Failing to profile the AGX Xavier with nvidia nsight graphics using the host

Hi all,

I am trying to profile an application launched on the target Jetson AGX Xavier with NVIDIA Nsight Graphics from the host, but the tool is failing to profile the target, so i have the following output :

Preparing to launch...

Checking file deployment: libNvda.Graphics.Interception.so

Checking file deployment: libnvperfapi64.so

Checking file deployment: libnvperf_target.so

Checking file deployment: VK_LAYER_NV_nomad_release_public_2020_5_1.json

Checking file deployment: libNvPmApi.Core.so

/home/mkartobi/yolov3_onnx/onnx_to_tensorrt.py -i dog -m yolov3 -r 416 -p FP16 -b 1

Launching: /home/mkartobi/yolov3_onnx/onnx_to_tensorrt.py (host: 192.168.55.1)

Process launched

Trying to connect to process...

Searching for attachable processes on 192.168.55.1:49152-49216...

bash: /home/mkartobi/yolov3_onnx/onnx_to_tensorrt.py: Permission denied

That’s the configuration :

Environment

TensorRT Version : 7.7.3
GPU Type : Volta 512 CUDA Cores 64 TensorCores
Nvidia Driver Version :
CUDA Version : 10.2
CUDNN Version : 8.0
Operating System + Version : Ubuntu 18.04 L4T 32
Python Version (if applicable) :
TensorFlow Version (if applicable) :
PyTorch Version (if applicable) :
Baremetal or Container (if container which image + tag) :

How can i avoid this issue and can i profile with nsight graphics directly from the target ?

Hi,

Please noted that you need root authority to get the profiling data.
Have you accessed the device with the root account?

You can find this topic for how to setup the password and configure for root.

Thanks.