Ubuntu 20.04, GeForce GTX 1650 Mobile, 470.57.02, PRIME - GPU has fallen off the bus

Graphics / Linux Linux
1

Sometimes when playing with PRIME (for example, press Alt-Tab in The Wither over Steam Proton) the driver crashes, only a reboot helps.

Laptop Acer Nitro AN515-43 with GeForce GTX 1650

Game command: __NV_PRIME_RENDER_OFFLOAD=1 __GLX_VENDOR_LIBRARY_NAME="nvidia" PROTON_USE_WINED3D=1 PROTON_NO_ESYNC=1 %command%

$ nvidia-smi 
Unable to determine the device handle for GPU 0000:01:00.0: Unknown Error
$ LANG=c dmesg -T | grep NVRM
[Thu Aug  5 21:00:10 2021] NVRM: loading NVIDIA UNIX x86_64 Kernel Module  470.57.02  Tue Jul 13 16:14:05 UTC 2021
[Fri Aug  6 07:13:43 2021] NVRM: GPU at PCI:0000:01:00: GPU-add4201d-17de-2e2d-3156-4121aa80135d
[Fri Aug  6 07:13:43 2021] NVRM: Xid (PCI:0000:01:00): 79, pid=44387, GPU has fallen off the bus.
[Fri Aug  6 07:13:43 2021] NVRM: GPU 0000:01:00.0: GPU has fallen off the bus.
[Fri Aug  6 07:13:43 2021] NVRM: GPU 0000:01:00.0: GPU serial number is \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff.
[Fri Aug  6 07:13:43 2021] NVRM: A GPU crash dump has been created. If possible, please run
                           NVRM: nvidia-bug-report.sh as root to collect this data before
                           NVRM: the NVIDIA kernel module is unloaded.

nvidia-bug-report.log.gz (419.4 KB)

1 Like
3

I connect my GTX 1080 with type-c with NUC, Ubuntu 20.04, Xorg.
Nvidia driver version: 470.57.02

kern.log:
Aug 7 20:39:03 nuc kernel: [27494.512428] NVRM: GPU at PCI:0000:07:00: GPU-245dd2b5-9cc3-f0ad-353b-c18b71eec7fe
Aug 7 20:39:03 nuc kernel: [27494.512438] NVRM: Xid (PCI:0000:07:00): 32, pid=263876, Channel ID 00000039 intr 80064000
Aug 7 20:39:03 nuc kernel: [27494.554155] DMAR: DRHD: handling fault status reg 3
Aug 7 20:39:03 nuc kernel: [27494.554178] DMAR: [DMA Write] Request device [07:00.0] PASID ffffffff fault addr 0 [fault reason 05] PTE Write access is not set
Aug 7 20:39:03 nuc kernel: [27494.654385] NVRM: Xid (PCI:0000:07:00): 79, pid=263876, GPU has fallen off the bus.
Aug 7 20:39:03 nuc kernel: [27494.654386] NVRM: GPU 0000:07:00.0: GPU has fallen off the bus.
Aug 7 20:39:03 nuc kernel: [27494.654414] NVRM: GPU 0000:07:00.0: GPU serial number is .
Aug 7 20:39:03 nuc kernel: [27494.755072] NVRM: Xid (PCI:0000:07:00): 32, pid=263876, Channel ID 00000039 intr 00024000