Compile with tensorrt lib on jetson orin nano devkit in C++

[paloutic]

Hello,

I want to use the TensorRT library in a C++ program. So, I add the following line in my CMakeLists.txt file :
find_package(TensorRT REQUIRED)

When I run the cmake command, I’ve got an error :

CMake Error at CMakeLists.txt:18 (find_package):
  By not providing "FindTensorRT.cmake" in CMAKE_MODULE_PATH this project has
  asked CMake to find a package configuration file provided by "TensorRT",
  but CMake did not find one.

  Could not find a package configuration file provided by "TensorRT" with any
  of the following names:

    TensorRTConfig.cmake
    tensorrt-config.cmake

  Add the installation prefix of "TensorRT" to CMAKE_PREFIX_PATH or set
  "TensorRT_DIR" to a directory containing one of the above files.  If
  "TensorRT" provides a separate development package or SDK, be sure it has
  been installed.

My configurations is as follow :
Jetson Orin nano Developer Kit (install from the SD Card Image Method)
Jetpack 5.1.1
CUDA: 11.4.315
cuDNN: 8.6.0.166
TensorRT: 8.5.2.2 (the command dpkg-query -W tensorrt return tensorrt 8.5.2.2-1+cuda11.4)

I can’t find the TensorRTConfig.cmake file, the tensorrt-config.cmake file or the TensorRT-8.5.2.2 folder. I just have a tensorrt folder in /usr/src/ :

cor@wporin:/usr/src/tensorrt$ ls
bin  data  samples

Is the TensorRt library really installed? What am I missing here?

Thanks,

[AastaLLL]

Hi,

TensorRT library is called nvinfer and is located at:

$ ll /usr/lib/aarch64-linux-gnu/libnvinfer*
$ ll /usr/include/aarch64-linux-gnu/NvInfer*

Below is a sample cmake file for your reference:

dusty-nv/jetson-inference/blob/master/CMakeLists.txt

Thanks.

[paloutic]

@AastaLLL, thank you for your answer. I finally get it to work by creating a file like this one : https://github.com/smistad/FAST/blob/master/cmake/FindTensorRT.cmake

Then I copy it into the same directory as my CMakeLists.txt file and everything is fine.

Thanks