./visualizer:segmentation fault about semantic_suma

[Chen-Xieyuanli]

Thanks for using our code.

Have you already tried the example demo of rangenet_lib?

You may test the rangnet_lib first. It could be the problem of tensorrt of cuda version.

You could find more details here: https://github.com/PRBonn/rangenet_lib

[xuguangyun]

Thanks for the reply. Yes, the Rangnet_lib have the same error also.
Here is my environment：Ubuntu18.04, TensorRT 6.01+cuda10.1.

[Chen-Xieyuanli]

It would be much better if you could provide more details about the compiling and error message/log.

[xuguangyun]

This is the output of when I run the Rangnet_lib:

scan: /root/catkin_ws5/src/rangenet_lib/example/000000.bin
path: /root/catkin_ws5/src/rangenet_lib/example/model/
verbose: 1

Setting verbosity to: false
Trying to open model
Trying to deserialize previously stored: /root/catkin_ws5/src/rangenet_lib/example/model//model.trt
Could not deserialize TensorRT engine.
Generating from sratch... This may take a while...
Trying to generate trt engine from : /root/catkin_ws5/src/rangenet_lib/example/model//model.onnx
Segmentation fault (core dumped)

[xuguangyun]

I have debug to the error line: netTensorRT.cpp, line 509 builder->platformHasFastFp16();
the value of "builder" is 0;
The error from TensorRT is almost sure,but how can I solved this .
Thanks.

[Chen-Xieyuanli]

Yes, it seems to be a problem caused by the TensorRT.

I didn't meet such a problem so far. I just googled it. It could be a problem with the mismatching of your graphic card and the TensorRT version.

Could you please also report your graphic card and the corresponding driver setups?

You may also give it a try with another version TensorRT, e.g. TensorRT-5.0.

Best

[xuguangyun]

Thanks. I have try with TensorRT5.1 version, the last error had disappear,while another error occur.......
Here is the terminal output:
OpenGL Context version:4.6 core profile
GLEW initialized
OpenGL context version:4.6
OpenGL vendor string: NVIDIA Corporation
OpenGL renderer string GeForce GTX 750/PCIe/SSE2
terminate called after throwing an instance of "rv::XmlError"
what(): Error while parsing in line 1
Aborted(core dumped)

[Chen-Xieyuanli]

Sorry for the late reply. It seems it cannot find the configuration file or something wrong when it parses the file.
You could try to run the visualizer with a specific path of the configuration file, like:
./visualizer /path/to/config/file

[Chen-Xieyuanli]

Hi @xuguangyun

Is there any update about this issue?

[xuguangyun]

Thanks a lot! I have try to run the visualizer with a specific path of the configuration file, the problem solved！

[Dhruv2012]

Sorry for the late reply. It seems it cannot find the configuration file or something wrong when it parses the file.
You could try to run the visualizer with a specific path of the configuration file, like:
./visualizer /path/to/config/file

Hi @Chen-Xieyuanli I am facing the same issue. I have successfully built the RangenetLib. Could you help out? In this case it doesnot seem to a issue of file location.
It displays this output while running the visualizer script.

OpenGL Context Version 4.6 core profile
GLEW initialized.
OpenGL context version: 4.6
OpenGL vendor string : NVIDIA Corporation
OpenGL renderer string: GeForce 940MX/PCIe/SSE2
Segmentation fault (core dumped)

[Chen-Xieyuanli]

Sorry for the late reply. It seems it cannot find the configuration file or something wrong when it parses the file.
You could try to run the visualizer with a specific path of the configuration file, like:
./visualizer /path/to/config/file

Hi @Chen-Xieyuanli I am facing the same issue. I have successfully built the RangenetLib. Could you help out? In this case it doesnot seem to a issue of file location.
It displays this output while running the visualizer script.

OpenGL Context Version 4.6 core profile GLEW initialized. OpenGL context version: 4.6 OpenGL vendor string : NVIDIA Corporation OpenGL renderer string: GeForce 940MX/PCIe/SSE2 Segmentation fault (core dumped)

Hey @Dhruv2012, sorry for the late reply. I never encountered such a problem, but there are some solutions provide by other users #7. It seems to be a problem caused by GTSAM.