+2004-02-11 Stefan Olsson <stefan@xapa.se>
+
+ * docs/html/ext/mt_allocator.html: New.
+
+2004-02-11 Benjamin Kosnik <bkoz@redhat.com>
+
+ * docs/html/20_util/allocator.html: New file, consolidate
+ allocator information here. Revamp.
+ * docs/html/documentation.html: Change links.
+ * docs/html/20_util/howto.html: Same.
+ * docs/html/ext/howto.html: Same.
+
2004-02-11 Paolo Carlini <pcarlini@suse.de>
PR libstdc++/13731 (first part: write)
--- /dev/null
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!DOCTYPE html
+ PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
+ "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
+
+<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
+<head>
+ <meta name="AUTHOR" content="pme@gcc.gnu.org (Phil Edwards) and bkoz@gcc.gnu.org (Benjamin Kosnik)" />
+ <meta name="KEYWORDS" content="c++, libstdc++, g++, allocator, memory" />
+ <meta name="DESCRIPTION" content="Allocators and allocation" />
+ <meta name="GENERATOR" content="emacs and ten fingers" />
+ <title>Allocators and allocation</title>
+<link rel="StyleSheet" href="../lib3styles.css" type="text/css" />
+<link rel="Copyright" href="../17_intro/license.html" type="text/html" />
+</head>
+<body>
+
+<h1 class="centered"><a name="top">Allocators and allocation</a></h1>
+
+<p class="fineprint"><em>
+ The latest version of this document is always available at
+ <a href="http://gcc.gnu.org/onlinedocs/libstdc++/20_util/allocator.html">
+ http://gcc.gnu.org/onlinedocs/libstdc++/20_util/allocator.html</a>.
+</em></p>
+
+<p><em>
+ To the <a href="http://gcc.gnu.org/libstdc++/">libstdc++-v3 homepage</a>.
+</em></p>
+
+<!-- ####################################################### -->
+<hr />
+<p> The C++ Standard encapsulates memory management characteristics
+ for strings, container classes, and parts of iostreams in a
+ template class called <code>std::allocator</code>.
+</p>
+
+<h3 class="left">
+ <a name="standard requirements">Standard requirements</a>
+</h3>
+ <p>The C++ standard only gives a few directives in this area:
+ </p>
+ <ul>
+ <li>When you add elements to a container, and the container must allocate
+ more memory to hold them, the container makes the request via its
+ <code>Allocator</code> template parameter. This includes adding
+ chars to the string class, which acts as a regular STL container
+ in this respect.
+ </li>
+ <li>The default <code>Allocator</code> of every container-of-T is
+ <code>std::allocator<T></code>.
+ </li>
+ <li>The interface of the <code>allocator<T></code> class is
+ extremely simple. It has about 20 public declarations (nested
+ typedefs, member functions, etc), but the two which concern us most
+ are:
+ <pre>
+ T* allocate (size_type n, const void* hint = 0);
+ void deallocate (T* p, size_type n);</pre>
+ (This is a simplification; the real signatures use nested typedefs.)
+ The <code>"n"</code> arguments in both those functions is a
+ <em>count</em> of the number of T's to allocate space for,
+ <em>not their total size</em>.
+ </li>
+ <li>"The storage is obtained by calling
+ <code>::operator new(size_t)</code>, but it is unspecified when or
+ how often this function is called. The use of <code>hint</code>
+ is unspecified, but intended as an aid to locality if an
+ implementation so desires." [20.4.1.1]/6
+ </li>
+ </ul>
+
+ <p> Complete details cam be found in the C++ standard, look in
+ [20.4 Memory].
+ </p>
+
+<h3 class="left">
+ <a name="probs possibilities">Problems and Possibilities</a>
+</h3>
+ <p>The easiest way of fulfilling the requirements is to call operator new
+ each time a container needs memory, and to call operator delete each
+ time the container releases memory. <strong>BUT</strong>
+ <a href="http://gcc.gnu.org/ml/libstdc++/2001-05/msg00105.html">this
+ method is horribly slow</a>.
+ </p>
+ <p>Or we can keep old memory around, and reuse it in a pool to save time.
+ The old libstdc++-v2 used a memory pool, and so do we. As of 3.0,
+ <a href="http://gcc.gnu.org/ml/libstdc++/2001-05/msg00136.html">it's
+ on by default</a>. The pool is shared among all the containers in the
+ program: when your program's std::vector<int> gets cut in half
+ and frees a bunch of its storage, that memory can be reused by the
+ private std::list<WonkyWidget> brought in from a KDE library
+ that you linked against. And we don't have to call operators new and
+ delete to pass the memory on, either, which is a speed bonus.
+ <strong>BUT</strong>...
+ </p>
+ <p>What about threads? No problem: in a threadsafe environment, the
+ memory pool is manipulated atomically, so you can grow a container in
+ one thread and shrink it in another, etc. <strong>BUT</strong> what
+ if threads in libstdc++-v3 aren't set up properly?
+ <a href="../faq/index.html#5_6">That's been answered already</a>.
+ </p>
+ <p><strong>BUT</strong> what if you want to use your own allocator? What
+ if you plan on using a runtime-loadable version of malloc() which uses
+ shared telepathic anonymous mmap'd sections serializable over a
+ network, so that memory requests <em>should</em> go through malloc?
+ And what if you need to debug it?
+ </p>
+
+<h3 class="left">
+ <a name="stdallocator">Implementation details of <code>std::allocator</code></a>
+</h3>
+ <p> The implementation of <code> std::allocator</code> has continued
+ to evolve through successive releases. Here's a brief history.
+ </p>
+
+<h5 class="left">
+ <a name="30allocator"> 3.0, 3.1, 3.2, 3.3 </a>
+</h5>
+ <p> During this period, all allocators were written to the SGI
+ style, and all STL containers expected this interface. This
+ interface had a traits class called <code>_Alloc_traits</code> that
+ attempted to provide more information for compile-time allocation
+ selection and optimization. This traits class had another allocator
+ wrapper, <code>__simple_alloc<T,A></code>, which was a
+ wrapper around another allocator, A, which itself is an allocator
+ for instances of T. But wait, there's more:
+ <code>__allocator<T,A></code> is another adapter. Many of
+ the provided allocator classes were SGI style: such classes can be
+ changed to a conforming interface with this wrapper:
+ <code>__allocator<T, __alloc></code> is thus the same as
+ <code>allocator<T></code>.
+ </p>
+
+ <p> The class <code>std::allocator</code> use the typedef
+ <code>__alloc</code> to select an underlying allocator that
+ satisfied memory allocation requests. The selection of this
+ underlying allocator was not user-configurable.
+ </p>
+
+<h5 class="left">
+ <a name="34allocator"> 3.4 </a>
+</h5>
+ <p> For this and later releases, the only allocator interface that
+ is support is the standard C++ interface. As such, all STL
+ containers have been adjusted, and all external allocators have
+ been modified to support this change. Because of this,
+ <code>__simple_alloc, __allocator, __alloc, </code> and <code>
+ _Alloc_traits</code> have all been removed.
+ </p>
+
+ <p> The class <code>std::allocator</code> just has typedef,
+ constructor, and rebind members. It inherits from one of the
+ high-speed extension allocators, covered below. Thus, all
+ allocation and deallocation depends on the base class.
+ </p>
+
+ <p> The base class that <code>std::allocator</code> is derived from
+ is not user-configurable.
+ </p>
+
+<h5 class="left">
+ <a name="benchmarks"> How the default allocation strategy is selected.</a>
+</h5>
+ <p> It's difficult to pick an allocation strategy that will provide
+ maximum utility, without excessively penalizing some behavior. In
+ fact, it's difficult just deciding which typical actions to measure
+ for speed.
+ </p>
+
+ <p> Three synthetic benchmarks have been created that provide data
+ that is used to compare different C++ allocators. These tests are:
+ </p>
+
+ <ul>
+ <li>Insertion. Over multiple iterations, various STL container
+ objects have elements inserted to some maximum amount. A variety
+ of allocators are tested.
+ Test source <a
+ href="http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc%2b%2b-v3/testsuite/performance/20_util/allocator/insert.cc?only_with_tag=MAIN">here.</a>
+ </li>
+
+ <li>Insertion, clear, and re-insertion in a multi-threaded
+ environment. Over multiple iterations, several threads are
+ started that insert elements into a STL container, then assign a
+ null instance of the same type to clear memory, and then
+ re-insert the same number of elements. Several STL containers and
+ multiple allocators are tested. This test shows the ability of
+ the allocator to reclaim memory on a pre-thread basis, as well as
+ measuring thread contention for memory resources.
+ Test source
+ <a href="http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc%2b%2b-v3/testsuite/performance/20_util/allocator/insert_insert.cc">
+ here.</a>
+ </li>
+
+ <li>A threaded producer/consumer model.
+ Test source
+ <a href="http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc%2b%2b-v3/testsuite/performance/20_util/allocator/producer_consumer.cc">
+ here.</a>
+ </li>
+ </ul>
+
+<h5 class="left">
+ <a name="forcenew"> Disabling memory caching.</a>
+</h5>
+ <p> In use, <code>std::allocator</code> may allocate and deallocate
+ using implementation-specified strategies and heuristics. Because of
+ this, every call to an allocator object's <code> allocate</code>
+ member function may not actually call the global operator new. This
+ situation is also duplicated for calls to the <code>
+ deallocate</code> member function.
+ </p>
+
+ <p> This can be confusing.
+ </p>
+
+ <p> In particular, this can make debugging memory errors more
+ difficult, especially when using third party tools like valgrind or
+ debug versions of <code> new</code>.
+ </p>
+
+ <p> There are various ways to solve this problem. One would be to
+ use a custom allocator that just called operators <code> new
+ </code> and <code> delete</code> directly, for every
+ allocation. (See include/ext/new_allocator.h, for instance.)
+ However, that option would involve changing source code to use the a
+ non-default allocator. Another option is to force the default
+ allocator to remove caching and pools, and to directly allocate
+ with every call of <code> allocate</code> and directly deallocate
+ with every call of <code> deallocate</code>, regardless of
+ efficiency. As it turns out, this last option is available,
+ although the exact mechanism has evolved with time.
+ </p>
+
+ <p> For GCC releases from 2.95 through the 3.1 series, defining
+ <code>__USE_MALLOC</code> on the gcc command line would change the
+ default allocation strategy to instead use <code> malloc</code> and
+ <code> free</code>. See
+ <a href="../23_containers/howto.html#3">this note</a>
+ for details as to why this was something needing improvement.
+ </p>
+
+ <p>Starting with GCC 3.2, and continued in the 3.3 series, to
+ globally disable memory caching within the library for the
+ default allocator, merely set GLIBCPP_FORCE_NEW (at this time,
+ with any value) in the system's environment before running the
+ program. If your program crashes with GLIBCPP_FORCE_NEW in the
+ environment, it likely means that you linked against objects
+ built against the older library. Code to support this extension
+ is fully compatible with 3.2 code if GLIBCPP_FORCE_NEW is not in
+ the environment.
+ </p>
+
+ <p> As it turns out, the 3.4 code base continues to use this
+ mechanism, only the environment variable has been changed to
+ GLIBCXX_FORCE_NEW.
+ </p>
+
+<h3 class="left">
+ <a name="ext allocators">Other allocators</a>
+</h3>
+ <p> Several other allocators are provided as part of this
+ implementation. The location of the extension allocators and their
+ names have changed, but in all cases, functionality is
+ equivalent. Starting with gcc-3.4, all extension allocators are
+ standard style. Before this point, SGI style was the norm. Because of
+ this, the number of template arguments also changed. Here's a simple
+ chart to track the changes.
+ </p>
+
+<table title="extension allocators" border="1">
+ <tr>
+ <th>Allocator (3.4)</th>
+ <th>Header (3.4)</th>
+ <th>Allocator (3.[0-3])</th>
+ <th>Header (3.[0-3])</th>
+ </tr>
+ <tr>
+ <td>__gnu_cxx::new_allocator<T></td>
+ <td><ext/new_allocator.h></td>
+ <td>std::__new_alloc</td>
+ <td><memory></td>
+ </tr>
+ <tr>
+ <td>__gnu_cxx::malloc_allocator<T></td>
+ <td><ext/malloc_allocator.h></td>
+ <td>std::__malloc_alloc_template<int></td>
+ <td><memory></td>
+ </tr>
+ <tr>
+ <td>__gnu_cxx::debug_allocator<T></td>
+ <td><ext/debug_allocator.h></td>
+ <td>std::debug_alloc<T></td>
+ <td><memory></td>
+ </tr>
+ <tr>
+ <td>__gnu_cxx::__pool_alloc<bool, int></td>
+ <td><ext/pool_allocator.h></td>
+ <td>std::__default_alloc_template<bool,int></td>
+ <td><memory></td>
+ </tr>
+ <tr>
+ <td>__gnu_cxx::__mt_alloc<T></td>
+ <td><ext/mt_allocator.h></td>
+ <td></td>
+ <td></td>
+ </tr>
+</table>
+
+ <p>More details on each of these allocators follows. </p>
+ <ul>
+ <li><code>new_allocator</code>
+ <p>Simply wraps <code>::operator new</code>
+ and <code>::operator delete</code>.
+ </p>
+ </li>
+ <li><code>malloc_allocator</code>
+ <p>Simply wraps
+ <code>malloc</code> and <code>free</code>. There is also a hook
+ for an out-of-memory handler (for new/delete this is taken care of
+ elsewhere).
+ </p>
+ </li>
+ <li><code>debug_allocator</code>
+ <p> A wrapper around an
+ arbitrary allocator A. It passes on slightly increased size
+ requests to A, and uses the extra memory to store size information.
+ When a pointer is passed to <code>deallocate()</code>, the stored
+ size is checked, and assert() is used to guarantee they match.
+ </p>
+ </li>
+ <li><code>__pool_alloc</code>
+ <p> A high-performance, single pool allocator. The reusable
+ memory is shared among identical instantiations of this type.
+ It calls through <code>::operator new</code> to obtain new memory
+ when its lists run out. If a client container requests a block
+ larger than a certain threshold size, then the pool is bypassed,
+ and the allocate/deallocate request is passed to
+ <code>::operator new</code> directly. </p>
+
+ <p> This class take a boolean template parameter, called
+ <code>thr</code>, and an integer template parameter, called
+ <code>inst</code>.
+ </p>
+ <p>The <code>inst</code> number is used to track additional memory
+ pools. The point of the number is to allow multiple
+ instantiations of the classes without changing the semantics at
+ all. All three of
+ </p>
+
+ <pre>
+ typedef __pool_alloc<true,0> normal;
+ typedef __pool_alloc<true,1> private;
+ typedef __pool_alloc<true,42> also_private;</pre>
+ <p>behave exactly the same way. However, the memory pool for each type
+ (and remember that different instantiations result in different types)
+ remains separate.
+ </p>
+ <p>The library uses <strong>0</strong> in all its instantiations. If you
+ wish to keep separate free lists for a particular purpose, use a
+ different number.
+ </p>
+ <p>The <code>thr</code> boolean determines whether the pool should
+ be manipulated atomically or not. When thr=true, the allocator
+ is is threadsafe, while thr=false, and is slightly faster but
+ unsafe for multiple threads.
+ </p>
+ <p>(Note that the GCC thread abstraction layer allows us to provide safe
+ zero-overhead stubs for the threading routines, if threads were
+ disabled at configuration time.)
+ </p>
+
+ </li>
+
+ <li><code>__mt_alloc</code>
+ <p>A high-performance
+ fixed-size allocator. It has its own documentation, found <a
+ href="../ext/mt_allocator.html">here</a>.
+ </p>
+ </li>
+ </ul>
+
+
+<h3 class="left">
+ <a name="using custom allocators">Using a specific allocator</a>
+</h3>
+ <p>You can specify different memory management schemes on a
+ per-container basis, by overriding the default
+ <code>Allocator</code> template parameter. For example, an easy
+ (but non-portable) method of specifying that only malloc/free
+ should be used instead of the default node allocator is:
+ </p>
+ <pre>
+ std::list <int, __gnu_cxx::malloc_allocator<int> > malloc_list;</pre>
+ Likewise, a debugging form of whichever allocator is currently in use:
+ <pre>
+ std::deque <int, __gnu_cxx::debug_allocator<std::allocator<int> > > debug_deque;</pre>
+
+
+<h3 class="left">
+ <a name="custom allocators">Writing custom allocators</a>
+</h3>
+ <p> Writing a portable C++ allocator would dictate that the
+ interface would look much like the one specified for <code>
+ std::allocator</code>. Additional member functions, but not
+ subtractions, would be permissible.
+ </p>
+
+ <p> Probably the best place to start would be to copy one of the
+ extension allocators already shipped with libstdc++: say, <code>
+ new_allocator </code>.
+ </p>
+
+
+<h3 class="left">
+ <a name="biblio">Bibliography / Further Reading</a>
+</h3>
+ <p>
+ ISO/IEC 14882:1998 Programming languages - C++ [20.4 Memory]
+ </p>
+
+ <p>
+ Austern, Matt, C/C++ Users Journal.
+ <a href="http://www.cuj.com/documents/s=8000/cujcexp1812austern/">The Standard Librarian: What Are Allocators Good
+ For?</a>
+ </p>
+
+ <p>
+ Berger, Emery,
+ <a href="http://www.cs.umass.edu/~emery/hoard/"> The Hoard memory allocator </a>
+ </p>
+
+ <p>
+ Berger, Emery with Ben Zorn & Kathryn McKinley, OOPSLA 2002
+ <a href="http://www.cs.umass.edu/~emery/pubs/berger-oopsla2002.pdf">Reconsidering Custom Memory Allocation</a>
+ </p>
+
+ <p>
+ Kreft, Klaus and Angelika Langer, C++ Report, June 1998
+ <a href="http://www.langer.camelot.de/Articles/C++Report/Allocators/Allocators.html">Allocator Types</a>
+ </p>
+
+ <p>
+ Stroustrup, Bjarne, 19.4 Allocators, The C++ Programming
+ Language, Special Edition, Addison Wesley, Inc. 2000
+ </p>
+
+ <p>
+ Yen, Felix, <a href="http://home.earthlink.net/~brimar/yalloc/">Yalloc: A Recycling C++ Allocator</a>
+ </p>
+
+<hr />
+<p>Return <a href="#top">to the top of the page</a> or
+ <a href="http://gcc.gnu.org/libstdc++/">to the libstdc++ homepage</a>.
+</p>
+
+
+<!-- ####################################################### -->
+
+<hr />
+<p class="fineprint"><em>
+See <a href="../17_intro/license.html">license.html</a> for copying conditions.
+Comments and suggestions are welcome, and may be sent to
+<a href="mailto:libstdc++@gcc.gnu.org">the libstdc++ mailing list</a>.
+</em></p>
+
+
+</body>
+</html>
<hr />
<h2><a name="5">Memory allocators</a></h2>
<p>The available free store ("heap") management classes are
- described <a href="../ext/howto.html">here</a>.
+ described <a href="allocator.html">here</a>.
</p>
<p>Return <a href="#top">to top of page</a> or
<a href="../faq/index.html">to the FAQ</a>.
<li><a href="18_support/howto.html#2">Implementation properties</a></li>
<li><a href="18_support/howto.html#3">Start and Termination</a></li>
<li><a href="18_support/howto.html#4">Verbose <code>terminate</code></a></li>
- <li><a href="18_support/howto.html#6">Dynamic memory management</a></li>
- <li><a href="18_support/howto.html#7">RTTI, the ABI, and demangling</a></li>
+ <li><a href="18_support/howto.html#5">Dynamic memory management</a></li>
+ <li><a href="18_support/howto.html#6">RTTI, the ABI, and demangling</a></li>
</ul>
</li>
<li><a href="20_util/howto.html#2"><code>auto_ptr</code> inside container classes</a></li>
<li><a href="20_util/howto.html#3">Functors</a></li>
<li><a href="20_util/howto.html#4">Pairs</a></li>
- <li><a href="20_util/howto.html#5">Memory allocators</a></li>
+ <li><a href="20_util/allocator.html">Allocators and allocation</a></li>
</ul>
</li>
<ul>
<li><a href="ext/howto.html#1">Ropes and trees and hashes, oh my!</a></li>
<li><a href="ext/howto.html#2">Added members and types</a></li>
- <li><a href="ext/howto.html#3">Allocators (versions 3.0, 3.1, 3.2, 3.3)</a></li>
- <li><a href="ext/howto.html#6">Allocators (version 3.4)</a></li>
+ <li><a href="ext/mt_allocator.html"><code>__mt_alloc</code> </a></li>
<li><a href="ext/howto.html#4">Compile-time checks</a></li>
<li><a href="ext/howto.html#5">LWG Issues</a></li>
<li><a href="ext/../18_support/howto.html#5">Demangling</a></li>
<ul>
<li><a href="#1">Ropes and trees and hashes, oh my!</a></li>
<li><a href="#2">Added members and types</a></li>
- <li><a href="#3">Allocators (versions 3.0, 3.1, 3.2, 3.3)</a></li>
- <li><a href="#6">Allocators (version 3.4)</a></li>
+ <li><a href="mt_allocator.html"><code>__mt_alloc</code> </a></li>
<li><a href="#4">Compile-time checks</a></li>
<li><a href="#5">LWG Issues</a></li>
<li><a href="../18_support/howto.html#5">Demangling</a></li>
</p>
<hr />
-<h2><a name="3">Allocators (versions 3.0, 3.1, 3.2, 3.3)</a></h2>
- <p>Thread-safety, space efficiency, high speed, portability... this is a
- mess. Where to begin?
- </p>
- <h3>The Rules</h3>
- <p>The C++ standard only gives a few directives in this area:
- </p>
- <ul>
- <li>When you add elements to a container, and the container must allocate
- more memory to hold them, the container makes the request via its
- <code>Allocator</code> template parameter. This includes adding
- char's to the string class, which acts as a regular STL container
- in this respect.
- </li>
- <li>The default <code>Allocator</code> of every container-of-T is
- <code>std::allocator<T></code>.
- </li>
- <li>The interface of the <code>allocator<T></code> class is
- extremely simple. It has about 20 public declarations (nested
- typedefs, member functions, etc), but the two which concern us most
- are:
- <pre>
- T* allocate (size_type n, const void* hint = 0);
- void deallocate (T* p, size_type n);</pre>
- (This is a simplicifcation; the real signatures use nested typedefs.)
- The <code>"n"</code> arguments in both those functions is a
- <em>count</em> of the number of T's to allocate space for,
- <em>not their total size</em>.
- </li>
- <li>"The storage is obtained by calling
- <code>::operator new(size_t)</code>, but it is unspecified when or
- how often this function is called. The use of <code>hint</code>
- is unspecified, but intended as an aid to locality if an
- implementation so desires." [20.4.1.1]/6
- </li>
- </ul>
- <h3>Problems and Possibilities</h3>
- <p>The easiest way of fulfilling the requirements is to call operator new
- each time a container needs memory, and to call operator delete each
- time the container releases memory. <strong>BUT</strong>
- <a href="http://gcc.gnu.org/ml/libstdc++/2001-05/msg00105.html">this
- method is horribly slow</a>.
- </p>
- <p>Or we can keep old memory around, and reuse it in a pool to save time.
- The old libstdc++-v2 used a memory pool, and so do we. As of 3.0,
- <a href="http://gcc.gnu.org/ml/libstdc++/2001-05/msg00136.html">it's
- on by default</a>. The pool is shared among all the containers in the
- program: when your program's std::vector<int> gets cut in half
- and frees a bunch of its storage, that memory can be reused by the
- private std::list<WonkyWidget> brought in from a KDE library
- that you linked against. And we don't have to call operators new and
- delete to pass the memory on, either, which is a speed bonus.
- <strong>BUT</strong>...
- </p>
- <p>What about threads? No problem: in a threadsafe environment, the
- memory pool is manipulated atomically, so you can grow a container in
- one thread and shrink it in another, etc. <strong>BUT</strong> what
- if threads in libstdc++-v3 aren't set up properly?
- <a href="../faq/index.html#5_6">That's been answered already</a>.
- </p>
- <p><strong>BUT</strong> what if you want to use your own allocator? What
- if you plan on using a runtime-loadable version of malloc() which uses
- shared telepathic anonymous mmap'd sections serializable over a
- network, so that memory requests <em>should</em> go through malloc?
- And what if you need to debug it?
- </p>
- <p>Well then:
- </p>
- <h3>Available allocators in namespace std</h3>
- <p>First I'll describe the situation as it exists for the code which
- was released in GCC 3.1 and 3.2. Then I'll describe the differences
- for 3.0. The allocator classes also have source documentation,
- which is described <a href="../documentation.html#4">here</a> (you
- will need to retrieve the maintainer-level docs, as almost none of
- these entities are in the ISO standard).
- </p>
- <p>As a general rule of thumb, users are not allowed to use names which
- begin with an underscore. This means that to be portable between
- compilers, none of the following may be used in your program directly.
- (If you decide to be unportable, then you're free do do what you want,
- but it's not our fault if stuff breaks.) They are presented here for
- information for maintainers and contributors in addition to users.
- </p>
- <p>These classes are always available:
- </p>
- <ul>
- <li><code>__new_alloc</code> simply wraps <code>::operator new</code>
- and <code>::operator delete</code>.
- </li>
- <li><code>__malloc_alloc_template<int inst></code> simply wraps
- <code>malloc</code> and <code>free</code>. There is also a hook
- for an out-of-memory handler (for new/delete this is taken care of
- elsewhere). The <code>inst</code> parameter is described below.
- This class was called <code>malloc_alloc</code> in earlier versions.
- </li>
- <li><code>allocator<T></code> has already been described; it is
- The Standard Allocator for instances of T. It uses the internal
- <code>__alloc</code> typedef (see below) to satisy its requests.
- </li>
- <li><code>__simple_alloc<T,A></code> is a wrapper around another
- allocator, A, which itself is an allocator for instances of T.
- This is primarily used in an internal "allocator traits"
- class which helps encapsulate the different styles of allocators.
- </li>
- <li><code>__debug_alloc<A></code> is also a wrapper around an
- arbitrary allocator A. It passes on slightly increased size
- requests to A, and uses the extra memory to store size information.
- When a pointer is passed to <code>deallocate()</code>, the stored
- size is checked, and assert() is used to guarantee they match.
- </li>
- <li><code>__allocator<T,A></code> is an adaptor. Many of these
- allocator classes have a consistent yet non-standard interface.
- Such classes can be changed to a conforming interface with this
- wrapper: <code>__allocator<T, __alloc></code> is thus the
- same as <code>allocator<T></code>.
- </li>
- </ul>
- <p>Normally,
- <code> __default_alloc_template<bool thr, int inst> </code>
- is also available. This is the high-speed pool, called the default
- node allocator. The reusable memory is shared among identical
- instantiations of
- this type. It calls through <code>__new_alloc</code> to obtain
- new memory when its lists run out. If a client container requests a
- block larger than a certain threshold size, then the pool is bypassed,
- and the allocate/deallocate request is passed to
- <code>__new_alloc</code> directly.
- </p>
- <p>Its <code>inst</code> parameter is described below. The
- <code>thr</code> boolean determines whether the pool should be
- manipulated atomically or not. Two typedefs are provided:
- <code>__alloc</code> is defined as this node allocator with thr=true,
- and therefore is threadsafe, while <code>__single_client_alloc</code>
- defines thr=false, and is slightly faster but unsafe for multiple
- threads.
- </p>
- <p>(Note that the GCC thread abstraction layer allows us to provide safe
- zero-overhead stubs for the threading routines, if threads were
- disabled at configuration time. In this situation,
- <code>__alloc</code> should not be noticably slower than
- <code>__single_client_alloc</code>.)
- </p>
- <p>[Another threadsafe allocator where each thread keeps its own free
- list, so that no locking is needed, might be described here.]
- </p>
- <h3>A cannon to swat a fly:<code> __USE_MALLOC</code></h3>
- <p>If you've already read <a href="../23_containers/howto.html#3">this
- advice</a> but still think you remember how to use this macro from
- SGI STL days. We have removed it in gcc 3.3. See next section
- for the new way to get the same effect.
- </p>
- <h3>Globally disabling memory caching:<code> GLIBCXX_FORCE_NEW</code></h3>
- <p>Starting with gcc 3.3, if you want to globally disable memory
- caching within the library for the default allocator (i.e.
- the one you get for all library objects when you do not specify
- which one to use), merely set GLIBCXX_FORCE_NEW (at this time,
- with any value) into your environment before running the
- program. You will obtain a similar effect without having to
- recompile your entire program and the entire library (the new
- operator in gcc is a light wrapper around malloc). If your
- program crashes with GLIBCXX_FORCE_NEW in the environment,
- it likely means that you linked against objects built against
- the older library. Code to support this extension is fully
- compatible with 3.2 code if GLIBCXX_FORCE_NEW is not in the
- environment. Prior to GCC 3.4, this variable was spelt
- GLIBCPP_FORCE_NEW.
- </p>
- <h3>Writing your own allocators</h3>
- <p>Depending on your application (a specific program, a generic library,
- etc), allocator classes tend to be one of two styles: "SGI"
- or "standard". See the comments in stl_alloc.h for more
- information on this crucial difference.
- </p>
- <p>At the bottom of that header is a helper type,
- <code>_Alloc_traits</code>, and various specializations of it. This
- allows the container classes to make possible compile-time
- optimizations based on features of the allocator. You should provide
- a specialization of this type for your allocator (doing so takes only
- two or three statements).
- </p>
- <h3>Using non-default allocators</h3>
- <p>You can specify different memory management schemes on a per-container
- basis, by overriding the default <code>Allocator</code> template
- parameter. For example, an easy
- (but nonportable)
- method of specifying that only malloc/free should be used instead of
- the default node allocator is:
- </p>
- <pre>
- std::list <my_type, std::__malloc_alloc_template<0> > my_malloc_based_list;</pre>
- Likewise, a debugging form of whichever allocator is currently in use:
- <pre>
- std::deque <my_type, std::__debug_alloc<std::__alloc> > debug_deque;</pre>
- <h3><code>inst</code></h3>
- <p>The <code>__malloc_alloc_template</code> and
- <code>__default_alloc_template</code> classes take an integer parameter,
- called inst here. This number is completely unused.
- </p>
- <p>The point of the number is to allow multiple instantiations of the
- classes without changing the semantics at all. All three of
- </p>
- <pre>
- typedef __default_alloc_template<true,0> normal;
- typedef __default_alloc_template<true,1> private;
- typedef __default_alloc_template<true,42> also_private;</pre>
- <p>behave exactly the same way. However, the memory pool for each type
- (and remember that different instantiations result in different types)
- remains separate.
- </p>
- <p>The library uses <strong>0</strong> in all its instantiations. If you
- wish to keep separate free lists for a particular purpose, use a
- different number.
- </p>
- <h3>3.0.x</h3>
- <p>For 3.0.x, many of the names were incorrectly <em>not</em> prefixed
- with underscores. So symbols such as "std::single_client_alloc"
- are present. Be very careful to not depend on these names any more
- than you would depend on implementation-only names.
- </p>
- <p>Certain macros like <code>_NOTHREADS</code> and <code>__STL_THREADS</code>
- can affect the 3.0.x allocators. Do not use them. Those macros have
- been completely removed for 3.1.
- </p>
- <p>Return <a href="#top">to top of page</a> or
- <a href="../faq/index.html">to the FAQ</a>.
- </p>
-
-<hr />
-<h2><a name="6">Allocators (version 3.4)</a></h2>
- <p>Changes are coming...
- </p>
- <p>If you plan on writing your own allocators,
- <a href="../documentation.html#4">source documentation</a> is
- available. You'll need to get the "maintainers" collection
- in order to see the helper classes and extra notes.
- </p>
- <p>Return <a href="#top">to top of page</a> or
- <a href="../faq/index.html">to the FAQ</a>.
- </p>
-
-<hr />
<h2><a name="4">Compile-time checks</a></h2>
<p>Currently libstdc++-v3 uses the concept checkers from the Boost
library to perform <a href="../19_diagnostics/howto.html#3">optional
--- /dev/null
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!DOCTYPE html
+ PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
+ "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
+
+<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
+<head>
+ <meta name="AUTHOR" content="Stefan Olsson <stefan@xapa.se>" />
+ <meta name="KEYWORDS" content="c++, libstdc++, g++, allocator, memory" />
+ <meta name="DESCRIPTION" content="Allocators and allocation" />
+ <meta name="GENERATOR" content="emacs and ten fingers" />
+ <title>A fixed-size, multi-thread optimized allocator</title>
+<link rel="StyleSheet" href="../lib3styles.css" type="text/css" />
+<link rel="Copyright" href="../17_intro/license.html" type="text/html" />
+</head>
+<body>
+
+<h1 class="centered"><a name="top">A fixed-size, multi-thread optimized allocator</a></h1>
+
+<p class="fineprint"><em>
+ The latest version of this document is always available at
+ <a href="http://gcc.gnu.org/onlinedocs/libstdc++/ext/mt_allocator.html">
+ http://gcc.gnu.org/onlinedocs/libstdc++/ext/mt_allocator.html</a>.
+</em></p>
+
+<p><em>
+ To the <a href="http://gcc.gnu.org/libstdc++/">libstdc++-v3 homepage</a>.
+</em></p>
+
+<!-- ####################################################### -->
+<hr />
+<h3 class="left">
+ <a name="intro">Introduction</a>
+</h3>
+
+<p> The mt allocator [hereinafter referred to simply as "the
+allocator"] is a fixed size (power of two) allocator that was
+initially developed specifically to suit the needs of multi threaded
+applications [hereinafter referred to as an MT application]. Over time
+the allocator has evolved and been improved in many ways, one of the
+being that it now also does a good job in single threaded applications
+[hereinafter referred to as a ST application]. (Note: In this
+document, when referring to single threaded applications this also
+includes applications that are compiled with gcc without thread
+support enabled. This is accomplished using ifdef's on __GTHREADS)
+</p>
+
+<p>
+The aim of this document is to describe - from a application point of
+view - the "inner workings" of the allocator.
+</p>
+
+
+<h3 class="left">
+ <a name="init">Initialization</a>
+</h3>
+
+<p>
+The static variables (pointers to freelists, tuning parameters etc)
+are initialized to their default values at file scope, i.e.:
+</p>
+
+<pre>
+ template<typename _Tp> size_t
+ __mt_alloc<_Tp>::_S_freelist_headroom = 10;
+</pre>
+
+<p>
+The very first allocate() call will always call the _S_init() function.
+In order to make sure that this function is called exactly once we make use
+of a __gthread_once (with _S_once_mt and _S_init as arguments) call in MT
+applications and check a static bool (_S_initialized) in ST applications.
+</p>
+
+<p>
+The _S_init() function:
+- If the GLIBCXX_FORCE_NEW environment variable is set, it sets the bool
+ _S_force_new to true and then returns. This will cause subsequent calls to
+ allocate() to return memory directly from a new() call, and deallocate will
+ only do a delete() call.
+</p>
+
+<p>
+- If the GLIBCXX_FORCE_NEW environment variable is not set, both ST and MT
+ applications will:
+ - Calculate the number of bins needed. A bin is a specific power of two size
+ of bytes. I.e., by default the allocator will deal with requests of up to
+ 128 bytes (or whatever the value of _S_max_bytes is when _S_init() is
+ called). This means that there will be bins of the following sizes
+ (in bytes): 1, 2, 4, 8, 16, 32, 64, 128.
+
+ - Create the _S_binmap array. All requests are rounded up to the next
+ "large enough" bin. I.e., a request for 29 bytes will cause a block from
+ the "32 byte bin" to be returned to the application. The purpose of
+ _S_binmap is to speed up the process of finding out which bin to use.
+ I.e., the value of _S_binmap[ 29 ] is initialized to 5 (bin 5 = 32 bytes).
+</p>
+<p>
+ - Create the _S_bin array. This array consists of bin_records. There will be
+ as many bin_records in this array as the number of bins that we calculated
+ earlier. I.e., if _S_max_bytes = 128 there will be 8 entries.
+ Each bin_record is then initialized:
+ - bin_record->first = An array of pointers to block_records. There will be
+ as many block_records pointers as there are maximum number of threads
+ (in a ST application there is only 1 thread, in a MT application there
+ are _S_max_threads).
+ This holds the pointer to the first free block for each thread in this
+ bin. I.e., if we would like to know where the first free block of size 32
+ for thread number 3 is we would look this up by: _S_bin[ 5 ].first[ 3 ]
+ - bin_record->last = See above, the only difference being that this points
+ to the last record on the same freelist.
+
+ The above created block_record pointers members are now initialized to
+ their initial values. I.e. _S_bin[ n ].first[ n ] = NULL;
+</p>
+
+<p>
+- Additionally a MT application will:
+ - Create a list of free thread id's. The pointer to the first entry
+ is stored in _S_thread_freelist_first. The reason for this approach is
+ that the __gthread_self() call will not return a value that corresponds to
+ the maximum number of threads allowed but rather a process id number or
+ something else. So what we do is that we create a list of thread_records.
+ This list is _S_max_threads long and each entry holds a size_t thread_id
+ which is initialized to 1, 2, 3, 4, 5 and so on up to _S_max_threads.
+ Each time a thread calls allocate() or deallocate() we call
+ _S_get_thread_id() which looks at the value of _S_thread_key which is a
+ thread local storage pointer. If this is NULL we know that this is a newly
+ created thread and we pop the first entry from this list and saves the
+ pointer to this record in the _S_thread_key variable. The next time
+ we will get the pointer to the thread_record back and we use the
+ thread_record->thread_id as identification. I.e., the first thread that
+ calls allocate will get the first record in this list and thus be thread
+ number 1 and will then find the pointer to its first free 32 byte block
+ in _S_bin[ 5 ].first[ 1 ]
+ When we create the _S_thread_key we also define a destructor
+ (_S_thread_key_destr) which means that when the thread dies, this
+ thread_record is returned to the front of this list and the thread id
+ can then be reused if a new thread is created.
+ This list is protected by a mutex (_S_thread_freelist_mutex) which is only
+ locked when records are removed/added to the list.
+</p>
+<p>
+ - Initialize the free and used counters of each bin_record:
+ - bin_record->free = An array of size_t. This keeps track of the number
+ of blocks on a specific thread's freelist in each bin. I.e., if a thread
+ has 12 32-byte blocks on it's freelists and allocates one of these, this
+ counter would be decreased to 11.
+
+ - bin_record->used = An array of size_t. This keeps track of the number
+ of blocks currently in use of this size by this thread. I.e., if a thread
+ has made 678 requests (and no deallocations...) of 32-byte blocks this
+ counter will read 678.
+
+ The above created arrays are now initialized with their initial values.
+ I.e. _S_bin[ n ].free[ n ] = 0;
+</p>
+<p>
+ - Initialize the mutex of each bin_record:
+ The bin_record->mutex is used to protect the global freelist. This concept
+ of a global freelist is explained in more detail in the section
+ "A multi threaded example", but basically this mutex is locked whenever
+ a block of memory is retrieved or returned to the global freelist for this
+ specific bin. This only occurs when a number of blocks are grabbed from the
+ global list to a thread specific list or when a thread decides to return
+ some blocks to the global freelist.
+</p>
+
+<h3 class="left">
+ <a name="st_example">A single threaded example (and a primer for the multi threaded example!)</a>
+</h3>
+
+<p>
+Let's start by describing how the data on a freelist is laid out in memory.
+This is the first two blocks in freelist for thread id 3 in bin 3 (8 bytes):
+</p>
+<pre>
++----------------+
+| next* ---------|--+ (_S_bin[ 3 ].first[ 3 ] points here)
+| | |
+| | |
+| | |
++----------------+ |
+| thread_id = 3 | |
+| | |
+| | |
+| | |
++----------------+ |
+| DATA | | (A pointer to here is what is returned to the
+| | | the application when needed)
+| | |
+| | |
+| | |
+| | |
+| | |
+| | |
++----------------+ |
++----------------+ |
+| next* |<-+ (If next == NULL it's the last one on the list and
+| | then the _S_bin[ 3 ].last[ 3 ] pointer points to
+| | here as well)
+| |
++----------------+
+| thread_id = 3 |
+| |
+| |
+| |
++----------------+
+| DATA |
+| |
+| |
+| |
+| |
+| |
+| |
+| |
++----------------+
+</pre>
+
+<p>
+With this in mind we simplify things a bit for a while and say that there is
+only one thread (a ST application). In this case all operations are made to
+what is referred to as the global pool - thread id 0 (No thread may be
+assigned this id since they span from 1 to _S_max_threads in a MT application).
+</p>
+<p>
+When the application requests memory (calling allocate()) we first look at the
+requested size and if this is > _S_max_bytes we call new() directly and return.
+</p>
+<p>
+If the requested size is within limits we start by finding out from which
+bin we should serve this request by looking in _S_binmap.
+</p>
+<p>
+A quick look at _S_bin[ bin ].first[ 0 ] tells us if there are any blocks of
+this size on the freelist (0). If this is not NULL - fine, just remove the
+block that _S_bin[ bin ].first[ 0 ] points to from the list,
+update _S_bin[ bin ].first[ 0 ] and return a pointer to that blocks data.
+</p>
+<p>
+If the freelist is empty (the pointer is NULL) we must get memory from the
+system and build us a freelist within this memory. All requests for new memory
+is made in chunks of _S_chunk_size. Knowing the size of a block_record and
+the bytes that this bin stores we then calculate how many blocks we can create
+within this chunk, build the list, remove the first block, update the pointers
+(_S_bin[ bin ].first[ 0 ] and _S_bin[ bin ].last[ 0 ]) and return a pointer
+to that blocks data.
+</p>
+
+<p>
+Deallocation is equally simple; the pointer is casted back to a block_record
+pointer, lookup which bin to use based on the size, add the block to the end
+of the global freelist (with the next pointer set to NULL) and update the
+pointers as needed (_S_bin[ bin ].first[ 0 ] and _S_bin[ bin ].last[ 0 ]).
+</p>
+
+<h3 class="left">
+ <a name="mt_example">A multi threaded example</a>
+</h3>
+
+<p>
+In the ST example we never used the thread_id variable present in each block.
+Let's start by explaining the purpose of this in a MT application.
+</p>
+
+<p>
+The concept of "ownership" was introduced since many MT applications
+allocate and deallocate memory to shared containers from different
+threads (such as a cache shared amongst all threads). This introduces
+a problem if the allocator only returns memory to the current threads
+freelist (I.e., there might be one thread doing all the allocation and
+thus obtaining ever more memory from the system and another thread
+that is getting a longer and longer freelist - this will in the end
+consume all available memory).
+</p>
+
+<p>
+Each time a block is moved from the global list (where ownership is
+irrelevant), to a threads freelist (or when a new freelist is built
+from a chunk directly onto a threads freelist or when a deallocation
+occurs on a block which was not allocated by the same thread id as the
+one doing the deallocation) the thread id is set to the current one.
+</p>
+
+<p>
+What's the use? Well, when a deallocation occurs we can now look at
+the thread id and find out if it was allocated by another thread id
+and decrease the used counter of that thread instead, thus keeping the
+free and used counters correct. And keeping the free and used counters
+corrects is very important since the relationship between these two
+variables decides if memory should be returned to the global pool or
+not when a deallocation occurs.
+</p>
+
+<p>
+When the application requests memory (calling allocate()) we first
+look at the requested size and if this is > _S_max_bytes we call new()
+directly and return.
+</p>
+
+<p>
+If the requested size is within limits we start by finding out from which
+bin we should serve this request by looking in _S_binmap.
+</p>
+
+<p>
+A call to _S_get_thread_id() returns the thread id for the calling thread
+(and if no value has been set in _S_thread_key, a new id is assigned and
+returned).
+</p>
+
+<p>
+A quick look at _S_bin[ bin ].first[ thread_id ] tells us if there are
+any blocks of this size on the current threads freelist. If this is
+not NULL - fine, just remove the block that _S_bin[ bin ].first[
+thread_id ] points to from the list, update _S_bin[ bin ].first[
+thread_id ], update the free and used counters and return a pointer to
+that blocks data.
+</p>
+
+<p>
+If the freelist is empty (the pointer is NULL) we start by looking at
+the global freelist (0). If there are blocks available on the global
+freelist we lock this bins mutex and move up to block_count (the
+number of blocks of this bins size that will fit into a _S_chunk_size)
+or until end of list - whatever comes first - to the current threads
+freelist and at the same time change the thread_id ownership and
+update the counters and pointers. When the bins mutex has been
+unlocked, we remove the block that _S_bin[ bin ].first[ thread_id ]
+points to from the list, update _S_bin[ bin ].first[ thread_id ],
+update the free and used counters, and return a pointer to that blocks
+data.
+</p>
+
+<p>
+The reason that the number of blocks moved to the current threads
+freelist is limited to block_count is to minimize the chance that a
+subsequent deallocate() call will return the excess blocks to the
+global freelist (based on the _S_freelist_headroom calculation, see
+below).
+</p>
+
+<p>
+However if there isn't any memory on the global pool we need to get
+memory from the system - this is done in exactly the same way as in a
+single threaded application with one major difference; the list built
+in the newly allocated memory (of _S_chunk_size size) is added to the
+current threads freelist instead of to the global.
+</p>
+
+<p>
+The basic process of a deallocation call is simple: always add the
+block to the end of the current threads freelist and update the
+counters and pointers (as described earlier with the specific check of
+ownership that causes the used counter of the thread that originally
+allocated the block to be decreased instead of the current threads
+counter).
+</p>
+
+<p>
+And here comes the free and used counters to service. Each time a
+deallocation() call is made, the length of the current threads
+freelist is compared to the amount memory in use by this thread.
+</p>
+
+<p>
+Let's go back to the example of an application that has one thread
+that does all the allocations and one that deallocates. Both these
+threads use say 516 32-byte blocks that was allocated during thread
+creation for example. Their used counters will both say 516 at this
+point. The allocation thread now grabs 1000 32-byte blocks and puts
+them in a shared container. The used counter for this thread is now
+1516.
+</p>
+
+<p>
+The deallocation thread now deallocates 500 of these blocks. For each
+deallocation made the used counter of the allocating thread is
+decreased and the freelist of the deallocation thread gets longer and
+longer. But the calculation made in deallocate() will limit the length
+of the freelist in the deallocation thread to _S_freelist_headroom %
+of it's used counter. In this case, when the freelist (given that the
+_S_freelist_headroom is at it's default value of 10%) exceeds 52
+(516/10) blocks will be returned to the global pool where the
+allocating thread may pick them up and reuse them.
+</p>
+
+<p>
+In order to reduce lock contention (since this requires this bins
+mutex to be locked) this operation is also made in chunks of blocks
+(just like when chunks of blocks are moved from the global freelist to
+a threads freelist mentioned above). The "formula" used can probably
+be improved to further reduce the risk of blocks being "bounced back
+and forth" between freelists.
+</p>
+
+<hr />
+<p>Return <a href="#top">to the top of the page</a> or
+ <a href="http://gcc.gnu.org/libstdc++/">to the libstdc++ homepage</a>.
+</p>
+
+
+<!-- ####################################################### -->
+
+<hr />
+<p class="fineprint"><em>
+See <a href="../17_intro/license.html">license.html</a> for copying conditions.
+Comments and suggestions are welcome, and may be sent to
+<a href="mailto:libstdc++@gcc.gnu.org">the libstdc++ mailing list</a>.
+</em></p>
+
+
+</body>
+</html>
projects / platform / upstream / gcc.git / commitdiff