Difficulty implementing Nvidia library compatibility

AI & Data Science Deep Learning (Training & Inference) TensorRT
,
1

I’m currently trying to install tensoRT 8.6.1

I installed drive 570.124.06:

nvidia@nvidia-Dell-G15-5511:~$ nvidia-smi
Tue Apr 22 00:44:56 2025
±--------------------------------------------------------------------------------------------+
| NVIDIA-SMI 570.124.06 Driver Version: 570.124.06 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 RTX 3060 … Off | 00000000:01:00.0 Off | N/A |
| N/A 52C P8 12W / 115W | 15MiB / 6144MiB | 0% Default |
| | | N/A |
±--------------------------------------±------------------------------------±---------------------+

±--------------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|===========================================================================|
| 0 N/A N/A 1730 G /usr/lib/xorg/Xorg 4MiB |
±--------------------------------------------------------------------------------------------+

I already installed cuda 12.0:
~$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2022 NVIDIA Corporation
Built on Mon_Oct_24_19:12:58_PDT_2022
Cuda compilation tools, release 12.0, V12.0.76
Build cuda_12.0.r12.0/compiler.31968024_0

CudNN 8.8.0:
nvidia@nvidia-Dell-G15-5511:~$ cat /usr/include/cudnn_version.h | grep CUDNN_MAJOR -A 2
define CUDNN_MAJOR 8
define CUDNN_MINOR 8
define CUDNN_PATCHLEVEL 0

define CUDNN_VERSION (CUDNN_MAJOR * 1000 + CUDNN_MINOR * 100 + CUDNN_PATCHLEVEL)

/* cannot use constexpr here since this is a C-only file */

after I use:

sudo dpkg -i nv-tensorrt-local-repo-${os}-${tag}_1.0-1_amd64.deb
sudo cp /var/nv-tensorrt-local-repo-${os}-${tag}/*-keyring.gpg /usr/share/keyrings/
sudo apt-get update

and then:

nvidia@nvidia-Dell-G15-5511:~$ sudo apt-get install tensorrt
[sudo] password for nvidia:
Reading package lists… Done
Building dependency tree
Reading state information… Done
You might want to run ‘apt --fix-broken install’ to correct these.
The following packages have unmet dependencies:
libnvinfer-dev : Depends: libcublas.so.12-dev
libnvinfer-samples : Depends: libcudart.so.12-dev or
cuda-cudart-cross-amd64-12-0 but it is not installable
Depends: cuda-nvcc-12-1 but it is not going to be installed or
cuda-nvcc-12-0 but it is not going to be installed
libnvinfer8 : Depends: libcublas.so.12
tensorrt : Depends: libnvinfer10 (= 10.9.0.34-1+cuda12.8) but it is not going to be installed
Depends: libnvinfer-plugin10 (= 10.9.0.34-1+cuda12.8) but it is not going to be installed
Depends: libnvinfer-vc-plugin10 (= 10.9.0.34-1+cuda12.8) but it is not going to be installed
Depends: libnvinfer-lean10 (= 10.9.0.34-1+cuda12.8) but it is not going to be installed
Depends: libnvinfer-dispatch10 (= 10.9.0.34-1+cuda12.8) but it is not going to be installed
Depends: libnvonnxparsers10 (= 10.9.0.34-1+cuda12.8) but it is not going to be installed
Depends: libnvinfer-bin (= 10.9.0.34-1+cuda12.8) but 8.6.1.6-1+cuda12.0 is to be installed
Dependencies: libnvinfer-dev (= 10.9.0.34-1+cuda12.8) but 8.6.1.6-1+cuda12.0 is to be installed
Depends: libnvinfer-lean-dev (= 10.9.0.34-1+cuda12.8) but 8.6.1.6-1+cuda12.0 is to be installed
Depends: libnvinfer-plugin-dev (= 10.9.0.34-1+cuda12.8) but 8.6.1.6-1+cuda12.0 is to be installed
Depends: libnvinfer-vc-plugin-dev (= 10.9.0.34-1+cuda12.8) but 8.6.1.6-1+cuda12.0 is to be installed
Depends: libnvinfer-dispatch-dev (= 10.9.0.34-1+cuda12.8) but 8.6.1.6-1+cuda12.0 is to be installed
Depends: libnvonnxparsers-dev (= 10.9.0.34-1+cuda12.8) but 8.6.1.6-1+cuda12.0 is to be installed
Depends: libnvinfer-samples (= 10.9.0.34-1+cuda12.8) but 8.6.1.6-1+cuda12.0 is to be installed
Depends: python3-libnvinfer-dev (= 10.9.0.34-1+cuda12.8) but 8.6.1.6-1+cuda12.0 is to be installed
E: Unmet dependencies. Try ‘apt --fix-broken install’ with no packages (or specify a solution).
nvidia@nvidia-Dell-G15-5511:~$ apt --fix-broken install
E: Could not open lock file /var/lib/dpkg/lock-frontend - open (13: Permission denied)
E: Unable to acquire the dpkg frontend lock (/var/lib/dpkg/lock-frontend), are you root?
nvidia@nvidia-Dell-G15-5511:~$ sudo apt --fix-broken install
Reading package lists… Done
Building dependency tree
Reading state information… Done
Correcting dependencies… Done
The following additional packages will be installed:
cuda-cccl-12-1 cuda-cccl-12-8 cuda-cudart-12-1 cuda-cudart-12-8 cuda-cudart-dev-12-1 cuda-cudart-dev-12-8 cuda-driver-dev-12-1
cuda-driver-dev-12-8 cuda-nvcc-12-1 cuda-toolkit-12-1-config-common cuda-toolkit-12-8-config-common cuda-toolkit-12-config-common
cuda-toolkit-config-common libcublas-12-8 libnvinfer-bin libnvinfer-dev libnvinfer-dispatch-dev libnvinfer-dispatch10 libnvinfer-headers-dev
libnvinfer-headers-plugin-dev libnvinfer-lean-dev libnvinfer-lean10 libnvinfer-plugin-dev libnvinfer-plugin10 libnvinfer-samples
libnvinfer-vc-plugin-dev libnvinfer-vc-plugin10 libnvinfer10 libnvonnxparsers-dev libnvonnxparsers10 python3-libnvinfer python3-libnvinfer-dev
python3-libnvinfer-dispatch python3-libnvinfer-lean tensorrt tensorrt-dev
The following packages will be REMOVED:
libnvparsers-dev
The following NEW packages will be installed:
cuda-cccl-12-1 cuda-cccl-12-8 cuda-cudart-12-1 cuda-cudart-12-8 cuda-cudart-dev-12-1 cuda-cudart-dev-12-8 cuda-driver-dev-12-1
cuda-driver-dev-12-8 cuda-nvcc-12-1 cuda-toolkit-12-1-config-common cuda-toolkit-12-8-config-common cuda-toolkit-12-config-common
cuda-toolkit-config-common libcublas-12-8 libnvinfer-dispatch10 libnvinfer-lean10 libnvinfer-plugin10 libnvinfer-vc-plugin10 libnvinfer10
libnvonnxparsers10
The following packages will be upgraded:
libnvinfer-bin libnvinfer-dev libnvinfer-dispatch-dev libnvinfer-headers-dev libnvinfer-headers-plugin-dev libnvinfer-lean-dev
libnvinfer-plugin-dev libnvinfer-samples libnvinfer-vc-plugin-dev libnvonnxparsers-dev python3-libnvinfer python3-libnvinfer-dev
python3-libnvinfer-dispatch python3-libnvinfer-lean tensorrt tensorrt-dev
16 upgraded, 20 newly installed, 1 to remove and 41 not upgraded.
15 not fully installed or removed.
Need to get 4,972 MB of archives.
After this operation, 10.1 GB of additional disk space will be used.
Do you want to continue? [Y/n]

2

Hi @matheuscaltran can you share more info about what the issue is here? At a glance it looks like the installation is moving forward correctly - what happens when you continue the installation?

Thanks,

Sophie

3

but he wants to install another version of tensorRT and other versions of cuda even though I already have them installed

4

Hi @matheuscaltran please can you open an issue on the Github for TensorRT? See here for more details - Best practices for reporting an issue or bug relating to TensorRT Thanks!

Sophie

6

ok thanks, but have you ever thought about improving NVIDIA’s library system? I see a lot of people complaining

I spend more time trying to install compatible libraries than I do programming