Felix

I just read that in QP on POSIX each Actor is implemented as a separate pthread. Can the amount of memory reserved for each one be controlled. In my design it would probably be overkill to have a thread per Actor when some of the simple ones could share a thread and save the overhead of a stack etc. Does QP allow Actors to share a thread?

Hi everyone, I'm investigating using QP/C++ on an NVIDIA TX2 that runs a simple machine. High on our list of priorities is flexibility and reliability. We'll be doing a bit of motion control and some image handling as well as telemetry and sending data out over an unreliable connection. We plan on using the Jetson Linux distro Nvidia provides (based on Ubuntu). Has anyone compiled QP for this specific chip or other ARMv8 machines? I assume it would be a pretty safe bet to work out of the box?