On Thu, 08 Dec 2011 08:35:12 -0800, Keith Packard <keithp at keithp.com> wrote: > On Thu, 08 Dec 2011 14:10:06 +0000, Chris Wilson <chris at chris-wilson.co.uk> wrote: > > > I had to remind myself why both the driver and the kernel are both > > touching the backlight across DPMS; the answer as I see it is that the > > kernel only knows about the raw backlight interface whereas the driver > > adjusts it via the preferred interface (which should handle the cases > > where the backlight modulation is handled independently of the PWM > > registers). > > Something in the kernel should be managing the backlight or we won't be > doing this right when fbdev is running. User space should not have to > know about the vagaries of the kernel backlight land. The tricky part is figuring out the right interface to use. Matthew was working upon such a mechanism to provide userspace with the ability to actually find the backlight controller for a respective output, but I don't believe there was ever a kernel API for i915 to call. -Chris -- Chris Wilson, Intel Open Source Technology Centre
