		int unarrived = unarrivedOf(s) - 1;
530		if (parties >= 0) {
#	Line 488 | Line 543 | public class Phaser {
543		if (unarrived == 0) {
544		if (casState
545		(s,
546	<	trippedStateFor(onAdvance(phase, parties)? -1 :
546	>	trippedStateFor(onAdvance(phase, parties) ? -1 :
547		((phase + 1) & phaseMask), parties))) {
548		releaseWaiters(phase);
549		break;
550		}
551		continue;
552		}
498	–	if (phase < 0)
499	–	break;
553		if (par != null && phase != phaseOf(root.state)) {
554		reconcileState();
555		continue;
556		}
557		}
558	<	throw badBounds(parties, unarrived);
558	>	throw new IllegalStateException(badBounds(parties, unarrived));
559		}
560		return phase;
561		}
562
563		/**
564		* Arrives at the barrier and awaits others. Equivalent in effect
565	<	* to {@code awaitAdvance(arrive())}.  If you instead need to
566	<	* await with interruption of timeout, and/or deregister upon
567	<	* arrival, you can arrange them using analogous constructions.
568	<	* @return the phase on entry to this method
565	>	* to {@code awaitAdvance(arrive())}.  If you need to await with
566	>	* interruption or timeout, you can arrange this with an analogous
567	>	* construction using one of the other forms of the awaitAdvance
568	>	* method.  If instead you need to deregister upon arrival use
569	>	* {@code arriveAndDeregister}. It is an unenforced usage error
570	>	* for an unregistered party to invoke this method.
571	>	*
572	>	* @return the arrival phase number, or a negative number if terminated
573		* @throws IllegalStateException if not terminated and the number
574	<	* of unarrived parties would become negative.
574	>	* of unarrived parties would become negative
575		*/
576		public int arriveAndAwaitAdvance() {
577		return awaitAdvance(arrive());
578		}
579
580		/**
581	<	* Awaits the phase of the barrier to advance from the given
582	<	* value, or returns immediately if argument is negative or this
583	<	* barrier is terminated.
584	<	* @param phase the phase on entry to this method
585	<	* @return the phase on exit from this method
581	>	* Awaits the phase of the barrier to advance from the given phase
582	>	* value, returning immediately if the current phase of the
583	>	* barrier is not equal to the given phase value or this barrier
584	>	* is terminated.  It is an unenforced usage error for an
585	>	* unregistered party to invoke this method.
586	>	*
587	>	* @param phase an arrival phase number, or negative value if
588	>	* terminated; this argument is normally the value returned by a
589	>	* previous call to {@code arrive} or its variants
590	>	* @return the next arrival phase number, or a negative value
591	>	* if terminated or argument is negative
592		*/
593		public int awaitAdvance(int phase) {
594		if (phase < 0)
#	Line 534 | Line 597 | public class Phaser {
597		int p = phaseOf(s);
598		if (p != phase)
599		return p;
600	<	if (unarrivedOf(s) == 0)
600	>	if (unarrivedOf(s) == 0 && parent != null)
601		parent.awaitAdvance(phase);
602		// Fall here even if parent waited, to reconcile and help release
603		return untimedWait(phase);
604		}
605
606		/**
607	<	* Awaits the phase of the barrier to advance from the given
608	<	* value, or returns immediately if argument is negative or this
609	<	* barrier is terminated, or throws InterruptedException if
610	<	* interrupted while waiting.
611	<	* @param phase the phase on entry to this method
612	<	* @return the phase on exit from this method
607	>	* Awaits the phase of the barrier to advance from the given phase
608	>	* value, throwing {@code InterruptedException} if interrupted
609	>	* while waiting, or returning immediately if the current phase of
610	>	* the barrier is not equal to the given phase value or this
611	>	* barrier is terminated. It is an unenforced usage error for an
612	>	* unregistered party to invoke this method.
613	>	*
614	>	* @param phase an arrival phase number, or negative value if
615	>	* terminated; this argument is normally the value returned by a
616	>	* previous call to {@code arrive} or its variants
617	>	* @return the next arrival phase number, or a negative value
618	>	* if terminated or argument is negative
619		* @throws InterruptedException if thread interrupted while waiting
620		*/
621	<	public int awaitAdvanceInterruptibly(int phase) throws InterruptedException {
621	>	public int awaitAdvanceInterruptibly(int phase)
622	>	throws InterruptedException {
623		if (phase < 0)
624		return phase;
625		long s = getReconciledState();
626		int p = phaseOf(s);
627		if (p != phase)
628		return p;
629	<	if (unarrivedOf(s) != 0)
629	>	if (unarrivedOf(s) == 0 && parent != null)
630		parent.awaitAdvanceInterruptibly(phase);
631		return interruptibleWait(phase);
632		}
633
634		/**
635	<	* Awaits the phase of the barrier to advance from the given value
636	<	* or the given timeout elapses, or returns immediately if
637	<	* argument is negative or this barrier is terminated.
638	<	* @param phase the phase on entry to this method
639	<	* @return the phase on exit from this method
635	>	* Awaits the phase of the barrier to advance from the given phase
636	>	* value or the given timeout to elapse, throwing {@code
637	>	* InterruptedException} if interrupted while waiting, or
638	>	* returning immediately if the current phase of the barrier is
639	>	* not equal to the given phase value or this barrier is
640	>	* terminated.  It is an unenforced usage error for an
641	>	* unregistered party to invoke this method.
642	>	*
643	>	* @param phase an arrival phase number, or negative value if
644	>	* terminated; this argument is normally the value returned by a
645	>	* previous call to {@code arrive} or its variants
646	>	* @param timeout how long to wait before giving up, in units of
647	>	*        {@code unit}
648	>	* @param unit a {@code TimeUnit} determining how to interpret the
649	>	*        {@code timeout} parameter
650	>	* @return the next arrival phase number, or a negative value
651	>	* if terminated or argument is negative
652		* @throws InterruptedException if thread interrupted while waiting
653		* @throws TimeoutException if timed out while waiting
654		*/
655	<	public int awaitAdvanceInterruptibly(int phase, long timeout, TimeUnit unit)
655	>	public int awaitAdvanceInterruptibly(int phase,
656	>	long timeout, TimeUnit unit)
657		throws InterruptedException, TimeoutException {
658		if (phase < 0)
659		return phase;
#	Line 578 | Line 661 | public class Phaser {
661		int p = phaseOf(s);
662		if (p != phase)
663		return p;
664	<	if (unarrivedOf(s) == 0)
664	>	if (unarrivedOf(s) == 0 && parent != null)
665		parent.awaitAdvanceInterruptibly(phase, timeout, unit);
666		return timedWait(phase, unit.toNanos(timeout));
667		}
#	Line 611 | Line 694 | public class Phaser {
694		* Returns the current phase number. The maximum phase number is
695		* {@code Integer.MAX_VALUE}, after which it restarts at
696		* zero. Upon termination, the phase number is negative.
697	+	*
698		* @return the phase number, or a negative value if terminated
699		*/
700		public final int getPhase() {
#	Line 618 | Line 702 | public class Phaser {
702		}
703
704		/**
621	–	* Returns {@code true} if the current phase number equals the given phase.
622	–	* @param phase the phase
623	–	* @return {@code true} if the current phase number equals the given phase
624	–	*/
625	–	public final boolean hasPhase(int phase) {
626	–	return phaseOf(getReconciledState()) == phase;
627	–	}
628	–
629	–	/**
705		* Returns the number of parties registered at this barrier.
706	+	*
707		* @return the number of parties
708		*/
709		public int getRegisteredParties() {
#	Line 635 | Line 711 | public class Phaser {
711		}
712
713		/**
714	<	* Returns the number of parties that have arrived at the current
715	<	* phase of this barrier.
714	>	* Returns the number of registered parties that have arrived at
715	>	* the current phase of this barrier.
716	>	*
717		* @return the number of arrived parties
718		*/
719		public int getArrivedParties() {
#	Line 646 | Line 723 | public class Phaser {
723		/**
724		* Returns the number of registered parties that have not yet
725		* arrived at the current phase of this barrier.
726	+	*
727		* @return the number of unarrived parties
728		*/
729		public int getUnarrivedParties() {
#	Line 653 | Line 731 | public class Phaser {
731		}
732
733		/**
734	<	* Returns the parent of this phaser, or null if none.
735	<	* @return the parent of this phaser, or null if none
734	>	* Returns the parent of this phaser, or {@code null} if none.
735	>	*
736	>	* @return the parent of this phaser, or {@code null} if none
737		*/
738		public Phaser getParent() {
739		return parent;
#	Line 663 | Line 742 | public class Phaser {
742		/**
743		* Returns the root ancestor of this phaser, which is the same as
744		* this phaser if it has no parent.
745	+	*
746		* @return the root ancestor of this phaser
747		*/
748		public Phaser getRoot() {
#	Line 671 | Line 751 | public class Phaser {
751
752		/**
753		* Returns {@code true} if this barrier has been terminated.
754	+	*
755		* @return {@code true} if this barrier has been terminated
756		*/
757		public boolean isTerminated() {
#	Line 678 | Line 759 | public class Phaser {
759		}
760
761		/**
762	<	* Overridable method to perform an action upon phase advance, and
763	<	* to control termination. This method is invoked whenever the
764	<	* barrier is tripped (and thus all other waiting parties are
765	<	* dormant). If it returns true, then, rather than advance the
766	<	* phase number, this barrier will be set to a final termination
767	<	* state, and subsequent calls to {@code isTerminated} will
768	<	* return true.
762	>	* Overridable method to perform an action upon impending phase
763	>	* advance, and to control termination. This method is invoked
764	>	* upon arrival of the party tripping the barrier (when all other
765	>	* waiting parties are dormant).  If this method returns {@code
766	>	* true}, then, rather than advance the phase number, this barrier
767	>	* will be set to a final termination state, and subsequent calls
768	>	* to {@link #isTerminated} will return true. Any (unchecked)
769	>	* Exception or Error thrown by an invocation of this method is
770	>	* propagated to the party attempting to trip the barrier, in
771	>	* which case no advance occurs.
772	>	*
773	>	* <p>The arguments to this method provide the state of the phaser
774	>	* prevailing for the current transition. (When called from within
775	>	* an implementation of {@code onAdvance} the values returned by
776	>	* methods such as {@code getPhase} may or may not reliably
777	>	* indicate the state to which this transition applies.)
778		*
779	<	* <p> The default version returns true when the number of
779	>	* <p>The default version returns {@code true} when the number of
780		* registered parties is zero. Normally, overrides that arrange
781		* termination for other reasons should also preserve this
782		* property.
783		*
784	<	* <p> You may override this method to perform an action with side
785	<	* effects visible to participating tasks, but it is in general
786	<	* only sensible to do so in designs where all parties register
787	<	* before any arrive, and all {@code awaitAdvance} at each phase.
788	<	* Otherwise, you cannot ensure lack of interference. In
789	<	* particular, this method may be invoked more than once per
790	<	* transition if other parties successfully register while the
701	<	* invocation of this method is in progress, thus postponing the
702	<	* transition until those parties also arrive, re-triggering this
703	<	* method.
784	>	* <p>You may override this method to perform an action with side
785	>	* effects visible to participating tasks, but doing so requires
786	>	* care: Method {@code onAdvance} may be invoked more than once
787	>	* per transition.  Further, unless all parties register before
788	>	* any arrive, and all {@link #awaitAdvance} at each phase, then
789	>	* you cannot ensure lack of interference from other parties
790	>	* during the invocation of this method.
791		*
792		* @param phase the phase number on entering the barrier
793		* @param registeredParties the current number of registered parties
#	Line 729 | Line 816 | public class Phaser {
816
817		// methods for waiting
818
732	–	/** The number of CPUs, for spin control */
733	–	static final int NCPUS = Runtime.getRuntime().availableProcessors();
734	–
735	–	/**
736	–	* The number of times to spin before blocking in timed waits.
737	–	* The value is empirically derived.
738	–	*/
739	–	static final int maxTimedSpins = (NCPUS < 2)? 0 : 32;
740	–
741	–	/**
742	–	* The number of times to spin before blocking in untimed waits.
743	–	* This is greater than timed value because untimed waits spin
744	–	* faster since they don't need to check times on each spin.
745	–	*/
746	–	static final int maxUntimedSpins = maxTimedSpins * 32;
747	–
748	–	/**
749	–	* The number of nanoseconds for which it is faster to spin
750	–	* rather than to use timed park. A rough estimate suffices.
751	–	*/
752	–	static final long spinForTimeoutThreshold = 1000L;
753	–
819		/**
820	<	* Wait nodes for Treiber stack representing wait queue for non-FJ
756	<	* tasks.
820	>	* Wait nodes for Treiber stack representing wait queue
821		*/
822	<	static final class QNode {
823	<	QNode next;
822	>	static final class QNode implements ForkJoinPool.ManagedBlocker {
823	>	final Phaser phaser;
824	>	final int phase;
825	>	final long startTime;
826	>	final long nanos;
827	>	final boolean timed;
828	>	final boolean interruptible;
829	>	volatile boolean wasInterrupted = false;
830		volatile Thread thread; // nulled to cancel wait
831	<	QNode() {
831	>	QNode next;
832	>	QNode(Phaser phaser, int phase, boolean interruptible,
833	>	boolean timed, long startTime, long nanos) {
834	>	this.phaser = phaser;
835	>	this.phase = phase;
836	>	this.timed = timed;
837	>	this.interruptible = interruptible;
838	>	this.startTime = startTime;
839	>	this.nanos = nanos;
840		thread = Thread.currentThread();
841		}
842	+	public boolean isReleasable() {
843	+	return (thread == null ||
844	+	phaser.getPhase() != phase ||
845	+	(interruptible && wasInterrupted) ||
846	+	(timed && (nanos - (System.nanoTime() - startTime)) <= 0));
847	+	}
848	+	public boolean block() {
849	+	if (Thread.interrupted()) {
850	+	wasInterrupted = true;
851	+	if (interruptible)
852	+	return true;
853	+	}
854	+	if (!timed)
855	+	LockSupport.park(this);
856	+	else {
857	+	long waitTime = nanos - (System.nanoTime() - startTime);
858	+	if (waitTime <= 0)
859	+	return true;
860	+	LockSupport.parkNanos(this, waitTime);
861	+	}
862	+	return isReleasable();
863	+	}
864		void signal() {
865		Thread t = thread;
866		if (t != null) {
#	Line 768 | Line 868 | public class Phaser {
868		LockSupport.unpark(t);
869		}
870		}
871	+	boolean doWait() {
872	+	if (thread != null) {
873	+	try {
874	+	ForkJoinPool.managedBlock(this, false);
875	+	} catch (InterruptedException ie) {
876	+	}
877	+	}
878	+	return wasInterrupted;
879	+	}
880	+
881		}
882
883		/**
884	<	* Removes and signals waiting threads from wait queue
884	>	* Removes and signals waiting threads from wait queue.
885		*/
886		private void releaseWaiters(int phase) {
887		AtomicReference<QNode> head = queueFor(phase);
#	Line 783 | Line 893 | public class Phaser {
893		}
894
895		/**
896	+	* Tries to enqueue given node in the appropriate wait queue.
897	+	*
898	+	* @return true if successful
899	+	*/
900	+	private boolean tryEnqueue(QNode node) {
901	+	AtomicReference<QNode> head = queueFor(node.phase);
902	+	return head.compareAndSet(node.next = head.get(), node);
903	+	}
904	+
905	+	/**
906		* Enqueues node and waits unless aborted or signalled.
907	+	*
908	+	* @return current phase
909		*/
910		private int untimedWait(int phase) {
789	–	int spins = maxUntimedSpins;
911		QNode node = null;
791	–	boolean interrupted = false;
912		boolean queued = false;
913	+	boolean interrupted = false;
914		int p;
915		while ((p = getPhase()) == phase) {
916	<	interrupted = Thread.interrupted();
917	<	if (node != null) {
918	<	if (!queued) {
919	<	AtomicReference<QNode> head = queueFor(phase);
920	<	queued = head.compareAndSet(node.next = head.get(), node);
921	<	}
801	<	else if (node.thread != null)
802	<	LockSupport.park(this);
803	<	}
804	<	else if (spins <= 0)
805	<	node = new QNode();
916	>	if (Thread.interrupted())
917	>	interrupted = true;
918	>	else if (node == null)
919	>	node = new QNode(this, phase, false, false, 0, 0);
920	>	else if (!queued)
921	>	queued = tryEnqueue(node);
922		else
923	<	--spins;
923	>	interrupted = node.doWait();
924		}
925		if (node != null)
926		node.thread = null;
927	+	releaseWaiters(phase);
928		if (interrupted)
929		Thread.currentThread().interrupt();
813	–	releaseWaiters(phase);
930		return p;
931		}
932
933		/**
934	<	* Messier interruptible version
934	>	* Interruptible version
935	>	* @return current phase
936		*/
937		private int interruptibleWait(int phase) throws InterruptedException {
821	–	int spins = maxUntimedSpins;
938		QNode node = null;
939		boolean queued = false;
940		boolean interrupted = false;
941		int p;
942	<	while ((p = getPhase()) == phase) {
943	<	if (interrupted = Thread.interrupted())
944	<	break;
945	<	if (node != null) {
946	<	if (!queued) {
947	<	AtomicReference<QNode> head = queueFor(phase);
948	<	queued = head.compareAndSet(node.next = head.get(), node);
833	<	}
834	<	else if (node.thread != null)
835	<	LockSupport.park(this);
836	<	}
837	<	else if (spins <= 0)
838	<	node = new QNode();
942	>	while ((p = getPhase()) == phase && !interrupted) {
943	>	if (Thread.interrupted())
944	>	interrupted = true;
945	>	else if (node == null)
946	>	node = new QNode(this, phase, true, false, 0, 0);
947	>	else if (!queued)
948	>	queued = tryEnqueue(node);
949		else
950	<	--spins;
950	>	interrupted = node.doWait();
951		}
952		if (node != null)
953		node.thread = null;
954	+	if (p != phase || (p = getPhase()) != phase)
955	+	releaseWaiters(phase);
956		if (interrupted)
957		throw new InterruptedException();
846	–	releaseWaiters(phase);
958		return p;
959		}
960
961		/**
962	<	* Even messier timeout version.
962	>	* Timeout version.
963	>	* @return current phase
964		*/
965		private int timedWait(int phase, long nanos)
966		throws InterruptedException, TimeoutException {
967	+	long startTime = System.nanoTime();
968	+	QNode node = null;
969	+	boolean queued = false;
970	+	boolean interrupted = false;
971		int p;
972	<	if ((p = getPhase()) == phase) {
973	<	long lastTime = System.nanoTime();
974	<	int spins = maxTimedSpins;
975	<	QNode node = null;
976	<	boolean queued = false;
977	<	boolean interrupted = false;
978	<	while ((p = getPhase()) == phase) {
979	<	if (interrupted = Thread.interrupted())
980	<	break;
981	<	long now = System.nanoTime();
982	<	if ((nanos -= now - lastTime) <= 0)
867	<	break;
868	<	lastTime = now;
869	<	if (node != null) {
870	<	if (!queued) {
871	<	AtomicReference<QNode> head = queueFor(phase);
872	<	queued = head.compareAndSet(node.next = head.get(), node);
873	<	}
874	<	else if (node.thread != null &&
875	<	nanos > spinForTimeoutThreshold) {
876	<	LockSupport.parkNanos(this, nanos);
877	<	}
878	<	}
879	<	else if (spins <= 0)
880	<	node = new QNode();
881	<	else
882	<	--spins;
883	<	}
884	<	if (node != null)
885	<	node.thread = null;
886	<	if (interrupted)
887	<	throw new InterruptedException();
888	<	if (p == phase && (p = getPhase()) == phase)
889	<	throw new TimeoutException();
972	>	while ((p = getPhase()) == phase && !interrupted) {
973	>	if (Thread.interrupted())
974	>	interrupted = true;
975	>	else if (nanos - (System.nanoTime() - startTime) <= 0)
976	>	break;
977	>	else if (node == null)
978	>	node = new QNode(this, phase, true, true, startTime, nanos);
979	>	else if (!queued)
980	>	queued = tryEnqueue(node);
981	>	else
982	>	interrupted = node.doWait();
983		}
984	<	releaseWaiters(phase);
984	>	if (node != null)
985	>	node.thread = null;
986	>	if (p != phase || (p = getPhase()) != phase)
987	>	releaseWaiters(phase);
988	>	if (interrupted)
989	>	throw new InterruptedException();
990	>	if (p == phase)
991	>	throw new TimeoutException();
992		return p;
993		}
994
995	<	// Temporary Unsafe mechanics for preliminary release
995	>	// Unsafe mechanics
996
997	<	static final Unsafe _unsafe;
998	<	static final long stateOffset;
997	>	private static final sun.misc.Unsafe UNSAFE = getUnsafe();
998	>	private static final long stateOffset =
999	>	objectFieldOffset("state", Phaser.class);
1000
1001	<	static {
1001	>	private final boolean casState(long cmp, long val) {
1002	>	return UNSAFE.compareAndSwapLong(this, stateOffset, cmp, val);
1003	>	}
1004	>
1005	>	private static long objectFieldOffset(String field, Class<?> klazz) {
1006		try {
1007	<	if (Phaser.class.getClassLoader() != null) {
1008	<	Field f = Unsafe.class.getDeclaredField("theUnsafe");
1009	<	f.setAccessible(true);
1010	<	_unsafe = (Unsafe)f.get(null);
1011	<	}
1012	<	else
908	<	_unsafe = Unsafe.getUnsafe();
909	<	stateOffset = _unsafe.objectFieldOffset
910	<	(Phaser.class.getDeclaredField("state"));
911	<	} catch (Exception e) {
912	<	throw new RuntimeException("Could not initialize intrinsics", e);
1007	>	return UNSAFE.objectFieldOffset(klazz.getDeclaredField(field));
1008	>	} catch (NoSuchFieldException e) {
1009	>	// Convert Exception to corresponding Error
1010	>	NoSuchFieldError error = new NoSuchFieldError(field);
1011	>	error.initCause(e);
1012	>	throw error;
1013		}
1014		}
1015
1016	<	final boolean casState(long cmp, long val) {
1017	<	return _unsafe.compareAndSwapLong(this, stateOffset, cmp, val);
1016	>	/**
1017	>	* Returns a sun.misc.Unsafe.  Suitable for use in a 3rd party package.
1018	>	* Replace with a simple call to Unsafe.getUnsafe when integrating
1019	>	* into a jdk.
1020	>	*
1021	>	* @return a sun.misc.Unsafe
1022	>	*/
1023	>	private static sun.misc.Unsafe getUnsafe() {
1024	>	try {
1025	>	return sun.misc.Unsafe.getUnsafe();
1026	>	} catch (SecurityException se) {
1027	>	try {
1028	>	return java.security.AccessController.doPrivileged
1029	>	(new java.security
1030	>	.PrivilegedExceptionAction<sun.misc.Unsafe>() {
1031	>	public sun.misc.Unsafe run() throws Exception {
1032	>	java.lang.reflect.Field f = sun.misc
1033	>	.Unsafe.class.getDeclaredField("theUnsafe");
1034	>	f.setAccessible(true);
1035	>	return (sun.misc.Unsafe) f.get(null);
1036	>	}});
1037	>	} catch (java.security.PrivilegedActionException e) {
1038	>	throw new RuntimeException("Could not initialize intrinsics",
1039	>	e.getCause());
1040	>	}
1041	>	}
1042		}
1043		}

Diff Legend

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