Attention is currently required from: pespin.

View Change

3 comments:

• File s1gw/ConnHdlr.ttcn:

  • Patch Set #2, Line 185: float Tval := 5.0

    Because sometimes the timeouts were too tight, specially when creating tons of sessions where then a […]

    I see, but this function is all about the S1AP Setup. This particular message does not trigger any PFCP procedures; the PDUs are passed as-is. This is why I am wondering why and whether it really belongs to this patch logically.

  • Patch Set #2, Line 269: do_repeat := true

    only once, but since it's an activate, it has to be a repeat otherwise when code goes into whatever […]

    Ah, I see. Indeed, we need to `repeat` then.

• File s1gw/S1GW_Tests.ttcn:

  • Patch Set #2, Line 81: id) alive

    TBH I don't think it's worth spending time in splitting this ...

    On the one hand, yes. One the other, I oftentimes feel puzzled when I track some change in git-blame and end up with a patch doing something else and changing things by the way. In such case neither the commit name nor the description tells you *why*.

    id: I'm simply refactoring code since anyway i need to add a lot of infra when adding the pfcp side.

    This is just moving `testcasename()` from within the function to the caller.
    Does the PFCP related logic require to pass custom `id` here? I don't think so.

To view, visit change 37920. To unsubscribe, or for help writing mail filters, visit settings.