Attention is currently required from: neels, pespin.
15 comments:
File include/osmocom/hnbgw/hnbgw.h:
Patch Set #2, Line 193: /* Return true when the user configured GTP mapping to be enabled, by configuring a PFCP link to a UPF.
I'm not sure we really want to disable this feature. […]
I'm not following you, pau. Which feature do you want ot disable? The feature to use the external UPF for gtp-mapping? Or the feature to bypass that external UPF? I think there are valid use cases in both scenrarios.
Paging should not really be something the HNBGW has to decide upon, but the SGSN. The SGSN knows if there's an Iu connection or not, and it should know if it has to buffer packets and page, or if it can send them to the RNC/HNBGW. Am I missing something?
File include/osmocom/hnbgw/ps_rab_fsm.h:
Patch Set #1, Line 8: struct addr_teid {
Maybe add a one line documenttion. […]
Done
Patch Set #1, Line 14: /* Two of this make up a GTP mapping, one for the Access (HNB) side, one for the Core (CN) side.
what does "this" mean here?
Done
Patch Set #1, Line 24: struct half_gtp_map {
gtp_tunnel would be accurate, except that the FAR (Forwarding Action Rule) is closely related to the […]
Done
Patch Set #1, Line 36: /* Whether the RANAP message this RAB's remote address was obtained from encoded the address in x213_nsap */
it is accurate but i'll rephrase to less ambiguous grammar
Done
Patch Set #1, Line 41: struct osmo_fsm_inst *fi;
document the type of fsm? because I see 2 more "fi" below.
Done
Patch Set #1, Line 54: /* Backpointer to the ps_rab_ass_fsm for the RAB Assignment Request from Core that created this RAB. */
commenting in code...
Done
Patch Set #1, Line 79: bool ps_rab_is_established(struct ps_rab *rab);
constify param
Done
File src/osmo-hnbgw/hnbgw_cn.c:
Patch Set #1, Line 373: if (!map->is_ps) {
we (i?) usually list CS above PS ... […]
Done
Patch Set #1, Line 396: rc = ranap_ran_rx_co_decode(map, message, msgb_l2(oph->msg), msgb_l2len(oph->msg));
i think i even considered that when writing the patch, but left it like this to show that the CS cod […]
Done
File src/osmo-hnbgw/hnbgw_pfcp.c:
Patch Set #1, Line 77: if (!hnb_gw_is_gtp_mapping_enabled(hnb_gw)) {
that's a matter of taste ... […]
agrering with neels.
File src/osmo-hnbgw/hnbgw_rua.c:
Patch Set #1, Line 284: if (!map->is_ps) {
idk i like CS above PS better, also before this patch there was "!is_ps" above
Ack
File src/osmo-hnbgw/hnbgw_vty.c:
Patch Set #1, Line 474: if (g_hnb_gw->config.pfcp.local_addr)
same order here as where the cmds are added to vty, but ok whatever
not aware of any specific ordering requirement.
File src/osmo-hnbgw/ps_rab_ass_fsm.c:
Patch Set #1, Line 490: /* See whether all RABs are done establishing, and replace RTP info in the RAB Assignment Response message, so that we
lol thx
Done
Patch Set #1, Line 625: void hnbgw_gtpmap_release(struct hnbgw_context_map *map)
I'm reading "gsmtap" all the time ;)
Done
To view, visit change 28816. To unsubscribe, or for help writing mail filters, visit settings.