Re: [PATCH] acpi: video: improve quirk check

[Felipe Contreras]

On Sat, Aug 3, 2013 at 8:47 PM, Aaron Lu <aaron.lwe@xxxxxxxxx> wrote:
> On Sun, Aug 4, 2013 at 6:20 AM, Felipe Contreras
> <felipe.contreras@xxxxxxxxx> wrote:
>> On Sat, Aug 3, 2013 at 4:40 PM, Rafael J. Wysocki <rjw@xxxxxxx> wrote:

>>> Do we still need to revert commit efaa14c if this patch is applied?
>>
>> I guess not. At least in this machine changing the backlight works
>> correctly, whereas in v3.11-rc3 it was all weird, and v3.7-v3.10
>> didn't work at all. I cannot see how it would affect negatively other
>> machines.
>
> That commit makes hotkey emit notifications, and it's not the
> problem of "booting into a black screen", that problem is due to
> broken _BQC.

The broken _BQC has been there for quite some time, hasn't it?

Either way, without efaa14c, changing the backlight doesn't work at
all either way, so there's no black screen, because there cannot be.

> BTW, the efaa14c will also make screen off at level 0 according
> to Felipe, who consider this is a bug. But since it is required to
> let firmware emit notifications on hotkey press, I think user will
> want it.

With or without efaa14c, level 0 makes the screen off, or at least it
would, if the control worked at all. So efaa14c is one step forward,
but two back, my patch removes the two steps back, but we are still
not at the level of what my blacklisting patch does, for that we would
need to fix two issues:

1. Fix the retrieval of the last level at boot
2. Fix level 0 (yes, I consider that a regression)

But we cannot achieve either of those for v3.11, the only
possibilities seem to be either a) revert efaa14c, or b) keep it and
apply my patch. Anything else doesn't seem to be a possible or
sensible option, and I vote for b).

-- 
Felipe Contreras
--
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/