slip_filter.hpp

// FEAT3: Finite Element Analysis Toolbox, Version 3
// Copyright (C) 2010 by Stefan Turek & the FEAT group
// FEAT3 is released under the GNU General Public License version 3,
// see the file 'copyright.txt' in the top level directory for details.
 
#pragma once
#ifndef KERNEL_LAFEM_ARCH_SLIP_FILTER_HPP
#define KERNEL_LAFEM_ARCH_SLIP_FILTER_HPP 1
 
// includes, FEAT
#include <kernel/base_header.hpp>
#include <kernel/backend.hpp>
 
namespace FEAT
{
  namespace LAFEM
  {
    namespace Arch
    {
      struct SlipFilter
      {
        template <int BlockSize_, typename DT_, typename IT_>
        static void filter_rhs(DT_ * v, const DT_ * const nu_elements, const IT_ * const sv_indices, const Index ue)
        {
          filter_rhs_generic<BlockSize_, DT_, IT_>(v, nu_elements, sv_indices, ue);
        }
 
        template <int BlockSize_>
        static void filter_rhs(float * v, const float * const nu_elements, const std::uint64_t * const sv_indices, const Index ue)
        {
          BACKEND_SKELETON_VOID_T1(BlockSize_, filter_rhs_cuda, filter_rhs_generic, filter_rhs_generic, v, nu_elements, sv_indices, ue)
        }
 
        template <int BlockSize_>
        static void filter_rhs(double * v, const double * const nu_elements, const std::uint64_t * const sv_indices, const Index ue)
        {
          BACKEND_SKELETON_VOID_T1(BlockSize_, filter_rhs_cuda, filter_rhs_generic, filter_rhs_generic, v, nu_elements, sv_indices, ue)
        }
 
        template <int BlockSize_>
        static void filter_rhs(float * v, const float * const nu_elements, const std::uint32_t * const sv_indices, const Index ue)
        {
          BACKEND_SKELETON_VOID_T1(BlockSize_, filter_rhs_cuda, filter_rhs_generic, filter_rhs_generic, v, nu_elements, sv_indices, ue)
        }
 
        template <int BlockSize_>
        static void filter_rhs(double * v, const double * const nu_elements, const std::uint32_t * const sv_indices, const Index ue)
        {
          BACKEND_SKELETON_VOID_T1(BlockSize_, filter_rhs_cuda, filter_rhs_generic, filter_rhs_generic, v, nu_elements, sv_indices, ue)
        }
 
        template <int BlockSize_, typename DT_, typename IT_>
        static void filter_def(DT_ * v, const DT_ * const nu_elements, const IT_ * const sv_indices, const Index ue)
        {
          filter_def_generic<BlockSize_, DT_, IT_>(v, nu_elements, sv_indices, ue);
        }
 
        template <int BlockSize_>
        static void filter_def(float * v, const float * const nu_elements, const std::uint64_t * const sv_indices, const Index ue)
        {
          BACKEND_SKELETON_VOID_T1(BlockSize_, filter_def_cuda, filter_def_generic, filter_def_generic, v, nu_elements, sv_indices, ue)
        }
 
        template <int BlockSize_>
        static void filter_def(double * v, const double * const nu_elements, const std::uint64_t * const sv_indices, const Index ue)
        {
          BACKEND_SKELETON_VOID_T1(BlockSize_, filter_def_cuda, filter_def_generic, filter_def_generic, v, nu_elements, sv_indices, ue)
        }
 
        template <int BlockSize_>
        static void filter_def(float * v, const float * const nu_elements, const std::uint32_t * const sv_indices, const Index ue)
        {
          BACKEND_SKELETON_VOID_T1(BlockSize_, filter_def_cuda, filter_def_generic, filter_def_generic, v, nu_elements, sv_indices, ue)
        }
 
        template <int BlockSize_>
        static void filter_def(double * v, const double * const nu_elements, const std::uint32_t * const sv_indices, const Index ue)
        {
          BACKEND_SKELETON_VOID_T1(BlockSize_, filter_def_cuda, filter_def_generic, filter_def_generic, v, nu_elements, sv_indices, ue)
        }
 
        template <int BlockSize_, typename DT_, typename IT_>
        static void filter_rhs_generic(DT_ * v, const DT_ * const nu_elements, const IT_ * const sv_indices, const Index ue);
 
        template <int BlockSize_, typename DT_, typename IT_>
        static void filter_def_generic(DT_ * v, const DT_ * const nu_elements, const IT_ * const sv_indices, const Index ue);
 
        template <int BlockSize_, typename DT_, typename IT_>
        static void filter_rhs_cuda(DT_ * v, const DT_ * const nu_elements, const IT_ * const sv_indices, const Index ue);
 
        template <int BlockSize_, typename DT_, typename IT_>
        static void filter_def_cuda(DT_ * v, const DT_ * const nu_elements, const IT_ * const sv_indices, const Index ue);
      }; // SlipFilter
 
      // Do not instantiate the following templates as this is done in slip_filter_generic.cpp and then linked
      // into the shared library
#ifdef FEAT_EICKT
      extern template void SlipFilter::filter_rhs_generic<2, float, std::uint64_t>(float * v, const float * const nu_elements, const std::uint64_t * const sv_indices, const Index ue);
      extern template void SlipFilter::filter_rhs_generic<2, double, std::uint64_t>(double * v, const double * const nu_elements, const std::uint64_t * const sv_indices, const Index ue);
      extern template void SlipFilter::filter_rhs_generic<2, float, std::uint32_t>(float * v, const float * const nu_elements, const std::uint32_t * const sv_indices, const Index ue);
      extern template void SlipFilter::filter_rhs_generic<2, double, std::uint32_t>(double * v, const double * const nu_elements, const std::uint32_t * const sv_indices, const Index ue);
 
      extern template void SlipFilter::filter_def_generic<2, float, std::uint64_t>(float * v, const float * const nu_elements, const std::uint64_t * const sv_indices, const Index ue);
      extern template void SlipFilter::filter_def_generic<2, double, std::uint64_t>(double * v, const double * const nu_elements, const std::uint64_t * const sv_indices, const Index ue);
      extern template void SlipFilter::filter_def_generic<2, float, std::uint32_t>(float * v, const float * const nu_elements, const std::uint32_t * const sv_indices, const Index ue);
      extern template void SlipFilter::filter_def_generic<2, double, std::uint32_t>(double * v, const double * const nu_elements, const std::uint32_t * const sv_indices, const Index ue);
 
      extern template void SlipFilter::filter_rhs_generic<3, float, std::uint64_t>(float * v, const float * const nu_elements, const std::uint64_t * const sv_indices, const Index ue);
      extern template void SlipFilter::filter_rhs_generic<3, double, std::uint64_t>(double * v, const double * const nu_elements, const std::uint64_t * const sv_indices, const Index ue);
      extern template void SlipFilter::filter_rhs_generic<3, float, std::uint32_t>(float * v, const float * const nu_elements, const std::uint32_t * const sv_indices, const Index ue);
      extern template void SlipFilter::filter_rhs_generic<3, double, std::uint32_t>(double * v, const double * const nu_elements, const std::uint32_t * const sv_indices, const Index ue);
 
      extern template void SlipFilter::filter_def_generic<3, float, std::uint64_t>(float * v, const float * const nu_elements, const std::uint64_t * const sv_indices, const Index ue);
      extern template void SlipFilter::filter_def_generic<3, double, std::uint64_t>(double * v, const double * const nu_elements, const std::uint64_t * const sv_indices, const Index ue);
      extern template void SlipFilter::filter_def_generic<3, float, std::uint32_t>(float * v, const float * const nu_elements, const std::uint32_t * const sv_indices, const Index ue);
      extern template void SlipFilter::filter_def_generic<3, double, std::uint32_t>(double * v, const double * const nu_elements, const std::uint32_t * const sv_indices, const Index ue);
#endif
    } // namespace Arch
  } // namespace LAFEM
} // namespace FEAT
 
#ifndef  __CUDACC__
#include <kernel/lafem/arch/slip_filter_generic.hpp>
#endif
#endif // KERNEL_LAFEM_ARCH_SLIP_FILTER_HPP