Segmentation fault (core dumped)

[shift093]

( running Ubuntu 18.04 GTX 1080Ti CUDA 10.1 NVIDIA 430.26)

The first run is very good. The second time a Segmentation fault (core dumped) occurs.

root@7afa36add325:/mnt# python SSD_Model/detect_objects_webcam.py
WARNING: Logging before flag parsing goes to stderr.
W0810 08:00:50.809057 140163187611392 deprecation_wrapper.py:119] From /usr/lib/python3.5/dist-packages/graphsurgeon/_utils.py:2: The name tf.NodeDef is deprecated. Please use tf.compat.v1.NodeDef instead.

W0810 08:00:50.809650 140163187611392 deprecation_wrapper.py:119] From /usr/lib/python3.5/dist-packages/graphsurgeon/DynamicGraph.py:4: The name tf.GraphDef is deprecated. Please use tf.compat.v1.GraphDef instead.

TensorRT inference engine settings:

• Inference precision - DataType.FLOAT
• Max batch size - 1

Loading cached TensorRT engine from /mnt/SSD_Model/utils/../workspace/engines/FLOAT/engine_bs_1.buf
TRT ENGINE PATH /mnt/SSD_Model/utils/../workspace/engines/FLOAT/engine_bs_1.buf
Running webcam: True
Segmentation fault (core dumped)

Any help?

[GLBurnett]

Hey @shift093,

I'm working on a fix for this. In the meantime, removing your cached engines (by deleting your .buf files in the /engines folder) can solve the problem for a few runs.

[shift093]

@GLBurnett It works! Btw it works on my tx2 ,and "engine file" works very well.
If there is further news, please inform me again.Thanks!

[HazemAbdelhafez]

Hello, is there any update on this issue? I am still facing the segmentation fault whenever I re-run and I have to remove the .buf file. Strangely this occurs on the Jetson Xavier but not my laptop where I can re-run many times.