On Sat, 2007-02-03 at 10:32 +0100, Frank Barknecht wrote:
First, make sure you're not really delete()ing a NULL pointer ;-)
That
said, I do think we've stumbled onto the same bug, and that there's something going horribly wrong (possibly because of mis-use and/or misunderstanding of flext's conventions) particularly when flext
objects
are not self-contained, and call functions from 3rd-party shared libraries (e.g. fluid~ --> libfluid; readanysf~ --> libogg,
libvorbis,
libmad, libflac, etc.; your resoncomb~ --> libSndObj).
Hm, maybe fluid~ should go back to being a pure C/Pd external like it was in its prevous life as iiwu~? Max-users didn't seem interested in it (or just not willing to compile) anyways, one even wrote a new, pure C Max external for fluidsynth instead of compiling the flext-fluid~. So the cross-platformness I was hoping to achieve didn't work out in practice.
the overloading of the memory allocation functions in flext seems to be only a problem on linux (iirc, thomas could never really reproduce it on osx/windows). i've had the same problems one or two years ago, and for me i fixed it, by writing my own custom scons build system for flext/flexternals, that enforces same preprocessor definitons on _all_ externals and ensures the rebuilding of objects, in case of header file changes.
cheers ... tim
-- tim@klingt.org ICQ: 96771783 http://www.mokabar.tk
Avoid the world, it's just a lot of dust and drag and means nothing in the end. Jack Kerouac