Attention is currently required from: fixeria, neels, nt2mku.
falconia has posted comments on this change. ( https://gerrit.osmocom.org/c/libosmocore/+/36784?usp=email )
Change subject: gsm48_encode_bearer_cap(): omit octet 3a if only GSM-FR/GSM-HR v1 is supported ......................................................................
Patch Set 7:
(3 comments)
Commit Message:
https://gerrit.osmocom.org/c/libosmocore/+/36784/comment/0c07b980_536d49fc PS7, Line 9: Some early GSM phones (like the Siemens P1 Porty) do not accept a mobile-terminating call setup if octet 3a is present. The logic of the patch is correct, as we now know thanks to @vyanitskiy@sysmocom.de, but this commit message needs to be changed. We should not be talking about putting this bearer cap IE in the MT SETUP message, as doing so is wrong in itself, instead we should be focusing on MS-side implementations using this libosmocore function to construct outgoing bearer cap IE in MO SETUP messages, or in CALL CONFIRM responses to MT calls - but still encoding in MS implementations, not by networks.
File src/gsm/gsm48_ie.c:
https://gerrit.osmocom.org/c/libosmocore/+/36784/comment/8cba7a2b_b52d98dc PS7, Line 322: int encode_3a = 0; Make this bool instead of int?
https://gerrit.osmocom.org/c/libosmocore/+/36784/comment/0749540a_d484d8cc PS7, Line 340: encode_3a == 1 If encode_3a is changed to bool, then this check should take the bool directly, without == operator.