rmhd_riemann_solver.tgz contains all the source files (written in
fortran 90), a Makefile, a README file with some basic instructions
and a file named RInput.txt that can be used to pass to the code
different initial conditions.

The code is able to reproduce without any problem all the results
showed in Giacomazzo and Rezzolla 2006 with the accuracy we used (the
exact value for the test number 5 of Balsara is 3.3e-4).

If you want to use the solver to find the exact solution to other Riemann 
problems then you have to distinguish between two different cases: one in 
which Bx=0 and one in which Bx/=0.

In the first case the code should work without any problem and you should 
be able to use different initial conditions modifying the file RInput.txt

In the latter (Bx different from zero) you should provide an approximate 
solution, we used HLLE with 800 points to get it for our tests, and it 
could be also necessary to modify the code, especially if slow shocks are 
present. This is due to the fact that the numerical algorithm implemented 
in the code could not be able to find a solution for the slow shock 
velocity and for the post-shock value of the x component of the velocity 
if it starts from a value which is too far from the exact one. I will work 
on this point in the future when I'll have some time. The same can happen 
with Alfven discontinuities; some details on these points can be found in 
the README file.

If you are not able to find a solution you are interested in, tell me
and I will be happy to help you. You can also find here the exact
solutions used for the figures in our paper: KO_exact_solutions.tgz
contains the exact solutions of the tests by Komissarov while in
exact_solutions.tgz you can find the data for the tests by Balsara.

I would also ask you to cite our paper: J. Fluid Mech. 562(2006),
223-259.

Note also that the exact solver can work with two equations of state:
the usual ideal fluid EOS that we used in our paper and a Synge type
EOS (see appendix of Meliani et al 2008, A&A 491, 321 if
interested). The code asks which one you want to use. In case you are
interested in using the Synge type EOS then please cite also Meliani
et al 2008, A&A 491, 321.