2006-02-01 18:26:27

by Oleg Nesterov

Subject: [PATCH rc1-mm] simplify exec from init's subthread

[ On top of "exec: allow init to exec from any thread.",
filename exec-allow-init-to-exec-from-any-thread.patch ]

I think it is enough to take tasklist_lock for reading while
changing child_reaper:

Reparenting needs write_lock(tasklist_lock)

Only one thread in a thread group can do exec()

sighand->siglock garantees that get_signal_to_deliver()
will not see a stale value of child_reaper.

This means that we can change child_reaper earlier, without
calling zap_other_threads() twice.

"child_reaper = current" is a NOOP when init does exec from
main thread, we don't care.

Signed-off-by: Oleg Nesterov <[email protected]>

--- RC-1/fs/exec.c~ 2006-02-01 23:38:20.000000000 +0300
+++ RC-1/fs/exec.c 2006-02-02 00:39:40.000000000 +0300
@@ -616,6 +616,15 @@ static int de_thread(struct task_struct
kmem_cache_free(sighand_cachep, newsighand);
return -EAGAIN;
}
+
+ /*
+ * child_reaper ignores SIGKILL, change it now.
+ * Reparenting needs write_lock on tasklist_lock,
+ * so it is safe to do it under read_lock.
+ */
+ if (unlikely(current->group_leader == child_reaper))
+ child_reaper = current;
+
zap_other_threads(current);
read_unlock(&tasklist_lock);

@@ -660,23 +669,12 @@ static int de_thread(struct task_struct
struct dentry *proc_dentry1, *proc_dentry2;
unsigned long ptrace;

- leader = current->group_leader;
- /*
- * If our leader is the child_reaper become
- * the child_reaper and resend SIGKILL signal.
- */
- if (unlikely(leader == child_reaper)) {
- write_lock(&tasklist_lock);
- child_reaper = current;
- zap_other_threads(current);
- write_unlock(&tasklist_lock);
- }
-
/*
* Wait for the thread group leader to be a zombie.
* It should already be zombie at this point, most
* of the time.
*/
+ leader = current->group_leader;
while (leader->exit_state != EXIT_ZOMBIE)
yield();

2006-02-02 04:40:21

by Eric W. Biederman

Subject: Re: [PATCH rc1-mm] simplify exec from init's subthread

Oleg Nesterov <[email protected]> writes:

> [ On top of "exec: allow init to exec from any thread.",
> filename exec-allow-init-to-exec-from-any-thread.patch ]
>
> I think it is enough to take tasklist_lock for reading while
> changing child_reaper:
>
> Reparenting needs write_lock(tasklist_lock)
>
> Only one thread in a thread group can do exec()
>
> sighand->siglock garantees that get_signal_to_deliver()
> will not see a stale value of child_reaper.
>
> This means that we can change child_reaper earlier, without
> calling zap_other_threads() twice.
>
> "child_reaper = current" is a NOOP when init does exec from
> main thread, we don't care.

Looks good. I fat fingered my locking anyway, and this
looks like a good fix, as well as the more obvious place,
to do this.

Acked-by: Eric Biederman <[email protected]>