I don't know the history of threads/NPTL on Linux. However, the distinction betw...

cbd1984 · on May 22, 2015

> (Not sure if that's how fork() and friends are actually implemented under the hood.)

If you use strace you can see that it is. fork(2) and pthread_create(3) both show up as calls to clone(2).

cbd1984 · on May 22, 2015

Just for the edification of the masses:

pthread_create(3) looks something like this:

    clone(child_stack=0x7f79d754bff0,
    flags=CLONE_VM|CLONE_FS|CLONE_FILES|CLONE_SIGHAND|CLONE_THREAD|CLONE_SYSVSEM|CLONE_SETTLS|CLONE_PARENT_SETTID|CLONE_CHILD_CLEARTID,
    parent_tidptr=0x7f79d754c9d0,
    tls=0x7f79d754c700, child_tidptr=0x7f79d754c9d0) = 31230

(Newlines are mine.)

Of course, those pointers are only that size on a 64-bit architecture. The flags are where the real point of interest is.

fork(2) is like this:

    clone(child_stack=0,
    flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD,
    child_tidptr=0x7f543a9aaa10) = 17978

When implementing fork(2), the return value from clone(2) is the child's PID from the context of the parent process. When implementing pthread_create(3), the return value for the parent is still an integer value which is unique to the thread, and which strace uses as if it were a PID when it's tracing down the system calls of individual threads in separate files, which strace can do because it's awesome.

Some more information:

> Linux has a unique implementation of threads. To the Linux kernel, there is no concept of a thread. Linux implements all threads as standard processes. The Linux kernel does not provide any special scheduling semantics or data structures to represent threads. Instead, a thread is merely a process that shares certain resources with other processes. Each thread has a unique task_struct and appears to the kernel as a normal process (which just happens to share resources, such as an address space, with other processes).

http://www.makelinux.net/books/lkd2/ch03lev1sec3