Advanced Operating Systems: Lecture 37 - Mr. Farhan Zaidi

Advanced Operating Systems - Lecture 37: Interrupt handlers. This lecture will cover the following: interrupt handlers; interrupts and exceptions; linux interrupt handling; top halfs, bottom halfs and tasklets; timings and timer devices; linux kernel timers and interval timers;...

CS703 ­ Advanced  Operating Systems

By Mr Farhan Zaidi

37

 Interrupt handlers

 Interrupts and exceptions

 Linux interrupt handling

 Top halfs, bottom halfs and tasklets

 Timings and timer devices

 Linux kernel timers and interval timers

 Interrupt handlers are best hidden

 Interrupt procedure does its task

 Varying terminology but for Intel:

 Interrupt (device generated)

 Maskable: device-generated, associated with IRQs (interrupt

request lines); may be temporarily disabled (still pending)

 Nonmaskable: some critical hardware failures

 Exceptions (Processor-detected)

 Faults – correctable (restartable); e.g page fault

 Traps – no reexecution needed; e.g breakpoint

 Aborts – severe error; process usually terminated (by

signal)

 Programmed exceptions (software interrupts)

 int (system call), int3 (breakpoint)

 into (overflow), bounds (address check)

 Hardware support for getting CPUs attention

 Often transfers from user to kernel mode

 Asynchronous: device or timer generated

 Synchronous: immediate result of last instruction

 Intel terminology and hardware

 Irqs, vectors, IDT, gates, PIC, APIC

 More complex than exceptions

 Requires registry, deferred processing, etc

 Some issues:

 IRQs are often shared; all handlers (ISRs) are executed so they must query device

 Three types of actions:

 Critical: Top-half (interrupts disabled – briefly!)

 Non-critical: Top-half (interrupts enabled)

 Non-critical deferrable: Do it “later” (interrupts enabled)

Memory Bus

INTR

vector 0

15

IRQs

IDT

0

255

handler

idtr

Mask points

 Accurate timing crucial for many aspects of OS

 Device-related timeouts

 File timestamps (created, accessed, written)

 Time-of-day (gettimeofday()

 High-precision timers (code profiling, etc.)

 Scheduling, cpu usage, etc

 Intel timer hardware

 RTC: Real Time Clock

 PIT: Programmable Interrupt Timer

 TSC: TimeStamp Counter (cycle counter)

 Local APIC Timer: per-cpu alarms

 Timer implementations

 Kernel timers (dynamic timers)

 User “interval” timers (alarm(), setitimer())

 Dynamic timers may be dynamically created and destroyed No limit is

placed on the number of currently active dynamic timers

 A dynamic timer is stored in the following timer_list structure:

struct timer_list

{

struct list_head list;

unsigned long expires;

unsigned long data;

void (*function)(unsigned long);

};

 Each timer contains a field that indicates how far in the future the timer should expire This field is initially calculated by adding the right number

of ticks to the current value of jiffies

 The field does not change Every time the kernel checks a timer, it

compares the expiration field to the value of jiffies at the current

moment, and the timer expires when jiffies is greater or equal to the

stored value

 setitimer(): 3 distinct user mode interval timers

 can set for one-time or periodic alarm

 sends signals on timer expiry

 three timers:

 ITIMER_REAL: elapsed time (SIGALRM)

 ITIMER_VIRT: user (SIGVTALRM)

 ITIMER_PROF: user + kernel (SIGPROF)

 implementations:

 VIRT, PROF – updates by PIT or APIC

 REAL – requires kernel timer

 may need to deliver to blocked process

 current->real_timer

 shared by alarm() API so can’t use both