11

Periodically (every few reboots by the look of it) my external VGA output on my Laptop changes name between VGA1 and VGA2. For example right now xrandr is displaying:

timp@helez:~$ xrandr | grep VGA
VGA2 connected 1680x1050+0+0 (normal left inverted right x axis y axis) 433mm x 271mm

But next time I reboot my system that could easily swap to VGA1 instead.

I do have an Optimus graphics card, so that may be related, and I did in the past have Bumblebee installed, but I have since removed all the related packages and/or downgraded them back to their standard versions.

Has anyone seen anything like this or have any ideas how to fix it?

More info on my system:

  • Laptop Model: Acer Aspire 5830TG
  • Distribution: Slackware 14.1 64-bit
  • GPU: NVIDIA GF 520M:

    timp@helez:~$ /sbin/lspci | grep VGA
    00:02.0 VGA compatible controller: Intel Corporation 2nd Generation Core Processor Family Integrated Graphics Controller (rev 09)
    01:00.0 VGA compatible controller: NVIDIA Corporation GF119M [GeForce GT 520M] (rev a1)
    
  • Xorg Version Info:

    timp@helez:~$ Xorg -version
    
    X.Org X Server 1.14.3
    Release Date: 2013-09-12
    X Protocol Version 11, Revision 0
    Build Operating System: Slackware 14.1 Slackware Linux Project
    Current Operating System: Linux helez 3.10.17 #2 SMP Wed Oct 23 16:34:38 CDT 2013 x86_64
    Kernel command line: auto BOOT_IMAGE=Linux ro root=801 vt.default_utf8=1
    Build Date: 09 October 2013  08:27:11PM
    
    Current version of pixman: 0.30.2
        Before reporting problems, check http://wiki.x.org
        to make sure that you have the latest version.
    
  • Video Driver: Intel:

    timp@helez:~$ grep intel /var/log/Xorg.0.log
    [   285.689] (II) LoadModule: "intel"
    [   285.691] (II) Loading /usr/lib64/xorg/modules/drivers/intel_drv.so
    [   285.733] (II) Module intel: vendor="X.Org Foundation"
    [   285.733] (II) intel: Driver for Intel(R) Integrated Graphics Chipsets:
    [snip a bunch more lines of output]
    
  • Window Manager: i3 version 4.7.2
  • 1
    What problem is this actually causing you? I assume you are using the screen's name in a script or similar, it might be easier to fix that rather than the name. – terdon Feb 20 '14 at 15:22
  • As part of my startx command I run a script to change the screen layout, but that's easy enough to fix. The more serious problem for me is that i3 uses the screen name in it's configuration files. (So that I can make sure certain desktops and windows go to certain screens) – TimP Feb 20 '14 at 21:32
  • You might be able to run a little script after starting X and before launching i3 (just before the i3 command in your startx) that gets the screen name from xrand and modifies the conf files accordingly. – terdon Feb 21 '14 at 2:45
  • @terdon, Yeah, that's certainly possible, and I probably should do it, but I keep just doing it manually, and I hope I can get a proper solution instead. – TimP Feb 23 '14 at 5:38
  • 1
    I have the same problem, I'm building a machine to drive a monitor wall that has 4 graphics cards in it. I think the root cause is that the kernel attaches the KMS/DRM drivers to devices in parallel so that it can sometimes attach one driver before the other which changes the monitor names. Finding a way to force a particular attachment order would fix the problem, but I don't know how to do that :/ – JasperWallace Mar 29 '15 at 15:29
1

You need to parse xrandr output (or write a program, or whatever) to find the correct names, and massage your i3 configuration appropriately.

The order of device detection is random, you can't depend on it.

  • Yeah, that's essentially what I ended up doing. – TimP Feb 25 '16 at 4:01
0

use cvs command to make a new resolution in your system for example:cvs 1280 768 60 and the give the commands xrandr --new mode,--add mode followed by your desktop resolution and then paste both the commands in linux directory nano /etc/gdm/Init/Default so that it does not change on reboot.

0

You can see in your dmesg the order of discovery of the graphics chips. If they vary with the time, you could perhaps fix this compiling your stuff in your kernel not as a module. Or trying a workaround putting a rule that indicate the order of loading for the modules concerned...

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.