This is merely a historical archive of years 2008-2021, before the migration to mailman3.
A maintained and still updated list archive can be found at https://lists.osmocom.org/hyperkitty/list/gerrit-log@lists.osmocom.org/.
daniel gerrit-no-reply at lists.osmocom.orgdaniel has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-sgsn/+/15487 ) Change subject: gprs_gmm_fsm.c: Implement RAT change between 2g and 3g ...................................................................... Patch Set 12: (6 comments) https://gerrit.osmocom.org/c/osmo-sgsn/+/15487/11/src/sgsn/gprs_gmm.c File src/sgsn/gprs_gmm.c: https://gerrit.osmocom.org/c/osmo-sgsn/+/15487/11/src/sgsn/gprs_gmm.c@1232 PS11, Line 1232: if (MSG_IU_UE_CTX(msg) && ctx->ran_type != MM_CTX_T_UTRAN_Iu) > Would be nice adding a comment here like: /* Detect MS switched to new a Radio Access technology */ Done https://gerrit.osmocom.org/c/osmo-sgsn/+/15487/11/src/sgsn/gprs_gmm.c@1644 PS11, Line 1644: if (MSG_IU_UE_CTX(msg) && mmctx->ran_type != MM_CTX_T_UTRAN_Iu) > The problem overall is, we have to refactor the full Attach Request/RAU logic. […] Thanks for the clarification. Refactoring the logic belongs in a different patch(set), let's keep this for handling RAT changes and fixing the SGSN crash. https://gerrit.osmocom.org/c/osmo-sgsn/+/15487/11/src/sgsn/gprs_gmm.c@1649 PS11, Line 1649: osmo_fsm_inst_dispatch(mmctx->gmm_fsm, E_GMM_COMMON_PROC_INIT_REQ, NULL); > So we need to find out before merging this patch blindly. Seems Lynxis moved the call here from line 1607 so we don't accidentally send messages (and crash) before handling the RAT change. I re-added the call to the code above with a check whether the RAT has changed. https://gerrit.osmocom.org/c/osmo-sgsn/+/15487/11/src/sgsn/gprs_gmm.c@1663 PS11, Line 1663: if (!mmctx->gb.llme) > Agree, you probably want to pass new_rat_type, llme and msg. Actually the IMPLICIT_DETACH event causes the gb fsm to to to idle which in turn unassigns the llme and sets it to NULL. However, implicit detach is only handled from the mm_ready state and not from mm_standby. I think we should implement add that to the the gb fsm so we don't need to do this here or in the gmm_fsm. https://gerrit.osmocom.org/c/osmo-sgsn/+/15487/9/src/sgsn/gprs_gmm_fsm.c File src/sgsn/gprs_gmm_fsm.c: https://gerrit.osmocom.org/c/osmo-sgsn/+/15487/9/src/sgsn/gprs_gmm_fsm.c@173 PS9, Line 173: (enum sgsn_ran_type) data; > I am pretty sure Coverity is not gonna like this this hack... When we pass a struct * as user data this will be resolved. https://gerrit.osmocom.org/c/osmo-sgsn/+/15487/11/src/sgsn/gprs_gmm_fsm.c File src/sgsn/gprs_gmm_fsm.c: https://gerrit.osmocom.org/c/osmo-sgsn/+/15487/11/src/sgsn/gprs_gmm_fsm.c@165 PS11, Line 165: gmm_fsm_state_chg(fi, ST_GMM_DEREGISTERED); > I'm wondering whether the mmctx ran type needs to be set here too, or some stuff needs to be dealloc […] It looks to me like we should just have a fall-through to the default: case in addition to setting the state to deregistered. There is nothing allocated in this fsm anyway it looks like, so I don't think there's anything to free? -- To view, visit https://gerrit.osmocom.org/c/osmo-sgsn/+/15487 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-sgsn Gerrit-Branch: master Gerrit-Change-Id: I3fc614da6ba137e871ee0fe86ca22b6a4a354dd2 Gerrit-Change-Number: 15487 Gerrit-PatchSet: 12 Gerrit-Owner: lynxis lazus <lynxis at fe80.eu> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: daniel <dwillmann at sysmocom.de> Gerrit-Reviewer: lynxis lazus <lynxis at fe80.eu> Gerrit-CC: fixeria <axilirator at gmail.com> Gerrit-CC: laforge <laforge at osmocom.org> Gerrit-CC: pespin <pespin at sysmocom.de> Gerrit-Comment-Date: Thu, 30 Jan 2020 17:00:25 +0000 Gerrit-HasComments: Yes Gerrit-Has-Labels: No Comment-In-Reply-To: lynxis lazus <lynxis at fe80.eu> Comment-In-Reply-To: pespin <pespin at sysmocom.de> Comment-In-Reply-To: fixeria <axilirator at gmail.com> Comment-In-Reply-To: daniel <dwillmann at sysmocom.de> Gerrit-MessageType: comment -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.osmocom.org/pipermail/gerrit-log/attachments/20200130/e186e46a/attachment.htm>