Facing issue in creating development environment for Deepstream

AI & Data Science Deep Learning (Training & Inference)
1

I was able to run sample Deepstream applications after fetching it’s docker image from ngc. However, I am not able to establish proper development environment with debugging enabled.

I have tried setting up gdbserver for remote debugging and then attaching to opened port (#5000 in my case) via eclipse-CDT. Debugger does gets attached to gdbserver, as can be inferred from “connected” message displayed over deepstream-container, however only some assembly code shows up.

In short, rather than breaking at my source code - it breaks at some assembly code.

Have also asked for help on:

  1. [url]Need to debug mixed "C and C++" code running in docker - Stack Overflow
  2. [url]https://www.reddit.com/r/docker/comments/azdfsz/need_to_debug_cc_code_running_in_docker/[/url]

as it seems to be a generic issue.

Any pointer on how you or your team established development workflow for deepstream development will be much appreciated.