Re: [PATCH] seqlock, procfs: Fix scoped_seqlock_read() critical section in do_task_stat()

[Oleg Nesterov]

On 12/02, Ingo Molnar wrote:
>
> RCU read-lock should not nest inside a read-seqlock
> irqsave ->stats_lock IRQs-off critical section,

Hmm... I agree with this patch, but is it actually wrong?

I thought that rcu_read_lock/unlock is safe under spin_lock_irq...

> +	{
> +		guard(rcu)();
> +		scoped_seqlock_read (&sig->stats_lock, ss_lock_irqsave) {

how about

	scoped_guard(rcu)  {
		scoped_seqlock_read (&sig->stats_lock, ss_lock_irqsave) {
			...
		}
	}
		
?

Oleg.