My colleague and I have studied some time on those man pages. As it is in many industrial setups. There are several userspace j implementations doing so. What I wanted to say here is that when you start issuing j CAN frames from the same host, the local j stack will not recognize those j frames as local, and therefore, the SA must differ. We had ported it into 3.

Uploader: Mezizahn
Date Added: 11 April 2016
File Size: 19.3 Mb
Operating Systems: Windows NT/2000/XP/2003/2003/7/8/10 MacOS 10/X
Downloads: 33214
Price: Free* [*Free Regsitration Required]

And since they are assembled way before consulting sockets, I don’t see how to turn it into a sockopt. This does not control the CAN loopback, but more the J loopback. Need to create a macro for that. This serves as a base for filter liunx. Things that could be handled outside the kernel should be handled outside the kernel.

Yet, I think the addressing is a leap forward. It’s not mandatory I do not require it neither. Traffic not destined for the ‘localhost’ should be dropped normally, but with this ‘engineering’ option, one can spy the bus for others traffic too.


lniux Make it a constant that can be overridden via sockopt. IMO it must not. I addressed a subset in this reply. Sign up using Email and Password.

[v4,4/5] can-j add documentation – Patchwork

As you stated somewhere above at “2. Please keep off attaching j addresses to network devices. This can have several reasons: The kernel considers the arbitration, but does i1939 act. If a retry is needed, the delay is IMO link specific.

In reply to this post by Kurt Van Dijck This contains the j documentation.

J1939 Stack for Linux (Object)

They are not only determined by the platform, but also on the bus that the system is connected to. This is a bad approach.

Do you mean bind and connect here?

It would be not good to have the j stack complaining then. I fitted j to existing ones. You’re right in that you’re quite some prototype of a potential user.

The thing is, I did not invent these semantics. The destination address would be yourself, unless you’re spying Oliver, Your comments are very ‘diverse’. Though, if you google for it you may find the following link: My colleague and I have studied some time on those man pages.


First some general remarks inside the text: Even when it’s a nice idea to handle all the address claiming infrastructure inside the kernel: I strip the skb to contain only [dlc] bytes. The lack of multiple ECU support on a single host coming hand-in-hand with linud attached to network interfaces needs some rework. The kernel just follows the process and hold traffic during the process.

It was a key feature of the design. That is something for netdev mailing list. A constant is a bit rude I think, but I don’t think this is a heavily used metric.