[Bug 58396] [i945GM regression] in 3.7.0 kernel on macbook1,1 (kernel crash)

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=58396

--- Comment #8 from Peter Ujfalusi <peter.ujfalusi at gmail.com> ---
(In reply to comment #7)
> Created attachment 71656 [details] [review]
> dump encoder responsible for modeset
> 
> Also please apply this little debug patch on top of a broken kernel and
> readd a new dmesg with drm debug enabled.
> 
> Analysis thus far is that we do a completely bogus modeset on the 2nd crtc,
> but while actually disabling the 1st crtc. Hence caller supplies no mode,
> and we fall over with a NULL deref trying to prepare the modeset on the 2nd
> crtc. Above patch might shed clue on why we try to do that bogus modeset.

Will do. Meanwhile I looked at the offending patch:
fa55583 drm/i915: fixup the plane->pipe fixup code

The new check does not match the removed one so the code in
intel_sanitize_crtc()
will going to take different path.
Snippet from the patch:

+
+       if ((val & DISPLAY_PLANE_ENABLE) &&
+           (!!(val & DISPPLANE_SEL_PIPE_MASK) == crtc->pipe))
+               return false;
+
...
-
-               if ((val & DISPLAY_PLANE_ENABLE) == 0 &&
-                   (!!(val & DISPPLANE_SEL_PIPE_MASK) == crtc->pipe))
-                       goto ok;
-

I'm mostly confident that this is the root cause of the issue I'm seeing.
I'l attach a patch to the bug to fix it. I have not tested it, but going to do
that as soon as I got home.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20121217/5455f924/attachment.html>