Communication CFT Weekly [2025-04-28]

[LittleHuba]

Participants

@crimson11
@alexandruiulian10
Braun Holger
Haraga Mihai
Kaluza Andreas
@qor-lb
@HartmannNico
@arsibo
@vinodreddy-g
@LittleHuba

Agenda

1. Discussion of communication steps for network communication

Meeting Minutes

to 1.:

• Discussion what communication steps happen in network communication

• SOME/IP is just an example network protocol (other protocols should fit mostly, if not please provide feedback)

• Order of steps in the send/receive pipeline is not fixed

• Feedback until next Monday from every company (@FScholPer your company was not represented today, please find somebody who can provide this feedback)

  • Do you see missing steps in the communication? (See slide 1)
  • How do you deploy the different steps in your architecture? (See slide 3)
  • What are your communication usecases (mainly raised to @LittleHuba and @FScholPer)?
    • Do multiple applications really listen to the same communication? How many, how often (1:1 vs. 1:n)?
    • How often does it happen that communication must be sent out via IPC and network at the same time?

Slides presented by @crimson11 :
mw_com_layers_networking.pptx

Misc

Alignment between Rust and C++ workstreams requested (30min)

Will be scheduled for next week
Participants: @HartmannNico @darkwisebear @LittleHuba @crimson11

[dasoho]

Continental feedback:

• Does scheduling (throtteling, debounce) need to have a representation in the API? Do we need to set the corresponding properties dynamically per topic?
• MACsec: Do we need it as part of the stack? https://eclipse-score.github.io/score/main/requirements/stakeholder/index.html#stkh_req__dependability__security_features
• Do we need another transport next to shared memory to support low-latency/high-throughput inter-partition communication on top of a hypervisor, or can we assume that shared memory is supported and accepted across partitions?
• The integration of parts of the stack into QNX io-pkt makes these parts OS-specific. How about other Linux safety supporting OS'es like RHIVOS (Red Hat in Vehicle OS) or EB-LISA (Linux for Safety Applications)? Is there a decision in S-CORE to focus on QNX?

[FScholPer]

MB Feedback:

• VM To VM communication
  • QNX Hypervisor integration -> Using Vsock as one option?
  • Virtual Interrupt needed?
• Do we already have a filtering or rate limiting?
• Differences between datatypes(order,compiler, compiler flags) have to be addressed
• 1:n communication is often when states are shared in the system. So specific to domain can be often the case.
• There are edge cases sending out data ipc and network at the same time.
• Usecase high bandwith streaming
• External communication through dds or zenoh protocol or usage of dynamic transport like in uProtocol