11.6. Device and Module Handling on a CLFS System

In Installing Basic System Software, we installed the Eudev package. Before we go into the details regarding how this works, a brief history of previous methods of handling devices is in order.

11.6.1. History Static Device Nodes

Linux systems in general traditionally use a static device creation method, whereby a great many device nodes are created under /dev (sometimes literally thousands of nodes), regardless of whether the corresponding hardware devices actually exist. This is typically done via a MAKEDEV script, which contains a number of calls to the mknod program with the relevant major and minor device numbers for every possible device that might exist in the world. Devfs

In February 2000, a new filesystem called devfs, which dynamically created device nodes as devices were found by the kernel, 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.

The main problem with the approach adopted by devfs 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 devfs file system also suffered from race conditions that were inherent in its design and could not be fixed without a substantial revision to the kernel. It was marked deprecated with the release of the 2.6 kernel series, and was removed entirely as of version 2.6.18. Sysfs

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 sysfs came to be. The job of sysfs is to export a view of the system's hardware configuration to userspace processes. Drivers that have been compiled into the kernel directly register their objects with sysfs as they are detected by the kernel. For drivers compiled as modules, this registration will happen when the module is loaded. Once the sysfs filesystem is mounted (on /sys), data which the built-in drivers registered with sysfs are available to userspace processes. With this userspace-visible representation, the possibility of seeing a userspace replacement for devfs became much more realistic. Udev Implementation

Shortly after the introduction of sysfs, work began on a program called Udev to advantage of it. The udev daemon made calls to mknod() to create device nodes in /dev dynamically, based on the information from sysfs, in /sys. For example, /sys/class/tty/vcs/dev contains the string “7:0”. This string was used by udev to create a device node with major number 7 and minor number 0.

Linux kernel version 2.6.32 introduced a new virtual file system called devtmpfs, an improved replacement for devfs. This allows device nodes to once again be dynamically created by the kernel, without many of the problems of devfs. As of version 176, Udev no longer creates device nodes itself, instead relying on devtmpfs to do so. Systemd and Eudev

In 2010, development began on systemd, an alternate init implementation. Starting with Udev 183, Udev's source tree was merged with systemd. Several Gentoo developers who disagreed with this merge announced a project fork called Eudev in December 2012, created by extracting the Udev code from systemd. One of the goals of Eudev is to allow for easier installation and usage of udevd without the need for the rest of systemd.

11.6.2. Device Node Creation

By default, device nodes created by the kernel in a devtmpfs are owned by root:root and have 600 permissions. udevd can modify ownership and permissions of the nodes under the /dev directory, and can also create additional symlinks, based on rules specified in the files within the /etc/udev/rules.d, /lib/udev/rules.d, and /run/udev/rules.d directories. The names for these files start with a number, to indicate the order in which they are run, and they have a .rules extension (udevd will ignore files with any other extension). All of the rules files from these directories are combined into a single list, sorted by filename, and run in that order. In the event of a conflict, where a rules file with the same name exists in two or more of these directories, the rules in /etc take the highest priority, followed by rules files in /run, and finally /lib. Any device for which a rule cannot be found will just be ignored by udevd and be left at the defaults defined by the kernel, as described above.

11.6.3. Module Loading

Device drivers compiled as modules may have aliases built into them. Aliases are visible in the output of the modinfo program and are usually related to the bus-specific identifiers of devices supported by a module. For example, the snd-fm801 driver supports PCI devices with vendor ID 0x1319 and device ID 0x0801, and has an alias of “pci:v00001319d00000801sv*sd*bc04sc01i*”. For most devices, the bus driver exports the alias of the driver that would handle the device via sysfs. E.g., the /sys/bus/pci/devices/0000:00:0d.0/modalias file might contain the string “pci:v00001319d00000801sv00001319sd00001319bc04sc01i00”. The default rules provided by Eudev will cause udevd to call out to /sbin/modprobe with the contents of the MODALIAS uevent environment variable (that should be the same as the contents of the modalias file in sysfs), thus loading all modules whose aliases match this string after wildcard expansion.

In this example, this means that, in addition to snd-fm801, the obsolete (and unwanted) forte driver will be loaded if it is available. See below for ways in which the loading of unwanted drivers can be prevented.

The kernel itself is also able to load modules for network protocols, filesystems and NLS support on demand.

11.6.4. Problems with Loading Modules and Creating Devices

There are a few possible problems when it comes to automatically creating device nodes. A kernel module is not loaded automatically

Eudev will only load a module if it has a bus-specific alias and the bus driver properly exports the necessary aliases to sysfs. In other cases, one should arrange module loading by other means. With Linux-3.14.21, Eudev is known to load properly-written drivers for INPUT, IDE, PCI, USB, SCSI, SERIO and FireWire devices.

To determine if the device driver you require has the necessary support for Eudev, run modinfo with the module name as the argument. Now try locating the device directory under /sys/bus and check whether there is a modalias file there.

If the modalias file exists in sysfs, the driver supports the device and can talk to it directly, but doesn't have the alias, it is a bug in the driver. Load the driver without the help from Eudev and expect the issue to be fixed later.

If there is no modalias file in the relevant directory under /sys/bus, this means that the kernel developers have not yet added modalias support to this bus type. With Linux-3.14.21, this is the case with ISA busses. Expect this issue to be fixed in later kernel versions.

Eudev is not intended to load “wrapper” drivers such as snd-pcm-oss and non-hardware drivers such as loop at all. A kernel module is not loaded automatically, and Eudev is not intended to load it

If the “wrapper” module only enhances the functionality provided by some other module (e.g., snd-pcm-oss enhances the functionality of snd-pcm by making the sound cards available to OSS applications), configure modprobe to load the wrapper after Eudev loads the wrapped module. To do this, add an “install” line to a file in /etc/modprobe.d. For example:

install snd-pcm /sbin/modprobe -i snd-pcm ; \
    /sbin/modprobe snd-pcm-oss ; true

If the module in question is not a wrapper and is useful by itself, configure the S05modules bootscript to load this module on system boot. To do this, add the module name to the /etc/sysconfig/modules file on a separate line. This works for wrapper modules too, but is suboptimal in that case. Eudev loads some unwanted module

Either don't build the module, or blacklist it in /etc/modprobe.d file as done with the forte module in the example below:

blacklist forte

Blacklisted modules can still be loaded manually with the explicit modprobe command. Eudev makes a wrong symlink

This usually happens if a rule unexpectedly matches a device. For example, a poorly-written rule can match both a SCSI disk (as desired) and the corresponding SCSI generic device (incorrectly) by vendor. Find the offending rule and make it more specific, with the help of udevadm info. Eudev rule works unreliably

This may be another manifestation of the previous problem. If not, and your rule uses sysfs attributes, it may be a kernel timing issue, to be fixed in later kernels. For now, you can work around it by creating a rule that waits for the used sysfs attribute and appending it to the /etc/udev/rules.d/10-wait_for_sysfs.rules file. Please notify the CLFS Development list if you do so and it helps. Device naming order changes randomly after rebooting

This is due to the fact that Eudev, by design, handles uevents and loads modules in parallel, and thus in an unpredictable order. This will never be “fixed”. You should not rely upon the kernel device names being stable. Instead, create your own rules that make symlinks with stable names based on some stable attributes of the device, such as a serial number or the output of various *_id utilities installed by Eudev. See Section 11.7, “Creating custom symlinks to devices” and Networking Configuration for examples.

11.6.5. Useful Reading

Additional helpful documentation is available at the following sites: