Re: [RFC PATCH] cpu_pm/rt: replace rt rwlock with raw spinlock

[Alex Shi]

It's a serious bug which cause arm/arm64 rt boot failed.
Anyone like to give a glance?

Thanks
Alex

On 06/14/2017 09:22 PM, Alex Shi wrote:
> This is a quick fix for a bug as 'scheduling while atomic' or
> 'scheduling from the idle thread' on arm/arm64.
> 
> On arm/arm64, rwlock cpu_pm_notifier_lock in cpu_pm cause a potential
> schedule after irq disable in idle call chain:
> 
> cpu_startup_entry
>   cpu_idle_loop
>     local_irq_disable()
>     cpuidle_idle_call
>       call_cpuidle
>         cpuidle_enter
>           cpuidle_enter_state
>             ->enter :arm_enter_idle_state
> 	      cpu_pm_enter/exit
> 	        CPU_PM_CPU_IDLE_ENTER
>                   read_lock(&cpu_pm_notifier_lock); <-- sleep in idle
> 	             __rt_spin_lock();
> 			schedule();
> 
> The kernel panic is here:
> [    4.609601] BUG: scheduling while atomic: swapper/1/0/0x00000002
> [    4.609608] [<ffff0000086fae70>] arm_enter_idle_state+0x18/0x70
> [    4.609614] Modules linked in:
> [    4.609615] [<ffff0000086f9298>] cpuidle_enter_state+0xf0/0x218
> [    4.609620] [<ffff0000086f93f8>] cpuidle_enter+0x18/0x20
> [    4.609626] Preemption disabled at:
> [    4.609627] [<ffff0000080fa234>] call_cpuidle+0x24/0x40
> [    4.609635] [<ffff000008882fa4>] schedule_preempt_disabled+0x1c/0x28
> [    4.609639] [<ffff0000080fa49c>] cpu_startup_entry+0x154/0x1f8
> [    4.609645] [<ffff00000808e004>] secondary_start_kernel+0x15c/0x1a0
> 
> Daniel Lezcano said this notification is needed on arm/arm64 platforms.
> I also tried use local_lock_irq to replace local_irq_disable, but my 2
> boards just die without any output. So maybe it's only quick way to
> make rt kernel work on arm/arm64.
> 
> Since this is quick fix, instead of split out the raw rwlock, to use
> raw_spin_lock is simple and don't cost much.
> 
> Signed-off-by: Alex Shi <alex.shi@xxxxxxxxxx>
> Cc: Sebastian Andrzej Siewior <bigeasy@xxxxxxxxxxxxx>
> Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> Cc: Anders Roxell <anders.roxell@xxxxxxxxxx>
> Cc: Daniel Lezcano <daniel.lezcano@xxxxxxxxxx>
> Cc: linux-rt-users <linux-rt-users@xxxxxxxxxxxxxxx>
> ---
>  kernel/cpu_pm.c | 26 +++++++++++++-------------
>  1 file changed, 13 insertions(+), 13 deletions(-)
> 
> diff --git a/kernel/cpu_pm.c b/kernel/cpu_pm.c
> index 009cc9a..8ffa13e3 100644
> --- a/kernel/cpu_pm.c
> +++ b/kernel/cpu_pm.c
> @@ -22,7 +22,7 @@
>  #include <linux/spinlock.h>
>  #include <linux/syscore_ops.h>
>  
> -static DEFINE_RWLOCK(cpu_pm_notifier_lock);
> +static DEFINE_RAW_SPINLOCK(cpu_pm_notifier_lock);
>  static RAW_NOTIFIER_HEAD(cpu_pm_notifier_chain);
>  
>  static int cpu_pm_notify(enum cpu_pm_event event, int nr_to_call, int *nr_calls)
> @@ -50,9 +50,9 @@ int cpu_pm_register_notifier(struct notifier_block *nb)
>  	unsigned long flags;
>  	int ret;
>  
> -	write_lock_irqsave(&cpu_pm_notifier_lock, flags);
> +	raw_spin_lock_irqsave(&cpu_pm_notifier_lock, flags);
>  	ret = raw_notifier_chain_register(&cpu_pm_notifier_chain, nb);
> -	write_unlock_irqrestore(&cpu_pm_notifier_lock, flags);
> +	raw_spin_unlock_irqrestore(&cpu_pm_notifier_lock, flags);
>  
>  	return ret;
>  }
> @@ -72,9 +72,9 @@ int cpu_pm_unregister_notifier(struct notifier_block *nb)
>  	unsigned long flags;
>  	int ret;
>  
> -	write_lock_irqsave(&cpu_pm_notifier_lock, flags);
> +	raw_spin_lock_irqsave(&cpu_pm_notifier_lock, flags);
>  	ret = raw_notifier_chain_unregister(&cpu_pm_notifier_chain, nb);
> -	write_unlock_irqrestore(&cpu_pm_notifier_lock, flags);
> +	raw_spin_unlock_irqrestore(&cpu_pm_notifier_lock, flags);
>  
>  	return ret;
>  }
> @@ -100,7 +100,7 @@ int cpu_pm_enter(void)
>  	int nr_calls;
>  	int ret = 0;
>  
> -	read_lock(&cpu_pm_notifier_lock);
> +	raw_spin_lock(&cpu_pm_notifier_lock);
>  	ret = cpu_pm_notify(CPU_PM_ENTER, -1, &nr_calls);
>  	if (ret)
>  		/*
> @@ -108,7 +108,7 @@ int cpu_pm_enter(void)
>  		 * PM entry who are notified earlier to prepare for it.
>  		 */
>  		cpu_pm_notify(CPU_PM_ENTER_FAILED, nr_calls - 1, NULL);
> -	read_unlock(&cpu_pm_notifier_lock);
> +	raw_spin_unlock(&cpu_pm_notifier_lock);
>  
>  	return ret;
>  }
> @@ -130,9 +130,9 @@ int cpu_pm_exit(void)
>  {
>  	int ret;
>  
> -	read_lock(&cpu_pm_notifier_lock);
> +	raw_spin_lock(&cpu_pm_notifier_lock);
>  	ret = cpu_pm_notify(CPU_PM_EXIT, -1, NULL);
> -	read_unlock(&cpu_pm_notifier_lock);
> +	raw_spin_unlock(&cpu_pm_notifier_lock);
>  
>  	return ret;
>  }
> @@ -159,7 +159,7 @@ int cpu_cluster_pm_enter(void)
>  	int nr_calls;
>  	int ret = 0;
>  
> -	read_lock(&cpu_pm_notifier_lock);
> +	raw_spin_lock(&cpu_pm_notifier_lock);
>  	ret = cpu_pm_notify(CPU_CLUSTER_PM_ENTER, -1, &nr_calls);
>  	if (ret)
>  		/*
> @@ -167,7 +167,7 @@ int cpu_cluster_pm_enter(void)
>  		 * PM entry who are notified earlier to prepare for it.
>  		 */
>  		cpu_pm_notify(CPU_CLUSTER_PM_ENTER_FAILED, nr_calls - 1, NULL);
> -	read_unlock(&cpu_pm_notifier_lock);
> +	raw_spin_unlock(&cpu_pm_notifier_lock);
>  
>  	return ret;
>  }
> @@ -192,9 +192,9 @@ int cpu_cluster_pm_exit(void)
>  {
>  	int ret;
>  
> -	read_lock(&cpu_pm_notifier_lock);
> +	raw_spin_lock(&cpu_pm_notifier_lock);
>  	ret = cpu_pm_notify(CPU_CLUSTER_PM_EXIT, -1, NULL);
> -	read_unlock(&cpu_pm_notifier_lock);
> +	raw_spin_unlock(&cpu_pm_notifier_lock);
>  
>  	return ret;
>  }
>