What is the difference between clone and fork system calls in Linux?

In Linux, both clone and fork are system calls used to create new processes, but they have different behaviors and use cases:

fork

• Purpose: Creates a new process by duplicating the calling process.
• Process Creation: The new process (child) is an exact copy of the parent process, with its own memory space.
• Memory Sharing: The child process does not share memory with the parent; it gets its own copy of the parent's memory.
• Return Value: Returns the child's PID to the parent process and 0 to the child process.

clone

• Purpose: Creates a new process or thread with more control over what is shared between the parent and child.
• Process Creation: Can create a new process or a thread, depending on the flags passed to it.
• Memory Sharing: Allows sharing of memory, file descriptors, and other resources between the parent and child processes based on the specified flags.
• Return Value: Returns the child's PID to the parent process and 0 to the child process, similar to fork.

Summary

• Use fork when you want a completely independent process.
• Use clone when you need fine-grained control over resource sharing, such as when implementing threads.

Example

Here’s a simple example of using fork and clone:

#include <stdio.h>
#include <unistd.h>
#include <sys/types.h>
#include <sched.h>

int child_function() {
    printf("This is the child process created by clone.\n");
    return 0;
}

int main() {
    pid_t pid = fork();
    if (pid == 0) {
        // Child process
        printf("This is the child process created by fork.\n");
    } else {
        // Parent process
        printf("This is the parent process.\n");
    }

    // Using clone
    int stack[1024];
    pid_t clone_pid = clone(child_function, stack + sizeof(stack), SIGCHLD, NULL);
    if (clone_pid == 0) {
        // Child process created by clone
        printf("This is the child process created by clone.\n");
    }

    return 0;
}

In this example, fork creates a separate child process, while clone can be used to create a child process that shares resources with the parent.