Re: [RFC][PATCH 4/9] idle: Fix rcu_idle_*() usage

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, 19 May 2022 23:27:54 +0200 Peter Zijlstra wrote:
> @@ -150,12 +151,12 @@ static void enter_s2idle_proper(struct c
>  	 */
>  	stop_critical_timings();
>  	if (!(target_state->flags & CPUIDLE_FLAG_RCU_IDLE))
> -		rcu_idle_enter();
> +		cpuidle_rcu_enter();
>  	target_state->enter_s2idle(dev, drv, index);
>  	if (WARN_ON_ONCE(!irqs_disabled()))
> -		local_irq_disable();
> +		raw_local_irq_disable();
>  	if (!(target_state->flags & CPUIDLE_FLAG_RCU_IDLE))
> -		rcu_idle_exit();
> +		cpuidle_rcu_enter();

Why does it need to enter once more instead of exit, if it is not typo?

Hillf
>  	tick_unfreeze();
>  	start_critical_timings();
>  
> @@ -233,14 +234,14 @@ int cpuidle_enter_state(struct cpuidle_d
>  
>  	stop_critical_timings();
>  	if (!(target_state->flags & CPUIDLE_FLAG_RCU_IDLE))
> -		rcu_idle_enter();
> +		cpuidle_rcu_enter();
>  
>  	entered_state = target_state->enter(dev, drv, index);
>  	if (WARN_ON_ONCE(!irqs_disabled()))
>  		raw_local_irq_disable();
>  
>  	if (!(target_state->flags & CPUIDLE_FLAG_RCU_IDLE))
> -		rcu_idle_exit();
> +		cpuidle_rcu_exit();

This looks like the correct sequence in sanity.

>  	start_critical_timings();
>  
>  	sched_clock_idle_wakeup_event();




[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux