Neko  0.9.99
A portable framework for high-order spectral element flow simulations
fusedcg_device Module Reference

Defines a fused Conjugate Gradient method for accelerators.

Data Types

type  fusedcg_device_t
 Fused preconditioned conjugate gradient method. More...
 
interface  cuda_fusedcg_update_p
 
interface  cuda_fusedcg_update_x
 
interface  cuda_fusedcg_part2
 

Functions/Subroutines

subroutine device_fusedcg_update_p (p_d, z_d, po_d, beta, n)
 
subroutine device_fusedcg_update_x (x_d, p_d, alpha, p_cur, n)
 
real(kind=rp) function device_fusedcg_part2 (a_d, b_d, c_d, alpha_d, alpha, p_cur, n)
 
subroutine fusedcg_device_init (this, n, max_iter, M, rel_tol, abs_tol, monitor)
 Initialise a fused PCG solver. More...
 
subroutine fusedcg_device_free (this)
 Deallocate a pipelined PCG solver. More...
 
type(ksp_monitor_t) function fusedcg_device_solve (this, Ax, x, f, n, coef, blst, gs_h, niter)
 Pipelined PCG solve. More...
 
type(ksp_monitor_t) function, dimension(3) fusedcg_device_solve_coupled (this, Ax, x, y, z, fx, fy, fz, n, coef, blstx, blsty, blstz, gs_h, niter)
 Pipelined PCG solve coupled solve. More...
 

Variables

integer, parameter device_fusedcg_p_space = 10
 

Function/Subroutine Documentation

◆ device_fusedcg_part2()

real(kind=rp) function fusedcg_device::device_fusedcg_part2 ( type(c_ptr), value  a_d,
type(c_ptr), value  b_d,
type(c_ptr), value  c_d,
type(c_ptr), value  alpha_d,
real(c_rp)  alpha,
integer  p_cur,
integer  n 
)
private

Definition at line 170 of file fusedcg_device.F90.

Here is the call graph for this function:
Here is the caller graph for this function:

◆ device_fusedcg_update_p()

subroutine fusedcg_device::device_fusedcg_update_p ( type(c_ptr), value  p_d,
type(c_ptr), value  z_d,
type(c_ptr), value  po_d,
real(c_rp)  beta,
integer(c_int)  n 
)
private

Definition at line 145 of file fusedcg_device.F90.

Here is the call graph for this function:
Here is the caller graph for this function:

◆ device_fusedcg_update_x()

subroutine fusedcg_device::device_fusedcg_update_x ( type(c_ptr), value  x_d,
type(c_ptr), value  p_d,
type(c_ptr), value  alpha,
integer(c_int)  p_cur,
integer(c_int)  n 
)
private

Definition at line 158 of file fusedcg_device.F90.

Here is the call graph for this function:
Here is the caller graph for this function:

◆ fusedcg_device_free()

subroutine fusedcg_device::fusedcg_device_free ( class(fusedcg_device_t), intent(inout)  this)
private

Definition at line 258 of file fusedcg_device.F90.

◆ fusedcg_device_init()

subroutine fusedcg_device::fusedcg_device_init ( class(fusedcg_device_t), intent(inout), target  this,
integer, intent(in)  n,
integer, intent(in)  max_iter,
class(pc_t), intent(in), optional, target  M,
real(kind=rp), intent(in), optional  rel_tol,
real(kind=rp), intent(in), optional  abs_tol,
logical, intent(in), optional  monitor 
)
private

Definition at line 195 of file fusedcg_device.F90.

◆ fusedcg_device_solve()

type(ksp_monitor_t) function fusedcg_device::fusedcg_device_solve ( class(fusedcg_device_t), intent(inout)  this,
class(ax_t), intent(in)  Ax,
type(field_t), intent(inout)  x,
real(kind=rp), dimension(n), intent(in)  f,
integer, intent(in)  n,
type(coef_t), intent(inout)  coef,
type(bc_list_t), intent(in)  blst,
type(gs_t), intent(inout)  gs_h,
integer, intent(in), optional  niter 
)
private

Definition at line 319 of file fusedcg_device.F90.

Here is the call graph for this function:

◆ fusedcg_device_solve_coupled()

type(ksp_monitor_t) function, dimension(3) fusedcg_device::fusedcg_device_solve_coupled ( class(fusedcg_device_t), intent(inout)  this,
class(ax_t), intent(in)  Ax,
type(field_t), intent(inout)  x,
type(field_t), intent(inout)  y,
type(field_t), intent(inout)  z,
real(kind=rp), dimension(n), intent(in)  fx,
real(kind=rp), dimension(n), intent(in)  fy,
real(kind=rp), dimension(n), intent(in)  fz,
integer, intent(in)  n,
type(coef_t), intent(inout)  coef,
type(bc_list_t), intent(in)  blstx,
type(bc_list_t), intent(in)  blsty,
type(bc_list_t), intent(in)  blstz,
type(gs_t), intent(inout)  gs_h,
integer, intent(in), optional  niter 
)
private

Definition at line 403 of file fusedcg_device.F90.

Variable Documentation

◆ device_fusedcg_p_space

integer, parameter fusedcg_device::device_fusedcg_p_space = 10
private

Definition at line 50 of file fusedcg_device.F90.