Interrupts

Interrupts

1. What is an interrupt?

   • There are numerous ways in which an interrupt may occur.
     • Updating the real-time clock
     • A printer running out of paper.
     • Data arriving at a network card.

2. How does an interrupt work?

   • When a program is interrupted it must remember at what point the interruption occured.
     • This is done by saving the contents of the important registers to the stack.
       • The interrupting program can then be run
         • Once complete, the data to resume the interrupted program can be 'popped' from the stack and the program can then continue from the point that it was interrupted.
           • Sometimes this can be done so quickly that the interrupt is never noticed.

3. Types of interrupt

   • Hardware
     • Maskable - Can be delayed to make room for higher priorities.
     • Non Maskable - This is for hardware that can not be delayed and must be processed immediately.
   • Software
     • Normal - These are interrupts which are caused by the software and are part of the program
     • Exception - Unplanned interrupts while executing a program is called Exception.

4. Problems with interrupts

   • Interrupt Latency - The time interval between the occurrence of the interrupt and the start of execution of the ISR.

5. Priorities

   • If an interrupt is being processed and another interrupt (with higher priority) arises:
     • OS suspends the current interrupt routine
       • Runs the new higher priority interrupt routine.
         • The OS returns to the original interrupt routine.
   • Priority levels may depend on
     • Importance of the requesting device.
     • Length of time required to service the interrupt