Preemptive multitasking

Pre-emptive multitasking is a form of multitasking in which processes are not allowed to take an indefinitely long time to complete execution in the CPU. Each process, in turn, is granted a portion of CPU time (usually called a time slice, on the order of milliseconds). A process which does not relinquish the CPU voluntarily within its time slice is forcibly context-switched with the next process in scheduling queue.

Context-switching can usually happen for many different reasons. Some common reasons are:

• the process has consumed its time slice, and the system clock interrupt pre-empts the process
• the process goes to wait state (often due to planned sleeping, waiting for an I/O event to happen, or just Mutual exclusion)
• a higher priority process becomes ready for execution, which causes a pre-emption
• the process gives away its time slice voluntarily
• the process terminates itself

Some OSes permit processes to deny pre-emption for a period of time. e.g. AmigaOS would permit a process to refuse to context-switch in order to read system data structures that might be altered by other processes.

Some OSes always forbid lower priority processes to pre-empt higher priority processes. Some OSes allow it occasionally. See scheduling.