spitz: add gpio button support (fixes regression)

[Eric Miao]

On Fri, Oct 16, 2009 at 5:52 AM, Pavel Machek <pavel at ucw.cz> wrote:
> On Thu 2009-10-15 23:47:32, Stanislav Brabec wrote:
>> Pavel Machek wrote:
>> >Well, resume is hit by second regression.
>> >
>> >Suspend key.... I restored it to the level of functionality in
>> >2.6.31 :-).
>>
>> > > +         .type   = EV_KEY,
>> > > +         .code   = KEY_SUSPEND,
>>
>> I guess it should be EV_PWR. Then it will at least go to suspend.
>
> Does it actually go to suspend with that change?
>
> (Actually, I wonder how it could have ever worked: userspace should
> certainly be in a loop there; kernel should not suspend machine on its
> own. In -rp kernels, kernel _does_ suspend on its own, but... I'm not
> sure if that's suitable for mainline).

EV_PWR is what's in the original spitzkbd.c, so I assume that's
something we should keep. And the reason that EV_PWR makes
it actually suspend might related the APM emulation things,
which I'm not very sure about.