Summary: | Marble widget crashes while starting up [mem2chunk_check, free_check, Marble::GeoDataObject::~GeoDataObject] | ||
---|---|---|---|
Product: | [Applications] marble | Reporter: | Chris <c.lusby> |
Component: | data | Assignee: | marble-bugs |
Status: | RESOLVED UPSTREAM | ||
Severity: | crash | CC: | andresbajotierra, caulier.gilles, nienhueser, ps_ml, sonataarctica, vmikiv, weber-jonathan |
Priority: | NOR | ||
Version: | unspecified | ||
Target Milestone: | --- | ||
Platform: | Ubuntu | ||
OS: | Linux | ||
Latest Commit: | Version Fixed In: | 1.1.0 | |
Sentry Crash Report: |
Description
Chris
2009-12-16 03:28:12 UTC
Look like marble widget crash... Gilles Caulier From bug 220288: --- -- Information about the crash: Hi! Digikam built from today's (2009-12-27) sources crashes at startup, before loading main view. I've tried removing the old database file; the application starts rebuilding the index but never gets to the end of the process. I built Digikam following the instructions on http://www.digikam.org/drupal/download?q=download/KDE4; as for the KDE platform, I'm using the precompiled 4.3.85 packages from the kubuntu-ppa deb repository. *** Bug 220288 has been marked as a duplicate of this bug. *** I wonder if bug 194750 could be somehow related... One thing that is broken in GeoDataPlacemark is the constructor which takes a QString as parameter: It forgets to initialize the d-pointer, which means that the static_cast later in p() operates on a different object than it should. This got lost in http://websvn.kde.org/trunk/KDE/kdeedu/marble/src/lib/geodata/data/GeoDataPlacemark.cpp?r1=946665&r2=946666& As far as I can tell, this version of the constructor is nowhere used in Marble however, so it shouldn't cause problems. I'd say the d-pointer should be initialized anyway however or the constructor be removed. @Patrick: Can you look at that? *** Bug 221050 has been marked as a duplicate of this bug. *** r1069978 | nienhueser | 2010-01-04 17:04:17 +0100 (Mon, 04 Jan 2010) | 2 lines Initialize the d-pointer with the correct class so that the static_cast later operates on the right base. Probably this has no effect on fixing this bug though. It's likely caused by a glibc bug [1] which should be fixed in the current Ubuntu packages. See also [2]. Can you please make sure that you're running a fully updated system and see if the problem has gone away? [1] http://sourceware.org/bugzilla/show_bug.cgi?id=10282#c4 [2] https://bugs.launchpad.net/ubuntu/+source/kdevelop/+bug/425723 I upgraded to 4.4 SC RC1 - and it seems fine now. From my perspective this bug can be closed. Thanks. Thanks for the feedback :-) *** Bug 194750 has been marked as a duplicate of this bug. *** |