On 10/02, Sukadev Bhattiprolu wrote:
>
> --- linux-2.6.orig/kernel/exit.c 2009-10-02 16:36:47.000000000 -0700
> +++ linux-2.6/kernel/exit.c 2009-10-02 17:19:06.000000000 -0700
> @@ -738,8 +738,19 @@ static struct task_struct *find_new_reap
> static void reparent_thread(struct task_struct *father, struct task_struct *p,
> struct list_head *dead)
> {
> - if (p->pdeath_signal)
> - group_send_sig_info(p->pdeath_signal, SEND_SIG_NOINFO, p);
> + struct siginfo info;
> +
> + if (p->pdeath_signal) {
> + info.si_signo = p->pdeath_signal;
> + info.si_errno = 0;
> +
> + rcu_read_lock();
> + info.si_pid = task_pid_nr_ns(father, task_active_pid_ns(p));
> + info.si_uid = __task_cred(father)->uid;
> + rcu_read_unlock();
> +
> + group_send_sig_info(p->pdeath_signal, &info, p);
> + }
This patch forgets to set "info.si_code = SI_USER", this means SI_FROMXXX()
will return the random value.
Imho it would be more clean to move the declaration of "info" under
"if (pdeath_signal)".
Actually, perhaps we should consider SEND_SIG_NOINFO as SI_FROMUSER().
I'll try to check tomorrow.
Oleg.
_______________________________________________
Containers mailing list
Containers@xxxxxxxxxxxxxxxxxxxxxxxxxx
https://lists.linux-foundation.org/mailman/listinfo/containers
