Devices include hardware components such as, printers, drives, adapters, buses, and enclosures, as well as pseudo-devices, such as the error special file and null special file. This chapter provides an overview of the following methods used by the operating system to manage these devices:
Devices are organized into clusters known as nodes. Each node is a logical subsystem of devices, where lower-level devices have a dependency on upper-level devices in child-parent relationships. For example, the system node is the highest of all nodes and consists of all the physical devices in the system. The system device is the top of the node and below that are the bus and adapters that have a dependency on the higher-level system device. At the bottom of the hierarchy are devices to which no other devices are connected. These devices have dependencies on all devices above them in the hierarchy.
At startup time, parent-child dependencies are used to configure all devices that make up a node. Configuration occurs from the top node down and any device having a dependency on a higher-level device is not configured until the higher-level device is configured.
Managing devices requires the operating system to comprehend what device connections are allowed. The operating system classifies devices hierarchically into three groups:
Functional classes consist of devices that perform the same function. Printers, for example, comprise a functional class. Functional classes are grouped into subclasses according to certain device similarities. For example, printers have a serial or parallel interface. Serial printers are one subclass and parallel printers are another. Device types are classified according to their model and manufacturer.
Device classes define valid parent-child connections for the operating system. The hierarchy defines the possible subclasses that can be connected for each of the possible child connection locations. For example, the term RS-232 8-port adapter specifies that only devices belonging to the RS-232 subclass can be connected to any of the eight ports of the adapter.
Device classes and their hierarchical dependencies are maintained in an Object Data Manager (ODM) Device Configuration database.
Device information is contained in a predefined database or a customized database that makes up the device configuration database.
The predefined database contains configuration data for all possible devices supported by the system. The hierarchical device class information is contained in this database.
The customized database contains configuration data for all currently defined and configured devices in the system. A record is kept of each device currently connected to your system.
The Configuration Manager is a program that automatically configures devices on your system during system startup and run time. The Configuration Manger uses the information from the predefined and customized databases during this process, and updates the customized database afterwards.
Devices that are connected to the system can be in one of four
states:
If a tty device and a printer alternately use the same tty connector, both a tty device and a printer are defined on the same parent and port in the device configuration database. Only one of these devices can be configured at a time. When the tty connector is configured, the printer specific setup information is retained until it is configured again. The device is not removed; it is in the defined state. Maintaining a device in defined state retains customized information for a device that is not currently in use, either before it is first made available or while it is temporarily removed from the system.
If a device driver exists for a device, the device can be made available through the device driver.
Some devices, in particular TCP/IP pseudo-devices, need the stopped state.
You can use the Web-based System Manager Devices application, SMIT, or operating system commands to perform device management tasks such as deleting or adding a device.