DNDS Namespace Reference

the host side operators are provided as implemented More...

Namespaces
namespace	CFV
namespace	Config
	Namespace for config field tag kwargs.
namespace	Debug
namespace	detail
namespace	Direct
namespace	ETensor
namespace	Euler
namespace	EulerP
	Namespace for the EulerP alternative evaluator module with GPU support.
namespace	Geom
namespace	HardEigen
namespace	IdealGas
namespace	Linear
namespace	Meta
namespace	MPI
namespace	ODE
namespace	OMP
namespace	Scalar
namespace	Serializer
namespace	TermColor

Classes
class	AdjacencyRow
	Non-owning span (pointer, size) into an ArrayAdjacency row. More...
class	Array
	Core 2D variable-length array container, the storage foundation of DNDSR. More...
struct	array_hash
	Hash functor for std::array<T, s>, combining element hashes via XOR. More...
class	ArrayAdjacency
	Mesh-connectivity array: ParArray<index> whose operator[] yields an AdjacencyRow typed view. More...
class	ArrayAdjacencyDeviceView
	Device-callable adjacency view: extends ArrayDeviceView so that indexing into a row returns an AdjacencyRow of indices. More...
class	ArrayDeviceView
	Non-owning device-callable view of an Array, specialised per DeviceBackend. More...
class	ArrayDeviceView< DeviceBackend::Host, T, _row_size, _row_max, _align >
	Host specialisation of ArrayDeviceView; identical to the host-side ArrayView. More...
class	ArrayDof
	Primary solver state container: an ArrayEigenMatrix pair with MPI-collective vector-space operations. More...
class	ArrayDofDeviceView
	Mutable device view of an ArrayDof father/son pair. More...
class	ArrayDofDeviceViewConst
	Const device view of an ArrayDof father/son pair. More...
class	ArrayDofOp
class	ArrayDofOp< DeviceBackend::Host, n_m, n_n >
	Host-side static dispatcher: implements every vector-space operation declared in DNDS_ARRAY_DOF_OP_FUNC_LIST for CPU execution. More...
class	ArrayDofSinglePack
	Helper for constructing a single ArrayDof with father, son, and transformer. More...
class	ArrayEigenMatrix
	ParArray<real> whose operator[] returns an Eigen::Map<Matrix<real, Ni, Nj>>. More...
class	ArrayEigenMatrixBatch
	CSR array storing a variable-sized batch of Eigen matrices per row. More...
class	ArrayEigenMatrixBatchDeviceView
class	ArrayEigenMatrixDeviceView
	Device-callable view onto ArrayEigenMatrix rows. More...
class	ArrayEigenUniMatrixBatch
	CSR array whose rows store a batch of identically-sized Eigen matrices. More...
class	ArrayEigenUniMatrixBatchDeviceView
	Device-callable view onto ArrayEigenUniMatrixBatch rows. More...
class	ArrayEigenVector
	ParArray<real, N> whose operator[] returns an Eigen::Map<Vector>. More...
class	ArrayEigenVectorDeviceView
	Device-callable view onto ArrayEigenVector rows. More...
class	ArrayIndex
	ArrayAdjacency<1> specialisation with an integer-typed operator[] (returns the single index by reference instead of a row wrapper). More...
class	ArrayIteratorBase
	CRTP base for row-granularity iterators over an Array / ArrayView. More...
class	ArrayLayout
	Compile-time layout descriptor deducing the concrete DataLayout from element type and row-size template arguments. More...
struct	ArrayPair
	Convenience bundle of a father, son, and attached ArrayTransformer. More...
struct	ArrayPairDeviceView
	Mutable device view onto an ArrayPair (for CUDA kernels). More...
struct	ArrayPairDeviceView_Base
	CRTP base implementing the unified-index accessors shared by ArrayPairDeviceView and ArrayPairDeviceViewConst. More...
struct	ArrayPairDeviceViewConst
	Const device view of a father-son array pair. More...
class	ArrayTransformer
	Ghost-communication engine for a father / son ParArray pair. More...
struct	ArrayTransformerType
	Type trait computing the ArrayTransformer type for a given Array type. More...
class	ArrayView
	Non-owning, device-callable view onto an Array. More...
struct	CheckResult
	Result of a single validation check. More...
struct	ConfigContext
	Runtime context supplied to context-aware validation checks. More...
class	ConfigRegistry
	Per-type singleton registry of config field metadata and validation checks. More...
class	ConfigSectionBuilder
	Builder object passed to the user's registration function. More...
struct	ConfigTypeTagOf
struct	ConfigTypeTagOf< bool >
struct	ConfigTypeTagOf< nlohmann::ordered_json >
struct	ConfigTypeTagOf< std::array< T, N > >
struct	ConfigTypeTagOf< std::string >
struct	ConfigTypeTagOf< std::vector< T > >
struct	ConfigTypeTagOf< T, std::enable_if_t< detail::is_eigen_type< T >::value > >
struct	ConfigTypeTagOf< T, std::enable_if_t< std::is_enum_v< T > > >
	C++ enum types serialize as JSON strings via nlohmann, so map to Enum. More...
struct	ConfigTypeTagOf< T, std::enable_if_t< std::is_floating_point_v< T > > >
struct	ConfigTypeTagOf< T, std::enable_if_t< std::is_integral_v< T > &&!std::is_same_v< T, bool > > >
class	CsvLog
	Append-only CSV logger with automatic file rotation after a line-count limit. More...
class	data_vector_base
	CRTP base offering operator[] / at on top of a derived's data() and size() accessors. Used by both #host_device_vector_r1 and (implicitly) #host_device_vector_r0. More...
struct	device_storage_factory
	Factory functions for constructing DeviceStorageBase instances of a specific backend. Specialised per backend so that the concrete type creation can live in backend-specific translation units. More...
struct	device_storage_factory< DeviceBackend::Host >
class	DeviceHostSingleAllocationBase
	Abstract single-allocation owning byte buffer. More...
class	DeviceHostSingleAllocationDirect
	Concrete DeviceHostSingleAllocationBase using std::vector<uint8_t> for host memory and DeviceStorage for device memory. More...
class	DeviceStorage
	Compile-time-specialised storage class; one definition per DeviceBackend. More...
class	DeviceStorage< DeviceBackend::Host >
	how to resolve int-rowsize duplicate? More...
class	DeviceStorageBase
	Abstract interface to a byte buffer owned by a specific backend. More...
class	DeviceTransferable
	CRTP mixin giving a class uniform to_device / to_host / device / getDeviceArrayBytes methods. More...
struct	deviceViewVector
	Contiguous host_device_vector of non-owning views, mirrored on a device. More...
class	EigenMatrixView
struct	Empty
	Trivially-copyable empty placeholder type that accepts any assignment. More...
struct	EmptyNoDefault
	Empty placeholder type without a default constructor; accepts any assignment. More...
class	ExprtkWrapperEvaluator
	Thin RAII wrapper over the exprtk expression library. More...
struct	FieldMeta
	Descriptor for a single configuration field. More...
class	GlobalOffsetsMapping
	Table mapping rank-local row indices to the global index space. More...
struct	has_static_CommMult
	here are some reasons to upgrade to C++20... More...
struct	has_static_CommMult< T, std::void_t< decltype(T::CommMult())> >
struct	has_static_CommType
	SFINAE trait detecting a static CommType member in T. More...
struct	has_static_CommType< T, std::void_t< decltype(T::CommType())> >
struct	host_device_vector_r0
	Legacy std::vector<T> subclass with an optional device mirror. More...
class	host_device_vector_r1
	Host + optional device vector of trivially copyable T. More...
class	HostDeviceEigenMatrix
struct	is_ssp
	Type trait that detects whether a type is a std::shared_ptr wrapping. More...
struct	is_ssp< ssp< T > >
struct	LogSimpleDIValue
	Storage-agnostic "either int or double" value with CSV-friendly streaming. More...
class	MatrixBatch
	Packed variable-shape matrix-batch layout inside a flat buffer. More...
struct	MemberPtr
	Pointer-to-member wrapper with a symbolic name; the pointer-based cousin of MemberRef used when the object is known only at visit time. More...
struct	MemberRef
	Simple {reference, name} bundle for a struct member. More...
class	MPIBufferHandler
	Process-singleton managing the buffer attached to MPI for MPI_Bsend (buffered sends). More...
struct	MPIInfo
	Lightweight bundle of an MPI communicator and the calling rank's coordinates. More...
struct	MPIReqHolder
	RAII vector of MPI_Requests that frees each non-null handle when destroyed. More...
struct	MPITypePairHolder
	RAII vector of (count, MPI_Datatype) pairs that frees every committed datatype when destroyed. More...
class	ObjectNaming
	Mixin base class providing a runtime instance name for tracing/debugging. More...
class	ObjectPool
	Generic object pool: caches unique_ptr<T> instances and hands them out with RAII return-on-destruction semantics. More...
struct	ObjName
	Tag type for naming objects created via make_ssp. More...
class	OffsetAscendIndexMapping
	Mapping between a rank's main data and its ghost copies. More...
class	ParArray
	MPI-aware Array: adds a communicator, rank, and global index mapping. More...
class	PerformanceTimer
	Process-wide singleton aggregating wall-clock timings by category. More...
class	ScalarStatistics
	Running-statistics accumulator using Welford's online algorithm. More...
class	ScopedValueAlternator
	RAII sentinel: on construction replaces ref with a new value and on destruction restores the original value. More...
class	vector_DeviceView
	Non-owning device-callable view {pointer, size} over a typed array. More...
struct	vector_hash
	Hash functor for std::vector<T>, combining element hashes via XOR. More...

Typedefs
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
using	tPy_Array = py_class_ssp< Array< T, _row_size, _row_max, _align > >
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
using	tPy_ParArray = py_class_ssp< ParArray< T, _row_size, _row_max, _align > >
template<class TArray >
using	tPy_ArrayTransformer = py_class_ssp< ArrayTransformerType_t< TArray > >
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
using	tPy_ParArrayPair = py_class_ssp< ArrayPair< ParArray< T, _row_size, _row_max, _align > > >
template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
using	tPy_ArrayAdjacency = py_class_ssp< ArrayAdjacency< _row_size, _row_max, _align > >
template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
using	tPy_ArrayAdjacencyPair = py_class_ssp< ArrayAdjacencyPair< _row_size, _row_max, _align > >
template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>
using	tPy_ArrayEigenMatrix = py_class_ssp< ArrayEigenMatrix< _mat_ni, _mat_nj, _mat_ni_max, _mat_nj_max, _align > >
template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>
using	tPy_ArrayEigenMatrixPair = py_class_ssp< ArrayEigenMatrixPair< _mat_ni, _mat_nj, _mat_ni_max, _mat_nj_max, _align > >
using	tPy_ArrayEigenMatrixBatch = py_class_ssp< ArrayEigenMatrixBatch >
using	tPy_ArrayEigenMatrixBatchPair = py_class_ssp< ArrayEigenMatrixBatchPair >
template<int _n_row, int _n_col>
using	tPy_ArrayEigenUniMatrixBatch = py_class_ssp< ArrayEigenUniMatrixBatch< _n_row, _n_col > >
template<int _n_row, int _n_col>
using	tPy_ArrayEigenUniMatrixBatchPair = py_class_ssp< ArrayEigenUniMatrixBatchPair< _n_row, _n_col > >
template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>
using	tPy_ArrayEigenVector = py_class_ssp< ArrayEigenVector< _vec_size, _row_max, _align > >
template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>
using	tPy_ArrayEigenVectorPair = py_class_ssp< ArrayEigenVectorPair< _vec_size, _row_max, _align > >
template<int n_m, int n_n>
using	tPy_ArrayDOF = py_class_ssp< ArrayDof< n_m, n_n > >
template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
using	ArrayAdjacencyPair = ArrayPair< ArrayAdjacency< _row_size, _row_max, _align > >
	ArrayPair alias for mesh adjacency (variable-width integer rows).
template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>
using	ArrayEigenVectorPair = ArrayPair< ArrayEigenVector< _vec_size, _row_max, _align > >
	ArrayPair alias for per-row Eigen vectors (e.g., node coords with N=3).
template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>
using	ArrayEigenMatrixPair = ArrayPair< ArrayEigenMatrix< _mat_ni, _mat_nj, _mat_ni_max, _mat_nj_max, _align > >
	ArrayPair alias for per-row Eigen matrices.
using	ArrayEigenMatrixBatchPair = ArrayPair< ArrayEigenMatrixBatch >
	ArrayPair alias for per-row variable-size Eigen matrix batches.
template<int _n_row, int _n_col>
using	ArrayEigenUniMatrixBatchPair = ArrayPair< ArrayEigenUniMatrixBatch< _n_row, _n_col > >
	ArrayPair alias for per-row batches of uniform _n_row x _n_col matrices.
using	t_pLGlobalMapping = ssp< GlobalOffsetsMapping >
	Shared pointer to a GlobalOffsetsMapping (globally replicated).
using	t_pLGhostMapping = ssp< OffsetAscendIndexMapping >
	Shared pointer to an OffsetAscendIndexMapping (per-rank ghost layout).
template<class TArray >
using	ArrayTransformerType_t = typename ArrayTransformerType< TArray >::Type
using	CrossFieldCheck = std::function< CheckResult(const void *obj)>
	A cross-field validation check that does not require runtime context.
using	ContextualCheck = std::function< CheckResult(const void *obj, const ConfigContext &ctx)>
	A cross-field validation check that requires runtime context.
using	tLogSimpleDIValueMap = std::map< std::string, LogSimpleDIValue >
	Convenience alias: the title -> value map type passed to CsvLog::WriteLine.
using	real = double
	Canonical floating-point scalar used throughout DNDSR (double precision).
using	index = int64_t
	Global row / DOF index type (signed 64-bit; handles multi-billion-cell meshes).
using	rowsize = int32_t
	Row-width / per-row element-count type (signed 32-bit).
using	real_sized_index = int64_t
	Integer type with the same width as real (used for type-punning / packing).
using	real_half_sized_index = int32_t
	Integer type with half the width of real.
template<typename T >
using	ssp = std::shared_ptr< T >
	Shortened alias for std::shared_ptr used pervasively in DNDSR.
template<typename T >
using	remove_cvref_t = std::remove_cv_t< std::remove_reference_t< T > >
	Convenience remove_cv + remove_reference composition (C++17 port of C++20's std::remove_cvref_t).
using	t_RowsizeVec = std::vector< rowsize >
	Vector of row widths (one rowsize per row).
using	t_IndexVec = std::vector< index >
	Vector of index values (global offsets, local ids, etc.).
using	t_pIndexVec = ssp< t_IndexVec >
	Shared pointer alias to t_IndexVec (used by mapping tables shared between arrays, see IndexMapping.hpp).
using	t_indexerPair = std::tuple< index, index >
	Paired indices, typically (start, size) or (first, last).
using	tDiFj = Eigen::Matrix< real, -1, -1, Eigen::RowMajor >
	Row-major dynamic Eigen matrix used by quadrature / basis tables.
using	MatrixXR = Eigen::Matrix< real, Eigen::Dynamic, Eigen::Dynamic >
	Column-major dynamic Eigen matrix of reals (default layout).
using	VectorXR = Eigen::Vector< real, Eigen::Dynamic >
	Dynamic Eigen vector of reals.
using	RowVectorXR = Eigen::RowVector< real, Eigen::Dynamic >
	Dynamic row-vector of reals.
template<class T >
using	py_class_ssp = py::classh< T >
using	t_supDeviceStorageBase = std::unique_ptr< DeviceStorageBase, std::function< void(DeviceStorageBase *)> >
	Owning unique pointer to a DeviceStorageBase with cross-DLL-safe deleter.
using	t_sspDeviceStorageBase = std::shared_ptr< DeviceStorageBase >
	Shared pointer equivalent of t_supDeviceStorageBase.
using	symbol_table_t = exprtk::symbol_table< real >
using	expression_t = exprtk::expression< real >
using	parser_t = exprtk::parser< real >
using	tExprVars = std::map< std::string, real >
	Scalar name -> value map fed into an ExprtkWrapperEvaluator.
using	tExprVarVecs = std::map< std::string, Eigen::Vector< real, Eigen::Dynamic > >
	Vector name -> Eigen dense vector map fed into an ExprtkWrapperEvaluator.
using	t_jsonconfig = nlohmann::ordered_json
	Project-wide JSON type alias: nlohmann/json with ordered keys.
using	MPI_int = int
	MPI counterpart type for MPI_int (= C int). Used for counts and ranks in MPI calls.
using	MPI_index = MPI_Aint
	MPI-compatible address/offset type (= MPI_Aint, 64-bit on all supported platforms). Used by the hindexed datatype machinery.
using	tMPI_sizeVec = std::vector< MPI_int >
	Vector of MPI counts.
using	tMPI_intVec = tMPI_sizeVec
	Alias for tMPI_sizeVec; used where the name "int vec" reads better.
using	tMPI_indexVec = std::vector< MPI_index >
	Vector of MPI_Aint byte-offsets for hindexed datatypes.
using	tMPI_AintVec = tMPI_indexVec
	Alias for tMPI_indexVec to match MPI_Aint terminology.
using	tMPI_statVec = std::vector< MPI_Status >
	Vector of MPI_Status, for MPI_Waitall / MPI_Testall.
using	tMPI_reqVec = std::vector< MPI_Request >
	Vector of MPI_Request, for persistent / nonblocking calls.
using	tMPI_typePairVec = std::vector< std::pair< MPI_int, MPI_Datatype > >
using	tpMPITypePairHolder = ssp< MPITypePairHolder >
	Shared-pointer alias to MPITypePairHolder.
template<class T >
using	host_device_vector = host_device_vector_r1< T >
	Primary public alias: host_device_vector<T> = #host_device_vector_r1<T>. Prefer this name throughout the code base.

Enumerations
enum	DataLayout {   ErrorLayout , TABLE_StaticFixed , TABLE_Fixed , TABLE_Max ,   TABLE_StaticMax , CSR }
	Enumeration of the five concrete data layouts supported by Array. More...
enum class	ConfigTypeTag {   Bool , Int , Real , String ,   Enum , Array , Object , ArrayOfObjects ,   MapOfObjects , Json }
	Enumerates the JSON Schema type associated with a config field. More...
enum class	DeviceBackend { Unknown = 0 , Host = 1 , Custom1 = 101 }
	Enumerates the backends a DeviceStorage / Array can live on. More...

Functions
void	pybind11_bind_Array_All (py::module_ m)
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
std::string	pybind11_Array_name_appends ()
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
std::string	pybind11_Array_name ()
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
std::string	pybind11_ParArray_name ()
template<class TArray >
std::string	pybind11_ArrayTransformer_name ()
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
std::string	pybind11_ParArrayPair_name ()
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
tPy_Array< T, _row_size, _row_max, _align >	pybind11_Array_declare (py::module_ &m)
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
tPy_Array< T, _row_size, _row_max, _align >	pybind11_Array_get_class (py::module_ &m)
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
void	pybind11_Array_define (py::module_ &m)
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
void	_pybind11_Array_define_dispatch (py::module_ &m)
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
tPy_ParArray< T, _row_size, _row_max, _align >	pybind11_ParArray_declare (py::module_ &m)
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
tPy_ParArray< T, _row_size, _row_max, _align >	pybind11_ParArray_get_class (py::module_ &m)
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
void	pybind11_ParArray_define (py::module_ &m)
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
void	_pybind11_ParArray_define_dispatch (py::module_ &m)
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
tPy_ParArrayPair< T, _row_size, _row_max, _align >	pybind11_ParArrayPair_declare (py::module_ &m)
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
tPy_ParArrayPair< T, _row_size, _row_max, _align >	pybind11_ParArrayPair_get_class (py::module_ &m)
template<class TPair , class TPy_Pair >
void	pybind11_ArrayPairGenericBindBasics (TPy_Pair &Pair_)
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
void	pybind11_ParArrayPair_define (py::module_ &m)
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
void	_pybind11_ParArrayPair_define_dispatch (py::module_ &m)
template<class TArray >
tPy_ArrayTransformer< TArray >	pybind11_ArrayTransformer_declare (py::module_ &m)
template<class TArray >
tPy_ArrayTransformer< TArray >	pybind11_ArrayTransformer_get_class (py::module_ &m)
template<class TArray >
void	pybind11_ArrayTransformer_define (py::module_ &m)
template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
void	_pybind11_ArrayTransformer_define_dispatch (py::module_ &m)
template<int offset = 0>
constexpr auto	_get_pybind11_arrayRowsizeInstantiationList ()
template<class T , size_t N, std::array< int, N > const & Arr, size_t... Is>
void	__pybind11_callBindArrays_rowsizes_sequence (py::module_ &m, std::index_sequence< Is... >)
template<class T , int offset = 0>
void	pybind11_callBindArrays_rowsizes (py::module_ &m)
template<class T , size_t N, std::array< int, N > const & Arr, size_t... Is>
void	__pybind11_callBindParArrays_rowsizes_sequence (py::module_ &m, std::index_sequence< Is... >)
template<class T , int offset = 0>
void	pybind11_callBindParArrays_rowsizes (py::module_ &m)
template<class T , size_t N, std::array< int, N > const & Arr, size_t... Is>
void	__pybind11_callBindArrayTransformers_rowsizes_sequence (py::module_ &m, std::index_sequence< Is... >)
template<class T , int offset = 0>
void	pybind11_callBindArrayTransformers_rowsizes (py::module_ &m)
template<class T , size_t N, std::array< int, N > const & Arr, size_t... Is>
void	__pybind11_callBindParArrayPairs_rowsizes_sequence (py::module_ &m, std::index_sequence< Is... >)
template<class T , int offset = 0>
void	pybind11_callBindParArrayPairs_rowsizes (py::module_ &m)
	pybind11_bind_Array_All_X_declare (1)
	pybind11_bind_Array_All_X_declare (2)
	pybind11_bind_Array_All_X_declare (3)
	pybind11_bind_Array_All_X_declare (4)
	pybind11_bind_Array_All_X_declare (5)
	pybind11_bind_Array_All_X_declare (6)
	pybind11_bind_Array_All_X_declare (7)
void	pybind11_bind_Array_Offsets (py::module_ m)
constexpr bool	isTABLE (DataLayout lo)
	Whether the layout uses a TABLE (padded) representation (vs CSR).
constexpr bool	isTABLE_Fixed (DataLayout lo)
	Whether the layout has a uniform row width (no per-row size needed).
constexpr bool	isTABLE_Max (DataLayout lo)
	Whether the layout is a padded-max variant (uses _pRowSizes).
constexpr bool	isTABLE_Static (DataLayout lo)
	Whether the layout has a compile-time row-size constant.
constexpr bool	isTABLE_Dynamic (DataLayout lo)
	Whether the layout carries a runtime row-size parameter.
template<class T >
constexpr bool	array_comp_acceptable ()
	Whether T is legal as an Array element type (trivially copyable or a fixed-size real Eigen matrix). Controls CUDA/MPI copy paths.
void	pybind11_bind_ArrayAdjacency_All (py::module_ &m)
template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
std::string	pybind11_ArrayAdjacency_name_appends ()
template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
std::string	pybind11_ArrayAdjacency_name ()
template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
std::string	pybind11_ArrayAdjacencyPair_name ()
template<class TArray = ArrayAdjacency<1>>
auto	pybind11_ArrayAdjacency_setitem (TArray &self, index index_, py::buffer row)
template<class TArray = ArrayAdjacency<1>>
auto	pybind11_ArrayAdjacency_getitem (TArray &self, index index_)
template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
tPy_ArrayAdjacency< _row_size, _row_max, _align >	pybind11_ArrayAdjacency_declare (py::module_ &m)
template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
tPy_ArrayAdjacency< _row_size, _row_max, _align >	pybind11_ArrayAdjacency_get_class (py::module_ &m)
template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
void	pybind11_ArrayAdjacency_define (py::module_ &m)
template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
void	_pybind11_ArrayAdjacency_define_dispatch (py::module_ &m)
template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
tPy_ArrayAdjacencyPair< _row_size, _row_max, _align >	pybind11_ArrayAdjacencyPair_declare (py::module_ &m)
template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
tPy_ArrayAdjacencyPair< _row_size, _row_max, _align >	pybind11_ArrayAdjacencyPair_get_class (py::module_ &m)
template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
void	pybind11_ArrayAdjacencyPair_define (py::module_ &m)
template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>
void	_pybind11_ArrayAdjacencyPair_define_dispatch (py::module_ &m)
template<size_t N, std::array< int, N > const & Arr, size_t... Is>
void	__pybind11_callBindArrayAdjacencys_rowsizes_sequence (py::module_ &m, std::index_sequence< Is... >)
void	pybind11_callBindArrayAdjacencys_rowsizes (py::module_ &m)
void	pybind11_bind_ArrayEigenMatrix_All (py::module_ &m)
template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>
std::string	pybind11_ArrayEigenMatrix_name_appends ()
template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>
std::string	pybind11_ArrayEigenMatrix_name ()
template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>
std::string	pybind11_ArrayEigenMatrixPair_name ()
template<class TArrayEigenMatrix = ArrayEigenMatrix<3, 3>>
auto	pybind11_ArrayEigenMatrix_getitem (TArrayEigenMatrix &self, index index_)
template<class TArrayEigenMatrix = ArrayEigenMatrix<3, 3>>
auto	pybind11_ArrayEigenMatrix_setitem (TArrayEigenMatrix &self, index index_, py::buffer row)
template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>
tPy_ArrayEigenMatrix< _mat_ni, _mat_nj, _mat_ni_max, _mat_nj_max, _align >	pybind11_ArrayEigenMatrix_declare (py::module_ &m)
template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>
tPy_ArrayEigenMatrix< _mat_ni, _mat_nj, _mat_ni_max, _mat_nj_max, _align >	pybind11_ArrayEigenMatrix_get_class (py::module_ &m)
template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>
void	pybind11_ArrayEigenMatrix_define (py::module_ &m)
template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>
void	_pybind11_ArrayEigenMatrix_define_dispatch (py::module_ &m)
template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>
tPy_ArrayEigenMatrixPair< _mat_ni, _mat_nj, _mat_ni_max, _mat_nj_max, _align >	pybind11_ArrayEigenMatrixPair_declare (py::module_ &m)
template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>
tPy_ArrayEigenMatrixPair< _mat_ni, _mat_nj, _mat_ni_max, _mat_nj_max, _align >	pybind11_ArrayEigenMatrixPair_get_class (py::module_ &m)
template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>
void	pybind11_ArrayEigenMatrixPair_define (py::module_ &m)
template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>
void	_pybind11_ArrayEigenMatrixPair_define_dispatch (py::module_ &m)
template<rowsize mat_n, size_t N, std::array< int, N > const & Arr, size_t... Is>
void	__pybind11_callBindArrayEigenMatrixs_rowsizes_sequence (py::module_ &m, std::index_sequence< Is... >)
template<rowsize mat_n>
void	pybind11_callBindArrayEigenMatrixs_rowsizes (py::module_ &m)
template void	pybind11_callBindArrayEigenMatrixs_rowsizes< 1 > (py::module_ &m)
template void	pybind11_callBindArrayEigenMatrixs_rowsizes< 2 > (py::module_ &m)
template void	pybind11_callBindArrayEigenMatrixs_rowsizes< 3 > (py::module_ &m)
template void	pybind11_callBindArrayEigenMatrixs_rowsizes< 4 > (py::module_ &m)
template void	pybind11_callBindArrayEigenMatrixs_rowsizes< 5 > (py::module_ &m)
template void	pybind11_callBindArrayEigenMatrixs_rowsizes< 6 > (py::module_ &m)
template void	pybind11_callBindArrayEigenMatrixs_rowsizes< 7 > (py::module_ &m)
template void	pybind11_callBindArrayEigenMatrixs_rowsizes< 8 > (py::module_ &m)
template void	pybind11_callBindArrayEigenMatrixs_rowsizes< DynamicSize > (py::module_ &m)
template void	pybind11_callBindArrayEigenMatrixs_rowsizes< NonUniformSize > (py::module_ &m)
template<rowsize _mat_ni, rowsize _mat_nj>
constexpr rowsize	__OneMatGetRowSize ()
	Compute the underlying per-row element count for an Ni x Nj matrix cell.
std::string	pybind11_ArrayEigenMatrixBatch_name ()
std::string	pybind11_ArrayEigenMatrixBatchPair_name ()
auto	pybind11_ArrayEigenMatrixBatch_setitem_row (ArrayEigenMatrixBatch &self, index i, const py::list &matList)
auto	pybind11_ArrayEigenMatrixBatch_getitem_row (ArrayEigenMatrixBatch &self, index index_)
auto	pybind11_ArrayEigenMatrixBatch_getitem (ArrayEigenMatrixBatch &self, std::tuple< index, rowsize > index_)
auto	pybind11_ArrayEigenMatrixBatch_setitem (ArrayEigenMatrixBatch &self, std::tuple< index, rowsize > index_, py::buffer row)
tPy_ArrayEigenMatrixBatch	pybind11_ArrayEigenMatrixBatch_declare (py::module_ &m)
tPy_ArrayEigenMatrixBatch	pybind11_ArrayEigenMatrixBatch_get_class (py::module_ &m)
void	pybind11_ArrayEigenMatrixBatch_define (py::module_ &m)
tPy_ArrayEigenMatrixBatchPair	pybind11_ArrayEigenMatrixBatchPair_declare (py::module_ &m)
tPy_ArrayEigenMatrixBatchPair	pybind11_ArrayEigenMatrixBatchPair_get_class (py::module_ &m)
void	pybind11_ArrayEigenMatrixBatchPair_define (py::module_ &m)
void	pybind11_bind_ArrayEigenMatrixBatch_All (py::module_ &m)
void	pybind11_bind_ArrayEigenUniMatrixBatch_All (py::module_ &m)
template<int _n_row, int _n_col>
std::string	pybind11_ArrayEigenUniMatrixBatch_name ()
template<int _n_row, int _n_col>
std::string	pybind11_ArrayEigenUniMatrixBatchPair_name ()
template<class TArrayEigenUniMatrixBatch = ArrayEigenUniMatrixBatch<3, 3>>
auto	pybind11_ArrayEigenUniMatrixBatch_getitem (TArrayEigenUniMatrixBatch &self, std::tuple< index, rowsize > index_)
template<class TArrayEigenUniMatrixBatch = ArrayEigenUniMatrixBatch<3, 3>>
auto	pybind11_ArrayEigenUniMatrixBatch_setitem (TArrayEigenUniMatrixBatch &self, std::tuple< index, rowsize > index_, py::buffer row)
template<class TArrayEigenUniMatrixBatch = ArrayEigenUniMatrixBatch<3, 3>>
auto	pybind11_ArrayEigenUniMatrixBatch_getitem_row (TArrayEigenUniMatrixBatch &self, index index_)
template<class TArrayEigenUniMatrixBatch = ArrayEigenUniMatrixBatch<3, 3>>
auto	pybind11_ArrayEigenUniMatrixBatch_setitem_row (TArrayEigenUniMatrixBatch &self, index index_, py::buffer row)
template<int _n_row, int _n_col>
tPy_ArrayEigenUniMatrixBatch< _n_row, _n_col >	pybind11_ArrayEigenUniMatrixBatch_declare (py::module_ &m)
template<int _n_row, int _n_col>
tPy_ArrayEigenUniMatrixBatch< _n_row, _n_col >	pybind11_ArrayEigenUniMatrixBatch_get_class (py::module_ &m)
template<int _n_row, int _n_col>
void	pybind11_ArrayEigenUniMatrixBatch_define (py::module_ &m)
template<int _n_row, int _n_col>
void	_pybind11_ArrayEigenUniMatrixBatch_define_dispatch (py::module_ &m)
template<int _n_row, int _n_col>
tPy_ArrayEigenUniMatrixBatchPair< _n_row, _n_col >	pybind11_ArrayEigenUniMatrixBatchPair_declare (py::module_ &m)
template<int _n_row, int _n_col>
tPy_ArrayEigenUniMatrixBatchPair< _n_row, _n_col >	pybind11_ArrayEigenUniMatrixBatchPair_get_class (py::module_ &m)
template<int _n_row, int _n_col>
void	pybind11_ArrayEigenUniMatrixBatchPair_define (py::module_ &m)
template<int _n_row, int _n_col>
void	_pybind11_ArrayEigenUniMatrixBatchPair_define_dispatch (py::module_ &m)
template<rowsize mat_n, size_t N, std::array< int, N > const & Arr, size_t... Is>
void	__pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes_sequence (py::module_ &m, std::index_sequence< Is... >)
template<rowsize mat_n>
void	pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes (py::module_ &m)
template void	pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 1 > (py::module_ &m)
template void	pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 2 > (py::module_ &m)
template void	pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 3 > (py::module_ &m)
template void	pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 4 > (py::module_ &m)
template void	pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 5 > (py::module_ &m)
template void	pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 6 > (py::module_ &m)
template void	pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 7 > (py::module_ &m)
template void	pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 8 > (py::module_ &m)
template void	pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< DynamicSize > (py::module_ &m)
template<int a, int b>
constexpr rowsize	EigenSize_Mul_RowSize ()
	a * b if both are compile-time sizes, DynamicSize if either is Eigen::Dynamic, otherwise a sentinel.
void	pybind11_bind_ArrayEigenVector_All (py::module_ &m)
template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>
std::string	pybind11_ArrayEigenVector_name_appends ()
template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>
std::string	pybind11_ArrayEigenVector_name ()
template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>
std::string	pybind11_ArrayEigenVectorPair_name ()
template<class TArrayEigenVector = ArrayEigenVector<1>>
auto	pybind11_ArrayEigenVector_getitem (TArrayEigenVector &self, index index_)
template<class TArrayEigenVector = ArrayEigenVector<1>>
auto	pybind11_ArrayEigenVector_setitem (TArrayEigenVector &self, index index_, py::buffer row)
template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>
tPy_ArrayEigenVector< _vec_size, _row_max, _align >	pybind11_ArrayEigenVector_declare (py::module_ &m)
template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>
tPy_ArrayEigenVector< _vec_size, _row_max, _align >	pybind11_ArrayEigenVector_get_class (py::module_ &m)
template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>
void	pybind11_ArrayEigenVector_define (py::module_ &m)
template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>
void	_pybind11_ArrayEigenVector_define_dispatch (py::module_ &m)
template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>
tPy_ArrayEigenVectorPair< _vec_size, _row_max, _align >	pybind11_ArrayEigenVectorPair_declare (py::module_ &m)
template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>
tPy_ArrayEigenVector< _vec_size, _row_max, _align >	pybind11_ArrayEigenVectorPair_get_class (py::module_ &m)
template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>
void	pybind11_ArrayEigenVectorPair_define (py::module_ &m)
template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>
void	_pybind11_ArrayEigenVectorPair_define_dispatch (py::module_ &m)
template<size_t N, std::array< int, N > const & Arr, size_t... Is>
void	__pybind11_callBindArrayEigenVectors_rowsizes_sequence (py::module_ &m, std::index_sequence< Is... >)
void	pybind11_callBindArrayEigenVectors_rowsizes (py::module_ &m)
void	pybind11_bind_ArrayDOF_All (py::module_ &m)
template<int n_m, int n_n>
std::string	pybind11_ArrayDOF_name ()
template<int n_m, int n_n>
tPy_ArrayDOF< n_m, n_n >	pybind11_ArrayDOF_declare (py::module_ &m)
template<int n_m, int n_n>
tPy_ArrayDOF< n_m, n_n >	pybind11_ArrayDOF_get_class (py::module_ &m)
template<int n_m, int n_n>
void	pybind11_ArrayDOF_define (py::module_ &m)
template<int n_m, int n_n>
void	_pybind11_ArrayDOF_define_dispatch (py::module_ &m)
template<rowsize mat_n, size_t N, std::array< int, N > const & Arr, size_t... Is>
void	__pybind11_callBindArrayDOFs_rowsizes_sequence (py::module_ &m, std::index_sequence< Is... >)
template<rowsize mat_n>
void	pybind11_callBindArrayDOF_rowsizes (py::module_ &m)
template void	pybind11_callBindArrayDOF_rowsizes< 1 > (py::module_ &m)
template void	pybind11_callBindArrayDOF_rowsizes< 2 > (py::module_ &m)
template void	pybind11_callBindArrayDOF_rowsizes< 3 > (py::module_ &m)
template void	pybind11_callBindArrayDOF_rowsizes< 4 > (py::module_ &m)
template void	pybind11_callBindArrayDOF_rowsizes< 5 > (py::module_ &m)
template void	pybind11_callBindArrayDOF_rowsizes< 6 > (py::module_ &m)
template void	pybind11_callBindArrayDOF_rowsizes< 7 > (py::module_ &m)
template void	pybind11_callBindArrayDOF_rowsizes< 8 > (py::module_ &m)
template void	pybind11_callBindArrayDOF_rowsizes< DynamicSize > (py::module_ &m)
template void	pybind11_callBindArrayDOF_rowsizes< NonUniformSize > (py::module_ &m)
std::vector< index >	BuildRedistributionPullingIndex (const MPIInfo &mpi, const std::vector< index > &readOrigIndex, const std::vector< index > &newOrigIndex, const ssp< GlobalOffsetsMapping > &readGlobalMapping)
	Redistributes ArrayPair data between different MPI partitions using ArrayTransformer.
template<class TArray >
void	RedistributeArrayWithTransformer (const MPIInfo &mpi, ssp< TArray > readFather, const std::vector< index > &readOrigIndex, const std::vector< index > &newOrigIndex, ssp< TArray > outFather)
	Redistributes an ArrayPair from a source partition to a target partition.
std::string	schemaTypeString (ConfigTypeTag tag)
bool	ostreamIsTTY (std::ostream &ostream)
	Heuristic detection of whether ostream is attached to a terminal.
std::ostream &	log ()
	Return the current DNDSR log stream (either std::cout or the installed file).
bool	logIsTTY ()
	Convenience: ostreamIsTTY applied to the current log() stream.
void	setLogStream (ssp< std::ostream > nstream)
	Redirect log() output to a user-supplied stream. Ownership is shared.
void	setLogStreamCout ()
	Restore the default std::cout routing for log().
int	get_terminal_width ()
	Terminal width in columns (falls back to a fixed default when not a TTY).
void	print_progress (std::ostream &os, double progress)
	Render a textual progress bar to os for progress in [0, 1].
std::string	getStringForceWString (const std::wstring &v)
	Convert a wstring to string (UTF-8 on Windows, byte-cast elsewhere).
int	get_env_OMP_NUM_THREADS ()
	Read OMP_NUM_THREADS env var, returning 1 if unset / invalid.
int	get_env_DNDS_DIST_OMP_NUM_THREADS ()
	Read the DNDSR-specific DNDS_DIST_OMP_NUM_THREADS override, falling back to get_env_OMP_NUM_THREADS.
std::string	GetSetVersionName (const std::string &ver="")
	Read/set the build version string accessible from code.
void	RegisterSignalHandler ()
	Install SEGV / ABRT handlers that print a backtrace via DNDS_signal_handler.
std::pair< index, index >	EvenSplitRange (int rank, int nRanks, index nGlobal)
	Split a global range [0, nGlobal) evenly among nRanks workers.
bool	IsUnInitReal (real v)
	Whether v equals the NaN sentinel UnInitReal (tested via isnan).
template<typename T , typename... Args>
ssp< T >	make_ssp (Args &&...args)
	Type-safe replacement for DNDS_MAKE_SSP. Creates ssp<T> with forwarded args.
template<typename T , typename... Args>
ssp< T >	make_ssp (ObjName objName, Args &&...args)
	Named variant of make_ssp. If T inherits ObjectNaming, sets the name.
constexpr int	RowSize_To_EigenSize (rowsize rs)
	Convert a rowsize constant to the corresponding Eigen compile-time size. Fixed >= 0 -> the value; DynamicSize / NonUniformSize -> Eigen::Dynamic.
std::string	RowSize_To_PySnippet (rowsize rs)
	Encode a rowsize constant as a short Python-binding snippet: "<number>" for fixed, "D" for DynamicSize, "I" for NonUniformSize. Used when generating pybind11 class names.
std::string	Align_To_PySnippet (rowsize al)
	Encode an alignment value as a Python-binding snippet: "N" for NoAlign, the number otherwise.
template<typename T >
bool	signedIntWillAddOverflow (T value, T increment)
	Overflow-detecting test for value + increment on signed integers.
template<typename T >
T	signedIntSafeAdd (T value, T increment)
	Add two signed integers, asserting on overflow instead of silently wrapping.
template<typename T >
T	size_t_to_signed (size_t v)
	Narrowing size_t -> T conversion with range check.
index	size_to_index (size_t v)
	Range-checked conversion from size_t to DNDS::index.
rowsize	size_to_rowsize (size_t v)
	Range-checked conversion from size_t to DNDS::rowsize.
template<class TtRowsizeVec , class TtIndexVec >
void	AccumulateRowSize (const TtRowsizeVec &rowsizes, TtIndexVec &rowstarts)
	Build a prefix-sum table from a row-size vector.
template<class T >
bool	checkUniformVector (const std::vector< T > &dat, T &value)
	Whether all elements of dat are equal; if so, stores the value into value.
template<class T , class TP = T>
void	PrintVec (const std::vector< T > &dat, std::ostream &out)
	Print a vector to out with outputDelim between elements.
template<class TL , class TR >
constexpr auto	divCeil (TL l, TR r)
	Integer ceiling division. l must be non-negative, r positive.
template<typename T >
constexpr T	sqr (const T &a)
	a * a, constexpr. Works for all arithmetic types.
template<typename T >
constexpr T	cube (const T &a)
	a * a * a, constexpr.
constexpr real	sign (real a)
	Signum function: +1, 0, or -1.
constexpr real	signTol (real a, real tol)
	Tolerant signum: returns 0 inside [-tol, tol].
constexpr real	signP (real a)
	"Signum, biased toward +1": treats 0 as positive.
constexpr real	signM (real a)
	"Signum, biased toward -1": treats 0 as negative.
template<typename T >
constexpr T	mod (T a, T b)
	Mathematical modulo that always returns a non-negative result. Unlike % in C++ where (-1) % 3 == -1, mod(-1, 3) == 2.
template<typename T >
constexpr T	divide_ceil (T a, T b)
	Integer ceiling division ceil(a / b). Correct for all signs.
real	float_mod (real a, real b)
	Floating-point modulo matching Python's % (result has sign of b).
template<class tIt1 , class tIt1end , class tIt2 , class tIt2end , class tF >
bool	iterateIdentical (tIt1 it1, tIt1end it1end, tIt2 it2, tIt2end it2end, tF F)
	Walk two ordered ranges in lockstep, calling F on each match.
int32_t	checkedIndexTo32 (index v)
	Narrow index to int32_t with range check; dies on overflow.
std::string	getStringForcePath (const std::filesystem::path::string_type &v)
	Portable conversion of a platform-native path string to std::string.
std::vector< std::string >	splitSString (const std::string &str, char delim)
std::vector< std::string >	splitSStringClean (const std::string &str, char delim)
bool	sstringHasSuffix (const std::string &str, const std::string &suffix)
template<class T >
bool	py_buffer_contains_T (const py::buffer_info &info)
bool	py_buffer_is_contigious_c (const py::buffer_info &info)
bool	py_buffer_is_contigious_f (const py::buffer_info &info)
std::tuple< ssize_t, char >	py_buffer_get_contigious_size (const py::buffer_info &info)
template<typename T >
py::memoryview	py_vector_as_memory_view (std::vector< T > &vec, bool readonly)
void	pybind11_bind_defines (py::module_ &m)
void	deviceStorageBase_deleter (DeviceStorageBase *p)
	Stateless deleter for DeviceStorageBase that works across shared-library boundaries where the vtable of unique_ptr's default deleter would not.
const char *	device_backend_name (DeviceBackend B)
	Canonical string name for a DeviceBackend (used in log messages).
DeviceBackend	device_backend_name_to_enum (std::string_view s)
	Inverse of device_backend_name. Returns Unknown for unrecognised names.
t_supDeviceStorageBase	null_supDeviceStorageBase ()
	Null-value helper for t_supDeviceStorageBase.
t_supDeviceStorageBase	device_storage_create (DeviceBackend backend, size_t n_bytes)
	Top-level factory: dispatches to the per-backend factory based on backend. Returns a null unique_ptr for DeviceBackendUnknown.
void	pybind11_bind_deviceStorage (py::module_ &m)
void	pybind11_bind_device_controls (py::module_ &m)
template<DeviceBackend B, class T , rowsize _row_size, rowsize _row_max, rowsize _align>
auto	ArrayDeviceView_build (index n_size, T *n_data, index n_data_size, const index *n_rowstart, index n_rowstart_size, const rowsize *n_rowsizes, index n_rowsizes_size, rowsize n_row_size_dynamic, T *n_data_device, const index *n_rowstart_device, const rowsize *n_rowsizes_device)
	Factory for ArrayDeviceView: selects host or device pointers based on B and forwards everything else.
template<class dir >
std::string	to_string (const Eigen::DenseBase< dir > &v, int precision=5, bool scientific=false)
	Render an Eigen::DenseBase to a string via operator<<.
std::string	getTraceString ()
	Return a symbolicated stack trace for the calling thread.
void	assert_false (const char *expr, const char *file, int line)
	Low-level: print a red "DNDS_assertion failed" line and abort.
void	assert_false_info (const char *expr, const char *file, int line, const std::string &info)
	Variant of assert_false that prints an extra info string.
void	assert_false_infof (const char *expr, const char *file, int line, const char *info,...)
	printf-style variant of assert_false. Used by DNDS_assert_infof.
template<class TException = std::runtime_error>
void	assert_false_info_throw (const char *expr, const char *file, int line, const std::string &info)
	Throwing variant of assert_false_info. Used by DNDS_check_throw.
hid_t	DNDS_H5T_INDEX ()
	HDF5 native datatype matching DNDS index (currently H5T_NATIVE_INT64).
hid_t	DNDS_H5T_ROWSIZE ()
	HDF5 native datatype matching DNDS rowsize (currently H5T_NATIVE_INT32).
hid_t	DNDS_H5T_REAL ()
	HDF5 native datatype matching DNDS real (currently H5T_NATIVE_DOUBLE).
auto	pybind11_GlobalOffsetsMapping_declare (py::module_ m)
auto	pybind11_GlobalOffsetsMapping_get_class (py::module_ m)
void	pybind11_GlobalOffsetsMapping_define (py::module_ m)
auto	pybind11_OffsetAscendIndexMapping_declare (py::module_ m)
auto	pybind11_OffsetAscendIndexMapping_get_class (py::module_ m)
void	pybind11_OffsetAscendIndexMapping_define (py::module_ m)
void	pybind11_bind_IndexMapping_All (py::module_ m)
Eigen::VectorXd	JsonGetEigenVector (const nlohmann::json &arr)
	Parse a JSON array into an Eigen::VectorXd. Throws a descriptive assertion on any JSON error.
Eigen::VectorFMTSafe< real, -1 >	JsonGetEigenVectorFMTSafe (const nlohmann::json &arr)
	Parse a JSON array into an Eigen::VectorFMTSafe (fixed-point-aware wrapper).
auto	EigenVectorGetJson (const Eigen::VectorXd &ve)
	Dump an Eigen::VectorXd into a JSON array of doubles.
auto	EigenVectorFMTSafeGetJson (const Eigen::VectorFMTSafe< real, -1 > &ve)
	Dump an Eigen::VectorFMTSafe into a JSON array of doubles.
void	to_json (nlohmann::ordered_json &j, const host_device_vector< real > &v)
void	from_json (const nlohmann::ordered_json &j, host_device_vector< real > &v)
void	assert_false_info_mpi (const char *expr, const char *file, int line, const std::string &info, const DNDS::MPIInfo &mpi)
	MPI-aware assertion-failure reporter.
std::string	getTimeStamp (const MPIInfo &mpi)
	Format a human-readable timestamp using the calling rank as context.
template<class Tbasic >
constexpr MPI_Datatype	__DNDSToMPITypeInt ()
	Map a DNDS integer type size to an MPI signed-integer datatype.
template<class Tbasic >
constexpr MPI_Datatype	__DNDSToMPITypeFloat ()
	Map a DNDS floating-point type size to an MPI datatype.
template<class T >
std::pair< MPI_Datatype, MPI_int >	BasicType_To_MPIIntType_Custom ()
	Dispatch to a user-provided CommPair / CommMult+ CommType pair on T.
template<class T >
std::pair< MPI_Datatype, MPI_int >	BasicType_To_MPIIntType ()
	Deduce an (MPI_Datatype, count) pair that represents a T value.
MPI_int	MPIWorldSize ()
	Convenience: MPI_Comm_size(MPI_COMM_WORLD).
MPI_int	MPIWorldRank ()
	Convenience: MPI_Comm_rank(MPI_COMM_WORLD).
template<class F >
void	MPISerialDo (const MPIInfo &mpi, F f)
	Execute f on each rank serially, in rank order.
void	InsertCheck (const MPIInfo &mpi, const std::string &info="", const std::string &FUNCTION="", const std::string &FILE="", int LINE=-1)
	Barrier + annotated print used by DNDS_MPI_InsertCheck.
void	pybind11_MPIInfo (py::module_ &m)
void	pybind11_bind_MPI_All (py::module_ &m)
MPI_Datatype	py_get_buffer_basic_mpi_datatype (const py::buffer_info &info)
MPI_Op	py_get_simple_mpi_op_by_name (const std::string &op)
template<typename TList , typename F >
void	for_each_member_list (TList &&obj_member_list, F &&f)
	Invoke f(member) for every element of a std::tuple of members.
template<typename T >
	MemberRef (T &, const char *) -> MemberRef< T >
template<typename Class , typename TList , typename F >
void	for_each_member_ptr_list (Class &obj, TList &&obj_member_ptr_list, F &&f)
	Invoke f(name, obj.*ptr) for every member in a list of MemberPtr.
template<typename TList , typename F >
void	for_each_member_ptr_list_raw (TList &&obj_member_ptr_list, F &&f)
	Low-level variant that passes each MemberPtr object through to f directly (for callers that need access to both ptr and name).
template<typename Class , typename T >
	MemberPtr (T Class::*, const char *) -> MemberPtr< Class, T >
PerformanceTimer &	Timer ()
	Short-hand accessor to the PerformanceTimer singleton.
template<class TBlkSiz , class TDisp , class TSizeof >
auto	optimize_hindexed_layout (index o_size, TBlkSiz *blk_sizes, TDisp *disps, TSizeof sizeofElem)
	Coalesce contiguous blocks in an MPI_Type_create_hindexed layout.

Variables
ssp< std::ostream >	logStream
	Shared output stream: where DNDS::log() writes progress / diagnostics.
bool	useCout = true
	Whether DNDS::log() is currently routed to std::cout.
template<typename T >
constexpr bool	is_ssp_v = is_ssp<T>::value
DNDS_CONSTANT const index	indexMin = INT64_MIN
	Minimum representable index value (= INT64_MIN).
DNDS_CONSTANT const real	UnInitReal = NAN
	Sentinel "not initialised" real value (NaN). Cheap to detect with std::isnan or IsUnInitReal; survives MPI transport unchanged.
DNDS_CONSTANT const index	UnInitIndex = INT64_MIN
	Sentinel "not initialised" index value (= INT64_MIN).
DNDS_CONSTANT const rowsize	UnInitRowsize = INT32_MIN
	Sentinel "not initialised" rowsize value (= INT32_MIN).
DNDS_CONSTANT const real	veryLargeReal = 3e200
	Catch-all upper bound ("practically infinity") for physical scalars.
DNDS_CONSTANT const real	largeReal = 3e10
	Loose upper bound (e.g., for non-dimensional limits).
DNDS_CONSTANT const real	verySmallReal = 1e-200
	Catch-all lower bound ("effectively zero").
DNDS_CONSTANT const real	smallReal = 1e-10
	Loose lower bound (for iterative-solver tolerances etc.).
DNDS_CONSTANT const real	pi = DNDS_E_PI
	π in double precision (matches DNDS_E_PI macro).
DNDS_CONSTANT const rowsize	DynamicSize = -1
	Template parameter flag: "row width is set at runtime but uniform".
DNDS_CONSTANT const rowsize	NonUniformSize = -2
	Template parameter flag: "each row has an independent width".
DNDS_CONSTANT const rowsize	NoAlign = -1024
	Alignment flag: no padding applied to rows (the only currently-supported value).
const char *	__EigenPCH = "EigenPCH"
const char *	__ExprtkPCH = "ExprtkPCH"
std::mutex	HDF_mutex
	Global mutex serialising host-side HDF5 calls.
const MPI_Datatype	DNDS_MPI_INDEX = __DNDSToMPITypeInt<index>()
	MPI datatype matching index (= MPI_INT64_T).
const MPI_Datatype	DNDS_MPI_REAL = __DNDSToMPITypeFloat<real>()
	MPI datatype matching real (= MPI_REAL8).

Detailed Description

the host side operators are provided as implemented

Typedef Documentation

ArrayAdjacencyPair

template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

using DNDS::ArrayAdjacencyPair = typedef ArrayPair<ArrayAdjacency<_row_size, _row_max, _align> >

ArrayPair alias for mesh adjacency (variable-width integer rows).

Definition at line 704 of file ArrayPair.hpp.

ArrayEigenMatrixBatchPair

using DNDS::ArrayEigenMatrixBatchPair = typedef ArrayPair<ArrayEigenMatrixBatch>

ArrayPair alias for per-row variable-size Eigen matrix batches.

Definition at line 716 of file ArrayPair.hpp.

ArrayEigenMatrixPair

template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>

using DNDS::ArrayEigenMatrixPair = typedef ArrayPair<ArrayEigenMatrix<_mat_ni, _mat_nj, _mat_ni_max, _mat_nj_max, _align> >

ArrayPair alias for per-row Eigen matrices.

Definition at line 713 of file ArrayPair.hpp.

ArrayEigenUniMatrixBatchPair

template<int _n_row, int _n_col>

using DNDS::ArrayEigenUniMatrixBatchPair = typedef ArrayPair<ArrayEigenUniMatrixBatch<_n_row, _n_col> >

ArrayPair alias for per-row batches of uniform _n_row x _n_col matrices.

Used by FiniteVolume / VariationalReconstruction to store per-quadrature-point Jacobians and basis coefficients.

Definition at line 722 of file ArrayPair.hpp.

ArrayEigenVectorPair

template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>

using DNDS::ArrayEigenVectorPair = typedef ArrayPair<ArrayEigenVector<_vec_size, _row_max, _align> >

ArrayPair alias for per-row Eigen vectors (e.g., node coords with N=3).

Definition at line 708 of file ArrayPair.hpp.

ArrayTransformerType_t

template<class TArray >

using DNDS::ArrayTransformerType_t = typedef typename ArrayTransformerType<TArray>::Type

Definition at line 1484 of file ArrayTransformer.hpp.

ContextualCheck

using DNDS::ContextualCheck = typedef std::function<CheckResult(const void *obj, const ConfigContext &ctx)>

A cross-field validation check that requires runtime context.

The function receives a const void* pointing to the config section struct and a ConfigContext carrying runtime values (nVars, dim, model, etc.).

Definition at line 208 of file ConfigRegistry.hpp.

CrossFieldCheck

using DNDS::CrossFieldCheck = typedef std::function<CheckResult(const void *obj)>

A cross-field validation check that does not require runtime context.

The function receives a const void* pointing to the config section struct. It should cast to const T& and inspect the fields, returning CheckResult.

Definition at line 202 of file ConfigRegistry.hpp.

expression_t

using DNDS::expression_t = typedef exprtk::expression<real>

Definition at line 13 of file ExprtkWrapper.cpp.

host_device_vector

template<class T >

using DNDS::host_device_vector = typedef host_device_vector_r1<T>

Primary public alias: host_device_vector<T> = #host_device_vector_r1<T>. Prefer this name throughout the code base.

Definition at line 527 of file Vector.hpp.

index

using DNDS::index = typedef int64_t

Global row / DOF index type (signed 64-bit; handles multi-billion-cell meshes).

Definition at line 107 of file Defines.hpp.

MatrixXR

using DNDS::MatrixXR = typedef Eigen::Matrix<real, Eigen::Dynamic, Eigen::Dynamic>

Column-major dynamic Eigen matrix of reals (default layout).

Definition at line 205 of file Defines.hpp.

MPI_index

using DNDS::MPI_index = typedef MPI_Aint

MPI-compatible address/offset type (= MPI_Aint, 64-bit on all supported platforms). Used by the hindexed datatype machinery.

Definition at line 46 of file MPI.hpp.

MPI_int

using DNDS::MPI_int = typedef int

MPI counterpart type for MPI_int (= C int). Used for counts and ranks in MPI calls.

Definition at line 43 of file MPI.hpp.

parser_t

using DNDS::parser_t = typedef exprtk::parser<real>

Definition at line 14 of file ExprtkWrapper.cpp.

py_class_ssp

template<class T >

using DNDS::py_class_ssp = typedef py::classh<T>

Definition at line 21 of file Defines_bind.hpp.

real

using DNDS::real = typedef double

Canonical floating-point scalar used throughout DNDSR (double precision).

Definition at line 105 of file Defines.hpp.

real_half_sized_index

using DNDS::real_half_sized_index = typedef int32_t

Integer type with half the width of real.

Definition at line 113 of file Defines.hpp.

real_sized_index

using DNDS::real_sized_index = typedef int64_t

Integer type with the same width as real (used for type-punning / packing).

Definition at line 111 of file Defines.hpp.

remove_cvref_t

template<typename T >

using DNDS::remove_cvref_t = typedef std::remove_cv_t<std::remove_reference_t<T> >

Convenience remove_cv + remove_reference composition (C++17 port of C++20's std::remove_cvref_t).

Definition at line 156 of file Defines.hpp.

rowsize

using DNDS::rowsize = typedef int32_t

Row-width / per-row element-count type (signed 32-bit).

Definition at line 109 of file Defines.hpp.

RowVectorXR

using DNDS::RowVectorXR = typedef Eigen::RowVector<real, Eigen::Dynamic>

Dynamic row-vector of reals.

Definition at line 209 of file Defines.hpp.

ssp

template<typename T >

using DNDS::ssp = typedef std::shared_ptr<T>

Shortened alias for std::shared_ptr used pervasively in DNDSR.

Definition at line 138 of file Defines.hpp.

symbol_table_t

using DNDS::symbol_table_t = typedef exprtk::symbol_table<real>

Definition at line 12 of file ExprtkWrapper.cpp.

t_indexerPair

using DNDS::t_indexerPair = typedef std::tuple<index, index>

Paired indices, typically (start, size) or (first, last).

Definition at line 167 of file Defines.hpp.

t_IndexVec

using DNDS::t_IndexVec = typedef std::vector<index>

Vector of index values (global offsets, local ids, etc.).

Definition at line 161 of file Defines.hpp.

t_jsonconfig

using DNDS::t_jsonconfig = typedef nlohmann::ordered_json

Project-wide JSON type alias: nlohmann/json with ordered keys.

Order preservation makes generated config files diffable across re-saves. Use this for all DNDSR-facing configuration objects; reach for the unordered nlohmann::json only where compatibility with third-party producers demands it.

Definition at line 19 of file JsonUtil.hpp.

t_pIndexVec

using DNDS::t_pIndexVec = typedef ssp<t_IndexVec>

Shared pointer alias to t_IndexVec (used by mapping tables shared between arrays, see IndexMapping.hpp).

Definition at line 164 of file Defines.hpp.

t_pLGhostMapping

using DNDS::t_pLGhostMapping = typedef ssp<OffsetAscendIndexMapping>

Shared pointer to an OffsetAscendIndexMapping (per-rank ghost layout).

Definition at line 33 of file ArrayTransformer.hpp.

t_pLGlobalMapping

using DNDS::t_pLGlobalMapping = typedef ssp<GlobalOffsetsMapping>

Shared pointer to a GlobalOffsetsMapping (globally replicated).

Definition at line 31 of file ArrayTransformer.hpp.

t_RowsizeVec

using DNDS::t_RowsizeVec = typedef std::vector<rowsize>

Vector of row widths (one rowsize per row).

Definition at line 159 of file Defines.hpp.

t_sspDeviceStorageBase

using DNDS::t_sspDeviceStorageBase = typedef std::shared_ptr<DeviceStorageBase>

Shared pointer equivalent of t_supDeviceStorageBase.

Definition at line 89 of file DeviceStorage.hpp.

t_supDeviceStorageBase

using DNDS::t_supDeviceStorageBase = typedef std::unique_ptr<DeviceStorageBase, std::function<void(DeviceStorageBase *)> >

Owning unique pointer to a DeviceStorageBase with cross-DLL-safe deleter.

Definition at line 87 of file DeviceStorage.hpp.

tDiFj

using DNDS::tDiFj = typedef Eigen::Matrix<real, -1, -1, Eigen::RowMajor>

Row-major dynamic Eigen matrix used by quadrature / basis tables.

Definition at line 202 of file Defines.hpp.

tExprVars

using DNDS::tExprVars = typedef std::map<std::string, real>

Scalar name -> value map fed into an ExprtkWrapperEvaluator.

Definition at line 11 of file ExprtkWrapper.hpp.

tExprVarVecs

using DNDS::tExprVarVecs = typedef std::map<std::string, Eigen::Vector<real, Eigen::Dynamic> >

Vector name -> Eigen dense vector map fed into an ExprtkWrapperEvaluator.

Definition at line 13 of file ExprtkWrapper.hpp.

tLogSimpleDIValueMap

using DNDS::tLogSimpleDIValueMap = typedef std::map<std::string, LogSimpleDIValue>

Convenience alias: the title -> value map type passed to CsvLog::WriteLine.

Definition at line 141 of file CsvLog.hpp.

tMPI_AintVec

using DNDS::tMPI_AintVec = typedef tMPI_indexVec

Alias for tMPI_indexVec to match MPI_Aint terminology.

Definition at line 58 of file MPI.hpp.

tMPI_indexVec

using DNDS::tMPI_indexVec = typedef std::vector<MPI_index>

Vector of MPI_Aint byte-offsets for hindexed datatypes.

Definition at line 56 of file MPI.hpp.

tMPI_intVec

using DNDS::tMPI_intVec = typedef tMPI_sizeVec

Alias for tMPI_sizeVec; used where the name "int vec" reads better.

Definition at line 54 of file MPI.hpp.

tMPI_reqVec

using DNDS::tMPI_reqVec = typedef std::vector<MPI_Request>

Vector of MPI_Request, for persistent / nonblocking calls.

Definition at line 63 of file MPI.hpp.

tMPI_sizeVec

using DNDS::tMPI_sizeVec = typedef std::vector<MPI_int>

Vector of MPI counts.

Definition at line 52 of file MPI.hpp.

tMPI_statVec

using DNDS::tMPI_statVec = typedef std::vector<MPI_Status>

Vector of MPI_Status, for MPI_Waitall / MPI_Testall.

Definition at line 61 of file MPI.hpp.

tMPI_typePairVec

using DNDS::tMPI_typePairVec = typedef std::vector<std::pair<MPI_int, MPI_Datatype> >

Definition at line 323 of file MPI.hpp.

tpMPITypePairHolder

using DNDS::tpMPITypePairHolder = typedef ssp<MPITypePairHolder>

Shared-pointer alias to MPITypePairHolder.

Definition at line 378 of file MPI.hpp.

tPy_Array

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

using DNDS::tPy_Array = typedef py_class_ssp<Array<T, _row_size, _row_max, _align> >

Definition at line 65 of file Array_bind.hpp.

tPy_ArrayAdjacency

template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

using DNDS::tPy_ArrayAdjacency = typedef py_class_ssp<ArrayAdjacency<_row_size, _row_max, _align> >

Definition at line 30 of file ArrayAdjacency_bind.hpp.

tPy_ArrayAdjacencyPair

template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

using DNDS::tPy_ArrayAdjacencyPair = typedef py_class_ssp<ArrayAdjacencyPair<_row_size, _row_max, _align> >

Definition at line 33 of file ArrayAdjacency_bind.hpp.

tPy_ArrayDOF

template<int n_m, int n_n>

using DNDS::tPy_ArrayDOF = typedef py_class_ssp<ArrayDof<n_m, n_n> >

Definition at line 23 of file ArrayDOF_bind.hpp.

tPy_ArrayEigenMatrix

template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>

using DNDS::tPy_ArrayEigenMatrix = typedef py_class_ssp<ArrayEigenMatrix<_mat_ni, _mat_nj, _mat_ni_max, _mat_nj_max, _align> >

Definition at line 37 of file ArrayEigenMatrix_bind.hpp.

tPy_ArrayEigenMatrixBatch

using DNDS::tPy_ArrayEigenMatrixBatch = typedef py_class_ssp<ArrayEigenMatrixBatch>

Definition at line 19 of file ArrayEigenMatrixBatch_bind.hpp.

tPy_ArrayEigenMatrixBatchPair

using DNDS::tPy_ArrayEigenMatrixBatchPair = typedef py_class_ssp<ArrayEigenMatrixBatchPair>

Definition at line 21 of file ArrayEigenMatrixBatch_bind.hpp.

tPy_ArrayEigenMatrixPair

template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>

using DNDS::tPy_ArrayEigenMatrixPair = typedef py_class_ssp<ArrayEigenMatrixPair<_mat_ni, _mat_nj, _mat_ni_max, _mat_nj_max, _align> >

Definition at line 41 of file ArrayEigenMatrix_bind.hpp.

tPy_ArrayEigenUniMatrixBatch

template<int _n_row, int _n_col>

using DNDS::tPy_ArrayEigenUniMatrixBatch = typedef py_class_ssp<ArrayEigenUniMatrixBatch<_n_row, _n_col> >

Definition at line 26 of file ArrayEigenUniMatrixBatch_bind.hpp.

tPy_ArrayEigenUniMatrixBatchPair

template<int _n_row, int _n_col>

using DNDS::tPy_ArrayEigenUniMatrixBatchPair = typedef py_class_ssp<ArrayEigenUniMatrixBatchPair<_n_row, _n_col> >

Definition at line 29 of file ArrayEigenUniMatrixBatch_bind.hpp.

tPy_ArrayEigenVector

template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>

using DNDS::tPy_ArrayEigenVector = typedef py_class_ssp<ArrayEigenVector<_vec_size, _row_max, _align> >

Definition at line 30 of file ArrayEigenVector_bind.hpp.

tPy_ArrayEigenVectorPair

template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>

using DNDS::tPy_ArrayEigenVectorPair = typedef py_class_ssp<ArrayEigenVectorPair<_vec_size, _row_max, _align> >

Definition at line 33 of file ArrayEigenVector_bind.hpp.

tPy_ArrayTransformer

template<class TArray >

using DNDS::tPy_ArrayTransformer = typedef py_class_ssp<ArrayTransformerType_t<TArray> >

Definition at line 72 of file Array_bind.hpp.

tPy_ParArray

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

using DNDS::tPy_ParArray = typedef py_class_ssp<ParArray<T, _row_size, _row_max, _align> >

Definition at line 68 of file Array_bind.hpp.

tPy_ParArrayPair

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

using DNDS::tPy_ParArrayPair = typedef py_class_ssp<ArrayPair<ParArray<T, _row_size, _row_max, _align> >>

Definition at line 76 of file Array_bind.hpp.

VectorXR

using DNDS::VectorXR = typedef Eigen::Vector<real, Eigen::Dynamic>

Dynamic Eigen vector of reals.

Definition at line 207 of file Defines.hpp.

Enumeration Type Documentation

ConfigTypeTag

enum class DNDS::ConfigTypeTag

strong

Enumerates the JSON Schema type associated with a config field.

Used by FieldMeta::schemaEntry to emit the correct "type" value in the generated JSON Schema. The mapping is:

Tag	JSON Schema type	C++ types
Bool	"boolean"	bool
Int	"integer"	int, int32_t, int64_t, uint8_t
Real	"number"	real (double), float
String	"string"	std::string
Enum	"string" + enum	any enum with DNDS_DEFINE_ENUM_JSON
Array	"array"	std::vector<scalar>, Eigen::VectorXd
Object	"object"	nested config section
ArrayOfObjects	"array" of objects	std::vector<Section>
MapOfObjects	"object" (additionalProperties)	std::map<string,Section>
Json	{} (any)	nlohmann::ordered_json

Enumerator
Bool
Int
Real
String
Enum
Array
Object
ArrayOfObjects
MapOfObjects
Json

Definition at line 105 of file ConfigRegistry.hpp.

DataLayout

enum DNDS::DataLayout

Enumeration of the five concrete data layouts supported by Array.

The value is determined at compile time from _row_size and _row_max template parameters. See the layout table in Array.hpp.

Enumerator
ErrorLayout	Invalid combination of template parameters.
TABLE_StaticFixed	Fixed row width, known at compile time.
TABLE_Fixed	Fixed row width, set at runtime (uniform across rows).
TABLE_Max	Padded variable rows; max width set at runtime.
TABLE_StaticMax	Padded variable rows; max width fixed at compile time.
CSR	Compressed Sparse Row (flat buffer + row-start index).

Definition at line 17 of file ArrayBasic.hpp.

DeviceBackend

enum class DNDS::DeviceBackend

strong

Enumerates the backends a DeviceStorage / Array can live on.

Host is always available; CUDA is compiled in when DNDS_USE_CUDA is defined. Additional slots (Custom1, ...) are placeholders for future backends (e.g., HIP, SYCL) that can be plugged in by providing new factory specialisations.

Enumerator
Unknown	Unset / sentinel.
Host	Plain CPU memory.
Custom1	Reserved slot for a project-specific backend.

Definition at line 27 of file DeviceStorage.hpp.

Function Documentation

__DNDSToMPITypeFloat()

template<class Tbasic >

constexpr MPI_Datatype DNDS::__DNDSToMPITypeFloat ( )

constexpr

Map a DNDS floating-point type size to an MPI datatype.

Compile-time selects MPI_REAL8 or MPI_REAL4 based on sizeof(Tbasic). Used by DNDS_MPI_REAL.

Definition at line 83 of file MPI.hpp.

__DNDSToMPITypeInt()

template<class Tbasic >

constexpr MPI_Datatype DNDS::__DNDSToMPITypeInt ( )

constexpr

Map a DNDS integer type size to an MPI signed-integer datatype.

Compile-time selects MPI_INT64_T or MPI_INT32_T based on sizeof(Tbasic). Used by DNDS_MPI_INDEX.

Definition at line 71 of file MPI.hpp.

__OneMatGetRowSize()

template<rowsize _mat_ni, rowsize _mat_nj>

constexpr rowsize DNDS::__OneMatGetRowSize ( )

constexpr

Compute the underlying per-row element count for an Ni x Nj matrix cell.

Resolves to Ni*Nj when both are compile-time fixed, NonUniformSize when either uses per-row sizing, or DynamicSize for the remaining runtime-determined cases.

Definition at line 20 of file ArrayEigenMatrix_DeviceView.hpp.

__pybind11_callBindArrayAdjacencys_rowsizes_sequence()

template<size_t N, std::array< int, N > const & Arr, size_t... Is>

void DNDS::__pybind11_callBindArrayAdjacencys_rowsizes_sequence	(	py::module_ &	m,
		std::index_sequence< Is... >
	)

Definition at line 203 of file ArrayAdjacency_bind.hpp.

Here is the caller graph for this function:

__pybind11_callBindArrayDOFs_rowsizes_sequence()

template<rowsize mat_n, size_t N, std::array< int, N > const & Arr, size_t... Is>

void DNDS::__pybind11_callBindArrayDOFs_rowsizes_sequence	(	py::module_ &	m,
		std::index_sequence< Is... >
	)

Definition at line 130 of file ArrayDOF_bind.hpp.

Here is the caller graph for this function:

__pybind11_callBindArrayEigenMatrixs_rowsizes_sequence()

template<rowsize mat_n, size_t N, std::array< int, N > const & Arr, size_t... Is>

void DNDS::__pybind11_callBindArrayEigenMatrixs_rowsizes_sequence	(	py::module_ &	m,
		std::index_sequence< Is... >
	)

Definition at line 249 of file ArrayEigenMatrix_bind.hpp.

Here is the caller graph for this function:

__pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes_sequence()

template<rowsize mat_n, size_t N, std::array< int, N > const & Arr, size_t... Is>

void DNDS::__pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes_sequence	(	py::module_ &	m,
		std::index_sequence< Is... >
	)

Definition at line 297 of file ArrayEigenUniMatrixBatch_bind.hpp.

Here is the caller graph for this function:

__pybind11_callBindArrayEigenVectors_rowsizes_sequence()

template<size_t N, std::array< int, N > const & Arr, size_t... Is>

void DNDS::__pybind11_callBindArrayEigenVectors_rowsizes_sequence	(	py::module_ &	m,
		std::index_sequence< Is... >
	)

Definition at line 196 of file ArrayEigenVector_bind.hpp.

Here is the caller graph for this function:

__pybind11_callBindArrays_rowsizes_sequence()

template<class T , size_t N, std::array< int, N > const & Arr, size_t... Is>

void DNDS::__pybind11_callBindArrays_rowsizes_sequence	(	py::module_ &	m,
		std::index_sequence< Is... >
	)

Definition at line 564 of file Array_bind.hpp.

Here is the caller graph for this function:

__pybind11_callBindArrayTransformers_rowsizes_sequence()

template<class T , size_t N, std::array< int, N > const & Arr, size_t... Is>

void DNDS::__pybind11_callBindArrayTransformers_rowsizes_sequence	(	py::module_ &	m,
		std::index_sequence< Is... >
	)

Definition at line 592 of file Array_bind.hpp.

Here is the caller graph for this function:

__pybind11_callBindParArrayPairs_rowsizes_sequence()

template<class T , size_t N, std::array< int, N > const & Arr, size_t... Is>

void DNDS::__pybind11_callBindParArrayPairs_rowsizes_sequence	(	py::module_ &	m,
		std::index_sequence< Is... >
	)

Definition at line 606 of file Array_bind.hpp.

Here is the caller graph for this function:

__pybind11_callBindParArrays_rowsizes_sequence()

template<class T , size_t N, std::array< int, N > const & Arr, size_t... Is>

void DNDS::__pybind11_callBindParArrays_rowsizes_sequence	(	py::module_ &	m,
		std::index_sequence< Is... >
	)

Definition at line 578 of file Array_bind.hpp.

Here is the caller graph for this function:

_get_pybind11_arrayRowsizeInstantiationList()

template<int offset = 0>

constexpr auto DNDS::_get_pybind11_arrayRowsizeInstantiationList ( )

constexpr

Definition at line 550 of file Array_bind.hpp.

_pybind11_Array_define_dispatch()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

void DNDS::_pybind11_Array_define_dispatch

(

py::module_ &

m

)

Definition at line 223 of file Array_bind.hpp.

_pybind11_ArrayAdjacency_define_dispatch()

template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

void DNDS::_pybind11_ArrayAdjacency_define_dispatch

(

py::module_ &

m

)

Definition at line 125 of file ArrayAdjacency_bind.hpp.

_pybind11_ArrayAdjacencyPair_define_dispatch()

template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

void DNDS::_pybind11_ArrayAdjacencyPair_define_dispatch

(

py::module_ &

m

)

Definition at line 191 of file ArrayAdjacency_bind.hpp.

_pybind11_ArrayDOF_define_dispatch()

template<int n_m, int n_n>

void DNDS::_pybind11_ArrayDOF_define_dispatch

(

py::module_ &

m

)

Definition at line 117 of file ArrayDOF_bind.hpp.

_pybind11_ArrayEigenMatrix_define_dispatch()

template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>

void DNDS::_pybind11_ArrayEigenMatrix_define_dispatch

(

py::module_ &

m

)

Definition at line 169 of file ArrayEigenMatrix_bind.hpp.

_pybind11_ArrayEigenMatrixPair_define_dispatch()

template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>

void DNDS::_pybind11_ArrayEigenMatrixPair_define_dispatch

(

py::module_ &

m

)

Definition at line 236 of file ArrayEigenMatrix_bind.hpp.

_pybind11_ArrayEigenUniMatrixBatch_define_dispatch()

template<int _n_row, int _n_col>

void DNDS::_pybind11_ArrayEigenUniMatrixBatch_define_dispatch

(

py::module_ &

m

)

Definition at line 201 of file ArrayEigenUniMatrixBatch_bind.hpp.

_pybind11_ArrayEigenUniMatrixBatchPair_define_dispatch()

template<int _n_row, int _n_col>

void DNDS::_pybind11_ArrayEigenUniMatrixBatchPair_define_dispatch

(

py::module_ &

m

)

Definition at line 283 of file ArrayEigenUniMatrixBatch_bind.hpp.

_pybind11_ArrayEigenVector_define_dispatch()

template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>

void DNDS::_pybind11_ArrayEigenVector_define_dispatch

(

py::module_ &

m

)

Definition at line 120 of file ArrayEigenVector_bind.hpp.

_pybind11_ArrayEigenVectorPair_define_dispatch()

template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>

void DNDS::_pybind11_ArrayEigenVectorPair_define_dispatch

(

py::module_ &

m

)

Definition at line 184 of file ArrayEigenVector_bind.hpp.

_pybind11_ArrayTransformer_define_dispatch()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

void DNDS::_pybind11_ArrayTransformer_define_dispatch

(

py::module_ &

m

)

Definition at line 538 of file Array_bind.hpp.

_pybind11_ParArray_define_dispatch()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

void DNDS::_pybind11_ParArray_define_dispatch

(

py::module_ &

m

)

Definition at line 296 of file Array_bind.hpp.

_pybind11_ParArrayPair_define_dispatch()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

void DNDS::_pybind11_ParArrayPair_define_dispatch

(

py::module_ &

m

)

Definition at line 437 of file Array_bind.hpp.

AccumulateRowSize()

template<class TtRowsizeVec , class TtIndexVec >

void DNDS::AccumulateRowSize ( const TtRowsizeVec &  rowsizes, TtIndexVec &  rowstarts  )

inline

Build a prefix-sum table from a row-size vector.

Resizes rowstarts to rowsizes.size() + 1 and fills it so that rowstarts[i] = sum_{k<i} rowsizes[k]. Asserts on overflow. Used extensively by CSR Array / ghost-indexing to turn row-count vectors into offset vectors.

Template Parameters

TtRowsizeVec	std::vector-like of row sizes (e.g. rowsize).
TtIndexVec	std::vector-like of offsets (e.g. index), must have a strictly wider range than the row-size type.

Definition at line 452 of file Defines.hpp.

Here is the caller graph for this function:

Align_To_PySnippet()

std::string DNDS::Align_To_PySnippet ( rowsize  al )

inline

Encode an alignment value as a Python-binding snippet: "N" for NoAlign, the number otherwise.

Definition at line 310 of file Defines.hpp.

Here is the caller graph for this function:

array_comp_acceptable()

template<class T >

constexpr bool DNDS::array_comp_acceptable ( )

constexpr

Whether T is legal as an Array element type (trivially copyable or a fixed-size real Eigen matrix). Controls CUDA/MPI copy paths.

Definition at line 56 of file ArrayBasic.hpp.

ArrayDeviceView_build()

template<DeviceBackend B, class T , rowsize _row_size, rowsize _row_max, rowsize _align>

auto DNDS::ArrayDeviceView_build	(	index	n_size,
		T *	n_data,
		index	n_data_size,
		const index *	n_rowstart,
		index	n_rowstart_size,
		const rowsize *	n_rowsizes,
		index	n_rowsizes_size,
		rowsize	n_row_size_dynamic,
		T *	n_data_device,
		const index *	n_rowstart_device,
		const rowsize *	n_rowsizes_device
	)

Factory for ArrayDeviceView: selects host or device pointers based on B and forwards everything else.

Used by Array::deviceView<B>(); rarely called directly.

Definition at line 107 of file DeviceView.hpp.

assert_false()

void DNDS::assert_false ( const char *  expr, const char *  file, int  line  )

inline

Low-level: print a red "DNDS_assertion failed" line and abort.

Definition at line 40 of file Errors.hpp.

Here is the call graph for this function:

assert_false_info()

void DNDS::assert_false_info ( const char *  expr, const char *  file, int  line, const std::string &  info  )

inline

Variant of assert_false that prints an extra info string.

Definition at line 48 of file Errors.hpp.

Here is the call graph for this function:

assert_false_info_mpi()

void DNDS::assert_false_info_mpi	(	const char *	expr,
		const char *	file,
		int	line,
		const std::string &	info,
		const DNDS::MPIInfo &	mpi
	)

MPI-aware assertion-failure reporter.

Barriers before abort so every rank flushes its output; if Debug::isDebugging is set, busy-waits to allow debugger attachment.

Definition at line 95 of file MPI.cpp.

Here is the call graph for this function:

assert_false_info_throw()

template<class TException = std::runtime_error>

void DNDS::assert_false_info_throw	(	const char *	expr,
		const char *	file,
		int	line,
		const std::string &	info
	)

Throwing variant of assert_false_info. Used by DNDS_check_throw.

Template Parameters

TException

Exception type to throw (defaults to std::runtime_error). Currently the implementation ignores the template parameter and always throws std::runtime_error; kept for future customisation.

Definition at line 76 of file Errors.hpp.

Here is the call graph for this function:

assert_false_infof()

void DNDS::assert_false_infof ( const char *  expr, const char *  file, int  line, const char *  info,   ...  )

inline

printf-style variant of assert_false. Used by DNDS_assert_infof.

Definition at line 57 of file Errors.hpp.

Here is the call graph for this function:

BasicType_To_MPIIntType()

template<class T >

std::pair< MPI_Datatype, MPI_int > DNDS::BasicType_To_MPIIntType

(

)

Deduce an (MPI_Datatype, count) pair that represents a T value.

Compile-time dispatch:

• builtin float / int types map to their obvious MPI_* datatypes, count = 1;
• C-style arrays (T[N]) recurse into the element type and multiply the count;
• std::array<T, N> recurses into T::value_type and multiplies by N;
• fixed-size real Eigen matrices map to DNDS_MPI_REAL with count sizeof(T)/sizeof(real);
• otherwise falls back to BasicType_To_MPIIntType_Custom.

Used throughout ghost-communication and collective code to avoid hand- writing datatype mapping for every MPI call.

Note

Not constexpr because OpenMPI handles are not constant-expressions. Warning, not const-expr since OpenMPI disallows it

Definition at line 149 of file MPI.hpp.

Here is the call graph for this function:

BasicType_To_MPIIntType_Custom()

template<class T >

std::pair< MPI_Datatype, MPI_int > DNDS::BasicType_To_MPIIntType_Custom

(

)

Dispatch to a user-provided CommPair / CommMult+ CommType pair on T.

Last resort for types that are not scalars, plain arrays, or real Eigen matrices. T must either expose static CommMult() + CommType() methods, or a static CommPair() returning the same pair.

Definition at line 124 of file MPI.hpp.

BuildRedistributionPullingIndex()

std::vector< index > DNDS::BuildRedistributionPullingIndex ( const MPIInfo &  mpi, const std::vector< index > &  readOrigIndex, const std::vector< index > &  newOrigIndex, const ssp< GlobalOffsetsMapping > &  readGlobalMapping  )

inline

Redistributes ArrayPair data between different MPI partitions using ArrayTransformer.

The caller provides:

• readOrigIndex: the original indices for data this rank read from HDF5 (even-split)
• readArray: the data this rank read (father of a temporary ArrayPair)
• newOrigIndex: the original indices this rank needs (from current mesh's cell2cellOrig)

The redistributor builds a temporary ArrayTransformer to pull needed data:

1. Build a global mapping on the "read" arrays (readArray serves as father).
2. For each entry in newOrigIndex, find which rank's read slice contains it, and compute the corresponding global index in the read array's global space.
3. Use ArrayTransformer::createGhostMapping + pullOnce to fetch the data.
4. Copy the pulled data into the output array in newOrigIndex order.

This works for fixed-row-size, dynamic-row-size, and CSR arrays alike, since ArrayTransformer handles all these layouts internally.

Builds the pulling index mapping for redistribution.

Given the origIndex arrays from the even-split read (distributed across all ranks), and the newOrigIndex for this rank, determines which global indices in the read array space need to be pulled.

Uses a 3-round MPI_Alltoallv rendezvous: Round 1: Send (origIdx, globalReadIdx) pairs to directory ranks. Round 2: Send query origIndex entries to directory ranks. Round 3: Directory ranks reply with resolved globalReadIdx.

Memory scaling

Temporary memory per rank is O(nLocal) where nLocal is the larger of readOrigIndex.size() and newOrigIndex.size(). Specifically:

• 7 index vectors of size O(nLocal): sendBuf, recvBuf, querySendBuf, queryRecvBuf, queryReplyBuf, replyRecvBuf, pullingIndexGlobal
• 1 unordered_map<index,index> of size O(nGlobal/nRanks) for the directory
• Total: ~7 × nLocal × sizeof(index) + O(nGlobal/nRanks) hash table overhead For a 100M-cell mesh with 8 ranks: ~12.5M entries × 8 B × 7 ≈ 700 MB per rank, plus ~200 MB for the directory hash table.

Parameters

mpi	MPI communicator info
readOrigIndex	Original indices from the even-split read (this rank's slice)
newOrigIndex	Original indices this rank needs (from current partition)
readGlobalMapping	Global mapping of the read array (prefix-sum offsets)

Returns

pullingIndexGlobal: vector of global read-array indices to pull (ordered to match newOrigIndex)

Definition at line 70 of file ArrayRedistributor.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

checkedIndexTo32()

int32_t DNDS::checkedIndexTo32 ( index  v )

inline

Narrow index to int32_t with range check; dies on overflow.

Todo:

//TODO: overflow_assign_int64_to_32

Definition at line 619 of file Defines.hpp.

checkUniformVector()

template<class T >

bool DNDS::checkUniformVector ( const std::vector< T > &  dat, T &  value  )

inline

Whether all elements of dat are equal; if so, stores the value into value.

Returns

false if the vector is empty or contains at least two distinct values.

Definition at line 469 of file Defines.hpp.

cube()

template<typename T >

constexpr T DNDS::cube ( const T &  a )

constexpr

a * a * a, constexpr.

Definition at line 525 of file Defines.hpp.

Here is the caller graph for this function:

device_backend_name()

const char * DNDS::device_backend_name ( DeviceBackend  B )

inline

Canonical string name for a DeviceBackend (used in log messages).

Definition at line 51 of file DeviceStorage.hpp.

Here is the caller graph for this function:

device_backend_name_to_enum()

DeviceBackend DNDS::device_backend_name_to_enum ( std::string_view  s )

inline

Inverse of device_backend_name. Returns Unknown for unrecognised names.

Definition at line 70 of file DeviceStorage.hpp.

Here is the caller graph for this function:

device_storage_create()

t_supDeviceStorageBase DNDS::device_storage_create ( DeviceBackend  backend, size_t  n_bytes  )

inline

Top-level factory: dispatches to the per-backend factory based on backend. Returns a null unique_ptr for DeviceBackendUnknown.

Definition at line 159 of file DeviceStorage.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

deviceStorageBase_deleter()

void DNDS::deviceStorageBase_deleter

(

DeviceStorageBase *

p

)

Stateless deleter for DeviceStorageBase that works across shared-library boundaries where the vtable of unique_ptr's default deleter would not.

Definition at line 12 of file DeviceStorage.cpp.

Here is the caller graph for this function:

divCeil()

template<class TL , class TR >

constexpr auto DNDS::divCeil ( TL  l, TR  r  )

constexpr

Integer ceiling division. l must be non-negative, r positive.

Bug:

The current expression has incorrect precedence; see divide_ceil for the corrected version. Do not use new code with divCeil.

l must be non-negative, r must be positive. integers

Definition at line 495 of file Defines.hpp.

Here is the call graph for this function:

divide_ceil()

template<typename T >

constexpr T DNDS::divide_ceil ( T  a, T  b  )

constexpr

Integer ceiling division ceil(a / b). Correct for all signs.

Definition at line 569 of file Defines.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

DNDS_H5T_INDEX()

hid_t DNDS::DNDS_H5T_INDEX ( )

inline

HDF5 native datatype matching DNDS index (currently H5T_NATIVE_INT64).

Static-asserts that the DNDS type has not drifted from int64_t.

Definition at line 12 of file HDF5.hpp.

Here is the caller graph for this function:

DNDS_H5T_REAL()

hid_t DNDS::DNDS_H5T_REAL ( )

inline

HDF5 native datatype matching DNDS real (currently H5T_NATIVE_DOUBLE).

Definition at line 36 of file HDF5.hpp.

Here is the caller graph for this function:

DNDS_H5T_ROWSIZE()

hid_t DNDS::DNDS_H5T_ROWSIZE ( )

inline

HDF5 native datatype matching DNDS rowsize (currently H5T_NATIVE_INT32).

Definition at line 24 of file HDF5.hpp.

Here is the caller graph for this function:

EigenSize_Mul_RowSize()

template<int a, int b>

constexpr rowsize DNDS::EigenSize_Mul_RowSize ( )

constexpr

a * b if both are compile-time sizes, DynamicSize if either is Eigen::Dynamic, otherwise a sentinel.

Helper used by ArrayEigenUniMatrixBatchDeviceView to compute the underlying ParArray row width from the matrix shape.

Definition at line 17 of file ArrayEigenUniMatrixBatch_DeviceView.hpp.

Here is the call graph for this function:

EigenVectorFMTSafeGetJson()

auto DNDS::EigenVectorFMTSafeGetJson ( const Eigen::VectorFMTSafe< real, -1 > &  ve )

inline

Dump an Eigen::VectorFMTSafe into a JSON array of doubles.

Definition at line 77 of file JsonUtil.hpp.

Here is the caller graph for this function:

EigenVectorGetJson()

auto DNDS::EigenVectorGetJson ( const Eigen::VectorXd &  ve )

inline

Dump an Eigen::VectorXd into a JSON array of doubles.

Definition at line 67 of file JsonUtil.hpp.

Here is the caller graph for this function:

EvenSplitRange()

std::pair< index, index > DNDS::EvenSplitRange ( int  rank, int  nRanks, index  nGlobal  )

inline

Split a global range [0, nGlobal) evenly among nRanks workers.

Used by the EvenSplit serialization-offset mode and by redistribution routines to pick per-rank slabs without exchanging sizes. The partition is dense and balanced to within one item.

Returns

[start, end) for the caller's rank.

Definition at line 129 of file Defines.hpp.

Here is the caller graph for this function:

float_mod()

real DNDS::float_mod ( real  a, real  b  )

inline

Floating-point modulo matching Python's % (result has sign of b).

Parameters

b	Must be positive.

Definition at line 580 of file Defines.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

for_each_member_list()

template<typename TList , typename F >

void DNDS::for_each_member_list	(	TList &&	obj_member_list,
		F &&	f
	)

Invoke f(member) for every element of a std::tuple of members.

Companion to MemberRef / MemberPtr below; used to build auto-generated JSON / diagnostic dumps of struct contents.

Definition at line 17 of file ObjectUtils.hpp.

Here is the caller graph for this function:

for_each_member_ptr_list()

template<typename Class , typename TList , typename F >

void DNDS::for_each_member_ptr_list	(	Class &	obj,
		TList &&	obj_member_ptr_list,
		F &&	f
	)

Invoke f(name, obj.*ptr) for every member in a list of MemberPtr.

Definition at line 42 of file ObjectUtils.hpp.

Here is the caller graph for this function:

for_each_member_ptr_list_raw()

template<typename TList , typename F >

void DNDS::for_each_member_ptr_list_raw	(	TList &&	obj_member_ptr_list,
		F &&	f
	)

Low-level variant that passes each MemberPtr object through to f directly (for callers that need access to both ptr and name).

Definition at line 51 of file ObjectUtils.hpp.

from_json()

void DNDS::from_json ( const nlohmann::ordered_json &  j, host_device_vector< real > &  v  )

inline

Definition at line 183 of file JsonUtil.hpp.

Here is the caller graph for this function:

get_env_DNDS_DIST_OMP_NUM_THREADS()

int DNDS::get_env_DNDS_DIST_OMP_NUM_THREADS

(

)

Read the DNDSR-specific DNDS_DIST_OMP_NUM_THREADS override, falling back to get_env_OMP_NUM_THREADS.

Definition at line 139 of file Defines.cpp.

Here is the caller graph for this function:

get_env_OMP_NUM_THREADS()

int DNDS::get_env_OMP_NUM_THREADS

(

)

Read OMP_NUM_THREADS env var, returning 1 if unset / invalid.

Definition at line 119 of file Defines.cpp.

get_terminal_width()

int DNDS::get_terminal_width

(

)

Terminal width in columns (falls back to a fixed default when not a TTY).

Definition at line 57 of file Defines.cpp.

Here is the caller graph for this function:

GetSetVersionName()

std::string DNDS::GetSetVersionName

(

const std::string &

ver = ""

)

Read/set the build version string accessible from code.

When ver is non-empty it replaces the stored value; always returns the current value. Written into HDF5 attributes for traceability.

Definition at line 172 of file Defines.cpp.

Here is the caller graph for this function:

getStringForcePath()

std::string DNDS::getStringForcePath ( const std::filesystem::path::string_type &  v )

inline

Portable conversion of a platform-native path string to std::string.

Definition at line 630 of file Defines.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

getStringForceWString()

std::string DNDS::getStringForceWString

(

const std::wstring &

v

)

Convert a wstring to string (UTF-8 on Windows, byte-cast elsewhere).

Definition at line 104 of file Defines.cpp.

Here is the caller graph for this function:

getTimeStamp()

std::string DNDS::getTimeStamp

(

const MPIInfo &

mpi

)

Format a human-readable timestamp using the calling rank as context.

Definition at line 116 of file MPI.cpp.

Here is the caller graph for this function:

getTraceString()

std::string DNDS::getTraceString

(

)

Return a symbolicated stack trace for the calling thread.

Host-only, implemented with boost::stacktrace (or similar). Used by the assert_false* helpers below.

Definition at line 162 of file Defines.cpp.

Here is the caller graph for this function:

InsertCheck()

void DNDS::InsertCheck ( const MPIInfo &  mpi, const std::string &  info = "", const std::string &  FUNCTION = "", const std::string &  FILE = "", int  LINE = -1  )

inline

Barrier + annotated print used by DNDS_MPI_InsertCheck.

No-op in release builds (NDEBUG or NINSERT defined).

Definition at line 779 of file MPI.hpp.

Here is the call graph for this function:

isTABLE()

constexpr bool DNDS::isTABLE ( DataLayout  lo )

constexpr

Whether the layout uses a TABLE (padded) representation (vs CSR).

Definition at line 28 of file ArrayBasic.hpp.

isTABLE_Dynamic()

constexpr bool DNDS::isTABLE_Dynamic ( DataLayout  lo )

constexpr

Whether the layout carries a runtime row-size parameter.

Definition at line 48 of file ArrayBasic.hpp.

isTABLE_Fixed()

constexpr bool DNDS::isTABLE_Fixed ( DataLayout  lo )

constexpr

Whether the layout has a uniform row width (no per-row size needed).

Definition at line 33 of file ArrayBasic.hpp.

Here is the caller graph for this function:

isTABLE_Max()

constexpr bool DNDS::isTABLE_Max ( DataLayout  lo )

constexpr

Whether the layout is a padded-max variant (uses _pRowSizes).

Definition at line 38 of file ArrayBasic.hpp.

Here is the caller graph for this function:

isTABLE_Static()

constexpr bool DNDS::isTABLE_Static ( DataLayout  lo )

constexpr

Whether the layout has a compile-time row-size constant.

Definition at line 43 of file ArrayBasic.hpp.

IsUnInitReal()

bool DNDS::IsUnInitReal ( real  v )

inline

Whether v equals the NaN sentinel UnInitReal (tested via isnan).

Definition at line 183 of file Defines.hpp.

iterateIdentical()

template<class tIt1 , class tIt1end , class tIt2 , class tIt2end , class tF >

bool DNDS::iterateIdentical	(	tIt1	it1,
		tIt1end	it1end,
		tIt2	it2,
		tIt2end	it2end,
		tF	F
	)

Walk two ordered ranges in lockstep, calling F on each match.

Two-pointer algorithm. Requires both ranges to be sorted in ascending order; advances the smaller side at each step. Exits early (returning true) if F(value, pos1, pos2) returns true.

Template Parameters

tF	Callable bool(value, size_t pos1, size_t pos2).

Returns

true if the functor short-circuited, false if ranges were walked to completion.

Definition at line 596 of file Defines.hpp.

Here is the caller graph for this function:

JsonGetEigenVector()

Eigen::VectorXd DNDS::JsonGetEigenVector ( const nlohmann::json &  arr )

inline

Parse a JSON array into an Eigen::VectorXd. Throws a descriptive assertion on any JSON error.

Definition at line 27 of file JsonUtil.hpp.

Here is the caller graph for this function:

JsonGetEigenVectorFMTSafe()

Eigen::VectorFMTSafe< real, -1 > DNDS::JsonGetEigenVectorFMTSafe ( const nlohmann::json &  arr )

inline

Parse a JSON array into an Eigen::VectorFMTSafe (fixed-point-aware wrapper).

Used by configuration paths that feed values into code paths compiled with the fixed-point Eigen shim.

Definition at line 48 of file JsonUtil.hpp.

Here is the caller graph for this function:

log()

std::ostream & DNDS::log

(

)

Return the current DNDSR log stream (either std::cout or the installed file).

Definition at line 49 of file Defines.cpp.

Here is the caller graph for this function:

logIsTTY()

bool DNDS::logIsTTY

(

)

Convenience: ostreamIsTTY applied to the current log() stream.

Definition at line 51 of file Defines.cpp.

Here is the call graph for this function:

make_ssp() [1/2]

template<typename T , typename... Args>

ssp< T > DNDS::make_ssp

(

Args &&...

args

)

Type-safe replacement for DNDS_MAKE_SSP. Creates ssp<T> with forwarded args.

Definition at line 255 of file Defines.hpp.

Here is the caller graph for this function:

make_ssp() [2/2]

template<typename T , typename... Args>

ssp< T > DNDS::make_ssp	(	ObjName	objName,
		Args &&...	args
	)

Named variant of make_ssp. If T inherits ObjectNaming, sets the name.

Usage: auto p = make_ssp<ParArray<index>>(ObjName{"cell2node"}, mpi);

Definition at line 264 of file Defines.hpp.

MemberPtr()

template<typename Class , typename T >

DNDS::MemberPtr	(	T Class::*	,
		const char *
	)		-> MemberPtr< Class, T >

MemberRef()

template<typename T >

DNDS::MemberRef	(	T &	,
		const char *
	)		-> MemberRef< T >

mod()

template<typename T >

constexpr T DNDS::mod ( T  a, T  b  )

constexpr

Mathematical modulo that always returns a non-negative result. Unlike % in C++ where (-1) % 3 == -1, mod(-1, 3) == 2.

Definition at line 558 of file Defines.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

MPISerialDo()

template<class F >

void DNDS::MPISerialDo ( const MPIInfo &  mpi, F  f  )

inline

Execute f on each rank serially, in rank order.

Inserts an MPI_Barrier before each rank's turn so that output interleaving is deterministic. Useful for diagnostics where every rank prints something about its own state.

Template Parameters

F	Callable with no arguments.

need some improvement: order could be bad?

Definition at line 698 of file MPI.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

MPIWorldRank()

MPI_int DNDS::MPIWorldRank ( )

inline

Convenience: MPI_Comm_rank(MPI_COMM_WORLD).

Definition at line 307 of file MPI.hpp.

Here is the caller graph for this function:

MPIWorldSize()

MPI_int DNDS::MPIWorldSize ( )

inline

Convenience: MPI_Comm_size(MPI_COMM_WORLD).

Definition at line 299 of file MPI.hpp.

Here is the caller graph for this function:

null_supDeviceStorageBase()

t_supDeviceStorageBase DNDS::null_supDeviceStorageBase ( )

inline

Null-value helper for t_supDeviceStorageBase.

Definition at line 92 of file DeviceStorage.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

optimize_hindexed_layout()

template<class TBlkSiz , class TDisp , class TSizeof >

auto DNDS::optimize_hindexed_layout	(	index	o_size,
		TBlkSiz *	blk_sizes,
		TDisp *	disps,
		TSizeof	sizeofElem
	)

Coalesce contiguous blocks in an MPI_Type_create_hindexed layout.

MPI derived-type performance depends heavily on the number of blocks. If two consecutive blocks happen to be adjacent in memory (disps[i+1] == disps[i] + blk_sizes[i] * sizeofElem), they can be merged into a single larger block. This helper walks the input arrays and returns a compacted (n_size, blk_sizes, disps) tuple.

Used by ArrayTransformercreateMPITypes before calling the actual MPI_Type_create_hindexed.

Template Parameters

TBlkSiz	Element type of the block-size array (e.g., MPI_int).
TDisp	Element type of the displacement array (e.g., MPI_Aint).
TSizeof	Element-size type (usually MPI_Aint).

Parameters

o_size	Number of input blocks.
blk_sizes	Block sizes (in element count).
disps	Displacements (in bytes).
sizeofElem	Element size in bytes (matches MPI_Type_extent).

Returns

(new_size, new_blk_sizes, new_disps) – a merged layout.

Definition at line 33 of file VectorUtils.hpp.

Here is the caller graph for this function:

ostreamIsTTY()

bool DNDS::ostreamIsTTY

(

std::ostream &

ostream

)

Heuristic detection of whether ostream is attached to a terminal.

Definition at line 36 of file Defines.cpp.

Here is the caller graph for this function:

print_progress()

void DNDS::print_progress	(	std::ostream &	os,
		double	progress
	)

Render a textual progress bar to os for progress in [0, 1].

Definition at line 75 of file Defines.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

PrintVec()

template<class T , class TP = T>

void DNDS::PrintVec ( const std::vector< T > &  dat, std::ostream &  out  )

inline

Print a vector to out with outputDelim between elements.

Template Parameters

TP	Optional cast type (e.g., promote int8_t to int to avoid character printing).

Definition at line 484 of file Defines.hpp.

py_buffer_contains_T()

template<class T >

bool DNDS::py_buffer_contains_T

(

const py::buffer_info &

info

)

Definition at line 24 of file Defines_bind.hpp.

py_buffer_get_contigious_size()

std::tuple< ssize_t, char > DNDS::py_buffer_get_contigious_size ( const py::buffer_info &  info )

inline

Definition at line 62 of file Defines_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

py_buffer_is_contigious_c()

bool DNDS::py_buffer_is_contigious_c ( const py::buffer_info &  info )

inline

Definition at line 30 of file Defines_bind.hpp.

Here is the caller graph for this function:

py_buffer_is_contigious_f()

bool DNDS::py_buffer_is_contigious_f ( const py::buffer_info &  info )

inline

Definition at line 46 of file Defines_bind.hpp.

Here is the caller graph for this function:

py_get_buffer_basic_mpi_datatype()

MPI_Datatype DNDS::py_get_buffer_basic_mpi_datatype ( const py::buffer_info &  info )

inline

Definition at line 13 of file MPI_bind.hpp.

Here is the caller graph for this function:

py_get_simple_mpi_op_by_name()

MPI_Op DNDS::py_get_simple_mpi_op_by_name ( const std::string &  op )

inline

Definition at line 43 of file MPI_bind.hpp.

Here is the caller graph for this function:

py_vector_as_memory_view()

template<typename T >

py::memoryview DNDS::py_vector_as_memory_view	(	std::vector< T > &	vec,
		bool	readonly
	)

Definition at line 77 of file Defines_bind.hpp.

Here is the caller graph for this function:

pybind11_Array_declare()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

tPy_Array< T, _row_size, _row_max, _align > DNDS::pybind11_Array_declare

(

py::module_ &

m

)

Definition at line 84 of file Array_bind.hpp.

pybind11_Array_define()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

void DNDS::pybind11_Array_define

(

py::module_ &

m

)

Definition at line 100 of file Array_bind.hpp.

Here is the call graph for this function:

pybind11_Array_get_class()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

tPy_Array< T, _row_size, _row_max, _align > DNDS::pybind11_Array_get_class

(

py::module_ &

m

)

Definition at line 94 of file Array_bind.hpp.

pybind11_Array_name()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

std::string DNDS::pybind11_Array_name

(

)

Definition at line 41 of file Array_bind.hpp.

pybind11_Array_name_appends()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

std::string DNDS::pybind11_Array_name_appends

(

)

Definition at line 24 of file Array_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayAdjacency_declare()

template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

tPy_ArrayAdjacency< _row_size, _row_max, _align > DNDS::pybind11_ArrayAdjacency_declare

(

py::module_ &

m

)

Definition at line 65 of file ArrayAdjacency_bind.hpp.

pybind11_ArrayAdjacency_define()

template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

void DNDS::pybind11_ArrayAdjacency_define

(

py::module_ &

m

)

Definition at line 80 of file ArrayAdjacency_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayAdjacency_get_class()

template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

tPy_ArrayAdjacency< _row_size, _row_max, _align > DNDS::pybind11_ArrayAdjacency_get_class

(

py::module_ &

m

)

Definition at line 74 of file ArrayAdjacency_bind.hpp.

pybind11_ArrayAdjacency_getitem()

template<class TArray = ArrayAdjacency<1>>

auto DNDS::pybind11_ArrayAdjacency_getitem	(	TArray &	self,
		index	index_
	)

Definition at line 50 of file ArrayAdjacency_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_ArrayAdjacency_name()

template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

std::string DNDS::pybind11_ArrayAdjacency_name

(

)

Definition at line 18 of file ArrayAdjacency_bind.hpp.

pybind11_ArrayAdjacency_name_appends()

template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

std::string DNDS::pybind11_ArrayAdjacency_name_appends

(

)

Definition at line 9 of file ArrayAdjacency_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayAdjacency_setitem()

template<class TArray = ArrayAdjacency<1>>

auto DNDS::pybind11_ArrayAdjacency_setitem	(	TArray &	self,
		index	index_,
		py::buffer	row
	)

Definition at line 39 of file ArrayAdjacency_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_ArrayAdjacencyPair_declare()

template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

tPy_ArrayAdjacencyPair< _row_size, _row_max, _align > DNDS::pybind11_ArrayAdjacencyPair_declare

(

py::module_ &

m

)

Definition at line 138 of file ArrayAdjacency_bind.hpp.

pybind11_ArrayAdjacencyPair_define()

template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

void DNDS::pybind11_ArrayAdjacencyPair_define

(

py::module_ &

m

)

Definition at line 152 of file ArrayAdjacency_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayAdjacencyPair_get_class()

template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

tPy_ArrayAdjacencyPair< _row_size, _row_max, _align > DNDS::pybind11_ArrayAdjacencyPair_get_class

(

py::module_ &

m

)

Definition at line 146 of file ArrayAdjacency_bind.hpp.

pybind11_ArrayAdjacencyPair_name()

template<rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

std::string DNDS::pybind11_ArrayAdjacencyPair_name

(

)

Definition at line 24 of file ArrayAdjacency_bind.hpp.

pybind11_ArrayDOF_declare()

template<int n_m, int n_n>

tPy_ArrayDOF< n_m, n_n > DNDS::pybind11_ArrayDOF_declare

(

py::module_ &

m

)

Definition at line 30 of file ArrayDOF_bind.hpp.

pybind11_ArrayDOF_define()

template<int n_m, int n_n>

void DNDS::pybind11_ArrayDOF_define

(

py::module_ &

m

)

Definition at line 45 of file ArrayDOF_bind.hpp.

pybind11_ArrayDOF_get_class()

template<int n_m, int n_n>

tPy_ArrayDOF< n_m, n_n > DNDS::pybind11_ArrayDOF_get_class

(

py::module_ &

m

)

Definition at line 39 of file ArrayDOF_bind.hpp.

pybind11_ArrayDOF_name()

template<int n_m, int n_n>

std::string DNDS::pybind11_ArrayDOF_name

(

)

Definition at line 15 of file ArrayDOF_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayEigenMatrix_declare()

template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>

tPy_ArrayEigenMatrix< _mat_ni, _mat_nj, _mat_ni_max, _mat_nj_max, _align > DNDS::pybind11_ArrayEigenMatrix_declare

(

py::module_ &

m

)

Definition at line 91 of file ArrayEigenMatrix_bind.hpp.

pybind11_ArrayEigenMatrix_define()

template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>

void DNDS::pybind11_ArrayEigenMatrix_define

(

py::module_ &

m

)

Definition at line 111 of file ArrayEigenMatrix_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayEigenMatrix_get_class()

template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>

tPy_ArrayEigenMatrix< _mat_ni, _mat_nj, _mat_ni_max, _mat_nj_max, _align > DNDS::pybind11_ArrayEigenMatrix_get_class

(

py::module_ &

m

)

Definition at line 104 of file ArrayEigenMatrix_bind.hpp.

pybind11_ArrayEigenMatrix_getitem()

template<class TArrayEigenMatrix = ArrayEigenMatrix<3, 3>>

auto DNDS::pybind11_ArrayEigenMatrix_getitem	(	TArrayEigenMatrix &	self,
		index	index_
	)

Definition at line 47 of file ArrayEigenMatrix_bind.hpp.

Here is the caller graph for this function:

pybind11_ArrayEigenMatrix_name()

template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>

std::string DNDS::pybind11_ArrayEigenMatrix_name

(

)

Definition at line 23 of file ArrayEigenMatrix_bind.hpp.

pybind11_ArrayEigenMatrix_name_appends()

template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>

std::string DNDS::pybind11_ArrayEigenMatrix_name_appends

(

)

Definition at line 11 of file ArrayEigenMatrix_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayEigenMatrix_setitem()

template<class TArrayEigenMatrix = ArrayEigenMatrix<3, 3>>

auto DNDS::pybind11_ArrayEigenMatrix_setitem	(	TArrayEigenMatrix &	self,
		index	index_,
		py::buffer	row
	)

Definition at line 59 of file ArrayEigenMatrix_bind.hpp.

Here is the caller graph for this function:

pybind11_ArrayEigenMatrixBatch_declare()

tPy_ArrayEigenMatrixBatch DNDS::pybind11_ArrayEigenMatrixBatch_declare ( py::module_ &  m )

inline

Definition at line 113 of file ArrayEigenMatrixBatch_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_ArrayEigenMatrixBatch_define()

void DNDS::pybind11_ArrayEigenMatrixBatch_define ( py::module_ &  m )

inline

Definition at line 125 of file ArrayEigenMatrixBatch_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_ArrayEigenMatrixBatch_get_class()

tPy_ArrayEigenMatrixBatch DNDS::pybind11_ArrayEigenMatrixBatch_get_class ( py::module_ &  m )

inline

Definition at line 120 of file ArrayEigenMatrixBatch_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayEigenMatrixBatch_getitem()

auto DNDS::pybind11_ArrayEigenMatrixBatch_getitem ( ArrayEigenMatrixBatch &  self, std::tuple< index, rowsize >  index_  )

inline

Definition at line 74 of file ArrayEigenMatrixBatch_bind.hpp.

Here is the caller graph for this function:

pybind11_ArrayEigenMatrixBatch_getitem_row()

auto DNDS::pybind11_ArrayEigenMatrixBatch_getitem_row ( ArrayEigenMatrixBatch &  self, index  index_  )

inline

warning: deleting list and Array could make the items dangling

Definition at line 55 of file ArrayEigenMatrixBatch_bind.hpp.

Here is the caller graph for this function:

pybind11_ArrayEigenMatrixBatch_name()

std::string DNDS::pybind11_ArrayEigenMatrixBatch_name ( )

inline

Definition at line 9 of file ArrayEigenMatrixBatch_bind.hpp.

Here is the caller graph for this function:

pybind11_ArrayEigenMatrixBatch_setitem()

auto DNDS::pybind11_ArrayEigenMatrixBatch_setitem ( ArrayEigenMatrixBatch &  self, std::tuple< index, rowsize >  index_, py::buffer  row  )

inline

Definition at line 86 of file ArrayEigenMatrixBatch_bind.hpp.

Here is the caller graph for this function:

pybind11_ArrayEigenMatrixBatch_setitem_row()

auto DNDS::pybind11_ArrayEigenMatrixBatch_setitem_row ( ArrayEigenMatrixBatch &  self, index  i, const py::list &  matList  )

inline

Definition at line 26 of file ArrayEigenMatrixBatch_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_ArrayEigenMatrixBatchPair_declare()

tPy_ArrayEigenMatrixBatchPair DNDS::pybind11_ArrayEigenMatrixBatchPair_declare ( py::module_ &  m )

inline

Definition at line 185 of file ArrayEigenMatrixBatch_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_ArrayEigenMatrixBatchPair_define()

void DNDS::pybind11_ArrayEigenMatrixBatchPair_define ( py::module_ &  m )

inline

Definition at line 196 of file ArrayEigenMatrixBatch_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_ArrayEigenMatrixBatchPair_get_class()

tPy_ArrayEigenMatrixBatchPair DNDS::pybind11_ArrayEigenMatrixBatchPair_get_class ( py::module_ &  m )

inline

Definition at line 191 of file ArrayEigenMatrixBatch_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayEigenMatrixBatchPair_name()

std::string DNDS::pybind11_ArrayEigenMatrixBatchPair_name ( )

inline

Definition at line 14 of file ArrayEigenMatrixBatch_bind.hpp.

Here is the caller graph for this function:

pybind11_ArrayEigenMatrixPair_declare()

template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>

tPy_ArrayEigenMatrixPair< _mat_ni, _mat_nj, _mat_ni_max, _mat_nj_max, _align > DNDS::pybind11_ArrayEigenMatrixPair_declare

(

py::module_ &

m

)

Definition at line 184 of file ArrayEigenMatrix_bind.hpp.

pybind11_ArrayEigenMatrixPair_define()

template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>

void DNDS::pybind11_ArrayEigenMatrixPair_define

(

py::module_ &

m

)

maybe should not use here

Definition at line 199 of file ArrayEigenMatrix_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayEigenMatrixPair_get_class()

template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>

tPy_ArrayEigenMatrixPair< _mat_ni, _mat_nj, _mat_ni_max, _mat_nj_max, _align > DNDS::pybind11_ArrayEigenMatrixPair_get_class

(

py::module_ &

m

)

Definition at line 192 of file ArrayEigenMatrix_bind.hpp.

pybind11_ArrayEigenMatrixPair_name()

template<rowsize _mat_ni = 1, rowsize _mat_nj = 1, rowsize _mat_ni_max = _mat_ni, rowsize _mat_nj_max = _mat_nj, rowsize _align = NoAlign>

std::string DNDS::pybind11_ArrayEigenMatrixPair_name

(

)

Definition at line 30 of file ArrayEigenMatrix_bind.hpp.

pybind11_ArrayEigenUniMatrixBatch_declare()

template<int _n_row, int _n_col>

tPy_ArrayEigenUniMatrixBatch< _n_row, _n_col > DNDS::pybind11_ArrayEigenUniMatrixBatch_declare

(

py::module_ &

m

)

Definition at line 110 of file ArrayEigenUniMatrixBatch_bind.hpp.

pybind11_ArrayEigenUniMatrixBatch_define()

template<int _n_row, int _n_col>

void DNDS::pybind11_ArrayEigenUniMatrixBatch_define

(

py::module_ &

m

)

Definition at line 125 of file ArrayEigenUniMatrixBatch_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayEigenUniMatrixBatch_get_class()

template<int _n_row, int _n_col>

tPy_ArrayEigenUniMatrixBatch< _n_row, _n_col > DNDS::pybind11_ArrayEigenUniMatrixBatch_get_class

(

py::module_ &

m

)

Definition at line 119 of file ArrayEigenUniMatrixBatch_bind.hpp.

pybind11_ArrayEigenUniMatrixBatch_getitem()

template<class TArrayEigenUniMatrixBatch = ArrayEigenUniMatrixBatch<3, 3>>

auto DNDS::pybind11_ArrayEigenUniMatrixBatch_getitem	(	TArrayEigenUniMatrixBatch &	self,
		std::tuple< index, rowsize >	index_
	)

Definition at line 35 of file ArrayEigenUniMatrixBatch_bind.hpp.

Here is the caller graph for this function:

pybind11_ArrayEigenUniMatrixBatch_getitem_row()

template<class TArrayEigenUniMatrixBatch = ArrayEigenUniMatrixBatch<3, 3>>

auto DNDS::pybind11_ArrayEigenUniMatrixBatch_getitem_row	(	TArrayEigenUniMatrixBatch &	self,
		index	index_
	)

warning, assume mat0.rowStride() applies to following matrices

Definition at line 70 of file ArrayEigenUniMatrixBatch_bind.hpp.

Here is the caller graph for this function:

pybind11_ArrayEigenUniMatrixBatch_name()

template<int _n_row, int _n_col>

std::string DNDS::pybind11_ArrayEigenUniMatrixBatch_name

(

)

Definition at line 10 of file ArrayEigenUniMatrixBatch_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayEigenUniMatrixBatch_setitem()

template<class TArrayEigenUniMatrixBatch = ArrayEigenUniMatrixBatch<3, 3>>

auto DNDS::pybind11_ArrayEigenUniMatrixBatch_setitem	(	TArrayEigenUniMatrixBatch &	self,
		std::tuple< index, rowsize >	index_,
		py::buffer	row
	)

Definition at line 47 of file ArrayEigenUniMatrixBatch_bind.hpp.

Here is the caller graph for this function:

pybind11_ArrayEigenUniMatrixBatch_setitem_row()

template<class TArrayEigenUniMatrixBatch = ArrayEigenUniMatrixBatch<3, 3>>

auto DNDS::pybind11_ArrayEigenUniMatrixBatch_setitem_row	(	TArrayEigenUniMatrixBatch &	self,
		index	index_,
		py::buffer	row
	)

Definition at line 83 of file ArrayEigenUniMatrixBatch_bind.hpp.

Here is the caller graph for this function:

pybind11_ArrayEigenUniMatrixBatchPair_declare()

template<int _n_row, int _n_col>

tPy_ArrayEigenUniMatrixBatchPair< _n_row, _n_col > DNDS::pybind11_ArrayEigenUniMatrixBatchPair_declare

(

py::module_ &

m

)

Definition at line 215 of file ArrayEigenUniMatrixBatch_bind.hpp.

pybind11_ArrayEigenUniMatrixBatchPair_define()

template<int _n_row, int _n_col>

void DNDS::pybind11_ArrayEigenUniMatrixBatchPair_define

(

py::module_ &

m

)

Definition at line 228 of file ArrayEigenUniMatrixBatch_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayEigenUniMatrixBatchPair_get_class()

template<int _n_row, int _n_col>

tPy_ArrayEigenUniMatrixBatchPair< _n_row, _n_col > DNDS::pybind11_ArrayEigenUniMatrixBatchPair_get_class

(

py::module_ &

m

)

Definition at line 222 of file ArrayEigenUniMatrixBatch_bind.hpp.

pybind11_ArrayEigenUniMatrixBatchPair_name()

template<int _n_row, int _n_col>

std::string DNDS::pybind11_ArrayEigenUniMatrixBatchPair_name

(

)

Definition at line 18 of file ArrayEigenUniMatrixBatch_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayEigenVector_declare()

template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>

tPy_ArrayEigenVector< _vec_size, _row_max, _align > DNDS::pybind11_ArrayEigenVector_declare

(

py::module_ &

m

)

Definition at line 66 of file ArrayEigenVector_bind.hpp.

pybind11_ArrayEigenVector_define()

template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>

void DNDS::pybind11_ArrayEigenVector_define

(

py::module_ &

m

)

Definition at line 81 of file ArrayEigenVector_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayEigenVector_get_class()

template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>

tPy_ArrayEigenVector< _vec_size, _row_max, _align > DNDS::pybind11_ArrayEigenVector_get_class

(

py::module_ &

m

)

Definition at line 75 of file ArrayEigenVector_bind.hpp.

pybind11_ArrayEigenVector_getitem()

template<class TArrayEigenVector = ArrayEigenVector<1>>

auto DNDS::pybind11_ArrayEigenVector_getitem	(	TArrayEigenVector &	self,
		index	index_
	)

Definition at line 40 of file ArrayEigenVector_bind.hpp.

Here is the caller graph for this function:

pybind11_ArrayEigenVector_name()

template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>

std::string DNDS::pybind11_ArrayEigenVector_name

(

)

Definition at line 18 of file ArrayEigenVector_bind.hpp.

pybind11_ArrayEigenVector_name_appends()

template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>

std::string DNDS::pybind11_ArrayEigenVector_name_appends

(

)

Definition at line 9 of file ArrayEigenVector_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayEigenVector_setitem()

template<class TArrayEigenVector = ArrayEigenVector<1>>

auto DNDS::pybind11_ArrayEigenVector_setitem	(	TArrayEigenVector &	self,
		index	index_,
		py::buffer	row
	)

Definition at line 51 of file ArrayEigenVector_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_ArrayEigenVectorPair_declare()

template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>

tPy_ArrayEigenVectorPair< _vec_size, _row_max, _align > DNDS::pybind11_ArrayEigenVectorPair_declare

(

py::module_ &

m

)

Definition at line 133 of file ArrayEigenVector_bind.hpp.

pybind11_ArrayEigenVectorPair_define()

template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>

void DNDS::pybind11_ArrayEigenVectorPair_define

(

py::module_ &

m

)

Definition at line 146 of file ArrayEigenVector_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayEigenVectorPair_get_class()

template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>

tPy_ArrayEigenVector< _vec_size, _row_max, _align > DNDS::pybind11_ArrayEigenVectorPair_get_class

(

py::module_ &

m

)

Definition at line 140 of file ArrayEigenVector_bind.hpp.

pybind11_ArrayEigenVectorPair_name()

template<rowsize _vec_size = 1, rowsize _row_max = _vec_size, rowsize _align = NoAlign>

std::string DNDS::pybind11_ArrayEigenVectorPair_name

(

)

Definition at line 24 of file ArrayEigenVector_bind.hpp.

pybind11_ArrayPairGenericBindBasics()

template<class TPair , class TPy_Pair >

void DNDS::pybind11_ArrayPairGenericBindBasics

(

TPy_Pair &

Pair_

)

Definition at line 326 of file Array_bind.hpp.

Here is the call graph for this function:

pybind11_ArrayTransformer_declare()

template<class TArray >

tPy_ArrayTransformer< TArray > DNDS::pybind11_ArrayTransformer_declare

(

py::module_ &

m

)

Definition at line 450 of file Array_bind.hpp.

pybind11_ArrayTransformer_define()

template<class TArray >

void DNDS::pybind11_ArrayTransformer_define

(

py::module_ &

m

)

Definition at line 464 of file Array_bind.hpp.

pybind11_ArrayTransformer_get_class()

template<class TArray >

tPy_ArrayTransformer< TArray > DNDS::pybind11_ArrayTransformer_get_class

(

py::module_ &

m

)

Definition at line 458 of file Array_bind.hpp.

pybind11_ArrayTransformer_name()

template<class TArray >

std::string DNDS::pybind11_ArrayTransformer_name

(

)

Definition at line 53 of file Array_bind.hpp.

pybind11_bind_Array_All()

void DNDS::pybind11_bind_Array_All

(

py::module_

m

)

Definition at line 5 of file Array_bind.cpp.

Here is the caller graph for this function:

pybind11_bind_Array_All_X_declare() [1/7]

DNDS::pybind11_bind_Array_All_X_declare

(

1

)

pybind11_bind_Array_All_X_declare() [2/7]

DNDS::pybind11_bind_Array_All_X_declare

(

2

)

pybind11_bind_Array_All_X_declare() [3/7]

DNDS::pybind11_bind_Array_All_X_declare

(

3

)

pybind11_bind_Array_All_X_declare() [4/7]

DNDS::pybind11_bind_Array_All_X_declare

(

4

)

pybind11_bind_Array_All_X_declare() [5/7]

DNDS::pybind11_bind_Array_All_X_declare

(

5

)

pybind11_bind_Array_All_X_declare() [6/7]

DNDS::pybind11_bind_Array_All_X_declare

(

6

)

pybind11_bind_Array_All_X_declare() [7/7]

DNDS::pybind11_bind_Array_All_X_declare

(

7

)

pybind11_bind_Array_Offsets()

void DNDS::pybind11_bind_Array_Offsets ( py::module_  m )

inline

Definition at line 639 of file Array_bind.hpp.

Here is the caller graph for this function:

pybind11_bind_ArrayAdjacency_All()

void DNDS::pybind11_bind_ArrayAdjacency_All

(

py::module_ &

m

)

Definition at line 5 of file ArrayAdjacency_bind.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_bind_ArrayDOF_All()

void DNDS::pybind11_bind_ArrayDOF_All

(

py::module_ &

m

)

Definition at line 5 of file ArrayDOF_bind.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_bind_ArrayEigenMatrix_All()

void DNDS::pybind11_bind_ArrayEigenMatrix_All

(

py::module_ &

m

)

Definition at line 5 of file ArrayEigenMatrix_bind.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_bind_ArrayEigenMatrixBatch_All()

void DNDS::pybind11_bind_ArrayEigenMatrixBatch_All ( py::module_ &  m )

inline

Definition at line 246 of file ArrayEigenMatrixBatch_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_bind_ArrayEigenUniMatrixBatch_All()

void DNDS::pybind11_bind_ArrayEigenUniMatrixBatch_All

(

py::module_ &

m

)

Definition at line 5 of file ArrayEigenUniMatrixBatch_bind.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_bind_ArrayEigenVector_All()

void DNDS::pybind11_bind_ArrayEigenVector_All

(

py::module_ &

m

)

Definition at line 5 of file ArrayEigenVector_bind.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_bind_defines()

void DNDS::pybind11_bind_defines ( py::module_ &  m )

inline

Definition at line 86 of file Defines_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_bind_device_controls()

void DNDS::pybind11_bind_device_controls ( py::module_ &  m )

inline

Definition at line 27 of file DeviceStorage_bind.hpp.

pybind11_bind_deviceStorage()

void DNDS::pybind11_bind_deviceStorage ( py::module_ &  m )

inline

Definition at line 10 of file DeviceStorage_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_bind_IndexMapping_All()

void DNDS::pybind11_bind_IndexMapping_All ( py::module_  m )

inline

Definition at line 149 of file IndexMapping_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_bind_MPI_All()

void DNDS::pybind11_bind_MPI_All

(

py::module_ &

m

)

Definition at line 135 of file MPI_bind.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_callBindArrayAdjacencys_rowsizes()

void DNDS::pybind11_callBindArrayAdjacencys_rowsizes ( py::module_ &  m )

inline

Definition at line 209 of file ArrayAdjacency_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_callBindArrayDOF_rowsizes()

template<rowsize mat_n>

void DNDS::pybind11_callBindArrayDOF_rowsizes

(

py::module_ &

m

)

Definition at line 138 of file ArrayDOF_bind.hpp.

Here is the call graph for this function:

pybind11_callBindArrayDOF_rowsizes< 1 >()

template void DNDS::pybind11_callBindArrayDOF_rowsizes< 1 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayDOF_rowsizes< 2 >()

template void DNDS::pybind11_callBindArrayDOF_rowsizes< 2 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayDOF_rowsizes< 3 >()

template void DNDS::pybind11_callBindArrayDOF_rowsizes< 3 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayDOF_rowsizes< 4 >()

template void DNDS::pybind11_callBindArrayDOF_rowsizes< 4 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayDOF_rowsizes< 5 >()

template void DNDS::pybind11_callBindArrayDOF_rowsizes< 5 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayDOF_rowsizes< 6 >()

template void DNDS::pybind11_callBindArrayDOF_rowsizes< 6 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayDOF_rowsizes< 7 >()

template void DNDS::pybind11_callBindArrayDOF_rowsizes< 7 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayDOF_rowsizes< 8 >()

template void DNDS::pybind11_callBindArrayDOF_rowsizes< 8 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayDOF_rowsizes< DynamicSize >()

template void DNDS::pybind11_callBindArrayDOF_rowsizes< DynamicSize > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayDOF_rowsizes< NonUniformSize >()

template void DNDS::pybind11_callBindArrayDOF_rowsizes< NonUniformSize > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenMatrixs_rowsizes()

template<rowsize mat_n>

void DNDS::pybind11_callBindArrayEigenMatrixs_rowsizes

(

py::module_ &

m

)

Definition at line 256 of file ArrayEigenMatrix_bind.hpp.

Here is the call graph for this function:

pybind11_callBindArrayEigenMatrixs_rowsizes< 1 >()

template void DNDS::pybind11_callBindArrayEigenMatrixs_rowsizes< 1 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenMatrixs_rowsizes< 2 >()

template void DNDS::pybind11_callBindArrayEigenMatrixs_rowsizes< 2 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenMatrixs_rowsizes< 3 >()

template void DNDS::pybind11_callBindArrayEigenMatrixs_rowsizes< 3 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenMatrixs_rowsizes< 4 >()

template void DNDS::pybind11_callBindArrayEigenMatrixs_rowsizes< 4 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenMatrixs_rowsizes< 5 >()

template void DNDS::pybind11_callBindArrayEigenMatrixs_rowsizes< 5 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenMatrixs_rowsizes< 6 >()

template void DNDS::pybind11_callBindArrayEigenMatrixs_rowsizes< 6 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenMatrixs_rowsizes< 7 >()

template void DNDS::pybind11_callBindArrayEigenMatrixs_rowsizes< 7 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenMatrixs_rowsizes< 8 >()

template void DNDS::pybind11_callBindArrayEigenMatrixs_rowsizes< 8 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenMatrixs_rowsizes< DynamicSize >()

template void DNDS::pybind11_callBindArrayEigenMatrixs_rowsizes< DynamicSize > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenMatrixs_rowsizes< NonUniformSize >()

template void DNDS::pybind11_callBindArrayEigenMatrixs_rowsizes< NonUniformSize > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes()

template<rowsize mat_n>

void DNDS::pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes

(

py::module_ &

m

)

Definition at line 304 of file ArrayEigenUniMatrixBatch_bind.hpp.

Here is the call graph for this function:

pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 1 >()

template void DNDS::pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 1 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 2 >()

template void DNDS::pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 2 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 3 >()

template void DNDS::pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 3 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 4 >()

template void DNDS::pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 4 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 5 >()

template void DNDS::pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 5 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 6 >()

template void DNDS::pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 6 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 7 >()

template void DNDS::pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 7 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 8 >()

template void DNDS::pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< 8 > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< DynamicSize >()

template void DNDS::pybind11_callBindArrayEigenUniMatrixBatchs_rowsizes< DynamicSize > ( py::module_ &  m )

extern

Here is the caller graph for this function:

pybind11_callBindArrayEigenVectors_rowsizes()

void DNDS::pybind11_callBindArrayEigenVectors_rowsizes ( py::module_ &  m )

inline

Definition at line 202 of file ArrayEigenVector_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_callBindArrays_rowsizes()

template<class T , int offset = 0>

void DNDS::pybind11_callBindArrays_rowsizes

(

py::module_ &

m

)

Definition at line 570 of file Array_bind.hpp.

Here is the call graph for this function:

pybind11_callBindArrayTransformers_rowsizes()

template<class T , int offset = 0>

void DNDS::pybind11_callBindArrayTransformers_rowsizes

(

py::module_ &

m

)

Definition at line 598 of file Array_bind.hpp.

Here is the call graph for this function:

pybind11_callBindParArrayPairs_rowsizes()

template<class T , int offset = 0>

void DNDS::pybind11_callBindParArrayPairs_rowsizes

(

py::module_ &

m

)

Definition at line 612 of file Array_bind.hpp.

Here is the call graph for this function:

pybind11_callBindParArrays_rowsizes()

template<class T , int offset = 0>

void DNDS::pybind11_callBindParArrays_rowsizes

(

py::module_ &

m

)

Definition at line 584 of file Array_bind.hpp.

Here is the call graph for this function:

pybind11_GlobalOffsetsMapping_declare()

auto DNDS::pybind11_GlobalOffsetsMapping_declare ( py::module_  m )

inline

Definition at line 15 of file IndexMapping_bind.hpp.

Here is the caller graph for this function:

pybind11_GlobalOffsetsMapping_define()

void DNDS::pybind11_GlobalOffsetsMapping_define ( py::module_  m )

inline

Definition at line 25 of file IndexMapping_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_GlobalOffsetsMapping_get_class()

auto DNDS::pybind11_GlobalOffsetsMapping_get_class ( py::module_  m )

inline

Definition at line 20 of file IndexMapping_bind.hpp.

pybind11_MPIInfo()

void DNDS::pybind11_MPIInfo

(

py::module_ &

m

)

Definition at line 10 of file MPI_bind.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_OffsetAscendIndexMapping_declare()

auto DNDS::pybind11_OffsetAscendIndexMapping_declare ( py::module_  m )

inline

Definition at line 53 of file IndexMapping_bind.hpp.

Here is the caller graph for this function:

pybind11_OffsetAscendIndexMapping_define()

void DNDS::pybind11_OffsetAscendIndexMapping_define ( py::module_  m )

inline

Definition at line 63 of file IndexMapping_bind.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

pybind11_OffsetAscendIndexMapping_get_class()

auto DNDS::pybind11_OffsetAscendIndexMapping_get_class ( py::module_  m )

inline

Definition at line 58 of file IndexMapping_bind.hpp.

pybind11_ParArray_declare()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

tPy_ParArray< T, _row_size, _row_max, _align > DNDS::pybind11_ParArray_declare

(

py::module_ &

m

)

Definition at line 237 of file Array_bind.hpp.

pybind11_ParArray_define()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

void DNDS::pybind11_ParArray_define

(

py::module_ &

m

)

Definition at line 257 of file Array_bind.hpp.

pybind11_ParArray_get_class()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

tPy_ParArray< T, _row_size, _row_max, _align > DNDS::pybind11_ParArray_get_class

(

py::module_ &

m

)

Definition at line 251 of file Array_bind.hpp.

pybind11_ParArray_name()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

std::string DNDS::pybind11_ParArray_name

(

)

Definition at line 47 of file Array_bind.hpp.

pybind11_ParArrayPair_declare()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

tPy_ParArrayPair< T, _row_size, _row_max, _align > DNDS::pybind11_ParArrayPair_declare

(

py::module_ &

m

)

Definition at line 310 of file Array_bind.hpp.

pybind11_ParArrayPair_define()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

void DNDS::pybind11_ParArrayPair_define

(

py::module_ &

m

)

Definition at line 402 of file Array_bind.hpp.

pybind11_ParArrayPair_get_class()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

tPy_ParArrayPair< T, _row_size, _row_max, _align > DNDS::pybind11_ParArrayPair_get_class

(

py::module_ &

m

)

Definition at line 320 of file Array_bind.hpp.

pybind11_ParArrayPair_name()

template<class T , rowsize _row_size = 1, rowsize _row_max = _row_size, rowsize _align = NoAlign>

std::string DNDS::pybind11_ParArrayPair_name

(

)

Definition at line 59 of file Array_bind.hpp.

RedistributeArrayWithTransformer()

template<class TArray >

void DNDS::RedistributeArrayWithTransformer	(	const MPIInfo &	mpi,
		ssp< TArray >	readFather,
		const std::vector< index > &	readOrigIndex,
		const std::vector< index > &	newOrigIndex,
		ssp< TArray >	outFather
	)

Redistributes an ArrayPair from a source partition to a target partition.

Template-free core logic: given a "read" father array (from even-split H5 read) with known origIndex, and a target partition's newOrigIndex, uses ArrayTransformer to pull data from wherever it was read to where it's needed.

Template Parameters

TArray

The ParArray type (e.g., ParArray<real, DynamicSize>)

Parameters

mpi	MPI communicator info
readFather	Father array populated from even-split H5 read
readOrigIndex	Original indices for each row of readFather
newOrigIndex	Original indices this rank needs (from current partition)
outFather	Output father array, pre-allocated with correct size and row structure

Definition at line 235 of file ArrayRedistributor.hpp.

Here is the call graph for this function:

RegisterSignalHandler()

void DNDS::RegisterSignalHandler ( )

inline

Install SEGV / ABRT handlers that print a backtrace via DNDS_signal_handler.

Optional opt-in; not installed automatically so unit test drivers keep full control of their own handlers. Typical drivers call this once after MPI_Init in main().

Definition at line 55 of file Defines.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

RowSize_To_EigenSize()

constexpr int DNDS::RowSize_To_EigenSize ( rowsize  rs )

constexpr

Convert a rowsize constant to the corresponding Eigen compile-time size. Fixed >= 0 -> the value; DynamicSize / NonUniformSize -> Eigen::Dynamic.

Definition at line 286 of file Defines.hpp.

RowSize_To_PySnippet()

std::string DNDS::RowSize_To_PySnippet ( rowsize  rs )

inline

Encode a rowsize constant as a short Python-binding snippet: "<number>" for fixed, "D" for DynamicSize, "I" for NonUniformSize. Used when generating pybind11 class names.

Definition at line 294 of file Defines.hpp.

Here is the caller graph for this function:

schemaTypeString()

std::string DNDS::schemaTypeString ( ConfigTypeTag  tag )

inline

Definition at line 166 of file ConfigParam.hpp.

Here is the caller graph for this function:

setLogStream()

void DNDS::setLogStream

(

ssp< std::ostream >

nstream

)

Redirect log() output to a user-supplied stream. Ownership is shared.

Definition at line 53 of file Defines.cpp.

setLogStreamCout()

void DNDS::setLogStreamCout

(

)

Restore the default std::cout routing for log().

Definition at line 55 of file Defines.cpp.

sign()

constexpr real DNDS::sign ( real  a )

constexpr

Signum function: +1, 0, or -1.

Definition at line 532 of file Defines.hpp.

Here is the caller graph for this function:

signedIntSafeAdd()

template<typename T >

T DNDS::signedIntSafeAdd	(	T	value,
		T	increment
	)

Add two signed integers, asserting on overflow instead of silently wrapping.

Exceptions

Dies	via DNDS_assert_info if the result would overflow.

Definition at line 394 of file Defines.hpp.

signedIntWillAddOverflow()

template<typename T >

bool DNDS::signedIntWillAddOverflow	(	T	value,
		T	increment
	)

Overflow-detecting test for value + increment on signed integers.

Returns

true if the addition would overflow.

Definition at line 380 of file Defines.hpp.

signM()

constexpr real DNDS::signM ( real  a )

constexpr

"Signum, biased toward -1": treats 0 as negative.

Definition at line 550 of file Defines.hpp.

Here is the caller graph for this function:

signP()

constexpr real DNDS::signP ( real  a )

constexpr

"Signum, biased toward +1": treats 0 as positive.

Definition at line 544 of file Defines.hpp.

Here is the caller graph for this function:

signTol()

constexpr real DNDS::signTol ( real  a, real  tol  )

constexpr

Tolerant signum: returns 0 inside [-tol, tol].

Definition at line 538 of file Defines.hpp.

Here is the caller graph for this function:

size_t_to_signed()

template<typename T >

T DNDS::size_t_to_signed

(

size_t

v

)

Narrowing size_t -> T conversion with range check.

Template Parameters

T	Signed integral target type.

Exceptions

Dies	via DNDS_assert_info if v > std::numeric_limits<T>::max().

Definition at line 409 of file Defines.hpp.

size_to_index()

index DNDS::size_to_index ( size_t  v )

inline

Range-checked conversion from size_t to DNDS::index.

Exceptions

std::runtime_error

(via DNDS_assert_info) if value exceeds index max. Use this for safe conversions from container sizes (e.g., std::vector::size()) to DNDS index types, avoiding sign comparison warnings and overflow bugs.

Definition at line 426 of file Defines.hpp.

Here is the caller graph for this function:

size_to_rowsize()

rowsize DNDS::size_to_rowsize ( size_t  v )

inline

Range-checked conversion from size_t to DNDS::rowsize.

Exceptions

std::runtime_error

(via DNDS_assert_info) if value exceeds rowsize max.

Definition at line 433 of file Defines.hpp.

splitSString()

std::vector< std::string > DNDS::splitSString ( const std::string &  str, char  delim  )

inline

Definition at line 772 of file Defines.hpp.

Here is the caller graph for this function:

splitSStringClean()

std::vector< std::string > DNDS::splitSStringClean ( const std::string &  str, char  delim  )

inline

Definition at line 790 of file Defines.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

sqr()

template<typename T >

constexpr T DNDS::sqr ( const T &  a )

constexpr

a * a, constexpr. Works for all arithmetic types.

Definition at line 517 of file Defines.hpp.

Here is the caller graph for this function:

sstringHasSuffix()

bool DNDS::sstringHasSuffix ( const std::string &  str, const std::string &  suffix  )

inline

Definition at line 800 of file Defines.hpp.

Here is the caller graph for this function:

Timer()

PerformanceTimer & DNDS::Timer ( )

inline

Short-hand accessor to the PerformanceTimer singleton.

Definition at line 136 of file Profiling.hpp.

Here is the call graph for this function:

Here is the caller graph for this function:

to_json()

void DNDS::to_json ( nlohmann::ordered_json &  j, const host_device_vector< real > &  v  )

inline

Definition at line 177 of file JsonUtil.hpp.

Here is the caller graph for this function:

to_string()

template<class dir >

std::string DNDS::to_string	(	const Eigen::DenseBase< dir > &	v,
		int	precision = 5,
		bool	scientific = false
	)

Render an Eigen::DenseBase to a string via operator<<.

Parameters

precision	Setprecision applied if > 0.
scientific	Whether to use std::scientific notation.

Definition at line 29 of file EigenUtil.hpp.

Variable Documentation

__EigenPCH

const char* DNDS::__EigenPCH = "EigenPCH"

Definition at line 5 of file EigenPCH.cpp.

__ExprtkPCH

const char* DNDS::__ExprtkPCH = "ExprtkPCH"

Definition at line 5 of file ExprtkPCH.cpp.

DNDS_MPI_INDEX

const MPI_Datatype DNDS::DNDS_MPI_INDEX = __DNDSToMPITypeInt<index>()

MPI datatype matching index (= MPI_INT64_T).

Definition at line 90 of file MPI.hpp.

DNDS_MPI_REAL

const MPI_Datatype DNDS::DNDS_MPI_REAL = __DNDSToMPITypeFloat<real>()

MPI datatype matching real (= MPI_REAL8).

Definition at line 92 of file MPI.hpp.

DynamicSize

DNDS_CONSTANT const rowsize DNDS::DynamicSize = -1

Template parameter flag: "row width is set at runtime but uniform".

Definition at line 277 of file Defines.hpp.

HDF_mutex

std::mutex DNDS::HDF_mutex

Global mutex serialising host-side HDF5 calls.

HDF5 is not thread-safe by default; this mutex guards all DNDSR HDF5 I/O when multiple CPU threads might touch the same file.

Definition at line 449 of file MPI.cpp.

indexMin

DNDS_CONSTANT const index DNDS::indexMin = INT64_MIN

Minimum representable index value (= INT64_MIN).

Definition at line 170 of file Defines.hpp.

is_ssp_v

template<typename T >

constexpr bool DNDS::is_ssp_v = is_ssp<T>::value

inlineconstexpr

Definition at line 152 of file Defines.hpp.

largeReal

DNDS_CONSTANT const real DNDS::largeReal = 3e10

Loose upper bound (e.g., for non-dimensional limits).

Definition at line 192 of file Defines.hpp.

logStream

ssp< std::ostream > DNDS::logStream

Shared output stream: where DNDS::log() writes progress / diagnostics.

Defaults to std::cout when useCout is true; otherwise refers to the file/stream installed via setLogStream.

Definition at line 45 of file Defines.cpp.

NoAlign

DNDS_CONSTANT const rowsize DNDS::NoAlign = -1024

Alignment flag: no padding applied to rows (the only currently-supported value).

Definition at line 282 of file Defines.hpp.

NonUniformSize

DNDS_CONSTANT const rowsize DNDS::NonUniformSize = -2

Template parameter flag: "each row has an independent width".

Definition at line 279 of file Defines.hpp.

pi

DNDS_CONSTANT const real DNDS::pi = DNDS_E_PI

π in double precision (matches DNDS_E_PI macro).

Definition at line 199 of file Defines.hpp.

smallReal

DNDS_CONSTANT const real DNDS::smallReal = 1e-10

Loose lower bound (for iterative-solver tolerances etc.).

Definition at line 196 of file Defines.hpp.

UnInitIndex

DNDS_CONSTANT const index DNDS::UnInitIndex = INT64_MIN

Sentinel "not initialised" index value (= INT64_MIN).

Definition at line 176 of file Defines.hpp.

UnInitReal

DNDS_CONSTANT const real DNDS::UnInitReal = NAN

Sentinel "not initialised" real value (NaN). Cheap to detect with std::isnan or IsUnInitReal; survives MPI transport unchanged.

Definition at line 174 of file Defines.hpp.

UnInitRowsize

DNDS_CONSTANT const rowsize DNDS::UnInitRowsize = INT32_MIN

Sentinel "not initialised" rowsize value (= INT32_MIN).

Definition at line 179 of file Defines.hpp.

useCout

bool DNDS::useCout = true

Whether DNDS::log() is currently routed to std::cout.

Definition at line 47 of file Defines.cpp.

veryLargeReal

DNDS_CONSTANT const real DNDS::veryLargeReal = 3e200

Catch-all upper bound ("practically infinity") for physical scalars.

Definition at line 190 of file Defines.hpp.

verySmallReal

DNDS_CONSTANT const real DNDS::verySmallReal = 1e-200

Catch-all lower bound ("effectively zero").

Definition at line 194 of file Defines.hpp.