tri11paragon
/
Insane_DNS
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
Insane_DNS/libraries/asio-1.28.1/doc/asio/overview/implementation.html

554 lines
44 KiB
HTML
Raw Normal View History

can extract domain names
2023-10-24 21:59:30 -04:00
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<title>Platform-Specific Implementation Notes</title>
<link rel="stylesheet" href="../../boostbook.css" type="text/css">
<meta name="generator" content="DocBook XSL Stylesheets V1.75.2">
<link rel="home" href="../../index.html" title="Asio">
<link rel="up" href="../overview.html" title="Overview">
<link rel="prev" href="cpp2011/chrono.html" title="Chrono">
<link rel="next" href="../using.html" title="Using, Building, and Configuring Asio">
<meta name="viewport" content="width=device-width, initial-scale=1">
</head>
<body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF">
<table cellpadding="2" width="100%"><tr><td valign="top"><img alt="asio C++ library" width="250" height="60" src="../../asio.png"></td></tr></table>
<hr>
<div class="spirit-nav">
<a accesskey="p" href="cpp2011/chrono.html"><img src="../../prev.png" alt="Prev"></a><a accesskey="u" href="../overview.html"><img src="../../up.png" alt="Up"></a><a accesskey="h" href="../../index.html"><img src="../../home.png" alt="Home"></a><a accesskey="n" href="../using.html"><img src="../../next.png" alt="Next"></a>
</div>
<div class="section">
<div class="titlepage"><div><div><h3 class="title">
<a name="asio.overview.implementation"></a><a class="link" href="implementation.html" title="Platform-Specific Implementation Notes">Platform-Specific Implementation
Notes</a>
</h3></div></div></div>
<p>
This section lists platform-specific implementation details, such as the
default demultiplexing mechanism, the number of threads created internally,
and when threads are created.
</p>
<h5>
<a name="asio.overview.implementation.h0"></a>
<span><a name="asio.overview.implementation.linux_kernel_2_4"></a></span><a class="link" href="implementation.html#asio.overview.implementation.linux_kernel_2_4">Linux
Kernel 2.4</a>
</h5>
<p>
Demultiplexing mechanism:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
Uses <code class="computeroutput"><span class="identifier">select</span></code> for demultiplexing.
This means that the number of file descriptors in the process cannot
be permitted to exceed <code class="computeroutput"><span class="identifier">FD_SETSIZE</span></code>.
</li></ul></div>
<p>
Threads:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
Demultiplexing using <code class="computeroutput"><span class="identifier">select</span></code>
is performed in one of the threads that calls <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run_one</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll_one</span><span class="special">()</span></code>.
</li>
<li class="listitem">
An additional thread per <code class="computeroutput"><span class="identifier">io_context</span></code>
is used to emulate asynchronous host resolution. This thread is created
on the first call to either <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">tcp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">udp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code>.
</li>
</ul></div>
<p>
Scatter-Gather:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
At most <code class="computeroutput"><span class="identifier">min</span><span class="special">(</span><span class="number">64</span><span class="special">,</span><span class="identifier">IOV_MAX</span><span class="special">)</span></code> buffers may be transferred in a single
operation.
</li></ul></div>
<h5>
<a name="asio.overview.implementation.h1"></a>
<span><a name="asio.overview.implementation.linux_kernel_2_6"></a></span><a class="link" href="implementation.html#asio.overview.implementation.linux_kernel_2_6">Linux
Kernel 2.6</a>
</h5>
<p>
Demultiplexing mechanism:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
Uses <code class="computeroutput"><span class="identifier">epoll</span></code> for demultiplexing.
</li></ul></div>
<p>
Threads:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
Demultiplexing using <code class="computeroutput"><span class="identifier">epoll</span></code>
is performed in one of the threads that calls <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run_one</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll_one</span><span class="special">()</span></code>.
</li>
<li class="listitem">
An additional thread per <code class="computeroutput"><span class="identifier">io_context</span></code>
is used to emulate asynchronous host resolution. This thread is created
on the first call to either <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">tcp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">udp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code>.
</li>
</ul></div>
<p>
Scatter-Gather:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
At most <code class="computeroutput"><span class="identifier">min</span><span class="special">(</span><span class="number">64</span><span class="special">,</span><span class="identifier">IOV_MAX</span><span class="special">)</span></code> buffers may be transferred in a single
operation.
</li></ul></div>
<h5>
<a name="asio.overview.implementation.h2"></a>
<span><a name="asio.overview.implementation.linux_kernel_5_10"></a></span><a class="link" href="implementation.html#asio.overview.implementation.linux_kernel_5_10">Linux
Kernel 5.10</a>
</h5>
<p>
Demultiplexing mechanism:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
If <code class="computeroutput"><span class="identifier">ASIO_HAS_IO_URING</span></code>
is defined, uses <code class="computeroutput"><span class="identifier">io_uring</span></code>
for file-related asynchonous operations.
</li>
<li class="listitem">
Uses <code class="computeroutput"><span class="identifier">epoll</span></code> for demultiplexing
other event sources.
</li>
<li class="listitem">
Optionally uses <code class="computeroutput"><span class="identifier">io_uring</span></code>
for all asynchronous operations if, in addition to <code class="computeroutput"><span class="identifier">ASIO_HAS_IO_URING</span></code>,
<code class="computeroutput"><span class="identifier">ASIO_DISABLE_EPOLL</span></code> is
defined to disable <code class="computeroutput"><span class="identifier">epoll</span></code>.
</li>
</ul></div>
<p>
Threads:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
Demultiplexing using <code class="computeroutput"><span class="identifier">epoll</span></code>
and <code class="computeroutput"><span class="identifier">io_uring</span></code> is performed
in one of the threads that calls <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run_one</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll_one</span><span class="special">()</span></code>.
</li>
<li class="listitem">
An additional thread per <code class="computeroutput"><span class="identifier">io_context</span></code>
is used to emulate asynchronous host resolution. This thread is created
on the first call to either <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">tcp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">udp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code>.
</li>
</ul></div>
<p>
Scatter-Gather:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
At most <code class="computeroutput"><span class="identifier">min</span><span class="special">(</span><span class="number">64</span><span class="special">,</span><span class="identifier">IOV_MAX</span><span class="special">)</span></code> buffers may be transferred in a single
operation.
</li></ul></div>
<h5>
<a name="asio.overview.implementation.h3"></a>
<span><a name="asio.overview.implementation.solaris"></a></span><a class="link" href="implementation.html#asio.overview.implementation.solaris">Solaris</a>
</h5>
<p>
Demultiplexing mechanism:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
Uses <code class="literal">/dev/poll</code> for demultiplexing.
</li></ul></div>
<p>
Threads:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
Demultiplexing using <code class="literal">/dev/poll</code> is performed in one
of the threads that calls <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run_one</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll_one</span><span class="special">()</span></code>.
</li>
<li class="listitem">
An additional thread per <code class="computeroutput"><span class="identifier">io_context</span></code>
is used to emulate asynchronous host resolution. This thread is created
on the first call to either <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">tcp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">udp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code>.
</li>
</ul></div>
<p>
Scatter-Gather:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
At most <code class="computeroutput"><span class="identifier">min</span><span class="special">(</span><span class="number">64</span><span class="special">,</span><span class="identifier">IOV_MAX</span><span class="special">)</span></code> buffers may be transferred in a single
operation.
</li></ul></div>
<h5>
<a name="asio.overview.implementation.h4"></a>
<span><a name="asio.overview.implementation.qnx_neutrino"></a></span><a class="link" href="implementation.html#asio.overview.implementation.qnx_neutrino">QNX
Neutrino</a>
</h5>
<p>
Demultiplexing mechanism:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
Uses <code class="computeroutput"><span class="identifier">select</span></code> for demultiplexing.
This means that the number of file descriptors in the process cannot
be permitted to exceed <code class="computeroutput"><span class="identifier">FD_SETSIZE</span></code>.
</li></ul></div>
<p>
Threads:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
Demultiplexing using <code class="computeroutput"><span class="identifier">select</span></code>
is performed in one of the threads that calls <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run_one</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll_one</span><span class="special">()</span></code>.
</li>
<li class="listitem">
An additional thread per <code class="computeroutput"><span class="identifier">io_context</span></code>
is used to emulate asynchronous host resolution. This thread is created
on the first call to either <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">tcp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">udp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code>.
</li>
</ul></div>
<p>
Scatter-Gather:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
At most <code class="computeroutput"><span class="identifier">min</span><span class="special">(</span><span class="number">64</span><span class="special">,</span><span class="identifier">IOV_MAX</span><span class="special">)</span></code> buffers may be transferred in a single
operation.
</li></ul></div>
<h5>
<a name="asio.overview.implementation.h5"></a>
<span><a name="asio.overview.implementation.mac_os_x"></a></span><a class="link" href="implementation.html#asio.overview.implementation.mac_os_x">Mac
OS X</a>
</h5>
<p>
Demultiplexing mechanism:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
Uses <code class="computeroutput"><span class="identifier">kqueue</span></code> for demultiplexing.
</li></ul></div>
<p>
Threads:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
Demultiplexing using <code class="computeroutput"><span class="identifier">kqueue</span></code>
is performed in one of the threads that calls <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run_one</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll_one</span><span class="special">()</span></code>.
</li>
<li class="listitem">
An additional thread per <code class="computeroutput"><span class="identifier">io_context</span></code>
is used to emulate asynchronous host resolution. This thread is created
on the first call to either <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">tcp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">udp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code>.
</li>
</ul></div>
<p>
Scatter-Gather:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
At most <code class="computeroutput"><span class="identifier">min</span><span class="special">(</span><span class="number">64</span><span class="special">,</span><span class="identifier">IOV_MAX</span><span class="special">)</span></code> buffers may be transferred in a single
operation.
</li></ul></div>
<h5>
<a name="asio.overview.implementation.h6"></a>
<span><a name="asio.overview.implementation.freebsd"></a></span><a class="link" href="implementation.html#asio.overview.implementation.freebsd">FreeBSD</a>
</h5>
<p>
Demultiplexing mechanism:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
Uses <code class="computeroutput"><span class="identifier">kqueue</span></code> for demultiplexing.
</li></ul></div>
<p>
Threads:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
Demultiplexing using <code class="computeroutput"><span class="identifier">kqueue</span></code>
is performed in one of the threads that calls <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run_one</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll_one</span><span class="special">()</span></code>.
</li>
<li class="listitem">
An additional thread per <code class="computeroutput"><span class="identifier">io_context</span></code>
is used to emulate asynchronous host resolution. This thread is created
on the first call to either <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">tcp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">udp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code>.
</li>
</ul></div>
<p>
Scatter-Gather:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
At most <code class="computeroutput"><span class="identifier">min</span><span class="special">(</span><span class="number">64</span><span class="special">,</span><span class="identifier">IOV_MAX</span><span class="special">)</span></code> buffers may be transferred in a single
operation.
</li></ul></div>
<h5>
<a name="asio.overview.implementation.h7"></a>
<span><a name="asio.overview.implementation.aix"></a></span><a class="link" href="implementation.html#asio.overview.implementation.aix">AIX</a>
</h5>
<p>
Demultiplexing mechanism:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
Uses <code class="computeroutput"><span class="identifier">select</span></code> for demultiplexing.
This means that the number of file descriptors in the process cannot
be permitted to exceed <code class="computeroutput"><span class="identifier">FD_SETSIZE</span></code>.
</li></ul></div>
<p>
Threads:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
Demultiplexing using <code class="computeroutput"><span class="identifier">select</span></code>
is performed in one of the threads that calls <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run_one</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll_one</span><span class="special">()</span></code>.
</li>
<li class="listitem">
An additional thread per <code class="computeroutput"><span class="identifier">io_context</span></code>
is used to emulate asynchronous host resolution. This thread is created
on the first call to either <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">tcp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">udp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code>.
</li>
</ul></div>
<p>
Scatter-Gather:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
At most <code class="computeroutput"><span class="identifier">min</span><span class="special">(</span><span class="number">64</span><span class="special">,</span><span class="identifier">IOV_MAX</span><span class="special">)</span></code> buffers may be transferred in a single
operation.
</li></ul></div>
<h5>
<a name="asio.overview.implementation.h8"></a>
<span><a name="asio.overview.implementation.hp_ux"></a></span><a class="link" href="implementation.html#asio.overview.implementation.hp_ux">HP-UX</a>
</h5>
<p>
Demultiplexing mechanism:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
Uses <code class="computeroutput"><span class="identifier">select</span></code> for demultiplexing.
This means that the number of file descriptors in the process cannot
be permitted to exceed <code class="computeroutput"><span class="identifier">FD_SETSIZE</span></code>.
</li></ul></div>
<p>
Threads:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
Demultiplexing using <code class="computeroutput"><span class="identifier">select</span></code>
is performed in one of the threads that calls <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run_one</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll_one</span><span class="special">()</span></code>.
</li>
<li class="listitem">
An additional thread per <code class="computeroutput"><span class="identifier">io_context</span></code>
is used to emulate asynchronous host resolution. This thread is created
on the first call to either <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">tcp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">udp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code>.
</li>
</ul></div>
<p>
Scatter-Gather:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
At most <code class="computeroutput"><span class="identifier">min</span><span class="special">(</span><span class="number">64</span><span class="special">,</span><span class="identifier">IOV_MAX</span><span class="special">)</span></code> buffers may be transferred in a single
operation.
</li></ul></div>
<h5>
<a name="asio.overview.implementation.h9"></a>
<span><a name="asio.overview.implementation.tru64"></a></span><a class="link" href="implementation.html#asio.overview.implementation.tru64">Tru64</a>
</h5>
<p>
Demultiplexing mechanism:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
Uses <code class="computeroutput"><span class="identifier">select</span></code> for demultiplexing.
This means that the number of file descriptors in the process cannot
be permitted to exceed <code class="computeroutput"><span class="identifier">FD_SETSIZE</span></code>.
</li></ul></div>
<p>
Threads:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
Demultiplexing using <code class="computeroutput"><span class="identifier">select</span></code>
is performed in one of the threads that calls <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run_one</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll_one</span><span class="special">()</span></code>.
</li>
<li class="listitem">
An additional thread per <code class="computeroutput"><span class="identifier">io_context</span></code>
is used to emulate asynchronous host resolution. This thread is created
on the first call to either <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">tcp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">udp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code>.
</li>
</ul></div>
<p>
Scatter-Gather:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
At most <code class="computeroutput"><span class="identifier">min</span><span class="special">(</span><span class="number">64</span><span class="special">,</span><span class="identifier">IOV_MAX</span><span class="special">)</span></code> buffers may be transferred in a single
operation.
</li></ul></div>
<h5>
<a name="asio.overview.implementation.h10"></a>
<span><a name="asio.overview.implementation.windows_95__98_and_me"></a></span><a class="link" href="implementation.html#asio.overview.implementation.windows_95__98_and_me">Windows
95, 98 and Me</a>
</h5>
<p>
Demultiplexing mechanism:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
Uses <code class="computeroutput"><span class="identifier">select</span></code> for demultiplexing.
</li></ul></div>
<p>
Threads:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
Demultiplexing using <code class="computeroutput"><span class="identifier">select</span></code>
is performed in one of the threads that calls <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run_one</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll_one</span><span class="special">()</span></code>.
</li>
<li class="listitem">
An additional thread per <code class="computeroutput"><span class="identifier">io_context</span></code>
is used to emulate asynchronous host resolution. This thread is created
on the first call to either <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">tcp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">udp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code>.
</li>
</ul></div>
<p>
Scatter-Gather:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
For sockets, at most 16 buffers may be transferred in a single operation.
</li></ul></div>
<h5>
<a name="asio.overview.implementation.h11"></a>
<span><a name="asio.overview.implementation.windows_nt__2000__xp__2003__vista__7__8__10_and_11"></a></span><a class="link" href="implementation.html#asio.overview.implementation.windows_nt__2000__xp__2003__vista__7__8__10_and_11">Windows
NT, 2000, XP, 2003, Vista, 7, 8, 10 and 11</a>
</h5>
<p>
Demultiplexing mechanism:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
Uses overlapped I/O and I/O completion ports for all asynchronous socket
operations except for asynchronous connect.
</li>
<li class="listitem">
Uses <code class="computeroutput"><span class="identifier">select</span></code> for emulating
asynchronous connect.
</li>
</ul></div>
<p>
Threads:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
Demultiplexing using I/O completion ports is performed in all threads
that call <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">run_one</span><span class="special">()</span></code>, <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">io_context</span><span class="special">::</span><span class="identifier">poll_one</span><span class="special">()</span></code>.
</li>
<li class="listitem">
An additional thread per <code class="computeroutput"><span class="identifier">io_context</span></code>
is used to trigger timers. This thread is created on construction of
the first <code class="computeroutput"><span class="identifier">basic_deadline_timer</span></code>
or <code class="computeroutput"><span class="identifier">basic_waitable_timer</span></code>
objects.
</li>
<li class="listitem">
An additional thread per <code class="computeroutput"><span class="identifier">io_context</span></code>
may be used for <code class="computeroutput"><span class="identifier">select</span></code>
demultiplexing. This thread is created on the first call to:
<div class="itemizedlist"><ul class="itemizedlist" type="circle">
<li class="listitem">
A socket <code class="computeroutput"><span class="identifier">async_wait</span><span class="special">()</span></code> function, except when using
<code class="computeroutput"><span class="identifier">wait_read</span></code> on a
stream-oriented socket. (For <code class="computeroutput"><span class="identifier">wait_read</span></code>
on a stream-oriented socket, the overlapped I/O operation <code class="computeroutput"><span class="identifier">WSARecv</span></code> is used and no additional
thread is required.)
</li>
<li class="listitem">
A socket <code class="computeroutput"><span class="identifier">async_connect</span><span class="special">()</span></code> operation, if the overlapped
I/O operation <code class="computeroutput"><span class="identifier">ConnectEx</span></code>
is unavailable. (On recent versions of Windows, <code class="computeroutput"><span class="identifier">ConnectEx</span></code>
is used and no additional thread is required.)
</li>
</ul></div>
</li>
<li class="listitem">
An additional thread per <code class="computeroutput"><span class="identifier">io_context</span></code>
is used to emulate asynchronous host resolution. This thread is created
on the first call to either <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">tcp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code> or <code class="computeroutput"><span class="identifier">ip</span><span class="special">::</span><span class="identifier">udp</span><span class="special">::</span><span class="identifier">resolver</span><span class="special">::</span><span class="identifier">async_resolve</span><span class="special">()</span></code>.
</li>
</ul></div>
<p>
Scatter-Gather:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
For sockets, at most 64 buffers may be transferred in a single operation.
</li>
<li class="listitem">
For stream-oriented handles, only one buffer may be transferred in a
single operation.
</li>
</ul></div>
<h5>
<a name="asio.overview.implementation.h12"></a>
<span><a name="asio.overview.implementation.windows_runtime"></a></span><a class="link" href="implementation.html#asio.overview.implementation.windows_runtime">Windows
Runtime</a>
</h5>
<p>
Asio provides limited support for the Windows Runtime. It requires that the
language extensions be enabled. Due to the restricted facilities exposed
by the Windows Runtime API, the support comes with the following caveats:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
The core facilities such as the <code class="computeroutput"><span class="identifier">io_context</span></code>,
<code class="computeroutput"><span class="identifier">strand</span></code>, buffers, composed
operations, timers, etc., should all work as normal.
</li>
<li class="listitem">
For sockets, only client-side TCP is supported.
</li>
<li class="listitem">
Explicit binding of a client-side TCP socket is not supported.
</li>
<li class="listitem">
The <code class="computeroutput"><span class="identifier">cancel</span><span class="special">()</span></code>
function is not supported for sockets. Asynchronous operations may only
be cancelled by closing the socket.
</li>
<li class="listitem">
Operations that use <code class="computeroutput"><span class="identifier">null_buffers</span></code>
are not supported.
</li>
<li class="listitem">
Only <code class="computeroutput"><span class="identifier">tcp</span><span class="special">::</span><span class="identifier">no_delay</span></code> and <code class="computeroutput"><span class="identifier">socket_base</span><span class="special">::</span><span class="identifier">keep_alive</span></code>
options are supported.
</li>
<li class="listitem">
Resolvers do not support service names, only numbers. I.e. you must use
"80" rather than "http".
</li>
<li class="listitem">
Most resolver query flags have no effect.
</li>
</ul></div>
<p>
Demultiplexing mechanism:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
Uses the <code class="computeroutput"><span class="identifier">Windows</span><span class="special">::</span><span class="identifier">Networking</span><span class="special">::</span><span class="identifier">Sockets</span><span class="special">::</span><span class="identifier">StreamSocket</span></code> class to implement asynchronous
TCP socket operations.
</li></ul></div>
<p>
Threads:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">
Event completions are delivered to the Windows thread pool and posted
to the <code class="computeroutput"><span class="identifier">io_context</span></code> for
the handler to be executed.
</li>
<li class="listitem">
An additional thread per <code class="computeroutput"><span class="identifier">io_context</span></code>
is used to trigger timers. This thread is created on construction of
the first timer objects.
</li>
</ul></div>
<p>
Scatter-Gather:
</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem">
For sockets, at most one buffer may be transferred in a single operation.
</li></ul></div>
</div>
<div class="copyright-footer">Copyright © 2003-2023 Christopher M. Kohlhoff<p>
Distributed under the Boost Software License, Version 1.0. (See accompanying
file LICENSE_1_0.txt or copy at <a href="http://www.boost.org/LICENSE_1_0.txt" target="_top">http://www.boost.org/LICENSE_1_0.txt</a>)
</p>
</div>
<hr>
<div class="spirit-nav">
<a accesskey="p" href="cpp2011/chrono.html"><img src="../../prev.png" alt="Prev"></a><a accesskey="u" href="../overview.html"><img src="../../up.png" alt="Up"></a><a accesskey="h" href="../../index.html"><img src="../../home.png" alt="Home"></a><a accesskey="n" href="../using.html"><img src="../../next.png" alt="Next"></a>
</div>
</body>
</html>