Re: [path][rfc] add PR_DETACH prctl command [2/2]

[Stas Sergeev]

20.04.2011 23:33, Oleg Nesterov wrote:

> I still do not understand the point of PR_DETACH. Why do you think it is
> needed without reparenting? OK, I do not really care ;)
Hmm, but that's really interesting, I wonder why do
you ask this. In my eyes, the reparenting was just a
"trick", which is now avoided for good. And I saved
a lot of per-thread memory by throwing that, and
shrunk the patch twice. Why do you think no-reparenting
changed something real? Could this change any use case?

> > > And. To hide the pr_detached task from do_wait(). you changed
> > > do_notify_parent() to returnd DEATH_REAP.
> > No, its hidden by the check in wait_consider_task().
> > do_notify_parent() was changed only to not allow the
> > second notification to the same parent.
> Not only. Please look at your own code ;) wait_consider_task() checks
> exit_state == EXIT_ZOMBIE before p->pr_detached, and thus do_notify_parent()
> haas to return DEATH_REAP so that the caller will set EXIT_DEAD. Otherwise
> the old parent could see EXIT_ZOMBIE&&  pr_detached task again.
Yes, but that's not to hide from do_wait().
At least as far as I understand, exit_notify() does
release_task() in this case, so that's not hiding: I
literally terminate the child this way.
Or am I missing something?

> But. What if the child stops after PR_DETACH? It will notify the parent
> anyway, no?
Ah, so that should be disallowed too, will fix.

> > > task_pid_vnr(p). Hmm, the change in reparent_leader doesn't look right,
> > > at least in case we reparent to sub-thread.
> > Why not? Whereever we reparented, I allow reparenting again.
> Even if the child reparents to the original parent's sub-thread?
OK. :)

> Everything. ptrace relies on do_wait(). wait_consider_task() doesn't
> work if pr_detached (except for WEXITED of course).
OK.

> > I am not that sure the last ones are very buggy.
> Well, I am not sure.
OK, at least the above 2 bugs are trivial to fix... :))
Thanks!

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