I would love to see something like this getting in somehow, but as Nate described, we run into a problem with literally the first person that tested the patch. That sounds statistically bad.

I did not find a way to ensure that a given minimal step would turn on the screen when starting from 0. My guess is that there are some graphical cards out there that will behave weird and turn on the screen at an arbitrary level. Unless testing this first on a LOT of hardware, it seems difficult to find a value suitable for all.

In D19678#429348, @thsurrel wrote:

I would love to see something like this getting in somehow, but as Nate described, we run into a problem with literally the first person that tested the patch. That sounds statistically bad.

I did not find a way to ensure that a given minimal step would turn on the screen when starting from 0. My guess is that there are some graphical cards out there that will behave weird and turn on the screen at an arbitrary level. Unless testing this first on a LOT of hardware, it seems difficult to find a value suitable for all.

Maybe we should look a level deeper and figure out why we're getting nonzero values that still have the backlight off? Could be a kernel bug.

Yeah, that would help if I had a laptop that has this behavior, but I couldn't find any.

Well I've got one! Unfortunately my kernel-fu is non-existent, but I can follow instructions and commands like a trained monkey pretty well. :)

And I could guide you around the kernel except in the graphic area ...