Bug 81422 - [G33 SDVO] blank screen with KMS on DVI output
Summary: [G33 SDVO] blank screen with KMS on DVI output
Status: CLOSED WORKSFORME
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: unspecified
Hardware: x86-64 (AMD64) Linux (All)
: medium minor
Assignee: Intel GFX Bugs mailing list
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2014-07-16 09:44 UTC by Krzysztof Mazur
Modified: 2017-07-24 22:52 UTC (History)
1 user (show)

See Also:
i915 platform: G33
i915 features: display/Other


Attachments
dmesg (Linux drm-intel-fixes-2014-07-09) (85.25 KB, text/plain)
2014-07-16 09:44 UTC, Krzysztof Mazur
no flags Details

Description Krzysztof Mazur 2014-07-16 09:44:50 UTC
Created attachment 102911 [details]
dmesg (Linux drm-intel-fixes-2014-07-09)

Hi,

the kernel with KMS does not initialize the display correctly on some old all-in-one PC with G33 chipset. The kernel reports some errors, including "SDVO pixel multiplier mismatch" and "VGA-1: EDID block 0 invalid", however the display seems to be connected using DVI1 port.

The monitor seems to report some strange mode 1024x819 which is used by default. However setting the mode to 1024x768 at 60 Hz does not fix the problem. The screen is blank both with the fbcon and the X server.

# xrandr
Screen 0: minimum 320 x 200, current 1024 x 768, maximum 32767 x 32767
VGA1 disconnected (normal left inverted right x axis y axis)
DVI1 connected 1024x768+0+0 (normal left inverted right x axis y axis) 376mm x 301mm
   1024x768       75.1 +   75.0     70.1     60.0*
   1024x819       75.0
   800x640        75.0
   800x600        75.0     60.3
   640x480        75.0     72.8     60.0
   720x400        70.1
HDMI1 disconnected (normal left inverted right x axis y axis)

I've tested Linux 3.12.9, 3.15.4 and drm-intel-fixes-2014-07-09. I'm attaching the dmesg from Linux drm-intel-fixes-2014-07-09 with drm.debug=0xe.
Comment 1 Chris Wilson 2014-07-19 10:12:23 UTC
Can you try even older kernels, perhaps almost as far back as 2.6.32? I doubt it is a regression as it looks like the panel is just returning a garbage EDID - so my first suspicion is that there was an override in their windows driver. But if you can find a regressing commit, we stand a chance of figuring out how this broke.
Comment 2 Jani Nikula 2014-09-08 15:55:56 UTC
Krzysztof, ping. Have you had the chance to try older kernels per comment #1?
Comment 3 Krzysztof Mazur 2014-09-13 07:43:26 UTC
I haven't had access to that hardware for some time. Recently I've tried Linux 2.6.32.63 and it also does not work.
Comment 4 Jesse Barnes 2015-03-30 20:48:37 UTC
Krzysztof, any news?  Maybe updated kernels are able to get the EDID from this sink?
Comment 5 Jani Nikula 2015-10-23 09:54:35 UTC
Timeout, closing. Please reopen if the problem persists with latest kernels.


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.