Nvidia-smi cannot find Tesla T4 on ubuntu 18.04

Graphics / Linux Linux
1

Hello all,

my setup:

  • Ubuntu 18.04
  • Intel(R) Core™ i7-6700K CPU @ 4.00GHz
  • 16 GB RAM
  • BeQuiet 400W PSU
  • Tesla T4 Compute Module

I want to use the compute module to run ISAAC SIM.

I installed the driver via:

sudo add-apt-repository ppa:graphics-drivers
sudo apt-get update
sudo apt-get install nvidia-driver-440

nvidia-smi outputs:
No devices were found

dmesg | grep -i nvidia outputs:

[    2.134397] nvidia: loading out-of-tree module taints kernel.
[    2.134405] nvidia: module license 'NVIDIA' taints kernel.
[    2.146271] nvidia: module verification failed: signature and/or required key missing - tainting kernel
[    2.155928] nvidia-nvlink: Nvlink Core is being initialized, major device number 237
[    3.334679] NVRM: loading NVIDIA UNIX x86_64 Kernel Module  460.84  Wed May 26 20:14:59 UTC 2021
[    3.337455] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms  460.84  Wed May 26 20:01:59 UTC 2021
[    3.338525] [drm] [nvidia-drm] [GPU ID 0x00000100] Loading driver
[    3.642238] caller os_map_kernel_space.part.10+0x98/0xa0 [nvidia] mapping multiple BARs
[    3.892030] [drm:nv_drm_load [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Failed to allocate NvKmsKapiDevice
[    3.892110] [drm:nv_drm_probe_devices [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Failed to register device
[    3.900025] nvidia-uvm: Loaded the UVM driver, major device number 235.
[    4.060238] caller os_map_kernel_space.part.10+0x98/0xa0 [nvidia] mapping multiple BARs
[    4.397666] caller os_map_kernel_space.part.10+0x98/0xa0 [nvidia] mapping multiple BARs
[   13.763567] caller os_map_kernel_space.part.10+0x98/0xa0 [nvidia] mapping multiple BARs
[   14.124830] caller os_map_kernel_space.part.10+0x98/0xa0 [nvidia] mapping multiple BARs
[   20.727715] caller os_map_kernel_space.part.10+0x98/0xa0 [nvidia] mapping multiple BARs
[   21.072616] caller os_map_kernel_space.part.10+0x98/0xa0 [nvidia] mapping multiple BARs
[   32.864519] caller os_map_kernel_space.part.10+0x98/0xa0 [nvidia] mapping multiple BARs
[   33.210691] caller os_map_kernel_space.part.10+0x98/0xa0 [nvidia] mapping multiple BARs
[  217.683689] caller os_map_kernel_space.part.10+0x98/0xa0 [nvidia] mapping multiple BARs
[  218.029766] caller os_map_kernel_space.part.10+0x98/0xa0 [nvidia] mapping multiple BARs
[  632.074290] caller os_map_kernel_space.part.10+0x98/0xa0 [nvidia] mapping multiple BARs
[  632.420370] caller os_map_kernel_space.part.10+0x98/0xa0 [nvidia] mapping multiple BARs

dmesg | grep -i nvrm outputs:

[    3.334679] NVRM: loading NVIDIA UNIX x86_64 Kernel Module  460.84  Wed May 26 20:14:59 UTC 2021
[    3.891769] NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x26:0xffff:1290)
[    3.891940] NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0
[    4.302782] NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x26:0xffff:1290)
[    4.302937] NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0
[    4.640048] NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x26:0xffff:1290)
[    4.640225] NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0
[   14.024478] NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x26:0xffff:1290)
[   14.024547] NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0
[   14.374264] NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x26:0xffff:1290)
[   14.374352] NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0
[   20.974518] NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x26:0xffff:1290)
[   20.974656] NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0
[   21.319343] NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x26:0xffff:1290)
[   21.319511] NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0
[   33.112809] NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x26:0xffff:1290)
[   33.112862] NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0
[   33.458035] NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x26:0xffff:1290)
[   33.458201] NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0
[  217.931411] NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x26:0xffff:1290)
[  217.931818] NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0
[  218.276897] NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x26:0xffff:1290)
[  218.277090] NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0
[  632.322032] NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x26:0xffff:1290)
[  632.322190] NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0
[  632.668198] NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x26:0xffff:1290)
[  632.668645] NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0

I attached the result of sudo nvidia-bug-report.sh to this post.

Possibly related topics: Nvidia-smi error no device and dmesg show "rm_init_adapter failed, device minor number 0"

What is the problem in my setup?

Is it even feasible to have ISAAC Sim running on the Tesla T4 GPU while having a physical monitor attached to the Intel GPU?

nvidia-bug-report.log.gz (149.6 KB)

1 Like
2

Please enable “Above 4G decoding” or “large/64bit BARs” in bios, disable CSM and reinstall the whole OS in EFI mode. CSM boot doesn’t work with Teslas.

1 Like
3

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.