Bug 17884 - mouse motion not restricted to active display in Plasma5
Summary: mouse motion not restricted to active display in Plasma5
Status: RESOLVED FIXED
Alias: None
Product: Mageia
Classification: Unclassified
Component: RPM Packages (show other bugs)
Version: 8
Hardware: x86_64 Linux
Priority: Normal normal
Target Milestone: ---
Assignee: KDE maintainers
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2016-03-05 17:10 CET by Pierre Fortin
Modified: 2021-07-09 01:14 CEST (History)
4 users (show)

See Also:
Source RPM:
CVE:
Status comment:


Attachments
Display Configuration (48.56 KB, image/jpeg)
2021-07-07 17:56 CEST, Pierre Fortin
Details
/etc/X11/xorg.conf (4.55 KB, text/plain)
2021-07-07 18:00 CEST, Pierre Fortin
Details

Description Pierre Fortin 2016-03-05 17:10:22 CET
Description of problem: Re-installed mga5 on new HD & switched to UEFI boot mode.  This bug was seen on previous install -- thought it was reported; but can't find bug report(s).

Platform: Dell Precision M6800 i7-4710MQ CPU @ 2.50GHz w/32GBram
This system has two display interfaces:
00:02.0 VGA compatible controller: Intel Corporation 4th Gen Core Processor Integrated Graphics Controller (rev 06)
01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Saturn XT [FirePro M6100] (rev ff)

The install detected/enabled both displays; but I only need/want the main (1920x1080) laptop screen.  However, while disabling the external display (1366x768) gives me the desired screen, the mouse continues to act as though the external display is not disabled.  It moves offscreen to the right unrestricted. Interestingly, even though the screen bottoms are aligned, the mouse does not notice that the upper 312 (1080-768) display rows don't exist -- no boundaries on the right edge; unless I enable the external display on the left side (highly horrible display mode).

Version-Release number of selected component (if applicable):


How reproducible: always.


Steps to Reproduce:
1. Use 2 display system
2. Disable non-main display
3.
Comment 1 Marja Van Waes 2016-03-08 11:44:58 CET
Which Desktop environment do you use? (KDE, Gnome, or ....?)

Keywords: (none) => NEEDINFO
CC: (none) => marja11

Comment 2 Pierre Fortin 2016-03-09 04:49:04 CET
Sorry for forgetting to specify...  KDE.  

There are other side-effects, such as:
- with screen saver set to clock: I get 2 overlapping clocks (both set to
  upper-left) when dark screen restored on input (kbrd or mouse)
- external screen on rare occasion comes on [1]

[1] always comes on when switching to vtys (Ctrl+Alt+F[2-6])
Comment 3 Pierre Fortin 2016-03-09 06:14:57 CET
More side-effects:
The external monitor, even though 'disabled', displays a copy of the upper-left portion of the laptop screen when at the kdm login. (both displays on)
When login starts, a full, smaller copy of the splash screen appears on the laptop screen (upper-left) (only laptop display on -- like clock in comment 2).
Marja Van Waes 2016-03-28 22:40:18 CEST

Summary: mouse motion not restricted to active display => mouse motion not restricted to active display in KDE
Assignee: bugsquad => mageia
Keywords: NEEDINFO => (none)

Samuel Verschelde 2016-08-25 16:23:59 CEST

Assignee: mageia => kde

Comment 4 Pierre Fortin 2017-09-18 22:22:16 CEST
Problem persists on mga6 with today's kernel update
$ uname -a
Linux prf.pfortin.com 4.9.50-desktop-1.mga6 #1 SMP Wed Sep 13 23:14:20 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux
Comment 5 Pierre Fortin 2018-02-05 22:28:54 CET
Problem persists on mga6 with today's kernel update
$ uname -a                                                                                                                                                                   
Linux prf.pfortin.com 4.14.13-desktop-1.mga6 #1 SMP Wed Jan 10 12:48:53 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
Comment 6 Marja Van Waes 2018-10-07 14:56:10 CEST
(In reply to Pierre Fortin from comment #5)
> Problem persists on mga6 with today's kernel update
> $ uname -a                                                                  
> 
> Linux prf.pfortin.com 4.14.13-desktop-1.mga6 #1 SMP Wed Jan 10 12:48:53 UTC
> 2018 x86_64 x86_64 x86_64 GNU/Linux

Changing Version: to 6, since 5 is no longer maintained.

Pierre, this issue still persists?

I assume you use Plasma5 as Desktop Environment now, and SDDM as Display Manager?

Version: 5 => 6

Comment 7 Pierre Fortin 2018-10-07 17:20:24 CEST
Still persists.  Plasma5 & sddm are defaults.
The over-lapping screens at startup/login are nits; but the mouse being lost on a non-existent display is irritating. Worse, since external is smaller -- trying to aim mouse at lower 1/4 right edge of main screen, if mouse steps one pixel off screen, the mouse pointer warps to the bottom of external 'display' and returns ~3/4 of the way up the side of real display.  
Settings->Displays shows only:
 "Laptop Screen eDP1"
 Primary display: Laptop Screen
 Display: [X] Enabled
 Resolution: 1920x1080
 Orientation: Normal
 Refresh rate: Auto
 Unify Outputs: [greyed out]
 Scale Display: minimum
Marja Van Waes 2018-10-07 17:43:51 CEST

Summary: mouse motion not restricted to active display in KDE => mouse motion not restricted to active display in PLasma5

Marja Van Waes 2018-10-07 17:44:01 CEST

Summary: mouse motion not restricted to active display in PLasma5 => mouse motion not restricted to active display in Plasma5

Comment 8 Morgan Leijström 2020-04-23 09:52:09 CEST
Do you still see this problem?

CC: (none) => fri

Comment 9 Pierre Fortin 2020-04-23 16:21:40 CEST
Yes. Fully updated system. The most irritating thing is when the mouse accidentally leaves the lower right portion of the only screen (1920x1080), when I bring it back, it "warps" higher up on the left -- more specifically: if the mouse exits anywhere between 721 & 1080, it always re-enters at 720. Then, if I move the mouse slowly farther to the right, I have to move the mouse that same "distance" back for it to reappear on the screen.  There is definitely a ghost screen which is nowhere to be found in the system settings...
Comment 10 Pierre Fortin 2020-04-23 18:14:56 CEST
Correction (using "watch -p -t -n 0 xdotool getmouselocation"): 
   if the mouse exits anywhere between 767 & 1080, it always re-enters at 768.
With xdotool, I see:
   x: 0 y:0 screen:0 window:<various> through x:1919 y:1079 screen:0 window:<various>
and:
   x: 0 y:0 screen:1 window:2000 through x:1023 y:767 screen:1 window:2000
The entirety of screen:1 shows window:2000
Tried using xwininfo to identify the ; but it restricts the mouse to screen:0

More:
$ xrandr -q --screen 0
Screen 0: minimum 320 x 200, current 1920 x 1080, maximum 16384 x 16384
eDP1 connected primary 1920x1080+0+0 (normal left inverted right x axis y axis) 382mm x 215mm
   1920x1080     60.01*+  59.96    59.93  
   1680x1050     59.95    59.88  
   1400x1050     59.98  
   1600x900      59.95    59.82  
   1280x1024     76.25    75.02    60.02  
   1400x900      59.96    59.88  
   1280x960      60.00  
   1368x768      59.88    59.85  
   1280x800      59.81    59.91  
   1280x720      59.86    59.74  
   1152x768      68.35  
   1024x768      98.16    74.81    60.00    60.00  
   1024x576      59.90    59.82  
   832x624       74.18  
   960x540       59.63    59.82  
   800x600      186.01    94.87    60.32    56.25  
   768x576       99.99    79.37  
   864x486       59.92    59.57  
   640x480      116.65    59.94  
   720x405       59.51    58.99  
   640x360       59.84    59.32  
VGA1 disconnected (normal left inverted right x axis y axis)
DP1 disconnected (normal left inverted right x axis y axis)
HDMI1 disconnected (normal left inverted right x axis y axis)
$ xrandr -q --screen 1
Screen 1: minimum 320 x 200, current 1024 x 768, maximum 16384 x 16384
DisplayPort-1 disconnected primary (normal left inverted right x axis y axis)
DisplayPort-2 disconnected (normal left inverted right x axis y axis)
DisplayPort-3 disconnected (normal left inverted right x axis y axis)
VGA-1 disconnected (normal left inverted right x axis y axis)

Connecting a real screen:
$ xwininfo -display :0

xwininfo: Please select the window about which you
          would like information by clicking the
          mouse in that window.

xwininfo: Window id: 0x220064a "Desktop — Plasma"

  Absolute upper-left X:  1920
  Absolute upper-left Y:  0
  Relative upper-left X:  0
  Relative upper-left Y:  0
  Width: 1366
  Height: 768
  Depth: 32
  Visual: 0xc0
  Visual Class: TrueColor
  Border width: 0
  Class: InputOutput
  Colormap: 0x2200649 (not installed)
  Bit Gravity State: NorthWestGravity
  Window Gravity State: NorthWestGravity
  Backing Store State: NotUseful
  Save Under State: no
  Map State: IsViewable
  Override Redirect State: no
  Corners:  +1920+0  -0+0  -0-312  +1920-312
  -geometry 1366x768-0+0

Yes, this is different physical display than what I tried circa mga5.  There is possibly related issue (can't find where I reported this way back...) where the boot and lock screens show a small version of the display in the upper-left; its size was not affected (yet) by the attachment of a larger monitor. Will be interesting to see if that changes to 1366x768 at the next reboot...  Like this:
+-------+------+  1080x768 over 1920x1080
|       |      |  The two "screens" timeout individually. Moving the 
|       |      |  mouse over either brings the login dialog back up
+-------+      |  on that "screen".
|              |
+--------------+
Comment 11 Pierre Fortin 2020-04-27 02:09:15 CEST
Rebooted for kernel 5.6.6. The overlapping small screen stayed at the old size of 1080x768 -- I was expecting it to change to 1366x768...
Comment 12 Marja Van Waes 2020-07-31 13:00:21 CEST
Changing Version: to "7" because Pierre still has the mouse motion issue in Mageia 7

Version: 6 => 7

Comment 13 Aurelien Oudelet 2021-07-06 13:17:21 CEST
Mageia 7 is EOL since July 1st 2021.
There will not have any further bugfix for this release.

You are encouraged to upgrade to Mageia 8 as soon as possible.

@reporter, if this bug still apply with Mageia 8, please let us know it.

@packager, if you work on the Mageia 7 version of your package, please check the Mageia 8 package if issue is also present. In this case, please fix the Mageia 8 version instead.

This bug report will be closed OLD if there is no further notice within 1st September 2021.
Comment 14 Pierre Fortin 2021-07-07 03:51:24 CEST
Nothing has changed since initial report.

Version: 7 => 8

Comment 15 Aurelien Oudelet 2021-07-07 10:19:28 CEST
Hum.

You don't want the mouse cursor to go on the external display while it is "deactivated". If I understand, this external display is connected with VGA/DVI or whatever connector. You use Plasma.

Go to: systemsettings5 > Hardware > Display > Monitor.
There, make sure the internal is set to Main/Primary and then switch off the external display.
Mouse cursor should no longer go away.

In the case it is still an issue, you should consider remove physically the connector to external display.

There is nothing we can do more on this, as it is Desktop dependant, not a Kernels/Driver issue, nor a X issue.

Status: NEW => NEEDINFO
CC: (none) => ouaurelien

Comment 16 Pierre Fortin 2021-07-07 17:56:13 CEST
Created attachment 12849 [details]
Display Configuration

I only ever connected an external monitor for maybe an hour years ago, and decided against using it because the different geometry did not provide usable (IMO) additional real estate. Ever since then, I have NEVER connected an external monitor; but the system thinks one exists. The external display was switched off back then. I have tried every new release to sort this out; but there is NO external display in systemsettings5 > Hardware > Display > Monitor...

So... no external display connected, no sign of external monitor in system settings; yet system still allows mouse to move outside main/primary display into what was a once-upon-a-time short duration of a connection...  

Changing from:
   For any display arrangement
to:
   For only this specific display arrangement
likewise has no effect.
Comment 17 Pierre Fortin 2021-07-07 18:00:45 CEST
Created attachment 12850 [details]
/etc/X11/xorg.conf

If systemsettings5 > Hardware > Display does not alter /etc/X11/xorg.conf; then what file(s) does it control?  

/etc/X11/xorg.conf contains the second screen info that is likely causing this issue...
Comment 18 Aurelien Oudelet 2021-07-07 18:06:36 CEST
(In reply to Pierre Fortin from comment #17)
> Created attachment 12850 [details]
> /etc/X11/xorg.conf
> 
> If systemsettings5 > Hardware > Display does not alter /etc/X11/xorg.conf;
> then what file(s) does it control?  
> 
> /etc/X11/xorg.conf contains the second screen info that is likely causing
> this issue...

/etc/X11/xorg.conf should not contain more than 1 display informations. Please remove the ones for the "strange" display.

Moreover, with today code, Xorg should detect all devices available at load time and while running the system.

The systemsettings5 stuff is responsible to deal with multi monitors. It uses KScreen. I don't know which file is.
Comment 19 Dave Hodgins 2021-07-07 19:37:56 CEST
xorg.conf is the system wide settings set by XFdrake, aka mcc/Hardware/Set up the graphical server).

systemsettings5 is used by kde/plasma to override the systemwide settings for
a user. It mostly updates ~/.config/systemsettingsrc, though other config files
may be updated too.

As xorg can now usually detect the hardware correctly during startup, try
moving /etc/X11/xorg.conf to a location where it will not be used, such as
/root.

CC: (none) => davidwhodgins

Comment 20 Dave Hodgins 2021-07-07 19:39:38 CEST
Forgot to add, either restart X (hold down ctrl, press the backspace key twice)
or reboot after moving xorg.conf.
Comment 21 Pierre Fortin 2021-07-08 20:19:14 CEST
Renamed xorg.conf and restarted. System is now working correctly and no new xorg.conf file was created. It appears that this file is used if it exists; and none is [re-]created if it doesn't... Maybe it should have been removed when xorg didn't need it anymore...  Closing.

Status: NEEDINFO => RESOLVED
Resolution: (none) => FIXED

Comment 22 Morgan Leijström 2021-07-08 21:12:54 CEST
(In reply to Pierre Fortin from comment #21)
> It appears that this file is used if it exists;
> and none is [re-]created if it doesn't... Maybe it should have been removed
> when xorg didn't need it anymore...  Closing.

Possibly for errata then?
- Manual fix for then who experience this problem is to remove (move/rename) xorg.cong?

Keywords: (none) => FOR_ERRATA8

Comment 23 Aurelien Oudelet 2021-07-08 21:17:57 CEST
(In reply to Morgan Leijström from comment #22)
> (In reply to Pierre Fortin from comment #21)
> > It appears that this file is used if it exists;
> > and none is [re-]created if it doesn't... Maybe it should have been removed
> > when xorg didn't need it anymore...  Closing.
> 
> Possibly for errata then?
> - Manual fix for then who experience this problem is to remove (move/rename)
> xorg.cong?

Nope. This is an old bug that a possible workaround is to remove
/etc/X11/xorg.conf file.

The "bug" was in that file that add a mystery display in dual displays configuration.
Comment 24 Morgan Leijström 2021-07-08 21:20:16 CEST
OK

Keywords: FOR_ERRATA8 => (none)

Comment 25 Dave Hodgins 2021-07-08 21:33:17 CEST
(In reply to Pierre Fortin from comment #21)
> Renamed xorg.conf and restarted. System is now working correctly and no new
> xorg.conf file was created. It appears that this file is used if it exists;
> and none is [re-]created if it doesn't... Maybe it should have been removed
> when xorg didn't need it anymore...  Closing.

Thanks for the status update. Just fyi, xorg.conf is still needed on systems
where the monitor either doesn't return edid information (Extended Display
Identification Data), or it returns incorrect information.

Not very common anymore but there are still some old monitors in use that
need it.
Comment 26 Pierre Fortin 2021-07-08 23:52:31 CEST
Not sure what to suggest as a long term solution...  The display was connected once a few years ago. Since then, there has been no monitor to return EDID info, so xorg assumed xorg.conf was valid. Way back (mga 5 or 6), I could see the ghost monitor via MCC and disable it; but apparently, that did not remove the xorg.conf statements which have been plaguing me since...

Worse, the ghost monitor ended up displaying on the main screen on occasion, usually during login.

Thanks for the clue that finally restricted the mouse!
Comment 27 Dave Hodgins 2021-07-09 01:14:07 CEST
Another case where the edid info is not available that still happens is due to
poorly designed kvm devices that, while they allow sharing the same screen,
keyboard, and mouse with two or more computers, they do not allow the computer
to access the edid info from the monitor. Same with some devices to allow
connecting a monitor with a vga connector to a computer with hdmi or other video
output. The properly designed (more expensive ones) do allow it.

So xorg.conf is still needed some of the time.

Note You need to log in before you can comment on or make changes to this bug.