Re: [REGRESSION/PATCH] acpi: blacklist win8 OSI for ASUS Zenbok Prime UX31A

[Felipe Contreras <felipe.contreras@gmail.com>]

On Tue, Jul 30, 2013 at 9:22 PM, Aaron Lu <aaron.lwe@gmail.com> wrote:
> On 07/31/2013 10:07 AM, Felipe Contreras wrote:
>> On Tue, Jul 30, 2013 at 8:36 PM, Aaron Lu <aaron.lwe@gmail.com> wrote:
>>> On 07/31/2013 08:11 AM, Felipe Contreras wrote:
>>>> On Tue, Jul 30, 2013 at 6:13 PM, Rafael J. Wysocki <rjw@sisk.pl> wrote:

>>>>> Yes, that will break backlight on your system and *then* you can complain to
>>>>> Linus if you wish.
>>>>
>>>> It is already broken in v3.11-rc3, in fact I just booted that to try
>>>> it out and it booted with the screen completely black (fortunately I
>>>> knew exactly what to type to change that).
>>>
>>> That is bad, can you please file a bug for that? I'll need to take a
>>> look at your ACPI tables, thanks.
>>
>> File a bug where?
>
> https://bugzilla.kernel.org/
> For component, please choose ACPI->Power_Video.

Here you go:
https://bugzilla.kernel.org/show_bug.cgi?id=60677

>>>> Apparently this commit also needs to be reverted: efaa14c (ACPI /
>>>> video: no automatic brightness changes by win8-compatible firmware).
>>>> In this machine it makes the backlight work again (without
>>>> acpi_osi="!Windows 2012"), but by doing so the ACPI driver also turns
>>>> off the screen completely at level 0. Also, each time I change the
>>>
>>> So with rc3 and commit efaa14c reverted, when you set level 0 to ACPI
>>> video's backlight interface, the screen will be off now? And this is not
>>> the case in 3.10, right?
>>
>> No, setting level 0 turns it off if efaa14c is *not* reverted. In 3.10
>> 0 doesn't turn it off.
>
> AFAIK, Win8 firmware will check a flag to decide what to do on hotkey
> press and backlight brightness level change. Common behavior is:
> If that flag is set:
>   - on hotkey press, send out event;
>   - on brightness level change, use GPU driver's interface to change
>     backlight level following operation region spec.
> If that flag is not set:
>   - on hotkey press, adjust the brightness level without sending out
>     event;
>   - on brightness level change, use EC to change the brightness level.
>
> The said commit sets the flag, so it seems with the flag set now,
> it is possible the firmware interface will also give a screen off
> result. But since we want to have notification event, we will have to
> set that flag I think.

Probably, but changing the brightness should still work even without
the flag, right? It doesn't work here.

Setting the flag makes it work, but not properly.

>>> Please attach acpidump output to the to be opened bugzilla page, thanks.
>>
>> I looked at the code in the DCDT, it appears to me that they store
>> different levels depending on the OSI version, so I don't think the
>> problem is in the ACPI driver. Yet, the issue remains there.
>
> It's good to know what is the problem, so I would like to see the table.
> Most of the bugs I solved in ACPI video driver's backlight control are
> caused by firmware, so yes, it's very unlikely a bug of the ACPI video
> driver itself.

I hope you meant the output of this:
https://aur.archlinux.org/packages/acpidump/

-- 
Felipe Contreras