Bug 27207

Created attachment 34257 [details]
lspci from drm-intel-next branch

Created attachment 34258 [details]
inte_reg_dump with drm-intel-next branch

Created attachment 34259 [details]
dmesg with drm-intel-next branch and drm.debug=0x06

Hi, Jan
    Will you please add the boot option of "nomodeset" and attach the following outputs?
    a. intel_reg_dumper
    b. acpidump
    c. vbios.dump

The acpidump can be obtained by using pmtools-20071116, which can be downloaded from: 
    http://www.lesswatts.org/projects/acpi/utilities.php
   
The vbios.dump can be obtained by using the following command:
    > 1. echo 1 > /sys/devices/pci0000:00/0000:00:02.0/rom
    2. cat /sys/devices/pci0000:00/0000:00:02.0/rom >vbios.dump
    3. echo 0 > /sys/devices/pci0000:00/0000:00:02.0/rom

Thanks.

Created attachment 34830 [details]
intel_reg_dump with drm-intel-next and nomodeset

Created attachment 34831 [details]
acpidump with drm-intel-next, pmtools-20071116 and nomodeset

Created attachment 34832 [details]
vbios dump with drm-intel-next and nomodeset

Sounds like we're failing to set dithering properly on your platform.

Yakui, can you take a look?

(In reply to comment #8)
> Sounds like we're failing to set dithering properly on your platform.
> 
> Yakui, can you take a look?


sure. I will look at this issue. Maybe the blur issue is related with the incorrect dithering setting. Thanks for pointing out the direction.

Thanks.

Hi, Jan
    It seems that this is a laptop with switchable graphics. I am not sure how the panel is controlled. IGD or Nvidia card?
    
    Can you boot the system into windows and switch the graphics into integrated graphics device? Then boot into the linux and attach the following output(please add the boot option of "drm.debug=0x04"):
    a. intel_reg_dumper
    b. dmesg

Thanks.

Created attachment 34838 [details]
dmesg with drm.debug=0x04 after windows

Created attachment 34839 [details]
intel_reg_dump after booting windows

Okay, I attached both files. Under Windows the graphics cards can be switched with a 3-way slider above the keyboard (SPEED, STAMINA, AUTO).

The corresponding LED turns off after rebooting, but that might just be some driver issue.

I have the same vaio Z as yours, and when I reboot from an older kernel (a 2.6.27 like the kernel26-lts from archlinux), the graphic switch works, ie if it's on stamina mode (IGD), the stamina led is on and there is only the intel card in lspci.

The panel is controlled by the IGD. I'm sure because the brightness is controlled by it and when both cards are in lspci the nvidia can't work because the intel is still here so the nvidia "fails to copy vbios to system memory."

By the way, on my model, text is not blurred but I have totally random colors like in that bug report :
https://bugs.freedesktop.org/show_bug.cgi?id=27220 (but in my case, white is not pink but green..)

Will you please try the following command and see whether the issue still exists?
  xrandr --output DP3 --mode 1600x900 --crtc 1 --refresh 60

Thanks
    Yakui

I tried:

DISPLAY=:0 xrandr --output DP3 --mode 1600x900 --crtc 1 --refresh 60 --verbose

but only got the following:

crtc 0: disable
xrandr: Configure crtc 0 failed
crtc 0: disable
crtc 1: disable
screen 0: revert
crtc 0: revert
crtc 1: revert

There's nothing new in dmesg after executing the command.

(In reply to comment #16)
> I tried:
> 
> DISPLAY=:0 xrandr --output DP3 --mode 1600x900 --crtc 1 --refresh 60 --verbose
> 
> but only got the following:
> 
> crtc 0: disable
> xrandr: Configure crtc 0 failed
> crtc 0: disable
> crtc 1: disable
> screen 0: revert
> crtc 0: revert
> crtc 1: revert
> 
> There's nothing new in dmesg after executing the command.

Doh to myself. Running the command via xterm works, but the problem still exists.

Created attachment 35845 [details] [review]
[RFC patch]: try to add the support of eDP on DP-D

Will you please try the debug patch on 2.6.34 kernel and see whether the issue still exists?
   Please add the boot option of "drm.debug=0x04" and attach the output of dmesg, intel_reg_dumper.

Thanks.

*** This bug has been marked as a duplicate of bug 27220 ***