When I am installing tensorrt 8.2.4.2 with rpm installation, 8.4.1.5 is getting installed

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

Description

When I am installing tensorrt 8.2.4.2 with rpm installation, 8.4.1.5 is getting installed.
In fact if I try any other version with RPM installation, it is installing 8.4.1.5 only.

I have not seen this problem couple of days back.
I am seeing this problem from today.
and 8.4.1.5 is not available with pip installation.

Environment

TensorRT Version: 8.2.4.2
GPU Type: TITAN V
Nvidia Driver Version: 460.73.01
CUDA Version: 11.2.2
CUDNN Version: 8.2.1.32
Operating System + Version: ubuntu-20.04.1
Python Version (if applicable): 3.7
TensorFlow Version (if applicable): 2.7
PyTorch Version (if applicable): n/a
Baremetal or Container (if container which image + tag): container

Relevant Files

Please find that attached log
tensorrt.rtf (28.8 KB)

#2

Hi,
Please refer to the installation steps from the below link if in case you are missing on anything

Also, we suggest you to use TRT NGC containers to avoid any system dependency related issues.

Thanks!

#4

Can you please acknowledge if someone looking into it…

#5

Hi,

Could you please let us know the rpm file you’re using.

Thank you.

#6

Hi

https://developer.download.nvidia.com/compute/machine-learning/tensorrt/secure/8.2.4/Local_repos/nv-tensorrt-repo-rhel7-cuda11.4-trt8.2.4.2-ga-20220324-1-1.x86_64.rpm

thanks

#7

Hi,

We don’t see similar behavior, could you please try TensorRT 8.2 GA Update 4.
Also, we recommend always using the latest TensorRT version for better performance.

Thank you.

#8

I see same problem again. If I try to install 8.4.1.5, it Is installing latest 8.5 version
I am following RPM installation instructions given in installation guide as below.

From installation guide:-

  1. Download the TensorRT local repo file that matches the RHEL/CentOS version and CPU architecture you are using.
  2. Install TensorRT from the RPM local repo package.

os=“rhelx” tag=“8.x.x-cuda-x.x” sudo rpm -Uvh nv-tensorrt-local-repo-${os}-${tag}-1.0-1.x86_64.rpm sudo yum clean expire-cache

The packages which can be installed are:

graphsurgeon-tf.x86_64 libnvinfer-bin.x86_64 libnvinfer-devel.x86_64 libnvinfer-plugin-devel.x86_64 libnvinfer-plugin8.x86_64 libnvinfer-samples.x86_64 libnvinfer8.x86_64 libnvonnxparsers-devel.x86_64 libnvonnxparsers8.x86_64 libnvparsers-devel.x86_64 libnvparsers8.x86_64 python3-libnvinfer.x86_64 python3-libnvinfer-devel.x86_64 tensorrt.x86_64 tensorrt-devel.x86_64 tensorrt-libs.x86_64 uff-converter-tf.x86_64 onnx-graphsurgeon.x86_64

Install TensorRT.

sudo yum install tensorrt