device_mma_math.f90 File Reference

Go to the source code of this file.

Functions/Subroutines
subroutine, public	device_mma_math::device_update_hessian_z (hess_d, a_d, m)
	Update Hessian for dual solver with z-term contribution: Hess -= a * a^T.
subroutine, public	device_mma_math::device_prepare_aa_matrix (aa_d, s_d, lambda_d, d_d, mu_d, y_d, a_d, zeta, z, m)
	Prepare AA matrix for dual-primal solver on device.
subroutine, public	device_mma_math::device_prepare_hessian (hess_d, y_d, mu_d, lambda_d, m)
	Solve linear system Ax = b on device.
subroutine, public	device_mma_math::device_solve_linear_system (a_d, b_d, n, info)
subroutine, public	device_mma_math::device_delta_1dbeam (delta_d, l_total, le, offset, n)
	A device support to do the following calculation for 1D beam elements: Delta(k) = ((L_total - Le*(offset+k-1))**3 - & (L_total - Le*(offset+k))**3) / 3.0_rp Where k ranges from 1 to n.
subroutine, public	device_mma_math::device_hess (hess_d, hijx_d, ljjxinv_d, n, m)
subroutine, public	device_mma_math::device_mma_ljjxinv (ljjxinv_d, pjlambda_d, qjlambda_d, x_d, low_d, upp_d, alpha_d, beta_d, n)
	A device support to do the following calculation needed for the dualsubsolve for MMA: Ljjxinv= - 1 / ( (2*pjlambda/(upp - x)**3) + & (2*qjlambda/(x - low)**3))
subroutine, public	device_mma_math::device_mma_dipsolvesub1 (x_d, pjlambda_d, qjlambda_d, low_d, upp_d, alpha_d, beta_d, n)
	A device support to do the following calculation needed for the dualsubsolve for MMA: x = (sqrt(pjlambda) * low + sqrt(qjlambda) * upp) / & (sqrt(pjlambda) + sqrt(qjlambda))
subroutine, public	device_mma_math::device_mattrans_v_mul (output_d, pij_d, lambda_d, m, n)
	A device support to do the following matrix multiplication output = matmul(transpose(pij), lambda) where matrix pij is mxn, vector lambda is of size m and the output vector is of size n.
subroutine, public	device_mma_math::device_mma_gensub1 (low_d, upp_d, x_d, xmin_d, xmax_d, asyinit, n)
subroutine, public	device_mma_math::device_mma_gensub2 (low_d, upp_d, x_d, xold1_d, xold2_d, xdiff_d, asydecr, asyincr, n)
subroutine, public	device_mma_math::device_mma_gensub3 (x_d, df0dx_d, dfdx_d, low_d, upp_d, min_d, max_d, alpha_d, beta_d, p0j_d, q0j_d, pij_d, qij_d, n, m)
subroutine, public	device_mma_math::device_mma_gensub4 (x_d, low_d, upp_d, pij_d, qij_d, n, m, bi_d)
subroutine, public	device_mma_math::device_mma_max (xsi_d, x_d, alpha_d, n)
subroutine, public	device_mma_math::device_max2 (a_d, b, c_d, d, n)
subroutine, public	device_mma_math::device_rex (rex_d, x_d, low_d, upp_d, pij_d, p0j_d, qij_d, q0j_d, lambda_d, xsi_d, eta_d, n, m)
real(kind=rp) function, public	device_mma_math::device_lcsc2 (a_d, b_d, n)
subroutine, public	device_mma_math::device_relambda (relambda_d, x_d, upp_d, low_d, pij_d, qij_d, n, m)
subroutine, public	device_mma_math::device_sub2cons2 (rexsi_d, xsi_d, x_d, alpha_d, epsi, n)
real(kind=rp) function, public	device_mma_math::device_maxval (rex_d, n)
real(kind=rp) function, public	device_mma_math::device_norm (rex_d, n)
subroutine, public	device_mma_math::device_delx (delx_d, x_d, low_d, upp_d, pij_d, qij_d, p0j_d, q0j_d, alpha_d, beta_d, lambda_d, epsi, n, m)
subroutine, public	device_mma_math::device_add2inv2 (a_d, b_d, c, n)
subroutine, public	device_mma_math::device_gg (gg_d, x_d, low_d, upp_d, pij_d, qij_d, n, m)
subroutine, public	device_mma_math::device_diagx (diagx_d, x_d, xsi_d, low_d, upp_d, p0j_d, q0j_d, pij_d, qij_d, alpha_d, beta_d, eta_d, lambda_d, n, m)
subroutine, public	device_mma_math::device_bb (bb_d, gg_d, delx_d, diagx_d, n, m)
subroutine, public	device_mma_math::device_updatebb (bb_d, dellambda_d, dely_d, d_d, mu_d, y_d, delz, m)
subroutine, public	device_mma_math::device_aa (aa_d, gg_d, diagx_d, n, m)
subroutine, public	device_mma_math::device_updateaa (aa_d, globaltmp_mm_d, s_d, lambda_d, d_d, mu_d, y_d, a_d, zeta, z, m)
subroutine, public	device_mma_math::device_dx (dx_d, delx_d, diagx_d, gg_d, dlambda_d, n, m)
subroutine, public	device_mma_math::device_dy (dy_d, dely_d, dlambda_d, d_d, mu_d, y_d, n)
subroutine, public	device_mma_math::device_dxsi (dxsi_d, xsi_d, dx_d, x_d, alpha_d, epsi, n)
subroutine, public	device_mma_math::device_deta (deta_d, eta_d, dx_d, x_d, beta_d, epsi, n)
real(kind=rp) function, public	device_mma_math::device_maxval2 (dxx_d, xx_d, cons, n)
real(kind=rp) function, public	device_mma_math::device_maxval3 (dx_d, x_d, alpha_d, cons, n)
subroutine, public	device_mma_math::device_kkt_rex (rex_d, df0dx_d, dfdx_d, xsi_d, eta_d, lambda_d, n, m)

Detailed Description

Copyright

Copyright (c) 2025, The Neko-TOP Authors All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

• Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
• Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
• Neither the name of the authors nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

Definition in file device_mma_math.f90.

Function/Subroutine Documentation

device_aa()

subroutine, public device_mma_math::device_aa	(	type(c_ptr)	aa_d,
		type(c_ptr)	gg_d,
		type(c_ptr)	diagx_d,
		integer(c_int)	n,
		integer(c_int)	m
	)

Definition at line 544 of file device_mma_math.f90.

device_add2inv2()

subroutine, public device_mma_math::device_add2inv2	(	type(c_ptr)	a_d,
		type(c_ptr)	b_d,
		real(kind=rp)	c,
		integer(c_int)	n
	)

Definition at line 465 of file device_mma_math.f90.

device_bb()

subroutine, public device_mma_math::device_bb	(	type(c_ptr)	bb_d,
		type(c_ptr)	gg_d,
		type(c_ptr)	delx_d,
		type(c_ptr)	diagx_d,
		integer(c_int)	n,
		integer(c_int)	m
	)

Definition at line 515 of file device_mma_math.f90.

device_delta_1dbeam()

subroutine, public device_mma_math::device_delta_1dbeam	(	type(c_ptr)	delta_d,
		real(c_rp)	l_total,
		real(c_rp)	le,
		integer(c_int), value	offset,
		integer(c_int), value	n
	)

Definition at line 150 of file device_mma_math.f90.

device_delx()

subroutine, public device_mma_math::device_delx	(	type(c_ptr)	delx_d,
		type(c_ptr)	x_d,
		type(c_ptr)	low_d,
		type(c_ptr)	upp_d,
		type(c_ptr)	pij_d,
		type(c_ptr)	qij_d,
		type(c_ptr)	p0j_d,
		type(c_ptr)	q0j_d,
		type(c_ptr)	alpha_d,
		type(c_ptr)	beta_d,
		type(c_ptr)	lambda_d,
		real(kind=rp)	epsi,
		integer(c_int)	n,
		integer(c_int)	m
	)

Definition at line 446 of file device_mma_math.f90.

device_deta()

subroutine, public device_mma_math::device_deta	(	type(c_ptr)	deta_d,
		type(c_ptr)	eta_d,
		type(c_ptr)	dx_d,
		type(c_ptr)	x_d,
		type(c_ptr)	beta_d,
		real(c_rp)	epsi,
		integer(c_int)	n
	)

Definition at line 619 of file device_mma_math.f90.

device_diagx()

subroutine, public device_mma_math::device_diagx	(	type(c_ptr)	diagx_d,
		type(c_ptr)	x_d,
		type(c_ptr)	xsi_d,
		type(c_ptr)	low_d,
		type(c_ptr)	upp_d,
		type(c_ptr)	p0j_d,
		type(c_ptr)	q0j_d,
		type(c_ptr)	pij_d,
		type(c_ptr)	qij_d,
		type(c_ptr)	alpha_d,
		type(c_ptr)	beta_d,
		type(c_ptr)	eta_d,
		type(c_ptr)	lambda_d,
		integer(c_int)	n,
		integer(c_int)	m
	)

Definition at line 495 of file device_mma_math.f90.

device_dx()

subroutine, public device_mma_math::device_dx	(	type(c_ptr)	dx_d,
		type(c_ptr)	delx_d,
		type(c_ptr)	diagx_d,
		type(c_ptr)	gg_d,
		type(c_ptr)	dlambda_d,
		integer(c_int)	n,
		integer(c_int)	m
	)

Definition at line 576 of file device_mma_math.f90.

device_dxsi()

subroutine, public device_mma_math::device_dxsi	(	type(c_ptr)	dxsi_d,
		type(c_ptr)	xsi_d,
		type(c_ptr)	dx_d,
		type(c_ptr)	x_d,
		type(c_ptr)	alpha_d,
		real(c_rp)	epsi,
		integer(c_int)	n
	)

Definition at line 604 of file device_mma_math.f90.

device_dy()

subroutine, public device_mma_math::device_dy	(	type(c_ptr)	dy_d,
		type(c_ptr)	dely_d,
		type(c_ptr)	dlambda_d,
		type(c_ptr)	d_d,
		type(c_ptr)	mu_d,
		type(c_ptr)	y_d,
		integer(c_int)	n
	)

Definition at line 590 of file device_mma_math.f90.

device_gg()

subroutine, public device_mma_math::device_gg	(	type(c_ptr)	gg_d,
		type(c_ptr)	x_d,
		type(c_ptr)	low_d,
		type(c_ptr)	upp_d,
		type(c_ptr)	pij_d,
		type(c_ptr)	qij_d,
		integer(c_int)	n,
		integer(c_int)	m
	)

Definition at line 481 of file device_mma_math.f90.

device_hess()

subroutine, public device_mma_math::device_hess	(	type(c_ptr)	hess_d,
		type(c_ptr)	hijx_d,
		type(c_ptr)	ljjxinv_d,
		integer(c_int)	n,
		integer(c_int)	m
	)

Definition at line 166 of file device_mma_math.f90.

device_kkt_rex()

subroutine, public device_mma_math::device_kkt_rex	(	type(c_ptr)	rex_d,
		type(c_ptr)	df0dx_d,
		type(c_ptr)	dfdx_d,
		type(c_ptr)	xsi_d,
		type(c_ptr)	eta_d,
		type(c_ptr)	lambda_d,
		integer(c_int)	n,
		integer(c_int)	m
	)

Definition at line 673 of file device_mma_math.f90.

device_lcsc2()

real(kind=rp) function, public device_mma_math::device_lcsc2	(	type(c_ptr)	a_d,
		type(c_ptr)	b_d,
		integer(c_int)	n
	)

Definition at line 366 of file device_mma_math.f90.

device_mattrans_v_mul()

subroutine, public device_mma_math::device_mattrans_v_mul	(	type(c_ptr)	output_d,
		type(c_ptr)	pij_d,
		type(c_ptr)	lambda_d,
		integer	m,
		integer	n
	)

Definition at line 235 of file device_mma_math.f90.

device_max2()

subroutine, public device_mma_math::device_max2	(	type(c_ptr)	a_d,
		real(c_rp)	b,
		type(c_ptr)	c_d,
		real(c_rp)	d,
		integer	n
	)

Definition at line 331 of file device_mma_math.f90.

device_maxval()

real(kind=rp) function, public device_mma_math::device_maxval	(	type(c_ptr)	rex_d,
		integer(c_int)	n
	)

Definition at line 412 of file device_mma_math.f90.

device_maxval2()

real(kind=rp) function, public device_mma_math::device_maxval2	(	type(c_ptr)	dxx_d,
		type(c_ptr)	xx_d,
		real(kind=rp), intent(in)	cons,
		integer	n
	)

Definition at line 634 of file device_mma_math.f90.

device_maxval3()

real(kind=rp) function, public device_mma_math::device_maxval3	(	type(c_ptr)	dx_d,
		type(c_ptr)	x_d,
		type(c_ptr)	alpha_d,
		real(kind=rp), intent(in)	cons,
		integer(c_int)	n
	)

Definition at line 653 of file device_mma_math.f90.

device_mma_dipsolvesub1()

subroutine, public device_mma_math::device_mma_dipsolvesub1	(	type(c_ptr)	x_d,
		type(c_ptr)	pjlambda_d,
		type(c_ptr)	qjlambda_d,
		type(c_ptr)	low_d,
		type(c_ptr)	upp_d,
		type(c_ptr)	alpha_d,
		type(c_ptr)	beta_d,
		integer(c_int)	n
	)

Definition at line 212 of file device_mma_math.f90.

device_mma_gensub1()

subroutine, public device_mma_math::device_mma_gensub1	(	type(c_ptr)	low_d,
		type(c_ptr)	upp_d,
		type(c_ptr)	x_d,
		type(c_ptr)	xmin_d,
		type(c_ptr)	xmax_d,
		real(c_rp)	asyinit,
		integer	n
	)

Definition at line 251 of file device_mma_math.f90.

device_mma_gensub2()

subroutine, public device_mma_math::device_mma_gensub2	(	type(c_ptr)	low_d,
		type(c_ptr)	upp_d,
		type(c_ptr)	x_d,
		type(c_ptr)	xold1_d,
		type(c_ptr)	xold2_d,
		type(c_ptr)	xdiff_d,
		real(c_rp)	asydecr,
		real(c_rp)	asyincr,
		integer	n
	)

Definition at line 267 of file device_mma_math.f90.

device_mma_gensub3()

subroutine, public device_mma_math::device_mma_gensub3	(	type(c_ptr)	x_d,
		type(c_ptr)	df0dx_d,
		type(c_ptr)	dfdx_d,
		type(c_ptr)	low_d,
		type(c_ptr)	upp_d,
		type(c_ptr)	min_d,
		type(c_ptr)	max_d,
		type(c_ptr)	alpha_d,
		type(c_ptr)	beta_d,
		type(c_ptr)	p0j_d,
		type(c_ptr)	q0j_d,
		type(c_ptr)	pij_d,
		type(c_ptr)	qij_d,
		integer	n,
		integer	m
	)

Definition at line 285 of file device_mma_math.f90.

device_mma_gensub4()

subroutine, public device_mma_math::device_mma_gensub4	(	type(c_ptr)	x_d,
		type(c_ptr)	low_d,
		type(c_ptr)	upp_d,
		type(c_ptr)	pij_d,
		type(c_ptr)	qij_d,
		integer	n,
		integer	m,
		type(c_ptr)	bi_d
	)

Definition at line 303 of file device_mma_math.f90.

device_mma_ljjxinv()

subroutine, public device_mma_math::device_mma_ljjxinv	(	type(c_ptr)	ljjxinv_d,
		type(c_ptr)	pjlambda_d,
		type(c_ptr)	qjlambda_d,
		type(c_ptr)	x_d,
		type(c_ptr)	low_d,
		type(c_ptr)	upp_d,
		type(c_ptr)	alpha_d,
		type(c_ptr)	beta_d,
		integer(c_int)	n
	)

And then remove the sensitivity for the active primal constraints Ljjxinv = merge(0.0_rp, Ljjxinv, x .eq. alpha) Ljjxinv = merge(0.0_rp, Ljjxinv, x .eq. beta)

Definition at line 189 of file device_mma_math.f90.

device_mma_max()

subroutine, public device_mma_math::device_mma_max	(	type(c_ptr)	xsi_d,
		type(c_ptr)	x_d,
		type(c_ptr)	alpha_d,
		integer	n
	)

Definition at line 317 of file device_mma_math.f90.

device_norm()

real(kind=rp) function, public device_mma_math::device_norm	(	type(c_ptr)	rex_d,
		integer(c_int)	n
	)

Definition at line 429 of file device_mma_math.f90.

device_prepare_aa_matrix()

subroutine, public device_mma_math::device_prepare_aa_matrix	(	type(c_ptr)	aa_d,
		type(c_ptr)	s_d,
		type(c_ptr)	lambda_d,
		type(c_ptr)	d_d,
		type(c_ptr)	mu_d,
		type(c_ptr)	y_d,
		type(c_ptr)	a_d,
		real(c_rp)	zeta,
		real(c_rp)	z,
		integer, value	m
	)

Definition at line 95 of file device_mma_math.f90.

device_prepare_hessian()

subroutine, public device_mma_math::device_prepare_hessian	(	type(c_ptr)	hess_d,
		type(c_ptr)	y_d,
		type(c_ptr)	mu_d,
		type(c_ptr)	lambda_d,
		integer, value	m
	)

Definition at line 114 of file device_mma_math.f90.

device_relambda()

subroutine, public device_mma_math::device_relambda	(	type(c_ptr)	relambda_d,
		type(c_ptr)	x_d,
		type(c_ptr)	upp_d,
		type(c_ptr)	low_d,
		type(c_ptr)	pij_d,
		type(c_ptr)	qij_d,
		integer(c_int)	n,
		integer(c_int)	m
	)

Definition at line 383 of file device_mma_math.f90.

device_rex()

subroutine, public device_mma_math::device_rex	(	type(c_ptr)	rex_d,
		type(c_ptr)	x_d,
		type(c_ptr)	low_d,
		type(c_ptr)	upp_d,
		type(c_ptr)	pij_d,
		type(c_ptr)	p0j_d,
		type(c_ptr)	qij_d,
		type(c_ptr)	q0j_d,
		type(c_ptr)	lambda_d,
		type(c_ptr)	xsi_d,
		type(c_ptr)	eta_d,
		integer(c_int)	n,
		integer(c_int)	m
	)

Definition at line 348 of file device_mma_math.f90.

device_solve_linear_system()

subroutine, public device_mma_math::device_solve_linear_system	(	type(c_ptr)	a_d,
		type(c_ptr)	b_d,
		integer(c_int), value	n,
		integer(c_int)	info
	)

Definition at line 128 of file device_mma_math.f90.

device_sub2cons2()

subroutine, public device_mma_math::device_sub2cons2	(	type(c_ptr)	rexsi_d,
		type(c_ptr)	xsi_d,
		type(c_ptr)	x_d,
		type(c_ptr)	alpha_d,
		real(kind=rp)	epsi,
		integer(c_int)	n
	)

Definition at line 397 of file device_mma_math.f90.

device_update_hessian_z()

subroutine, public device_mma_math::device_update_hessian_z	(	type(c_ptr), intent(in)	hess_d,
		type(c_ptr), intent(in)	a_d,
		integer, value	m
	)

Definition at line 77 of file device_mma_math.f90.

device_updateaa()

subroutine, public device_mma_math::device_updateaa	(	type(c_ptr)	aa_d,
		type(c_ptr)	globaltmp_mm_d,
		type(c_ptr)	s_d,
		type(c_ptr)	lambda_d,
		type(c_ptr)	d_d,
		type(c_ptr)	mu_d,
		type(c_ptr)	y_d,
		type(c_ptr)	a_d,
		real(c_rp)	zeta,
		real(c_rp)	z,
		integer(c_int)	m
	)

Definition at line 558 of file device_mma_math.f90.

device_updatebb()

subroutine, public device_mma_math::device_updatebb	(	type(c_ptr)	bb_d,
		type(c_ptr)	dellambda_d,
		type(c_ptr)	dely_d,
		type(c_ptr)	d_d,
		type(c_ptr)	mu_d,
		type(c_ptr)	y_d,
		real(c_rp)	delz,
		integer(c_int)	m
	)

Definition at line 529 of file device_mma_math.f90.