Pointer goes offscreen on non-rectangular dual-head

Bug #119417 reported by Marcus Sundman
22
This bug affects 3 people
Affects Status Importance Assigned to Milestone
xorg (Ubuntu)
Expired
Low
Unassigned

Bug Description

On a non-rectangular multi-monitor layout (e.g. when using monitors with different resolutions) the mouse pointer can be positioned outside the areas displayed by the monitors.

My preferred multi-head layouts are ones where the viewports touch only at the corner. (E.g., if you have two 800x600 displays then you have a "virtual" screen of 1600x1200 where the viewport of one display is in the top left part and the viewport of the other display is in the bottom right part.) This keeps stuff att the viewport borders still fast to access (you know, fitt's law and how things at the border have infinite width/height, enabling things such as Opera's "panel toggle" etc). This layout works very well on other OSes, but on Ubuntu widgets along the borders are annoyingly easy to miss because the pointer overshoots off the screen so easily.

(I think the proprietary ati driver has an option to keep the pointer within the bounds of the monitors' viewports, but I have only nvidia cards myself so I can't confirm this. However, even if one driver has implemented a workaround for this deficiency doesn't mean it should be implemented at the driver level.)

Revision history for this message
Brian Murray (brian-murray) wrote :

Thank you for taking the time to report this bug and helping to make Ubuntu better. You reported this bug a while ago and there hasn't been any activity in recently. We were wondering if this is still an issue for you? Thanks in advance.

Revision history for this message
Marcus Sundman (sundman) wrote :

Yes, it is. I'm using TwinView which has its own xinerama implementation. (AFAIK TwinView is the only way to make dual-head work properly (e.g., with 2D acceleration) with nvidia cards.) Here is the TwinView MetaMode line I'm currently using:

Option "MetaModes" "CRT-0: 1600x1200 @1600x1200 +1280+0, DFP-0: 1280x1024 @1280x1024 +0+0"

I'd much rather use "CRT-0: 1600x1200 @1600x1200 +1280+1024, DFP-0: 1280x1024 @1280x1024 +0+0", but that's quite unpractical if the pointer can go into the dead zones.

Revision history for this message
Brian Murray (brian-murray) wrote :

Thank you for taking the time to report this bug and helping to make Ubuntu better. You reported this bug a while ago and there hasn't been any activity in recently. We were wondering if this is still an issue for you? Thanks in advance.

Revision history for this message
Marcus Sundman (sundman) wrote :

What's this? There has been activity less than two days ago, and I already wrote that it is still an issue for me.

Revision history for this message
Timo Aaltonen (tjaalton) wrote :

You really should report this to NVidia, since we can't do anything about it.

Changed in xorg:
status: New → Triaged
Revision history for this message
Marcus Sundman (sundman) wrote :

Is it really impossible for the window manager or the xserver core to keep the pointer inside the viewports? I find this hard to believe.

Revision history for this message
Timo Aaltonen (tjaalton) wrote :

Hmm, I thought it was a bug in TwinView, but then came across with this:

[patch] xserver-xorg-core: Fixes Xinerama Not Respecting Display Boundries
https://bugs.edge.launchpad.net/ubuntu/+source/xorg-server/+bug/110809

Does that sound like your problem?

Revision history for this message
Marcus Sundman (sundman) wrote :

No. That's about how windows are maximized. My problem is that the mouse pointer can go outside display boundaries.

Revision history for this message
Ludovico Cavedon (cavedon) wrote :

I have this problem also with the intel driver using xrandr and dual head
1280x800 + 1920x1200

I am moving this bug to xorg, as it is not a kernel issue.

affects: linux-restricted-modules-2.6.22 (Ubuntu) → xorg (Ubuntu)
Bryce Harrington (bryce)
affects: xorg (Ubuntu) → nvidia-graphics-drivers-180 (Ubuntu)
Revision history for this message
Ludovico Cavedon (cavedon) wrote : Re: [Bug 119417] Re: Pointer goes offscreen on non-rectangular dual-head

Bryce Harrington wrote:
> ** Package changed: xorg (Ubuntu) => nvidia-graphics-drivers-180
> (Ubuntu)

Why did you move the bug? I have to nvidia? I have an itel chipset.
Do you mean that it is not xorg dependent by every affected video driver
needs to fix it?

Thanks,
Ludovico

Revision history for this message
Bryce Harrington (bryce) wrote :

> Why did you move the bug? I have to nvidia? I have an itel chipset.

The original reporter was using -nvidia. Dual head implementation for -nvidia is completely unrelated to intel.

You've probably made a mistake in assuming you have the same bug as the original reporter based on coincidentally having similar symptoms.

Revision history for this message
Marcus Sundman (sundman) wrote :

I don't think this has anything to do with any particular graphics chipset implementation. At least it shouldn't be. It has to do with how the pointing devices relate to a desktop that expands beyond the viewports.

Revision history for this message
Ludovico Cavedon (cavedon) wrote :

That is what I thought.
However I opened a separate bug #373367

Thanks

Marcus Sundman (sundman)
description: updated
Revision history for this message
Christopher M. Peñalver (penalvch) wrote :

Marcus Sundman, thank you for taking the time to report this bug and helping to make Ubuntu better. Please execute the following command, as it will automatically gather debugging information, in a terminal:
apport-collect 119417

Please ensure you have xdiagnose installed, and that you click the Yes button for attaching additional debugging information.

affects: nvidia-graphics-drivers-180 (Ubuntu) → xorg (Ubuntu)
Changed in xorg (Ubuntu):
importance: Undecided → Low
status: Triaged → Incomplete
Revision history for this message
Launchpad Janitor (janitor) wrote :

[Expired for xorg (Ubuntu) because there has been no activity for 60 days.]

Changed in xorg (Ubuntu):
status: Incomplete → Expired
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers