No backlight since linux-4.2.4 - drivers/gpu/drm/radeon/atombios_encoders.c

[Alex Deucher]

On Mon, Oct 26, 2015 at 5:32 AM, Michel Dänzer <michel at daenzer.net> wrote:
> On 25.10.2015 04:25, Michael Burian wrote:
>> hi,
>>
>> after updating the kernel from linux-stable[0] v4.2.3 to v4.2.4 on my
>> hp compaq nx9420 laptop (Debian Jessie, amd64), the backlight is turned off
>> within the first couple of seconds during boot.
>>
>> The situation persists (Xorg seems to be running fine, icons can be recognized
>> by illuminating the screen with a flashlight) until reboot using an earlier
>> kernel.
>>
>> current mainline[1] as of now (4.3.0-rc6-00277-g01815536) is also affected
>>
>> Workaround: use earlier kernel (v4.2.3) or (see below)
>>
>> Hardware:
>>
>> lspci | grep adeo
>> 01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] RV530/M56-P [Mobility Radeon X1600]
>>
>> git bisect log
>> # bad: [190bd21bba09ed476d9359d3e8be20e8d9dcd8a4] Linux 4.2.4
>> # good: [fcba09f2b0bf27eeaa1d4d439edb649585f35040] Linux 4.2.3
>> git bisect start 'v4.2.4' 'v4.2.3' 'drivers/gpu/'
>> # good: [885838eef785ff810dac945be8132856f2113cf1] drm/amdgpu: make UVD handle checking more strict
>> git bisect good 885838eef785ff810dac945be8132856f2113cf1
>> # good: [230f6fd628c83fcb44ea364d69a9ad88826ef8c4] drm/i915/bios: handle MIPI Sequence Block v3+ gracefully
>> git bisect good 230f6fd628c83fcb44ea364d69a9ad88826ef8c4
>> # bad: [5a633828b4b2bef343826afcb0a70770c4911c55] drm/radeon: Restore LCD backlight level on resume (>= R5xx)
>> git bisect bad 5a633828b4b2bef343826afcb0a70770c4911c55
>> # good: [60a09aef456ef590762d47da9c25650e230e2f85] drm: Reject DRI1 hw lock ioctl functions for kms drivers
>> git bisect good 60a09aef456ef590762d47da9c25650e230e2f85
>> # first bad commit: [5a633828b4b2bef343826afcb0a70770c4911c55] drm/radeon: Restore LCD backlight level on resume (>= R5xx)
>>
>> Workaround 2:
>>
>> reverting this single commit 5a633828b4b2bef343826afcb0a70770c4911c55 fixes the problem in linux-stable[0]
>> as well as mainline[1] 4.3.0-rc6-00277-g01815536
>
> Please provide the dmesg output from running with the attached debugging
> patch.
>
>
> Alex, any other ideas what might go wrong?

Not off hand.  Does the driver exposed backlight control work on your
system?  Backlight control in the r5xx days was mostly handled by
platform specific ACPI methods or the standard ACPI backlight method
under other OSes (depending on the OSI string).

Alex