Radeon drm: dpms backlight problem / question / bug?

[Michel Dänzer]

On Sam, 2010-11-06 at 15:36 +0100, Prof. Dr. Klaus Kusche wrote: 
> On 2010-11-02 08:20, Michel Dänzer wrote:
> > On Mon, 2010-11-01 at 16:09 +0100, Prof. Dr. Klaus Kusche wrote:
> >> On my Dell Precision M6500, backlight power switching does not work,
> >> neither by using bl_power in /sys/backlight/...
> >
> > Please provide the full path of the file you're using.
> >
> >> nor by using DPMS:
> >> * Any changes of bl_power are simply and silently ignored
> >>     (no effect, except that brightness is set to minimum),
> >>     although brightness tuning works fine
> >>     (in other words: Both the generic ACPI BIOS driver and the in-kernel
> >>     Dell laptop backlight driver can control backlight brightness,
> >>     but not backlight power).
> >> * If DPMS becomes active, the screen contents on the internal display
> >>     slowly becomes garbled or fades away (so video output drivers
> >>     seem to be off), but the backlight remains on
> >>     (which is very bad for battery power).
> >>     The external screen (on display port) behaves even worse:
> >>     It enters energy saving mode (backlight off) and immediately wakes up
> >>     again, continuously cycling the backlight power every few seconds.
> >>
> >> Another problem which optically looks the same:
> >> Zapping the X server (Xorg 1.9.1, radeon 6.13.2) with Ctrl-Alt-Bksp
> >> leaves the system in an undefined state:
> >> * Similar to DPMS, the display fades away or gets garbled
> >>     (video output off), but the backlight remains on.
> >> * It does not switch to the text console automatically,
> >>     nor is it possible to switch manually with Ctrl-Alt-Fn.
> >>
> >> Configuration:
> >> * Kernel 2.6.35.7-grsec
> >> * Radeon DRM with KMS
> >> * Radeon framebuffer with backlight control enabled
> >> * The Radeon chip is a JUNIPER 0x1002:0x68A0,
> >>     the kernel should contain all the microcode needed for it.
> >> * The displays are LCD1: INTERNAL_UNIPHY and DFP1: INTERNAL_UNIPHY1
> >
> > Please provide the full Xorg.0.log file and dmesg output.
> 
> I've sent the full logs on Nov 2nd.
> Did they arrive?

Yes, as should be evident from the list archives.

> Did they provide any insight?
> 
> Anything else I can do / try / provide?

I'm out of ideas unfortunately, hopefully someeone else can jump in.


-- 
Earthling Michel Dänzer           |                http://www.vmware.com
Libre software enthusiast         |          Debian, X and DRI developer