https://bugzilla.novell.com/show_bug.cgi?id=680085
https://bugzilla.novell.com/show_bug.cgi?id=680085#c30
Egbert Eich changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |ASSIGNED
--- Comment #30 from Egbert Eich 2011-04-18 15:31:56 UTC ---
Sorry, there are several issues reported in this bug which are not really
related to each other:
1. Bad EDID block. The patch suggested in comment #1 doesn't apply here as this
is for DP to DVI connectors while a DMS59 is pure DVI.
2. Patches disabling EDID checking are a strategy out of desperation: EDID
blocks from monitors are usually good (unless one managed to destroy the EDID
data on a monitor - happened to me once).
3. Andre suggests to fix/update the X driver. This is not responsible for the
wrong EDID readings in the kernel.
4. Crash:
Mar 28 18:06:24 test114 kernel: [23777.388002] kernel BUG at
/usr/src/packages/BUILD/kernel-desktop-2.6.37.1/linux-2.6.37/drivers/gpu/drm/i915/i915_gem.c:4201!
This is due to the code:
BUG_ON(obj_priv->pin_count == DRM_I915_GEM_OBJECT_MAX_PIN_COUNT);
in i915_gem_object_pin(). the pin_count exceeds 15 possibly due to constant
attempts to reset the mode.
On the other hand I cannot identify a code path where an fb_object accidentally
remains pinned. I don't agree with the assessment in
https://bugs.freedesktop.org/show_bug.cgi?id=35758 about the cause.
While I agree that 007c80a5 is an important patch it doesn't fix the crash seen
here.
5. Andre suggests to boot without vga=... or with nomodeset. The former will
avoid the bootsplash. That this will avoid hangs seems to be related to bnc
#672008. Using nomodeset will cause KMS to be disabled - the intel driver will
not work with that!
6. The EDID read out problems don't seem to go way with reprobing the EDID on
failure. I can investigate if I find a system that shows such issues.
--
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.