THIS FILE IS AUTO GENERATED DONT MODIFY IT. More...

Detailed Description

THIS FILE IS AUTO GENERATED DONT MODIFY IT.

Namespaces
	coreneuron

	detail

	Instrumentor

	lfputils

	nrn_partrans

	scopmath

	units

Classes
struct	alloc_deleter
	Allocator-aware deleter for use with std::unique_ptr. More...

struct	BAMech

class	BinaryReportHandler

class	BinQ

struct	CellMapping
	Compartment mapping information for a cell. More...

class	CheckPoints

struct	cmp_str

struct	comp

class	ConditionEvent

class	CoreNeuron
	A class representing the CoreNEURON state, holding pointers to the various data structures. More...

struct	corenrn_parameters

struct	corenrn_parameters_data

struct	DiscreteEvent

struct	DoubScal

struct	DoubVec

struct	Elm

class	FileHandler

class	fixed_vector

class	InputPreSyn

class	InterleaveInfo

struct	InterThreadEvent

struct	Item

struct	less_time

struct	LFPCalculator
	LFPCalculator allows calculation of LFP given membrane currents. More...

struct	Memb_func

struct	Memb_list

struct	mpi_function

struct	mpi_function< std::integral_constant< function_ptr, fptr > >

struct	mpi_function_base

struct	mpi_manager_t

class	NetCon

class	NetCvode

class	NetCvodeThreadData

class	NetParEvent

struct	NetReceiveBuffer_t

struct	NetSendBuffer_t

struct	NewtonSpace

struct	NrnFastImem

struct	nrnmpi_init_ret_t

struct	NRNMPI_Spike

struct	NRNMPI_Spikebuf

struct	nrnran123_State

struct	NrnThread

struct	NrnThreadBAList

struct	NrnThreadChkpnt

struct	NrnThreadMappingInfo
	Compartment mapping information for NrnThread. More...

struct	NrnThreadMembList

class	Phase1

class	Phase2

class	PlayRecord

class	PlayRecordEvent

struct	Point_process

class	PreSyn

struct	PreSynHelper

struct	ReportConfiguration

class	ReportHandler

struct	SecMapping
	Section to segment mapping. More...

class	SelfEvent

class	SidInfo

class	SonataReportHandler

struct	SparseObj

struct	SpikesInfo

struct	SPTREE

struct	SummationReport

struct	SummationReportMapping

union	ThreadDatum

class	TNode
	TNode is the tree node that represents the tree of the compartments. More...

class	TQItem

class	TQueue

struct	TrajectoryRequests

struct	unified_allocator
	C++ allocator that uses [de]allocate_unified. More...

struct	UserParams
	This structure is data needed is several part of nrn_setup, phase1 and phase2. More...

class	VecPlayContinuous

struct	version
	Project version information. More...

struct	VoidFunc

Typedefs
using	DependencyTable = std::vector< std::vector< int > >

using	NRB_P = std::pair< int, int >

using	Core2NrnWatchInfoItem = std::vector< std::pair< int, bool > >
	All activated WATCH statements need activation on NEURON side. More...

using	Core2NrnWatchInfo = std::vector< Core2NrnWatchInfoItem >

using	SelfEventWeightMap = std::map< int, std::vector< TQItem * > >

using	PSD = std::pair< std::size_t, double * >

using	N2V = std::map< std::string, PSD >

using	segvec_type = std::vector< int >
	type to store every section and associated segments More...

using	secseg_map_type = std::map< int, segvec_type >

using	secseg_it_type = secseg_map_type::iterator

using	List = Item

using	Offset = size_t

using	MechId = int

using	VariableName = const char *

using	MechNamesMapping = std::map< MechId, std::map< VariableName, Offset, cmp_str > >

using	SerializedNames = const char **

using	Pfrpdat = Datum ()(void)

using	mod_alloc_t = void()(double , Datum *, int)

using	mod_f_t = void()(NrnThread , Memb_list *, int)

using	pnt_receive_t = void()(Point_process , int, double)

using	thread_table_check_t = void()(int, int, double , Datum , ThreadDatum , NrnThread , Memb_list , int)

using	NetBufReceive_t = void()(NrnThread )

using	nrn_watch_check_t = void()(NrnThread , Memb_list *)

using	bbcore_read_t = void()(double , int , int , int , int, int, double , Datum , ThreadDatum , NrnThread , Memb_list , double)

using	bbcore_write_t = void()(double , int , int , int , int, int, double , Datum , ThreadDatum , NrnThread , Memb_list , double)

using	Pfrv = void(*)()

using	VecTNode = std::vector< TNode * >

using	TQPair = std::pair< double, TQItem * >

using	Datum = int

using	Pfri = int(*)()

using	Symbol = char

using	TNI = std::pair< TNode *, int >

using	HashCnt = std::map< size_t, std::pair< TNode *, int > >

using	TNIVec = std::vector< TNI >

using	VTN = VecTNode

using	VVTN = std::vector< VTN >

using	VVVTN = std::vector< VVTN >

using	IvocVect = fixed_vector< double >

Enumerations
enum	LFPCalculatorType { LineSource, PointSource }

enum	TargetType { TargetType::Compartment = 0, Cell = 1, TargetType::SectionSoma = 2, TargetType::SectionAxon = 3, TargetType::SectionDendrite = 4, TargetType::SectionApical = 5, TargetType::SectionSomaAll = 6, TargetType::SectionAxonAll = 7, TargetType::SectionDendriteAll = 8, TargetType::SectionApicalAll = 9 }

enum	ReportType { SomaReport, CompartmentReport, SynapseReport, IMembraneReport, SectionReport, SummationReport }

enum	SectionType { Cell, Cell = 1, Soma, Axon, Dendrite, Apical, All }

enum	container { spltree, pq_que }

enum	mech_type { voltage = -1, i_membrane_ = -2 }
	Mechanism type to be used from stdindex2ptr and nrn_dblpntr2nrncore (in Neuron) Values of the mechanism types should be negative numbers to avoid any conflict with mechanism types of Memb_list(>0) or time(0) passed from Neuron. More...

enum	Layout { SoA = 0, AoS = 1 }

Functions
std::string	cnrn_version ()

std::ostream &	operator<< (std::ostream &os, const corenrn_parameters &corenrn_param)

void	call_prcellstate_for_prcellgid (int prcellgid, int compute_gpu, int is_init)

void	get_nrn_trajectory_requests (int bsize)

void	nrn_init_and_load_data (int argc, char *argv[], CheckPoints &checkPoints, bool is_mapping_needed, bool run_setup_cleanup)

void	handle_forward_skip (double forwardskip, int prcellgid)

static void	trajectory_return ()

std::unique_ptr< ReportHandler >	create_report_handler (const ReportConfiguration &config, const SpikesInfo &spikes_info)

void	nrn_init_capacitance (NrnThread , Memb_list , int)

void	nrn_cur_capacitance (NrnThread _nt, Memb_list ml, int type)

void	nrn_alloc_capacitance (double data, Datum pdata, int type)

void	nrn_init_ion (NrnThread , Memb_list , int)

void	nrn_cur_ion (NrnThread _nt, Memb_list ml, int type)

void	nrn_alloc_ion (double data, Datum pdata, int type)

void	second_order_cur (NrnThread *_nt, int secondorder)

void	nrn_ion_global_map_copyto_device ()

void	nrn_ion_global_map_delete_from_device ()

void	nrn_VecPlay_copyto_device (NrnThread nt, void *d_vecplay)

void	nrn_VecPlay_delete_from_device (NrnThread *nt)

void	cnrn_target_copyin_debug (std::string_view file, int line, std::size_t sizeof_T, std::type_info const &typeid_T, void const h_ptr, std::size_t len, void d_ptr)

void	cnrn_target_delete_debug (std::string_view file, int line, std::size_t sizeof_T, std::type_info const &typeid_T, void const *h_ptr, std::size_t len)

void	cnrn_target_deviceptr_debug (std::string_view file, int line, std::type_info const &typeid_T, void const h_ptr, void d_ptr)

void	cnrn_target_is_present_debug (std::string_view file, int line, std::type_info const &typeid_T, void const h_ptr, void d_ptr)

void	cnrn_target_memcpy_to_device_debug (std::string_view file, int line, std::size_t sizeof_T, std::type_info const &typeid_T, void const h_ptr, std::size_t len, void d_ptr)

int	cnrn_target_get_num_devices ()

void	cnrn_target_set_default_device (int device_num)

void	setup_nrnthreads_on_device (NrnThread *threads, int nthreads)

void	copy_ivoc_vect_to_device (const IvocVect &from, IvocVect &to)

void	delete_ivoc_vect_from_device (IvocVect &vec)

void	realloc_net_receive_buffer (NrnThread nt, Memb_list ml)

static void	net_receive_buffer_order (NetReceiveBuffer_t *nrb)

void	update_net_receive_buffer (NrnThread *nt)

void	update_net_send_buffer_on_host (NrnThread nt, NetSendBuffer_t nsb)

void	update_nrnthreads_on_host (NrnThread *threads, int nthreads)

void	update_weights_from_gpu (NrnThread *threads, int nthreads)
	Copy weights from GPU to CPU. More...

void	delete_nrnthreads_on_device (NrnThread *threads, int nthreads)
	Cleanup device memory that is being tracked by the OpenACC runtime. More...

void	nrn_newtonspace_copyto_device (NewtonSpace *ns)

void	nrn_newtonspace_delete_from_device (NewtonSpace *ns)

void	nrn_sparseobj_copyto_device (SparseObj *so)

void	nrn_sparseobj_delete_from_device (SparseObj *so)

void	init_gpu ()

static void	inverse_permute_copy (size_t n, double permuted_src, double dest, int *permute)
	permuted array copied to unpermuted array If permute is NULL then just a copy More...

static void	soa2aos_inverse_permute_copy (size_t n, int sz, int stride, double src, double dest, int permute)
	SoA permuted mechanism data copied to unpermuted AoS data. More...

static void	soa2aos_unpermuted_copy (size_t n, int sz, int stride, double src, double *dest)
	SoA unpermuted mechanism data copied to unpermuted AoS data. More...

static void	aos2aos_copy (size_t n, int sz, double src, double *dest)
	AoS mechanism data copied to AoS data. More...

static void	core2nrn_corepointer (int tid, NrnThreadMembList *tml)
	Copy back COREPOINTER info to NEURON. More...

static void	core2nrn_tqueue (NrnThread &)
	Copy event queue and related state back to NEURON. More...

static void	core2nrn_watch ()
	Callbacks into NEURON for WatchCondition. More...

static void	core2nrn_vecplay ()
	Transfer VecPlay indices to NEURON. More...

void	core2nrn_data_return ()
	copy data back to NEURON. More...

static void	core2nrn_PreSyn_flag (NrnThread &nt)

void	nrn2core_PreSyn_flag_receive (int tid)

static void	clear_inv_perm_for_selfevent_targets ()

static bool	core2nrn_tqueue_item (TQItem *q, SelfEventWeightMap &sewm, NrnThread &nt)

void	watch_datum_indices (int type, int &first, int &last)
	return first and last datum indices of WATCH statements More...

void	hoc_register_var (DoubScal ds, DoubVec dv, VoidFunc *)

void	set_globals (const char *path, bool cli_global_seed, int cli_global_seed_value)

void	write_mech_report ()
	display global mechanism count More...

int	nrn_soa_padded_size (int cnt, int layout)
	calculate size after padding for specific memory layout More...

size_t	nrn_soa_byte_align (size_t size)
	return the new offset considering the byte aligment settings More...

int	nrn_i_layout (int icnt, int cnt, int isz, int sz, int layout)
	This function return the index in a flat array of a matrix coordinate (icnt, isz). More...

int	nrn_param_layout (int i, int mtype, Memb_list *ml)

static void	mk_mech ()

static void	mk_mech (std::istream &)

void	mk_mech (const char *datpath)
	Read meta data about the mechanisms and allocate corresponding mechanism management data structures. More...

int	nrn_get_mechtype (const char *name)
	Get mechanism type by the mechanism name. More...

const char *	nrn_get_mechname (int type)

static void	nrn2core_tqueue ()
	Copy each thread's queue from NEURON. More...

static void	watch_activate_clear ()

static void	nrn2core_transfer_watch_condition (int, int, int, int, int)

static void	vec_play_activate ()

static void	nrn2core_patstim_share_info ()

void	direct_mode_initialize ()
	All state from NEURON necessary to continue a run. More...

static void	setup_type2semantics ()

void **	pattern_stim_info_ref (int icnt, int cnt, double _p, Datum _ppvar, ThreadDatum _thread, NrnThread _nt, Memb_list *ml, double v)

int	checkpoint_save_patternstim (_threadargsproto_)

void	checkpoint_restore_patternstim (int, double, _threadargsproto_)

static int	nrn_original_aos_index (int etype, int ix, NrnThread &nt, int **ml_pinv)

int *	inverse_permute (int *p, int n)

void	nrn_inverse_i_layout (int i, int &icnt, int cnt, int &isz, int sz, int layout)

void	nrn_read_filesdat (int &ngrp, int &grp, const char filesdat)

void	netpar_tid_gid2ps (int tid, int gid, PreSyn ps, InputPreSyn psi)

void	determine_inputpresyn ()

void	nrn_setup_cleanup ()
	Clean up. More...

void	nrn_setup (const char filesdat, bool is_mapping_needed, CheckPoints &checkPoints, bool run_setup_cleanup, const char datpath, const char restore_path, double mindelay)

void	setup_ThreadData (NrnThread &nt)

void	read_phasegap (NrnThread &nt, UserParams &userParams)

double *	stdindex2ptr (int mtype, int index, NrnThread &nt)

void	nrn_cleanup_ion_map ()
	Cleanup global ion map created during mechanism registration. More...

void	delete_fornetcon_info (NrnThread &nt)

void	nrn_cleanup ()

void	delete_trajectory_requests (NrnThread &nt)

void	read_phase1 (NrnThread &nt, UserParams &userParams)

void	read_phase2 (NrnThread &nt, UserParams &userParams)

void	read_phase3 (NrnThread &nt, UserParams &userParams)
	read mapping information for neurons More...

size_t	memb_list_size (NrnThreadMembList *tml, bool include_data)

size_t	output_presyn_size (void)
	Approximate count of number of bytes for the gid2out map. More...

size_t	input_presyn_size (void)

size_t	model_size (bool detailed_report)

void	nrn_init_and_load_data (int argc, char **argv, CheckPoints &checkPoints, bool is_mapping_needed=false, bool run_setup_cleanup=true)

void	allocate_data_in_mechanism_nrn_init ()

void	mk_spikevec_buffer (int sz)

void	spikevec_lock ()

void	spikevec_unlock ()

static void	local_spikevec_sort (std::vector< double > &isvect, std::vector< int > &isvecg, std::vector< double > &osvect, std::vector< int > &osvecg)

static void	output_spikes_serial (const char *outpath)

void	output_spikes (const char *outpath, const SpikesInfo &spikes_info)

void	clear_spike_vectors ()

void	validation (std::vector< std::pair< double, int >> &res)

template<typename T >
void	mech_data_layout_transform (T *data, int cnt, int sz, int layout)

static int	permute (int i, NrnThread &nt)

static int	inv_permute (int i, NrnThread &nt)

static int	ml_permute (int i, Memb_list *ml)

static void	pr_memb (int type, Memb_list ml, int cellnodes, NrnThread &nt, FILE *f)

static void	pr_netcon (NrnThread &nt, FILE *f)

static void	pr_realcell (PreSyn &ps, NrnThread &nt, FILE *f)

int	prcellstate (int gid, const char *suffix)

void	nrn_flush_reports (double t)

void	setup_report_engine (double dt_report, double mindelay)
	in the current implementation, we call flush during every spike exchange interval. More...

void	set_report_buffer_size (int n)

void	finalize_report ()

std::vector< ReportConfiguration >	create_report_configurations (const std::string &filename, const std::string &output_dir, SpikesInfo &spikes_info)

void	parse_filter_string (const std::string &filter, ReportConfiguration &config)

void	register_target_type (ReportConfiguration &report, ReportType report_type)

template<typename T >
std::vector< T >	intersection_gids (const NrnThread &nt, std::vector< T > &target_gids)

static int *	fornetcon_slot (const int mtype, const int instance, const int fnslot, const NrnThread &nt)
	If FOR_NETCON in use, setup NrnThread fornetcon related info. More...

void	setup_fornetcon_info (NrnThread &nt)
	If FOR_NETCON in use, setup NrnThread fornetcon related info. More...

void	nrn_jacob_capacitance (NrnThread , Memb_list , int)

void	nrn_div_capacity (NrnThread , Memb_list , int)

void	nrn_mul_capacity (NrnThread , Memb_list , int)

void	capacitance_reg (void)

int	nrn_is_ion (int)

void	ion_reg (const char *, double)

void	_passive_reg (void)

void	_stim_reg (void)

void	_hh_reg (void)

void	_netstim_reg (void)

void	_expsyn_reg (void)

void	_exp2syn_reg (void)

void	_svclmp_reg (void)

void	modl_reg ()
	Mechanism registration function. More...

void	_nrn_destroy_sparseobj_thread (SparseObj *so)

template<typename F >
int	euler_thread (int neqn, int var, int der, F fun, _threadargsproto_)

template<typename F >
int	derivimplicit_thread (int n, int slist, int dlist, F fun, _threadargsproto_)

static void	set_an_offset (int mech_id, const char *variable_name, int offset)

double *	get_var_location_from_var_name (int mech_id, const char variable_name, Memb_list ml, int node_index)

void	register_all_variables_offsets (int mech_id, SerializedNames variable_names)

int	register_mech (const char **m, mod_alloc_t alloc, mod_f_t cur, mod_f_t jacob, mod_f_t stat, mod_f_t initialize, mod_f_t private_constructor, mod_f_t private_destructor, int nrnpointerindex, int vectorized)

int	point_register_mech (const char **, mod_alloc_t alloc, mod_f_t cur, mod_f_t jacob, mod_f_t stat, mod_f_t initialize, mod_f_t private_constructor, mod_f_t private_destructor, int nrnpointerindex, mod_f_t constructor, mod_f_t destructor, int vectorized)

void	register_constructor (mod_f_t constructor)

void	hoc_register_net_receive_buffering (NetBufReceive_t, int)

void	hoc_register_net_send_buffering (int)

void	hoc_register_watch_check (nrn_watch_check_t, int)

void	nrn_writes_conc (int, int)

constexpr double	ktf (double celsius)

constexpr double	nrn_nernst (double ci, double co, double z, double celsius)

constexpr void	nrn_wrote_conc (int type, double p1, int p2, int it, double *gimap, double celsius, int _cntml_padded)

double	nrn_ghk (double v, double ci, double co, double z, double celsius)

void	hoc_register_prop_size (int, int, int)

void	hoc_register_dparam_semantics (int type, int, const char *name)

void	hoc_reg_ba (int, mod_f_t, int)

void	_nrn_layout_reg (int, int)

void	_nrn_thread_reg0 (int i, void(f)(ThreadDatum ))

void	_nrn_thread_reg1 (int i, void(f)(ThreadDatum ))

int	nrn_mech_depend (int type, int *dependencies)

void	add_nrn_fornetcons (int, int)

void	add_nrn_has_net_event (int)

void	net_event (Point_process *, double)

void	net_send (void *, int, Point_process , double, double)

void	net_move (void *, Point_process , double)

void	artcell_net_send (void *, int, Point_process , double, double)

void	artcell_net_move (void *, Point_process , double)

void	nrn2ncs_outputevent (int netcon_output_index, double firetime)

void	net_sem_from_gpu (int sendtype, int i_vdata, int, int ith, int ipnt, double, double)

void	hoc_malchk (void)

void *	hoc_Emalloc (size_t)

void	_pattern_reg (void)

void	pattern_stim_setup_helper (int size, double tvec, int gidvec, int icnt, int cnt, double _p, Datum _ppvar, ThreadDatum _thread, NrnThread _nt, Memb_list *ml, double v)

static size_t	read_raster_file (const char fname, double tvec, int *gidvec, double tstop)

void	nrn_set_extra_thread0_vdata ()

void	nrn_mkPatternStim (const char *fname, double tstop)

static NrnThreadMembList *	alloc_nrn_thread_memb (NrnThread *nt, int type)

Point_process *	nrn_artcell_instantiate (const char *mechname)

static void	ion_write_depend (int type, int etype)

void	hoc_reg_bbcore_read (int type, bbcore_read_t f)

void	hoc_reg_bbcore_write (int type, bbcore_write_t f)

void	add_nrn_artcell (int type, int qi)

void	set_pnt_receive (int type, pnt_receive_t pnt_receive, pnt_receive_t pnt_receive_init, short size)

void	alloc_mech (int memb_func_size_)

void	initnrn ()

static int	depend_append (int idep, int *dependencies, int deptype, int type)

void	register_destructor (mod_f_t d)

int	point_reg_helper (const Symbol *s2)

void	_modl_cleanup ()

void	state_discontinuity (int, double *pd, double d)

void	_nrn_thread_table_reg (int i, thread_table_check_t f)

void	_nrn_setdata_reg (int i, void(call)(double , Datum *))

void	_nrn_thread_table_reg (int i, void(f)(int, int, double , Datum , ThreadDatum , NrnThread , Memb_list , int))

static void *	emalloc (size_t size)

void	nrnmpi_spike_initialize ()

void	wait_before_spike_exchange ()

int	nrnmpi_spike_exchange_impl (int nin, NRNMPI_Spike spikeout, int icapacity, NRNMPI_Spike *spikein, int &ovfl, int nout, NRNMPI_Spikebuf spbufout, NRNMPI_Spikebuf *spbufin)

int	nrnmpi_spike_exchange_compressed_impl (int localgid_size, unsigned char &spfixin_ovfl, int send_nspike, int nin, int ovfl_capacity, unsigned char spikeout_fixed, int ag_send_size, unsigned char spikein_fixed, int &ovfl)

int	nrnmpi_int_allmax_impl (int x)

void	nrnmpi_int_alltoall_impl (int s, int r, int n)

void	nrnmpi_int_alltoallv_impl (const int s, const int scnt, const int sdispl, int r, int rcnt, int rdispl)

void	nrnmpi_dbl_alltoallv_impl (double s, int scnt, int sdispl, double r, int rcnt, int rdispl)

void	nrnmpi_int_allgather_impl (int s, int r, int n)

double	nrnmpi_dbl_allmin_impl (double x)

double	nrnmpi_dbl_allmax_impl (double x)

void	nrnmpi_barrier_impl ()

double	nrnmpi_dbl_allreduce_impl (double x, int type)

void	nrnmpi_dbl_allreduce_vec_impl (double src, double dest, int cnt, int type)

void	nrnmpi_long_allreduce_vec_impl (long src, long dest, int cnt, int type)

static void	nrn_fatal_error (const char *msg)

nrnmpi_init_ret_t	nrnmpi_init_impl (int pargc, char **pargv, bool is_quiet)

void	nrnmpi_finalize_impl (void)

void	nrnmpi_check_threading_support_impl ()

bool	nrnmpi_initialized_impl ()

void	nrnmpi_abort_impl (int errcode)

double	nrnmpi_wtime_impl ()

int	nrnmpi_local_rank_impl ()
	Return local mpi rank within a shared memory node. More...

int	nrnmpi_local_size_impl ()
	Return number of ranks running on single shared memory node. More...

void	nrnmpi_write_file_impl (const std::string &filename, const char *buffer, size_t length)
	Write given buffer to a new file using MPI collective I/O. More...

mpi_manager_t &	mpi_manager ()

void	deliver_net_events (NrnThread *nt)

void	nrn_deliver_events (NrnThread *nt)

void	clear_event_queue ()

void	init_net_events ()

void	nrn_play_init ()

void	fixed_play_continuous (NrnThread *nt)

static int	default_rendezvous (HAVEWANT_t key)

static int *	cnt2displ (int *cnt)

static int *	srccnt2destcnt (int *srccnt)

static void	rendezvous_rank_get (HAVEWANT_t data, int size, HAVEWANT_t &sdata, int &scnt, int &sdispl, HAVEWANT_t &rdata, int &rcnt, int &rdispl, int(rendezvous_rank)(HAVEWANT_t))

static void	have_to_want (HAVEWANT_t have, int have_size, HAVEWANT_t want, int want_size, HAVEWANT_t &send_to_want, int &send_to_want_cnt, int &send_to_want_displ, HAVEWANT_t &recv_from_have, int &recv_from_have_cnt, int &recv_from_have_displ, int(*rendezvous_rank)(HAVEWANT_t))

void	nrn_multisend_send (PreSyn , double t, NrnThread )

void	nrn_multisend_receive (NrnThread *)

void	nrn_multisend_advance ()

void	nrn_multisend_init ()

void	nrn_multisend_cleanup ()

void	nrn_multisend_setup ()

void	nrn_multisend_setup_targets (bool use_phase2, int &targets_phase1, int &targets_phase2)

void	mk_netcvode ()

void	interthread_enqueue (NrnThread *nt)

void	nrn_p_construct ()

void	ncs2nrn_integrate (double tstop)

static bool	pscheck (double var, double thresh, int *flag)

void	nrn_spike_exchange_init ()

static void	alloc_mpi_space ()
	Allocate space for spikes: 200 structs of {int gid; double time} coming from nrnmpi.h and array of int of the global domain size. More...

static bool	nrn_need_npe ()

void	nrn_fake_fire (int gid, double spiketime, int fake_out)

int	nrn_set_timeout (int timeout)

void	BBS_netpar_solve (double tstop)

double	set_mindelay (double maxdelay)

int	nrnmpi_spike_compress (int nspike, bool gid_compress, int xchng_meth)

void	nrn_spike_exchange (NrnThread *nt)

void	nrnmpi_v_transfer ()

void	nrnthread_v_transfer (NrnThread *_nt)

size_t	level_from_leaf (VecTNode &)

size_t	level_from_root (VecTNode &)

void	group_order2 (VecTNode &, size_t groupsize, size_t ncell)
	Implementation of the advanced interleaving strategy (interleave_permute_type == 2) More...

size_t	dist2child (TNode *nd)

size_t	warp_balance (size_t ncell, VecTNode &nodevec)
	Use of the LPT (Least Processing Time) algorithm to create balanced groups of cells. More...

void	spinit (SPTREE *q)

SPBLK *	spenq (SPBLK n, SPTREE q)

SPBLK *	spdeq (SPBLK **np)

void	splay (SPBLK n, SPTREE q)

SPBLK *	sphead (SPTREE *q)

void	spdelete (SPBLK n, SPTREE q)

void *	nrn_cacheline_alloc (void **memptr, size_t size)

void *	emalloc_align (size_t size, size_t alignment)

void *	ecalloc_align (size_t n, size_t size, size_t alignment)

void	check_bbcore_write_version (const char *)

void	nrn_outputevent (unsigned char, double)

bool	warpcmp (const TNode a, const TNode b)

void	create_interleave_info ()

void	destroy_interleave_info ()

static void	print_quality2 (int iwarp, InterleaveInfo &ii, int *p)

static void	print_quality1 (int iwarp, InterleaveInfo &ii, int ncell, int *p)

static void	warp_balance (int ith, InterleaveInfo &ii)

int *	interleave_order (int ith, int ncell, int nnode, int *parent)
	Function that performs the permutation of the cells such that the execution threads access coalesced memory. More...

static void	triang_interleaved (NrnThread nt, int icell, int icellsize, int nstride, int stride, int *lastnode)

static void	bksub_interleaved (NrnThread nt, int icell, int icellsize, int, int stride, int *firstnode)

	nrn_pragma_acc (routine vector) static void triang_interleaved2(NrnThread *nt

	nrn_pragma_acc (loop seq) for(

	if (ncell==0)

	nrn_pragma_acc (parallel loop present(nt[0:1], stride[0:nstride], firstnode[0:ncell], lastnode[0:ncell], cellsize[0:ncell]) if(nt->compute_gpu) async(nt->stream_id)) nrn_pragma_omp(target teams distribute parallel for simd if(nt -> compute_gpu)) for(int icell=0

void	solve_interleaved (int ith)
	Solve the Hines matrices based on the interleave_permute_type (1 or 2). More...

void	solve_interleaved2_launcher (NrnThread nt, InterleaveInfo info, int ncore, void *stream)
	CUDA branch of the solve_interleaved with interleave_permute_type == 2. More...

int *	node_order (int ncell, int nnode, int parents, int &nwarp, int &nstride, int &stride, int &firstnode, int &lastnode, int &cellsize, int &stridedispl)
	Function that returns a permutation of length nnode. More...

template<typename T >
void	copy_array (T &dest, T src, size_t n)

template<typename T >
void	copy_align_array (T &dest, T src, size_t n)

static bool	tnode_earlier (TNode a, TNode b)
	Function to order trees by size, hash and nodeindex. More...

static bool	ptr_tnode_earlier (TNode a, TNode b)

static void	tree_analysis (int *parent, int nnode, int ncell, VecTNode &nodevec)
	Perform tree preparation for interleaving strategies. More...

static void	node_interleave_order (int ncell, VecTNode &nodevec)
	Naive interleaving strategy (interleave_permute_type == 1) More...

static void	admin1 (int ncell, VecTNode &nodevec, int &nwarp, int &nstride, int &stride, int &firstnode, int &lastnode, int &cellsize)

static void	admin2 (int ncell, VecTNode &nodevec, int &nwarp, int &nstride, int &stridedispl, int &strides, int &rootbegin, int &nodebegin, int *&ncycles)
	Prepare for solve_interleaved2. More...

static void	check (VecTNode &)

static void	quality (VecTNode &nodevec, size_t max=32)

static void	set_cellindex (int ncell, VecTNode &nodevec)
	Set the cellindex to distinguish the different cells. More...

static void	set_groupindex (VecTNode &nodevec)
	Initialization of the groupindex (groups) More...

static void	ident_statistic (VecTNode &nodevec, size_t ncell)

static bool	interleave_comp (TNode a, TNode b)

static size_t	stride_length (size_t begin, size_t end, VecTNode &nodevec)

void	chklevel (VTN &level, size_t nident=8)

static bool	sortlevel_cmp (TNode a, TNode b)

static void	sortlevel (VTN &level)

static void	set_treenode_order (VVTN &levels)

static bool	is_parent_race2 (TNode *nd)

static bool	is_child_race2 (TNode *nd)

template<typename T >
static void	move_range (size_t start, size_t length, size_t dst, std::vector< T > &v)

static void	move_nodes (size_t start, size_t length, size_t dst, VTN &nodes)

static size_t	next_leaf (TNode *nd, VTN &nodes)

static void	checkrace (TNode *nd, VTN &nodes)

static bool	eliminate_race (TNode nd, size_t d, VTN &nodes, TNode look)

static void	eliminate_prace (TNode *nd, VTN &nodes)

static void	eliminate_crace (TNode *nd, VTN &nodes)

static void	question2 (VVTN &levels)

static void	analyze (VVTN &levels)

void	prgroupsize (VVVTN &groups)

static bool	final_nodevec_cmp (TNode a, TNode b)

static void	set_nodeindex (VecTNode &nodevec)

int	get_data_index (int node_index, int variable_index, int mtype, Memb_list *ml)

template<typename T >
void	permute (T data, int cnt, int sz, int layout, int p)

static void	invert_permute (int *p, int n)

static int	full_search (NrnThread &nt, double *pd)

int	type_of_ntdata (NrnThread &nt, int i, bool reset)

static void	update_pdata_values (Memb_list *ml, int type, NrnThread &nt)

void	node_permute (int vec, int n, int permute)

void	permute_ptr (int vec, int n, int p)

void	permute_data (double vec, int n, int p)

void	permute_ml (Memb_list *ml, int type, NrnThread &nt)

int	nrn_index_permute (int ix, int type, Memb_list *ml)

static bool	nrn_index_sort_cmp (const std::pair< int, int > &a, const std::pair< int, int > &b)

static int *	nrn_index_sort (int *values, int n)

void	permute_nodeindices (Memb_list ml, int p)

static void *	nrn_fixed_step_thread (NrnThread *)

static void	nrn_fixed_step_group_thread (NrnThread *, int, int, int &)

void	dt2thread (double adt)

void	nrn_fixed_step_minimal ()

void	nrn_fixed_single_steps_minimal (int total_sim_steps, double tstop)
	--> Coreneuron More...

void	nrn_fixed_step_group_minimal (int total_sim_steps)

void	update (NrnThread *_nt)

void	nonvint (NrnThread *_nt)

void	nrn_ba (NrnThread *nt, int bat)

void	nrncore2nrn_send_init ()

void	nrncore2nrn_send_values (NrnThread *nth)

void *	nrn_fixed_step_lastpart (NrnThread *nth)

void	fast_imem_free ()

void	nrn_fast_imem_alloc ()

void	nrn_calc_fast_imem (NrnThread *nt)

void	nrn_calc_fast_imem_init (NrnThread *nt)

void	nrn_finitialize (int setv, double v)

NrnThreadMembList *	create_tml (NrnThread &nt, int mech_id, Memb_func &memb_func, int &shadow_rhs_cnt, const std::vector< int > &mech_types, const std::vector< int > &nodecounts)

void	nrn_threads_create (int n)

void	nrn_threads_free ()

void	nrn_mk_table_check ()

void	nrn_thread_table_check ()

template<typename F , typename... Args>
void	nrn_multithread_job (F &&job, Args &&... args)

void	nrn_solve_minimal (NrnThread *)

void *	setup_tree_matrix_minimal (NrnThread *)

constexpr int	at_time (NrnThread *nt, double te)

int	abort_run (int code)

int	nrn_crout_thread (NewtonSpace ns, int n, double a, int perm, _threadargsproto_)
	Performs an LU triangular factorization of a real matrix by the Crout algorithm using partial pivoting. More...

void	nrn_scopmath_solve_thread (int n, double *a, double b, int perm, double p, int *y, _threadargsproto_)
	Performs forward substitution algorithm to transform the constant vector in the linear simultaneous equations to be consistent with the factored matrix. More...

NewtonSpace *	nrn_cons_newtonspace (int n, int n_instance)

void	nrn_destroy_newtonspace (NewtonSpace *ns)

template<typename F >
int	nrn_newton_thread (NewtonSpace ns, int n, int s, F func, double *value, _threadargsproto_)
	Iteratively solves simultaneous nonlinear equations by Newton's method, using a Jacobian matrix computed by finite differences. More...

template<typename SPFUN >
void *	nrn_cons_sparseobj (SPFUN fun, int n, Memb_list *ml, _threadargsproto_)
	sparse matrix dynamic allocation: create_coef_list makes a list for fast setup, does minimum ordering and ensures all elements needed are present. More...

template<typename F >
int	sparse_thread (SparseObj so, int n, int s, int d, double t, double dt, F fun, int linflag, _threadargsproto_)
	This is an experimental numerical method for SCoP-3 which integrates kinetic rate equations. More...

template<typename SPFUN >
int	_cvode_sparse_thread (void *vpr, int n, int x, SPFUN fun, _threadargsproto_)

static int	check_state (int n, int *s, _threadargsproto_)

template<typename SPFUN >
int	_ss_sparse_thread (SparseObj so, int n, int s, int d, double t, double dt, SPFUN fun, int linflag, _threadargsproto_)

template<typename DIFUN >
int	_ss_derivimplicit_thread (int n, int slist, int dlist, DIFUN fun, _threadargsproto_)

static void	triang (NrnThread *)

static void	bksub (NrnThread *)

static void	nrn_rhs (NrnThread *_nt)

static void	nrn_lhs (NrnThread *_nt)

IvocVect *	vector_new (int n)

int	vector_capacity (IvocVect *v)

double *	vector_vec (IvocVect *v)

IvocVect *	vector_new1 (int n)

bool	gpu_enabled ()
	Check if GPU support is enabled. More...

void *	allocate_unified (std::size_t num_bytes)
	Allocate unified memory in GPU builds iff GPU enabled, otherwise new. More...

void	deallocate_unified (void *ptr, std::size_t num_bytes)
	Deallocate memory allocated by `allocate_unified`. More...

template<typename T , typename U >
bool	operator== (unified_allocator< T > const &, unified_allocator< U > const &) noexcept

template<typename T , typename U >
bool	operator!= (unified_allocator< T > const &x, unified_allocator< U > const &y) noexcept

template<typename T , typename Alloc , typename... Args>
auto	allocate_unique (const Alloc &alloc, Args &&... args)

double	nrn_mallinfo (void)
	Returns current memory usage in KBs. More...

void	report_mem_usage (const char *message, bool all_ranks=false)
	Reports current memory usage of the simulator to stdout. More...

void	report_cell_stats ()
	Reports global cell statistics of the simulation. More...

char *	pnt_name (Point_process *pnt)

void	nrn_exit (int err)

void	hoc_execerror (const char s1, const char s2)

void	hoc_warning (const char s1, const char s2)

double *	makevector (size_t size)

void	freevector (double *p)

double **	makematrix (size_t nrows, size_t ncols)

void	freematrix (double **matrix)

void *	ecalloc (size_t n, size_t size)

void *	erealloc (void *ptr, size_t size)

double	hoc_Exp (double x)

template<typename T >
T *	cnrn_target_deviceptr_or_present (std::string_view file, int line, bool must_be_present_or_null, const T *h_ptr)

template<typename T >
T *	cnrn_target_copyin (std::string_view file, int line, const T *h_ptr, std::size_t len=1)

template<typename T >
void	cnrn_target_delete (std::string_view file, int line, T *h_ptr, std::size_t len=1)

template<typename T >
void	cnrn_target_memcpy_to_device (std::string_view file, int line, T d_ptr, const T h_ptr, std::size_t len=1)

template<typename T >
void	cnrn_target_update_on_device (std::string_view file, int line, const T *h_ptr, std::size_t len=1)

std::size_t	nrnran123_instance_count ()

uint32_t	nrnran123_get_globalindex ()

void	nrnran123_set_globalindex (uint32_t gix)

void	nrnran123_initialise_global_state_on_device ()

void	nrnran123_destroy_global_state_on_device ()

nrnran123_State *	nrnran123_newstream3 (uint32_t id1, uint32_t id2, uint32_t id3, bool use_unified_memory)
	Allocate a new Random123 stream. More...

void	nrnran123_deletestream (nrnran123_State *s, bool use_unified_memory)

void	nrn_abort (int errcode)

double	nrn_wtime ()

template<typename... Args>
void	nrn_fatal_error (const char *msg, Args &&... args)

Variables
corenrn_parameters	corenrn_param
	Printing method. More...

int	nrn_nobanner_ {0}
	Declaring global corenrn_parameters object for this instance of CoreNeuron. More...

CoreNeuron	corenrn

InterleaveInfo *	interleave_info

void(*	core2nrn_clear_queues_ )(double t)
	Callback to clear NEURON thread queues. More...

void(*	core2nrn_watch_clear_ )()

void(*	core2nrn_watch_activate_ )(int tid, int type, int watch_begin, Core2NrnWatchInfo &)

void(*	core2nrn_vecplay_ )(int tid, int i_nrn, int last, int discon, int ubound)
	VecPlay indices back to NEURON. More...

void(*	core2nrn_vecplay_events_ )()

void(*	core2nrn_NetCon_event_ )(int tid, double td, size_t nc_index)
	Callbacks into NEURON for queue event types. More...

void(*	core2nrn_SelfEvent_event_ )(int tid, double td, int tar_type, int tar_index, double flag, size_t nc_index, int is_movable)

void(*	core2nrn_SelfEvent_event_noweight_ )(int tid, double td, int tar_type, int tar_index, double flag, int is_movable)

void(*	core2nrn_PreSyn_flag_ )(int tid, std::set< int > presyns_flag_true)

void(*	nrn2core_transfer_PreSyn_flag_ )(int tid, std::set< int > &presyns_flag_true)

std::map< int, int * >	type2invperm

static N2V *	n2v

std::map< std::string, int >	mech2type

void(*	nrn2core_mkmech_info_ )(std::ostream &)

void(*	nrn2core_transfer_watch_ )(void(*cb)(int, int, int, int, int))
	Pointer to function in NEURON that iterates over activated WATCH statements, sending each item to ... More...

NrnCoreTransferEvents (	nrn2core_transfer_tqueue_ )(int tid)
	Pointer to function in NEURON that iterates over its tqeueue. More...

static std::unordered_map< int, int >	type2movable

void(*	nrn2core_patternstim_ )(void **info)

NrnThreadChkpnt *	nrnthread_chkpnt

int	patstimtype

const int	max_line_length = 1024
	Encapsulate low-level reading of coreneuron input data files. More...

static OMP_Mutex	mut

std::vector< std::map< int, PreSyn * > >	neg_gid2out
	Vector of maps for negative presyns. More...

std::map< int, PreSyn * >	gid2out
	Maps for ouput and input presyns. More...

std::map< int, InputPreSyn * >	gid2in

std::vector< NetCon * >	netcon_in_presyn_order_
	InputPreSyn.nc_index_ to + InputPreSyn.nc_cnt_ give the NetCon*. More...

std::vector< int * >	nrnthreads_netcon_srcgid
	Only for setup vector of netcon source gids. More...

std::vector< std::vector< int > >	nrnthreads_netcon_negsrcgid_tid
	If a nrnthreads_netcon_srcgid is negative, need to determine the thread when in order to use the correct neg_gid2out[tid] map. More...

std::vector< double >	spikevec_time
	--> Coreneuron as SpikeBuffer class More...

std::vector< int >	spikevec_gid

static OMP_Mutex	mut

static std::map< Point_process *, int >	pnt2index

static int	pntindex

static std::map< NetCon , DiscreteEvent >	map_nc2src

static std::vector< int > *	inv_permute_

static int	size_report_buffer = 4

static const char *	mechanism [] = {"0", "capacitance", "cm", 0, "i_cap", 0, 0}

static void(*	mechanism [])(void)

bool	nrn_have_gaps
	variables defined in coreneuron library More...

bool	nrn_use_fast_imem

static MechNamesMapping	mechNamesMapping

int	nrn_ion_global_map_size

double **	nrn_ion_global_map

const int	ion_global_map_member_size

int	nrn_fornetcon_cnt_

int *	nrn_fornetcon_type_

int *	nrn_fornetcon_index_

bool	nrn_use_localgid_

int	nrn_extra_thread0_vdata

int	secondorder = 0

double	t

double	dt

double	celsius

double	pi

int	rev_dt

int	state_discon_allowed_

int	state_discon_flag_ = 0

int	nrnmpi_numprocs = 1

int	nrnmpi_myid = 0

mpi_function< cnrn_make_integral_constant_t(nrnmpi_init_impl)>	nrnmpi_init {"nrnmpi_init_impl"}

mpi_function< cnrn_make_integral_constant_t(nrnmpi_finalize_impl)>	nrnmpi_finalize

mpi_function< cnrn_make_integral_constant_t(nrnmpi_check_threading_support_impl)>	nrnmpi_check_threading_support {"nrnmpi_check_threading_support_impl"}

mpi_function< cnrn_make_integral_constant_t(nrnmpi_write_file_impl)>	nrnmpi_write_file

mpi_function< cnrn_make_integral_constant_t(nrnmpi_spike_exchange_impl)>	nrnmpi_spike_exchange

mpi_function< cnrn_make_integral_constant_t(nrnmpi_spike_exchange_compressed_impl)>	nrnmpi_spike_exchange_compressed {"nrnmpi_spike_exchange_compressed_impl"}

mpi_function< cnrn_make_integral_constant_t(nrnmpi_int_allmax_impl)>	nrnmpi_int_allmax

mpi_function< cnrn_make_integral_constant_t(nrnmpi_int_allgather_impl)>	nrnmpi_int_allgather

mpi_function< cnrn_make_integral_constant_t(nrnmpi_int_alltoall_impl)>	nrnmpi_int_alltoall

mpi_function< cnrn_make_integral_constant_t(nrnmpi_int_alltoallv_impl)>	nrnmpi_int_alltoallv

mpi_function< cnrn_make_integral_constant_t(nrnmpi_dbl_alltoallv_impl)>	nrnmpi_dbl_alltoallv

mpi_function< cnrn_make_integral_constant_t(nrnmpi_dbl_allmin_impl)>	nrnmpi_dbl_allmin

mpi_function< cnrn_make_integral_constant_t(nrnmpi_dbl_allmax_impl)>	nrnmpi_dbl_allmax

mpi_function< cnrn_make_integral_constant_t(nrnmpi_barrier_impl)>	nrnmpi_barrier

mpi_function< cnrn_make_integral_constant_t(nrnmpi_dbl_allreduce_impl)>	nrnmpi_dbl_allreduce

mpi_function< cnrn_make_integral_constant_t(nrnmpi_dbl_allreduce_vec_impl)>	nrnmpi_dbl_allreduce_vec

mpi_function< cnrn_make_integral_constant_t(nrnmpi_long_allreduce_vec_impl)>	nrnmpi_long_allreduce_vec {"nrnmpi_long_allreduce_vec_impl"}

mpi_function< cnrn_make_integral_constant_t(nrnmpi_initialized_impl)>	nrnmpi_initialized

mpi_function< cnrn_make_integral_constant_t(nrnmpi_abort_impl)>	nrnmpi_abort {"nrnmpi_abort_impl"}

mpi_function< cnrn_make_integral_constant_t(nrnmpi_wtime_impl)>	nrnmpi_wtime {"nrnmpi_wtime_impl"}

mpi_function< cnrn_make_integral_constant_t(nrnmpi_local_rank_impl)>	nrnmpi_local_rank

mpi_function< cnrn_make_integral_constant_t(nrnmpi_local_size_impl)>	nrnmpi_local_size

MPI_Comm	nrnmpi_comm

static int	np

static int *	displs {nullptr}

static int *	byteovfl {nullptr}

static MPI_Datatype	spike_type

MPI_Comm	nrnmpi_world_comm

int	nrnmpi_numprocs_

int	nrnmpi_myid_

static bool	nrnmpi_under_nrncontrol_ {false}

bool	use_multisend_

bool	use_phase2_

int	n_multisend_interval = 2

NetCvode *	net_cvode_instance

bool	cvode_active_

bool	nrn_use_bin_queue_ = 0
	Flag to use the bin queue. More...

static bool	active_ = false

static double	usable_mindelay_

static double	mindelay_

static double	last_maxstep_arg_

static std::vector< NetParEvent >	npe_

static int	timeout_ = 0

bool	stoprun

const char *	bbcore_write_version = "1.6"

int	interleave_permute_type

int	cellorder_nwarp = 0

int	icore

int int	ncycle

int int int *	stride = ii.stride

int int int int	lastnode

int	istride = stride[icycle]

int	i = lastnode - istride + icore

int	ii = i

bool	has_subtrees_to_compute = true

int	root

int int	lastroot

int int int int int int	firstnode

icycle< ncycle;++icycle) { int istride=stride[icycle];nrn_pragma_acc(loop vector) nrn_pragma_omp(loop bind(parallel)) for(int icore=0;icore< warpsize;++icore) { int i=ii+icore;if(icore< istride) { int ip=GPU_PARENT(i);GPU_RHS(i) -=GPU_B(i) GPU_RHS(ip);GPU_RHS(i)/=GPU_D(i);} i+=istride;} ii+=istride;}}void solve_interleaved2(int ith) { NrnThread nt=nrn_threads+ith;InterleaveInfo &ii=interleave_info[ith];int nwarp=ii.nwarp;if(nwarp==0) return;int ncore=nwarp warpsize;int ncycles=ii.cellsize;int stridedispl=ii.stridedispl;int strides=ii.stride;int rootbegin=ii.firstnode;int nodebegin=ii.lastnode;nrn_pragma_acc(parallel loop gang present(nt[0:1], strides[0:nstride], ncycles[0:nwarp], stridedispl[0:nwarp+1], rootbegin[0:nwarp+1], nodebegin[0:nwarp+1]) if(nt->compute_gpu) async(nt->stream_id)) nrn_pragma_omp(target teams loop if(nt->compute_gpu)) for(int icore=0;icore< ncore;icore+=warpsize) { int iwarp=icore/warpsize;int ic=icore &(warpsize - 1);int ncycle=ncycles[iwarp];int stride=strides+stridedispl[iwarp];int root=rootbegin[iwarp];int lastroot=rootbegin[iwarp+1];int firstnode=nodebegin[iwarp];int lastnode=nodebegin[iwarp+1];triang_interleaved2(nt, ic, ncycle, stride, lastnode);bksub_interleaved2(nt, root+ic, lastroot, ic, ncycle, stride, firstnode);} nrn_pragma_acc(wait(nt->stream_id))}void solve_interleaved1(int ith) { NrnThread nt=nrn_threads+ith;int ncell=nt->	ncell

int	nstride = ii.nstride

int *	cellsize = ii.cellsize

static size_t	groupsize = 32

static std::vector< int >	type_hints

int	nrn_nthread = 0

NrnThread *	nrn_threads = nullptr

bool	_nrn_skip_initmodel

void(*	nrn_mk_transfer_thread_data_ )()

static int	table_check_cnt_
	--> CoreNeuron class More...

static ThreadDatum *	table_check_

bool	use_solve_interleave

const int	NUM_STATS = 13

int	v_structure_change

int	diam_changed

int	hoc_errno_count

int	structure_change_cnt

Typedef Documentation

◆ bbcore_read_t

using coreneuron::bbcore_read_t = typedef void (*)(double*, int*, int*, int*, int, int, double*, Datum*, ThreadDatum*, NrnThread*, Memb_list*, double)

Definition at line 205 of file membfunc.hpp.

◆ bbcore_write_t

using coreneuron::bbcore_write_t = typedef void (*)(double*, int*, int*, int*, int, int, double*, Datum*, ThreadDatum*, NrnThread*, Memb_list*, double)

Definition at line 218 of file membfunc.hpp.

◆ Core2NrnWatchInfo

using coreneuron::Core2NrnWatchInfo = typedef std::vector<Core2NrnWatchInfoItem>

Definition at line 196 of file core2nrn_data_return.cpp.

◆ Core2NrnWatchInfoItem

using coreneuron::Core2NrnWatchInfoItem = typedef std::vector<std::pair<int, bool> >

All activated WATCH statements need activation on NEURON side.

Definition at line 195 of file core2nrn_data_return.cpp.

◆ Datum

using coreneuron::Datum = typedef int

Definition at line 23 of file nrnconf.h.

◆ DependencyTable

using coreneuron::DependencyTable = typedef std::vector<std::vector<int> >

Definition at line 51 of file coreneuron.hpp.

◆ HashCnt

using coreneuron::HashCnt = typedef std::map<size_t, std::pair<TNode*, int> >

Definition at line 100 of file cellorder1.cpp.

◆ IvocVect

using coreneuron::IvocVect = typedef fixed_vector<double>

Definition at line 72 of file ivocvect.hpp.

◆ List

using coreneuron::List = typedef Item

Definition at line 45 of file mod2c_core_thread.hpp.

◆ MechId

using coreneuron::MechId = typedef int

Definition at line 20 of file mech_mapping.cpp.

◆ MechNamesMapping

using coreneuron::MechNamesMapping = typedef std::map<MechId, std::map<VariableName, Offset, cmp_str> >

Definition at line 32 of file mech_mapping.cpp.

◆ mod_alloc_t

using coreneuron::mod_alloc_t = typedef void (*)(double*, Datum*, int)

Definition at line 23 of file membfunc.hpp.

◆ mod_f_t

using coreneuron::mod_f_t = typedef void (*)(NrnThread*, Memb_list*, int)

Definition at line 24 of file membfunc.hpp.

◆ N2V

using coreneuron::N2V = typedef std::map<std::string, PSD>

Definition at line 27 of file global_vars.cpp.

◆ NetBufReceive_t

using coreneuron::NetBufReceive_t = typedef void (*)(NrnThread*)

Definition at line 114 of file membfunc.hpp.

◆ NRB_P

using coreneuron::NRB_P = typedef std::pair<int, int>

Definition at line 877 of file nrn_acc_manager.cpp.

◆ nrn_watch_check_t

using coreneuron::nrn_watch_check_t = typedef void (*)(NrnThread*, Memb_list*)

Definition at line 119 of file membfunc.hpp.

◆ Offset

using coreneuron::Offset = typedef size_t

Definition at line 19 of file mech_mapping.cpp.

◆ Pfri

using coreneuron::Pfri = typedef int (*)()

Definition at line 24 of file nrnconf.h.

◆ Pfrpdat

using coreneuron::Pfrpdat = typedef Datum* (*) (void)

Definition at line 19 of file membfunc.hpp.

◆ Pfrv

using coreneuron::Pfrv = typedef void (*)()

Definition at line 25 of file register_mech.cpp.

◆ pnt_receive_t

using coreneuron::pnt_receive_t = typedef void (*)(Point_process*, int, double)

Definition at line 25 of file membfunc.hpp.

◆ PSD

using coreneuron::PSD = typedef std::pair<std::size_t, double*>

Definition at line 26 of file global_vars.cpp.

◆ secseg_it_type

using coreneuron::secseg_it_type = typedef secseg_map_type::iterator

Definition at line 23 of file nrnsection_mapping.hpp.

◆ secseg_map_type

using coreneuron::secseg_map_type = typedef std::map<int, segvec_type>

Definition at line 22 of file nrnsection_mapping.hpp.

◆ segvec_type

using coreneuron::segvec_type = typedef std::vector<int>

type to store every section and associated segments

Definition at line 21 of file nrnsection_mapping.hpp.

◆ SelfEventWeightMap

using coreneuron::SelfEventWeightMap = typedef std::map<int, std::vector<TQItem*> >

Definition at line 440 of file core2nrn_data_return.cpp.

◆ SerializedNames

using coreneuron::SerializedNames = typedef const char**

Definition at line 31 of file mech_mapping.hpp.

◆ Symbol

using coreneuron::Symbol = typedef char

Definition at line 25 of file nrnconf.h.

◆ thread_table_check_t

using coreneuron::thread_table_check_t = typedef void (*)(int, int, double*, Datum*, ThreadDatum*, NrnThread*, Memb_list*, int)

Definition at line 27 of file membfunc.hpp.

◆ TNI

using coreneuron::TNI = typedef std::pair<TNode*, int>

Definition at line 99 of file cellorder1.cpp.

◆ TNIVec

using coreneuron::TNIVec = typedef std::vector<TNI>

Definition at line 101 of file cellorder1.cpp.

◆ TQPair

using coreneuron::TQPair = typedef std::pair<double, TQItem*>

Definition at line 79 of file tqueue.hpp.

◆ VariableName

using coreneuron::VariableName = typedef const char*

Definition at line 21 of file mech_mapping.cpp.

◆ VecTNode

using coreneuron::VecTNode = typedef std::vector<TNode*>

Definition at line 17 of file tnode.hpp.

◆ VTN

using coreneuron::VTN = typedef VecTNode

Definition at line 25 of file cellorder2.cpp.

◆ VVTN

using coreneuron::VVTN = typedef std::vector<VTN>

Definition at line 26 of file cellorder2.cpp.

◆ VVVTN

using coreneuron::VVVTN = typedef std::vector<VVTN>

Definition at line 27 of file cellorder2.cpp.

Enumeration Type Documentation

◆ container

enum coreneuron::container

Enumerator
spltree
pq_que

Definition at line 120 of file tqueue.hpp.

◆ Layout

enum coreneuron::Layout

Enumerator
SoA
AoS

Definition at line 69 of file nrniv_decl.h.

◆ LFPCalculatorType

enum coreneuron::LFPCalculatorType

Enumerator
LineSource
PointSource

Definition at line 66 of file lfp.hpp.

◆ mech_type

enum coreneuron::mech_type

Mechanism type to be used from stdindex2ptr and nrn_dblpntr2nrncore (in Neuron) Values of the mechanism types should be negative numbers to avoid any conflict with mechanism types of Memb_list(>0) or time(0) passed from Neuron.

Enumerator
voltage
i_membrane_

Definition at line 19 of file nrniv_decl.h.

◆ ReportType

enum coreneuron::ReportType

Enumerator
SomaReport
CompartmentReport
SynapseReport
IMembraneReport
SectionReport
SummationReport

Definition at line 73 of file nrnreport.hpp.

◆ SectionType

enum coreneuron::SectionType

Enumerator
Cell
Cell
Soma
Axon
Dendrite
Apical
All

Definition at line 83 of file nrnreport.hpp.

◆ TargetType

enum coreneuron::TargetType

strong

Enumerator
Compartment
Cell
SectionSoma
SectionAxon
SectionDendrite
SectionApical
SectionSomaAll
SectionAxonAll
SectionDendriteAll
SectionApicalAll

Definition at line 59 of file nrnreport.hpp.

Function Documentation

◆ _cvode_sparse_thread()

template<typename SPFUN >

int coreneuron::_cvode_sparse_thread	(	void **	vpr,
		int	n,
		int *	x,
		SPFUN	fun,
		_threadargsproto_
	)

Definition at line 608 of file sparse_thread.hpp.

◆ _exp2syn_reg()

void coreneuron::_exp2syn_reg ( void )

◆ _expsyn_reg()

void coreneuron::_expsyn_reg ( void )

◆ _hh_reg()

void coreneuron::_hh_reg ( void )

◆ _modl_cleanup()

void coreneuron::_modl_cleanup ( )

Definition at line 367 of file register_mech.cpp.

◆ _netstim_reg()

void coreneuron::_netstim_reg ( void )

◆ _nrn_destroy_sparseobj_thread()

void coreneuron::_nrn_destroy_sparseobj_thread ( SparseObj * so )

inline

Definition at line 628 of file sparse_thread.hpp.

◆ _nrn_layout_reg()

void coreneuron::_nrn_layout_reg	(	int	type,
		int	layout
	)

Definition at line 176 of file register_mech.cpp.

◆ _nrn_setdata_reg()

void coreneuron::_nrn_setdata_reg	(	int	i,
		void()(double , Datum *)	call
	)

Definition at line 440 of file register_mech.cpp.

◆ _nrn_thread_reg0()

void coreneuron::_nrn_thread_reg0	(	int	i,
		void()(ThreadDatum )	f
	)

Definition at line 419 of file register_mech.cpp.

◆ _nrn_thread_reg1()

void coreneuron::_nrn_thread_reg1	(	int	i,
		void()(ThreadDatum )	f
	)

Definition at line 426 of file register_mech.cpp.

◆ _nrn_thread_table_reg() [1/2]

void coreneuron::_nrn_thread_table_reg	(	int	i,
		thread_table_check_t	f
	)

Definition at line 433 of file register_mech.cpp.

◆ _nrn_thread_table_reg() [2/2]

void coreneuron::_nrn_thread_table_reg	(	int	i,
		void()(int, int, double , Datum , ThreadDatum , NrnThread , Memb_list , int)	f
	)

◆ _passive_reg()

void coreneuron::_passive_reg ( void )

◆ _pattern_reg()

void coreneuron::_pattern_reg ( void )

◆ _ss_derivimplicit_thread()

template<typename DIFUN >

int coreneuron::_ss_derivimplicit_thread	(	int	n,
		int *	slist,
		int *	dlist,
		DIFUN	fun,
		_threadargsproto_
	)

Definition at line 67 of file ssimplic_thread.hpp.

◆ _ss_sparse_thread()

template<typename SPFUN >

int coreneuron::_ss_sparse_thread	(	SparseObj *	so,
		int	n,
		int *	s,
		int *	d,
		double *	t,
		double	dt,
		SPFUN	fun,
		int	linflag,
		_threadargsproto_
	)

Definition at line 30 of file ssimplic_thread.hpp.

◆ _stim_reg()

void coreneuron::_stim_reg ( void )

◆ _svclmp_reg()

void coreneuron::_svclmp_reg ( void )

◆ abort_run()

int coreneuron::abort_run ( int code )

Definition at line 40 of file abort.cpp.

◆ add_nrn_artcell()

void coreneuron::add_nrn_artcell	(	int	type,
		int	qi
	)

Definition at line 65 of file register_mech.cpp.

◆ add_nrn_fornetcons()

void coreneuron::add_nrn_fornetcons	(	int	type,
		int	indx
	)

Definition at line 54 of file register_mech.cpp.

◆ add_nrn_has_net_event()

void coreneuron::add_nrn_has_net_event ( int type )

Definition at line 42 of file register_mech.cpp.

◆ admin1()

static void coreneuron::admin1	(	int	ncell,
		VecTNode &	nodevec,
		int &	nwarp,
		int &	nstride,
		int *&	stride,
		int *&	firstnode,
		int *&	lastnode,
		int *&	cellsize
	)

static

Definition at line 499 of file cellorder1.cpp.

◆ admin2()

static void coreneuron::admin2	(	int	ncell,
		VecTNode &	nodevec,
		int &	nwarp,
		int &	nstride,
		int *&	stridedispl,
		int *&	strides,
		int *&	rootbegin,
		int *&	nodebegin,
		int *&	ncycles
	)

static

Prepare for solve_interleaved2.

One group of cells per warp.

warp[i] has a number of compute cycles (ncycle[i]) the index of its first root (rootbegin[i], last rootbegin[nwarp] = ncell) the index of its first node (nodebegin[i], last nodebegin[nwarp] = nnode)

Each compute cycle has a stride A stride is how many nodes are processed by a warp in one compute cycle There are nstride strides. nstride is the sum of ncycles of all warps. warp[i] has ncycle[i] strides same as sum of ncycle warp[i] has a stridedispl[i] which is stridedispl[i-1] + ncycle[i]. ie. The zeroth cycle of warp[j] works on stride[stridedispl[j]] The value of a stride beginning at node i (node i is computed by core 0 of some warp for some cycle) is determined by stride_length(i, j, nodevec)

Definition at line 601 of file cellorder1.cpp.

◆ alloc_mech()

void coreneuron::alloc_mech ( int memb_func_size_ )

Definition at line 86 of file register_mech.cpp.

◆ alloc_mpi_space()

static void coreneuron::alloc_mpi_space ( )

static

Allocate space for spikes: 200 structs of {int gid; double time} coming from nrnmpi.h and array of int of the global domain size.

Definition at line 96 of file netpar.cpp.

◆ alloc_nrn_thread_memb()

static NrnThreadMembList* coreneuron::alloc_nrn_thread_memb	(	NrnThread *	nt,
		int	type
	)

static

Definition at line 140 of file patternstim.cpp.

◆ allocate_data_in_mechanism_nrn_init()

void coreneuron::allocate_data_in_mechanism_nrn_init ( )

Definition at line 21 of file finitialize.cpp.

◆ allocate_unified()

void* coreneuron::allocate_unified ( std::size_t num_bytes )

Allocate unified memory in GPU builds iff GPU enabled, otherwise new.

Definition at line 26 of file memory.cpp.

◆ allocate_unique()

template<typename T , typename Alloc , typename... Args>

auto coreneuron::allocate_unique	(	const Alloc &	alloc,
		Args &&...	args
	)

Definition at line 98 of file memory.h.

◆ analyze()

static void coreneuron::analyze ( VVTN & levels )

static

Definition at line 405 of file cellorder2.cpp.

◆ aos2aos_copy()

static void coreneuron::aos2aos_copy	(	size_t	n,
		int	sz,
		double *	src,
		double **	dest
	)

static

AoS mechanism data copied to AoS data.

dest is an array of n pointers to the beginning of each sz length array. src is a contiguous array of n segments of size sz.

Definition at line 104 of file core2nrn_data_return.cpp.

◆ artcell_net_move()

void coreneuron::artcell_net_move	(	void **	v,
		Point_process *	pnt,
		double	tt
	)

Definition at line 310 of file netcvode.cpp.

◆ artcell_net_send()

void coreneuron::artcell_net_send	(	void **	v,
		int	weight_index_,
		Point_process *	pnt,
		double	td,
		double	flag
	)

Definition at line 103 of file netcvode.cpp.

◆ at_time()

constexpr int coreneuron::at_time	(	NrnThread *	nt,
		double	te
	)

constexpr

Definition at line 198 of file multicore.hpp.

◆ BBS_netpar_solve()

void coreneuron::BBS_netpar_solve ( double tstop )

Definition at line 604 of file netpar.cpp.

◆ bksub()

static void coreneuron::bksub ( NrnThread * _nt )

Definition at line 54 of file solve_core.cpp.

◆ bksub_interleaved()

static void coreneuron::bksub_interleaved	(	NrnThread *	nt,
		int	icell,
		int	icellsize,
		int	,
		int *	stride,
		int *	firstnode
	)

static

Definition at line 461 of file cellorder.cpp.

◆ call_prcellstate_for_prcellgid()

void coreneuron::coreneuron::call_prcellstate_for_prcellgid	(	int	prcellgid,
		int	compute_gpu,
		int	is_init
	)

Definition at line 358 of file main1.cpp.

◆ capacitance_reg()

void coreneuron::capacitance_reg ( void )

Definition at line 31 of file capac.cpp.

◆ check()

void coreneuron::check ( VecTNode & nodevec )

static

Check that the first compartments of nodevec are the root nodes (cells)

Definition at line 375 of file cellorder1.cpp.

◆ check_bbcore_write_version()

void coreneuron::check_bbcore_write_version ( const char * version )

Definition at line 128 of file nrnoc_aux.cpp.

◆ check_state()

static int coreneuron::check_state	(	int	n,
		int *	s,
		_threadargsproto_
	)

static

Definition at line 17 of file ssimplic_thread.hpp.

◆ checkpoint_restore_patternstim()

void coreneuron::checkpoint_restore_patternstim	(	int	,
		double	,
		_threadargsproto_
	)

◆ checkpoint_save_patternstim()

int coreneuron::checkpoint_save_patternstim ( _threadargsproto_ )

◆ checkrace()

static void coreneuron::checkrace	(	TNode *	nd,
		VTN &	nodes
	)

static

Definition at line 265 of file cellorder2.cpp.

◆ chklevel()

void coreneuron::chklevel	(	VTN &	level,
		size_t	nident = `8`
	)

Definition at line 30 of file cellorder2.cpp.

◆ clear_event_queue()

void coreneuron::clear_event_queue ( )

Definition at line 47 of file cvodestb.cpp.

◆ clear_inv_perm_for_selfevent_targets()

static void coreneuron::clear_inv_perm_for_selfevent_targets ( )

static

Definition at line 432 of file core2nrn_data_return.cpp.

◆ clear_spike_vectors()

void coreneuron::clear_spike_vectors ( )

Definition at line 296 of file output_spikes.cpp.

◆ cnrn_target_copyin()

template<typename T >

T* coreneuron::cnrn_target_copyin	(	std::string_view	file,
		int	line,
		const T *	h_ptr,
		std::size_t	len = `1`
	)

Definition at line 110 of file offload.hpp.

◆ cnrn_target_copyin_debug()

void coreneuron::cnrn_target_copyin_debug	(	std::string_view	file,
		int	line,
		std::size_t	sizeof_T,
		std::type_info const &	typeid_T,
		void const *	h_ptr,
		std::size_t	len,
		void *	d_ptr
	)

Definition at line 97 of file nrn_acc_manager.cpp.

◆ cnrn_target_delete()

template<typename T >

void coreneuron::cnrn_target_delete	(	std::string_view	file,
		int	line,
		T *	h_ptr,
		std::size_t	len = `1`
	)

Definition at line 132 of file offload.hpp.

◆ cnrn_target_delete_debug()

void coreneuron::cnrn_target_delete_debug	(	std::string_view	file,
		int	line,
		std::size_t	sizeof_T,
		std::type_info const &	typeid_T,
		void const *	h_ptr,
		std::size_t	len
	)

Definition at line 111 of file nrn_acc_manager.cpp.

◆ cnrn_target_deviceptr_debug()

void coreneuron::cnrn_target_deviceptr_debug	(	std::string_view	file,
		int	line,
		std::type_info const &	typeid_T,
		void const *	h_ptr,
		void *	d_ptr
	)

Definition at line 124 of file nrn_acc_manager.cpp.

◆ cnrn_target_deviceptr_or_present()

template<typename T >

T* coreneuron::cnrn_target_deviceptr_or_present	(	std::string_view	file,
		int	line,
		bool	must_be_present_or_null,
		const T *	h_ptr
	)

Definition at line 71 of file offload.hpp.

◆ cnrn_target_get_num_devices()

int coreneuron::cnrn_target_get_num_devices ( )

Definition at line 227 of file nrn_acc_manager.cpp.

◆ cnrn_target_is_present_debug()

void coreneuron::cnrn_target_is_present_debug	(	std::string_view	file,
		int	line,
		std::type_info const &	typeid_T,
		void const *	h_ptr,
		void *	d_ptr
	)

Definition at line 135 of file nrn_acc_manager.cpp.

◆ cnrn_target_memcpy_to_device()

template<typename T >

void coreneuron::cnrn_target_memcpy_to_device	(	std::string_view	file,
		int	line,
		T *	d_ptr,
		const T *	h_ptr,
		std::size_t	len = `1`
	)

Definition at line 150 of file offload.hpp.

◆ cnrn_target_memcpy_to_device_debug()

void coreneuron::cnrn_target_memcpy_to_device_debug	(	std::string_view	file,
		int	line,
		std::size_t	sizeof_T,
		std::type_info const &	typeid_T,
		void const *	h_ptr,
		std::size_t	len,
		void *	d_ptr
	)

Definition at line 146 of file nrn_acc_manager.cpp.

◆ cnrn_target_set_default_device()

void coreneuron::cnrn_target_set_default_device ( int device_num )

Definition at line 243 of file nrn_acc_manager.cpp.

◆ cnrn_target_update_on_device()

template<typename T >

void coreneuron::cnrn_target_update_on_device	(	std::string_view	file,
		int	line,
		const T *	h_ptr,
		std::size_t	len = `1`
	)

Definition at line 175 of file offload.hpp.

◆ cnrn_version()

std::string coreneuron::cnrn_version ( )

Definition at line 408 of file main1.cpp.

◆ cnt2displ()

static int* coreneuron::cnt2displ ( int * cnt )

static

Definition at line 59 of file have2want.h.

◆ copy_align_array()

template<typename T >

void coreneuron::copy_align_array	(	T *&	dest,
		T *	src,
		size_t	n
	)

Definition at line 118 of file cellorder.hpp.

◆ copy_array()

template<typename T >

void coreneuron::copy_array	(	T *&	dest,
		T *	src,
		size_t	n
	)

Definition at line 111 of file cellorder.hpp.

◆ copy_ivoc_vect_to_device()

void coreneuron::copy_ivoc_vect_to_device	(	const IvocVect &	from,
		IvocVect &	to
	)

Definition at line 784 of file nrn_acc_manager.cpp.

◆ core2nrn_corepointer()

static void coreneuron::core2nrn_corepointer	(	int	tid,
		NrnThreadMembList *	tml
	)

static

Copy back COREPOINTER info to NEURON.

Definition at line 114 of file core2nrn_data_return.cpp.

◆ core2nrn_data_return()

void coreneuron::core2nrn_data_return ( )

copy data back to NEURON.

Copies back to NEURON everything needed to analyze and continue simulation.

Copies t, voltage, i_membrane_ if it used, and mechanism param data. Copies event queue and related state, e.g. WATCH, VecPlayContinuous.

I.e. voltage, i_membrane_, mechanism data, event queue, WATCH state, Play state, etc.

Definition at line 217 of file core2nrn_data_return.cpp.

◆ core2nrn_PreSyn_flag()

static void coreneuron::core2nrn_PreSyn_flag ( NrnThread & nt )

static

Definition at line 381 of file core2nrn_data_return.cpp.

◆ core2nrn_tqueue()

void coreneuron::core2nrn_tqueue ( NrnThread & nt )

static

Copy event queue and related state back to NEURON.

Definition at line 522 of file core2nrn_data_return.cpp.

◆ core2nrn_tqueue_item()

static bool coreneuron::core2nrn_tqueue_item	(	TQItem *	q,
		SelfEventWeightMap &	sewm,
		NrnThread &	nt
	)

static

Definition at line 443 of file core2nrn_data_return.cpp.

◆ core2nrn_vecplay()

static void coreneuron::core2nrn_vecplay ( )

static

Transfer VecPlay indices to NEURON.

Definition at line 329 of file core2nrn_data_return.cpp.

◆ core2nrn_watch()

static void coreneuron::core2nrn_watch ( )

static

Callbacks into NEURON for WatchCondition.

Definition at line 287 of file core2nrn_data_return.cpp.

◆ create_interleave_info()

void coreneuron::create_interleave_info ( )

Definition at line 96 of file cellorder.cpp.

◆ create_report_configurations()

std::vector< ReportConfiguration > coreneuron::create_report_configurations	(	const std::string &	filename,
		const std::string &	output_dir,
		SpikesInfo &	spikes_info
	)

Definition at line 106 of file report_configuration_parser.cpp.

◆ create_report_handler()

std::unique_ptr<ReportHandler> coreneuron::create_report_handler	(	const ReportConfiguration &	config,
		const SpikesInfo &	spikes_info
	)

Definition at line 425 of file main1.cpp.

◆ create_tml()

NrnThreadMembList * coreneuron::create_tml	(	NrnThread &	nt,
		int	mech_id,
		Memb_func &	memb_func,
		int &	shadow_rhs_cnt,
		const std::vector< int > &	mech_types,
		const std::vector< int > &	nodecounts
	)

Definition at line 64 of file multicore.cpp.

◆ deallocate_unified()

void coreneuron::deallocate_unified	(	void *	ptr,
		std::size_t	num_bytes
	)

Deallocate memory allocated by allocate_unified.

Definition at line 44 of file memory.cpp.

◆ default_rendezvous()

static int coreneuron::default_rendezvous ( HAVEWANT_t key )

static

Definition at line 55 of file have2want.h.

◆ delete_fornetcon_info()

void coreneuron::delete_fornetcon_info ( NrnThread & nt )

Definition at line 705 of file nrn_setup.cpp.

◆ delete_ivoc_vect_from_device()

void coreneuron::delete_ivoc_vect_from_device ( IvocVect & vec )

Definition at line 800 of file nrn_acc_manager.cpp.

◆ delete_nrnthreads_on_device()

void coreneuron::delete_nrnthreads_on_device	(	NrnThread *	threads,
		int	nthreads
	)

Cleanup device memory that is being tracked by the OpenACC runtime.

This function painstakingly calls cnrn_target_delete in reverse order on all pointers that were passed to cnrn_target_copyin in setup_nrnthreads_on_device. This cleanup ensures that if the GPU is initialised multiple times from the same process then the OpenACC runtime will not be polluted with old pointers, which can cause errors. In particular if we do:

{
  // ... some_ptr is dynamically allocated ...
  cnrn_target_copyin(some_ptr, some_size);
  // ... do some work ...
  // cnrn_target_delete(some_ptr);
  free(some_ptr);
}
{
  // ... same_ptr_again is dynamically allocated at the same address ...
  cnrn_target_copyin(same_ptr_again, some_other_size); // ERROR
}

the application will/may abort with an error such as: FATAL ERROR: variable in data clause is partially present on the device. The pattern above is typical of calling CoreNEURON on GPU multiple times in the same process.

Definition at line 1151 of file nrn_acc_manager.cpp.

◆ delete_trajectory_requests()

void coreneuron::delete_trajectory_requests ( NrnThread & nt )

Definition at line 899 of file nrn_setup.cpp.

◆ deliver_net_events()

void coreneuron::deliver_net_events ( NrnThread * nt )

Definition at line 24 of file cvodestb.cpp.

◆ depend_append()

static int coreneuron::depend_append	(	int	idep,
		int *	dependencies,
		int	deptype,
		int	type
	)

static

Definition at line 273 of file register_mech.cpp.

◆ derivimplicit_thread()

template<typename F >

int coreneuron::derivimplicit_thread	(	int	n,
		int *	slist,
		int *	dlist,
		F	fun,
		_threadargsproto_
	)

Definition at line 86 of file mod2c_core_thread.hpp.

◆ destroy_interleave_info()

void coreneuron::destroy_interleave_info ( )

Definition at line 101 of file cellorder.cpp.

◆ determine_inputpresyn()

void coreneuron::determine_inputpresyn ( )

If PreSyn or InputPreSyn is already in the map

Increase PreSyn count

Increase InputPreSyn count

Create InputPreSyn and increase its count

Increase negative PreSyn count

Resize the vector to its actual size of the netcons put in it

Definition at line 248 of file nrn_setup.cpp.

◆ direct_mode_initialize()

void coreneuron::direct_mode_initialize ( )

All state from NEURON necessary to continue a run.

In NEURON direct mode, we desire the exact behavior of ParallelContext.psolve(tstop). I.e. a sequence of such calls with and without intervening calls to h.finitialize(). Most state (structure and data of the substantive model) has been copied from NEURON during nrn_setup. Now we need to copy the event queue and set up any other invalid internal structures. I.e basically the nrn_finitialize above but without changing any simulation data. We follow some of the strategy of checkpoint_initialize.

Definition at line 50 of file nrn2core_data_init.cpp.

◆ dist2child()

size_t coreneuron::dist2child ( TNode * nd )

Definition at line 153 of file cellorder2.cpp.

◆ dt2thread()

void coreneuron::dt2thread ( double adt )

Definition at line 70 of file fadvance_core.cpp.

◆ ecalloc()

void * coreneuron::ecalloc	(	size_t	n,
		size_t	size
	)

Definition at line 85 of file nrnoc_aux.cpp.

◆ ecalloc_align()

void* coreneuron::ecalloc_align	(	size_t	n,
		size_t	size,
		size_t	alignment
	)

◆ eliminate_crace()

static void coreneuron::eliminate_crace	(	TNode *	nd,
		VTN &	nodes
	)

static

Definition at line 309 of file cellorder2.cpp.

◆ eliminate_prace()

static void coreneuron::eliminate_prace	(	TNode *	nd,
		VTN &	nodes
	)

static

Definition at line 296 of file cellorder2.cpp.

◆ eliminate_race()

static bool coreneuron::eliminate_race	(	TNode *	nd,
		size_t	d,
		VTN &	nodes,
		TNode *	look
	)

static

Definition at line 273 of file cellorder2.cpp.

◆ emalloc()

void * coreneuron::emalloc ( size_t size )

static

Definition at line 30 of file mpispike.cpp.

◆ emalloc_align()

void* coreneuron::emalloc_align	(	size_t	size,
		size_t	alignment
	)

◆ erealloc()

void * coreneuron::erealloc	(	void *	ptr,
		size_t	size
	)

Definition at line 94 of file nrnoc_aux.cpp.

◆ euler_thread()

template<typename F >

int coreneuron::euler_thread	(	int	neqn,
		int *	var,
		int *	der,
		F	fun,
		_threadargsproto_
	)

Definition at line 74 of file mod2c_core_thread.hpp.

◆ fast_imem_free()

void coreneuron::fast_imem_free ( )

Definition at line 21 of file fast_imem.cpp.

◆ final_nodevec_cmp()

static bool coreneuron::final_nodevec_cmp	(	TNode *	a,
		TNode *	b
	)

static

Definition at line 442 of file cellorder2.cpp.

◆ finalize_report()

void coreneuron::finalize_report ( )

Definition at line 79 of file nrnreport.cpp.

◆ fixed_play_continuous()

void coreneuron::fixed_play_continuous ( NrnThread * nt )

Definition at line 81 of file cvodestb.cpp.

◆ fornetcon_slot()

static int* coreneuron::fornetcon_slot	(	const int	mtype,
		const int	instance,
		const int	fnslot,
		const NrnThread &	nt
	)

static

If FOR_NETCON in use, setup NrnThread fornetcon related info.

i.e NrnThread._fornetcon_perm_indices, NrnThread._fornetcon_weight_perm, and the relevant dparam element of each mechanism instance that uses a FOR_NETCONS statement.

Makes use of nrn_fornetcon_cnt_, nrn_fornetcon_type_, and nrn_fornetcon_index_ that were specified during registration of mechanisms that use FOR_NETCONS.

nrn_fornetcon_cnt_ is the number of mechanisms that use FOR_NETCONS, nrn_fornetcon_type_ is an int array of size nrn_fornetcon_cnt, that specifies the mechanism type. nrn_fornetcon_index_ is an int array of size nrn_fornetcon_cnt, that specifies the index into an instance's dparam int array having the fornetcon semantics.

FOR_NETCONS (args) means to loop over all NetCon connecting to this target instance and args are the names of the items of each NetCon's weight vector (same as the enclosing NET_RECEIVE but possible different local names).

NrnThread._weights is a vector of weight groups where the number of groups is the number of NetCon in this thread and each group has a size equal to the number of args in the target NET_RECEIVE block. The order of these groups is the NetCon Object order in HOC (the construction order). So the weight vector indices for the NetCons in the FOR_NETCONS loop are not adjacent.

NrnThread._fornetcon_weight_perm is an index vector into the NrnThread._weight vector such that the list of indices that targets a mechanism instance are adjacent. NrnThread._fornetcon_perm_indices is an index vector into the NrnThread._fornetcon_weight_perm to the first of the list of NetCon weights that target the instance. The index of _fornetcon_perm_indices containing this first in the list is stored in the mechanism instances dparam at the dparam's semantic fornetcon slot. (Note that the next index points to the first index of the next target instance.)

Definition at line 60 of file setup_fornetcon.cpp.

◆ freematrix()

void coreneuron::freematrix ( double ** matrix )

Definition at line 66 of file nrnoc_aux.cpp.

◆ freevector()

void coreneuron::freevector ( double * p )

Definition at line 52 of file nrnoc_aux.cpp.

◆ full_search()

static int coreneuron::full_search	(	NrnThread &	nt,
		double *	pd
	)

static

Definition at line 162 of file node_permute.cpp.

◆ get_data_index()

int coreneuron::get_data_index	(	int	node_index,
		int	variable_index,
		int	mtype,
		Memb_list *	ml
	)

Definition at line 19 of file data_layout.cpp.

◆ get_nrn_trajectory_requests()

void coreneuron::get_nrn_trajectory_requests ( int bsize )

Definition at line 130 of file main1.cpp.

◆ get_var_location_from_var_name()

double * coreneuron::get_var_location_from_var_name	(	int	mech_id,
		const char *	variable_name,
		Memb_list *	ml,
		int	node_index
	)

Definition at line 39 of file mech_mapping.cpp.

◆ gpu_enabled()

bool coreneuron::gpu_enabled ( )

Check if GPU support is enabled.

This returns true if GPU support was enabled at compile time and at runtime via coreneuron.gpu = True and/or –gpu, otherwise it returns false.

Definition at line 18 of file memory.cpp.

◆ group_order2()

void coreneuron::group_order2	(	VecTNode &	nodevec,
		size_t	groupsize,
		size_t	ncell
	)

Implementation of the advanced interleaving strategy (interleave_permute_type == 2)

The main steps are the following:

warp_balance function creates balanced groups of cells.
The compartments/tree nodes populate the groups vector (VVVTN) based on their groudindex and their level (see level_from_root).
The analyze() & question2() functions (operating per group) make sure that each cell is still a tree (treenode_order) and that the dependent nodes belong to separate warps.

Definition at line 460 of file cellorder2.cpp.

◆ handle_forward_skip()

void coreneuron::handle_forward_skip	(	double	forwardskip,
		int	prcellgid
	)

Definition at line 384 of file main1.cpp.

◆ have_to_want()

static void coreneuron::have_to_want	(	HAVEWANT_t *	have,
		int	have_size,
		HAVEWANT_t *	want,
		int	want_size,
		HAVEWANT_t *&	send_to_want,
		int *&	send_to_want_cnt,
		int *&	send_to_want_displ,
		HAVEWANT_t *&	recv_from_have,
		int *&	recv_from_have_cnt,
		int *&	recv_from_have_displ,
		int(*)(HAVEWANT_t)	rendezvous_rank
	)

static

Definition at line 128 of file have2want.h.

◆ hoc_Emalloc()

void * coreneuron::hoc_Emalloc ( size_t size )

Definition at line 80 of file nrnoc_aux.cpp.

◆ hoc_execerror()

void coreneuron::hoc_execerror	(	const char *	s1,
		const char *	s2
	)

Definition at line 39 of file nrnoc_aux.cpp.

◆ hoc_Exp()

double coreneuron::hoc_Exp ( double x )

Definition at line 109 of file nrnoc_aux.cpp.

◆ hoc_malchk()

void coreneuron::hoc_malchk ( void )

Definition at line 83 of file nrnoc_aux.cpp.

◆ hoc_reg_ba()

void coreneuron::hoc_reg_ba	(	int	mt,
		mod_f_t	f,
		int	type
	)

Definition at line 378 of file register_mech.cpp.

◆ hoc_reg_bbcore_read()

void coreneuron::hoc_reg_bbcore_read	(	int	type,
		bbcore_read_t	f
	)

Definition at line 29 of file register_mech.cpp.

◆ hoc_reg_bbcore_write()

void coreneuron::hoc_reg_bbcore_write	(	int	type,
		bbcore_write_t	f
	)

Definition at line 35 of file register_mech.cpp.

◆ hoc_register_dparam_semantics()

void coreneuron::hoc_register_dparam_semantics	(	int	type,
		int	ix,
		const char *	name
	)

Definition at line 207 of file register_mech.cpp.

◆ hoc_register_net_receive_buffering()

void coreneuron::hoc_register_net_receive_buffering	(	NetBufReceive_t	f,
		int	type
	)

Definition at line 180 of file register_mech.cpp.

◆ hoc_register_net_send_buffering()

void coreneuron::hoc_register_net_send_buffering ( int type )

Definition at line 184 of file register_mech.cpp.

◆ hoc_register_prop_size()

void coreneuron::hoc_register_prop_size	(	int	type,
		int	psize,
		int	dpsize
	)

Definition at line 192 of file register_mech.cpp.

◆ hoc_register_var()

void coreneuron::hoc_register_var	(	DoubScal *	ds,
		DoubVec *	dv,
		VoidFunc *
	)

Definition at line 31 of file global_vars.cpp.

◆ hoc_register_watch_check()

void coreneuron::hoc_register_watch_check	(	nrn_watch_check_t	nwc,
		int	type
	)

Definition at line 188 of file register_mech.cpp.

◆ hoc_warning()

void coreneuron::hoc_warning	(	const char *	s1,
		const char *	s2
	)

Definition at line 44 of file nrnoc_aux.cpp.

◆ ident_statistic()

static void coreneuron::ident_statistic	(	VecTNode &	nodevec,
		size_t	ncell
	)

static

Definition at line 276 of file cellorder1.cpp.

◆ if()

coreneuron::if ( ncell = = 0 )

Definition at line 637 of file cellorder.cpp.

◆ init_gpu()

void coreneuron::init_gpu ( )

◆ init_net_events()

void coreneuron::init_net_events ( )

Definition at line 53 of file cvodestb.cpp.

◆ initnrn()

void coreneuron::initnrn ( )

Definition at line 102 of file register_mech.cpp.

◆ input_presyn_size()

size_t coreneuron::input_presyn_size ( void )

Definition at line 1035 of file nrn_setup.cpp.

◆ interleave_comp()

static bool coreneuron::interleave_comp	(	TNode *	a,
		TNode *	b
	)

static

Definition at line 451 of file cellorder1.cpp.

◆ interleave_order()

int* coreneuron::interleave_order	(	int	ith,
		int	ncell,
		int	nnode,
		int *	parent
	)

Function that performs the permutation of the cells such that the execution threads access coalesced memory.

Parameters

ith	NrnThread to access
ncell	number of cells in NrnThread
nnode	number of compartments in the ncells
parent	parent indices of cells

Returns: int* order, interleaved order of the cells

Definition at line 290 of file cellorder.cpp.

◆ intersection_gids()

template<typename T >

std::vector<T> coreneuron::intersection_gids	(	const NrnThread &	nt,
		std::vector< T > &	target_gids
	)

Definition at line 17 of file report_handler.cpp.

◆ interthread_enqueue()

void coreneuron::interthread_enqueue ( NrnThread * nt )

Definition at line 138 of file netcvode.cpp.

◆ inv_permute()

static int coreneuron::inv_permute	(	int	i,
		NrnThread &	nt
	)

static

Definition at line 32 of file prcellstate.cpp.

◆ inverse_permute()

int * coreneuron::inverse_permute	(	int *	p,
		int	n
	)

Definition at line 131 of file node_permute.cpp.

◆ inverse_permute_copy()

static void coreneuron::inverse_permute_copy	(	size_t	n,
		double *	permuted_src,
		double *	dest,
		int *	permute
	)

static

permuted array copied to unpermuted array If permute is NULL then just a copy

Definition at line 50 of file core2nrn_data_return.cpp.

◆ invert_permute()

static void coreneuron::invert_permute	(	int *	p,
		int	n
	)

static

Definition at line 139 of file node_permute.cpp.

◆ ion_reg()

void coreneuron::ion_reg	(	const char *	,
		double
	)

◆ ion_write_depend()

static void coreneuron::ion_write_depend	(	int	type,
		int	etype
	)

static

Definition at line 259 of file register_mech.cpp.

◆ is_child_race2()

static bool coreneuron::is_child_race2 ( TNode * nd )

static

Definition at line 132 of file cellorder2.cpp.

◆ is_parent_race2()

static bool coreneuron::is_parent_race2 ( TNode * nd )

static

Definition at line 102 of file cellorder2.cpp.

◆ ktf()

constexpr double coreneuron::ktf ( double celsius )

constexpr

Definition at line 124 of file membfunc.hpp.

◆ level_from_leaf()

size_t coreneuron::level_from_leaf ( VecTNode & nodevec )

Definition at line 218 of file cellorder1.cpp.

◆ level_from_root()

size_t coreneuron::level_from_root ( VecTNode & nodevec )

Definition at line 203 of file cellorder1.cpp.

◆ local_spikevec_sort()

static void coreneuron::local_spikevec_sort	(	std::vector< double > &	isvect,
		std::vector< int > &	isvecg,
		std::vector< double > &	osvect,
		std::vector< int > &	osvecg
	)

static

Definition at line 67 of file output_spikes.cpp.

◆ makematrix()

double ** coreneuron::makematrix	(	size_t	nrows,
		size_t	ncols
	)

Definition at line 58 of file nrnoc_aux.cpp.

◆ makevector()

double * coreneuron::makevector ( size_t size )

Definition at line 48 of file nrnoc_aux.cpp.

◆ mech_data_layout_transform()

template<typename T >

void coreneuron::mech_data_layout_transform	(	T *	data,
		int	cnt,
		int	sz,
		int	layout
	)

inline

Definition at line 89 of file phase2.cpp.

◆ memb_list_size()

size_t coreneuron::memb_list_size	(	NrnThreadMembList *	tml,
		bool	include_data
	)

Definition at line 989 of file nrn_setup.cpp.

◆ mk_mech() [1/3]

static void coreneuron::mk_mech ( )

static

Definition at line 68 of file mk_mech.cpp.

◆ mk_mech() [2/3]

void coreneuron::mk_mech ( const char * datpath )

Read meta data about the mechanisms and allocate corresponding mechanism management data structures.

Definition at line 43 of file mk_mech.cpp.

◆ mk_mech() [3/3]

static void coreneuron::mk_mech ( std::istream & s )

static

Allocate space for mechanism related data structures

Read all the mechanisms and their meta data

Calling _reg functions for the default mechanisms from the file mech/cfile/cabvars.h

Allocate space for mechanism related data structures

Read all the mechanisms and their meta data

Calling _reg functions for the default mechanisms from the file mech/cfile/cabvars.h

Definition at line 80 of file mk_mech.cpp.

◆ mk_netcvode()

void coreneuron::mk_netcvode ( )

Definition at line 41 of file netcvode.cpp.

◆ mk_spikevec_buffer()

void coreneuron::mk_spikevec_buffer ( int sz )

Definition at line 50 of file output_spikes.cpp.

◆ ml_permute()

static int coreneuron::ml_permute	(	int	i,
		Memb_list *	ml
	)

static

Definition at line 46 of file prcellstate.cpp.

◆ model_size()

size_t coreneuron::model_size ( bool detailed_report )

Definition at line 1047 of file nrn_setup.cpp.

◆ modl_reg()

void coreneuron::modl_reg ( )

Mechanism registration function.

If external mechanisms present then use modl_reg function generated in mod_func.cpp otherwise use empty one.

Definition at line 31 of file enginemech.cpp.

◆ move_nodes()

static void coreneuron::move_nodes	(	size_t	start,
		size_t	length,
		size_t	dst,
		VTN &	nodes
	)

static

Definition at line 181 of file cellorder2.cpp.

◆ move_range()

template<typename T >

static void coreneuron::move_range	(	size_t	start,
		size_t	length,
		size_t	dst,
		std::vector< T > &	v
	)

static

Definition at line 167 of file cellorder2.cpp.

◆ mpi_manager()

mpi_manager_t& coreneuron::mpi_manager ( )

inline

Definition at line 51 of file nrnmpi.h.

◆ ncs2nrn_integrate()

void coreneuron::ncs2nrn_integrate ( double tstop )

Definition at line 488 of file netcvode.cpp.

◆ net_event()

void coreneuron::net_event	(	Point_process *	pnt,
		double	time
	)

Definition at line 107 of file netcvode.cpp.

◆ net_move()

void coreneuron::net_move	(	void **	v,
		Point_process *	pnt,
		double	tt
	)

Definition at line 296 of file netcvode.cpp.

◆ net_receive_buffer_order()

static void coreneuron::net_receive_buffer_order ( NetReceiveBuffer_t * nrb )

static

Definition at line 888 of file nrn_acc_manager.cpp.

◆ net_sem_from_gpu()

void coreneuron::net_sem_from_gpu	(	int	sendtype,
		int	i_vdata,
		int	weight_index_,
		int	ith,
		int	ipnt,
		double	td,
		double	flag
	)

Definition at line 59 of file netcvode.cpp.

◆ net_send()

void coreneuron::net_send	(	void **	v,
		int	weight_index_,
		Point_process *	pnt,
		double	td,
		double	flag
	)

Definition at line 77 of file netcvode.cpp.

◆ netpar_tid_gid2ps()

void coreneuron::netpar_tid_gid2ps	(	int	tid,
		int	gid,
		PreSyn **	ps,
		InputPreSyn **	psi
	)

for gid < 0 returns the PreSyn* in the thread (tid) specific map.

Definition at line 225 of file nrn_setup.cpp.

◆ next_leaf()

static size_t coreneuron::next_leaf	(	TNode *	nd,
		VTN &	nodes
	)

static

Definition at line 254 of file cellorder2.cpp.

◆ node_interleave_order()

static void coreneuron::node_interleave_order	(	int	ncell,
		VecTNode &	nodevec
	)

static

Naive interleaving strategy (interleave_permute_type == 1)

Sort so nodevec[ncell:nnode] cell instances are interleaved. Keep the secondary ordering with respect to treenode_order so each cell is still a tree.

Parameters

ncell	number of cells (trees)
nodevec	vector that contains compartments (nodes of the trees)

Definition at line 472 of file cellorder1.cpp.

◆ node_order()

int * coreneuron::node_order	(	int	ncell,
		int	nnode,
		int *	parents,
		int &	nwarp,
		int &	nstride,
		int *&	stride,
		int *&	firstnode,
		int *&	lastnode,
		int *&	cellsize,
		int *&	stridedispl
	)

Function that returns a permutation of length nnode.

There are two permutation strategies: For interleave_permute_type == 1 : Naive interleaving -> Each execution thread deals with one Hines matrix (cell) For interleave_permute_type == 2 : Advanced interleaving -> Each Hines matrix is solved by multiple execution threads (with coalesced memory access as well)

Parameters

ncell	number of cells
nnode	number of compartments in the ncells
parents	parent indices of the cells
nwarp	number of warps
nstride	nstride is the maximum cell size (not counting root)
stride	stride[i] is the number of cells with an ith node: using stride[i] we know how many positions to move in order to access the next element of the same cell (given that the cells are ordered with the treenode_order).
firstnode	firstnode[i] is the index of the first nonroot node of the cell
lastnode	lastnode[i] is the index of the last node of the cell
cellsize	cellsize is the number of nodes in the cell not counting root.
stridedispl

Returns: int* : a permutation of length nnode

Definition at line 300 of file cellorder1.cpp.

◆ node_permute()

void coreneuron::node_permute	(	int *	vec,
		int	n,
		int *	permute
	)

Definition at line 337 of file node_permute.cpp.

◆ nonvint()

void coreneuron::nonvint ( NrnThread * _nt )

Definition at line 232 of file fadvance_core.cpp.

◆ nrn2core_patstim_share_info()

void coreneuron::nrn2core_patstim_share_info ( )

static

Definition at line 419 of file nrn2core_data_init.cpp.

◆ nrn2core_PreSyn_flag_receive()

void coreneuron::nrn2core_PreSyn_flag_receive ( int tid )

Definition at line 399 of file core2nrn_data_return.cpp.

◆ nrn2core_tqueue()

static void coreneuron::nrn2core_tqueue ( )

static

Copy each thread's queue from NEURON.

Definition at line 154 of file nrn2core_data_init.cpp.

◆ nrn2core_transfer_watch_condition()

void coreneuron::nrn2core_transfer_watch_condition	(	int	tid,
		int	pnttype,
		int	pntindex,
		int	watch_index,
		int	triggered
	)

static

Definition at line 356 of file nrn2core_data_init.cpp.

◆ nrn2ncs_outputevent()

void coreneuron::nrn2ncs_outputevent	(	int	netcon_output_index,
		double	firetime
	)

◆ nrn_abort()

void coreneuron::nrn_abort ( int errcode )

Definition at line 13 of file utils.cpp.

◆ nrn_alloc_capacitance()

void coreneuron::nrn_alloc_capacitance	(	double *	data,
		Datum *	pdata,
		int	type
	)

Definition at line 130 of file capac.cpp.

◆ nrn_alloc_ion()

void coreneuron::nrn_alloc_ion	(	double *	data,
		Datum *	pdata,
		int	type
	)

◆ nrn_artcell_instantiate()

Point_process * coreneuron::nrn_artcell_instantiate ( const char * mechname )

Definition at line 178 of file patternstim.cpp.

◆ nrn_ba()

void coreneuron::nrn_ba	(	NrnThread *	nt,
		int	bat
	)

Definition at line 258 of file fadvance_core.cpp.

◆ nrn_cacheline_alloc()

void * coreneuron::nrn_cacheline_alloc	(	void **	memptr,
		size_t	size
	)

Definition at line 103 of file nrnoc_aux.cpp.

◆ nrn_calc_fast_imem()

void coreneuron::nrn_calc_fast_imem ( NrnThread * nt )

Definition at line 44 of file fast_imem.cpp.

◆ nrn_calc_fast_imem_init()

void coreneuron::nrn_calc_fast_imem_init ( NrnThread * nt )

Definition at line 62 of file fast_imem.cpp.

◆ nrn_cleanup()

void coreneuron::nrn_cleanup ( )

Definition at line 714 of file nrn_setup.cpp.

◆ nrn_cleanup_ion_map()

void coreneuron::nrn_cleanup_ion_map ( )

Cleanup global ion map created during mechanism registration.

function defined in coreneuron library

In case of coreneuron standalone execution nrn_ion_global_map can be deleted at the end of execution. But in case embedded run via neuron, mechanisms are registered only once i.e. during first call to coreneuron. This is why we call cleanup only in case of standalone coreneuron execution via nrniv-core or special-core.

Todo:: coreneuron should have finalise callback which can be called from NEURON for final memory cleanup including global state like registered mechanisms and ions map.

Definition at line 696 of file nrn_setup.cpp.

◆ nrn_cons_newtonspace()

NewtonSpace * coreneuron::nrn_cons_newtonspace	(	int	n,
		int	n_instance
	)

Definition at line 15 of file newton_thread.cpp.

◆ nrn_cons_sparseobj()

template<typename SPFUN >

void* coreneuron::nrn_cons_sparseobj	(	SPFUN	fun,
		int	n,
		Memb_list *	ml,
		_threadargsproto_
	)

sparse matrix dynamic allocation: create_coef_list makes a list for fast setup, does minimum ordering and ensures all elements needed are present.

This could easily be made recursive but it isn't right now.

Definition at line 531 of file sparse_thread.hpp.

◆ nrn_crout_thread()

int coreneuron::nrn_crout_thread	(	NewtonSpace *	ns,
		int	n,
		double **	a,
		int *	perm,
		_threadargsproto_
	)

inline

Performs an LU triangular factorization of a real matrix by the Crout algorithm using partial pivoting.

Rows are not normalized; implicit equilibration is used. ROUNDOFF is the minimal value for a pivot element without its being considered too close to zero (currently set to 1.0E-20).

Returns: 0 if no error; 2 if matrix is singular or ill-conditioned

Parameters

	n	number of rows of the matrix
	a	double precision matrix to be factored
[out]	a	factors required to transform the constant vector in the set of simultaneous equations are stored in the lower triangle; factors for back substitution are stored in the upper triangle.
[out]	perm	permutation vector to store row interchanges

Note: Having a differnt permutation per instance may not be a good idea.

Definition at line 40 of file crout_thread.hpp.

◆ nrn_cur_capacitance()

void coreneuron::nrn_cur_capacitance	(	NrnThread *	_nt,
		Memb_list *	ml,
		int	type
	)

Definition at line 101 of file capac.cpp.

◆ nrn_cur_ion()

void coreneuron::nrn_cur_ion	(	NrnThread *	_nt,
		Memb_list *	ml,
		int	type
	)

◆ nrn_deliver_events()

void coreneuron::nrn_deliver_events ( NrnThread * nt )

Definition at line 32 of file cvodestb.cpp.

◆ nrn_destroy_newtonspace()

void coreneuron::nrn_destroy_newtonspace ( NewtonSpace * ns )

Definition at line 29 of file newton_thread.cpp.

◆ nrn_div_capacity()

void coreneuron::nrn_div_capacity	(	NrnThread *	_nt,
		Memb_list *	ml,
		int	type
	)

Definition at line 136 of file capac.cpp.

◆ nrn_exit()

void coreneuron::nrn_exit ( int err )

Definition at line 30 of file nrnoc_aux.cpp.

◆ nrn_fake_fire()

void coreneuron::nrn_fake_fire	(	int	gid,
		double	spiketime,
		int	fake_out
	)

Definition at line 578 of file netpar.cpp.

◆ nrn_fast_imem_alloc()

void coreneuron::nrn_fast_imem_alloc ( )

Definition at line 32 of file fast_imem.cpp.

◆ nrn_fatal_error() [1/2]

static void coreneuron::nrn_fatal_error ( const char * msg )

static

Definition at line 30 of file nrnmpi.cpp.

◆ nrn_fatal_error() [2/2]

template<typename... Args>

void coreneuron::nrn_fatal_error	(	const char *	msg,
		Args &&...	args
	)

Definition at line 18 of file utils.hpp.

◆ nrn_finitialize()

void coreneuron::nrn_finitialize	(	int	setv,
		double	v
	)

Definition at line 40 of file finitialize.cpp.

◆ nrn_fixed_single_steps_minimal()

void coreneuron::nrn_fixed_single_steps_minimal	(	int	total_sim_steps,
		double	tstop
	)

--> Coreneuron

Definition at line 128 of file fadvance_core.cpp.

◆ nrn_fixed_step_group_minimal()

void coreneuron::nrn_fixed_step_group_minimal ( int total_sim_steps )

Definition at line 148 of file fadvance_core.cpp.

◆ nrn_fixed_step_group_thread()

static void coreneuron::nrn_fixed_step_group_thread	(	NrnThread *	nth,
		int	step_group_max,
		int	step_group_begin,
		int &	step_group_end
	)

static

Definition at line 180 of file fadvance_core.cpp.

◆ nrn_fixed_step_lastpart()

void * coreneuron::nrn_fixed_step_lastpart ( NrnThread * nth )

Definition at line 376 of file fadvance_core.cpp.

◆ nrn_fixed_step_minimal()

void coreneuron::nrn_fixed_step_minimal ( )

Definition at line 91 of file fadvance_core.cpp.

◆ nrn_fixed_step_thread()

static void * coreneuron::nrn_fixed_step_thread ( NrnThread * nth )

static

Definition at line 332 of file fadvance_core.cpp.

◆ nrn_flush_reports()

void coreneuron::nrn_flush_reports ( double t )

Definition at line 36 of file nrnreport.cpp.

◆ nrn_get_mechname()

const char * coreneuron::nrn_get_mechname ( int type )

Definition at line 145 of file mk_mech.cpp.

◆ nrn_get_mechtype()

int coreneuron::nrn_get_mechtype ( const char * name )

Get mechanism type by the mechanism name.

Definition at line 138 of file mk_mech.cpp.

◆ nrn_ghk()

double coreneuron::nrn_ghk	(	double	v,
		double	ci,
		double	co,
		double	z,
		double	celsius
	)

inline

Definition at line 158 of file membfunc.hpp.

◆ nrn_i_layout()

int coreneuron::nrn_i_layout	(	int	icnt,
		int	cnt,
		int	isz,
		int	sz,
		int	layout
	)

This function return the index in a flat array of a matrix coordinate (icnt, isz).

The matrix size is (cnt, sz) Depending of the layout some padding can be calculated

Definition at line 32 of file mem_layout_util.cpp.

◆ nrn_index_permute()

int coreneuron::nrn_index_permute	(	int	ix,
		int	type,
		Memb_list *	ml
	)

Definition at line 363 of file node_permute.cpp.

◆ nrn_index_sort()

static int* coreneuron::nrn_index_sort	(	int *	values,
		int	n
	)

static

Definition at line 417 of file node_permute.cpp.

◆ nrn_index_sort_cmp()

static bool coreneuron::nrn_index_sort_cmp	(	const std::pair< int, int > &	a,
		const std::pair< int, int > &	b
	)

static

Definition at line 405 of file node_permute.cpp.

◆ nrn_init_and_load_data() [1/2]

void coreneuron::nrn_init_and_load_data	(	int	argc,
		char **	argv,
		CheckPoints &	checkPoints,
		bool	is_mapping_needed = `false`,
		bool	run_setup_cleanup = `true`
	)

◆ nrn_init_and_load_data() [2/2]

void coreneuron::nrn_init_and_load_data	(	int	argc,
		char *	argv[],
		CheckPoints &	checkPoints,
		bool	is_mapping_needed,
		bool	run_setup_cleanup
	)

profiler like tau/vtune : do not measure from begining

Setting the timeout

Definition at line 168 of file main1.cpp.

◆ nrn_init_capacitance()

void coreneuron::nrn_init_capacitance	(	NrnThread *	_nt,
		Memb_list *	ml,
		int
	)

Definition at line 83 of file capac.cpp.

◆ nrn_init_ion()

void coreneuron::nrn_init_ion	(	NrnThread *	,
		Memb_list *	,
		int
	)

◆ nrn_inverse_i_layout()

void coreneuron::nrn_inverse_i_layout	(	int	i,
		int &	icnt,
		int	cnt,
		int &	isz,
		int	sz,
		int	layout
	)

Definition at line 669 of file nrn_setup.cpp.

◆ nrn_ion_global_map_copyto_device()

void coreneuron::nrn_ion_global_map_copyto_device ( )

◆ nrn_ion_global_map_delete_from_device()

void coreneuron::nrn_ion_global_map_delete_from_device ( )

◆ nrn_is_ion()

int coreneuron::nrn_is_ion ( int )

◆ nrn_jacob_capacitance()

void coreneuron::nrn_jacob_capacitance	(	NrnThread *	,
		Memb_list *	,
		int
	)

Definition at line 58 of file capac.cpp.

◆ nrn_lhs()

static void coreneuron::nrn_lhs ( NrnThread * _nt )

static

Definition at line 117 of file treeset_core.cpp.

◆ nrn_mallinfo()

double coreneuron::nrn_mallinfo ( void )

Returns current memory usage in KBs.

Parameters

Void

Returns: memory usage in KBs

Definition at line 43 of file memory_utils.cpp.

◆ nrn_mech_depend()

int coreneuron::nrn_mech_depend	(	int	type,
		int *	dependencies
	)

Definition at line 294 of file register_mech.cpp.

◆ nrn_mk_table_check()

void coreneuron::nrn_mk_table_check ( )

Definition at line 133 of file multicore.cpp.

◆ nrn_mkPatternStim()

void coreneuron::nrn_mkPatternStim	(	const char *	fname,
		double	tstop
	)

Definition at line 60 of file patternstim.cpp.

◆ nrn_mul_capacity()

void coreneuron::nrn_mul_capacity	(	NrnThread *	_nt,
		Memb_list *	ml,
		int	type
	)

Definition at line 157 of file capac.cpp.

◆ nrn_multisend_advance()

void coreneuron::nrn_multisend_advance ( )

◆ nrn_multisend_cleanup()

void coreneuron::nrn_multisend_cleanup ( )

◆ nrn_multisend_init()

void coreneuron::nrn_multisend_init ( )

◆ nrn_multisend_receive()

void coreneuron::nrn_multisend_receive ( NrnThread * )

◆ nrn_multisend_send()

void coreneuron::nrn_multisend_send	(	PreSyn *	,
		double	t,
		NrnThread *
	)

◆ nrn_multisend_setup()

void coreneuron::nrn_multisend_setup ( )

◆ nrn_multisend_setup_targets()

void coreneuron::nrn_multisend_setup_targets	(	bool	use_phase2,
		int *&	targets_phase1,
		int *&	targets_phase2
	)

◆ nrn_multithread_job()

template<typename F , typename... Args>

void coreneuron::nrn_multithread_job	(	F &&	job,
		Args &&...	args
	)

Definition at line 161 of file multicore.hpp.

◆ nrn_need_npe()

static bool coreneuron::nrn_need_npe ( )

static

Definition at line 221 of file netpar.cpp.

◆ nrn_nernst()

constexpr double coreneuron::nrn_nernst	(	double	ci,
		double	co,
		double	z,
		double	celsius
	)

constexpr

Definition at line 129 of file membfunc.hpp.

◆ nrn_newton_thread()

template<typename F >

int coreneuron::nrn_newton_thread	(	NewtonSpace *	ns,
		int	n,
		int *	s,
		F	func,
		double *	value,
		_threadargsproto_
	)

inline

Iteratively solves simultaneous nonlinear equations by Newton's method, using a Jacobian matrix computed by finite differences.

Returns: 0 if no error; 2 if matrix is singular or ill-conditioned; 1 if maximum iterations exceeded.

Parameters

	n	number of variables to solve for
	x	pointer to array of the solution vector elements possibly indexed by index
	p	array of parameter values
	func	callable that computes the deviation from zero of each equation in the model
	value	pointer to array to array of the function values
[out]	x	contains the solution value or the most recent iteration's result in the event of an error.

Definition at line 101 of file newton_thread.hpp.

◆ nrn_newtonspace_copyto_device()

void coreneuron::nrn_newtonspace_copyto_device ( NewtonSpace * ns )

Definition at line 1248 of file nrn_acc_manager.cpp.

◆ nrn_newtonspace_delete_from_device()

void coreneuron::nrn_newtonspace_delete_from_device ( NewtonSpace * ns )

Definition at line 1290 of file nrn_acc_manager.cpp.

◆ nrn_original_aos_index()

static int coreneuron::nrn_original_aos_index	(	int	etype,
		int	ix,
		NrnThread &	nt,
		int **	ml_pinv
	)

static

Definition at line 93 of file nrn_checkpoint.cpp.

◆ nrn_outputevent()

void coreneuron::nrn_outputevent	(	unsigned char	,
		double
	)

◆ nrn_p_construct()

void coreneuron::nrn_p_construct ( )

Definition at line 175 of file netcvode.cpp.

◆ nrn_param_layout()

int coreneuron::nrn_param_layout	(	int	i,
		int	mtype,
		Memb_list *	ml
	)

Definition at line 52 of file mem_layout_util.cpp.

◆ nrn_play_init()

void coreneuron::nrn_play_init ( )

Definition at line 72 of file cvodestb.cpp.

◆ nrn_pragma_acc() [1/3]

coreneuron::nrn_pragma_acc ( loop seq )

◆ nrn_pragma_acc() [2/3]

coreneuron::nrn_pragma_acc ( parallel loop presentnt[0:1], stride[0:nstride], firstnode[0:ncell], lastnode[0:ncell], cellsize[0:ncell]) if(nt->compute_gpu) async(nt->stream_id ) -> compute_gpu)) for(int icell

pure virtual

◆ nrn_pragma_acc() [3/3]

coreneuron::nrn_pragma_acc ( routine vector )

Definition at line 30 of file ivocvect.cpp.

◆ nrn_read_filesdat()

void coreneuron::nrn_read_filesdat	(	int &	ngrp,
		int *&	grp,
		const char *	filesdat
	)

Definition at line 171 of file nrn_setup.cpp.

◆ nrn_rhs()

static void coreneuron::nrn_rhs ( NrnThread * _nt )

static

Definition at line 23 of file treeset_core.cpp.

◆ nrn_scopmath_solve_thread()

void coreneuron::nrn_scopmath_solve_thread	(	int	n,
		double **	a,
		double *	b,
		int *	perm,
		double *	p,
		int *	y,
		_threadargsproto_
	)

inline

Performs forward substitution algorithm to transform the constant vector in the linear simultaneous equations to be consistent with the factored matrix.

Then performs back substitution to find the solution to the simultaneous linear equations.

Parameters

	n	number of rows of the matrix
	a	double precision matrix containing the factored matrix of coefficients of the linear equations
	b	vector of function values
	perm	permutation vector to store row interchanges
[out]	p[y[i]]	contains the solution vector

Definition at line 127 of file crout_thread.hpp.

◆ nrn_set_extra_thread0_vdata()

void coreneuron::nrn_set_extra_thread0_vdata ( )

Definition at line 48 of file patternstim.cpp.

◆ nrn_set_timeout()

int coreneuron::nrn_set_timeout ( int timeout )

Definition at line 598 of file netpar.cpp.

◆ nrn_setup()

void coreneuron::nrn_setup	(	const char *	filesdat,
		bool	is_mapping_needed,
		CheckPoints &	checkPoints,
		bool	run_setup_cleanup,
		const char *	datpath,
		const char *	restore_path,
		double *	mindelay
	)

Reserve vector of maps of size ngroup for negative gid-s std::vector< std::map<int, PreSyn*> > neg_gid2out;

gid2out - map of output presyn-s std::map<int, PreSyn*> gid2out;

Allocate memory for fast_imem calculation

Generally, tables depend on a few parameters. And if those parameters change, then the table needs to be recomputed. This is obviously important in NEURON since the user can change those parameters at any time. However, there is no c example for CoreNEURON so can't see what it looks like in that context. Boils down to setting up a function pointer of the function _check_table_thread(), which is only executed by StochKV.c.

Definition at line 401 of file nrn_setup.cpp.

◆ nrn_setup_cleanup()

void coreneuron::nrn_setup_cleanup ( )

Clean up.

Definition at line 391 of file nrn_setup.cpp.

◆ nrn_soa_byte_align()

size_t coreneuron::nrn_soa_byte_align ( size_t size )

return the new offset considering the byte aligment settings

Definition at line 20 of file mem_layout_util.cpp.

◆ nrn_soa_padded_size()

int coreneuron::nrn_soa_padded_size	(	int	cnt,
		int	layout
	)

calculate size after padding for specific memory layout

Definition at line 15 of file mem_layout_util.cpp.

◆ nrn_solve_minimal()

void coreneuron::nrn_solve_minimal ( NrnThread * _nt )

Definition at line 18 of file solve_core.cpp.

◆ nrn_sparseobj_copyto_device()

void coreneuron::nrn_sparseobj_copyto_device ( SparseObj * so )

Definition at line 1309 of file nrn_acc_manager.cpp.

◆ nrn_sparseobj_delete_from_device()

void coreneuron::nrn_sparseobj_delete_from_device ( SparseObj * so )

Definition at line 1392 of file nrn_acc_manager.cpp.

◆ nrn_spike_exchange()

void coreneuron::nrn_spike_exchange ( NrnThread * nt )

◆ nrn_spike_exchange_init()

void coreneuron::nrn_spike_exchange_init ( )

Definition at line 238 of file netpar.cpp.

◆ nrn_thread_table_check()

void coreneuron::nrn_thread_table_check ( )

Definition at line 168 of file multicore.cpp.

◆ nrn_threads_create()

void coreneuron::nrn_threads_create ( int n )

Definition at line 102 of file multicore.cpp.

◆ nrn_threads_free()

void coreneuron::nrn_threads_free ( )

Definition at line 125 of file multicore.cpp.

◆ nrn_VecPlay_copyto_device()

void coreneuron::nrn_VecPlay_copyto_device	(	NrnThread *	nt,
		void **	d_vecplay
	)

◆ nrn_VecPlay_delete_from_device()

void coreneuron::nrn_VecPlay_delete_from_device ( NrnThread * nt )

◆ nrn_writes_conc()

void coreneuron::nrn_writes_conc	(	int	type,
		int
	)

Definition at line 163 of file register_mech.cpp.

◆ nrn_wrote_conc()

constexpr void coreneuron::nrn_wrote_conc	(	int	type,
		double *	p1,
		int	p2,
		int	it,
		double **	gimap,
		double	celsius,
		int	_cntml_padded
	)

constexpr

Definition at line 141 of file membfunc.hpp.

◆ nrn_wtime()

double coreneuron::nrn_wtime ( )

Definition at line 22 of file utils.cpp.

◆ nrncore2nrn_send_init()

void coreneuron::nrncore2nrn_send_init ( )

Definition at line 267 of file fadvance_core.cpp.

◆ nrncore2nrn_send_values()

void coreneuron::nrncore2nrn_send_values ( NrnThread * nth )

Definition at line 284 of file fadvance_core.cpp.

◆ nrnmpi_abort_impl()

void coreneuron::nrnmpi_abort_impl ( int errcode )

Definition at line 102 of file nrnmpi.cpp.

◆ nrnmpi_barrier_impl()

void coreneuron::nrnmpi_barrier_impl ( )

Definition at line 288 of file mpispike.cpp.

◆ nrnmpi_check_threading_support_impl()

void coreneuron::nrnmpi_check_threading_support_impl ( )

Definition at line 86 of file nrnmpi.cpp.

◆ nrnmpi_dbl_allmax_impl()

double coreneuron::nrnmpi_dbl_allmax_impl ( double x )

Definition at line 282 of file mpispike.cpp.

◆ nrnmpi_dbl_allmin_impl()

double coreneuron::nrnmpi_dbl_allmin_impl ( double x )

Definition at line 276 of file mpispike.cpp.

◆ nrnmpi_dbl_allreduce_impl()

double coreneuron::nrnmpi_dbl_allreduce_impl	(	double	x,
		int	type
	)

Definition at line 292 of file mpispike.cpp.

◆ nrnmpi_dbl_allreduce_vec_impl()

void coreneuron::nrnmpi_dbl_allreduce_vec_impl	(	double *	src,
		double *	dest,
		int	cnt,
		int	type
	)

Definition at line 306 of file mpispike.cpp.

◆ nrnmpi_dbl_alltoallv_impl()

void coreneuron::nrnmpi_dbl_alltoallv_impl	(	double *	s,
		int *	scnt,
		int *	sdispl,
		double *	r,
		int *	rcnt,
		int *	rdispl
	)

Definition at line 261 of file mpispike.cpp.

◆ nrnmpi_finalize_impl()

void coreneuron::nrnmpi_finalize_impl ( void )

Definition at line 75 of file nrnmpi.cpp.

◆ nrnmpi_init_impl()

nrnmpi_init_ret_t coreneuron::nrnmpi_init_impl	(	int *	pargc,
		char ***	pargv,
		bool	is_quiet
	)

Definition at line 37 of file nrnmpi.cpp.

◆ nrnmpi_initialized_impl()

bool coreneuron::nrnmpi_initialized_impl ( )

Definition at line 96 of file nrnmpi.cpp.

◆ nrnmpi_int_allgather_impl()

void coreneuron::nrnmpi_int_allgather_impl	(	int *	s,
		int *	r,
		int	n
	)

Definition at line 272 of file mpispike.cpp.

◆ nrnmpi_int_allmax_impl()

int coreneuron::nrnmpi_int_allmax_impl ( int x )

Definition at line 242 of file mpispike.cpp.

◆ nrnmpi_int_alltoall_impl()

void coreneuron::nrnmpi_int_alltoall_impl	(	int *	s,
		int *	r,
		int	n
	)

Definition at line 248 of file mpispike.cpp.

◆ nrnmpi_int_alltoallv_impl()

void coreneuron::nrnmpi_int_alltoallv_impl	(	const int *	s,
		const int *	scnt,
		const int *	sdispl,
		int *	r,
		int *	rcnt,
		int *	rdispl
	)

Definition at line 252 of file mpispike.cpp.

◆ nrnmpi_local_rank_impl()

int coreneuron::nrnmpi_local_rank_impl ( )

Return local mpi rank within a shared memory node.

When performing certain operations, we need to know the rank of mpi process on a given node. This function uses MPI 3 MPI_Comm_split_type function and MPI_COMM_TYPE_SHARED key to find out the local rank.

Definition at line 117 of file nrnmpi.cpp.

◆ nrnmpi_local_size_impl()

int coreneuron::nrnmpi_local_size_impl ( )

Return number of ranks running on single shared memory node.

We use MPI 3 MPI_Comm_split_type function and MPI_COMM_TYPE_SHARED key to determine number of mpi ranks within a shared memory node.

Definition at line 135 of file nrnmpi.cpp.

◆ nrnmpi_long_allreduce_vec_impl()

void coreneuron::nrnmpi_long_allreduce_vec_impl	(	long *	src,
		long *	dest,
		int	cnt,
		int	type
	)

Definition at line 320 of file mpispike.cpp.

◆ nrnmpi_spike_compress()

int coreneuron::nrnmpi_spike_compress	(	int	nspike,
		bool	gid_compress,
		int	xchng_meth
	)

Definition at line 753 of file netpar.cpp.

◆ nrnmpi_spike_exchange_compressed_impl()

int coreneuron::nrnmpi_spike_exchange_compressed_impl	(	int	localgid_size,
		unsigned char *&	spfixin_ovfl,
		int	send_nspike,
		int *	nin,
		int	ovfl_capacity,
		unsigned char *	spikeout_fixed,
		int	ag_send_size,
		unsigned char *	spikein_fixed,
		int &	ovfl
	)

Definition at line 177 of file mpispike.cpp.

◆ nrnmpi_spike_exchange_impl()

int coreneuron::nrnmpi_spike_exchange_impl	(	int *	nin,
		NRNMPI_Spike *	spikeout,
		int	icapacity,
		NRNMPI_Spike **	spikein,
		int &	ovfl,
		int	nout,
		NRNMPI_Spikebuf *	spbufout,
		NRNMPI_Spikebuf *	spbufin
	)

Definition at line 82 of file mpispike.cpp.

◆ nrnmpi_spike_initialize()

void coreneuron::nrnmpi_spike_initialize ( )

Definition at line 37 of file mpispike.cpp.

◆ nrnmpi_v_transfer()

void coreneuron::nrnmpi_v_transfer ( )

Definition at line 35 of file partrans.cpp.

◆ nrnmpi_write_file_impl()

void coreneuron::nrnmpi_write_file_impl	(	const std::string &	filename,
		const char *	buffer,
		size_t	length
	)

Write given buffer to a new file using MPI collective I/O.

For output like spikes, each rank has to write spike timing information to a single file. This routine writes buffers of length len1, len2, len3... at the offsets 0, 0+len1, 0+len1+len2... offsets. This write op is a collective across all ranks of the common MPI communicator used for spike exchange.

Parameters

filename	Name of the file to write
buffer	Buffer to write
length	Length of the buffer to write

Definition at line 160 of file nrnmpi.cpp.

◆ nrnmpi_wtime_impl()

double coreneuron::nrnmpi_wtime_impl ( )

Definition at line 106 of file nrnmpi.cpp.

◆ nrnran123_deletestream()

void coreneuron::nrnran123_deletestream	(	nrnran123_State *	s,
		bool	use_unified_memory
	)

Definition at line 201 of file nrnran123.cpp.

◆ nrnran123_destroy_global_state_on_device()

void coreneuron::nrnran123_destroy_global_state_on_device ( )

Definition at line 154 of file nrnran123.cpp.

◆ nrnran123_get_globalindex()

uint32_t coreneuron::nrnran123_get_globalindex ( )

Definition at line 108 of file nrnran123.cpp.

◆ nrnran123_initialise_global_state_on_device()

void coreneuron::nrnran123_initialise_global_state_on_device ( )

Definition at line 146 of file nrnran123.cpp.

◆ nrnran123_instance_count()

std::size_t coreneuron::nrnran123_instance_count ( )

Definition at line 103 of file nrnran123.cpp.

◆ nrnran123_newstream3()

nrnran123_State* coreneuron::nrnran123_newstream3	(	uint32_t	id1,
		uint32_t	id2,
		uint32_t	id3,
		bool	use_unified_memory
	)

Allocate a new Random123 stream.

Todo:: It would be nicer if the API return type was std::unique_ptr<nrnran123_State, ...not specified...>, so we could use a custom allocator/deleter and avoid the (fragile) need for matching nrnran123_deletestream calls.

Definition at line 168 of file nrnran123.cpp.

◆ nrnran123_set_globalindex()

void coreneuron::nrnran123_set_globalindex ( uint32_t gix )

Definition at line 113 of file nrnran123.cpp.

◆ nrnthread_v_transfer()

void coreneuron::nrnthread_v_transfer ( NrnThread * _nt )

Definition at line 108 of file partrans.cpp.

◆ operator!=()

template<typename T , typename U >

bool coreneuron::operator!=	(	unified_allocator< T > const &	x,
		unified_allocator< U > const &	y
	)

noexcept

Definition at line 67 of file memory.h.

◆ operator<<()

std::ostream & coreneuron::operator<<	(	std::ostream &	os,
		const corenrn_parameters &	corenrn_param
	)

Definition at line 218 of file corenrn_parameters.cpp.

◆ operator==()

template<typename T , typename U >

bool coreneuron::operator==	(	unified_allocator< T > const &	,
		unified_allocator< U > const &
	)

noexcept

Definition at line 62 of file memory.h.

◆ output_presyn_size()

size_t coreneuron::output_presyn_size ( void )

Approximate count of number of bytes for the gid2out map.

Definition at line 1023 of file nrn_setup.cpp.

◆ output_spikes()

void coreneuron::output_spikes	(	const char *	outpath,
		const SpikesInfo &	spikes_info
	)

Definition at line 279 of file output_spikes.cpp.

◆ output_spikes_serial()

static void coreneuron::output_spikes_serial ( const char * outpath )

static

Definition at line 253 of file output_spikes.cpp.

◆ parse_filter_string()

void coreneuron::parse_filter_string	(	const std::string &	filter,
		ReportConfiguration &	config
	)

Definition at line 32 of file report_configuration_parser.cpp.

◆ pattern_stim_info_ref()

void** coreneuron::pattern_stim_info_ref	(	int	icnt,
		int	cnt,
		double *	_p,
		Datum *	_ppvar,
		ThreadDatum *	_thread,
		NrnThread *	_nt,
		Memb_list *	ml,
		double	v
	)

◆ pattern_stim_setup_helper()

void coreneuron::pattern_stim_setup_helper	(	int	size,
		double *	tvec,
		int *	gidvec,
		int	icnt,
		int	cnt,
		double *	_p,
		Datum *	_ppvar,
		ThreadDatum *	_thread,
		NrnThread *	_nt,
		Memb_list *	ml,
		double	v
	)

◆ permute() [1/2]

static int coreneuron::permute	(	int	i,
		NrnThread &	nt
	)

static

Definition at line 28 of file prcellstate.cpp.

◆ permute() [2/2]

template<typename T >

void coreneuron::permute	(	T *	data,
		int	cnt,
		int	sz,
		int	layout,
		int *	p
	)

Definition at line 98 of file node_permute.cpp.

◆ permute_data()

void coreneuron::permute_data	(	double *	vec,
		int	n,
		int *	p
	)

Definition at line 349 of file node_permute.cpp.

◆ permute_ml()

void coreneuron::permute_ml	(	Memb_list *	ml,
		int	type,
		NrnThread &	nt
	)

Definition at line 353 of file node_permute.cpp.

◆ permute_nodeindices()

void coreneuron::permute_nodeindices	(	Memb_list *	ml,
		int *	p
	)

Definition at line 431 of file node_permute.cpp.

◆ permute_ptr()

void coreneuron::permute_ptr	(	int *	vec,
		int	n,
		int *	p
	)

Definition at line 345 of file node_permute.cpp.

◆ pnt_name()

char * coreneuron::pnt_name ( Point_process * pnt )

Definition at line 26 of file nrnoc_aux.cpp.

◆ point_reg_helper()

int coreneuron::point_reg_helper ( const Symbol * s2 )

Definition at line 325 of file register_mech.cpp.

◆ point_register_mech()

int coreneuron::point_register_mech	(	const char **	m,
		mod_alloc_t	alloc,
		mod_f_t	cur,
		mod_f_t	jacob,
		mod_f_t	stat,
		mod_f_t	initialize,
		mod_f_t	private_constructor,
		mod_f_t	private_destructor,
		int	nrnpointerindex,
		mod_f_t	constructor,
		mod_f_t	destructor,
		int	vectorized
	)

Definition at line 339 of file register_mech.cpp.

◆ pr_memb()

static void coreneuron::pr_memb	(	int	type,
		Memb_list *	ml,
		int *	cellnodes,
		NrnThread &	nt,
		FILE *	f
	)

static

Definition at line 52 of file prcellstate.cpp.

◆ pr_netcon()

static void coreneuron::pr_netcon	(	NrnThread &	nt,
		FILE *	f
	)

static

Fill the NetCon <-> DiscreteEvent map with PreSyn-s

Fill the NetCon <-> DiscreteEvent map with InputPreSyn-s Traverse gid <-> InputPreSyn map and loop over NetCon-s of the correspondent InputPreSyn. If NetCon is in the nc2src map, remember its ips and the gid

input presyn

src gid of the input presyn

Definition at line 86 of file prcellstate.cpp.

◆ pr_realcell()

static void coreneuron::pr_realcell	(	PreSyn &	ps,
		NrnThread &	nt,
		FILE *	f
	)

static

Definition at line 199 of file prcellstate.cpp.

◆ prcellstate()

int coreneuron::prcellstate	(	int	gid,
		const char *	suffix
	)

Definition at line 271 of file prcellstate.cpp.

◆ prgroupsize()

void coreneuron::prgroupsize ( VVVTN & groups )

Definition at line 429 of file cellorder2.cpp.

◆ print_quality1()

static void coreneuron::print_quality1	(	int	iwarp,
		InterleaveInfo &	ii,
		int	ncell,
		int *	p
	)

static

Definition at line 175 of file cellorder.cpp.

◆ print_quality2()

static void coreneuron::print_quality2	(	int	iwarp,
		InterleaveInfo &	ii,
		int *	p
	)

static

Definition at line 110 of file cellorder.cpp.

◆ pscheck()

static bool coreneuron::pscheck	(	double	var,
		double	thresh,
		int *	flag
	)

static

Definition at line 509 of file netcvode.cpp.

◆ ptr_tnode_earlier()

static bool coreneuron::ptr_tnode_earlier	(	TNode *	a,
		TNode *	b
	)

static

Definition at line 44 of file cellorder1.cpp.

◆ quality()

static void coreneuron::quality	(	VecTNode &	nodevec,
		size_t	max = `32`
	)

static

Definition at line 110 of file cellorder1.cpp.

◆ question2()

static void coreneuron::question2 ( VVTN & levels )

static

Definition at line 325 of file cellorder2.cpp.

◆ read_phase1()

void coreneuron::read_phase1	(	NrnThread &	nt,
		UserParams &	userParams
	)

Definition at line 917 of file nrn_setup.cpp.

◆ read_phase2()

void coreneuron::read_phase2	(	NrnThread &	nt,
		UserParams &	userParams
	)

Definition at line 924 of file nrn_setup.cpp.

◆ read_phase3()

void coreneuron::read_phase3	(	NrnThread &	nt,
		UserParams &	userParams
	)

read mapping information for neurons

restore checkpoint state (before restoring queue items

mapping information for all neurons in single NrnThread

number of cells in mapping file should equal to cells in NrnThread

for every neuron

Definition at line 935 of file nrn_setup.cpp.

◆ read_phasegap()

void coreneuron::read_phasegap	(	NrnThread &	nt,
		UserParams &	userParams
	)

Definition at line 588 of file nrn_setup.cpp.

◆ read_raster_file()

size_t coreneuron::read_raster_file	(	const char *	fname,
		double **	tvec,
		int **	gidvec,
		double	tstop
	)

static

Definition at line 100 of file patternstim.cpp.

◆ realloc_net_receive_buffer()

void coreneuron::realloc_net_receive_buffer	(	NrnThread *	nt,
		Memb_list *	ml
	)

Definition at line 811 of file nrn_acc_manager.cpp.

◆ register_all_variables_offsets()

void coreneuron::register_all_variables_offsets	(	int	mech_id,
		SerializedNames	variable_names
	)

Definition at line 58 of file mech_mapping.cpp.

◆ register_constructor()

void coreneuron::register_constructor ( mod_f_t constructor )

Definition at line 317 of file register_mech.cpp.

◆ register_destructor()

void coreneuron::register_destructor ( mod_f_t d )

Definition at line 321 of file register_mech.cpp.

◆ register_mech()

int coreneuron::register_mech	(	const char **	m,
		mod_alloc_t	alloc,
		mod_f_t	cur,
		mod_f_t	jacob,
		mod_f_t	stat,
		mod_f_t	initialize,
		mod_f_t	private_constructor,
		mod_f_t	private_destructor,
		int	nrnpointerindex,
		int	vectorized
	)

Definition at line 112 of file register_mech.cpp.

◆ register_target_type()

void coreneuron::register_target_type	(	ReportConfiguration &	report,
		ReportType	report_type
	)

Definition at line 57 of file report_configuration_parser.cpp.

◆ rendezvous_rank_get()

static void coreneuron::rendezvous_rank_get	(	HAVEWANT_t *	data,
		int	size,
		HAVEWANT_t *&	sdata,
		int *&	scnt,
		int *&	sdispl,
		HAVEWANT_t *&	rdata,
		int *&	rcnt,
		int *&	rdispl,
		int(*)(HAVEWANT_t)	rendezvous_rank
	)

static

Definition at line 83 of file have2want.h.

◆ report_cell_stats()

void coreneuron::report_cell_stats ( )

Reports global cell statistics of the simulation.

This routine prints the global number of cells, synapses of the simulation

Parameters

void

Returns: void

Definition at line 30 of file nrn_stats.cpp.

◆ report_mem_usage()

void coreneuron::report_mem_usage	(	const char *	message,
		bool	all_ranks = `false`
	)

Reports current memory usage of the simulator to stdout.

Current implementation is based on mallinfo. This routine prints min, max and avg memory usage across mpi comm world

Parameters

message	string indicating current stage of the simulation
all_ranks	indicate whether to print info from all ranks

Returns: Void

Definition at line 79 of file memory_utils.cpp.

◆ second_order_cur()

void coreneuron::second_order_cur	(	NrnThread *	_nt,
		int	secondorder
	)

◆ set_an_offset()

static void coreneuron::set_an_offset	(	int	mech_id,
		const char *	variable_name,
		int	offset
	)

static

Definition at line 35 of file mech_mapping.cpp.

◆ set_cellindex()

static void coreneuron::set_cellindex	(	int	ncell,
		VecTNode &	nodevec
	)

static

Set the cellindex to distinguish the different cells.

Definition at line 242 of file cellorder1.cpp.

◆ set_globals()

void coreneuron::set_globals	(	const char *	path,
		bool	cli_global_seed,
		int	cli_global_seed_value
	)

Definition at line 43 of file global_vars.cpp.

◆ set_groupindex()

static void coreneuron::set_groupindex ( VecTNode & nodevec )

static

Initialization of the groupindex (groups)

The cells are groupped at a later stage based on a load balancing algorithm. This is just an initialization function.

Definition at line 261 of file cellorder1.cpp.

◆ set_mindelay()

double coreneuron::set_mindelay ( double maxdelay )

we have removed nt_ from PreSyn. Build local map of PreSyn and NrnThread which will be used to find out if src in same thread as NetCon

Definition at line 643 of file netpar.cpp.

◆ set_nodeindex()

static void coreneuron::set_nodeindex ( VecTNode & nodevec )

static

Definition at line 454 of file cellorder2.cpp.

◆ set_pnt_receive()

void coreneuron::set_pnt_receive	(	int	type,
		pnt_receive_t	pnt_receive,
		pnt_receive_t	pnt_receive_init,
		short	size
	)

Definition at line 74 of file register_mech.cpp.

◆ set_report_buffer_size()

void coreneuron::set_report_buffer_size ( int n )

Definition at line 69 of file nrnreport.cpp.

◆ set_treenode_order()

static void coreneuron::set_treenode_order ( VVTN & levels )

static

Definition at line 73 of file cellorder2.cpp.

◆ setup_fornetcon_info()

void coreneuron::setup_fornetcon_info ( NrnThread & nt )

If FOR_NETCON in use, setup NrnThread fornetcon related info.

Definition at line 77 of file setup_fornetcon.cpp.

◆ setup_nrnthreads_on_device()

void coreneuron::setup_nrnthreads_on_device	(	NrnThread *	threads,
		int	nthreads
	)

Definition at line 466 of file nrn_acc_manager.cpp.

◆ setup_report_engine()

void coreneuron::setup_report_engine	(	double	dt_report,
		double	mindelay
	)

in the current implementation, we call flush during every spike exchange interval.

Hence there should be sufficient buffer to hold all reports for the duration of mindelay interval. In the below call we specify the number of timesteps that we have to buffer. TODO: revisit this because spike exchange can happen few steps before/after mindelay interval and hence adding two extra timesteps to buffer.

Definition at line 53 of file nrnreport.cpp.

◆ setup_ThreadData()

void coreneuron::setup_ThreadData ( NrnThread & nt )

Definition at line 570 of file nrn_setup.cpp.

◆ setup_tree_matrix_minimal()

void * coreneuron::setup_tree_matrix_minimal ( NrnThread * _nt )

Definition at line 178 of file treeset_core.cpp.

◆ setup_type2semantics()

static void coreneuron::setup_type2semantics ( )

static

Definition at line 136 of file nrn2core_data_init.cpp.

◆ soa2aos_inverse_permute_copy()

static void coreneuron::soa2aos_inverse_permute_copy	(	size_t	n,
		int	sz,
		int	stride,
		double *	src,
		double **	dest,
		int *	permute
	)

static

SoA permuted mechanism data copied to unpermuted AoS data.

dest is an array of n pointers to the beginning of each sz length array. src is a contiguous array of sz segments of size stride. The stride may be slightly greater than n for purposes of alignment. Each of the sz segments of src are permuted.

Definition at line 66 of file core2nrn_data_return.cpp.

◆ soa2aos_unpermuted_copy()

static void coreneuron::soa2aos_unpermuted_copy	(	size_t	n,
		int	sz,
		int	stride,
		double *	src,
		double **	dest
	)

static

SoA unpermuted mechanism data copied to unpermuted AoS data.

dest is an array of n pointers to the beginning of each sz length array. src is a contiguous array of sz segments of size stride. The stride may be slightly greater than n for purposes of alignment. Each of the sz segments of src have the same order as the n pointers of dest.

Definition at line 89 of file core2nrn_data_return.cpp.

◆ solve_interleaved()

void coreneuron::solve_interleaved ( int ith )

Solve the Hines matrices based on the interleave_permute_type (1 or 2).

For interleave_permute_type == 1 : Naive interleaving -> Each execution thread deals with one Hines matrix (cell) For interleave_permute_type == 2 : Advanced interleaving -> Each Hines matrix is solved by multiple execution threads (with coalesced memory access as well)

◆ solve_interleaved2_launcher()

void coreneuron::solve_interleaved2_launcher	(	NrnThread *	nt,
		InterleaveInfo *	info,
		int	ncore,
		void *	stream
	)

CUDA branch of the solve_interleaved with interleave_permute_type == 2.

This branch is activated in runtime with the –cuda-interface CLI flag

◆ sortlevel()

static void coreneuron::sortlevel ( VTN & level )

static

Definition at line 64 of file cellorder2.cpp.

◆ sortlevel_cmp()

static bool coreneuron::sortlevel_cmp	(	TNode *	a,
		TNode *	b
	)

static

Definition at line 37 of file cellorder2.cpp.

◆ sparse_thread()

template<typename F >

int coreneuron::sparse_thread	(	SparseObj *	so,
		int	n,
		int *	s,
		int *	d,
		double *	t,
		double	dt,
		F	fun,
		int	linflag,
		_threadargsproto_
	)

This is an experimental numerical method for SCoP-3 which integrates kinetic rate equations.

It is intended to be used only by models generated by MODL, and its identity is meant to be concealed from the user.

Parameters

n	number of state variables
s	array of pointers to the state variables
d	array of pointers to the derivatives of states
t	pointer to the independent variable
dt	the time step
fun	callable corresponding to the kinetic block equations
prhs	pointer to right hand side vector (answer on return) does not have to be allocated by caller. (this is no longer quite right)
linflag	solve as linear equations, when nonlinear, all states are forced >= 0

Definition at line 561 of file sparse_thread.hpp.

◆ spdelete()

void coreneuron::spdelete	(	SPBLK *	n,
		SPTREE *	q
	)

Definition at line 532 of file tqueue.cpp.

◆ spdeq()

SPBLK * coreneuron::spdeq ( SPBLK ** np )

Definition at line 325 of file tqueue.cpp.

◆ spenq()

SPBLK * coreneuron::spenq	(	SPBLK *	n,
		SPTREE *	q
	)

Definition at line 197 of file tqueue.cpp.

◆ sphead()

SPBLK * coreneuron::sphead ( SPTREE * q )

Definition at line 503 of file tqueue.cpp.

◆ spikevec_lock()

void coreneuron::spikevec_lock ( )

Definition at line 59 of file output_spikes.cpp.

◆ spikevec_unlock()

void coreneuron::spikevec_unlock ( )

Definition at line 63 of file output_spikes.cpp.

◆ spinit()

void coreneuron::spinit ( SPTREE * q )

Definition at line 180 of file tqueue.cpp.

◆ splay()

void coreneuron::splay	(	SPBLK *	n,
		SPTREE *	q
	)

Definition at line 402 of file tqueue.cpp.

◆ srccnt2destcnt()

static int* coreneuron::srccnt2destcnt ( int * srccnt )

static

Definition at line 68 of file have2want.h.

◆ state_discontinuity()

void coreneuron::state_discontinuity	(	int	,
		double *	pd,
		double	d
	)

Definition at line 371 of file register_mech.cpp.

◆ stdindex2ptr()

double * coreneuron::stdindex2ptr	(	int	mtype,
		int	index,
		NrnThread &	nt
	)

Definition at line 636 of file nrn_setup.cpp.

◆ stride_length()

static size_t coreneuron::stride_length	(	size_t	begin,
		size_t	end,
		VecTNode &	nodevec
	)

static

Definition at line 563 of file cellorder1.cpp.

◆ tnode_earlier()

static bool coreneuron::tnode_earlier	(	TNode *	a,
		TNode *	b
	)

static

Function to order trees by size, hash and nodeindex.

Definition at line 30 of file cellorder1.cpp.

◆ trajectory_return()

static void coreneuron::trajectory_return ( )

static

Definition at line 413 of file main1.cpp.

◆ tree_analysis()

static void coreneuron::tree_analysis	(	int *	parent,
		int	nnode,
		int	ncell,
		VecTNode &	nodevec
	)

static

Perform tree preparation for interleaving strategies.

Parameters

parent	vector of parent indices
nnode	number of compartments in the cells
ncell	number of cells

Definition at line 428 of file cellorder1.cpp.

◆ triang()

static void coreneuron::triang ( NrnThread * _nt )

static

Todo:: OpenACC GPU offload is sequential/slow.

Because –cell-permute=0 and –gpu is forbidden anyway, no OpenMP target offload equivalent is implemented.

Definition at line 32 of file solve_core.cpp.

◆ triang_interleaved()

static void coreneuron::triang_interleaved	(	NrnThread *	nt,
		int	icell,
		int	icellsize,
		int	nstride,
		int *	stride,
		int *	lastnode
	)

static

Definition at line 438 of file cellorder.cpp.

◆ type_of_ntdata()

int coreneuron::type_of_ntdata	(	NrnThread &	nt,
		int	i,
		bool	reset
	)

Definition at line 186 of file node_permute.cpp.

◆ update()

void coreneuron::update ( NrnThread * _nt )

Definition at line 201 of file fadvance_core.cpp.

◆ update_net_receive_buffer()

void coreneuron::update_net_receive_buffer ( NrnThread * nt )

Definition at line 928 of file nrn_acc_manager.cpp.

◆ update_net_send_buffer_on_host()

void coreneuron::update_net_send_buffer_on_host	(	NrnThread *	nt,
		NetSendBuffer_t *	nsb
	)

Definition at line 974 of file nrn_acc_manager.cpp.

◆ update_nrnthreads_on_host()

void coreneuron::update_nrnthreads_on_host	(	NrnThread *	threads,
		int	nthreads
	)

Definition at line 1012 of file nrn_acc_manager.cpp.

◆ update_pdata_values()

static void coreneuron::update_pdata_values	(	Memb_list *	ml,
		int	type,
		NrnThread &	nt
	)

static

Definition at line 212 of file node_permute.cpp.

◆ update_weights_from_gpu()

void coreneuron::update_weights_from_gpu	(	NrnThread *	threads,
		int	nthreads
	)

Copy weights from GPU to CPU.

User may record NetCon weights at the end of simulation. For this purpose update weights of all NrnThread objects from GPU to CPU.

Definition at line 1112 of file nrn_acc_manager.cpp.

◆ validation()

void coreneuron::validation ( std::vector< std::pair< double, int >> & res )

Definition at line 305 of file output_spikes.cpp.

◆ vec_play_activate()

void coreneuron::vec_play_activate ( )

static

Definition at line 98 of file nrn2core_data_init.cpp.

◆ vector_capacity()

int coreneuron::vector_capacity ( IvocVect * v )

Definition at line 16 of file ivocvect.cpp.

◆ vector_new()

IvocVect * coreneuron::vector_new ( int n )

Definition at line 13 of file ivocvect.cpp.

◆ vector_new1()

IvocVect * coreneuron::vector_new1 ( int n )

Definition at line 26 of file ivocvect.cpp.

◆ vector_vec()

double * coreneuron::vector_vec ( IvocVect * v )

Definition at line 19 of file ivocvect.cpp.

◆ wait_before_spike_exchange()

void coreneuron::wait_before_spike_exchange ( )

Definition at line 78 of file mpispike.cpp.

◆ warp_balance() [1/2]

static void coreneuron::warp_balance	(	int	ith,
		InterleaveInfo &	ii
	)

static

Definition at line 243 of file cellorder.cpp.

◆ warp_balance() [2/2]

size_t coreneuron::warp_balance	(	size_t	ncell,
		VecTNode &	nodevec
	)

Use of the LPT (Least Processing Time) algorithm to create balanced groups of cells.

Competing objectives are to keep identical cells together and also balance warps.

Parameters

ncell	number of cells
nodevec	vector of compartments from all cells

Returns: number of warps

Definition at line 43 of file balance.cpp.

◆ warpcmp()

bool coreneuron::warpcmp	(	const TNode *	a,
		const TNode *	b
	)

Definition at line 27 of file balance.cpp.

◆ watch_activate_clear()

void coreneuron::watch_activate_clear ( )

static

Definition at line 318 of file nrn2core_data_init.cpp.

◆ watch_datum_indices()

void coreneuron::watch_datum_indices	(	int	type,
		int &	first,
		int &	last
	)

return first and last datum indices of WATCH statements

Definition at line 301 of file nrn2core_data_init.cpp.

◆ write_mech_report()

void coreneuron::write_mech_report ( )

display global mechanism count

write mechanism counts to stdout

mechanim count across all gids, local to rank

each gid record goes on separate row, only check non-empty threads

print global stats to stdout

Definition at line 19 of file mech_report.cpp.

Variable Documentation

◆ _nrn_skip_initmodel

bool coreneuron::_nrn_skip_initmodel

Definition at line 19 of file finitialize.cpp.

◆ active_

bool coreneuron::active_ = false

static

Definition at line 83 of file netpar.cpp.

◆ bbcore_write_version

const char * coreneuron::bbcore_write_version = "1.6"

Definition at line 24 of file nrnoc_aux.cpp.

◆ byteovfl

int* coreneuron::byteovfl {nullptr}

static

Definition at line 27 of file mpispike.cpp.

◆ cellorder_nwarp

int coreneuron::cellorder_nwarp = 0

Definition at line 24 of file balance.cpp.

◆ cellsize

int* coreneuron::cellsize = ii.cellsize

Definition at line 645 of file cellorder.cpp.

◆ celsius

double coreneuron::celsius

Definition at line 22 of file register_mech.cpp.

◆ core2nrn_clear_queues_

void(* coreneuron::core2nrn_clear_queues_) (double t)

Callback to clear NEURON thread queues.

In particular need to initialize bin queues to the current time before transferring events.

Definition at line 188 of file core2nrn_data_return.cpp.

◆ core2nrn_NetCon_event_

void(* coreneuron::core2nrn_NetCon_event_) (int tid, double td, size_t nc_index)

Callbacks into NEURON for queue event types.

Definition at line 352 of file core2nrn_data_return.cpp.

◆ core2nrn_PreSyn_flag_

void(* coreneuron::core2nrn_PreSyn_flag_) (int tid, std::set< int > presyns_flag_true)

Definition at line 376 of file core2nrn_data_return.cpp.

◆ core2nrn_SelfEvent_event_

void(* coreneuron::core2nrn_SelfEvent_event_) (int tid, double td, int tar_type, int tar_index, double flag, size_t nc_index, int is_movable)

Definition at line 355 of file core2nrn_data_return.cpp.

◆ core2nrn_SelfEvent_event_noweight_

void(* coreneuron::core2nrn_SelfEvent_event_noweight_) (int tid, double td, int tar_type, int tar_index, double flag, int is_movable)

Definition at line 363 of file core2nrn_data_return.cpp.

◆ core2nrn_vecplay_

void(* coreneuron::core2nrn_vecplay_) (int tid, int i_nrn, int last, int discon, int ubound)

VecPlay indices back to NEURON.

Definition at line 207 of file core2nrn_data_return.cpp.

◆ core2nrn_vecplay_events_

void(* coreneuron::core2nrn_vecplay_events_) ()

Definition at line 208 of file core2nrn_data_return.cpp.

◆ core2nrn_watch_activate_

void(* coreneuron::core2nrn_watch_activate_) (int tid, int type, int watch_begin, Core2NrnWatchInfo &)

Definition at line 200 of file core2nrn_data_return.cpp.

◆ core2nrn_watch_clear_

void(* coreneuron::core2nrn_watch_clear_) ()

Definition at line 199 of file core2nrn_data_return.cpp.

◆ corenrn

CoreNeuron coreneuron::corenrn

Definition at line 53 of file multicore.cpp.

◆ corenrn_param

corenrn_parameters coreneuron::corenrn_param

Printing method.

Definition at line 268 of file corenrn_parameters.cpp.

◆ cvode_active_

bool coreneuron::cvode_active_

Definition at line 36 of file netcvode.cpp.

◆ diam_changed

int coreneuron::diam_changed

Definition at line 21 of file nrnoc_aux.cpp.

◆ displs

int* coreneuron::displs {nullptr}

static

Definition at line 26 of file mpispike.cpp.

◆ dt

double coreneuron::dt

Definition at line 22 of file register_mech.cpp.

◆ firstnode

int * coreneuron::firstnode

Initial value:

{
    nrn_pragma_acc(loop seq)
    for (int i = root; i < lastroot; i += 1) {
        GPU_RHS(i) /= GPU_D(i);  
    }
 
    int i = firstnode + icore

Definition at line 532 of file cellorder.cpp.

◆ gid2in

std::map< int, InputPreSyn * > coreneuron::gid2in

Definition at line 158 of file nrn_setup.cpp.

◆ gid2out

std::map<int, PreSyn*> coreneuron::gid2out

Maps for ouput and input presyns.

Definition at line 157 of file nrn_setup.cpp.

◆ groupsize

size_t coreneuron::groupsize = 32

static

Definition at line 25 of file cellorder1.cpp.

◆ has_subtrees_to_compute

coreneuron::has_subtrees_to_compute = true

Definition at line 489 of file cellorder.cpp.

◆ hoc_errno_count

int coreneuron::hoc_errno_count

Definition at line 23 of file nrnoc_aux.cpp.

◆ i

coreneuron::i = lastnode - istride + icore

Definition at line 485 of file cellorder.cpp.

◆ icore

int int int coreneuron::icore

Definition at line 482 of file cellorder.cpp.

◆ ii

InterleaveInfo & coreneuron::ii = i

Definition at line 486 of file cellorder.cpp.

◆ interleave_info

InterleaveInfo * coreneuron::interleave_info

Definition at line 29 of file cellorder.cpp.

◆ interleave_permute_type

int coreneuron::interleave_permute_type

Definition at line 28 of file cellorder.cpp.

◆ inv_permute_

std::vector<int>* coreneuron::inv_permute_

static

Definition at line 26 of file prcellstate.cpp.

◆ ion_global_map_member_size

const int coreneuron::ion_global_map_member_size

◆ istride

coreneuron::istride = stride[icycle]

Definition at line 484 of file cellorder.cpp.

◆ last_maxstep_arg_

double coreneuron::last_maxstep_arg_

static

Definition at line 86 of file netpar.cpp.

◆ lastnode

int * coreneuron::lastnode

Initial value:

{

int icycle = ncycle - 1

Definition at line 482 of file cellorder.cpp.

◆ lastroot

int int coreneuron::lastroot

Definition at line 528 of file cellorder.cpp.

◆ map_nc2src

std::map<NetCon*, DiscreteEvent*> coreneuron::map_nc2src

static

Definition at line 25 of file prcellstate.cpp.

◆ max_line_length

const int coreneuron::max_line_length = 1024

Encapsulate low-level reading of coreneuron input data files.

Error handling is simple: abort()!

Reader will abort() if native integer size is not 4 bytes.

All automatic allocations performed by read_int_array() and read_dbl_array() methods use new [].

Definition at line 30 of file nrn_filehandler.hpp.

◆ mech2type

std::map<std::string, int> coreneuron::mech2type

Definition at line 33 of file mk_mech.cpp.

◆ mechanism [1/2]

const char* coreneuron::mechanism[] = {"0", "capacitance", "cm", 0, "i_cap", 0, 0}

static

Definition at line 22 of file capac.cpp.

◆ mechanism [2/2]

void(* coreneuron::mechanism[])(void)

static

Initial value:

= {
                                    capacitance_reg,
                                    _passive_reg,
 
                                    nullptr}

Definition at line 18 of file cabvars.h.

◆ mechNamesMapping

MechNamesMapping coreneuron::mechNamesMapping

static

Definition at line 33 of file mech_mapping.cpp.

◆ mindelay_

double coreneuron::mindelay_

static

Definition at line 85 of file netpar.cpp.

◆ mut [1/2]

OMP_Mutex coreneuron::mut

static

Definition at line 48 of file output_spikes.cpp.

◆ mut [2/2]

OMP_Mutex coreneuron::mut

static

Definition at line 152 of file nrn_setup.cpp.

◆ n2v

N2V* coreneuron::n2v

static

Definition at line 29 of file global_vars.cpp.

◆ n_multisend_interval

int coreneuron::n_multisend_interval = 2

Definition at line 55 of file multisend.cpp.

◆ ncell

icycle< ncycle; ++icycle) { int istride = stride[icycle]; nrn_pragma_acc(loop vector) nrn_pragma_omp(loop bind(parallel)) for (int icore = 0; icore < warpsize; ++icore) { int i = ii + icore; if (icore < istride) { int ip = GPU_PARENT(i); GPU_RHS(i) -= GPU_B(i) * GPU_RHS(ip); GPU_RHS(i) /= GPU_D(i); } i += istride; } ii += istride; }}void solve_interleaved2(int ith) { NrnThread* nt = nrn_threads + ith; InterleaveInfo& ii = interleave_info[ith]; int nwarp = ii.nwarp; if (nwarp == 0) return; int ncore = nwarp * warpsize; int* ncycles = ii.cellsize; int* stridedispl = ii.stridedispl; int* strides = ii.stride; int* rootbegin = ii.firstnode; int* nodebegin = ii.lastnode; nrn_pragma_acc(parallel loop gang present(nt [0:1], strides [0:nstride], ncycles [0:nwarp], stridedispl [0:nwarp + 1], rootbegin [0:nwarp + 1], nodebegin [0:nwarp + 1]) if (nt->compute_gpu) async(nt->stream_id)) nrn_pragma_omp(target teams loop if(nt->compute_gpu)) for (int icore = 0; icore < ncore; icore += warpsize) { int iwarp = icore / warpsize; int ic = icore & (warpsize - 1); int ncycle = ncycles[iwarp]; int* stride = strides + stridedispl[iwarp]; int root = rootbegin[iwarp]; int lastroot = rootbegin[iwarp + 1]; int firstnode = nodebegin[iwarp]; int lastnode = nodebegin[iwarp + 1]; triang_interleaved2(nt, ic, ncycle, stride, lastnode); bksub_interleaved2(nt, root + ic, lastroot, ic, ncycle, stride, firstnode); } nrn_pragma_acc(wait(nt->stream_id))}void solve_interleaved1(int ith) { NrnThread* nt = nrn_threads + ith; int ncell = nt-> coreneuron::ncell

Definition at line 636 of file cellorder.cpp.

◆ ncycle

int int int int coreneuron::ncycle

Definition at line 482 of file cellorder.cpp.

◆ neg_gid2out

std::vector<std::map<int, PreSyn*> > coreneuron::neg_gid2out

Vector of maps for negative presyns.

Definition at line 155 of file nrn_setup.cpp.

◆ net_cvode_instance

NetCvode * coreneuron::net_cvode_instance

Definition at line 35 of file netcvode.cpp.

◆ netcon_in_presyn_order_

std::vector<NetCon*> coreneuron::netcon_in_presyn_order_

InputPreSyn.nc_index_ to + InputPreSyn.nc_cnt_ give the NetCon*.

Definition at line 161 of file nrn_setup.cpp.

◆ np

int coreneuron::np

static

Definition at line 25 of file mpispike.cpp.

◆ npe_

std::vector<NetParEvent> coreneuron::npe_

static

Definition at line 87 of file netpar.cpp.

◆ nrn2core_mkmech_info_

void(* coreneuron::nrn2core_mkmech_info_) (std::ostream &)

Definition at line 36 of file mk_mech.cpp.

◆ nrn2core_patternstim_

void(* coreneuron::nrn2core_patternstim_) (void **info)

Definition at line 414 of file nrn2core_data_init.cpp.

◆ nrn2core_transfer_PreSyn_flag_

void(* coreneuron::nrn2core_transfer_PreSyn_flag_) (int tid, std::set< int > &presyns_flag_true)

Definition at line 378 of file core2nrn_data_return.cpp.

◆ nrn2core_transfer_tqueue_

NrnCoreTransferEvents*(* coreneuron::nrn2core_transfer_tqueue_) (int tid)

Pointer to function in NEURON that iterates over its tqeueue.

Definition at line 131 of file nrn2core_data_init.cpp.

◆ nrn2core_transfer_watch_

void(* coreneuron::nrn2core_transfer_watch_) (void(*cb)(int, int, int, int, int))

Pointer to function in NEURON that iterates over activated WATCH statements, sending each item to ...

Definition at line 35 of file nrn2core_data_init.cpp.

◆ nrn_extra_thread0_vdata

int coreneuron::nrn_extra_thread0_vdata

Definition at line 46 of file patternstim.cpp.

◆ nrn_fornetcon_cnt_

int coreneuron::nrn_fornetcon_cnt_

Definition at line 50 of file register_mech.cpp.

◆ nrn_fornetcon_index_

int * coreneuron::nrn_fornetcon_index_

Definition at line 52 of file register_mech.cpp.

◆ nrn_fornetcon_type_

int * coreneuron::nrn_fornetcon_type_

Definition at line 51 of file register_mech.cpp.

◆ nrn_have_gaps

bool coreneuron::nrn_have_gaps

variables defined in coreneuron library

Definition at line 21 of file partrans.cpp.

◆ nrn_ion_global_map

double** coreneuron::nrn_ion_global_map

◆ nrn_ion_global_map_size

int coreneuron::nrn_ion_global_map_size

◆ nrn_mk_transfer_thread_data_

void(* coreneuron::nrn_mk_transfer_thread_data_) ()

Definition at line 57 of file multicore.cpp.

◆ nrn_nobanner_

int coreneuron::nrn_nobanner_ {0}

Declaring global corenrn_parameters object for this instance of CoreNeuron.

Definition at line 269 of file corenrn_parameters.cpp.

◆ nrn_nthread

int coreneuron::nrn_nthread = 0

Definition at line 55 of file multicore.cpp.

◆ nrn_threads

NrnThread * coreneuron::nrn_threads = nullptr

Definition at line 56 of file multicore.cpp.

◆ nrn_use_bin_queue_

bool coreneuron::nrn_use_bin_queue_ = 0

Flag to use the bin queue.

Definition at line 39 of file netcvode.cpp.

◆ nrn_use_fast_imem

bool coreneuron::nrn_use_fast_imem

Definition at line 19 of file fast_imem.cpp.

◆ nrn_use_localgid_

bool coreneuron::nrn_use_localgid_

◆ nrnmpi_abort

mpi_function< cnrn_make_integral_constant_t(nrnmpi_abort_impl)> coreneuron::nrnmpi_abort {"nrnmpi_abort_impl"}

Definition at line 52 of file nrnmpidec.cpp.

◆ nrnmpi_barrier

mpi_function< cnrn_make_integral_constant_t(nrnmpi_barrier_impl)> coreneuron::nrnmpi_barrier

Initial value:

{

"nrnmpi_barrier_impl"}

Definition at line 42 of file nrnmpidec.cpp.

◆ nrnmpi_check_threading_support

mpi_function< cnrn_make_integral_constant_t(nrnmpi_check_threading_support_impl)> coreneuron::nrnmpi_check_threading_support {"nrnmpi_check_threading_support_impl"}

Definition at line 19 of file nrnmpidec.cpp.

◆ nrnmpi_comm

MPI_Comm coreneuron::nrnmpi_comm

Definition at line 24 of file nrnmpi.cpp.

◆ nrnmpi_dbl_allmax

mpi_function< cnrn_make_integral_constant_t(nrnmpi_dbl_allmax_impl)> coreneuron::nrnmpi_dbl_allmax

Initial value:

{

"nrnmpi_dbl_allmax_impl"}

Definition at line 40 of file nrnmpidec.cpp.

◆ nrnmpi_dbl_allmin

mpi_function< cnrn_make_integral_constant_t(nrnmpi_dbl_allmin_impl)> coreneuron::nrnmpi_dbl_allmin

Initial value:

{

"nrnmpi_dbl_allmin_impl"}

Definition at line 38 of file nrnmpidec.cpp.

◆ nrnmpi_dbl_allreduce

mpi_function< cnrn_make_integral_constant_t(nrnmpi_dbl_allreduce_impl)> coreneuron::nrnmpi_dbl_allreduce

Initial value:

{

"nrnmpi_dbl_allreduce_impl"}

Definition at line 44 of file nrnmpidec.cpp.

◆ nrnmpi_dbl_allreduce_vec

mpi_function< cnrn_make_integral_constant_t(nrnmpi_dbl_allreduce_vec_impl)> coreneuron::nrnmpi_dbl_allreduce_vec

Initial value:

{

"nrnmpi_dbl_allreduce_vec_impl"}

Definition at line 46 of file nrnmpidec.cpp.

◆ nrnmpi_dbl_alltoallv

mpi_function< cnrn_make_integral_constant_t(nrnmpi_dbl_alltoallv_impl)> coreneuron::nrnmpi_dbl_alltoallv

Initial value:

{

"nrnmpi_dbl_alltoallv_impl"}

Definition at line 36 of file nrnmpidec.cpp.

◆ nrnmpi_finalize

mpi_function< cnrn_make_integral_constant_t(nrnmpi_finalize_impl)> coreneuron::nrnmpi_finalize

Initial value:

{

"nrnmpi_finalize_impl"}

Definition at line 16 of file nrnmpidec.cpp.

◆ nrnmpi_init

mpi_function< cnrn_make_integral_constant_t(nrnmpi_init_impl)> coreneuron::nrnmpi_init {"nrnmpi_init_impl"}

Definition at line 15 of file nrnmpidec.cpp.

◆ nrnmpi_initialized

mpi_function< cnrn_make_integral_constant_t(nrnmpi_initialized_impl)> coreneuron::nrnmpi_initialized

Initial value:

{

"nrnmpi_initialized_impl"}

Definition at line 50 of file nrnmpidec.cpp.

◆ nrnmpi_int_allgather

mpi_function< cnrn_make_integral_constant_t(nrnmpi_int_allgather_impl)> coreneuron::nrnmpi_int_allgather

Initial value:

{

"nrnmpi_int_allgather_impl"}

Definition at line 30 of file nrnmpidec.cpp.

◆ nrnmpi_int_allmax

mpi_function< cnrn_make_integral_constant_t(nrnmpi_int_allmax_impl)> coreneuron::nrnmpi_int_allmax

Initial value:

{

"nrnmpi_int_allmax_impl"}

Definition at line 28 of file nrnmpidec.cpp.

◆ nrnmpi_int_alltoall

mpi_function< cnrn_make_integral_constant_t(nrnmpi_int_alltoall_impl)> coreneuron::nrnmpi_int_alltoall

Initial value:

{

"nrnmpi_int_alltoall_impl"}

Definition at line 32 of file nrnmpidec.cpp.

◆ nrnmpi_int_alltoallv

mpi_function< cnrn_make_integral_constant_t(nrnmpi_int_alltoallv_impl)> coreneuron::nrnmpi_int_alltoallv

Initial value:

{

"nrnmpi_int_alltoallv_impl"}

Definition at line 34 of file nrnmpidec.cpp.

◆ nrnmpi_local_rank

mpi_function< cnrn_make_integral_constant_t(nrnmpi_local_rank_impl)> coreneuron::nrnmpi_local_rank

Initial value:

{

"nrnmpi_local_rank_impl"}

Definition at line 54 of file nrnmpidec.cpp.

◆ nrnmpi_local_size

mpi_function< cnrn_make_integral_constant_t(nrnmpi_local_size_impl)> coreneuron::nrnmpi_local_size

Initial value:

{

"nrnmpi_local_size_impl"}

Definition at line 56 of file nrnmpidec.cpp.

◆ nrnmpi_long_allreduce_vec

mpi_function< cnrn_make_integral_constant_t(nrnmpi_long_allreduce_vec_impl)> coreneuron::nrnmpi_long_allreduce_vec {"nrnmpi_long_allreduce_vec_impl"}

Definition at line 49 of file nrnmpidec.cpp.

◆ nrnmpi_myid

int coreneuron::nrnmpi_myid = 0

Definition at line 11 of file nrnmpi_def_cinc.cpp.

◆ nrnmpi_myid_

int coreneuron::nrnmpi_myid_

Definition at line 26 of file nrnmpi.cpp.

◆ nrnmpi_numprocs

int coreneuron::nrnmpi_numprocs = 1

Definition at line 10 of file nrnmpi_def_cinc.cpp.

◆ nrnmpi_numprocs_

int coreneuron::nrnmpi_numprocs_

Definition at line 25 of file nrnmpi.cpp.

◆ nrnmpi_spike_exchange

mpi_function< cnrn_make_integral_constant_t(nrnmpi_spike_exchange_impl)> coreneuron::nrnmpi_spike_exchange

Initial value:

{

"nrnmpi_spike_exchange_impl"}

Definition at line 24 of file nrnmpidec.cpp.

◆ nrnmpi_spike_exchange_compressed

mpi_function< cnrn_make_integral_constant_t(nrnmpi_spike_exchange_compressed_impl)> coreneuron::nrnmpi_spike_exchange_compressed {"nrnmpi_spike_exchange_compressed_impl"}

Definition at line 27 of file nrnmpidec.cpp.

◆ nrnmpi_under_nrncontrol_

bool coreneuron::nrnmpi_under_nrncontrol_ {false}

static

Definition at line 28 of file nrnmpi.cpp.

◆ nrnmpi_world_comm

MPI_Comm coreneuron::nrnmpi_world_comm

Definition at line 23 of file nrnmpi.cpp.

◆ nrnmpi_write_file

mpi_function< cnrn_make_integral_constant_t(nrnmpi_write_file_impl)> coreneuron::nrnmpi_write_file

Initial value:

{

"nrnmpi_write_file_impl"}

Definition at line 20 of file nrnmpidec.cpp.

◆ nrnmpi_wtime

mpi_function< cnrn_make_integral_constant_t(nrnmpi_wtime_impl)> coreneuron::nrnmpi_wtime {"nrnmpi_wtime_impl"}

Definition at line 53 of file nrnmpidec.cpp.

◆ nrnthread_chkpnt

NrnThreadChkpnt * coreneuron::nrnthread_chkpnt

Definition at line 651 of file nrn_checkpoint.cpp.

◆ nrnthreads_netcon_negsrcgid_tid

std::vector<std::vector<int> > coreneuron::nrnthreads_netcon_negsrcgid_tid

If a nrnthreads_netcon_srcgid is negative, need to determine the thread when in order to use the correct neg_gid2out[tid] map.

Companion to nrnthreads_netcon_srcgid when src gid is negative to allow determination of the NrnThread of the source PreSyn.

Definition at line 168 of file nrn_setup.cpp.

◆ nrnthreads_netcon_srcgid

std::vector<int*> coreneuron::nrnthreads_netcon_srcgid

Only for setup vector of netcon source gids.

Only for setup vector of netcon source gids and mindelay determination.

Definition at line 164 of file nrn_setup.cpp.

◆ nstride

int coreneuron::nstride = ii.nstride

Definition at line 641 of file cellorder.cpp.

◆ NUM_STATS

const int coreneuron::NUM_STATS = 13

Definition at line 28 of file nrn_stats.cpp.

◆ patstimtype

int coreneuron::patstimtype

Definition at line 653 of file nrn_checkpoint.cpp.

◆ pi

double coreneuron::pi

Definition at line 22 of file register_mech.cpp.

◆ pnt2index

std::map<Point_process*, int> coreneuron::pnt2index

static

Definition at line 23 of file prcellstate.cpp.

◆ pntindex

int coreneuron::pntindex

static

Definition at line 24 of file prcellstate.cpp.

◆ rev_dt

int coreneuron::rev_dt

Definition at line 23 of file register_mech.cpp.

◆ root

int coreneuron::root

Definition at line 527 of file cellorder.cpp.

◆ secondorder

int coreneuron::secondorder = 0

Definition at line 21 of file register_mech.cpp.

◆ size_report_buffer

int coreneuron::size_report_buffer = 4

static

Definition at line 34 of file nrnreport.cpp.

◆ spike_type

MPI_Datatype coreneuron::spike_type

static

Definition at line 28 of file mpispike.cpp.

◆ spikevec_gid

std::vector< int > coreneuron::spikevec_gid

Definition at line 46 of file output_spikes.cpp.

◆ spikevec_time

std::vector< double > coreneuron::spikevec_time

--> Coreneuron as SpikeBuffer class

Definition at line 45 of file output_spikes.cpp.

◆ state_discon_allowed_

int coreneuron::state_discon_allowed_

Definition at line 369 of file register_mech.cpp.

◆ state_discon_flag_

int coreneuron::state_discon_flag_ = 0

Definition at line 370 of file register_mech.cpp.

◆ stoprun

bool coreneuron::stoprun

Definition at line 19 of file nrnoc_aux.cpp.

◆ stride

int * coreneuron::stride = ii.stride

Definition at line 482 of file cellorder.cpp.

◆ structure_change_cnt

int coreneuron::structure_change_cnt

◆ t

double coreneuron::t

Definition at line 22 of file register_mech.cpp.

◆ table_check_

ThreadDatum* coreneuron::table_check_

static

Definition at line 61 of file multicore.cpp.

◆ table_check_cnt_

int coreneuron::table_check_cnt_

static

--> CoreNeuron class

Definition at line 60 of file multicore.cpp.

◆ timeout_

int coreneuron::timeout_ = 0

static

Definition at line 597 of file netpar.cpp.

◆ type2invperm

std::map<int, int*> coreneuron::type2invperm

Definition at line 430 of file core2nrn_data_return.cpp.

◆ type2movable

std::unordered_map<int, int> coreneuron::type2movable

static

Definition at line 135 of file nrn2core_data_init.cpp.

◆ type_hints

std::vector<int> coreneuron::type_hints

static

Definition at line 160 of file node_permute.cpp.

◆ usable_mindelay_

double coreneuron::usable_mindelay_

static

Definition at line 84 of file netpar.cpp.

◆ use_multisend_

bool coreneuron::use_multisend_

Definition at line 53 of file multisend.cpp.

◆ use_phase2_

bool coreneuron::use_phase2_

Definition at line 54 of file multisend.cpp.

◆ use_solve_interleave

bool coreneuron::use_solve_interleave

Definition at line 13 of file solve_core.cpp.

◆ v_structure_change

int coreneuron::v_structure_change

Definition at line 20 of file nrnoc_aux.cpp.

Detailed Description

Namespaces

Classes

Typedefs

Enumerations

Functions

Variables

Typedef Documentation

◆ bbcore_read_t

◆ bbcore_write_t

◆ Core2NrnWatchInfo

◆ Core2NrnWatchInfoItem

◆ Datum

◆ DependencyTable

◆ HashCnt

◆ IvocVect

◆ List

◆ MechId

◆ MechNamesMapping

◆ mod_alloc_t

◆ mod_f_t

◆ N2V

◆ NetBufReceive_t

◆ NRB_P

◆ nrn_watch_check_t

◆ Offset

◆ Pfri

◆ Pfrpdat

◆ Pfrv

◆ pnt_receive_t

◆ PSD

◆ secseg_it_type

◆ secseg_map_type

◆ segvec_type

◆ SelfEventWeightMap

◆ SerializedNames

◆ Symbol

◆ thread_table_check_t

◆ TNI

◆ TNIVec

◆ TQPair

◆ VariableName

◆ VecTNode

◆ VTN

◆ VVTN

◆ VVVTN

Enumeration Type Documentation

◆ container

◆ Layout

◆ LFPCalculatorType

◆ mech_type

◆ ReportType

◆ SectionType

◆ TargetType

Function Documentation

◆ _cvode_sparse_thread()

◆ _exp2syn_reg()

◆ _expsyn_reg()

◆ _hh_reg()

◆ _modl_cleanup()

◆ _netstim_reg()

◆ _nrn_destroy_sparseobj_thread()

◆ _nrn_layout_reg()

◆ _nrn_setdata_reg()

◆ _nrn_thread_reg0()

◆ _nrn_thread_reg1()

◆ _nrn_thread_table_reg() [1/2]

◆ _nrn_thread_table_reg() [2/2]

◆ _passive_reg()

◆ _pattern_reg()

◆ _ss_derivimplicit_thread()

◆ _ss_sparse_thread()

◆ _stim_reg()

◆ _svclmp_reg()

◆ abort_run()

◆ add_nrn_artcell()

◆ add_nrn_fornetcons()

◆ add_nrn_has_net_event()

◆ admin1()

◆ admin2()