lkml.org 
[lkml]   [2000]   [Aug]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    Date
    From
    SubjectRe: SCO: "thread creation is about a thousand times faster than on
    On 27 Aug 2000, Linus Torvalds wrote:


    >
    > Btw, here's a simple thread-group patch. It doesn't actually _do_
    > anything with the thread group part, but this is the kind of frame-work
    > I've envisioned. Completely untested, of course, but with these kinds
    > of security issues taken into account.
    >
    > Comments? At least it's small and simple..
    >
    > Linus



    And what will be done, when thread-group leader (tgig==pid) will exec() ?
    If I read your core correctly, it will be de_threaded by seting tgid=pid,
    but it was before! So the process will _not_ be removed from thread group.
    and it will remain in thread group _after_ exec.
    Am I mistaken?


    Martin





    >
    > ----
    > diff -u --recursive --new-file penguin/linux/fs/exec.c linux/fs/exec.c
    > --- penguin/linux/fs/exec.c Mon Aug 14 10:19:55 2000
    > +++ linux/fs/exec.c Sun Aug 27 17:29:52 2000
    > @@ -496,6 +496,25 @@
    > write_unlock(&files->file_lock);
    > }
    >
    > +/*
    > + * An execve() will automatically "de-thread" the process.
    > + * Note: we don't have to hold the tasklist_lock to test
    > + * whether we migth need to do this. If we're not part of
    > + * a thread group, there is no way we can become one
    > + * dynamically. And if we are, we only need to protect the
    > + * unlink - even if we race with the last other thread exit,
    > + * at worst the list_del_init() might end up being a no-op.
    > + */
    > +static inline void de_thread(struct task_struct *tsk)
    > +{
    > + if (!list_empty(&tsk->thread_group)) {
    > + write_lock_irq(&tasklist_lock);
    > + list_del_init(&tsk->thread_group);
    > + write_unlock_irq(&tasklist_lock);
    > + }
    > + tsk->tgid = tsk->pid;
    > +}
    > +
    > int flush_old_exec(struct linux_binprm * bprm)
    > {
    > char * name;
    > @@ -533,6 +552,8 @@
    > current->comm[i] = '\0';
    >
    > flush_thread();
    > +
    > + de_thread(current);
    >
    > if (bprm->e_uid != current->euid || bprm->e_gid != current->egid ||


    --------------------------------------------------------------------------------
    Martin "MaD" Douda
    WEB: http://martin.douda.net/ PHONE:+420603752779 ICQ# 86467013
    EMAIL: <martin@douda.net>, <mad@gate.mobil.cz> (160 characters only)
    PGP:ID=0x6FE43023 Fingerprint:E495 11DA EF6E 0DD6 965A 54F3 888E CC9E 6FE4 3023
    --------------------------------------------------------------------------------
    The only thing good about "standards" in computer science is that there are so
    many to choose from.

    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    Please read the FAQ at http://www.tux.org/lkml/

    \
     
     \ /
      Last update: 2005-03-22 12:38    [W:0.028 / U:30.908 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site