Deepstream Installation not Working

[AndresGodoy]

Please provide complete information as applicable to your setup.

• Hardware Platform (GPU)
• DeepStream Version 6.0
• TensorRT Version8.0.1
• NVIDIA GPU Driver Version (valid for GPU only) 470.63.01
• Issue Type( questions)

I’m trying to install deepstream SDK on a dGPU following the documentation.

When asked to test the installation

deepstream-app --help

promts the next error:

deepstream-app: error while loading shared libraries: libnppc.so.11: cannot open shared object file: No such file or directory

1Tensor RT installation proposed in the documentation uses Cuda11.3 and the proposed Cuda version is 11.4. Is that okay or we should install Tensor RT 8.X that matches Cuda 11.4 or downgrade Cuda to 11.3?

Is there something else I should check to validate the installation steps?

Thanks beforehand.

[Amycao]

You need to install cuda 11.4 and TensorRT 8.0.1 GA for Ubuntu 18.04 and CUDA 11.3 DEB local repo package
https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_Quickstart.html#dgpu-setup-for-ubuntu

[AndresGodoy]

amycao, thanks for your answer, we effectively followed the instructions on the docs, but still have issues running python deepstream-test1:

python3 deepstream_test_1.py /opt/nvidia/deepstream/deepstream-6.0/samples/streams/sample_720p.h264```

Which outputs this error: 

```bash
Cuda failure: status=1 in cuResData at line 316
Cuda failure: status=1 in cuResData at line 316
Cuda failure: status=700 in cuResData at line 316
Cuda failure: status=700 in cuResData at line 316

[Amycao]

Did you run original test1 app?

[AndresGodoy]

Fixed by checking the CUDA installation folders.

Thanks