SUMMARY I've been having lots of issues with my computer going to sleep lately. When I let the system go to sleep on it's own from inactivity, it will turn the monitors off for a few seconds, then immediately wake up. It'll then be in a state in which if I play a video, it will freeze for several seconds or minutes. To get out of this state, I have to type "systemctl --user restart plasma-powerdevil.service", and it'll be able to play videos without freezing, but it still won't be able to sleep. Note: This only seems to happen with "Turn off screen", not with actually sleeping. These are the logs from the last time I tried to let the system turn off the screen on it's own: Jun 22 08:47:38 arch-desktop dbus-broker[328292]: A security policy denied :1.32 to send method call /org/freedesktop/login1/seat/seat0:org.freedesktop.login1.Seat.Inhibit to org.freedesktop.login1. Jun 22 08:47:38 arch-desktop kwin_wayland[328622]: kwin_core: Failed to delay sleep: Sender is not authorized to send message Jun 22 08:47:38 arch-desktop dbus-broker[328292]: A security policy denied :1.32 to send method call /org/freedesktop/login1/seat/seat0:org.freedesktop.login1.Seat.Inhibit to org.freedesktop.login1. Jun 22 08:47:38 arch-desktop kwin_wayland[328622]: kwin_core: Failed to delay sleep: Sender is not authorized to send message Jun 22 08:47:51 arch-desktop kernel: amdgpu 0000:03:00.0: [drm] *ERROR* dpcd_set_link_settings:1118: core_link_write_dpcd (DP_DOWNSPREAD_CTRL) failed Jun 22 08:47:51 arch-desktop kernel: amdgpu 0000:03:00.0: [drm] *ERROR* dpcd_set_link_settings:1123: core_link_write_dpcd (DP_LANE_COUNT_SET) failed Jun 22 08:47:51 arch-desktop kernel: amdgpu 0000:03:00.0: [drm] *ERROR* dpcd_set_link_settings:1151: core_link_write_dpcd (DP_LINK_BW_SET) failed Jun 22 08:47:52 arch-desktop kernel: amdgpu 0000:03:00.0: [drm] *ERROR* dpcd_set_link_settings:1118: core_link_write_dpcd (DP_DOWNSPREAD_CTRL) failed Jun 22 08:47:52 arch-desktop kernel: amdgpu 0000:03:00.0: [drm] *ERROR* dpcd_set_link_settings:1123: core_link_write_dpcd (DP_LANE_COUNT_SET) failed Jun 22 08:47:52 arch-desktop kernel: amdgpu 0000:03:00.0: [drm] *ERROR* dpcd_set_link_settings:1151: core_link_write_dpcd (DP_LINK_BW_SET) failed STEPS TO REPRODUCE 1. Wait for computer to sleep (Amount of time specified in "Settings/Power Management/Turn off screen", for me it's 30 minutes) OBSERVED RESULT Computer doesn't go to sleep EXPECTED RESULT Computer goes to sleep SOFTWARE/OS VERSIONS Operating System: Arch Linux KDE Plasma Version: 6.4.0 KDE Frameworks Version: 6.15.0 Qt Version: 6.9.1 Kernel Version: 6.15.2-arch1-1 (64-bit) Graphics Platform: Wayland Processors: 12 × AMD Ryzen 5 7600X 6-Core Processor Memory: 32 GiB of RAM (30.5 GiB usable) Graphics Processor 1: AMD Radeon RX 7800 XT Graphics Processor 2: AMD Radeon Graphics ADDITIONAL INFORMATION This seems similar to https://bugs.kde.org/show_bug.cgi?id=480026. The issue says "resolve/fixed", but I still have this problem, so I don't know if it's the same one. If I run the same command as in the bug, `dbus-send --session --print-reply --dest=org.kde.kglobalaccel /component/org_kde_powerdevil org.kde.kglobalaccel.Component.invokeShortcut string:'Turn Off Screen'`, I get a similar looking issue, and this shows up in my journalctl: Jun 22 22:16:04 arch-desktop dbus-broker[599]: A security policy denied :1.39 to send method call /org/freedesktop/login1/seat/seat0:org.freedesktop.login1.Seat.Inhibit to org.freedesktop.login1. Jun 22 22:16:04 arch-desktop kwin_wayland[1102]: kwin_core: Failed to delay sleep: Sender is not authorized to send message Jun 22 22:16:04 arch-desktop dbus-broker[599]: A security policy denied :1.39 to send method call /org/freedesktop/login1/seat/seat0:org.freedesktop.login1.Seat.Inhibit to org.freedesktop.login1. Jun 22 22:16:04 arch-desktop kwin_wayland[1102]: kwin_core: Failed to delay sleep: Sender is not authorized to send message Jun 22 22:16:17 arch-desktop org_kde_powerdevil[8676]: [ 8695] Removing connected display on bus 9 Jun 22 22:16:17 arch-desktop org_kde_powerdevil[8676]: [ 8695] Emitting DDCA_Display_Status_Event[1096.040: DDCA_EVENT_DISPLAY_DISCONNECTED, card0-HDMI-A-1, dref: DDCA_Display_Ref[1], io_path:/dev/i2c-9, ddc working: false] Jun 22 22:16:17 arch-desktop org_kde_powerdevil[8676]: [ 8695] libddcutil callback thread 0x7f017402c010 started Jun 22 22:16:17 arch-desktop org_kde_powerdevil[8676]: [ 8695] Started 1 event callback thread(s) Jun 22 22:16:17 arch-desktop flatpak[2877]: gsr error: gsr_capture_kms_update_connector_ids: failed to find monitor by name "DP-3" Jun 22 22:16:18 arch-desktop org_kde_powerdevil[8676]: [ 8695] Adding connected display with bus 9 Jun 22 22:16:19 arch-desktop flatpak[2877]: gsr error: gsr_capture_kms_update_connector_ids: failed to find monitor by name "DP-3" Jun 22 22:16:20 arch-desktop org_kde_powerdevil[8676]: [ 8695] Emitting DDCA_Display_Status_Event[1099.084: DDCA_EVENT_DISPLAY_CONNECTED, card0-HDMI-A-1, dref: DDCA_Display_Ref[3], io_path:/dev/i2c-9, ddc working: true] Jun 22 22:16:20 arch-desktop org_kde_powerdevil[8676]: [ 8695] libddcutil callback thread 0x7f017400bf90 started Jun 22 22:16:20 arch-desktop org_kde_powerdevil[8676]: [ 8695] Started 1 event callback thread(s) Jun 22 22:16:20 arch-desktop org_kde_powerdevil[8676]: [ 8676] Quiescing libddcutil API... Jun 22 22:16:20 arch-desktop org_kde_powerdevil[8676]: [ 8676] Quiesce libddcutil API complete Jun 22 22:16:20 arch-desktop org_kde_powerdevil[8676]: [ 8676] Display redetection starting. Jun 22 22:16:20 arch-desktop org_kde_powerdevil[8676]: [ 8694] recheck thread terminating because watch thread terminated Jun 22 22:16:20 arch-desktop org_kde_powerdevil[8676]: [ 8676] Watch thread terminated. Jun 22 22:16:21 arch-desktop org_kde_powerdevil[8676]: Traced function stack on entry to i2c_check_open_bus_alive() Jun 22 22:16:21 arch-desktop org_kde_powerdevil[8676]: No traced function stack for current thread Jun 22 22:16:21 arch-desktop flatpak[2877]: gsr error: gsr_capture_kms_update_connector_ids: failed to find monitor by name "DP-3" Jun 22 22:16:22 arch-desktop org_kde_powerdevil[8676]: [ 8676] (i2c_check_open_bus_alive) Retrying i2c_check_edid_exists_by_dh() (c): Slept for 1000 millisec Jun 22 22:16:22 arch-desktop org_kde_powerdevil[8676]: [ 8676] /dev/i2c-12: Checking EDID succeeded after 3 tries (G) Jun 22 22:16:22 arch-desktop org_kde_powerdevil[8676]: [ 8676] (check_supported_feature) busno=12, sleep-multiplier = 0.50. Testing for supported feature 0x10 returned Error_Info[DDCRC_DPMS_ASLEEP in i2c_check_open_bus_alive] Jun 22 22:16:22 arch-desktop org_kde_powerdevil[8676]: [ 8676] Watching for display connection changes, resolved watch mode = Watch_Mode_Xevent, poll loop interval = 100 millisec Jun 22 22:16:22 arch-desktop org_kde_powerdevil[8676]: [ 8676] extra_stabilization_millisec: 0, stabilization_poll_millisec: 100 Jun 22 22:16:22 arch-desktop org_kde_powerdevil[8676]: [ 8676] libddcutil recheck thread (nil) started Jun 22 22:16:22 arch-desktop org_kde_powerdevil[8676]: [ 8676] libddcutil watch thread 0x5636eb61ee60 started Jun 22 22:16:22 arch-desktop org_kde_powerdevil[8676]: [ 8676] Display redetection finished. Jun 22 22:16:22 arch-desktop org_kde_powerdevil[8676]: [ 8676] Unquiescing libddcutil API... Jun 22 22:16:22 arch-desktop org_kde_powerdevil[8676]: [ 11980] (dw_recheck_displays_func) Recheck interval: Slept for 200 millisec Jun 22 22:16:22 arch-desktop kwin_wayland[1102]: kwin_xwl: Could not find a matching X RandR CRTC/output to set as primary for KWin::DrmOutput(0x55baa9561f30, name="DP-3", geometry=QRect(0,0 1920x1080), scale=1) Jun 22 22:16:22 arch-desktop kwin_wayland_wrapper[1196]: XWAYLAND: mode -1x-1 is not available Jun 22 22:16:22 arch-desktop flatpak[10682]: [44:0622/221622.974027:ERROR:ui/gl/gl_display.cc:508] EGL Driver message (Error) eglGetMscRateANGLE: glXGetMscRateOML failed. Jun 22 22:16:22 arch-desktop kwin_wayland_wrapper[1196]: XWAYLAND: mode -1x-1 is not available Jun 22 22:16:22 arch-desktop flatpak[10682]: [44:0622/221622.987079:ERROR:ui/gl/gl_display.cc:508] EGL Driver message (Error) eglGetMscRateANGLE: glXGetMscRateOML failed. Jun 22 22:16:22 arch-desktop kwin_wayland_wrapper[1196]: XWAYLAND: mode -1x-1 is not available Jun 22 22:16:22 arch-desktop flatpak[10682]: [44:0622/221622.992956:ERROR:ui/gl/gl_display.cc:508] EGL Driver message (Error) eglGetMscRateANGLE: glXGetMscRateOML failed. Jun 22 22:16:23 arch-desktop kwin_wayland_wrapper[1196]: XWAYLAND: mode -1x-1 is not available Jun 22 22:16:23 arch-desktop flatpak[10682]: [44:0622/221623.008419:ERROR:ui/gl/gl_display.cc:508] EGL Driver message (Error) eglGetMscRateANGLE: glXGetMscRateOML failed. Jun 22 22:16:23 arch-desktop kwin_wayland[1102]: kwin_scene_opengl: 0x3: GL_INVALID_VALUE in glTexSubImage2D(yoffset 0 + height 1036 > 982)
I'm not able to reproduce this on a system running git-master that has an NVIDIA GPU Running the dbus command, this is all that's written to the journal `dbus-send --session --print-reply --dest=org.kde.kglobalaccel /component/org_kde_powerdevil org.kde.kglobalaccel.Component.invokeShortcut string:'Turn Off Screen'` Jun 24 17:11:56 tlc-xps17-solus dbus-broker[1249]: A security policy denied :1.43 to send method call /org/freedesktop/login1/seat/seat0:org.freedesktop.login1.Seat.Inhibit to org.freedesktop.login1. Jun 24 17:11:56 tlc-xps17-solus kwin_wayland[2247]: kwin_core: Failed to delay sleep: Sender is not authorized to send message Jun 24 17:11:56 tlc-xps17-solus dbus-broker[1249]: A security policy denied :1.43 to send method call /org/freedesktop/login1/seat/seat0:org.freedesktop.login1.Seat.Inhibit to org.freedesktop.login1. Jun 24 17:11:56 tlc-xps17-solus kwin_wayland[2247]: kwin_core: Failed to delay sleep: Sender is not authorized to send message Jun 24 17:11:56 tlc-xps17-solus dbus-broker[1249]: A security policy denied :1.43 to send method call /org/freedesktop/login1/seat/seat0:org.freedesktop.login1.Seat.Inhibit to org.freedesktop.login1. Jun 24 17:11:56 tlc-xps17-solus kwin_wayland[2247]: kwin_core: Failed to delay sleep: Sender is not authorized to send message The screens are turned off and stay off until I press a key or move the mouse. I waited 30 seconds. Same behavior on a system on git-master with an AMD GPU, and on that system with a 6.3.5 Plasma session I'll leave this open so others can try reproducing
I'm trying to hunt down the same problem and I've found something that may be useful for others to reproduce it easier. First, (in my case) you don't need to wait for system to sleep but instead set in System->Power Management "Turn off screen when locked: 20 seconds". Then simply lock the screen and wait if the screen properly turns off as it should, but sometimes the screen goes black and after a couple of seconds it is woke up, then after 20 seconds goes black and again woke up. Now the conditions in my case: I have Fedora 42 with Plasma 6.4.1 installed on my external SSD that I plug into 2 separate laptops. First laptop is with Intel 11th gen iGPU only, the second laptop is Intel 12th gen iGPU + nVidia dGPU. 1. If I have external monitor unplugged it *always* works - not reproducible 2. If I plug external monitor to iGPU only laptop as USB-C/Display Port/Thunderbolt - not reproducible 3. If I plug external monitor to iGPU only laptop as HDMI - *always reproducible* 4. If I plug external monitor to 12th gen laptop to USB-C which is hardwired to Intel iGPU - not reproducible 5. If I plug external monitor to 12th gen laptop to HDMI which is hardwired to nVidia dGPU - not reproducible So as of now I'd say someone needs to try plugging external monitor through HDMI that is connected to intel iGPU. I was using Fedora 42 with Gnome+GDM and there was not such a problem in any of the above configuration so I don't think the problem is somewhere in drivers/firmware.
(In reply to SzymonP from comment #2) > I'm trying to hunt down the same problem and I've found something that may > be useful for others to reproduce it easier. Thank you for the detailed testing! > So as of now I'd say someone needs to try plugging external monitor through > HDMI that is connected to intel iGPU. I tried re-testing with a laptop that only has an integrated AMD GPU connected to HDMI, with your testing steps, and couldn't reproduce. This lends weight to this potentially being tied to having an integrated Intel GPU connected to HDMI. Hopefully someone else with this hardware can confirm.
> Hopefully someone else with this hardware can confirm. Tried it with a Kaby Lake powered laptop, but could not reproduce it on Plasma 6.4.1 + Kernel 6.15.3 (Fedora 42).
Based on the symptoms and some of the kwin log entries this looks like bug 493879. I'm going to merge this report into that older one, please follow that one for updates. *** This bug has been marked as a duplicate of bug 493879 ***