How to translate xx/x scripts of TensorRT installation?

I have downloaded nv-tensorrt-local-repo-ubuntu2004-8.6.1-cuda-12.0_1.0-1_arm64.deb

But I don’t know how to translate to those xx x.x?

os="ubuntuxx04"
tag="10.x.x-cuda-x.x"
sudo dpkg -i nv-tensorrt-local-repo-${os}-${tag}_1.0-1_arm64.deb
sudo cp /var/nv-tensorrt-local-repo-${os}-${tag}/*-keyring.gpg /usr/share/keyrings/
sudo apt-get update

My env:

Software part of jetson-stats 4.2.12 - (c) 2024, Raffaello Bonghi
Model: NVIDIA Orin Nano Developer Kit - Jetpack 5.1.4 [L4T 35.6.0]
NV Power Mode[0]: 15W
Serial Number: [XXX Show with: jetson_release -s XXX]
Hardware:
 - P-Number: p3767-0005
 - Module: NVIDIA Jetson Orin Nano (Developer kit)
Platform:
 - Distribution: Ubuntu 20.04 focal
 - Release: 5.10.216-tegra
jtop:
 - Version: 4.2.12
 - Service: Active
Libraries:
 - CUDA: 12.3.107
 - cuDNN: 8.6.0.166
 - TensorRT: 8.5.2.2
 - VPI: 2.4.8
 - Vulkan: 1.3.204
 - OpenCV: 4.9.0 - with CUDA: YES
DeepStream C/C++ SDK version: 6.3

Python Environment:
Python 3.8.10
    GStreamer:                   YES (1.16.3)
  NVIDIA CUDA:                   YES (ver 11.4, CUFFT CUBLAS FAST_MATH)
         OpenCV version: 4.9.0  CUDA True
           YOLO version: 8.3.33
         PYCUDA version: 2024.1.2
          Torch version: 2.5.1+l4t35.6
    Torchvision version: 0.20.1a0+3ac97aa
 DeepStream SDK version: 1.1.8
onnxruntime     version: 1.16.3
onnxruntime-gpu version: 1.18.0

Hi,

As mentioned in the below comment, TensorRT is not upgradable on JetPack 5.
So please use the TensorRT 8.5 of the JetPack 5.1.4.

Thanks.

Are you sure about TensorRT 8.5 of the JetPack 5.1.4 is NOT upgradable.

As we have setHardwareCompatibilityLevel issue that needs to upgrade TensorRT 8.5. It doesn’t have this API.

Hi,

The latest TensorRT for JetPack 5 is 8.5.2.
Thanks.

I didn’t find 8.5.2 for jetson also. As you have claimed that ARM SBSA is NOT for jetson.

So which 8.5.2 is for jetson, it’s confusing?

Hi,

TensorRT 8.5 is included in the JetPack 5.1.4.
You can find it in the SDKmananger or below link:

https://repo.download.nvidia.com/jetson#Jetpack%205.1.4

Thanks.

Can you help to check setHardwareCompatibilityLevel which version is implemented this API?

Currently RT8.5 don’t have this API.

Hi,

For JetPack 5, only CUDA is upgradable but up to v12.2.
From JetPack 6, CUDA/cuDNN/TensorRT are upgradable.

Upgradable packages can be found in the website link.
The default packages are included in the JetPack, which can be found in SDKmanger or our apt server.

Thanks.

1 Like

Just FYI.
The feature you are interested in is not supported on Jetson.

https://docs.nvidia.com/deeplearning/tensorrt/developer-guide/index.html#hardware-compat

By default, TensorRT engines are only compatible with the type of device where they were built. With build-time configuration, engines that are compatible with other types of devices can be built. Currently, hardware compatibility is supported only for Ampere and later device architectures and is not supported on NVIDIA DRIVE OS or JetPack.

Thanks.

1 Like

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