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/.
dexter gerrit-no-reply at lists.osmocom.orgPatch Set 1: > llist_del Form an empty list is a safe Operation. So is tzhis a > purely cosmetic fix orcdid you See any real world issue? Yes, https://gerrit.osmocom.org/5832 causes a segfault in osmo_fsm_inst_term() when osmo_fsm_inst_unlink_parent() was called before. I see a problem with llist_del() in linuxlist.h. llist_del() overwrites entry->next and entry->prev with LLIST_POISONn values. But before doing this it executes __list_del() which dereferences entry->prev and entry->next. When we run llist_del() a second time we segfault because we try to dereference ethe LLIST_POISONn addresses we assigned to the pointers last time. We could check the addresses in llist_del() but I did not want to change the behaviour of llist_del(), probably there is a good reason why it is built that way. -- To view, visit https://gerrit.osmocom.org/5831 To unsubscribe, visit https://gerrit.osmocom.org/settings Gerrit-MessageType: comment Gerrit-Change-Id: I4b33d508c8a11b72fbf30125088a882894d9e6ac Gerrit-PatchSet: 1 Gerrit-Project: libosmocore Gerrit-Branch: master Gerrit-Owner: dexter <pmaier at sysmocom.de> Gerrit-Reviewer: Harald Welte <laforge at gnumonks.org> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: dexter <pmaier at sysmocom.de> Gerrit-HasComments: No