fork() vs. clone(): What is the difference?

fork()

• fork() is the traditional system call that creates a new process.
• It does not take any parameters.

1
2
 
#include <unistd.h>
pid_t fork(void);
 

clone()

The problem is clone(). It also creates a new (“child”) process, then how is it different from fork()?

By contrast with fork(2), these system calls provide more precise control over what pieces of execution context are shared between the calling process and the child process.¹

That is why it is more granular, since it provides control over the resources shared between the parent and child. The fork system call did not allow any additional control other than creating a process.

The main use of clone() is to implement threads, ² not create threads.³

While pthread_create() is the standard POSIX way to create threads, it often utilizes clone() internally. In other words, clone() is the base which is used for the implementation of pthread_create() and all the family of the fork() system calls.

Conclusion

• Use fork() if you only need to simply create a process in your code.
• clone() is something that is usually internally in Linux -> because it gives more control.
• Use pthread_create() to create a thread.

Reference