Re: [RFC v1 PATCH 4/6] input: pmic8058_pwrkey: Add support for powerkey

[Dmitry Torokhov]

On Fri, Nov 12, 2010 at 02:26:28PM +0530, Trilok Soni wrote:
> Hi Dmitry,
> 
> On 11/12/2010 6:27 AM, Dmitry Torokhov wrote:
> > On Thu, Nov 11, 2010 at 05:30:21PM +0530, Trilok Soni wrote:
> >> Hi Dmitry,
> >>
> >> On 11/11/2010 12:51 PM, Dmitry Torokhov wrote:
> >>> Hi Trilkok,
> >>>
> >>> On Wed, Nov 10, 2010 at 06:17:59PM +0530, Trilok Soni wrote:
> >>>> Add support for PMIC8058 power key driven over dedicated KYPD_PWR_N
> >>>> pin. It allows the user to specify the amount of time by which the
> >>>> power key reporting can be delayed.
> >>>>
> >>>
> >>> Why do we need to delay KEY_POWER reporting? Do we need to use high
> >>> resolution timers or regular timers would do as well? KEY_END
> >>> appears to be abused (you don't want to move your cursor to the end
> >>> of line, do you?). Also I wonder if header file should reside in
> >>> linux/mfd with the rest of pmic8058 components.
> >>
> >> Most of the time Mobile devices come with single physical key for
> >> POWER, which if pressed for less than 500ms (configurable) then it
> >> will only report KEY_END (which say locks the screen on mobile) and if
> >> it pressed more than 500ms then it will also report KEY_POWER event
> >> too, which will say display menu on your mobile for asking you to
> >> suspend/switch off/etc, operations.
> >>
> > 
> > I see,. If you would have used KEY_SCREENLOCK iinstead of KEY_END I
> > would likely not ask this question ;)
> > 
> 
> KEY_SCRENNLOCK looks good, let me analyze the impact on userspace framework
> which I have. I will come back on this in a day.
> 
> >> For the timers I can move from hrtimers to regular timers.
> >>
> >> For the header file, I can move them to include/linux/mfd too. No
> >> problem on that.
> >>
> > 
> > I am not even sure we need to keep them in separate header files, but it
> > is up to you.
> 
> Do you suggest that all the MFD sub-devices's platform data structures should come from single
> header file?
> 

It is an option, depends on how many external headers are needed, etc.

When I looked at this particular file I got the feeling that it could be
folded together with the rest. I expect that the board code that will
specify platform resources will include every one of this sub-files
anyway. But maybe if I was presented with the combined header I'd say
"wow, thats too big"...

Like I said, it is up to you.

-- 
Dmitry
--
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/