Public bug reported:
On a three-monitor setup, locking the screen (Super+L) makes application windows
migrate to the rightmost monitor (DP-1). Window positions are not restored after
unlocking; they have to be moved back by hand every time.
This is a regression. Kernel 6.8.0-136-generic reproduces it. Kernel
6.8.0-124-generic, booted on the same otherwise unchanged system, does not.
## System
Ubuntu 24.04.4 LTS (noble)
GNOME Shell 46.0, Wayland session (mutter 46.0-0ubuntu6~24.04.14)
Two GPUs:
00:02.0 Intel RocketLake-S GT1 [UHD Graphics 750] [8086:4c8a] — driver i915
01:00.0 AMD/ATI Caicos [Radeon HD 6450/7450/8450 / R5 230 OEM] [1002:6779] —
driver radeon
Monitor layout (xrandr):
HDMI-1 1920x1080+0+0 (i915) leftmost
HDMI-4 1920x1080+1920+0 (radeon) centre, primary
DP-1 1920x1080+3840+0 (i915) rightmost <-- windows end up here
DRM connectors:
card1-DVI-D-1 disconnected [radeon]
card1-HDMI-A-4 connected [radeon]
card1-VGA-1 disconnected [radeon]
card2-DP-1 connected [i915]
card2-HDMI-A-1 connected [i915]
card2-HDMI-A-2 disconnected [i915]
card2-HDMI-A-3 disconnected [i915]
## Kernel versions tested
linux-image-6.8.0-136-generic 6.8.0-136.136 REPRODUCES
linux-image-6.8.0-124-generic 6.8.0-124.124 does NOT reproduce
## How it was tested
A polling watcher sampled /sys/class/drm/card*-*/status every 100 ms, together
with xrandr output geometry, while the screen was locked and unlocked.
On 6.8.0-136: card2-DP-1 was observed transitioning to "disconnected" and back
to "connected" — a link drop of roughly 1.2 seconds — coinciding with the lock,
and windows had moved to DP-1 afterwards.
On 6.8.0-124: three consecutive lock/unlock cycles produced no change at all —
DP-1 stayed "connected" throughout, xrandr geometry was unchanged, and no window
moved. The kernel log contains no DRM/HPD/link messages for that period.
## Note on the mechanism
I have not established the mechanism, and I want to be explicit that the
following is a hypothesis rather than a finding.
The affected connector (DP-1) belongs to i915, but the only suggestive line in
the 6.8.0-136 kernel log is from the radeon driver:
workqueue: radeon_hotplug_work_func [radeon] hogged CPU for >10000us 4 times,
consider switching to WQ_UNBOUND
No explicit DP-1 link-loss message was logged on 6.8.0-136 — the drop was only
visible through /sys connector status polling, as the system was running without
drm.debug enabled. So it is possible that repeated radeon hotplug processing
triggers a re-probe of all outputs, mutter recomputes the monitor configuration,
and windows land on the rightmost output as a side effect. That would make this
a
radeon-side regression rather than an i915 one. I cannot confirm either way from
the data I have.
I am happy to reboot into 6.8.0-136 with drm.debug=0x1e and attach a full debug
log if that would help narrow it down.
## Workaround in use
Pinning 6.8.0-124-generic as the default boot entry.
ProblemType: Bug
DistroRelease: Ubuntu 24.04
Package: linux-image-6.8.0-124-generic 6.8.0-124.124
ProcVersionSignature: Ubuntu 6.8.0-124.124-generic 6.8.12
Uname: Linux 6.8.0-124-generic x86_64
ApportVersion: 2.28.2-0ubuntu0.1
Architecture: amd64
AudioDevicesInUse:
USER PID ACCESS COMMAND
/dev/snd/controlC1: ruslan 7993 F.... wireplumber
/dev/snd/controlC0: ruslan 7993 F.... wireplumber
/dev/snd/seq: ruslan 7986 F.... pipewire
CRDA: N/A
CasperMD5CheckResult: pass
CurrentDesktop: ubuntu:GNOME
Date: Wed Aug 5 07:19:37 2026
HibernationDevice: RESUME=UUID=f2686375-0636-44de-a71b-65e576abbeac
resume_offset=42792960
InstallationDate: Installed on 2022-07-21 (1476 days ago)
InstallationMedia: Ubuntu 22.04 LTS "Jammy Jellyfish" - Release amd64 (20220419)
MachineType: ASUS System Product Name
ProcFB:
0 radeondrmfb
1 i915drmfb
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-6.8.0-124-generic
root=UUID=f2686375-0636-44de-a71b-65e576abbeac ro quiet splash
resume=f2686375-0636-44de-a71b-65e576abbeac resume_offset=42792960 vt.handoff=7
PulseList: Error: command ['pacmd', 'list'] failed with exit code 1: No
PulseAudio daemon running, or not running as session daemon.
RelatedPackageVersions:
linux-restricted-modules-6.8.0-124-generic N/A
linux-backports-modules-6.8.0-124-generic N/A
linux-firmware 20240318.git3b128b60-0ubuntu2.27
RfKill:
0: hci0: Bluetooth
Soft blocked: no
Hard blocked: no
SourcePackage: linux
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 01/26/2021
dmi.bios.release: 4.3
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: 0403
dmi.board.asset.tag: Default string
dmi.board.name: PRIME B560M-A
dmi.board.vendor: ASUSTeK COMPUTER INC.
dmi.board.version: Rev 1.xx
dmi.chassis.asset.tag: Default string
dmi.chassis.type: 3
dmi.chassis.vendor: Default string
dmi.chassis.version: Default string
dmi.modalias:
dmi:bvnAmericanMegatrendsInc.:bvr0403:bd01/26/2021:br4.3:svnASUS:pnSystemProductName:pvrSystemVersion:rvnASUSTeKCOMPUTERINC.:rnPRIMEB560M-A:rvrRev1.xx:cvnDefaultstring:ct3:cvrDefaultstring:skuSKU:
dmi.product.family: To be filled by O.E.M.
dmi.product.name: System Product Name
dmi.product.sku: SKU
dmi.product.version: System Version
dmi.sys.vendor: ASUS
** Affects: linux (Ubuntu)
Importance: Undecided
Status: New
** Tags: amd64 apport-bug noble wayland-session
** Attachment added: "журнал ядра сессии, где проблема воспроизводилась (04.08)"
https://bugs.launchpad.net/bugs/2162881/+attachment/5989621/+files/kernel-6.8.0-136.log
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2162881
Title:
Windows migrate to rightmost DisplayPort monitor on screen lock —
regression between 6.8.0-124 and 6.8.0-136
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2162881/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
