Bug 25223

Summary: LVDS detected, but disabled on GeForce Go 7300
Product: xorg Reporter: Martin Holec <martix.cz>
Component: Driver/nouveauAssignee: Nouveau Project <nouveau>
Status: RESOLVED INVALID QA Contact: Xorg Project Team <xorg-team>
Severity: blocker    
Priority: highest CC: 67b0226d, piotrdrag
Version: 7.5 (2009.10)   
Hardware: Other   
OS: All   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
dmesg
none
Xorg log
none
Forcibly try DDC on LVDS if i2c table index valid
none
Force the use of strap-based FP mode if present, and fp_no_ddc is true
none
log with modeset=0
none
log with modeset=1 none

Description Martin Holec 2009-11-22 05:23:36 UTC
Xorg doesn't start after boot. Internal display connected to LVDS fails, but external display with D-Sub works.
Tested on Fedora 12 LiveCD with nouveau.modeset=1.

Hardware: http://www.smolts.org/client/show/pub_63cc826b-b57b-401b-ad0c-a27b0cb5b5d3
Comment 1 Martin Holec 2009-11-22 05:24:22 UTC
Created attachment 31382 [details]
dmesg
Comment 2 Martin Holec 2009-11-22 05:24:47 UTC
Created attachment 31383 [details]
Xorg log
Comment 3 Ben Skeggs 2009-11-22 14:03:38 UTC
Does the display work with nouveau.modeset=0?  And, what resolution does your panel have?
Comment 4 Martin Holec 2009-11-23 07:29:26 UTC
No it doesn't. Resolution is 1024x768.
Comment 5 Ben Skeggs 2009-11-23 15:22:18 UTC
The info in the VBIOS about your panel is confusing for nouveau, we end up with *no* valid method for detecting your panel's mode.

Your VBIOS has a 1024x768 mode listed for your panel, and nouveau detects that it should use it if it's using those mode tables.  However, another VBIOS table tells nouveau to *not* use those mode tables.  I'll attach panel-strap-mode.diff to force the use of the tables.

Now, the entry for your panel gives a valid index into the DDC table so it's possible it reports an EDID that we should use.  I suspect this is probably the case here, and we make a bad decision on when to use DDC on your card.  I'll attach panel-ddc-mode.diff for this.

Can you try each patch separately and report which (if any) work for you.

Thanks!
Comment 6 Ben Skeggs 2009-11-23 15:23:09 UTC
Created attachment 31425 [details] [review]
Forcibly try DDC on LVDS if i2c table index valid
Comment 7 Ben Skeggs 2009-11-23 15:24:06 UTC
Created attachment 31426 [details] [review]
Force the use of strap-based FP mode if present, and fp_no_ddc is true
Comment 8 Ben Skeggs 2009-12-03 02:59:42 UTC
Any update?  This should be fixed in latest git too..
Comment 9 67b0226d 2009-12-25 07:00:20 UTC
(In reply to comment #8)
> Any update?  This should be fixed in latest git too..
I have a similar, if not the same problem. Your second patch already seems to be in git, and I also tried applying your first patch, but nothing changes. I have tried with both modeset=0 and modeset=1, Xorg.0.log and dmesg attached respectively.
Comment 10 67b0226d 2009-12-25 07:08:46 UTC
Created attachment 32292 [details]
log with modeset=0
Comment 11 67b0226d 2009-12-25 07:09:35 UTC
Created attachment 32293 [details]
log with modeset=1
Comment 12 Lucas Stach 2011-02-15 02:53:37 UTC
This report seems to be against a very old version of nouveau. Is this still an issue with recent nouveau?

If we get no status update within one month I will close this bug as unreproducible.
Comment 13 Lucas Stach 2011-03-31 03:23:44 UTC
Closing as we got no new info from reporter and the bug is against ancient version of nouveau.

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.