Transcript PPT Chapter 02

Chapter 2
The OS, the Computer,
and User Programs
Copyright © 2008
Introduction
• Fundamental Principles of OS Operation
• The Computer
• OS Interaction with the Computer and User Programs
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.2
Fundamental Principles of OS
Operation
• The kernel of the OS is the collection of routines that
form the core of the operating system
– Implements control functions
– Set of services to user programs
– Exists in memory during operation of the OS
• An interrupt diverts the CPU to execution of the kernel
code
• A software interrupt is used by programs to
communicate their requests to the kernel
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.3
Fundamental Principles of OS
Operation (continued)
• CPU has two modes of operation:
– Kernel mode
• CPU can execute all instructions
• Kernel operates with CPU in this mode so that it can control
computer operations
– User mode
• CPU cannot execute instructions that could interfere with other
programs or with the OS if used indiscriminately
• CPU is put in this mode to execute user programs
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.4
The Computer
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.5
The Computer (continued)
•
•
•
•
•
The CPU
Memory Management Unit (MMU)
Memory Hierarchy
Input/Output
Interrupts
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.6
The CPU
• Two features of the CPU are visible to user programs
or the OS:
– General-purpose registers (GPRs)
• Also called program-accessible registers
• Hold data, addresses, index values, or the stack pointer
during execution of a program
– Control registers
• Contain information that controls or influences operation of
the CPU
• Set of control registers is called the program status word
(PSW)
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.7
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.8
The CPU (continued)
• CPU can operate in two modes:
– Kernel mode
• Can execute privileged instructions
• OS puts CPU in kernel mode when it is executing
instructions in the kernel
– User mode
• Cannot execute privileged instructions
• OS puts CPU in user mode while executing user programs
• Mode (M) field of PSW contains 0 if CPU is in privileged
mode and 1 if it is in user mode
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.9
State of the CPU
• GPRs and PSW contain the information needed to
know what the CPU is doing
– State of the CPU
• Kernel saves state of CPU when it takes away the CPU
from program
– When program is to be resumed, it reloads the saved
CPU state into GPRs and PSW
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.10
Example 2.1: State of the CPU (Contd)
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.11
Memory Management Unit (MMU)
• Virtual memory is an illusion of memory that may be
larger than the real memory of a computer
– Implemented using noncontiguous memory allocation
and the MMU
• CPU passes the address of data or instruction used in an
instruction to MMU
– It is called the logical addresses
• MMU translates logical address to physical address
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.12
Memory hierarchy
• The memory hierarchy provides a large and fast
memory, at a low cost
– It is an arrangement of several memories with different
access speeds and sizes
• The CPU accesses only the fastest memory; i.e., the cache
• If a required byte is not present in the memory being
accessed, it is loaded there from a slower memory
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.13
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.14
Memory Hierarchy (continued)
• When CPU performs a cache lookup, a cache hit or
miss may occur
– Hit ratio (h) of the cache is the fraction of bytes accessed
by the CPU that score a hit in the cache
tema = h × tcache + (1 – h) × (ttra + tcache)
= tcache + (1 – h) × ttra
where
tema = effective memory access time,
tcache = access time of cache, and
ttra
= time taken to transfer a cache block from memory to
cache.
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.15
Memory Hierarchy (continued)
• Operation of memory is analogous to operation of a
cache
– Blocks of bytes (pages) are transferred from disk to
memory or from memory to disk
– But,
• Memory management and transfer of blocks between
memory and disk are performed by SW
• In the cache, the transfer is performed by HW
• Memory hierarchy comprising MMU, memory, and the
disk is called the virtual memory
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.16
Memory Hierarchy (continued)
• Memory protection is implemented by checking whether
a memory address used by a program lies outside the
memory area allocated to it
– Control registers used:
• base and size (also called limit)
– Address of first byte = <base>
– Address of last byte = <base> +<size> – 1
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.17
Example 2.2: Basics of Memory
Protection
•interrupt
• Execution of Load instruction causes protection violation
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.18
Input/Output
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.19
Interrupts
• An event is a situation that requires OS’s attention
– Designer associates an interrupt with each event
• Purpose is to report occurrence of the event to OS and
enable it to perform event handling actions
• Interrupt action saves CPU state and loads new
contents into the PSW and GPRs
– CPU starts executing instructions of an interrupt servicing
routine (ISR) in the kernel
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.20
Interrupts (continued)
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.21
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.22
OS Interaction with the Computer and
User Programs
• OS interacts with the computer to
– know information about events, so that it can service
them
– restore CPU state to resume a program after servicing an
interrupt
• Programs need to use services of the OS for purposes
such as initiating an I/O operation
– The method to cause an interrupt and pass requirement
to the OS is known as a System call
• Will learn about:
– Controlling Execution of Programs
– Interrupt Servicing
– System Calls
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.23
Controlling Execution of Programs
1. When user program starts, PSW should contain:
a. Program counter (PC) field
b. Mode (M) field, set to user mode (1)
c. Memory protection information (MPI) field contains start
address in memory and size of program
d. Interrupt mask (IM) field, set to enable interrupts
2. When program is interrupted, CPU state (PSW and
GPRs) are saved
•
Program table or process control block (PCB)
3. When program is resumed, its CPU state is restored
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.24
Interrupt Servicing
• Context save saves CPU state
of the program
• The scheduler selects a
a program for execution
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.25
Interrupt Servicing (continued)
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.26
Interrupt Servicing (continued)
• Two approaches for nested interrupt servicing:
– Disable nested interrupts through masking
– Handle nested interrupts─preemptible kernel
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.28
System Calls
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.29
Example 2.4: System Call in a
Hypothetical OS
• CPU provides the SI instruction to cause a software
interrupt
• OS provides system call for obtaining current time
– Code is 78: Instruction SI 78 causes a SW interrupt
• 78 is entered in IC field of PSW before it is saved
• Interrupt vector for program contains aaa in PC
– CPU is switched to routine with start address aaa
– It finds that IC is 78 and determines that program needs
time of day
• Time is returned to program in a standard location, typically
a data register
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.30
System Calls (continued)
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.31
Summary
• An interrupt is a special signal sent to CPU to indicate
occurrence of an event
– It transfers control to the OS
– A system call is a SW interrupt
• A program uses it to request OS services
• CPU’s control registers govern its functioning
– Program status word (PSW) is the collection of these
• OS kernel saves CPU state when interrupt occurs
– i.e., PSW and GPRs
• CPU has two modes of operation controlled by mode
(M) field of PSW
– User mode and kernel mode
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.32
Summary (continued)
• Memory hierarchy provides the same effect as a fast
and large memory, at a low cost
– Contains:
• A very fast and small cache memory
• A slower and larger RAM
• A disk
• Effective memory access time depends on cache hit
ratio
• I/O system uses direct memory access (DMA) to permit
CPU and I/O system to operate independently
Operating Systems, by Dhananjay Dhamdhere
Copyright © 2008
2.33