author | dl |
Thu, 07 Apr 2011 15:06:32 +0100 | |
changeset 9242 | ef138d47df58 |
parent 5506 | 202f599c92aa |
child 18576 | 7a5c231327af |
permissions | -rw-r--r-- |
2 | 1 |
/* |
2 |
* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. |
|
3 |
* |
|
4 |
* This code is free software; you can redistribute it and/or modify it |
|
5 |
* under the terms of the GNU General Public License version 2 only, as |
|
5506 | 6 |
* published by the Free Software Foundation. Oracle designates this |
2 | 7 |
* particular file as subject to the "Classpath" exception as provided |
5506 | 8 |
* by Oracle in the LICENSE file that accompanied this code. |
2 | 9 |
* |
10 |
* This code is distributed in the hope that it will be useful, but WITHOUT |
|
11 |
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or |
|
12 |
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License |
|
13 |
* version 2 for more details (a copy is included in the LICENSE file that |
|
14 |
* accompanied this code). |
|
15 |
* |
|
16 |
* You should have received a copy of the GNU General Public License version |
|
17 |
* 2 along with this work; if not, write to the Free Software Foundation, |
|
18 |
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. |
|
19 |
* |
|
5506 | 20 |
* Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA |
21 |
* or visit www.oracle.com if you need additional information or have any |
|
22 |
* questions. |
|
2 | 23 |
*/ |
24 |
||
25 |
/* |
|
26 |
* This file is available under and governed by the GNU General Public |
|
27 |
* License version 2 only, as published by the Free Software Foundation. |
|
28 |
* However, the following notice accompanied the original version of this |
|
29 |
* file: |
|
30 |
* |
|
31 |
* Written by Doug Lea with assistance from members of JCP JSR-166 |
|
32 |
* Expert Group and released to the public domain, as explained at |
|
9242
ef138d47df58
7034657: Update Creative Commons license URL in legal notices
dl
parents:
5506
diff
changeset
|
33 |
* http://creativecommons.org/publicdomain/zero/1.0/ |
2 | 34 |
*/ |
35 |
||
36 |
package java.util.concurrent.locks; |
|
37 |
||
38 |
/** |
|
39 |
* A <tt>ReadWriteLock</tt> maintains a pair of associated {@link |
|
40 |
* Lock locks}, one for read-only operations and one for writing. |
|
41 |
* The {@link #readLock read lock} may be held simultaneously by |
|
42 |
* multiple reader threads, so long as there are no writers. The |
|
43 |
* {@link #writeLock write lock} is exclusive. |
|
44 |
* |
|
45 |
* <p>All <tt>ReadWriteLock</tt> implementations must guarantee that |
|
46 |
* the memory synchronization effects of <tt>writeLock</tt> operations |
|
47 |
* (as specified in the {@link Lock} interface) also hold with respect |
|
48 |
* to the associated <tt>readLock</tt>. That is, a thread successfully |
|
49 |
* acquiring the read lock will see all updates made upon previous |
|
50 |
* release of the write lock. |
|
51 |
* |
|
52 |
* <p>A read-write lock allows for a greater level of concurrency in |
|
53 |
* accessing shared data than that permitted by a mutual exclusion lock. |
|
54 |
* It exploits the fact that while only a single thread at a time (a |
|
55 |
* <em>writer</em> thread) can modify the shared data, in many cases any |
|
56 |
* number of threads can concurrently read the data (hence <em>reader</em> |
|
57 |
* threads). |
|
58 |
* In theory, the increase in concurrency permitted by the use of a read-write |
|
59 |
* lock will lead to performance improvements over the use of a mutual |
|
60 |
* exclusion lock. In practice this increase in concurrency will only be fully |
|
61 |
* realized on a multi-processor, and then only if the access patterns for |
|
62 |
* the shared data are suitable. |
|
63 |
* |
|
64 |
* <p>Whether or not a read-write lock will improve performance over the use |
|
65 |
* of a mutual exclusion lock depends on the frequency that the data is |
|
66 |
* read compared to being modified, the duration of the read and write |
|
67 |
* operations, and the contention for the data - that is, the number of |
|
68 |
* threads that will try to read or write the data at the same time. |
|
69 |
* For example, a collection that is initially populated with data and |
|
70 |
* thereafter infrequently modified, while being frequently searched |
|
71 |
* (such as a directory of some kind) is an ideal candidate for the use of |
|
72 |
* a read-write lock. However, if updates become frequent then the data |
|
73 |
* spends most of its time being exclusively locked and there is little, if any |
|
74 |
* increase in concurrency. Further, if the read operations are too short |
|
75 |
* the overhead of the read-write lock implementation (which is inherently |
|
76 |
* more complex than a mutual exclusion lock) can dominate the execution |
|
77 |
* cost, particularly as many read-write lock implementations still serialize |
|
78 |
* all threads through a small section of code. Ultimately, only profiling |
|
79 |
* and measurement will establish whether the use of a read-write lock is |
|
80 |
* suitable for your application. |
|
81 |
* |
|
82 |
* |
|
83 |
* <p>Although the basic operation of a read-write lock is straight-forward, |
|
84 |
* there are many policy decisions that an implementation must make, which |
|
85 |
* may affect the effectiveness of the read-write lock in a given application. |
|
86 |
* Examples of these policies include: |
|
87 |
* <ul> |
|
88 |
* <li>Determining whether to grant the read lock or the write lock, when |
|
89 |
* both readers and writers are waiting, at the time that a writer releases |
|
90 |
* the write lock. Writer preference is common, as writes are expected to be |
|
91 |
* short and infrequent. Reader preference is less common as it can lead to |
|
92 |
* lengthy delays for a write if the readers are frequent and long-lived as |
|
93 |
* expected. Fair, or "in-order" implementations are also possible. |
|
94 |
* |
|
95 |
* <li>Determining whether readers that request the read lock while a |
|
96 |
* reader is active and a writer is waiting, are granted the read lock. |
|
97 |
* Preference to the reader can delay the writer indefinitely, while |
|
98 |
* preference to the writer can reduce the potential for concurrency. |
|
99 |
* |
|
100 |
* <li>Determining whether the locks are reentrant: can a thread with the |
|
101 |
* write lock reacquire it? Can it acquire a read lock while holding the |
|
102 |
* write lock? Is the read lock itself reentrant? |
|
103 |
* |
|
104 |
* <li>Can the write lock be downgraded to a read lock without allowing |
|
105 |
* an intervening writer? Can a read lock be upgraded to a write lock, |
|
106 |
* in preference to other waiting readers or writers? |
|
107 |
* |
|
108 |
* </ul> |
|
109 |
* You should consider all of these things when evaluating the suitability |
|
110 |
* of a given implementation for your application. |
|
111 |
* |
|
112 |
* @see ReentrantReadWriteLock |
|
113 |
* @see Lock |
|
114 |
* @see ReentrantLock |
|
115 |
* |
|
116 |
* @since 1.5 |
|
117 |
* @author Doug Lea |
|
118 |
*/ |
|
119 |
public interface ReadWriteLock { |
|
120 |
/** |
|
121 |
* Returns the lock used for reading. |
|
122 |
* |
|
123 |
* @return the lock used for reading. |
|
124 |
*/ |
|
125 |
Lock readLock(); |
|
126 |
||
127 |
/** |
|
128 |
* Returns the lock used for writing. |
|
129 |
* |
|
130 |
* @return the lock used for writing. |
|
131 |
*/ |
|
132 |
Lock writeLock(); |
|
133 |
} |