mod_eos_lte Module Reference

LTE (Saha-table) EoS kernels and finalise for the eos% family. More...

Functions/Subroutines
subroutine, public	eos_init_lte ()
	update_eos_LTE, get_Te_LTE, get_temperature_from_eint_LTE and Rfactor_from_LTE are PRIVATE: all bound to eos% pointers inside eos_finalise_LTE, so no external module names them. get_temperature_from_eint_fast_LTE stays public – the seam binds it to tc_flget_temperature_from_eint (fast TC path).
subroutine, public	eos_finalise_lte ()
	LTE arm of eos_finalise: wire the LTE runtime pointer targets, then dispatch on eos_method to the per-method finaliser, each owned by its method module (analytic->mod_eos_LTE_saha, entropy->mod_eos_LTE_entropy, tables->mod_eos_LTE_tables).
subroutine, public	get_temperature_from_eint_fast_lte (w, x, ixil, ixol, res)
double precision function, public	y_from_nh_eint (nh, eint_nh)
	Ionization fraction from (log10 nH, log10 eint/nH) in code units. Dispatches: analytic -> Saha quadratic, tables -> PCHIP interpolation.
double precision function, public	t_from_nh_eint (nh, eint_nh)
	Temperature from (log10 nH, log10 eint/nH) in code units. Dispatches: analytic -> Saha bisection/Newton, tables -> PCHIP interpolation.
double precision function, public	p2eint_from_nh_p (nh, ponh)
	Pressure-to-eint ratio from (log10 nH, log10 p/nH) in code units. Dispatches: analytic -> Saha solve for eint/p, tables -> PCHIP interpolation.
double precision function, public	gamma1_from_nh_p (log_nh, log_p_nh)
	Gamma_1 from pressure-indexed table: (log10 nH, log10 p/nH) -> Gamma_1. For 'entropy' the conversion p -> eint -> Gamma_1 via formula is intended to keep Maxwell consistency with the forward at runtime. The p->eint inverse is one bisection per cell – non-trivial cost; this function is in the hot path via hd_get_csound2_LTE.
double precision function, public	p_nh_from_eint (log_nh, log_eint_nh)
	p/nH from (log10 nH, log10 eint/nH) in code units. Returns (1+He+y)*T directly – single lookup replaces T + y lookups.
double precision function, public	eint_nh_from_t (log_nh, log_t)
	Internal energy per nH from (log10 nH, log10 T) in code units. Uses the bisection-built inverse table (H+He, machine precision). Fallback: H-only Saha if table not built.
subroutine, public	eint_from_p_bisect (log_nh_val, log_p_val, log_eint_nh_out)
	Given log10(nH) and log10(p), find log10(eint/nH) by table-guessed bisection on the forward pressure table. Wraps the bracketing logic and log_p_bisect_cached into a single call for use by both HD and MHD from_conserved routines.
subroutine, public	eos_get_eintt_grid (n_nh, lg_nh_min, lg_nh_max)
	log_nH grid metadata of the (log_nH, log_T) inverse table (eint from T), choosing the container by method ('tables'->eint_from_T, 'entropy'->eintT). n_nH=0 if no such table (analytic/FI). Lets cooling's build_Y_mod_table get its grid without reaching into EoS table internals.

Detailed Description

LTE (Saha-table) EoS kernels and finalise for the eos% family.

Carved out of mod_eos.t (the per-type split). Owns everything LTE: the per-substep update_eos_LTE (caches Te_/Ne_), the cached-Te/Rfactor getters, the scalar table kernels (T, y, p2eint, gamma1, p/nH, eint, bisections), the per-method table finalisers, and eos_finalise_LTE (the LTE arm of the eos_finalise dispatcher). Depends only on the EoS leaf modules + shared accessors, never on mod_eos -> no circular use. mod_eos re-exports the public kernels so existing use mod_eos callers are unaffected.

Function/Subroutine Documentation

eint_from_p_bisect()

subroutine, public mod_eos_lte::eint_from_p_bisect	(	double precision, intent(in)	log_nh_val,
		double precision, intent(in)	log_p_val,
		double precision, intent(out)	log_eint_nh_out
	)

Given log10(nH) and log10(p), find log10(eint/nH) by table-guessed bisection on the forward pressure table. Wraps the bracketing logic and log_p_bisect_cached into a single call for use by both HD and MHD from_conserved routines.

Definition at line 758 of file mod_eos_LTE.t.

Here is the call graph for this function:

eint_nh_from_t()

double precision function, public mod_eos_lte::eint_nh_from_t	(	double precision, intent(in)	log_nh,
		double precision, intent(in)	log_t
	)

Internal energy per nH from (log10 nH, log10 T) in code units. Uses the bisection-built inverse table (H+He, machine precision). Fallback: H-only Saha if table not built.

Definition at line 536 of file mod_eos_LTE.t.

Here is the call graph for this function:

eos_finalise_lte()

subroutine, public mod_eos_lte::eos_finalise_lte

LTE arm of eos_finalise: wire the LTE runtime pointer targets, then dispatch on eos_method to the per-method finaliser, each owned by its method module (analytic->mod_eos_LTE_saha, entropy->mod_eos_LTE_entropy, tables->mod_eos_LTE_tables).

R-factor: physics-independent (cached-Ne lookup), bound here rather than in the hd/mhd/ffhd link arms so the target stays private.

Definition at line 59 of file mod_eos_LTE.t.

Here is the call graph for this function:

eos_get_eintt_grid()

subroutine, public mod_eos_lte::eos_get_eintt_grid	(	integer, intent(out)	n_nh,
		double precision, intent(out)	lg_nh_min,
		double precision, intent(out)	lg_nh_max
	)

log_nH grid metadata of the (log_nH, log_T) inverse table (eint from T), choosing the container by method ('tables'->eint_from_T, 'entropy'->eintT). n_nH=0 if no such table (analytic/FI). Lets cooling's build_Y_mod_table get its grid without reaching into EoS table internals.

Table-metadata helpers: expose inverse-table grid bounds without reaching into EoS table internals.

Definition at line 813 of file mod_eos_LTE.t.

eos_init_lte()

subroutine, public mod_eos_lte::eos_init_lte

update_eos_LTE, get_Te_LTE, get_temperature_from_eint_LTE and Rfactor_from_LTE are PRIVATE: all bound to eos% pointers inside eos_finalise_LTE, so no external module names them. get_temperature_from_eint_fast_LTE stays public – the seam binds it to tc_flget_temperature_from_eint (fast TC path).

Scalar table kernels (re-exported by mod_eos for the hd/mhd/ffhd seams). T_and_y_from_nH_eint is internal-only (entropy update path). LTE arms of the eos_init / eos_finalise dispatchers LTE arm of eos_init (before units are known): dispatch on eos_method to the chosen method's loader, each owned by its method module (analytic loads nothing). The LTE temperature-from-pressure getter is a later pass, so unlike FI/PI it is left unset here.

Definition at line 47 of file mod_eos_LTE.t.

Here is the call graph for this function:

gamma1_from_nh_p()

double precision function, public mod_eos_lte::gamma1_from_nh_p	(	double precision, intent(in)	log_nh,
		double precision, intent(in)	log_p_nh
	)

Gamma_1 from pressure-indexed table: (log10 nH, log10 p/nH) -> Gamma_1. For 'entropy' the conversion p -> eint -> Gamma_1 via formula is intended to keep Maxwell consistency with the forward at runtime. The p->eint inverse is one bisection per cell – non-trivial cost; this function is in the hot path via hd_get_csound2_LTE.

Definition at line 492 of file mod_eos_LTE.t.

Here is the call graph for this function:

get_temperature_from_eint_fast_lte()

subroutine, public mod_eos_lte::get_temperature_from_eint_fast_lte	(	double precision, dimension(ixi^s,1:nw), intent(in)	w,
		double precision, dimension(ixi^s,1:ndim), intent(in)	x,
		integer, intent(in)	ixi,
		integer, intent(in)	l,
		integer, intent(in)	ixo,
			l,
		double precision, dimension(ixi^s), intent(out)	res
	)

Parameters

[in]

ixi

Fast TC variant: two-pass regime-aware bypass.

Pass 1 (vectorised): compute FI formula for ALL cells as array ops. The compiler vectorises this with SVML (no branches, pure arithmetic).

Pass 2 (scalar): overwrite ionisation-zone cells with bilinear table lookup + dexp. Threshold check uses multiply (eint <= threshold * rho) to avoid a 14-cycle scalar division.

Parameters

[in]

l

Fast TC variant: two-pass regime-aware bypass.

Pass 1 (vectorised): compute FI formula for ALL cells as array ops. The compiler vectorises this with SVML (no branches, pure arithmetic).

Pass 2 (scalar): overwrite ionisation-zone cells with bilinear table lookup + dexp. Threshold check uses multiply (eint <= threshold * rho) to avoid a 14-cycle scalar division.

Parameters

[in]

ixo

Fast TC variant: two-pass regime-aware bypass.

Pass 1 (vectorised): compute FI formula for ALL cells as array ops. The compiler vectorises this with SVML (no branches, pure arithmetic).

Pass 2 (scalar): overwrite ionisation-zone cells with bilinear table lookup + dexp. Threshold check uses multiply (eint <= threshold * rho) to avoid a 14-cycle scalar division.

Parameters

[in]

l

Fast TC variant: two-pass regime-aware bypass.

Pass 1 (vectorised): compute FI formula for ALL cells as array ops. The compiler vectorises this with SVML (no branches, pure arithmetic).

Pass 2 (scalar): overwrite ionisation-zone cells with bilinear table lookup + dexp. Threshold check uses multiply (eint <= threshold * rho) to avoid a 14-cycle scalar division.

Precompute scalar constants (avoid per-cell divisions)

Pass 1: FI formula for ALL cells (vectorisable array operations). T = (gamma-1) * (eint - eion_per_rho * rho) / (Rfactor_FI * rho)

Pass 2: overwrite ionisation-zone cells.

Inlined uniform bilinear (hot path: no function call).

Non-uniform: dispatch to the binary-search bilinear. Slightly slower per cell (binary search adds ~8 cmps per axis) but unavoidable when the grid is non-uniform.

Definition at line 265 of file mod_eos_LTE.t.

Here is the call graph for this function:

p2eint_from_nh_p()

double precision function, public mod_eos_lte::p2eint_from_nh_p	(	double precision, intent(in)	nh,
		double precision, intent(in)	ponh
	)

Pressure-to-eint ratio from (log10 nH, log10 p/nH) in code units. Dispatches: analytic -> Saha solve for eint/p, tables -> PCHIP interpolation.

Definition at line 460 of file mod_eos_LTE.t.

Here is the call graph for this function:

p_nh_from_eint()

double precision function, public mod_eos_lte::p_nh_from_eint	(	double precision, intent(in)	log_nh,
		double precision, intent(in)	log_eint_nh
	)

p/nH from (log10 nH, log10 eint/nH) in code units. Returns (1+He+y)*T directly – single lookup replaces T + y lookups.

Definition at line 515 of file mod_eos_LTE.t.

Here is the call graph for this function:

t_from_nh_eint()

double precision function, public mod_eos_lte::t_from_nh_eint	(	double precision, intent(in)	nh,
		double precision, intent(in)	eint_nh
	)

Temperature from (log10 nH, log10 eint/nH) in code units. Dispatches: analytic -> Saha bisection/Newton, tables -> PCHIP interpolation.

Definition at line 390 of file mod_eos_LTE.t.

Here is the call graph for this function:

y_from_nh_eint()

double precision function, public mod_eos_lte::y_from_nh_eint	(	double precision, intent(in)	nh,
		double precision, intent(in)	eint_nh
	)

Ionization fraction from (log10 nH, log10 eint/nH) in code units. Dispatches: analytic -> Saha quadratic, tables -> PCHIP interpolation.

Scalar EoS kernels: pointwise lookups on (log10 nH, log10 eint/nH) or (log10 nH, log10 p/nH) in code units, called by the hd/mhd EoS layer. Each dispatches analytic / entropy / table. Ionisation fraction and temperature first, then pressure / Gamma_1 / eint.

Definition at line 363 of file mod_eos_LTE.t.

Here is the call graph for this function: