Routines needed for kpoint calculation. More...

Data Types
type	kp_transform_plan_type

type	rskp_grid_type

Functions/Subroutines
subroutine, public	kpoint_initialize (kpoint, particle_set, cell)
	Generate the kpoints and initialize the kpoint environment.

subroutine, public	kpoint_env_initialize (kpoint, para_env, blacs_env, with_aux_fit)
	Initialize the kpoint environment.

subroutine, public	kpoint_initialize_mos (kpoint, mos, added_mos, for_aux_fit)
	Initialize a set of MOs and density matrix for each kpoint (kpoint group)

subroutine, public	kpoint_initialize_mo_set (kpoint)
	...

subroutine, public	kpoint_init_cell_index (kpoint, sab_nl, para_env, nimages)
	Generates the mapping of cell indices and linear RS index CELL (0,0,0) is always mapped to index 1.

subroutine, public	rskp_transform (rmatrix, cmatrix, rsmat, ispin, xkp, cell_to_index, sab_nl, is_complex, rs_sign)
	Transformation of real space matrices to a kpoint.

subroutine, public	rskp_transform_grid_prepare (grid, rmatrix, rsmat, ispin, xkp, nkp_grid, cell_to_index, sab_nl, used_fft, is_complex, rs_sign, max_storage_bytes)
	Prepare a batched real-cell to complete reciprocal-grid transform for local DBCSR blocks. The stored blocks remain MPI local. Irregular or too-large grids return used_fft=.FALSE.

subroutine, public	rskp_transform_grid_extract (grid, ikp, rmatrix, cmatrix)
	Extract one reciprocal-grid matrix from a prepared local DBCSR block cache.

subroutine, public	rskp_transform_grid_release (grid)
	Release a reciprocal-grid transformation cache.

subroutine, public	kpoint_set_mo_occupation (kpoint, smear, probe, added_mos_auto, added_mos_auto_grow, separate_spin_occupations)
	Given the eigenvalues of all kpoints, calculates the occupation numbers.

subroutine, public	kpoint_smearing_edge_status (wocc, wkp, smear, nspin, has_weight, first_fractional, last_occupied, last_occupied_spin)
	summarize whether weighted k-point smearing reaches the available band edges

subroutine, public	kpoint_density_matrices (kpoint, energy_weighted, for_aux_fit)
	Calculate kpoint density matrices (rho(k), owned by kpoint groups)

subroutine, public	kpoint_ot_energy_weighted_density (coeff_re, coeff_im, hc_re, hc_im, occupation, wmat_re, wmat_im)
	Build W(k) for noncanonical complex OT orbitals from H(k) C(k). The occupied-space Lagrange multiplier is Hermitian[(C^H H C) f], which reduces to the usual eigenvalue-weighted expression for canonical orbitals.

subroutine, public	kpoint_ot_energy_weighted_matrices (kpoint, matrix_ks)
	Build energy-weighted density matrices for complex K-point OT. H(k) is transformed only while W(k) is needed, avoiding a persistent dense AO-by-AO Hamiltonian for every spin and K point.

subroutine, public	lowdin_kp_trans (kpoint, pmat_diag)
	Calculate Lowdin transformation of density matrix S^1/2 P S^1/2 Integrate diagonal elements over k-points to get Lowdin charges.

subroutine, public	lowdin_kp_mo_coeff (kp, ispin, use_real_wfn, shalfc, cshalfc)
	Calculate S(k)^1/2 C(k) for real or complex k-point wavefunctions.

subroutine, public	kpoint_density_transform (kpoint, denmat, wtype, tempmat, sab_nl, fmwork, for_aux_fit, pmat_ext, overlap_rs)
	generate real space density matrices in DBCSR format

subroutine, public	kp_transform_plan_create (plan, sab_nl, cell_to_index, nimg, block_template, group_entries)
	Build the immutable neighbor-list traversal shared by R-to-K and K-to-R transforms.

subroutine, public	kp_transform_plan_release (plan)
	Release a K-to-R traversal plan.

Detailed Description

Routines needed for kpoint calculation.

History: 2014.07 created [JGH] 2014.11 unified k-point and gamma-point code [Ole Schuett]

Author: JGH

Function/Subroutine Documentation

◆ kpoint_initialize()

subroutine, public kpoint_methods::kpoint_initialize	(	type(kpoint_type), pointer	kpoint,
		type(particle_type), dimension(:), pointer	particle_set,
		type(cell_type), pointer	cell
	)

Generate the kpoints and initialize the kpoint environment.

Parameters

kpoint	The kpoint environment
particle_set	Particle types and coordinates
cell	Computational cell information

Definition at line 155 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ kpoint_env_initialize()

subroutine, public kpoint_methods::kpoint_env_initialize	(	type(kpoint_type), intent(inout)	kpoint,
		type(mp_para_env_type), intent(in), target	para_env,
		type(cp_blacs_env_type), intent(in), target	blacs_env,
		logical, intent(in), optional	with_aux_fit
	)

Initialize the kpoint environment.

Parameters

kpoint	Kpoint environment
para_env	...
blacs_env	...
with_aux_fit	...

Definition at line 656 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ kpoint_initialize_mos()

subroutine, public kpoint_methods::kpoint_initialize_mos	(	type(kpoint_type), pointer	kpoint,
		type(mo_set_type), dimension(:), intent(inout)	mos,
		integer, intent(in), optional	added_mos,
		logical, optional	for_aux_fit
	)

Initialize a set of MOs and density matrix for each kpoint (kpoint group)

Parameters

kpoint	Kpoint environment
mos	Reference MOs (global)
added_mos	...
for_aux_fit	...

Definition at line 844 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ kpoint_initialize_mo_set()

subroutine, public kpoint_methods::kpoint_initialize_mo_set ( type(kpoint_type), pointer kpoint )

...

Parameters

kpoint ...

Definition at line 1000 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ kpoint_init_cell_index()

subroutine, public kpoint_methods::kpoint_init_cell_index	(	type(kpoint_type), pointer	kpoint,
		type(neighbor_list_set_p_type), dimension(:), pointer	sab_nl,
		type(mp_para_env_type), pointer	para_env,
		integer, intent(out)	nimages
	)

Generates the mapping of cell indices and linear RS index CELL (0,0,0) is always mapped to index 1.

Parameters

kpoint	Kpoint environment
sab_nl	Defining neighbour list
para_env	Parallel environment
nimages	[output]

Definition at line 1040 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ rskp_transform()

subroutine, public kpoint_methods::rskp_transform	(	type(dbcsr_type)	rmatrix,
		type(dbcsr_type), optional	cmatrix,
		type(dbcsr_p_type), dimension(:, :), pointer	rsmat,
		integer, intent(in)	ispin,
		real(kind=dp), dimension(3), intent(in)	xkp,
		integer, dimension(:, :, :), pointer	cell_to_index,
		type(neighbor_list_set_p_type), dimension(:), pointer	sab_nl,
		logical, intent(in), optional	is_complex,
		real(kind=dp), intent(in), optional	rs_sign
	)

Transformation of real space matrices to a kpoint.

Parameters

rmatrix	Real part of kpoint matrix
cmatrix	Complex part of kpoint matrix (optional)
rsmat	Real space matrices
ispin	Spin index
xkp	Kpoint coordinates
cell_to_index	mapping of cell indices to RS index
sab_nl	Defining neighbor list
is_complex	Matrix to be transformed is imaginary
rs_sign	Matrix to be transformed is csaled by rs_sign

Definition at line 1173 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ rskp_transform_grid_prepare()

subroutine, public kpoint_methods::rskp_transform_grid_prepare	(	type(rskp_grid_type), intent(inout)	grid,
		type(dbcsr_type), intent(inout)	rmatrix,
		type(dbcsr_p_type), dimension(:, :), pointer	rsmat,
		integer, intent(in)	ispin,
		real(kind=dp), dimension(:, :), intent(in)	xkp,
		integer, dimension(3), intent(in)	nkp_grid,
		integer, dimension(:, :, :), pointer	cell_to_index,
		type(neighbor_list_set_p_type), dimension(:), pointer	sab_nl,
		logical, intent(out)	used_fft,
		logical, intent(in), optional	is_complex,
		real(kind=dp), intent(in), optional	rs_sign,
		integer(kind=int_8), intent(in), optional	max_storage_bytes
	)

Prepare a batched real-cell to complete reciprocal-grid transform for local DBCSR blocks. The stored blocks remain MPI local. Irregular or too-large grids return used_fft=.FALSE.

Parameters

grid	cached transformed local blocks and their DBCSR coordinates
rmatrix	allocated output template defining the local block layout
rsmat	real-space matrix set
ispin	spin component of rsmat
xkp	complete reciprocal grid in arbitrary order
nkp_grid	reciprocal grid dimensions
cell_to_index	real-cell coordinate mapping
sab_nl	neighbor list defining the stored block orientation
used_fft	whether the cache was prepared
is_complex	whether the real-space operator is imaginary
rs_sign	optional overall sign
max_storage_bytes	optional conservative per-rank memory limit

Definition at line 1284 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ rskp_transform_grid_extract()

subroutine, public kpoint_methods::rskp_transform_grid_extract	(	type(rskp_grid_type), intent(in)	grid,
		integer, intent(in)	ikp,
		type(dbcsr_type), intent(inout)	rmatrix,
		type(dbcsr_type), intent(inout), optional	cmatrix
	)

Extract one reciprocal-grid matrix from a prepared local DBCSR block cache.

Parameters

grid	...
ikp	...
rmatrix	...
cmatrix	...

Definition at line 1474 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ rskp_transform_grid_release()

subroutine, public kpoint_methods::rskp_transform_grid_release ( type(rskp_grid_type), intent(inout) grid )

Release a reciprocal-grid transformation cache.

Parameters

grid ...

Definition at line 1512 of file kpoint_methods.F.

Here is the caller graph for this function:

◆ kpoint_set_mo_occupation()

subroutine, public kpoint_methods::kpoint_set_mo_occupation	(	type(kpoint_type), pointer	kpoint,
		type(smear_type)	smear,
		type(hairy_probes_type), dimension(:), optional, pointer	probe,
		logical, dimension(:), intent(in), optional	added_mos_auto,
		logical, intent(out), optional	added_mos_auto_grow,
		logical, intent(in), optional	separate_spin_occupations
	)

Given the eigenvalues of all kpoints, calculates the occupation numbers.

Parameters

kpoint	Kpoint environment
smear	Smearing information
probe	...
added_mos_auto	...
added_mos_auto_grow	...
separate_spin_occupations	enforce the electron count of each spin channel

Definition at line 1534 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ kpoint_smearing_edge_status()

subroutine, public kpoint_methods::kpoint_smearing_edge_status	(	real(kind=dp), dimension(:, :, :), intent(in)	wocc,
		real(kind=dp), dimension(:), intent(in)	wkp,
		type(smear_type), intent(in)	smear,
		integer, intent(in)	nspin,
		logical, intent(out)	has_weight,
		logical, intent(out)	first_fractional,
		logical, intent(out)	last_occupied,
		logical, dimension(:), intent(out), optional	last_occupied_spin
	)

summarize whether weighted k-point smearing reaches the available band edges

Parameters

wocc	...
wkp	...
smear	...
nspin	...
has_weight	...
first_fractional	...
last_occupied	...
last_occupied_spin	...

Definition at line 1819 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ kpoint_density_matrices()

subroutine, public kpoint_methods::kpoint_density_matrices	(	type(kpoint_type), pointer	kpoint,
		logical, optional	energy_weighted,
		logical, optional	for_aux_fit
	)

Calculate kpoint density matrices (rho(k), owned by kpoint groups)

Parameters

kpoint	kpoint environment
energy_weighted	calculate energy weighted density matrix
for_aux_fit	...

Definition at line 1917 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ kpoint_ot_energy_weighted_density()

subroutine, public kpoint_methods::kpoint_ot_energy_weighted_density	(	type(cp_fm_type), intent(in)	coeff_re,
		type(cp_fm_type), intent(in)	coeff_im,
		type(cp_fm_type), intent(in)	hc_re,
		type(cp_fm_type), intent(in)	hc_im,
		real(kind=dp), dimension(:), intent(in)	occupation,
		type(cp_fm_type), intent(in)	wmat_re,
		type(cp_fm_type), intent(in)	wmat_im
	)

Build W(k) for noncanonical complex OT orbitals from H(k) C(k). The occupied-space Lagrange multiplier is Hermitian[(C^H H C) f], which reduces to the usual eigenvalue-weighted expression for canonical orbitals.

Parameters

coeff_re	real part of C(k)
coeff_im	imaginary part of C(k)
hc_re	real part of H(k) C(k)
hc_im	imaginary part of H(k) C(k)
occupation	orbital occupations
wmat_re	real part of W(k)
wmat_im	imaginary part of W(k)

Definition at line 2057 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ kpoint_ot_energy_weighted_matrices()

subroutine, public kpoint_methods::kpoint_ot_energy_weighted_matrices	(	type(kpoint_type), pointer	kpoint,
		type(dbcsr_p_type), dimension(:, :), pointer	matrix_ks
	)

Build energy-weighted density matrices for complex K-point OT. H(k) is transformed only while W(k) is needed, avoiding a persistent dense AO-by-AO Hamiltonian for every spin and K point.

Parameters

kpoint	K-point environment
matrix_ks	real-space Kohn-Sham matrices

Definition at line 2115 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ lowdin_kp_trans()

subroutine, public kpoint_methods::lowdin_kp_trans	(	type(kpoint_type), pointer	kpoint,
		real(kind=dp), dimension(:, :), intent(inout)	pmat_diag
	)

Calculate Lowdin transformation of density matrix S^1/2 P S^1/2 Integrate diagonal elements over k-points to get Lowdin charges.

Parameters

kpoint	kpoint environment
pmat_diag	Sum over kpoints of diagonal elements

History: 04.2026 created [JGH]

Definition at line 2325 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ lowdin_kp_mo_coeff()

subroutine, public kpoint_methods::lowdin_kp_mo_coeff	(	type(kpoint_env_type), pointer	kp,
		integer, intent(in)	ispin,
		logical, intent(in)	use_real_wfn,
		type(cp_fm_type), intent(inout), optional	shalfc,
		type(cp_cfm_type), intent(inout), optional	cshalfc
	)

Calculate S(k)^1/2 C(k) for real or complex k-point wavefunctions.

Parameters

kp	K-point environment for one local k point
ispin	Spin index
use_real_wfn	Use real k-point wavefunctions
shalfc	Output matrix containing S(k)^1/2 C(k) for real wavefunctions
cshalfc	Output matrix containing S(k)^1/2 C(k) for complex wavefunctions

Definition at line 2412 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ kpoint_density_transform()

subroutine, public kpoint_methods::kpoint_density_transform	(	type(kpoint_type), pointer	kpoint,
		type(dbcsr_p_type), dimension(:, :)	denmat,
		logical, intent(in)	wtype,
		type(dbcsr_type), pointer	tempmat,
		type(neighbor_list_set_p_type), dimension(:), pointer	sab_nl,
		type(cp_fm_type), dimension(:), intent(in)	fmwork,
		logical, optional	for_aux_fit,
		type(cp_fm_type), dimension(:, :, :), intent(in), optional	pmat_ext,
		type(dbcsr_p_type), dimension(:, :), optional, pointer	overlap_rs
	)

generate real space density matrices in DBCSR format

Parameters

kpoint	Kpoint environment
denmat	Real space (DBCSR) density matrices
wtype	True = energy weighted density matrix False = normal density matrix
tempmat	DBCSR matrix to be used as template
sab_nl	...
fmwork	FM work matrices (kpoint group)
for_aux_fit	...
pmat_ext	...
overlap_rs	...

Definition at line 2482 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ kp_transform_plan_create()

subroutine, public kpoint_methods::kp_transform_plan_create	(	type(kp_transform_plan_type), intent(out)	plan,
		type(neighbor_list_set_p_type), dimension(:), pointer	sab_nl,
		integer, dimension(:, :, :), pointer	cell_to_index,
		integer, intent(in)	nimg,
		type(dbcsr_type), intent(in), optional	block_template,
		logical, intent(in), optional	group_entries
	)

Build the immutable neighbor-list traversal shared by R-to-K and K-to-R transforms.

Parameters

plan	...
sab_nl	...
cell_to_index	...
nimg	...
block_template	...
group_entries	...

Definition at line 2875 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ kp_transform_plan_release()

subroutine, public kpoint_methods::kp_transform_plan_release ( type(kp_transform_plan_type), intent(inout) plan )

Release a K-to-R traversal plan.

Parameters

plan ...

Definition at line 2988 of file kpoint_methods.F.

Here is the call graph for this function:

Here is the caller graph for this function:

Data Types

Functions/Subroutines

Detailed Description

Function/Subroutine Documentation

◆ kpoint_initialize()

◆ kpoint_env_initialize()

◆ kpoint_initialize_mos()

◆ kpoint_initialize_mo_set()

◆ kpoint_init_cell_index()

◆ rskp_transform()

◆ rskp_transform_grid_prepare()

◆ rskp_transform_grid_extract()

◆ rskp_transform_grid_release()

◆ kpoint_set_mo_occupation()

◆ kpoint_smearing_edge_status()

◆ kpoint_density_matrices()

◆ kpoint_ot_energy_weighted_density()

◆ kpoint_ot_energy_weighted_matrices()

◆ lowdin_kp_trans()

◆ lowdin_kp_mo_coeff()

◆ kpoint_density_transform()

◆ kp_transform_plan_create()

◆ kp_transform_plan_release()