Attention is currently required from: laforge, dexter. pespin has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-hnbgw/+/26795 )
Change subject: mgw_fsm: add MGW support to osmo-hnbgw ......................................................................
Patch Set 25:
(6 comments)
File src/osmo-hnbgw/context_map.c:
https://gerrit.osmocom.org/c/osmo-hnbgw/+/26795/comment/d8015f32_fed1c4a0 PS25, Line 142: /* a possibly still existing MGW FSM must be termiated when the context terminated
File src/osmo-hnbgw/mgw_fsm.c:
https://gerrit.osmocom.org/c/osmo-hnbgw/+/26795/comment/4d2caeb3_d7cfa334 PS25, Line 50: * do not support at the moemnt. */ moment
https://gerrit.osmocom.org/c/osmo-hnbgw/+/26795/comment/0e787d32_8885d161 PS25, Line 52: /* Send Iu Release request, this is done in errornous cases from which we cannot recover */ Request
https://gerrit.osmocom.org/c/osmo-hnbgw/+/26795/comment/265e6df2_45574422 PS25, Line 53: static void tx_release_req(struct hnbgw_context_map *map) This function doesn't really look like belonging in this file. It should be oved somewhere Iu related, not here which is MGW related.
https://gerrit.osmocom.org/c/osmo-hnbgw/+/26795/comment/8297121e_e3e32ff0 PS25, Line 424: rc = msgb_resize_area(mgw_fsm_priv->ranap_rab_ass_resp_oph->msg, I'm still not understanding why is this resize here needed. Your previous answer didn't really explain it imho. Why do you need to resize if ranap_rab_ass_resp_encode gets passed msgb_l2len to make sure it doesn't overflow?
https://gerrit.osmocom.org/c/osmo-hnbgw/+/26795/comment/0e23e35b_29de19b0 PS25, Line 627: static int handle_rab_release(struct hnbgw_context_map *map, struct osmo_prim_hdr *oph, ranap_message *message) Same, this function doesn't really look like belonging in this file.