Cross Compilation for CUDA Applications on Windows host systems and Jetson Nano target systems

Hi,

I’m a member of the faculty of electronic engineering at the UAS Technikum Wien. I’m using the Jetson Nano in combination with multicopters. I’m currently trying to setup a development environment, which enables me to implement applications based on CUDALibs on my windows host system and cross-compile them for the Jetson Nano (target system).

Hence, I’m searching for a way to transfer the cross-compiled applications to the Jetson Nano, let them run on the Jetson Nano and debug them on the Jetson Nano.

My Jetson Nano already offers a VNC connection, an sftp connection and an ssh connection. I’m guessing I’m not far away from a cross-compilation setup. My questions are:

1.) Which tools are required on the host computer (Windows 10 OS) to develop and cross-compile applications?
2.) Which tools are required on the target system (Jetson Nano) to actually receive the compiled applications from the host computer, run them and debug them on the target system.

Any help is kindly appreciated,

Patrick

Hi,
Please check the revelant post:
https://devtalk.nvidia.com/default/topic/1049299/jetson-nano/cross-compile-cuda-engines/post/5325342/#5325342