You can manually execute each command in the shell individually; the installation script for the current repository has only been tested under the root user.
The script works up to
truenas_admin@truenas[~]$ sudo systemd-sysext merge
No extensions found.
The card is not installed yet. Still in the old machine waiting…
Is it normal behavior ?
Props to zzzhouuu for providing a fix for my old GTX 1080s. It looks like many of you were similarly pinched into figuring something out. The cards I have are sufficient for my needs and I wouldn’t have been able to justify upgrading my GPUs. Consequently, I would have just lost out on functionality.
I do have one question.
Before driver change that dropped GTX 1080 support:
- both cards appear in UI when deploying apps
After zzzhouuu’s fix applies the older drivers:
- one checkbox appears in UI when deploying apps
lspci shows both cards, nvidia-smi only seems to show one. The cards have always had the SLI bridge attached both before and after the update (mostly so I don’t lose it)
lspci -k
0b:00.0 VGA compatible controller: NVIDIA Corporation GP104 [GeForce GTX 1080] (rev a1)
Subsystem: ASUSTeK Computer Inc. GP104 [GeForce GTX 1080]
Kernel driver in use: vfio-pci
Kernel modules: nouveau, nvidia_drm, nvidia
0b:00.1 Audio device: NVIDIA Corporation GP104 High Definition Audio Controller (rev a1)
Subsystem: ASUSTeK Computer Inc. GP104 High Definition Audio Controller
Kernel driver in use: vfio-pci
Kernel modules: snd_hda_intel
0c:00.0 VGA compatible controller: NVIDIA Corporation GP104 [GeForce GTX 1080] (rev a1)
Subsystem: ASUSTeK Computer Inc. GP104 [GeForce GTX 1080]
Kernel driver in use: nvidia
Kernel modules: nouveau, nvidia_drm, nvidia
0c:00.1 Audio device: NVIDIA Corporation GP104 High Definition Audio Controller (rev a1)
Subsystem: ASUSTeK Computer Inc. GP104 High Definition Audio Controller
Kernel driver in use: snd_hda_intel
Kernel modules: snd_hda_intel
nvidia-smi
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 570.172.08 Driver Version: 570.172.08 CUDA Version: 12.8 |
|-----------------------------------------+------------------------+----------------------+
| 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 GTX 1080 Off | 00000000:0C:00.0 Off | N/A |
| 0% 44C P0 43W / 198W | 0MiB / 8192MiB | 0% Default |
| | | N/A |
+-----------------------------------------+------------------------+----------------------+
+-----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|
| No running processes found |
+-----------------------------------------------------------------------------------------+
Any ideas? Is this expected or is something wrong?
Have you tried restarting the system? Does the issue persist after restarting? Sometimes, installations may encounter minor problems that require a restart to resolve.
I appreciate the reply. I restarted several times and also found a likely culprit. My VM was grabbing my card even though I didn’t have it isolated anymore. When I turned it off and rebooted, both cards were listed in nvidia-smi. This seems like completely normal and expected behavior to me, I just failed to understand that removing the isolation didn’t stop the VM from grabbing the device. If I understand correctly now, the isolation is just there to keep my docker containers from snatching the GPU away from my VM (I assume this causes errors)
Did I understand that correctly?
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 570.172.08 Driver Version: 570.172.08 CUDA Version: 12.8 |
|-----------------------------------------+------------------------+----------------------+
| 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 GTX 1080 Off | 00000000:0B:00.0 Off | N/A |
| 0% 45C P0 37W / 198W | 0MiB / 8192MiB | 1% Default |
| | | N/A |
+-----------------------------------------+------------------------+----------------------+
| 1 NVIDIA GeForce GTX 1080 Off | 00000000:0C:00.0 Off | N/A |
| 0% 44C P5 11W / 198W | 0MiB / 8192MiB | 0% Default |
| | | N/A |
+-----------------------------------------+------------------------+----------------------+
+-----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|
| No running processes found |
+-----------------------------------------------------------------------------------------+