cub/device/device_adjacent_difference.cuh

File members: cub/device/device_adjacent_difference.cuh

/******************************************************************************
 * Copyright (c) 2011-2021, NVIDIA CORPORATION.  All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 *     * Redistributions of source code must retain the above copyright
 *       notice, this list of conditions and the following disclaimer.
 *     * Redistributions in binary form must reproduce the above copyright
 *       notice, this list of conditions and the following disclaimer in the
 *       documentation and/or other materials provided with the distribution.
 *     * Neither the name of the NVIDIA CORPORATION nor the
 *       names of its contributors may be used to endorse or promote products
 *       derived from this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
 * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
 * DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE FOR ANY
 * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
 * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 *
 ******************************************************************************/

#pragma once

#include <cub/config.cuh>

#if defined(_CCCL_IMPLICIT_SYSTEM_HEADER_GCC)
#  pragma GCC system_header
#elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_CLANG)
#  pragma clang system_header
#elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_MSVC)
#  pragma system_header
#endif // no system header

#include <cub/detail/choose_offset.cuh>
#include <cub/detail/nvtx.cuh>
#include <cub/detail/type_traits.cuh>
#include <cub/device/dispatch/dispatch_adjacent_difference.cuh>
#include <cub/util_deprecated.cuh>
#include <cub/util_namespace.cuh>

#include <thrust/detail/integer_traits.h>

#include <cstdint>

CUB_NAMESPACE_BEGIN

struct DeviceAdjacentDifference
{
private:
  template <bool may_alias,
            bool read_left,
            typename NumItemsT,
            typename InputIteratorT,
            typename OutputIteratorT,
            typename DifferenceOpT>
  static CUB_RUNTIME_FUNCTION cudaError_t AdjacentDifference(
    void* d_temp_storage,
    std::size_t& temp_storage_bytes,
    InputIteratorT d_input,
    OutputIteratorT d_output,
    NumItemsT num_items,
    DifferenceOpT difference_op,
    cudaStream_t stream)
  {
    using OffsetT = detail::choose_offset_t<NumItemsT>;

    using DispatchT =
      DispatchAdjacentDifference<InputIteratorT, OutputIteratorT, DifferenceOpT, OffsetT, may_alias, read_left>;

    return DispatchT::Dispatch(
      d_temp_storage, temp_storage_bytes, d_input, d_output, static_cast<OffsetT>(num_items), difference_op, stream);
  }

public:
  template <typename InputIteratorT,
            typename OutputIteratorT,
            typename DifferenceOpT = ::cuda::std::minus<>,
            typename NumItemsT     = std::uint32_t>
  static CUB_RUNTIME_FUNCTION cudaError_t SubtractLeftCopy(
    void* d_temp_storage,
    std::size_t& temp_storage_bytes,
    InputIteratorT d_input,
    OutputIteratorT d_output,
    NumItemsT num_items,
    DifferenceOpT difference_op = {},
    cudaStream_t stream         = 0)
  {
    CUB_DETAIL_NVTX_RANGE_SCOPE_IF(d_temp_storage, "cub::DeviceAdjacentDifference::SubtractLeftCopy");

    constexpr bool may_alias = false;
    constexpr bool read_left = true;

    return AdjacentDifference<may_alias, read_left>(
      d_temp_storage, temp_storage_bytes, d_input, d_output, num_items, difference_op, stream);
  }

#ifndef DOXYGEN_SHOULD_SKIP_THIS // Do not document
  template <typename InputIteratorT, typename OutputIteratorT, typename DifferenceOpT, typename NumItemsT = std::uint32_t>
  CUB_DETAIL_RUNTIME_DEBUG_SYNC_IS_NOT_SUPPORTED static CUB_RUNTIME_FUNCTION cudaError_t SubtractLeftCopy(
    void* d_temp_storage,
    std::size_t& temp_storage_bytes,
    InputIteratorT d_input,
    OutputIteratorT d_output,
    NumItemsT num_items,
    DifferenceOpT difference_op,
    cudaStream_t stream,
    bool debug_synchronous)
  {
    CUB_DETAIL_RUNTIME_DEBUG_SYNC_USAGE_LOG

    return SubtractLeftCopy(d_temp_storage, temp_storage_bytes, d_input, d_output, num_items, difference_op, stream);
  }
#endif // DOXYGEN_SHOULD_SKIP_THIS

  template <typename RandomAccessIteratorT,
            typename DifferenceOpT = ::cuda::std::minus<>,
            typename NumItemsT     = std::uint32_t>
  static CUB_RUNTIME_FUNCTION cudaError_t SubtractLeft(
    void* d_temp_storage,
    std::size_t& temp_storage_bytes,
    RandomAccessIteratorT d_input,
    NumItemsT num_items,
    DifferenceOpT difference_op = {},
    cudaStream_t stream         = 0)
  {
    CUB_DETAIL_NVTX_RANGE_SCOPE_IF(d_temp_storage, "cub::DeviceAdjacentDifference::SubtractLeft");

    constexpr bool may_alias = true;
    constexpr bool read_left = true;

    return AdjacentDifference<may_alias, read_left>(
      d_temp_storage, temp_storage_bytes, d_input, d_input, num_items, difference_op, stream);
  }

#ifndef DOXYGEN_SHOULD_SKIP_THIS // Do not document
  template <typename RandomAccessIteratorT, typename DifferenceOpT, typename NumItemsT = std::uint32_t>
  CUB_DETAIL_RUNTIME_DEBUG_SYNC_IS_NOT_SUPPORTED static CUB_RUNTIME_FUNCTION cudaError_t SubtractLeft(
    void* d_temp_storage,
    std::size_t& temp_storage_bytes,
    RandomAccessIteratorT d_input,
    NumItemsT num_items,
    DifferenceOpT difference_op,
    cudaStream_t stream,
    bool debug_synchronous)
  {
    CUB_DETAIL_RUNTIME_DEBUG_SYNC_USAGE_LOG

    return SubtractLeft(d_temp_storage, temp_storage_bytes, d_input, num_items, difference_op, stream);
  }
#endif // DOXYGEN_SHOULD_SKIP_THIS

  template <typename InputIteratorT,
            typename OutputIteratorT,
            typename DifferenceOpT = ::cuda::std::minus<>,
            typename NumItemsT     = std::uint32_t>
  static CUB_RUNTIME_FUNCTION cudaError_t SubtractRightCopy(
    void* d_temp_storage,
    std::size_t& temp_storage_bytes,
    InputIteratorT d_input,
    OutputIteratorT d_output,
    NumItemsT num_items,
    DifferenceOpT difference_op = {},
    cudaStream_t stream         = 0)
  {
    CUB_DETAIL_NVTX_RANGE_SCOPE_IF(d_temp_storage, "cub::DeviceAdjacentDifference::SubtractRightCopy");

    constexpr bool may_alias = false;
    constexpr bool read_left = false;

    return AdjacentDifference<may_alias, read_left>(
      d_temp_storage, temp_storage_bytes, d_input, d_output, num_items, difference_op, stream);
  }

#ifndef DOXYGEN_SHOULD_SKIP_THIS // Do not document
  template <typename InputIteratorT, typename OutputIteratorT, typename DifferenceOpT, typename NumItemsT = std::uint32_t>
  CUB_DETAIL_RUNTIME_DEBUG_SYNC_IS_NOT_SUPPORTED static CUB_RUNTIME_FUNCTION cudaError_t SubtractRightCopy(
    void* d_temp_storage,
    std::size_t& temp_storage_bytes,
    InputIteratorT d_input,
    OutputIteratorT d_output,
    NumItemsT num_items,
    DifferenceOpT difference_op,
    cudaStream_t stream,
    bool debug_synchronous)
  {
    CUB_DETAIL_RUNTIME_DEBUG_SYNC_USAGE_LOG

    return SubtractRightCopy(d_temp_storage, temp_storage_bytes, d_input, d_output, num_items, difference_op, stream);
  }
#endif // DOXYGEN_SHOULD_SKIP_THIS

  template <typename RandomAccessIteratorT,
            typename DifferenceOpT = ::cuda::std::minus<>,
            typename NumItemsT     = std::uint32_t>
  static CUB_RUNTIME_FUNCTION cudaError_t SubtractRight(
    void* d_temp_storage,
    std::size_t& temp_storage_bytes,
    RandomAccessIteratorT d_input,
    NumItemsT num_items,
    DifferenceOpT difference_op = {},
    cudaStream_t stream         = 0)
  {
    CUB_DETAIL_NVTX_RANGE_SCOPE_IF(d_temp_storage, "cub::DeviceAdjacentDifference::SubtractRight");

    constexpr bool may_alias = true;
    constexpr bool read_left = false;

    return AdjacentDifference<may_alias, read_left>(
      d_temp_storage, temp_storage_bytes, d_input, d_input, num_items, difference_op, stream);
  }

#ifndef DOXYGEN_SHOULD_SKIP_THIS // Do not document
  template <typename RandomAccessIteratorT, typename DifferenceOpT, typename NumItemsT>
  CUB_DETAIL_RUNTIME_DEBUG_SYNC_IS_NOT_SUPPORTED static CUB_RUNTIME_FUNCTION cudaError_t SubtractRight(
    void* d_temp_storage,
    std::size_t& temp_storage_bytes,
    RandomAccessIteratorT d_input,
    NumItemsT num_items,
    DifferenceOpT difference_op,
    cudaStream_t stream,
    bool debug_synchronous)
  {
    CUB_DETAIL_RUNTIME_DEBUG_SYNC_USAGE_LOG

    return SubtractRight(d_temp_storage, temp_storage_bytes, d_input, num_items, difference_op, stream);
  }
#endif // DOXYGEN_SHOULD_SKIP_THIS
};

CUB_NAMESPACE_END