Re: [PATCH 08/16] freezer: use dedicated lock instead oftask_lock() + memory barrier

[Oleg Nesterov]

On 08/28, Oleg Nesterov wrote:
>
> Off-topic, but fake_signal_wake_up() is not safe if the caller
> try_to_freeze_cgroup(). Unlike try_to_freeze_tasks() (which holds
> tasklist) we can race with the exiting thread, ->sighand can be
> NULL.

Although with PF_NOFREEZE in do_exit() from 6/16, this race is only
theoretical.

And, forgot to mention, __thaw_task() looks racy too, but i think
this recalc_sigpending_and_wake() can simply go away.

Oleg.

--
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/