2014-10-02 09:45:46

by David Vrabel

Subject: Re: [Xen-devel] [RFC PATCH 14/16] x86/xen: support poweroff through poweroff handler call chain

On 30/09/14 19:00, Guenter Roeck wrote:
> The kernel core now supports a poweroff handler call chain
> to remove power from the system. Call it if pm_power_off
> is set to NULL.
>
> Cc: Konrad Rzeszutek Wilk <[email protected]>
> Cc: Boris Ostrovsky <[email protected]>
> Cc: Thomas Gleixner <[email protected]>
> Cc: Ingo Molnar <[email protected]>
> Cc: H. Peter Anvin <[email protected]>
> Signed-off-by: Guenter Roeck <[email protected]>
> ---
> arch/x86/xen/enlighten.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
> index c0cb11f..645d00f 100644
> --- a/arch/x86/xen/enlighten.c
> +++ b/arch/x86/xen/enlighten.c
> @@ -1322,6 +1322,8 @@ static void xen_machine_power_off(void)
> {
> if (pm_power_off)
> pm_power_off();
> + else
> + do_kernel_poweroff();

Why isn't this if (pm_power_off) check in do_kernel_poweroff()?

That way when you finally remove pm_power_off you need only update one
place. A quick skim of the other archs suggest this would work for them
too.

David

2014-10-02 13:29:12

by Guenter Roeck

Subject: Re: [Xen-devel] [RFC PATCH 14/16] x86/xen: support poweroff through poweroff handler call chain

On 10/02/2014 02:45 AM, David Vrabel wrote:
> On 30/09/14 19:00, Guenter Roeck wrote:
>> The kernel core now supports a poweroff handler call chain
>> to remove power from the system. Call it if pm_power_off
>> is set to NULL.
>>
>> Cc: Konrad Rzeszutek Wilk <[email protected]>
>> Cc: Boris Ostrovsky <[email protected]>
>> Cc: Thomas Gleixner <[email protected]>
>> Cc: Ingo Molnar <[email protected]>
>> Cc: H. Peter Anvin <[email protected]>
>> Signed-off-by: Guenter Roeck <[email protected]>
>> ---
>> arch/x86/xen/enlighten.c | 2 ++
>> 1 file changed, 2 insertions(+)
>>
>> diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
>> index c0cb11f..645d00f 100644
>> --- a/arch/x86/xen/enlighten.c
>> +++ b/arch/x86/xen/enlighten.c
>> @@ -1322,6 +1322,8 @@ static void xen_machine_power_off(void)
>> {
>> if (pm_power_off)
>> pm_power_off();
>> + else
>> + do_kernel_poweroff();
>
> Why isn't this if (pm_power_off) check in do_kernel_poweroff()?
>
> That way when you finally remove pm_power_off you need only update one
> place. A quick skim of the other archs suggest this would work for them
> too.
>

Good idea. I'll do that for the next version of the patch set.

Guenter