为什么我在 C 中调用 fork() 而是使用 sys_clone() 系统调用？ 为什么不是 sys_fork() 系统调用？

Question

Here is my source code test.c:

#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>

int main(){
    pid_t pid;

    pid=fork();
    if(pid==0){
        printf("new process\n");
    }else{
        printf("old process\n");
    }

    return 0;

}

I compiled it by:

gcc test.c -o fork

And I found it make sys_clone() syscall instead of sys_fork() by strace.

$strace -o my ./fork
new process
old process

$cat my | grep clone
clone(child_stack=NULL, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0x7f459f17d810) = 4415

$cat my | grep fork
execve("./fork", ["./fork"], 0x7ffc76ccbf00 /* 52 vars */) = 0

see clone() in the scripts? why not fork?

Answer 1

clone(2) is a Linux specific system call that superseded fork(2) system call. So clone is used for creating all the "process-like" entities (for creating threads, implementing vfork, posix_spawn and so on) on Linux. Basically, fork is equivalent to clone with specific flags. So that's why you see clone under strace .

Of course, the POSIX fork isn't deprecated and isn't going to disappear. So use of fork isn't an issue. Just that C library implementations on Linux use clone system call to implement fork .

Answer 2

In some versions of several C libraries for Linux (eg GNU libc or musl-libc ), the fork function is actually implemented by a call to clone

Since these libraries are open source, you could download their source code and study it (or even improve that source code)

See also ldd(1) , syscalls(2) , fork(2) , clone(2) , uname(2) , errno(3) , execve(2) , elf(5) , credentials(7) , ld.so(8) etc.... and read Advanced Linux Programming

Perhaps some C libraries are using clone or fork depending of the result of some initial call to uname(2)

Read also some good textbook on operating systems .