Happened to me today on Xorg (so no Wayland)
(gdb) bt -full
#0 0x00007f962af48818 in abort () at /lib64/libc.so.6
#1 0x00007f96279ba6d0 in amdgpu_ctx_set_sw_reset_status(radeon_winsys_ctx*,
pipe_reset_status, char const*, ...) ()
at /usr/lib64/dri-freeworld/libgallium-26.1.0.so
#2 0x00007f96279bf8a8 in void amdgpu_cs_submit_ib<(queue_type)0>(void*, void*,
int) () at /usr/lib64/dri-freeworld/libgallium-26.1.0.so
#3 0x00007f96273c6bfe in util_queue_thread_func () at
/usr/lib64/dri-freeworld/libgallium-26.1.0.so
#4 0x00007f962741188c in impl_thrd_routine () at
/usr/lib64/dri-freeworld/libgallium-26.1.0.so
#5 0x00007f962afba939 in start_thread () at /lib64/libc.so.6
#6 0x00007f962b03f09c in __clone3 () at /lib64/libc.so.6
Theoretically this can be related from kernel log:
[532758.533021] amdgpu 0000:c4:00.0: [drm] DMUB HPD RX IRQ callback:
link_index=4
[532758.547428] amdgpu 0000:c4:00.0: [drm] REG_WAIT timeout 1us * 100 tries -
dcn31_program_compbuf_size line:141
[532758.548554] amdgpu 0000:c4:00.0: [drm] DMUB HPD RX IRQ callback:
link_index=4
[532758.909663] amdgpu 0000:c4:00.0: [drm] DMUB HPD RX IRQ callback:
link_index=4
Like I used to have probably stable Xorg - until I've recently added
mesa-va-drivers-freeworld-26.1.0-1.fc45.x86_64 to get accelerated video
playback.
Anyway if this will be frequent I'll dump va drivers and keep CPU
decoding.
Worth to at the moment of crash machine was mostly idle doing some
'terminal' work.
--
You received this bug notification because you are a member of Ubuntu
Desktop Bugs, which is subscribed to mutter in Ubuntu.
https://bugs.launchpad.net/bugs/2101816
Title:
[amdgpu] Crash with SIGABRT in amdgpu_ctx_set_sw_reset_status() from
amdgpu_cs_submit_ib() from util_queue_thread_func() from
impl_thrd_routine() from start_thread()
To manage notifications about this bug go to:
https://bugs.launchpad.net/mesa/+bug/2101816/+subscriptions
--
desktop-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/desktop-bugs
