embedding octave - floating point exception

[John W. Eaton]

On  9-Mar-2009, Marco Driusso wrote:

| I'm embedding the octave interpreter in a c++ program. I convert my c++
| variables to strings to pass them to octave (using the eval_string function
| like explained in the octave_embed.tar.gz example) and I take them back
| writing a file from octave and reading it from c++.
| I have to use the gcc compiler and not the mkoctfile (like is done the
| example) because I'm embedding octave in a big program that has an huge
| Makefile, so what I do is just to modify this makefile adding the needed
| libs.
| Once compiled and linked (no problems doing that), I run my program and just
| after called the octave_main(int argc, char*[]argv,int embedded) function I
| have a floating point exception.
| In the octave manual I read that if octave crashes immediately with a
| floating point exception it could be that octave fail to initialize  the
| IEEE floating point values for infinity and NaN.
| But octave run correctly on my system, I've not such error when I run it
| normally from a shell.
| Does anyone have an idea to solve this problem?

Did you try running your program under a debugger to see precisely
where it is crashing?

jwe