Attention is currently required from: pespin.

neels has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-mgw/+/35176?usp=email )

Change subject: IuUP: Allow Initialization with set rem IP address and unset rem port ......................................................................

Patch Set 4: Code-Review+1

(5 comments)

Patchset:

PS4: Ok, I see, and agree with this. Below some petty details below.

First petty detail: we could squash it into that other patch.

File src/libosmo-mgcp/mgcp_network.c:

https://gerrit.osmocom.org/c/osmo-mgw/+/35176/comment/e0c4ce52_25437aa2 PS4, Line 838: osmo_sockaddr_port(&conn->end.addr.u.sa) == 0) { the most concise restriction would be: - validate the IP address if present, - and validate the port if present, independently.

I mean, in this patch we skip both checks if one of them is not set. Instead we can skip only the port number check if port == 0, and skip only the address check if address == ANY.

(That's mainly why it's so compelling to just skip sender validation for early IuUP entirely; doing checks properly rejiggers this entire function.)

https://gerrit.osmocom.org/c/osmo-mgw/+/35176/comment/6ec9f8d2_73ec40b1 PS4, Line 841: announce "not all hNodeB", but some are happily sending RAB Assignment success before IuUP Initialization.

I find "MGCP client" confusing here, the hNodeB obviously does not talk MGCP protocol.

https://gerrit.osmocom.org/c/osmo-mgw/+/35176/comment/a34c5956_389c6fc8 PS4, Line 842: ASs (RAB Assignment Response at HNBGW)

https://gerrit.osmocom.org/c/osmo-mgw/+/35176/comment/f783b56a_3eb116ae PS4, Line 843: MDCX Writing "MDCX" is too specific: it is up to the client to do CRCX with SDP all-in-one or adjust with MDCX later.

I'd rather write:

Hence the MGW may not yet know the remote IuUP address and port at the time of receiving IuUP Initialization from the hNodeB.