Red Hat Bugzilla – Bug 204599
i810: cursor disappears after VT switch
Last modified: 2018-04-11 14:10:40 EDT
Description of problem:
Originally wrongly entered as Bug #204175.
After switching to and back from a text-based console
(with Ctrl-Alt-F1) mouse cursor disappeared (from gdm or
fluxbox). It happens with the touchpad, and also with
an external usb mouse.
Computer is a Dell Latitude D505. System is rawhide up-to-date.
# lspci | grep VGA
00:02.0 VGA compatible controller: Intel Corporation 82852/855GM Integrated
Graphics Device (rev 02)
It also happens when suspending on disk, but I guess this is because
it switch to text-mode before suspend.
Even if the cursor has disappeared, the mouse is still functionning
normally, moving and clicking do the right things, although it is
not very easy to use since the cursor isn't visible...
Version-Release number of selected component (if applicable):
Steps to Reproduce:
Created attachment 135194 [details]
Xorg.log before switching to console
Created attachment 135195 [details]
Xorg.log after coming back from the console
Created attachment 135196 [details]
difference between the Xorg.log before going to console and after coming back
Similar problem here -- cursor doesn't disapper, but it gets locked in the left
top corner of the display and cannot be moved. This is my graphic card (PCI ID
00:02.0 VGA compatible controller: Intel Corporation Mobile 915GM/GMS/910GML
Express Graphics Controller (rev 03)
Created attachment 139712 [details]
Created attachment 139713 [details]
Difference between /var/log/Xorg.0.log before and after switching to console and back
Created attachment 139719 [details]
Xorg.0.log without any switching to console
*** Bug 213176 has been marked as a duplicate of this bug. ***
May have DUPS bug 206832 and probably blocks RHELRalpha1 URGENT bug 213050
Similar symptoms also reported as bug 183096.
I cannot reproduce it when System/Configuration/Desktop Effects is off. Patrice,
do you have it on? And if yes, could try to reproduce it with compiz switched
off, please? Thanks.
I don't know exactly what compiz is, I use fluxbox and xfce and
it also happens for gdm or wdm.
I have googled a bit about compiz, and certainly don't use
it. I don't have the package installed, and anyway fluxbox wouldn't
Yes, of course, if you have fluxbox, then you don't care about such silliness as
compiz. Just thought worthy to ask, some other user (and me) have been able to
reproduce this only with compiz switched on.
*** Bug 214481 has been marked as a duplicate of this bug. ***
*** Bug 206832 has been marked as a duplicate of this bug. ***
Any action on this bug? I can't use suspend/resume on my laptop because after
coming out of suspend, the mouse is broken. I just tried and can reproduce it
with just metacity. FC6 with all updates, Intel video. I can provide any other
details you need. For reference, my original bug was 213176 which got marked as
a dup to this bug (Sorry, I didn't see the request for more info after the bug
was marked a dup and closed).
This appears to be fixed in FC7. I am running up to date fc7t2+rawhide on my
laptop and the cursor doesn't seem to have any problems reappearing when
switching to a VT and back. I just tried it several times.
Fedora Core 5 and Fedora Core 6 are, as we're sure you've noticed, no longer
test releases. We're cleaning up the bug database and making sure important bug
reports filed against these test releases don't get lost. It would be helpful if
you could test this issue with a released version of Fedora or with the latest
development / test release. Thanks for your help and for your patience.
[This is a bulk message for all open FC5/FC6 test release bugs. I'm adding
myself to the CC list for each bug, so I'll see any comments you make after this
and do my best to make sure every issue gets proper attention.]
This bug is still present for me in
Patrice, does the intel driver work for you?
Yes, the intel driver doesn't have this bug.
*** Bug 238674 has been marked as a duplicate of this bug. ***
Just added my own bug as a duplicate of this one.
Relevant information from my duplicate, that doesn't seem to be mentioned in
"You can work around this bug, by setting the Option SWCursor, however this
disable direct rendering."
This will probably explain why some users have only noticed it in compiz: They
didn't have direct rendering on before setting up their X for compiz.
Note that while the intel driver does not seem to have the problem, the intel
driver does not work with my Dell Latitude D505 with a 1400x1050 LCD display.
Sounds like this one can be closed as 'wontfix' since the i810 driver is no
longer maintained. Is there another bug open either here or at
freedesktop.org on your Dell Latitude problem?
I still use the i810 driver to connect to an external screen, it
doesn't seems to work with the intel driver (I haven't reported,
I don't know if I am doing things rightly to switch to the external
screen). But I can live with this bug since my use of i810 is very
occasional. As a side note I tried rawhide X a bit but then reverted
to F-8 X because rawhide X was very broken (and I don't think my
testing would have made sense it was too obvious that it was broken
to be useful to report), so maybe my informations are obsolete.
If you get a chance, it would be good to try the intel driver again, using the
instructions at http://www.intellinuxgraphics.org/dualhead.html to setup your
dual head configuration. If that doesn't work I'd like to hear about it.
In fact it seems to me that the 'Clone mode' doesn't work. When
I hit Fn-F8 (CRT/LCD) nothing happens. I don't have an external
screen right now, I only tested it with video projectors.
Yeah, Fn-F8 probably won't work with the new driver yet (and if it does you
may be in danger of crashing). Until we get the hotkeys wired up with the new
driver, you'll need to use the 'xrandr' command or xorg.conf options to setup
I see this still on F8 with integrated nvidia chipsets, using either the nv or
binary nvidia driver. Should I open a new bug or generalize this one?
(In reply to comment #31)
> Yeah, Fn-F8 probably won't work with the new driver yet (and if it does you
> may be in danger of crashing). Until we get the hotkeys wired up with the new
> driver, you'll need to use the 'xrandr' command or xorg.conf options to setup
> clone mode.
Clone mode works with
xrandr --output LVDS --auto --output VGA --auto --same-as LVDS
(In reply to comment #32)
> I see this still on F8 with integrated nvidia chipsets, using either the nv or
> binary nvidia driver. Should I open a new bug or generalize this one?
Since it is not the same driver, I think it makes more sense to add
a new bug. Closing.