Core dump error when running DFS on WSL2

General
1

I am using Ubuntu 22.04 with Python 3.10.12.
I installed the DFS using the .whl package inside a Python virtualenv (virtual environment).
I do not need to use a GPU. I only want to convert an ONNX or TensorFlow model to an HEF file using CPU only.

However, when I run the following command:

(hailo_env) user@user:~/hailo/hailo_ai_sw_suite$ hailo --help
[info] No GPU chosen and no suitable GPU found, falling back to CPU.
Aborted (core dumped)

The process aborts with a core dump even though I want to run everything on CPU.

Is there any configuration to force CPU-only mode, or any known issue that causes the tool to crash when no GPU is present?

How can I run the Hailo model conversion pipeline without a GPU?

2

Hey @Kim_Taehyun ,

Welcome to the Hailo Community!

Don’t worry about the “No GPU found” message—it’s normal, as DFC works perfectly on CPU only for model conversion. The crash afterward, however, points to an environment issue.

My suggestion: Clean Reinstall

Please try a fresh install in a new virtual environment using the following specs:

  • OS: Ubuntu 22.04
  • Python: 3.10
  • RAM: $\ge 16$ GB
  • Dev Packages: python3.10-dev, python3-tk, graphviz, libgraphviz-dev

Once your environment is set up, you can run the standard three-step conversion on your CPU: Parse $\rightarrow$ Optimize $\rightarrow$ Compile to HEF. It will be slower than using a GPU, but it works.

Also, it looks like hailo -h (or hailo --help) is working fine, which is a good sign!

(hailo_env) omria@had-1127-dsk:~$ hailo -h
[info] No GPU chosen, Selected GPU 0
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
E0000 00:00:1764081123.691593  899937 cuda_dnn.cc:8310] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
E0000 00:00:1764081123.694973  899937 cuda_blas.cc:1418] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
[info] Current Time: 16:32:06, 11/25/25
[info] CPU: Architecture: x86_64, Model: AMD Ryzen 7 5700G with Radeon Graphics, Number Of Cores: 16, Utilization: 0.1%
[info] Memory: Total: 30GB, Available: 28GB
[info] System info: OS: Linux, Kernel: 6.5.0-27-generic
[info] Hailo DFC Version: 3.32.0
[info] HailoRT Version: Not Installed
[info] PCIe: No Hailo PCIe device was found
[info] Running `hailo -h`
usage: hailo [-h] [--version] {analyze-noise,compiler,params-csv,parser,profiler,optimize,visualizer,tutorial,har,join,har-onnx-rt,runtime-profiler,dfc-studio,help} ...

Hailo Command Line Utility

positional arguments:
  {analyze-noise,compiler,params-csv,parser,profiler,optimize,visualizer,tutorial,har,join,har-onnx-rt,runtime-profiler,dfc-studio,help}
                        Hailo utilities aimed to help with everything you need
    analyze-noise       Analyze network quantization noise
    compiler            Compile Hailo model to HEF binary files
    params-csv          Convert translated params to csv
    parser              Translate network to Hailo network
    profiler            Hailo models Profiler
    optimize            Optimize model
    visualizer          HAR visualization tool
    tutorial            Runs the tutorials in jupyter notebook
    har                 Query and extract information from Hailo Archive file
    join                Join two Hailo models to a single model
    har-onnx-rt         Generates ONNX-Runtime model including pre/post processing
    runtime-profiler    Hailo Runtime Profiler
    dfc-studio          Start DFC Studio
    help                Show the list of commands

options:
  -h, --help            show this help message and exit
  --version             show program's version number and exit

If the issues persist after a clean reinstall, please send us your DFC version number, pip list output, and the error logs so we can investigate further.