Operating Systems Lab. (#1) University of Tehran – ECE Dept. Fall 2005 Reza Shokri [email protected].

Operating Systems Lab.(#1)

University of Tehran – ECE Dept.Fall 2005

Reza [email protected]

Operating Systems Lab - UT ECE Dept Fall 2005

2/40

Outline

The Evolution of Linux Kernel Organization Linux Versions The Booting Procedure Process and Resource Management Memory Management Device Management File Management An Overview of the Unix Filesystem


3/40

The Evolution of Linux

By 1985, two primary versions of UNIX were running on many different hardware platforms: The main line version from AT&T Bell Labs (System V) Alternative version from the University of California at

Berkeley (BSD UNIX) By 1989, a committee had been formed to develop a

standardized system call interfaces. The result was and API, a system call interface, called POSIX.

Linux developed by “Linus Trovald” in 1991 It inherits from UNIX architecture

BUT, not completely


4/40

Kernel Organization

UNIX kernels, is designed and implemented as a monolithic kernel.

Data structures for each aspect of the kernel are generally accessible to all other aspects of the kernel.

Device Drivers Modules Interrupts


5/40

Kernel Organization


6/40

Kernel Organization

Trap System call

For a system call, F, a stub procedure is used to invoke F (The stub is also named F)

The stub is linked into the (user-space) calling program. When a process executes the call to F at runtime, control is

transferred to the stub procedure rather than directly to the kernel.

The stub procedure validates the parameter values being passed to the kernel procedure.

The stub procedure executes a trap instruction that switches the CPU to supervisory mode. It then branches to the entry point for the target kernel function.


7/40

Linux Versions


8/40

The Booting Procedure The BIOS first runs the manufacturer’s hardware

booting sequence by executing the Power-On Self Test (POST). The POST performs various diagnostic tests to check presence of resources.

The BIOS launches the first stage boot loader on the MBR of the primary hard disk.

The first stage boot loader loads itself into memory and launches the second stage boot loader from the /boot/ partition.

The second stage boot loader loads the kernel into memory, which in turn loads any necessary modules and mounts the root partition read-only.


9/40

The Booting Procedure

The kernel transfers control of the boot process to the /sbin/init program.

The /sbin/init program loads all services and user-space tools, and mounts all partitions listed in /etc/fstab.

The user is presented with a login prompt for the freshly booted Linux system.


10/40

The Booting Procedure (in detail)

BIOS On Boot, Processor looks at the end of system

memory for the Basic Input/Output System or BIOS program and runs it.

The BIOS controls the first step of the boot process.

BIOS provides the lowest level interface to peripheral devices.

Tests the system, looks for and checks peripherals, and then locates a valid device with which to boot the system.


11/40


BIOS loads into memory whatever program is residing in the first sector of this device, called the Master Boot Record or MBR. The MBR is only 512 bytes in size and contains machine

code instructions for booting the machine, called a boot loader

BIOS also loads the partition table. Once the BIOS finds and loads the boot loader

program into memory, it yields control of the boot process to it.


12/40


Boot Loader X86 Arch.

LILO GRUB

Boot loaders are broken into 2 stages Firsts stage just locates the second stage and loads first

part of it into memory. Second Stage displays a graphical screen showing

different OS options to boot. Then, places the kernel from /boot and loads initial RAM

Disk initrd<kernel-version>.img The rest of boot process is done by kernel


13/40


The Kernel Initializes and configures the memory and various

hardware attached to system I/O, storage devices, all processors

Decompresses and loads initrd which contains necessary drivers.

Initializes virtual devices LVM Software RAID

Unmounts initrd and frees up all memory used by initrd


14/40


Then, Kernel creates a root device, mounts the root partition read-only

NOW, it is operational, BUT there are no user applications

To set up the user environment, kernel executes /sbin/init program

init program Coordinates the rest of boot process Is the grandparent of all other processes


15/40


At the first, runs /etc/rc.d/rc.sysinit script Sets environment path, starts swap, checks file

system and initialize terminals all things! At the first executes /etc/rc.d/init.rd/functions Sets up graphical interface Starts number of minimum consoles (mingetty) that all

display login screen and sets their default font Shows the message “Welcome to RedHat Linux …

Press I to Interactive Mode” Mounts /proc file system, configures kernel

parameters, sets up system clock, keyboard, keymap


16/40


Initializes USB Checks root file system

In failure runs fsck Remount file system as read-write Activate swap partition Runs system wide configuration scripts in

/etc/sysconfig Then checks run level parameter in

/etc/inittab file to start or stop appropriate services by running /etc/rc.d/rc script


17/40


Run Levels 0 - halt (Do NOT set initdefault to this) 1 - Single user mode 2 - Multiuser, without NFS 3 - Full multiuser mode 4 - unused 5 - X11 6 - reboot (Do NOT set initdefault to this)


18/40


Related files /etc/inittab (id:3:initdefault) /etc/rc. d/init.d

Contains scripts to start and stop all services Each /etc/rc.d/rcX.d

Contains two type of soft links K<a number><service name> : k20nfs S<a number><service name> : S05nfs (Numbers keeps order of scripts execution)

The Last script which init runs is /etc/rc.d/rc.local This file is useful for system customization

The telinit command is used to change run-levels on-the-fly on a running Linux system. telinit 5


19/40

The Booting Procedure (in more details)

Hardware fetch-execute cycle begins Executing Hardware Process Executing POST Reading boot record and load the OS into primary memory Run kernel to initialize the computer hardware

Setting the CPU in supervisor mode. Branching to the main entry point in the kernel.

Kernel initializes the trap table, the interrupt handler, the scheduler, the clock, the modules, and so on.

Kernel initializes the process manager. The hardware process then creates the initial process. The initial process is allocated the first entry in the kernel’s

process descriptor table. Is referred to task[0] or INIT_TASK.


20/40

The Booting Procedure (in more details)

The initial process then creates the first useful Linux process to run the init program and begins to execute an idle loop. (init process also called the idle process)

Running other init programs from /etc, /bin, /sbin, /etc/rc. During initialization, the kernel creates one process on each

communication port that can be used to support user logins. Each of these processes copy of the getty program.

The getty process initializes itself and then waits for a user to begin using the port.

When the user begins to use the port, getty runs the login program, expecting authentication Information.

If the authentication is successful, then the login process changes the current directory to the user’s home directory and executes the specified shell program.


21/40

Process and Resource Management

Each user program run as a Process Process Manager

Creates a collection of Virtual Machines as host CPU (fork())

Uses hardware level timers, interrupts, protection mechanisms, IPC, scheduling and appropriate data structures


22/40


Process Manager Is created as a first process (PID = 0)

Other processes are forked from this process Runs only when a process begins supervisor

mode execution (Syscall or Interrupt) Each process is represented by a task_struct

data structure. The task_vector is an array of pointers to

every task_struct data structure in the system.


23/40


Process States Running

The process is either running or it is ready to run. Waiting

The process is waiting for an event or for a resource. Stopped

The process has been stopped, usually by receiving a signal.

Zombie This is a halted process which, for some reason, still has a

task_struct data structure in the task vector.


24/40

Memory Management

Memory Manager Allocates blocks of the system’s primary memory on

request Ensures exclusive control of a block it has been allocated Provide a means by which cooperating processes share

blocks of memory Automatically move information between the primary and

secondary memories

Linux uses Demand Paged Virtual Memory Model


25/40

Memory Management

Physical memory is divided into segments of equal size, called pages. Type of memory:

Physical memory Swap memory

It is a partition in hard disk that is treated as physical memory

It is used for storing infrequent data and also hardware caching

Virtual memory System defines 32-bit memory address (4 GB) virtually


26/40

Memory Management

Processes address a virtual address and kernel memory manager maps it into physical address and handles page misses

Each virtual memory is divided into 2 segments 3GB user space 1GB kernel space

Kernel segment is permanently mapped and associated with physical address


27/40

Device Management

Device Management Using Polling Using Interrupts

With each device, a Device Driver Program is associated which does Setting up hardware on initialization. Bringing the associated devices into and out of services. Receiving data from the hardware and passing it back to

the kernel. Sending data from the kernel to the device. Detecting and handling device errors.


28/40

Device Management

Each device treated as a file In Linux everything is abstracted by file (except

CPU and Memory) Linux are divided to

Character Devices Are read and written on character at a time (byte

stream) like: Mouse, Keyboard Block Devices

Are accessed by blocks of data (i.e. 1 kilo byte at a time) like: disks


29/40

Device Management

Devices and Device Drivers are referenced by two numbers Major Number:

Defines the class of device. SCSI=21 Minor Number

A Device Driver can be used with multiple devices. Minor number is used to distinguish among the physical devices that use a particular device driver


30/40

File Management

It is responsible for storing and retrieving information on disk Provide a single internal view of file

File is named byte stream that can be stored in a secondary storage as a set of blocks

Linux file manager defines common set of functions specified by POSIX open(), read(), lseek(), write(), close(), ioctl()


31/40

File Management

File Manager is built on an abstract file model Virtual File System (VFS) VFS does common, system-independent tasks

Checking access permission When data to be read or written

File System Designers provide extensions to VFS to implement system-dependent portions


32/40

File Management

The Unix system has the following types of files: Ordinary Files

Contain information entered into them by a user, an application or …

Directory Files Manage the cataloging of the file system

Special Files (devices) Used to access the peripheral devices

FIFO Files for Pipes


33/40

File Management

Each file has an inode structure that is identified by an i-number.

The inode contains the information required to access the file (File Descriptor). Access rights, owner, group, size, creation time,

… It doesn’t contain file name


34/40

An Overview of the Linux Filesystem

An operating system's le system structure is its most basic level of organization.

Almost all of the ways an operating system interacts with its users, applications, and security model are dependent upon the way it stores its les on a storage device.

A file system can be seen in terms of two different logical categories of les: Shareable vs. unsharable les Variable vs. static files


35/40


Filesystem Hierarchy Standard (FHS), a collaborative document that denes the names and locations of many les and directories.

/dev/ Directory The /dev/ directory contains le system entries which

represent devices that are attached to the system.

/etc/ Directory The /etc/ directory is reserved for conguration les that are

local to the machine. No binaries are to be put in /etc/.


36/40


The /lib/ Directory The /lib/ directory should contain only those

libraries that are needed to execute the binaries in /bin/ and /sbin/.

These shared library images are particularly important for booting the system and executing commands within the root le system.

/mnt/ Directory The /mnt/ directory is for temporarily mounted le

systems, such as CD-ROMs and oppy disks.


37/40


/opt/ Directory The /opt/ directory provides storage for large,

static application software packages. A package placing files in the /opt/ directory

creates a directory bearing the same name as the package. This directory in turn holds les that otherwise would be scattered throughout the file system, giving the system administrator an easy way to determine the role of each file within a particular package.


38/40


/proc/ Directory The /proc/ directory contains special files that

either extract information from or send information to the kernel.

/sbin/ Directory The /sbin/ directory is for executables used only

by the root user. The executables in /sbin/ are only used to boot

and mount /usr/ and perform system recovery operations.


39/40


/usr/ Directory The /usr/ directory is for les that can be shared

across a whole site. The /usr/ directory usually has its own partition, and it should be mountable read-only.

/usr/local/ Directory The /usr/local hierarchy is for use by the system

administrator when installing software locally. It may be used for programs and data that are shareable

among a group of hosts, but not found in /usr.“ is similar in structure to the /usr/ directory.


40/40


/var/ Directory Since the FHS requires Linux to mount /usr/ read-

only, any programs that write log les or need spool/ or lock/ directories should write them to the /var/ directory.

Operating Systems Lab. (#1) University of Tehran – ECE Dept. Fall 2005 Reza Shokri [email protected].

Documents

kernel procedure

monolithic kernel

boot process

stub procedure

university of tehran

stage boot loader

boot partition

standardized system