Attention is currently required from: pespin.
4 comments:
Commit Message:
Patch Set #1, Line 27: a separate SCCP CR without payload.
Not sure if it really makes sense to use any value != 0 here,so it probably makes more sense to have […]
Sure.
With a RANAP payload size, a user can still experiment with a bound below which to still avoid the additional CR->CC roundtrip, so I thought this is more interesting than a boolean...
Would be trivial to change this to a boolean flag, basically shows only in the cfg.
Opinions anyone else?
File include/osmocom/hnbgw/context_map.h:
Patch Set #1, Line 42: RUA_CN_DomainIndicator_t cached_domain_indicator;
isn't this the same as is_ps?
I guess so.
rua_to_scu() wants a RUA_CN_ComainIndicator_t arg, so I just added it ...
Ok, so indeed it seems a hnbgw_context_map is always separate for CS and PS, so can replace with
(is_ps ? RUA_CN_DomainIndicator_ps_domain : RUA_CN_DomainIndicator_cs_domain)
File src/osmo-hnbgw/hnbgw.c:
Patch Set #1, Line 93: gw->config.max_sccp_cr_payload_len = 999999;
this really looks like a hack. Use -1 value and a "no sccp..." VTY command. […]
on the VTY, we often use max values like 99999, that's the only reason.
Given a usual MTU is like 1500, I assume 999k is sufficiently close to infinity.
File src/osmo-hnbgw/hnbgw_rua.c:
Patch Set #1, Line 274: if (data && len && map && !map->is_ps && !release_context_map) {
This new check looks non related? why did you add it?
Because ranap_cn_rx_co_decode() crashes when there is no payload data.
To view, visit change 28230. To unsubscribe, or for help writing mail filters, visit settings.