Markus Niessen writes:
> Dear Sirs,
> my name is Markus Niessen. I=B4m working as post doc at the Lehrstuh=
l=20
> Lasertechnik at the RWTH Aachen (Germany). I developed apdaptive FEM=
's=20
> and started to parallelize the code. Studiing the webpage of Pets=
c I=20
> found a link to the libmesh webpage. At a first look I found that a =
lot=20
> the work has already done whithin your brilliant projekt.
I'm glad you found the library. Thanks for stopping by!
> I realy would like to learn more about libmesh. Is there more=20
> documentation availible then the one generated with doxygen?
Not really. There are several example problems accessible from
the libmesh webpage that should help you if you are trying to
develop an application. If you just want to learn more about the
mechanics of the library, a combination of web-based Doxygen
documentation and reading through the actual source files is
probably the way to go. I guess there won't be a comprehensive
Programmers and Developers manual until we are sponsored by a
large corporation to create one ;)
> While trying to build libmesh on Suse 9.1 I encountered a problem wh=
en=20
> trying to enable petsc (petsc version 2.2.1). The make command stopp=
ed=20
> with an error 1:=20
> /usr/lib/gcc-lib/i586-suse-linux/3.3.3/../../../../i586-suse-linux/b=
in/ld:=20
> cannot find -lpetscksp
> collect2: ld returned 1 exit status
> make: *** [bin/meshtool] Error 1
>=20
> Is there any chance to fix this problem?
Sure. What version (CVS or 0.4.3rc2) of libmesh are you attempting to
build, and how was your PETSc installation created? One possibility
is that the version of petsc installed on your computer is newer
(2.2.x) than the version supported in the last release of libmesh. In
that case, try the CVS version of libmesh. On the other hand, it may
be that your PETSC_DIR and PETSC_ARCH variables are not set correctly.
In this case, you will probably have to set them by hand and
reconfigure.
Cheers,
John
|