solver_interface.h File Reference

Detailed Description

Interface and wrapper to matrix/factorization (ladel) functions.

Author

Ben Hermans

This file includes all calls to ladel functions apart from scaling in scaling.c and memory allocation/deallocation in the main functions in qpalm.c. It includes all matrix operations, such as matrix vector products, row- and columnwise norms, cholesky factorizations, factorization updates and solving the linear system.

Definition in file solver_interface.h.

#include <qpalm/global_opts.h>
#include <qpalm/constants.h>
#include <qpalm/types.h>
#include <ladel.h>

Include dependency graph for solver_interface.h:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Macros
#define	mat_inf_norm_cols   ladel_infinity_norm_columns
#define	mat_inf_norm_rows   ladel_infinity_norm_rows

Functions
void	mat_vec (solver_sparse *A, solver_dense *x, solver_dense *y, solver_common *c)
	Matrix-vector multiplication.
void	mat_tpose_vec (solver_sparse *A, solver_dense *x, solver_dense *y, solver_common *c)
	Matrix-transpose-vector multiplication.
void	qpalm_set_factorization_method (QPALMWorkspace *work, solver_common *c)
	Choose the linear systems solver method based on the problem data sizes.
void	qpalm_form_kkt (QPALMWorkspace *work)
	Form the KKT system $\left[\begin{array}{cc}Q& A^T\\ A& -{\mathrm{\Sigma }}^{-1}\end{array}\right]$.
void	qpalm_reform_kkt (QPALMWorkspace *work)
	Reform the KKT system (i.e.
void	kkt_update_entering_constraints (QPALMWorkspace *work, solver_common *c)
	Perform a factorization update for the entering constraints.
void	kkt_update_leaving_constraints (QPALMWorkspace *work, solver_common *c)
	Perform a factorization update for the leaving constraints.
void	kkt_solve (QPALMWorkspace *work, solver_common *c)
	Solve the KKT system $\left[\begin{array}{cc}Q+\frac{1}{\gamma }I& A^T\\ A& -{\mathrm{\Sigma }}^{-1}\end{array}\right]\left[\begin{array}{c}d\\ -\lambda \end{array}\right]=\left[\begin{array}{c}-\mathrm{\nabla }\phi \\ 0\end{array}\right]$.
void	ldlchol (solver_sparse *M, QPALMWorkspace *work, solver_common *c)
	Calculate $LD{L}^T$ factorization of a matrix $M$.
void	ldlcholQAtsigmaA (QPALMWorkspace *work, solver_common *c)
	Calculate $LD{L}^T$ factorization of $Q+A{(a,:)}^T\ast \mathrm{\Sigma }(a,a)\ast A(a,:)$, with $\mathrm{\Sigma }=diag(\sigma )$ and $a$ the set of active constraints.
void	ldlupdate_entering_constraints (QPALMWorkspace *work, solver_common *c)
	Update the $LD{L}^T$ factorization given a set of entering constraints.
void	ldldowndate_leaving_constraints (QPALMWorkspace *work, solver_common *c)
	Downdate the $LD{L}^T$ factorization given a set of leaving constraints.
void	ldlupdate_sigma_changed (QPALMWorkspace *work, solver_common *c)
	Update the $LD{L}^T$ factorization given a set of indexes where $sigma$ has been updated.
void	ldlsolveLD_neg_dphi (QPALMWorkspace *work, solver_common *c)
	Solve the linear system $LD{L}^T\ast d=-\mathrm{\nabla }\phi$.

Macro Definition Documentation

mat_inf_norm_cols

#define mat_inf_norm_cols   ladel_infinity_norm_columns

Definition at line 66 of file solver_interface.h.

mat_inf_norm_rows

#define mat_inf_norm_rows   ladel_infinity_norm_rows

Definition at line 67 of file solver_interface.h.

Function Documentation

mat_vec()

void mat_vec	(	solver_sparse *	A,
		solver_dense *	x,
		solver_dense *	y,
		solver_common *	c
	)

Matrix-vector multiplication.

$y=A\ast x$

Parameters

A	Sparse matrix
x	Dense input vector
y	Dense output vector
c	Solver environment

Definition at line 73 of file solver_interface.c.

Here is the caller graph for this function:

mat_tpose_vec()

void mat_tpose_vec	(	solver_sparse *	A,
		solver_dense *	x,
		solver_dense *	y,
		solver_common *	c
	)

Matrix-transpose-vector multiplication.

$y=A^T\ast x$

Parameters

A	Sparse matrix
x	Dense input vector
y	Dense output vector
c	Solver environment

Definition at line 93 of file solver_interface.c.

Here is the caller graph for this function:

qpalm_set_factorization_method()

void qpalm_set_factorization_method	(	QPALMWorkspace *	work,
		solver_common *	c
	)

Choose the linear systems solver method based on the problem data sizes.

This chooses between forming and solving the KKT system or the SCHUR complement. The resulting method is in work->solver->factorization_method.

Parameters

work	Workspace
c	Linear systems solver environment

Definition at line 18 of file solver_interface.c.

Here is the caller graph for this function:

qpalm_form_kkt()

void qpalm_form_kkt

(

QPALMWorkspace *

work

)

Form the KKT system $\left[\begin{array}{cc}Q& A^T\\ A& -{\mathrm{\Sigma }}^{-1}\end{array}\right]$.

The result is in work->solver->kkt.

Note

Only the rows of A corresponding to active constraints are included in the system above. This routine also saves some structures that are useful in updating the kkt system later.

Parameters

work

Workspace

Definition at line 114 of file solver_interface.c.

Here is the caller graph for this function:

qpalm_reform_kkt()

void qpalm_reform_kkt

(

QPALMWorkspace *

work

)

Reform the KKT system (i.e.

delete constraints which are no longer active and add those that are now active).

Parameters

work

Workspace

Definition at line 172 of file solver_interface.c.

Here is the caller graph for this function:

kkt_update_entering_constraints()

void kkt_update_entering_constraints	(	QPALMWorkspace *	work,
		solver_common *	c
	)

Perform a factorization update for the entering constraints.

Parameters

work	Workspace
c	Linear systems solver environment

Definition at line 197 of file solver_interface.c.

Here is the caller graph for this function:

kkt_update_leaving_constraints()

void kkt_update_leaving_constraints	(	QPALMWorkspace *	work,
		solver_common *	c
	)

Perform a factorization update for the leaving constraints.

Parameters

work	Workspace
c	Linear systems solver environment

Definition at line 215 of file solver_interface.c.

Here is the caller graph for this function:

kkt_solve()

void kkt_solve	(	QPALMWorkspace *	work,
		solver_common *	c
	)

Solve the KKT system $\left[\begin{array}{cc}Q+\frac{1}{\gamma }I& A^T\\ A& -{\mathrm{\Sigma }}^{-1}\end{array}\right]\left[\begin{array}{c}d\\ -\lambda \end{array}\right]=\left[\begin{array}{c}-\mathrm{\nabla }\phi \\ 0\end{array}\right]$.

Parameters

work	Workspace
c	Linear systems solver environment

Definition at line 233 of file solver_interface.c.

Here is the call graph for this function:

Here is the caller graph for this function:

ldlchol()

void ldlchol	(	solver_sparse *	M,
		QPALMWorkspace *	work,
		solver_common *	c
	)

Calculate $LD{L}^T$ factorization of a matrix $M$.

If work->settings->proximal = true, use $M+\frac{1}{\gamma }\ast I$ instead.

Parameters

M	Matrix to be factorized
work	Workspace
c	Solver environment

Definition at line 246 of file solver_interface.c.

Here is the caller graph for this function:

ldlcholQAtsigmaA()

void ldlcholQAtsigmaA	(	QPALMWorkspace *	work,
		solver_common *	c
	)

Calculate $LD{L}^T$ factorization of $Q+A{(a,:)}^T\ast \mathrm{\Sigma }(a,a)\ast A(a,:)$, with $\mathrm{\Sigma }=diag(\sigma )$ and $a$ the set of active constraints.

If work->settings->proximal = true, use $Q+\frac{1}{\gamma }\ast I+A{(a,:)}^T\ast \mathrm{\Sigma }(a,a)\ast A(a,:)$ instead.

Parameters

work	Workspace
c	Solver environment

Definition at line 274 of file solver_interface.c.

Here is the call graph for this function:

Here is the caller graph for this function:

ldlupdate_entering_constraints()

void ldlupdate_entering_constraints	(	QPALMWorkspace *	work,
		solver_common *	c
	)

Update the $LD{L}^T$ factorization given a set of entering constraints.

The index set of entering constraints is assumed to be set in work->solver->enter.

Parameters

work	Workspace
c	Solver environment

Definition at line 297 of file solver_interface.c.

Here is the caller graph for this function:

ldldowndate_leaving_constraints()

void ldldowndate_leaving_constraints	(	QPALMWorkspace *	work,
		solver_common *	c
	)

Downdate the $LD{L}^T$ factorization given a set of leaving constraints.

The index set of leaving constraints is assumed to be set in work->solver->leave.

Parameters

work	Workspace
c	Solver environment

Definition at line 306 of file solver_interface.c.

Here is the caller graph for this function:

ldlupdate_sigma_changed()

void ldlupdate_sigma_changed	(	QPALMWorkspace *	work,
		solver_common *	c
	)

Update the $LD{L}^T$ factorization given a set of indexes where $sigma$ has been updated.

The index set of changed $sigma$ is assumed to be set in work->solver->enter.

Parameters

work	Workspace
c	Solver environment

Definition at line 315 of file solver_interface.c.

Here is the caller graph for this function:

ldlsolveLD_neg_dphi()

void ldlsolveLD_neg_dphi	(	QPALMWorkspace *	work,
		solver_common *	c
	)

Solve the linear system $LD{L}^T\ast d=-\mathrm{\nabla }\phi$.

Parameters

work	Workspace
c	Solver environment

Definition at line 359 of file solver_interface.c.

Here is the call graph for this function:

Here is the caller graph for this function: