Download interrupt - Universidade de Coimbra

Operating
Systems
2006/2007
1. Introduction
Paulo Marques
Departamento de Eng. Informática
Universidade de Coimbra
[email protected]
Disclaimer

This slides and notes are heavily based on the companion material of
[Silberschatz05].The original material can be found at:


In some cases, material from [Stallings04] may also be used. The
original material can be found at:


http://codex.cs.yale.edu/avi/os-book/os7/slide-dir/index.html
http://williamstallings.com/OS/OS5e.html
The respective copyrights belong to their owners.
2
What’s an Operating System?
User Applications
Operating System
Hardware
3
What Operating Systems Do

Controls and coordinates use of hardware among various
applications and users


The OS is a resource allocator: it manages all resources and
decides between conflicting requests for efficient and fair resource
use
OS is a control program: controls execution of programs to
prevent errors and improper use of the computer
4
Two Modes of Operation
User Mode
Kernel Mode
UNIX Operating System Structure
5
Input/Output versus CPU Usage






I/O devices and the CPU can execute concurrently
Each device controller is in charge of a particular device
type
Each device controller has a local buffer
CPU moves data from/to main memory to/from local
buffers
I/O is from the device to local buffer of controller
Device controller informs CPU that it has finished its
operation by causing an interrupt
6
Interrupt Timeline
7
Interrupt Operation





Interrupt transfers control to the interrupt service routine
generally, through the interrupt vector, which contains
the addresses of all the service routines
Interrupt architecture must save the address of the
interrupted instruction and the relevant context of the
executing program
Incoming interrupts are disabled while another interrupt is
being processed to prevent a lost interrupt
A trap is a software-generated interrupt caused either by
an error or a user request
An operating system is interrupt driven
8
Interrupt Operation (2)

The operating system preserves the state of the CPU by
storing registers and the program counter

Determines which type of interrupt has occurred:



polling
vectored interrupt system
Separate segments of code determine what action should
be taken for each type of interrupt
9
Different Types of I/O
10
Different Types of I/O (2)

After I/O starts, control returns to user program only upon
I/O completion:




Wait instruction idles the CPU until the next interrupt
Wait loop (contention for memory access)
At most one I/O request is outstanding at a time,
no simultaneous I/O processing or even CPU processing
After I/O starts, control returns to user program without
waiting for I/O completion:




System call – request to the operating system to allow user to
wait for I/O completion
Device-status table contains an entry for each I/O device
indicating its type, address, and state
Operating system indexes into I/O device table to determine
device status and to modify table entry to include interrupt
Multiple outstanding I/O operations over time.
Also, CPU processing can carry on.
11
Device Status Table
12
Direct Memory Access (DMA)

Used for high-speed I/O devices able to transmit
information at close to memory speeds

Device controller transfers blocks of data from buffer
storage directly to main memory without CPU
intervention

Implies the capability of arbitrating the system bus

Only one interrupt is generated per block, rather than the
one interrupt per byte
13
DMA (2)
14
Storage Hierarchy

Caches, Buffers and a Storage Hierarchy are Essential
15
Storage Hierarchy

A great part of this hierarchy is controlled by the
Operating System



Cache Management Policies
Buffer Management Policies
Memory Management Policies
Principles of Locality:
- Temporal Locality
- Spatial Locality
16
Protection

Dual Mode of Operation





I/O Protection


All I/O Instructions are privileged.
For doing I/O a trap into the operating system must be generated.
Memory Protection


“User Mode” and “Kernel Mode”
In “User Mode” only common instructions can be performed (e.g.
arithmetic and logic). In “Kernel Mode” can do anything.
A mode bit contains the current mode of operation
Kernel mode assumed on traps and on interrupts
Each program can only access its memory
CPU Protection

The CPU must remain in control of the Operating System, even if the
applications don’t want to let go.
17
Dual Mode of Operation
18
System Calls, Traps and Interrupts
Traps
System calls
Operating
System
Interrupts
19
Structure of a System Call
20
Example with a “Hello World” Application!
#include <stdio.h>
int main() {
printf(“Hello World\n”);
return 0;
}

The “strace” program is very useful:

It allows you to see what system calls are made and with what
parameters!
21
“strace” in Action!
22
System Call Implementation

Typically, a number associated with each system call



System-call interface maintains a table indexed according to these
numbers
The system call interface invokes intended system call in
OS kernel and returns status of the system call and any
return values
The caller needs to know nothing about how the system
call is implemented


Just needs to obey an API and understand what the OS will do as
a result call
Most details of OS interface are hidden from the programmer by
the API
23
System Call Example (Win32 API)
Quick Question:
Why use function libraries instead of system calls??
24
System Call Implementation (2)

Often, more information is required than simply the
identity of the desired system call


Exact type and amount of information vary according to OS and
call
Three general methods used to pass parameters to the OS

Simplest: pass the parameters in registers


Parameters stored in a block, or table, in memory, and address of
block passed as a parameter in a register



In some cases, there may be more parameters than registers
This approach is taken by Linux and Solaris
Parameters placed, or pushed, onto the stack by the program and
popped off the stack by the operating system
Block and stack methods do not limit the number or length of
parameters being passed
25
System Call Parameter Passing
26
Memory Protection

Each program must only be able to access its own
memory



Segmentation is based on having a base pointer and a
limit for each process running


Segmentation
Virtual Memory (Mostly Used Today)
Any access made outside of its “space” generates a trap and
normally leads to the process being killed
Virtual Memory is based on having a table that translates
“virtual addresses” into real ones



Normally, there is such a table for each process
Each process sees all the address space
An access made to an non existing page generates a trap and
normally leads to the process being killed
27
Segmentation
28
Virtual Memory
4Gb
4Gb
Disk
5000
1000
1000
0
0
Address Space
of Process A
256Mb
Address Space
of Process B
Address Translation
Table
0
Physical Memory
(simplified)
29
Multiprogramming


Multiprogramming: maintaining several jobs in
memory, active at the same time, so that when the
current program can no longer run (for example because
it’s blocked waiting for I/O), another is available to run
Optimizes resource utilization: CPU and I/O
Time Multiplexed CPU
CPU
E.g. Windows 95/98!
30
Multitasking

Multitasking is an extension of multiprogramming in
which the execution of the active programs is time-sliced:


Each program runs for a short period of time, then another is run.
When a program is running and is forcibly replaced by another,
typically with a higher priority, it is said to have been preempted,
thus the term Preemptive Operating Systems
31
Multitasking

For implementing multitasking, a non-maskable
interrupt must connected to a hardware timer

Every few milliseconds (e.g. 100ms), the interrupt causes a
task (or process) switch
User Level
total = 0;
for (int i=0; i<20000; i++)
total = total + i;
printf(“total=%d\n”, total);
Kernel Level
while (!feof(fd))
{
if (fscanf(fd, “%d”, &d) == 1)
printf(“%d\n”, d);
}
(...)
Interrupt Handler
32
Important Note
Mode Switch != Task Switch

Although they are both quite heavy!
33
Nowadays…

Windows XP, Linux 2.6.xx:


Preemptive multitasking operating systems using virtual
memory
Each process thinks it has the whole computer for itself
 “Virtual Machine”
34
Operating System Architecture and Structures

Currently, there are two dominant architectures…

Microkernel:

Moves as much as possible from the kernel into “user” space
Communication takes place between user modules
using message passing

Benefits:






Easier to extend a microkernel
Easier to port the operating system to new architectures
More reliable (less code is running in kernel mode)
More secure
Detriments:

SLOW: Performance overhead of user space to kernel space
communication
35
Operating System Architecture and Structures

Currently, there are two dominant architectures…

Monolithic:

Consists of everything below the system-call interface and above
the physical hardware
Provides the file system, CPU scheduling, memory management,
and other operating-system functions; a large number of functions
for one level

Benefits:



FAST!
Detriments:



Less reliable (more code is running in kernel mode)
Less secure
No so easy to port to new architectures (… it depends)
36
Operating System Architecture and Structures

Most modern operating systems implement
kernel modules





Uses “object-oriented approach”
Each core component is separate
Each talks to the others over known interfaces
Each is loadable as needed within the kernel
Benefits


It’s easier to extend the operating system
Allows to isolate functionality in well defined software entities,
making the OS more maintainable and reliability
37
Kernel Modules in Solaris
38
Operating System Architecture and Structures

Virtual Machines

Facet A:



“Abstract Machine” with a certain ISA (e.g. Java, MS.NET)
Good for portability, allowing running the same unmodified
application in different physical architectures
Facet B:


Software that allows different operating systems to be run in
parallel giving them the illusion of “having a whole machine”
(e.g. VMWARE, Xen)
Allows for server consolidation, software testing, software fault
tolerance, etc.
39
Microsoft .NET CLR
.NET Application
The CLR
virtual machine
Host Operating System
Hardware
40
VMWARE Architecture
41
Operating System Booting

Every time the computer wakes up, it starts executing code
at a certain address



At that address there is no RAM! What’s mapped there it’s
an EPROM with the BIOS!
The BIOS contains a small program that performs a certain
number of functions:






On your PC, it’s address 0x000FFFF0!
Executes the POST test (Power On Self Test)…
Looks and initializes the Graph Card… and other devices
(…)
Locates the device where to boot from… (assume a disk)
Tries to execute the code at cylinder 0, head 0, sector 1!
At the first sector of the disk resides the MBR:

Master Boot Record
42
Operating System Booting (2)


The program in the MBR checks the primary partition
table of the disk
If it finds a primary partition marked active, it starts
executing the code of its volume boot sector

This volume boot sector normally consists in a boot
loader which, eventually, loads the code of the operating
system and starts to execute it

The operating system has to setup all the necessary data
structures (e.g. special registers of the processor).
The last step before entering its “main execution stage” is
switching to protected mode
And… that’s it!


43
Reference

Chapter 1: Introduction


1.1, 1.2, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9
Chapter 2: Operating Systems Structures


All chapter 2! 
2.1, 2.2, 2.3, 2.4, 2.5, 2.6, 2.7, 2.8, 2.9, 2.10
44