On 02/09/12 12:19, Gleb Natapov wrote:
> On Thu, Feb 09, 2012 at 11:45:21AM +0100, Gerd Hoffmann wrote:
>> On 02/09/12 09:48, Gleb Natapov wrote:
>>> On Wed, Feb 08, 2012 at 12:00:14PM +0100, Gerd Hoffmann wrote:
>>>>  * qemu_system_wakeup_request is supposed to be called on events which
>>>>    should wake up the guest.
>>>>
>>> qemu_system_wakeup_request() should get wakeup source as a parameter.
>>> There are ways to report it to a guest.
>>
>> Can we do that incrementally, when we actually implement the guest
>> reporting?
>>
> What do you mean by "when we actually implement the guest
> reporting"? The guest reporting is part of ACPI spec and implemented
> by all relevant guests. I think that adding wakeup source parameter to
> qemu_system_wakeup_request() and reporting RTC_STS and PWRBTN_STS should
> not complicate your patch series to much. I agree that DSDT magic required
> by other devices can wait for later.

Incremental patch (just infrastructure, no acpi windup yet) attached.
Something like this?

cheers,
  Gerd