Bug 1778886 - built-in laptop screen remains "on" when docked and lid closed
Summary: built-in laptop screen remains "on" when docked and lid closed
Keywords:
Status: CLOSED DUPLICATE of bug 1748997
Alias: None
Product: Fedora
Classification: Fedora
Component: kernel
Version: 30
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
Assignee: Kernel Maintainer List
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2019-12-02 18:00 UTC by Gabriel Somlo
Modified: 2019-12-03 13:30 UTC (History)
23 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2019-12-03 13:30:38 UTC
Type: Bug
Embargoed:


Attachments (Terms of Use)
RPMs present on the live image that are not installed on my custom kickstarted system (1.86 KB, text/plain)
2019-12-02 18:00 UTC, Gabriel Somlo
no flags Details
list of modules used on the laptop (output of lsmod) (4.46 KB, text/plain)
2019-12-02 18:27 UTC, Gabriel Somlo
no flags Details

Description Gabriel Somlo 2019-12-02 18:00:37 UTC
Created attachment 1641460 [details]
RPMs present on the live image that are not installed on my custom kickstarted system

Description of problem:
When docked (via usb-c) with an external monitor connected, the built-in laptop screen remains turned on when the lid is closed.

Version-Release number of selected component (if applicable):
systemd-241-12.git323cdf4.fc30.x86_64

How reproducible:
connect external screen (via usb-c dock in my case), and close the lid

Actual results:
built-in screen remains active in "join displays" mode, same as when lid is open

Expected results:
built-in screen turns off, only external display remains available in "single display" mode

Additional info:
This worked when I tried it via the F30 Workstation "Live" image, but stopped working after I installed a custom kickstart file which includes @gnome-desktop
in its %package section. The custom kickstart is at http://mirror.ini.cmu.edu/ks/errol30.ks, and I'm attaching the list of RPMs that are installed on the Live image and NOT on the system I kickstarted -- none of them looks like it might contribute additional configuration that results in the right behavior, though.

At this point I think it's rather a bug in either systemd-logind or whatever it calls to turn screens on/off when the lid switch is triggered.

Comment 1 Zbigniew Jędrzejewski-Szmek 2019-12-02 18:08:10 UTC
Please report the details of your kernel version and graphics card and the
drivers you are using.

It might also help to boot into text mode (add "3" on the kernel command line),
and see if the issue still occurs then. AFAIK it's the kernel drivers that
turns of the screen (or firmware?), and the userspace does not control this.

Comment 2 Gabriel Somlo 2019-12-02 18:25:55 UTC
"uname -a": "Linux errol.ini.cmu.edu 5.3.12-200.fc30.x86_64 #1 SMP Thu Nov 21 22:47:04 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux"

"lspci | grep -i vga": "00:02.0 VGA compatible controller: Intel Corporation UHD Graphics 620 (Whiskey Lake)"

The laptop is a System76 Galago Pro.

After running "init 3" and logging in as root, the screens were mirrored (same stuff I typed in one showed up in the other), and closing the lid still did not change the resolution on the external monitor, or give any indication that the built-in screen was turned off. Although I'm not sure wayland (or X11) is a valid comparison w.r.t. text mode...

The annoying thing is that this worked fine on the Live image, but 1. I'm using up-to-date RPMs (so a regression somewhere in there is likely), and 2. I'm not running with a strict superset of the packages that were installed on Live (the ones I don't have currently installed are in the attached list). My bet is on regression (#1), but would be happy if we narrowed it down to some package that should (but currently isn't) listed under @gnome-desktop in comps... Thanks!

Comment 3 Gabriel Somlo 2019-12-02 18:27:07 UTC
Created attachment 1641463 [details]
list of modules used on the laptop (output of lsmod)

Comment 4 Gabriel Somlo 2019-12-02 21:12:30 UTC
re. first attachment (rpm packages installed by default with the Live image that didn't get pulled in by my custom kickstart): I manually installed all but two of the RPM packages mentioned there (fedora-release-workstation and pulseaudio-module-bluetooth, which conflicted with already installed packages), rebooted, and still experienced the problem. So, unless the magic is in "fedora-release-workstation" itself, I think this is strong evidence that we're looking at a regression.

Comment 5 Gabriel Somlo 2019-12-03 13:30:38 UTC
thanks to logwatch, I figured out what the problem is -- failure of upower.service to start properly. This is a duplicate of #1748997.

*** This bug has been marked as a duplicate of bug 1748997 ***


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