MPI-parallel distributed array of per-cell gradient matrices. More...

#include <Euler.hpp>

Inheritance diagram for DNDS::Euler::ArrayGRADV< nVarsFixed, gDim >:

Collaboration diagram for DNDS::Euler::ArrayGRADV< nVarsFixed, gDim >:

Public Types
using	t_self = ArrayGRADV< nVarsFixed, gDim >
	Self type alias.

using	t_base = CFV::tUGrad< nVarsFixed, gDim >
	Base distributed gradient array type.

Public Types inherited from DNDS::ArrayDof< n_m, n_n >
using	t_base = ArrayEigenMatrixPair< n_m, n_n >

template<DeviceBackend B>
using	t_deviceView = ArrayDofDeviceView< B, n_m, n_n >
	Mutable device view alias.

template<DeviceBackend B>
using	t_deviceViewConst = ArrayDofDeviceViewConst< B, n_m, n_n >
	Const device view alias.

using	t_self = ArrayDof< n_m, n_n >

template<DeviceBackend B>
using	t_ops = ArrayDofOp< B, n_m, n_n >
	Static dispatcher alias selecting host / CUDA implementation.

using	t_element_mat = Eigen::Matrix< real, RowSize_To_EigenSize(n_m), RowSize_To_EigenSize(n_n)>
	Shape of one DOF row as an Eigen matrix.

Public Types inherited from DNDS::ArrayPair< TArray >
using	t_self = ArrayPair< TArray >

using	t_arr = TArray

using	TTrans = typename ArrayTransformerType< TArray >::Type

template<DeviceBackend B>
using	t_deviceView = ArrayPairDeviceView< B, TArray >
	Device-view template alias: `t_deviceView<DeviceBackend::CUDA>` gives the mutable CUDA view type for this pair.

template<DeviceBackend B>
using	t_deviceViewConst = ArrayPairDeviceViewConst< B, TArray >
	Const-device-view template alias.

Public Member Functions
void	setConstant (real R)
	Set all gradient entries to a uniform scalar value.

template<class TR >
void	setConstant (const TR &R)
	Set all cells' gradient matrices to copies of a given matrix.

void	operator+= (t_self &R)
	Element-wise addition: `this[i] += R[i]` for every cell.

void	operator-= (t_self &R)
	Element-wise subtraction: `this[i] -= R[i]` for every cell.

void	operator*= (real R)
	Uniform scalar multiplication: `this[i] *= R` for every cell.

void	operator*= (std::vector< real > &R)
	Per-cell scalar multiplication from a vector of cell-wise weights.

void	operator*= (ArrayDOFV< 1 > &R)
	Component-wise multiplication by a scalar-per-cell ArrayDOFV<1>.

void	operator*= (const Eigen::Array< real, 1, nVarsFixed > &R)
	Column-wise scaling by a per-variable multiplier row-vector.

void	operator= (t_self &R)
	Deep copy assignment from another ArrayGRADV.

Public Member Functions inherited from DNDS::ArrayDof< n_m, n_n >
template<DeviceBackend B>
t_deviceView< B >	deviceView ()
	Build a mutable device view (wraps the base-class implementation).

template<DeviceBackend B>
t_deviceViewConst< B >	deviceView () const
	Build a const device view.

void	clone (const t_self &R)
	Deep copy from another ArrayDof. Delegates to the base-class clone.

void	setConstant (real R)
	Set every entry of every (father+son) row to the scalar `R`.

void	setConstant (const Eigen::Ref< const t_element_mat > &R)
	Set every row to the matrix `R` (must have shape `n_m x n_n`).

void	operator+= (const t_self &R)
	In-place element-wise add: `this += R`.

void	operator+= (real R)
	Add the scalar `R` to every entry.

void	operator+= (const Eigen::Ref< const t_element_mat > &R)
	Add a per-row matrix `R` (same to every row).

void	operator-= (const t_self &R)
	In-place element-wise subtract: `this -= R`.

void	operator*= (real R)
	Scalar multiply in place.

template<int n_m_T = n_m>
std::enable_if_t<!(n_m_T==1 &&n_n==1)>	operator*= (const ArrayDof< 1, 1 > &R)
	Scale each row by a corresponding scalar stored in `R` (a `1x1` ArrayDof).

void	operator*= (const Eigen::Ref< const t_element_mat > &R)
	In-place multiplication by a small fixed matrix (same applied to every row).

void	operator*= (const t_self &R)
	Element-wise multiply: `this *= R` (Hadamard).

void	operator/= (const t_self &R)
	Element-wise divide: `this /= R`.

void	operator= (const t_self &R)
	Value-copy assignment from another ArrayDof of identical layout.

void	addTo (const t_self &R, real r)
	AXPY: `this += r * R`. One of the hot-path solver primitives.

real	norm2 ()
	Global L2 norm (MPI-collective). `sqrt(sum_i sum_j x_ij^2)`.

real	norm2 (const t_self &R)
	Global L2 distance between `this` and `R` (collective).

real	min ()
	Global minimum across all entries (collective).

real	max ()
	Global maximum across all entries (collective).

real	sum ()
	Global sum of all entries (collective).

t_element_mat	componentWiseNorm1 ()
	Per-component global L1 norm, returned as an `n_m x n_n` matrix (collective).

t_element_mat	componentWiseNorm1 (const t_self &R)
	Per-component global L1 distance between `this` and `R` (collective).

real	dot (const t_self &R)
	Global inner product: `sum_i sum_j x_ij * R_ij` (collective).

void	to_device (DeviceBackend backend)
	Mirror both father and son to the given device backend.

void	to_host ()
	Bring both father and son mirrors back to host memory.

Public Member Functions inherited from DNDS::ArrayPair< TArray >
void	clone (const t_self &R)
	Deep-copy: allocate new father / son and copy their data; rebind trans.

decltype(father->operator[](index(0)))	operator[] (index i) const
	Read-only row-pointer access in the combined address space.

decltype(father->operator[](index(0)))	operator[] (index i)
	Mutable row-pointer access in the combined address space.

template<class... TOthers>
decltype(auto)	operator() (index i, TOthers... aOthers)
	N-ary element access in the combined space (mutable). Arguments after the row index are forwarded to the underlying `operator()`.

template<class... TOthers>
decltype(auto)	operator() (index i, TOthers... aOthers) const
	N-ary element access (const).

template<class TF >
auto	runFunctionAppendedIndex (index i, TF &&F)
	Invoke `F(array, localIndex)` on either father or son depending on which range `i` falls into.

auto	RowSize () const
	Uniform row width (delegates to father).

auto	RowSize (index i) const
	Per-row width in the combined address space.

void	ResizeRow (index i, rowsize rs)
	Resize a single row in the combined address space.

template<class... TOthers>
void	ResizeRow (index i, TOthers... aOthers)
	Variadic ResizeRow overload that forwards extra args.

index	Size () const
	Combined row count (`father->Size() + son->Size()`).

void	TransAttach ()
	Bind the transformer to the current father / son pointers.

template<typename... Args>
void	InitPair (const std::string &name, Args &&...args)
	Allocate both father and son arrays, forwarding all args to TArray constructor.

template<class TPrimaryPair >
void	BorrowAndPull (TPrimaryPair &primary)
	Attach, borrow ghost indexing from a primary pair, create MPI types, and pull once.

template<class TPrimaryPair >
void	BorrowSetup (TPrimaryPair &primary)
	Attach, borrow ghost indexing from a primary pair, and create MPI types (no pull).

void	CompressBoth ()
	Compress both father and son CSR arrays (no-op for non-CSR layouts).

void	CopyFather (t_self &R)
	Copy only the father's data from another pair (shallow).

void	SwapDataFatherSon (t_self &R)
	Swap both father and son data with another pair of the same type.

std::size_t	hash ()
	Combined hash across ranks. Used for determinism / equality checks in tests.

void	WriteSerialize (Serializer::SerializerBaseSSP serializerP, const std::string &name, bool includePIG=true, bool includeSon=true)
	Writes the ArrayPair (father, optional son, optional ghost mapping).

void	WriteSerialize (Serializer::SerializerBaseSSP serializerP, const std::string &name, const std::vector< index > &origIndex, bool includePIG=true, bool includeSon=true)
	Writes the ArrayPair with an origIndex companion dataset for redistribution support.

void	ReadSerialize (Serializer::SerializerBaseSSP serializerP, const std::string &name, bool includePIG=true, bool includeSon=true)
	Reads an ArrayPair written by WriteSerialize (same partition count).

void	ReadSerializeRedistributed (Serializer::SerializerBaseSSP serializerP, const std::string &name, const std::vector< index > &newOrigIndex)
	Reads ArrayPair data from HDF5 with redistribution support.

template<DeviceBackend B>
auto	deviceView ()
	Produce a mutable device view; both father and son must be allocated.

template<DeviceBackend B>
auto	deviceView () const
	Produce a const device view.

void	to_device (DeviceBackend backend)
	Mirror both father and son to the given device backend.

void	to_host ()
	Bring both father and son mirrors back to host memory.

Additional Inherited Members
Static Public Member Functions inherited from DNDS::ArrayPair< TArray >
static constexpr bool	IsCSR ()
	Whether the underlying array uses CSR storage.

Public Attributes inherited from DNDS::ArrayPair< TArray >
ssp< TArray >	father
	Owned-side array (must be resized before ghost setup).

ssp< TArray >	son
	Ghost-side array (sized automatically by createMPITypes / BorrowAndPull).

TTrans	trans
	Ghost-communication engine bound to father and son.

Detailed Description

template<int nVarsFixed, int gDim>
class DNDS::Euler::ArrayGRADV< nVarsFixed, gDim >

MPI-parallel distributed array of per-cell gradient matrices.

Wraps CFV::tUGrad<nVarsFixed, gDim> (which is ArrayDof<gDim, nVarsFixed>), adding element-wise arithmetic operators for gradient manipulation.

Each cell stores a (gDim x nVars) matrix where row d contains the partial derivative of all conservative variables with respect to spatial coordinate d:

\[ \mathbf{G}_i = \begin{bmatrix} \partial u_1/\partial x_1 & \cdots & \partial u_{nVars}/\partial x_1 \\ \vdots & \ddots & \vdots \\ \partial u_1/\partial x_{gDim} & \cdots & \partial u_{nVars}/\partial x_{gDim} \end{bmatrix} \]

Template Parameters

nVarsFixed	Compile-time number of conservative variables (matrix columns), or `Eigen::Dynamic` for runtime-sized.
gDim	Geometry (mesh) spatial dimension (2 or 3), determining the number of gradient rows.

See also: ArrayDOFV for per-cell DOF vectors.; ArrayRECV for reconstruction coefficient storage.

Definition at line 642 of file Euler.hpp.

Member Typedef Documentation

◆ t_base

template<int nVarsFixed, int gDim>

using DNDS::Euler::ArrayGRADV< nVarsFixed, gDim >::t_base = CFV::tUGrad<nVarsFixed, gDim>

Base distributed gradient array type.

Definition at line 646 of file Euler.hpp.

◆ t_self

template<int nVarsFixed, int gDim>

using DNDS::Euler::ArrayGRADV< nVarsFixed, gDim >::t_self = ArrayGRADV<nVarsFixed, gDim>

Self type alias.

Definition at line 645 of file Euler.hpp.

Member Function Documentation

◆ operator*=() [1/4]

template<int nVarsFixed, int gDim>

void DNDS::Euler::ArrayGRADV< nVarsFixed, gDim >::operator*= ( ArrayDOFV< 1 > & R )

inline

Component-wise multiplication by a scalar-per-cell ArrayDOFV<1>.

Each cell's gradient matrix is scaled by the single scalar stored in the corresponding cell of R.

Parameters

R	Scalar-per-cell array (ArrayDOFV<1>).

Definition at line 729 of file Euler.hpp.

Here is the call graph for this function:

◆ operator*=() [2/4]

template<int nVarsFixed, int gDim>

void DNDS::Euler::ArrayGRADV< nVarsFixed, gDim >::operator*= ( const Eigen::Array< real, 1, nVarsFixed > & R )

inline

Column-wise scaling by a per-variable multiplier row-vector.

Each column (variable) of every cell's gradient matrix is multiplied by the corresponding element of R: this[i].col(j) *= R(j).

Parameters

R	Row array of per-variable multipliers (1 x nVarsFixed).

Note: Only operates on host memory. Asserts if device storage is active.; OpenMP parallelized when DNDS_DIST_MT_USE_OMP is defined.

Definition at line 744 of file Euler.hpp.

Here is the call graph for this function:

◆ operator*=() [3/4]

template<int nVarsFixed, int gDim>

void DNDS::Euler::ArrayGRADV< nVarsFixed, gDim >::operator*= ( real R )

inline

Uniform scalar multiplication: this[i] *= R for every cell.

Parameters

R	Scalar multiplier applied to all entries of all cells.

Definition at line 697 of file Euler.hpp.

Here is the call graph for this function:

◆ operator*=() [4/4]

template<int nVarsFixed, int gDim>

void DNDS::Euler::ArrayGRADV< nVarsFixed, gDim >::operator*= ( std::vector< real > & R )

inline

Per-cell scalar multiplication from a vector of cell-wise weights.

Multiplies each cell's gradient matrix by the corresponding scalar in R: this[i] *= R[i]. Only operates on father (locally-owned) cells.

Parameters

R	Vector of per-cell scalar multipliers. Must have size >= number of locally-owned cells.

Note: OpenMP parallelized when DNDS_DIST_MT_USE_OMP is defined.

Definition at line 712 of file Euler.hpp.

Here is the call graph for this function:

◆ operator+=()

template<int nVarsFixed, int gDim>

void DNDS::Euler::ArrayGRADV< nVarsFixed, gDim >::operator+= ( t_self & R )

inline

Element-wise addition: this[i] += R[i] for every cell.

Parameters

R	Source gradient array with the same distribution layout.

Definition at line 681 of file Euler.hpp.

Here is the call graph for this function:

◆ operator-=()

template<int nVarsFixed, int gDim>

void DNDS::Euler::ArrayGRADV< nVarsFixed, gDim >::operator-= ( t_self & R )

inline

Element-wise subtraction: this[i] -= R[i] for every cell.

Parameters

R	Source gradient array with the same distribution layout.

Definition at line 689 of file Euler.hpp.

Here is the call graph for this function:

◆ operator=()

template<int nVarsFixed, int gDim>

void DNDS::Euler::ArrayGRADV< nVarsFixed, gDim >::operator= ( t_self & R )

inline

Deep copy assignment from another ArrayGRADV.

Parameters

R	Source array to copy from. Must have the same distribution layout.

Definition at line 758 of file Euler.hpp.

Here is the call graph for this function:

◆ setConstant() [1/2]

template<int nVarsFixed, int gDim>

template<class TR >

void DNDS::Euler::ArrayGRADV< nVarsFixed, gDim >::setConstant ( const TR & R )

inline

Set all cells' gradient matrices to copies of a given matrix.

Template Parameters

TR	Matrix type compatible with per-cell gradient storage.

Parameters

R	Reference matrix; each cell's gradient is set to this value.

Note: Only operates on host memory. Asserts if device storage is active.; OpenMP parallelized when DNDS_DIST_MT_USE_OMP is defined.

Definition at line 666 of file Euler.hpp.

Here is the call graph for this function:

◆ setConstant() [2/2]

template<int nVarsFixed, int gDim>

void DNDS::Euler::ArrayGRADV< nVarsFixed, gDim >::setConstant ( real R )

inline

Set all gradient entries to a uniform scalar value.

Parameters

R	Scalar value to fill every matrix entry with.

Definition at line 652 of file Euler.hpp.

Here is the call graph for this function:

The documentation for this class was generated from the following file:

/home/runner/work/DNDSR/DNDSR/src/Euler/Euler.hpp

Public Types

Public Member Functions

Additional Inherited Members

Detailed Description

Member Typedef Documentation

◆ t_base

◆ t_self

Member Function Documentation

◆ operator*=() [1/4]

◆ operator*=() [2/4]

◆ operator*=() [3/4]

◆ operator*=() [4/4]

◆ operator+=()

◆ operator-=()

◆ operator=()

◆ setConstant() [1/2]

◆ setConstant() [2/2]