Opm::gpuistl::GpuDILU< M, X, Y, l > Class Template Reference

DILU preconditioner on the GPU. More...

#include <GpuDILU.hpp>

Inheritance diagram for Opm::gpuistl::GpuDILU< M, X, Y, l >:

Public Types
using	matrix_type = typename std::remove_const< M >::type
	The matrix type the preconditioner is for.
using	domain_type = X
	The domain type of the preconditioner.
using	range_type = Y
	The range type of the preconditioner.
using	field_type = typename X::field_type
	The field type of the preconditioner.
using	CuMat = GpuSparseMatrix< field_type >
	The GPU matrix type.
using	FloatMat = GpuSparseMatrix< float >
using	FloatVec = GpuVector< float >

Public Member Functions
	GpuDILU (const M &A, bool splitMatrix, bool tuneKernels, int mixedPrecisionScheme)
	Constructor.
void	pre (X &x, Y &b) override
	Prepare the preconditioner.
void	apply (X &v, const Y &d) override
	Apply the preconditoner.
void	post (X &x) override
	Post processing.
Dune::SolverCategory::Category	category () const override
	Category of the preconditioner (see SolverCategory::Category)
void	update () final
	Updates the matrix data.
void	reorderAndSplitMatrix (int moveThreadBlockSize)
	perform matrix splitting and reordering
void	computeDiagonal (int factorizationThreadBlockSize)
	Compute the diagonal of the DILU, and update the data of the reordered matrix.
void	tuneThreadBlockSizes ()
	function that will experimentally tune the thread block sizes of the important cuda kernels
virtual bool	hasPerfectUpdate () const override

Static Public Member Functions
static constexpr bool	shouldCallPre ()
static constexpr bool	shouldCallPost ()

Detailed Description

template<class M, class X, class Y, int l = 1>
class Opm::gpuistl::GpuDILU< M, X, Y, l >

DILU preconditioner on the GPU.

Template Parameters

M	The matrix type to operate on
X	Type of the update
Y	Type of the defect
l	Ignored. Just there to have the same number of template arguments as other preconditioners.

Note

We assume X and Y are both GpuVector<real_type>, but we leave them as template arguments in case of future additions.

Constructor & Destructor Documentation

GpuDILU()

template<class M , class X , class Y , int l>

Opm::gpuistl::GpuDILU< M, X, Y, l >::GpuDILU ( const M &  A, bool  splitMatrix, bool  tuneKernels, int  mixedPrecisionScheme  )

explicit

Constructor.

Constructor gets all parameters to operate the prec.

Parameters

A	The matrix to operate on.
w	The relaxation factor.

Member Function Documentation

hasPerfectUpdate()

template<class M , class X , class Y , int l = 1>

virtual bool Opm::gpuistl::GpuDILU< M, X, Y, l >::hasPerfectUpdate ( ) const

inlineoverridevirtual

Implements Dune::PreconditionerWithUpdate< X, Y >.

post()

template<class M , class X , class Y , int l>

void Opm::gpuistl::GpuDILU< M, X, Y, l >::post ( X &  x )

override

Post processing.

Note

Does nothing at the moment

pre()

template<class M , class X , class Y , int l>

void Opm::gpuistl::GpuDILU< M, X, Y, l >::pre ( X &  x, Y &  b  )

override

Prepare the preconditioner.

Note

Does nothing at the time being.

shouldCallPost()

template<class M , class X , class Y , int l = 1>

static constexpr bool Opm::gpuistl::GpuDILU< M, X, Y, l >::shouldCallPost ( )

inlinestaticconstexpr

Returns

false

shouldCallPre()

template<class M , class X , class Y , int l = 1>

static constexpr bool Opm::gpuistl::GpuDILU< M, X, Y, l >::shouldCallPre ( )

inlinestaticconstexpr

Returns

false

update()

template<class M , class X , class Y , int l>

void Opm::gpuistl::GpuDILU< M, X, Y, l >::update ( )

finalvirtual

Updates the matrix data.

Implements Dune::PreconditionerWithUpdate< X, Y >.

---

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

• opm/simulators/linalg/gpuistl/GpuDILU.hpp
• opm/simulators/linalg/gpuistl/GpuDILU.cpp