Crash in [@ NvRmShimImportMemContextFromFd]
Categories
(Core :: Widget: Gtk, defect, P5)
Tracking
()
People
(Reporter: emilio, Unassigned)
References
(Blocks 2 open bugs)
Details
(Keywords: crash, Whiteboard: [tbird crash])
Crash Data
Crash report: https://crash-stats.mozilla.org/report/index/08281a3b-527b-4806-b69f-a60ce0230203
Reason: SIGSEGV / SI_KERNEL
Top 10 frames of crashing thread:
0 libnvidia-allocator.so.1 NvRmShimImportMemContextFromFd
1 libgbm.so.1 backend_create_device /usr/src/debug/mesa/mesa-22.3.4/src/gbm/main/backend.c:101
1 libgbm.so.1 load_backend /usr/src/debug/mesa/mesa-22.3.4/src/gbm/main/backend.c:133
2 libgbm.so.1 override_backend /usr/src/debug/mesa/mesa-22.3.4/src/gbm/main/backend.c:186
2 libgbm.so.1 _gbm_create_device /usr/src/debug/mesa/mesa-22.3.4/src/gbm/main/backend.c:219
3 libgbm.so.1 gbm_create_device /usr/src/debug/mesa/mesa-22.3.4/src/gbm/main/gbm.c:138
4 libxul.so mozilla::widget::nsGbmLib::CreateDevice widget/gtk/DMABufLibWrapper.h:64
5 libxul.so mozilla::widget::nsDMABufDevice::Configure widget/gtk/DMABufLibWrapper.cpp:247
6 libxul.so gfxPlatformGtk::InitDmabufConfig gfx/thebes/gfxPlatformGtk.cpp:215
7 libxul.so gfxPlatformGtk::gfxPlatformGtk gfx/thebes/gfxPlatformGtk.cpp:111
I hit this all the time when trying to create a gbm device with nvidia
and nvidia-utils
installed on my arch laptop.
Seems like glxtest fails with:
glxtest: VA-API test failed: failed to initialise VAAPI connection.
But we still try to use dmabuf and crash somehow?
Reporter | ||
Comment 1•2 years ago
|
||
I think we try to use the intel GPU but try to load the nvidia GBM device, which causes things to go boom.
Updated•2 years ago
|
Comment 2•2 years ago
|
||
But we still try to use dmabuf and crash somehow?
Well, WebGL?
Comment 3•2 years ago
|
||
We can't create GBM device on NVIDIA, we use EGL_MESA_image_dma_buf_export to get dmabuf surface from EGL texture.
And yes, dmabuf is used for WebGL.
Updated•2 years ago
|
Comment 4•2 years ago
|
||
I think we should add this test to glxtest to avoid Firefox itself crash.
Comment 5•2 years ago
|
||
Comment 6•2 years ago
|
||
This crash comes from broken systems, looks like there are some leftovers from old NVIDIA drivers while system is running on Intel.
Mesa (Intel) drivers (/usr/src/debug/mesa/mesa-22.3.5/src/gbm/main/backend.c:101) calls NvRmShimImportMemContextFromFd() from libnvidia-allocator.so.1.
No idea why libnvidia-allocator.so.1 is installed on the system.
Updated•2 years ago
|
Updated•2 years ago
|
Description
•