//                                MFEM Example 1
//                             Caliper Modification
//
// Compile with: make ex1
//
// Sample runs:  ex1 -m ../data/square-disc.mesh
//               ex1 -m ../data/star.mesh
//               ex1 -m ../data/star-mixed.mesh
//               ex1 -m ../data/escher.mesh
//               ex1 -m ../data/fichera.mesh
//               ex1 -m ../data/fichera-mixed.mesh
//               ex1 -m ../data/toroid-wedge.mesh
//               ex1 -m ../data/periodic-annulus-sector.msh
//               ex1 -m ../data/periodic-torus-sector.msh
//               ex1 -m ../data/square-disc-p2.vtk -o 2
//               ex1 -m ../data/square-disc-p3.mesh -o 3
//               ex1 -m ../data/square-disc-nurbs.mesh -o -1
//               ex1 -m ../data/star-mixed-p2.mesh -o 2
//               ex1 -m ../data/disc-nurbs.mesh -o -1
//               ex1 -m ../data/pipe-nurbs.mesh -o -1
//               ex1 -m ../data/fichera-mixed-p2.mesh -o 2
//               ex1 -m ../data/star-surf.mesh
//               ex1 -m ../data/square-disc-surf.mesh
//               ex1 -m ../data/inline-segment.mesh
//               ex1 -m ../data/amr-quad.mesh
//               ex1 -m ../data/amr-hex.mesh
//               ex1 -m ../data/fichera-amr.mesh
//               ex1 -m ../data/mobius-strip.mesh
//               ex1 -m ../data/mobius-strip.mesh -o -1 -sc
//
// Device sample runs:
//               ex1 -pa -d cuda
//               ex1 -pa -d raja-cuda
//               ex1 -pa -d occa-cuda
//               ex1 -pa -d raja-omp
//               ex1 -pa -d occa-omp
//               ex1 -pa -d ceed-cpu
//             * ex1 -pa -d ceed-cuda
//               ex1 -pa -d ceed-cuda:/gpu/cuda/shared
//               ex1 -m ../data/beam-hex.mesh -pa -d cuda
//               ex1 -m ../data/beam-tet.mesh -pa -d ceed-cpu
//               ex1 -m ../data/beam-tet.mesh -pa -d ceed-cuda:/gpu/cuda/ref
//
// Description: This example is a copy of Example 1 instrumented with the
//              Caliper performance profilinh library. Any option supported by
//              the Caliper ConfigManager can be passed to the code using a
//              configuration string after -p or --caliper flag. For more
//              information, see the Caliper documentation.
//
// Examples: ex1 --caliper runtime-report
//           ex1 --caliper runtime-report,mem.highwatermark
//
// The first run will return the default report. The second run will also output
// the memory high-water mark and time spent in MPI routines.

#include "mfem.hpp"
#include <fstream>
#include <iostream>

using namespace std;
using namespace mfem;

int main(int argc, char *argv[])
{
   // Define Caliper ConfigManager
   cali::ConfigManager mgr;
   // Caliper instrumentation
   MFEM_PERF_FUNCTION;

   // 1. Parse command-line options.
   const char *mesh_file = "../../data/star.mesh";
   int order = 1;
   bool static_cond = false;
   bool pa = false;
   const char *device_config = "cpu";
   bool visualization = true;
   const char* cali_config = "runtime-report";

   OptionsParser args(argc, argv);
   args.AddOption(&mesh_file, "-m", "--mesh",
                  "Mesh file to use.");
   args.AddOption(&order, "-o", "--order",
                  "Finite element order (polynomial degree) or -1 for"
                  " isoparametric space.");
   args.AddOption(&static_cond, "-sc", "--static-condensation", "-no-sc",
                  "--no-static-condensation", "Enable static condensation.");
   args.AddOption(&pa, "-pa", "--partial-assembly", "-no-pa",
                  "--no-partial-assembly", "Enable Partial Assembly.");
   args.AddOption(&device_config, "-d", "--device",
                  "Device configuration string, see Device::Configure().");
   args.AddOption(&visualization, "-vis", "--visualization", "-no-vis",
                  "--no-visualization",
                  "Enable or disable GLVis visualization.");
   args.AddOption(&cali_config, "-p", "--caliper",
                  "Caliper configuration string.");

   args.Parse();
   if (!args.Good())
   {
      args.PrintUsage(cout);
      return 1;
   }
   args.PrintOptions(cout);

   // 2. Enable hardware devices such as GPUs, and programming models such as
   //    CUDA, OCCA, RAJA and OpenMP based on command line options.
   Device device(device_config);
   device.Print();

   // Caliper configuration
   mgr.add(cali_config);
   mgr.start();

   // 3. Read the mesh from the given mesh file. We can handle triangular,
   //    quadrilateral, tetrahedral, hexahedral, surface and volume meshes with
   //    the same code.
   Mesh mesh(mesh_file, 1, 1);
   int dim = mesh.Dimension();

   // 4. Refine the mesh to increase the resolution. In this example we do
   //    'ref_levels' of uniform refinement. We choose 'ref_levels' to be the
   //    largest number that gives a final mesh with no more than 50,000
   //    elements.
   {
      int ref_levels =
         (int)floor(log(50000./mesh.GetNE())/log(2.)/dim);
      for (int l = 0; l < ref_levels; l++)
      {
         mesh.UniformRefinement();
      }
   }

   // 5. Define a finite element space on the mesh. Here we use continuous
   //    Lagrange finite elements of the specified order. If order < 1, we
   //    instead use an isoparametric/isogeometric space.
   FiniteElementCollection *fec;
   bool delete_fec;
   if (order > 0)
   {
      fec = new H1_FECollection(order, dim);
      delete_fec = true;
   }
   else if (mesh.GetNodes())
   {
      fec = mesh.GetNodes()->OwnFEC();
      delete_fec = false;
      cout << "Using isoparametric FEs: " << fec->Name() << endl;
   }
   else
   {
      fec = new H1_FECollection(order = 1, dim);
      delete_fec = true;
   }
   FiniteElementSpace fespace(&mesh, fec);
   cout << "Number of finite element unknowns: "
        << fespace.GetTrueVSize() << endl;

   // 6. Determine the list of true (i.e. conforming) essential boundary dofs.
   //    In this example, the boundary conditions are defined by marking all
   //    the boundary attributes from the mesh as essential (Dirichlet) and
   //    converting them to a list of true dofs.
   Array<int> ess_tdof_list;
   if (mesh.bdr_attributes.Size())
   {
      Array<int> ess_bdr(mesh.bdr_attributes.Max());
      ess_bdr = 1;
      fespace.GetEssentialTrueDofs(ess_bdr, ess_tdof_list);
   }

   // 7. Set up the linear form b(.) which corresponds to the right-hand side of
   //    the FEM linear system, which in this case is (1,phi_i) where phi_i are
   //    the basis functions in the finite element fespace.
   MFEM_PERF_BEGIN("Set up the linear form");
   LinearForm b(&fespace);
   ConstantCoefficient one(1.0);
   b.AddDomainIntegrator(new DomainLFIntegrator(one));
   b.Assemble();
   MFEM_PERF_END("Set up the linear form");

   // 8. Define the solution vector x as a finite element grid function
   //    corresponding to fespace. Initialize x with initial guess of zero,
   //    which satisfies the boundary conditions.
   GridFunction x(&fespace);
   x = 0.0;

   // 9. Set up the bilinear form a(.,.) on the finite element space
   //    corresponding to the Laplacian operator -Delta, by adding the Diffusion
   //    domain integrator.
   MFEM_PERF_BEGIN("Set up the bilinear form");
   BilinearForm a(&fespace);
   if (pa) { a.SetAssemblyLevel(AssemblyLevel::PARTIAL); }
   a.AddDomainIntegrator(new DiffusionIntegrator(one));

   // 10. Assemble the bilinear form and the corresponding linear system,
   //     applying any necessary transformations such as: eliminating boundary
   //     conditions, applying conforming constraints for non-conforming AMR,
   //     static condensation, etc.
   if (static_cond) { a.EnableStaticCondensation(); }
   a.Assemble();

   OperatorPtr A;
   Vector B, X;
   a.FormLinearSystem(ess_tdof_list, x, b, A, X, B);
   MFEM_PERF_END("Set up the bilinear form");

   cout << "Size of linear system: " << A->Height() << endl;

   // 11. Solve the linear system A X = B.
   if (!pa)
   {
      MFEM_PERF_SCOPE("Solve A X=B (FA)");
#ifndef MFEM_USE_SUITESPARSE
      // Use a simple symmetric Gauss-Seidel preconditioner with PCG.
      GSSmoother M((SparseMatrix&)(*A));
      PCG(*A, M, B, X, 1, 200, 1e-12, 0.0);
#else
      // If MFEM was compiled with SuiteSparse, use UMFPACK to solve the system.
      UMFPackSolver umf_solver;
      umf_solver.Control[UMFPACK_ORDERING] = UMFPACK_ORDERING_METIS;
      umf_solver.SetOperator(*A);
      umf_solver.Mult(B, X);
#endif
   }
   else // Jacobi preconditioning in partial assembly mode
   {
      MFEM_PERF_SCOPE("Solve A X=B (PA)");
      if (UsesTensorBasis(fespace))
      {
         OperatorJacobiSmoother M(a, ess_tdof_list);
         PCG(*A, M, B, X, 1, 400, 1e-12, 0.0);
      }
      else
      {
         CG(*A, B, X, 1, 400, 1e-12, 0.0);
      }
   }
   // 12. Recover the solution as a finite element grid function.
   a.RecoverFEMSolution(X, b, x);

   // 13. Save the refined mesh and the solution. This output can be viewed later
   //     using GLVis: "glvis -m refined.mesh -g sol.gf".
   MFEM_PERF_BEGIN("Save the results");
   ofstream mesh_ofs("refined.mesh");
   mesh_ofs.precision(8);
   mesh.Print(mesh_ofs);
   ofstream sol_ofs("sol.gf");
   sol_ofs.precision(8);
   x.Save(sol_ofs);
   MFEM_PERF_END("Save the results");
   // 14. Send the solution by socket to a GLVis server.
   if (visualization)
   {
      char vishost[] = "localhost";
      int  visport   = 19916;
      socketstream sol_sock(vishost, visport);
      sol_sock.precision(8);
      sol_sock << "solution\n" << mesh << x << flush;
   }

   // 15. Free the used memory.
   if (delete_fec)
   {
      delete fec;
   }

   // Flush output
   mgr.flush();

   return 0;
}