No way to use nvprof on l4t tensorrt docker image?

Hello, I’m newbie in tensorrt.

I want to use nvprof to trace gpu for my tensorrt application on docker image.

As you know, there’s only one tensorrt docker image for jetson.
So I pulled the image and tested nvprof.

Then I faced following message.


nvprof: error while loading shared libraries: cannot open shared object file: No such file or directory

So I tried to install libcupti-dev on container.
But, I noticed that there’s no libcupti for bionic(ubuntu18.04lts) / arm64(aarch64).

Then, can’t I use nvprof on l4t tensorrt container?
But it’s weird because I can get result of nvprof on native jetson environment (my device is jetson agx xavier) like following.

~$ nvprof -h
Usage: nvprof [options] [application] [application-arguments] Options:
–aggregate-mode <on|off> Turn on/off aggregate mode for events and metrics specified by subsequent “–events” and “–metrics” options. Those event/metric values will be collected for each domain instance, instead of the whole device. Allowed values: on - turn on aggregate mode (default)
off - turnoff aggregate mode

So, I wonder if I really need to install libcupti to use nvprof on l4t tensorrt container, or not, how can I use nvprof on the container.

Thank you.


TensorRT Version:
GPU Type:
Nvidia Driver Version: 510
CUDA Version: 10.2
CUDNN Version: 8.2.1
Operating System + Version: Ubuntu18.04 lts
Python Version (if applicable):
TensorFlow Version (if applicable):
PyTorch Version (if applicable):
Baremetal or Container (if container which image + tag): l4t-tensorrt:r8.0.1-runtime

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

This looks like a Jetson issue. Please refer to the below samlples in case useful.

For any further assistance, we recommend you to raise it to the respective platform from the below link