Chapter 10. Timers and Time Management
The passing of time is very important to the kernel. A large number of kernel functions are time driven, as opposed to event driven. Some of these functions are periodic, such as balancing the scheduler runqueues or refreshing the screen. They occur on a fixed schedule, such as 100 times per second. The kernel schedules other functions, such as delayed disk I/O, at a relative time in the future. For example, the kernel might schedule work for 500 milliseconds from now. Finally, the kernel must also manage the system uptime and the current date and time.
Note the differences between relative and absolute time. Scheduling an event for five seconds in the future requires no concept of the absolute timeonly the relative time (for example, five seconds from now). Conversely, managing the current time of day requires the kernel to understand not just the passing of time, but also some absolute measurement of it. Both these concepts are crucial to the management of time.
Also note the differences between events that occur periodically and events the kernel schedules for a fixed point in the future. Events that occur periodicallysay, every 10 millisecondsare driven by the system timer. The system timer is a programmable piece of hardware that issues an interrupt at a fixed frequency. The interrupt handler for this timercalled the timer interruptupdates the system time and performs periodic work. The system timer and its timer interrupt are central to Linux, and a large focus of this chapter.
The other focus is dynamic timersthe facility used to schedule events that run once after a specified time has elapsed. For example, the floppy device driver uses a timer to shut off the floppy drive motor after a specified period of inactivity. The kernel can create and destroy timers dynamically. This chapter covers the kernel implementation of dynamic timers, as well as the interface available for their use in your code.