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/.
Vadim Yanitskiy gerrit-no-reply at lists.osmocom.orgVadim Yanitskiy has posted comments on this change. ( https://gerrit.osmocom.org/12556 ) Change subject: Add tests for transaction routines ...................................................................... Patch Set 9: (3 comments) https://gerrit.osmocom.org/#/c/12556/9/tests/trans/trans_test.c File tests/trans/trans_test.c: https://gerrit.osmocom.org/#/c/12556/9/tests/trans/trans_test.c@68 PS9, Line 68: base_callref > What's the point of using static counter? What would be the advantage compared to explicitly passing […] We would avoid one meaningless parameter ;) As a bonus, you could reflect values of this static counter in the logging output, instead of counting the amount of allocated transactions manually. https://gerrit.osmocom.org/#/c/12556/9/tests/trans/trans_test.c@94 PS9, Line 94: vlr_subscr_name > I don't think we have function to print ran-conn itself and there's 1:1 mapping between VLR subscrib […] Ok, I see. What about this? LOG_RAN_CONN_CAT(t->conn, DCC, LOGL_DEBUG, "Connection assigned to transaction\n"); Off-topic: Now I think that we need something like LOG_TRANS and LOG_TRANS_CAT() for transactions. https://gerrit.osmocom.org/#/c/12556/9/tests/trans/trans_test.c@134 PS9, Line 134: base_callref >> The key idea is that allocating multiple transactions with >> same callref is wrong, and we shouldn't do this at least >> in tests. > > I strongly disagree. We should never try to "fix" issues by > making tests behave differently from the code. Actually, your test case behaves differently from the code. You won't see any existing code parts, which *intentionally use same callref* for new transactions. Despite potentially this may happen (e.g. triggered by external MNCC / EUSE). So, I don't ask you to fix trans_alloc(), I ask you not to use the same callref value for multiple transactions. > The right way is to add test which illustrate current code > behavior (even if it's wrong), than fix the code and update > the test accordingly. Yes, and this is what I would like to see in this (or in a subsequent) change. Please illustrate the current behavior (e.g. in a separate test_trans_callref function) by allocating a few transactions with same callref, and print some warning until trans_alloc() is fixed. -- To view, visit https://gerrit.osmocom.org/12556 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-msc Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I78dfb7cd35073a305cf668beda7d9d58d5a5a713 Gerrit-Change-Number: 12556 Gerrit-PatchSet: 9 Gerrit-Owner: Max <msuraev at sysmocom.de> Gerrit-Reviewer: Harald Welte <laforge at gnumonks.org> Gerrit-Reviewer: Jenkins Builder (1000002) Gerrit-Reviewer: Max <msuraev at sysmocom.de> Gerrit-Reviewer: Neels Hofmeyr <nhofmeyr at sysmocom.de> Gerrit-Reviewer: Vadim Yanitskiy <axilirator at gmail.com> Gerrit-Comment-Date: Tue, 15 Jan 2019 12:10:41 +0000 Gerrit-HasComments: Yes Gerrit-HasLabels: No -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.osmocom.org/pipermail/gerrit-log/attachments/20190115/0c7f3ec8/attachment.htm>