In Unix like operating systems, what does the fork() system call do when a process invokes it?

Introduction / Context:
Process creation is a fundamental operation in Unix like operating systems. The fork() system call is the primary mechanism by which new processes are created. Understanding its behavior, including what is copied and how the return values differ in the parent and child, is essential for writing multi process applications and for interpreting exam questions about process control.

Given Data / Assumptions:

• A process executing in user space calls the fork() system call.
• The operating system supports virtual memory and separate process address spaces.
• The process has an open file table, registers, and other state.

Concept / Approach:
When fork() is called, the kernel creates a new process known as the child. The child receives a copy of the parent process address space, including code, data, and stack, although modern systems implement this copy efficiently using copy on write techniques. The child also inherits many attributes such as open file descriptors and environment. After fork() returns, two processes are running: the original parent and the new child. The important detail is that fork() returns different values in each process: zero in the child and the child process identifier in the parent. This allows programs to distinguish parent and child paths and perform different actions.

Step-by-Step Solution:
Step 1: Recall that the purpose of fork() is process creation, not process termination or thread creation.Step 2: Understand that the child process is almost an exact duplicate of the parent at the moment of fork(), with separate address space and independent program counter.Step 3: After fork(), both processes continue executing the next instruction after the system call, but they see different return values from fork().Step 4: Remember that additional calls such as exec() are often used in the child to load a new program image.Step 5: Choose the option that clearly describes fork() creating a new child process by duplicating the calling process.

Verification / Alternative check:
Simple C programs that call fork() and print the process identifier and return value demonstrate this behavior. Running such a program shows that the print statement executes twice, once in the parent and once in the child, with different process identifiers. Documentation and man pages for fork() state that on success, the call returns twice, once in the parent and once in the child, which matches the explanation given here.

Why Other Options Are Wrong:
Option B: Describes process termination, which is handled by exit() and related system calls, not by fork().Option C: Mentions formatting a disk, which is unrelated to process creation.Option D: Refers to thread creation within a process; thread libraries typically use functions such as pthread_create rather than fork() for that purpose.

Common Pitfalls:
One common confusion is between processes and threads. fork() creates a new process with its own address space, whereas threads share address space. Another mistake is to assume that fork() duplicates everything including process identifiers, when in fact the child has a new identifier. Understanding the different return values of fork() is also crucial; failing to test these correctly leads to bugs in parent or child process logic.

Final Answer:
The correct answer is It creates a new child process by duplicating the calling process, so that both parent and child continue execution with separate process identifiers.