This code solves

   a alpha soln - b div (beta grad soln) = rhs,

where a and b are constants, alpha, beta, rhs, and soln are arrays.

To solve the linear system, it can use either the linear solvers in
BoxLib or the Hypre library at LLNL.  

The directory includes a makefile and a sample inputs file.  

To build the code, you need to specify in the makefile the location
where you have placed BoxLib.  To use Hypre, you need to specify the
location of the Hypre library and set USE_HYPRE to TRUE in the
makefile.

The inputs file contains some adjustable parameters.  See the comments
in the file for details.

Plotfiles can be generated by setting the plot_* parameters in the
inputs file.  They can be viewed with amrvis3d (CCSE's native vis /
spreadsheet tool) or with visit.


================= SPECIAL NOTES FOR THE HPGMG SOLVER =================

If you want to use the HPGMG solver, you must use my (Brian F's) fork of HPGMG.
The base code maintained by Sam Williams does not have hooks for calling HPGMG
as a library, so I had to put them in manually. While the functionality and
performance of HPGMG remains unchanged in the "library" version (data copies
notwithstanding), there nevertheless exists a fair amount of change to the
code. Some of it is now in MultiFab.cpp, and some is in the library fork of
HPGMG. The ugliness of the hacked up library version is due in part to the
notion that HPGMG is a C code which does some very C-ish things which C++
compilers do not like very much. Also, name-mangling between C and C++ symbols
adds to the headache.
