Layer 3&4 isn’t that bad, but some people refer to CAN when they actually mean devicenet which goes right up to layer 7, and is considerably more complex…
This driver is for real CAN-traffic, not for higer level protocol. I think we have to skip this project since the schedule would be quite insane (only month to make things happen).
But thanks for answers, and nice to see that there are some sollutions allready on a market.
JUKO, this depends on the workforce you have available and on the funds of course. There are protocol stacks commercially available includung C source code that usually port quite easily to QNX, and the aforementioned CAN driver has a generic pathname interface. One month should be sufficient to make this work, unless there are lots of other things to do which are not discussed here. We did also a CANopen stack under QNX6 in a custom engineering project, maybe this can be adopted to your needs. As I said, it all depends on funds.
My workforce is really limited and so are other recourses. It’s better to leave that project alone or some other company do it. Than gamble with fines, when project is late. But thanks for ideas and help anyway.
I hope we will get a anotherd QNX/RTOS-project someday, the enviroment looked quite facinating.