• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
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/publicdomain/zero/1.0/
5  */
6 
7 package java.util.concurrent.locks;
8 
9 import java.util.concurrent.TimeUnit;
10 
11 /**
12  * {@code Lock} implementations provide more extensive locking
13  * operations than can be obtained using {@code synchronized} methods
14  * and statements.  They allow more flexible structuring, may have
15  * quite different properties, and may support multiple associated
16  * {@link Condition} objects.
17  *
18  * <p>A lock is a tool for controlling access to a shared resource by
19  * multiple threads. Commonly, a lock provides exclusive access to a
20  * shared resource: only one thread at a time can acquire the lock and
21  * all access to the shared resource requires that the lock be
22  * acquired first. However, some locks may allow concurrent access to
23  * a shared resource, such as the read lock of a {@link ReadWriteLock}.
24  *
25  * <p>The use of {@code synchronized} methods or statements provides
26  * access to the implicit monitor lock associated with every object, but
27  * forces all lock acquisition and release to occur in a block-structured way:
28  * when multiple locks are acquired they must be released in the opposite
29  * order, and all locks must be released in the same lexical scope in which
30  * they were acquired.
31  *
32  * <p>While the scoping mechanism for {@code synchronized} methods
33  * and statements makes it much easier to program with monitor locks,
34  * and helps avoid many common programming errors involving locks,
35  * there are occasions where you need to work with locks in a more
36  * flexible way. For example, some algorithms for traversing
37  * concurrently accessed data structures require the use of
38  * &quot;hand-over-hand&quot; or &quot;chain locking&quot;: you
39  * acquire the lock of node A, then node B, then release A and acquire
40  * C, then release B and acquire D and so on.  Implementations of the
41  * {@code Lock} interface enable the use of such techniques by
42  * allowing a lock to be acquired and released in different scopes,
43  * and allowing multiple locks to be acquired and released in any
44  * order.
45  *
46  * <p>With this increased flexibility comes additional
47  * responsibility. The absence of block-structured locking removes the
48  * automatic release of locks that occurs with {@code synchronized}
49  * methods and statements. In most cases, the following idiom
50  * should be used:
51  *
52  * <pre> {@code
53  * Lock l = ...;
54  * l.lock();
55  * try {
56  *   // access the resource protected by this lock
57  * } finally {
58  *   l.unlock();
59  * }}</pre>
60  *
61  * When locking and unlocking occur in different scopes, care must be
62  * taken to ensure that all code that is executed while the lock is
63  * held is protected by try-finally or try-catch to ensure that the
64  * lock is released when necessary.
65  *
66  * <p>{@code Lock} implementations provide additional functionality
67  * over the use of {@code synchronized} methods and statements by
68  * providing a non-blocking attempt to acquire a lock ({@link
69  * #tryLock()}), an attempt to acquire the lock that can be
70  * interrupted ({@link #lockInterruptibly}, and an attempt to acquire
71  * the lock that can timeout ({@link #tryLock(long, TimeUnit)}).
72  *
73  * <p>A {@code Lock} class can also provide behavior and semantics
74  * that is quite different from that of the implicit monitor lock,
75  * such as guaranteed ordering, non-reentrant usage, or deadlock
76  * detection. If an implementation provides such specialized semantics
77  * then the implementation must document those semantics.
78  *
79  * <p>Note that {@code Lock} instances are just normal objects and can
80  * themselves be used as the target in a {@code synchronized} statement.
81  * Acquiring the
82  * monitor lock of a {@code Lock} instance has no specified relationship
83  * with invoking any of the {@link #lock} methods of that instance.
84  * It is recommended that to avoid confusion you never use {@code Lock}
85  * instances in this way, except within their own implementation.
86  *
87  * <p>Except where noted, passing a {@code null} value for any
88  * parameter will result in a {@link NullPointerException} being
89  * thrown.
90  *
91  * <h3>Memory Synchronization</h3>
92  *
93  * <p>All {@code Lock} implementations <em>must</em> enforce the same
94  * memory synchronization semantics as provided by the built-in monitor
95  * lock, as described in
96  * <a href="https://docs.oracle.com/javase/specs/jls/se8/html/jls-17.html#jls-17.4">
97  * Chapter 17 of
98  * <cite>The Java&trade; Language Specification</cite></a>:
99  * <ul>
100  * <li>A successful {@code lock} operation has the same memory
101  * synchronization effects as a successful <em>Lock</em> action.
102  * <li>A successful {@code unlock} operation has the same
103  * memory synchronization effects as a successful <em>Unlock</em> action.
104  * </ul>
105  *
106  * Unsuccessful locking and unlocking operations, and reentrant
107  * locking/unlocking operations, do not require any memory
108  * synchronization effects.
109  *
110  * <h3>Implementation Considerations</h3>
111  *
112  * <p>The three forms of lock acquisition (interruptible,
113  * non-interruptible, and timed) may differ in their performance
114  * characteristics, ordering guarantees, or other implementation
115  * qualities.  Further, the ability to interrupt the <em>ongoing</em>
116  * acquisition of a lock may not be available in a given {@code Lock}
117  * class.  Consequently, an implementation is not required to define
118  * exactly the same guarantees or semantics for all three forms of
119  * lock acquisition, nor is it required to support interruption of an
120  * ongoing lock acquisition.  An implementation is required to clearly
121  * document the semantics and guarantees provided by each of the
122  * locking methods. It must also obey the interruption semantics as
123  * defined in this interface, to the extent that interruption of lock
124  * acquisition is supported: which is either totally, or only on
125  * method entry.
126  *
127  * <p>As interruption generally implies cancellation, and checks for
128  * interruption are often infrequent, an implementation can favor responding
129  * to an interrupt over normal method return. This is true even if it can be
130  * shown that the interrupt occurred after another action may have unblocked
131  * the thread. An implementation should document this behavior.
132  *
133  * @see ReentrantLock
134  * @see Condition
135  * @see ReadWriteLock
136  *
137  * @since 1.5
138  * @author Doug Lea
139  */
140 public interface Lock {
141 
142     /**
143      * Acquires the lock.
144      *
145      * <p>If the lock is not available then the current thread becomes
146      * disabled for thread scheduling purposes and lies dormant until the
147      * lock has been acquired.
148      *
149      * <p><b>Implementation Considerations</b>
150      *
151      * <p>A {@code Lock} implementation may be able to detect erroneous use
152      * of the lock, such as an invocation that would cause deadlock, and
153      * may throw an (unchecked) exception in such circumstances.  The
154      * circumstances and the exception type must be documented by that
155      * {@code Lock} implementation.
156      */
lock()157     void lock();
158 
159     /**
160      * Acquires the lock unless the current thread is
161      * {@linkplain Thread#interrupt interrupted}.
162      *
163      * <p>Acquires the lock if it is available and returns immediately.
164      *
165      * <p>If the lock is not available then the current thread becomes
166      * disabled for thread scheduling purposes and lies dormant until
167      * one of two things happens:
168      *
169      * <ul>
170      * <li>The lock is acquired by the current thread; or
171      * <li>Some other thread {@linkplain Thread#interrupt interrupts} the
172      * current thread, and interruption of lock acquisition is supported.
173      * </ul>
174      *
175      * <p>If the current thread:
176      * <ul>
177      * <li>has its interrupted status set on entry to this method; or
178      * <li>is {@linkplain Thread#interrupt interrupted} while acquiring the
179      * lock, and interruption of lock acquisition is supported,
180      * </ul>
181      * then {@link InterruptedException} is thrown and the current thread's
182      * interrupted status is cleared.
183      *
184      * <p><b>Implementation Considerations</b>
185      *
186      * <p>The ability to interrupt a lock acquisition in some
187      * implementations may not be possible, and if possible may be an
188      * expensive operation.  The programmer should be aware that this
189      * may be the case. An implementation should document when this is
190      * the case.
191      *
192      * <p>An implementation can favor responding to an interrupt over
193      * normal method return.
194      *
195      * <p>A {@code Lock} implementation may be able to detect
196      * erroneous use of the lock, such as an invocation that would
197      * cause deadlock, and may throw an (unchecked) exception in such
198      * circumstances.  The circumstances and the exception type must
199      * be documented by that {@code Lock} implementation.
200      *
201      * @throws InterruptedException if the current thread is
202      *         interrupted while acquiring the lock (and interruption
203      *         of lock acquisition is supported)
204      */
lockInterruptibly()205     void lockInterruptibly() throws InterruptedException;
206 
207     /**
208      * Acquires the lock only if it is free at the time of invocation.
209      *
210      * <p>Acquires the lock if it is available and returns immediately
211      * with the value {@code true}.
212      * If the lock is not available then this method will return
213      * immediately with the value {@code false}.
214      *
215      * <p>A typical usage idiom for this method would be:
216      * <pre> {@code
217      * Lock lock = ...;
218      * if (lock.tryLock()) {
219      *   try {
220      *     // manipulate protected state
221      *   } finally {
222      *     lock.unlock();
223      *   }
224      * } else {
225      *   // perform alternative actions
226      * }}</pre>
227      *
228      * This usage ensures that the lock is unlocked if it was acquired, and
229      * doesn't try to unlock if the lock was not acquired.
230      *
231      * @return {@code true} if the lock was acquired and
232      *         {@code false} otherwise
233      */
tryLock()234     boolean tryLock();
235 
236     /**
237      * Acquires the lock if it is free within the given waiting time and the
238      * current thread has not been {@linkplain Thread#interrupt interrupted}.
239      *
240      * <p>If the lock is available this method returns immediately
241      * with the value {@code true}.
242      * If the lock is not available then
243      * the current thread becomes disabled for thread scheduling
244      * purposes and lies dormant until one of three things happens:
245      * <ul>
246      * <li>The lock is acquired by the current thread; or
247      * <li>Some other thread {@linkplain Thread#interrupt interrupts} the
248      * current thread, and interruption of lock acquisition is supported; or
249      * <li>The specified waiting time elapses
250      * </ul>
251      *
252      * <p>If the lock is acquired then the value {@code true} is returned.
253      *
254      * <p>If the current thread:
255      * <ul>
256      * <li>has its interrupted status set on entry to this method; or
257      * <li>is {@linkplain Thread#interrupt interrupted} while acquiring
258      * the lock, and interruption of lock acquisition is supported,
259      * </ul>
260      * then {@link InterruptedException} is thrown and the current thread's
261      * interrupted status is cleared.
262      *
263      * <p>If the specified waiting time elapses then the value {@code false}
264      * is returned.
265      * If the time is
266      * less than or equal to zero, the method will not wait at all.
267      *
268      * <p><b>Implementation Considerations</b>
269      *
270      * <p>The ability to interrupt a lock acquisition in some implementations
271      * may not be possible, and if possible may
272      * be an expensive operation.
273      * The programmer should be aware that this may be the case. An
274      * implementation should document when this is the case.
275      *
276      * <p>An implementation can favor responding to an interrupt over normal
277      * method return, or reporting a timeout.
278      *
279      * <p>A {@code Lock} implementation may be able to detect
280      * erroneous use of the lock, such as an invocation that would cause
281      * deadlock, and may throw an (unchecked) exception in such circumstances.
282      * The circumstances and the exception type must be documented by that
283      * {@code Lock} implementation.
284      *
285      * @param time the maximum time to wait for the lock
286      * @param unit the time unit of the {@code time} argument
287      * @return {@code true} if the lock was acquired and {@code false}
288      *         if the waiting time elapsed before the lock was acquired
289      *
290      * @throws InterruptedException if the current thread is interrupted
291      *         while acquiring the lock (and interruption of lock
292      *         acquisition is supported)
293      */
tryLock(long time, TimeUnit unit)294     boolean tryLock(long time, TimeUnit unit) throws InterruptedException;
295 
296     /**
297      * Releases the lock.
298      *
299      * <p><b>Implementation Considerations</b>
300      *
301      * <p>A {@code Lock} implementation will usually impose
302      * restrictions on which thread can release a lock (typically only the
303      * holder of the lock can release it) and may throw
304      * an (unchecked) exception if the restriction is violated.
305      * Any restrictions and the exception
306      * type must be documented by that {@code Lock} implementation.
307      */
unlock()308     void unlock();
309 
310     /**
311      * Returns a new {@link Condition} instance that is bound to this
312      * {@code Lock} instance.
313      *
314      * <p>Before waiting on the condition the lock must be held by the
315      * current thread.
316      * A call to {@link Condition#await()} will atomically release the lock
317      * before waiting and re-acquire the lock before the wait returns.
318      *
319      * <p><b>Implementation Considerations</b>
320      *
321      * <p>The exact operation of the {@link Condition} instance depends on
322      * the {@code Lock} implementation and must be documented by that
323      * implementation.
324      *
325      * @return A new {@link Condition} instance for this {@code Lock} instance
326      * @throws UnsupportedOperationException if this {@code Lock}
327      *         implementation does not support conditions
328      */
newCondition()329     Condition newCondition();
330 }
331