Problems building Jetson-Inference from source

Robotics & Edge Computing Jetson & Embedded Systems Jetson AGX Orin
1

Hi, I am new using the Jetson AGX-Orin; well, I am completely new using any Jetson.

I flashed the Jetson-AGX-Orin using the Nvidia SDKmanager with the JetPack 6.2.1 (rev. 1) using Ubuntu 22.04. But while building the jetson-inference project from source, and running make, the following error arose

imagen
imagen732×202 37.1 KB

Hoping you can help me to figure out this issue.

Irandi Gutiérrez Carmona

2

*** Please note that this reply is generated by LLM automatically ***

Based on the provided context, it appears that you are experiencing an issue with building the jetsoninference project from source on your Jetson AGX Orin device, which was flashed using the Nvidia SDK Manager with JetPack 6.2.1 and Ubuntu 22.04.

The error message you provided is not explicitly stated, but based on the image file name (imagen_732202_371KB), it seems that the issue might be related to a memory or storage problem.

Here are a few potential solutions to help you resolve the issue:

  1. Memory and Storage: Ensure that your Jetson AGX Orin device has sufficient memory and storage to build and run the jetsoninference project. You can check the available memory and storage using the free -h and df -h commands, respectively.
  2. Dependency Issues: Verify that all dependencies required by the jetsoninference project are installed and up-to-date. You can check the dependencies using the dpkg -l command and update them using apt-get update and apt-get upgrade.
  3. Build Configuration: Check the build configuration for the jetsoninference project to ensure that it is set up correctly for your Jetson AGX Orin device. You can refer to the project’s documentation or seek help from the project’s community for guidance on build configuration.
  4. JetPack Version: Although you mentioned that you are using JetPack 6.2.1, it’s essential to ensure that you are using the latest version of JetPack. You can check for updates using the Nvidia SDK Manager.
  5. Ubuntu Version: Verify that you are using the correct version of Ubuntu (22.04) and that it is up-to-date.

Comparing your issue to others in the provided context, there are a few similar issues related to building and running projects on Jetson devices:

  • CSI-0 with MIPI 4 lane is not working properly (Topic: CSI-0 with MIPI 4 lane is not working properly): This issue is related to a camera sensor problem, but it might be helpful to check the camera sensor configuration and dependencies.
  • Porting new camera sensor driver (Topic: Porting new camera sensor driver): This issue is related to porting a new camera sensor driver, but it might be helpful to check the driver configuration and dependencies.
  • Jetson Orin Nano - IMX219 not detected on JetPack 6 (Error -121) (Topic: Jetson Orin Nano - IMX219 not detected on JetPack 6 (Error -121)): This issue is related to a camera sensor detection problem, but it might be helpful to check the camera sensor configuration and dependencies.

To further assist you, I would like to request more information about the error message you are experiencing. Please provide the exact error message, and I will do my best to help you resolve the issue.

*** Please give the thumbs up if you get answers you like. Or provide feedback to help us improve the answer. ***

4

I have updated and upgraded the system, everything seems up to date. Running the

make -j$(nproc)

command the following error arose:

[ 92%] Built target camera-capture
/usr/bin/ld: cannot find -lnpymath: No such file or directory
collect2: error: ld returned 1 exit status
make[2]: *** [python/bindings_python_3.10/CMakeFiles/jetson-inference-python-310.dir/build.make:239: aarch64/lib/python/3.10/jetson_inference_python.so] Error 1
make[1]: *** [CMakeFiles/Makefile2:744: python/bindings_python_3.10/CMakeFiles/jetson-inference-python-310.dir/all] Error 2
make: *** [Makefile:136: all] Error 2

5

you could try

pip install numpy==1.26.4

Delete the build directory. Then compile again.

7

Hi,

jetson-inference and jetson-utils are deprecated.
If you want to give it a try, please set up the device with JetPack 6.0.

Or please refer to our tutorial below for the example that runs on JetPack 6.2.

Thanks.

1 Like
8

I am running the Tutorial - Ultralytics YOLOv8 in the recommended link. When running the following command

reComputer run ultralytics-yolo

I receive the following error:

L4T VERSION 36.4.7 is not in the allowed versions list.
The JetPack versions currently supported by this container are: 32.6.1 35.3.1 35.4.1 35.5.0 36.3.0 36.4.0.

Is it recommended to downgrade JetsonPack to some specific version? I flashed the Jetson JetPack6.2.1 (rev.1) using the Nvidia in SDK Manager in Ubuntu 22.04, is it recommended to modify these versions to improve compatibility with the available tutorials?

9

Hi,

Sorry for the missing.
You will also need a WAR. Please find the WAR information below:

Thanks.

11

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.