On Mon, 5 Oct 2009, Petry Stefan wrote: > I recompiled libmesh with the compiler flags "-DDEBUG > -D_GLIBCXX_DEBUG_PEDANTIC" and linked it to our program. > Unfortunately, I had to skip the flag "-D_GLIBCXX_DEBUG", because > linker errors occured, when I tried to link a libmesh dll compiled > with this flag to our program. But If i understood your remarks > correctly the most important switch for debug assertions is "DDEBUG > ". However, rerunning our program with this debug version of libmesh > produced no assertions.
Turning on DEBUG will activate the libMesh assertions. It won't catch array bounds errors, though. It does change the libstdc++ ABI, so to use it you'd have to recompile your own program with the same flag on. I think that's worth a try. There's not much else we can do for you unless you can create a test case that reproduces the problem and is small enough to post. > Concerning our use of libmesh on Windows I have to explain that we > only employ a very limited subset of libmesh's capabilities. > Especially we are only using a single processor. We were not able to > port a parallel-processing version of libmesh to Windows. For our 2d > application we can live with such a restriction, but I think in > general this will be a knock-out criterion. The tools that we use > are mingw and gcc. How far does the port attempt get in parallel? If PETSc works but libMesh doesn't, it shouldn't take too much effort to go the rest of the way. --- Roy ------------------------------------------------------------------------------ Come build with us! The BlackBerry® Developer Conference in SF, CA is the only developer event you need to attend this year. Jumpstart your developing skills, take BlackBerry mobile applications to market and stay ahead of the curve. Join us from November 9-12, 2009. Register now! http://p.sf.net/sfu/devconf _______________________________________________ Libmesh-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/libmesh-users
