Skip to content

Chapter 4: The Abstraction: The Process

"A process is a running program. The program itself is a lifeless thing: it just sits there on the disk, a bunch of instructions waiting to spring into action. It is the operating system that transforms these bytes into something useful."

1. The Crux: The Illusion of Many CPUs

How can the OS provide the illusion of a nearly-endless supply of CPUs when there are only a few physical ones available?

The OS achieves this through Virtualizing the CPU. By running one process, then stopping it and running another, the OS creates the illusion of many virtual CPUs. This technique is known as Time Sharing.

[!CAUTION] Performance Trade-off: Sharing the CPU inevitably leads to slower performance for each individual process, as they must share the hardware.

2. Machine State: What Constitutes a Process?

To understand a process, we must understand its Machine State—everything the program can read or update during execution.

Key Components:

  • Memory (Address Space): Where instructions lie and where the data the program reads/writes sits.
  • Registers:
    • Program Counter (PC) / Instruction Pointer (IP): Tells us which instruction will execute next.
    • Stack Pointer & Frame Pointer: Used to manage the stack for function parameters, local variables, and return addresses.
  • Persistent Storage: Includes a list of files the process currently has open.

3. Process API: The OS Interface

Any modern operating system must include these fundamental interfaces for process management: 1. Create: A way to create new processes. 2. Destroy: Forcefully stop a process. 3. Wait: Wait for a process to stop running. 4. Miscellaneous Control: Suspending or resuming a process. 5. Status: Getting information about a process (how long it has run, its state, etc.).

4. Process Creation: From Disk to Memory

How does the OS get a program up and running? It follows a systematic procedure:

  1. Loading: The OS loads the code and static data from disk into the address space of the process.
    • Modern optimization: Loading is often done lazily, only bringing in pieces of code/data as needed.
  2. Stack Allocation: The OS allocates memory for the program's run-time stack (for local variables, function parameters, and return addresses).
  3. Heap Allocation: The OS allocates memory for the heap (for explicitly requested dynamic data via malloc()).
  4. I/O Setup: Initializing tasks related to standard input, output, and error.
  5. Entry Point: The OS starts execution at main(), handing off control of the CPU to the newly created process.

5. Process States

A process is always in one of three primary states:

stateDiagram-v2
    [*] --> Ready: Create
    Ready --> Running: Scheduled
    Running --> Ready: Descheduled
    Running --> Blocked: I/O Request
    Blocked --> Ready: I/O Done
    Running --> [*]: Exit
  • Running: The process is executing instructions on a processor.
  • Ready: The process is prepared to run but the OS hasn't chosen to run it yet.
  • Blocked: The process has performed some operation (like I/O) that makes it unready to run until some event happens.

6. Mechanisms vs. Policies

  • Context Switch (Mechanism): The low-level piece of code that gives the OS the ability to stop running one program and start running another.
  • Scheduling Policy (Policy): The high-level algorithms used to make decisions (e.g., which program should the OS run next?).

Last Updated: May 14, 2026

End Note: Understanding the process abstraction is the first step in mastering how an operating system virtualizes hardware. By managing machine state and transitioning between states, the OS provides a robust environment for multiple programs to coexist.

Comments