Process creation (fork and exec) in Operating Systems - Step-by-Step Execution

Concept Flow - Process creation (fork and exec)

Start: Running Process

↓

Call fork()

↓

Create Child Process

↓

Child: Returns 0 from fork()

↓

Call exec() to run new program

↓

Replace Child's code with new program

↓

Parent: Returns child's PID from fork()

↓

Parent continues original program

The original process calls fork() to create a child. The child gets 0 return and can call exec() to run a new program. The parent gets child's ID and continues.

Execution Sample

Operating Systems

pid = fork()
if pid == 0:
    exec("/bin/ls")
else:
    wait()

This code creates a child process with fork. The child runs 'ls' using exec, the parent waits for child to finish.

Analysis Table

Step	Process	Action	fork() Return	exec() Called	Resulting State
1	Parent	Calls fork()	N/A	No	Parent running, fork called
2	Parent	fork() returns child's PID	Child PID (e.g. 1234)	No	Parent continues original code
2	Child	fork() returns 0	0	No	Child starts execution
3	Child	Calls exec("/bin/ls")	0	Yes	Child replaced by 'ls' program
4	Parent	Calls wait()	Child PID	No	Parent waits for child to finish
5	Child	Runs 'ls' program	0	Yes	Child runs new program until exit
6	Parent	wait() returns after child exits	Child PID	No	Parent resumes after child ends

💡 Execution stops when child finishes exec program and parent finishes wait.

State Tracker

Variable	Start	After fork()	After exec()	After wait()	Final
pid	undefined	0 in child, child's PID in parent	0 in child (exec replaces code, pid unchanged)	child's PID in parent	child's PID in parent

Key Insights - 3 Insights

Why does fork() return 0 in the child but a positive number in the parent?

fork() returns 0 in the child process to indicate it is the new child. The parent gets the child's process ID (a positive number). See execution_table step 2.

What happens to the child's code after exec() is called?

exec() replaces the child's current program code with a new program. The child's process ID stays the same, but the code and data change. See execution_table step 3 and 5.

Why does the parent call wait() after fork()?

The parent calls wait() to pause and let the child finish its new program before continuing. This prevents the child from becoming a 'zombie' process. See execution_table step 4 and 6.

Visual Quiz - 3 Questions

Look at the execution_table at step 2. What value does fork() return in the child process?

A -1

B Child's PID

C 0

D Undefined

💡 Hint

Check the 'fork() Return' column for the child process at step 2.

At which step does the child's program get replaced by the new program?

A Step 3

B Step 1

C Step 5

D Step 6

💡 Hint

Look for when exec() is called in the 'exec() Called' column.

If the parent did not call wait(), what would happen according to the execution flow?

A Parent would immediately exit before child finishes

B Child would become a zombie process

C Parent would wait indefinitely

D Child would never run exec()

💡 Hint

Refer to key_moments about why wait() is important to prevent zombie processes.

Concept Snapshot

fork() creates a new child process by duplicating the parent.
In the child, fork() returns 0; in the parent, it returns child's PID.
exec() replaces the child's code with a new program.
Parent can call wait() to pause until child finishes.
This sequence creates a new process running a different program.

Full Transcript

Process creation in operating systems often uses fork and exec. First, a running process calls fork(), which creates a new child process. The fork call returns 0 in the child and the child's process ID in the parent. The child can then call exec() to replace its program with a new one, such as running a different command. Meanwhile, the parent can call wait() to pause until the child finishes. This method allows one process to create another and run a different program, managing execution order and resources.