On 30 Oct 2008, at 18:22, James Turner wrote: > Having some weirdness here - using OSG in Debug mode, it appends a > 'd' suffix to the library and plugin names. Fair enough, except at > runtime, it's failing to load any of the plugins, since it's looking > for names without the 'd'. Eg 'osgdb_qt.so' where it needs to use > 'osgdb_qtd.so'.
Hacked around this issue for now. So, I can confirm that disabling hardware mip-map generation does fix my initial issue on the 7300 GT, which is great news. Unfortunately, my secondary issue is still present: #0 0x94e31401 in std::basic_string<char, std::char_traits<char>, std::allocator<char> >::basic_string () #1 0x016d5269 in osg::Object::Object (this=0x28b31290, [EMAIL PROTECTED], [EMAIL PROTECTED]) at /Users/jmt/Code/OSG/OpenSceneGraph/src/osg/ Object.cpp:22 #2 0x0214909b in osgDB::ReaderWriter::Options::Options (this=0x28b31290, [EMAIL PROTECTED], [EMAIL PROTECTED]) at ReaderWriter:127 #3 0x021e979e in SGReaderWriterBTGOptions::SGReaderWriterBTGOptions (this=0x28b31290, [EMAIL PROTECTED], [EMAIL PROTECTED]) at SGReaderWriterBTGOptions.hxx:36 #4 0x021e71d1 in simgear::TileEntry::loadTileByName ([EMAIL PROTECTED], options=0x24a6c7f0) at /Users/jmt/Code/FGFS/ SimGear/source/simgear/scene/tgdb/TileEntry.cxx:390 #5 0x021efc61 in simgear::ReaderWriterSTG::readNode (this=0x271f860, [EMAIL PROTECTED], options=0x24a6c7f0) at /Users/jmt/Code/FGFS/ SimGear/source/simgear/scene/tgdb/ReaderWriterSTG.cxx:71 #6 0x021f7396 in simgear::ModelRegistryCallback<simgear::DefaultProcessPolicy, simgear::DefaultCachePolicy, simgear::OptimizeModelPolicy, simgear::DefaultCopyPolicy, simgear::OSGSubstitutePolicy>::loadUsingReaderWriter ([EMAIL PROTECTED], opt=0x24a6c7f0) at ModelRegistry.hxx:115 #7 0x021f9a2c in simgear::ModelRegistryCallback<simgear::DefaultProcessPolicy, simgear::DefaultCachePolicy, simgear::OptimizeModelPolicy, simgear::DefaultCopyPolicy, simgear::OSGSubstitutePolicy>::readNode (this=0x271fdb0, [EMAIL PROTECTED], opt=0x24a6c7f0) at ModelRegistry.hxx:91 #8 0x021f159b in simgear::ModelRegistry::readNode (this=0x271fd50, [EMAIL PROTECTED], opt=0x24a6c7f0) at /Users/jmt/Code/FGFS/SimGear/ source/simgear/scene/model/ModelRegistry.cxx:505 #9 0x00bd8bf8 in osgDB::Registry::readNode (this=0x271cda0, [EMAIL PROTECTED], options=0x24a6c7f0, buildKdTreeIfRequired=false) at Registry:237 #10 0x00bd0580 in osgDB::DatabasePager::DatabaseThread::run (this=0x27dd690) at /Users/jmt/Code/OSG/OpenSceneGraph/src/osgDB/ DatabasePager.cpp:539 #11 0x01be6edc in OpenThreads::ThreadPrivateActions::StartThread (data=0x27dd69c) at /Users/jmt/Code/OSG/OpenSceneGraph/src/OpenThreads/ pthreads/PThread.c++:170 #12 0x923fa6f5 in _pthread_start () What happens (based on earlier testing), is that the dynamic cast is failing (returning null): SGReaderWriterBTGOptions *opt = new SGReaderWriterBTGOptions(*dynamic_cast<const SGReaderWriterBTGOptions *>(options)); Last time I checked, I logged the pointer address of the SGReaderWriterBTGOptions object, and it was sane, but somehow the dynamic cast failed. I hoped this was the GCC 4.0 bug that Tat has previously mentioned, but I'm using GCC 4.2, and it made no change. This is why I think there's some memory corruption going - something is trashing the RTTI data associated with the object, is my feeling. Ack, James ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Flightgear-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/flightgear-devel
