KASAN: use-after-free Read in task_is_descendant

[Kees Cook]

On Thu, Oct 25, 2018 at 12:13 PM, Oleg Nesterov <oleg at redhat.com> wrote:
> So again, suppose that "child" is already dead. Its task_struct can't be freed,
> but child->real_parent can point to the already freed memory.

I can't find a path for "child" to be released. I see task_lock()
always called on it before it ends up in Yama.

(Well, I can't find the lock for switch_mm(), but I assume that's safe
because it's switching to the task.)

> This means that the 1st walker = rcu_dereference(walker->real_parent) is fine,
> this simply reads the child->real_parent pointer, but on the second iteration
>
>         walker = rcu_dereference(walker->real_parent);
>
> reads the alredy freed memory.

What does rcu_read_lock() protect actually protect here? I thought
none of the task structs would be freed until after all
rcu_read_unlock() finished.

> OK. Lets ignore ptracer_exception_found() for the moment. Why do you think the
> patch below can't help?
>
> Oleg.
>
> --- x/security/yama/yama_lsm.c
> +++ x/security/yama/yama_lsm.c
> @@ -368,7 +368,8 @@ static int yama_ptrace_access_check(stru
>                         break;
>                 case YAMA_SCOPE_RELATIONAL:
>                         rcu_read_lock();
> -                       if (!task_is_descendant(current, child) &&
> +                       if (!pid_alive(child) ||
> +                           !task_is_descendant(current, child) &&
>                             !ptracer_exception_found(current, child) &&
>                             !ns_capable(__task_cred(child)->user_ns, CAP_SYS_PTRACE))
>                                 rc = -EPERM;
>

Hm, documentation there says:
 * If pid_alive fails, then pointers within the task structure
 * can be stale and must not be dereferenced.

What is the safe pattern for checking vs rcu?

-Kees

-- 
Kees Cook