Re: Q: perf_event && event->owner

[Oleg Nesterov]

On 11/09, Peter Zijlstra wrote:
>
> I think you're right, how about something like this?

I need to read it with a fresh head ;)

At first glance,

> @@ -2254,6 +2249,12 @@ static int perf_release(struct inode *in
>
>  	file->private_data = NULL;
>
> +	if (event->owner) {
> +		mutex_lock(&event->owner->perf_event_mutex);
> +		list_del_init(&event->owner_entry);
> +		mutex_unlock(&event->owner->perf_event_mutex);
> +	}

Agreed, it is better to do this in perf_release().

But, this can use the already freed task_struct, event->owner.

Either sys_perf_open() should do get_task_struct() like we currently
do, or perf_event_exit_task() should clear event->owner and then
perf_release() should do something like

	rcu_read_lock();
	owner = event->owner;
	if (owner)
		get_task_struct(owner);
	rcu_read_unlock();

	if (owner) {
		mutex_lock(&event->owner->perf_event_mutex);
		list_del_init(&event->owner_entry);
		mutex_unlock(&event->owner->perf_event_mutex);
		put_task_struct(owner);
	}

Probably this can be simplified...

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/