[Bug 77351] [855GM] Mouse cursors show wrong default black cursors and/or do not update correctly

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=77351

--- Comment #12 from Chris Wilson <chris at chris-wilson.co.uk> ---
(In reply to comment #11)
> (In reply to comment #10)
> > The log behaves as I expect. The change of cursors is being sent to the
> > kernel, but you observe that the updates do not occur.
> 
> Okay, but it was working with the older driver versions?! Just mentioning in
> case you did not see it.

Right, and the only difference between then and now would appear to be use of a
single cursor for both CRTCs. (Previously each CRTC had an independent cursor
allocated for it.) It just may be that our cursor programming sequence is wrong
in the kernel, we keep finding bugs there so it would not be too surprising.

> > Can you please try this kernel patch:
> 
> Uuh, a kernel patch... Never did this (also never compiled the kernel
> myself). Sorry, can you tell me how in a few words? Do I need to get the
> kernel, apply the patch and then compile the kernel myself? ... If it is
> worth the effort - apparently I am running into a seldom bug here. Thank you
> anyway!

Getting to the bottom of this is most certainly worth the effort!

First of to compile the kernel, run

$ git clone --depth 1 --branch drm-intel-fixes
git://people.freedesktop.org/~danvet/drm-intel
$ cd drm-intel
$ cp /boot/config-`uname -r` .config
$ make oldconfig
$ make
$ sudo make modules_install install

That should make the new kernel available for booting. Check that it does boot!
Applying the patch should then be easier (you can probably do it faster using a
text editor than pasting it into patch -p1) and then you just need to repeat
make && sudo make modules_install install.

> > The other thing may be the two CRTCs using the same physical address. It's
> > not prohibited by the spec, but it may just be upsetting the hw?
> 
> Do you mean my both screens (VGA1 LVDS1)? One is the laptop's built-in
> screen and another one is attached via VGA port. I could not find a physical
> address in my Xorg.0.log. Do I (or my distro's defaults) use a not-so-good
> configuration? If yes, sorry about this.

No, it's a quirk of this particular hardware that it reads the cursor using a
physical address to a contiguous chunk of memory.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20140414/d3edaff6/attachment.html>