📄 udev.html
字号:
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"><html xmlns="http://www.w3.org/1999/xhtml"> <head> <meta http-equiv="Content-Type" content= "application/xhtml+xml; charset=iso-8859-1" /> <title> 11.5. Device and Module Handling on an LFS System </title> <link rel="stylesheet" href="../stylesheets/lfs.css" type="text/css" /> <meta name="generator" content="DocBook XSL Stylesheets V1.69.1" /> <link rel="stylesheet" href="../stylesheets/lfs-print.css" type= "text/css" media="print" /> </head> <body id="lfs" class="CLFS-SVN-20060417-MIPS"> <div class="navheader"> <div class="headertitles"> <h4> Cross-Compiled Linux From Scratch - Version CLFS-SVN-20060417-MIPS </h4> <h3> Chapter 11. Setting Up System Bootscripts </h3> </div> <ul class="headerlinks"> <li class="prev"> <a accesskey="p" href="usage.html" title= "How Do These Bootscripts Work?">Prev</a> <p> How Do These Bootscripts Work? </p> </li> <li class="next"> <a accesskey="n" href="setclock.html" title= "Configuring the setclock Script">Next</a> <p> Configuring the setclock Script </p> </li> <li class="up"> <a accesskey="u" href="chapter.html" title= "Chapter 11. Setting Up System Bootscripts">Up</a>. </li> <li class="home"> <a accesskey="h" href="../index.html" title= "Cross-Compiled Linux From Scratch - Version CLFS-SVN-20060417-MIPS"> Home</a> </li> </ul> </div> <div class="sect1" lang="en" xml:lang="en"> <div class="titlepage"> <h1 class="sect1"> 11.5. Device and Module Handling on an LFS System </h1> </div> <p> In <a href="../final-system/chapter.html">Installing Basic System Software</a>, we installed the Udev package. Before we go into the details regarding how this works, a brief history of previous methods of handling devices is in order. </p> <p> Linux systems in general traditionally use a static device creation method, whereby a great many device nodes are created under <tt class="filename">/dev</tt> (sometimes literally thousands of nodes), regardless of whether the corresponding hardware devices actually exist. This is typically done via a <span><strong class= "command">MAKEDEV</strong></span> script, which contains a number of calls to the <span><strong class="command">mknod</strong></span> program with the relevant major and minor device numbers for every possible device that might exist in the world. Using the Udev method, only those devices which are detected by the kernel get device nodes created for them. Because these device nodes will be created each time the system boots, they will be stored on a <tt class= "systemitem">tmpfs</tt> (a virtual file system that resides entirely in system memory). Device nodes do not require much space, so the memory that is used is negligible. </p> <div class="sect2" lang="en" xml:lang="en"> <div class="titlepage"> <h2 class="sect2"> 11.5.1. History </h2> </div> <p> In February 2000, a new filesystem called <tt class= "systemitem">devfs</tt> was merged into the 2.3.46 kernel and was made available during the 2.4 series of stable kernels. Although it was present in the kernel source itself, this method of creating devices dynamically never received overwhelming support from the core kernel developers. </p> <p> The main problem with the approach adopted by <tt class= "systemitem">devfs</tt> was the way it handled device detection, creation, and naming. The latter issue, that of device node naming, was perhaps the most critical. It is generally accepted that if device names are allowed to be configurable, then the device naming policy should be up to a system administrator, not imposed on them by any particular developer(s). The <tt class= "systemitem">devfs</tt> file system also suffers from race conditions that are inherent in its design and cannot be fixed without a substantial revision to the kernel. It has also been marked as deprecated due to a lack of recent maintenance. </p> <p> With the development of the unstable 2.5 kernel tree, later released as the 2.6 series of stable kernels, a new virtual filesystem called <tt class="systemitem">sysfs</tt> came to be. The job of <tt class="systemitem">sysfs</tt> is to export a view of the system's hardware configuration to userspace processes. With this userspace-visible representation, the possibility of seeing a userspace replacement for <tt class="systemitem">devfs</tt> became much more realistic. </p> </div> <div class="sect2" lang="en" xml:lang="en"> <div class="titlepage"> <h2 class="sect2"> 11.5.2. Udev Implementation </h2> </div> <p> The <tt class="systemitem">sysfs</tt> filesystem was mentioned briefly above. One may wonder how <tt class="systemitem">sysfs</tt> knows about the devices present on a system and what device numbers should be used for them. Drivers that have been compiled into the kernel directly register their objects with <tt class= "systemitem">sysfs</tt> as they are detected by the kernel. For drivers compiled as modules, this registration will happen when the module is loaded. Once the <tt class="systemitem">sysfs</tt> filesystem is mounted (on <tt class="filename">/sys</tt>), data which the built-in drivers registered with <tt class= "systemitem">sysfs</tt> are available to userspace processes and to <span><strong class="command">udev</strong></span> for device node creation. </p> <p> The <span><strong class="command">S10udev</strong></span> initscript takes care of creating these device nodes when Linux is booted. This script starts by registering <span><strong class= "command">/sbin/udevsend</strong></span> as a hotplug event handler. Hotplug events (discussed below) are not usually generated during this stage, but <span><strong class= "command">udev</strong></span> is registered just in case they do occur. The <span><strong class="command">udevstart</strong></span> program then walks through the <tt class="systemitem">/sys</tt> filesystem and creates devices under <tt class="filename">/dev</tt> that match the descriptions. For example, <tt class= "filename">/sys/class/tty/vcs/dev</tt> contains the string “<span class="quote">7:0</span>” This string is used by <span><strong class="command">udevstart</strong></span> to create <tt class="filename">/dev/vcs</tt> with major number <span class= "emphasis"><em>7</em></span> and minor <span class= "emphasis"><em>0</em></span>. The names and permissions of the nodes created under the <tt class="filename">/dev</tt> directory are configured according to the rules specified in the files within the <tt class="filename">/etc/udev/rules.d/</tt> directory. These are numbered in a similar fashion to the LFS-Bootscripts package. If <span><strong class="command">udev</strong></span> can't find a rule for the device it is creating, it will default permissions to <span class="emphasis"><em>660</em></span> and ownership to <span class="emphasis"><em>root:root</em></span>. </p> <p> Once the above stage is complete, all devices that were already present and have compiled-in drivers will be available for use. This leads us to the devices that have modular drivers. </p> <p> Earlier, we mentioned the concept of a “<span class= "quote">hotplug event handler.</span>” When a new device connection is detected by the kernel, the kernel will generate a hotplug event and look at the file <tt class= "filename">/proc/sys/kernel/hotplug</tt> to determine the userspace program that handles the device's connection. The <span><strong class="command">udev</strong></span> bootscript registered <span><strong class="command">udevsend</strong></span> as this handler. When these hotplug events are generated, the kernel will tell <span><strong class="command">udev</strong></span> to check the <tt class="filename">/sys</tt> filesystem for the information pertaining to this new device and create the <tt class=⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -