What's the right nvidia driver to support both Geforce 3080Ti and Tesla P40 on my Ubuntu 22.04?

It looks like that a P40 needs a driver w/o “-open” suffix but a 3080Ti needs a driver w “-open”… what’s the right driver to install?

Is it possible to support both Geforce 3080Ti and Tesla P40 on my Ubuntu 22.04?

$ ubuntu-drivers devices
== /sys/devices/pci0000:00/0000:00:15.0/0000:03:00.0 ==
modalias : pci:v000010DEd00001B38sv000010DEsd000011D9bc03sc02i00
vendor : NVIDIA Corporation
model : GP102GL [Tesla P40]
manual_install: True
driver : nvidia-driver-470 - distro non-free
driver : nvidia-driver-515-server - distro non-free
driver : nvidia-driver-470-server - distro non-free
driver : nvidia-driver-525 - third-party non-free
driver : nvidia-driver-515 - distro non-free
driver : nvidia-driver-510 - distro non-free
driver : nvidia-driver-525-server - distro non-free
driver : nvidia-driver-418-server - distro non-free
driver : nvidia-driver-450-server - distro non-free
driver : nvidia-driver-390 - distro non-free
driver : xserver-xorg-video-nouveau - distro free builtin

== /sys/devices/pci0000:00/0000:00:15.1/0000:04:00.0 ==
modalias : pci:v000010DEd00002208sv00001043sd00008802bc03sc00i00
vendor : NVIDIA Corporation
model : GA102 [GeForce RTX 3080 Ti]
driver : nvidia-driver-515-open - distro non-free
driver : nvidia-driver-525-open - third-party non-free recommended
driver : xserver-xorg-video-nouveau - distro free builtin

== /sys/devices/pci0000:00/0000:00:0f.0 ==
modalias : pci:v000015ADd00000405sv000015ADsd00000405bc03sc00i00
vendor : VMware
model : SVGA II Adapter
manual_install: True
driver : open-vm-tools-desktop - distro free

$ lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu 22.04.2 LTS
Release: 22.04
Codename: jammy

$ lspci | grep -i nvidia
03:00.0 3D controller: NVIDIA Corporation GP102GL [Tesla P40] (rev a1)
04:00.0 VGA compatible controller: NVIDIA Corporation GA102 [GeForce RTX 3080 Ti] (rev a1)
04:00.1 Audio device: NVIDIA Corporation GA102 High Definition Audio Controller (rev a1)
0b:00.0 3D controller: NVIDIA Corporation GP102GL [Tesla P40] (rev a1)

ubuntu-drivers output bogus results. Please use nvidia-driver-525 as it supports both devices.

I installed nvidia-driver-525-open driver, the P40 GPUs didn’t show up and only 3080Ti made it… Let me try nvidia-driver-525 driver instead…

$ nvidia-smi
Tue Mar 21 15:55:49 2023
±----------------------------------------------------------------------------+
| NVIDIA-SMI 525.89.02 Driver Version: 525.89.02 CUDA Version: 12.0 |
|-------------------------------±---------------------±---------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|===============================+======================+======================|
| 0 NVIDIA GeForce … Off | 00000000:04:00.0 Off | N/A |
| 0% 28C P8 11W / 350W | 1MiB / 12288MiB | 0% Default |
| | | N/A |
±------------------------------±---------------------±---------------------+

±----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
| No running processes found |
±----------------------------------------------------------------------------+

Please run nvidia-bug-report.sh as root and attach the resulting nvidia-bug-report.log.gz file to your post.

Please see the file enclosed.

nvidia-bug-report.log.gz (223.6 KB)

You can’t use the -open kernel modules since the Pascal based P40 is not supported.

After installing nvidia-driver-525, the Geforce 3080Ti disappears and the Tesla P40 just shows up…
Please see the attached bug report zip file.

$ sudo apt install nvidia-driver-525
$ nvidia-smi
Wed Mar 22 19:46:41 2023
±----------------------------------------------------------------------------+
| NVIDIA-SMI 525.89.02 Driver Version: 525.89.02 CUDA Version: 12.0 |
|-------------------------------±---------------------±---------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|===============================+======================+======================|
| 0 Tesla P40 Off | 00000000:03:00.0 Off | Off |
| N/A 28C P8 8W / 250W | 0MiB / 24576MiB | 0% Default |
| | | N/A |
±------------------------------±---------------------±---------------------+
| 1 Tesla P40 Off | 00000000:0B:00.0 Off | Off |
| N/A 34C P8 10W / 250W | 0MiB / 24576MiB | 0% Default |
| | | N/A |
±------------------------------±---------------------±---------------------+

±----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
| No running processes found |
±----------------------------------------------------------------------------+

nvidia-bug-report.log.gz (530.2 KB)

It’s an esxi passthrough system
RmInitAdapter failed! (0x26:0x56:1469)
known issue, you might try driver version 470.57 - 470.82 or upgrade to esxi 8 to check if that has been resolved meanwhile.
Otherwise, please check the esxi thread in this forum for additional info.

Thanks a lot.

Any idea where I can download an old version driver (470.57 - 470.82) ?
Also, do you know how much performance dropback will the old driver have?

Performance should be equal, affected are graphics features (wayland gbm, vulkan) and cuda level.
https://http.download.nvidia.com/XFree86/Linux-x86_64/

nice! let me try old driver first~

Yepp, that makes it, after downloading nvidia driver to 470.82.1 and cuda to 11.4 ~
Thanks a lot.

$ nvidia-smi
Thu Mar 23 11:03:36 2023
±----------------------------------------------------------------------------+
| NVIDIA-SMI 470.82.01 Driver Version: 470.82.01 CUDA Version: 11.4 |
|-------------------------------±---------------------±---------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|===============================+======================+======================|
| 0 Tesla P40 Off | 00000000:03:00.0 Off | Off |
| N/A 30C P8 9W / 250W | 0MiB / 24451MiB | 0% Default |
| | | N/A |
±------------------------------±---------------------±---------------------+
| 1 NVIDIA GeForce … Off | 00000000:04:00.0 Off | N/A |
| 0% 41C P8 28W / 350W | 1MiB / 12053MiB | 0% Default |
| | | N/A |
±------------------------------±---------------------±---------------------+
| 2 Tesla P40 Off | 00000000:0B:00.0 Off | Off |
| N/A 32C P8 10W / 250W | 0MiB / 24451MiB | 0% Default |
| | | N/A |
±------------------------------±---------------------±---------------------+

±----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
| No running processes found |
±----------------------------------------------------------------------------+

$ nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2021 NVIDIA Corporation
Built on Mon_Oct_11_21:27:02_PDT_2021
Cuda compilation tools, release 11.4, V11.4.152
Build cuda_11.4.r11.4/compiler.30521435_0

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.