Operating Systems: Chapter 1

• Chapter 1 introduces the fundamental concepts of operating systems.

What is an Operating System?

• An operating system (OS) acts as an intermediary between a computer user and the computer hardware.
• The goals of an OS are to execute user programs, make computing easier, and utilize hardware efficiently.

What Operating Systems Do

• Operating systems manage hardware and software, acting as communication channels.
• They control computer resources, provide services to programs, coordinate program execution, and offer a user interface (virtual machine).
• Operating Systems also hide software complexity, support multiple execution modes, and monitor program execution for errors.

Functions of Operating Systems

• Operating systems handle memory, processors, devices, and files.
• They also manage I/O, user interfaces, computer booting, security, system performance, job accounting, error detection, coordination between software/users, basic computer tasks, and network.

Computer System Structure

• A computer system consists of hardware, operating systems, application programs, and users.
• Hardware provides computing resources and the operating system controls and coordinates hardware use.
• Application programs define how system resources are used, and users interact with the system.

Computer System Organization

• Modern systems use CPUs and device controllers connected via a common bus for shared memory access.
• Concurrent execution enables CPUs and devices to compete for memory cycles.

Computer System Operation

• I/O devices and the CPU execute concurrently.
• Device controllers manage specific device types, each with a local buffer.
• The device driver manages this, and the CPU moves data between main memory and local buffers.
• I/O occurs from the device to the local buffer and the device controller signals completion by causing an interrupt.

Common Functions of Interrupts

• Interrupt architecture saves the address of interrupted instructions.
• Traps or exceptions are software-generated interrupts caused by errors or user requests.
• Operating systems are interrupt-driven.

Interrupt Handling

• The operating system saves the CPU state (registers, program counter) upon interrupt.
• It identifies the interrupt type and executes specific code to handle each interrupt.

Interrupt-driven I/O Cycle

• The device driver initiates I/O.
• While the CPU executes other code, the I/O controller initiates I/O and the CPU checks for interrupts.
• When complete, the controller sends an interrupt signal.
• The CPU transfers control to an interrupt handler, data is processed, and the CPU resumes the interrupted task.

Storage Structure

• Main memory is accessible by the CPU, allowing random access.
• Main memory is typically volatile while secondary storage provides larger, nonvolatile capacity.

Storage Hierarchy

• Storage systems are organized in a hierarchy based on speed, cost, and volatility.
• Caching involves copying information into faster storage for quicker access.
• A device driver is used for each device controller to manage I/O and provide a uniform interface between the controller and the kernel

Operating-System Operations

• The bootstrap program initializes the system and loads the kernel, typically residing in the ROM.
• The operating system is then loaded into RAM, and device drivers are started.

Kernel Loading

• The kernel is the core software module of an operating system, loading first and residing in main memory permanently.
• It is placed in protected memory and controls device functions, memory/task management, and disk management.
• The kernel also provides software infrastructure, like drivers to integrate system hardware with application software.

Daemons

• Daemons are system software running in the background without direct user interaction.
  • System Daemons manage system-wide functions like time synchronization
  • Network Daemons handle network-related processes such as DNS
  • Device Daemons manage processes specific to a particular device, i.e. Bluetooth

Kernel Operations

• Kernel operations are interrupt-driven, activated by hardware and software.
• The OS is activated by an interrupt and suspends a running program and transfers control to the OS.
• A program resumes once the service is completed.
• Hardware interrupts are caused by devices, while software interrupts (exceptions or traps) occur due to software errors/system calls.

Multiprogramming (Batch system)

• Single users cannot always keep CPU and I/O busy.
• Multiprogramming organizes jobs so that CPU always has one to execute, and a job is selected and run by job scheduling.
• In order to do this, a subset of total jobs in the system is kept in memory
• When a job waits (e.g., for I/O), the operating system switches to another job.

Multitasking (Timesharing)

• Multitasking allows frequent switching between jobs enabling interactive computing.
• Each user has at least one process (task) executing in memory.
• CPU scheduling takes place if several jobs are ready to run at the same time.
• Swapping moves processes in and out of memory and Virtual memory allows the execution of processes not completely in memory.

Multiprogramming vs Multitasking

• Multiprogramming loads multiple programs to share the CPU, switching between them, to increase CPU utilization. It usually does not involve direct user interaction.
• Multitasking executes multiple tasks simultaneously by rapidly switching between them. This improves user responsiveness and allows for interactive computing.

Dual-mode Operation

• Dual-mode operation protects the OS and system components.
• User mode runs user applications, while kernel mode executes OS code with unrestricted hardware access.

Transition from User to Kernel Mode

• System calls switch the mode to kernel, and return calls reset the mode to user.
• Some privileged instructions can only be executed in kernel mode.
• The mode bit provided by hardware distinguishes between user and kernel code.
• When a user runs, the code bit is "user" while if the kernel is running the mode bit is set to "kernel".

User Mode vs Kernel Mode

• User Mode is restricted, with application programs executing. An interrupt will cause only one process to fail and cannot access kernel programs directly.
• Kernel Mode is privileged when accessing hardware resources.
• In kernel mode, an interrupt could cause the whole operating system to fail, although both user and kernel programs can perform access.

Timer

• A timer can prevent infinite loops by interrupting the computer after a set time period.
• After an interrupt the operating system sets a counter (privileged instruction), and generates an interrupt upon zero.
• Timers are setup before scheduling so as to regain control or terminate a program.
• The Timer is privileged, only the OS can load it.

Process Management

• A process is an active program in execution, requiring resources to accomplish its task (CPU, memory, I/O, files).
• Every single-threaded process has a program counter, while a multi-threaded process has one program counter per thread.
• Processes can run concurrently using concurrency by multiplexing CPUs among threads.

Process Management Activities

• Process Management Activities are the responsibility of the operating system.
• O/S activities includes creating/deleting processes, suspending/resuming, providing mechanisms for synchronization, communication, and preventing deadlocks.

Memory Management

• Instructions must be in memory to execute programs.
• The data that needed by programs must be in memory.
• Memory management optimizes CPU utilization and computer response to users.
• Memory must also be tracked for usage, decide which data to move in or out, and allocate/deallocate as needed.

File-system Management

• The operation system provides a uniform, logical view of information storage through files.
• Managing a file-system includes files usually organized into directories, access control to determine who can access what, and OS activities like creating/deleting, primitives to manipulate files and directories, mapping/backup, and files onto secondary storage.

Mass-Storage Management

• Disks are usually used to store data that must be kept for a "long" period.
• It is important to properly manage storage.
• Storage activities include: mounting/unmounting, fee-space management, storage allocation, disk scheduling, partitioning, and protection.

Caching

• Caching is used performantly at multiple computer levels (hardware, operating system, software).
• Information in use is copied from slower to faster storage temporarily.
• If information is available, it is used directly from the (cache)
• Otherwise, data is copied to cache and used there.
• Issues include caching management and size / replacement policy

Types of Storage

• There are different levels of storage with different characteristics like name, size, implementation, technology, access-time, and transfer speed.
• The main levels are registers, cache, main memory, solid-state disk, and magnetic disk.
• Moving between levels of storage hierarchy can be explicit or implicit.

I/O Subsystem

• An operating system hides hardware device peculiarities.
• I/O Subsystem provides Memory management, General device-driver interface and Drivers for specific hardware devices
• Memory management includes I/O request scheduling, buffering, caching, spooling, error handling and I/O protection

Protection and Security

• Protection mechanism controls access of processes or users to defined resources.
• Security defends the system internally and externally (denial-of-service, worms, viruses, identity theft, theft).
• Systems distinguish users via user identities (user IDs, security IDs).
• Systems also utilize group identifier which allows a set of users to be defined and managed, then associated with each process, file
• Security is escalated using Privilege escalation to allow the user to the affected ID with more rights