0
|
1 |
/****************************************************************************
|
|
2 |
**
|
|
3 |
** Copyright (C) 2009 Nokia Corporation and/or its subsidiary(-ies).
|
|
4 |
** All rights reserved.
|
|
5 |
** Contact: Nokia Corporation (qt-info@nokia.com)
|
|
6 |
**
|
|
7 |
** This file is part of the documentation of the Qt Toolkit.
|
|
8 |
**
|
|
9 |
** $QT_BEGIN_LICENSE:LGPL$
|
|
10 |
** No Commercial Usage
|
|
11 |
** This file contains pre-release code and may not be distributed.
|
|
12 |
** You may use this file in accordance with the terms and conditions
|
|
13 |
** contained in the Technology Preview License Agreement accompanying
|
|
14 |
** this package.
|
|
15 |
**
|
|
16 |
** GNU Lesser General Public License Usage
|
|
17 |
** Alternatively, this file may be used under the terms of the GNU Lesser
|
|
18 |
** General Public License version 2.1 as published by the Free Software
|
|
19 |
** Foundation and appearing in the file LICENSE.LGPL included in the
|
|
20 |
** packaging of this file. Please review the following information to
|
|
21 |
** ensure the GNU Lesser General Public License version 2.1 requirements
|
|
22 |
** will be met: http://www.gnu.org/licenses/old-licenses/lgpl-2.1.html.
|
|
23 |
**
|
|
24 |
** In addition, as a special exception, Nokia gives you certain additional
|
|
25 |
** rights. These rights are described in the Nokia Qt LGPL Exception
|
|
26 |
** version 1.1, included in the file LGPL_EXCEPTION.txt in this package.
|
|
27 |
**
|
|
28 |
** If you have questions regarding the use of this file, please contact
|
|
29 |
** Nokia at qt-info@nokia.com.
|
|
30 |
**
|
|
31 |
**
|
|
32 |
**
|
|
33 |
**
|
|
34 |
**
|
|
35 |
**
|
|
36 |
**
|
|
37 |
**
|
|
38 |
** $QT_END_LICENSE$
|
|
39 |
**
|
|
40 |
****************************************************************************/
|
|
41 |
|
|
42 |
/*!
|
|
43 |
\example threads/waitconditions
|
|
44 |
\title Wait Conditions Example
|
|
45 |
|
|
46 |
The Wait Conditions example shows how to use QWaitCondition and
|
|
47 |
QMutex to control access to a circular buffer shared by a
|
|
48 |
producer thread and a consumer thread.
|
|
49 |
|
|
50 |
The producer writes data to the buffer until it reaches the end
|
|
51 |
of the buffer, at which point it restarts from the beginning,
|
|
52 |
overwriting existing data. The consumer thread reads the data as
|
|
53 |
it is produced and writes it to standard error.
|
|
54 |
|
|
55 |
Wait conditions make it possible to have a higher level of
|
|
56 |
concurrency than what is possible with mutexes alone. If accesses
|
|
57 |
to the buffer were simply guarded by a QMutex, the consumer
|
|
58 |
thread couldn't access the buffer at the same time as the
|
|
59 |
producer thread. Yet, there is no harm in having both threads
|
|
60 |
working on \e{different parts} of the buffer at the same time.
|
|
61 |
|
|
62 |
The example comprises two classes: \c Producer and \c Consumer.
|
|
63 |
Both inherit from QThread. The circular buffer used for
|
|
64 |
communicating between these two classes and the synchronization
|
|
65 |
tools that protect it are global variables.
|
|
66 |
|
|
67 |
An alternative to using QWaitCondition and QMutex to solve the
|
|
68 |
producer-consumer problem is to use QSemaphore. This is what the
|
|
69 |
\l{threads/semaphores}{Semaphores} example does.
|
|
70 |
|
|
71 |
\section1 Global Variables
|
|
72 |
|
|
73 |
Let's start by reviewing the circular buffer and the associated
|
|
74 |
synchronization tools:
|
|
75 |
|
|
76 |
\snippet examples/threads/waitconditions/waitconditions.cpp 0
|
|
77 |
|
|
78 |
\c DataSize is the amount of data that the producer will generate.
|
|
79 |
To keep the example as simple as possible, we make it a constant.
|
|
80 |
\c BufferSize is the size of the circular buffer. It is less than
|
|
81 |
\c DataSize, meaning that at some point the producer will reach
|
|
82 |
the end of the buffer and restart from the beginning.
|
|
83 |
|
|
84 |
To synchronize the producer and the consumer, we need two wait
|
|
85 |
conditions and one mutex. The \c bufferNotEmpty condition is
|
|
86 |
signalled when the producer has generated some data, telling the
|
|
87 |
consumer that it can start reading it. The \c bufferNotFull
|
|
88 |
condition is signalled when the consumer has read some data,
|
|
89 |
telling the producer that it can generate more. The \c numUsedBytes
|
|
90 |
is the number of bytes in the buffer that contain data.
|
|
91 |
|
|
92 |
Together, the wait conditions, the mutex, and the \c numUsedBytes
|
|
93 |
counter ensure that the producer is never more than \c BufferSize
|
|
94 |
bytes ahead of the consumer, and that the consumer never reads
|
|
95 |
data that the consumer hasn't generated yet.
|
|
96 |
|
|
97 |
\section1 Producer Class
|
|
98 |
|
|
99 |
Let's review the code for the \c Producer class:
|
|
100 |
|
|
101 |
\snippet examples/threads/waitconditions/waitconditions.cpp 1
|
|
102 |
\snippet examples/threads/waitconditions/waitconditions.cpp 2
|
|
103 |
|
|
104 |
The producer generates \c DataSize bytes of data. Before it
|
|
105 |
writes a byte to the circular buffer, it must first check whether
|
|
106 |
the buffer is full (i.e., \c numUsedBytes equals \c BufferSize).
|
|
107 |
If the buffer is full, the thread waits on the \c bufferNotFull
|
|
108 |
condition.
|
|
109 |
|
|
110 |
At the end, the producer increments \c numUsedBytes and signalls
|
|
111 |
that the condition \c bufferNotEmpty is true, since \c
|
|
112 |
numUsedBytes is necessarily greater than 0.
|
|
113 |
|
|
114 |
We guard all accesses to the \c numUsedBytes variable with a
|
|
115 |
mutex. In addition, the QWaitCondition::wait() function accepts a
|
|
116 |
mutex as its argument. This mutex is unlocked before the thread
|
|
117 |
is put to sleep and locked when the thread wakes up. Furthermore,
|
|
118 |
the transition from the locked state to the wait state is atomic,
|
|
119 |
to prevent race conditions from occurring.
|
|
120 |
|
|
121 |
\section1 Consumer Class
|
|
122 |
|
|
123 |
Let's turn to the \c Consumer class:
|
|
124 |
|
|
125 |
\snippet examples/threads/waitconditions/waitconditions.cpp 3
|
|
126 |
\snippet examples/threads/waitconditions/waitconditions.cpp 4
|
|
127 |
|
|
128 |
The code is very similar to the producer. Before we read the
|
|
129 |
byte, we check whether the buffer is empty (\c numUsedBytes is 0)
|
|
130 |
instead of whether it's full and wait on the \c bufferNotEmpty
|
|
131 |
condition if it's empty. After we've read the byte, we decrement
|
|
132 |
\c numUsedBytes (instead of incrementing it), and we signal the
|
|
133 |
\c bufferNotFull condition (instead of the \c bufferNotEmpty
|
|
134 |
condition).
|
|
135 |
|
|
136 |
\section1 The main() Function
|
|
137 |
|
|
138 |
In \c main(), we create the two threads and call QThread::wait()
|
|
139 |
to ensure that both threads get time to finish before we exit:
|
|
140 |
|
|
141 |
\snippet examples/threads/waitconditions/waitconditions.cpp 5
|
|
142 |
\snippet examples/threads/waitconditions/waitconditions.cpp 6
|
|
143 |
|
|
144 |
So what happens when we run the program? Initially, the producer
|
|
145 |
thread is the only one that can do anything; the consumer is
|
|
146 |
blocked waiting for the \c bufferNotEmpty condition to be
|
|
147 |
signalled (\c numUsedBytes is 0). Once the producer has put one
|
|
148 |
byte in the buffer, \c numUsedBytes is \c BufferSize - 1 and the
|
|
149 |
\c bufferNotEmpty condition is signalled. At that point, two
|
|
150 |
things can happen: Either the consumer thread takes over and
|
|
151 |
reads that byte, or the consumer gets to produce a second byte.
|
|
152 |
|
|
153 |
The producer-consumer model presented in this example makes it
|
|
154 |
possible to write highly concurrent multithreaded applications.
|
|
155 |
On a multiprocessor machine, the program is potentially up to
|
|
156 |
twice as fast as the equivalent mutex-based program, since the
|
|
157 |
two threads can be active at the same time on different parts of
|
|
158 |
the buffer.
|
|
159 |
|
|
160 |
Be aware though that these benefits aren't always realized.
|
|
161 |
Locking and unlocking a QMutex has a cost. In practice, it would
|
|
162 |
probably be worthwhile to divide the buffer into chunks and to
|
|
163 |
operate on chunks instead of individual bytes. The buffer size is
|
|
164 |
also a parameter that must be selected carefully, based on
|
|
165 |
experimentation.
|
|
166 |
*/
|