https://bugs.kde.org/show_bug.cgi?id=525970
Bug ID: 525970
Summary: plasmashell crashes after AMDGPU GPU reset during
Diablo IV on Wayland
Classification: Plasma
Product: plasmashell
Version First 6.7.5
Reported In:
Platform: Fedora RPMs
OS: Linux
Status: REPORTED
Keywords: drkonqi
Severity: crash
Priority: NOR
Component: generic-crash
Assignee: [email protected]
Reporter: [email protected]
Target Milestone: 1.0
Application: plasmashell (6.7.5)
ApplicationNotResponding [ANR]: false
Qt Version: 6.11.2
Frameworks Version: 6.30.0
Operating System: Linux 7.2.5-200.fc44.x86_64 x86_64
Windowing System: Wayland
Distribution: "Fedora Linux 44 (KDE Plasma Desktop Edition)"
DrKonqi: 6.7.5 [CoredumpBackend]
-- Information about the crash:
plasmashell crashes after an AMDGPU GPU reset while playing Diablo IV under KDE
Plasma Wayland.
System:
- Fedora Linux 44
- KDE Plasma 6.7.5
- KDE Frameworks 6.30.0
- Qt 6.11.2
- Wayland
- Kernel 7.2.4-200.fc44.x86_64 (kernel at the time of the crash)
- Mesa 26.2.2-6.fc44
- Xwayland 24.1.13-1.fc44
- AMD Radeon RX 7900 XT
The crash occurred while Diablo IV was running. The kernel log shows a GPU hang
on the gfx ring:
amdgpu: ring gfx_0.0.0 timeout
Process Diablo IV. exe pid 18130 thread vkd3d_queue pid 18258
This was followed by:
amdgpu: gfx_v11_0_bad_op_irq: Illegal opcode in command stream
amdgpu: MES failed to respond to RESET
amdgpu: GPU reset begin!
amdgpu: GPU reset succeeded
amdgpu: VRAM is lost due to GPU reset!
After the GPU reset, KDE and other graphical applications started losing their
graphics contexts. The log contains:
amdgpu: The CS has cancelled because the context is lost. This context is
innocent.
KWin also reported:
A graphics reset not attributable to the current GL context occurred.
Xwayland then crashed with SIGABRT inside Mesa/AMD GPU code while handling a
Present/EGL fence. plasmashell subsequently also crashed.
The plasmashell crash therefore appears to happen as a consequence of the GPU
reset/context loss rather than being the initial cause of the GPU failure.
The full kernel log shows the GPU reset and the subsequent Xwayland/plasmashell
graphics context failures.
The system has since been booted with kernel 7.2.5-200.fc44.x86_64, so I am
currently testing whether the problem also occurs with the newer kernel.
The reporter is unsure if this crash is reproducible.
-- Backtrace (Reduced):
#5 0x00007fed9be86777 in unlink_chunk (p=p@entry=0x7fec905e0de0,
av=0x7fec90000030) at malloc.c:1600
#6 0x00007fed9be89173 in _int_malloc (av=av@entry=0x7fec90000030,
bytes=bytes@entry=4) at malloc.c:4246
#7 0x00007fed9be893ee in __libc_malloc2 (bytes=4) at malloc.c:3261
#8 0x00007fed9be8a423 in __GI___libc_malloc (bytes=bytes@entry=4) at
malloc.c:3306
#9 0x00007fed9be8aef8 in __GI___libc_realloc (oldmem=<optimized out>,
bytes=bytes@entry=4) at malloc.c:3388
#10 0x00007fed74da0034 in util_idalloc_resize (buf=buf@entry=0x7fed4c7cf390,
new_num_elements=new_num_elements@entry=1) at ../src/util/u_idalloc.c:50
#11 0x00007fed74da05a0 in util_idalloc_resize (buf=0x7fed4c7cf390,
new_num_elements=1) at ../src/util/u_idalloc.c:245
#12 util_idalloc_init (buf=0x7fed4c7cf390, initial_num_ids=1) at
../src/util/u_idalloc.c:62
#13 util_idalloc_sparse_init (buf=buf@entry=0x7fed4c7cd1d0) at
../src/util/u_idalloc.c:246
#14 0x00007fed748df04a in _mesa_InitHashTable
(table=table@entry=0x7fed4c7cd1b8) at ../src/mesa/main/hash.c:52
#15 0x00007fed74b9409f in _mesa_init_pipeline (ctx=ctx@entry=0x7fed4c729010) at
../src/mesa/main/pipelineobj.c:96
#16 0x00007fed74a987cd in init_attrib_groups (ctx=0x7fed4c729010) at
../src/mesa/main/context.c:676
#17 _mesa_initialize_context (ctx=ctx@entry=0x7fed4c729010,
api=api@entry=API_OPENGL_COMPAT, no_error=no_error@entry=false,
visual=visual@entry=0x7fecb1de7cc0, share_list=share_list@entry=0x7fed51e74010,
driverFunctions=driverFunctions@entry=0x7fecb1de7c10, options=<optimized out>)
at ../src/mesa/main/context.c:1029
#18 0x00007fed74927cee in st_create_context (api=API_OPENGL_COMPAT,
pipe=pipe@entry=0x7fec900c6930, visual=0x7fecb1de7cc0,
share=share@entry=0x7fed480ee010, options=options@entry=0x7fecb1de7dd8,
no_error=no_error@entry=false, has_egl_image_validate=true) at
../src/mesa/state_tracker/st_context.c:787
#19 0x00007fed74938c11 in st_api_create_context
(fscreen=fscreen@entry=0x561380befde0, attribs=attribs@entry=0x7fecb1de7db0,
error=error@entry=0x7fecb1de7dac, shared_ctx=shared_ctx@entry=0x7fed480ee010)
at ../src/mesa/state_tracker/st_manager.c:1020
#20 0x00007fed748a52d7 in dri_create_context
(screen=screen@entry=0x561380befde0, api=<optimized out>,
visual=visual@entry=0x561380d98930, ctx_config=ctx_config@entry=0x7fecb1de7eb0,
error=error@entry=0x7fecb1de7f6c,
sharedContextPrivate=sharedContextPrivate@entry=0x7fed48004cf0,
loaderPrivate=0x7fec904a1160, thread_safe=true) at
../src/gallium/frontends/dri/dri_context.c:180
Reported using DrKonqi
--
You are receiving this mail because:
You are watching all bug changes.
