KokkosSparse::spadd_symbolic

Defined in header KokkosSparse_spadd.hpp

template <typename ExecSpace, typename KernelHandle, typename alno_row_view_t_,
          typename alno_nnz_view_t_, typename blno_row_view_t_,
          typename blno_nnz_view_t_, typename clno_row_view_t_>
void spadd_symbolic(const ExecSpace &exec, KernelHandle *handle,
                    typename KernelHandle::const_nnz_lno_t m,  //same type as column indices
                    typename KernelHandle::const_nnz_lno_t n,
                    const alno_row_view_t_ a_rowmap,
                    const alno_nnz_view_t_ a_entries, const blno_row_view_t_ b_rowmap,
                    const blno_nnz_view_t_ b_entries, clno_row_view_t_ c_rowmap)

template <typename KernelHandle, typename... Args>
void spadd_symbolic(KernelHandle *handle, Args... args)

template <typename ExecSpace, typename KernelHandle, typename AMatrix, typename BMatrix,
          typename CMatrix>
void spadd_symbolic(const ExecSpace &exec, KernelHandle *handle, const AMatrix &A,
                    const BMatrix &B, CMatrix &C);

template <typename KernelHandle, typename AMatrix, typename BMatrix, typename CMatrix>
void spadd_symbolic(KernelHandle *handle, const AMatrix &A, const BMatrix &B, CMatrix &C);

Performs the symbolic phase of the addition of two sparse matrices.

\[C = \beta*C + \alpha*(A+B)\]
  1. Computes the row map of C using the resources of exec and store the total number of non-zeros of C in the handle.

  2. Same as 1. but use the resources of KernelHandle::HandleExecSpace{}.

  3. Compute the row map of C, create and allocate views to hold column indices and values of C without initializing them using the resources of exec. Construct matrix C from the row map and the newly created column indices and values views.

  4. Same as 3. but use the resources of KernelHandle::HandleExecSpace{}.

Parameters

exec:

execution space instance.

handle:

spadd kernels handle obtained from an instance of KokkosKernels::KokkosKernelsHandle.

m, n:

number of rows and column of the matrices A, B and C.

a_rowmap, b_rowmap:

row maps of the input matrices A and B.

a_entries, b_entries:

column indices of the entries in each row of A and B.

c_rowmap:

row map of the output matrix C.

A, B, C:

three crs matrices.

Type Requirements

Checks are performed in the KokkosKernelsHandle that was used to obtain the spadd handle.

Example

#include "Kokkos_Core.hpp"

#include "KokkosKernels_default_types.hpp"
#include "KokkosSparse_spadd.hpp"

#include "KokkosKernels_Test_Structured_Matrix.hpp"

using Scalar  = default_scalar;
using Ordinal = default_lno_t;
using Offset  = default_size_type;
using Layout  = default_layout;

int main(int argc, char* argv[]) {
  Kokkos::initialize();

  using device_type = typename Kokkos::Device<
      Kokkos::DefaultExecutionSpace,
      typename Kokkos::DefaultExecutionSpace::memory_space>;
  using execution_space = typename device_type::execution_space;
  using memory_space    = typename device_type::memory_space;
  using matrix_type =
      typename KokkosSparse::CrsMatrix<Scalar, Ordinal, device_type, void,
                                     Offset>;

  int return_value = 0;

  {
    // The mat_structure view is used to generate a matrix using
    // finite difference (FD) or finite element (FE) discretization
    // on a cartesian grid.
    // Each row corresponds to an axis (x, y and z)
    // In each row the first entry is the number of grid point in
    // that direction, the second and third entries are used to apply
    // BCs in that direction.
    Kokkos::View<Ordinal* [3], Kokkos::HostSpace> mat_structure(
        "Matrix Structure", 2);
    mat_structure(0, 0) = 10;  // Request 10 grid point in 'x' direction
    mat_structure(0, 1) = 1;   // Add BC to the left
    mat_structure(0, 2) = 1;   // Add BC to the right
    mat_structure(1, 0) = 10;  // Request 10 grid point in 'y' direction
    mat_structure(1, 1) = 1;   // Add BC to the bottom
    mat_structure(1, 2) = 1;   // Add BC to the top

    matrix_type A =
        Test::generate_structured_matrix2D<matrix_type>("FD", mat_structure);
    matrix_type B =
        Test::generate_structured_matrix2D<matrix_type>("FE", mat_structure);
    matrix_type C;

    // Create KokkosKernelHandle
    using KernelHandle = KokkosKernels::Experimental::KokkosKernelsHandle<
        Offset, Ordinal, Scalar, execution_space, memory_space, memory_space>;
    KernelHandle kh;
    kh.create_spadd_handle(false);

    const Scalar alpha = 2.5;
    const Scalar beta  = 1.2;


    KokkosSparse::spadd_symbolic(&kh, A, B, C);
    KokkosSparse::spadd_numeric(&kh, alpha, A, beta, B, C);
    kh.destroy_spadd_handle();

    std::cout << "spadd was performed correctly!" << std::endl;
  }

  Kokkos::finalize();

  return return_value;
}