1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
|
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
<html xmlns="http://www.w3.org/1999/xhtml"><head><title>Chapter 28. Concurrency</title><meta name="generator" content="DocBook XSL-NS Stylesheets V1.76.1"/><meta name="keywords" content=" ISO C++ , library "/><meta name="keywords" content=" ISO C++ , library "/><link rel="home" href="../spine.html" title="The GNU C++ Library"/><link rel="up" href="extensions.html" title="Part III. Extensions"/><link rel="prev" href="ext_demangling.html" title="Chapter 27. Demangling"/><link rel="next" href="bk01pt03ch28s02.html" title="Implementation"/></head><body><div class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="3" align="center">Chapter 28. Concurrency</th></tr><tr><td align="left"><a accesskey="p" href="ext_demangling.html">Prev</a> </td><th width="60%" align="center">Part III.
Extensions
</th><td align="right"> <a accesskey="n" href="bk01pt03ch28s02.html">Next</a></td></tr></table><hr/></div><div class="chapter" title="Chapter 28. Concurrency"><div class="titlepage"><div><div><h2 class="title"><a id="manual.ext.concurrency"/>Chapter 28. Concurrency</h2></div></div></div><div class="toc"><p><strong>Table of Contents</strong></p><dl><dt><span class="section"><a href="ext_concurrency.html#manual.ext.concurrency.design">Design</a></span></dt><dd><dl><dt><span class="section"><a href="ext_concurrency.html#manual.ext.concurrency.design.threads">Interface to Locks and Mutexes</a></span></dt><dt><span class="section"><a href="ext_concurrency.html#manual.ext.concurrency.design.atomics">Interface to Atomic Functions</a></span></dt></dl></dd><dt><span class="section"><a href="bk01pt03ch28s02.html">Implementation</a></span></dt><dd><dl><dt><span class="section"><a href="bk01pt03ch28s02.html#manual.ext.concurrency.impl.atomic_fallbacks">Using Builtin Atomic Functions</a></span></dt><dt><span class="section"><a href="bk01pt03ch28s02.html#manual.ext.concurrency.impl.thread">Thread Abstraction</a></span></dt></dl></dd><dt><span class="section"><a href="bk01pt03ch28s03.html">Use</a></span></dt></dl></div><div class="section" title="Design"><div class="titlepage"><div><div><h2 class="title"><a id="manual.ext.concurrency.design"/>Design</h2></div></div></div><div class="section" title="Interface to Locks and Mutexes"><div class="titlepage"><div><div><h3 class="title"><a id="manual.ext.concurrency.design.threads"/>Interface to Locks and Mutexes</h3></div></div></div><p>The file <ext/concurrence.h> contains all the higher-level
constructs for playing with threads. In contrast to the atomics layer,
the concurrence layer consists largely of types. All types are defined within <code class="code">namespace __gnu_cxx</code>.
</p><p>
These types can be used in a portable manner, regardless of the
specific environment. They are carefully designed to provide optimum
efficiency and speed, abstracting out underlying thread calls and
accesses when compiling for single-threaded situations (even on hosts
that support multiple threads.)
</p><p>The enumerated type <code class="code">_Lock_policy</code> details the set of
available locking
policies: <code class="code">_S_single</code>, <code class="code">_S_mutex</code>,
and <code class="code">_S_atomic</code>.
</p><div class="itemizedlist"><ul class="itemizedlist"><li class="listitem"><p><code class="code">_S_single</code></p><p>Indicates single-threaded code that does not need locking.
</p></li><li class="listitem"><p><code class="code">_S_mutex</code></p><p>Indicates multi-threaded code using thread-layer abstractions.
</p></li><li class="listitem"><p><code class="code">_S_atomic</code></p><p>Indicates multi-threaded code using atomic operations.
</p></li></ul></div><p>The compile-time constant <code class="code">__default_lock_policy</code> is set
to one of the three values above, depending on characteristics of the
host environment and the current compilation flags.
</p><p>Two more datatypes make up the rest of the
interface: <code class="code">__mutex</code>, and <code class="code">__scoped_lock</code>.
</p><p>
</p><p>The scoped lock idiom is well-discussed within the C++
community. This version takes a <code class="code">__mutex</code> reference, and
locks it during construction of <code class="code">__scoped_locke</code> and
unlocks it during destruction. This is an efficient way of locking
critical sections, while retaining exception-safety.
</p></div><div class="section" title="Interface to Atomic Functions"><div class="titlepage"><div><div><h3 class="title"><a id="manual.ext.concurrency.design.atomics"/>Interface to Atomic Functions</h3></div></div></div><p>
Two functions and one type form the base of atomic support.
</p><p>The type <code class="code">_Atomic_word</code> is a signed integral type
supporting atomic operations.
</p><p>
The two functions functions are:
</p><pre class="programlisting">
_Atomic_word
__exchange_and_add_dispatch(volatile _Atomic_word*, int);
void
__atomic_add_dispatch(volatile _Atomic_word*, int);
</pre><p>Both of these functions are declared in the header file
<ext/atomicity.h>, and are in <code class="code">namespace __gnu_cxx</code>.
</p><div class="itemizedlist"><ul class="itemizedlist"><li class="listitem"><p>
<code class="code">
__exchange_and_add_dispatch
</code>
</p><p>Adds the second argument's value to the first argument. Returns the old value.
</p></li><li class="listitem"><p>
<code class="code">
__atomic_add_dispatch
</code>
</p><p>Adds the second argument's value to the first argument. Has no return value.
</p></li></ul></div><p>
These functions forward to one of several specialized helper
functions, depending on the circumstances. For instance,
</p><p>
<code class="code">
__exchange_and_add_dispatch
</code>
</p><p>
Calls through to either of:
</p><div class="itemizedlist"><ul class="itemizedlist"><li class="listitem"><p><code class="code">__exchange_and_add</code>
</p><p>Multi-thread version. Inlined if compiler-generated builtin atomics
can be used, otherwise resolved at link time to a non-builtin code
sequence.
</p></li><li class="listitem"><p><code class="code">__exchange_and_add_single</code>
</p><p>Single threaded version. Inlined.</p></li></ul></div><p>However, only <code class="code">__exchange_and_add_dispatch</code>
and <code class="code">__atomic_add_dispatch</code> should be used. These functions
can be used in a portable manner, regardless of the specific
environment. They are carefully designed to provide optimum efficiency
and speed, abstracting out atomic accesses when they are not required
(even on hosts that support compiler intrinsics for atomic
operations.)
</p><p>
In addition, there are two macros
</p><p>
<code class="code">
_GLIBCXX_READ_MEM_BARRIER
</code>
</p><p>
<code class="code">
_GLIBCXX_WRITE_MEM_BARRIER
</code>
</p><p>
Which expand to the appropriate write and read barrier required by the
host hardware and operating system.
</p></div></div></div><div class="navfooter"><hr/><table width="100%" summary="Navigation footer"><tr><td align="left"><a accesskey="p" href="ext_demangling.html">Prev</a> </td><td align="center"><a accesskey="u" href="extensions.html">Up</a></td><td align="right"> <a accesskey="n" href="bk01pt03ch28s02.html">Next</a></td></tr><tr><td align="left" valign="top">Chapter 27. Demangling </td><td align="center"><a accesskey="h" href="../spine.html">Home</a></td><td align="right" valign="top"> Implementation</td></tr></table></div></body></html>
|
