[Intel-gfx] New Samsung driver for backlight control.

[Greg KH]

On Fri, Sep 11, 2009 at 03:14:46AM +0100, Mike Lothian wrote:
> 2009/9/9 Greg KH <greg at kroah.com>:
> > On Wed, Sep 09, 2009 at 01:04:15PM +0200, Matthias Hopf wrote:
> >> On Sep 08, 09 09:31:24 -0700, Greg KH wrote:
> >> > > AFAIR at one point you deliberately decided to expose only 8 levels to
> >> > > the user - while the hardware was capable of more levels (I think it was
> >> > > something about some user land tool expecting something like that).
> >> > > But I may be utterly wrong.
> >> >
> >> > Yes, that was when we were blindly poking in the PCI config space.  We
> >> > aren't supposed to be doing that :)
> >> >
> >> > > I'd love to have more fine grained control on my NC10 as well.
> >> >
> >> > I have reports that Windows also only supports 8 levels, so that is what
> >> > this hardware controller is possible of supporting, sorry.
> >>
> >> Hm. This is contradicting. On the one hand the hardware is capable of
> >> more levels. On the other it isn't *reporting* more levels.
> >>
> >> I wouldn't rely on anything what Windows supports. There can be
> >> arbitrary limitations in the driver.
> >
> > The hardware seems to allow multiple levels if you poke a raw value into
> > a random PCI config location.
> >
> > However, this is not the correct way to control this hardware, a SMI
> > call needs to be made instead.  And for that call, only 8 levels are
> > available.
> >
> > So it's not a limitation in the Windows driver, it is a limitation in
> > the BIOS provided by the manufacturer.
> >
> > thanks,
> >
> > greg k-h
> >
> 
> Who's the best person to talk to about getting my laptop added to
> 
> drivers/input/keyboard/atkbd.c
> 
> To stop the stuck keys

The input maintainer and developers as found in the kernel MAINTAINERS
file?