Attention is currently required from: neels, laforge, fixeria, msuraev. pespin has posted comments on this change. ( https://gerrit.osmocom.org/c/libosmo-sccp/+/29084 )

Change subject: SIGTRAN: cache Optional Data for SCCP CR/CC/RLSD ......................................................................

Patch Set 15:

(7 comments)

File src/sccp_scoc.c:

https://gerrit.osmocom.org/c/libosmo-sccp/+/29084/comment/99187a07_486bf644 PS15, Line 642: /* optional: importance */ why is this line removed here? this is non related.

https://gerrit.osmocom.org/c/libosmo-sccp/+/29084/comment/22584621_ed05b35e PS15, Line 600: if (

in whihc situation would we have a zero-length opt_data_cache?

Ack

https://gerrit.osmocom.org/c/libosmo-sccp/+/29084/comment/95f59ccb_f6e0401e PS15, Line 606: } else remove curly braces in the if.

https://gerrit.osmocom.org/c/libosmo-sccp/+/29084/comment/76465829_27cd148a PS15, Line 630: if (xua_drop_data_check(prim, SCCP_MAX_DATA, "cache overrun")) ...check_drop(), otherwise it's not clear what you are checking and it's confusing since you actually check for drop and return the opposite to keep it. Actually, since you call this function _keep(), it would make sense to reverse the loginc on the xua_drop_data_check() function... I would even drop the function completely tbh.

https://gerrit.osmocom.org/c/libosmo-sccp/+/29084/comment/1e4a2119_e4a3eef7 PS15, Line 635: LOGP(DLSCCP, LOGL_ERROR, "replacing unsent %u bytes of optional data cache with %s optional data\n", NOTICE

https://gerrit.osmocom.org/c/libosmo-sccp/+/29084/comment/fba00b9f_9b6214fd PS15, Line 639: msgb_alloc(SCCP_MAX_DATA, "SCCP optional data cache for CR/CC/RLSD");

If we ever should run into memory leaks, it would be better to use msgb_alloc_c(conn) to attach this […]

Ack

https://gerrit.osmocom.org/c/libosmo-sccp/+/29084/comment/d53cddbe_6bbb5c85 PS15, Line 784: xua_msg_add_sccp_addr(xua, SUA_IEI_DEST_ADDR, &conn->calling_addr); why is this line removed here? this is non related.