neels has posted comments on this change. ( https://gerrit.osmocom.org/c/libosmocore/+/16030 )
Change subject: osmo_sockaddr_str: fix 32bit addr mixup of host/network byte order ...................................................................... Patch Set 2: (1 comment) https://gerrit.osmocom.org/c/libosmocore/+/16030/2/src/sockaddr_str.c File src/sockaddr_str.c: https://gerrit.osmocom.org/c/libosmocore/+/16030/2/src/sockaddr_str.c@409 PS2, Line 409: /*! Convert osmo_sockaddr_str address string to IPv4 address data in host-byte-order. > The API (osmo_sockaddr_str_to_32n) was already defined correctly, since it > states it converted from […] no, that is not an option. osmo_sockaddr_str_to_32n() is already being used in programs. Regardless of what the API doc said and what I was thinking when using that function, if we modify the effect of this function, then we break those other programs. They are using host-byte-order and they need host-byte-order, and are calling osmo_sockaddr_str_{to,from}_32n() to obtain that. They should move to *_32h() whenever ready, we cannot change the behavior of this function without renaming. osmo_sockaddr_str_to_32() already does network byte order correctly. There we can just fix the API doc, because it remains ABI compatible. Actually almost no-one anywhere needs host byte order, except our MNCC socket it seems. I wish I had never added _32n()... but now that it's there, I followed through with the _32h() version. (should be just deprecate without replacement instead?) -- To view, visit https://gerrit.osmocom.org/c/libosmocore/+/16030 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: libosmocore Gerrit-Branch: master Gerrit-Change-Id: I3cf150cc0cc06dd36039fbde091bc71b01697322 Gerrit-Change-Number: 16030 Gerrit-PatchSet: 2 Gerrit-Owner: neels <nhofm...@sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: neels <nhofm...@sysmocom.de> Gerrit-Reviewer: osmith <osm...@sysmocom.de> Gerrit-Reviewer: pespin <pes...@sysmocom.de> Gerrit-Comment-Date: Wed, 13 Nov 2019 15:45:50 +0000 Gerrit-HasComments: Yes Gerrit-Has-Labels: No Comment-In-Reply-To: pespin <pes...@sysmocom.de> Gerrit-MessageType: comment
