GNOME Bugzilla – Bug 628652
Memory leak in recurrence_page_set_dates()
Last modified: 2013-09-14 16:54:16 UTC
==21464== 216 bytes in 4 blocks are definitely lost in loss record 27,646 of 33,527 ==21464== at 0x4A0615D: malloc (vg_replace_malloc.c:195) ==21464== by 0x359A87FD51: strdup (strdup.c:43) ==21464== by 0x3EFB41A64A: icalparameter_set_tzid (icalderivedparameter.c:497) ==21464== by 0x6D97017: e_cal_component_set_dtend (e-cal-component.c:2466) ==21464== by 0xECAB499: recurrence_page_set_dates (recurrence-page.c:1944) ==21464== by 0xECAB874: recurrence_page_fill_widgets (recurrence-page.c:1593) ==21464== by 0xEC99F50: comp_editor_page_fill_widgets (comp-editor-page.c:326) ==21464== by 0xEC97F1D: real_edit_comp (comp-editor.c:2646) ==21464== by 0xEC9D4D9: event_editor_edit_comp (event-editor.c:621) ==21464== by 0xEC2AB55: e_calendar_view_open_event_with_flags (e-calendar-view.c:1550) ==21464== by 0xEC2BBD7: e_calendar_view_new_appointment_for (e-calendar-view.c:1440) ==21464== by 0xEC2C199: e_calendar_view_new_appointment_full (e-calendar-view.c:1511)
Strange, I do not see this one. I suppose you just opened New appointment dialog and closed it? I see a similar one, but it seems to me like a fake, because I added some debug prints in the recurrence page and the component is properly freed, with a ref_count being 1. The other option can be that my libical is pretty old, I have libical-0.43-5 here. ==5626== 108 bytes in 2 blocks are definitely lost in loss record 24,787 of 28,171 ==5626== at 0x4005BDC: malloc (vg_replace_malloc.c:195) ==5626== by 0xA1C51F: strdup (strdup.c:43) ==5626== by 0x466895D: icalmemory_strdup (icalmemory.c:244) ==5626== by 0x4669BCA: icalparameter_new_clone (icalparameter.c:132) ==5626== by 0x466C9A4: icalproperty_new_clone (icalproperty.c:157) ==5626== by 0x4667CBF: icalcomponent_new_clone (icalcomponent.c:199) ==5626== by 0x44D7BC8: e_cal_component_clone (e-cal-component.c:493) ==5626== by 0x57F5946: recurrence_page_fill_widgets (recurrence-page.c:1578) ==5626== by 0x57E06FB: comp_editor_page_fill_widgets (comp-editor-page.c:326) ==5626== by 0x57DE888: fill_widgets (comp-editor.c:2646) ==5626== by 0x57DE9FE: real_edit_comp (comp-editor.c:2677) ==5626== by 0x57E4FB2: event_editor_edit_comp (event-editor.c:621) ==5626== by 0x57DF151: comp_editor_edit_comp (comp-editor.c:2869) ==5626== by 0x5755C77: e_calendar_view_open_event_with_flags (e-calendar-view.c:1550) ==5626== by 0x5755852: e_calendar_view_new_appointment_for (e-calendar-view.c:1440) ==5626== by 0x5755AD0: e_calendar_view_new_appointment_full (e-calendar-view.c:1511) ==5626== by 0x5755B63: e_calendar_view_new_appointment (e-calendar-view.c:1519) ==5626== by 0x5E8FE4E: action_event_new_cb (e-cal-shell-view-actions.c:838)
I don't remember how I did it. I didn't notice at the time -- just found it in the valgrind log later. It's quite possible that I opened the new appointment dialog and then closed it, as you suggest.
any news for this report?
I'm not running evo in valgrind at the moment, so I wouldn't know if it's still happening.
Closing this bug report as no further information has been provided. Please feel free to reopen this bug if you can provide the information asked for. Thanks!