Bug 108618 - Setting minimal brightness level on HP EliteBook Folio G1 and HP Elite x2 1013 G3 turn off backlight
Summary: Setting minimal brightness level on HP EliteBook Folio G1 and HP Elite x2 101...
Status: CLOSED NOTOURBUG
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: XOrg git
Hardware: x86-64 (AMD64) Linux (All)
: medium major
Assignee: Intel GFX Bugs mailing list
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2018-11-01 08:01 UTC by russianneuromancer
Modified: 2019-04-06 10:32 UTC (History)
1 user (show)

See Also:
i915 platform: KBL
i915 features: display/backlight

Attachments
dmesg drm-tip 2018-11-01 from HP Elite x2 1013 G3 (1.18 MB, text/plain)
2018-11-01 08:01 UTC, russianneuromancer 		no flags Details
View All

Description russianneuromancer 2018-11-01 08:01:55 UTC 
Created attachment 142318 [details]
dmesg drm-tip 2018-11-01 from HP Elite x2 1013 G3

Hello!

Some HP devices (probably many) such as HP EliteBook Folio G1 and HP Elite x2 1013 G3 turn off backlight if user set minimal brightness level. Due to unresolved issues with hp_wmi brightness hotkeys doesn't work, so user left with black screen and no convenient option that would let him turn backlight on again. (Hence major severity.)

I propose to limit minimal brightness level for such devices to prevent backlight from turning off.

debug dmesg from drm-tip 2018-11-01 (9458f0c67b519ddf1c2b4eec9110e826c4e7d2f3) run on HP Elite x2 1013 G3 is attached. I set maximum brightness, then minimal (backlight disabled) and then increase backlight to save dmesg log.

Issue is reproducible at least since Linux 4.13 (was tested on HP EliteBook Folio G1 , earlier releases not tested).
Comment 1 Jani Nikula 2018-11-02 09:15:43 UTC 
Do you boot in UEFI mode? If not, please switch to UEFI boot.

The VBT (which may be different for UEFI vs. legacy boot) indicates minimum brightness of 0 which we respect. The vendor could have chosen non-zero and set it to VBT if so desired.

The i915 kernel driver is not the place to work around hp_wmi hotkey issues.
Comment 2 russianneuromancer 2018-11-04 06:34:37 UTC 
Jani, if possible, could you please reconsider your resolution on this issue?

> The i915 kernel driver is not the place to work around hp_wmi hotkey issues.

Sure, but this two HP devices was just example of case when issue is more severe due to not working hotkeys. I just disabled backlight on GPD Win 2 for example, just by decreasing brightness, but in GPD case I at least can enable it back by using working hotkeys. I guess many other vendors affected too.

Also, don't you think it would be easier to solve this on kernel driver level, instead of fix every desktop environment separately? Latter will take a few years in best case scenario.

> Do you boot in UEFI mode? If not, please switch to UEFI boot.

All three devices boot in UEFI mode.
Comment 3 Jani Nikula 2019-01-02 08:41:23 UTC 
(In reply to russianneuromancer from comment #2)
> Jani, if possible, could you please reconsider your resolution on this issue?
> 
> > The i915 kernel driver is not the place to work around hp_wmi hotkey issues.
> 
> Sure, but this two HP devices was just example of case when issue is more
> severe due to not working hotkeys. I just disabled backlight on GPD Win 2
> for example, just by decreasing brightness, but in GPD case I at least can
> enable it back by using working hotkeys. I guess many other vendors affected
> too.
> 
> Also, don't you think it would be easier to solve this on kernel driver
> level, instead of fix every desktop environment separately? Latter will take
> a few years in best case scenario.

Experience has taught me going the "easy but wrong" route usually gets more complicated over the years. The kernel driver will have to carry this type of changes essentially indefinitely, and any further changes will need to take them into account.
Comment 4 Jan-Michael Brummer 2019-04-06 10:32:28 UTC 
Jani, i agree with you that it is always a burden to add workarounds to a sane driver. In this case we do have another aspect to consider:

Those devices are 2-in-1.

There is a high chance that those devices are used without a keyboard attached and then there is not even a option to use the keyboard backlight hardkeys: Doomed.

What would be your proposal here? Let every desktop environment fix it in their controls and multiply the quirk n-times? Or fix it somewhere in the driver code, let it be intel gfx or whereever?

I would like to hear your thoughts on this one.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.