SUMMARY Before 6.3.2 update, I had no issue reaching SDDM login page. After updating from 6.3.1 to 6.3.2, SDDM login page does not appear, it is just a blackscreen. Switching to TTY and launching a Plasma Wayland session from there result in the same black screen. I run 4 screens: Unplugging one of those screen let me get access to SDDM. Important to note that the issue was not present in 6.3.1 STEPS TO REPRODUCE 1. Upgrade to 6.3.2 Plasma 2. Start the PC with multiple monitors 3. -- OBSERVED RESULT Blackscreen Log: Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: Invalid framebuffer status: "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT" Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: 0x502: GL_INVALID_OPERATION error generated. <image> and <target> are inco> Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: Invalid framebuffer status: "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT" Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: 0x502: GL_INVALID_OPERATION error generated. <image> and <target> are inco> Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: Invalid framebuffer status: "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT" Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: 0x502: GL_INVALID_OPERATION error generated. <image> and <target> are inco> Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: Invalid framebuffer status: "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT" Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: 0x502: GL_INVALID_OPERATION error generated. <image> and <target> are inco> Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: Invalid framebuffer status: "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT" Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: 0x502: GL_INVALID_OPERATION error generated. <image> and <target> are inco> Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: Invalid framebuffer status: "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT" Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: 0x502: GL_INVALID_OPERATION error generated. <image> and <target> are inco> Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: Invalid framebuffer status: "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT" Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: 0x502: GL_INVALID_OPERATION error generated. <image> and <target> are inco> Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_core: Applying output config failed! Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: Invalid framebuffer status: "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT" Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: 0x502: GL_INVALID_OPERATION error generated. <image> and <target> are inco> Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: Invalid framebuffer status: "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT" Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: 0x502: GL_INVALID_OPERATION error generated. <image> and <target> are inco> Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: Invalid framebuffer status: "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT" Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: 0x502: GL_INVALID_OPERATION error generated. <image> and <target> are inco> Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: Invalid framebuffer status: "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT" Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: 0x502: GL_INVALID_OPERATION error generated. <image> and <target> are inco> Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_core: Applying output config failed! Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: Invalid framebuffer status: "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT" Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: 0x502: GL_INVALID_OPERATION error generated. <image> and <target> are inco> Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: Invalid framebuffer status: "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT" Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: 0x502: GL_INVALID_OPERATION error generated. <image> and <target> are inco> Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: Invalid framebuffer status: "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT" Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: 0x502: GL_INVALID_OPERATION error generated. <image> and <target> are inco> Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: Invalid framebuffer status: "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT" Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_scene_opengl: 0x502: GL_INVALID_OPERATION error generated. <image> and <target> are inco> Feb 27 11:12:21 CachyOSBTW kwin_wayland[2990]: kwin_core: Applying output config failed! EXPECTED RESULT Normal access to SDDM / Desktop SOFTWARE/OS VERSIONS Linux/KDE Plasma: CachyOS / Arch Linux (tested with both repos) / KDE Plasma 6.3.2 KDE Plasma Version: 6.3.2 KDE Frameworks Version: 6.11.0 Qt Version: 6.8.2 ADDITIONAL INFORMATION Unplugging and re pluggin the 4th screen while in the desktop causes the exact same issue. At disposition if more info is required. Cheers!
Want to add more info: - I can reproduce the blackscreen when on the desktop by just adding a 4th screen. It looks like it is the hard limit for whatever reason - I use an Nvidia card. One member of my community using an Intel Arc card was able to reproduce the same bug with his hardware. Hope this helps!
Can confirm something is going on connecting a fourth monitors. Connecting a fourth monitor during a plasma session or sitting at/booting into SDDM login: Instant black screen. Switching ports around on the graphics card and therefore the added, fourth monitor isn't the same monitor (considering port enumeration or physical connector [DisplayPort vs HDMI]) doesn't seem to make a difference. --- Core was generated by `/usr/bin/kwin_wayland --wayland-fd 7 --socket wayland-0 --xwayland-fd 8 --xwayland-fd 9 --xwayland-display :0 --xwayland-xauthority /run/user/1000/xauth_jNIYIK --xwayland'. Program terminated with signal SIGSEGV, Segmentation fault. #0 0x000071ee89ab0b0b in pthread_kill () from /usr/lib/libc.so.6 [Current thread is 1 (Thread 0x71ee833a1a40 (LWP 2963))] Cannot QML trace cores :( /usr/share/drkonqi/gdb/python/gdb_preamble/preamble.py:547: DeprecationWarning: datetime.datetime.utcfromtimestamp() is deprecated and scheduled for removal in a future version. Use timezone-aware objects to represent datetimes in UTC: datetime.datetime.fromtimestamp(timestamp, datetime.UTC). boot_time = datetime.utcfromtimestamp(psutil.boot_time()).strftime('%Y-%m-%dT%H:%M:%S') /usr/share/drkonqi/gdb/python/gdb_preamble/preamble.py:564: DeprecationWarning: datetime.datetime.utcnow() is deprecated and scheduled for removal in a future version. Use timezone-aware objects to represent datetimes in UTC: datetime.datetime.now(datetime.UTC). 'timestamp': datetime.utcnow().isoformat(), [Current thread is 1 (Thread 0x71ee833a1a40 (LWP 2963)) Thread 1 (Thread 0x71ee833a1a40 (LWP 2963)): [KCrash Handler] #4 0x000071ee8d0ca593 in ??? () at /usr/lib/libkwin.so.6 #5 0x000071ee8d0ce2ee in ??? () at /usr/lib/libkwin.so.6 #6 0x000071ee8d0c65f3 in KWin::DrmOutput::present(std::shared_ptr<KWin::OutputFrame> const&) () at /usr/lib/libkwin.so.6 #7 0x000071ee8cd58b12 in KWin::WaylandCompositor::composite(KWin::RenderLoop*) () at /usr/lib/libkwin.so.6 #8 0x000071ee8a1c4d27 in ??? () at /usr/lib/libQt6Core.so.6 #9 0x000071ee8cd73636 in ??? () at /usr/lib/libkwin.so.6 #10 0x000071ee8a1c4d27 in ??? () at /usr/lib/libQt6Core.so.6 #11 0x000071ee8a1cf3e4 in QTimer::timerEvent(QTimerEvent*) () at /usr/lib/libQt6Core.so.6 #12 0x000071ee8a1b34b9 in QObject::event(QEvent*) () at /usr/lib/libQt6Core.so.6 #13 0x000071ee8b6fefe7 in QApplicationPrivate::notify_helper(QObject*, QEvent*) () at /usr/lib/libQt6Widgets.so.6 #14 0x000071ee8a160350 in QCoreApplication::notifyInternal2(QObject*, QEvent*) () at /usr/lib/libQt6Core.so.6 #15 0x000071ee8a2f4c08 in QTimerInfoList::activateTimers() () at /usr/lib/libQt6Core.so.6 #16 0x000071ee8a2fe323 in QEventDispatcherUNIX::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () at /usr/lib/libQt6Core.so.6 #17 0x000071ee8af30d32 in QUnixEventDispatcherQPA::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () at /usr/lib/libQt6Gui.so.6 #18 0x000071ee8a16ada5 in QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) () at /usr/lib/libQt6Core.so.6 #19 0x000071ee8a1631bf in QCoreApplication::exec() () at /usr/lib/libQt6Core.so.6 #20 0x00005a4b171ae2c9 in ??? () #21 0x000071ee89a3690e in ??? () at /usr/lib/libc.so.6 #22 0x000071ee89a369ca in __libc_start_main () at /usr/lib/libc.so.6 #23 0x00005a4b171b6a55 in ??? () --- Operating System: CachyOS Linux KDE Plasma Version: 6.3.2 KDE Frameworks Version: 6.11.0 Qt Version: 6.8.2 Kernel Version: 6.13.4-2-cachyos (64-bit) Graphics Platform: Wayland Processors: 12 × 11th Gen Intel® Core™ i5-11400 @ 2.60GHz Memory: 31,2 GiB of RAM Graphics Processor: Intel® Arc Mesa: 25.0.0 PS: Aforementioned Intel GPU user
Created attachment 178959 [details] Trace of crash connecting fourth monitor with kwin debug symbols
Searchable backtrace with kwin symbols Thread 1 (Thread 0x7e2916b9ea40 (LWP 11636)): [KCrash Handler] #4 KWin::DrmPipeline::prepareAtomicPresentation (this=this@entry=0x5900536e4e60, commit=commit@entry=0x7e28e8006130, frame=std::shared_ptr<KWin::OutputFrame> (use count 1, weak count 0) = {...}) at /usr/src/debug/kwin/kwin-6.3.2.1/src/backends/drm/drm_property.cpp:153 #5 0x00007e29208ce2ee in KWin::DrmPipeline::present (this=0x5900536e4e60, frame=std::shared_ptr<KWin::OutputFrame> (use count 1, weak count 0) = {...}) at /usr/src/debug/kwin/kwin-6.3.2.1/src/backends/drm/drm_pipeline.cpp:61 #6 KWin::DrmPipeline::present (this=0x5900536e4e60, frame=std::shared_ptr<KWin::OutputFrame> (use count 1, weak count 0) = {...}) at /usr/src/debug/kwin/kwin-6.3.2.1/src/backends/drm/drm_pipeline.cpp:51 #7 0x00007e29208c65f3 in KWin::DrmOutput::present (this=0x590056f6a440, frame=std::shared_ptr<KWin::OutputFrame> (use count 1, weak count 0) = {...}) at /usr/src/debug/kwin/kwin-6.3.2.1/src/backends/drm/drm_output.cpp:326 #8 0x00007e2920558b12 in KWin::WaylandCompositor::composite (this=0x5900539f6050, renderLoop=<optimized out>) at /usr/src/debug/kwin/kwin-6.3.2.1/src/compositor_wayland.cpp:465 #9 0x00007e291d9c4d27 in ??? () at /usr/lib/libQt6Core.so.6 #10 0x00007e2920573636 in KWin::RenderLoop::frameRequested (this=<optimized out>, _t1=<optimized out>) at /usr/src/debug/kwin/build/src/kwin_autogen/include/moc_renderloop.cpp:211 #11 KWin::RenderLoopPrivate::dispatch (this=0x590056fbc760) at /usr/src/debug/kwin/kwin-6.3.2.1/src/core/renderloop.cpp:197 #12 operator() (__closure=<optimized out>) at /usr/src/debug/kwin/kwin-6.3.2.1/src/core/renderloop.cpp:37 #13 operator() (__closure=<optimized out>) at /usr/include/qt6/QtCore/qobjectdefs_impl.h:141 #14 QtPrivate::FunctorCallBase::call_internal<void, QtPrivate::FunctorCall<QtPrivate::IndexesList<>, QtPrivate::List<>, void, KWin::RenderLoopPrivate::RenderLoopPrivate(KWin::RenderLoop*, KWin::Output*)::<lambda()> >::call(KWin::RenderLoopPrivate::RenderLoopPrivate(KWin::RenderLoop*, KWin::Output*)::<lambda()>&, void**)::<lambda()> > (args=<optimized out>, fn=<optimized out>) at /usr/include/qt6/QtCore/qobjectdefs_impl.h:65 #15 QtPrivate::FunctorCall<QtPrivate::IndexesList<>, QtPrivate::List<>, void, KWin::RenderLoopPrivate::RenderLoopPrivate(KWin::RenderLoop*, KWin::Output*)::<lambda()> >::call (f=<optimized out>, arg=<optimized out>) at /usr/include/qt6/QtCore/qobjectdefs_impl.h:140 #16 QtPrivate::FunctorCallable<KWin::RenderLoopPrivate::RenderLoopPrivate(KWin::RenderLoop*, KWin::Output*)::<lambda()> >::call<QtPrivate::List<>, void> (f=<optimized out>, arg=<optimized out>) at /usr/include/qt6/QtCore/qobjectdefs_impl.h:362 #17 QtPrivate::QCallableObject<KWin::RenderLoopPrivate::RenderLoopPrivate(KWin::RenderLoop*, KWin::Output*)::<lambda()>, QtPrivate::List<>, void>::impl(int, QtPrivate::QSlotObjectBase *, QObject *, void **, bool *) (which=<optimized out>, this_=<optimized out>, r=<optimized out>, a=<optimized out>, ret=<optimized out>) at /usr/include/qt6/QtCore/qobjectdefs_impl.h:572 #18 0x00007e291d9c4d27 in ??? () at /usr/lib/libQt6Core.so.6 #19 0x00007e291d9cf3e4 in QTimer::timerEvent(QTimerEvent*) () at /usr/lib/libQt6Core.so.6 #20 0x00007e291d9b34b9 in QObject::event(QEvent*) () at /usr/lib/libQt6Core.so.6 #21 0x00007e291eefefe7 in QApplicationPrivate::notify_helper(QObject*, QEvent*) () at /usr/lib/libQt6Widgets.so.6 #22 0x00007e291d960350 in QCoreApplication::notifyInternal2(QObject*, QEvent*) () at /usr/lib/libQt6Core.so.6 #23 0x00007e291daf4c08 in QTimerInfoList::activateTimers() () at /usr/lib/libQt6Core.so.6 #24 0x00007e291dafe323 in QEventDispatcherUNIX::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () at /usr/lib/libQt6Core.so.6 #25 0x00007e291e730d32 in QUnixEventDispatcherQPA::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () at /usr/lib/libQt6Gui.so.6 #26 0x00007e291d96ada5 in QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) () at /usr/lib/libQt6Core.so.6 #27 0x00007e291d9631bf in QCoreApplication::exec() () at /usr/lib/libQt6Core.so.6 #28 0x000059003597b2c9 in main (argc=<optimized out>, argv=<optimized out>) at /usr/src/debug/kwin/kwin-6.3.2.1/src/main_wayland.cpp:624
Hi, same issue here. I have a laptop with two external monitors connected. Login from sddm to wayland session did not work and stuck at login-screen after a few seconds. My workarounds: - Disconnect one external monitor, login to wayland session, after desktop appears, plugin the disconnected monitor, everything works just fine - Alternative: start a x11 session instead of wayland. Works also just fine.
Setting to confirmed since multiple people can reproduce this
This should be fixed with KWin 6.3.2.1. Can affected people try that?
(In reply to Nate Graham from comment #7) > This should be fixed with KWin 6.3.2.1. Can affected people try that? I will try when the fix is available. I'm on kde neon stable, all up to date.
(In reply to Nate Graham from comment #7) > This should be fixed with KWin 6.3.2.1. Can affected people try that? Unfortunatelly was crashing for me even with 6.3.2.1 (installed from Fedora 41 test updates). Reverting just kwin packages back to 6.3.1 and everything is working...
> This should be fixed with KWin 6.3.2.1. Can't confirm. Checked by logs that I was on 6.3.2.1 when reporting yesterday. Tested for good measure right now on 6.3.2.1 and issue persists.
Created attachment 178984 [details] latest log from cachyos Hi, since I tried the change the resolution on my second screen via HDMI for 2560x1440 to 1920x1080, I got a bug with the two screens. The laptop screen and the second screen are black. I must to disconnect the HDMI cable to have the screen working on the laptop. I have start multiple type the computer today and have no problem when the second screen via HDMI was on is default resolution 2560x1440. The bug happened when I tried to change the resolution 2560x1440 to 1920x1080 on the second screen. Here the log files. Computer : Lenovo Legion 5, Ryzen 5 5600H, RTX 3060 laptop mobil OS : CachyOS Thanks
I just tested an another screen connected via displayport and I had no bug. So the culprit seems to be HDMI. So I connected the third screens via HDMI, the scrren stayed black until I change back the resolution 1920x1080 to 2560x1440. So now my second screen via HDMI worked like before. I will stay away to change the resolution on it until the bufg with HDMI will resolve. Here little video with second screens connected via displayport : https://youtu.be/G51IrIXEmm0
Created attachment 178987 [details] screenshot
In my case, it is also the displayport display. if i disconnect the display, login, plugin again, it worked. I checked the sddm log in my home-folder: Error: could not determine $DISPLAY. Error: Can not contact kdeinit5! org.kde.startup: "kdeinit5_shutdown" QList() exited with code 255 startplasma-wayland: Shutting down... startplasmacompositor: Shutting down... startplasmacompositor: Done.
I can confirm this bug still exists in 6.3.2.1. With 4 screens connected I got black screen. If I login with 3 screens, then plug in the 4th, kwin will also turn into black screen. The error I got from `~/.local/share/sddm/wayland-session.log` is "Applying output config failed"
I can confirm the same symptoms and error messages after upgrade from 6.3.1 to 6.3.2.1. Rollback to 6.3.1 resolves the issue. Operating System: Fedora Linux 41 Product Name: HP EliteBook x360 Graphics Platform: Wayland Kernel Version: 6.13.5-200.fc41.x86_64 (64-bit) (also tested using kernel 6.12.15-200 with same result) - 3 total displays: laptop screen, 1 HDMI external monitor, and 1 DisplayPort external monitor - External displays are connected via a single USB-C into a Dell dock - Booting with external displays connected results in black screen with a single underscore (_) in upper left corner. - Switching to TTY allows text login as expected (i.e. the OS is fully booted), but cannot manually launch Plasma Wayland - Disconnecting external displays allows a normal boot to graphical login and into Plasma - Connecting displays after login works as expected
I can confirm this too on Fedora 41. But additionally I am only able to connect one external display. I can boot normally with one external (DisplayPort via Dock) and the internal display. If I connect the second DisplayPort-display KWin crashes and all displays are black. Operating System: Fedora Linux 41 Product Name: Fujitsu E756 Graphics Platform: Wayland Kernel Version: 6.13.5-200.fc41.x86_64 (64-bit)
Also seeing this on old Hardware. AMD Radeon APU Notebook with 1080p LCD TV connected using HDMI. KWin 6.3.2.1 cashes at SDDM login with just that TV connected (also crashes when connected during login screen). Also can't activate the HDMI device when desktop is loaded. KScreen shows: "Couldn't apply display configuration: The driver rejected the output configuration" Everything working, when using KWin 6.3.1. SDDM login shows up on booth displays. HDMI device can be activated/configured and deactivated from desktop. So some heavy regression. Operating System: Fedora Linux 40 Device: HP 635, AMD E-450 with Wrestler Radeon HD 6320 Graphics Platform: Wayland Kernel Version: 6.13.5-100.fc40.x86_64
I'd guess, the driver rejecting is triggered here. https://github.com/KDE/kwin/commit/9dd70310eb016625313a0e81b6b49e98e64fff6a There might be external devices, that behave somewhat non-regular, but anyway, that results in an output configuration that can actually be enabled. So maybe that's not the perfect place to fix a crash with devices that won't ever activate.
OS: archlinux Plasma version: 6.3.2 I have 3 monitors. I am able to see the login screen, after enter the password, everything goes black. If I remove any of the monitors, everything works, If I reattach it, it goes black again. I have two monitors of 1366x768 and one 1024x768. on ksystemlog I can noticed this "3/2/25 5:43 PM kwin_wayland kwin_core: Applying output config failed! "
(In reply to samoht0 from comment #19) > I'd guess, the driver rejecting is triggered here. > > https://github.com/KDE/kwin/commit/9dd70310eb016625313a0e81b6b49e98e64fff6a I can confirm, reverting the commit fixed the issue for me. As long as the logic getting the output configurations that can't be powered at all is not *rock solid*, stating an hard error seems to cause severe regressions. Note: The old Radeon HD 6320 supports two output devices, so activating one external device is working at the hardware limit.
(In reply to samoht0 from comment #19) > I'd guess, the driver rejecting is triggered here. > > https://github.com/KDE/kwin/commit/9dd70310eb016625313a0e81b6b49e98e64fff6a > > There might be external devices, that behave somewhat non-regular, but > anyway, that results in an output configuration that can actually > be enabled. So maybe that's not the perfect place to fix a crash with > devices that won't ever activate. I also confirm reverting the commit fixed the issue. It seems the error stops all other pipelines?
(In reply to samoht0 from comment #19) > I'd guess, the driver rejecting is triggered here. > > https://github.com/KDE/kwin/commit/9dd70310eb016625313a0e81b6b49e98e64fff6a I can confirm reverting this single commit fixes the issue on my end as well with a 4 screen setup on of the screen being HDMI on a desktop GPU Nvidia RTX 2070. Linux/KDE Plasma: Arch Linux | KDE Plasma 6.3.2 KDE Frameworks Version: 6.11.0 Qt Version: 6.8.2 KWin Version : 6.3.2.1-1 Kernel version: 6.13.5-arch1-1
I confirm the issue. Software: - KDE Plasma 6.3.2 - Qt: 6.8.2-3 - KWin : 6.3.2.1-1 - Kernel: 6.13.5-arch1-1 Hardware: - ThinkPad P14s g5 AMD Ryzen 7 PRO 8840HS w/ Radeon 780M Graphics - built-in 2880×1800 LCD display, - external 2560×1440 display via DisplayPort connected via TB4 UBS-C dock - external 1920×1080 display via DisplayPort connected via TB4 UBS-C dock - external 1920×1080 display via HDMI connected via TB4 UBS-C dock Connecting USB-C TB4 dock with 3 external displays causes blank screen on all displays. Disconnecting it brings back built-in display with crash report. Connectung USB-C dock with 2 external displays works fine. Connecting the third exernal display causes crash. Reverting back to KDE 6.3.1 fixes the issue.
Confirmed crash with Laptop and two external displays: Connecting Thunderbolt 3 dock (Dell WD19TBS) with two 2560x1440 monitors connected via Display Port results in kwin_wayland crash. If I close the laptop to disable the laptop display, the system recovers and the desktop is shown on the two external monitors. If I then re-open the laptop lid then try to enable the laptop display in my display settings, the message "Couldn't apply display configuration: The driver rejected the output configuration" is shown in the display settings UI. Operating System: Arch Linux KDE Plasma Version: 6.3.2 kwin Version: 6.3.2.1 KDE Frameworks Version: 6.11.0 Qt Version: 6.8.2 Kernel Version: 6.13.5-arch1-1 (64-bit) Graphics Platform: Wayland Processors: 12 × Intel® Core™ i7-8850H CPU @ 2.60GHz Memory: 31.0 GiB of RAM Graphics Processor 1: Mesa Intel® UHD Graphics 630 Graphics Processor 2: llvmpipe Manufacturer: Dell Inc. Product Name: Latitude 5591
(In reply to samoht0 from comment #19) > I'd guess, the driver rejecting is triggered here. > > https://github.com/KDE/kwin/commit/9dd70310eb016625313a0e81b6b49e98e64fff6a > > There might be external devices, that behave somewhat non-regular, but > anyway, that results in an output configuration that can actually > be enabled. So maybe that's not the perfect place to fix a crash with > devices that won't ever activate. I can confirm that I'm experiencing the same issue with 6.3.2.1 on ArchLinux with 4 screens with i915 driver and reverting this commit fixes the issue.
*** Bug 500139 has been marked as a duplicate of this bug. ***
A possibly relevant merge request was started @ https://invent.kde.org/plasma/kwin/-/merge_requests/7254
*** Bug 500094 has been marked as a duplicate of this bug. ***
I've just tried the code on the merge request (https://invent.kde.org/plasma/kwin/-/merge_requests/7254), and I confirm that it fixes the issue on my side. Thanks Xaver!
Git commit e652bdd3118fa55563caf0125deb0999f5ebd503 by Xaver Hugl. Committed on 03/03/2025 at 20:33. Pushed by zamundaaa into branch 'Plasma/6.3'. backends/drm: fix testing for more connectors than CRTCs The connectors may not actually be enabled, so the checks could sometimes result in wrongly returning Error::NotEnoughCrtcs. Related: bug 500797 (cherry picked from commit aa11d89a90bc4eb77bfe191a2cadede4d957d2a9) M +7 -7 src/backends/drm/drm_gpu.cpp https://invent.kde.org/plasma/kwin/-/commit/e652bdd3118fa55563caf0125deb0999f5ebd503
I tried out the commits from the 6.3 branch dated feb 28 and march 3 to drm_gpu.cpp on the fedora source rpm and rebuilt+installed the resulting packages. Now it just works normal again, like before. Seems it fixed the problem.
*** Bug 501022 has been marked as a duplicate of this bug. ***
Confirming change from https://invent.kde.org/plasma/kwin/-/merge_requests/7254 works also for me on Fedora 41, everything is ok now.
I had the same problem but with only one monitor. I have tried downgrading kwin, plasma-desktop and sddm but the problem remains. What else can I do to solve the problem?
(In reply to vitoproscia.1704 from comment #35) > I had the same problem but with only one monitor. I have tried downgrading > kwin, plasma-desktop and sddm but the problem remains. > What else can I do to solve the problem? If you are on kwin version 6.3.1 or older, then your issue is probably not related to this one.
(In reply to e from comment #36) > (In reply to vitoproscia.1704 from comment #35) > > I had the same problem but with only one monitor. I have tried downgrading > > kwin, plasma-desktop and sddm but the problem remains. > > What else can I do to solve the problem? > > If you are on kwin version 6.3.1 or older, then your issue is probably not > related to this one. I changed from kwin 6.3.2 to 6.3.1 with “downgrade kwin” but the situation does not change. Particularly at boot after grub comes out black screen before login, only sometimes after several attempts of upgrades, kernel recomplications it booted allowing me to login, but at reboot again black screen, it's after the last arch aggirornament that this problem coarse. Any idea how to solve it ?
Thanks for the fix, are we going to see a hotfix release for this ? It's quite the regression
*** Bug 500884 has been marked as a duplicate of this bug. ***
I created patched version for openSUSE Tumbleweed and can confirm that it works. Thanks https://build.opensuse.org/package/show/home:hlavki/kwin6
(In reply to Seïfane Idouchach from comment #38) > Thanks for the fix, are we going to see a hotfix release for this ? It's > quite the regression The next bugfix release is in less than a week, but I sent an email recommending distros to backport the commit immediately as well.
(In reply to Zamundaaa from comment #41) > (In reply to Seïfane Idouchach from comment #38) > > Thanks for the fix, are we going to see a hotfix release for this ? It's > > quite the regression > The next bugfix release is in less than a week, but I sent an email > recommending distros to backport the commit immediately as well. Thanks a lot for your work. I confirm that on archlinux the patch was applied on kwin 6.3.2.1-4 !
Hi, there, I've just updated kwin to version 6.3.2.1-3. I always have the problem on my laptop when I change the resolution of an external screen connected to HDMI. But this time I don't have a display only on the external screen, the display configuration window greyed out in a loop every 5 seconds or so. Without removing the HDMI cable, I waited until the window was accessible to reset the external screen to its original resolution, so that it could be operational again. I'll wait for kwin 6.3.2.1-4 to see if it's solved.
(In reply to monbureaulinux from comment #43) > Hi, there, > > I've just updated kwin to version 6.3.2.1-3. > I always have the problem on my laptop when I change the resolution of an > external screen connected to HDMI. > But this time I don't have a display only on the external screen, the > display configuration window greyed out in a loop every 5 seconds or so. > Without removing the HDMI cable, I waited until the window was accessible to > reset the external screen to its original resolution, so that it could be > operational again. > I'll wait for kwin 6.3.2.1-4 to see if it's solved. That sounds unrelated to this issue, please make a separate bug report if it still happens then.
*** Bug 501088 has been marked as a duplicate of this bug. ***
today I updated and tested the kwin 6.3.2.1-4 and all my 3 monitors back to work as spected. thank you very much everyone.
Same! GG everyone, thanks for the quick turnover!
Fedora (41, probably other current versions too) now has new updates to kwin packages, version 6.3.2.1-2, which seem to include this fix and work normally, at least with my setup.
*** Bug 501123 has been marked as a duplicate of this bug. ***
(In reply to Henrik from comment #48) > Fedora (41, probably other current versions too) now has new updates to kwin > packages, version 6.3.2.1-2, which seem to include this fix and work > normally, at least with my setup. It seems that there was a change in the fedora kwin 6.3.2.1-3 package, that causes kwin to crash again for me. Downgrading to my locally built package based on 6.3.2.1-1 + patches mentioned in my previous comment above works.
I also have the issue with "kwin-6.3.2.1-3.fc41.x86_64". I have all latest updates installed to this point in time. Operating System: Fedora Linux 41 KDE Plasma Version: 6.3.2 KDE Frameworks Version: 6.11.0 Qt Version: 6.8.2 Kernel Version: 6.12.15-200.fc41.x86_64 (64-bit) Graphics Platform: Wayland Processors: 32 × AMD Ryzen 9 5950X 16-Core Processor Memory: 62.7 GiB of RAM Graphics Processor: AMD Radeon RX 6700 XT Manufacturer: Micro-Star International Co., Ltd. Product Name: MS-7D52 System Version: 1.0
This is not an HDMI only issue, as I can confirm the same issue with VGA. My old Dell laptop drives an older 1080p TV with VGA as an external monitor and the laptop monitor. It's old school Intel Business Graphics 4 GPU soldered to the mother board. Running Fedora 41 KDE, with the latest update, both screens are black. Unplug the VGA cable, and the laptop screen comes back. If you plug in the VGA cable after your logged in, the laptop screen goes black, and the external screen comes up. I'm using unmodified Fedora 41 KDE, so it's on Wayland. System is 8GB of RAM, 1 TB SSD, Intel 990 Core Two Duo CPU, Intel Business Graphics 4 GPU - maxed out my Dell 1540 and works great for what I use it for. Wish I could provide more data, but I'm limited.
*** Bug 501405 has been marked as a duplicate of this bug. ***
If you switch to X11 (I'm using Fedora 41 and had to install it), when it boots up with a second monitor, it will give you a blank screen with a cursor. If you type your password and hit enter, which will not echo, it will load the desktop and work as normal. But it only does this on X11 not Wayland. Wayland just hangs.
These sound like different issues. Please open new bug reports for them. Thanks!
(In reply to Henrik from comment #50) > (In reply to Henrik from comment #48) > > Fedora (41, probably other current versions too) now has new updates to kwin > > packages, version 6.3.2.1-2, which seem to include this fix and work > > normally, at least with my setup. > > It seems that there was a change in the fedora kwin 6.3.2.1-3 package, that > causes kwin to crash again for me. Downgrading to my locally built package > based on 6.3.2.1-1 + patches mentioned in my previous comment above works. Now Fedora has update packages with kwin 6.3.3-1, and these work for me. I presume they contain the fixes to this bug, though I haven'd checked the source code.
The latest Fedora update(s) fixed the issue on my end as well. By the way, Plasma on Wayland is awesome! I haven't used the X11 session for a long time, never looking back :)