| Summary: | kwin_wayland with nvidia 545 driver drops user back to login after a few seconds | ||
|---|---|---|---|
| Product: | [Plasma] kwin | Reporter: | Barry Scott <barry> |
| Component: | wayland-generic | Assignee: | KWin default assignee <kwin-bugs-null> |
| Status: | RESOLVED WORKSFORME | ||
| Severity: | grave | CC: | ekurzinger, fanzhuyifan, nate, ngompa, warp-spam_kde, xaver.hugl |
| Priority: | NOR | ||
| Version First Reported In: | 5.27.9 | ||
| Target Milestone: | --- | ||
| Platform: | Other | ||
| OS: | Other | ||
| See Also: | https://bugzilla.redhat.com/show_bug.cgi?id=2252447 | ||
| Latest Commit: | Version Fixed/Implemented In: | ||
| Sentry Crash Report: | |||
|
Description
Barry Scott
2023-12-08 09:28:00 UTC
I am seeing what appears to be the same failure. To provide a bit more information on what is failing, there appears to be two different things going wrong, though it is distinctly possible that the second problem is related to the first. This is all in kwin 5.57.9 as packaged in Debian sid. In src/backends/drm/drm_gbm_surface.cpp there is GbmSurface::swapBuffers, this calls eglSwapBuffers, and then calls gbm_surface_lock_front_buffer. The latter call to gbm_surface_lock_front_buffer is failing, without any indication as to why. When that fails, we eventually get to GbmSurface::~GbmSurface, which calls gbm_surface_destroy, which triggers a very unhappy error about something either being freed twice, or there being memory corruption in play. That is a fatal error inside libc6, resulting in kwin dying. What I can't figure out, so far at least, is _why_. I am also on a nVidia RTX 3060, running the 545.23.08 driver. I would love a minimum viable test application for EGL on GBM to make it easier to test success/failure. This isn't actionable as-is, though it sounds like a bug in the driver. Please attach a backtrace of the crash. Barry, can you try: sudo ln -s /usr/lib/x86_64-linux-gnu/nvidia/current/nvidia-drm_gbm.so /usr/lib/x86_64-linux-gnu/gbm/
As far as the stack trace, creating the above symlink fixed kwin not working on wayland, but the strack trace I was getting when it wasn't is below.
Of significant note, I added a bunch of debug log entries, so the offsets won't exactly match lines in an unmodified build.
Module libsystemd.so.0 from deb systemd-255-1.amd64
Module libzstd.so.1 from deb libzstd-1.5.5+dfsg2-2.amd64
Module libudev.so.1 from deb systemd-255-1.amd64
Stack trace of thread 3325:
#0 0x00007f0d23ca80fc __pthread_kill_implementation (libc.so.6 + 0x8a0fc)
#1 0x00007f0d23c5a472 __GI_raise (libc.so.6 + 0x3c472)
#2 0x00007f0d23c444b2 __GI_abort (libc.so.6 + 0x264b2)
#3 0x00007f0d23c451ed __libc_message (libc.so.6 + 0x271ed)
#4 0x00007f0d23cb1a75 malloc_printerr (libc.so.6 + 0x93a75)
#5 0x00007f0d23cb3ad0 _int_free (libc.so.6 + 0x95ad0)
#6 0x00007f0d23cb616f __GI___libc_free (libc.so.6 + 0x9816f)
#7 0x00007f0d269d27e9 _ZN4KWin10GbmSurfaceD2Ev (libkwin.so.5 + 0x3d27e9)
#8 0x00007f0d267c682e _ZNSt16_Sp_counted_baseILN9__gnu_cxx12_Lock_policyE2EE19_M_release_last_useEv (libkwin.so.5 + 0x1c682e)
#9 0x00007f0d269cf438 _ZNSt16_Sp_counted_baseILN9__gnu_cxx12_Lock_policyE2EE10_M_releaseEv (libkwin.so.5 + 0x3cf438)
#10 0x00007f0d269cf712 operator() (libkwin.so.5 + 0x3cf712)
#11 0x00007f0d269cfec4 _ZNK4KWin18EglGbmLayerSurface13createSurfaceERK5QSizeRK4QMapIj7QVectorImEE (libkwin.so.5 + 0x3cfec4)
#12 0x00007f0d269d0b9d _ZN4KWin18EglGbmLayerSurface12checkSurfaceERK5QSizeRK4QMapIj7QVectorImEE (libkwin.so.5 + 0x3d0b9d)
#13 0x00007f0d269d13fe _ZN4KWin18EglGbmLayerSurface16renderTestBufferERK5QSizeRK4QMapIj7QVectorImEE (libkwin.so.5 + 0x3d13fe)
#14 0x00007f0d269c78c6 _ZN4KWin11EglGbmLayer15checkTestBufferEv (libkwin.so.5 + 0x3c78c6)
#15 0x00007f0d269eab28 _ZN4KWin11DrmPipeline21commitPipelinesAtomicERK7QVectorIPS0_ENS0_10CommitModeERKS1_IPNS_9DrmObjectEE (libkwin.so.5 + 0x3eab28)
#16 0x00007f0d269da33f _ZN4KWin6DrmGpu13testPipelinesEv (libkwin.so.5 + 0x3da33f)
#17 0x00007f0d269db424 _ZN4KWin6DrmGpu19checkCrtcAssignmentE7QVectorIPNS_12DrmConnectorEERKS1_IPNS_7DrmCrtcEE (libkwin.so.5 + 0x3db424)
#18 0x00007f0d269db38f _ZN4KWin6DrmGpu19checkCrtcAssignmentE7QVectorIPNS_12DrmConnectorEERKS1_IPNS_7DrmCrtcEE (libkwin.so.5 + 0x3db38f)
#19 0x00007f0d269db38f _ZN4KWin6DrmGpu19checkCrtcAssignmentE7QVectorIPNS_12DrmConnectorEERKS1_IPNS_7DrmCrtcEE (libkwin.so.5 + 0x3db38f)
#20 0x00007f0d269db424 _ZN4KWin6DrmGpu19checkCrtcAssignmentE7QVectorIPNS_12DrmConnectorEERKS1_IPNS_7DrmCrtcEE (libkwin.so.5 + 0x3db424)
#21 0x00007f0d269dbdd0 _ZN4KWin6DrmGpu24testPendingConfigurationEv (libkwin.so.5 + 0x3dbdd0)
#22 0x00007f0d269dc9db _ZN4KWin6DrmGpu13updateOutputsEv (libkwin.so.5 + 0x3dc9db)
#23 0x00007f0d269b520c _ZN4KWin10DrmBackend13updateOutputsEv (libkwin.so.5 + 0x3b520c)
#24 0x00007f0d24d062b2 n/a (libQt5Core.so.5 + 0x3062b2)
#25 0x00007f0d267db7f4 _ZN4KWin10Compositor10setupStartEv (libkwin.so.5 + 0x1db7f4)
#26 0x00007f0d267dcdd4 _ZN4KWin17WaylandCompositor5startEv (libkwin.so.5 + 0x1dcdd4)
#27 0x00007f0d24cf9940 _ZN7QObject5eventEP6QEvent (libQt5Core.so.5 + 0x2f9940)
#28 0x00007f0d24362f32 _ZN19QApplicationPrivate13notify_helperEP7QObjectP6QEvent (libQt5Widgets.so.5 + 0x162f32)
#29 0x00007f0d24ccc748 _ZN16QCoreApplication15notifyInternal2EP7QObjectP6QEvent (libQt5Core.so.5 + 0x2cc748)
#30 0x00007f0d24ccfe51 _ZN23QCoreApplicationPrivate16sendPostedEventsEP7QObjectiP11QThreadData (libQt5Core.so.5 + 0x2cfe51)
#31 0x00007f0d24d25115 _ZN20QEventDispatcherUNIX13processEventsE6QFlagsIN10QEventLoop17ProcessEventsFlagEE (libQt5Core.so.5 + 0x325115)
#32 0x000055bc0100e8c1 _ZN23QUnixEventDispatcherQPA13processEventsE6QFlagsIN10QEventLoop17ProcessEventsFlagEE (kwin_wayland + 0x1508c1)
#33 0x00007f0d24ccb0fb _ZN10QEventLoop4execE6QFlagsINS_17ProcessEventsFlagEE (libQt5Core.so.5 + 0x2cb0fb)
#34 0x00007f0d24cd38a4 _ZN16QCoreApplication4execEv (libQt5Core.so.5 + 0x2d38a4)
#35 0x000055bc00f1636f main (kwin_wayland + 0x5836f)
#36 0x00007f0d23c456ca __libc_start_call_main (libc.so.6 + 0x276ca)
#37 0x00007f0d23c45785 __libc_start_main_impl (libc.so.6 + 0x27785)
#38 0x000055bc00f18461 _start (kwin_wayland + 0x5a461)
Stack trace of thread 3335:
#0 0x00007f0d23d19a1f __GI___poll (libc.so.6 + 0xfba1f)
#1 0x00007f0d22714277 n/a (libglib-2.0.so.0 + 0x5a277)
#2 0x00007f0d22714930 g_main_context_iteration (libglib-2.0.so.0 + 0x5a930)
#3 0x00007f0d24d27d4a _ZN20QEventDispatcherGlib13processEventsE6QFlagsIN10QEventLoop17ProcessEventsFlagEE (libQt5Core.so.5 + 0x327d4a)
#4 0x00007f0d24ccb0fb _ZN10QEventLoop4execE6QFlagsINS_17ProcessEventsFlagEE (libQt5Core.so.5 + 0x2cb0fb)
#5 0x00007f0d24ad9c52 _ZN7QThread4execEv (libQt5Core.so.5 + 0xd9c52)
#6 0x00007f0d26da87ab n/a (libQt5DBus.so.5 + 0x177ab)
#7 0x00007f0d24adaeb1 n/a (libQt5Core.so.5 + 0xdaeb1)
#8 0x00007f0d23ca63ec start_thread (libc.so.6 + 0x883ec)
#9 0x00007f0d23d26a5c __clone3 (libc.so.6 + 0x108a5c)
Stack trace of thread 3339:
#0 0x00007f0d23d19a1f __GI___poll (libc.so.6 + 0xfba1f)
#1 0x00007f0d22714277 n/a (libglib-2.0.so.0 + 0x5a277)
#2 0x00007f0d22714930 g_main_context_iteration (libglib-2.0.so.0 + 0x5a930)
#3 0x00007f0d24d27d4a _ZN20QEventDispatcherGlib13processEventsE6QFlagsIN10QEventLoop17ProcessEventsFlagEE (libQt5Core.so.5 + 0x327d4a)
#4 0x00007f0d24ccb0fb _ZN10QEventLoop4execE6QFlagsINS_17ProcessEventsFlagEE (libQt5Core.so.5 + 0x2cb0fb)
#5 0x00007f0d24ad9c52 _ZN7QThread4execEv (libQt5Core.so.5 + 0xd9c52)
#6 0x00007f0d24adaeb1 n/a (libQt5Core.so.5 + 0xdaeb1)
#7 0x00007f0d23ca63ec start_thread (libc.so.6 + 0x883ec)
#8 0x00007f0d23d26a5c __clone3 (libc.so.6 + 0x108a5c)
ELF object binary architecture: AMD x86-64
```
Dear Bug Submitter, This bug has been in NEEDSINFO status with no change for at least 15 days. Please provide the requested information as soon as possible and set the bug status as REPORTED. Due to regular bug tracker maintenance, if the bug is still in NEEDSINFO status with no change in 30 days the bug will be closed as RESOLVED > WORKSFORME due to lack of needed information. For more information about our bug triaging procedures please read the wiki located here: https://community.kde.org/Guidelines_and_HOWTOs/Bug_triaging If you have already provided the requested information, please mark the bug as REPORTED so that the KDE team knows that the bug is ready to be confirmed. Thank you for helping us make KDE software even better for everyone! This bug has been in NEEDSINFO status with no change for at least 30 days. The bug is now closed as RESOLVED > WORKSFORME due to lack of needed information. For more information about our bug triaging procedures please read the wiki located here: https://community.kde.org/Guidelines_and_HOWTOs/Bug_triaging Thank you for helping us make KDE software even better for everyone! |