Holds displacements for applying operators to avoid replicating for all operators. More...

#include <displacements.h>

Collaboration diagram for madness::Displacements< NDIM >:

[legend]

Classes
struct	DispEntry

struct	PeriodicLists
	the displacements for kernels lattice-summed along one set of axes, per level, in the two orders consumers use More...

Public Member Functions
	Displacements ()

const std::vector< Key< NDIM > > &	get_disp ()
	return the standard displacements appropriate for operators w/o lattice summation

const std::vector< Key< NDIM > > &	get_disp (Level n, const array_of_bools< NDIM > &kernel_lattice_sum_axes)

const std::vector< Key< NDIM > > &	get_disp_images (Level n, const array_of_bools< NDIM > &kernel_lattice_sum_axes)

Static Public Member Functions
static int	bmax_default ()

static void	make_disp (int bmax)

static void	make_disp_images (PeriodicLists &lists)
	fills lists.by_image_distance (requires the lock)

static void	make_disp_periodic (int bmax, Level n, PeriodicLists &lists)
	fills lists.by_distance[n] for lists.axes

static PeriodicLists &	periodic (const array_of_bools< NDIM > &axes)
	the lists for kernels lattice-summed along `axes`, built on the first request

static void	reset_periodic_axes (const array_of_bools< NDIM > &periodic_axes)
	builds the displacements for kernels lattice-summed along `periodic_axes` if they do not exist yet

static void	set_width (const Tensor< double > &width)

static void	sort_displacements (std::vector< Key< NDIM > > &d, const Tensor< double > &w)

static void	sort_displacements_images (std::vector< Key< NDIM > > &d, const array_of_bools< NDIM > &paxes, const Tensor< double > &w)

static void	sort_displacements_periodic (std::vector< Key< NDIM > > &d, const array_of_bools< NDIM > &paxes, const Tensor< double > &w)

Static Private Member Functions
static std::size_t	axes_mask (const array_of_bools< NDIM > &axes)

Static Private Attributes
static std::vector< Key< NDIM > >	disp = {}
	standard displacements to be used with standard kernels (range-unrestricted, no lattice sum)

static constexpr Level	nlevels = 8 * sizeof(Translation) - 2
	levels for which lists are built

static std::array< std::atomic< PeriodicLists * >,(std::size_t(1)<< NDIM)>	periodic_lists {}

static Mutex	periodic_lists_mutex {}
	serializes building the lists and set_width()

static std::vector< std::unique_ptr< PeriodicLists > >	periodic_lists_owner {}
	owns what periodic_lists points to

static Tensor< double >	widths {NDIM}
	cell width, used to order displacements from least to most real space distance

Detailed Description

template<std::size_t NDIM>
class madness::Displacements< NDIM >

Holds displacements for applying operators to avoid replicating for all operators.

Constructor & Destructor Documentation

◆ Displacements()

template<std::size_t NDIM>

madness::Displacements< NDIM >::Displacements ( )

inline

The first call generates the standard displacements and, if the default boundary conditions have periodic axes, the displacements for kernels lattice-summed along those axes. Lists for any other set of lattice-summed axes are built on their first request (get_disp, get_disp_images), which SeparatedConvolution makes when it is constructed.

References madness::Displacements< NDIM >::bmax_default(), madness::Displacements< NDIM >::disp, e(), madness::FunctionDefaults< NDIM >::get_bc(), MADNESS_PRAGMA_CLANG, madness::Displacements< NDIM >::make_disp(), NDIM, madness::Tensor< T >::normf(), madness::Displacements< NDIM >::periodic(), madness::pop(), and madness::Displacements< NDIM >::widths.

Member Function Documentation

◆ axes_mask()

template<std::size_t NDIM>

static std::size_t madness::Displacements< NDIM >::axes_mask ( const array_of_bools< NDIM > & axes )

inlinestaticprivate

References d, mask, and NDIM.

Referenced by madness::Displacements< NDIM >::periodic().

◆ bmax_default()

template<std::size_t NDIM>

static int madness::Displacements< NDIM >::bmax_default ( )

inlinestatic

References NDIM.

Referenced by madness::Displacements< NDIM >::Displacements(), madness::BoxSurfaceDisplacementRange< NDIM >::compute_probing_displacement(), madness::BoxSurfaceDisplacementValidator< NDIM >::operator()(), and madness::Displacements< NDIM >::periodic().

◆ get_disp() [1/2]

template<std::size_t NDIM>

const std::vector< Key< NDIM > > & madness::Displacements< NDIM >::get_disp ( )

inline

return the standard displacements appropriate for operators w/o lattice summation

References madness::Displacements< NDIM >::disp, MADNESS_PRAGMA_CLANG, and madness::pop().

◆ get_disp() [2/2]

template<std::size_t NDIM>

const std::vector< Key< NDIM > > & madness::Displacements< NDIM >::get_disp	(	Level	n,
		const array_of_bools< NDIM > &	kernel_lattice_sum_axes
	)

inline

the displacements for a kernel lattice-summed along kernel_lattice_sum_axes (the standard displacements if none), ordered by the real-space distance to the source modulo the lattice (Key::real_distsq_bc), as FunctionImpl::do_apply screens a kernel that decays away from the source

References madness::array_of_bools< N >::any(), madness::Displacements< NDIM >::PeriodicLists::by_distance, madness::Displacements< NDIM >::disp, MADNESS_CHECK, MADNESS_PRAGMA_CLANG, madness::Displacements< NDIM >::nlevels, madness::Displacements< NDIM >::periodic(), and madness::pop().

Referenced by madness::FunctionImpl< T, NDIM >::do_apply_directed_screening(), and madness::SeparatedConvolution< Q, NDIM >::get_disp().

◆ get_disp_images()

template<std::size_t NDIM>

const std::vector< Key< NDIM > > & madness::Displacements< NDIM >::get_disp_images	(	Level	n,
		const array_of_bools< NDIM > &	kernel_lattice_sum_axes
	)

inline

the displacements of get_disp(n, kernel_lattice_sum_axes) ordered by the distance to the nearest lattice image other than the home cell (Key::real_distsq_images), for kernels that sum only those images (OperatorInfo::images_only); built on the first request for these axes

References madness::Displacements< NDIM >::PeriodicLists::by_image_distance, madness::Displacements< NDIM >::PeriodicLists::images_built, MADNESS_CHECK, MADNESS_PRAGMA_CLANG, madness::Displacements< NDIM >::make_disp_images(), madness::Displacements< NDIM >::nlevels, madness::Displacements< NDIM >::periodic(), madness::Displacements< NDIM >::periodic_lists_mutex, and madness::pop().

Referenced by madness::SeparatedConvolution< Q, NDIM >::get_disp().

◆ make_disp()

template<std::size_t NDIM>

static void madness::Displacements< NDIM >::make_disp ( int bmax )

inlinestatic

References d, madness::Displacements< NDIM >::disp, MADNESS_EXCEPTION, NDIM, madness::Displacements< NDIM >::sort_displacements(), and madness::Displacements< NDIM >::widths.

Referenced by madness::Displacements< NDIM >::Displacements().

◆ make_disp_images()

template<std::size_t NDIM>

static void madness::Displacements< NDIM >::make_disp_images ( PeriodicLists & lists )

inlinestatic

fills lists.by_image_distance (requires the lock)

References madness::Displacements< NDIM >::PeriodicLists::axes, madness::Displacements< NDIM >::PeriodicLists::by_distance, madness::Displacements< NDIM >::PeriodicLists::by_image_distance, madness::Displacements< NDIM >::nlevels, madness::Displacements< NDIM >::sort_displacements_images(), and madness::Displacements< NDIM >::widths.

Referenced by madness::Displacements< NDIM >::get_disp_images(), and madness::Displacements< NDIM >::set_width().

◆ make_disp_periodic()

template<std::size_t NDIM>

static void madness::Displacements< NDIM >::make_disp_periodic	(	int	bmax,
		Level	n,
		PeriodicLists &	lists
	)

inlinestatic

fills lists.by_distance[n] for lists.axes

References madness::array_of_bools< N >::any(), madness::Displacements< NDIM >::PeriodicLists::axes, madness::Displacements< NDIM >::PeriodicLists::by_distance, d, MADNESS_ASSERT, MADNESS_PRAGMA_CLANG, NDIM, madness::pop(), madness::Displacements< NDIM >::sort_displacements_periodic(), and madness::Displacements< NDIM >::widths.

Referenced by madness::Displacements< NDIM >::periodic().

◆ periodic()

template<std::size_t NDIM>

static PeriodicLists & madness::Displacements< NDIM >::periodic ( const array_of_bools< NDIM > & axes )

inlinestatic

the lists for kernels lattice-summed along axes, built on the first request

References madness::array_of_bools< N >::any(), madness::Displacements< NDIM >::axes_mask(), madness::Displacements< NDIM >::bmax_default(), MADNESS_CHECK_THROW, MADNESS_PRAGMA_CLANG, madness::Displacements< NDIM >::make_disp_periodic(), NDIM, madness::Displacements< NDIM >::nlevels, p(), madness::Displacements< NDIM >::periodic_lists, madness::Displacements< NDIM >::periodic_lists_mutex, madness::Displacements< NDIM >::periodic_lists_owner, and madness::pop().

Referenced by madness::Displacements< NDIM >::Displacements(), madness::Displacements< NDIM >::get_disp(), madness::Displacements< NDIM >::get_disp_images(), and madness::Displacements< NDIM >::reset_periodic_axes().

◆ reset_periodic_axes()

template<std::size_t NDIM>

static void madness::Displacements< NDIM >::reset_periodic_axes ( const array_of_bools< NDIM > & periodic_axes )

inlinestatic

builds the displacements for kernels lattice-summed along periodic_axes if they do not exist yet

Lists are built per set of lattice-summed axes on first request, so this is not needed for correctness; it moves the build (all levels, a few tens of milliseconds) out of the first apply.

Parameters

periodic_axes the lattice-summed axes

References madness::array_of_bools< N >::any(), MADNESS_ASSERT, and madness::Displacements< NDIM >::periodic().

◆ set_width()

template<std::size_t NDIM>

static void madness::Displacements< NDIM >::set_width ( const Tensor< double > & width )

inlinestatic

Sets the cell widths used to order the displacements by real-space distance, and reorders them if the widths changed. Called by FunctionDefaults::recompute_cell_info whenever the cell is set.

Warning: reorders the lists in place: must not be called while operators are being applied (see FunctionDefaults::set_cell)

Referenced by madness::FunctionDefaults< NDIM >::recompute_cell_info(), test_opdir(), and testgradG().

◆ sort_displacements()

template<std::size_t NDIM>

static void madness::Displacements< NDIM >::sort_displacements	(	std::vector< Key< NDIM > > &	d,
		const Tensor< double > &	w
	)

inlinestatic

References d, k, and w().

Referenced by madness::Displacements< NDIM >::make_disp(), and madness::Displacements< NDIM >::set_width().

◆ sort_displacements_images()

template<std::size_t NDIM>

static void madness::Displacements< NDIM >::sort_displacements_images	(	std::vector< Key< NDIM > > &	d,
		const array_of_bools< NDIM > &	paxes,
		const Tensor< double > &	w
	)

inlinestatic

orders by the distance to the nearest non-home image: least distance between the boxes, then boxes, then the distance between the box centers. The last breaks the tie among the displacements one box from an image, which are all at least distance 0: the narrowest summed axis comes first, and its block is the operator's largest (SeparatedConvolution::norm_bound visits the leading image-adjacent run).

References d, k, and w().

Referenced by madness::Displacements< NDIM >::make_disp_images().

◆ sort_displacements_periodic()

template<std::size_t NDIM>

static void madness::Displacements< NDIM >::sort_displacements_periodic	(	std::vector< Key< NDIM > > &	d,
		const array_of_bools< NDIM > &	paxes,
		const Tensor< double > &	w
	)

inlinestatic

References d, k, and w().

Referenced by madness::Displacements< NDIM >::make_disp_periodic(), and madness::Displacements< NDIM >::set_width().

Member Data Documentation

◆ disp

template<std::size_t NDIM>

std::vector< Key<NDIM> > madness::Displacements< NDIM >::disp = {}

inlinestaticprivate

standard displacements to be used with standard kernels (range-unrestricted, no lattice sum)

Referenced by madness::Displacements< NDIM >::Displacements(), madness::Displacements< NDIM >::get_disp(), madness::Displacements< NDIM >::get_disp(), madness::Displacements< NDIM >::make_disp(), and madness::Displacements< NDIM >::set_width().

◆ nlevels

template<std::size_t NDIM>

constexpr Level madness::Displacements< NDIM >::nlevels = 8 * sizeof(Translation) - 2

staticconstexprprivate

levels for which lists are built

Referenced by madness::Displacements< NDIM >::get_disp(), madness::Displacements< NDIM >::get_disp_images(), madness::Displacements< NDIM >::make_disp_images(), madness::Displacements< NDIM >::periodic(), and madness::Displacements< NDIM >::set_width().

◆ periodic_lists

template<std::size_t NDIM>

std::array<std::atomic<PeriodicLists*>, (std::size_t(1) << NDIM)> madness::Displacements< NDIM >::periodic_lists {}

inlinestaticprivate

the lists for each set of lattice-summed axes, indexed by the axes' bitmask; a slot is filled on the first request for its axes (see periodic()) and published through the atomic once complete, so lookups take no lock. Sorting the lists with the metric of the axes they are requested for is what makes the shell-wise screening of FunctionImpl::do_apply sound: a list sorted for a superset of the kernel's axes would place a displacement wrapped along an axis the kernel does not sum next to the source, where for that kernel it is a cell away. A set holds every level's list in each order it was asked for (for NDIM = 3 up to (4 bmax + 1)^3 keys per level, some ten MB per order), so a process that uses operators summed along several different axis sets pays that per set; before, one set served them all.

Referenced by madness::Displacements< NDIM >::periodic().

◆ periodic_lists_mutex

template<std::size_t NDIM>

Mutex madness::Displacements< NDIM >::periodic_lists_mutex {}

inlinestaticprivate

serializes building the lists and set_width()

Referenced by madness::Displacements< NDIM >::get_disp_images(), madness::Displacements< NDIM >::periodic(), and madness::Displacements< NDIM >::set_width().

◆ periodic_lists_owner

template<std::size_t NDIM>

std::vector<std::unique_ptr<PeriodicLists> > madness::Displacements< NDIM >::periodic_lists_owner {}

inlinestaticprivate

owns what periodic_lists points to

Referenced by madness::Displacements< NDIM >::periodic(), and madness::Displacements< NDIM >::set_width().

◆ widths

template<std::size_t NDIM>

Tensor<double> madness::Displacements< NDIM >::widths {NDIM}

inlinestaticprivate

cell width, used to order displacements from least to most real space distance

Referenced by madness::Displacements< NDIM >::Displacements(), madness::Displacements< NDIM >::make_disp(), madness::Displacements< NDIM >::make_disp_images(), madness::Displacements< NDIM >::make_disp_periodic(), and madness::Displacements< NDIM >::set_width().

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

displacements.h

Classes

Public Member Functions

Static Public Member Functions

Static Private Member Functions

Static Private Attributes

Detailed Description

Constructor & Destructor Documentation

◆ Displacements()

Member Function Documentation

◆ axes_mask()

◆ bmax_default()

◆ get_disp() [1/2]

◆ get_disp() [2/2]

◆ get_disp_images()

◆ make_disp()

◆ make_disp_images()

◆ make_disp_periodic()

◆ periodic()

◆ reset_periodic_axes()

◆ set_width()

◆ sort_displacements()

◆ sort_displacements_images()

◆ sort_displacements_periodic()

Member Data Documentation

◆ disp

◆ nlevels

◆ periodic_lists

◆ periodic_lists_mutex

◆ periodic_lists_owner

◆ widths