Description of problem: While logging in. Version-Release number of selected component: sddm-0.12.0-5.fc22 Additional info: reporter: libreport-2.6.3 backtrace_rating: 4 cmdline: /usr/bin/sddm-greeter --socket /tmp/sddm-:0-VvygWe --theme /usr/share/sddm/themes/01-breeze-fedora crash_function: qt_message_fatal executable: /usr/bin/sddm-greeter global_pid: 1394 kernel: 4.2.6-200.fc22.x86_64 runlevel: N 5 type: CCpp uid: 990 Truncated backtrace: Thread no. 1 (10 frames) #2 qt_message_fatal at global/qlogging.cpp:1578 #3 QMessageLogger::fatal at global/qlogging.cpp:781 #4 QSGRenderLoop::handleContextCreationFailure at /usr/src/debug/qtdeclarative-opensource-src-5.5.1/src/quick/scenegraph/qsgrenderloop.cpp:244 #5 QSGGuiThreadRenderLoop::renderWindow at /usr/src/debug/qtdeclarative-opensource-src-5.5.1/src/quick/scenegraph/qsgrenderloop.cpp:333 #6 QSGGuiThreadRenderLoop::exposureChanged at /usr/src/debug/qtdeclarative-opensource-src-5.5.1/src/quick/scenegraph/qsgrenderloop.cpp:422 #7 QWindow::event at kernel/qwindow.cpp:2053 #8 QQuickWindow::event at /usr/src/debug/qtdeclarative-opensource-src-5.5.1/src/quick/items/qquickwindow.cpp:1413 #9 QCoreApplication::notify at kernel/qcoreapplication.cpp:1038 #10 QCoreApplication::notifyInternal at kernel/qcoreapplication.cpp:965 #11 sendSpontaneousEvent at ../../src/corelib/kernel/qcoreapplication.h:227
Created attachment 1098082 [details] File: backtrace
Created attachment 1098083 [details] File: cgroup
Created attachment 1098084 [details] File: core_backtrace
Created attachment 1098085 [details] File: dso_list
Created attachment 1098086 [details] File: environ
Created attachment 1098087 [details] File: limits
Created attachment 1098088 [details] File: maps
Created attachment 1098089 [details] File: mountinfo
Created attachment 1098090 [details] File: open_fds
Created attachment 1098091 [details] File: proc_pid_status
Created attachment 1098092 [details] File: var_log_messages
It appears opengl isn't working with your nvidia driver, and unfortunately, not much we can do about that.
I think you mean: OpenGL isn't handling a(n error) condition thrown by the NVIDIA driver--the only way to get CUDA, y'know--and because of our not-wanting-to-work-with-proprietary-drivers fetish, we'll say that "opengl isn't working with your nvidia driver." Moreover, the correct answer for software handling an error is, at the very least, to exit gracefully.