Ubuntu
linux package

External monitor turns off for short period of times / until moving mouse

Bug #1822854 reported by Julian Andres Klode on 2019-04-02

This bug affects 1 person

Affects		Status	Importance	Assigned to	Milestone
	linux (Ubuntu)	Triaged	Low	Unassigned

Bug Description

In disco; I have this issue where my external monitor (a ThinkVision P24h-10) connected to my ThinkPad T480s turns off for short times while actively working on the system. I noticed that if I don't move my mouse at all while it's off, it stays off.

The T480s is connected to the monitor via USB-C; the monitor provides DP, USB, and Power over that connection. When the display turns off, USB devices continue working.

I have attached /sys/debug/dri/0 output from when the monitor was off (a) vs on (b).

Other related bits:
* This also happens on HDMI, but much less often
* There are no log messages when the display turns on/off in the journal.

See original description

Tags:

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02: AlsaInfo.txt

AlsaInfo.txt Edit (45.5 KiB, text/plain)

apport information

tags:	added: apport-collected disco
description:	updated

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02: CRDA.txt

CRDA.txt Edit (2.9 KiB, text/plain)

apport information

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02: CurrentDmesg.txt

CurrentDmesg.txt Edit (236.9 KiB, text/plain)

apport information

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02: IwConfig.txt

IwConfig.txt Edit (1.6 KiB, text/plain)

apport information

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02: Lspci.txt

Lspci.txt Edit (15.7 KiB, text/plain)

apport information

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02: Lsusb.txt

Lsusb.txt Edit (1.0 KiB, text/plain)

apport information

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02: ProcCpuinfo.txt

ProcCpuinfo.txt Edit (10.0 KiB, text/plain)

apport information

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02: ProcCpuinfoMinimal.txt

ProcCpuinfoMinimal.txt Edit (1.2 KiB, text/plain)

apport information

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02: ProcEnviron.txt

ProcEnviron.txt Edit (311 bytes, text/plain)

apport information

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02: ProcInterrupts.txt

#10

ProcInterrupts.txt Edit (5.6 KiB, text/plain)

apport information

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02: ProcModules.txt

#11

ProcModules.txt Edit (10.4 KiB, text/plain)

apport information

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02: PulseList.txt

#12

PulseList.txt Edit (50.8 KiB, text/plain)

apport information

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02: RfKill.txt

#13

RfKill.txt Edit (183 bytes, text/plain)

apport information

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02: UdevDb.txt

#14

UdevDb.txt Edit (292.7 KiB, text/plain)

apport information

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02: WifiSyslog.txt

#15

WifiSyslog.txt Edit (1.5 MiB, text/plain)

apport information

description:

updated

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02:

#16

Display-off debug state Edit (4.8 MiB, application/octet-stream)

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02:

#17

Display-on debug state (b) Edit (4.9 MiB, application/octet-stream)

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02:

#18

Interesting (?) difference between the i915 debug files:

--- display off
+++ display on
- Active context: [0] user_handle 0 hw_id 0, prio 0, ban score 0 (unbannable) guilty 0 active 0
+ ELSP[0]: pid 12429, ban score 0, seqno 5:010cd84c, prio 0, emitted 0ms, start 00e21000, head 00003c50, tail 00003cd8
+ Active context: Xorg[12429] user_handle 1 hw_id 5, prio 0, ban score 0 guilty 0 active 0

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02:

#19

head of dmesg (journalctl -k) for that boot Edit (221.7 KiB, text/plain)

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-02:

#20

Also, FYI: There are no log messages when the display turns on/off in the journal.

description:

updated

Revision history for this message

Ubuntu Kernel Bot (ubuntu-kernel-bot) wrote on 2019-04-02: Status changed to Confirmed

#21

This change was made by a bot.

Changed in linux (Ubuntu):
status:	New → Confirmed

Revision history for this message

Daniel van Vugt (vanvugt) wrote on 2019-04-03:

#22

I don't think this is going to be a gnome-shell or mutter bug. But in case it is, please test some other desktop environments to verify.

Changed in gnome-shell (Ubuntu):
status:	New → Incomplete

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-03:

#23

It might be a kernel bug related to get/iommu which I had enabled to test with accelerated graphics in VMs. I think that the Kabylake are not really supported for that yet. I turned those off now.

Switching desktop environments would be hard. I can't work in a different de! My assumption was that maybe it was gnome powering down the monitor because it wrongly thought I was inactive or something.

IIRC, the same problem occurs in Wayland sessions, so it's at least not an X bug.

Revision history for this message

Hui Wang (hui.wang) wrote on 2019-04-03:

#24

@Julian,

Could you please tell the detailed steps to reproduce this problem? It looks like "you enabled the test with accelerated graphics in VMs", does it introduce this issue?

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-03:

#25

I had added i915.enable_gvt=1 intel_iommu=on on the kernel commandline to enable Intel GVT-g, and connected an external monitor (with USB devices attached, and supplying power to the laptop) via USB C.

It _seems_ that dropping the options makes it stable again, unless there were some more unrelated changes during the reboot, or I just did not hit it yet (though that seems unlikely).

Daniel van Vugt (vanvugt) on 2019-04-04

summary:	External monitor turns off for short period of times / until moving - mouse + mouse when using kernel options i915.enable_gvt=1 intel_iommu=on
no longer affects:	gnome-shell (Ubuntu)

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-04: Re: External monitor turns off for short period of times / until moving mouse when using kernel options i915.enable_gvt=1 intel_iommu=on

#26

Marking this as Triaged / Low, as it seems to be only caused by these unusual options.

Changed in linux (Ubuntu):
importance:	Undecided → Low
status:	Confirmed → Triaged

Revision history for this message

Julian Andres Klode (juliank) wrote on 2019-04-05:

#27

Just happened without those options. But I'm not sure if it's a kernel bug, or a hardware/cable bug. Gotta investigate more I guess. It certainly happened less, the last two days have been issue free, but maybe I unplugged more and the cable got cleaner or something.

summary:

External monitor turns off for short period of times / until moving
- mouse when using kernel options i915.enable_gvt=1 intel_iommu=on
+ mouse

Report a bug

This report contains Public information

Everyone can see this information.

You are

Subscribing...

Edit bug mail

Other bug subscribers

Bug attachments

Add attachment

Remote bug watches

Bug watches keep track of this bug in other bug trackers.

Ubuntulinux package

External monitor turns off for short period of times / until moving mouse

Bug Description

Other bug subscribers

Bug attachments

Remote bug watches

Ubuntu
linux package