Re: [PATCH v2 1/3] PM / Sleep: Make [un]lock_system_sleep() generic

From: Rafael J. Wysocki
Date: Tue Dec 06 2011 - 17:33:02 EST


On Tuesday, December 06, 2011, Srivatsa S. Bhat wrote:
> The [un]lock_system_sleep() APIs were originally introduced to mutually
> exclude memory hotplug and hibernation.
>
> Directly using mutex_lock(&pm_mutex) to achieve mutual exclusion with
> suspend or hibernation code can lead to freezing failures. However, the
> APIs [un]lock_system_sleep() can be safely used to achieve the same,
> without causing freezing failures.
>
> So, since it would be beneficial to modify all the existing users of
> mutex_lock(&pm_mutex) (in all parts of the kernel), so that they use these
> safe APIs intead, make these APIs generic by removing the restriction that
> they work only when CONFIG_HIBERNATE_CALLBACKS is set. Moreover, that
> restriction didn't buy us anything anyway.
>
> v2: Rebased on top of the patchset posted at:
> https://lkml.org/lkml/2011/12/6/334
>
> Suggested-by: Tejun Heo <tj@xxxxxxxxxx>
> Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@xxxxxxxxxxxxxxxxxx>

All patches in this series applied to linux-pm/linux-next.

Thanks,
Rafael


> ---
>
> include/linux/suspend.h | 36 ++++++++++++++++--------------------
> 1 files changed, 16 insertions(+), 20 deletions(-)
>
> diff --git a/include/linux/suspend.h b/include/linux/suspend.h
> index 906d62c..95040cc 100644
> --- a/include/linux/suspend.h
> +++ b/include/linux/suspend.h
> @@ -332,6 +332,8 @@ static inline bool system_entering_hibernation(void) { return false; }
> #define PM_RESTORE_PREPARE 0x0005 /* Going to restore a saved image */
> #define PM_POST_RESTORE 0x0006 /* Restore failed */
>
> +extern struct mutex pm_mutex;
> +
> #ifdef CONFIG_PM_SLEEP
> void save_processor_state(void);
> void restore_processor_state(void);
> @@ -352,6 +354,19 @@ extern bool events_check_enabled;
> extern bool pm_wakeup_pending(void);
> extern bool pm_get_wakeup_count(unsigned int *count);
> extern bool pm_save_wakeup_count(unsigned int count);
> +
> +static inline void lock_system_sleep(void)
> +{
> + freezer_do_not_count();
> + mutex_lock(&pm_mutex);
> +}
> +
> +static inline void unlock_system_sleep(void)
> +{
> + mutex_unlock(&pm_mutex);
> + freezer_count();
> +}
> +
> #else /* !CONFIG_PM_SLEEP */
>
> static inline int register_pm_notifier(struct notifier_block *nb)
> @@ -367,30 +382,11 @@ static inline int unregister_pm_notifier(struct notifier_block *nb)
> #define pm_notifier(fn, pri) do { (void)(fn); } while (0)
>
> static inline bool pm_wakeup_pending(void) { return false; }
> -#endif /* !CONFIG_PM_SLEEP */
> -
> -extern struct mutex pm_mutex;
>
> -#ifndef CONFIG_HIBERNATE_CALLBACKS
> static inline void lock_system_sleep(void) {}
> static inline void unlock_system_sleep(void) {}
>
> -#else
> -
> -/* Let some subsystems like memory hotadd exclude hibernation */
> -
> -static inline void lock_system_sleep(void)
> -{
> - freezer_do_not_count();
> - mutex_lock(&pm_mutex);
> -}
> -
> -static inline void unlock_system_sleep(void)
> -{
> - mutex_unlock(&pm_mutex);
> - freezer_count();
> -}
> -#endif
> +#endif /* !CONFIG_PM_SLEEP */
>
> #ifdef CONFIG_ARCH_SAVE_PAGE_KEYS
> /*
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-pm" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
>

--
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/