Bug 97295

Summary:	Regression backlight control broken on Dell XPS13 w/ kernel 4.7
Product:	DRI	Reporter:	Martin <bugs>
Component:	DRM/Intel	Assignee:	Intel GFX Bugs mailing list <intel-gfx-bugs>
Status:	CLOSED FIXED	QA Contact:	Intel GFX Bugs mailing list <intel-gfx-bugs>
Severity:	major
Priority:	medium	CC:	aaron.lu, intel-gfx-bugs
Version:	unspecified
Hardware:	Other
OS:	All
Whiteboard:
i915 platform:	IVB	i915 features:	display/backlight

Description Martin 2016-08-11 10:57:39 UTC

I reported this bug on kernel bugzilla, but assumed it might be wise to inform this bugtracker as well. On kernel 4.7 intel_backlight takes precedence over acpi_video0 as backlight device, breaking backlight support on my IVB XPS13 again (1Hz pulsating brightness).

(New) Kernel bug: https://bugzilla.kernel.org/show_bug.cgi?id=151731
(Old) Freedesktop bug: https://bugs.freedesktop.org/show_bug.cgi?id=76276

Quick solution is the same as in comment #15, hard-disable the initialisation of intel_panel_setup_backlight in intel_panel.c

Comment 1 Jani Nikula 2016-08-12 08:06:17 UTC

I don't recall any recent drm/i915 changes that would cause intel_backlight taking precedence over acpi_video. I presume the changes is on the ACPI side.


(In reply to Martin from comment #0)
> (Old) Freedesktop bug: https://bugs.freedesktop.org/show_bug.cgi?id=76276

This has no relevance these days.

Comment 2 Martin 2016-08-12 09:37:16 UTC

Well, the quickhack (disabling initialisation of intel backlight) solves the problem anyway, that's why I mentioned the bug. This laptop was affected back then as well.

Comment 3 linuxchr1s 2016-08-13 08:37:23 UTC

I got the same issue with MSI GE60-i7. intel_backlight is populated, but changing brightness doesn’t have any effect.

Comment 4 Jani Nikula 2016-09-20 09:46:00 UTC

Please try current drm-intel-nightly branch of http://cgit.freedesktop.org/drm-intel

Comment 5 fclad 2016-09-24 00:20:51 UTC

It seems to work in 4.8.0-1-drm-intel-nightly

/sys/class/backlight/acpi_video0 is back and backlight works properly.

Comment 6 yann 2016-09-27 09:40:29 UTC

Martin, can you also re-test on your side with latest drm-nightly?

Comment 7 Martin 2016-09-28 08:03:58 UTC

I can confirum that drm-intel-nightly branch of http://cgit.freedesktop.org/drm-intel fixes the backlight problem (correct acpi_video0 device detected).

Graphical glitches are so bad however that this branch is unusable for daily work atm, but that's not the subject of this bug I guess.

Comment 8 Martin 2016-10-03 11:06:14 UTC

Running freshly compiled 4.8 without any problems. Thx!!

Comment 9 Jani Nikula 2016-10-03 11:23:18 UTC

(In reply to Martin from comment #8)
> Running freshly compiled 4.8 without any problems. Thx!!

Thanks for the follow-up, closing.

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.