Re: [PATCH] Allow memory hotplug and hibernation in the same kernel

[Rafael J. Wysocki]

On Friday 13 November 2009, Andi Kleen wrote:
> Allow memory hotplug and hibernation in the same kernel
> 
> Memory hotplug and hibernation was excluded in Kconfig. This is obviously
> a problem for distribution kernels who want to support both in the same
> image.
> 
> After some discussions with Rafael and others the only problem is 
> with parallel memory hotadd or removal while a hibernation operation
> is in process. It was also working for s390 before.
> 
> This patch removes the Kconfig level exclusion, and simply
> makes the memory add / remove functions grab the pm_mutex
> to exclude against hibernation.
> 
> This is a 2.6.32 candidate.
> 
> Cc: gerald.schaefer@xxxxxxxxxx
> Cc: rjw@xxxxxxx
> 
> Signed-off-by: Andi Kleen <ak@xxxxxxxxxxxxxxx>
> 
> ---
>  include/linux/suspend.h |   21 +++++++++++++++++++--
>  mm/Kconfig              |    5 +----
>  mm/memory_hotplug.c     |   21 +++++++++++++++++----
>  3 files changed, 37 insertions(+), 10 deletions(-)
> 
> Index: linux-2.6.32-rc6-ak/include/linux/suspend.h
> ===================================================================
> --- linux-2.6.32-rc6-ak.orig/include/linux/suspend.h
> +++ linux-2.6.32-rc6-ak/include/linux/suspend.h
> @@ -301,6 +301,8 @@ static inline int unregister_pm_notifier
>  #define pm_notifier(fn, pri)	do { (void)(fn); } while (0)
>  #endif /* !CONFIG_PM_SLEEP */
>  
> +extern struct mutex pm_mutex;
> +
>  #ifndef CONFIG_HIBERNATION
>  static inline void register_nosave_region(unsigned long b, unsigned long e)
>  {
> @@ -308,8 +310,23 @@ static inline void register_nosave_regio
>  static inline void register_nosave_region_late(unsigned long b, unsigned long e)
>  {
>  }
> -#endif
>  
> -extern struct mutex pm_mutex;
> +static inline void lock_hibernation(void) {}
> +static inline void unlock_hibernation(void) {}
> +
> +#else
> +
> +/* Let some subsystems like memory hotadd exclude hibernation */
> +
> +static inline void lock_hibernation(void)
> +{
> +	mutex_lock(&pm_mutex);
> +}
> +
> +static inline void unlock_hibernation(void)
> +{
> +	mutex_unlock(&pm_mutex);
> +}

This also is going to affect suspend to RAM, which kind of makes sense BTW,
so I'd not put it under the #ifdef.  Also, the names should reflect the fact
that suspend is affected too.  What about block|unblock_system_sleep()?

> +#endif
>  
>  #endif /* _LINUX_SUSPEND_H */

The rest of the patch is fine by me.

Thanks,
Rafael
--
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/