Attention is currently required from: osmith.
pespin has posted comments on this change. ( https://gerrit.osmocom.org/c/osmocom-bb/+/32596 )
Change subject: layer23: Introduce apn_fsm ......................................................................
Patch Set 2:
(2 comments)
File src/host/layer23/src/common/apn_fsm.c:
https://gerrit.osmocom.org/c/osmocom-bb/+/32596/comment/eca4f8ba_9327ce0b PS2, Line 103: apn_fsm_state_chg(fi, APN_ST_DISABLED);
Don't we need to check *((bool *)data)? I guess the assumption is that it is false here, as I unders […]
In theory it's not really needed, since in this state we know for sure the current state is to have GPRS enabled, so I'd expect the event to be "disable" here, but yeah, let's test it to make sure.
https://gerrit.osmocom.org/c/osmocom-bb/+/32596/comment/5ec1a663_fa8a80e9 PS2, Line 139: apn_fsm_state_chg(fi, APN_ST_INACTIVE);
- is this correct, receiving CTX_ACC and then going to inactive here? […]
Yes, because it's "DEACT". I left out REJ for now, we are not even implementing most of deactivation anyway so far, and I need to figure out how to trigger app-specific code to instruct GMM/SM layers to do stuff from the common/ code.
So yes, this FSM will probably be extended in the future. The problem is that I can't simply implement everything at the same time in one commit, that would be impossible to review. Pushing this in small bits allows other people already testing and contributing too.
-
pespin