Reply to topic

KDE 4.11 with xf8x-video-intel 2.21.14 brightness problem.

firewalker
Registered Member
Posts
45
Karma
0
OS
After upgrading to xf86-video-intel to 2.21.14 KDE can no longer control the brightness. It display the brightness level info, but no change. Downgrading to 2.21.13 it works.

issuing as root:

echo _value_ > /sys/class/backlight/intel_backlight/brightness

changes the brightness.

_value_: 0 to 4882

What mechanism KDE uses to set brightness?

I did a git bisect between xf86-video-intel 2.21.13 and 2.21.14 but could find any good revision. The result was:

b6bcb7ae20b92e48b8a2fa096c0e33b6b97b1b30 is the first bad commit
commit b6bcb7ae20b92e48b8a2fa096c0e33b6b97b1b30
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Thu Jul 25 21:22:25 2013 +0100

intel: Source our PCI IDs table from the copy in the kernel

Rather than duplicating the information we already use in the kernel, we
can reuse the pci-id tables so long as we apply a little fuzz.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

:040000 040000 14fb19dba1e993587a006921497c475438ebd102 1481104f860239ad6da456eb0ebddf6d66feed38 M src


It doesn't seem right to me.

log:
git bisect start
# good: [bb8484da487008df8b69beb8086167bc272e85bb] 2.21.13 release
git bisect good bb8484da487008df8b69beb8086167bc272e85bb
# bad: [0cd154039ab02799dc972d93c415e762226df1aa] 2.21.14 release
git bisect bad 0cd154039ab02799dc972d93c415e762226df1aa
# bad: [1dd79c225cedf5614a77e5db9f9d9700bf957bb8] sna: Restore driver info upon FreeScreen
git bisect bad 1dd79c225cedf5614a77e5db9f9d9700bf957bb8
# bad: [8afb5438f3479df39752cbda680c13d80f81f42f] intel: Always define the PCI-IDs
git bisect bad 8afb5438f3479df39752cbda680c13d80f81f42f
# bad: [6fd37e644d9c662776aff14cd51a27b7e52472de] intel: Replace the codename in the user facing string for unknown Haswell parts
git bisect bad 6fd37e644d9c662776aff14cd51a27b7e52472de
# bad: [4c6871684b0006a89be6760db99fda9b47790226] intel: Cross-check an unnamed chipset against the list of known PCI-IDs
git bisect bad 4c6871684b0006a89be6760db99fda9b47790226
# bad: [b6bcb7ae20b92e48b8a2fa096c0e33b6b97b1b30] intel: Source our PCI IDs table from the copy in the kernel
git bisect bad b6bcb7ae20b92e48b8a2fa096c0e33b6b97b1b30
# first bad commit: [b6bcb7ae20b92e48b8a2fa096c0e33b6b97b1b30] intel: Source our PCI IDs table from the copy in the kernel
User avatar bcooksley
Administrator
Posts
18586
Karma
83
OS
This is unusual - are you certain the Intel driver is being used by X?
If that is the case, then it seems the Intel driver is no longer advertising screen brightness capabilities via DPMS as it should be.

As such, it is likely this is a bug in the Intel driver itself. Can you report this to them, so they can investigate what change on their side changed how they handle/provide brightness changing?


System Settings and Device Actions KCM maintainer
Image
firewalker
Registered Member
Posts
45
Karma
0
OS
I will make a report to them.
firewalker
Registered Member
Posts
45
Karma
0
OS
The answer fron xf86.

The dell_backlight interface is dysfunctional and should not be exposed for your system i.e. don't specify acpi_backlight=vendor. Also i915.i915_enable_fbc=1 is known to cause issues on your system.


What mechanism does KDE use to set the brightness?
firewalker
Registered Member
Posts
45
Karma
0
OS
Solved by adding

Code: Select all
Section "Device"
        Identifier      "HD 3000"
        Driver  "Intel"
        Option "Backlight" "intel_backlight"
EndSection


to /etc/X11/xorg.conf.d/20-intel.conf file.
User avatar bcooksley
Administrator
Posts
18586
Karma
83
OS
KDE uses a variety of mechanisms, including DPMS (which is what the backlight option in xorg.conf fixed) and interacting with various vendor backlight devices under /sys if that is not present. Once it has found a mechanism it believes to work, it does not continue to check further devices.


System Settings and Device Actions KCM maintainer
Image

 
Reply to topic

Bookmarks



Who is online

Registered users: Alexa [Bot], Baidu [Spider], Bing [Bot], claydoh, Exabot [Bot], Google [Bot], koriun, Majestic-12 [Bot], mmistretta, MSNbot Media, scummos, Wiz, Yahoo [Bot]