repo_id
stringlengths
21
96
file_path
stringlengths
31
155
content
stringlengths
1
92.9M
__index_level_0__
int64
0
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/mdarray.hpp
/* * Copyright (2019) Sandia Corporation * * The source code is licensed under the 3-clause BSD license found in the LICENSE file * thirdparty/LICENSES/mdarray.license */ /* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <stddef.h> #include <raft/core/detail/macros.hpp> #include <raft/core/host_device_accessor.hpp> #include <raft/core/mdspan.hpp> #include <raft/core/mdspan_types.hpp> #include <raft/core/memory_type.hpp> #include <raft/core/resources.hpp> namespace raft { /** * @defgroup mdarray_apis multi-dimensional memory-owning type * @{ */ /** * @brief Interface to implement an owning multi-dimensional array * * raft::array_interace is an interface to owning container types for mdspan. * Check implementation of raft::mdarray which implements raft::array_interface * using Curiously Recurring Template Pattern. * This interface calls into method `view()` whose implementation is provided by * the implementing class. `view()` must return an object of type raft::host_mdspan * or raft::device_mdspan or any types derived from the them. */ template <typename Base> class array_interface { /** * @brief Get an mdspan */ auto view() noexcept { return static_cast<Base*>(this)->view(); } /** * @brief Get an mdspan<const T> */ auto view() const noexcept { return static_cast<Base*>(this)->view(); } }; namespace detail { template <typename T, typename = void> struct is_array_interface : std::false_type {}; template <typename T> struct is_array_interface<T, std::void_t<decltype(std::declval<T>().view())>> : std::bool_constant<is_mdspan_v<decltype(std::declval<T>().view())>> {}; template <typename T> using is_array_interface_t = is_array_interface<std::remove_const_t<T>>; /** * @\brief Boolean to determine if template type T is raft::array_interface or derived type * or any type that has a member function `view()` that returns either * raft::host_mdspan or raft::device_mdspan */ template <typename T> inline constexpr bool is_array_interface_v = is_array_interface<std::remove_const_t<T>>::value; } // namespace detail template <typename...> struct is_array_interface : std::true_type {}; template <typename T1> struct is_array_interface<T1> : detail::is_array_interface_t<T1> {}; template <typename T1, typename... Tn> struct is_array_interface<T1, Tn...> : std::conditional_t<detail::is_array_interface_v<T1>, is_array_interface<Tn...>, std::false_type> {}; /** * @\brief Boolean to determine if variadic template types Tn are raft::array_interface * or derived type or any type that has a member function `view()` that returns either * raft::host_mdspan or raft::device_mdspan */ template <typename... Tn> inline constexpr bool is_array_interface_v = is_array_interface<Tn...>::value; /** * @brief Modified from the c++ mdarray proposal * * https://isocpp.org/files/papers/D1684R0.html * * mdarray is a container type for mdspan with similar template arguments. However there * are some inconsistencies in between them. We have made some modificiations to fit our * needs, which are listed below. * * - Layout policy is different, the mdarray in raft uses `std::experimental::extent` directly just * like `mdspan`, while the `mdarray` in the reference implementation uses varidic * template. * * - Most of the constructors from the reference implementation is removed to make sure * CUDA stream is honored. Note that this class is not coupled to CUDA and therefore * will only be used in the case where the device variant is used. * * - unique_size is not implemented, which is still working in progress in the proposal * * - For container policy, we adopt the alternative approach documented in the proposal * [sec 2.4.3], which requires an additional make_accessor method for it to be used in * mdspan. The container policy reference implementation has multiple `access` methods * that accommodate needs for both mdarray and mdspan. This is more difficult for us * since the policy might contain states that are unwanted inside a CUDA kernel. Also, * on host we return a proxy to the actual value as `device_ref` so different access * methods will have different return type, which is less desirable. * * - For the above reasons, copying from other mdarray with different policy type is also * removed. */ template <typename ElementType, typename Extents, typename LayoutPolicy, typename ContainerPolicy> class mdarray : public array_interface<mdarray<ElementType, Extents, LayoutPolicy, ContainerPolicy>> { static_assert(!std::is_const<ElementType>::value, "Element type for container must not be const."); public: using extents_type = Extents; using layout_type = LayoutPolicy; using mapping_type = typename layout_type::template mapping<extents_type>; using element_type = ElementType; using value_type = std::remove_cv_t<element_type>; using index_type = typename extents_type::index_type; using difference_type = std::ptrdiff_t; using rank_type = typename extents_type::rank_type; // Naming: ref impl: container_policy_type, proposal: container_policy using container_policy_type = ContainerPolicy; using container_type = typename container_policy_type::container_type; using pointer = typename container_policy_type::pointer; using const_pointer = typename container_policy_type::const_pointer; using reference = typename container_policy_type::reference; using const_reference = typename container_policy_type::const_reference; private: template <typename E, typename ViewAccessorPolicy = std::conditional_t<std::is_const_v<E>, typename container_policy_type::const_accessor_policy, typename container_policy_type::accessor_policy>> using view_type_impl = mdspan<E, extents_type, layout_type, host_device_accessor<ViewAccessorPolicy, container_policy_type::mem_type>>; public: /** * \brief the mdspan type returned by view method. */ using view_type = view_type_impl<element_type>; using const_view_type = view_type_impl<element_type const>; public: constexpr mdarray(raft::resources const& handle) noexcept( std::is_nothrow_default_constructible_v<container_type>) : cp_{}, c_{cp_.create(handle, 0)} {}; constexpr mdarray(mdarray const&) noexcept(std::is_nothrow_copy_constructible_v<container_type>) = default; constexpr mdarray(mdarray&&) noexcept(std::is_nothrow_move_constructible<container_type>::value) = default; constexpr auto operator=(mdarray const&) noexcept( std::is_nothrow_copy_assignable<container_type>::value) -> mdarray& = default; constexpr auto operator=(mdarray&&) noexcept( std::is_nothrow_move_assignable<container_type>::value) -> mdarray& = default; ~mdarray() noexcept(std::is_nothrow_destructible<container_type>::value) = default; #ifndef RAFT_MDARRAY_CTOR_CONSTEXPR #if !(__CUDACC_VER_MAJOR__ == 11 && __CUDACC_VER_MINOR__ <= 2) // 11.0: // Error: Internal Compiler Error (codegen): "there was an error in verifying the lgenfe output!" // // 11.2: // Call parameter type does not match function signature! // i8** null // i8* %call14 = call i32 null(void (i8*)* null, i8* null, i8** null), !dbg !1060 // <unnamed>: parse Invalid record (Producer: 'LLVM7.0.1' Reader: 'LLVM 7.0.1') #define RAFT_MDARRAY_CTOR_CONSTEXPR constexpr #else #define RAFT_MDARRAY_CTOR_CONSTEXPR #endif // !(__CUDACC_VER_MAJOR__ == 11 && __CUDACC_VER_MINOR__ <= 2) #endif // RAFT_MDARRAY_CTOR_CONSTEXPR /** * @brief The only constructor that can create storage, raft::resources is accepted * so that the device implementation can make sure the relevant CUDA stream is * being used for allocation. */ RAFT_MDARRAY_CTOR_CONSTEXPR mdarray(raft::resources const& handle, mapping_type const& m, container_policy_type const& cp) : cp_(cp), map_(m), c_(cp_.create(handle, map_.required_span_size())) { } RAFT_MDARRAY_CTOR_CONSTEXPR mdarray(raft::resources const& handle, mapping_type const& m, container_policy_type& cp) : cp_(cp), map_(m), c_(cp_.create(handle, map_.required_span_size())) { } #undef RAFT_MDARRAY_CTOR_CONSTEXPR /** * @brief Get an mdspan */ auto view() noexcept { return view_type(c_.data(), map_, cp_.make_accessor_policy()); } /** * @brief Get an mdspan<const T> */ auto view() const noexcept { return const_view_type(c_.data(), map_, cp_.make_accessor_policy()); } [[nodiscard]] constexpr auto size() const noexcept -> std::size_t { return this->view().size(); } [[nodiscard]] auto data_handle() noexcept -> pointer { return c_.data(); } [[nodiscard]] constexpr auto data_handle() const noexcept -> const_pointer { return c_.data(); } /** * @brief Indexing operator, use it sparingly since it triggers a device<->host copy. */ template <typename... IndexType> auto operator()(IndexType&&... indices) -> std::enable_if_t<sizeof...(IndexType) == extents_type::rank() && (std::is_convertible_v<IndexType, index_type> && ...) && std::is_constructible_v<extents_type, IndexType...>, /* device policy is not default constructible due to requirement for CUDA stream. */ /* std::is_default_constructible_v<container_policy_type> */ reference> { return cp_.access(c_, map_(std::forward<IndexType>(indices)...)); } /** * @brief Indexing operator, use it sparingly since it triggers a device<->host copy. */ template <typename... IndexType> auto operator()(IndexType&&... indices) const -> std::enable_if_t<sizeof...(IndexType) == extents_type::rank() && (std::is_convertible_v<IndexType, index_type> && ...) && std::is_constructible_v<extents_type, IndexType...> && std::is_constructible<mapping_type, extents_type>::value, /* device policy is not default constructible due to requirement for CUDA stream. */ /* std::is_default_constructible_v<container_policy_type> */ const_reference> { return cp_.access(c_, map_(std::forward<IndexType>(indices)...)); } // basic_mdarray observers of the domain multidimensional index space (also in basic_mdspan) [[nodiscard]] RAFT_INLINE_FUNCTION static constexpr auto rank() noexcept -> rank_type { return extents_type::rank(); } [[nodiscard]] RAFT_INLINE_FUNCTION static constexpr auto rank_dynamic() noexcept -> rank_type { return extents_type::rank_dynamic(); } [[nodiscard]] RAFT_INLINE_FUNCTION static constexpr auto static_extent(size_t r) noexcept -> index_type { return extents_type::static_extent(r); } [[nodiscard]] RAFT_INLINE_FUNCTION constexpr auto extents() const noexcept -> extents_type { return map_.extents(); } /** * @brief the extent of rank r */ [[nodiscard]] RAFT_INLINE_FUNCTION constexpr auto extent(size_t r) const noexcept -> index_type { return map_.extents().extent(r); } // mapping [[nodiscard]] RAFT_INLINE_FUNCTION constexpr auto mapping() const noexcept -> mapping_type { return map_; } [[nodiscard]] RAFT_INLINE_FUNCTION constexpr auto is_unique() const noexcept -> bool { return map_.is_unique(); } [[nodiscard]] RAFT_INLINE_FUNCTION constexpr auto is_exhaustive() const noexcept -> bool { return map_.is_exhaustive(); } [[nodiscard]] RAFT_INLINE_FUNCTION constexpr auto is_strided() const noexcept -> bool { return map_.is_strided(); } [[nodiscard]] RAFT_INLINE_FUNCTION constexpr auto stride(size_t r) const -> index_type { return map_.stride(r); } [[nodiscard]] RAFT_INLINE_FUNCTION static constexpr auto is_always_unique() noexcept -> bool { return mapping_type::is_always_unique(); } [[nodiscard]] RAFT_INLINE_FUNCTION static constexpr auto is_always_exhaustive() noexcept -> bool { return mapping_type::is_always_exhaustive(); } [[nodiscard]] RAFT_INLINE_FUNCTION static constexpr auto is_always_strided() noexcept -> bool { return mapping_type::is_always_strided(); } private: template <typename, typename, typename, typename> friend class mdarray; private: container_policy_type cp_; mapping_type map_; container_type c_; }; /** @} */ /** * @defgroup mdarray_reshape Row- or Col-norm computation * @{ */ /** * @brief Flatten object implementing raft::array_interface into a 1-dim array view * * @tparam array_interface_type Expected type implementing raft::array_interface * @param mda raft::array_interace implementing object * @return Either raft::host_mdspan or raft::device_mdspan with vector_extent * depending on the underlying ContainerPolicy */ template <typename array_interface_type, std::enable_if_t<is_array_interface_v<array_interface_type>>* = nullptr> auto flatten(const array_interface_type& mda) { return flatten(mda.view()); } /** * @brief Reshape object implementing raft::array_interface * * @tparam array_interface_type Expected type implementing raft::array_interface * @tparam Extents raft::extents for dimensions * @tparam IndexType the index type of the extents * @param mda raft::array_interace implementing object * @param new_shape Desired new shape of the input * @return raft::host_mdspan or raft::device_mdspan, depending on the underlying * ContainerPolicy */ template <typename array_interface_type, typename IndexType = std::uint32_t, size_t... Extents, std::enable_if_t<is_array_interface_v<array_interface_type>>* = nullptr> auto reshape(const array_interface_type& mda, extents<IndexType, Extents...> new_shape) { return reshape(mda.view(), new_shape); } /** @} */ } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/host_csr_matrix.hpp
/* * Copyright (c) 2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/csr_matrix.hpp> #include <raft/core/host_container_policy.hpp> #include <raft/core/host_span.hpp> #include <raft/core/resources.hpp> #include <raft/core/sparse_types.hpp> #include <type_traits> namespace raft { /** * \defgroup host_csr_matrix Host CSR Matrix * @{ */ /** * Specialization for a sparsity-preserving compressed structure view which uses host memory */ template <typename IndptrType, typename IndicesType, typename NZType> using host_compressed_structure_view = compressed_structure_view<IndptrType, IndicesType, NZType, false>; /** * Specialization for a sparsity-owning compressed structure which uses host memory */ template <typename IndptrType, typename IndicesType, typename NZType, template <typename T> typename ContainerPolicy = host_vector_policy> using host_compressed_structure = compressed_structure<IndptrType, IndicesType, NZType, false, ContainerPolicy>; /** * Specialization for a csr matrix view which uses host memory */ template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType> using host_csr_matrix_view = csr_matrix_view<ElementType, IndptrType, IndicesType, NZType, false>; template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType, template <typename T> typename ContainerPolicy = host_vector_policy, SparsityType sparsity_type = SparsityType::OWNING> using host_csr_matrix = csr_matrix<ElementType, IndptrType, IndicesType, NZType, false, ContainerPolicy, sparsity_type>; /** * Specialization for a sparsity-owning csr matrix which uses host memory */ template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType, template <typename T> typename ContainerPolicy = host_vector_policy> using host_sparsity_owning_csr_matrix = csr_matrix<ElementType, IndptrType, IndicesType, NZType, false, ContainerPolicy>; /** * Specialization for a sparsity-preserving csr matrix which uses host memory */ template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType, template <typename T> typename ContainerPolicy = host_vector_policy> using host_sparsity_preserving_csr_matrix = csr_matrix<ElementType, IndptrType, IndicesType, NZType, false, ContainerPolicy, SparsityType::PRESERVING>; template <typename T> struct is_host_csr_matrix_view : std::false_type {}; template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType> struct is_host_csr_matrix_view<host_csr_matrix_view<ElementType, IndptrType, IndicesType, NZType>> : std::true_type {}; template <typename T> constexpr bool is_host_csr_matrix_view_v = is_host_csr_matrix_view<T>::value; template <typename T> struct is_host_csr_matrix : std::false_type {}; template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType, template <typename T> typename ContainerPolicy, SparsityType sparsity_type> struct is_host_csr_matrix< host_csr_matrix<ElementType, IndptrType, IndicesType, NZType, ContainerPolicy, sparsity_type>> : std::true_type {}; template <typename T> constexpr bool is_host_csr_matrix_v = is_host_csr_matrix<T>::value; template <typename T> constexpr bool is_host_csr_sparsity_owning_v = is_host_csr_matrix<T>::value and T::get_sparsity_type() == OWNING; template <typename T> constexpr bool is_host_csr_sparsity_preserving_v = std::disjunction_v< is_host_csr_matrix_view<T>, std::bool_constant<is_host_csr_matrix<T>::value and T::get_sparsity_type() == PRESERVING>>; /** * Create a sparsity-owning sparse matrix in the compressed-sparse row format. sparsity-owning * means that all of the underlying vectors (data, indptr, indices) are owned by the csr_matrix * instance. If not known up front, the sparsity can be ignored in this factory function and * `resize()` invoked on the instance once the sparsity is known. * * @code{.cpp} * #include <raft/core/host_resources.hpp> * #include <raft/core/host_csr_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * * raft::resources handle; * csr_matrix = raft::make_host_csr_matrix(handle, n_rows, n_cols); * ... * // compute expected sparsity * ... * int nnz = 5000; * csr_matrix.initialize_sparsity(nnz); * @endcode * * @tparam ElementType * @tparam IndptrType * @tparam IndicesType * @tparam NZType * @param[in] handle raft handle for managing expensive resources * @param[in] n_rows total number of rows in the matrix * @param[in] n_cols total number of columns in the matrix * @param[in] nnz number of non-zeros in the matrix if known [optional] * @return a sparsity-owning sparse matrix in compressed (csr) format */ template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType = uint64_t> auto make_host_csr_matrix(raft::resources const& handle, IndptrType n_rows, IndicesType n_cols, NZType nnz = 0) { return host_sparsity_owning_csr_matrix<ElementType, IndptrType, IndicesType, NZType>( handle, n_rows, n_cols, nnz); } /** * Create a sparsity-preserving sparse matrix in the compressed-sparse row format. * sparsity-preserving means that a view of the csr sparsity is supplied, allowing the values in * the sparsity to change but not the sparsity itself. The csr_matrix instance does not own the * sparsity, the sparsity must be known up front, and cannot be resized later. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/host_csr_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * * raft::resources handle; * coo_structure = raft::make_host_compressed_structure(handle, n_rows, n_cols); * ... * // compute expected sparsity * ... * csr_structure.initialize_sparsity(nnz); * csr_matrix = raft::make_host_csr_matrix(handle, csr_structure.view()); * @endcode * * @tparam ElementType * @tparam IndptrType * @tparam IndicesType * @tparam NZType * @param[in] handle raft handle for managing expensive resources * @param[in] structure a sparsity-preserving compressed structural view * @return a sparsity-preserving sparse matrix in compressed (csr) format */ template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType = uint64_t> auto make_host_csr_matrix(raft::resources const& handle, host_compressed_structure_view<IndptrType, IndicesType, NZType> structure) { return host_sparsity_preserving_csr_matrix<ElementType, IndptrType, IndicesType, NZType>( handle, structure); } /** * Create a non-owning sparse matrix view in the coordinate format. This is sparsity-preserving, * meaning that the underlying sparsity is known and cannot be changed. Use the sparsity-owning * coo_matrix if sparsity needs to be mutable. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/host_csr_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * int nnz = 5000; * * // The following pointer is assumed to reference device memory for a size of nnz * float* h_elm_ptr = ...; * * raft::resources handle; * csr_structure = raft::make_host_compressed_structure(handle, n_rows, n_cols, nnz); * csr_matrix_view = raft::make_host_csr_matrix_view(handle, h_elm_ptr, csr_structure.view()); * @endcode * * @tparam ElementType * @tparam IndptrType * @tparam IndicesType * @tparam NZType * @param[in] ptr a pointer to array of nonzero matrix elements on host (size nnz) * @param[in] structure a sparsity-preserving compressed sparse structural view * @return a sparsity-preserving csr matrix view */ template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType = uint64_t> auto make_host_csr_matrix_view( ElementType* ptr, host_compressed_structure_view<IndptrType, IndicesType, NZType> structure) { return host_csr_matrix_view<ElementType, IndptrType, IndicesType, NZType>( raft::host_span<ElementType>(ptr, structure.get_nnz()), structure); } /** * Create a non-owning sparse matrix view in the compressed-sparse row format. This is * sparsity-preserving, meaning that the underlying sparsity is known and cannot be changed. Use the * sparsity-owning coo_matrix if sparsity needs to be mutable. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/host_span.hpp> * #include <raft/core/host_csr_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * int nnz = 5000; * * // The following span is assumed to be of size nnz * raft::host_span<float> h_elm_ptr; * * raft::resources handle; * csr_structure = raft::make_host_compressed_structure(handle, n_rows, n_cols, nnz); * csr_matrix_view = raft::make_host_csr_matrix_view(handle, h_elm_ptr, csr_structure.view()); * @endcode * * @tparam ElementType * @tparam IndptrType * @tparam IndicesType * @tparam NZType * @param[in] elements host span containing array of matrix elements (size nnz) * @param[in] structure a sparsity-preserving structural view * @return a sparsity-preserving csr matrix view */ template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType = uint64_t> auto make_host_csr_matrix_view( raft::host_span<ElementType> elements, host_compressed_structure_view<IndptrType, IndicesType, NZType> structure) { RAFT_EXPECTS(elements.size() == structure.get_nnz(), "Size of elements must be equal to the nnz from the structure"); return host_csr_matrix_view<ElementType, IndptrType, IndicesType, NZType>(elements, structure); } /** * Create a sparsity-owning compressed structure. This is not sparsity-preserving, meaning that * the underlying sparsity does not need to be known upon construction. When not known up front, * the allocation of the underlying indices array is delayed until `resize(nnz)` is invoked. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/host_csr_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * int nnz = 5000; * * raft::resources handle; * csr_structure = raft::make_host_compressed_structure(handle, n_rows, n_cols, nnz); * ... * // compute expected sparsity * ... * csr_structure.initialize_sparsity(nnz); * @endcode * * * @tparam IndptrType * @tparam IndicesType * @tparam NZType * @param[in] handle raft handle for managing expensive resources * @param[in] n_rows total number of rows * @param[in] n_cols total number of cols * @param[in] nnz total number of nonzeros, if known * @return a sparsity-owning compressed structure instance */ template <typename IndptrType, typename IndicesType, typename NZType = uint64_t> auto make_host_compressed_structure(raft::resources const& handle, IndptrType n_rows, IndicesType n_cols, NZType nnz = 0) { return host_compressed_structure<IndptrType, IndicesType, NZType>(handle, n_rows, n_cols, nnz); } /** * Create a non-owning sparsity-preserved compressed structure view. Sparsity-preserving means that * the underlying sparsity is known and cannot be changed. Use the sparsity-owning version if the * sparsity is not known up front. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/host_csr_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * int nnz = 5000; * * // The following pointer is assumed to reference host-accessible memory of size n_rows+1 * int *indptr = ...; * * // The following pointer is assumed to reference host-accessible memory of size nnz * int *indices = ...; * * raft::resources handle; * csr_structure = raft::make_host_compressed_structure_view(handle, indptr, indices, n_rows, * n_cols, nnz); * @endcode * * * @tparam ElementType * @tparam IndptrType * @tparam IndicesType * @tparam NZType * @param[in] indptr structural indptr (size n_rows+1) * @param[in] indices structural indices (size nnz) * @param[in] n_rows total number of rows * @param[in] n_cols total number of columns * @param[in] nnz number of non-zeros * @return a sparsity-preserving compressed structural view */ template <typename IndptrType, typename IndicesType, typename NZType = uint64_t> auto make_host_compressed_structure_view( IndptrType* indptr, IndicesType* indices, IndptrType n_rows, IndicesType n_cols, NZType nnz) { return host_compressed_structure_view<IndptrType, IndicesType, NZType>( raft::host_span<IndptrType>(indptr, n_rows + 1), raft::host_span<IndicesType>(indices, nnz), n_cols); } /** * Create a non-owning sparsity-preserved compressed structure view. Sparsity-preserving means that * the underlying sparsity is known and cannot be changed. Use the sparsity-owning version if the * sparsity is not known up front. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/host_csr_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * int nnz = 5000; * * // The following host span is assumed to be of size n_rows+1 * raft::host_span<int> indptr; * * // The following host span is assumed to be of size nnz * raft::host_span<int> indices; * * raft::resources handle; * csr_structure = raft::make_host_compressed_structure_view(handle, indptr, indices, n_rows, * n_cols); * @endcode * * @tparam IndptrType * @tparam IndicesType * @tparam NZType * @param[in] indptr structural indptr (size n_rows+1) * @param[in] indices structural indices (size nnz) * @param[in] n_cols total number of columns * @return a sparsity-preserving compressed structural view * */ template <typename IndptrType, typename IndicesType, typename NZType = uint64_t> auto make_host_compressed_structure_view(raft::host_span<IndptrType> indptr, raft::host_span<IndicesType> indices, IndicesType n_cols) { return host_compressed_structure_view<IndptrType, IndicesType, NZType>(indptr, indices, n_cols); } /** @} */ }; // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/host_mdarray.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cstdint> #include <raft/core/host_mdspan.hpp> #include <raft/core/resources.hpp> #include <raft/core/host_container_policy.hpp> #include <raft/core/mdarray.hpp> namespace raft { /** * @brief mdarray with host container policy * @tparam ElementType the data type of the elements * @tparam Extents defines the shape * @tparam LayoutPolicy policy for indexing strides and layout ordering * @tparam ContainerPolicy storage and accessor policy */ template <typename ElementType, typename Extents, typename LayoutPolicy = layout_c_contiguous, typename ContainerPolicy = host_vector_policy<ElementType>> using host_mdarray = mdarray<ElementType, Extents, LayoutPolicy, host_accessor<ContainerPolicy>>; /** * @brief Shorthand for 0-dim host mdarray (scalar). * @tparam ElementType the data type of the scalar element * @tparam IndexType the index type of the extents */ template <typename ElementType, typename IndexType = std::uint32_t> using host_scalar = host_mdarray<ElementType, scalar_extent<IndexType>>; /** * @brief Shorthand for 1-dim host mdarray. * @tparam ElementType the data type of the vector elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> using host_vector = host_mdarray<ElementType, vector_extent<IndexType>, LayoutPolicy>; /** * @brief Shorthand for c-contiguous host matrix. * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> using host_matrix = host_mdarray<ElementType, matrix_extent<IndexType>, LayoutPolicy>; /** * @defgroup host_mdarray_factories factories to create host mdarrays * @{ */ /** * @brief Create a host mdarray. * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering * @param[in] res raft handle for managing expensive resources * @param[in] exts dimensionality of the array (series of integers) * @return raft::host_mdarray */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous, size_t... Extents> auto make_host_mdarray(raft::resources& res, extents<IndexType, Extents...> exts) { using mdarray_t = host_mdarray<ElementType, decltype(exts), LayoutPolicy>; typename mdarray_t::mapping_type layout{exts}; typename mdarray_t::container_policy_type policy; return mdarray_t{res, layout, policy}; } /** * @} */ /** * @brief Create a host mdarray. * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering * @param exts dimensionality of the array (series of integers) * Note: This function is deprecated and will be removed in a future version. Please use version * that accepts raft::resources. * * @return raft::host_mdarray */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous, size_t... Extents> auto make_host_mdarray(extents<IndexType, Extents...> exts) { using mdarray_t = host_mdarray<ElementType, decltype(exts), LayoutPolicy>; typename mdarray_t::mapping_type layout{exts}; typename mdarray_t::container_policy_type policy; raft::resources res; return mdarray_t{res, layout, policy}; } /** * @ingroup host_mdarray_factories * @brief Create a 2-dim c-contiguous host mdarray. * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering * @param[in] res raft handle for managing expensive resources * @param[in] n_rows number or rows in matrix * @param[in] n_cols number of columns in matrix * @return raft::host_matrix */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> auto make_host_matrix(raft::resources& res, IndexType n_rows, IndexType n_cols) { return make_host_mdarray<ElementType, IndexType, LayoutPolicy>( res, make_extents<IndexType>(n_rows, n_cols)); } /** * @brief Create a 2-dim c-contiguous host mdarray. * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering * @param[in] n_rows number or rows in matrix * @param[in] n_cols number of columns in matrix * Note: This function is deprecated and will be removed in a future version. Please use version * that accepts raft::resources. * * @return raft::host_matrix */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> auto make_host_matrix(IndexType n_rows, IndexType n_cols) { return make_host_mdarray<ElementType, IndexType, LayoutPolicy>( make_extents<IndexType>(n_rows, n_cols)); } /** * @ingroup host_mdarray_factories * @brief Create a host scalar from v. * * @tparam ElementType the data type of the scalar element * @tparam IndexType the index type of the extents * @param[in] res raft handle for managing expensive resources * @param[in] v scalar type to wrap * @return raft::host_scalar */ template <typename ElementType, typename IndexType = std::uint32_t> auto make_host_scalar(raft::resources& res, ElementType const& v) { // FIXME(jiamingy): We can optimize this by using std::array as container policy, which // requires some more compile time dispatching. This is enabled in the ref impl but // hasn't been ported here yet. scalar_extent<IndexType> extents; using policy_t = typename host_scalar<ElementType>::container_policy_type; policy_t policy; auto scalar = host_scalar<ElementType>{res, extents, policy}; scalar(0) = v; return scalar; } /** * @brief Create a host scalar from v. * * @tparam ElementType the data type of the scalar element * @tparam IndexType the index type of the extents * @param[in] v scalar type to wrap * Note: This function is deprecated and will be removed in a future version. Please use version * that accepts raft::resources. * * @return raft::host_scalar */ template <typename ElementType, typename IndexType = std::uint32_t> auto make_host_scalar(ElementType const& v) { // FIXME(jiamingy): We can optimize this by using std::array as container policy, which // requires some more compile time dispatching. This is enabled in the ref impl but // hasn't been ported here yet. scalar_extent<IndexType> extents; using policy_t = typename host_scalar<ElementType>::container_policy_type; policy_t policy; raft::resources handle; auto scalar = host_scalar<ElementType>{handle, extents, policy}; scalar(0) = v; return scalar; } /** * @ingroup host_mdarray_factories * @brief Create a 1-dim host mdarray. * @tparam ElementType the data type of the vector elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering * @param[in] res raft handle for managing expensive resources * @param[in] n number of elements in vector * @return raft::host_vector */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> auto make_host_vector(raft::resources& res, IndexType n) { return make_host_mdarray<ElementType, IndexType, LayoutPolicy>(res, make_extents<IndexType>(n)); } /** * @brief Create a 1-dim host mdarray. * @tparam ElementType the data type of the vector elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering * @param[in] n number of elements in vector * * Note: This function is deprecated and will be removed in a future version. Please use version * that accepts raft::resources. * @return raft::host_vector */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> auto make_host_vector(IndexType n) { return make_host_mdarray<ElementType, IndexType, LayoutPolicy>(make_extents<IndexType>(n)); } } // end namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/device_span.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/span.hpp> namespace raft { /** * @defgroup device_span one-dimensional device span type * @{ */ /** * @brief A span class for device pointer. */ template <typename T, size_t extent = std::experimental::dynamic_extent> using device_span = span<T, true, extent>; /** * @} */ } // end namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/logger.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include "logger-macros.hpp" #include "logger-ext.hpp" #if !defined(RAFT_COMPILED) #include "logger-inl.hpp" #endif
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/logger-macros.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once /** * @defgroup logging levels used in raft * * @note exactly match the corresponding ones (but reverse in terms of value) * in spdlog for wrapping purposes * * @{ */ #define RAFT_LEVEL_TRACE 6 #define RAFT_LEVEL_DEBUG 5 #define RAFT_LEVEL_INFO 4 #define RAFT_LEVEL_WARN 3 #define RAFT_LEVEL_ERROR 2 #define RAFT_LEVEL_CRITICAL 1 #define RAFT_LEVEL_OFF 0 /** @} */ #if !defined(RAFT_ACTIVE_LEVEL) #define RAFT_ACTIVE_LEVEL RAFT_LEVEL_INFO #endif /** * @defgroup loggerMacros Helper macros for dealing with logging * @{ */ #if (RAFT_ACTIVE_LEVEL >= RAFT_LEVEL_TRACE) #define RAFT_LOG_TRACE(fmt, ...) \ do { \ std::stringstream ss; \ ss << raft::detail::format("%s:%d ", __FILE__, __LINE__); \ ss << raft::detail::format(fmt, ##__VA_ARGS__); \ raft::logger::get(RAFT_NAME).log(RAFT_LEVEL_TRACE, ss.str().c_str()); \ } while (0) #else #define RAFT_LOG_TRACE(fmt, ...) void(0) #endif #if (RAFT_ACTIVE_LEVEL >= RAFT_LEVEL_TRACE) #define RAFT_LOG_TRACE_VEC(ptr, len) \ do { \ std::stringstream ss; \ ss << raft::detail::format("%s:%d ", __FILE__, __LINE__); \ print_vector(#ptr, ptr, len, ss); \ raft::logger::get(RAFT_NAME).log(RAFT_LEVEL_TRACE, ss.str().c_str()); \ } while (0) #else #define RAFT_LOG_TRACE_VEC(ptr, len) void(0) #endif #if (RAFT_ACTIVE_LEVEL >= RAFT_LEVEL_DEBUG) #define RAFT_LOG_DEBUG(fmt, ...) \ do { \ std::stringstream ss; \ ss << raft::detail::format("%s:%d ", __FILE__, __LINE__); \ ss << raft::detail::format(fmt, ##__VA_ARGS__); \ raft::logger::get(RAFT_NAME).log(RAFT_LEVEL_DEBUG, ss.str().c_str()); \ } while (0) #else #define RAFT_LOG_DEBUG(fmt, ...) void(0) #endif #if (RAFT_ACTIVE_LEVEL >= RAFT_LEVEL_INFO) #define RAFT_LOG_INFO(fmt, ...) \ raft::logger::get(RAFT_NAME).log(RAFT_LEVEL_INFO, fmt, ##__VA_ARGS__) #else #define RAFT_LOG_INFO(fmt, ...) void(0) #endif #if (RAFT_ACTIVE_LEVEL >= RAFT_LEVEL_WARN) #define RAFT_LOG_WARN(fmt, ...) \ raft::logger::get(RAFT_NAME).log(RAFT_LEVEL_WARN, fmt, ##__VA_ARGS__) #else #define RAFT_LOG_WARN(fmt, ...) void(0) #endif #if (RAFT_ACTIVE_LEVEL >= RAFT_LEVEL_ERROR) #define RAFT_LOG_ERROR(fmt, ...) \ raft::logger::get(RAFT_NAME).log(RAFT_LEVEL_ERROR, fmt, ##__VA_ARGS__) #else #define RAFT_LOG_ERROR(fmt, ...) void(0) #endif #if (RAFT_ACTIVE_LEVEL >= RAFT_LEVEL_CRITICAL) #define RAFT_LOG_CRITICAL(fmt, ...) \ raft::logger::get(RAFT_NAME).log(RAFT_LEVEL_CRITICAL, fmt, ##__VA_ARGS__) #else #define RAFT_LOG_CRITICAL(fmt, ...) void(0) #endif /** @} */
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/resources.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include "resource/resource_types.hpp" #include <algorithm> #include <mutex> #include <raft/core/error.hpp> // RAFT_EXPECTS #include <raft/core/logger.hpp> #include <string> #include <vector> namespace raft { /** * @brief Resource container which allows lazy-loading and registration * of resource_factory implementations, which in turn generate resource instances. * * This class is intended to be agnostic of the resources it contains and * does not, itself, differentiate between host and device resources. Downstream * accessor functions can then register and load resources as needed in order * to keep its usage somewhat opaque to end-users. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/resource/cuda_stream.hpp> * #include <raft/core/resource/cublas_handle.hpp> * * raft::resources res; * auto stream = raft::resource::get_cuda_stream(res); * auto cublas_handle = raft::resource::get_cublas_handle(res); * @endcode */ class resources { public: template <typename T> using pair_res = std::pair<resource::resource_type, std::shared_ptr<T>>; using pair_res_factory = pair_res<resource::resource_factory>; using pair_resource = pair_res<resource::resource>; resources() : factories_(resource::resource_type::LAST_KEY), resources_(resource::resource_type::LAST_KEY) { for (int i = 0; i < resource::resource_type::LAST_KEY; ++i) { factories_.at(i) = std::make_pair(resource::resource_type::LAST_KEY, std::make_shared<resource::empty_resource_factory>()); resources_.at(i) = std::make_pair(resource::resource_type::LAST_KEY, std::make_shared<resource::empty_resource>()); } } /** * @brief Shallow copy of underlying resources instance. * Note that this does not create any new resources. */ resources(const resources& res) : factories_(res.factories_), resources_(res.resources_) {} resources(resources&&) = delete; resources& operator=(resources&&) = delete; /** * @brief Returns true if a resource_factory has been registered for the * given resource_type, false otherwise. * @param resource_type resource type to check * @return true if resource_factory is registered for the given resource_type */ bool has_resource_factory(resource::resource_type resource_type) const { std::lock_guard<std::mutex> _(mutex_); return factories_.at(resource_type).first != resource::resource_type::LAST_KEY; } /** * @brief Register a resource_factory with the current instance. * This will overwrite any existing resource factories. * @param factory resource factory to register on the current instance */ void add_resource_factory(std::shared_ptr<resource::resource_factory> factory) const { std::lock_guard<std::mutex> _(mutex_); resource::resource_type rtype = factory.get()->get_resource_type(); RAFT_EXPECTS(rtype != resource::resource_type::LAST_KEY, "LAST_KEY is a placeholder and not a valid resource factory type."); factories_.at(rtype) = std::make_pair(rtype, factory); // Clear the corresponding resource, so that on next `get_resource` the new factory is used if (resources_.at(rtype).first != resource::resource_type::LAST_KEY) { resources_.at(rtype) = std::make_pair(resource::resource_type::LAST_KEY, std::make_shared<resource::empty_resource>()); } } /** * @brief Retrieve a resource for the given resource_type and cast to given pointer type. * Note that the resources are loaded lazily on-demand and resources which don't yet * exist on the current instance will be created using the corresponding factory, if * it exists. * @tparam res_t pointer type for which retrieved resource will be casted * @param resource_type resource type to retrieve * @return the given resource, if it exists. */ template <typename res_t> res_t* get_resource(resource::resource_type resource_type) const { std::lock_guard<std::mutex> _(mutex_); if (resources_.at(resource_type).first == resource::resource_type::LAST_KEY) { RAFT_EXPECTS(factories_.at(resource_type).first != resource::resource_type::LAST_KEY, "No resource factory has been registered for the given resource %d.", resource_type); resource::resource_factory* factory = factories_.at(resource_type).second.get(); resources_.at(resource_type) = std::make_pair( resource_type, std::shared_ptr<resource::resource>(factory->make_resource())); } resource::resource* res = resources_.at(resource_type).second.get(); return reinterpret_cast<res_t*>(res->get_resource()); } protected: mutable std::mutex mutex_; mutable std::vector<pair_res_factory> factories_; mutable std::vector<pair_resource> resources_; }; } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/copy.cuh
/* * Copyright (c) 2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/detail/copy.hpp> namespace raft { /** * @brief Copy data from one mdspan to another with the same extents * * This function copies data from one mdspan to another, regardless of whether * or not the mdspans have the same layout, memory type (host/device/managed) * or data type. So long as it is possible to convert the data type from source * to destination, and the extents are equal, this function should be able to * perform the copy. Any necessary device operations will be stream-ordered via the CUDA stream * provided by the `raft::resources` argument. * * This header includes a custom kernel used for copying data between * completely arbitrary mdspans on device. To compile this function in a * non-CUDA translation unit, `raft/core/copy.hpp` may be used instead. The * pure C++ header will correctly compile even without a CUDA compiler. * Depending on the specialization, this CUDA header may invoke the kernel and * therefore require a CUDA compiler. * * Limitations: Currently this function does not support copying directly * between two arbitrary mdspans on different CUDA devices. It is assumed that the caller sets the * correct CUDA device. Furthermore, host-to-host copies that require a transformation of the * underlying memory layout are currently not performant, although they are supported. * * Note that when copying to an mdspan with a non-unique layout (i.e. the same * underlying memory is addressed by different element indexes), the source * data must contain non-unique values for every non-unique destination * element. If this is not the case, the behavior is undefined. Some copies * to non-unique layouts which are well-defined will nevertheless fail with an * exception to avoid race conditions in the underlying copy. * * @tparam DstType An mdspan type for the destination container. * @tparam SrcType An mdspan type for the source container * @param res raft::resources used to provide a stream for copies involving the * device. * @param dst The destination mdspan. * @param src The source mdspan. */ template <typename DstType, typename SrcType> detail::mdspan_copyable_with_kernel_t<DstType, SrcType> copy(resources const& res, DstType&& dst, SrcType&& src) { detail::copy(res, std::forward<DstType>(dst), std::forward<SrcType>(src)); } #ifndef RAFT_NON_CUDA_COPY_IMPLEMENTED #define RAFT_NON_CUDA_COPY_IMPLEMENTED template <typename DstType, typename SrcType> detail::mdspan_copyable_not_with_kernel_t<DstType, SrcType> copy(resources const& res, DstType&& dst, SrcType&& src) { detail::copy(res, std::forward<DstType>(dst), std::forward<SrcType>(src)); } #endif } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/cuda_support.hpp
/* * Copyright (c) 2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once namespace raft { #ifndef RAFT_DISABLE_CUDA auto constexpr static const CUDA_ENABLED = true; #else auto constexpr static const CUDA_ENABLED = false; #endif } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/error.hpp
/* * Copyright (c) 2019-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #ifndef __RAFT_RT_ERROR #define __RAFT_RT_ERROR #pragma once #include <cstdio> #include <execinfo.h> #include <iostream> #include <memory> #include <sstream> #include <stdexcept> #include <string> #include <vector> namespace raft { /** * @defgroup error_handling Exceptions & Error Handling * @{ */ /** base exception class for the whole of raft */ class exception : public std::exception { public: /** default ctor */ explicit exception() noexcept : std::exception(), msg_() {} /** copy ctor */ exception(exception const& src) noexcept : std::exception(), msg_(src.what()) { collect_call_stack(); } /** ctor from an input message */ explicit exception(std::string const msg) noexcept : std::exception(), msg_(std::move(msg)) { collect_call_stack(); } /** get the message associated with this exception */ char const* what() const noexcept override { return msg_.c_str(); } private: /** message associated with this exception */ std::string msg_; /** append call stack info to this exception's message for ease of debug */ // Courtesy: https://www.gnu.org/software/libc/manual/html_node/Backtraces.html void collect_call_stack() noexcept { #ifdef __GNUC__ constexpr int kMaxStackDepth = 64; void* stack[kMaxStackDepth]; // NOLINT auto depth = backtrace(stack, kMaxStackDepth); std::ostringstream oss; oss << std::endl << "Obtained " << depth << " stack frames" << std::endl; char** strings = backtrace_symbols(stack, depth); if (strings == nullptr) { oss << "But no stack trace could be found!" << std::endl; msg_ += oss.str(); return; } ///@todo: support for demangling of C++ symbol names for (int i = 0; i < depth; ++i) { oss << "#" << i << " in " << strings[i] << std::endl; } free(strings); msg_ += oss.str(); #endif // __GNUC__ } }; /** * @brief Exception thrown when logical precondition is violated. * * This exception should not be thrown directly and is instead thrown by the * RAFT_EXPECTS and RAFT_FAIL macros. * */ struct logic_error : public raft::exception { explicit logic_error(char const* const message) : raft::exception(message) {} explicit logic_error(std::string const& message) : raft::exception(message) {} }; /** * @brief Exception thrown when attempting to use CUDA features from a non-CUDA * build * */ struct non_cuda_build_error : public raft::exception { explicit non_cuda_build_error(char const* const message) : raft::exception(message) {} explicit non_cuda_build_error(std::string const& message) : raft::exception(message) {} }; /** * @} */ } // namespace raft // FIXME: Need to be replaced with RAFT_FAIL /** macro to throw a runtime error */ #define THROW(fmt, ...) \ do { \ int size1 = \ std::snprintf(nullptr, 0, "exception occurred! file=%s line=%d: ", __FILE__, __LINE__); \ int size2 = std::snprintf(nullptr, 0, fmt, ##__VA_ARGS__); \ if (size1 < 0 || size2 < 0) \ throw raft::exception("Error in snprintf, cannot handle raft exception."); \ auto size = size1 + size2 + 1; /* +1 for final '\0' */ \ auto buf = std::make_unique<char[]>(size_t(size)); \ std::snprintf(buf.get(), \ size1 + 1 /* +1 for '\0' */, \ "exception occurred! file=%s line=%d: ", \ __FILE__, \ __LINE__); \ std::snprintf(buf.get() + size1, size2 + 1 /* +1 for '\0' */, fmt, ##__VA_ARGS__); \ std::string msg(buf.get(), buf.get() + size - 1); /* -1 to remove final '\0' */ \ throw raft::exception(msg); \ } while (0) // FIXME: Need to be replaced with RAFT_EXPECTS /** macro to check for a conditional and assert on failure */ #define ASSERT(check, fmt, ...) \ do { \ if (!(check)) THROW(fmt, ##__VA_ARGS__); \ } while (0) /** * Macro to append error message to first argument. * This should only be called in contexts where it is OK to throw exceptions! */ #define SET_ERROR_MSG(msg, location_prefix, fmt, ...) \ do { \ int size1 = std::snprintf(nullptr, 0, "%s", location_prefix); \ int size2 = std::snprintf(nullptr, 0, "file=%s line=%d: ", __FILE__, __LINE__); \ int size3 = std::snprintf(nullptr, 0, fmt, ##__VA_ARGS__); \ if (size1 < 0 || size2 < 0 || size3 < 0) \ throw raft::exception("Error in snprintf, cannot handle raft exception."); \ auto size = size1 + size2 + size3 + 1; /* +1 for final '\0' */ \ std::vector<char> buf(size); \ std::snprintf(buf.data(), size1 + 1 /* +1 for '\0' */, "%s", location_prefix); \ std::snprintf( \ buf.data() + size1, size2 + 1 /* +1 for '\0' */, "file=%s line=%d: ", __FILE__, __LINE__); \ std::snprintf(buf.data() + size1 + size2, size3 + 1 /* +1 for '\0' */, fmt, ##__VA_ARGS__); \ msg += std::string(buf.data(), buf.data() + size - 1); /* -1 to remove final '\0' */ \ } while (0) /** * @defgroup assertion Assertion and error macros * @{ */ /** * @brief Macro for checking (pre-)conditions that throws an exception when a condition is false * * @param[in] cond Expression that evaluates to true or false * @param[in] fmt String literal description of the reason that cond is expected to be true with * optional format tagas * @throw raft::logic_error if the condition evaluates to false. */ #define RAFT_EXPECTS(cond, fmt, ...) \ do { \ if (!(cond)) { \ std::string msg{}; \ SET_ERROR_MSG(msg, "RAFT failure at ", fmt, ##__VA_ARGS__); \ throw raft::logic_error(msg); \ } \ } while (0) /** * @brief Indicates that an erroneous code path has been taken. * * @param[in] fmt String literal description of the reason that this code path is erroneous with * optional format tagas * @throw always throws raft::logic_error */ #define RAFT_FAIL(fmt, ...) \ do { \ std::string msg{}; \ SET_ERROR_MSG(msg, "RAFT failure at ", fmt, ##__VA_ARGS__); \ throw raft::logic_error(msg); \ } while (0) /** * @} */ #endif
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/temporary_device_buffer.hpp
/* * Copyright (c) 2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include "device_mdarray.hpp" #include "device_mdspan.hpp" #include <raft/core/resource/cuda_stream.hpp> #include <raft/util/cudart_utils.hpp> #include <variant> namespace raft { /** * \defgroup temporary_device_buffer `raft::temporary_device_buffer` * @{ */ /** * @brief An object which provides temporary access on-device to memory from either a host or device * pointer. This object provides a `view()` method that will provide a `raft::device_mdspan` that * may be read-only depending on const-qualified nature of the input pointer. * * @tparam ElementType type of the input * @tparam Extents raft::extents * @tparam LayoutPolicy layout of the input * @tparam ContainerPolicy container to be used to own device memory if needed */ template <typename ElementType, typename Extents, typename LayoutPolicy = layout_c_contiguous, template <typename> typename ContainerPolicy = device_uvector_policy> class temporary_device_buffer { using view_type = device_mdspan<ElementType, Extents, LayoutPolicy>; using index_type = typename Extents::index_type; using element_type = std::remove_cv_t<ElementType>; using container_policy = ContainerPolicy<element_type>; using owning_device_buffer = device_mdarray<element_type, Extents, LayoutPolicy, container_policy>; using data_store = std::variant<ElementType*, owning_device_buffer>; static constexpr bool is_const_pointer_ = std::is_const_v<ElementType>; public: temporary_device_buffer(temporary_device_buffer const&) = delete; temporary_device_buffer& operator=(temporary_device_buffer const&) = delete; constexpr temporary_device_buffer(temporary_device_buffer&&) = default; constexpr temporary_device_buffer& operator=(temporary_device_buffer&&) = default; /** * @brief Construct a new temporary device buffer object * * @param handle raft::resources * @param data input pointer * @param extents dimensions of input array * @param write_back if true, any writes to the `view()` of this object will be copid * back if the original pointer was in host memory */ temporary_device_buffer(resources const& handle, ElementType* data, Extents extents, bool write_back = false) : stream_(resource::get_cuda_stream(handle)), original_data_(data), extents_{extents}, write_back_(write_back), length_([this]() { std::size_t length = 1; for (std::size_t i = 0; i < extents_.rank(); ++i) { length *= extents_.extent(i); } return length; }()), device_id_{get_device_for_address(data)} { if (device_id_ == -1) { typename owning_device_buffer::mapping_type layout{extents_}; typename owning_device_buffer::container_policy_type policy{}; owning_device_buffer device_data{handle, layout, policy}; raft::copy(device_data.data_handle(), data, length_, resource::get_cuda_stream(handle)); data_ = data_store{std::in_place_index<1>, std::move(device_data)}; } else { data_ = data_store{std::in_place_index<0>, data}; } } ~temporary_device_buffer() noexcept(is_const_pointer_) { // only need to write data back for non const pointers // when write_back=true and original pointer is in // host memory if constexpr (not is_const_pointer_) { if (write_back_ && device_id_ == -1) { raft::copy(original_data_, std::get<1>(data_).data_handle(), length_, stream_); } } } /** * @brief Returns a `raft::device_mdspan` * * @return raft::device_mdspan */ auto view() -> view_type { if (device_id_ == -1) { return std::get<1>(data_).view(); } else { return make_mdspan<ElementType, index_type, LayoutPolicy, false, true>(original_data_, extents_); } } private: rmm::cuda_stream_view stream_; ElementType* original_data_; data_store data_; Extents extents_; bool write_back_; std::size_t length_; int device_id_; }; /**@}*/ /** * \defgroup temporary_device_buffer_factories Temporary device buffer factories * @{ */ /** * @brief Factory to create a `raft::temporary_device_buffer` * * @code{.cpp} * #include <raft/core/resources.hpp> * * raft::resources handle; * * // Initialize raft::device_mdarray and raft::extents * // Can be either raft::device_mdarray or raft::host_mdarray * auto exts = raft::make_extents<int>(5); * auto array = raft::make_device_mdarray<int, int>(handle, exts); * * auto d_buf = raft::make_temporary_device_buffer(handle, array.data_handle(), exts); * @endcode * * @tparam ElementType type of the input * @tparam IndexType index type of `raft::extents` * @tparam LayoutPolicy layout of the input * @tparam ContainerPolicy container to be used to own device memory if needed * @tparam Extents variadic dimensions for `raft::extents` * @param handle raft::resources * @param data input pointer * @param extents dimensions of input array * @param write_back if true, any writes to the `view()` of this object will be copid * back if the original pointer was in host memory * @return raft::temporary_device_buffer */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous, template <typename> typename ContainerPolicy = device_uvector_policy, size_t... Extents> auto make_temporary_device_buffer(raft::resources const& handle, ElementType* data, raft::extents<IndexType, Extents...> extents, bool write_back = false) { return temporary_device_buffer<ElementType, decltype(extents), LayoutPolicy, ContainerPolicy>( handle, data, extents, write_back); } /** * @brief Factory to create a `raft::temporary_device_buffer` which produces a * read-only `raft::device_mdspan` from `view()` method with * `write_back=false` * * @code{.cpp} * #include <raft/core/resources.hpp> * * raft::resources handle; * * // Initialize raft::device_mdarray and raft::extents * // Can be either raft::device_mdarray or raft::host_mdarray * auto exts = raft::make_extents<int>(5); * auto array = raft::make_device_mdarray<int, int>(handle, exts); * * auto d_buf = raft::make_readonly_temporary_device_buffer(handle, array.data_handle(), exts); * @endcode * * @tparam ElementType type of the input * @tparam IndexType index type of `raft::extents` * @tparam LayoutPolicy layout of the input * @tparam ContainerPolicy container to be used to own device memory if needed * @tparam Extents variadic dimensions for `raft::extents` * @param handle raft::resources * @param data input pointer * @param extents dimensions of input array * @return raft::temporary_device_buffer */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous, template <typename> typename ContainerPolicy = device_uvector_policy, size_t... Extents> auto make_readonly_temporary_device_buffer(raft::resources const& handle, ElementType* data, raft::extents<IndexType, Extents...> extents) { return temporary_device_buffer<std::add_const_t<ElementType>, decltype(extents), LayoutPolicy, ContainerPolicy>(handle, data, extents, false); } /** * @brief Factory to create a `raft::temporary_device_buffer` which produces a * writeable `raft::device_mdspan` from `view()` method with * `write_back=true` * * @code{.cpp} * #include <raft/core/resources.hpp> * * raft::resources handle; * * // Initialize raft::host_mdarray and raft::extents * // Can be either raft::device_mdarray or raft::host_mdarray * auto exts = raft::make_extents<int>(5); * auto array = raft::make_host_mdarray<int, int>(handle, exts); * * auto d_buf = raft::make_writeback_temporary_device_buffer(handle, array.data_handle(), exts); * @endcode * * @tparam ElementType type of the input * @tparam IndexType index type of `raft::extents` * @tparam LayoutPolicy layout of the input * @tparam ContainerPolicy container to be used to own device memory if needed * @tparam Extents variadic dimensions for `raft::extents` * @param handle raft::resources * @param data input pointer * @param extents dimensions of input array * @return raft::temporary_device_buffer */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous, template <typename> typename ContainerPolicy = device_uvector_policy, size_t... Extents, typename = std::enable_if_t<not std::is_const_v<ElementType>>> auto make_writeback_temporary_device_buffer(raft::resources const& handle, ElementType* data, raft::extents<IndexType, Extents...> extents) { return temporary_device_buffer<ElementType, decltype(extents), LayoutPolicy, ContainerPolicy>( handle, data, extents, true); } /**@}*/ } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/serialize.hpp
/* * Copyright (c) 2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/detail/mdspan_numpy_serializer.hpp> #include <raft/core/device_mdspan.hpp> #include <raft/core/host_mdspan.hpp> #include <raft/core/resource/cuda_stream.hpp> #include <raft/core/resources.hpp> #include <iostream> #include <vector> /** * Collection of serialization functions for RAFT data types */ namespace raft { template <typename ElementType, typename Extents, typename LayoutPolicy, typename AccessorPolicy> inline void serialize_mdspan( const raft::resources&, std::ostream& os, const raft::host_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>& obj) { detail::numpy_serializer::serialize_host_mdspan(os, obj); } template <typename ElementType, typename Extents, typename LayoutPolicy, typename AccessorPolicy> inline void serialize_mdspan( const raft::resources& handle, std::ostream& os, const raft::device_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>& obj) { static_assert(std::is_same_v<LayoutPolicy, raft::layout_c_contiguous> || std::is_same_v<LayoutPolicy, raft::layout_f_contiguous>, "The serializer only supports row-major and column-major layouts"); using obj_t = raft::device_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>; // Copy to host before serializing // For contiguous layouts, size() == product of dimensions std::vector<typename obj_t::value_type> tmp(obj.size()); cudaStream_t stream = resource::get_cuda_stream(handle); raft::update_host(tmp.data(), obj.data_handle(), obj.size(), stream); resource::sync_stream(handle); using inner_accessor_type = typename obj_t::accessor_type::accessor_type; auto tmp_mdspan = raft::host_mdspan<ElementType, Extents, LayoutPolicy, raft::host_accessor<inner_accessor_type>>( tmp.data(), obj.extents()); detail::numpy_serializer::serialize_host_mdspan(os, tmp_mdspan); } template <typename ElementType, typename Extents, typename LayoutPolicy, typename AccessorPolicy> inline void serialize_mdspan( const raft::resources&, std::ostream& os, const raft::managed_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>& obj) { using obj_t = raft::managed_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>; using inner_accessor_type = typename obj_t::accessor_type::accessor_type; auto tmp_mdspan = raft::host_mdspan<ElementType, Extents, LayoutPolicy, raft::host_accessor<inner_accessor_type>>( obj.data_handle(), obj.extents()); detail::numpy_serializer::serialize_host_mdspan(os, tmp_mdspan); } template <typename ElementType, typename Extents, typename LayoutPolicy, typename AccessorPolicy> inline void deserialize_mdspan( const raft::resources&, std::istream& is, raft::host_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>& obj) { detail::numpy_serializer::deserialize_host_mdspan(is, obj); } template <typename ElementType, typename Extents, typename LayoutPolicy, typename AccessorPolicy> inline void deserialize_mdspan( const raft::resources& handle, std::istream& is, raft::device_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>& obj) { static_assert(std::is_same_v<LayoutPolicy, raft::layout_c_contiguous> || std::is_same_v<LayoutPolicy, raft::layout_f_contiguous>, "The serializer only supports row-major and column-major layouts"); using obj_t = raft::device_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>; // Copy to device after serializing // For contiguous layouts, size() == product of dimensions std::vector<typename obj_t::value_type> tmp(obj.size()); using inner_accessor_type = typename obj_t::accessor_type::accessor_type; auto tmp_mdspan = raft::host_mdspan<ElementType, Extents, LayoutPolicy, raft::host_accessor<inner_accessor_type>>( tmp.data(), obj.extents()); detail::numpy_serializer::deserialize_host_mdspan(is, tmp_mdspan); cudaStream_t stream = resource::get_cuda_stream(handle); raft::update_device(obj.data_handle(), tmp.data(), obj.size(), stream); resource::sync_stream(handle); } template <typename ElementType, typename Extents, typename LayoutPolicy, typename AccessorPolicy> inline void deserialize_mdspan( const raft::resources& handle, std::istream& is, raft::host_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>&& obj) { deserialize_mdspan(handle, is, obj); } template <typename ElementType, typename Extents, typename LayoutPolicy, typename AccessorPolicy> inline void deserialize_mdspan( const raft::resources& handle, std::istream& is, raft::managed_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>& obj) { using obj_t = raft::managed_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>; using inner_accessor_type = typename obj_t::accessor_type::accessor_type; auto tmp_mdspan = raft::host_mdspan<ElementType, Extents, LayoutPolicy, raft::host_accessor<inner_accessor_type>>( obj.data_handle(), obj.extents()); detail::numpy_serializer::deserialize_host_mdspan(is, tmp_mdspan); } template <typename ElementType, typename Extents, typename LayoutPolicy, typename AccessorPolicy> inline void deserialize_mdspan( const raft::resources& handle, std::istream& is, raft::managed_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>&& obj) { deserialize_mdspan(handle, is, obj); } template <typename ElementType, typename Extents, typename LayoutPolicy, typename AccessorPolicy> inline void deserialize_mdspan( const raft::resources& handle, std::istream& is, raft::device_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>&& obj) { deserialize_mdspan(handle, is, obj); } template <typename T> inline void serialize_scalar(const raft::resources&, std::ostream& os, const T& value) { detail::numpy_serializer::serialize_scalar(os, value); } template <typename T> inline T deserialize_scalar(const raft::resources&, std::istream& is) { return detail::numpy_serializer::deserialize_scalar<T>(is); } } // end namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/sparse_types.hpp
/* * Copyright (c) 2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/logger.hpp> #include <raft/core/resources.hpp> #include <raft/core/span.hpp> #include <raft/core/sparse_types.hpp> namespace raft { /** * \defgroup sparse_types Sparse API vocabulary * @{ */ enum SparsityType { OWNING, PRESERVING }; /** * Maintains metadata about the structure and sparsity of a sparse matrix. * @tparam RowType * @tparam ColType * @tparam NZType * @tparam is_device */ template <typename RowType, typename ColType, typename NZType, int is_device> class sparse_structure { public: using row_type = RowType; using col_type = ColType; using nnz_type = NZType; /** * Constructor when sparsity is already known * @param n_rows total number of rows in matrix * @param n_cols total number of columns in matrix * @param nnz sparsity of matrix */ sparse_structure(row_type n_rows, col_type n_cols, nnz_type nnz) : n_rows_(n_rows), n_cols_(n_cols), nnz_(nnz){}; /** * Constructor when sparsity is not yet known * @param n_rows total number of rows in matrix * @param n_cols total number of columns in matrix */ sparse_structure(row_type n_rows, col_type n_cols) : n_rows_(n_rows), n_cols_(n_cols), nnz_(0) {} /** * Return the sparsity of the matrix (this will be 0 when sparsity is not yet known) * @return sparsity of matrix */ nnz_type get_nnz() { return nnz_; } /** * Return the total number of rows in the matrix * @return total number of rows in the matriz */ row_type get_n_rows() { return n_rows_; } /** * Return the total number of columns in the matrix * @return total number of columns */ col_type get_n_cols() { return n_cols_; } /** * Initialize the matrix sparsity when it was not known * upon construction. * @param nnz */ virtual void initialize_sparsity(nnz_type nnz) { nnz_ = nnz; } protected: row_type n_rows_; col_type n_cols_; nnz_type nnz_; }; /** * A non-owning view of a sparse matrix, which includes a * structure component coupled with its elements/weights * * @tparam ElementType * @tparam sparse_structure */ template <typename ElementType, typename StructureType, bool is_device> class sparse_matrix_view { public: using element_type = ElementType; using structure_view_type = typename StructureType::view_type; sparse_matrix_view(raft::span<ElementType, is_device> element_span, structure_view_type structure_view) : element_span_(element_span), structure_view_(structure_view) { // FIXME: Validate structure sizes match span size. } /** * Return a view of the structure underlying this matrix * @return */ structure_view_type structure_view() { return structure_view_; } /** * Return a span of the nonzero elements of the matrix * @return span of the nonzero elements of the matrix */ span<element_type, is_device> get_elements() { return element_span_; } protected: raft::span<element_type, is_device> element_span_; structure_view_type structure_view_; }; /** * TODO: Need to support the following types of configurations: * 1. solid: immutable_sparse_matrix_view<const ElementType, const StructureType> * - This is an immutable view type, nothing can change. * 2. liquid: sparse_matrix<ElementType, const StructureType> * - sparse_matrix owning container w/ StructureType=immutable view? * 3. gas: sparse_matrix<ElementType, StructureType> * - sparse_matrix owning container w/ StructureType owning container? */ /** * An owning container for a sparse matrix, which includes a * structure component coupled with its elements/weights * @tparam ElementType * @tparam sparse_structure * @tparam ContainerPolicy */ template <typename ElementType, typename StructureType, typename ViewType, bool is_device, template <typename T> typename ContainerPolicy> class sparse_matrix { public: using view_type = ViewType; using element_type = typename view_type::element_type; using structure_type = StructureType; using row_type = typename structure_type::row_type; using col_type = typename structure_type::col_type; using nnz_type = typename structure_type::nnz_type; using structure_view_type = typename structure_type::view_type; using container_policy_type = ContainerPolicy<element_type>; using container_type = typename container_policy_type::container_type; // constructor that owns the data and the structure sparse_matrix(raft::resources const& handle, row_type n_rows, col_type n_cols, nnz_type nnz = 0) noexcept(std::is_nothrow_default_constructible_v<container_type>) : structure_{handle, n_rows, n_cols, nnz}, cp_{}, c_elements_{cp_.create(handle, 0)} {}; // Constructor that owns the data but not the structure // This constructor is only callable with a `structure_type == *_structure_view` // which makes it okay to copy sparse_matrix(raft::resources const& handle, structure_type structure) noexcept( std::is_nothrow_default_constructible_v<container_type>) : structure_{structure}, cp_{}, c_elements_{cp_.create(handle, structure_.get_nnz())} {}; constexpr sparse_matrix(sparse_matrix const&) noexcept( std::is_nothrow_copy_constructible_v<container_type>) = default; constexpr sparse_matrix(sparse_matrix&&) noexcept( std::is_nothrow_move_constructible<container_type>::value) = default; constexpr auto operator=(sparse_matrix const&) noexcept( std::is_nothrow_copy_assignable<container_type>::value) -> sparse_matrix& = default; constexpr auto operator=(sparse_matrix&&) noexcept( std::is_nothrow_move_assignable<container_type>::value) -> sparse_matrix& = default; ~sparse_matrix() noexcept(std::is_nothrow_destructible<container_type>::value) = default; void initialize_sparsity(nnz_type nnz) { c_elements_.resize(nnz); }; raft::span<ElementType, is_device> get_elements() { return raft::span<ElementType, is_device>(c_elements_.data(), structure_.get_nnz()); } /** * Return a view of the structure underlying this matrix * @return */ virtual structure_view_type structure_view() = 0; /** * Return a sparsity-preserving view of this sparse matrix * @return view of this sparse matrix */ view_type view() { auto struct_view = structure_view(); auto element_span = raft::span<ElementType, is_device>(c_elements_.data(), struct_view.get_nnz()); return view_type(element_span, struct_view); } protected: structure_type structure_; container_policy_type cp_; container_type c_elements_; }; /* @} */ } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/device_container_policy.hpp
/* * Copyright (2019) Sandia Corporation * * The source code is licensed under the 3-clause BSD license found in the LICENSE file * thirdparty/LICENSES/mdarray.license */ /* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/device_mdspan.hpp> #include <raft/util/cudart_utils.hpp> #include <raft/core/detail/span.hpp> // dynamic_extent #include <raft/core/host_device_accessor.hpp> #include <raft/core/resource/cuda_stream.hpp> #include <raft/core/resource/device_memory_resource.hpp> #include <rmm/cuda_stream_view.hpp> #include <rmm/device_uvector.hpp> #include <rmm/mr/device/device_memory_resource.hpp> #include <thrust/device_ptr.h> namespace raft { /** * @brief A simplified version of thrust::device_reference with support for CUDA stream. */ template <typename T> class device_reference { public: using value_type = typename std::remove_cv_t<T>; using pointer = thrust::device_ptr<T>; using const_pointer = thrust::device_ptr<T const>; private: std::conditional_t<std::is_const<T>::value, const_pointer, pointer> ptr_; rmm::cuda_stream_view stream_; public: device_reference(thrust::device_ptr<T> ptr, rmm::cuda_stream_view stream) : ptr_{ptr}, stream_{stream} { } operator value_type() const // NOLINT { auto* raw = ptr_.get(); value_type v{}; update_host(&v, raw, 1, stream_); return v; } auto operator=(T const& other) -> device_reference& { auto* raw = ptr_.get(); update_device(raw, &other, 1, stream_); return *this; } }; /** * @brief A thin wrapper over rmm::device_uvector for implementing the mdarray container policy. * */ template <typename T> class device_uvector { rmm::device_uvector<T> data_; public: using value_type = T; using size_type = std::size_t; using reference = device_reference<T>; using const_reference = device_reference<T const>; using pointer = value_type*; using const_pointer = value_type const*; using iterator = pointer; using const_iterator = const_pointer; public: ~device_uvector() = default; device_uvector(device_uvector&&) noexcept = default; device_uvector(device_uvector const& that) : data_{that.data_, that.data_.stream()} {} auto operator=(device_uvector<T> const& that) -> device_uvector<T>& { data_ = rmm::device_uvector<T>{that.data_, that.data_.stream()}; return *this; } auto operator=(device_uvector<T>&& that) noexcept -> device_uvector<T>& = default; /** * @brief Default ctor is deleted as it doesn't accept stream. */ device_uvector() = delete; /** * @brief Ctor that accepts a size, stream and an optional mr. */ explicit device_uvector(std::size_t size, rmm::cuda_stream_view stream) : data_{size, stream} {} /** * @brief Ctor that accepts a size, stream and a memory resource. */ explicit device_uvector(std::size_t size, rmm::cuda_stream_view stream, rmm::mr::device_memory_resource* mr) : data_{size, stream, mr} { } /** * @brief Index operator that returns a proxy to the actual data. */ template <typename Index> auto operator[](Index i) noexcept -> reference { return device_reference<T>{thrust::device_ptr<T>{data_.data() + i}, data_.stream()}; } /** * @brief Index operator that returns a proxy to the actual data. */ template <typename Index> auto operator[](Index i) const noexcept { return device_reference<T const>{thrust::device_ptr<T const>{data_.data() + i}, data_.stream()}; } void resize(size_type size) { data_.resize(size, data_.stream()); } [[nodiscard]] auto data() noexcept -> pointer { return data_.data(); } [[nodiscard]] auto data() const noexcept -> const_pointer { return data_.data(); } }; /** * @brief A container policy for device mdarray. */ template <typename ElementType> class device_uvector_policy { public: using element_type = ElementType; using container_type = device_uvector<element_type>; // FIXME(jiamingy): allocator type is not supported by rmm::device_uvector using pointer = typename container_type::pointer; using const_pointer = typename container_type::const_pointer; using reference = device_reference<element_type>; using const_reference = device_reference<element_type const>; using accessor_policy = std::experimental::default_accessor<element_type>; using const_accessor_policy = std::experimental::default_accessor<element_type const>; public: auto create(raft::resources const& res, size_t n) -> container_type { if (mr_ == nullptr) { // NB: not using the workspace resource by default! // The workspace resource is for short-lived temporary allocations. return container_type(n, resource::get_cuda_stream(res)); } else { return container_type(n, resource::get_cuda_stream(res), mr_); } } constexpr device_uvector_policy() = default; constexpr explicit device_uvector_policy(rmm::mr::device_memory_resource* mr) noexcept : mr_(mr) { } [[nodiscard]] constexpr auto access(container_type& c, size_t n) const noexcept -> reference { return c[n]; } [[nodiscard]] constexpr auto access(container_type const& c, size_t n) const noexcept -> const_reference { return c[n]; } [[nodiscard]] auto make_accessor_policy() noexcept { return accessor_policy{}; } [[nodiscard]] auto make_accessor_policy() const noexcept { return const_accessor_policy{}; } private: rmm::mr::device_memory_resource* mr_{nullptr}; }; } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/copy.hpp
/* * Copyright (c) 2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/detail/copy.hpp> namespace raft { #ifndef RAFT_NON_CUDA_COPY_IMPLEMENTED #define RAFT_NON_CUDA_COPY_IMPLEMENTED /** * @brief Copy data from one mdspan to another with the same extents * * This function copies data from one mdspan to another, regardless of whether * or not the mdspans have the same layout, memory type (host/device/managed) * or data type. So long as it is possible to convert the data type from source * to destination, and the extents are equal, this function should be able to * perform the copy. * * This header does _not_ include the custom kernel used for copying data * between completely arbitrary mdspans on device. For arbitrary copies of this * kind, `#include <raft/core/copy.cuh>` instead. Specializations of this * function that require the custom kernel will be SFINAE-omitted when this * header is used instead of `copy.cuh`. This header _does_ support * device-to-device copies that can be performed with cuBLAS or a * straightforward cudaMemcpy. Any necessary device operations will be stream-ordered via the CUDA * stream provided by the `raft::resources` argument. * * Limitations: Currently this function does not support copying directly * between two arbitrary mdspans on different CUDA devices. It is assumed that the caller sets the * correct CUDA device. Furthermore, host-to-host copies that require a transformation of the * underlying memory layout are currently not performant, although they are supported. * * Note that when copying to an mdspan with a non-unique layout (i.e. the same * underlying memory is addressed by different element indexes), the source * data must contain non-unique values for every non-unique destination * element. If this is not the case, the behavior is undefined. Some copies * to non-unique layouts which are well-defined will nevertheless fail with an * exception to avoid race conditions in the underlying copy. * * @tparam DstType An mdspan type for the destination container. * @tparam SrcType An mdspan type for the source container * @param res raft::resources used to provide a stream for copies involving the * device. * @param dst The destination mdspan. * @param src The source mdspan. */ template <typename DstType, typename SrcType> detail::mdspan_copyable_not_with_kernel_t<DstType, SrcType> copy(resources const& res, DstType&& dst, SrcType&& src) { detail::copy(res, std::forward<DstType>(dst), std::forward<SrcType>(src)); } #endif } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/host_coo_matrix.hpp
/* * Copyright (c) 2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/coo_matrix.hpp> #include <raft/core/host_container_policy.hpp> #include <raft/core/host_span.hpp> #include <raft/core/sparse_types.hpp> namespace raft { /** * \defgroup host_coo_matrix Host COO Matrix * @{ */ /** * Specialization for a sparsity-preserving coordinate structure view which uses host memory */ template <typename RowType, typename ColType, typename NZType> using host_coordinate_structure_view = coordinate_structure_view<RowType, ColType, NZType, false>; /** * Specialization for a sparsity-owning coordinate structure which uses host memory */ template <typename RowType, typename ColType, typename NZType, template <typename T> typename ContainerPolicy = host_vector_policy> using host_coordinate_structure = coordinate_structure<RowType, ColType, NZType, false, ContainerPolicy>; /** * Specialization for a coo matrix view which uses host memory */ template <typename ElementType, typename RowType, typename ColType, typename NZType> using host_coo_matrix_view = coo_matrix_view<ElementType, RowType, ColType, NZType, false>; template <typename ElementType, typename RowType, typename ColType, typename NZType, template <typename T> typename ContainerPolicy = host_vector_policy, SparsityType sparsity_type = SparsityType::OWNING> using host_coo_matrix = coo_matrix<ElementType, RowType, ColType, NZType, false, ContainerPolicy, sparsity_type>; /** * Specialization for a sparsity-owning coo matrix which uses host memory */ template <typename ElementType, typename RowType, typename ColType, typename NZType, template <typename T> typename ContainerPolicy = host_vector_policy> using host_sparsity_owning_coo_matrix = coo_matrix<ElementType, RowType, ColType, NZType, false, ContainerPolicy>; template <typename ElementType, typename RowType, typename ColType, typename NZType, template <typename T> typename ContainerPolicy = host_vector_policy> using host_sparsity_preserving_coo_matrix = coo_matrix<ElementType, RowType, ColType, NZType, false, ContainerPolicy, SparsityType::PRESERVING>; template <typename T> struct is_host_coo_matrix_view : std::false_type {}; template <typename ElementType, typename RowType, typename ColType, typename NZType> struct is_host_coo_matrix_view<host_coo_matrix_view<ElementType, RowType, ColType, NZType>> : std::true_type {}; template <typename T> constexpr bool is_host_coo_matrix_view_v = is_host_coo_matrix_view<T>::value; template <typename T> struct is_host_coo_matrix : std::false_type {}; template <typename ElementType, typename RowType, typename ColType, typename NZType, template <typename T> typename ContainerPolicy, SparsityType sparsity_type> struct is_host_coo_matrix< host_coo_matrix<ElementType, RowType, ColType, NZType, ContainerPolicy, sparsity_type>> : std::true_type {}; template <typename T> constexpr bool is_host_coo_matrix_v = is_host_coo_matrix<T>::value; template <typename T> constexpr bool is_host_coo_sparsity_owning_v = is_host_coo_matrix<T>::value and T::get_sparsity_type() == OWNING; template <typename T> constexpr bool is_host_coo_sparsity_preserving_v = is_host_coo_matrix<T>::value and T::get_sparsity_type() == PRESERVING; /** * Create a sparsity-owning sparse matrix in the coordinate format. sparsity-owning means that * all of the underlying vectors (data, indptr, indices) are owned by the coo_matrix instance. If * not known up front, the sparsity can be ignored in this factory function and `resize()` invoked * on the instance once the sparsity is known. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/host_coo_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * * raft::resources handle; * coo_matrix = raft::make_host_coo_matrix(handle, n_rows, n_cols); * ... * // compute expected sparsity * ... * int nnz = 5000; * coo_matrix.initialize_sparsity(nnz); * @endcode * * @tparam ElementType * @tparam RowType * @tparam ColType * @tparam NZType * @param[in] handle raft handle for managing expensive resources * @param[in] n_rows total number of rows in the matrix * @param[in] n_cols total number of columns in the matrix * @param[in] nnz number of non-zeros in the matrix if known [optional] * @return a sparsity-owning sparse matrix in coordinate (coo) format */ template <typename ElementType, typename RowType, typename ColType, typename NZType> auto make_host_coo_matrix(raft::resources const& handle, RowType n_rows, ColType n_cols, NZType nnz = 0) { return host_sparsity_owning_coo_matrix<ElementType, RowType, ColType, NZType>( handle, n_rows, n_cols, nnz); } /** * Create a sparsity-preserving sparse matrix in the coordinate format. sparsity-preserving means * that a view of the coo sparsity is supplied, allowing the values in the sparsity to change but * not the sparsity itself. The coo_matrix instance does not own the sparsity, the sparsity must * be known up front, and cannot be resized later. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/host_coo_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * * raft::resources handle; * coo_structure = raft::make_host_coordinate_structure(handle, n_rows, n_cols); * ... * // compute expected sparsity * ... * coo_structure.initialize_sparsity(nnz); * coo_matrix = raft::make_host_coo_matrix(handle, coo_structure.view()); * @endcode * * @tparam ElementType * @tparam RowType * @tparam ColType * @tparam NZType * @param[in] handle raft handle for managing expensive resources * @param[in] structure a sparsity-preserving coordinate structural view * @return a sparsity-preserving sparse matrix in coordinate (coo) format */ template <typename ElementType, typename RowType, typename ColType, typename NZType> auto make_host_coo_matrix(raft::resources const& handle, host_coordinate_structure_view<RowType, ColType, NZType> structure) { return host_sparsity_preserving_coo_matrix<ElementType, RowType, ColType, NZType>(handle, structure); } /** * Create a non-owning sparse matrix view in the coordinate format. This is sparsity-preserving, * meaning that the underlying sparsity is known and cannot be changed. Use the sparsity-owning * coo_matrix if sparsity needs to be mutable. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/host_coo_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * int nnz = 5000; * * // The following pointer is assumed to reference host-accessible memory for a size of nnz * float* h_elm_ptr = ...; * * raft::resources handle; * coo_structure = raft::make_host_coordinate_structure(handle, n_rows, n_cols, nnz); * coo_matrix_view = raft::make_host_coo_matrix_view(handle, h_elm_ptr, coo_structure.view()); * @endcode * * @tparam ElementType * @tparam RowType * @tparam ColType * @tparam NZType * @param[in] ptr a pointer to array of nonzero matrix elements on host (size nnz) * @param[in] structure a sparsity-preserving coordinate structural view * @return a sparsity-preserving sparse matrix in coordinate (coo) format */ template <typename ElementType, typename RowType, typename ColType, typename NZType> auto make_host_coo_matrix_view(ElementType* ptr, host_coordinate_structure_view<RowType, ColType, NZType> structure) { return host_coo_matrix_view<ElementType, RowType, ColType, NZType>( raft::host_span<ElementType>(ptr, structure.get_nnz()), structure); } /** * Create a non-owning sparse matrix view in the coordinate format. This is sparsity-preserving, * meaning that the underlying sparsity is known and cannot be changed. Use the sparsity-owning * coo_matrix if sparsity needs to be mutable. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/host_span.hpp> * #include <raft/core/host_coo_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * int nnz = 5000; * * // The following span is assumed to be of size nnz * raft::host_span<float> h_elm_ptr; * * raft::resources handle; * coo_structure = raft::make_host_coordinate_structure(handle, n_rows, n_cols, nnz); * coo_matrix_view = raft::make_host_coo_matrix_view(handle, h_elm_ptr, coo_structure.view()); * @endcode * * @tparam ElementType * @tparam RowType * @tparam ColType * @tparam NZType * @param[in] elements a host span containing nonzero matrix elements (size nnz) * @param[in] structure a sparsity-preserving coordinate structural view * @return */ template <typename ElementType, typename RowType, typename ColType, typename NZType> auto make_host_coo_matrix_view(raft::host_span<ElementType> elements, host_coordinate_structure_view<RowType, ColType, NZType> structure) { RAFT_EXPECTS(elements.size() == structure.get_nnz(), "Size of elements must be equal to the nnz from the structure"); return host_coo_matrix_view<ElementType, RowType, ColType, NZType>(elements, structure); } /** * Create a sparsity-owning coordinate structure object. If not known up front, this object can be * resized() once the sparsity (number of non-zeros) is known, postponing the allocation of the * underlying data arrays. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/host_coo_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * int nnz = 5000; * * raft::resources handle; * coo_structure = raft::make_host_coordinate_structure(handle, n_rows, n_cols, nnz); * * ... * // compute expected sparsity * ... * coo_structure.initialize_sparsity(nnz); * @endcode * * @tparam RowType * @tparam ColType * @tparam NZType * @param[in] handle raft handle for managing expensive resources on host * @param[in] n_rows total number of rows * @param[in] n_cols total number of cols * @param[in] nnz number of non-zeros * @return a sparsity-owning coordinate structure instance */ template <typename RowType, typename ColType, typename NZType> auto make_host_coordinate_structure(raft::resources const& handle, RowType n_rows, ColType n_cols, NZType nnz = 0) { return host_coordinate_structure<RowType, ColType, NZType>(handle, n_rows, n_cols, nnz); } /** * Create a non-owning sparsity-preserved coordinate structure view. Sparsity-preserving means that * the underlying sparsity is known and cannot be changed. Use the sparsity-owning version if the * sparsity is not known up front. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/host_coo_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * int nnz = 5000; * * // The following pointers are assumed to reference host-accessible memory of size nnz * int *rows = ...; * int *cols = ...; * * raft::resources handle; * coo_structure = raft::make_host_coordinate_structure_view(handle, rows, cols, n_rows, n_cols, * nnz); * @endcode * * @tparam RowType * @tparam ColType * @tparam NZType * @param[in] rows pointer to row indices array on host (size nnz) * @param[in] cols pointer to column indices array on host (size nnz) * @param[in] n_rows total number of rows * @param[in] n_cols total number of columns * @param[in] nnz number of non-zeros * @return a sparsity-preserving coordinate structural view */ template <typename RowType, typename ColType, typename NZType> auto make_host_coordinate_structure_view( RowType* rows, ColType* cols, RowType n_rows, ColType n_cols, NZType nnz) { return host_coordinate_structure_view<RowType, ColType, NZType>( raft::host_span<RowType>(rows, nnz), raft::host_span<ColType>(cols, nnz), n_rows, n_cols); } /** * Create a non-owning sparsity-preserved coordinate structure view. Sparsity-preserving means that * the underlying sparsity is known and cannot be changed. Use the sparsity-owning version if the * sparsity is not known up front. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/host_coo_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * int nnz = 5000; * * // The following host spans are assumed to be of size nnz * raft::host_span<int> rows; * raft::host_span<int> cols; * * raft::resources handle; * coo_structure = raft::make_host_coordinate_structure_view(handle, rows, cols, n_rows, n_cols); * @endcode * * @tparam RowType * @tparam ColType * @tparam NZType * @param[in] rows a host span containing row indices (size nnz) * @param[in] cols a host span containing column indices (size nnz) * @param[in] n_rows total number of rows * @param[in] n_cols total number of columns * @return a sparsity-preserving coordinate structural view */ template <typename RowType, typename ColType, typename NZType> auto make_host_coordinate_structure_view(raft::host_span<RowType> rows, raft::host_span<ColType> cols, RowType n_rows, ColType n_cols) { return host_coordinate_structure_view<RowType, ColType, NZType>(rows, cols, n_rows, n_cols); } /** @} */ }; // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/device_mdspan.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cstdint> #include <raft/core/host_device_accessor.hpp> #include <raft/core/mdspan.hpp> #include <raft/core/memory_type.hpp> namespace raft { template <typename AccessorPolicy> using device_accessor = host_device_accessor<AccessorPolicy, memory_type::device>; template <typename AccessorPolicy> using managed_accessor = host_device_accessor<AccessorPolicy, memory_type::managed>; /** * @brief std::experimental::mdspan with device tag to avoid accessing incorrect memory location. */ template <typename ElementType, typename Extents, typename LayoutPolicy = layout_c_contiguous, typename AccessorPolicy = std::experimental::default_accessor<ElementType>> using device_mdspan = mdspan<ElementType, Extents, LayoutPolicy, device_accessor<AccessorPolicy>>; template <typename ElementType, typename Extents, typename LayoutPolicy = layout_c_contiguous, typename AccessorPolicy = std::experimental::default_accessor<ElementType>> using managed_mdspan = mdspan<ElementType, Extents, LayoutPolicy, managed_accessor<AccessorPolicy>>; template <typename T, bool B> struct is_device_mdspan : std::false_type {}; template <typename T> struct is_device_mdspan<T, true> : std::bool_constant<T::accessor_type::is_device_accessible> {}; /** * @\brief Boolean to determine if template type T is either raft::device_mdspan or a derived type */ template <typename T> using is_device_mdspan_t = is_device_mdspan<T, is_mdspan_v<T>>; template <typename T> using is_input_device_mdspan_t = is_device_mdspan<T, is_input_mdspan_v<T>>; template <typename T> using is_output_device_mdspan_t = is_device_mdspan<T, is_output_mdspan_v<T>>; template <typename T, bool B> struct is_managed_mdspan : std::false_type {}; template <typename T> struct is_managed_mdspan<T, true> : std::bool_constant<T::accessor_type::is_managed_accessible> {}; /** * @\brief Boolean to determine if template type T is either raft::managed_mdspan or a derived type */ template <typename T> using is_managed_mdspan_t = is_managed_mdspan<T, is_mdspan_v<T>>; template <typename T> using is_input_managed_mdspan_t = is_managed_mdspan<T, is_input_mdspan_v<T>>; template <typename T> using is_output_managed_mdspan_t = is_managed_mdspan<T, is_output_mdspan_v<T>>; /** * @\brief Boolean to determine if variadic template types Tn are either raft::device_mdspan or a * derived type */ template <typename... Tn> inline constexpr bool is_device_mdspan_v = std::conjunction_v<is_device_mdspan_t<Tn>...>; template <typename... Tn> inline constexpr bool is_input_device_mdspan_v = std::conjunction_v<is_input_device_mdspan_t<Tn>...>; template <typename... Tn> inline constexpr bool is_output_device_mdspan_v = std::conjunction_v<is_output_device_mdspan_t<Tn>...>; template <typename... Tn> using enable_if_device_mdspan = std::enable_if_t<is_device_mdspan_v<Tn...>>; template <typename... Tn> using enable_if_input_device_mdspan = std::enable_if_t<is_input_device_mdspan_v<Tn...>>; template <typename... Tn> using enable_if_output_device_mdspan = std::enable_if_t<is_output_device_mdspan_v<Tn...>>; /** * @\brief Boolean to determine if variadic template types Tn are either raft::managed_mdspan or a * derived type */ template <typename... Tn> inline constexpr bool is_managed_mdspan_v = std::conjunction_v<is_managed_mdspan_t<Tn>...>; template <typename... Tn> inline constexpr bool is_input_managed_mdspan_v = std::conjunction_v<is_input_managed_mdspan_t<Tn>...>; template <typename... Tn> inline constexpr bool is_output_managed_mdspan_v = std::conjunction_v<is_output_managed_mdspan_t<Tn>...>; template <typename... Tn> using enable_if_managed_mdspan = std::enable_if_t<is_managed_mdspan_v<Tn...>>; template <typename... Tn> using enable_if_input_managed_mdspan = std::enable_if_t<is_input_managed_mdspan_v<Tn...>>; template <typename... Tn> using enable_if_output_managed_mdspan = std::enable_if_t<is_output_managed_mdspan_v<Tn...>>; /** * @brief Shorthand for 0-dim host mdspan (scalar). * @tparam ElementType the data type of the scalar element * @tparam IndexType the index type of the extents */ template <typename ElementType, typename IndexType = std::uint32_t> using device_scalar_view = device_mdspan<ElementType, scalar_extent<IndexType>>; /** * @brief Shorthand for 1-dim device mdspan. * @tparam ElementType the data type of the vector elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> using device_vector_view = device_mdspan<ElementType, vector_extent<IndexType>, LayoutPolicy>; /** * @brief Shorthand for c-contiguous device matrix view. * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> using device_matrix_view = device_mdspan<ElementType, matrix_extent<IndexType>, LayoutPolicy>; /** * @brief Shorthand for 128 byte aligned device matrix view. * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy must be of type layout_{left/right}_padded */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_right_padded<ElementType>, typename = enable_if_layout_padded<ElementType, LayoutPolicy>> using device_aligned_matrix_view = device_mdspan<ElementType, matrix_extent<IndexType>, LayoutPolicy, std::experimental::aligned_accessor<ElementType, detail::alignment::value>>; /** * @brief Create a 2-dim 128 byte aligned mdspan instance for device pointer. It's * expected that the given layout policy match the layout of the underlying * pointer. * @tparam ElementType the data type of the matrix elements * @tparam LayoutPolicy must be of type layout_{left/right}_padded * @tparam IndexType the index type of the extents * @param[in] ptr on device to wrap * @param[in] n_rows number of rows in pointer * @param[in] n_cols number of columns in pointer */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_right_padded<ElementType>> auto make_device_aligned_matrix_view(ElementType* ptr, IndexType n_rows, IndexType n_cols) { using data_handle_type = typename std::experimental::aligned_accessor<ElementType, detail::alignment::value>::data_handle_type; static_assert(std::is_same<LayoutPolicy, layout_left_padded<ElementType>>::value || std::is_same<LayoutPolicy, layout_right_padded<ElementType>>::value); assert(reinterpret_cast<std::uintptr_t>(ptr) == std::experimental::details::alignTo(reinterpret_cast<std::uintptr_t>(ptr), detail::alignment::value)); data_handle_type aligned_pointer = ptr; matrix_extent<IndexType> extents{n_rows, n_cols}; return device_aligned_matrix_view<ElementType, IndexType, LayoutPolicy>{aligned_pointer, extents}; } /** * @brief Create a raft::managed_mdspan * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering * @param ptr Pointer to the data * @param exts dimensionality of the array (series of integers) * @return raft::managed_mdspan */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous, size_t... Extents> auto make_managed_mdspan(ElementType* ptr, extents<IndexType, Extents...> exts) { return make_mdspan<ElementType, IndexType, LayoutPolicy, true, true>(ptr, exts); } /** * @brief Create a 0-dim (scalar) mdspan instance for device value. * * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @param[in] ptr on device to wrap */ template <typename ElementType, typename IndexType = std::uint32_t> auto make_device_scalar_view(ElementType* ptr) { scalar_extent<IndexType> extents; return device_scalar_view<ElementType, IndexType>{ptr, extents}; } /** * @brief Create a 2-dim c-contiguous mdspan instance for device pointer. It's * expected that the given layout policy match the layout of the underlying * pointer. * @tparam ElementType the data type of the matrix elements * @tparam LayoutPolicy policy for strides and layout ordering * @tparam IndexType the index type of the extents * @param[in] ptr on device to wrap * @param[in] n_rows number of rows in pointer * @param[in] n_cols number of columns in pointer */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> auto make_device_matrix_view(ElementType* ptr, IndexType n_rows, IndexType n_cols) { matrix_extent<IndexType> extents{n_rows, n_cols}; return device_matrix_view<ElementType, IndexType, LayoutPolicy>{ptr, extents}; } /** * @brief Create a 2-dim mdspan instance for device pointer with a strided layout * that is restricted to stride 1 in the trailing dimension. It's * expected that the given layout policy match the layout of the underlying * pointer. * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering * @param[in] ptr on device to wrap * @param[in] n_rows number of rows in pointer * @param[in] n_cols number of columns in pointer * @param[in] stride leading dimension / stride of data */ template <typename ElementType, typename IndexType, typename LayoutPolicy = layout_c_contiguous> auto make_device_strided_matrix_view(ElementType* ptr, IndexType n_rows, IndexType n_cols, IndexType stride) { constexpr auto is_row_major = std::is_same_v<LayoutPolicy, layout_c_contiguous>; IndexType stride0 = is_row_major ? (stride > 0 ? stride : n_cols) : 1; IndexType stride1 = is_row_major ? 1 : (stride > 0 ? stride : n_rows); assert(is_row_major ? stride0 >= n_cols : stride1 >= n_rows); matrix_extent<IndexType> extents{n_rows, n_cols}; auto layout = make_strided_layout(extents, std::array<IndexType, 2>{stride0, stride1}); return device_matrix_view<ElementType, IndexType, layout_stride>{ptr, layout}; } /** * @brief Create a 1-dim mdspan instance for device pointer. * @tparam ElementType the data type of the vector elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering * @param[in] ptr on device to wrap * @param[in] n number of elements in pointer * @return raft::device_vector_view */ template <typename ElementType, typename IndexType, typename LayoutPolicy = layout_c_contiguous> auto make_device_vector_view(ElementType* ptr, IndexType n) { return device_vector_view<ElementType, IndexType, LayoutPolicy>{ptr, n}; } /** * @brief Create a 1-dim mdspan instance for device pointer. * @tparam ElementType the data type of the vector elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering * @param[in] ptr on device to wrap * @param[in] mapping The layout mapping to use for this vector * @return raft::device_vector_view */ template <typename ElementType, typename IndexType, typename LayoutPolicy = layout_c_contiguous> auto make_device_vector_view( ElementType* ptr, const typename LayoutPolicy::template mapping<vector_extent<IndexType>>& mapping) { return device_vector_view<ElementType, IndexType, LayoutPolicy>{ptr, mapping}; } /** * @brief Construct a strided vector layout mapping * * Usage example: * @code{.cpp} * #include <raft/core/device_mdspan.hpp> * * int n_elements = 10; * int stride = 10; * auto vector = raft::make_device_vector_view(vector_ptr, * raft::make_vector_strided_layout(n_elements, stride)); * @endcode * * @tparam IndexType the index type of the extents * @param[in] n the number of elements in the vector * @param[in] stride the stride between elements in the vector */ template <typename IndexType> auto make_vector_strided_layout(IndexType n, IndexType stride) { return make_strided_layout(vector_extent<IndexType>{n}, std::array<IndexType, 1>{stride}); } } // end namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/cusparse_macros.hpp
/* * Copyright (c) 2019-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cusparse.h> #include <raft/core/error.hpp> ///@todo: enable this once logging is enabled // #include <cuml/common/logger.hpp> #define _CUSPARSE_ERR_TO_STR(err) \ case err: return #err; // Notes: //(1.) CUDA_VER_10_1_UP aggregates all the CUDA version selection logic; //(2.) to enforce a lower version, // //`#define CUDA_ENFORCE_LOWER // #include <raft/sparse/detail/cusparse_wrappers.h>` // // (i.e., before including this header) // #define CUDA_VER_10_1_UP (CUDART_VERSION >= 10100) namespace raft { /** * @ingroup error_handling * @{ */ /** * @brief Exception thrown when a cuSparse error is encountered. */ struct cusparse_error : public raft::exception { explicit cusparse_error(char const* const message) : raft::exception(message) {} explicit cusparse_error(std::string const& message) : raft::exception(message) {} }; /** * @} */ namespace sparse { namespace detail { inline const char* cusparse_error_to_string(cusparseStatus_t err) { #if defined(CUDART_VERSION) && CUDART_VERSION >= 10100 return cusparseGetErrorString(err); #else // CUDART_VERSION switch (err) { _CUSPARSE_ERR_TO_STR(CUSPARSE_STATUS_SUCCESS); _CUSPARSE_ERR_TO_STR(CUSPARSE_STATUS_NOT_INITIALIZED); _CUSPARSE_ERR_TO_STR(CUSPARSE_STATUS_ALLOC_FAILED); _CUSPARSE_ERR_TO_STR(CUSPARSE_STATUS_INVALID_VALUE); _CUSPARSE_ERR_TO_STR(CUSPARSE_STATUS_ARCH_MISMATCH); _CUSPARSE_ERR_TO_STR(CUSPARSE_STATUS_EXECUTION_FAILED); _CUSPARSE_ERR_TO_STR(CUSPARSE_STATUS_INTERNAL_ERROR); _CUSPARSE_ERR_TO_STR(CUSPARSE_STATUS_MATRIX_TYPE_NOT_SUPPORTED); default: return "CUSPARSE_STATUS_UNKNOWN"; }; #endif // CUDART_VERSION } } // namespace detail } // namespace sparse } // namespace raft #undef _CUSPARSE_ERR_TO_STR /** * @ingroup assertion * @{ */ /** * @brief Error checking macro for cuSparse runtime API functions. * * Invokes a cuSparse runtime API function call, if the call does not return * CUSPARSE_STATUS_SUCCESS, throws an exception detailing the cuSparse error that occurred */ #define RAFT_CUSPARSE_TRY(call) \ do { \ cusparseStatus_t const status = (call); \ if (CUSPARSE_STATUS_SUCCESS != status) { \ std::string msg{}; \ SET_ERROR_MSG(msg, \ "cuSparse error encountered at: ", \ "call='%s', Reason=%d:%s", \ #call, \ status, \ raft::sparse::detail::cusparse_error_to_string(status)); \ throw raft::cusparse_error(msg); \ } \ } while (0) /** * @} */ // FIXME: Remove after consumer rename #ifndef CUSPARSE_TRY #define CUSPARSE_TRY(call) RAFT_CUSPARSE_TRY(call) #endif // FIXME: Remove after consumer rename #ifndef CUSPARSE_CHECK #define CUSPARSE_CHECK(call) CUSPARSE_TRY(call) #endif /** * @ingroup assertion * @{ */ //@todo: use logger here once logging is enabled /** check for cusparse runtime API errors but do not assert */ #define RAFT_CUSPARSE_TRY_NO_THROW(call) \ do { \ cusparseStatus_t err = call; \ if (err != CUSPARSE_STATUS_SUCCESS) { \ printf("CUSPARSE call='%s' got errorcode=%d err=%s", \ #call, \ err, \ raft::sparse::detail::cusparse_error_to_string(err)); \ } \ } while (0) /** * @} */ // FIXME: Remove after consumer rename #ifndef CUSPARSE_CHECK_NO_THROW #define CUSPARSE_CHECK_NO_THROW(call) RAFT_CUSPARSE_TRY_NO_THROW(call) #endif
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/host_span.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/span.hpp> namespace raft { /** * @defgroup host_span one-dimensional device span type * @{ */ /** * @brief A span class for host pointer. */ template <typename T, size_t extent = std::experimental::dynamic_extent> using host_span = span<T, false, extent>; /** * @} */ } // end namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/device_setter.hpp
/* * Copyright (c) 2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cuda_runtime_api.h> #include <raft/core/logger.hpp> #include <raft/util/cuda_rt_essentials.hpp> namespace raft { /** * @brief A scoped setter for the active CUDA device * * On construction, the device_setter will set the active CUDA device to the * indicated value. On deletion, the active CUDA device will be set back to * its previous value. If the call to set the new active device fails, an * exception will be thrown. If the call to set the device back to its * previously selected value throws, an error will be logged, but no * exception will be thrown. * * @param int device_id The ID of the CUDA device to make active * */ struct device_setter { /** * Return the id of the current device as an integer */ static auto get_current_device() { auto result = int{}; RAFT_CUDA_TRY(cudaGetDevice(&result)); return result; } /** * Return the count of currently available CUDA devices */ static auto get_device_count() { auto result = int{}; RAFT_CUDA_TRY(cudaGetDeviceCount(&result)); return result; } explicit device_setter(int new_device) : prev_device_{get_current_device()} { RAFT_CUDA_TRY(cudaSetDevice(new_device)); } ~device_setter() { RAFT_CUDA_TRY_NO_THROW(cudaSetDevice(prev_device_)); } private: int prev_device_; }; } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/comms.hpp
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cuda_runtime.h> #include <memory> #include <raft/core/error.hpp> #include <vector> namespace raft { namespace comms { /** * @defgroup comms_types Common mnmg comms types * @{ */ typedef unsigned int request_t; enum class datatype_t { CHAR, UINT8, INT32, UINT32, INT64, UINT64, FLOAT32, FLOAT64 }; enum class op_t { SUM, PROD, MIN, MAX }; /** * The resulting status of distributed stream synchronization */ enum class status_t { SUCCESS, // Synchronization successful ERROR, // An error occurred querying sync status ABORT // A failure occurred in sync, queued operations aborted }; template <typename value_t> constexpr datatype_t get_type(); template <> constexpr datatype_t get_type<char>() { return datatype_t::CHAR; } template <> constexpr datatype_t get_type<uint8_t>() { return datatype_t::UINT8; } template <> constexpr datatype_t get_type<int>() { return datatype_t::INT32; } template <> constexpr datatype_t get_type<uint32_t>() { return datatype_t::UINT32; } template <> constexpr datatype_t get_type<int64_t>() { return datatype_t::INT64; } template <> constexpr datatype_t get_type<uint64_t>() { return datatype_t::UINT64; } template <> constexpr datatype_t get_type<float>() { return datatype_t::FLOAT32; } template <> constexpr datatype_t get_type<double>() { return datatype_t::FLOAT64; } /** * @} */ /** * @defgroup comms_iface MNMG Communicator Interface * @{ */ class comms_iface { public: virtual ~comms_iface() {} virtual int get_size() const = 0; virtual int get_rank() const = 0; virtual std::unique_ptr<comms_iface> comm_split(int color, int key) const = 0; virtual void barrier() const = 0; virtual status_t sync_stream(cudaStream_t stream) const = 0; virtual void isend(const void* buf, size_t size, int dest, int tag, request_t* request) const = 0; virtual void irecv(void* buf, size_t size, int source, int tag, request_t* request) const = 0; virtual void waitall(int count, request_t array_of_requests[]) const = 0; virtual void allreduce(const void* sendbuff, void* recvbuff, size_t count, datatype_t datatype, op_t op, cudaStream_t stream) const = 0; virtual void bcast( void* buff, size_t count, datatype_t datatype, int root, cudaStream_t stream) const = 0; virtual void bcast(const void* sendbuff, void* recvbuff, size_t count, datatype_t datatype, int root, cudaStream_t stream) const = 0; virtual void reduce(const void* sendbuff, void* recvbuff, size_t count, datatype_t datatype, op_t op, int root, cudaStream_t stream) const = 0; virtual void allgather(const void* sendbuff, void* recvbuff, size_t sendcount, datatype_t datatype, cudaStream_t stream) const = 0; virtual void allgatherv(const void* sendbuf, void* recvbuf, const size_t* recvcounts, const size_t* displs, datatype_t datatype, cudaStream_t stream) const = 0; virtual void gather(const void* sendbuff, void* recvbuff, size_t sendcount, datatype_t datatype, int root, cudaStream_t stream) const = 0; virtual void gatherv(const void* sendbuf, void* recvbuf, size_t sendcount, const size_t* recvcounts, const size_t* displs, datatype_t datatype, int root, cudaStream_t stream) const = 0; virtual void reducescatter(const void* sendbuff, void* recvbuff, size_t recvcount, datatype_t datatype, op_t op, cudaStream_t stream) const = 0; // if a thread is sending & receiving at the same time, use device_sendrecv to avoid deadlock virtual void device_send(const void* buf, size_t size, int dest, cudaStream_t stream) const = 0; // if a thread is sending & receiving at the same time, use device_sendrecv to avoid deadlock virtual void device_recv(void* buf, size_t size, int source, cudaStream_t stream) const = 0; virtual void device_sendrecv(const void* sendbuf, size_t sendsize, int dest, void* recvbuf, size_t recvsize, int source, cudaStream_t stream) const = 0; virtual void device_multicast_sendrecv(const void* sendbuf, std::vector<size_t> const& sendsizes, std::vector<size_t> const& sendoffsets, std::vector<int> const& dests, void* recvbuf, std::vector<size_t> const& recvsizes, std::vector<size_t> const& recvoffsets, std::vector<int> const& sources, cudaStream_t stream) const = 0; virtual void group_start() const = 0; virtual void group_end() const = 0; }; /** * @} */ /** * @defgroup comms_t Base Communicator Proxy * @{ */ class comms_t { public: comms_t(std::unique_ptr<comms_iface> impl) : impl_(impl.release()) { ASSERT(nullptr != impl_.get(), "ERROR: Invalid comms_iface used!"); } /** * Virtual Destructor to enable polymorphism */ virtual ~comms_t() {} /** * Returns the size of the communicator clique */ int get_size() const { return impl_->get_size(); } /** * Returns the local rank */ int get_rank() const { return impl_->get_rank(); } /** * Splits the current communicator clique into sub-cliques matching * the given color and key * * @param color ranks w/ the same color are placed in the same communicator * @param key controls rank assignment */ std::unique_ptr<comms_iface> comm_split(int color, int key) const { return impl_->comm_split(color, key); } /** * Performs a collective barrier synchronization */ void barrier() const { impl_->barrier(); } /** * Some collective communications implementations (eg. NCCL) might use asynchronous * collectives that are explicitly synchronized. It's important to always synchronize * using this method to allow failures to propagate, rather than `cudaStreamSynchronize()`, * to prevent the potential for deadlocks. * * @param stream the cuda stream to sync collective operations on */ status_t sync_stream(cudaStream_t stream) const { return impl_->sync_stream(stream); } /** * Performs an asynchronous point-to-point send * @tparam value_t the type of data to send * @param buf pointer to array of data to send * @param size number of elements in buf * @param dest destination rank * @param tag a tag to use for the receiver to filter * @param request pointer to hold returned request_t object. * This will be used in `waitall()` to synchronize until the message is delivered (or fails). */ template <typename value_t> void isend(const value_t* buf, size_t size, int dest, int tag, request_t* request) const { impl_->isend(static_cast<const void*>(buf), size * sizeof(value_t), dest, tag, request); } /** * Performs an asynchronous point-to-point receive * @tparam value_t the type of data to be received * @param buf pointer to (initialized) array that will hold received data * @param size number of elements in buf * @param source source rank * @param tag a tag to use for message filtering * @param request pointer to hold returned request_t object. * This will be used in `waitall()` to synchronize until the message is delivered (or fails). */ template <typename value_t> void irecv(value_t* buf, size_t size, int source, int tag, request_t* request) const { impl_->irecv(static_cast<void*>(buf), size * sizeof(value_t), source, tag, request); } /** * Synchronize on an array of request_t objects returned from isend/irecv * @param count number of requests to synchronize on * @param array_of_requests an array of request_t objects returned from isend/irecv */ void waitall(int count, request_t array_of_requests[]) const { impl_->waitall(count, array_of_requests); } /** * Perform an allreduce collective * @tparam value_t datatype of underlying buffers * @param sendbuff data to reduce * @param recvbuff buffer to hold the reduced result * @param count number of elements in sendbuff * @param op reduction operation to perform * @param stream CUDA stream to synchronize operation */ template <typename value_t> void allreduce( const value_t* sendbuff, value_t* recvbuff, size_t count, op_t op, cudaStream_t stream) const { impl_->allreduce(static_cast<const void*>(sendbuff), static_cast<void*>(recvbuff), count, get_type<value_t>(), op, stream); } /** * Broadcast data from one rank to the rest * @tparam value_t datatype of underlying buffers * @param buff buffer to send * @param count number of elements if buff * @param root the rank initiating the broadcast * @param stream CUDA stream to synchronize operation */ template <typename value_t> void bcast(value_t* buff, size_t count, int root, cudaStream_t stream) const { impl_->bcast(static_cast<void*>(buff), count, get_type<value_t>(), root, stream); } /** * Broadcast data from one rank to the rest * @tparam value_t datatype of underlying buffers * @param sendbuff buffer containing data to broadcast (only used in root) * @param recvbuff buffer to receive broadcasted data * @param count number of elements if buff * @param root the rank initiating the broadcast * @param stream CUDA stream to synchronize operation */ template <typename value_t> void bcast( const value_t* sendbuff, value_t* recvbuff, size_t count, int root, cudaStream_t stream) const { impl_->bcast(static_cast<const void*>(sendbuff), static_cast<void*>(recvbuff), count, get_type<value_t>(), root, stream); } /** * Reduce data from many ranks down to a single rank * @tparam value_t datatype of underlying buffers * @param sendbuff buffer containing data to reduce * @param recvbuff buffer containing reduced data (only needs to be initialized on root) * @param count number of elements in sendbuff * @param op reduction operation to perform * @param root rank to store the results * @param stream CUDA stream to synchronize operation */ template <typename value_t> void reduce(const value_t* sendbuff, value_t* recvbuff, size_t count, op_t op, int root, cudaStream_t stream) const { impl_->reduce(static_cast<const void*>(sendbuff), static_cast<void*>(recvbuff), count, get_type<value_t>(), op, root, stream); } /** * Gathers data from each rank onto all ranks * @tparam value_t datatype of underlying buffers * @param sendbuff buffer containing data to gather * @param recvbuff buffer containing gathered data from all ranks * @param sendcount number of elements in send buffer * @param stream CUDA stream to synchronize operation */ template <typename value_t> void allgather(const value_t* sendbuff, value_t* recvbuff, size_t sendcount, cudaStream_t stream) const { impl_->allgather(static_cast<const void*>(sendbuff), static_cast<void*>(recvbuff), sendcount, get_type<value_t>(), stream); } /** * Gathers data from all ranks and delivers to combined data to all ranks * @tparam value_t datatype of underlying buffers * @param sendbuf buffer containing data to send * @param recvbuf buffer containing data to receive * @param recvcounts pointer to an array (of length num_ranks size) containing the number of * elements that are to be received from each rank * @param displs pointer to an array (of length num_ranks size) to specify the displacement * (relative to recvbuf) at which to place the incoming data from each rank * @param stream CUDA stream to synchronize operation */ template <typename value_t> void allgatherv(const value_t* sendbuf, value_t* recvbuf, const size_t* recvcounts, const size_t* displs, cudaStream_t stream) const { impl_->allgatherv(static_cast<const void*>(sendbuf), static_cast<void*>(recvbuf), recvcounts, displs, get_type<value_t>(), stream); } /** * Gathers data from each rank onto all ranks * @tparam value_t datatype of underlying buffers * @param sendbuff buffer containing data to gather * @param recvbuff buffer containing gathered data from all ranks * @param sendcount number of elements in send buffer * @param root rank to store the results * @param stream CUDA stream to synchronize operation */ template <typename value_t> void gather(const value_t* sendbuff, value_t* recvbuff, size_t sendcount, int root, cudaStream_t stream) const { impl_->gather(static_cast<const void*>(sendbuff), static_cast<void*>(recvbuff), sendcount, get_type<value_t>(), root, stream); } /** * Gathers data from all ranks and delivers to combined data to all ranks * @tparam value_t datatype of underlying buffers * @param sendbuf buffer containing data to send * @param recvbuf buffer containing data to receive * @param sendcount number of elements in send buffer * @param recvcounts pointer to an array (of length num_ranks size) containing the number of * elements that are to be received from each rank * @param displs pointer to an array (of length num_ranks size) to specify the displacement * (relative to recvbuf) at which to place the incoming data from each rank * @param root rank to store the results * @param stream CUDA stream to synchronize operation */ template <typename value_t> void gatherv(const value_t* sendbuf, value_t* recvbuf, size_t sendcount, const size_t* recvcounts, const size_t* displs, int root, cudaStream_t stream) const { impl_->gatherv(static_cast<const void*>(sendbuf), static_cast<void*>(recvbuf), sendcount, recvcounts, displs, get_type<value_t>(), root, stream); } /** * Reduces data from all ranks then scatters the result across ranks * @tparam value_t datatype of underlying buffers * @param sendbuff buffer containing data to send (size recvcount * num_ranks) * @param recvbuff buffer containing received data * @param recvcount number of items to receive * @param op reduction operation to perform * @param stream CUDA stream to synchronize operation */ template <typename value_t> void reducescatter(const value_t* sendbuff, value_t* recvbuff, size_t recvcount, op_t op, cudaStream_t stream) const { impl_->reducescatter(static_cast<const void*>(sendbuff), static_cast<void*>(recvbuff), recvcount, get_type<value_t>(), op, stream); } /** * Performs a point-to-point send * * if a thread is sending & receiving at the same time, use device_sendrecv to avoid deadlock. * * @tparam value_t the type of data to send * @param buf pointer to array of data to send * @param size number of elements in buf * @param dest destination rank * @param stream CUDA stream to synchronize operation */ template <typename value_t> void device_send(const value_t* buf, size_t size, int dest, cudaStream_t stream) const { impl_->device_send(static_cast<const void*>(buf), size * sizeof(value_t), dest, stream); } /** * Performs a point-to-point receive * * if a thread is sending & receiving at the same time, use device_sendrecv to avoid deadlock. * * @tparam value_t the type of data to be received * @param buf pointer to (initialized) array that will hold received data * @param size number of elements in buf * @param source source rank * @param stream CUDA stream to synchronize operation */ template <typename value_t> void device_recv(value_t* buf, size_t size, int source, cudaStream_t stream) const { impl_->device_recv(static_cast<void*>(buf), size * sizeof(value_t), source, stream); } /** * Performs a point-to-point send/receive * * @tparam value_t the type of data to be sent & received * @param sendbuf pointer to array of data to send * @param sendsize number of elements in sendbuf * @param dest destination rank * @param recvbuf pointer to (initialized) array that will hold received data * @param recvsize number of elements in recvbuf * @param source source rank * @param stream CUDA stream to synchronize operation */ template <typename value_t> void device_sendrecv(const value_t* sendbuf, size_t sendsize, int dest, value_t* recvbuf, size_t recvsize, int source, cudaStream_t stream) const { impl_->device_sendrecv(static_cast<const void*>(sendbuf), sendsize * sizeof(value_t), dest, static_cast<void*>(recvbuf), recvsize * sizeof(value_t), source, stream); } /** * Performs a multicast send/receive * * @tparam value_t the type of data to be sent & received * @param sendbuf pointer to array of data to send * @param sendsizes numbers of elements to send * @param sendoffsets offsets in a number of elements from sendbuf * @param dests destination ranks * @param recvbuf pointer to (initialized) array that will hold received data * @param recvsizes numbers of elements to recv * @param recvoffsets offsets in a number of elements from recvbuf * @param sources source ranks * @param stream CUDA stream to synchronize operation */ template <typename value_t> void device_multicast_sendrecv(const value_t* sendbuf, std::vector<size_t> const& sendsizes, std::vector<size_t> const& sendoffsets, std::vector<int> const& dests, value_t* recvbuf, std::vector<size_t> const& recvsizes, std::vector<size_t> const& recvoffsets, std::vector<int> const& sources, cudaStream_t stream) const { auto sendbytesizes = sendsizes; auto sendbyteoffsets = sendoffsets; for (size_t i = 0; i < sendsizes.size(); ++i) { sendbytesizes[i] *= sizeof(value_t); sendbyteoffsets[i] *= sizeof(value_t); } auto recvbytesizes = recvsizes; auto recvbyteoffsets = recvoffsets; for (size_t i = 0; i < recvsizes.size(); ++i) { recvbytesizes[i] *= sizeof(value_t); recvbyteoffsets[i] *= sizeof(value_t); } impl_->device_multicast_sendrecv(static_cast<const void*>(sendbuf), sendbytesizes, sendbyteoffsets, dests, static_cast<void*>(recvbuf), recvbytesizes, recvbyteoffsets, sources, stream); } /** * Multiple collectives & device send/receive operations placed between group_start() and * group_end() are merged into one big operation. Internally, this function is a wrapper for * ncclGroupStart(). */ void group_start() const { impl_->group_start(); } /** * Multiple collectives & device send/receive operations placed between group_start() and * group_end() are merged into one big operation. Internally, this function is a wrapper for * ncclGroupEnd(). */ void group_end() const { impl_->group_end(); } private: std::unique_ptr<comms_iface> impl_; }; /** * @} */ } // namespace comms } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/device_mdarray.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cstdint> #include <raft/core/device_container_policy.hpp> #include <raft/core/device_mdspan.hpp> #include <raft/core/mdarray.hpp> #include <raft/core/resources.hpp> namespace raft { /** * @brief mdarray with device container policy * @tparam ElementType the data type of the elements * @tparam Extents defines the shape * @tparam LayoutPolicy policy for indexing strides and layout ordering * @tparam ContainerPolicy storage and accessor policy */ template <typename ElementType, typename Extents, typename LayoutPolicy = layout_c_contiguous, typename ContainerPolicy = device_uvector_policy<ElementType>> using device_mdarray = mdarray<ElementType, Extents, LayoutPolicy, device_accessor<ContainerPolicy>>; /** * @brief Shorthand for 0-dim host mdarray (scalar). * @tparam ElementType the data type of the scalar element * @tparam IndexType the index type of the extents */ template <typename ElementType, typename IndexType = std::uint32_t> using device_scalar = device_mdarray<ElementType, scalar_extent<IndexType>>; /** * @brief Shorthand for 1-dim device mdarray. * @tparam ElementType the data type of the vector elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> using device_vector = device_mdarray<ElementType, vector_extent<IndexType>, LayoutPolicy>; /** * @brief Shorthand for c-contiguous device matrix. * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> using device_matrix = device_mdarray<ElementType, matrix_extent<IndexType>, LayoutPolicy>; /** * @brief Create a device mdarray. * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering * @param handle raft::resources * @param exts dimensionality of the array (series of integers) * @return raft::device_mdarray */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous, size_t... Extents> auto make_device_mdarray(raft::resources const& handle, extents<IndexType, Extents...> exts) { using mdarray_t = device_mdarray<ElementType, decltype(exts), LayoutPolicy>; typename mdarray_t::mapping_type layout{exts}; typename mdarray_t::container_policy_type policy{}; return mdarray_t{handle, layout, policy}; } /** * @brief Create a device mdarray. * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering * @param handle raft::resources * @param mr rmm memory resource used for allocating the memory for the array * @param exts dimensionality of the array (series of integers) * @return raft::device_mdarray */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous, size_t... Extents> auto make_device_mdarray(raft::resources const& handle, rmm::mr::device_memory_resource* mr, extents<IndexType, Extents...> exts) { using mdarray_t = device_mdarray<ElementType, decltype(exts), LayoutPolicy>; typename mdarray_t::mapping_type layout{exts}; typename mdarray_t::container_policy_type policy{mr}; return mdarray_t{handle, layout, policy}; } /** * @brief Create a 2-dim c-contiguous device mdarray. * * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering * @param[in] handle raft handle for managing expensive resources * @param[in] n_rows number or rows in matrix * @param[in] n_cols number of columns in matrix * @return raft::device_matrix */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> auto make_device_matrix(raft::resources const& handle, IndexType n_rows, IndexType n_cols) { return make_device_mdarray<ElementType, IndexType, LayoutPolicy>( handle, make_extents<IndexType>(n_rows, n_cols)); } /** * @brief Create a device scalar from v. * * @tparam ElementType the data type of the scalar element * @tparam IndexType the index type of the extents * @param[in] handle raft handle for managing expensive cuda resources * @param[in] v scalar to wrap on device * @return raft::device_scalar */ template <typename ElementType, typename IndexType = std::uint32_t> auto make_device_scalar(raft::resources const& handle, ElementType const& v) { scalar_extent<IndexType> extents; using policy_t = typename device_scalar<ElementType>::container_policy_type; policy_t policy{}; auto scalar = device_scalar<ElementType>{handle, extents, policy}; scalar(0) = v; return scalar; } /** * @brief Create a 1-dim device mdarray. * @tparam ElementType the data type of the vector elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering * @param[in] handle raft handle for managing expensive cuda resources * @param[in] n number of elements in vector * @return raft::device_vector */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> auto make_device_vector(raft::resources const& handle, IndexType n) { return make_device_mdarray<ElementType, IndexType, LayoutPolicy>(handle, make_extents<IndexType>(n)); } } // end namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/kvp.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/detail/macros.hpp> #ifdef _RAFT_HAS_CUDA #include <cub/cub.cuh> #include <raft/util/cuda_utils.cuh> // raft::shfl_xor #endif namespace raft { /** * \brief A key identifier paired with a corresponding value * */ template <typename _Key, typename _Value> struct KeyValuePair { typedef _Key Key; ///< Key data type typedef _Value Value; ///< Value data type Key key; ///< Item key Value value; ///< Item value /// Constructor KeyValuePair() = default; #ifdef _RAFT_HAS_CUDA /// Conversion Constructor to allow integration w/ cub RAFT_INLINE_FUNCTION KeyValuePair(cub::KeyValuePair<_Key, _Value> kvp) : key(kvp.key), value(kvp.value) { } RAFT_INLINE_FUNCTION operator cub::KeyValuePair<_Key, _Value>() { return cub::KeyValuePair(key, value); } #endif /// Constructor RAFT_INLINE_FUNCTION KeyValuePair(Key const& key, Value const& value) : key(key), value(value) {} /// Inequality operator RAFT_INLINE_FUNCTION bool operator!=(const KeyValuePair& b) { return (value != b.value) || (key != b.key); } RAFT_INLINE_FUNCTION bool operator<(const KeyValuePair<_Key, _Value>& b) const { return (key < b.key) || ((key == b.key) && value < b.value); } RAFT_INLINE_FUNCTION bool operator>(const KeyValuePair<_Key, _Value>& b) const { return (key > b.key) || ((key == b.key) && value > b.value); } }; #ifdef _RAFT_HAS_CUDA template <typename _Key, typename _Value> RAFT_INLINE_FUNCTION KeyValuePair<_Key, _Value> shfl_xor(const KeyValuePair<_Key, _Value>& input, int laneMask, int width = WarpSize, uint32_t mask = 0xffffffffu) { return KeyValuePair<_Key, _Value>(shfl_xor(input.key, laneMask, width, mask), shfl_xor(input.value, laneMask, width, mask)); } #endif } // end namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/memory_type.hpp
/* * Copyright (c) 2022, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once namespace raft { enum class memory_type { host, device, managed, pinned }; auto constexpr is_device_accessible(memory_type mem_type) { return (mem_type == memory_type::device || mem_type == memory_type::managed); } auto constexpr is_host_accessible(memory_type mem_type) { return (mem_type == memory_type::host || mem_type == memory_type::managed || mem_type == memory_type::pinned); } auto constexpr is_host_device_accessible(memory_type mem_type) { return is_device_accessible(mem_type) && is_host_accessible(mem_type); } namespace detail { template <bool is_host_accessible, bool is_device_accessible> auto constexpr memory_type_from_access() { if constexpr (is_host_accessible && is_device_accessible) { return memory_type::managed; } else if constexpr (is_host_accessible) { return memory_type::host; } else if constexpr (is_device_accessible) { return memory_type::device; } static_assert(is_host_accessible || is_device_accessible, "Must be either host or device accessible to return a valid memory type"); } } // end namespace detail } // end namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/cudart_utils.hpp
/* * Copyright (c) 2019-2022, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ /** * This file is deprecated and will be removed in release 22.06. * Please use util/cudart_utils.hpp instead. */ #pragma once #include <raft/util/cudart_utils.hpp>
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/cusolver_macros.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #ifndef __RAFT_RT_CUSOLVER_MACROS_H #define __RAFT_RT_CUSOLVER_MACROS_H #pragma once #include <cusolverDn.h> #include <cusolverSp.h> ///@todo: enable this once logging is enabled // #include <cuml/common/logger.hpp> #include <raft/util/cudart_utils.hpp> #include <type_traits> #define _CUSOLVER_ERR_TO_STR(err) \ case err: return #err; namespace raft { /** * @ingroup error_handling * @{ */ /** * @brief Exception thrown when a cuSOLVER error is encountered. */ struct cusolver_error : public raft::exception { explicit cusolver_error(char const* const message) : raft::exception(message) {} explicit cusolver_error(std::string const& message) : raft::exception(message) {} }; /** * @} */ namespace linalg { namespace detail { inline const char* cusolver_error_to_string(cusolverStatus_t err) { switch (err) { _CUSOLVER_ERR_TO_STR(CUSOLVER_STATUS_SUCCESS); _CUSOLVER_ERR_TO_STR(CUSOLVER_STATUS_NOT_INITIALIZED); _CUSOLVER_ERR_TO_STR(CUSOLVER_STATUS_ALLOC_FAILED); _CUSOLVER_ERR_TO_STR(CUSOLVER_STATUS_INVALID_VALUE); _CUSOLVER_ERR_TO_STR(CUSOLVER_STATUS_ARCH_MISMATCH); _CUSOLVER_ERR_TO_STR(CUSOLVER_STATUS_EXECUTION_FAILED); _CUSOLVER_ERR_TO_STR(CUSOLVER_STATUS_INTERNAL_ERROR); _CUSOLVER_ERR_TO_STR(CUSOLVER_STATUS_MATRIX_TYPE_NOT_SUPPORTED); _CUSOLVER_ERR_TO_STR(CUSOLVER_STATUS_ZERO_PIVOT); _CUSOLVER_ERR_TO_STR(CUSOLVER_STATUS_NOT_SUPPORTED); default: return "CUSOLVER_STATUS_UNKNOWN"; }; } } // namespace detail } // namespace linalg } // namespace raft #undef _CUSOLVER_ERR_TO_STR /** * @ingroup assertion * @{ */ /** * @brief Error checking macro for cuSOLVER runtime API functions. * * Invokes a cuSOLVER runtime API function call, if the call does not return * CUSolver_STATUS_SUCCESS, throws an exception detailing the cuSOLVER error that occurred */ #define RAFT_CUSOLVER_TRY(call) \ do { \ cusolverStatus_t const status = (call); \ if (CUSOLVER_STATUS_SUCCESS != status) { \ std::string msg{}; \ SET_ERROR_MSG(msg, \ "cuSOLVER error encountered at: ", \ "call='%s', Reason=%d:%s", \ #call, \ status, \ raft::linalg::detail::cusolver_error_to_string(status)); \ throw raft::cusolver_error(msg); \ } \ } while (0) // FIXME: remove after consumer rename #ifndef CUSOLVER_TRY #define CUSOLVER_TRY(call) RAFT_CUSOLVER_TRY(call) #endif // /** // * @brief check for cuda runtime API errors but log error instead of raising // * exception. // */ #define RAFT_CUSOLVER_TRY_NO_THROW(call) \ do { \ cusolverStatus_t const status = call; \ if (CUSOLVER_STATUS_SUCCESS != status) { \ printf("CUSOLVER call='%s' at file=%s line=%d failed with %s\n", \ #call, \ __FILE__, \ __LINE__, \ raft::linalg::detail::cusolver_error_to_string(status)); \ } \ } while (0) /** * @} */ // FIXME: remove after cuml rename #ifndef CUSOLVER_CHECK #define CUSOLVER_CHECK(call) CUSOLVER_TRY(call) #endif #ifndef CUSOLVER_CHECK_NO_THROW #define CUSOLVER_CHECK_NO_THROW(call) CUSOLVER_TRY_NO_THROW(call) #endif #endif
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/bitset.cuh
/* * Copyright (c) 2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/detail/mdspan_util.cuh> // native_popc #include <raft/core/device_container_policy.hpp> #include <raft/core/device_mdarray.hpp> #include <raft/core/resource/thrust_policy.hpp> #include <raft/core/resources.hpp> #include <raft/linalg/map.cuh> #include <raft/linalg/reduce.cuh> #include <raft/util/device_atomics.cuh> #include <thrust/for_each.h> namespace raft::core { /** * @defgroup bitset Bitset * @{ */ /** * @brief View of a RAFT Bitset. * * This lightweight structure stores a pointer to a bitset in device memory with it's length. * It provides a test() device function to check if a given index is set in the bitset. * * @tparam bitset_t Underlying type of the bitset array. Default is uint32_t. * @tparam index_t Indexing type used. Default is uint32_t. */ template <typename bitset_t = uint32_t, typename index_t = uint32_t> struct bitset_view { static constexpr index_t bitset_element_size = sizeof(bitset_t) * 8; _RAFT_HOST_DEVICE bitset_view(bitset_t* bitset_ptr, index_t bitset_len) : bitset_ptr_{bitset_ptr}, bitset_len_{bitset_len} { } /** * @brief Create a bitset view from a device vector view of the bitset. * * @param bitset_span Device vector view of the bitset * @param bitset_len Number of bits in the bitset */ _RAFT_HOST_DEVICE bitset_view(raft::device_vector_view<bitset_t, index_t> bitset_span, index_t bitset_len) : bitset_ptr_{bitset_span.data_handle()}, bitset_len_{bitset_len} { } /** * @brief Device function to test if a given index is set in the bitset. * * @param sample_index Single index to test * @return bool True if index has not been unset in the bitset */ inline _RAFT_DEVICE auto test(const index_t sample_index) const -> bool { const bitset_t bit_element = bitset_ptr_[sample_index / bitset_element_size]; const index_t bit_index = sample_index % bitset_element_size; const bool is_bit_set = (bit_element & (bitset_t{1} << bit_index)) != 0; return is_bit_set; } /** * @brief Device function to test if a given index is set in the bitset. * * @param sample_index Single index to test * @return bool True if index has not been unset in the bitset */ inline _RAFT_DEVICE auto operator[](const index_t sample_index) const -> bool { return test(sample_index); } /** * @brief Device function to set a given index to set_value in the bitset. * * @param sample_index index to set * @param set_value Value to set the bit to (true or false) */ inline _RAFT_DEVICE void set(const index_t sample_index, bool set_value) const { const index_t bit_element = sample_index / bitset_element_size; const index_t bit_index = sample_index % bitset_element_size; const bitset_t bitmask = bitset_t{1} << bit_index; if (set_value) { atomicOr(bitset_ptr_ + bit_element, bitmask); } else { const bitset_t bitmask2 = ~bitmask; atomicAnd(bitset_ptr_ + bit_element, bitmask2); } } /** * @brief Get the device pointer to the bitset. */ inline _RAFT_HOST_DEVICE auto data() -> bitset_t* { return bitset_ptr_; } inline _RAFT_HOST_DEVICE auto data() const -> const bitset_t* { return bitset_ptr_; } /** * @brief Get the number of bits of the bitset representation. */ inline _RAFT_HOST_DEVICE auto size() const -> index_t { return bitset_len_; } /** * @brief Get the number of elements used by the bitset representation. */ inline _RAFT_HOST_DEVICE auto n_elements() const -> index_t { return raft::ceildiv(bitset_len_, bitset_element_size); } inline auto to_mdspan() -> raft::device_vector_view<bitset_t, index_t> { return raft::make_device_vector_view<bitset_t, index_t>(bitset_ptr_, n_elements()); } inline auto to_mdspan() const -> raft::device_vector_view<const bitset_t, index_t> { return raft::make_device_vector_view<const bitset_t, index_t>(bitset_ptr_, n_elements()); } private: bitset_t* bitset_ptr_; index_t bitset_len_; }; /** * @brief RAFT Bitset. * * This structure encapsulates a bitset in device memory. It provides a view() method to get a * device-usable lightweight view of the bitset. * Each index is represented by a single bit in the bitset. The total number of bytes used is * ceil(bitset_len / 8). * @tparam bitset_t Underlying type of the bitset array. Default is uint32_t. * @tparam index_t Indexing type used. Default is uint32_t. */ template <typename bitset_t = uint32_t, typename index_t = uint32_t> struct bitset { static constexpr index_t bitset_element_size = sizeof(bitset_t) * 8; /** * @brief Construct a new bitset object with a list of indices to unset. * * @param res RAFT resources * @param mask_index List of indices to unset in the bitset * @param bitset_len Length of the bitset * @param default_value Default value to set the bits to. Default is true. */ bitset(const raft::resources& res, raft::device_vector_view<const index_t, index_t> mask_index, index_t bitset_len, bool default_value = true) : bitset_{std::size_t(raft::ceildiv(bitset_len, bitset_element_size)), raft::resource::get_cuda_stream(res)}, bitset_len_{bitset_len} { reset(res, default_value); set(res, mask_index, !default_value); } /** * @brief Construct a new bitset object * * @param res RAFT resources * @param bitset_len Length of the bitset * @param default_value Default value to set the bits to. Default is true. */ bitset(const raft::resources& res, index_t bitset_len, bool default_value = true) : bitset_{std::size_t(raft::ceildiv(bitset_len, bitset_element_size)), resource::get_cuda_stream(res)}, bitset_len_{bitset_len} { reset(res, default_value); } // Disable copy constructor bitset(const bitset&) = delete; bitset(bitset&&) = default; bitset& operator=(const bitset&) = delete; bitset& operator=(bitset&&) = default; /** * @brief Create a device-usable view of the bitset. * * @return bitset_view<bitset_t, index_t> */ inline auto view() -> raft::core::bitset_view<bitset_t, index_t> { return bitset_view<bitset_t, index_t>(to_mdspan(), bitset_len_); } [[nodiscard]] inline auto view() const -> raft::core::bitset_view<const bitset_t, index_t> { return bitset_view<const bitset_t, index_t>(to_mdspan(), bitset_len_); } /** * @brief Get the device pointer to the bitset. */ inline auto data() -> bitset_t* { return bitset_.data(); } inline auto data() const -> const bitset_t* { return bitset_.data(); } /** * @brief Get the number of bits of the bitset representation. */ inline auto size() const -> index_t { return bitset_len_; } /** * @brief Get the number of elements used by the bitset representation. */ inline auto n_elements() const -> index_t { return raft::ceildiv(bitset_len_, bitset_element_size); } /** @brief Get an mdspan view of the current bitset */ inline auto to_mdspan() -> raft::device_vector_view<bitset_t, index_t> { return raft::make_device_vector_view<bitset_t, index_t>(bitset_.data(), n_elements()); } [[nodiscard]] inline auto to_mdspan() const -> raft::device_vector_view<const bitset_t, index_t> { return raft::make_device_vector_view<const bitset_t, index_t>(bitset_.data(), n_elements()); } /** @brief Resize the bitset. If the requested size is larger, new memory is allocated and set to * the default value. * @param res RAFT resources * @param new_bitset_len new size of the bitset * @param default_value default value to initialize the new bits to */ void resize(const raft::resources& res, index_t new_bitset_len, bool default_value = true) { auto old_size = raft::ceildiv(bitset_len_, bitset_element_size); auto new_size = raft::ceildiv(new_bitset_len, bitset_element_size); bitset_.resize(new_size); bitset_len_ = new_bitset_len; if (old_size < new_size) { // If the new size is larger, set the new bits to the default value thrust::fill_n(resource::get_thrust_policy(res), bitset_.data() + old_size, new_size - old_size, default_value ? ~bitset_t{0} : bitset_t{0}); } } /** * @brief Test a list of indices in a bitset. * * @tparam output_t Output type of the test. Default is bool. * @param res RAFT resources * @param queries List of indices to test * @param output List of outputs */ template <typename output_t = bool> void test(const raft::resources& res, raft::device_vector_view<const index_t, index_t> queries, raft::device_vector_view<output_t, index_t> output) const { RAFT_EXPECTS(output.extent(0) == queries.extent(0), "Output and queries must be same size"); auto bitset_view = view(); raft::linalg::map( res, output, [bitset_view] __device__(index_t query) { return output_t(bitset_view.test(query)); }, queries); } /** * @brief Set a list of indices in a bitset to set_value. * * @param res RAFT resources * @param mask_index indices to remove from the bitset * @param set_value Value to set the bits to (true or false) */ void set(const raft::resources& res, raft::device_vector_view<const index_t, index_t> mask_index, bool set_value = false) { auto this_bitset_view = view(); thrust::for_each_n(resource::get_thrust_policy(res), mask_index.data_handle(), mask_index.extent(0), [this_bitset_view, set_value] __device__(const index_t sample_index) { this_bitset_view.set(sample_index, set_value); }); } /** * @brief Flip all the bits in a bitset. * @param res RAFT resources */ void flip(const raft::resources& res) { auto bitset_span = this->to_mdspan(); raft::linalg::map( res, bitset_span, [] __device__(bitset_t element) { return bitset_t(~element); }, raft::make_const_mdspan(bitset_span)); } /** * @brief Reset the bits in a bitset. * * @param res RAFT resources * @param default_value Value to set the bits to (true or false) */ void reset(const raft::resources& res, bool default_value = true) { thrust::fill_n(resource::get_thrust_policy(res), bitset_.data(), n_elements(), default_value ? ~bitset_t{0} : bitset_t{0}); } /** * @brief Returns the number of bits set to true in count_gpu_scalar. * * @param[in] res RAFT resources * @param[out] count_gpu_scalar Device scalar to store the count */ void count(const raft::resources& res, raft::device_scalar_view<index_t> count_gpu_scalar) { auto n_elements_ = n_elements(); auto count_gpu = raft::make_device_vector_view<index_t, index_t>(count_gpu_scalar.data_handle(), 1); auto bitset_matrix_view = raft::make_device_matrix_view<const bitset_t, index_t, col_major>( bitset_.data(), n_elements_, 1); bitset_t n_last_element = (bitset_len_ % bitset_element_size); bitset_t last_element_mask = n_last_element ? (bitset_t)((bitset_t{1} << n_last_element) - bitset_t{1}) : ~bitset_t{0}; raft::linalg::coalesced_reduction( res, bitset_matrix_view, count_gpu, index_t{0}, false, [last_element_mask, n_elements_] __device__(bitset_t element, index_t index) { index_t result = 0; if constexpr (bitset_element_size == 64) { if (index == n_elements_ - 1) result = index_t(raft::detail::popc(element & last_element_mask)); else result = index_t(raft::detail::popc(element)); } else { // Needed because popc is not overloaded for 16 and 8 bit elements if (index == n_elements_ - 1) result = index_t(raft::detail::popc(uint32_t{element} & last_element_mask)); else result = index_t(raft::detail::popc(uint32_t{element})); } return result; }); } /** * @brief Returns the number of bits set to true. * * @param res RAFT resources * @return index_t Number of bits set to true */ auto count(const raft::resources& res) -> index_t { auto count_gpu_scalar = raft::make_device_scalar<index_t>(res, 0.0); count(res, count_gpu_scalar.view()); index_t count_cpu = 0; raft::update_host( &count_cpu, count_gpu_scalar.data_handle(), 1, resource::get_cuda_stream(res)); resource::sync_stream(res); return count_cpu; } /** * @brief Checks if any of the bits are set to true in the bitset. * @param res RAFT resources */ bool any(const raft::resources& res) { return count(res) > 0; } /** * @brief Checks if all of the bits are set to true in the bitset. * @param res RAFT resources */ bool all(const raft::resources& res) { return count(res) == bitset_len_; } /** * @brief Checks if none of the bits are set to true in the bitset. * @param res RAFT resources */ bool none(const raft::resources& res) { return count(res) == 0; } private: raft::device_uvector<bitset_t> bitset_; index_t bitset_len_; }; /** @} */ } // end namespace raft::core
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/device_resources_manager.hpp
/* * Copyright (c) 2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <algorithm> #include <memory> #include <optional> #include <raft/core/device_resources.hpp> #include <raft/core/device_setter.hpp> #include <rmm/cuda_stream.hpp> #include <rmm/cuda_stream_pool.hpp> #include <rmm/mr/device/cuda_memory_resource.hpp> #include <rmm/mr/device/per_device_resource.hpp> #include <thrust/optional.h> namespace raft { /** * @brief A singleton used to easily generate a raft::device_resources object * * Many calls to RAFT functions require a `raft::device_resources` object * to provide CUDA resources like streams and stream pools. The * `raft::device_resources_manager` singleton provides a straightforward method to create those * objects in a way that allows consumers of RAFT to limit total consumption of device resources * without actively managing streams or other CUDA-specific objects. * * To control the resources a consuming application will use, the * resource manager provides setters for a variety of values. For * instance, to ensure that no more than `N` CUDA streams are used per * device, a consumer might call * `raft::device_resources_manager::set_streams_per_device(N)`. Note that all of these * setters must be used prior to retrieving the first `device_resources` from * the manager. Setters invoked after this will log a warning but have no * effect. * * After calling all desired setters, consumers can simply call * `auto res = raft::device_resources_manager::get_device_resources();` to get a valid * device_resources object for the current device based on previously-set * parameters. Importantly, calling `get_device_resources()` again from the same * thread is guaranteed to return a `device_resources` object with the same * underlying CUDA stream and (if a non-zero number of stream pools has been * requested) stream pool. * * Typical usage might look something like the following: * @code * void initialize_application() { * raft::device_resources_manager::set_streams_per_device(16); * } * * void foo_called_from_multiple_threads() { * auto res = raft::device_resources_manager::get_device_resources(); * // Call RAFT function using res * res.sync_stream() // Ensure work completes before returning * } * @endcode * * Note that all public methods of the `device_resources_manager` are thread-safe, * but the manager is designed to minimize locking required for * retrieving `device_resources` objects. Each thread must acquire a lock * exactly once per device when calling `get_device_resources`. Subsequent calls * will still be thread-safe but will not require a lock. * * All public methods of the `device_resources_manager` are static. Please see * documentation of those methods for additional usage information. * */ struct device_resources_manager { device_resources_manager(device_resources_manager const&) = delete; void operator=(device_resources_manager const&) = delete; private: device_resources_manager() {} ~device_resources_manager() { // Ensure that we destroy any pool memory resources before CUDA context is // lost per_device_components_.clear(); } // Get an id used to identify this thread for the purposes of assigning // (in round-robin fashion) the same resources to the thread on subsequent calls to // `get_device_resources` static auto get_thread_id() { static std::atomic<std::size_t> thread_counter{}; thread_local std::size_t id = ++thread_counter; return id; } // This struct holds the various parameters used to control // construction of the underlying resources shared by all // `device_resources` objects returned by `get_device_resources` struct resource_params { // The total number of primary streams to be used by the // application. If no value is provided, the default stream per thread // is used. std::optional<std::size_t> stream_count{std::nullopt}; // The total number of stream pools to be used by the application std::size_t pool_count{}; // How many streams to assign to each pool std::size_t pool_size{rmm::cuda_stream_pool::default_size}; // If a memory pool is requested (max_mem_pool_size is non-zero), use // this initial size for the pool in bytes. Must be a multiple of 256. // If nullopt, use half of the available memory on the current // device. thrust::optional<std::size_t> init_mem_pool_size{thrust::nullopt}; // If set to any non-zero value, create a memory pool with this // maximum size. If nullopt, use up to the entire available memory of the // device thrust::optional<std::size_t> max_mem_pool_size{std::size_t{}}; // Limit on workspace memory for the returned device_resources object std::optional<std::size_t> workspace_allocation_limit{std::nullopt}; // Optional specification of separate workspace memory resources for each // device. The integer in each pair indicates the device for this memory // resource. std::vector<std::pair<std::shared_ptr<rmm::mr::device_memory_resource>, int>> workspace_mrs{}; auto get_workspace_memory_resource(int device_id) {} } params_; // This struct stores the underlying resources to be shared among // `device_resources` objects returned by this manager. struct resource_components { // Construct all underlying resources indicated by `params` for the // indicated device. This includes primary streams, stream pools, and // a memory pool if requested. resource_components(int device_id, resource_params const& params) : device_id_{device_id}, streams_{[&params, this]() { auto scoped_device = device_setter{device_id_}; auto result = std::unique_ptr<rmm::cuda_stream_pool>{nullptr}; if (params.stream_count) { result = std::make_unique<rmm::cuda_stream_pool>(*params.stream_count); } return result; }()}, pools_{[&params, this]() { auto scoped_device = device_setter{device_id_}; auto result = std::vector<std::shared_ptr<rmm::cuda_stream_pool>>{}; if (params.pool_size != 0) { for (auto i = std::size_t{}; i < params.pool_count; ++i) { result.push_back(std::make_shared<rmm::cuda_stream_pool>(params.pool_size)); } } else if (params.pool_count != 0) { RAFT_LOG_WARN("Stream pools of size 0 requested; no pools will be created"); } return result; }()}, pool_mr_{[&params, this]() { auto scoped_device = device_setter{device_id_}; auto result = std::shared_ptr<rmm::mr::pool_memory_resource<rmm::mr::cuda_memory_resource>>{nullptr}; // If max_mem_pool_size is nullopt or non-zero, create a pool memory // resource if (params.max_mem_pool_size.value_or(1) != 0) { auto* upstream = dynamic_cast<rmm::mr::cuda_memory_resource*>(rmm::mr::get_current_device_resource()); if (upstream != nullptr) { result = std::make_shared<rmm::mr::pool_memory_resource<rmm::mr::cuda_memory_resource>>( upstream, params.init_mem_pool_size, params.max_mem_pool_size); rmm::mr::set_current_device_resource(result.get()); } else { RAFT_LOG_WARN( "Pool allocation requested, but other memory resource has already been set and " "will not be overwritten"); } } return result; }()}, workspace_mr_{[&params, this]() { auto result = std::shared_ptr<rmm::mr::device_memory_resource>{nullptr}; auto iter = std::find_if(std::begin(params.workspace_mrs), std::end(params.workspace_mrs), [this](auto&& pair) { return pair.second == device_id_; }); if (iter != std::end(params.workspace_mrs)) { result = iter->first; } return result; }()} { } // Get the id of the device associated with the constructed resource // components [[nodiscard]] auto get_device_id() const { return device_id_; } // Get the total number of streams available for this application [[nodiscard]] auto stream_count() const { auto result = std::size_t{}; if (streams_) { result = streams_->get_pool_size(); } return result; } // Get the stream assigned to this host thread. Note that the same stream // may be used by multiple threads, but any given thread will always use // the same stream [[nodiscard]] auto get_stream() const { auto result = rmm::cuda_stream_per_thread; if (stream_count() != 0) { result = streams_->get_stream(get_thread_id() % stream_count()); } return result; } // Get the total number of stream pools available for this // application [[nodiscard]] auto pool_count() const { return pools_.size(); } // Get the stream pool assigned to this host thread. Note that the same stream pool // may be used by multiple threads, but any given thread will always use // the same stream pool [[nodiscard]] auto get_pool() const { auto result = std::shared_ptr<rmm::cuda_stream_pool>{nullptr}; if (pool_count() != 0) { result = pools_[get_thread_id() % pool_count()]; } return result; } // Return a (possibly null) shared_ptr to the pool memory resource // created for this device by the manager [[nodiscard]] auto get_pool_memory_resource() const { return pool_mr_; } // Return the RAFT workspace allocation limit that will be used by // `device_resources` returned from this manager [[nodiscard]] auto get_workspace_allocation_limit() const { return workspace_allocation_limit_; } // Return a (possibly null) shared_ptr to the memory resource that will // be used for workspace allocations by `device_resources` returned from // this manager [[nodiscard]] auto get_workspace_memory_resource() { return workspace_mr_; } private: int device_id_; std::unique_ptr<rmm::cuda_stream_pool> streams_; std::vector<std::shared_ptr<rmm::cuda_stream_pool>> pools_; std::shared_ptr<rmm::mr::pool_memory_resource<rmm::mr::cuda_memory_resource>> pool_mr_; std::shared_ptr<rmm::mr::device_memory_resource> workspace_mr_; std::optional<std::size_t> workspace_allocation_limit_{std::nullopt}; }; // Mutex used to lock access to shared data until after the first // `get_device_resources` call in each thread mutable std::mutex manager_mutex_{}; // Indicates whether or not `get_device_resources` has been called by any // host thread bool params_finalized_{}; // Container for underlying device resources to be re-used across host // threads for each device std::vector<resource_components> per_device_components_; // Container for device_resources objects shared among threads. The index // of the outer vector is the thread id of the thread requesting resources // modulo the total number of resources managed by this object. The inner // vector contains all resources associated with that id across devices // in any order. std::vector<std::vector<raft::device_resources>> resources_{}; // Return a lock for accessing shared data [[nodiscard]] auto get_lock() const { return std::unique_lock{manager_mutex_}; } // Retrieve the underlying resources to be shared across the // application for the indicated device. This method acquires a lock the // first time it is called in each thread for a specific device to ensure that the // underlying resources have been correctly initialized exactly once across // all host threads. auto const& get_device_resources_(int device_id) { // Each thread maintains an independent list of devices it has // accessed. If it has not marked a device as initialized, it // acquires a lock to initialize it exactly once. This means that each // thread will lock once for a particular device and not proceed until // some thread has actually generated the corresponding device // components thread_local auto initialized_devices = std::vector<int>{}; auto res_iter = decltype(std::end(resources_[0])){}; if (std::find(std::begin(initialized_devices), std::end(initialized_devices), device_id) == std::end(initialized_devices)) { // Only lock if we have not previously accessed this device on this // thread auto lock = get_lock(); initialized_devices.push_back(device_id); // If we are building components, do not allow any further changes to // resource parameters. params_finalized_ = true; if (resources_.empty()) { // We will potentially need as many device_resources objects as there are combinations of // streams and pools on a given device. resources_.resize(std::max(params_.stream_count.value_or(1), std::size_t{1}) * std::max(params_.pool_count, std::size_t{1})); } auto res_idx = get_thread_id() % resources_.size(); // Check to see if we have constructed device_resources for the // requested device at the index assigned to this thread res_iter = std::find_if(std::begin(resources_[res_idx]), std::end(resources_[res_idx]), [device_id](auto&& res) { return res.get_device() == device_id; }); if (res_iter == std::end(resources_[res_idx])) { // Even if we have not yet built device_resources for the current // device, we may have already built the underlying components, since // multiple device_resources may point to the same components. auto component_iter = std::find_if( std::begin(per_device_components_), std::end(per_device_components_), [device_id](auto&& components) { return components.get_device_id() == device_id; }); if (component_iter == std::end(per_device_components_)) { // Build components for this device if we have not yet done so on // another thread per_device_components_.emplace_back(device_id, params_); component_iter = std::prev(std::end(per_device_components_)); } auto scoped_device = device_setter(device_id); // Build the device_resources object for this thread out of shared // components resources_[res_idx].emplace_back(component_iter->get_stream(), component_iter->get_pool(), component_iter->get_workspace_memory_resource(), component_iter->get_workspace_allocation_limit()); res_iter = std::prev(std::end(resources_[res_idx])); } } else { auto res_idx = get_thread_id() % resources_.size(); // If we have previously accessed this device on this thread, we do not // need to lock. We know that this thread already initialized the // resources it requires for this device if no other thread had already done so, so we simply // retrieve the previously-generated resources. res_iter = std::find_if(std::begin(resources_[res_idx]), std::end(resources_[res_idx]), [device_id](auto&& res) { return res.get_device() == device_id; }); } return *res_iter; } // Thread-safe setter for the number of streams void set_streams_per_device_(std::optional<std::size_t> num_streams) { auto lock = get_lock(); if (params_finalized_) { RAFT_LOG_WARN( "Attempted to set device_resources_manager properties after resources have already been " "retrieved"); } else { params_.stream_count = num_streams; } } // Thread-safe setter for the number and size of stream pools void set_stream_pools_per_device_(std::size_t num_pools, std::size_t num_streams) { auto lock = get_lock(); if (params_finalized_) { RAFT_LOG_WARN( "Attempted to set device_resources_manager properties after resources have already been " "retrieved"); } else { params_.pool_count = num_pools; params_.pool_size = num_streams; } } // Thread-safe setter for the RAFT workspace allocation limit void set_workspace_allocation_limit_(std::size_t memory_limit) { auto lock = get_lock(); if (params_finalized_) { RAFT_LOG_WARN( "Attempted to set device_resources_manager properties after resources have already been " "retrieved"); } else { params_.workspace_allocation_limit.emplace(memory_limit); } } // Thread-safe setter for the maximum memory pool size void set_max_mem_pool_size_(std::optional<std::size_t> memory_limit) { auto lock = get_lock(); if (params_finalized_) { RAFT_LOG_WARN( "Attempted to set device_resources_manager properties after resources have already been " "retrieved"); } else { if (memory_limit) { params_.max_mem_pool_size.emplace(*memory_limit); } else { params_.max_mem_pool_size = thrust::nullopt; } } } // Thread-safe setter for the initial memory pool size void set_init_mem_pool_size_(std::optional<std::size_t> init_memory) { auto lock = get_lock(); if (params_finalized_) { RAFT_LOG_WARN( "Attempted to set device_resources_manager properties after resources have already been " "retrieved"); } else { if (init_memory) { params_.init_mem_pool_size.emplace(*init_memory); } else { params_.init_mem_pool_size = thrust::nullopt; } } } // Thread-safe setter for workspace memory resources void set_workspace_memory_resource_(std::shared_ptr<rmm::mr::device_memory_resource> mr, int device_id) { auto lock = get_lock(); if (params_finalized_) { RAFT_LOG_WARN( "Attempted to set device_resources_manager properties after resources have already been " "retrieved"); } else { auto iter = std::find_if(std::begin(params_.workspace_mrs), std::end(params_.workspace_mrs), [device_id](auto&& pair) { return pair.second == device_id; }); if (iter != std::end(params_.workspace_mrs)) { iter->first = mr; } else { params_.workspace_mrs.emplace_back(mr, device_id); } } } // Retrieve the instance of this singleton static auto& get_manager() { static auto manager = device_resources_manager{}; return manager; } public: /** * @brief Retrieve device_resources to be used with the RAFT API * * This thread-safe method ensures that a `device_resources` object with * the same underlying stream and stream pool is returned every time it is * called by the same host thread. This means that if `get_device_resources` is * used to provide all `device_resources` in an application, then * `raft::get_device_resources().sync_stream()` and (if a stream pool is used) * raft::get_device_resources().sync_stream_pool() are guaranteed to synchronize all * work previously submitted to the device by this host thread. * * If the max memory pool size set with `set_max_mem_pool_size` is non-zero, * the first call of this method will also create a memory pool to be used * for all RMM-based allocations on device. * * @param device_id int If provided, the device for which resources should * be returned. Defaults to active CUDA device. */ static auto const& get_device_resources(int device_id = device_setter::get_current_device()) { return get_manager().get_device_resources_(device_id); } /** * @brief Set the total number of CUDA streams to be used per device * * If nullopt, the default stream per thread will be used * (essentially allowing as many streams as there are host threads). * Otherwise, all returned `device_resources` will draw their streams from this * limited pool. * * Limiting the total number of streams can be desirable for a number of * reasons, but it is most often used in consuming applications to * prevent a large number of host threads from flooding the device with * simultaneous requests that may exhaust device memory or other * resources. * * If called after the first call to * `raft::device_resources_manager::get_device_resources`, no change will be made, * and a warning will be emitted. */ static void set_streams_per_device(std::optional<std::size_t> num_streams) { get_manager().set_streams_per_device_(num_streams); } /** * @brief Set the total number and size of CUDA stream pools to be used per device * * Setting the number of stream pools to a non-zero value will provide a * pool of stream pools that can be shared among host threads. This can be * useful for the same reason it is useful to limit the total number of * primary streams assigned to `device_resoures` for each host thread. * Repeated calls to `get_device_resources` on a given host thread are * guaranteed to return `device_resources` with the same underlying stream * pool. * * If called after the first call to * `raft::device_resources_manager::get_device_resources`, no change will be made, * and a warning will be emitted. */ static void set_stream_pools_per_device( std::size_t num_pools, std::size_t num_streams = rmm::cuda_stream_pool::default_size) { get_manager().set_stream_pools_per_device_(num_pools, num_streams); } /** * @brief Set the maximum size of temporary RAFT workspaces * * Note that this limits only the size of temporary workspace * allocations. To cap the device memory generally available for all device * allocations made with RMM, use * `raft::device_manager::set_max_mem_pool_size` * * If called after the first call to * `raft::device_resources_manager::get_device_resources`, no change will be made, * and a warning will be emitted. */ static void set_workspace_allocation_limit(std::size_t memory_limit) { get_manager().set_workspace_allocation_limit_(memory_limit); } /** * @brief Set the maximum size of the device memory pool * * If set to 0, no memory pool will be used. If set to nullopt, the memory * pool is allowed to grow to the size of available device memory. * * Note that the pool will not actually be created until the first call * to `raft::device_manager::get_device_resources(device_id)`, after which it will become * the current RMM device memory resource for the indicated device. If the * current RMM device memory resource has already been set to some * non-default resource, no pool resource will be created and a warning will be emitted. It is * assumed that applications which have set a memory resource already wish to manage RMM * themselves. * * If called after the first call to * `raft::device_resources_manager::get_device_resources`, no change will be made, * and a warning will be emitted. */ static void set_max_mem_pool_size(std::optional<std::size_t> max_mem) { get_manager().set_max_mem_pool_size_(max_mem); } /** * @brief Set the initial size of the device memory pool * * If set to nullopt, the memory pool starts with half of the available * device memory. * * If called after the first call to * `raft::device_resources_manager::get_device_resources`, no change will be made, * and a warning will be emitted. */ static void set_init_mem_pool_size(std::optional<std::size_t> init_mem) { get_manager().set_init_mem_pool_size_(init_mem); } /** * @brief Request a device memory pool with specified parameters * * This convenience method essentially combines * `set_init_mem_pool_size` and `set_max_mem_pool_size`. It is provided * primarily to allow users who want a memory pool but do not want to choose * specific pool sizes to simply call * `raft::device_manager::set_memory_pool()` and enable a memory pool using * RMM defaults (initialize with half of available memory, allow to grow * to all available memory). * * If called after the first call to * `raft::device_resources_manager::get_device_resources`, no change will be made, * and a warning will be emitted. */ static void set_mem_pool(std::optional<std::size_t> init_mem = std::nullopt, std::optional<std::size_t> max_mem = std::nullopt) { set_init_mem_pool_size(init_mem); set_max_mem_pool_size(max_mem); } /** * @brief Set the workspace memory resource to be used on a specific device * * RAFT device_resources objects can be built with a separate memory * resource for allocating temporary workspaces. If a (non-nullptr) memory * resource is provided by this setter, it will be used as the * workspace memory resource for all `device_resources` returned for the * indicated device. * * If called after the first call to * `raft::device_resources_manager::get_device_resources`, no change will be made, * and a warning will be emitted. */ static void set_workspace_memory_resource(std::shared_ptr<rmm::mr::device_memory_resource> mr, int device_id = device_setter::get_current_device()) { get_manager().set_workspace_memory_resource_(mr, device_id); } }; } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/nvtx.hpp
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <optional> #include <raft/core/detail/nvtx.hpp> /** * \section Usage * * To add NVTX ranges to your code, use the `nvtx::range` RAII object. A * range begins when the object is created, and ends when the object is * destroyed. * * The example below creates nested NVTX ranges. The range `fun_scope` spans * the whole function, while the range `epoch_scope` spans an iteration * (and appears 5 times in the timeline). * \code{.cpp} * #include <raft/common/nvtx.hpp> * void some_function(int k){ * // Begins a NVTX range with the message "some_function_{k}" * // The range ends when some_function() returns * common::nvtx::range fun_scope( r{"some_function_%d", k}; * * for(int i = 0; i < 5; i++){ * common::nvtx::range epoch_scope{"epoch-%d", i}; * // some logic inside the loop * } * } * \endcode * * \section Domains * * All NVTX ranges are assigned to domains. A domain defines a named timeline in * the Nsight Systems view. By default, we put all ranges into a domain `domain::app` * named "application". This is controlled by the template parameter `Domain`. * * The example below defines a domain and uses it in a function. * \code{.cpp} * #include <raft/common/nvtx.hpp> * * struct my_app_domain { * static constexpr char const* name{"my application"}; * } * * void some_function(int k){ * // This NVTX range appears in the timeline named "my application" in Nsight Systems. * common::nvtx::range<my_app_domain> fun_scope( r{"some_function_%d", k}; * // some logic inside the loop * } * \endcode */ namespace raft::common::nvtx { namespace domain { /** @brief The default NVTX domain. */ struct app { static constexpr char const* name{"application"}; }; /** @brief This NVTX domain is supposed to be used within raft. */ struct raft { static constexpr char const* name{"raft"}; }; } // namespace domain /** * @brief Push a named NVTX range. * * @tparam Domain optional struct that defines the NVTX domain message; * You can create a new domain with a custom message as follows: * \code{.cpp} * struct custom_domain { static constexpr char const* name{"custom message"}; } * \endcode * NB: make sure to use the same domain for `push_range` and `pop_range`. * @param format range name format (accepts printf-style arguments) * @param args the arguments for the printf-style formatting */ template <typename Domain = domain::app, typename... Args> inline void push_range(const char* format, Args... args) { detail::push_range<Domain, Args...>(format, args...); } /** * @brief Pop the latest range. * * @tparam Domain optional struct that defines the NVTX domain message; * You can create a new domain with a custom message as follows: * \code{.cpp} * struct custom_domain { static constexpr char const* name{"custom message"}; } * \endcode * NB: make sure to use the same domain for `push_range` and `pop_range`. */ template <typename Domain = domain::app> inline void pop_range() { detail::pop_range<Domain>(); } /** * @brief Push a named NVTX range that would be popped at the end of the object lifetime. * * Refer to \ref Usage for the usage examples. * * @tparam Domain optional struct that defines the NVTX domain message; * You can create a new domain with a custom message as follows: * \code{.cpp} * struct custom_domain { static constexpr char const* name{"custom message"}; } * \endcode */ template <typename Domain = domain::app> class range { public: /** * Push a named NVTX range. * At the end of the object lifetime, pop the range back. * * @param format range name format (accepts printf-style arguments) * @param args the arguments for the printf-style formatting */ template <typename... Args> explicit range(const char* format, Args... args) { push_range<Domain, Args...>(format, args...); } ~range() { pop_range<Domain>(); } /* This object is not meant to be touched. */ range(const range&) = delete; range(range&&) = delete; auto operator=(const range&) -> range& = delete; auto operator=(range&&) -> range& = delete; static auto operator new(std::size_t) -> void* = delete; static auto operator new[](std::size_t) -> void* = delete; }; } // namespace raft::common::nvtx
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/host_container_policy.hpp
/* * Copyright (2019) Sandia Corporation * * The source code is licensed under the 3-clause BSD license found in the LICENSE file * thirdparty/LICENSES/mdarray.license */ /* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/mdspan_types.hpp> #include <raft/core/resources.hpp> #include <vector> namespace raft { /** * @brief A container policy for host mdarray. */ template <typename ElementType, typename Allocator = std::allocator<ElementType>> class host_vector_policy { public: using element_type = ElementType; using container_type = std::vector<element_type, Allocator>; using allocator_type = typename container_type::allocator_type; using pointer = typename container_type::pointer; using const_pointer = typename container_type::const_pointer; using reference = element_type&; using const_reference = element_type const&; using accessor_policy = std::experimental::default_accessor<element_type>; using const_accessor_policy = std::experimental::default_accessor<element_type const>; public: auto create(raft::resources const&, size_t n) -> container_type { return container_type(n); } constexpr host_vector_policy() noexcept(std::is_nothrow_default_constructible_v<ElementType>) = default; [[nodiscard]] constexpr auto access(container_type& c, size_t n) const noexcept -> reference { return c[n]; } [[nodiscard]] constexpr auto access(container_type const& c, size_t n) const noexcept -> const_reference { return c[n]; } [[nodiscard]] auto make_accessor_policy() noexcept { return accessor_policy{}; } [[nodiscard]] auto make_accessor_policy() const noexcept { return const_accessor_policy{}; } }; } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/math.hpp
/* * Copyright (c) 2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <algorithm> #include <cmath> #include <type_traits> #include <raft/core/detail/macros.hpp> #if defined(_RAFT_HAS_CUDA) #include <cuda_bf16.h> #include <cuda_fp16.h> #endif namespace raft { /** * @defgroup math_functions Mathematical Functions * @{ */ template <typename T> RAFT_INLINE_FUNCTION auto abs(T x) -> std::enable_if_t<std::is_same_v<float, T> || std::is_same_v<double, T> || std::is_same_v<int, T> || std::is_same_v<long int, T> || std::is_same_v<long long int, T>, T> { #ifdef __CUDA_ARCH__ return ::abs(x); #else return std::abs(x); #endif } template <typename T> constexpr RAFT_INLINE_FUNCTION auto abs(T x) -> std::enable_if_t<!std::is_same_v<float, T> && !std::is_same_v<double, T> && !std::is_same_v<int, T> && !std::is_same_v<long int, T> && !std::is_same_v<long long int, T>, T> { return x < T{0} ? -x : x; } /** Inverse cosine */ template <typename T> RAFT_INLINE_FUNCTION auto acos(T x) { #ifdef __CUDA_ARCH__ return ::acos(x); #else return std::acos(x); #endif } /** Inverse sine */ template <typename T> RAFT_INLINE_FUNCTION auto asin(T x) { #ifdef __CUDA_ARCH__ return ::asin(x); #else return std::asin(x); #endif } /** Inverse hyperbolic tangent */ template <typename T> RAFT_INLINE_FUNCTION auto atanh(T x) { #ifdef __CUDA_ARCH__ return ::atanh(x); #else return std::atanh(x); #endif } /** Cosine */ template <typename T, std::enable_if_t<CUDA_CONDITION_ELSE_TRUE(((!std::is_same_v<T, __half> && (!std::is_same_v<T, nv_bfloat16>)))), int> = 0> RAFT_INLINE_FUNCTION auto cos(T x) { #ifdef __CUDA_ARCH__ return ::cos(x); #else return std::cos(x); #endif } #if defined(_RAFT_HAS_CUDA) template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, __half>, __half> cos(T x) { #if (__CUDA_ARCH__ >= 530) return ::hcos(x); #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "__half is only supported on __CUDA_ARCH__ >= 530"); return T{}; #endif } template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, nv_bfloat16>, nv_bfloat16> cos(T x) { #if (__CUDA_ARCH__ >= 800) return ::hcos(x); #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "nv_bfloat16 is only supported on __CUDA_ARCH__ >= 800"); return T{}; #endif } #endif /** Sine */ template <typename T, std::enable_if_t<CUDA_CONDITION_ELSE_TRUE(((!std::is_same_v<T, __half> && (!std::is_same_v<T, nv_bfloat16>)))), int> = 0> RAFT_INLINE_FUNCTION auto sin(T x) { #ifdef __CUDA_ARCH__ return ::sin(x); #else return std::sin(x); #endif } #if defined(_RAFT_HAS_CUDA) template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, __half>, __half> sin(T x) { #if (__CUDA_ARCH__ >= 530) return ::hsin(x); #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "__half is only supported on __CUDA_ARCH__ >= 530"); return T{}; #endif } template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, nv_bfloat16>, nv_bfloat16> sin(T x) { #if (__CUDA_ARCH__ >= 800) return ::hsin(x); #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "nv_bfloat16 is only supported on __CUDA_ARCH__ >= 800"); return T{}; #endif } #endif /** Sine and cosine */ template <typename T> RAFT_INLINE_FUNCTION std::enable_if_t<std::is_same_v<float, T> || std::is_same_v<double, T>> sincos( const T& x, T* s, T* c) { #ifdef __CUDA_ARCH__ ::sincos(x, s, c); #else *s = std::sin(x); *c = std::cos(x); #endif } /** Hyperbolic tangent */ template <typename T> RAFT_INLINE_FUNCTION auto tanh(T x) { #ifdef __CUDA_ARCH__ return ::tanh(x); #else return std::tanh(x); #endif } /** Exponential function */ template <typename T, std::enable_if_t<CUDA_CONDITION_ELSE_TRUE(((!std::is_same_v<T, __half> && (!std::is_same_v<T, nv_bfloat16>)))), int> = 0> RAFT_INLINE_FUNCTION auto exp(T x) { #ifdef __CUDA_ARCH__ return ::exp(x); #else return std::exp(x); #endif } #if defined(_RAFT_HAS_CUDA) template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, __half>, __half> exp(T x) { #if (__CUDA_ARCH__ >= 530) return ::hexp(x); #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "__half is only supported on __CUDA_ARCH__ >= 530"); return T{}; #endif } template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, nv_bfloat16>, nv_bfloat16> exp(T x) { #if (__CUDA_ARCH__ >= 800) return ::hexp(x); #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "nv_bfloat16 is only supported on __CUDA_ARCH__ >= 800"); return T{}; #endif } #endif /** Natural logarithm */ template <typename T, std::enable_if_t<CUDA_CONDITION_ELSE_TRUE(((!std::is_same_v<T, __half> && (!std::is_same_v<T, nv_bfloat16>)))), int> = 0> RAFT_INLINE_FUNCTION auto log(T x) { #ifdef __CUDA_ARCH__ return ::log(x); #else return std::log(x); #endif } #if defined(_RAFT_HAS_CUDA) template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, __half>, __half> log(T x) { #if (__CUDA_ARCH__ >= 530) return ::hlog(x); #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "__half is only supported on __CUDA_ARCH__ >= 530"); return T{}; #endif } template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, nv_bfloat16>, nv_bfloat16> log(T x) { #if (__CUDA_ARCH__ >= 800) return ::hlog(x); #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "nv_bfloat16 is only supported on __CUDA_ARCH__ >= 800"); return T{}; #endif } #endif /** * @brief The CUDA Math API has overloads for all combinations of float/double. We provide similar * functionality while wrapping around std::max, which only supports arguments of the same type. * However, though the CUDA Math API supports combinations of unsigned and signed integers, this is * very error-prone so we do not support that and require the user to cast instead. (e.g the max of * -1 and 1u is 4294967295u...) * * When no overload matches, we provide a generic implementation but require that both types be the * same (and that the less-than operator be defined). * @{ */ template < typename T1, typename T2, std::enable_if_t<CUDA_CONDITION_ELSE_TRUE(RAFT_DEPAREN( ((!std::is_same_v<T1, __half> && !std::is_same_v<T2, __half>) || (!std::is_same_v<T1, nv_bfloat16> && !std::is_same_v<T2, nv_bfloat16>)))), int> = 0> RAFT_INLINE_FUNCTION auto max(const T1& x, const T2& y) { #ifdef __CUDA_ARCH__ // Combinations of types supported by the CUDA Math API if constexpr ((std::is_integral_v<T1> && std::is_integral_v<T2> && std::is_same_v<T1, T2>) || ((std::is_same_v<T1, float> || std::is_same_v<T1, double>)&&( std::is_same_v<T2, float> || std::is_same_v<T2, double>))) { return ::max(x, y); } // Else, check that the types are the same and provide a generic implementation else { static_assert( std::is_same_v<T1, T2>, "No native max overload for these types. Both argument types must be the same to use " "the generic max. Please cast appropriately."); return (x < y) ? y : x; } #else if constexpr (std::is_same_v<T1, float> && std::is_same_v<T2, double>) { return std::max(static_cast<double>(x), y); } else if constexpr (std::is_same_v<T1, double> && std::is_same_v<T2, float>) { return std::max(x, static_cast<double>(y)); } else { static_assert( std::is_same_v<T1, T2>, "std::max requires that both argument types be the same. Please cast appropriately."); return std::max(x, y); } #endif } #if defined(_RAFT_HAS_CUDA) template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, __half>, __half> max(T x, T y) { #if (__CUDA_ARCH__ >= 530) return ::__hmax(x, y); #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "__half is only supported on __CUDA_ARCH__ >= 530"); return T{}; #endif } template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, nv_bfloat16>, nv_bfloat16> max(T x, T y) { #if (__CUDA_ARCH__ >= 800) return ::__hmax(x, y); #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "nv_bfloat16 is only supported on __CUDA_ARCH__ >= 800"); return T{}; #endif } #endif /** Many-argument overload to avoid verbose nested calls or use with variadic arguments */ template <typename T1, typename T2, typename... Args> RAFT_INLINE_FUNCTION auto max(const T1& x, const T2& y, Args&&... args) { return raft::max(x, raft::max(y, std::forward<Args>(args)...)); } /** One-argument overload for convenience when using with variadic arguments */ template <typename T> constexpr RAFT_INLINE_FUNCTION auto max(const T& x) { return x; } #if defined(_RAFT_HAS_CUDA) template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, __half>, __half> max(T x) { #if (__CUDA_ARCH__ >= 530) return x; #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "__half is only supported on __CUDA_ARCH__ >= 530"); return T{}; #endif } template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, nv_bfloat16>, nv_bfloat16> max(T x) { #if (__CUDA_ARCH__ >= 800) return x; #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "nv_bfloat16 is only supported on __CUDA_ARCH__ >= 800"); return T{}; #endif } #endif /** * @brief Minimum Minimum of two or more values. * * The CUDA Math API has overloads for all combinations of float/double. We provide similar * functionality while wrapping around std::min, which only supports arguments of the same type. * However, though the CUDA Math API supports combinations of unsigned and signed integers, this is * very error-prone so we do not support that and require the user to cast instead. (e.g the min of * -1 and 1u is 1u...) * * When no overload matches, we provide a generic implementation but require that both types be the * same (and that the less-than operator be defined). * @{ */ template < typename T1, typename T2, std::enable_if_t<CUDA_CONDITION_ELSE_TRUE(RAFT_DEPAREN( ((!std::is_same_v<T1, __half> && !std::is_same_v<T2, __half>) || (!std::is_same_v<T1, nv_bfloat16> && !std::is_same_v<T2, nv_bfloat16>)))), int> = 0> RAFT_INLINE_FUNCTION auto min(const T1& x, const T2& y) { #ifdef __CUDA_ARCH__ // Combinations of types supported by the CUDA Math API if constexpr ((std::is_integral_v<T1> && std::is_integral_v<T2> && std::is_same_v<T1, T2>) || ((std::is_same_v<T1, float> || std::is_same_v<T1, double>)&&( std::is_same_v<T2, float> || std::is_same_v<T2, double>))) { return ::min(x, y); } // Else, check that the types are the same and provide a generic implementation else { static_assert( std::is_same_v<T1, T2>, "No native min overload for these types. Both argument types must be the same to use " "the generic min. Please cast appropriately."); return (y < x) ? y : x; } #else if constexpr (std::is_same_v<T1, float> && std::is_same_v<T2, double>) { return std::min(static_cast<double>(x), y); } else if constexpr (std::is_same_v<T1, double> && std::is_same_v<T2, float>) { return std::min(x, static_cast<double>(y)); } else { static_assert( std::is_same_v<T1, T2>, "std::min requires that both argument types be the same. Please cast appropriately."); return std::min(x, y); } #endif } #if defined(_RAFT_HAS_CUDA) template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, __half>, __half> min(T x, T y) { #if (__CUDA_ARCH__ >= 530) return ::__hmin(x, y); #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "__half is only supported on __CUDA_ARCH__ >= 530"); return T{}; #endif } template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, nv_bfloat16>, nv_bfloat16> min(T x, T y) { #if (__CUDA_ARCH__ >= 800) return ::__hmin(x, y); #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "nv_bfloat16 is only supported on __CUDA_ARCH__ >= 800"); return T{}; #endif } #endif /** Many-argument overload to avoid verbose nested calls or use with variadic arguments */ template <typename T1, typename T2, typename... Args> RAFT_INLINE_FUNCTION auto min(const T1& x, const T2& y, Args&&... args) { return raft::min(x, raft::min(y, std::forward<Args>(args)...)); } /** One-argument overload for convenience when using with variadic arguments */ template <typename T> constexpr RAFT_INLINE_FUNCTION auto min(const T& x) { return x; } #if defined(_RAFT_HAS_CUDA) template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, nv_bfloat16>, __half> min( T x) { #if (__CUDA_ARCH__ >= 530) return x; #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "__half is only supported on __CUDA_ARCH__ >= 530"); return T{}; #endif } template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, nv_bfloat16>, nv_bfloat16> min(T x) { #if (__CUDA_ARCH__ >= 800) return x; #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "nv_bfloat16 is only supported on __CUDA_ARCH__ >= 800"); return T{}; #endif } #endif /** Power */ template <typename T1, typename T2> RAFT_INLINE_FUNCTION auto pow(T1 x, T2 y) { #ifdef __CUDA_ARCH__ return ::pow(x, y); #else return std::pow(x, y); #endif } /** Square root */ template <typename T, std::enable_if_t<CUDA_CONDITION_ELSE_TRUE(((!std::is_same_v<T, __half> && (!std::is_same_v<T, nv_bfloat16>)))), int> = 0> RAFT_INLINE_FUNCTION auto sqrt(T x) { #ifdef __CUDA_ARCH__ return ::sqrt(x); #else return std::sqrt(x); #endif } #if defined(_RAFT_HAS_CUDA) template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, __half>, __half> sqrt(T x) { #if (__CUDA_ARCH__ >= 530) return ::hsqrt(x); #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "__half is only supported on __CUDA_ARCH__ >= 530"); return T{}; #endif } template <typename T> RAFT_DEVICE_INLINE_FUNCTION typename std::enable_if_t<std::is_same_v<T, nv_bfloat16>, nv_bfloat16> sqrt(T x) { #if (__CUDA_ARCH__ >= 800) return ::hsqrt(x); #else // Fail during template instantiation if the compute capability doesn't support this operation static_assert(sizeof(T) != sizeof(T), "nv_bfloat16 is only supported on __CUDA_ARCH__ >= 800"); return T{}; #endif } #endif /** Sign */ template <typename T> RAFT_INLINE_FUNCTION auto sgn(T val) -> int { return (T(0) < val) - (val < T(0)); } /** @} */ } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/host_mdspan.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cstdint> #include <raft/core/mdspan.hpp> #include <raft/core/memory_type.hpp> #include <raft/core/host_device_accessor.hpp> namespace raft { template <typename AccessorPolicy> using host_accessor = host_device_accessor<AccessorPolicy, memory_type::host>; /** * @brief std::experimental::mdspan with host tag to avoid accessing incorrect memory location. */ template <typename ElementType, typename Extents, typename LayoutPolicy = layout_c_contiguous, typename AccessorPolicy = std::experimental::default_accessor<ElementType>> using host_mdspan = mdspan<ElementType, Extents, LayoutPolicy, host_accessor<AccessorPolicy>>; template <typename T, bool B> struct is_host_mdspan : std::false_type {}; template <typename T> struct is_host_mdspan<T, true> : std::bool_constant<T::accessor_type::is_host_accessible> {}; /** * @\brief Boolean to determine if template type T is either raft::host_mdspan or a derived type */ template <typename T> using is_host_mdspan_t = is_host_mdspan<T, is_mdspan_v<T>>; template <typename T> using is_input_host_mdspan_t = is_host_mdspan<T, is_input_mdspan_v<T>>; template <typename T> using is_output_host_mdspan_t = is_host_mdspan<T, is_output_mdspan_v<T>>; /** * @\brief Boolean to determine if variadic template types Tn are either raft::host_mdspan or a * derived type */ template <typename... Tn> inline constexpr bool is_host_mdspan_v = std::conjunction_v<is_host_mdspan_t<Tn>...>; template <typename... Tn> inline constexpr bool is_input_host_mdspan_v = std::conjunction_v<is_input_host_mdspan_t<Tn>...>; template <typename... Tn> inline constexpr bool is_output_host_mdspan_v = std::conjunction_v<is_output_host_mdspan_t<Tn>...>; template <typename... Tn> using enable_if_host_mdspan = std::enable_if_t<is_input_mdspan_v<Tn...>>; template <typename... Tn> using enable_if_input_host_mdspan = std::enable_if_t<is_input_host_mdspan_v<Tn...>>; template <typename... Tn> using enable_if_output_host_mdspan = std::enable_if_t<is_output_host_mdspan_v<Tn...>>; /** * @brief Shorthand for 0-dim host mdspan (scalar). * @tparam ElementType the data type of the scalar element * @tparam IndexType the index type of the extents */ template <typename ElementType, typename IndexType = std::uint32_t> using host_scalar_view = host_mdspan<ElementType, scalar_extent<IndexType>>; /** * @brief Shorthand for 1-dim host mdspan. * @tparam ElementType the data type of the vector elements * @tparam IndexType the index type of the extents */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> using host_vector_view = host_mdspan<ElementType, vector_extent<IndexType>, LayoutPolicy>; /** * @brief Shorthand for c-contiguous host matrix view. * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> using host_matrix_view = host_mdspan<ElementType, matrix_extent<IndexType>, LayoutPolicy>; /** * @brief Shorthand for 128 byte aligned host matrix view. * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy must be of type layout_{left/right}_padded */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_right_padded<ElementType>, typename = enable_if_layout_padded<ElementType, LayoutPolicy>> using host_aligned_matrix_view = host_mdspan<ElementType, matrix_extent<IndexType>, LayoutPolicy, std::experimental::aligned_accessor<ElementType, detail::alignment::value>>; /** * @brief Create a 2-dim 128 byte aligned mdspan instance for host pointer. It's * expected that the given layout policy match the layout of the underlying * pointer. * @tparam ElementType the data type of the matrix elements * @tparam LayoutPolicy must be of type layout_{left/right}_padded * @tparam IndexType the index type of the extents * @param[in] ptr on host to wrap * @param[in] n_rows number of rows in pointer * @param[in] n_cols number of columns in pointer */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_right_padded<ElementType>> auto make_host_aligned_matrix_view(ElementType* ptr, IndexType n_rows, IndexType n_cols) { using data_handle_type = typename std::experimental::aligned_accessor<ElementType, detail::alignment::value>::data_handle_type; static_assert(std::is_same<LayoutPolicy, layout_left_padded<ElementType>>::value || std::is_same<LayoutPolicy, layout_right_padded<ElementType>>::value); assert(reinterpret_cast<std::uintptr_t>(ptr) == std::experimental::details::alignTo(reinterpret_cast<std::uintptr_t>(ptr), detail::alignment::value)); data_handle_type aligned_pointer = ptr; matrix_extent<IndexType> extents{n_rows, n_cols}; return host_aligned_matrix_view<ElementType, IndexType, LayoutPolicy>{aligned_pointer, extents}; } /** * @brief Create a 0-dim (scalar) mdspan instance for host value. * * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @param[in] ptr on device to wrap */ template <typename ElementType, typename IndexType = std::uint32_t> auto make_host_scalar_view(ElementType* ptr) { scalar_extent<IndexType> extents; return host_scalar_view<ElementType, IndexType>{ptr, extents}; } /** * @brief Create a 2-dim c-contiguous mdspan instance for host pointer. It's * expected that the given layout policy match the layout of the underlying * pointer. * @tparam ElementType the data type of the matrix elements * @tparam IndexType the index type of the extents * @tparam LayoutPolicy policy for strides and layout ordering * @param[in] ptr on host to wrap * @param[in] n_rows number of rows in pointer * @param[in] n_cols number of columns in pointer */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> auto make_host_matrix_view(ElementType* ptr, IndexType n_rows, IndexType n_cols) { matrix_extent<IndexType> extents{n_rows, n_cols}; return host_matrix_view<ElementType, IndexType, LayoutPolicy>{ptr, extents}; } /** * @brief Create a 1-dim mdspan instance for host pointer. * @tparam ElementType the data type of the vector elements * @tparam IndexType the index type of the extents * @param[in] ptr on host to wrap * @param[in] n number of elements in pointer * @return raft::host_vector_view */ template <typename ElementType, typename IndexType = std::uint32_t, typename LayoutPolicy = layout_c_contiguous> auto make_host_vector_view(ElementType* ptr, IndexType n) { return host_vector_view<ElementType, IndexType, LayoutPolicy>{ptr, n}; } } // end namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/core/device_csr_matrix.hpp
/* * Copyright (c) 2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/csr_matrix.hpp> #include <raft/core/device_container_policy.hpp> #include <raft/core/device_span.hpp> #include <raft/core/resource/cuda_stream.hpp> #include <raft/core/resources.hpp> #include <raft/core/sparse_types.hpp> #include <type_traits> namespace raft { /** * \defgroup device_csr_matrix Device CSR Matrix Types * @{ */ /** * Specialization for a sparsity-preserving compressed structure view which uses device memory */ template <typename IndptrType, typename IndicesType, typename NZType> using device_compressed_structure_view = compressed_structure_view<IndptrType, IndicesType, NZType, true>; /** * Specialization for a sparsity-owning compressed structure which uses device memory */ template <typename IndptrType, typename IndicesType, typename NZType, template <typename T> typename ContainerPolicy = device_uvector_policy> using device_compressed_structure = compressed_structure<IndptrType, IndicesType, NZType, true, ContainerPolicy>; /** * Specialization for a csr matrix view which uses device memory */ template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType> using device_csr_matrix_view = csr_matrix_view<ElementType, IndptrType, IndicesType, NZType, true>; template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType, template <typename T> typename ContainerPolicy = device_uvector_policy, SparsityType sparsity_type = SparsityType::OWNING> using device_csr_matrix = csr_matrix<ElementType, IndptrType, IndicesType, NZType, true, ContainerPolicy, sparsity_type>; /** * Specialization for a sparsity-owning csr matrix which uses device memory */ template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType, template <typename T> typename ContainerPolicy = device_uvector_policy> using device_sparsity_owning_csr_matrix = csr_matrix<ElementType, IndptrType, IndicesType, NZType, true, ContainerPolicy>; /** * Specialization for a sparsity-preserving csr matrix which uses device memory */ template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType, template <typename T> typename ContainerPolicy = device_uvector_policy> using device_sparsity_preserving_csr_matrix = csr_matrix<ElementType, IndptrType, IndicesType, NZType, true, ContainerPolicy, SparsityType::PRESERVING>; template <typename T> struct is_device_csr_matrix_view : std::false_type {}; template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType> struct is_device_csr_matrix_view< device_csr_matrix_view<ElementType, IndptrType, IndicesType, NZType>> : std::true_type {}; template <typename T> constexpr bool is_device_csr_matrix_view_v = is_device_csr_matrix_view<T>::value; template <typename T> struct is_device_csr_matrix : std::false_type {}; template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType, template <typename T> typename ContainerPolicy, SparsityType sparsity_type> struct is_device_csr_matrix< device_csr_matrix<ElementType, IndptrType, IndicesType, NZType, ContainerPolicy, sparsity_type>> : std::true_type {}; template <typename T> constexpr bool is_device_csr_matrix_v = is_device_csr_matrix<T>::value; template <typename T> constexpr bool is_device_csr_sparsity_owning_v = is_device_csr_matrix<T>::value and T::get_sparsity_type() == OWNING; template <typename T> constexpr bool is_device_csr_sparsity_preserving_v = is_device_csr_matrix<T>::value and T::get_sparsity_type() == PRESERVING; /** * Create a sparsity-owning sparse matrix in the compressed-sparse row format. sparsity-owning * means that all of the underlying vectors (data, indptr, indices) are owned by the csr_matrix * instance. If not known up front, the sparsity can be ignored in this factory function and * `resize()` invoked on the instance once the sparsity is known. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/device_csr_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * * raft::resources handle; * csr_matrix = raft::make_device_csr_matrix(handle, n_rows, n_cols); * ... * // compute expected sparsity * ... * int nnz = 5000; * csr_matrix.initialize_sparsity(nnz); * @endcode * @tparam ElementType * @tparam IndptrType * @tparam IndicesType * @tparam NZType * @param[in] handle a raft handle for managing expensive device resources * @param[in] n_rows total number of rows in the matrix * @param[in] n_cols total number of columns in the matrix * @param[in] nnz number of non-zeros in the matrix if known [optional] * @return a sparsity-owning sparse matrix in compressed (csr) format */ template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType = uint64_t> auto make_device_csr_matrix(raft::resources const& handle, IndptrType n_rows, IndicesType n_cols, NZType nnz = 0) { return device_sparsity_owning_csr_matrix<ElementType, IndptrType, IndicesType, NZType>( handle, n_rows, n_cols, nnz); } /** * Create a sparsity-preserving sparse matrix in the compressed-sparse row format. * sparsity-preserving means that a view of the csr sparsity is supplied, allowing the values in * the sparsity to change but not the sparsity itself. The csr_matrix instance does not own the * sparsity, the sparsity must be known up front, and cannot be resized later. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/device_csr_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * * raft::resources handle; * coo_structure = raft::make_device_compressed_structure(handle, n_rows, n_cols); * ... * // compute expected sparsity * ... * csr_structure.initialize_sparsity(nnz); * csr_matrix = raft::make_device_csr_matrix(handle, csr_structure.view()); * @endcode * * @tparam ElementType * @tparam IndptrType * @tparam IndicesType * @tparam NZType * @param[in] handle raft handle for managing expensive device resources * @param[in] structure a sparsity-preserving compressed structural view * @return a sparsity-preserving sparse matrix in compressed (csr) format */ template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType = uint64_t> auto make_device_csr_matrix( raft::resources const& handle, device_compressed_structure_view<IndptrType, IndicesType, NZType> structure) { return device_sparsity_preserving_csr_matrix<ElementType, IndptrType, IndicesType, NZType>( handle, structure); } /** * Create a non-owning sparse matrix view in the coordinate format. This is sparsity-preserving, * meaning that the underlying sparsity is known and cannot be changed. Use the sparsity-owning * coo_matrix if sparsity needs to be mutable. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/device_csr_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * int nnz = 5000; * * // The following pointer is assumed to reference device memory for a size of nnz * float* d_elm_ptr = ...; * * raft::resources handle; * csr_structure = raft::make_device_compressed_structure(handle, n_rows, n_cols, nnz); * csr_matrix_view = raft::make_device_csr_matrix_view(handle, d_elm_ptr, csr_structure.view()); * @endcode * * @tparam ElementType * @tparam IndptrType * @tparam IndicesType * @tparam NZType * @param[in] ptr a pointer to array of nonzero matrix elements on device (size nnz) * @param[in] structure a sparsity-preserving compressed sparse structural view * @return a sparsity-preserving csr matrix view */ template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType = uint64_t> auto make_device_csr_matrix_view( ElementType* ptr, device_compressed_structure_view<IndptrType, IndicesType, NZType> structure) { return device_csr_matrix_view<ElementType, IndptrType, IndicesType, NZType>( raft::device_span<ElementType>(ptr, structure.get_nnz()), structure); } /** * Create a non-owning sparse matrix view in the compressed-sparse row format. This is * sparsity-preserving, meaning that the underlying sparsity is known and cannot be changed. Use the * sparsity-owning coo_matrix if sparsity needs to be mutable. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/device_span.hpp> * #include <raft/core/device_csr_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * int nnz = 5000; * * // The following span is assumed to be of size nnz * raft::device_span<float> d_elm_ptr; * * raft::resources handle; * csr_structure = raft::make_device_compressed_structure(handle, n_rows, n_cols, nnz); * csr_matrix_view = raft::make_device_csr_matrix_view(handle, d_elm_ptr, csr_structure.view()); * @endcode * * @tparam ElementType * @tparam IndptrType * @tparam IndicesType * @tparam NZType * @param[in] elements device span containing array of matrix elements (size nnz) * @param[in] structure a sparsity-preserving structural view * @return a sparsity-preserving csr matrix view */ template <typename ElementType, typename IndptrType, typename IndicesType, typename NZType = uint64_t> auto make_device_csr_matrix_view( raft::device_span<ElementType> elements, device_compressed_structure_view<IndptrType, IndicesType, NZType> structure) { RAFT_EXPECTS(elements.size() == structure.get_nnz(), "Size of elements must be equal to the nnz from the structure"); return device_csr_matrix_view<ElementType, IndptrType, IndicesType, NZType>(elements, structure); } /** * Create a sparsity-owning compressed structure. This is not sparsity-preserving, meaning that * the underlying sparsity does not need to be known upon construction. When not known up front, * the allocation of the underlying indices array is delayed until `resize(nnz)` is invoked. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/device_csr_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * int nnz = 5000; * * raft::resources handle; * csr_structure = raft::make_device_compressed_structure(handle, n_rows, n_cols, nnz); * ... * // compute expected sparsity * ... * csr_structure.initialize_sparsity(nnz); * @endcode * * @tparam IndptrType * @tparam IndicesType * @tparam NZType * @param[in] handle raft handle for managing expensive device resources * @param[in] n_rows total number of rows * @param[in] n_cols total number of cols * @param[in] nnz total number of nonzeros, if known * @return a sparsity-owning compressed structure instance */ template <typename IndptrType, typename IndicesType, typename NZType = uint64_t> auto make_device_compressed_structure(raft::resources const& handle, IndptrType n_rows, IndicesType n_cols, NZType nnz = 0) { return device_compressed_structure<IndptrType, IndicesType, NZType>(handle, n_rows, n_cols, nnz); } /** * Create a non-owning sparsity-preserved compressed structure view. Sparsity-preserving means that * the underlying sparsity is known and cannot be changed. Use the sparsity-owning version if the * sparsity is not known up front. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/device_csr_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * int nnz = 5000; * * // The following pointer is assumed to reference device memory of size n_rows+1 * int *indptr = ...; * * // The following pointer is assumed to reference device memory of size nnz * int *indices = ...; * * raft::resources handle; * csr_structure = raft::make_device_compressed_structure_view(handle, indptr, indices, n_rows, * n_cols, nnz); * @endcode * * * @tparam ElementType * @tparam IndptrType * @tparam IndicesType * @tparam NZType * @param[in] indptr structural indptr (size n_rows+1) * @param[in] indices structural indices (size nnz) * @param[in] n_rows total number of rows * @param[in] n_cols total number of columns * @param[in] nnz number of non-zeros * @return a sparsity-preserving compressed structural view */ template <typename IndptrType, typename IndicesType, typename NZType = uint64_t> auto make_device_compressed_structure_view( IndptrType* indptr, IndicesType* indices, IndptrType n_rows, IndicesType n_cols, NZType nnz) { return device_compressed_structure_view<IndptrType, IndicesType, NZType>( raft::device_span<IndptrType>(indptr, n_rows + 1), raft::device_span<IndicesType>(indices, nnz), n_cols); } /** * Create a non-owning sparsity-preserved compressed structure view. Sparsity-preserving means that * the underlying sparsity is known and cannot be changed. Use the sparsity-owning version if the * sparsity is not known up front. * * @code{.cpp} * #include <raft/core/resources.hpp> * #include <raft/core/device_csr_matrix.hpp> * * int n_rows = 100000; * int n_cols = 10000; * int nnz = 5000; * * // The following device spans is assumed to be of size n_rows+1 * raft::device_span<int> indptr; * * // The following device span is assumed to be of size nnz * raft::device_span<int> indices; * * raft::resources handle; * csr_structure = raft::make_device_compressed_structure_view(handle, indptr, indices, n_rows, * n_cols); * @endcode * * @tparam IndptrType * @tparam IndicesType * @tparam NZType * @param[in] indptr structural indptr (size n_rows+1) * @param[in] indices structural indices (size nnz) * @param[in] n_cols total number of columns * @return a sparsity-preserving compressed structural view * */ template <typename IndptrType, typename IndicesType, typename NZType = uint64_t> auto make_device_compressed_structure_view(raft::device_span<IndptrType> indptr, raft::device_span<IndicesType> indices, IndicesType n_cols) { return device_compressed_structure_view<IndptrType, IndicesType, NZType>(indptr, indices, n_cols); } /** @} */ }; // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/resource/device_properties.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cuda_runtime.h> #include <raft/core/resource/device_id.hpp> #include <raft/core/resource/resource_types.hpp> #include <raft/core/resources.hpp> #include <raft/util/cudart_utils.hpp> namespace raft::resource { class device_properties_resource : public resource { public: device_properties_resource(int dev_id) { RAFT_CUDA_TRY_NO_THROW(cudaGetDeviceProperties(&prop_, dev_id)); } void* get_resource() override { return &prop_; } ~device_properties_resource() override {} private: cudaDeviceProp prop_; }; /** * @defgroup resource_device_props Device properties resource functions * @{ */ /** * Factory that knows how to construct a * specific raft::resource to populate * the res_t. */ class device_properties_resource_factory : public resource_factory { public: device_properties_resource_factory(int dev_id) : dev_id_(dev_id) {} resource_type get_resource_type() override { return resource_type::DEVICE_PROPERTIES; } resource* make_resource() override { return new device_properties_resource(dev_id_); } private: int dev_id_; }; /** * Load a cudaDeviceProp from a res (and populate it on the res if needed). * @param res raft res object for managing resources * @return populated cuda device properties instance */ inline cudaDeviceProp& get_device_properties(resources const& res) { if (!res.has_resource_factory(resource_type::DEVICE_PROPERTIES)) { int dev_id = get_device_id(res); res.add_resource_factory(std::make_shared<device_properties_resource_factory>(dev_id)); } return *res.get_resource<cudaDeviceProp>(resource_type::DEVICE_PROPERTIES); }; /** * @} */ } // namespace raft::resource
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/resource/cusolver_dn_handle.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include "cuda_stream.hpp" #include <cusolverDn.h> #include <raft/core/cusolver_macros.hpp> #include <raft/core/resource/resource_types.hpp> #include <raft/core/resources.hpp> #include <rmm/cuda_stream_view.hpp> namespace raft::resource { /** * */ class cusolver_dn_resource : public resource { public: cusolver_dn_resource(rmm::cuda_stream_view stream) { RAFT_CUSOLVER_TRY_NO_THROW(cusolverDnCreate(&cusolver_res)); RAFT_CUSOLVER_TRY_NO_THROW(cusolverDnSetStream(cusolver_res, stream)); } void* get_resource() override { return &cusolver_res; } ~cusolver_dn_resource() override { RAFT_CUSOLVER_TRY_NO_THROW(cusolverDnDestroy(cusolver_res)); } private: cusolverDnHandle_t cusolver_res; }; /** * @defgroup resource_cusolver_dn cuSolver DN handle resource functions * @{ */ /** * Factory that knows how to construct a * specific raft::resource to populate * the res_t. */ class cusolver_dn_resource_factory : public resource_factory { public: cusolver_dn_resource_factory(rmm::cuda_stream_view stream) : stream_(stream) {} resource_type get_resource_type() override { return resource_type::CUSOLVER_DN_HANDLE; } resource* make_resource() override { return new cusolver_dn_resource(stream_); } private: rmm::cuda_stream_view stream_; }; /** * Load a cusolverSpres_t from raft res if it exists, otherwise * add it and return it. * @param[in] res the raft resources object * @return cusolver dn handle */ inline cusolverDnHandle_t get_cusolver_dn_handle(resources const& res) { if (!res.has_resource_factory(resource_type::CUSOLVER_DN_HANDLE)) { cudaStream_t stream = get_cuda_stream(res); res.add_resource_factory(std::make_shared<cusolver_dn_resource_factory>(stream)); } return *res.get_resource<cusolverDnHandle_t>(resource_type::CUSOLVER_DN_HANDLE); }; /** * @} */ } // namespace raft::resource
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/resource/stream_view.hpp
/* * Copyright (c) 2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/resources.hpp> #include <raft/core/stream_view.hpp> #ifndef RAFT_DISABLE_CUDA #include <raft/core/resource/cuda_stream.hpp> #endif namespace raft::resource { struct stream_view_resource : public resource { stream_view_resource(raft::stream_view view = raft::stream_view_per_thread) : stream(view) {} void* get_resource() override { return &stream; } ~stream_view_resource() override {} private: raft::stream_view stream; }; /** * Factory that knows how to construct a specific raft::resource to populate * the resources instance. */ struct stream_view_resource_factory : public resource_factory { public: stream_view_resource_factory(raft::stream_view view = raft::stream_view_per_thread) : stream(view) { } resource_type get_resource_type() override { return resource_type::STREAM_VIEW; } resource* make_resource() override { return new stream_view_resource(stream); } private: raft::stream_view stream; }; /** * @defgroup resource_stream_view stream resource functions compatible with * non-CUDA builds * @{ */ /** * Load a raft::stream_view from a resources instance (and populate it on the res * if needed). * @param res raft res object for managing resources * @return */ inline raft::stream_view get_stream_view(resources const& res) { if (!res.has_resource_factory(resource_type::STREAM_VIEW)) { res.add_resource_factory(std::make_shared<stream_view_resource_factory>()); } return *res.get_resource<raft::stream_view>(resource_type::STREAM_VIEW); }; /** * Load a raft::stream__view from a resources instance (and populate it on the res * if needed). * @param[in] res raft resources object for managing resources * @param[in] view raft stream view */ inline void set_stream_view(resources const& res, raft::stream_view view) { res.add_resource_factory(std::make_shared<stream_view_resource_factory>(view)); }; /** * @brief synchronize a specific stream * * @param[in] res the raft resources object * @param[in] stream stream to synchronize */ inline void sync_stream_view(const resources& res, raft::stream_view stream) { stream.interruptible_synchronize(); } /** * @brief synchronize main stream on the resources instance */ inline void sync_stream_view(const resources& res) { sync_stream_view(res, get_stream_view(res)); } /** * @} */ } // namespace raft::resource
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/resource/cuda_stream_pool.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cuda_runtime.h> #include <raft/core/resource/cuda_event.hpp> #include <raft/core/resource/cuda_stream.hpp> #include <raft/core/resource/detail/stream_sync_event.hpp> #include <raft/core/resource/resource_types.hpp> #include <raft/core/resources.hpp> #include <rmm/cuda_stream_pool.hpp> namespace raft::resource { class cuda_stream_pool_resource : public resource { public: cuda_stream_pool_resource(std::shared_ptr<rmm::cuda_stream_pool> stream_pool) : stream_pool_(stream_pool) { } ~cuda_stream_pool_resource() override {} void* get_resource() override { return &stream_pool_; } private: std::shared_ptr<rmm::cuda_stream_pool> stream_pool_{nullptr}; }; /** * Factory that knows how to construct a * specific raft::resource to populate * the res_t. */ class cuda_stream_pool_resource_factory : public resource_factory { public: cuda_stream_pool_resource_factory(std::shared_ptr<rmm::cuda_stream_pool> stream_pool = {nullptr}) : stream_pool_(stream_pool) { } resource_type get_resource_type() override { return resource_type::CUDA_STREAM_POOL; } resource* make_resource() override { return new cuda_stream_pool_resource(stream_pool_); } private: std::shared_ptr<rmm::cuda_stream_pool> stream_pool_{nullptr}; }; inline bool is_stream_pool_initialized(const resources& res) { return *res.get_resource<std::shared_ptr<rmm::cuda_stream_pool>>( resource_type::CUDA_STREAM_POOL) != nullptr; } /** * @defgroup resource_stream_pool CUDA Stream pool resource functions * @{ */ /** * Load a cuda_stream_pool, and create a new one if it doesn't already exist * @param res raft res object for managing resources * @return */ inline const rmm::cuda_stream_pool& get_cuda_stream_pool(const resources& res) { if (!res.has_resource_factory(resource_type::CUDA_STREAM_POOL)) { res.add_resource_factory(std::make_shared<cuda_stream_pool_resource_factory>()); } return *( *res.get_resource<std::shared_ptr<rmm::cuda_stream_pool>>(resource_type::CUDA_STREAM_POOL)); }; /** * Explicitly set a stream pool on the current res. Note that this will overwrite * an existing stream pool on the res. * @param res * @param stream_pool */ inline void set_cuda_stream_pool(const resources& res, std::shared_ptr<rmm::cuda_stream_pool> stream_pool) { res.add_resource_factory(std::make_shared<cuda_stream_pool_resource_factory>(stream_pool)); }; inline std::size_t get_stream_pool_size(const resources& res) { return is_stream_pool_initialized(res) ? get_cuda_stream_pool(res).get_pool_size() : 0; } /** * @brief return stream from pool */ inline rmm::cuda_stream_view get_stream_from_stream_pool(const resources& res) { RAFT_EXPECTS(is_stream_pool_initialized(res), "ERROR: rmm::cuda_stream_pool was not initialized"); return get_cuda_stream_pool(res).get_stream(); } /** * @brief return stream from pool at index */ inline rmm::cuda_stream_view get_stream_from_stream_pool(const resources& res, std::size_t stream_idx) { RAFT_EXPECTS(is_stream_pool_initialized(res), "ERROR: rmm::cuda_stream_pool was not initialized"); return get_cuda_stream_pool(res).get_stream(stream_idx); } /** * @brief return stream from pool if size > 0, else main stream on res */ inline rmm::cuda_stream_view get_next_usable_stream(const resources& res) { return is_stream_pool_initialized(res) ? get_stream_from_stream_pool(res) : get_cuda_stream(res); } /** * @brief return stream from pool at index if size > 0, else main stream on res * * @param[in] res the raft resources object * @param[in] stream_idx the required index of the stream in the stream pool if available */ inline rmm::cuda_stream_view get_next_usable_stream(const resources& res, std::size_t stream_idx) { return is_stream_pool_initialized(res) ? get_stream_from_stream_pool(res, stream_idx) : get_cuda_stream(res); } /** * @brief synchronize the stream pool on the res * * @param[in] res the raft resources object */ inline void sync_stream_pool(const resources& res) { for (std::size_t i = 0; i < get_stream_pool_size(res); i++) { sync_stream(res, get_cuda_stream_pool(res).get_stream(i)); } } /** * @brief synchronize subset of stream pool * * @param[in] res the raft resources object * @param[in] stream_indices the indices of the streams in the stream pool to synchronize */ inline void sync_stream_pool(const resources& res, const std::vector<std::size_t> stream_indices) { RAFT_EXPECTS(is_stream_pool_initialized(res), "ERROR: rmm::cuda_stream_pool was not initialized"); for (const auto& stream_index : stream_indices) { sync_stream(res, get_cuda_stream_pool(res).get_stream(stream_index)); } } /** * @brief ask stream pool to wait on last event in main stream * * @param[in] res the raft resources object */ inline void wait_stream_pool_on_stream(const resources& res) { if (!res.has_resource_factory(resource_type::CUDA_STREAM_POOL)) { res.add_resource_factory(std::make_shared<cuda_stream_pool_resource_factory>()); } cudaEvent_t event = detail::get_cuda_stream_sync_event(res); RAFT_CUDA_TRY(cudaEventRecord(event, get_cuda_stream(res))); for (std::size_t i = 0; i < get_stream_pool_size(res); i++) { RAFT_CUDA_TRY(cudaStreamWaitEvent(get_cuda_stream_pool(res).get_stream(i), event, 0)); } } /** * @} */ } // namespace raft::resource
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/resource/thrust_policy.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/resource/cuda_stream.hpp> #include <raft/core/resource/resource_types.hpp> #include <raft/core/resources.hpp> #include <rmm/exec_policy.hpp> namespace raft::resource { class thrust_policy_resource : public resource { public: thrust_policy_resource(rmm::cuda_stream_view stream_view) : thrust_policy_(std::make_unique<rmm::exec_policy>(stream_view)) { } void* get_resource() override { return thrust_policy_.get(); } ~thrust_policy_resource() override {} private: std::unique_ptr<rmm::exec_policy> thrust_policy_; }; /** * Factory that knows how to construct a * specific raft::resource to populate * the res_t. */ class thrust_policy_resource_factory : public resource_factory { public: thrust_policy_resource_factory(rmm::cuda_stream_view stream_view) : stream_view_(stream_view) {} resource_type get_resource_type() override { return resource_type::THRUST_POLICY; } resource* make_resource() override { return new thrust_policy_resource(stream_view_); } private: rmm::cuda_stream_view stream_view_; }; /** * @defgroup resource_thrust_policy Thrust policy resource functions * @{ */ /** * Load a thrust policy from a res (and populate it on the res if needed). * @param res raft res object for managing resources * @return thrust execution policy */ inline rmm::exec_policy& get_thrust_policy(resources const& res) { if (!res.has_resource_factory(resource_type::THRUST_POLICY)) { rmm::cuda_stream_view stream = get_cuda_stream(res); res.add_resource_factory(std::make_shared<thrust_policy_resource_factory>(stream)); } return *res.get_resource<rmm::exec_policy>(resource_type::THRUST_POLICY); }; /** * @} */ } // namespace raft::resource
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/resource/cuda_stream.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cuda_runtime.h> #include <raft/core/interruptible.hpp> #include <raft/core/resource/resource_types.hpp> #include <raft/core/resources.hpp> #include <raft/util/cudart_utils.hpp> #include <rmm/cuda_stream_view.hpp> namespace raft::resource { class cuda_stream_resource : public resource { public: cuda_stream_resource(rmm::cuda_stream_view stream_view = rmm::cuda_stream_per_thread) : stream(stream_view) { } void* get_resource() override { return &stream; } ~cuda_stream_resource() override {} private: rmm::cuda_stream_view stream; }; /** * Factory that knows how to construct a specific raft::resource to populate * the resources instance. */ class cuda_stream_resource_factory : public resource_factory { public: cuda_stream_resource_factory(rmm::cuda_stream_view stream_view = rmm::cuda_stream_per_thread) : stream(stream_view) { } resource_type get_resource_type() override { return resource_type::CUDA_STREAM_VIEW; } resource* make_resource() override { return new cuda_stream_resource(stream); } private: rmm::cuda_stream_view stream; }; /** * @defgroup resource_cuda_stream CUDA stream resource functions * @{ */ /** * Load a rmm::cuda_stream_view from a resources instance (and populate it on the res * if needed). * @param res raft res object for managing resources * @return */ inline rmm::cuda_stream_view get_cuda_stream(resources const& res) { if (!res.has_resource_factory(resource_type::CUDA_STREAM_VIEW)) { res.add_resource_factory(std::make_shared<cuda_stream_resource_factory>()); } return *res.get_resource<rmm::cuda_stream_view>(resource_type::CUDA_STREAM_VIEW); }; /** * Load a rmm::cuda_stream_view from a resources instance (and populate it on the res * if needed). * @param[in] res raft resources object for managing resources * @param[in] stream_view cuda stream view */ inline void set_cuda_stream(resources const& res, rmm::cuda_stream_view stream_view) { res.add_resource_factory(std::make_shared<cuda_stream_resource_factory>(stream_view)); }; /** * @brief synchronize a specific stream * * @param[in] res the raft resources object * @param[in] stream stream to synchronize */ inline void sync_stream(const resources& res, rmm::cuda_stream_view stream) { interruptible::synchronize(stream); } /** * @brief synchronize main stream on the resources instance */ inline void sync_stream(const resources& res) { sync_stream(res, get_cuda_stream(res)); } /** * @} */ } // namespace raft::resource
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/resource/cusolver_sp_handle.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cusolverSp.h> #include <raft/core/cusolver_macros.hpp> #include <raft/core/resource/cuda_stream.hpp> #include <raft/core/resource/resource_types.hpp> #include <raft/core/resources.hpp> namespace raft::resource { /** * */ class cusolver_sp_resource : public resource { public: cusolver_sp_resource(rmm::cuda_stream_view stream) { RAFT_CUSOLVER_TRY_NO_THROW(cusolverSpCreate(&cusolver_res)); RAFT_CUSOLVER_TRY_NO_THROW(cusolverSpSetStream(cusolver_res, stream)); } void* get_resource() override { return &cusolver_res; } ~cusolver_sp_resource() override { RAFT_CUSOLVER_TRY_NO_THROW(cusolverSpDestroy(cusolver_res)); } private: cusolverSpHandle_t cusolver_res; }; /** * Factory that knows how to construct a * specific raft::resource to populate * the res_t. */ class cusolver_sp_resource_factory : public resource_factory { public: cusolver_sp_resource_factory(rmm::cuda_stream_view stream) : stream_(stream) {} resource_type get_resource_type() override { return resource_type::CUSOLVER_SP_HANDLE; } resource* make_resource() override { return new cusolver_sp_resource(stream_); } private: rmm::cuda_stream_view stream_; }; /** * @defgroup resource_cusolver_sp cuSolver SP handle resource functions * @{ */ /** * Load a cusolverSpres_t from raft res if it exists, otherwise * add it and return it. * @param[in] res the raft resources object * @return cusolver sp handle */ inline cusolverSpHandle_t get_cusolver_sp_handle(resources const& res) { if (!res.has_resource_factory(resource_type::CUSOLVER_SP_HANDLE)) { cudaStream_t stream = get_cuda_stream(res); res.add_resource_factory(std::make_shared<cusolver_sp_resource_factory>(stream)); } return *res.get_resource<cusolverSpHandle_t>(resource_type::CUSOLVER_SP_HANDLE); }; /** * @} */ } // namespace raft::resource
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/resource/cuda_event.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cuda_runtime.h> #include <raft/core/resource/resource_types.hpp> #include <raft/core/resources.hpp> #include <raft/util/cudart_utils.hpp> namespace raft::resource { class cuda_event_resource : public resource { public: cuda_event_resource() { RAFT_CUDA_TRY_NO_THROW(cudaEventCreateWithFlags(&event_, cudaEventDisableTiming)); } void* get_resource() override { return &event_; } ~cuda_event_resource() override { RAFT_CUDA_TRY_NO_THROW(cudaEventDestroy(event_)); } private: cudaEvent_t event_; }; } // namespace raft::resource
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/resource/device_id.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cuda_runtime.h> #include <raft/core/resource/resource_types.hpp> #include <raft/core/resources.hpp> #include <raft/util/cudart_utils.hpp> namespace raft::resource { class device_id_resource : public resource { public: device_id_resource() : dev_id_([]() -> int { int cur_dev = -1; RAFT_CUDA_TRY_NO_THROW(cudaGetDevice(&cur_dev)); return cur_dev; }()) { } void* get_resource() override { return &dev_id_; } ~device_id_resource() override {} private: int dev_id_; }; /** * Factory that knows how to construct a * specific raft::resource to populate * the res_t. */ class device_id_resource_factory : public resource_factory { public: resource_type get_resource_type() override { return resource_type::DEVICE_ID; } resource* make_resource() override { return new device_id_resource(); } }; /** * @defgroup resource_device_id Device ID resource functions * @{ */ /** * Load a device id from a res (and populate it on the res if needed). * @param res raft res object for managing resources * @return device id */ inline int get_device_id(resources const& res) { if (!res.has_resource_factory(resource_type::DEVICE_ID)) { res.add_resource_factory(std::make_shared<device_id_resource_factory>()); } return *res.get_resource<int>(resource_type::DEVICE_ID); }; /** * @} */ } // namespace raft::resource
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/resource/cusparse_handle.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cusparse_v2.h> #include <raft/core/cusparse_macros.hpp> #include <raft/core/resource/cuda_stream.hpp> #include <raft/core/resource/resource_types.hpp> #include <raft/core/resources.hpp> namespace raft::resource { class cusparse_resource : public resource { public: cusparse_resource(rmm::cuda_stream_view stream) { RAFT_CUSPARSE_TRY_NO_THROW(cusparseCreate(&cusparse_res)); RAFT_CUSPARSE_TRY_NO_THROW(cusparseSetStream(cusparse_res, stream)); } ~cusparse_resource() { RAFT_CUSPARSE_TRY_NO_THROW(cusparseDestroy(cusparse_res)); } void* get_resource() override { return &cusparse_res; } private: cusparseHandle_t cusparse_res; }; /** * Factory that knows how to construct a * specific raft::resource to populate * the res_t. */ class cusparse_resource_factory : public resource_factory { public: cusparse_resource_factory(rmm::cuda_stream_view stream) : stream_(stream) {} resource_type get_resource_type() override { return resource_type::CUSPARSE_HANDLE; } resource* make_resource() override { return new cusparse_resource(stream_); } private: rmm::cuda_stream_view stream_; }; /** * @defgroup resource_cusparse cuSparse handle resource functions * @{ */ /** * Load a cusparseres_t from raft res if it exists, otherwise * add it and return it. * @param[in] res the raft resources object * @return cusparse handle */ inline cusparseHandle_t get_cusparse_handle(resources const& res) { if (!res.has_resource_factory(resource_type::CUSPARSE_HANDLE)) { rmm::cuda_stream_view stream = get_cuda_stream(res); res.add_resource_factory(std::make_shared<cusparse_resource_factory>(stream)); } return *res.get_resource<cusparseHandle_t>(resource_type::CUSPARSE_HANDLE); }; /** * @} */ } // namespace raft::resource
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/resource/sub_comms.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/comms.hpp> #include <raft/core/resource/resource_types.hpp> #include <raft/core/resources.hpp> namespace raft::resource { class sub_comms_resource : public resource { public: sub_comms_resource() : communicators_() {} void* get_resource() override { return &communicators_; } ~sub_comms_resource() override {} private: std::unordered_map<std::string, std::shared_ptr<comms::comms_t>> communicators_; }; /** * Factory that knows how to construct a * specific raft::resource to populate * the res_t. */ class sub_comms_resource_factory : public resource_factory { public: resource_type get_resource_type() override { return resource_type::SUB_COMMUNICATOR; } resource* make_resource() override { return new sub_comms_resource(); } }; /** * @defgroup resource_sub_comms Subcommunicator resource functions * @{ */ inline const comms::comms_t& get_subcomm(const resources& res, std::string key) { if (!res.has_resource_factory(resource_type::SUB_COMMUNICATOR)) { res.add_resource_factory(std::make_shared<sub_comms_resource_factory>()); } auto sub_comms = res.get_resource<std::unordered_map<std::string, std::shared_ptr<comms::comms_t>>>( resource_type::SUB_COMMUNICATOR); auto sub_comm = sub_comms->at(key); RAFT_EXPECTS(nullptr != sub_comm.get(), "ERROR: Subcommunicator was not initialized"); return *sub_comm; } inline void set_subcomm(resources const& res, std::string key, std::shared_ptr<comms::comms_t> subcomm) { if (!res.has_resource_factory(resource_type::SUB_COMMUNICATOR)) { res.add_resource_factory(std::make_shared<sub_comms_resource_factory>()); } auto sub_comms = res.get_resource<std::unordered_map<std::string, std::shared_ptr<comms::comms_t>>>( resource_type::SUB_COMMUNICATOR); sub_comms->insert(std::make_pair(key, subcomm)); } /** * @} */ } // namespace raft::resource
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/resource/cublas_handle.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cublas_v2.h> #include <raft/core/cublas_macros.hpp> #include <raft/core/resource/cuda_stream.hpp> #include <raft/core/resource/resource_types.hpp> #include <raft/core/resources.hpp> namespace raft::resource { class cublas_resource : public resource { public: cublas_resource(rmm::cuda_stream_view stream) { RAFT_CUBLAS_TRY_NO_THROW(cublasCreate(&cublas_res)); RAFT_CUBLAS_TRY_NO_THROW(cublasSetStream(cublas_res, stream)); } ~cublas_resource() override { RAFT_CUBLAS_TRY_NO_THROW(cublasDestroy(cublas_res)); } void* get_resource() override { return &cublas_res; } private: cublasHandle_t cublas_res; }; /** * Factory that knows how to construct a * specific raft::resource to populate * the res_t. */ class cublas_resource_factory : public resource_factory { public: cublas_resource_factory(rmm::cuda_stream_view stream) : stream_(stream) {} resource_type get_resource_type() override { return resource_type::CUBLAS_HANDLE; } resource* make_resource() override { return new cublas_resource(stream_); } private: rmm::cuda_stream_view stream_; }; /** * @defgroup resource_cublas cuBLAS handle resource functions * @{ */ /** * Load a cublasres_t from raft res if it exists, otherwise * add it and return it. * @param[in] res the raft resources object * @return cublas handle */ inline cublasHandle_t get_cublas_handle(resources const& res) { if (!res.has_resource_factory(resource_type::CUBLAS_HANDLE)) { cudaStream_t stream = get_cuda_stream(res); res.add_resource_factory(std::make_shared<cublas_resource_factory>(stream)); } auto ret = *res.get_resource<cublasHandle_t>(resource_type::CUBLAS_HANDLE); RAFT_CUBLAS_TRY(cublasSetStream(ret, get_cuda_stream(res))); return ret; }; /** * @} */ } // namespace raft::resource
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/resource/comms.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/comms.hpp> #include <raft/core/resource/resource_types.hpp> #include <raft/core/resources.hpp> namespace raft::resource { class comms_resource : public resource { public: comms_resource(std::shared_ptr<comms::comms_t> comnumicator) : communicator_(comnumicator) {} void* get_resource() override { return &communicator_; } ~comms_resource() override {} private: std::shared_ptr<comms::comms_t> communicator_; }; /** * Factory that knows how to construct a * specific raft::resource to populate * the res_t. */ class comms_resource_factory : public resource_factory { public: comms_resource_factory(std::shared_ptr<comms::comms_t> communicator) : communicator_(communicator) { } resource_type get_resource_type() override { return resource_type::COMMUNICATOR; } resource* make_resource() override { return new comms_resource(communicator_); } private: std::shared_ptr<comms::comms_t> communicator_; }; /** * @defgroup resource_comms Comms resource functions * @{ */ inline bool comms_initialized(resources const& res) { return res.has_resource_factory(resource_type::COMMUNICATOR); } inline comms::comms_t const& get_comms(resources const& res) { RAFT_EXPECTS(comms_initialized(res), "ERROR: Communicator was not initialized\n"); return *(*res.get_resource<std::shared_ptr<comms::comms_t>>(resource_type::COMMUNICATOR)); } inline void set_comms(resources const& res, std::shared_ptr<comms::comms_t> communicator) { res.add_resource_factory(std::make_shared<comms_resource_factory>(communicator)); } /** * @} */ } // namespace raft::resource
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/resource/device_memory_resource.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/operators.hpp> #include <raft/core/resource/resource_types.hpp> #include <raft/core/resources.hpp> #include <raft/util/cudart_utils.hpp> #include <rmm/mr/device/device_memory_resource.hpp> #include <rmm/mr/device/limiting_resource_adaptor.hpp> #include <rmm/mr/device/per_device_resource.hpp> #include <rmm/mr/device/pool_memory_resource.hpp> #include <cstddef> #include <optional> namespace raft::resource { /** * \defgroup device_memory_resource Device memory resources * @{ */ class limiting_memory_resource : public resource { public: limiting_memory_resource(std::shared_ptr<rmm::mr::device_memory_resource> mr, std::size_t allocation_limit, std::optional<std::size_t> alignment) : upstream_(mr), mr_(make_adaptor(mr, allocation_limit, alignment)) { } auto get_resource() -> void* override { return &mr_; } ~limiting_memory_resource() override = default; private: std::shared_ptr<rmm::mr::device_memory_resource> upstream_; rmm::mr::limiting_resource_adaptor<rmm::mr::device_memory_resource> mr_; static inline auto make_adaptor(std::shared_ptr<rmm::mr::device_memory_resource> upstream, std::size_t limit, std::optional<std::size_t> alignment) -> rmm::mr::limiting_resource_adaptor<rmm::mr::device_memory_resource> { auto p = upstream.get(); if (alignment.has_value()) { return rmm::mr::limiting_resource_adaptor(p, limit, alignment.value()); } else { return rmm::mr::limiting_resource_adaptor(p, limit); } } }; /** * Factory that knows how to construct a specific raft::resource to populate * the resources instance. */ class workspace_resource_factory : public resource_factory { public: explicit workspace_resource_factory( std::shared_ptr<rmm::mr::device_memory_resource> mr = {nullptr}, std::optional<std::size_t> allocation_limit = std::nullopt, std::optional<std::size_t> alignment = std::nullopt) : allocation_limit_(allocation_limit.value_or(default_allocation_limit())), alignment_(alignment), mr_(mr ? mr : default_plain_resource()) { } auto get_resource_type() -> resource_type override { return resource_type::WORKSPACE_RESOURCE; } auto make_resource() -> resource* override { return new limiting_memory_resource(mr_, allocation_limit_, alignment_); } /** Construct a sensible default pool memory resource. */ static inline auto default_pool_resource(std::size_t limit) -> std::shared_ptr<rmm::mr::device_memory_resource> { // Set the default granularity to 1 GiB constexpr std::size_t kOneGb = 1024lu * 1024lu * 1024lu; // The initial size of the pool. The choice of this value only affects the performance a little // bit. Heuristics: // 1) the pool shouldn't be too big from the beginning independently of the limit; // 2) otherwise, set it to half the max size to avoid too many resize calls. auto min_size = std::min<std::size_t>(kOneGb, limit / 2lu); // The pool is going to be place behind the limiting resource adaptor. This means the user won't // be able to allocate more than 'limit' bytes of memory anyway. At the same time, the pool // itself may consume a little bit more memory than the 'limit' due to memory fragmentation. // Therefore, we look for a compromise, such that: // 1) 'limit' is accurate - the user should be more likely to run into the limiting // resource adaptor bad_alloc error than into the pool bad_alloc error. // 2) The pool doesn't grab too much memory on top of the 'limit'. auto max_size = std::min<std::size_t>(limit + kOneGb / 2lu, limit * 3lu / 2lu); auto upstream = rmm::mr::get_current_device_resource(); RAFT_LOG_DEBUG( "Setting the workspace pool resource; memory limit = %zu, initial pool size = %zu, max pool " "size = %zu.", limit, min_size, max_size); return std::make_shared<rmm::mr::pool_memory_resource<rmm::mr::device_memory_resource>>( upstream, min_size, max_size); } /** * Get the global memory resource wrapped into an unmanaged shared_ptr (with no deleter). * * Note: the lifetime of the underlying `rmm::mr::get_current_device_resource()` is managed * somewhere else, since it's passed by a raw pointer. Hence, this shared_ptr wrapper is not * allowed to delete the pointer on destruction. */ static inline auto default_plain_resource() -> std::shared_ptr<rmm::mr::device_memory_resource> { return std::shared_ptr<rmm::mr::device_memory_resource>{rmm::mr::get_current_device_resource(), void_op{}}; } private: std::size_t allocation_limit_; std::optional<std::size_t> alignment_; std::shared_ptr<rmm::mr::device_memory_resource> mr_; static inline auto default_allocation_limit() -> std::size_t { std::size_t free_size{}; std::size_t total_size{}; RAFT_CUDA_TRY(cudaMemGetInfo(&free_size, &total_size)); // Note, the workspace does not claim all this memory from the start, so it's still usable by // the main resource as well. // This limit is merely an order for algorithm internals to plan the batching accordingly. return total_size / 2; } }; /** * Load a temp workspace resource from a resources instance (and populate it on the res * if needed). * * @param res raft resources object for managing resources * @return device memory resource object */ inline auto get_workspace_resource(resources const& res) -> rmm::mr::limiting_resource_adaptor<rmm::mr::device_memory_resource>* { if (!res.has_resource_factory(resource_type::WORKSPACE_RESOURCE)) { res.add_resource_factory(std::make_shared<workspace_resource_factory>()); } return res.get_resource<rmm::mr::limiting_resource_adaptor<rmm::mr::device_memory_resource>>( resource_type::WORKSPACE_RESOURCE); }; /** Get the total size of the workspace resource. */ inline auto get_workspace_total_bytes(resources const& res) -> size_t { return get_workspace_resource(res)->get_allocation_limit(); }; /** Get the already allocated size of the workspace resource. */ inline auto get_workspace_used_bytes(resources const& res) -> size_t { return get_workspace_resource(res)->get_allocated_bytes(); }; /** Get the available size of the workspace resource. */ inline auto get_workspace_free_bytes(resources const& res) -> size_t { const auto* p = get_workspace_resource(res); return p->get_allocation_limit() - p->get_allocated_bytes(); }; /** * Set a temporary workspace resource on a resources instance. * * @param res raft resources object for managing resources * @param mr an optional RMM device_memory_resource * @param allocation_limit * the total amount of memory in bytes available to the temporary workspace resources. * @param alignment optional alignment requirements passed to RMM allocations * */ inline void set_workspace_resource(resources const& res, std::shared_ptr<rmm::mr::device_memory_resource> mr = {nullptr}, std::optional<std::size_t> allocation_limit = std::nullopt, std::optional<std::size_t> alignment = std::nullopt) { res.add_resource_factory( std::make_shared<workspace_resource_factory>(mr, allocation_limit, alignment)); }; /** * Set the temporary workspace resource to a pool on top of the global memory resource * (`rmm::mr::get_current_device_resource()`. * * @param res raft resources object for managing resources * @param allocation_limit * the total amount of memory in bytes available to the temporary workspace resources; * if not provided, a last used or default limit is used. * */ inline void set_workspace_to_pool_resource( resources const& res, std::optional<std::size_t> allocation_limit = std::nullopt) { if (!allocation_limit.has_value()) { allocation_limit = get_workspace_total_bytes(res); } res.add_resource_factory(std::make_shared<workspace_resource_factory>( workspace_resource_factory::default_pool_resource(*allocation_limit), allocation_limit, std::nullopt)); }; /** * Set the temporary workspace resource the same as the global memory resource * (`rmm::mr::get_current_device_resource()`. * * Note, the workspace resource is always limited; the limit here defines how much of the global * memory resource can be consumed by the workspace allocations. * * @param res raft resources object for managing resources * @param allocation_limit * the total amount of memory in bytes available to the temporary workspace resources. */ inline void set_workspace_to_global_resource( resources const& res, std::optional<std::size_t> allocation_limit = std::nullopt) { res.add_resource_factory(std::make_shared<workspace_resource_factory>( workspace_resource_factory::default_plain_resource(), allocation_limit, std::nullopt)); }; /** @} */ } // namespace raft::resource
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/resource/resource_types.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once namespace raft::resource { /** * @defgroup resource_types Core resource vocabulary types * @{ */ /** * @brief Resource types can apply to any resource and don't have to be host- or device-specific. */ enum resource_type { // device-specific resource types CUBLAS_HANDLE = 0, // cublas handle CUSOLVER_DN_HANDLE, // cusolver dn handle CUSOLVER_SP_HANDLE, // cusolver sp handle CUSPARSE_HANDLE, // cusparse handle CUDA_STREAM_VIEW, // view of a cuda stream CUDA_STREAM_POOL, // cuda stream pool CUDA_STREAM_SYNC_EVENT, // cuda event for syncing streams COMMUNICATOR, // raft communicator SUB_COMMUNICATOR, // raft sub communicator DEVICE_PROPERTIES, // cuda device properties DEVICE_ID, // cuda device id STREAM_VIEW, // view of a cuda stream or a placeholder in // CUDA-free builds THRUST_POLICY, // thrust execution policy WORKSPACE_RESOURCE, // rmm device memory resource LAST_KEY // reserved for the last key }; /** * @brief A resource constructs and contains an instance of * some pre-determined object type and facades that object * behind a common API. */ class resource { public: virtual void* get_resource() = 0; virtual ~resource() {} }; class empty_resource : public resource { public: empty_resource() : resource() {} void* get_resource() override { return nullptr; } ~empty_resource() override {} }; /** * @brief A resource factory knows how to construct an instance of * a specific raft::resource::resource. */ class resource_factory { public: /** * @brief Return the resource_type associated with the current factory * @return resource_type corresponding to the current factory */ virtual resource_type get_resource_type() = 0; /** * @brief Construct an instance of the factory's underlying resource. * @return resource instance */ virtual resource* make_resource() = 0; virtual ~resource_factory() {} }; /** * @brief A resource factory knows how to construct an instance of * a specific raft::resource::resource. */ class empty_resource_factory : public resource_factory { public: empty_resource_factory() : resource_factory() {} /** * @brief Return the resource_type associated with the current factory * @return resource_type corresponding to the current factory */ resource_type get_resource_type() override { return resource_type::LAST_KEY; } /** * @brief Construct an instance of the factory's underlying resource. * @return resource instance */ resource* make_resource() override { return &res; } private: empty_resource res; }; /** * @} */ } // namespace raft::resource
0
rapidsai_public_repos/raft/cpp/include/raft/core/resource
rapidsai_public_repos/raft/cpp/include/raft/core/resource/detail/device_memory_resource.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/logger.hpp> #include <raft/core/resource/device_memory_resource.hpp> #include <rmm/mr/device/device_memory_resource.hpp> #include <mutex> #include <set> #include <string> namespace raft::resource::detail { /** * Warn a user of the calling algorithm if they use the default non-pooled memory allocator, * as it may hurt the performance. * * This helper function is designed to produce the warning once for a given `user_name`. * * @param[in] res * @param[in] user_name the name of the algorithm or any other identification. * */ inline void warn_non_pool_workspace(resources const& res, std::string user_name) { // Detect if the plain cuda memory resource is used for the workspace if (rmm::mr::cuda_memory_resource{}.is_equal(*get_workspace_resource(res)->get_upstream())) { static std::set<std::string> notified_names{}; static std::mutex mutex{}; std::lock_guard<std::mutex> guard(mutex); auto [it, inserted] = notified_names.insert(std::move(user_name)); if (inserted) { RAFT_LOG_WARN( "[%s] the default cuda resource is used for the raft workspace allocations. This may lead " "to a significant slowdown for this algorithm. Consider using the default pool resource " "(`raft::resource::set_workspace_to_pool_resource`) or set your own resource explicitly " "(`raft::resource::set_workspace_resource`).", it->c_str()); } } } } // namespace raft::resource::detail
0
rapidsai_public_repos/raft/cpp/include/raft/core/resource
rapidsai_public_repos/raft/cpp/include/raft/core/resource/detail/stream_sync_event.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cuda_runtime.h> #include <raft/core/resource/cuda_event.hpp> #include <raft/core/resource/resource_types.hpp> #include <raft/core/resources.hpp> #include <raft/util/cudart_utils.hpp> namespace raft::resource::detail { /** * Factory that knows how to construct a specific raft::resource to populate * the res_t. */ class cuda_stream_sync_event_resource_factory : public resource_factory { public: resource_type get_resource_type() override { return resource_type::CUDA_STREAM_SYNC_EVENT; } resource* make_resource() override { return new cuda_event_resource(); } }; /** * Load a cudaEvent from a resources instance (and populate it on the resources instance) * if needed) for syncing the main cuda stream. * @param res raft resources instance for managing resources * @return */ inline cudaEvent_t& get_cuda_stream_sync_event(resources const& res) { if (!res.has_resource_factory(resource_type::CUDA_STREAM_SYNC_EVENT)) { res.add_resource_factory(std::make_shared<cuda_stream_sync_event_resource_factory>()); } return *res.get_resource<cudaEvent_t>(resource_type::CUDA_STREAM_SYNC_EVENT); }; } // namespace raft::resource::detail
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/detail/callback_sink.hpp
/* * Copyright (c) 2020-2022, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <iostream> #include <mutex> #define SPDLOG_HEADER_ONLY #include <spdlog/common.h> #include <spdlog/details/log_msg.h> #include <spdlog/sinks/base_sink.h> namespace spdlog::sinks { typedef void (*LogCallback)(int lvl, const char* msg); template <class Mutex> class CallbackSink : public base_sink<Mutex> { public: explicit CallbackSink(std::string tag = "spdlog", LogCallback callback = nullptr, void (*flush)() = nullptr) : _callback{callback}, _flush{flush} {}; void set_callback(LogCallback callback) { _callback = callback; } void set_flush(void (*flush)()) { _flush = flush; } protected: void sink_it_(const details::log_msg& msg) override { spdlog::memory_buf_t formatted; base_sink<Mutex>::formatter_->format(msg, formatted); std::string msg_string = fmt::to_string(formatted); if (_callback) { _callback(static_cast<int>(msg.level), msg_string.c_str()); } else { std::cout << msg_string; } } void flush_() override { if (_flush) { _flush(); } else { std::cout << std::flush; } } LogCallback _callback; void (*_flush)(); }; using callback_sink_mt = CallbackSink<std::mutex>; using callback_sink_st = CallbackSink<details::null_mutex>; } // end namespace spdlog::sinks
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/detail/span.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <limits> // numeric_limits #include <raft/core/detail/macros.hpp> #include <raft/core/mdspan_types.hpp> #include <type_traits> namespace raft { template <class ElementType, bool is_device, std::size_t Extent> class span; namespace detail { /*! * The extent E of the span returned by subspan is determined as follows: * * - If Count is not dynamic_extent, Count; * - Otherwise, if Extent is not dynamic_extent, Extent - Offset; * - Otherwise, dynamic_extent. */ template <std::size_t Extent, std::size_t Offset, std::size_t Count> struct extent_value_t : public std::integral_constant< std::size_t, Count != dynamic_extent ? Count : (Extent != dynamic_extent ? Extent - Offset : Extent)> {}; /*! * If N is dynamic_extent, the extent of the returned span E is also * dynamic_extent; otherwise it is std::size_t(sizeof(T)) * N. */ template <typename T, std::size_t Extent> struct extent_as_bytes_value_t : public std::integral_constant<std::size_t, Extent == dynamic_extent ? Extent : sizeof(T) * Extent> {}; template <std::size_t From, std::size_t To> struct is_allowed_extent_conversion_t : public std::integral_constant<bool, From == To || From == dynamic_extent || To == dynamic_extent> {}; template <class From, class To> struct is_allowed_element_type_conversion_t : public std::integral_constant<bool, std::is_convertible<From (*)[], To (*)[]>::value> {}; template <class T> struct is_span_oracle_t : std::false_type {}; template <class T, bool is_device, std::size_t Extent> struct is_span_oracle_t<span<T, is_device, Extent>> : std::true_type {}; template <class T> struct is_span_t : public is_span_oracle_t<typename std::remove_cv<T>::type> {}; template <class InputIt1, class InputIt2, class Compare> _RAFT_HOST_DEVICE constexpr auto lexicographical_compare(InputIt1 first1, InputIt1 last1, InputIt2 first2, InputIt2 last2) -> bool { Compare comp; for (; first1 != last1 && first2 != last2; ++first1, ++first2) { if (comp(*first1, *first2)) { return true; } if (comp(*first2, *first1)) { return false; } } return first1 == last1 && first2 != last2; } template <typename T, std::size_t Extent> struct span_storage { private: T* ptr_{nullptr}; public: constexpr span_storage() noexcept = default; constexpr span_storage(T* ptr, std::size_t) noexcept : ptr_{ptr} {} [[nodiscard]] constexpr auto size() const noexcept -> std::size_t { return Extent; } [[nodiscard]] constexpr auto data() const noexcept -> T* { return ptr_; } }; template <typename T> struct span_storage<T, dynamic_extent> { private: T* ptr_{nullptr}; std::size_t size_{0}; public: constexpr span_storage() noexcept = default; constexpr span_storage(T* ptr, std::size_t size) noexcept : ptr_{ptr}, size_{size} {} [[nodiscard]] constexpr auto size() const noexcept -> std::size_t { return size_; } [[nodiscard]] constexpr auto data() const noexcept -> T* { return ptr_; } }; } // namespace detail } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/detail/mdspan_util.cuh
/* * Copyright (c) 2022, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/thirdparty/mdspan/include/experimental/mdspan> #include <raft/core/detail/macros.hpp> #include <tuple> #include <utility> namespace raft::detail { template <class T, std::size_t N, std::size_t... Idx> MDSPAN_INLINE_FUNCTION constexpr auto arr_to_tup(T (&arr)[N], std::index_sequence<Idx...>) { return std::make_tuple(arr[Idx]...); } template <class T, std::size_t N> MDSPAN_INLINE_FUNCTION constexpr auto arr_to_tup(T (&arr)[N]) { return arr_to_tup(arr, std::make_index_sequence<N>{}); } template <typename T> MDSPAN_INLINE_FUNCTION auto native_popc(T v) -> int32_t { int c = 0; for (; v != 0; v &= v - 1) { c++; } return c; } MDSPAN_INLINE_FUNCTION auto popc(uint32_t v) -> int32_t { #if defined(__CUDA_ARCH__) return __popc(v); #elif defined(__GNUC__) || defined(__clang__) return __builtin_popcount(v); #else return native_popc(v); #endif // compiler } MDSPAN_INLINE_FUNCTION auto popc(uint64_t v) -> int32_t { #if defined(__CUDA_ARCH__) return __popcll(v); #elif defined(__GNUC__) || defined(__clang__) return __builtin_popcountll(v); #else return native_popc(v); #endif // compiler } } // end namespace raft::detail
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/detail/mdspan_numpy_serializer.hpp
/* * Copyright (c) 2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/device_mdspan.hpp> #include <raft/core/host_mdspan.hpp> #include <raft/core/resources.hpp> #include <algorithm> #include <complex> #include <cstdint> #include <cstring> #include <iostream> #include <map> #include <sstream> #include <string> #include <type_traits> #include <vector> namespace raft { namespace detail { namespace numpy_serializer { /** * A small implementation of NumPy serialization format. * Reference: https://numpy.org/doc/1.23/reference/generated/numpy.lib.format.html * * Adapted from https://github.com/llohse/libnpy/blob/master/include/npy.hpp, using the following * license: * * MIT License * * Copyright (c) 2021 Leon Merten Lohse * * Permission is hereby granted, free of charge, to any person obtaining a copy * of this software and associated documentation files (the "Software"), to deal * in the Software without restriction, including without limitation the rights * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell * copies of the Software, and to permit persons to whom the Software is * furnished to do so, subject to the following conditions: * * The above copyright notice and this permission notice shall be included in all * copies or substantial portions of the Software. * * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. */ #define RAFT_NUMPY_LITTLE_ENDIAN_CHAR '<' #define RAFT_NUMPY_BIG_ENDIAN_CHAR '>' #define RAFT_NUMPY_NO_ENDIAN_CHAR '|' #define RAFT_NUMPY_MAGIC_STRING "\x93NUMPY" #define RAFT_NUMPY_MAGIC_STRING_LENGTH 6 #if RAFT_SYSTEM_LITTLE_ENDIAN == 1 #define RAFT_NUMPY_HOST_ENDIAN_CHAR RAFT_NUMPY_LITTLE_ENDIAN_CHAR #else // RAFT_SYSTEM_LITTLE_ENDIAN == 1 #define RAFT_NUMPY_HOST_ENDIAN_CHAR RAFT_NUMPY_BIG_ENDIAN_CHAR #endif // RAFT_SYSTEM_LITTLE_ENDIAN == 1 using ndarray_len_t = std::uint64_t; struct dtype_t { const char byteorder; const char kind; const unsigned int itemsize; std::string to_string() const { char buf[16] = {0}; std::sprintf(buf, "%c%c%u", byteorder, kind, itemsize); return std::string(buf); } bool operator==(const dtype_t& other) const { return (byteorder == other.byteorder && kind == other.kind && itemsize == other.itemsize); } }; struct header_t { const dtype_t dtype; const bool fortran_order; const std::vector<ndarray_len_t> shape; bool operator==(const header_t& other) const { return (dtype == other.dtype && fortran_order == other.fortran_order && shape == other.shape); } }; template <class T> struct is_complex : std::false_type {}; template <class T> struct is_complex<std::complex<T>> : std::true_type {}; template <typename T, typename std::enable_if_t<std::is_floating_point_v<T>, bool> = true> inline dtype_t get_numpy_dtype() { return {RAFT_NUMPY_HOST_ENDIAN_CHAR, 'f', sizeof(T)}; } template <typename T, typename std::enable_if_t<std::is_integral_v<T> && std::is_signed_v<T>, bool> = true> inline dtype_t get_numpy_dtype() { const char endian_char = (sizeof(T) == 1 ? RAFT_NUMPY_NO_ENDIAN_CHAR : RAFT_NUMPY_HOST_ENDIAN_CHAR); return {endian_char, 'i', sizeof(T)}; } template <typename T, typename std::enable_if_t<std::is_integral_v<T> && std::is_unsigned_v<T>, bool> = true> inline dtype_t get_numpy_dtype() { const char endian_char = (sizeof(T) == 1 ? RAFT_NUMPY_NO_ENDIAN_CHAR : RAFT_NUMPY_HOST_ENDIAN_CHAR); return {endian_char, 'u', sizeof(T)}; } template <typename T, typename std::enable_if_t<is_complex<T>{}, bool> = true> inline dtype_t get_numpy_dtype() { return {RAFT_NUMPY_HOST_ENDIAN_CHAR, 'c', sizeof(T)}; } template <typename T, typename std::enable_if_t<std::is_enum_v<T>, bool> = true> inline dtype_t get_numpy_dtype() { return get_numpy_dtype<std::underlying_type_t<T>>(); } template <typename T> inline std::string tuple_to_string(const std::vector<T>& tuple) { std::ostringstream oss; if (tuple.empty()) { oss << "()"; } else if (tuple.size() == 1) { oss << "(" << tuple.front() << ",)"; } else { oss << "("; for (std::size_t i = 0; i < tuple.size() - 1; ++i) { oss << tuple[i] << ", "; } oss << tuple.back() << ")"; } return oss.str(); } inline std::string header_to_string(const header_t& header) { std::ostringstream oss; oss << "{'descr': '" << header.dtype.to_string() << "', 'fortran_order': " << (header.fortran_order ? "True" : "False") << ", 'shape': " << tuple_to_string(header.shape) << "}"; return oss.str(); } inline std::string trim(const std::string& str) { const std::string whitespace = " \t"; auto begin = str.find_first_not_of(whitespace); if (begin == std::string::npos) { return ""; } auto end = str.find_last_not_of(whitespace); return str.substr(begin, end - begin + 1); } // A poor man's parser for Python dictionary // TODO(hcho3): Consider writing a proper parser // Limitation: can only parse a flat dictionary; all values are assumed to non-objects // Limitation: must know all the keys ahead of time; you get undefined behavior if you omit any key inline std::map<std::string, std::string> parse_pydict(std::string str, const std::vector<std::string>& keys) { std::map<std::string, std::string> result; // Unwrap dictionary str = trim(str); RAFT_EXPECTS(str.front() == '{' && str.back() == '}', "Expected a Python dictionary"); str = str.substr(1, str.length() - 2); // Get the position of each key and put it in the list std::vector<std::pair<std::size_t, std::string>> positions; for (auto const& key : keys) { std::size_t pos = str.find("'" + key + "'"); RAFT_EXPECTS(pos != std::string::npos, "Missing '%s' key.", key.c_str()); positions.emplace_back(pos, key); } // Sort the list std::sort(positions.begin(), positions.end()); // Extract each key-value pair for (std::size_t i = 0; i < positions.size(); ++i) { std::string key = positions[i].second; std::size_t begin = positions[i].first; std::size_t end = (i + 1 < positions.size() ? positions[i + 1].first : std::string::npos); std::string raw_value = trim(str.substr(begin, end - begin)); if (raw_value.back() == ',') { raw_value.pop_back(); } std::size_t sep_pos = raw_value.find_first_of(":"); if (sep_pos == std::string::npos) { result[key] = ""; } else { result[key] = trim(raw_value.substr(sep_pos + 1)); } } return result; } inline std::string parse_pystring(std::string str) { RAFT_EXPECTS(str.front() == '\'' && str.back() == '\'', "Invalid Python string: %s", str.c_str()); return str.substr(1, str.length() - 2); } inline bool parse_pybool(std::string str) { if (str == "True") { return true; } else if (str == "False") { return false; } else { RAFT_FAIL("Invalid Python boolean: %s", str.c_str()); } } inline std::vector<std::string> parse_pytuple(std::string str) { std::vector<std::string> result; str = trim(str); RAFT_EXPECTS(str.front() == '(' && str.back() == ')', "Invalid Python tuple: %s", str.c_str()); str = str.substr(1, str.length() - 2); std::istringstream iss(str); for (std::string token; std::getline(iss, token, ',');) { result.push_back(trim(token)); } return result; } inline dtype_t parse_descr(std::string typestr) { RAFT_EXPECTS(typestr.length() >= 3, "Invalid typestr: Too short"); char byteorder_c = typestr.at(0); char kind_c = typestr.at(1); std::string itemsize_s = typestr.substr(2); const char endian_chars[] = { RAFT_NUMPY_LITTLE_ENDIAN_CHAR, RAFT_NUMPY_BIG_ENDIAN_CHAR, RAFT_NUMPY_NO_ENDIAN_CHAR}; const char numtype_chars[] = {'f', 'i', 'u', 'c'}; RAFT_EXPECTS(std::find(std::begin(endian_chars), std::end(endian_chars), byteorder_c) != std::end(endian_chars), "Invalid typestr: unrecognized byteorder %c", byteorder_c); RAFT_EXPECTS(std::find(std::begin(numtype_chars), std::end(numtype_chars), kind_c) != std::end(numtype_chars), "Invalid typestr: unrecognized kind %c", kind_c); unsigned int itemsize = std::stoul(itemsize_s); return {byteorder_c, kind_c, itemsize}; } inline void write_magic(std::ostream& os) { os.write(RAFT_NUMPY_MAGIC_STRING, RAFT_NUMPY_MAGIC_STRING_LENGTH); RAFT_EXPECTS(os.good(), "Error writing magic string"); // Use version 1.0 os.put(1); os.put(0); RAFT_EXPECTS(os.good(), "Error writing magic string"); } inline void read_magic(std::istream& is) { char magic_buf[RAFT_NUMPY_MAGIC_STRING_LENGTH + 2] = {0}; is.read(magic_buf, RAFT_NUMPY_MAGIC_STRING_LENGTH + 2); RAFT_EXPECTS(is.good(), "Error reading magic string"); RAFT_EXPECTS(std::memcmp(magic_buf, RAFT_NUMPY_MAGIC_STRING, RAFT_NUMPY_MAGIC_STRING_LENGTH) == 0, "The given stream does not have a valid NumPy format."); std::uint8_t version_major = magic_buf[RAFT_NUMPY_MAGIC_STRING_LENGTH]; std::uint8_t version_minor = magic_buf[RAFT_NUMPY_MAGIC_STRING_LENGTH + 1]; RAFT_EXPECTS(version_major == 1 && version_minor == 0, "Unsupported NumPy version: %d.%d", version_major, version_minor); } inline void write_header(std::ostream& os, const header_t& header) { std::string header_dict = header_to_string(header); std::size_t preamble_length = RAFT_NUMPY_MAGIC_STRING_LENGTH + 2 + 2 + header_dict.length() + 1; RAFT_EXPECTS(preamble_length < 255 * 255, "Header too long"); // Enforce 64-byte alignment std::size_t padding_len = 64 - preamble_length % 64; std::string padding(padding_len, ' '); write_magic(os); // Write header length std::uint8_t header_len_le16[2]; std::uint16_t header_len = static_cast<std::uint16_t>(header_dict.length() + padding.length() + 1); header_len_le16[0] = (header_len >> 0) & 0xff; header_len_le16[1] = (header_len >> 8) & 0xff; os.put(header_len_le16[0]); os.put(header_len_le16[1]); RAFT_EXPECTS(os.good(), "Error writing HEADER_LEN"); os << header_dict << padding << "\n"; RAFT_EXPECTS(os.good(), "Error writing header dict"); } inline std::string read_header_bytes(std::istream& is) { read_magic(is); // Read header length std::uint8_t header_len_le16[2]; is.read(reinterpret_cast<char*>(header_len_le16), 2); RAFT_EXPECTS(is.good(), "Error while reading HEADER_LEN"); const std::uint32_t header_length = (header_len_le16[0] << 0) | (header_len_le16[1] << 8); std::vector<char> header_bytes(header_length); is.read(header_bytes.data(), header_length); RAFT_EXPECTS(is.good(), "Error while reading the header"); return std::string(header_bytes.data(), header_length); } inline header_t read_header(std::istream& is) { std::string header_bytes = read_header_bytes(is); // remove trailing newline RAFT_EXPECTS(header_bytes.back() == '\n', "Invalid NumPy header"); header_bytes.pop_back(); // parse the header dict auto header_dict = parse_pydict(header_bytes, {"descr", "fortran_order", "shape"}); dtype_t descr = parse_descr(parse_pystring(header_dict["descr"])); bool fortran_order = parse_pybool(header_dict["fortran_order"]); std::vector<ndarray_len_t> shape; auto shape_tup_str = parse_pytuple(header_dict["shape"]); for (const auto& e : shape_tup_str) { shape.push_back(static_cast<ndarray_len_t>(std::stoul(e))); } RAFT_EXPECTS( descr.byteorder == RAFT_NUMPY_HOST_ENDIAN_CHAR || descr.byteorder == RAFT_NUMPY_NO_ENDIAN_CHAR, "The mdspan was serialized on a %s machine but you're attempting to load it on " "a %s machine. This use case is not currently supported.", (RAFT_SYSTEM_LITTLE_ENDIAN ? "big-endian" : "little-endian"), (RAFT_SYSTEM_LITTLE_ENDIAN ? "little-endian" : "big-endian")); return {descr, fortran_order, shape}; } template <typename ElementType, typename Extents, typename LayoutPolicy, typename AccessorPolicy> inline void serialize_host_mdspan( std::ostream& os, const raft::host_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>& obj) { static_assert(std::is_same_v<LayoutPolicy, raft::layout_c_contiguous> || std::is_same_v<LayoutPolicy, raft::layout_f_contiguous>, "The serializer only supports row-major and column-major layouts"); using obj_t = raft::host_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>; const auto dtype = get_numpy_dtype<ElementType>(); const bool fortran_order = std::is_same_v<LayoutPolicy, raft::layout_f_contiguous>; std::vector<ndarray_len_t> shape; for (typename obj_t::rank_type i = 0; i < obj.rank(); ++i) { shape.push_back(obj.extent(i)); } const header_t header = {dtype, fortran_order, shape}; write_header(os, header); // For contiguous layouts, size() == product of dimensions os.write(reinterpret_cast<const char*>(obj.data_handle()), obj.size() * sizeof(ElementType)); RAFT_EXPECTS(os.good(), "Error writing content of mdspan"); } template <typename T> inline void serialize_scalar(std::ostream& os, const T& value) { const auto dtype = get_numpy_dtype<T>(); const bool fortran_order = false; const std::vector<ndarray_len_t> shape{}; const header_t header = {dtype, fortran_order, shape}; write_header(os, header); os.write(reinterpret_cast<const char*>(&value), sizeof(T)); RAFT_EXPECTS(os.good(), "Error serializing a scalar"); } template <typename ElementType, typename Extents, typename LayoutPolicy, typename AccessorPolicy> inline void deserialize_host_mdspan( std::istream& is, const raft::host_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>& obj) { static_assert(std::is_same_v<LayoutPolicy, raft::layout_c_contiguous> || std::is_same_v<LayoutPolicy, raft::layout_f_contiguous>, "The serializer only supports row-major and column-major layouts"); using obj_t = raft::host_mdspan<ElementType, Extents, LayoutPolicy, AccessorPolicy>; // Check if given dtype and fortran_order are compatible with the mdspan const auto expected_dtype = get_numpy_dtype<ElementType>(); const bool expected_fortran_order = std::is_same_v<LayoutPolicy, raft::layout_f_contiguous>; header_t header = read_header(is); RAFT_EXPECTS(header.dtype == expected_dtype, "Expected dtype %s but got %s instead", header.dtype.to_string().c_str(), expected_dtype.to_string().c_str()); RAFT_EXPECTS(header.fortran_order == expected_fortran_order, "Wrong matrix layout; expected %s but got a different layout", (expected_fortran_order ? "Fortran layout" : "C layout")); // Check if dimensions are correct RAFT_EXPECTS(obj.rank() == header.shape.size(), "Incorrect rank: expected %zu but got %zu", obj.rank(), header.shape.size()); for (typename obj_t::rank_type i = 0; i < obj.rank(); ++i) { RAFT_EXPECTS(static_cast<ndarray_len_t>(obj.extent(i)) == header.shape[i], "Incorrect dimension: expected %zu but got %zu", static_cast<ndarray_len_t>(obj.extent(i)), header.shape[i]); } // For contiguous layouts, size() == product of dimensions is.read(reinterpret_cast<char*>(obj.data_handle()), obj.size() * sizeof(ElementType)); RAFT_EXPECTS(is.good(), "Error while reading mdspan content"); } template <typename T> inline T deserialize_scalar(std::istream& is) { // Check if dtype is correct const auto expected_dtype = get_numpy_dtype<T>(); header_t header = read_header(is); RAFT_EXPECTS(header.dtype == expected_dtype, "Expected dtype %s but got %s instead", header.dtype.to_string().c_str(), expected_dtype.to_string().c_str()); // Check if dimensions are correct; shape should be () RAFT_EXPECTS(header.shape.empty(), "Incorrect rank: expected 0 but got %zu", header.shape.size()); T value; is.read(reinterpret_cast<char*>(&value), sizeof(T)); RAFT_EXPECTS(is.good(), "Error while deserializing scalar"); return value; } } // end namespace numpy_serializer } // end namespace detail } // end namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/detail/logger.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #pragma message(__FILE__ \ " is deprecated and will be removed in future releases." \ " Please use the <raft/core/logger.hpp> version instead.") #include <raft/core/logger.hpp>
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/detail/copy.hpp
/* * Copyright (c) 2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cstdio> #include <raft/core/cuda_support.hpp> #include <raft/core/device_mdspan.hpp> #include <raft/core/error.hpp> #include <raft/core/host_mdspan.hpp> #include <raft/core/logger.hpp> #include <raft/core/mdspan.hpp> #include <raft/core/resource/stream_view.hpp> #include <raft/core/resources.hpp> #include <type_traits> #ifndef RAFT_DISABLE_CUDA #include <raft/core/cudart_utils.hpp> #include <raft/core/device_mdarray.hpp> #include <raft/core/resource/cublas_handle.hpp> #include <raft/linalg/detail/cublas_wrappers.hpp> #ifdef __CUDACC__ #include <raft/util/cuda_dev_essentials.cuh> #endif #endif namespace raft { namespace detail { template <bool B, typename DstType = void, typename SrcType = void, typename T = void, typename = void> struct mdspan_copyable : std::false_type { auto static constexpr const custom_kernel_allowed = false; auto static constexpr const custom_kernel_not_allowed = false; }; /* * A helper struct used to determine whether one mdspan type can be copied to * another and if so how */ template <typename DstType, typename SrcType, typename T> struct mdspan_copyable<true, DstType, SrcType, T, std::enable_if_t<std::conjunction_v< std::bool_constant<is_mdspan_v<std::remove_reference_t<DstType>>>, std::bool_constant<is_mdspan_v<std::remove_reference_t<SrcType>>>>>> { using dst_type = std::remove_reference_t<DstType>; using src_type = std::remove_reference_t<SrcType>; // Extents properties using dst_extents_type = typename dst_type::extents_type; using src_extents_type = typename src_type::extents_type; using index_type = std::conditional_t<(std::numeric_limits<typename dst_extents_type::index_type>::max() > std::numeric_limits<typename src_extents_type::index_type>::max()), typename dst_extents_type::index_type, typename src_extents_type::index_type>; // Dtype properties using dst_value_type = typename dst_type::value_type; using src_value_type = typename src_type::value_type; using dst_element_type = typename dst_type::element_type; using src_element_type = typename src_type::element_type; auto static constexpr const same_dtype = std::is_same_v<dst_value_type, src_value_type>; auto static constexpr const compatible_dtype = std::is_assignable_v<typename dst_type::reference, typename src_type::reference>; auto static constexpr const dst_float = std::is_same_v<dst_value_type, float>; auto static constexpr const src_float = std::is_same_v<src_value_type, float>; auto static constexpr const dst_double = std::is_same_v<dst_value_type, double>; auto static constexpr const src_double = std::is_same_v<src_value_type, double>; auto static constexpr const both_float = dst_float && src_float; auto static constexpr const both_double = dst_double && src_double; auto static constexpr const both_float_or_both_double = both_float || both_double; // Ranks auto static constexpr const dst_rank = dst_extents_type::rank(); auto static constexpr const src_rank = src_extents_type::rank(); auto static constexpr const compatible_rank = (dst_rank == src_rank); auto static constexpr const has_vector_rank = (dst_rank == 1); auto static constexpr const has_matrix_rank = (dst_rank == 2); // Layout properties using dst_layout_type = typename dst_type::layout_type; using src_layout_type = typename src_type::layout_type; auto static constexpr const same_layout = std::is_same_v<dst_layout_type, src_layout_type>; auto static check_for_unique_dst(dst_type dst) { if constexpr (!dst_type::is_always_unique()) { RAFT_EXPECTS(dst.is_unique(), "Destination mdspan must be unique for parallelized copies"); } } auto static constexpr const src_contiguous = std::disjunction_v<std::is_same<src_layout_type, layout_c_contiguous>, std::is_same<src_layout_type, layout_f_contiguous>>; auto static constexpr const dst_contiguous = std::disjunction_v<std::is_same<dst_layout_type, layout_c_contiguous>, std::is_same<dst_layout_type, layout_f_contiguous>>; auto static constexpr const both_contiguous = src_contiguous && dst_contiguous; auto static constexpr const same_underlying_layout = std::disjunction_v<std::bool_constant<same_layout>, std::bool_constant<has_vector_rank && both_contiguous>>; // Layout for intermediate tile if copying through custom kernel using tile_layout_type = std::conditional_t<src_contiguous, src_layout_type, std::conditional_t<dst_contiguous, dst_layout_type, layout_c_contiguous>>; // Accessibility auto static constexpr const dst_device_accessible = is_device_mdspan_v<dst_type>; auto static constexpr const src_device_accessible = is_device_mdspan_v<src_type>; auto static constexpr const both_device_accessible = dst_device_accessible && src_device_accessible; auto static constexpr const dst_host_accessible = is_host_mdspan_v<dst_type>; auto static constexpr const src_host_accessible = is_host_mdspan_v<src_type>; auto static constexpr const both_host_accessible = dst_host_accessible && src_host_accessible; // Allowed copy codepaths auto static constexpr const can_use_host = both_host_accessible; #if (defined(__AVX__) || defined(__SSE__) || defined(__ARM_NEON)) // TODO(wphicks): Following should be only necessary restrictions. Test if // perf actually improves once fully implemented. // auto static constexpr const can_use_simd = can_use_host && both_contiguous && // both_float_or_both_double; auto static constexpr const can_use_simd = can_use_host && both_contiguous && both_float && has_matrix_rank; #else auto static constexpr const can_use_simd = false; #endif auto static constexpr const can_use_std_copy = std::conjunction_v<std::bool_constant<can_use_host>, std::bool_constant<compatible_dtype>, std::bool_constant<both_contiguous>, std::bool_constant<same_underlying_layout>>; auto static constexpr const can_use_raft_copy = std::conjunction_v<std::bool_constant<CUDA_ENABLED>, std::bool_constant<same_dtype>, std::bool_constant<both_contiguous>, std::bool_constant<same_underlying_layout>>; // Do we need intermediate storage on device in order to perform // non-trivial layout or dtype conversions after copying source from host or // before copying converted results back to host? auto static constexpr const requires_intermediate = !both_host_accessible && !both_device_accessible && !can_use_raft_copy; auto static constexpr const use_intermediate_dst = std::conjunction_v<std::bool_constant<requires_intermediate>, std::bool_constant<src_device_accessible>>; auto static constexpr const use_intermediate_src = std::conjunction_v<std::bool_constant<requires_intermediate>, std::bool_constant<!use_intermediate_dst>>; auto static constexpr const can_use_device = std::conjunction_v<std::bool_constant<CUDA_ENABLED>, std::disjunction<std::bool_constant<both_device_accessible>, std::bool_constant<requires_intermediate>, std::bool_constant<can_use_raft_copy>>>; auto static constexpr const can_use_cublas = std::conjunction_v<std::bool_constant<can_use_device>, std::bool_constant<compatible_dtype>, std::bool_constant<both_contiguous>, std::bool_constant<!same_underlying_layout>, std::bool_constant<has_matrix_rank>, std::bool_constant<both_float_or_both_double>>; auto static constexpr const custom_kernel_allowed = std::conjunction_v<std::bool_constant<can_use_device>, std::bool_constant<!(can_use_raft_copy || can_use_cublas)>>; auto static constexpr const custom_kernel_not_allowed = !custom_kernel_allowed; auto static constexpr const custom_kernel_required = std::conjunction_v<std::bool_constant<!can_use_host>, std::bool_constant<!(can_use_raft_copy || can_use_cublas)>>; // Viable overload? auto static constexpr const value = std::conjunction_v<std::bool_constant<is_mdspan_v<src_type>>, std::bool_constant<is_mdspan_v<dst_type>>, std::bool_constant<can_use_host || can_use_device>>; using type = std::enable_if_t<value, T>; }; template <typename DstType, typename SrcType, typename T = void> using mdspan_copyable_t = typename mdspan_copyable<true, DstType, SrcType, T>::type; template <typename DstType, typename SrcType> auto static constexpr const mdspan_copyable_v = mdspan_copyable<true, DstType, SrcType, void>::value; template <typename DstType, typename SrcType> auto static constexpr const mdspan_copyable_with_kernel_v = mdspan_copyable<true, DstType, SrcType, void>::custom_kernel_allowed; template <typename DstType, typename SrcType> auto static constexpr const mdspan_copyable_not_with_kernel_v = mdspan_copyable<true, DstType, SrcType, void>::custom_kernel_not_allowed; template <typename DstType, typename SrcType, typename T = void> using mdspan_copyable_with_kernel_t = std::enable_if_t<mdspan_copyable_with_kernel_v<DstType, SrcType>, T>; template <typename DstType, typename SrcType, typename T = void> using mdspan_copyable_not_with_kernel_t = std::enable_if_t<mdspan_copyable_not_with_kernel_v<DstType, SrcType>, T>; #ifdef __CUDACC__ auto static constexpr const mdspan_copy_tile_dim = 32; auto static constexpr const mdspan_copy_tile_elems = mdspan_copy_tile_dim * mdspan_copy_tile_dim; // Helper struct to work around lack of CUDA-native std::apply template <typename IdxType, IdxType... Idx> struct index_sequence {}; template <typename IdxType, IdxType N, IdxType... Idx> struct make_index_sequence : std::conditional_t<N == IdxType{}, index_sequence<IdxType, Idx...>, make_index_sequence<IdxType, N - IdxType{1}, N - IdxType{1}, Idx...>> {}; /* template <typename LambdaT, typename ContainerT, typename IdxT, IdxT... Idx> __host__ __device__ decltype(auto) apply(LambdaT&& lambda, ContainerT&& args, index_sequence<IdxT, Idx...>) { return lambda(args[Idx]...); } template <typename LambdaT, typename ContainerT, typename IdxT, IdxT size> __host__ __device__ decltype(auto) apply(LambdaT&& lambda, ContainerT&& args) { return apply(std::forward<LambdaT>(lambda), std::forward<ContainerT>(args), make_index_sequence<IdxT, size>{}); } */ /* * Given an mdspan and an array of indices, return a reference to the * indicated element. */ template <typename MdspanType, typename IdxType, IdxType... Idx> __device__ decltype(auto) get_mdspan_elem(MdspanType md, IdxType const* indices, index_sequence<IdxType, Idx...>) { return md(indices[Idx]...); } template <typename MdspanType, typename IdxType> __device__ decltype(auto) get_mdspan_elem(MdspanType md, IdxType const* indices) { return get_mdspan_elem( md, indices, make_index_sequence<IdxType, MdspanType::extents_type::rank()>{}); } /* Advance old_indices forward by the number of mdspan elements specified * by increment. Store the result in indices. Return true if the new * indices are valid for the input mdspan. */ template <typename MdspanType, typename IdxType, typename IncrType> __device__ auto increment_indices(IdxType* indices, MdspanType const& md, IdxType const* old_indices, IdxType const* index_strides, IncrType increment) { #pragma unroll for (auto i = typename MdspanType::extents_type::rank_type{}; i < md.rank(); ++i) { increment += index_strides[i] * old_indices[i]; } #pragma unroll for (auto i = typename MdspanType::extents_type::rank_type{}; i < md.rank(); ++i) { // Iterate through dimensions in order from slowest to fastest varying for // layout_right and layout_left. Otherwise, just iterate through dimensions // in order. // // TODO(wphicks): It is possible to always iterate through dimensions in // the slowest to fastest order. Consider this or at minimum expanding to // padded layouts. auto const real_index = [](auto ind) { if constexpr (std::is_same_v<typename MdspanType::layout_type, layout_f_contiguous>) { return MdspanType::rank() - ind - 1; } else { return ind; } }(i); auto cur_index = IdxType{}; while (cur_index < md.extent(real_index) - 1 && increment >= index_strides[real_index]) { increment -= index_strides[real_index]; ++cur_index; } indices[real_index] = cur_index; } return increment == IdxType{}; } /* * WARNING: This kernel _must_ be launched with mdspan_copy_tile_dim x * mdspan_copy_tile_dim threads per block. This restriction allows for * additional optimizations at the expense of generalized launch * parameters. */ template <typename DstType, typename SrcType> RAFT_KERNEL mdspan_copy_kernel(DstType dst, SrcType src) { using config = mdspan_copyable<true, DstType, SrcType>; // An intermediate storage location for the data to be copied. __shared__ typename config::dst_value_type tile[mdspan_copy_tile_dim][mdspan_copy_tile_dim + 1]; // Compute the cumulative product of extents in order from fastest to // slowest varying extent typename config::index_type index_strides[config::dst_rank]; auto cur_stride = typename config::index_type{1}; #pragma unroll for (auto i = typename SrcType::extents_type::rank_type{}; i < config::src_rank; ++i) { // Iterate through dimensions in order from fastest to slowest varying auto const real_index = [](auto ind) { if constexpr (std::is_same_v<typename config::src_layout_type, layout_c_contiguous>) { return config::src_rank - ind - 1; } else { return ind; } }(i); index_strides[real_index] = cur_stride; cur_stride *= src.extent(real_index); } // The index of the first element in the mdspan which will be copied via // the current tile for this block. typename config::index_type tile_offset[config::dst_rank] = {0}; typename config::index_type cur_indices[config::dst_rank]; auto valid_tile = increment_indices( tile_offset, src, tile_offset, index_strides, blockIdx.x * mdspan_copy_tile_elems); while (valid_tile) { auto tile_read_x = std::is_same_v<typename config::src_layout_type, layout_f_contiguous> ? threadIdx.x : threadIdx.y; auto tile_read_y = std::is_same_v<typename config::src_layout_type, layout_f_contiguous> ? threadIdx.y : threadIdx.x; auto valid_index = increment_indices(cur_indices, src, tile_offset, index_strides, tile_read_x * mdspan_copy_tile_dim + tile_read_y); if constexpr (config::same_underlying_layout || !config::dst_contiguous) { if (valid_index) { tile[tile_read_x][tile_read_y] = get_mdspan_elem(src, cur_indices); get_mdspan_elem(dst, cur_indices) = tile[tile_read_x][tile_read_y]; } } else { if (valid_index) { tile[tile_read_x][tile_read_y] = get_mdspan_elem(src, cur_indices); } __syncthreads(); valid_index = increment_indices(cur_indices, src, tile_offset, index_strides, tile_read_y * mdspan_copy_tile_dim + tile_read_x); if (valid_index) { get_mdspan_elem(dst, cur_indices) = tile[tile_read_y][tile_read_x]; } __syncthreads(); } valid_tile = increment_indices( tile_offset, src, tile_offset, index_strides, blockDim.x * mdspan_copy_tile_elems); } } #endif template <typename DstType, typename SrcType> mdspan_copyable_t<DstType, SrcType> copy(resources const& res, DstType&& dst, SrcType&& src) { using config = mdspan_copyable<true, DstType, SrcType>; for (auto i = std::size_t{}; i < config::src_rank; ++i) { RAFT_EXPECTS(src.extent(i) == dst.extent(i), "Must copy between mdspans of the same shape"); } if constexpr (config::use_intermediate_src) { #ifndef RAFT_DISABLE_CUDA // Copy to intermediate source on device, then perform necessary // changes in layout on device, directly into final destination using mdarray_t = device_mdarray<typename config::src_value_type, typename config::src_extents_type, typename config::src_layout_type>; auto intermediate = mdarray_t(res, typename mdarray_t::mapping_type{src.extents()}, typename mdarray_t::container_policy_type{}); detail::copy(res, intermediate.view(), src); detail::copy(res, dst, intermediate.view()); #else // Not possible to reach this due to enable_ifs. Included for safety. throw(raft::non_cuda_build_error("Copying to device in non-CUDA build")); #endif } else if constexpr (config::use_intermediate_dst) { #ifndef RAFT_DISABLE_CUDA // Perform necessary changes in layout on device, then copy to final // destination on host using mdarray_t = device_mdarray<typename config::dst_value_type, typename config::dst_extents_type, typename config::dst_layout_type>; auto intermediate = mdarray_t(res, typename mdarray_t::mapping_type{dst.extents()}, typename mdarray_t::container_policy_type{}); detail::copy(res, intermediate.view(), src); detail::copy(res, dst, intermediate.view()); #else throw(raft::non_cuda_build_error("Copying from device in non-CUDA build")); #endif } else if constexpr (config::can_use_raft_copy) { #ifndef RAFT_DISABLE_CUDA raft::copy(dst.data_handle(), src.data_handle(), dst.size(), resource::get_cuda_stream(res)); #else // Not possible to reach this due to enable_ifs. Included for safety. throw(raft::non_cuda_build_error("Copying to from or on device in non-CUDA build")); #endif } else if constexpr (config::can_use_cublas) { #ifndef RAFT_DISABLE_CUDA auto constexpr const alpha = typename std::remove_reference_t<DstType>::value_type{1}; auto constexpr const beta = typename std::remove_reference_t<DstType>::value_type{0}; if constexpr (std::is_same_v<typename config::dst_layout_type, layout_c_contiguous>) { CUBLAS_TRY(linalg::detail::cublasgeam(resource::get_cublas_handle(res), CUBLAS_OP_T, CUBLAS_OP_N, dst.extent(1), dst.extent(0), &alpha, src.data_handle(), src.extent(0), &beta, dst.data_handle(), dst.extent(1), dst.data_handle(), dst.extent(1), resource::get_cuda_stream(res))); } else { CUBLAS_TRY(linalg::detail::cublasgeam(resource::get_cublas_handle(res), CUBLAS_OP_T, CUBLAS_OP_N, dst.extent(0), dst.extent(1), &alpha, src.data_handle(), src.extent(1), &beta, dst.data_handle(), dst.extent(0), dst.data_handle(), dst.extent(0), resource::get_cuda_stream(res))); } #else // Not possible to reach this due to enable_ifs. Included for safety. throw(raft::non_cuda_build_error("Copying to from or on device in non-CUDA build")); #endif } else if constexpr (config::custom_kernel_allowed) { #ifdef __CUDACC__ config::check_for_unique_dst(dst); auto const blocks = std::min( // This maximum is somewhat arbitrary. Could query the device to see // how many blocks we could reasonably allow, but this is probably // sufficient considering that this kernel will likely overlap with // real computations for most use cases. typename config::index_type{32}, raft::ceildiv(typename config::index_type(dst.size()), typename config::index_type(mdspan_copy_tile_elems))); auto constexpr const threads = dim3{mdspan_copy_tile_dim, mdspan_copy_tile_dim, 1}; mdspan_copy_kernel<<<blocks, threads, 0, resource::get_cuda_stream(res)>>>(dst, src); #else // Should never actually reach this because of enable_ifs. Included for // safety. RAFT_FAIL( "raft::copy called in a way that requires custom kernel. Please use " "raft/core/copy.cuh and include the header in a .cu file"); #endif } else if constexpr (config::can_use_std_copy) { std::copy(src.data_handle(), src.data_handle() + dst.size(), dst.data_handle()); } else { // TODO(wphicks): Make the following cache-oblivious and add SIMD support auto indices = std::array<typename config::index_type, config::dst_rank>{}; for (auto i = std::size_t{}; i < dst.size(); ++i) { if (i != 0) { if constexpr (std::is_same_v<typename config::src_layout_type, layout_c_contiguous>) { // For layout_right/layout_c_contiguous, we iterate over the // rightmost extent fastest auto dim = config::src_rank - 1; while ((++indices[dim]) == src.extent(dim)) { indices[dim] = typename config::index_type{}; --dim; } } else { // For layout_left/layout_f_contiguous (and currently all other // layouts), we iterate over the leftmost extent fastest. The // cache-oblivious implementation should work through dimensions in // order of increasing stride. auto dim = std::size_t{}; while ((++indices[dim]) == src.extent(dim)) { indices[dim] = typename config::index_type{}; ++dim; } } } std::apply(dst, indices) = std::apply(src, indices); } } } } // namespace detail } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/detail/nvtx.hpp
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <rmm/cuda_stream_view.hpp> #ifdef NVTX_ENABLED #include <cstdint> #include <cstdlib> #include <mutex> #include <nvtx3/nvToolsExt.h> #include <string> #include <type_traits> #include <unordered_map> namespace raft::common::nvtx::detail { /** * @brief An internal struct to store associated state with the color * generator */ struct color_gen_state { /** collection of all tagged colors generated so far */ static inline std::unordered_map<std::string, uint32_t> all_colors_; /** mutex for accessing the above map */ static inline std::mutex map_mutex_; /** saturation */ static inline constexpr float kS = 0.9f; /** value */ static inline constexpr float kV = 0.85f; /** golden ratio */ static inline constexpr float kPhi = 1.61803f; /** inverse golden ratio */ static inline constexpr float kInvPhi = 1.f / kPhi; }; // all h, s, v are in range [0, 1] // Ref: http://en.wikipedia.org/wiki/HSL_and_HSV#Converting_to_RGB inline auto hsv2rgb(float h, float s, float v) -> uint32_t { uint32_t out = 0xff000000u; if (s <= 0.0f) { return out; } // convert hue from [0, 1] range to [0, 360] float h_deg = h * 360.f; if (0.f > h_deg || h_deg >= 360.f) h_deg = 0.f; h_deg /= 60.f; int h_range = static_cast<int>(h_deg); float h_mod = h_deg - h_range; float x = v * (1.f - s); float y = v * (1.f - (s * h_mod)); float z = v * (1.f - (s * (1.f - h_mod))); float r, g, b; switch (h_range) { case 0: r = v; g = z; b = x; break; case 1: r = y; g = v; b = x; break; case 2: r = x; g = v; b = z; break; case 3: r = x; g = y; b = v; break; case 4: r = z; g = x; b = v; break; case 5: default: r = v; g = x; b = y; break; } out |= (uint32_t(r * 256.f) << 16); out |= (uint32_t(g * 256.f) << 8); out |= uint32_t(b * 256.f); return out; } /** * @brief Helper method to generate 'visually distinct' colors. * Inspired from https://martin.ankerl.com/2009/12/09/how-to-create-random-colors-programmatically/ * However, if an associated tag is passed, it will look up in its history for * any generated color against this tag and if found, just returns it, else * generates a new color, assigns a tag to it and stores it for future usage. * Such a thing is very useful for nvtx markers where the ranges associated * with a specific tag should ideally get the same color for the purpose of * visualizing it on nsight-systems timeline. * @param tag look for any previously generated colors with this tag or * associate the currently generated color with it * @return returns 32b RGB integer with alpha channel set of 0xff */ inline auto generate_next_color(const std::string& tag) -> uint32_t { // std::unordered_map<std::string, uint32_t> color_gen_state::all_colors_; // std::mutex color_gen_state::map_mutex_; std::lock_guard<std::mutex> guard(color_gen_state::map_mutex_); if (!tag.empty()) { auto itr = color_gen_state::all_colors_.find(tag); if (itr != color_gen_state::all_colors_.end()) { return itr->second; } } auto h = static_cast<float>(rand()) / static_cast<float>(RAND_MAX); h += color_gen_state::kInvPhi; if (h >= 1.f) h -= 1.f; auto rgb = hsv2rgb(h, color_gen_state::kS, color_gen_state::kV); if (!tag.empty()) { color_gen_state::all_colors_[tag] = rgb; } return rgb; } template <typename Domain, typename = Domain> struct domain_store { /* If `Domain::name` does not exist, this default instance is used and throws the error. */ static_assert(sizeof(Domain) != sizeof(Domain), "Type used to identify a domain must contain a static member 'char const* name'"); static inline auto value() -> const nvtxDomainHandle_t { return nullptr; } }; template <typename Domain> struct domain_store< Domain, /* Check if there exists `Domain::name` */ std::enable_if_t< std::is_same<char const*, typename std::decay<decltype(Domain::name)>::type>::value, Domain>> { static inline auto value() -> const nvtxDomainHandle_t { // NB: static modifier ensures the domain is created only once static const nvtxDomainHandle_t kValue = nvtxDomainCreateA(Domain::name); return kValue; } }; template <typename Domain> inline void push_range_name(const char* name) { nvtxEventAttributes_t event_attrib = {0}; event_attrib.version = NVTX_VERSION; event_attrib.size = NVTX_EVENT_ATTRIB_STRUCT_SIZE; event_attrib.colorType = NVTX_COLOR_ARGB; event_attrib.color = generate_next_color(name); event_attrib.messageType = NVTX_MESSAGE_TYPE_ASCII; event_attrib.message.ascii = name; nvtxDomainRangePushEx(domain_store<Domain>::value(), &event_attrib); } template <typename Domain, typename... Args> inline void push_range(const char* format, Args... args) { if constexpr (sizeof...(args) > 0) { int length = std::snprintf(nullptr, 0, format, args...); assert(length >= 0); std::vector<char> buf(length + 1); std::snprintf(buf.data(), length + 1, format, args...); push_range_name<Domain>(buf.data()); } else { push_range_name<Domain>(format); } } template <typename Domain> inline void pop_range() { nvtxDomainRangePop(domain_store<Domain>::value()); } } // namespace raft::common::nvtx::detail #else // NVTX_ENABLED namespace raft::common::nvtx::detail { template <typename Domain, typename... Args> inline void push_range(const char* format, Args... args) { } template <typename Domain> inline void pop_range() { } } // namespace raft::common::nvtx::detail #endif // NVTX_ENABLED
0
rapidsai_public_repos/raft/cpp/include/raft/core
rapidsai_public_repos/raft/cpp/include/raft/core/detail/macros.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #ifndef _RAFT_HAS_CUDA #if defined(__CUDACC__) #define _RAFT_HAS_CUDA __CUDACC__ #endif #endif #if defined(_RAFT_HAS_CUDA) #define CUDA_CONDITION_ELSE_TRUE(condition) condition #define CUDA_CONDITION_ELSE_FALSE(condition) condition #else #define CUDA_CONDITION_ELSE_TRUE(condition) true #define CUDA_CONDITION_ELSE_FALSE(condition) false #endif #ifndef _RAFT_HOST_DEVICE #if defined(_RAFT_HAS_CUDA) #define _RAFT_DEVICE __device__ #define _RAFT_HOST __host__ #define _RAFT_FORCEINLINE __forceinline__ #else #define _RAFT_DEVICE #define _RAFT_HOST #define _RAFT_FORCEINLINE inline #endif #endif #define _RAFT_HOST_DEVICE _RAFT_HOST _RAFT_DEVICE #ifndef RAFT_INLINE_FUNCTION #define RAFT_INLINE_FUNCTION _RAFT_HOST_DEVICE _RAFT_FORCEINLINE #endif #ifndef RAFT_DEVICE_INLINE_FUNCTION #define RAFT_DEVICE_INLINE_FUNCTION _RAFT_DEVICE _RAFT_FORCEINLINE #endif // The RAFT_INLINE_CONDITIONAL is a conditional inline specifier that removes // the inline specification when RAFT_COMPILED is defined. // // When RAFT_COMPILED is not defined, functions may be defined in multiple // translation units and we do not want that to lead to linker errors. // // When RAFT_COMPILED is defined, this serves two purposes: // // 1. It triggers a multiple definition error message when memory_pool-inl.hpp // (for instance) is accidentally included in multiple translation units. // // 2. We function definitions to be non-inline, because non-inline functions // symbols are always exported in the object symbol table. For inline functions, // the compiler may elide the external symbol, which results in linker errors. #ifdef RAFT_COMPILED #define RAFT_INLINE_CONDITIONAL #else #define RAFT_INLINE_CONDITIONAL inline #endif // RAFT_COMPILED // The RAFT_WEAK_FUNCTION specificies that: // // 1. A function may be defined in multiple translation units (like inline) // // 2. Must still emit an external symbol (unlike inline). This enables declaring // a function signature in an `-ext` header and defining it in a source file. // // From // https://gcc.gnu.org/onlinedocs/gcc/Common-Function-Attributes.html#Common-Function-Attributes: // // "The weak attribute causes a declaration of an external symbol to be emitted // as a weak symbol rather than a global." #define RAFT_WEAK_FUNCTION __attribute__((weak)) // The RAFT_HIDDEN_FUNCTION specificies that the function will be hidden // and therefore not callable by consumers of raft when compiled as // a shared library. // // Hidden visibility also ensures that the linker doesn't de-duplicate the // symbol across multiple `.so`. This allows multiple libraries to embed raft // without issue #define RAFT_HIDDEN_FUNCTION __attribute__((visibility("hidden"))) // The RAFT_KERNEL specificies that a kernel has hidden visibility // // Raft needs to ensure that the visibility of its __global__ function // templates have hidden visibility ( default is weak visibility). // // When kernls have weak visibility it means that if two dynamic libraries // both contain identical instantiations of a RAFT template, then the linker // will discard one of the two instantiations and use only one of them. // // Do to unique requirements of how the CUDA works this de-deduplication // can lead to the wrong kernels being called ( SM version being wrong ), // silently no kernel being called at all, or cuda runtime errors being // thrown. // // https://github.com/rapidsai/raft/issues/1722 #if defined(__CUDACC_RDC__) #define RAFT_KERNEL RAFT_HIDDEN_FUNCTION __global__ void #elif defined(_RAFT_HAS_CUDA) #define RAFT_KERNEL static __global__ void #else #define RAFT_KERNEL static void #endif /** * Some macro magic to remove optional parentheses of a macro argument. * See https://stackoverflow.com/a/62984543 */ #ifndef RAFT_DEPAREN_MAGICRAFT_DEPAREN_H1 #define RAFT_DEPAREN(X) RAFT_DEPAREN_H2(RAFT_DEPAREN_H1 X) #define RAFT_DEPAREN_H1(...) RAFT_DEPAREN_H1 __VA_ARGS__ #define RAFT_DEPAREN_H2(...) RAFT_DEPAREN_H3(__VA_ARGS__) #define RAFT_DEPAREN_H3(...) RAFT_DEPAREN_MAGIC##__VA_ARGS__ #define RAFT_DEPAREN_MAGICRAFT_DEPAREN_H1 #endif #ifndef RAFT_STRINGIFY #define RAFT_STRINGIFY_DETAIL(...) #__VA_ARGS__ #define RAFT_STRINGIFY(...) RAFT_STRINGIFY_DETAIL(__VA_ARGS__) #endif
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/comms/std_comms.hpp
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/resource/comms.hpp> #include <raft/core/resource/cuda_stream.hpp> #include <raft/core/resources.hpp> #include <raft/comms/comms.hpp> #include <raft/comms/detail/std_comms.hpp> #include <iostream> #include <nccl.h> #include <ucp/api/ucp.h> namespace raft { namespace comms { using std_comms = detail::std_comms; /** * @defgroup std_comms_factory std_comms Factory functions * @{ */ /** * Factory function to construct a RAFT NCCL communicator and inject it into a * RAFT handle. * * @param handle raft::resources for injecting the comms * @param nccl_comm initialized NCCL communicator to use for collectives * @param num_ranks number of ranks in communicator clique * @param rank rank of local instance * * @code{.cpp} * #include <raft/comms/std_comms.hpp> * #include <raft/core/device_mdarray.hpp> * * ncclComm_t nccl_comm; * raft::raft::resources handle; * * build_comms_nccl_only(&handle, nccl_comm, 5, 0); * ... * const auto& comm = resource::get_comms(handle); * auto gather_data = raft::make_device_vector<float>(handle, comm.get_size()); * ... * comm.allgather((gather_data.data_handle())[comm.get_rank()], * gather_data.data_handle(), * 1, * resource::get_cuda_stream(handle)); * * comm.sync_stream(resource::get_cuda_stream(handle)); * @endcode */ void build_comms_nccl_only(resources* handle, ncclComm_t nccl_comm, int num_ranks, int rank) { cudaStream_t stream = resource::get_cuda_stream(*handle); auto communicator = std::make_shared<comms_t>( std::unique_ptr<comms_iface>(new raft::comms::std_comms(nccl_comm, num_ranks, rank, stream))); resource::set_comms(*handle, communicator); } /** * Factory function to construct a RAFT NCCL+UCX and inject it into a RAFT * handle. * * @param handle raft::resources for injecting the comms * @param nccl_comm initialized NCCL communicator to use for collectives * @param ucp_worker of local process * Note: This is purposefully left as void* so that the ucp_worker_h * doesn't need to be exposed through the cython layer * @param eps array of ucp_ep_h instances. * Note: This is purposefully left as void* so that * the ucp_ep_h doesn't need to be exposed through the cython layer. * @param num_ranks number of ranks in communicator clique * @param rank rank of local instance * * @code{.cpp} * #include <raft/comms/std_comms.hpp> * #include <raft/core/device_mdarray.hpp> * * ncclComm_t nccl_comm; * raft::raft::resources handle; * ucp_worker_h ucp_worker; * ucp_ep_h *ucp_endpoints_arr; * * build_comms_nccl_ucx(&handle, nccl_comm, &ucp_worker, ucp_endpoints_arr, 5, 0); * ... * const auto& comm = resource::get_comms(handle); * auto gather_data = raft::make_device_vector<float>(handle, comm.get_size()); * ... * comm.allgather((gather_data.data_handle())[comm.get_rank()], * gather_data.data_handle(), * 1, * resource::get_cuda_stream(handle)); * * comm.sync_stream(resource::get_cuda_stream(handle)); * @endcode */ void build_comms_nccl_ucx( resources* handle, ncclComm_t nccl_comm, void* ucp_worker, void* eps, int num_ranks, int rank) { auto eps_sp = std::make_shared<ucp_ep_h*>(new ucp_ep_h[num_ranks]); auto size_t_ep_arr = reinterpret_cast<size_t*>(eps); for (int i = 0; i < num_ranks; i++) { size_t ptr = size_t_ep_arr[i]; auto ucp_ep_v = reinterpret_cast<ucp_ep_h*>(*eps_sp); if (ptr != 0) { auto eps_ptr = reinterpret_cast<ucp_ep_h>(size_t_ep_arr[i]); ucp_ep_v[i] = eps_ptr; } else { ucp_ep_v[i] = nullptr; } } cudaStream_t stream = resource::get_cuda_stream(*handle); auto communicator = std::make_shared<comms_t>(std::unique_ptr<comms_iface>(new raft::comms::std_comms( nccl_comm, (ucp_worker_h)ucp_worker, eps_sp, num_ranks, rank, stream))); resource::set_comms(*handle, communicator); } /** * @} */ inline void nccl_unique_id_from_char(ncclUniqueId* id, char* uniqueId) { memcpy(id->internal, uniqueId, NCCL_UNIQUE_ID_BYTES); } inline void get_nccl_unique_id(char* uid) { ncclUniqueId id; ncclGetUniqueId(&id); memcpy(uid, id.internal, NCCL_UNIQUE_ID_BYTES); } }; // namespace comms }; // end namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/comms/mpi_comms.hpp
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/comms/comms.hpp> #include <raft/comms/detail/mpi_comms.hpp> #include <raft/core/resource/comms.hpp> #include <raft/core/resource/cuda_stream.hpp> namespace raft { namespace comms { using mpi_comms = detail::mpi_comms; /** * @defgroup mpi_comms_factory MPI Comms Factory Functions * @{ */ /** * Given a properly initialized MPI_Comm, construct an instance of RAFT's * MPI Communicator and inject it into the given RAFT handle instance * @param handle raft handle for managing expensive resources * @param comm an initialized MPI communicator * * @code{.cpp} * #include <raft/comms/mpi_comms.hpp> * #include <raft/core/device_mdarray.hpp> * * MPI_Comm mpi_comm; * raft::raft::resources handle; * * initialize_mpi_comms(&handle, mpi_comm); * ... * const auto& comm = resource::get_comms(handle); * auto gather_data = raft::make_device_vector<float>(handle, comm.get_size()); * ... * comm.allgather((gather_data.data_handle())[comm.get_rank()], * gather_data.data_handle(), * 1, * resource::get_cuda_stream(handle)); * * comm.sync_stream(resource::get_cuda_stream(handle)); * @endcode */ inline void initialize_mpi_comms(resources* handle, MPI_Comm comm) { auto communicator = std::make_shared<comms_t>( std::unique_ptr<comms_iface>(new mpi_comms(comm, false, resource::get_cuda_stream(*handle)))); resource::set_comms(*handle, communicator); }; /** * @} */ }; // namespace comms }; // end namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/comms/comms_test.hpp
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/comms/comms.hpp> #include <raft/comms/detail/test.hpp> #include <raft/core/resources.hpp> namespace raft { namespace comms { /** * @brief A simple sanity check that NCCL is able to perform a collective operation * * @param[in] handle the raft handle to use. This is expected to already have an * initialized comms instance. * @param[in] root the root rank id */ bool test_collective_allreduce(raft::resources const& handle, int root) { return detail::test_collective_allreduce(handle, root); } /** * @brief A simple sanity check that NCCL is able to perform a collective operation * * @param[in] handle the raft handle to use. This is expected to already have an * initialized comms instance. * @param[in] root the root rank id */ bool test_collective_broadcast(raft::resources const& handle, int root) { return detail::test_collective_broadcast(handle, root); } /** * @brief A simple sanity check that NCCL is able to perform a collective reduce * * @param[in] handle the raft handle to use. This is expected to already have an * initialized comms instance. * @param[in] root the root rank id */ bool test_collective_reduce(raft::resources const& handle, int root) { return detail::test_collective_reduce(handle, root); } /** * @brief A simple sanity check that NCCL is able to perform a collective allgather * * @param[in] handle the raft handle to use. This is expected to already have an * initialized comms instance. * @param[in] root the root rank id */ bool test_collective_allgather(raft::resources const& handle, int root) { return detail::test_collective_allgather(handle, root); } /** * @brief A simple sanity check that NCCL is able to perform a collective gather * * @param[in] handle the raft handle to use. This is expected to already have an * initialized comms instance. * @param[in] root the root rank id */ bool test_collective_gather(raft::resources const& handle, int root) { return detail::test_collective_gather(handle, root); } /** * @brief A simple sanity check that NCCL is able to perform a collective gatherv * * @param[in] handle the raft handle to use. This is expected to already have an * initialized comms instance. * @param[in] root the root rank id */ bool test_collective_gatherv(raft::resources const& handle, int root) { return detail::test_collective_gatherv(handle, root); } /** * @brief A simple sanity check that NCCL is able to perform a collective reducescatter * * @param[in] handle the raft handle to use. This is expected to already have an * initialized comms instance. * @param[in] root the root rank id */ bool test_collective_reducescatter(raft::resources const& handle, int root) { return detail::test_collective_reducescatter(handle, root); } /** * A simple sanity check that UCX is able to send messages between all ranks * * @param[in] h the raft handle to use. This is expected to already have an * initialized comms instance. * @param[in] numTrials number of iterations of all-to-all messaging to perform */ bool test_pointToPoint_simple_send_recv(raft::resources const& h, int numTrials) { return detail::test_pointToPoint_simple_send_recv(h, numTrials); } /** * A simple sanity check that device is able to send OR receive. * * @param h the raft handle to use. This is expected to already have an * initialized comms instance. * @param numTrials number of iterations of send or receive messaging to perform */ bool test_pointToPoint_device_send_or_recv(raft::resources const& h, int numTrials) { return detail::test_pointToPoint_device_send_or_recv(h, numTrials); } /** * A simple sanity check that device is able to send and receive at the same time. * * @param h the raft handle to use. This is expected to already have an * initialized comms instance. * @param numTrials number of iterations of send or receive messaging to perform */ bool test_pointToPoint_device_sendrecv(raft::resources const& h, int numTrials) { return detail::test_pointToPoint_device_sendrecv(h, numTrials); } /** * A simple sanity check that device is able to perform multiple concurrent sends and receives. * * @param h the raft handle to use. This is expected to already have an * initialized comms instance. * @param numTrials number of iterations of send or receive messaging to perform */ bool test_pointToPoint_device_multicast_sendrecv(raft::resources const& h, int numTrials) { return detail::test_pointToPoint_device_multicast_sendrecv(h, numTrials); } /** * A simple test that the comms can be split into 2 separate subcommunicators * * @param h the raft handle to use. This is expected to already have an * initialized comms instance. * @param n_colors number of different colors to test */ bool test_commsplit(raft::resources const& h, int n_colors) { return detail::test_commsplit(h, n_colors); } } // namespace comms }; // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/comms/comms.hpp
/* * Copyright (c) 2021-2022, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ /** * This file is deprecated and will be removed in release 22.06. * Please use core/comms.hpp instead. */ #pragma once #include <raft/core/comms.hpp>
0
rapidsai_public_repos/raft/cpp/include/raft/comms
rapidsai_public_repos/raft/cpp/include/raft/comms/detail/util.hpp
/* * Copyright (c) 2021-2022, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/interruptible.hpp> #include <nccl.h> #include <raft/core/error.hpp> #include <string> /** * @brief Error checking macro for NCCL runtime API functions. * * Invokes a NCCL runtime API function call, if the call does not return ncclSuccess, throws an * exception detailing the NCCL error that occurred */ #define RAFT_NCCL_TRY(call) \ do { \ ncclResult_t const status = (call); \ if (ncclSuccess != status) { \ std::string msg{}; \ SET_ERROR_MSG(msg, \ "NCCL error encountered at: ", \ "call='%s', Reason=%d:%s", \ #call, \ status, \ ncclGetErrorString(status)); \ throw raft::logic_error(msg); \ } \ } while (0); // FIXME: Remove after consumer rename #ifndef NCCL_TRY #define NCCL_TRY(call) RAFT_NCCL_TRY(call) #endif #define RAFT_NCCL_TRY_NO_THROW(call) \ do { \ ncclResult_t status = call; \ if (ncclSuccess != status) { \ printf("NCCL call='%s' failed. Reason:%s\n", #call, ncclGetErrorString(status)); \ } \ } while (0) // FIXME: Remove after consumer rename #ifndef NCCL_TRY_NO_THROW #define NCCL_TRY_NO_THROW(call) RAFT_NCCL_TRY_NO_THROW(call) #endif namespace raft { namespace comms { namespace detail { constexpr size_t get_datatype_size(const datatype_t datatype) { switch (datatype) { case datatype_t::CHAR: return sizeof(char); case datatype_t::UINT8: return sizeof(uint8_t); case datatype_t::INT32: return sizeof(int); case datatype_t::UINT32: return sizeof(unsigned int); case datatype_t::INT64: return sizeof(int64_t); case datatype_t::UINT64: return sizeof(uint64_t); case datatype_t::FLOAT32: return sizeof(float); case datatype_t::FLOAT64: return sizeof(double); default: throw "Unsupported datatype"; } } constexpr ncclDataType_t get_nccl_datatype(const datatype_t datatype) { switch (datatype) { case datatype_t::CHAR: return ncclChar; case datatype_t::UINT8: return ncclUint8; case datatype_t::INT32: return ncclInt; case datatype_t::UINT32: return ncclUint32; case datatype_t::INT64: return ncclInt64; case datatype_t::UINT64: return ncclUint64; case datatype_t::FLOAT32: return ncclFloat; case datatype_t::FLOAT64: return ncclDouble; default: throw "Unsupported datatype"; } } constexpr ncclRedOp_t get_nccl_op(const op_t op) { switch (op) { case op_t::SUM: return ncclSum; case op_t::PROD: return ncclProd; case op_t::MIN: return ncclMin; case op_t::MAX: return ncclMax; default: throw "Unsupported datatype"; } } inline status_t nccl_sync_stream(ncclComm_t comm, cudaStream_t stream) { cudaError_t cudaErr; ncclResult_t ncclErr, ncclAsyncErr; while (1) { cudaErr = cudaStreamQuery(stream); if (cudaErr == cudaSuccess) return status_t::SUCCESS; if (cudaErr != cudaErrorNotReady) { // An error occurred querying the status of the stream_ return status_t::ERROR; } ncclErr = ncclCommGetAsyncError(comm, &ncclAsyncErr); if (ncclErr != ncclSuccess) { // An error occurred retrieving the asynchronous error return status_t::ERROR; } if (ncclAsyncErr != ncclSuccess || !interruptible::yield_no_throw()) { // An asynchronous error happened. Stop the operation and destroy // the communicator ncclErr = ncclCommAbort(comm); if (ncclErr != ncclSuccess) // Caller may abort with an exception or try to re-create a new communicator. return status_t::ABORT; // TODO: shouldn't we place status_t::ERROR above under the condition, and // status_t::ABORT below here (i.e. after successful ncclCommAbort)? } // Let other threads (including NCCL threads) use the CPU. std::this_thread::yield(); } } }; // namespace detail }; // namespace comms }; // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/comms
rapidsai_public_repos/raft/cpp/include/raft/comms/detail/std_comms.hpp
/* * Copyright (c) 2020-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/comms/comms.hpp> #include <raft/comms/detail/ucp_helper.hpp> #include <raft/comms/detail/util.hpp> #include <raft/core/resources.hpp> #include <rmm/device_scalar.hpp> #include <rmm/device_uvector.hpp> #include <raft/core/error.hpp> #include <raft/util/cudart_utils.hpp> #include <thrust/iterator/zip_iterator.h> #include <cuda_runtime.h> #include <ucp/api/ucp.h> #include <ucp/api/ucp_def.h> #include <nccl.h> #include <unordered_map> #include <unordered_set> #include <utility> #include <algorithm> #include <chrono> #include <cstdio> #include <exception> #include <memory> #include <stdlib.h> #include <thread> #include <time.h> namespace raft { namespace comms { namespace detail { class std_comms : public comms_iface { public: std_comms() = delete; /** * @brief Constructor for collective + point-to-point operation. * @param nccl_comm initialized nccl comm * @param ucp_worker initialized ucp_worker instance * @param eps shared pointer to array of ucp endpoints * @param num_ranks number of ranks in the cluster * @param rank rank of the current worker * @param stream cuda stream for synchronizing and ordering collective operations * @param subcomms_ucp use ucp for subcommunicators */ std_comms(ncclComm_t nccl_comm, ucp_worker_h ucp_worker, std::shared_ptr<ucp_ep_h*> eps, int num_ranks, int rank, rmm::cuda_stream_view stream, bool subcomms_ucp = true) : nccl_comm_(nccl_comm), stream_(stream), status_(stream), num_ranks_(num_ranks), rank_(rank), subcomms_ucp_(subcomms_ucp), ucp_worker_(ucp_worker), ucp_eps_(eps), next_request_id_(0) { initialize(); }; /** * @brief constructor for collective-only operation * @param nccl_comm initialized nccl communicator * @param num_ranks size of the cluster * @param rank rank of the current worker * @param stream stream for ordering collective operations */ std_comms(const ncclComm_t nccl_comm, int num_ranks, int rank, rmm::cuda_stream_view stream) : nccl_comm_(nccl_comm), stream_(stream), status_(stream), num_ranks_(num_ranks), rank_(rank), subcomms_ucp_(false) { initialize(); }; void initialize() { status_.set_value_to_zero_async(stream_); buf_ = status_.data(); } ~std_comms() { requests_in_flight_.clear(); free_requests_.clear(); } int get_size() const { return num_ranks_; } int get_rank() const { return rank_; } std::unique_ptr<comms_iface> comm_split(int color, int key) const { rmm::device_uvector<int> d_colors(get_size(), stream_); rmm::device_uvector<int> d_keys(get_size(), stream_); update_device(d_colors.data() + get_rank(), &color, 1, stream_); update_device(d_keys.data() + get_rank(), &key, 1, stream_); allgather(d_colors.data() + get_rank(), d_colors.data(), 1, datatype_t::INT32, stream_); allgather(d_keys.data() + get_rank(), d_keys.data(), 1, datatype_t::INT32, stream_); this->sync_stream(stream_); std::vector<int> h_colors(get_size()); std::vector<int> h_keys(get_size()); update_host(h_colors.data(), d_colors.data(), get_size(), stream_); update_host(h_keys.data(), d_keys.data(), get_size(), stream_); this->sync_stream(stream_); ncclComm_t nccl_comm; // Create a structure to allgather... ncclUniqueId id{}; rmm::device_uvector<ncclUniqueId> d_nccl_ids(get_size(), stream_); if (key == 0) { RAFT_NCCL_TRY(ncclGetUniqueId(&id)); } update_device(d_nccl_ids.data() + get_rank(), &id, 1, stream_); allgather(d_nccl_ids.data() + get_rank(), d_nccl_ids.data(), sizeof(ncclUniqueId), datatype_t::UINT8, stream_); auto offset = std::distance(thrust::make_zip_iterator(h_colors.begin(), h_keys.begin()), std::find_if(thrust::make_zip_iterator(h_colors.begin(), h_keys.begin()), thrust::make_zip_iterator(h_colors.end(), h_keys.end()), [color](auto tuple) { return thrust::get<0>(tuple) == color; })); auto subcomm_size = std::count(h_colors.begin(), h_colors.end(), color); update_host(&id, d_nccl_ids.data() + offset, 1, stream_); this->sync_stream(stream_); RAFT_NCCL_TRY(ncclCommInitRank(&nccl_comm, subcomm_size, id, key)); return std::unique_ptr<comms_iface>(new std_comms(nccl_comm, subcomm_size, key, stream_)); } void barrier() const { allreduce(buf_, buf_, 1, datatype_t::INT32, op_t::SUM, stream_); ASSERT(sync_stream(stream_) == status_t::SUCCESS, "ERROR: syncStream failed. This can be caused by a failed rank_."); } void get_request_id(request_t* req) const { request_t req_id; if (this->free_requests_.empty()) req_id = this->next_request_id_++; else { auto it = this->free_requests_.begin(); req_id = *it; this->free_requests_.erase(it); } *req = req_id; } void isend(const void* buf, size_t size, int dest, int tag, request_t* request) const { ASSERT(ucp_worker_ != nullptr, "ERROR: UCX comms not initialized on communicator."); get_request_id(request); ucp_ep_h ep_ptr = (*ucp_eps_)[dest]; ucp_request* ucp_req = (ucp_request*)malloc(sizeof(ucp_request)); this->ucp_handler_.ucp_isend(ucp_req, ep_ptr, buf, size, tag, default_tag_mask, get_rank()); requests_in_flight_.insert(std::make_pair(*request, ucp_req)); } void irecv(void* buf, size_t size, int source, int tag, request_t* request) const { ASSERT(ucp_worker_ != nullptr, "ERROR: UCX comms not initialized on communicator."); get_request_id(request); ucp_ep_h ep_ptr = (*ucp_eps_)[source]; ucp_tag_t tag_mask = default_tag_mask; ucp_request* ucp_req = (ucp_request*)malloc(sizeof(ucp_request)); ucp_handler_.ucp_irecv(ucp_req, ucp_worker_, ep_ptr, buf, size, tag, tag_mask, source); requests_in_flight_.insert(std::make_pair(*request, ucp_req)); } void waitall(int count, request_t array_of_requests[]) const { ASSERT(ucp_worker_ != nullptr, "ERROR: UCX comms not initialized on communicator."); std::vector<ucp_request*> requests; requests.reserve(count); time_t start = time(NULL); for (int i = 0; i < count; ++i) { auto req_it = requests_in_flight_.find(array_of_requests[i]); ASSERT(requests_in_flight_.end() != req_it, "ERROR: waitall on invalid request: %d", array_of_requests[i]); requests.push_back(req_it->second); free_requests_.insert(req_it->first); requests_in_flight_.erase(req_it); } while (requests.size() > 0) { time_t now = time(NULL); // Timeout if we have not gotten progress or completed any requests // in 10 or more seconds. ASSERT(now - start < 10, "Timed out waiting for requests."); for (std::vector<ucp_request*>::iterator it = requests.begin(); it != requests.end();) { bool restart = false; // resets the timeout when any progress was made // Causes UCP to progress through the send/recv message queue while (ucp_worker_progress(ucp_worker_) != 0) { restart = true; } auto req = *it; // If the message needs release, we know it will be sent/received // asynchronously, so we will need to track and verify its state if (req->needs_release) { ASSERT(UCS_PTR_IS_PTR(req->req), "UCX Request Error. Request is not valid UCX pointer"); ASSERT(!UCS_PTR_IS_ERR(req->req), "UCX Request Error: %d\n", UCS_PTR_STATUS(req->req)); ASSERT(req->req->completed == 1 || req->req->completed == 0, "request->completed not a valid value: %d\n", req->req->completed); } // If a message was sent synchronously (eg. completed before // `isend`/`irecv` completed) or an asynchronous message // is complete, we can go ahead and clean it up. if (!req->needs_release || req->req->completed == 1) { restart = true; // perform cleanup ucp_handler_.free_ucp_request(req); // remove from pending requests it = requests.erase(it); } else { ++it; } // if any progress was made, reset the timeout start time if (restart) { start = time(NULL); } } } } void allreduce(const void* sendbuff, void* recvbuff, size_t count, datatype_t datatype, op_t op, cudaStream_t stream) const { RAFT_NCCL_TRY(ncclAllReduce( sendbuff, recvbuff, count, get_nccl_datatype(datatype), get_nccl_op(op), nccl_comm_, stream)); } void bcast(void* buff, size_t count, datatype_t datatype, int root, cudaStream_t stream) const { RAFT_NCCL_TRY( ncclBroadcast(buff, buff, count, get_nccl_datatype(datatype), root, nccl_comm_, stream)); } void bcast(const void* sendbuff, void* recvbuff, size_t count, datatype_t datatype, int root, cudaStream_t stream) const { RAFT_NCCL_TRY(ncclBroadcast( sendbuff, recvbuff, count, get_nccl_datatype(datatype), root, nccl_comm_, stream)); } void reduce(const void* sendbuff, void* recvbuff, size_t count, datatype_t datatype, op_t op, int root, cudaStream_t stream) const { RAFT_NCCL_TRY(ncclReduce(sendbuff, recvbuff, count, get_nccl_datatype(datatype), get_nccl_op(op), root, nccl_comm_, stream)); } void allgather(const void* sendbuff, void* recvbuff, size_t sendcount, datatype_t datatype, cudaStream_t stream) const { RAFT_NCCL_TRY(ncclAllGather( sendbuff, recvbuff, sendcount, get_nccl_datatype(datatype), nccl_comm_, stream)); } void allgatherv(const void* sendbuf, void* recvbuf, const size_t* recvcounts, const size_t* displs, datatype_t datatype, cudaStream_t stream) const { // From: "An Empirical Evaluation of Allgatherv on Multi-GPU Systems" - // https://arxiv.org/pdf/1812.05964.pdf Listing 1 on page 4. RAFT_EXPECTS(num_ranks_ <= 2048, "# NCCL operations between ncclGroupStart & ncclGroupEnd shouldn't exceed 2048."); RAFT_NCCL_TRY(ncclGroupStart()); for (int root = 0; root < num_ranks_; ++root) { size_t dtype_size = get_datatype_size(datatype); RAFT_NCCL_TRY(ncclBroadcast(sendbuf, static_cast<char*>(recvbuf) + displs[root] * dtype_size, recvcounts[root], get_nccl_datatype(datatype), root, nccl_comm_, stream)); } RAFT_NCCL_TRY(ncclGroupEnd()); } void gather(const void* sendbuff, void* recvbuff, size_t sendcount, datatype_t datatype, int root, cudaStream_t stream) const { size_t dtype_size = get_datatype_size(datatype); RAFT_NCCL_TRY(ncclGroupStart()); if (get_rank() == root) { for (int r = 0; r < get_size(); ++r) { RAFT_NCCL_TRY(ncclRecv(static_cast<char*>(recvbuff) + sendcount * r * dtype_size, sendcount, get_nccl_datatype(datatype), r, nccl_comm_, stream)); } } RAFT_NCCL_TRY( ncclSend(sendbuff, sendcount, get_nccl_datatype(datatype), root, nccl_comm_, stream)); RAFT_NCCL_TRY(ncclGroupEnd()); } void gatherv(const void* sendbuff, void* recvbuff, size_t sendcount, const size_t* recvcounts, const size_t* displs, datatype_t datatype, int root, cudaStream_t stream) const { size_t dtype_size = get_datatype_size(datatype); RAFT_NCCL_TRY(ncclGroupStart()); if (get_rank() == root) { for (int r = 0; r < get_size(); ++r) { RAFT_NCCL_TRY(ncclRecv(static_cast<char*>(recvbuff) + displs[r] * dtype_size, recvcounts[r], get_nccl_datatype(datatype), r, nccl_comm_, stream)); } } RAFT_NCCL_TRY( ncclSend(sendbuff, sendcount, get_nccl_datatype(datatype), root, nccl_comm_, stream)); RAFT_NCCL_TRY(ncclGroupEnd()); } void reducescatter(const void* sendbuff, void* recvbuff, size_t recvcount, datatype_t datatype, op_t op, cudaStream_t stream) const { RAFT_NCCL_TRY(ncclReduceScatter(sendbuff, recvbuff, recvcount, get_nccl_datatype(datatype), get_nccl_op(op), nccl_comm_, stream)); } status_t sync_stream(cudaStream_t stream) const { return nccl_sync_stream(nccl_comm_, stream); } // if a thread is sending & receiving at the same time, use device_sendrecv to avoid deadlock void device_send(const void* buf, size_t size, int dest, cudaStream_t stream) const { RAFT_NCCL_TRY(ncclSend(buf, size, ncclUint8, dest, nccl_comm_, stream)); } // if a thread is sending & receiving at the same time, use device_sendrecv to avoid deadlock void device_recv(void* buf, size_t size, int source, cudaStream_t stream) const { RAFT_NCCL_TRY(ncclRecv(buf, size, ncclUint8, source, nccl_comm_, stream)); } void device_sendrecv(const void* sendbuf, size_t sendsize, int dest, void* recvbuf, size_t recvsize, int source, cudaStream_t stream) const { // ncclSend/ncclRecv pair needs to be inside ncclGroupStart/ncclGroupEnd to avoid deadlock RAFT_NCCL_TRY(ncclGroupStart()); RAFT_NCCL_TRY(ncclSend(sendbuf, sendsize, ncclUint8, dest, nccl_comm_, stream)); RAFT_NCCL_TRY(ncclRecv(recvbuf, recvsize, ncclUint8, source, nccl_comm_, stream)); RAFT_NCCL_TRY(ncclGroupEnd()); } void device_multicast_sendrecv(const void* sendbuf, std::vector<size_t> const& sendsizes, std::vector<size_t> const& sendoffsets, std::vector<int> const& dests, void* recvbuf, std::vector<size_t> const& recvsizes, std::vector<size_t> const& recvoffsets, std::vector<int> const& sources, cudaStream_t stream) const { // ncclSend/ncclRecv pair needs to be inside ncclGroupStart/ncclGroupEnd to avoid deadlock RAFT_NCCL_TRY(ncclGroupStart()); for (size_t i = 0; i < sendsizes.size(); ++i) { RAFT_NCCL_TRY(ncclSend(static_cast<const char*>(sendbuf) + sendoffsets[i], sendsizes[i], ncclUint8, dests[i], nccl_comm_, stream)); } for (size_t i = 0; i < recvsizes.size(); ++i) { RAFT_NCCL_TRY(ncclRecv(static_cast<char*>(recvbuf) + recvoffsets[i], recvsizes[i], ncclUint8, sources[i], nccl_comm_, stream)); } RAFT_NCCL_TRY(ncclGroupEnd()); } void group_start() const { RAFT_NCCL_TRY(ncclGroupStart()); } void group_end() const { RAFT_NCCL_TRY(ncclGroupEnd()); } private: ncclComm_t nccl_comm_; cudaStream_t stream_; rmm::device_scalar<int32_t> status_; int32_t* buf_; int num_ranks_; int rank_; bool subcomms_ucp_; comms_ucp_handler ucp_handler_; ucp_worker_h ucp_worker_; std::shared_ptr<ucp_ep_h*> ucp_eps_; mutable request_t next_request_id_; mutable std::unordered_map<request_t, struct ucp_request*> requests_in_flight_; mutable std::unordered_set<request_t> free_requests_; }; } // namespace detail } // end namespace comms } // end namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/comms
rapidsai_public_repos/raft/cpp/include/raft/comms/detail/mpi_comms.hpp
/* * Copyright (c) 2019-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cstdio> #include <memory> #include <unordered_map> #include <unordered_set> #include <utility> #include <mpi.h> #include <nccl.h> #include <raft/comms/comms.hpp> #include <raft/comms/detail/util.hpp> #include <raft/core/error.hpp> #include <raft/core/resources.hpp> #include <raft/util/cudart_utils.hpp> #include <rmm/cuda_stream_view.hpp> #include <rmm/device_scalar.hpp> #define RAFT_MPI_TRY(call) \ do { \ int status = call; \ if (MPI_SUCCESS != status) { \ int mpi_error_string_lenght = 0; \ char mpi_error_string[MPI_MAX_ERROR_STRING]; \ MPI_Error_string(status, mpi_error_string, &mpi_error_string_lenght); \ RAFT_EXPECTS( \ MPI_SUCCESS == status, "ERROR: MPI call='%s'. Reason:%s\n", #call, mpi_error_string); \ } \ } while (0) // FIXME: Remove after consumer rename #ifndef MPI_TRY #define MPI_TRY(call) RAFT_MPI_TRY(call) #endif #define RAFT_MPI_TRY_NO_THROW(call) \ do { \ int status = call; \ if (MPI_SUCCESS != status) { \ int mpi_error_string_lenght = 0; \ char mpi_error_string[MPI_MAX_ERROR_STRING]; \ MPI_Error_string(status, mpi_error_string, &mpi_error_string_lenght); \ printf("MPI call='%s' at file=%s line=%d failed with %s ", \ #call, \ __FILE__, \ __LINE__, \ mpi_error_string); \ } \ } while (0) // FIXME: Remove after consumer rename #ifndef MPI_TRY_NO_THROW #define MPI_TRY_NO_THROW(call) RAFT_MPI_TRY_NO_THROW(call) #endif namespace raft { namespace comms { namespace detail { constexpr MPI_Datatype get_mpi_datatype(const datatype_t datatype) { switch (datatype) { case datatype_t::CHAR: return MPI_CHAR; case datatype_t::UINT8: return MPI_UNSIGNED_CHAR; case datatype_t::INT32: return MPI_INT; case datatype_t::UINT32: return MPI_UNSIGNED; case datatype_t::INT64: return MPI_LONG_LONG; case datatype_t::UINT64: return MPI_UNSIGNED_LONG_LONG; case datatype_t::FLOAT32: return MPI_FLOAT; case datatype_t::FLOAT64: return MPI_DOUBLE; default: // Execution should never reach here. This takes care of compiler warning. return MPI_DOUBLE; } } constexpr MPI_Op get_mpi_op(const op_t op) { switch (op) { case op_t::SUM: return MPI_SUM; case op_t::PROD: return MPI_PROD; case op_t::MIN: return MPI_MIN; case op_t::MAX: return MPI_MAX; default: // Execution should never reach here. This takes care of compiler warning. return MPI_MAX; } } class mpi_comms : public comms_iface { public: mpi_comms(MPI_Comm comm, const bool owns_mpi_comm, rmm::cuda_stream_view stream) : owns_mpi_comm_(owns_mpi_comm), mpi_comm_(comm), size_(0), rank_(1), status_(stream), next_request_id_(0), stream_(stream) { int mpi_is_initialized = 0; RAFT_MPI_TRY(MPI_Initialized(&mpi_is_initialized)); RAFT_EXPECTS(mpi_is_initialized, "ERROR: MPI is not initialized!"); RAFT_MPI_TRY(MPI_Comm_size(mpi_comm_, &size_)); RAFT_MPI_TRY(MPI_Comm_rank(mpi_comm_, &rank_)); // get NCCL unique ID at rank 0 and broadcast it to all others ncclUniqueId id; if (0 == rank_) RAFT_NCCL_TRY(ncclGetUniqueId(&id)); RAFT_MPI_TRY(MPI_Bcast((void*)&id, sizeof(id), MPI_BYTE, 0, mpi_comm_)); // initializing NCCL RAFT_NCCL_TRY(ncclCommInitRank(&nccl_comm_, size_, id, rank_)); initialize(); } void initialize() { status_.set_value_to_zero_async(stream_); buf_ = status_.data(); } virtual ~mpi_comms() { // finalizing NCCL RAFT_NCCL_TRY_NO_THROW(ncclCommDestroy(nccl_comm_)); if (owns_mpi_comm_) { RAFT_MPI_TRY_NO_THROW(MPI_Comm_free(&mpi_comm_)); } } int get_size() const { return size_; } int get_rank() const { return rank_; } std::unique_ptr<comms_iface> comm_split(int color, int key) const { MPI_Comm new_comm; RAFT_MPI_TRY(MPI_Comm_split(mpi_comm_, color, key, &new_comm)); return std::unique_ptr<comms_iface>(new mpi_comms(new_comm, true, stream_)); } void barrier() const { allreduce(buf_, buf_, 1, datatype_t::INT32, op_t::SUM, stream_); ASSERT(sync_stream(stream_) == status_t::SUCCESS, "ERROR: syncStream failed. This can be caused by a failed rank_."); } void isend(const void* buf, size_t size, int dest, int tag, request_t* request) const { MPI_Request mpi_req; request_t req_id; if (free_requests_.empty()) { req_id = next_request_id_++; } else { auto it = free_requests_.begin(); req_id = *it; free_requests_.erase(it); } RAFT_MPI_TRY(MPI_Isend(buf, size, MPI_BYTE, dest, tag, mpi_comm_, &mpi_req)); requests_in_flight_.insert(std::make_pair(req_id, mpi_req)); *request = req_id; } void irecv(void* buf, size_t size, int source, int tag, request_t* request) const { MPI_Request mpi_req; request_t req_id; if (free_requests_.empty()) { req_id = next_request_id_++; } else { auto it = free_requests_.begin(); req_id = *it; free_requests_.erase(it); } RAFT_MPI_TRY(MPI_Irecv(buf, size, MPI_BYTE, source, tag, mpi_comm_, &mpi_req)); requests_in_flight_.insert(std::make_pair(req_id, mpi_req)); *request = req_id; } void waitall(int count, request_t array_of_requests[]) const { std::vector<MPI_Request> requests; requests.reserve(count); for (int i = 0; i < count; ++i) { auto req_it = requests_in_flight_.find(array_of_requests[i]); RAFT_EXPECTS(requests_in_flight_.end() != req_it, "ERROR: waitall on invalid request: %d", array_of_requests[i]); requests.push_back(req_it->second); free_requests_.insert(req_it->first); requests_in_flight_.erase(req_it); } RAFT_MPI_TRY(MPI_Waitall(requests.size(), requests.data(), MPI_STATUSES_IGNORE)); } void allreduce(const void* sendbuff, void* recvbuff, size_t count, datatype_t datatype, op_t op, cudaStream_t stream) const { RAFT_NCCL_TRY(ncclAllReduce( sendbuff, recvbuff, count, get_nccl_datatype(datatype), get_nccl_op(op), nccl_comm_, stream)); } void bcast(void* buff, size_t count, datatype_t datatype, int root, cudaStream_t stream) const { RAFT_NCCL_TRY( ncclBroadcast(buff, buff, count, get_nccl_datatype(datatype), root, nccl_comm_, stream)); } void bcast(const void* sendbuff, void* recvbuff, size_t count, datatype_t datatype, int root, cudaStream_t stream) const { RAFT_NCCL_TRY(ncclBroadcast( sendbuff, recvbuff, count, get_nccl_datatype(datatype), root, nccl_comm_, stream)); } void reduce(const void* sendbuff, void* recvbuff, size_t count, datatype_t datatype, op_t op, int root, cudaStream_t stream) const { RAFT_NCCL_TRY(ncclReduce(sendbuff, recvbuff, count, get_nccl_datatype(datatype), get_nccl_op(op), root, nccl_comm_, stream)); } void allgather(const void* sendbuff, void* recvbuff, size_t sendcount, datatype_t datatype, cudaStream_t stream) const { RAFT_NCCL_TRY(ncclAllGather( sendbuff, recvbuff, sendcount, get_nccl_datatype(datatype), nccl_comm_, stream)); } void allgatherv(const void* sendbuf, void* recvbuf, const size_t* recvcounts, const size_t* displs, datatype_t datatype, cudaStream_t stream) const { RAFT_EXPECTS(size_ <= 2048, "# NCCL operations between ncclGroupStart & ncclGroupEnd shouldn't exceed 2048."); // From: "An Empirical Evaluation of Allgatherv on Multi-GPU Systems" - // https://arxiv.org/pdf/1812.05964.pdf Listing 1 on page 4. RAFT_NCCL_TRY(ncclGroupStart()); for (int root = 0; root < size_; ++root) { RAFT_NCCL_TRY( ncclBroadcast(sendbuf, static_cast<char*>(recvbuf) + displs[root] * get_datatype_size(datatype), recvcounts[root], get_nccl_datatype(datatype), root, nccl_comm_, stream)); } RAFT_NCCL_TRY(ncclGroupEnd()); } void gather(const void* sendbuff, void* recvbuff, size_t sendcount, datatype_t datatype, int root, cudaStream_t stream) const { size_t dtype_size = get_datatype_size(datatype); RAFT_NCCL_TRY(ncclGroupStart()); if (get_rank() == root) { for (int r = 0; r < get_size(); ++r) { RAFT_NCCL_TRY(ncclRecv(static_cast<char*>(recvbuff) + sendcount * r * dtype_size, sendcount, get_nccl_datatype(datatype), r, nccl_comm_, stream)); } } RAFT_NCCL_TRY( ncclSend(sendbuff, sendcount, get_nccl_datatype(datatype), root, nccl_comm_, stream)); RAFT_NCCL_TRY(ncclGroupEnd()); } void gatherv(const void* sendbuff, void* recvbuff, size_t sendcount, const size_t* recvcounts, const size_t* displs, datatype_t datatype, int root, cudaStream_t stream) const { size_t dtype_size = get_datatype_size(datatype); RAFT_NCCL_TRY(ncclGroupStart()); if (get_rank() == root) { for (int r = 0; r < get_size(); ++r) { RAFT_NCCL_TRY(ncclRecv(static_cast<char*>(recvbuff) + displs[r] * dtype_size, recvcounts[r], get_nccl_datatype(datatype), r, nccl_comm_, stream)); } } RAFT_NCCL_TRY( ncclSend(sendbuff, sendcount, get_nccl_datatype(datatype), root, nccl_comm_, stream)); RAFT_NCCL_TRY(ncclGroupEnd()); } void reducescatter(const void* sendbuff, void* recvbuff, size_t recvcount, datatype_t datatype, op_t op, cudaStream_t stream) const { RAFT_NCCL_TRY(ncclReduceScatter(sendbuff, recvbuff, recvcount, get_nccl_datatype(datatype), get_nccl_op(op), nccl_comm_, stream)); } status_t sync_stream(cudaStream_t stream) const { return nccl_sync_stream(nccl_comm_, stream); } // if a thread is sending & receiving at the same time, use device_sendrecv to avoid deadlock void device_send(const void* buf, size_t size, int dest, cudaStream_t stream) const { RAFT_NCCL_TRY(ncclSend(buf, size, ncclUint8, dest, nccl_comm_, stream)); } // if a thread is sending & receiving at the same time, use device_sendrecv to avoid deadlock void device_recv(void* buf, size_t size, int source, cudaStream_t stream) const { RAFT_NCCL_TRY(ncclRecv(buf, size, ncclUint8, source, nccl_comm_, stream)); } void device_sendrecv(const void* sendbuf, size_t sendsize, int dest, void* recvbuf, size_t recvsize, int source, cudaStream_t stream) const { // ncclSend/ncclRecv pair needs to be inside ncclGroupStart/ncclGroupEnd to avoid deadlock RAFT_NCCL_TRY(ncclGroupStart()); RAFT_NCCL_TRY(ncclSend(sendbuf, sendsize, ncclUint8, dest, nccl_comm_, stream)); RAFT_NCCL_TRY(ncclRecv(recvbuf, recvsize, ncclUint8, source, nccl_comm_, stream)); RAFT_NCCL_TRY(ncclGroupEnd()); } void device_multicast_sendrecv(const void* sendbuf, std::vector<size_t> const& sendsizes, std::vector<size_t> const& sendoffsets, std::vector<int> const& dests, void* recvbuf, std::vector<size_t> const& recvsizes, std::vector<size_t> const& recvoffsets, std::vector<int> const& sources, cudaStream_t stream) const { // ncclSend/ncclRecv pair needs to be inside ncclGroupStart/ncclGroupEnd to avoid deadlock RAFT_NCCL_TRY(ncclGroupStart()); for (size_t i = 0; i < sendsizes.size(); ++i) { RAFT_NCCL_TRY(ncclSend(static_cast<const char*>(sendbuf) + sendoffsets[i], sendsizes[i], ncclUint8, dests[i], nccl_comm_, stream)); } for (size_t i = 0; i < recvsizes.size(); ++i) { RAFT_NCCL_TRY(ncclRecv(static_cast<char*>(recvbuf) + recvoffsets[i], recvsizes[i], ncclUint8, sources[i], nccl_comm_, stream)); } RAFT_NCCL_TRY(ncclGroupEnd()); } void group_start() const { RAFT_NCCL_TRY(ncclGroupStart()); } void group_end() const { RAFT_NCCL_TRY(ncclGroupEnd()); } private: bool owns_mpi_comm_; MPI_Comm mpi_comm_; cudaStream_t stream_; rmm::device_scalar<int32_t> status_; int32_t* buf_; ncclComm_t nccl_comm_; int size_; int rank_; mutable request_t next_request_id_; mutable std::unordered_map<request_t, MPI_Request> requests_in_flight_; mutable std::unordered_set<request_t> free_requests_; }; } // end namespace detail }; // end namespace comms }; // end namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/comms
rapidsai_public_repos/raft/cpp/include/raft/comms/detail/test.hpp
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/comms/comms.hpp> #include <raft/core/resource/comms.hpp> #include <raft/core/resource/cuda_stream.hpp> #include <raft/core/resources.hpp> #include <rmm/device_scalar.hpp> #include <rmm/device_uvector.hpp> #include <iostream> #include <numeric> #include <iostream> #include <numeric> namespace raft { namespace comms { namespace detail { /** * @brief A simple sanity check that NCCL is able to perform a collective operation * * @param[in] handle the raft handle to use. This is expected to already have an * initialized comms instance. * @param[in] root the root rank id */ bool test_collective_allreduce(raft::resources const& handle, int root) { comms_t const& communicator = resource::get_comms(handle); int const send = 1; cudaStream_t stream = resource::get_cuda_stream(handle); rmm::device_scalar<int> temp_d(stream); RAFT_CUDA_TRY(cudaMemcpyAsync(temp_d.data(), &send, 1, cudaMemcpyHostToDevice, stream)); communicator.allreduce(temp_d.data(), temp_d.data(), 1, op_t::SUM, stream); int temp_h = 0; RAFT_CUDA_TRY(cudaMemcpyAsync(&temp_h, temp_d.data(), 1, cudaMemcpyDeviceToHost, stream)); resource::sync_stream(handle, stream); communicator.barrier(); std::cout << "Clique size: " << communicator.get_size() << std::endl; std::cout << "final_size: " << temp_h << std::endl; return temp_h == communicator.get_size(); } /** * @brief A simple sanity check that NCCL is able to perform a collective operation * * @param[in] handle the raft handle to use. This is expected to already have an * initialized comms instance. * @param[in] root the root rank id */ bool test_collective_broadcast(raft::resources const& handle, int root) { comms_t const& communicator = resource::get_comms(handle); int const send = root; cudaStream_t stream = resource::get_cuda_stream(handle); rmm::device_scalar<int> temp_d(stream); if (communicator.get_rank() == root) RAFT_CUDA_TRY( cudaMemcpyAsync(temp_d.data(), &send, sizeof(int), cudaMemcpyHostToDevice, stream)); communicator.bcast(temp_d.data(), 1, root, stream); communicator.sync_stream(stream); int temp_h = -1; // Verify more than one byte is being sent RAFT_CUDA_TRY( cudaMemcpyAsync(&temp_h, temp_d.data(), sizeof(int), cudaMemcpyDeviceToHost, stream)); resource::sync_stream(handle, stream); communicator.barrier(); std::cout << "Clique size: " << communicator.get_size() << std::endl; std::cout << "final_size: " << temp_h << std::endl; return temp_h == root; } /** * @brief A simple sanity check that NCCL is able to perform a collective reduce * * @param[in] handle the raft handle to use. This is expected to already have an * initialized comms instance. * @param[in] root the root rank id */ bool test_collective_reduce(raft::resources const& handle, int root) { comms_t const& communicator = resource::get_comms(handle); int const send = root; cudaStream_t stream = resource::get_cuda_stream(handle); rmm::device_scalar<int> temp_d(stream); RAFT_CUDA_TRY(cudaMemcpyAsync(temp_d.data(), &send, sizeof(int), cudaMemcpyHostToDevice, stream)); communicator.reduce(temp_d.data(), temp_d.data(), 1, op_t::SUM, root, stream); communicator.sync_stream(stream); int temp_h = -1; // Verify more than one byte is being sent RAFT_CUDA_TRY( cudaMemcpyAsync(&temp_h, temp_d.data(), sizeof(int), cudaMemcpyDeviceToHost, stream)); resource::sync_stream(handle, stream); communicator.barrier(); std::cout << "Clique size: " << communicator.get_size() << std::endl; std::cout << "final_size: " << temp_h << std::endl; if (communicator.get_rank() == root) return temp_h == root * communicator.get_size(); else return true; } /** * @brief A simple sanity check that NCCL is able to perform a collective allgather * * @param[in] handle the raft handle to use. This is expected to already have an * initialized comms instance. * @param[in] root the root rank id */ bool test_collective_allgather(raft::resources const& handle, int root) { comms_t const& communicator = resource::get_comms(handle); int const send = communicator.get_rank(); cudaStream_t stream = resource::get_cuda_stream(handle); rmm::device_scalar<int> temp_d(stream); rmm::device_uvector<int> recv_d(communicator.get_size(), stream); RAFT_CUDA_TRY(cudaMemcpyAsync(temp_d.data(), &send, sizeof(int), cudaMemcpyHostToDevice, stream)); communicator.allgather(temp_d.data(), recv_d.data(), 1, stream); communicator.sync_stream(stream); int temp_h[communicator.get_size()]; // Verify more than one byte is being sent RAFT_CUDA_TRY(cudaMemcpyAsync( &temp_h, recv_d.data(), sizeof(int) * communicator.get_size(), cudaMemcpyDeviceToHost, stream)); resource::sync_stream(handle, stream); communicator.barrier(); std::cout << "Clique size: " << communicator.get_size() << std::endl; std::cout << "final_size: " << temp_h << std::endl; for (int i = 0; i < communicator.get_size(); i++) { if (temp_h[i] != i) return false; } return true; } /** * @brief A simple sanity check that NCCL is able to perform a collective gather * * @param[in] handle the raft handle to use. This is expected to already have an * initialized comms instance. * @param[in] root the root rank id */ bool test_collective_gather(raft::resources const& handle, int root) { comms_t const& communicator = resource::get_comms(handle); int const send = communicator.get_rank(); cudaStream_t stream = resource::get_cuda_stream(handle); rmm::device_scalar<int> temp_d(stream); rmm::device_uvector<int> recv_d(communicator.get_rank() == root ? communicator.get_size() : 0, stream); RAFT_CUDA_TRY(cudaMemcpyAsync(temp_d.data(), &send, sizeof(int), cudaMemcpyHostToDevice, stream)); communicator.gather(temp_d.data(), recv_d.data(), 1, root, stream); communicator.sync_stream(stream); if (communicator.get_rank() == root) { std::vector<int> temp_h(communicator.get_size(), 0); RAFT_CUDA_TRY(cudaMemcpyAsync( temp_h.data(), recv_d.data(), sizeof(int) * temp_h.size(), cudaMemcpyDeviceToHost, stream)); resource::sync_stream(handle, stream); for (int i = 0; i < communicator.get_size(); i++) { if (temp_h[i] != i) return false; } } return true; } /** * @brief A simple sanity check that NCCL is able to perform a collective gatherv * * @param[in] handle the raft handle to use. This is expected to already have an * initialized comms instance. * @param[in] root the root rank id */ bool test_collective_gatherv(raft::resources const& handle, int root) { comms_t const& communicator = resource::get_comms(handle); std::vector<size_t> sendcounts(communicator.get_size()); std::iota(sendcounts.begin(), sendcounts.end(), size_t{1}); std::vector<size_t> displacements(communicator.get_size() + 1, 0); std::partial_sum(sendcounts.begin(), sendcounts.end(), displacements.begin() + 1); std::vector<int> sends( displacements[communicator.get_rank() + 1] - displacements[communicator.get_rank()], communicator.get_rank()); cudaStream_t stream = resource::get_cuda_stream(handle); rmm::device_uvector<int> temp_d(sends.size(), stream); rmm::device_uvector<int> recv_d(communicator.get_rank() == root ? displacements.back() : 0, stream); RAFT_CUDA_TRY(cudaMemcpyAsync( temp_d.data(), sends.data(), sends.size() * sizeof(int), cudaMemcpyHostToDevice, stream)); communicator.gatherv( temp_d.data(), recv_d.data(), temp_d.size(), communicator.get_rank() == root ? sendcounts.data() : static_cast<size_t*>(nullptr), communicator.get_rank() == root ? displacements.data() : static_cast<size_t*>(nullptr), root, stream); communicator.sync_stream(stream); if (communicator.get_rank() == root) { std::vector<int> temp_h(displacements.back(), 0); RAFT_CUDA_TRY(cudaMemcpyAsync(temp_h.data(), recv_d.data(), sizeof(int) * displacements.back(), cudaMemcpyDeviceToHost, stream)); resource::sync_stream(handle, stream); for (int i = 0; i < communicator.get_size(); i++) { if (std::count_if(temp_h.begin() + displacements[i], temp_h.begin() + displacements[i + 1], [i](auto val) { return val != i; }) != 0) { return false; } } } return true; } /** * @brief A simple sanity check that NCCL is able to perform a collective reducescatter * * @param[in] handle the raft handle to use. This is expected to already have an * initialized comms instance. * @param[in] root the root rank id */ bool test_collective_reducescatter(raft::resources const& handle, int root) { comms_t const& communicator = resource::get_comms(handle); std::vector<int> sends(communicator.get_size(), 1); cudaStream_t stream = resource::get_cuda_stream(handle); rmm::device_uvector<int> temp_d(sends.size(), stream); rmm::device_scalar<int> recv_d(stream); RAFT_CUDA_TRY(cudaMemcpyAsync( temp_d.data(), sends.data(), sends.size() * sizeof(int), cudaMemcpyHostToDevice, stream)); communicator.reducescatter(temp_d.data(), recv_d.data(), 1, op_t::SUM, stream); communicator.sync_stream(stream); int temp_h = -1; // Verify more than one byte is being sent RAFT_CUDA_TRY( cudaMemcpyAsync(&temp_h, recv_d.data(), sizeof(int), cudaMemcpyDeviceToHost, stream)); resource::sync_stream(handle, stream); communicator.barrier(); std::cout << "Clique size: " << communicator.get_size() << std::endl; std::cout << "final_size: " << temp_h << std::endl; return temp_h == communicator.get_size(); } /** * A simple sanity check that UCX is able to send messages between all ranks * * @param[in] h the raft handle to use. This is expected to already have an * initialized comms instance. * @param[in] numTrials number of iterations of all-to-all messaging to perform */ bool test_pointToPoint_simple_send_recv(raft::resources const& h, int numTrials) { comms_t const& communicator = resource::get_comms(h); int const rank = communicator.get_rank(); bool ret = true; for (int i = 0; i < numTrials; i++) { std::vector<int> received_data((communicator.get_size() - 1), -1); std::vector<request_t> requests; requests.resize(2 * (communicator.get_size() - 1)); int request_idx = 0; // post receives for (int r = 0; r < communicator.get_size(); ++r) { if (r != rank) { communicator.irecv( received_data.data() + request_idx, 1, r, 0, requests.data() + request_idx); ++request_idx; } } for (int r = 0; r < communicator.get_size(); ++r) { if (r != rank) { communicator.isend(&rank, 1, r, 0, requests.data() + request_idx); ++request_idx; } } communicator.waitall(requests.size(), requests.data()); communicator.barrier(); if (communicator.get_rank() == 0) { std::cout << "=========================" << std::endl; std::cout << "Trial " << i << std::endl; } for (int printrank = 0; printrank < communicator.get_size(); ++printrank) { if (communicator.get_rank() == printrank) { std::cout << "Rank " << communicator.get_rank() << " received: ["; for (size_t i = 0; i < received_data.size(); i++) { auto rec = received_data[i]; std::cout << rec; if (rec == -1) ret = false; communicator.barrier(); if (i < received_data.size() - 1) std::cout << ", "; } std::cout << "]" << std::endl; } communicator.barrier(); } if (communicator.get_rank() == 0) std::cout << "=========================" << std::endl; } return ret; } /** * A simple sanity check that device is able to send OR receive. * * @param h the raft handle to use. This is expected to already have an * initialized comms instance. * @param numTrials number of iterations of send or receive messaging to perform */ bool test_pointToPoint_device_send_or_recv(raft::resources const& h, int numTrials) { comms_t const& communicator = resource::get_comms(h); int const rank = communicator.get_rank(); cudaStream_t stream = resource::get_cuda_stream(h); bool ret = true; for (int i = 0; i < numTrials; i++) { if (communicator.get_rank() == 0) { std::cout << "=========================" << std::endl; std::cout << "Trial " << i << std::endl; } bool sender = (rank % 2) == 0 ? true : false; rmm::device_scalar<int> received_data(-1, stream); rmm::device_scalar<int> sent_data(rank, stream); if (sender) { if (rank + 1 < communicator.get_size()) { communicator.device_send(sent_data.data(), 1, rank + 1, stream); } } else { communicator.device_recv(received_data.data(), 1, rank - 1, stream); } communicator.sync_stream(stream); if (!sender && received_data.value(stream) != rank - 1) { ret = false; } if (communicator.get_rank() == 0) { std::cout << "=========================" << std::endl; } } return ret; } /** * A simple sanity check that device is able to send and receive at the same time. * * @param h the raft handle to use. This is expected to already have an * initialized comms instance. * @param numTrials number of iterations of send or receive messaging to perform */ bool test_pointToPoint_device_sendrecv(raft::resources const& h, int numTrials) { comms_t const& communicator = resource::get_comms(h); int const rank = communicator.get_rank(); cudaStream_t stream = resource::get_cuda_stream(h); bool ret = true; for (int i = 0; i < numTrials; i++) { if (communicator.get_rank() == 0) { std::cout << "=========================" << std::endl; std::cout << "Trial " << i << std::endl; } rmm::device_scalar<int> received_data(-1, stream); rmm::device_scalar<int> sent_data(rank, stream); if (rank % 2 == 0) { if (rank + 1 < communicator.get_size()) { communicator.device_sendrecv( sent_data.data(), 1, rank + 1, received_data.data(), 1, rank + 1, stream); } } else { communicator.device_sendrecv( sent_data.data(), 1, rank - 1, received_data.data(), 1, rank - 1, stream); } communicator.sync_stream(stream); if (((rank % 2 == 0) && (received_data.value(stream) != rank + 1)) || ((rank % 2 == 1) && (received_data.value(stream) != rank - 1))) { ret = false; } if (communicator.get_rank() == 0) { std::cout << "=========================" << std::endl; } } return ret; } /** * A simple sanity check that device is able to perform multiple concurrent sends and receives. * * @param h the raft handle to use. This is expected to already have an * initialized comms instance. * @param numTrials number of iterations of send or receive messaging to perform */ bool test_pointToPoint_device_multicast_sendrecv(raft::resources const& h, int numTrials) { comms_t const& communicator = resource::get_comms(h); int const rank = communicator.get_rank(); cudaStream_t stream = resource::get_cuda_stream(h); bool ret = true; for (int i = 0; i < numTrials; i++) { if (communicator.get_rank() == 0) { std::cout << "=========================" << std::endl; std::cout << "Trial " << i << std::endl; } rmm::device_uvector<int> received_data(communicator.get_size(), stream); rmm::device_scalar<int> sent_data(rank, stream); std::vector<size_t> sendsizes(communicator.get_size(), 1); std::vector<size_t> sendoffsets(communicator.get_size(), 0); std::vector<int> dests(communicator.get_size()); std::iota(dests.begin(), dests.end(), int{0}); std::vector<size_t> recvsizes(communicator.get_size(), 1); std::vector<size_t> recvoffsets(communicator.get_size()); std::iota(recvoffsets.begin(), recvoffsets.end(), size_t{0}); std::vector<int> srcs(communicator.get_size()); std::iota(srcs.begin(), srcs.end(), int{0}); communicator.device_multicast_sendrecv(sent_data.data(), sendsizes, sendoffsets, dests, received_data.data(), recvsizes, recvoffsets, srcs, stream); communicator.sync_stream(stream); std::vector<int> h_received_data(communicator.get_size()); raft::update_host(h_received_data.data(), received_data.data(), received_data.size(), stream); resource::sync_stream(h, stream); for (int i = 0; i < communicator.get_size(); ++i) { if (h_received_data[i] != i) { ret = false; } } if (communicator.get_rank() == 0) { std::cout << "=========================" << std::endl; } } return ret; } /** * A simple test that the comms can be split into 2 separate subcommunicators * * @param h the raft handle to use. This is expected to already have an * initialized comms instance. * @param n_colors number of different colors to test */ bool test_commsplit(raft::resources const& h, int n_colors) { comms_t const& communicator = resource::get_comms(h); int const rank = communicator.get_rank(); int const size = communicator.get_size(); if (n_colors > size) n_colors = size; // first we need to assign to a color, then assign the rank within the color int color = rank % n_colors; int key = rank / n_colors; auto stream_pool = std::make_shared<rmm::cuda_stream_pool>(1); handle_t new_handle(rmm::cuda_stream_default, stream_pool); auto shared_comm = std::make_shared<comms_t>(communicator.comm_split(color, key)); new_handle.set_comms(shared_comm); return test_collective_allreduce(new_handle, 0); } } // namespace detail } // namespace comms }; // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/comms
rapidsai_public_repos/raft/cpp/include/raft/comms/detail/ucp_helper.hpp
/* * Copyright (c) 2021-2022, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/util/cudart_utils.hpp> #include <stdio.h> #include <ucp/api/ucp.h> #include <ucp/api/ucp_def.h> namespace raft { namespace comms { namespace detail { /** * Standard UCX request object that will be passed * around asynchronously. This object is really * opaque and the comms layer only cares that it * has been completed. Because raft comms do not * initialize the ucx application context, it doesn't * own this object and thus it's important not to * modify this struct. */ struct ucx_context { int completed; }; /** * Wraps the `ucx_context` request and adds a few * other fields for trace logging and cleanup. */ class ucp_request { public: struct ucx_context* req; bool needs_release = true; int other_rank = -1; bool is_send_request = false; }; // by default, match the whole tag static const ucp_tag_t default_tag_mask = (ucp_tag_t)-1; /** * @brief Asynchronous send callback sets request to completed */ static void send_callback(void* request, ucs_status_t status) { struct ucx_context* context = (struct ucx_context*)request; context->completed = 1; } /** * @brief Asynchronous recv callback sets request to completed */ static void recv_callback(void* request, ucs_status_t status, ucp_tag_recv_info_t* info) { struct ucx_context* context = (struct ucx_context*)request; context->completed = 1; } /** * Helper class for interacting with ucp. */ class comms_ucp_handler { private: ucp_tag_t build_message_tag(int rank, int tag) const { // keeping the rank in the lower bits enables debugging. return ((uint32_t)tag << 31) | (uint32_t)rank; } public: /** * @brief Frees any memory underlying the given ucp request object */ void free_ucp_request(ucp_request* request) const { if (request->needs_release) { request->req->completed = 0; ucp_request_free(request->req); } free(request); } /** * @brief Asynchronously send data to the given endpoint using the given tag */ void ucp_isend(ucp_request* req, ucp_ep_h ep_ptr, const void* buf, size_t size, int tag, ucp_tag_t tag_mask, int rank) const { ucp_tag_t ucp_tag = build_message_tag(rank, tag); ucs_status_ptr_t send_result = ucp_tag_send_nb(ep_ptr, buf, size, ucp_dt_make_contig(1), ucp_tag, send_callback); struct ucx_context* ucp_req = (struct ucx_context*)send_result; if (UCS_PTR_IS_ERR(send_result)) { ASSERT(!UCS_PTR_IS_ERR(send_result), "unable to send UCX data message (%d)\n", UCS_PTR_STATUS(send_result)); /** * If the request didn't fail, but it's not OK, it is in flight. * Expect the handler to be invoked */ } else if (UCS_PTR_STATUS(send_result) != UCS_OK) { /** * If the request is OK, it's already been completed and we don't need to wait on it. * The request will be a nullptr, however, so we need to create a new request * and set it to completed to make the "waitall()" function work properly. */ req->needs_release = true; } else { req->needs_release = false; } req->other_rank = rank; req->is_send_request = true; req->req = ucp_req; } /** * @brief Asynchronously receive data from given endpoint with the given tag. */ void ucp_irecv(ucp_request* req, ucp_worker_h worker, ucp_ep_h ep_ptr, void* buf, size_t size, int tag, ucp_tag_t tag_mask, int sender_rank) const { ucp_tag_t ucp_tag = build_message_tag(sender_rank, tag); ucs_status_ptr_t recv_result = ucp_tag_recv_nb(worker, buf, size, ucp_dt_make_contig(1), ucp_tag, tag_mask, recv_callback); struct ucx_context* ucp_req = (struct ucx_context*)recv_result; req->req = ucp_req; req->needs_release = true; req->is_send_request = false; req->other_rank = sender_rank; ASSERT(!UCS_PTR_IS_ERR(recv_result), "unable to receive UCX data message (%d)\n", UCS_PTR_STATUS(recv_result)); } }; } // end namespace detail } // end namespace comms } // end namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/spatial
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/ivf_pq_types.hpp
/* * Copyright (c) 2020-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ /** * This file is deprecated and will be removed in release 22.06. * Please use the cuh version instead. */ /** * DISCLAIMER: this file is deprecated: use epsilon_neighborhood.cuh instead */ #pragma once #pragma message(__FILE__ \ " is deprecated and will be removed in a future release." \ " Please use the raft::neighbors version instead.") #include <raft/neighbors/ivf_pq_types.hpp> namespace raft::spatial::knn::ivf_pq { using raft::neighbors::ivf_pq::codebook_gen; using raft::neighbors::ivf_pq::index; using raft::neighbors::ivf_pq::index_params; using raft::neighbors::ivf_pq::search_params; } // namespace raft::spatial::knn::ivf_pq
0
rapidsai_public_repos/raft/cpp/include/raft/spatial
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/ann_common.h
/* * Copyright (c) 2020-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma message( \ __FILE__ \ " is deprecated and will be removed in a future release." \ " Please use the other approximate KNN implementations defined in spatial/knn/*.") #pragma once #include "detail/processing.hpp" #include "ivf_flat_types.hpp" #include <raft/neighbors/ivf_pq_types.hpp> #include <raft/distance/distance_types.hpp> namespace raft { namespace spatial { namespace knn { struct knnIndex { raft::distance::DistanceType metric; float metricArg; int nprobe; std::unique_ptr<MetricProcessor<float>> metric_processor; std::unique_ptr<const ivf_flat::index<float, int64_t>> ivf_flat_float_; std::unique_ptr<const ivf_flat::index<uint8_t, int64_t>> ivf_flat_uint8_t_; std::unique_ptr<const ivf_flat::index<int8_t, int64_t>> ivf_flat_int8_t_; std::unique_ptr<const raft::neighbors::ivf_pq::index<int64_t>> ivf_pq; int device; template <typename T, typename IdxT> auto ivf_flat() -> std::unique_ptr<const ivf_flat::index<T, IdxT>>&; }; template <> inline auto knnIndex::ivf_flat<float, int64_t>() -> std::unique_ptr<const ivf_flat::index<float, int64_t>>& { return ivf_flat_float_; } template <> inline auto knnIndex::ivf_flat<uint8_t, int64_t>() -> std::unique_ptr<const ivf_flat::index<uint8_t, int64_t>>& { return ivf_flat_uint8_t_; } template <> inline auto knnIndex::ivf_flat<int8_t, int64_t>() -> std::unique_ptr<const ivf_flat::index<int8_t, int64_t>>& { return ivf_flat_int8_t_; } struct knnIndexParam { virtual ~knnIndexParam() {} }; struct IVFParam : knnIndexParam { int nlist; int nprobe; }; struct IVFFlatParam : IVFParam {}; struct IVFPQParam : IVFParam { int M; int n_bits; bool usePrecomputedTables; }; inline auto from_legacy_index_params(const IVFFlatParam& legacy, raft::distance::DistanceType metric, float metric_arg) { ivf_flat::index_params params; params.metric = metric; params.metric_arg = metric_arg; params.n_lists = legacy.nlist; return params; } }; // namespace knn }; // namespace spatial }; // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/spatial
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/ivf_flat.cuh
/* * Copyright (c) 2020-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ /** * This file is deprecated and will be removed in release 22.06. * Please use the cuh version instead. */ /** * DISCLAIMER: this file is deprecated: use epsilon_neighborhood.cuh instead */ #pragma once #pragma message(__FILE__ \ " is deprecated and will be removed in a future release." \ " Please use the raft::neighbors version instead.") #include <raft/neighbors/ivf_flat.cuh> namespace raft::spatial::knn::ivf_flat { using raft::neighbors::ivf_flat::build; using raft::neighbors::ivf_flat::extend; using raft::neighbors::ivf_flat::search; }; // namespace raft::spatial::knn::ivf_flat
0
rapidsai_public_repos/raft/cpp/include/raft/spatial
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/epsilon_neighborhood.cuh
/* * Copyright (c) 2020-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ /** * This file is deprecated and will be removed in release 22.06. * Please use the cuh version instead. */ /** * DISCLAIMER: this file is deprecated: use epsilon_neighborhood.cuh instead */ #pragma once #pragma message(__FILE__ \ " is deprecated and will be removed in a future release." \ " Please use the raft::neighbors version instead.") #include <raft/neighbors/epsilon_neighborhood.cuh> namespace raft::spatial::knn { using raft::neighbors::epsilon_neighborhood::eps_neighbors_l2sq; using raft::neighbors::epsilon_neighborhood::epsUnexpL2SqNeighborhood; } // namespace raft::spatial::knn
0
rapidsai_public_repos/raft/cpp/include/raft/spatial
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/knn.cuh
/* * Copyright (c) 2020-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/device_mdspan.hpp> #include <raft/core/nvtx.hpp> #include <raft/matrix/detail/select_radix.cuh> #include <raft/matrix/detail/select_warpsort.cuh> #include <raft/neighbors/detail/knn_brute_force.cuh> #include <raft/neighbors/detail/selection_faiss.cuh> namespace raft::spatial::knn { /** * Performs a k-select across row partitioned index/distance * matrices formatted like the following: * row1: k0, k1, k2 * row2: k0, k1, k2 * row3: k0, k1, k2 * row1: k0, k1, k2 * row2: k0, k1, k2 * row3: k0, k1, k2 * * etc... * * @tparam idx_t * @tparam value_t * @param in_keys * @param in_values * @param out_keys * @param out_values * @param n_samples * @param n_parts * @param k * @param stream * @param translations */ template <typename idx_t = int64_t, typename value_t = float> inline void knn_merge_parts(const value_t* in_keys, const idx_t* in_values, value_t* out_keys, idx_t* out_values, size_t n_samples, int n_parts, int k, cudaStream_t stream, idx_t* translations) { raft::neighbors::detail::knn_merge_parts( in_keys, in_values, out_keys, out_values, n_samples, n_parts, k, stream, translations); } /** Choose an implementation for the select-top-k, */ enum class SelectKAlgo { /** Adapted from the faiss project. Result: sorted (not stable). */ FAISS, /** Incomplete series of radix sort passes, comparing 8 bits per pass. Result: unsorted. */ RADIX_8_BITS, /** Incomplete series of radix sort passes, comparing 11 bits per pass. Result: unsorted. */ RADIX_11_BITS, /** Filtering with a bitonic-sort-based priority queue. Result: sorted (not stable). */ WARP_SORT }; /** * Select k smallest or largest key/values from each row in the input data. * * If you think of the input data `in_keys` as a row-major matrix with input_len columns and * n_inputs rows, then this function selects k smallest/largest values in each row and fills * in the row-major matrix `out_keys` of size (n_inputs, k). * * Note, depending on the selected algorithm, the values within rows of `out_keys` are not * necessarily sorted. See the `SelectKAlgo` enumeration for more details. * * Note: This call is deprecated, please use `raft/matrix/select_k.cuh` * * @tparam idx_t * the payload type (what is being selected together with the keys). * @tparam value_t * the type of the keys (what is being compared). * * @param[in] in_keys * contiguous device array of inputs of size (input_len * n_inputs); * these are compared and selected. * @param[in] in_values * contiguous device array of inputs of size (input_len * n_inputs); * typically, these are indices of the corresponding in_keys. * You can pass `NULL` as an argument here; this would imply `in_values` is a homogeneous array * of indices from `0` to `input_len - 1` for every input and reduce the usage of memory * bandwidth. * @param[in] n_inputs * number of input rows, i.e. the batch size. * @param[in] input_len * length of a single input array (row); also sometimes referred as n_cols. * Invariant: input_len >= k. * @param[out] out_keys * contiguous device array of outputs of size (k * n_inputs); * the k smallest/largest values from each row of the `in_keys`. * @param[out] out_values * contiguous device array of outputs of size (k * n_inputs); * the payload selected together with `out_keys`. * @param[in] select_min * whether to select k smallest (true) or largest (false) keys. * @param[in] k * the number of outputs to select in each input row. * @param[in] stream * @param[in] algo * the implementation of the algorithm */ template <typename idx_t = int, typename value_t = float> [[deprecated("Use function `select_k` from `raft/matrix/select_k.cuh`")]] inline void select_k( const value_t* in_keys, const idx_t* in_values, size_t n_inputs, size_t input_len, value_t* out_keys, idx_t* out_values, bool select_min, int k, cudaStream_t stream, SelectKAlgo algo = SelectKAlgo::FAISS) { common::nvtx::range<common::nvtx::domain::raft> fun_scope("select-%s-%d (%zu, %zu) algo-%d", select_min ? "min" : "max", k, n_inputs, input_len, int(algo)); ASSERT(size_t(input_len) >= size_t(k), "Size of the input (input_len = %zu) must be not smaller than the selection (k = %zu).", size_t(input_len), size_t(k)); switch (algo) { case SelectKAlgo::FAISS: neighbors::detail::select_k( in_keys, in_values, n_inputs, input_len, out_keys, out_values, select_min, k, stream); break; case SelectKAlgo::RADIX_8_BITS: matrix::detail::select::radix::select_k<value_t, idx_t, 8, 512>( in_keys, in_values, n_inputs, input_len, k, out_keys, out_values, select_min, true, stream); break; case SelectKAlgo::RADIX_11_BITS: matrix::detail::select::radix::select_k<value_t, idx_t, 11, 512>( in_keys, in_values, n_inputs, input_len, k, out_keys, out_values, select_min, true, stream); break; case SelectKAlgo::WARP_SORT: matrix::detail::select::warpsort::select_k<value_t, idx_t>( in_keys, in_values, n_inputs, input_len, k, out_keys, out_values, select_min, stream); break; default: ASSERT(false, "Unknown algorithm (id = %d)", int(algo)); } } /** * @brief Flat C++ API function to perform a brute force knn on * a series of input arrays and combine the results into a single * output array for indexes and distances. * * @param[in] handle the cuml handle to use * @param[in] input vector of pointers to the input arrays * @param[in] sizes vector of sizes of input arrays * @param[in] D the dimensionality of the arrays * @param[in] search_items array of items to search of dimensionality D * @param[in] n number of rows in search_items * @param[out] res_I the resulting index array of size n * k * @param[out] res_D the resulting distance array of size n * k * @param[in] k the number of nearest neighbors to return * @param[in] rowMajorIndex are the index arrays in row-major order? * @param[in] rowMajorQuery are the query arrays in row-major order? * @param[in] metric distance metric to use. Euclidean (L2) is used by * default * @param[in] metric_arg the value of `p` for Minkowski (l-p) distances. This * is ignored if the metric_type is not Minkowski. * @param[in] translations starting offsets for partitions. should be the same size * as input vector. */ template <typename idx_t = std::int64_t, typename value_t = float, typename value_int = int> void brute_force_knn(raft::resources const& handle, std::vector<value_t*>& input, std::vector<value_int>& sizes, value_int D, value_t* search_items, value_int n, idx_t* res_I, value_t* res_D, value_int k, bool rowMajorIndex = true, bool rowMajorQuery = true, std::vector<idx_t>* translations = nullptr, distance::DistanceType metric = distance::DistanceType::L2Unexpanded, float metric_arg = 2.0f) { ASSERT(input.size() == sizes.size(), "input and sizes vectors must be the same size"); raft::neighbors::detail::brute_force_knn_impl(handle, input, sizes, D, search_items, n, res_I, res_D, k, rowMajorIndex, rowMajorQuery, translations, metric, metric_arg); } } // namespace raft::spatial::knn
0
rapidsai_public_repos/raft/cpp/include/raft/spatial
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/ann_types.hpp
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/distance/distance_types.hpp> namespace raft::spatial::knn { /** The base for approximate KNN index structures. */ struct index {}; /** The base for KNN index parameters. */ struct index_params { /** Distance type. */ raft::distance::DistanceType metric = distance::DistanceType::L2Expanded; /** The argument used by some distance metrics. */ float metric_arg = 2.0f; /** * Whether to add the dataset content to the index, i.e.: * * - `true` means the index is filled with the dataset vectors and ready to search after calling * `build`. * - `false` means `build` only trains the underlying model (e.g. quantizer or clustering), but * the index is left empty; you'd need to call `extend` on the index afterwards to populate it. */ bool add_data_on_build = true; }; struct search_params {}; }; // namespace raft::spatial::knn
0
rapidsai_public_repos/raft/cpp/include/raft/spatial
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/ivf_pq.cuh
/* * Copyright (c) 2020-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ /** * This file is deprecated and will be removed in release 22.06. * Please use the cuh version instead. */ /** * DISCLAIMER: this file is deprecated: use epsilon_neighborhood.cuh instead */ #pragma once #pragma message(__FILE__ \ " is deprecated and will be removed in a future release." \ " Please use the raft::neighbors version instead.") #include <raft/neighbors/ivf_pq.cuh> namespace raft::spatial::knn::ivf_pq { using raft::neighbors::ivf_pq::build; using raft::neighbors::ivf_pq::extend; using raft::neighbors::ivf_pq::search; } // namespace raft::spatial::knn::ivf_pq
0
rapidsai_public_repos/raft/cpp/include/raft/spatial
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/ball_cover.cuh
/* * Copyright (c) 2020-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ /** * This file is deprecated and will be removed in release 22.06. * Please use the cuh version instead. */ /** * DISCLAIMER: this file is deprecated: use epsilon_neighborhood.cuh instead */ #pragma once #pragma message(__FILE__ \ " is deprecated and will be removed in a future release." \ " Please use the raft::neighbors version instead.") #include <raft/neighbors/ball_cover.cuh> #include <raft/spatial/knn/ball_cover_types.hpp> namespace raft::spatial::knn { template <typename idx_t, typename value_t, typename int_t, typename matrix_idx_t> void rbc_build_index(raft::resources const& handle, BallCoverIndex<idx_t, value_t, int_t, matrix_idx_t>& index) { raft::neighbors::ball_cover::build_index(handle, index); } template <typename idx_t, typename value_t, typename int_t, typename matrix_idx_t> void rbc_all_knn_query(raft::resources const& handle, BallCoverIndex<idx_t, value_t, int_t, matrix_idx_t>& index, int_t k, idx_t* inds, value_t* dists, bool perform_post_filtering = true, float weight = 1.0) { raft::neighbors::ball_cover::all_knn_query( handle, index, k, inds, dists, perform_post_filtering, weight); } template <typename idx_t, typename value_t, typename int_t> void rbc_knn_query(raft::resources const& handle, const BallCoverIndex<idx_t, value_t, int_t>& index, int_t k, const value_t* query, int_t n_query_pts, idx_t* inds, value_t* dists, bool perform_post_filtering = true, float weight = 1.0) { raft::neighbors::ball_cover::knn_query( handle, index, k, query, n_query_pts, inds, dists, perform_post_filtering, weight); } } // namespace raft::spatial::knn
0
rapidsai_public_repos/raft/cpp/include/raft/spatial
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/specializations.cuh
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #pragma message( \ __FILE__ \ " is deprecated and will be removed." \ " Including specializations is not necessary any more." \ " For more information, see: https://docs.rapids.ai/api/raft/nightly/using_libraft.html")
0
rapidsai_public_repos/raft/cpp/include/raft/spatial
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/ann.cuh
/* * Copyright (c) 2020-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include "ann_common.h" #include "detail/ann_quantized.cuh" #include <raft/core/nvtx.hpp> namespace raft::spatial::knn { /** * @brief Flat C++ API function to build an approximate nearest neighbors index * from an index array and a set of parameters. * * @param[in] handle RAFT handle * @param[out] index index to be built * @param[in] params parametrization of the index to be built * @param[in] metric distance metric to use. Euclidean (L2) is used by default * @param[in] metricArg metric argument * @param[in] index_array the index array to build the index with * @param[in] n number of rows in the index array * @param[in] D the dimensionality of the index array */ template <typename T = float, typename value_idx = int> [[deprecated("Consider using new-style raft::spatial::knn::*::build functions")]] inline void approx_knn_build_index(raft::resources& handle, raft::spatial::knn::knnIndex* index, knnIndexParam* params, raft::distance::DistanceType metric, float metricArg, T* index_array, value_idx n, value_idx D) { common::nvtx::range<common::nvtx::domain::raft> fun_scope( "legacy approx_knn_build_index(n_rows = %u, dim = %u)", n, D); detail::approx_knn_build_index(handle, index, params, metric, metricArg, index_array, n, D); } /** * @brief Flat C++ API function to perform an approximate nearest neighbors * search from previously built index and a query array * * @param[in] handle RAFT handle * @param[out] distances distances of the nearest neighbors toward * their query point * @param[out] indices indices of the nearest neighbors * @param[in] index index to perform a search with * @param[in] k the number of nearest neighbors to search for * @param[in] query_array the query to perform a search with * @param[in] n number of rows in the query array */ template <typename T = float, typename value_idx = int> [[deprecated("Consider using new-style raft::spatial::knn::*::search functions")]] inline void approx_knn_search(raft::resources& handle, float* distances, int64_t* indices, raft::spatial::knn::knnIndex* index, value_idx k, T* query_array, value_idx n) { common::nvtx::range<common::nvtx::domain::raft> fun_scope( "legacy approx_knn_search(k = %u, n_queries = %u)", k, n); detail::approx_knn_search(handle, distances, indices, index, k, query_array, n); } } // namespace raft::spatial::knn
0
rapidsai_public_repos/raft/cpp/include/raft/spatial
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/ball_cover_types.hpp
/* * Copyright (c) 2020-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ /** * This file is deprecated and will be removed in release 22.06. * Please use the cuh version instead. */ /** * DISCLAIMER: this file is deprecated: use epsilon_neighborhood.cuh instead */ #pragma once #pragma message(__FILE__ \ " is deprecated and will be removed in a future release." \ " Please use the raft::neighbors version instead.") #include <raft/neighbors/ball_cover_types.hpp> namespace raft::spatial::knn { using raft::neighbors::ball_cover::BallCoverIndex; } // namespace raft::spatial::knn
0
rapidsai_public_repos/raft/cpp/include/raft/spatial
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/ivf_flat_types.hpp
/* * Copyright (c) 2020-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ /** * This file is deprecated and will be removed in release 22.06. * Please use the cuh version instead. */ /** * DISCLAIMER: this file is deprecated: use epsilon_neighborhood.cuh instead */ #pragma once #pragma message(__FILE__ \ " is deprecated and will be removed in a future release." \ " Please use the raft::neighbors version instead.") #include <raft/neighbors/ivf_flat_types.hpp> namespace raft::spatial::knn::ivf_flat { using raft::neighbors::ivf_flat::index; using raft::neighbors::ivf_flat::index_params; using raft::neighbors::ivf_flat::kIndexGroupSize; using raft::neighbors::ivf_flat::search_params; }; // namespace raft::spatial::knn::ivf_flat
0
rapidsai_public_repos/raft/cpp/include/raft/spatial
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/common.hpp
/* * Copyright (c) 2021-2022, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ /** * This file is deprecated and will be removed in a future release. * Please use the ann_types.hpp version instead. */ #pragma once #include <raft/spatial/knn/ann_types.hpp>
0
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail/fused_l2_knn-ext.cuh
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cstddef> // size_t #include <cstdint> // uint32_t #include <raft/distance/distance_types.hpp> // DistanceType #include <raft/util/raft_explicit.hpp> // RAFT_EXPLICIT #if defined(RAFT_EXPLICIT_INSTANTIATE_ONLY) namespace raft::spatial::knn::detail { template <typename value_idx, typename value_t, bool usePrevTopKs = false> void fusedL2Knn(size_t D, value_idx* out_inds, value_t* out_dists, const value_t* index, const value_t* query, size_t n_index_rows, size_t n_query_rows, int k, bool rowMajorIndex, bool rowMajorQuery, cudaStream_t stream, raft::distance::DistanceType metric, const value_t* index_norms = NULL, const value_t* query_norms = NULL) RAFT_EXPLICIT; } // namespace raft::spatial::knn::detail #endif // RAFT_EXPLICIT_INSTANTIATE_ONLY #define instantiate_raft_spatial_knn_detail_fusedL2Knn(Mvalue_idx, Mvalue_t, MusePrevTopKs) \ extern template void \ raft::spatial::knn::detail::fusedL2Knn<Mvalue_idx, Mvalue_t, MusePrevTopKs>( \ size_t D, \ Mvalue_idx * out_inds, \ Mvalue_t * out_dists, \ const Mvalue_t* index, \ const Mvalue_t* query, \ size_t n_index_rows, \ size_t n_query_rows, \ int k, \ bool rowMajorIndex, \ bool rowMajorQuery, \ cudaStream_t stream, \ raft::distance::DistanceType metric, \ const Mvalue_t* index_norms, \ const Mvalue_t* query_norms); instantiate_raft_spatial_knn_detail_fusedL2Knn(int32_t, float, true); instantiate_raft_spatial_knn_detail_fusedL2Knn(int32_t, float, false); instantiate_raft_spatial_knn_detail_fusedL2Knn(int64_t, float, true); instantiate_raft_spatial_knn_detail_fusedL2Knn(int64_t, float, false); // These are used by brute_force_knn: instantiate_raft_spatial_knn_detail_fusedL2Knn(uint32_t, float, true); instantiate_raft_spatial_knn_detail_fusedL2Knn(uint32_t, float, false); #undef instantiate_raft_spatial_knn_detail_fusedL2Knn
0
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail/haversine_distance.cuh
/* * Copyright (c) 2020-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/util/cuda_utils.cuh> #include <raft/util/cudart_utils.hpp> #include <raft/util/pow2_utils.cuh> #include <raft/core/resources.hpp> #include <raft/distance/distance_types.hpp> #include <raft/neighbors/detail/faiss_select/Select.cuh> namespace raft { namespace spatial { namespace knn { namespace detail { template <typename value_t> DI value_t compute_haversine(value_t x1, value_t y1, value_t x2, value_t y2) { value_t sin_0 = raft::sin(0.5 * (x1 - y1)); value_t sin_1 = raft::sin(0.5 * (x2 - y2)); value_t rdist = sin_0 * sin_0 + raft::cos(x1) * raft::cos(y1) * sin_1 * sin_1; return 2 * raft::asin(raft::sqrt(rdist)); } /** * @tparam value_idx data type of indices * @tparam value_t data type of values and distances * @tparam warp_q * @tparam thread_q * @tparam tpb * @param[out] out_inds output indices * @param[out] out_dists output distances * @param[in] index index array * @param[in] query query array * @param[in] n_index_rows number of rows in index array * @param[in] k number of closest neighbors to return */ template <typename value_idx, typename value_t, int warp_q = 1024, int thread_q = 8, int tpb = 128> RAFT_KERNEL haversine_knn_kernel(value_idx* out_inds, value_t* out_dists, const value_t* index, const value_t* query, size_t n_index_rows, int k) { constexpr int kNumWarps = tpb / WarpSize; __shared__ value_t smemK[kNumWarps * warp_q]; __shared__ value_idx smemV[kNumWarps * warp_q]; using namespace raft::neighbors::detail::faiss_select; BlockSelect<value_t, value_idx, false, Comparator<value_t>, warp_q, thread_q, tpb> heap( std::numeric_limits<value_t>::max(), std::numeric_limits<value_idx>::max(), smemK, smemV, k); // Grid is exactly sized to rows available int limit = Pow2<WarpSize>::roundDown(n_index_rows); const value_t* query_ptr = query + (blockIdx.x * 2); value_t x1 = query_ptr[0]; value_t x2 = query_ptr[1]; int i = threadIdx.x; for (; i < limit; i += tpb) { const value_t* idx_ptr = index + (i * 2); value_t y1 = idx_ptr[0]; value_t y2 = idx_ptr[1]; value_t dist = compute_haversine(x1, y1, x2, y2); heap.add(dist, i); } // Handle last remainder fraction of a warp of elements if (i < n_index_rows) { const value_t* idx_ptr = index + (i * 2); value_t y1 = idx_ptr[0]; value_t y2 = idx_ptr[1]; value_t dist = compute_haversine(x1, y1, x2, y2); heap.addThreadQ(dist, i); } heap.reduce(); for (int i = threadIdx.x; i < k; i += tpb) { out_dists[blockIdx.x * k + i] = smemK[i]; out_inds[blockIdx.x * k + i] = smemV[i]; } } /** * Conmpute the k-nearest neighbors using the Haversine * (great circle arc) distance. Input is assumed to have * 2 dimensions (latitude, longitude) in radians. * @tparam value_idx * @tparam value_t * @param[out] out_inds output indices array on device (size n_query_rows * k) * @param[out] out_dists output dists array on device (size n_query_rows * k) * @param[in] index input index array on device (size n_index_rows * 2) * @param[in] query input query array on device (size n_query_rows * 2) * @param[in] n_index_rows number of rows in index array * @param[in] n_query_rows number of rows in query array * @param[in] k number of closest neighbors to return * @param[in] stream stream to order kernel launch */ template <typename value_idx, typename value_t> void haversine_knn(value_idx* out_inds, value_t* out_dists, const value_t* index, const value_t* query, size_t n_index_rows, size_t n_query_rows, int k, cudaStream_t stream) { haversine_knn_kernel<<<n_query_rows, 128, 0, stream>>>( out_inds, out_dists, index, query, n_index_rows, k); } } // namespace detail } // namespace knn } // namespace spatial } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail/processing.hpp
/* * Copyright (c) 2020-2022, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once namespace raft { namespace spatial { namespace knn { /** * @brief A virtual class defining pre- and post-processing * for metrics. This class will temporarily modify its given * state in `preprocess()` and undo those modifications in * `postprocess()` */ template <typename math_t> class MetricProcessor { public: virtual void preprocess(math_t* data) {} virtual void revert(math_t* data) {} virtual void postprocess(math_t* data) {} virtual void set_num_queries(int k) {} virtual ~MetricProcessor() = default; }; } // namespace knn } // namespace spatial } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail/fused_l2_knn-inl.cuh
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <cub/cub.cuh> #include <limits> #include <raft/linalg/norm.cuh> #include <raft/neighbors/detail/faiss_select/Select.cuh> // TODO: Need to hide the PairwiseDistance class impl and expose to public API #include "processing.cuh" #include <raft/core/operators.hpp> #include <raft/distance/detail/distance.cuh> #include <raft/distance/detail/distance_ops/l2_exp.cuh> #include <raft/distance/detail/distance_ops/l2_unexp.cuh> #include <raft/distance/detail/pairwise_distance_base.cuh> #include <raft/util/cuda_utils.cuh> namespace raft { namespace spatial { namespace knn { namespace detail { template <typename Policy, typename Pair, typename myWarpSelect, typename IdxT> DI void loadAllWarpQShmem(myWarpSelect** heapArr, Pair* shDumpKV, const IdxT m, const unsigned int numOfNN) { const int lid = raft::laneId(); #pragma unroll for (int i = 0; i < Policy::AccRowsPerTh; ++i) { const auto rowId = (threadIdx.x / Policy::AccThCols) + i * Policy::AccThRows; if (rowId < m) { #pragma unroll for (int j = 0; j < myWarpSelect::kNumWarpQRegisters; ++j) { const int idx = j * warpSize + lid; if (idx < numOfNN) { Pair KVPair = shDumpKV[rowId * numOfNN + idx]; heapArr[i]->warpV[j] = KVPair.key; heapArr[i]->warpK[j] = KVPair.value; } } } } } template <typename Policy, typename Pair, typename myWarpSelect> DI void loadWarpQShmem(myWarpSelect* heapArr, Pair* shDumpKV, const int rowId, const unsigned int numOfNN) { const int lid = raft::laneId(); #pragma unroll for (int j = 0; j < myWarpSelect::kNumWarpQRegisters; ++j) { const int idx = j * warpSize + lid; if (idx < numOfNN) { Pair KVPair = shDumpKV[rowId * numOfNN + idx]; heapArr->warpV[j] = KVPair.key; heapArr->warpK[j] = KVPair.value; } } } template <typename Policy, typename Pair, typename myWarpSelect, typename IdxT> DI void storeWarpQShmem(myWarpSelect* heapArr, Pair* shDumpKV, const IdxT rowId, const unsigned int numOfNN) { const int lid = raft::laneId(); #pragma unroll for (int j = 0; j < myWarpSelect::kNumWarpQRegisters; ++j) { const int idx = j * warpSize + lid; if (idx < numOfNN) { Pair otherKV = Pair(heapArr->warpV[j], heapArr->warpK[j]); shDumpKV[rowId * numOfNN + idx] = otherKV; } } } template <typename Policy, typename Pair, typename myWarpSelect, typename IdxT, typename OutT> DI void storeWarpQGmem(myWarpSelect** heapArr, volatile OutT* out_dists, volatile IdxT* out_inds, const IdxT m, const unsigned int numOfNN, const IdxT starty) { const int lid = raft::laneId(); #pragma unroll for (int i = 0; i < Policy::AccRowsPerTh; ++i) { const auto gmemRowId = starty + i * Policy::AccThRows; if (gmemRowId < m) { #pragma unroll for (int j = 0; j < myWarpSelect::kNumWarpQRegisters; ++j) { const auto idx = j * warpSize + lid; if (idx < numOfNN) { out_dists[std::size_t(gmemRowId) * numOfNN + idx] = heapArr[i]->warpK[j]; out_inds[std::size_t(gmemRowId) * numOfNN + idx] = (IdxT)heapArr[i]->warpV[j]; } } } } } template <typename Policy, typename Pair, typename myWarpSelect, typename IdxT, typename OutT> DI void loadPrevTopKsGmemWarpQ(myWarpSelect** heapArr, volatile OutT* out_dists, volatile IdxT* out_inds, const IdxT m, const unsigned int numOfNN, const IdxT starty) { const int lid = raft::laneId(); #pragma unroll for (int i = 0; i < Policy::AccRowsPerTh; ++i) { const auto gmemRowId = starty + i * Policy::AccThRows; if (gmemRowId < m) { #pragma unroll for (int j = 0; j < myWarpSelect::kNumWarpQRegisters; ++j) { const auto idx = j * warpSize + lid; if (idx < numOfNN) { heapArr[i]->warpK[j] = out_dists[std::size_t(gmemRowId) * numOfNN + idx]; heapArr[i]->warpV[j] = (uint32_t)out_inds[std::size_t(gmemRowId) * numOfNN + idx]; } } static constexpr auto kLaneWarpKTop = myWarpSelect::kNumWarpQRegisters - 1; heapArr[i]->warpKTop = raft::shfl(heapArr[i]->warpK[kLaneWarpKTop], heapArr[i]->kLane); } } } template <typename Pair, int NumWarpQRegs, typename myWarpSelect> DI void updateSortedWarpQ( myWarpSelect& heapArr, Pair* allWarpTopKs, int rowId, int finalNumVals, int startId = 0) { constexpr uint32_t mask = 0xffffffffu; const int lid = raft::laneId(); // calculate srcLane such that tid 0 -> 31, 1 -> 0,... 31 -> 30. // warp around 0 to 31 required for NN > 32 const auto srcLane = (warpSize + (lid - 1)) & (warpSize - 1); for (int k = startId; k < finalNumVals; k++) { Pair KVPair = allWarpTopKs[rowId * (256) + k]; #pragma unroll for (int i = 0; i < NumWarpQRegs; i++) { unsigned activeLanes = __ballot_sync(mask, KVPair.value < heapArr->warpK[i]); if (activeLanes) { Pair tempKV; tempKV.value = raft::shfl(heapArr->warpK[i], srcLane); tempKV.key = raft::shfl(heapArr->warpV[i], srcLane); const auto firstActiveLane = __ffs(activeLanes) - 1; if (firstActiveLane == lid) { heapArr->warpK[i] = KVPair.value; heapArr->warpV[i] = KVPair.key; } else if (lid > firstActiveLane) { heapArr->warpK[i] = tempKV.value; heapArr->warpV[i] = tempKV.key; } if (i == 0 && NumWarpQRegs > 1) { heapArr->warpK[1] = __shfl_up_sync(mask, heapArr->warpK[1], 1); heapArr->warpV[1] = __shfl_up_sync(mask, heapArr->warpV[1], 1); if (lid == 0) { heapArr->warpK[1] = tempKV.value; heapArr->warpV[1] = tempKV.key; } break; } } } } } template <typename DataT, typename OutT, typename IdxT, typename Policy, typename OpT, typename FinalLambda, int NumWarpQ, int NumThreadQ, bool usePrevTopKs = false, bool isRowMajor = true> __launch_bounds__(Policy::Nthreads, 2) RAFT_KERNEL fusedL2kNN(const DataT* x, const DataT* y, const DataT* _xn, const DataT* _yn, const IdxT m, const IdxT n, const IdxT k, const IdxT lda, const IdxT ldb, const IdxT ldd, OpT distance_op, FinalLambda fin_op, unsigned int numOfNN, volatile int* mutexes, volatile OutT* out_dists, volatile IdxT* out_inds) { using AccT = typename OpT::AccT; extern __shared__ char smem[]; typedef cub::KeyValuePair<uint32_t, AccT> Pair; constexpr auto identity = std::numeric_limits<AccT>::max(); constexpr auto keyMax = std::numeric_limits<uint32_t>::max(); constexpr auto Dir = false; using namespace raft::neighbors::detail::faiss_select; typedef WarpSelect<AccT, uint32_t, Dir, Comparator<AccT>, NumWarpQ, NumThreadQ, 32> myWarpSelect; auto rowEpilog_lambda = [m, n, &distance_op, numOfNN, out_dists, out_inds, mutexes] __device__(IdxT gridStrideY) { if (gridDim.x == 1) { return; } // Use ::template to disambiguate (See: // https://en.cppreference.com/w/cpp/language/dependent_name) int smem_offset = OpT::template shared_mem_size<Policy>(); Pair* shDumpKV = (Pair*)(&smem[smem_offset]); const int lid = threadIdx.x % warpSize; const IdxT starty = gridStrideY + (threadIdx.x / Policy::AccThCols); // 0 -> consumer done consuming the buffer. // -1 -> consumer started consuming the buffer // -2 -> producer done filling the buffer // 1 -> prod acquired to fill the buffer if (blockIdx.x == 0) { auto cta_processed = 0; myWarpSelect heapArr1(identity, keyMax, numOfNN); myWarpSelect heapArr2(identity, keyMax, numOfNN); myWarpSelect* heapArr[] = {&heapArr1, &heapArr2}; __syncwarp(); loadAllWarpQShmem<Policy, Pair>(heapArr, &shDumpKV[0], m, numOfNN); while (cta_processed < gridDim.x - 1) { if (threadIdx.x == 0) { while (atomicCAS((int*)&mutexes[gridStrideY / Policy::Mblk], -2, -1) != -2) ; } __threadfence(); __syncthreads(); #pragma unroll for (int i = 0; i < Policy::AccRowsPerTh; ++i) { const auto rowId = starty + i * Policy::AccThRows; if (rowId < m) { #pragma unroll for (int j = 0; j < myWarpSelect::kNumWarpQRegisters; ++j) { Pair otherKV; otherKV.value = identity; otherKV.key = keyMax; const auto idx = j * warpSize + lid; if (idx < numOfNN) { otherKV.value = out_dists[rowId * numOfNN + idx]; otherKV.key = (uint32_t)out_inds[rowId * numOfNN + idx]; const auto shMemRowId = (threadIdx.x / Policy::AccThCols) + i * Policy::AccThRows; shDumpKV[shMemRowId * numOfNN + idx] = otherKV; } } } } __threadfence(); __syncthreads(); if (threadIdx.x == 0) { atomicExch((int*)&mutexes[gridStrideY / Policy::Mblk], 0); } __threadfence(); // Perform merging of otherKV with topk's across warp. #pragma unroll for (int i = 0; i < Policy::AccRowsPerTh; ++i) { const auto rowId = starty + i * Policy::AccThRows; if (rowId < m) { #pragma unroll for (int j = 0; j < myWarpSelect::kNumWarpQRegisters; ++j) { Pair otherKV; otherKV.value = identity; otherKV.key = keyMax; const auto idx = j * warpSize + lid; if (idx < numOfNN) { const auto shMemRowId = (threadIdx.x / Policy::AccThCols) + i * Policy::AccThRows; otherKV = shDumpKV[shMemRowId * numOfNN + idx]; } heapArr[i]->add(otherKV.value, otherKV.key); } } } cta_processed++; } #pragma unroll for (int i = 0; i < Policy::AccRowsPerTh; ++i) { const auto rowId = starty + i * Policy::AccThRows; if (rowId < m) { bool needSort = (heapArr[i]->numVals > 0); needSort = __any_sync(0xffffffff, needSort); if (needSort) { heapArr[i]->reduce(); } } } storeWarpQGmem<Policy, Pair>(heapArr, out_dists, out_inds, m, numOfNN, starty); } else { if (threadIdx.x == 0) { while (atomicCAS((int*)&mutexes[gridStrideY / Policy::Mblk], 0, 1) != 0) ; } __threadfence(); __syncthreads(); #pragma unroll for (int i = 0; i < Policy::AccRowsPerTh; ++i) { const auto rowId = starty + i * Policy::AccThRows; if (rowId < m) { for (int idx = lid; idx < numOfNN; idx += warpSize) { const auto shMemRowId = (threadIdx.x / Policy::AccThCols) + i * Policy::AccThRows; Pair KVPair = shDumpKV[shMemRowId * numOfNN + idx]; out_dists[rowId * numOfNN + idx] = KVPair.value; out_inds[rowId * numOfNN + idx] = (IdxT)KVPair.key; } } } __threadfence(); __syncthreads(); if (threadIdx.x == 0) { atomicExch((int*)&mutexes[gridStrideY / Policy::Mblk], -2); } __threadfence(); } }; // epilogue operation lambda for final value calculation auto epilog_lambda = [&distance_op, numOfNN, m, n, ldd, out_dists, out_inds, keyMax, identity] __device__( AccT acc[Policy::AccRowsPerTh][Policy::AccColsPerTh], DataT * regxn, DataT * regyn, IdxT gridStrideX, IdxT gridStrideY) { // Use ::template to disambiguate (See: // https://en.cppreference.com/w/cpp/language/dependent_name) int smem_offset = OpT::template shared_mem_size<Policy>(); Pair* shDumpKV = (Pair*)(&smem[smem_offset]); constexpr uint32_t mask = 0xffffffffu; const IdxT starty = gridStrideY + (threadIdx.x / Policy::AccThCols); const IdxT startx = gridStrideX + (threadIdx.x % Policy::AccThCols); const int lid = raft::laneId(); myWarpSelect heapArr1(identity, keyMax, numOfNN); myWarpSelect heapArr2(identity, keyMax, numOfNN); myWarpSelect* heapArr[] = {&heapArr1, &heapArr2}; if (usePrevTopKs) { if (gridStrideX == blockIdx.x * Policy::Nblk) { loadPrevTopKsGmemWarpQ<Policy, Pair>(heapArr, out_dists, out_inds, m, numOfNN, starty); } } if (gridStrideX > blockIdx.x * Policy::Nblk) { #pragma unroll for (int i = 0; i < Policy::AccRowsPerTh; ++i) { const auto rowId = (threadIdx.x / Policy::AccThCols) + i * Policy::AccThRows; Pair tempKV = shDumpKV[(rowId * numOfNN) + numOfNN - 1]; heapArr[i]->warpKTop = tempKV.value; } // total vals can atmost be 256, (32*8) int numValsWarpTopK[Policy::AccRowsPerTh]; int anyWarpTopKs = 0; #pragma unroll for (int i = 0; i < Policy::AccRowsPerTh; ++i) { const auto rowId = starty + i * Policy::AccThRows; numValsWarpTopK[i] = 0; if (rowId < m) { #pragma unroll for (int j = 0; j < Policy::AccColsPerTh; ++j) { const auto colId = startx + j * Policy::AccThCols; if (colId < ldd) { if (acc[i][j] < heapArr[i]->warpKTop) { numValsWarpTopK[i]++; } } } anyWarpTopKs += numValsWarpTopK[i]; } } anyWarpTopKs = __syncthreads_or(anyWarpTopKs > 0); if (anyWarpTopKs) { Pair* allWarpTopKs = (Pair*)(&smem[0]); uint32_t needScanSort[Policy::AccRowsPerTh]; #pragma unroll for (int i = 0; i < Policy::AccRowsPerTh; ++i) { const auto gmemRowId = starty + i * Policy::AccThRows; needScanSort[i] = 0; if (gmemRowId < m) { int myVals = numValsWarpTopK[i]; needScanSort[i] = __ballot_sync(mask, myVals > 0); if (needScanSort[i]) { #pragma unroll for (unsigned int k = 1; k <= 16; k *= 2) { const unsigned int n = __shfl_up_sync(mask, numValsWarpTopK[i], k); if (lid >= k) { numValsWarpTopK[i] += n; } } } // As each thread will know its total vals to write. // we only store its starting location. numValsWarpTopK[i] -= myVals; } if (needScanSort[i]) { const auto rowId = (threadIdx.x / Policy::AccThCols) + i * Policy::AccThRows; if (gmemRowId < m) { if (needScanSort[i] & ((uint32_t)1 << lid)) { #pragma unroll for (int j = 0; j < Policy::AccColsPerTh; ++j) { const auto colId = startx + j * Policy::AccThCols; if (colId < ldd) { if (acc[i][j] < heapArr[i]->warpKTop) { Pair otherKV = {colId, acc[i][j]}; allWarpTopKs[rowId * (256) + numValsWarpTopK[i]] = otherKV; numValsWarpTopK[i]++; } } } } __syncwarp(); const int finalNumVals = raft::shfl(numValsWarpTopK[i], 31); loadWarpQShmem<Policy, Pair>(heapArr[i], &shDumpKV[0], rowId, numOfNN); updateSortedWarpQ<Pair, myWarpSelect::kNumWarpQRegisters>( heapArr[i], &allWarpTopKs[0], rowId, finalNumVals); } } } __syncthreads(); #pragma unroll for (int i = 0; i < Policy::AccRowsPerTh; ++i) { if (needScanSort[i]) { const auto rowId = (threadIdx.x / Policy::AccThCols) + i * Policy::AccThRows; const auto gmemRowId = starty + i * Policy::AccThRows; if (gmemRowId < m) { storeWarpQShmem<Policy, Pair>(heapArr[i], shDumpKV, rowId, numOfNN); } } } } } else { #pragma unroll for (int i = 0; i < Policy::AccRowsPerTh; ++i) { const auto gmemRowId = starty + i * Policy::AccThRows; const auto shMemRowId = (threadIdx.x / Policy::AccThCols) + i * Policy::AccThRows; if (gmemRowId < m) { #pragma unroll for (int j = 0; j < Policy::AccColsPerTh; ++j) { const auto colId = startx + j * Policy::AccThCols; Pair otherKV = {keyMax, identity}; if (colId < ldd) { otherKV.value = acc[i][j]; otherKV.key = colId; } heapArr[i]->add(otherKV.value, otherKV.key); } bool needSort = (heapArr[i]->numVals > 0); needSort = __any_sync(mask, needSort); if (needSort) { heapArr[i]->reduce(); } storeWarpQShmem<Policy, Pair>(heapArr[i], shDumpKV, shMemRowId, numOfNN); } } } if (((gridStrideX + Policy::Nblk * gridDim.x) >= n) && gridDim.x == 1) { // This is last iteration of grid stride X loadAllWarpQShmem<Policy, Pair>(heapArr, &shDumpKV[0], m, numOfNN); storeWarpQGmem<Policy, Pair>(heapArr, out_dists, out_inds, m, numOfNN, starty); } }; constexpr bool write_out = false; raft::distance::detail::PairwiseDistances<DataT, OutT, IdxT, Policy, OpT, decltype(epilog_lambda), FinalLambda, decltype(rowEpilog_lambda), isRowMajor, write_out> obj(x, y, m, n, k, lda, ldb, ldd, _xn, _yn, nullptr, // output ptr, can be null as write_out == false. smem, distance_op, epilog_lambda, fin_op, rowEpilog_lambda); obj.run(); } template <typename DataT, typename AccT, typename OutT, typename IdxT, int VecLen, bool usePrevTopKs, bool isRowMajor> void fusedL2UnexpKnnImpl(const DataT* x, const DataT* y, IdxT m, IdxT n, IdxT k, IdxT lda, IdxT ldb, IdxT ldd, bool sqrt, OutT* out_dists, IdxT* out_inds, IdxT numOfNN, cudaStream_t stream, void* workspace, size_t& worksize) { typedef typename raft::linalg::Policy2x8<DataT, 1>::Policy RowPolicy; typedef typename raft::linalg::Policy4x4<DataT, VecLen>::ColPolicy ColPolicy; typedef typename std::conditional<true, RowPolicy, ColPolicy>::type KPolicy; ASSERT(isRowMajor, "Only Row major inputs are allowed"); dim3 blk(KPolicy::Nthreads); // Accumulation operation lambda typedef cub::KeyValuePair<uint32_t, AccT> Pair; raft::distance::detail::ops::l2_unexp_distance_op<DataT, AccT, IdxT> distance_op{sqrt}; raft::identity_op fin_op{}; if constexpr (isRowMajor) { constexpr auto fusedL2UnexpKnn32RowMajor = fusedL2kNN<DataT, OutT, IdxT, KPolicy, decltype(distance_op), decltype(fin_op), 32, 2, usePrevTopKs, isRowMajor>; constexpr auto fusedL2UnexpKnn64RowMajor = fusedL2kNN<DataT, OutT, IdxT, KPolicy, decltype(distance_op), decltype(fin_op), 64, 3, usePrevTopKs, isRowMajor>; auto fusedL2UnexpKnnRowMajor = fusedL2UnexpKnn32RowMajor; if (numOfNN <= 32) { fusedL2UnexpKnnRowMajor = fusedL2UnexpKnn32RowMajor; } else if (numOfNN <= 64) { fusedL2UnexpKnnRowMajor = fusedL2UnexpKnn64RowMajor; } else { ASSERT(numOfNN <= 64, "fusedL2kNN: num of nearest neighbors must be <= 64"); } const auto sharedMemSize = distance_op.template shared_mem_size<KPolicy>() + KPolicy::Mblk * numOfNN * sizeof(Pair); dim3 grid = raft::distance::detail::launchConfigGenerator<KPolicy>( m, n, sharedMemSize, fusedL2UnexpKnnRowMajor); if (grid.x > 1) { const auto numMutexes = raft::ceildiv<int>(m, KPolicy::Mblk); if (workspace == nullptr || worksize < (sizeof(int32_t) * numMutexes)) { worksize = sizeof(int32_t) * numMutexes; return; } else { RAFT_CUDA_TRY(cudaMemsetAsync(workspace, 0, sizeof(int32_t) * numMutexes, stream)); } } fusedL2UnexpKnnRowMajor<<<grid, blk, sharedMemSize, stream>>>(x, y, nullptr, nullptr, m, n, k, lda, ldb, ldd, distance_op, fin_op, (uint32_t)numOfNN, (int*)workspace, out_dists, out_inds); } else { } RAFT_CUDA_TRY(cudaGetLastError()); } template <typename DataT, typename AccT, typename OutT, typename IdxT, bool usePrevTopKs, bool isRowMajor> void fusedL2UnexpKnn(IdxT m, IdxT n, IdxT k, IdxT lda, IdxT ldb, IdxT ldd, const DataT* x, const DataT* y, bool sqrt, OutT* out_dists, IdxT* out_inds, IdxT numOfNN, cudaStream_t stream, void* workspace, size_t& worksize) { size_t bytesA = sizeof(DataT) * lda; size_t bytesB = sizeof(DataT) * ldb; if (16 % sizeof(DataT) == 0 && bytesA % 16 == 0 && bytesB % 16 == 0) { fusedL2UnexpKnnImpl<DataT, AccT, OutT, IdxT, 16 / sizeof(DataT), usePrevTopKs, isRowMajor>( x, y, m, n, k, lda, ldb, ldd, sqrt, out_dists, out_inds, numOfNN, stream, workspace, worksize); } else if (8 % sizeof(DataT) == 0 && bytesA % 8 == 0 && bytesB % 8 == 0) { fusedL2UnexpKnnImpl<DataT, AccT, OutT, IdxT, 8 / sizeof(DataT), usePrevTopKs, isRowMajor>( x, y, m, n, k, lda, ldb, ldd, sqrt, out_dists, out_inds, numOfNN, stream, workspace, worksize); } else { fusedL2UnexpKnnImpl<DataT, AccT, OutT, IdxT, 1, usePrevTopKs, isRowMajor>(x, y, m, n, k, lda, ldb, ldd, sqrt, out_dists, out_inds, numOfNN, stream, workspace, worksize); } } template <typename DataT, typename AccT, typename OutT, typename IdxT, int VecLen, bool usePrevTopKs, bool isRowMajor> void fusedL2ExpKnnImpl(const DataT* x, const DataT* y, const DataT* xn, const DataT* yn, IdxT m, IdxT n, IdxT k, IdxT lda, IdxT ldb, IdxT ldd, bool sqrt, OutT* out_dists, IdxT* out_inds, IdxT numOfNN, cudaStream_t stream, void* workspace, size_t& worksize) { typedef typename raft::linalg::Policy2x8<DataT, 1>::Policy RowPolicy; typedef typename raft::linalg::Policy4x4<DataT, VecLen>::ColPolicy ColPolicy; typedef typename std::conditional<true, RowPolicy, ColPolicy>::type KPolicy; ASSERT(isRowMajor, "Only Row major inputs are allowed"); ASSERT(!(((x != y) && (worksize < (m + n) * sizeof(AccT))) || (worksize < m * sizeof(AccT))), "workspace size error"); ASSERT(workspace != nullptr, "workspace is null"); dim3 blk(KPolicy::Nthreads); typedef cub::KeyValuePair<uint32_t, AccT> Pair; raft::distance::detail::ops::l2_exp_distance_op<DataT, AccT, IdxT> distance_op{sqrt}; raft::identity_op fin_op{}; if constexpr (isRowMajor) { constexpr auto fusedL2ExpKnn32RowMajor = fusedL2kNN<DataT, OutT, IdxT, KPolicy, decltype(distance_op), decltype(fin_op), 32, 2, usePrevTopKs, isRowMajor>; constexpr auto fusedL2ExpKnn64RowMajor = fusedL2kNN<DataT, OutT, IdxT, KPolicy, decltype(distance_op), decltype(fin_op), 64, 3, usePrevTopKs, isRowMajor>; auto fusedL2ExpKnnRowMajor = fusedL2ExpKnn32RowMajor; if (numOfNN <= 32) { fusedL2ExpKnnRowMajor = fusedL2ExpKnn32RowMajor; } else if (numOfNN <= 64) { fusedL2ExpKnnRowMajor = fusedL2ExpKnn64RowMajor; } else { ASSERT(numOfNN <= 64, "fusedL2kNN: num of nearest neighbors must be <= 64"); } const auto sharedMemSize = distance_op.template shared_mem_size<KPolicy>() + (KPolicy::Mblk * numOfNN * sizeof(Pair)); dim3 grid = raft::distance::detail::launchConfigGenerator<KPolicy>( m, n, sharedMemSize, fusedL2ExpKnnRowMajor); int32_t* mutexes = nullptr; if (grid.x > 1) { const auto numMutexes = raft::ceildiv<int>(m, KPolicy::Mblk); const auto normsSize = (x != y) ? (m + n) * sizeof(DataT) : n * sizeof(DataT); const auto requiredSize = sizeof(int32_t) * numMutexes + normsSize; if (worksize < requiredSize) { worksize = requiredSize; return; } else { mutexes = (int32_t*)((char*)workspace + normsSize); RAFT_CUDA_TRY(cudaMemsetAsync(mutexes, 0, sizeof(int32_t) * numMutexes, stream)); } } // calculate norms if they haven't been passed in if (!xn) { DataT* xn_ = (DataT*)workspace; workspace = xn_ + m; raft::linalg::rowNorm( xn_, x, k, m, raft::linalg::L2Norm, isRowMajor, stream, raft::identity_op{}); xn = xn_; } if (!yn) { if (x == y) { yn = xn; } else { DataT* yn_ = (DataT*)(workspace); raft::linalg::rowNorm( yn_, y, k, n, raft::linalg::L2Norm, isRowMajor, stream, raft::identity_op{}); yn = yn_; } } fusedL2ExpKnnRowMajor<<<grid, blk, sharedMemSize, stream>>>(x, y, xn, yn, m, n, k, lda, ldb, ldd, distance_op, fin_op, (uint32_t)numOfNN, mutexes, out_dists, out_inds); } else { } RAFT_CUDA_TRY(cudaGetLastError()); } template <typename DataT, typename AccT, typename OutT, typename IdxT, bool usePrevTopKs, bool isRowMajor> void fusedL2ExpKnn(IdxT m, IdxT n, IdxT k, IdxT lda, IdxT ldb, IdxT ldd, const DataT* x, const DataT* y, const DataT* xn, const DataT* yn, bool sqrt, OutT* out_dists, IdxT* out_inds, IdxT numOfNN, cudaStream_t stream, void* workspace, size_t& worksize) { size_t bytesA = sizeof(DataT) * lda; size_t bytesB = sizeof(DataT) * ldb; if (16 % sizeof(DataT) == 0 && bytesA % 16 == 0 && bytesB % 16 == 0) { fusedL2ExpKnnImpl<DataT, AccT, OutT, IdxT, 16 / sizeof(DataT), usePrevTopKs, isRowMajor>( x, y, xn, yn, m, n, k, lda, ldb, ldd, sqrt, out_dists, out_inds, numOfNN, stream, workspace, worksize); } else if (8 % sizeof(DataT) == 0 && bytesA % 8 == 0 && bytesB % 8 == 0) { fusedL2ExpKnnImpl<DataT, AccT, OutT, IdxT, 8 / sizeof(DataT), usePrevTopKs, isRowMajor>( x, y, xn, yn, m, n, k, lda, ldb, ldd, sqrt, out_dists, out_inds, numOfNN, stream, workspace, worksize); } else { fusedL2ExpKnnImpl<DataT, AccT, OutT, IdxT, 1, usePrevTopKs, isRowMajor>(x, y, xn, yn, m, n, k, lda, ldb, ldd, sqrt, out_dists, out_inds, numOfNN, stream, workspace, worksize); } } /** * Compute the k-nearest neighbors using L2 expanded/unexpanded distance. * @tparam value_idx * @tparam value_t * @param[out] out_inds output indices array on device (size n_query_rows * k) * @param[out] out_dists output dists array on device (size n_query_rows * k) * @param[in] index input index array on device (size n_index_rows * D) * @param[in] query input query array on device (size n_query_rows * D) * @param[in] n_index_rows number of rows in index array * @param[in] n_query_rows number of rows in query array * @param[in] k number of closest neighbors to return * @param[in] rowMajorIndex are the index arrays in row-major layout? * @param[in] rowMajorQuery are the query array in row-major layout? * @param[in] stream stream to order kernel launch */ template <typename value_idx, typename value_t, bool usePrevTopKs = false> void fusedL2Knn(size_t D, value_idx* out_inds, value_t* out_dists, const value_t* index, const value_t* query, size_t n_index_rows, size_t n_query_rows, int k, bool rowMajorIndex, bool rowMajorQuery, cudaStream_t stream, raft::distance::DistanceType metric, const value_t* index_norms = NULL, const value_t* query_norms = NULL) { // Validate the input data ASSERT(k > 0, "l2Knn: k must be > 0"); ASSERT(D > 0, "l2Knn: D must be > 0"); ASSERT(n_index_rows > 0, "l2Knn: n_index_rows must be > 0"); ASSERT(index, "l2Knn: index must be provided (passed null)"); ASSERT(n_query_rows > 0, "l2Knn: n_query_rows must be > 0"); ASSERT(query, "l2Knn: query must be provided (passed null)"); ASSERT(out_dists, "l2Knn: out_dists must be provided (passed null)"); ASSERT(out_inds, "l2Knn: out_inds must be provided (passed null)"); // Currently we only support same layout for x & y inputs. ASSERT(rowMajorIndex == rowMajorQuery, "l2Knn: rowMajorIndex and rowMajorQuery should have same layout"); // TODO: Add support for column major layout ASSERT(rowMajorIndex == true, "l2Knn: only rowMajor inputs are supported for now."); // Even for L2 Sqrt distance case we use non-sqrt version as FAISS bfKNN only support // non-sqrt metric & some tests in RAFT/cuML (like Linkage) fails if we use L2 sqrt. constexpr bool sqrt = false; size_t worksize = 0, tempWorksize = 0; rmm::device_uvector<char> workspace(worksize, stream); value_idx lda = D, ldb = D, ldd = n_index_rows; switch (metric) { case raft::distance::DistanceType::L2SqrtExpanded: case raft::distance::DistanceType::L2Expanded: tempWorksize = raft::distance::detail:: getWorkspaceSize<raft::distance::DistanceType::L2Expanded, float, float, float, value_idx>( query, index, n_query_rows, n_index_rows, D); worksize = tempWorksize; workspace.resize(worksize, stream); fusedL2ExpKnn<value_t, value_t, value_t, value_idx, usePrevTopKs, true>(n_query_rows, n_index_rows, D, lda, ldb, ldd, query, index, query_norms, index_norms, sqrt, out_dists, out_inds, k, stream, workspace.data(), worksize); if (worksize > tempWorksize) { workspace.resize(worksize, stream); fusedL2ExpKnn<value_t, value_t, value_t, value_idx, usePrevTopKs, true>(n_query_rows, n_index_rows, D, lda, ldb, ldd, query, index, query_norms, index_norms, sqrt, out_dists, out_inds, k, stream, workspace.data(), worksize); } break; case raft::distance::DistanceType::L2Unexpanded: case raft::distance::DistanceType::L2SqrtUnexpanded: fusedL2UnexpKnn<value_t, value_t, value_t, value_idx, usePrevTopKs, true>(n_query_rows, n_index_rows, D, lda, ldb, ldd, query, index, sqrt, out_dists, out_inds, k, stream, workspace.data(), worksize); if (worksize) { workspace.resize(worksize, stream); fusedL2UnexpKnn<value_t, value_t, value_t, value_idx, usePrevTopKs, true>(n_query_rows, n_index_rows, D, lda, ldb, ldd, query, index, sqrt, out_dists, out_inds, k, stream, workspace.data(), worksize); } break; default: printf("only L2 distance metric is supported\n"); break; }; } } // namespace detail } // namespace knn } // namespace spatial } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail/fused_l2_knn.cuh
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #ifndef RAFT_EXPLICIT_INSTANTIATE_ONLY #include "fused_l2_knn-inl.cuh" #endif #ifdef RAFT_COMPILED #include "fused_l2_knn-ext.cuh" #endif
0
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail/epsilon_neighborhood.cuh
/* * Copyright (c) 2020-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/linalg/contractions.cuh> #include <raft/util/device_utils.cuh> namespace raft { namespace spatial { namespace knn { namespace detail { template <typename DataT, typename IdxT, typename Policy, typename BaseClass = raft::linalg::Contractions_NT<DataT, IdxT, Policy>> struct EpsUnexpL2SqNeighborhood : public BaseClass { private: typedef Policy P; bool* adj; DataT eps; IdxT* vd; char* smem; // for final reductions DataT acc[P::AccRowsPerTh][P::AccColsPerTh]; public: DI EpsUnexpL2SqNeighborhood(bool* _adj, IdxT* _vd, const DataT* _x, const DataT* _y, IdxT _m, IdxT _n, IdxT _k, DataT _eps, char* _smem) : BaseClass(_x, _y, _m, _n, _k, _smem), adj(_adj), eps(_eps), vd(_vd), smem(_smem) { } DI void run() { prolog(); loop(); epilog(); } private: DI void prolog() { this->ldgXY(IdxT(blockIdx.x) * P::Mblk, IdxT(blockIdx.y) * P::Nblk, 0); #pragma unroll for (int i = 0; i < P::AccRowsPerTh; ++i) { #pragma unroll for (int j = 0; j < P::AccColsPerTh; ++j) { acc[i][j] = BaseClass::Zero; } } this->stsXY(); __syncthreads(); this->switch_write_buffer(); } DI void loop() { for (int kidx = P::Kblk; kidx < this->k; kidx += P::Kblk) { this->ldgXY(IdxT(blockIdx.x) * P::Mblk, IdxT(blockIdx.y) * P::Nblk, kidx); accumulate(); // on the previous k-block this->stsXY(); __syncthreads(); this->switch_write_buffer(); this->switch_read_buffer(); } accumulate(); // last iteration } DI void epilog() { IdxT startx = blockIdx.x * P::Mblk + this->accrowid; IdxT starty = blockIdx.y * P::Nblk + this->acccolid; auto lid = raft::laneId(); IdxT sums[P::AccColsPerTh]; #pragma unroll for (int j = 0; j < P::AccColsPerTh; ++j) { sums[j] = 0; } #pragma unroll for (int i = 0; i < P::AccRowsPerTh; ++i) { auto xid = startx + i * P::AccThRows; #pragma unroll for (int j = 0; j < P::AccColsPerTh; ++j) { auto yid = starty + j * P::AccThCols; auto is_neigh = acc[i][j] <= eps; ///@todo: fix uncoalesced writes using shared mem if (xid < this->m && yid < this->n) { adj[xid * this->n + yid] = is_neigh; sums[j] += is_neigh; } } } // perform reduction of adjacency values to compute vertex degrees if (vd != nullptr) { updateVertexDegree(sums); } } DI void accumulate() { #pragma unroll for (int ki = 0; ki < P::Kblk; ki += P::Veclen) { this->ldsXY(ki); #pragma unroll for (int i = 0; i < P::AccRowsPerTh; ++i) { #pragma unroll for (int j = 0; j < P::AccColsPerTh; ++j) { #pragma unroll for (int v = 0; v < P::Veclen; ++v) { auto diff = this->regx[i][v] - this->regy[j][v]; acc[i][j] += diff * diff; } } } } } DI void updateVertexDegree(IdxT (&sums)[P::AccColsPerTh]) { __syncthreads(); // so that we can safely reuse smem int gid = threadIdx.x / P::AccThCols; int lid = threadIdx.x % P::AccThCols; auto cidx = IdxT(blockIdx.y) * P::Nblk + lid; IdxT totalSum = 0; // update the individual vertex degrees #pragma unroll for (int i = 0; i < P::AccColsPerTh; ++i) { sums[i] = batchedBlockReduce<IdxT, P::AccThCols>(sums[i], smem); auto cid = cidx + i * P::AccThCols; if (gid == 0 && cid < this->n) { atomicUpdate(cid, sums[i]); totalSum += sums[i]; } __syncthreads(); // for safe smem reuse } // update the total edge count totalSum = raft::blockReduce<IdxT>(totalSum, smem); if (threadIdx.x == 0) { atomicUpdate(this->n, totalSum); } } DI void atomicUpdate(IdxT addrId, IdxT val) { if (sizeof(IdxT) == 4) { raft::myAtomicAdd<unsigned>((unsigned*)(vd + addrId), val); } else if (sizeof(IdxT) == 8) { raft::myAtomicAdd<unsigned long long>((unsigned long long*)(vd + addrId), val); } } }; // struct EpsUnexpL2SqNeighborhood template <typename DataT, typename IdxT, typename Policy> __launch_bounds__(Policy::Nthreads, 2) RAFT_KERNEL epsUnexpL2SqNeighKernel( bool* adj, IdxT* vd, const DataT* x, const DataT* y, IdxT m, IdxT n, IdxT k, DataT eps) { extern __shared__ char smem[]; EpsUnexpL2SqNeighborhood<DataT, IdxT, Policy> obj(adj, vd, x, y, m, n, k, eps, smem); obj.run(); } template <typename DataT, typename IdxT, int VecLen> void epsUnexpL2SqNeighImpl(bool* adj, IdxT* vd, const DataT* x, const DataT* y, IdxT m, IdxT n, IdxT k, DataT eps, cudaStream_t stream) { typedef typename raft::linalg::Policy4x4<DataT, VecLen>::Policy Policy; dim3 grid(raft::ceildiv<int>(m, Policy::Mblk), raft::ceildiv<int>(n, Policy::Nblk)); dim3 blk(Policy::Nthreads); epsUnexpL2SqNeighKernel<DataT, IdxT, Policy> <<<grid, blk, Policy::SmemSize, stream>>>(adj, vd, x, y, m, n, k, eps); RAFT_CUDA_TRY(cudaGetLastError()); } /** * @brief Computes epsilon neighborhood for the L2-Squared distance metric * * @tparam DataT IO and math type * @tparam IdxT Index type * * @param[out] adj adjacency matrix [row-major] [on device] [dim = m x n] * @param[out] vd vertex degree array [on device] [len = m + 1] * `vd + m` stores the total number of edges in the adjacency * matrix. Pass a nullptr if you don't need this info. * @param[in] x first matrix [row-major] [on device] [dim = m x k] * @param[in] y second matrix [row-major] [on device] [dim = n x k] * @param[in] eps defines epsilon neighborhood radius (should be passed as * squared as we compute L2-squared distance in this method) * @param[in] fop device lambda to do any other custom functions * @param[in] stream cuda stream */ template <typename DataT, typename IdxT> void epsUnexpL2SqNeighborhood(bool* adj, IdxT* vd, const DataT* x, const DataT* y, IdxT m, IdxT n, IdxT k, DataT eps, cudaStream_t stream) { size_t bytes = sizeof(DataT) * k; if (16 % sizeof(DataT) == 0 && bytes % 16 == 0) { epsUnexpL2SqNeighImpl<DataT, IdxT, 16 / sizeof(DataT)>(adj, vd, x, y, m, n, k, eps, stream); } else if (8 % sizeof(DataT) == 0 && bytes % 8 == 0) { epsUnexpL2SqNeighImpl<DataT, IdxT, 8 / sizeof(DataT)>(adj, vd, x, y, m, n, k, eps, stream); } else { epsUnexpL2SqNeighImpl<DataT, IdxT, 1>(adj, vd, x, y, m, n, k, eps, stream); } } } // namespace detail } // namespace knn } // namespace spatial } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail/ann_utils.cuh
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/logger.hpp> #include <raft/distance/distance_types.hpp> #include <raft/util/cuda_utils.cuh> #include <raft/util/cudart_utils.hpp> #include <raft/util/integer_utils.hpp> #include <rmm/cuda_stream_view.hpp> #include <rmm/device_scalar.hpp> #include <rmm/device_uvector.hpp> #include <memory> #include <optional> #include <cuda_fp16.hpp> namespace raft::spatial::knn::detail::utils { /** Whether pointers are accessible on the device or on the host. */ enum class pointer_residency { /** Some of the pointers are on the device, some on the host. */ mixed, /** All pointers accessible from both the device and the host. */ host_and_device, /** All pointers are host accessible. */ host_only, /** All poitners are device accessible. */ device_only }; template <typename... Types> struct pointer_residency_count {}; template <> struct pointer_residency_count<> { static inline auto run() -> std::tuple<int, int> { return std::make_tuple(0, 0); } }; template <typename Type, typename... Types> struct pointer_residency_count<Type, Types...> { static inline auto run(const Type* ptr, const Types*... ptrs) -> std::tuple<int, int> { auto [on_device, on_host] = pointer_residency_count<Types...>::run(ptrs...); cudaPointerAttributes attr; RAFT_CUDA_TRY(cudaPointerGetAttributes(&attr, ptr)); switch (attr.type) { case cudaMemoryTypeUnregistered: return std::make_tuple(on_device, on_host + 1); case cudaMemoryTypeHost: return std::make_tuple(on_device + int(attr.devicePointer == ptr), on_host + 1); case cudaMemoryTypeDevice: return std::make_tuple(on_device + 1, on_host); case cudaMemoryTypeManaged: return std::make_tuple(on_device + 1, on_host + 1); default: return std::make_tuple(on_device, on_host); } } }; /** Check if all argument pointers reside on the host or on the device. */ template <typename... Types> auto check_pointer_residency(const Types*... ptrs) -> pointer_residency { auto [on_device, on_host] = pointer_residency_count<Types...>::run(ptrs...); int n_args = sizeof...(Types); if (on_device == n_args && on_host == n_args) { return pointer_residency::host_and_device; } if (on_device == n_args) { return pointer_residency::device_only; } if (on_host == n_args) { return pointer_residency::host_only; } return pointer_residency::mixed; } /** RAII helper to access the host data from gpu when necessary. */ template <typename PtrT, typename Action> struct with_mapped_memory_t { with_mapped_memory_t(PtrT ptr, size_t size, Action action) : action_(action) { if (ptr == nullptr) { return; } switch (utils::check_pointer_residency(ptr)) { case utils::pointer_residency::device_only: case utils::pointer_residency::host_and_device: { dev_ptr_ = (void*)ptr; // NOLINT } break; default: { host_ptr_ = (void*)ptr; // NOLINT RAFT_CUDA_TRY(cudaHostRegister(host_ptr_, size, choose_flags(ptr))); RAFT_CUDA_TRY(cudaHostGetDevicePointer(&dev_ptr_, host_ptr_, 0)); } break; } } ~with_mapped_memory_t() { if (host_ptr_ != nullptr) { cudaHostUnregister(host_ptr_); } } auto operator()() { return action_((PtrT)dev_ptr_); } // NOLINT private: Action action_; void* host_ptr_ = nullptr; void* dev_ptr_ = nullptr; template <typename T> static auto choose_flags(const T*) -> unsigned int { int dev_id, readonly_supported; RAFT_CUDA_TRY(cudaGetDevice(&dev_id)); RAFT_CUDA_TRY(cudaDeviceGetAttribute( &readonly_supported, cudaDevAttrHostRegisterReadOnlySupported, dev_id)); if (readonly_supported) { return cudaHostRegisterMapped | cudaHostRegisterReadOnly; } else { return cudaHostRegisterMapped; } } template <typename T> static auto choose_flags(T*) -> unsigned int { return cudaHostRegisterMapped; } }; template <typename T> struct config {}; template <> struct config<double> { using value_t = double; static constexpr double kDivisor = 1.0; }; template <> struct config<float> { using value_t = float; static constexpr double kDivisor = 1.0; }; template <> struct config<half> { using value_t = half; static constexpr double kDivisor = 1.0; }; template <> struct config<uint8_t> { using value_t = uint32_t; static constexpr double kDivisor = 256.0; }; template <> struct config<int8_t> { using value_t = int32_t; static constexpr double kDivisor = 128.0; }; /** * @brief Converting values between the types taking into account scaling factors * for the integral types. * * @tparam T target type of the mapping. */ template <typename T> struct mapping { /** * @defgroup * @brief Cast and possibly scale a value of the source type `S` to the target type `T`. * * @tparam S source type * @param x source value * @{ */ template <typename S> HDI constexpr auto operator()(const S& x) const -> std::enable_if_t<std::is_same_v<S, T>, T> { return x; }; template <typename S> HDI constexpr auto operator()(const S& x) const -> std::enable_if_t<!std::is_same_v<S, T>, T> { constexpr double kMult = config<T>::kDivisor / config<S>::kDivisor; if constexpr (std::is_floating_point_v<S>) { return static_cast<T>(x * static_cast<S>(kMult)); } if constexpr (std::is_floating_point_v<T>) { return static_cast<T>(x) * static_cast<T>(kMult); } return static_cast<T>(static_cast<float>(x) * static_cast<float>(kMult)); }; /** @} */ }; /** * @brief Sets the first num bytes of the block of memory pointed by ptr to the specified value. * * @param[out] ptr host or device pointer * @param[in] value * @param[in] n_bytes */ template <typename T, typename IdxT> inline void memzero(T* ptr, IdxT n_elems, rmm::cuda_stream_view stream) { switch (check_pointer_residency(ptr)) { case pointer_residency::host_and_device: case pointer_residency::device_only: { RAFT_CUDA_TRY(cudaMemsetAsync(ptr, 0, n_elems * sizeof(T), stream)); } break; case pointer_residency::host_only: { stream.synchronize(); ::memset(ptr, 0, n_elems * sizeof(T)); } break; default: RAFT_FAIL("memset: unreachable code"); } } template <typename T, typename IdxT> RAFT_KERNEL outer_add_kernel(const T* a, IdxT len_a, const T* b, IdxT len_b, T* c) { IdxT gid = threadIdx.x + blockDim.x * static_cast<IdxT>(blockIdx.x); IdxT i = gid / len_b; IdxT j = gid % len_b; if (i >= len_a) return; c[gid] = (a == nullptr ? T(0) : a[i]) + (b == nullptr ? T(0) : b[j]); } template <typename T, typename IdxT> RAFT_KERNEL block_copy_kernel(const IdxT* in_offsets, const IdxT* out_offsets, IdxT n_blocks, const T* in_data, T* out_data, IdxT n_mult) { IdxT i = static_cast<IdxT>(blockDim.x) * static_cast<IdxT>(blockIdx.x) + threadIdx.x; // find the source offset using the binary search. uint32_t l = 0; uint32_t r = n_blocks; IdxT in_offset = 0; if (in_offsets[r] * n_mult <= i) return; while (l + 1 < r) { uint32_t c = (l + r) >> 1; IdxT o = in_offsets[c] * n_mult; if (o <= i) { l = c; in_offset = o; } else { r = c; } } // copy the data out_data[out_offsets[l] * n_mult - in_offset + i] = in_data[i]; } /** * Copy chunks of data from one array to another at given offsets. * * @tparam T element type * @tparam IdxT index type * * @param[in] in_offsets * @param[in] out_offsets * @param n_blocks size of the offset arrays minus one. * @param[in] in_data * @param[out] out_data * @param n_mult constant multiplier for offset values (such as e.g. `dim`) * @param stream */ template <typename T, typename IdxT> void block_copy(const IdxT* in_offsets, const IdxT* out_offsets, IdxT n_blocks, const T* in_data, T* out_data, IdxT n_mult, rmm::cuda_stream_view stream) { IdxT in_size; update_host(&in_size, in_offsets + n_blocks, 1, stream); stream.synchronize(); dim3 threads(128, 1, 1); dim3 blocks(ceildiv<IdxT>(in_size * n_mult, threads.x), 1, 1); block_copy_kernel<<<blocks, threads, 0, stream>>>( in_offsets, out_offsets, n_blocks, in_data, out_data, n_mult); } /** * @brief Fill matrix `c` with all combinations of sums of vectors `a` and `b`. * * NB: device-only function * * @tparam T element type * @tparam IdxT index type * * @param[in] a device pointer to a vector [len_a] * @param len_a number of elements in `a` * @param[in] b device pointer to a vector [len_b] * @param len_b number of elements in `b` * @param[out] c row-major matrix [len_a, len_b] * @param stream */ template <typename T, typename IdxT> void outer_add(const T* a, IdxT len_a, const T* b, IdxT len_b, T* c, rmm::cuda_stream_view stream) { dim3 threads(128, 1, 1); dim3 blocks(ceildiv<IdxT>(len_a * len_b, threads.x), 1, 1); outer_add_kernel<<<blocks, threads, 0, stream>>>(a, len_a, b, len_b, c); } template <typename T, typename S, typename IdxT, typename LabelT> RAFT_KERNEL copy_selected_kernel( IdxT n_rows, IdxT n_cols, const S* src, const LabelT* row_ids, IdxT ld_src, T* dst, IdxT ld_dst) { IdxT gid = threadIdx.x + blockDim.x * static_cast<IdxT>(blockIdx.x); IdxT j = gid % n_cols; IdxT i_dst = gid / n_cols; if (i_dst >= n_rows) return; auto i_src = static_cast<IdxT>(row_ids[i_dst]); dst[ld_dst * i_dst + j] = mapping<T>{}(src[ld_src * i_src + j]); } /** * @brief Copy selected rows of a matrix while mapping the data from the source to the target * type. * * @tparam T target type * @tparam S source type * @tparam IdxT index type * @tparam LabelT label type * * @param n_rows * @param n_cols * @param[in] src input matrix [..., ld_src] * @param[in] row_ids selection of rows to be copied [n_rows] * @param ld_src number of cols in the input (ld_src >= n_cols) * @param[out] dst output matrix [n_rows, ld_dst] * @param ld_dst number of cols in the output (ld_dst >= n_cols) * @param stream */ template <typename T, typename S, typename IdxT, typename LabelT> void copy_selected(IdxT n_rows, IdxT n_cols, const S* src, const LabelT* row_ids, IdxT ld_src, T* dst, IdxT ld_dst, rmm::cuda_stream_view stream) { switch (check_pointer_residency(src, dst, row_ids)) { case pointer_residency::host_and_device: case pointer_residency::device_only: { IdxT block_dim = 128; IdxT grid_dim = ceildiv(n_rows * n_cols, block_dim); copy_selected_kernel<T, S> <<<grid_dim, block_dim, 0, stream>>>(n_rows, n_cols, src, row_ids, ld_src, dst, ld_dst); } break; case pointer_residency::host_only: { stream.synchronize(); for (IdxT i_dst = 0; i_dst < n_rows; i_dst++) { auto i_src = static_cast<IdxT>(row_ids[i_dst]); for (IdxT j = 0; j < n_cols; j++) { dst[ld_dst * i_dst + j] = mapping<T>{}(src[ld_src * i_src + j]); } } stream.synchronize(); } break; default: RAFT_FAIL("All pointers must reside on the same side, host or device."); } } /** * A batch input iterator over the data source. * Given an input pointer, it decides whether the current device has the access to the data and * gives it back to the user in batches. Three scenarios are possible: * * 1. if `source == nullptr`: then `batch.data() == nullptr` * 2. if `source` is accessible from the device, `batch.data()` points directly at the source at * the proper offsets on each iteration. * 3. if `source` is not accessible from the device, `batch.data()` points to an intermediate * buffer; the corresponding data is copied in the given `stream` on every iterator dereference * (i.e. batches can be skipped). Dereferencing the same batch two times in a row does not force * the copy. * * In all three scenarios, the number of iterations, batch offsets and sizes are the same. * * The iterator can be reused. If the number of iterations is one, at most one copy will ever be * invoked (i.e. small datasets are not reloaded multiple times). */ template <typename T> struct batch_load_iterator { using size_type = size_t; /** A single batch of data residing in device memory. */ struct batch { /** Logical width of a single row in a batch, in elements of type `T`. */ [[nodiscard]] auto row_width() const -> size_type { return row_width_; } /** Logical offset of the batch, in rows (`row_width()`) */ [[nodiscard]] auto offset() const -> size_type { return pos_.value_or(0) * batch_size_; } /** Logical size of the batch, in rows (`row_width()`) */ [[nodiscard]] auto size() const -> size_type { return batch_len_; } /** Logical size of the batch, in rows (`row_width()`) */ [[nodiscard]] auto data() const -> const T* { return const_cast<const T*>(dev_ptr_); } /** Whether this batch copies the data (i.e. the source is inaccessible from the device). */ [[nodiscard]] auto does_copy() const -> bool { return needs_copy_; } private: batch(const T* source, size_type n_rows, size_type row_width, size_type batch_size, rmm::cuda_stream_view stream, rmm::mr::device_memory_resource* mr) : stream_(stream), buf_(0, stream, mr), source_(source), dev_ptr_(nullptr), n_rows_(n_rows), row_width_(row_width), batch_size_(std::min(batch_size, n_rows)), pos_(std::nullopt), n_iters_(raft::div_rounding_up_safe(n_rows, batch_size)), needs_copy_(false) { if (source_ == nullptr) { return; } cudaPointerAttributes attr; RAFT_CUDA_TRY(cudaPointerGetAttributes(&attr, source_)); dev_ptr_ = reinterpret_cast<T*>(attr.devicePointer); if (dev_ptr_ == nullptr) { buf_.resize(row_width_ * batch_size_, stream); dev_ptr_ = buf_.data(); needs_copy_ = true; } } rmm::cuda_stream_view stream_; rmm::device_uvector<T> buf_; const T* source_; size_type n_rows_; size_type row_width_; size_type batch_size_; size_type n_iters_; bool needs_copy_; std::optional<size_type> pos_; size_type batch_len_; T* dev_ptr_; friend class batch_load_iterator<T>; /** * Changes the state of the batch to point at the `pos` index. * If necessary, copies the data from the source in the registered stream. */ void load(const size_type& pos) { // No-op if the data is already loaded, or it's the end of the input. if (pos == pos_ || pos >= n_iters_) { return; } pos_.emplace(pos); batch_len_ = std::min(batch_size_, n_rows_ - std::min(offset(), n_rows_)); if (source_ == nullptr) { return; } if (needs_copy_) { if (size() > 0) { RAFT_LOG_TRACE("batch_load_iterator::copy(offset = %zu, size = %zu, row_width = %zu)", size_t(offset()), size_t(size()), size_t(row_width())); copy(dev_ptr_, source_ + offset() * row_width(), size() * row_width(), stream_); } } else { dev_ptr_ = const_cast<T*>(source_) + offset() * row_width(); } } }; using value_type = batch; using reference = const value_type&; using pointer = const value_type*; /** * Create a batch iterator over the data `source`. * * For convenience, the data `source` is read in logical units of size `row_width`; batch sizes * and offsets are calculated in logical rows. Hence, can interpret the data as a contiguous * row-major matrix of size [n_rows, row_width], and the batches are the sub-matrices of size * [x<=batch_size, n_rows]. * * @param source the input data -- host, device, or nullptr. * @param n_rows the size of the input in logical rows. * @param row_width the size of the logical row in the elements of type `T`. * @param batch_size the desired size of the batch. * @param stream the ordering for the host->device copies, if applicable. * @param mr a custom memory resource for the intermediate buffer, if applicable. */ batch_load_iterator(const T* source, size_type n_rows, size_type row_width, size_type batch_size, rmm::cuda_stream_view stream, rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource()) : cur_batch_(new batch(source, n_rows, row_width, batch_size, stream, mr)), cur_pos_(0) { } /** * Whether this iterator copies the data on every iteration * (i.e. the source is inaccessible from the device). */ [[nodiscard]] auto does_copy() const -> bool { return cur_batch_->does_copy(); } /** Reset the iterator position to `begin()` */ void reset() { cur_pos_ = 0; } /** Reset the iterator position to `end()` */ void reset_to_end() { cur_pos_ = cur_batch_->n_iters_; } [[nodiscard]] auto begin() const -> const batch_load_iterator<T> { batch_load_iterator<T> x(*this); x.reset(); return x; } [[nodiscard]] auto end() const -> const batch_load_iterator<T> { batch_load_iterator<T> x(*this); x.reset_to_end(); return x; } [[nodiscard]] auto operator*() const -> reference { cur_batch_->load(cur_pos_); return *cur_batch_; } [[nodiscard]] auto operator->() const -> pointer { cur_batch_->load(cur_pos_); return cur_batch_.get(); } friend auto operator==(const batch_load_iterator<T>& x, const batch_load_iterator<T>& y) -> bool { return x.cur_batch_ == y.cur_batch_ && x.cur_pos_ == y.cur_pos_; }; friend auto operator!=(const batch_load_iterator<T>& x, const batch_load_iterator<T>& y) -> bool { return x.cur_batch_ != y.cur_batch_ || x.cur_pos_ != y.cur_pos_; }; auto operator++() -> batch_load_iterator<T>& { ++cur_pos_; return *this; } auto operator++(int) -> batch_load_iterator<T> { batch_load_iterator<T> x(*this); ++cur_pos_; return x; } auto operator--() -> batch_load_iterator<T>& { --cur_pos_; return *this; } auto operator--(int) -> batch_load_iterator<T> { batch_load_iterator<T> x(*this); --cur_pos_; return x; } private: std::shared_ptr<value_type> cur_batch_; size_type cur_pos_; }; } // namespace raft::spatial::knn::detail::utils
0
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail/ball_cover.cuh
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/core/resource/cuda_stream.hpp> #include <raft/core/resource/thrust_policy.hpp> #include <raft/core/resources.hpp> #include "../ball_cover_types.hpp" #include "ball_cover/common.cuh" #include "ball_cover/registers.cuh" #include "haversine_distance.cuh" #include <cstdint> #include <limits.h> #include <raft/util/cuda_utils.cuh> #include <raft/neighbors/detail/faiss_select/key_value_block_select.cuh> #include <raft/matrix/copy.cuh> #include <raft/neighbors/brute_force.cuh> #include <raft/random/rng.cuh> #include <raft/sparse/convert/csr.cuh> #include <rmm/device_uvector.hpp> #include <rmm/exec_policy.hpp> #include <thrust/fill.h> #include <thrust/for_each.h> #include <thrust/functional.h> #include <thrust/iterator/counting_iterator.h> #include <thrust/iterator/zip_iterator.h> #include <thrust/reduce.h> #include <thrust/sequence.h> #include <thrust/sort.h> #include <thrust/tuple.h> namespace raft { namespace spatial { namespace knn { namespace detail { /** * Given a set of points in row-major order which are to be * used as a set of index points, uniformly samples a subset * of points to be used as landmarks. * @tparam value_idx * @tparam value_t * @param handle * @param index */ template <typename value_idx, typename value_t, typename value_int = std::uint32_t> void sample_landmarks(raft::resources const& handle, BallCoverIndex<value_idx, value_t, value_int>& index) { rmm::device_uvector<value_idx> R_1nn_cols2(index.n_landmarks, resource::get_cuda_stream(handle)); rmm::device_uvector<value_t> R_1nn_ones(index.m, resource::get_cuda_stream(handle)); rmm::device_uvector<value_idx> R_indices(index.n_landmarks, resource::get_cuda_stream(handle)); thrust::sequence(resource::get_thrust_policy(handle), index.get_R_1nn_cols().data_handle(), index.get_R_1nn_cols().data_handle() + index.m, (value_idx)0); thrust::fill(resource::get_thrust_policy(handle), R_1nn_ones.data(), R_1nn_ones.data() + R_1nn_ones.size(), 1.0); thrust::fill(resource::get_thrust_policy(handle), R_indices.data(), R_indices.data() + R_indices.size(), 0.0); /** * 1. Randomly sample sqrt(n) points from X */ raft::random::RngState rng_state(12345); raft::random::sampleWithoutReplacement(handle, rng_state, R_indices.data(), R_1nn_cols2.data(), index.get_R_1nn_cols().data_handle(), R_1nn_ones.data(), (value_idx)index.n_landmarks, (value_idx)index.m); // index.get_X() returns the wrong indextype (uint32_t where we need value_idx), so need to // create new device_matrix_view here auto x = index.get_X(); auto r = index.get_R(); raft::matrix::copy_rows<value_t, value_idx>( handle, make_device_matrix_view<const value_t, value_idx>(x.data_handle(), x.extent(0), x.extent(1)), make_device_matrix_view<value_t, value_idx>(r.data_handle(), r.extent(0), r.extent(1)), make_device_vector_view(R_1nn_cols2.data(), index.n_landmarks)); } /** * Constructs a 1-nn index mapping each landmark to their closest points. * @tparam value_idx * @tparam value_t * @param handle * @param R_knn_inds_ptr * @param R_knn_dists_ptr * @param k * @param index */ template <typename value_idx, typename value_t, typename value_int = std::uint32_t> void construct_landmark_1nn(raft::resources const& handle, const value_idx* R_knn_inds_ptr, const value_t* R_knn_dists_ptr, value_int k, BallCoverIndex<value_idx, value_t, value_int>& index) { rmm::device_uvector<value_idx> R_1nn_inds(index.m, resource::get_cuda_stream(handle)); thrust::fill(resource::get_thrust_policy(handle), R_1nn_inds.data(), R_1nn_inds.data() + index.m, std::numeric_limits<value_idx>::max()); value_idx* R_1nn_inds_ptr = R_1nn_inds.data(); value_t* R_1nn_dists_ptr = index.get_R_1nn_dists().data_handle(); auto idxs = thrust::make_counting_iterator<value_idx>(0); thrust::for_each( resource::get_thrust_policy(handle), idxs, idxs + index.m, [=] __device__(value_idx i) { R_1nn_inds_ptr[i] = R_knn_inds_ptr[i * k]; R_1nn_dists_ptr[i] = R_knn_dists_ptr[i * k]; }); auto keys = thrust::make_zip_iterator( thrust::make_tuple(R_1nn_inds.data(), index.get_R_1nn_dists().data_handle())); // group neighborhoods for each reference landmark and sort each group by distance thrust::sort_by_key(resource::get_thrust_policy(handle), keys, keys + index.m, index.get_R_1nn_cols().data_handle(), NNComp()); // convert to CSR for fast lookup raft::sparse::convert::sorted_coo_to_csr(R_1nn_inds.data(), index.m, index.get_R_indptr().data_handle(), index.n_landmarks + 1, resource::get_cuda_stream(handle)); } /** * Computes the k closest landmarks to a set of query points. * @tparam value_idx * @tparam value_t * @tparam value_int * @param handle * @param index * @param query_pts * @param n_query_pts * @param k * @param R_knn_inds * @param R_knn_dists */ template <typename value_idx, typename value_t, typename value_int = std::uint32_t> void k_closest_landmarks(raft::resources const& handle, const BallCoverIndex<value_idx, value_t, value_int>& index, const value_t* query_pts, value_int n_query_pts, value_int k, value_idx* R_knn_inds, value_t* R_knn_dists) { std::vector<raft::device_matrix_view<const value_t, value_int>> inputs = {index.get_R()}; raft::neighbors::brute_force::knn<value_idx, value_t, value_int>( handle, inputs, make_device_matrix_view(query_pts, n_query_pts, inputs[0].extent(1)), make_device_matrix_view(R_knn_inds, n_query_pts, k), make_device_matrix_view(R_knn_dists, n_query_pts, k), index.get_metric()); } /** * Uses the sorted data points in the 1-nn landmark index to compute * an array of radii for each landmark. * @tparam value_idx * @tparam value_t * @param handle * @param index */ template <typename value_idx, typename value_t, typename value_int = std::uint32_t> void compute_landmark_radii(raft::resources const& handle, BallCoverIndex<value_idx, value_t, value_int>& index) { auto entries = thrust::make_counting_iterator<value_idx>(0); const value_idx* R_indptr_ptr = index.get_R_indptr().data_handle(); const value_t* R_1nn_dists_ptr = index.get_R_1nn_dists().data_handle(); value_t* R_radius_ptr = index.get_R_radius().data_handle(); thrust::for_each(resource::get_thrust_policy(handle), entries, entries + index.n_landmarks, [=] __device__(value_idx input) { value_idx last_row_idx = R_indptr_ptr[input + 1] - 1; R_radius_ptr[input] = R_1nn_dists_ptr[last_row_idx]; }); } /** * 4. Perform k-select over original KNN, using L_r to filter distances * * a. Map 1 row to each warp/block * b. Add closest k R points to heap * c. Iterate through batches of R, having each thread in the warp load a set * of distances y from R (only if d(q, r) < 3 * distance to closest r) and * marking the distance to be computed between x, y only * if knn[k].distance >= d(x_i, R_k) + d(R_k, y) */ template <typename value_idx, typename value_t, typename value_int = std::uint32_t, typename dist_func> void perform_rbc_query(raft::resources const& handle, const BallCoverIndex<value_idx, value_t, value_int>& index, const value_t* query, value_int n_query_pts, std::uint32_t k, const value_idx* R_knn_inds, const value_t* R_knn_dists, dist_func dfunc, value_idx* inds, value_t* dists, value_int* dists_counter, value_int* post_dists_counter, float weight = 1.0, bool perform_post_filtering = true) { // initialize output inds and dists thrust::fill(resource::get_thrust_policy(handle), inds, inds + (k * n_query_pts), std::numeric_limits<value_idx>::max()); thrust::fill(resource::get_thrust_policy(handle), dists, dists + (k * n_query_pts), std::numeric_limits<value_t>::max()); if (index.n == 2) { // Compute nearest k for each neighborhood in each closest R rbc_low_dim_pass_one<value_idx, value_t, value_int, 2>(handle, index, query, n_query_pts, k, R_knn_inds, R_knn_dists, dfunc, inds, dists, weight, dists_counter); if (perform_post_filtering) { rbc_low_dim_pass_two<value_idx, value_t, value_int, 2>(handle, index, query, n_query_pts, k, R_knn_inds, R_knn_dists, dfunc, inds, dists, weight, post_dists_counter); } } else if (index.n == 3) { // Compute nearest k for each neighborhood in each closest R rbc_low_dim_pass_one<value_idx, value_t, value_int, 3>(handle, index, query, n_query_pts, k, R_knn_inds, R_knn_dists, dfunc, inds, dists, weight, dists_counter); if (perform_post_filtering) { rbc_low_dim_pass_two<value_idx, value_t, value_int, 3>(handle, index, query, n_query_pts, k, R_knn_inds, R_knn_dists, dfunc, inds, dists, weight, post_dists_counter); } } } /** * Similar to a ball tree, the random ball cover algorithm * uses the triangle inequality to prune distance computations * in any metric space with a guarantee of sqrt(n) * c^{3/2} * where `c` is an expansion constant based on the distance * metric. * * This function variant performs an all nearest neighbors * query which is useful for algorithms that need to perform * A * A.T. */ template <typename value_idx = std::int64_t, typename value_t, typename value_int = std::uint32_t, typename distance_func> void rbc_build_index(raft::resources const& handle, BallCoverIndex<value_idx, value_t, value_int>& index, distance_func dfunc) { ASSERT(index.n <= 3, "only 2d and 3d vectors are supported in current implementation"); ASSERT(!index.is_index_trained(), "index cannot be previously trained"); rmm::device_uvector<value_idx> R_knn_inds(index.m, resource::get_cuda_stream(handle)); // Initialize the uvectors thrust::fill(resource::get_thrust_policy(handle), R_knn_inds.begin(), R_knn_inds.end(), std::numeric_limits<value_idx>::max()); thrust::fill(resource::get_thrust_policy(handle), index.get_R_closest_landmark_dists().data_handle(), index.get_R_closest_landmark_dists().data_handle() + index.m, std::numeric_limits<value_t>::max()); /** * 1. Randomly sample sqrt(n) points from X */ sample_landmarks<value_idx, value_t>(handle, index); /** * 2. Perform knn = bfknn(X, R, k) */ value_int k = 1; k_closest_landmarks(handle, index, index.get_X().data_handle(), index.m, k, R_knn_inds.data(), index.get_R_closest_landmark_dists().data_handle()); /** * 3. Create L_r = knn[:,0].T (CSR) * * Slice closest neighboring R * Secondary sort by (R_knn_inds, R_knn_dists) */ construct_landmark_1nn( handle, R_knn_inds.data(), index.get_R_closest_landmark_dists().data_handle(), k, index); /** * Compute radius of each R for filtering: p(q, r) <= p(q, q_r) + radius(r) * (need to take the */ compute_landmark_radii(handle, index); } /** * Performs an all neighbors knn query (e.g. index == query) */ template <typename value_idx = std::int64_t, typename value_t, typename value_int = std::uint32_t, typename distance_func> void rbc_all_knn_query(raft::resources const& handle, BallCoverIndex<value_idx, value_t, value_int>& index, value_int k, value_idx* inds, value_t* dists, distance_func dfunc, // approximate nn options bool perform_post_filtering = true, float weight = 1.0) { ASSERT(index.n <= 3, "only 2d and 3d vectors are supported in current implementation"); ASSERT(index.n_landmarks >= k, "number of landmark samples must be >= k"); ASSERT(!index.is_index_trained(), "index cannot be previously trained"); rmm::device_uvector<value_idx> R_knn_inds(k * index.m, resource::get_cuda_stream(handle)); rmm::device_uvector<value_t> R_knn_dists(k * index.m, resource::get_cuda_stream(handle)); // Initialize the uvectors thrust::fill(resource::get_thrust_policy(handle), R_knn_inds.begin(), R_knn_inds.end(), std::numeric_limits<value_idx>::max()); thrust::fill(resource::get_thrust_policy(handle), R_knn_dists.begin(), R_knn_dists.end(), std::numeric_limits<value_t>::max()); thrust::fill(resource::get_thrust_policy(handle), inds, inds + (k * index.m), std::numeric_limits<value_idx>::max()); thrust::fill(resource::get_thrust_policy(handle), dists, dists + (k * index.m), std::numeric_limits<value_t>::max()); // For debugging / verification. Remove before releasing rmm::device_uvector<value_int> dists_counter(index.m, resource::get_cuda_stream(handle)); rmm::device_uvector<value_int> post_dists_counter(index.m, resource::get_cuda_stream(handle)); sample_landmarks<value_idx, value_t>(handle, index); k_closest_landmarks( handle, index, index.get_X().data_handle(), index.m, k, R_knn_inds.data(), R_knn_dists.data()); construct_landmark_1nn(handle, R_knn_inds.data(), R_knn_dists.data(), k, index); compute_landmark_radii(handle, index); perform_rbc_query(handle, index, index.get_X().data_handle(), index.m, k, R_knn_inds.data(), R_knn_dists.data(), dfunc, inds, dists, dists_counter.data(), post_dists_counter.data(), weight, perform_post_filtering); } /** * Performs a knn query against an index. This assumes the index has * already been built. */ template <typename value_idx = std::int64_t, typename value_t, typename value_int = std::uint32_t, typename distance_func> void rbc_knn_query(raft::resources const& handle, const BallCoverIndex<value_idx, value_t, value_int>& index, value_int k, const value_t* query, value_int n_query_pts, value_idx* inds, value_t* dists, distance_func dfunc, // approximate nn options bool perform_post_filtering = true, float weight = 1.0) { ASSERT(index.n <= 3, "only 2d and 3d vectors are supported in current implementation"); ASSERT(index.n_landmarks >= k, "number of landmark samples must be >= k"); ASSERT(index.is_index_trained(), "index must be previously trained"); rmm::device_uvector<value_idx> R_knn_inds(k * n_query_pts, resource::get_cuda_stream(handle)); rmm::device_uvector<value_t> R_knn_dists(k * n_query_pts, resource::get_cuda_stream(handle)); // Initialize the uvectors thrust::fill(resource::get_thrust_policy(handle), R_knn_inds.begin(), R_knn_inds.end(), std::numeric_limits<value_idx>::max()); thrust::fill(resource::get_thrust_policy(handle), R_knn_dists.begin(), R_knn_dists.end(), std::numeric_limits<value_t>::max()); thrust::fill(resource::get_thrust_policy(handle), inds, inds + (k * n_query_pts), std::numeric_limits<value_idx>::max()); thrust::fill(resource::get_thrust_policy(handle), dists, dists + (k * n_query_pts), std::numeric_limits<value_t>::max()); k_closest_landmarks(handle, index, query, n_query_pts, k, R_knn_inds.data(), R_knn_dists.data()); // For debugging / verification. Remove before releasing rmm::device_uvector<value_int> dists_counter(index.m, resource::get_cuda_stream(handle)); rmm::device_uvector<value_int> post_dists_counter(index.m, resource::get_cuda_stream(handle)); thrust::fill(resource::get_thrust_policy(handle), post_dists_counter.data(), post_dists_counter.data() + post_dists_counter.size(), 0); thrust::fill(resource::get_thrust_policy(handle), dists_counter.data(), dists_counter.data() + dists_counter.size(), 0); perform_rbc_query(handle, index, query, n_query_pts, k, R_knn_inds.data(), R_knn_dists.data(), dfunc, inds, dists, dists_counter.data(), post_dists_counter.data(), weight, perform_post_filtering); } }; // namespace detail }; // namespace knn }; // namespace spatial }; // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail/ann_quantized.cuh
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include "../ann_common.h" #include "../ivf_flat.cuh" #include <raft/core/resource/cuda_stream.hpp> #include "processing.cuh" #include <raft/core/operators.hpp> #include <raft/util/cuda_utils.cuh> #include <raft/util/cudart_utils.hpp> #include <raft/distance/distance.cuh> #include <raft/distance/distance_types.hpp> #include <raft/label/classlabels.cuh> #include <raft/neighbors/ivf_pq.cuh> #include <raft/core/device_mdspan.hpp> #include <rmm/cuda_stream_view.hpp> #include <thrust/iterator/transform_iterator.h> namespace raft::spatial::knn::detail { template <typename T = float, typename IntType = int> void approx_knn_build_index(raft::resources const& handle, knnIndex* index, knnIndexParam* params, raft::distance::DistanceType metric, float metricArg, T* index_array, IntType n, IntType D) { auto stream = resource::get_cuda_stream(handle); index->metric = metric; index->metricArg = metricArg; if (dynamic_cast<const IVFParam*>(params)) { index->nprobe = dynamic_cast<const IVFParam*>(params)->nprobe; } auto ivf_ft_pams = dynamic_cast<IVFFlatParam*>(params); auto ivf_pq_pams = dynamic_cast<IVFPQParam*>(params); if constexpr (std::is_same_v<T, float>) { index->metric_processor = create_processor<float>(metric, n, D, 0, false, stream); // For cosine/correlation distance, the metric processor translates distance // to inner product via pre/post processing - pass the translated metric to // ANN index if (metric == raft::distance::DistanceType::CosineExpanded || metric == raft::distance::DistanceType::CorrelationExpanded) { metric = index->metric = raft::distance::DistanceType::InnerProduct; } } if constexpr (std::is_same_v<T, float>) { index->metric_processor->preprocess(index_array); } if (ivf_ft_pams) { auto new_params = from_legacy_index_params(*ivf_ft_pams, metric, metricArg); index->ivf_flat<T, int64_t>() = std::make_unique<const ivf_flat::index<T, int64_t>>( ivf_flat::build(handle, new_params, index_array, int64_t(n), D)); } else if (ivf_pq_pams) { neighbors::ivf_pq::index_params params; params.metric = metric; params.metric_arg = metricArg; params.n_lists = ivf_pq_pams->nlist; params.pq_bits = ivf_pq_pams->n_bits; params.pq_dim = ivf_pq_pams->M; // TODO: handle ivf_pq_pams.usePrecomputedTables ? auto index_view = raft::make_device_matrix_view<const T, int64_t>(index_array, n, D); index->ivf_pq = std::make_unique<const neighbors::ivf_pq::index<int64_t>>( neighbors::ivf_pq::build(handle, params, index_view)); } else { RAFT_FAIL("Unrecognized index type."); } if constexpr (std::is_same_v<T, float>) { index->metric_processor->revert(index_array); } } template <typename T = float, typename IntType = int> void approx_knn_search(raft::resources const& handle, float* distances, int64_t* indices, knnIndex* index, IntType k, T* query_array, IntType n) { if constexpr (std::is_same_v<T, float>) { index->metric_processor->preprocess(query_array); index->metric_processor->set_num_queries(k); } // search if (index->ivf_flat<T, int64_t>()) { ivf_flat::search_params params; params.n_probes = index->nprobe; ivf_flat::search( handle, params, *(index->ivf_flat<T, int64_t>()), query_array, n, k, indices, distances); } else if (index->ivf_pq) { neighbors::ivf_pq::search_params params; params.n_probes = index->nprobe; auto query_view = raft::make_device_matrix_view<const T, uint32_t>(query_array, n, index->ivf_pq->dim()); auto indices_view = raft::make_device_matrix_view<int64_t, uint32_t>(indices, n, k); auto distances_view = raft::make_device_matrix_view<float, uint32_t>(distances, n, k); neighbors::ivf_pq::search( handle, params, *index->ivf_pq, query_view, indices_view, distances_view); } else { RAFT_FAIL("The model is not trained"); } // revert changes to the query if constexpr (std::is_same_v<T, float>) { index->metric_processor->revert(query_array); } // perform post-processing to show the real distances if (index->metric == raft::distance::DistanceType::L2SqrtExpanded || index->metric == raft::distance::DistanceType::L2SqrtUnexpanded || index->metric == raft::distance::DistanceType::LpUnexpanded) { /** * post-processing */ float p = 0.5; // standard l2 if (index->metric == raft::distance::DistanceType::LpUnexpanded) p = 1.0 / index->metricArg; raft::linalg::unaryOp<float>( distances, distances, n * k, raft::pow_const_op<float>(p), resource::get_cuda_stream(handle)); } if constexpr (std::is_same_v<T, float>) { index->metric_processor->postprocess(distances); } } } // namespace raft::spatial::knn::detail
0
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail/processing.cuh
/* * Copyright (c) 2020-2022, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include "processing.hpp" #include <raft/core/operators.hpp> #include <raft/distance/distance_types.hpp> #include <raft/linalg/matrix_vector_op.cuh> #include <raft/linalg/norm.cuh> #include <raft/linalg/unary_op.cuh> #include <raft/stats/mean.cuh> #include <raft/stats/mean_center.cuh> #include <rmm/device_uvector.hpp> namespace raft { namespace spatial { namespace knn { template <typename math_t> class CosineMetricProcessor : public MetricProcessor<math_t> { protected: int k_; bool row_major_; size_t n_rows_; size_t n_cols_; cudaStream_t stream_; rmm::device_uvector<math_t> colsums_; public: CosineMetricProcessor(size_t n_rows, size_t n_cols, int k, bool row_major, cudaStream_t stream) : stream_(stream), colsums_(n_rows, stream), n_cols_(n_cols), n_rows_(n_rows), row_major_(row_major), k_(k) { } void preprocess(math_t* data) { raft::linalg::rowNorm(colsums_.data(), data, n_cols_, n_rows_, raft::linalg::NormType::L2Norm, row_major_, stream_, raft::sqrt_op{}); raft::linalg::matrixVectorOp( data, data, colsums_.data(), n_cols_, n_rows_, row_major_, false, raft::div_op{}, stream_); } void revert(math_t* data) { raft::linalg::matrixVectorOp( data, data, colsums_.data(), n_cols_, n_rows_, row_major_, false, raft::mul_op{}, stream_); } void postprocess(math_t* data) { raft::linalg::unaryOp( data, data, k_ * n_rows_, [] __device__(math_t in) { return 1 - in; }, stream_); } void set_num_queries(int k) override { k_ = k; } ~CosineMetricProcessor() = default; }; template <typename math_t> class CorrelationMetricProcessor : public CosineMetricProcessor<math_t> { using cosine = CosineMetricProcessor<math_t>; public: CorrelationMetricProcessor( size_t n_rows, size_t n_cols, int k, bool row_major, cudaStream_t stream) : CosineMetricProcessor<math_t>(n_rows, n_cols, k, row_major, stream), means_(n_rows, stream) { } void preprocess(math_t* data) { math_t normalizer_const = 1.0 / (math_t)cosine::n_cols_; raft::linalg::reduce(means_.data(), data, cosine::n_cols_, cosine::n_rows_, (math_t)0.0, cosine::row_major_, true, cosine::stream_); raft::linalg::unaryOp(means_.data(), means_.data(), cosine::n_rows_, raft::mul_const_op<math_t>(normalizer_const), cosine::stream_); raft::stats::meanCenter(data, data, means_.data(), cosine::n_cols_, cosine::n_rows_, cosine::row_major_, false, cosine::stream_); CosineMetricProcessor<math_t>::preprocess(data); } void revert(math_t* data) { CosineMetricProcessor<math_t>::revert(data); raft::stats::meanAdd(data, data, means_.data(), cosine::n_cols_, cosine::n_rows_, cosine::row_major_, false, cosine::stream_); } void postprocess(math_t* data) { CosineMetricProcessor<math_t>::postprocess(data); } ~CorrelationMetricProcessor() = default; rmm::device_uvector<math_t> means_; }; template <typename math_t> class DefaultMetricProcessor : public MetricProcessor<math_t> { public: void preprocess(math_t* data) {} void revert(math_t* data) {} void postprocess(math_t* data) {} ~DefaultMetricProcessor() = default; }; template <typename math_t> inline std::unique_ptr<MetricProcessor<math_t>> create_processor( distance::DistanceType metric, int n, int D, int k, bool rowMajorQuery, cudaStream_t userStream) { MetricProcessor<math_t>* mp = nullptr; switch (metric) { case distance::DistanceType::CosineExpanded: mp = new CosineMetricProcessor<math_t>(n, D, k, rowMajorQuery, userStream); break; case distance::DistanceType::CorrelationExpanded: mp = new CorrelationMetricProcessor<math_t>(n, D, k, rowMajorQuery, userStream); break; default: mp = new DefaultMetricProcessor<math_t>(); } return std::unique_ptr<MetricProcessor<math_t>>(mp); } // Currently only being used by floats template class MetricProcessor<float>; template class CosineMetricProcessor<float>; template class CorrelationMetricProcessor<float>; template class DefaultMetricProcessor<float>; } // namespace knn } // namespace spatial } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail/ball_cover/registers-ext.cuh
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include "../../ball_cover_types.hpp" // BallCoverIndex #include "registers_types.cuh" // DistFunc #include <cstdint> // uint32_t #include <raft/util/raft_explicit.hpp> //RAFT_EXPLICIT #if defined(RAFT_EXPLICIT_INSTANTIATE_ONLY) namespace raft::spatial::knn::detail { template <typename value_idx, typename value_t, typename value_int = std::uint32_t, int dims = 2, typename dist_func> void rbc_low_dim_pass_one(raft::resources const& handle, const BallCoverIndex<value_idx, value_t, value_int>& index, const value_t* query, const value_int n_query_rows, value_int k, const value_idx* R_knn_inds, const value_t* R_knn_dists, dist_func& dfunc, value_idx* inds, value_t* dists, float weight, value_int* dists_counter) RAFT_EXPLICIT; template <typename value_idx, typename value_t, typename value_int = std::uint32_t, int dims = 2, typename dist_func> void rbc_low_dim_pass_two(raft::resources const& handle, const BallCoverIndex<value_idx, value_t, value_int>& index, const value_t* query, const value_int n_query_rows, value_int k, const value_idx* R_knn_inds, const value_t* R_knn_dists, dist_func& dfunc, value_idx* inds, value_t* dists, float weight, value_int* post_dists_counter) RAFT_EXPLICIT; }; // namespace raft::spatial::knn::detail #endif // RAFT_EXPLICIT_INSTANTIATE_ONLY #define instantiate_raft_spatial_knn_detail_rbc_low_dim_pass_one( \ Mvalue_idx, Mvalue_t, Mvalue_int, Mdims, Mdist_func) \ extern template void \ raft::spatial::knn::detail::rbc_low_dim_pass_one<Mvalue_idx, Mvalue_t, Mvalue_int, Mdims>( \ raft::resources const& handle, \ const BallCoverIndex<Mvalue_idx, Mvalue_t, Mvalue_int>& index, \ const Mvalue_t* query, \ const Mvalue_int n_query_rows, \ Mvalue_int k, \ const Mvalue_idx* R_knn_inds, \ const Mvalue_t* R_knn_dists, \ Mdist_func<Mvalue_t, Mvalue_int>& dfunc, \ Mvalue_idx* inds, \ Mvalue_t* dists, \ float weight, \ Mvalue_int* dists_counter) #define instantiate_raft_spatial_knn_detail_rbc_low_dim_pass_two( \ Mvalue_idx, Mvalue_t, Mvalue_int, Mdims, Mdist_func) \ extern template void \ raft::spatial::knn::detail::rbc_low_dim_pass_two<Mvalue_idx, Mvalue_t, Mvalue_int, Mdims>( \ raft::resources const& handle, \ const BallCoverIndex<Mvalue_idx, Mvalue_t, Mvalue_int>& index, \ const Mvalue_t* query, \ const Mvalue_int n_query_rows, \ Mvalue_int k, \ const Mvalue_idx* R_knn_inds, \ const Mvalue_t* R_knn_dists, \ Mdist_func<Mvalue_t, Mvalue_int>& dfunc, \ Mvalue_idx* inds, \ Mvalue_t* dists, \ float weight, \ Mvalue_int* dists_counter) instantiate_raft_spatial_knn_detail_rbc_low_dim_pass_one( std::int64_t, float, std::uint32_t, 2, raft::spatial::knn::detail::HaversineFunc); instantiate_raft_spatial_knn_detail_rbc_low_dim_pass_one( std::int64_t, float, std::uint32_t, 3, raft::spatial::knn::detail::HaversineFunc); instantiate_raft_spatial_knn_detail_rbc_low_dim_pass_one( std::int64_t, float, std::uint32_t, 2, raft::spatial::knn::detail::EuclideanFunc); instantiate_raft_spatial_knn_detail_rbc_low_dim_pass_one( std::int64_t, float, std::uint32_t, 3, raft::spatial::knn::detail::EuclideanFunc); instantiate_raft_spatial_knn_detail_rbc_low_dim_pass_one( std::int64_t, float, std::uint32_t, 2, raft::spatial::knn::detail::DistFunc); instantiate_raft_spatial_knn_detail_rbc_low_dim_pass_one( std::int64_t, float, std::uint32_t, 3, raft::spatial::knn::detail::DistFunc); instantiate_raft_spatial_knn_detail_rbc_low_dim_pass_two( std::int64_t, float, std::uint32_t, 2, raft::spatial::knn::detail::HaversineFunc); instantiate_raft_spatial_knn_detail_rbc_low_dim_pass_two( std::int64_t, float, std::uint32_t, 3, raft::spatial::knn::detail::HaversineFunc); instantiate_raft_spatial_knn_detail_rbc_low_dim_pass_two( std::int64_t, float, std::uint32_t, 2, raft::spatial::knn::detail::EuclideanFunc); instantiate_raft_spatial_knn_detail_rbc_low_dim_pass_two( std::int64_t, float, std::uint32_t, 3, raft::spatial::knn::detail::EuclideanFunc); instantiate_raft_spatial_knn_detail_rbc_low_dim_pass_two( std::int64_t, float, std::uint32_t, 2, raft::spatial::knn::detail::DistFunc); instantiate_raft_spatial_knn_detail_rbc_low_dim_pass_two( std::int64_t, float, std::uint32_t, 3, raft::spatial::knn::detail::DistFunc); #undef instantiate_raft_spatial_knn_detail_rbc_low_dim_pass_two #undef instantiate_raft_spatial_knn_detail_rbc_low_dim_pass_one
0
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail/ball_cover/registers.cuh
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #ifndef RAFT_EXPLICIT_INSTANTIATE_ONLY #include "registers-inl.cuh" #endif #ifdef RAFT_COMPILED #include "registers-ext.cuh" #endif
0
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail/ball_cover/registers-inl.cuh
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include "common.cuh" #include <raft/core/resource/cuda_stream.hpp> #include <raft/core/resource/thrust_policy.hpp> #include "../../ball_cover_types.hpp" #include "../haversine_distance.cuh" #include "registers_types.cuh" // DistFunc #include <cstdint> #include <limits.h> #include <raft/neighbors/detail/faiss_select/key_value_block_select.cuh> #include <raft/util/cuda_utils.cuh> #include <thrust/fill.h> namespace raft { namespace spatial { namespace knn { namespace detail { /** * To find exact neighbors, we perform a post-processing stage * that filters out those points which might have neighbors outside * of their k closest landmarks. This is usually a very small portion * of the total points. * @tparam value_idx * @tparam value_t * @tparam value_int * @tparam tpb * @param X * @param n_cols * @param R_knn_inds * @param R_knn_dists * @param R_radius * @param landmarks * @param n_landmarks * @param bitset_size * @param k * @param output * @param weight */ template <typename value_idx, typename value_t, typename value_int = std::uint32_t, int col_q = 2, int tpb = 32, typename distance_func> RAFT_KERNEL perform_post_filter_registers(const value_t* X, value_int n_cols, const value_idx* R_knn_inds, const value_t* R_knn_dists, const value_t* R_radius, const value_t* landmarks, int n_landmarks, value_int bitset_size, value_int k, distance_func dfunc, std::uint32_t* output, float weight = 1.0) { // allocate array of size n_landmarks / 32 ints extern __shared__ std::uint32_t shared_mem[]; // Start with all bits on for (value_int i = threadIdx.x; i < bitset_size; i += tpb) { shared_mem[i] = 0xffffffff; } __syncthreads(); // TODO: Would it be faster to use L1 for this? value_t local_x_ptr[col_q]; for (value_int j = 0; j < n_cols; ++j) { local_x_ptr[j] = X[n_cols * blockIdx.x + j]; } value_t closest_R_dist = R_knn_dists[blockIdx.x * k + (k - 1)]; // zero out bits for closest k landmarks for (value_int j = threadIdx.x; j < k; j += tpb) { _zero_bit(shared_mem, (std::uint32_t)R_knn_inds[blockIdx.x * k + j]); } __syncthreads(); // Discard any landmarks where p(q, r) > p(q, r_q) + radius(r) // That is, the distance between the current point and the current // landmark is > the distance between the current point and // its closest landmark + the radius of the current landmark. for (value_int l = threadIdx.x; l < n_landmarks; l += tpb) { // compute p(q, r) value_t dist = dfunc(local_x_ptr, landmarks + (n_cols * l), n_cols); if (dist > weight * (closest_R_dist + R_radius[l]) || dist > 3 * closest_R_dist) { _zero_bit(shared_mem, l); } } __syncthreads(); /** * Output bitset */ for (value_int l = threadIdx.x; l < bitset_size; l += tpb) { output[blockIdx.x * bitset_size + l] = shared_mem[l]; } } /** * @tparam value_idx * @tparam value_t * @tparam value_int * @tparam bitset_type * @tparam warp_q number of registers to use per warp * @tparam thread_q number of registers to use within each thread * @tparam tpb number of threads per block * @param X * @param n_cols * @param bitset * @param bitset_size * @param R_knn_dists * @param R_indptr * @param R_1nn_inds * @param R_1nn_dists * @param knn_inds * @param knn_dists * @param n_landmarks * @param k * @param dist_counter */ template <typename value_idx, typename value_t, typename value_int = std::uint32_t, typename bitset_type = std::uint32_t, typename dist_func, int warp_q = 32, int thread_q = 2, int tpb = 128, int col_q = 2> RAFT_KERNEL compute_final_dists_registers(const value_t* X_index, const value_t* X, const value_int n_cols, bitset_type* bitset, value_int bitset_size, const value_t* R_closest_landmark_dists, const value_idx* R_indptr, const value_idx* R_1nn_inds, const value_t* R_1nn_dists, value_idx* knn_inds, value_t* knn_dists, value_int n_landmarks, value_int k, dist_func dfunc, value_int* dist_counter) { static constexpr int kNumWarps = tpb / WarpSize; __shared__ value_t shared_memK[kNumWarps * warp_q]; __shared__ KeyValuePair<value_t, value_idx> shared_memV[kNumWarps * warp_q]; const value_t* x_ptr = X + (n_cols * blockIdx.x); value_t local_x_ptr[col_q]; for (value_int j = 0; j < n_cols; ++j) { local_x_ptr[j] = x_ptr[j]; } using namespace raft::neighbors::detail::faiss_select; KeyValueBlockSelect<value_t, value_idx, false, Comparator<value_t>, warp_q, thread_q, tpb> heap( std::numeric_limits<value_t>::max(), std::numeric_limits<value_t>::max(), -1, shared_memK, shared_memV, k); const value_int n_k = Pow2<WarpSize>::roundDown(k); value_int i = threadIdx.x; for (; i < n_k; i += tpb) { value_idx ind = knn_inds[blockIdx.x * k + i]; heap.add(knn_dists[blockIdx.x * k + i], R_closest_landmark_dists[ind], ind); } if (i < k) { value_idx ind = knn_inds[blockIdx.x * k + i]; heap.addThreadQ(knn_dists[blockIdx.x * k + i], R_closest_landmark_dists[ind], ind); } heap.checkThreadQ(); for (value_int cur_R_ind = 0; cur_R_ind < n_landmarks; ++cur_R_ind) { // if cur R overlaps cur point's closest R, it could be a // candidate if (_get_val(bitset + (blockIdx.x * bitset_size), cur_R_ind)) { value_idx R_start_offset = R_indptr[cur_R_ind]; value_idx R_stop_offset = R_indptr[cur_R_ind + 1]; value_idx R_size = R_stop_offset - R_start_offset; // Loop through R's neighborhood in parallel // Round R_size to the nearest warp threads so they can // all be computing in parallel. const value_int limit = Pow2<WarpSize>::roundDown(R_size); i = threadIdx.x; for (; i < limit; i += tpb) { value_idx cur_candidate_ind = R_1nn_inds[R_start_offset + i]; value_t cur_candidate_dist = R_1nn_dists[R_start_offset + i]; value_t z = heap.warpKTopRDist == 0.00 ? 0.0 : (abs(heap.warpKTop - heap.warpKTopRDist) * abs(heap.warpKTopRDist - cur_candidate_dist) - heap.warpKTop * cur_candidate_dist) / heap.warpKTopRDist; z = isnan(z) || isinf(z) ? 0.0 : z; // If lower bound on distance could possibly be in // the closest k neighbors, compute it and add to k-select value_t dist = std::numeric_limits<value_t>::max(); if (z <= heap.warpKTop) { const value_t* y_ptr = X_index + (n_cols * cur_candidate_ind); value_t local_y_ptr[col_q]; for (value_int j = 0; j < n_cols; ++j) { local_y_ptr[j] = y_ptr[j]; } dist = dfunc(local_x_ptr, local_y_ptr, n_cols); } heap.add(dist, cur_candidate_dist, cur_candidate_ind); } // second round guarantees to be only a single warp. if (i < R_size) { value_idx cur_candidate_ind = R_1nn_inds[R_start_offset + i]; value_t cur_candidate_dist = R_1nn_dists[R_start_offset + i]; value_t z = heap.warpKTopRDist == 0.00 ? 0.0 : (abs(heap.warpKTop - heap.warpKTopRDist) * abs(heap.warpKTopRDist - cur_candidate_dist) - heap.warpKTop * cur_candidate_dist) / heap.warpKTopRDist; z = isnan(z) || isinf(z) ? 0.0 : z; // If lower bound on distance could possibly be in // the closest k neighbors, compute it and add to k-select value_t dist = std::numeric_limits<value_t>::max(); if (z <= heap.warpKTop) { const value_t* y_ptr = X_index + (n_cols * cur_candidate_ind); value_t local_y_ptr[col_q]; for (value_int j = 0; j < n_cols; ++j) { local_y_ptr[j] = y_ptr[j]; } dist = dfunc(local_x_ptr, local_y_ptr, n_cols); } heap.addThreadQ(dist, cur_candidate_dist, cur_candidate_ind); } heap.checkThreadQ(); } } heap.reduce(); for (value_int i = threadIdx.x; i < k; i += tpb) { knn_dists[blockIdx.x * k + i] = shared_memK[i]; knn_inds[blockIdx.x * k + i] = shared_memV[i].value; } } /** * Random ball cover kernel for n_dims == 2 * @tparam value_idx * @tparam value_t * @tparam warp_q * @tparam thread_q * @tparam tpb * @tparam value_idx * @tparam value_t * @param R_knn_inds * @param R_knn_dists * @param m * @param k * @param R_indptr * @param R_1nn_cols * @param R_1nn_dists */ template <typename value_idx = std::int64_t, typename value_t, int warp_q = 32, int thread_q = 2, int tpb = 128, int col_q = 2, typename value_int = std::uint32_t, typename distance_func> RAFT_KERNEL block_rbc_kernel_registers(const value_t* X_index, const value_t* X, value_int n_cols, // n_cols should be 2 or 3 dims const value_idx* R_knn_inds, const value_t* R_knn_dists, value_int m, value_int k, const value_idx* R_indptr, const value_idx* R_1nn_cols, const value_t* R_1nn_dists, value_idx* out_inds, value_t* out_dists, value_int* dist_counter, const value_t* R_radius, distance_func dfunc, float weight = 1.0) { static constexpr value_int kNumWarps = tpb / WarpSize; __shared__ value_t shared_memK[kNumWarps * warp_q]; __shared__ KeyValuePair<value_t, value_idx> shared_memV[kNumWarps * warp_q]; // TODO: Separate kernels for different widths: // 1. Very small (between 3 and 32) just use registers for columns of "blockIdx.x" // 2. Can fit comfortably in shared memory (32 to a few thousand?) // 3. Load each time individually. const value_t* x_ptr = X + (n_cols * blockIdx.x); // Use registers only for 2d or 3d value_t local_x_ptr[col_q]; for (value_int i = 0; i < n_cols; ++i) { local_x_ptr[i] = x_ptr[i]; } // Each warp works on 1 R using namespace raft::neighbors::detail::faiss_select; KeyValueBlockSelect<value_t, value_idx, false, Comparator<value_t>, warp_q, thread_q, tpb> heap( std::numeric_limits<value_t>::max(), std::numeric_limits<value_t>::max(), -1, shared_memK, shared_memV, k); value_t min_R_dist = R_knn_dists[blockIdx.x * k + (k - 1)]; value_int n_dists_computed = 0; /** * First add distances for k closest neighbors of R * to the heap */ // Start iterating through elements of each set from closest R elements, // determining if the distance could even potentially be in the heap. for (value_int cur_k = 0; cur_k < k; ++cur_k) { // index and distance to current blockIdx.x's closest landmark value_t cur_R_dist = R_knn_dists[blockIdx.x * k + cur_k]; value_idx cur_R_ind = R_knn_inds[blockIdx.x * k + cur_k]; // Equation (2) in Cayton's paper- prune out R's which are > 3 * p(q, r_q) if (cur_R_dist > weight * (min_R_dist + R_radius[cur_R_ind])) continue; if (cur_R_dist > 3 * min_R_dist) return; // The whole warp should iterate through the elements in the current R value_idx R_start_offset = R_indptr[cur_R_ind]; value_idx R_stop_offset = R_indptr[cur_R_ind + 1]; value_idx R_size = R_stop_offset - R_start_offset; value_int limit = Pow2<WarpSize>::roundDown(R_size); value_int i = threadIdx.x; for (; i < limit; i += tpb) { // Index and distance of current candidate's nearest landmark value_idx cur_candidate_ind = R_1nn_cols[R_start_offset + i]; value_t cur_candidate_dist = R_1nn_dists[R_start_offset + i]; // Take 2 landmarks l_1 and l_2 where l_1 is the furthest point in the heap // and l_2 is the current landmark R. s is the current data point and // t is the new candidate data point. We know that: // d(s, t) cannot possibly be any smaller than | d(s, l_1) - d(l_1, l_2) | * | d(l_1, l_2) - // d(l_2, t) | - d(s, l_1) * d(l_2, t) // Therefore, if d(s, t) >= d(s, l_1) from the computation above, we know that the distance to // the candidate point cannot possibly be in the nearest neighbors. However, if d(s, t) < d(s, // l_1) then we should compute the distance because it's possible it could be smaller. // value_t z = heap.warpKTopRDist == 0.00 ? 0.0 : (abs(heap.warpKTop - heap.warpKTopRDist) * abs(heap.warpKTopRDist - cur_candidate_dist) - heap.warpKTop * cur_candidate_dist) / heap.warpKTopRDist; z = isnan(z) || isinf(z) ? 0.0 : z; value_t dist = std::numeric_limits<value_t>::max(); if (z <= heap.warpKTop) { const value_t* y_ptr = X_index + (n_cols * cur_candidate_ind); value_t local_y_ptr[col_q]; for (value_int j = 0; j < n_cols; ++j) { local_y_ptr[j] = y_ptr[j]; } dist = dfunc(local_x_ptr, local_y_ptr, n_cols); ++n_dists_computed; } heap.add(dist, cur_candidate_dist, cur_candidate_ind); } if (i < R_size) { value_idx cur_candidate_ind = R_1nn_cols[R_start_offset + i]; value_t cur_candidate_dist = R_1nn_dists[R_start_offset + i]; value_t z = heap.warpKTopRDist == 0.0 ? 0.0 : (abs(heap.warpKTop - heap.warpKTopRDist) * abs(heap.warpKTopRDist - cur_candidate_dist) - heap.warpKTop * cur_candidate_dist) / heap.warpKTopRDist; z = isnan(z) || isinf(z) ? 0.0 : z; value_t dist = std::numeric_limits<value_t>::max(); if (z <= heap.warpKTop) { const value_t* y_ptr = X_index + (n_cols * cur_candidate_ind); value_t local_y_ptr[col_q]; for (value_int j = 0; j < n_cols; ++j) { local_y_ptr[j] = y_ptr[j]; } dist = dfunc(local_x_ptr, local_y_ptr, n_cols); ++n_dists_computed; } heap.addThreadQ(dist, cur_candidate_dist, cur_candidate_ind); } heap.checkThreadQ(); } heap.reduce(); for (int i = threadIdx.x; i < k; i += tpb) { out_dists[blockIdx.x * k + i] = shared_memK[i]; out_inds[blockIdx.x * k + i] = shared_memV[i].value; } } template <typename value_idx, typename value_t, typename value_int = std::uint32_t, int dims = 2, typename dist_func> void rbc_low_dim_pass_one(raft::resources const& handle, const BallCoverIndex<value_idx, value_t, value_int>& index, const value_t* query, const value_int n_query_rows, value_int k, const value_idx* R_knn_inds, const value_t* R_knn_dists, dist_func& dfunc, value_idx* inds, value_t* dists, float weight, value_int* dists_counter) { if (k <= 32) block_rbc_kernel_registers<value_idx, value_t, 32, 2, 128, dims, value_int> <<<n_query_rows, 128, 0, resource::get_cuda_stream(handle)>>>( index.get_X().data_handle(), query, index.n, R_knn_inds, R_knn_dists, index.m, k, index.get_R_indptr().data_handle(), index.get_R_1nn_cols().data_handle(), index.get_R_1nn_dists().data_handle(), inds, dists, dists_counter, index.get_R_radius().data_handle(), dfunc, weight); else if (k <= 64) block_rbc_kernel_registers<value_idx, value_t, 64, 3, 128, 2, value_int> <<<n_query_rows, 128, 0, resource::get_cuda_stream(handle)>>>( index.get_X().data_handle(), query, index.n, R_knn_inds, R_knn_dists, index.m, k, index.get_R_indptr().data_handle(), index.get_R_1nn_cols().data_handle(), index.get_R_1nn_dists().data_handle(), inds, dists, dists_counter, index.get_R_radius().data_handle(), dfunc, weight); else if (k <= 128) block_rbc_kernel_registers<value_idx, value_t, 128, 3, 128, dims, value_int> <<<n_query_rows, 128, 0, resource::get_cuda_stream(handle)>>>( index.get_X().data_handle(), query, index.n, R_knn_inds, R_knn_dists, index.m, k, index.get_R_indptr().data_handle(), index.get_R_1nn_cols().data_handle(), index.get_R_1nn_dists().data_handle(), inds, dists, dists_counter, index.get_R_radius().data_handle(), dfunc, weight); else if (k <= 256) block_rbc_kernel_registers<value_idx, value_t, 256, 4, 128, dims, value_int> <<<n_query_rows, 128, 0, resource::get_cuda_stream(handle)>>>( index.get_X().data_handle(), query, index.n, R_knn_inds, R_knn_dists, index.m, k, index.get_R_indptr().data_handle(), index.get_R_1nn_cols().data_handle(), index.get_R_1nn_dists().data_handle(), inds, dists, dists_counter, index.get_R_radius().data_handle(), dfunc, weight); else if (k <= 512) block_rbc_kernel_registers<value_idx, value_t, 512, 8, 64, dims, value_int> <<<n_query_rows, 64, 0, resource::get_cuda_stream(handle)>>>( index.get_X().data_handle(), query, index.n, R_knn_inds, R_knn_dists, index.m, k, index.get_R_indptr().data_handle(), index.get_R_1nn_cols().data_handle(), index.get_R_1nn_dists().data_handle(), inds, dists, dists_counter, index.get_R_radius().data_handle(), dfunc, weight); else if (k <= 1024) block_rbc_kernel_registers<value_idx, value_t, 1024, 8, 64, dims, value_int> <<<n_query_rows, 64, 0, resource::get_cuda_stream(handle)>>>( index.get_X().data_handle(), query, index.n, R_knn_inds, R_knn_dists, index.m, k, index.get_R_indptr().data_handle(), index.get_R_1nn_cols().data_handle(), index.get_R_1nn_dists().data_handle(), inds, dists, dists_counter, index.get_R_radius().data_handle(), dfunc, weight); } template <typename value_idx, typename value_t, typename value_int = std::uint32_t, int dims = 2, typename dist_func> void rbc_low_dim_pass_two(raft::resources const& handle, const BallCoverIndex<value_idx, value_t, value_int>& index, const value_t* query, const value_int n_query_rows, value_int k, const value_idx* R_knn_inds, const value_t* R_knn_dists, dist_func& dfunc, value_idx* inds, value_t* dists, float weight, value_int* post_dists_counter) { const value_int bitset_size = ceil(index.n_landmarks / 32.0); rmm::device_uvector<std::uint32_t> bitset(bitset_size * n_query_rows, resource::get_cuda_stream(handle)); thrust::fill( resource::get_thrust_policy(handle), bitset.data(), bitset.data() + bitset.size(), 0); perform_post_filter_registers<value_idx, value_t, value_int, dims, 128> <<<n_query_rows, 128, bitset_size * sizeof(std::uint32_t), resource::get_cuda_stream(handle)>>>( query, index.n, R_knn_inds, R_knn_dists, index.get_R_radius().data_handle(), index.get_R().data_handle(), index.n_landmarks, bitset_size, k, dfunc, bitset.data(), weight); if (k <= 32) compute_final_dists_registers<value_idx, value_t, value_int, std::uint32_t, dist_func, 32, 2, 128, dims> <<<n_query_rows, 128, 0, resource::get_cuda_stream(handle)>>>( index.get_X().data_handle(), query, index.n, bitset.data(), bitset_size, index.get_R_closest_landmark_dists().data_handle(), index.get_R_indptr().data_handle(), index.get_R_1nn_cols().data_handle(), index.get_R_1nn_dists().data_handle(), inds, dists, index.n_landmarks, k, dfunc, post_dists_counter); else if (k <= 64) compute_final_dists_registers<value_idx, value_t, value_int, std::uint32_t, dist_func, 64, 3, 128, dims> <<<n_query_rows, 128, 0, resource::get_cuda_stream(handle)>>>( index.get_X().data_handle(), query, index.n, bitset.data(), bitset_size, index.get_R_closest_landmark_dists().data_handle(), index.get_R_indptr().data_handle(), index.get_R_1nn_cols().data_handle(), index.get_R_1nn_dists().data_handle(), inds, dists, index.n_landmarks, k, dfunc, post_dists_counter); else if (k <= 128) compute_final_dists_registers<value_idx, value_t, value_int, std::uint32_t, dist_func, 128, 3, 128, dims> <<<n_query_rows, 128, 0, resource::get_cuda_stream(handle)>>>( index.get_X().data_handle(), query, index.n, bitset.data(), bitset_size, index.get_R_closest_landmark_dists().data_handle(), index.get_R_indptr().data_handle(), index.get_R_1nn_cols().data_handle(), index.get_R_1nn_dists().data_handle(), inds, dists, index.n_landmarks, k, dfunc, post_dists_counter); else if (k <= 256) compute_final_dists_registers<value_idx, value_t, value_int, std::uint32_t, dist_func, 256, 4, 128, dims> <<<n_query_rows, 128, 0, resource::get_cuda_stream(handle)>>>( index.get_X().data_handle(), query, index.n, bitset.data(), bitset_size, index.get_R_closest_landmark_dists().data_handle(), index.get_R_indptr().data_handle(), index.get_R_1nn_cols().data_handle(), index.get_R_1nn_dists().data_handle(), inds, dists, index.n_landmarks, k, dfunc, post_dists_counter); else if (k <= 512) compute_final_dists_registers<value_idx, value_t, value_int, std::uint32_t, dist_func, 512, 8, 64, dims><<<n_query_rows, 64, 0, resource::get_cuda_stream(handle)>>>( index.get_X().data_handle(), query, index.n, bitset.data(), bitset_size, index.get_R_closest_landmark_dists().data_handle(), index.get_R_indptr().data_handle(), index.get_R_1nn_cols().data_handle(), index.get_R_1nn_dists().data_handle(), inds, dists, index.n_landmarks, k, dfunc, post_dists_counter); else if (k <= 1024) compute_final_dists_registers<value_idx, value_t, value_int, std::uint32_t, dist_func, 1024, 8, 64, dims><<<n_query_rows, 64, 0, resource::get_cuda_stream(handle)>>>( index.get_X().data_handle(), query, index.n, bitset.data(), bitset_size, index.get_R_closest_landmark_dists().data_handle(), index.get_R_indptr().data_handle(), index.get_R_1nn_cols().data_handle(), index.get_R_1nn_dists().data_handle(), inds, dists, index.n_landmarks, k, dfunc, post_dists_counter); } }; // namespace detail }; // namespace knn }; // namespace spatial }; // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail/ball_cover/common.cuh
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include "../haversine_distance.cuh" #include "registers_types.cuh" #include <cstdint> #include <thrust/functional.h> #include <thrust/tuple.h> namespace raft { namespace spatial { namespace knn { namespace detail { struct NNComp { template <typename one, typename two> __host__ __device__ bool operator()(const one& t1, const two& t2) { // sort first by each sample's reference landmark, if (thrust::get<0>(t1) < thrust::get<0>(t2)) return true; if (thrust::get<0>(t1) > thrust::get<0>(t2)) return false; // then by closest neighbor, return thrust::get<1>(t1) < thrust::get<1>(t2); } }; /** * Zeros the bit at location h in a one-hot encoded 32-bit int array */ __device__ inline void _zero_bit(std::uint32_t* arr, std::uint32_t h) { int bit = h % 32; int idx = h / 32; std::uint32_t assumed; std::uint32_t old = arr[idx]; do { assumed = old; old = atomicCAS(arr + idx, assumed, assumed & ~(1 << bit)); } while (assumed != old); } /** * Returns whether or not bit at location h is nonzero in a one-hot * encoded 32-bit in array. */ __device__ inline bool _get_val(std::uint32_t* arr, std::uint32_t h) { int bit = h % 32; int idx = h / 32; return (arr[idx] & (1 << bit)) > 0; } }; // namespace detail }; // namespace knn }; // namespace spatial }; // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/detail/ball_cover/registers_types.cuh
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include "../haversine_distance.cuh" // compute_haversine #include <cstdint> // uint32_t namespace raft { namespace spatial { namespace knn { namespace detail { template <typename value_t, typename value_int = std::uint32_t> struct DistFunc { virtual __device__ __host__ __forceinline__ value_t operator()(const value_t* a, const value_t* b, const value_int n_dims) { return -1; }; }; template <typename value_t, typename value_int = std::uint32_t> struct HaversineFunc : public DistFunc<value_t, value_int> { __device__ __host__ __forceinline__ value_t operator()(const value_t* a, const value_t* b, const value_int n_dims) override { return raft::spatial::knn::detail::compute_haversine(a[0], b[0], a[1], b[1]); } }; template <typename value_t, typename value_int = std::uint32_t> struct EuclideanFunc : public DistFunc<value_t, value_int> { __device__ __host__ __forceinline__ value_t operator()(const value_t* a, const value_t* b, const value_int n_dims) override { value_t sum_sq = 0; for (value_int i = 0; i < n_dims; ++i) { value_t diff = a[i] - b[i]; sum_sq += diff * diff; } return raft::sqrt(sum_sq); } }; }; // namespace detail }; // namespace knn }; // namespace spatial }; // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn
rapidsai_public_repos/raft/cpp/include/raft/spatial/knn/specializations/knn.cuh
/* * Copyright (c) 2021-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #pragma message( \ __FILE__ \ " is deprecated and will be removed." \ " Including specializations is not necessary any more." \ " For more information, see: https://docs.rapids.ai/api/raft/nightly/using_libraft.html")
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/spectral/cluster_solvers.cuh
/* * Copyright (c) 2019-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #ifndef __CLUSTER_SOLVERS_H #define __CLUSTER_SOLVERS_H #pragma once #include <raft/cluster/kmeans.cuh> #include <raft/core/resource/thrust_policy.hpp> #include <utility> // for std::pair namespace raft { namespace spectral { using namespace matrix; // aggregate of control params for Eigen Solver: // template <typename index_type_t, typename value_type_t, typename size_type_t = index_type_t> struct cluster_solver_config_t { size_type_t n_clusters; size_type_t maxIter; value_type_t tol; unsigned long long seed{123456}; }; template <typename index_type_t, typename value_type_t, typename size_type_t = index_type_t> struct kmeans_solver_t { explicit kmeans_solver_t( cluster_solver_config_t<index_type_t, value_type_t, size_type_t> const& config) : config_(config) { } std::pair<value_type_t, index_type_t> solve(raft::resources const& handle, size_type_t n_obs_vecs, size_type_t dim, value_type_t const* __restrict__ obs, index_type_t* __restrict__ codes) const { RAFT_EXPECTS(obs != nullptr, "Null obs buffer."); RAFT_EXPECTS(codes != nullptr, "Null codes buffer."); value_type_t residual{}; index_type_t iters{}; raft::cluster::KMeansParams km_params; km_params.n_clusters = config_.n_clusters; km_params.tol = config_.tol; km_params.max_iter = config_.maxIter; km_params.rng_state.seed = config_.seed; auto X = raft::make_device_matrix_view<const value_type_t>(obs, n_obs_vecs, dim); auto labels = raft::make_device_vector_view<index_type_t>(codes, n_obs_vecs); auto centroids = raft::make_device_matrix<value_type_t, index_type_t>(handle, config_.n_clusters, dim); auto weight = raft::make_device_vector<value_type_t, index_type_t>(handle, n_obs_vecs); thrust::fill(resource::get_thrust_policy(handle), weight.data_handle(), weight.data_handle() + n_obs_vecs, 1); auto sw = std::make_optional((raft::device_vector_view<const value_type_t>)weight.view()); raft::cluster::kmeans_fit_predict<value_type_t, index_type_t>( handle, km_params, X, sw, centroids.view(), labels, raft::make_host_scalar_view(&residual), raft::make_host_scalar_view(&iters)); return std::make_pair(residual, iters); } auto const& get_config(void) const { return config_; } private: cluster_solver_config_t<index_type_t, value_type_t, size_type_t> config_; }; } // namespace spectral } // namespace raft #endif
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/spectral/eigen_solvers.cuh
/* * Copyright (c) 2019-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #ifndef __EIGEN_SOLVERS_H #define __EIGEN_SOLVERS_H #pragma once #include <raft/sparse/solver/lanczos.cuh> #include <raft/spectral/matrix_wrappers.hpp> namespace raft { namespace spectral { // aggregate of control params for Eigen Solver: // template <typename index_type_t, typename value_type_t, typename size_type_t = index_type_t> struct eigen_solver_config_t { size_type_t n_eigVecs; size_type_t maxIter; size_type_t restartIter; value_type_t tol; bool reorthogonalize{false}; unsigned long long seed{ 1234567}; // CAVEAT: this default value is now common to all instances of using seed in // Lanczos; was not the case before: there were places where a default seed = 123456 // was used; this may trigger slightly different # solver iterations }; template <typename index_type_t, typename value_type_t, typename size_type_t = index_type_t> struct lanczos_solver_t { explicit lanczos_solver_t( eigen_solver_config_t<index_type_t, value_type_t, size_type_t> const& config) : config_(config) { } index_type_t solve_smallest_eigenvectors( raft::resources const& handle, matrix::sparse_matrix_t<index_type_t, value_type_t> const& A, value_type_t* __restrict__ eigVals, value_type_t* __restrict__ eigVecs) const { RAFT_EXPECTS(eigVals != nullptr, "Null eigVals buffer."); RAFT_EXPECTS(eigVecs != nullptr, "Null eigVecs buffer."); index_type_t iters{}; sparse::solver::computeSmallestEigenvectors(handle, A, config_.n_eigVecs, config_.maxIter, config_.restartIter, config_.tol, config_.reorthogonalize, iters, eigVals, eigVecs, config_.seed); return iters; } index_type_t solve_largest_eigenvectors( raft::resources const& handle, matrix::sparse_matrix_t<index_type_t, value_type_t> const& A, value_type_t* __restrict__ eigVals, value_type_t* __restrict__ eigVecs) const { RAFT_EXPECTS(eigVals != nullptr, "Null eigVals buffer."); RAFT_EXPECTS(eigVecs != nullptr, "Null eigVecs buffer."); index_type_t iters{}; sparse::solver::computeLargestEigenvectors(handle, A, config_.n_eigVecs, config_.maxIter, config_.restartIter, config_.tol, config_.reorthogonalize, iters, eigVals, eigVecs, config_.seed); return iters; } auto const& get_config(void) const { return config_; } private: eigen_solver_config_t<index_type_t, value_type_t, size_type_t> config_; }; } // namespace spectral } // namespace raft #endif
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/spectral/matrix_wrappers.hpp
/* * Copyright (c) 2020-2022, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #include <raft/spectral/detail/matrix_wrappers.hpp> // ========================================================= // Useful macros // ========================================================= namespace raft { namespace spectral { namespace matrix { using size_type = int; // for now; TODO: move it in appropriate header // specifies type of algorithm used // for SpMv: // using detail::sparse_mv_alg_t; // Vector "view"-like aggregate for linear algebra purposes // using detail::vector_view_t; using detail::vector_t; using detail::sparse_matrix_t; using detail::laplacian_matrix_t; using detail::modularity_matrix_t; } // namespace matrix } // namespace spectral } // namespace raft
0
rapidsai_public_repos/raft/cpp/include/raft
rapidsai_public_repos/raft/cpp/include/raft/spectral/specializations.cuh
/* * Copyright (c) 2022-2023, NVIDIA CORPORATION. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #pragma once #pragma message( \ __FILE__ \ " is deprecated and will be removed." \ " Including specializations is not necessary any more." \ " For more information, see: https://docs.rapids.ai/api/raft/nightly/using_libraft.html")
0