TensorRT was linked against cuBLAS/cuBLAS LT 11.3.0 but loaded cuBLAS/cuBLAS LT 11.2.1

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

Description

When I run the sampleMNIST application located in TensorRT-7.2.2.3/samples/ I get an error
TensorRT was linked against cuBLAS/cuBLAS LT 11.3.0 but loaded cuBLAS/cuBLAS LT 11.2.1
I had installed CUDA 11.2, so I also installed CUDA 11.1 as given in requirements.
I set the $PATH to cuda 11.1 and $LD_LIBRARY_PATH to 11.1 lib location.
I have also set path to TensorRT-7.2.2.3 lib in $LD_LIBRARY_PATH.
What am I missing?

Environment

Ubuntu 20.4

TensorRT Version: 7.2.2.3
GPU Type: T4
Nvidia Driver Version: 460
CUDA Version: 11.1
CUDNN Version: 8.0.3
Operating System + Version: Ubuntu 20.4
Python Version (if applicable):
TensorFlow Version (if applicable):
PyTorch Version (if applicable):
Baremetal or Container (if container which image + tag):

Relevant Files

Please attach or include links to any models, data, files, or scripts necessary to reproduce your issue. (Github repo, Google Drive, Dropbox, etc.)

Steps To Reproduce

Please include:

  • Exact steps/commands to build your repro
  • Exact steps/commands to run your repro
  • Full traceback of errors encountered
2 Likes
3

Hi @mehul.vast,
Can you please validate your installation steps with the following guide.

Also we recommend you to use the NGC container so as to avoid any system level dependency.

Thanks!

4

Same problem. Did you solve it?

5

No, not yet. You have any idea about it?

6

Setting the PATH and LD_LIBRARY_PATH to cuda 11.2 worked to remove that warning. However I still can’t execute the sampleMNIST application. I have this error:

terminate called after throwing an instance of ‘pwgen::PwgenException’
what(): Driver error:

is it the same for you?

7

No, not the same. What did you follow for installation of dependencies? Runfile, TAR or DEB? for CUDA, cuDNN?
It has something to do with the installations followed.

8

Hi @mehul.vast ,
We recommend you to try NVIDIA GPU Cloud (NGC) tensorrt optimized containers, which removes many of the host-side dependencies.

Thanks!

9

I tried to follow the guide to install Deepstream as best I could but yet here I am with this error also.

10

have you solve it? I meet the same error.

11

@2511858125 Check the Compatibility matrix of deepstream properly. Every version has to be intact