cudd/cuddLinear.c File Reference

Functions for BDD and ADD reduction by linear transformations. More...

#include "util.h"
#include "cuddInt.h"
Include dependency graph for cuddLinear.c:

Defines

#define CUDD_SWAP_MOVE   0
#define CUDD_LINEAR_TRANSFORM_MOVE   1
#define CUDD_INVERSE_TRANSFORM_MOVE   2
#define BPL   32
#define LOGBPL   5

Functions

int Cudd_PrintLinear (DdManager *table)
 Prints the linear transform matrix.
int Cudd_ReadLinear (DdManager *table, int x, int y)
 Reads an entry of the linear transform matrix.
int cuddLinearAndSifting (DdManager *table, int lower, int upper)
 BDD reduction based on combination of sifting and linear transformations.
int cuddLinearInPlace (DdManager *table, int x, int y)
 Linearly combines two adjacent variables.
void cuddUpdateInteractionMatrix (DdManager *table, int xindex, int yindex)
 Updates the interaction matrix.
int cuddInitLinear (DdManager *table)
 Initializes the linear transform matrix.
int cuddResizeLinear (DdManager *table)
 Resizes the linear transform matrix.
static int ddLinearUniqueCompare (void const *ptrX, void const *ptrY)
 Comparison function used by qsort.
static int ddLinearAndSiftingAux (DdManager *table, int x, int xLow, int xHigh)
 Given xLow <= x <= xHigh moves x up and down between the boundaries.
static MoveddLinearAndSiftingUp (DdManager *table, int y, int xLow, Move *prevMoves)
 Sifts a variable up and applies linear transformations.
static MoveddLinearAndSiftingDown (DdManager *table, int x, int xHigh, Move *prevMoves)
 Sifts a variable down and applies linear transformations.
static int ddLinearAndSiftingBackward (DdManager *table, int size, Move *moves)
 Given a set of moves, returns the DD heap to the order giving the minimum size.
static MoveddUndoMoves (DdManager *table, Move *moves)
 Given a set of moves, returns the DD heap to the order in effect before the moves.
static void cuddXorLinear (DdManager *table, int x, int y)
 XORs two rows of the linear transform matrix.

Detailed Description

Functions for BDD and ADD reduction by linear transformations.

Author:
Fabio Somenzi

Copyright (c) 1995-2015, Regents of the University of Colorado

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 University of Colorado 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.


Function Documentation

int Cudd_PrintLinear ( DdManager table  ) 

Prints the linear transform matrix.

Returns:
1 in case of success; 0 otherwise.
Side effects
none
int Cudd_ReadLinear ( DdManager table,
int  x,
int  y 
)

Reads an entry of the linear transform matrix.

Side effects
none
Parameters:
table CUDD manager
x row index
y column index
int cuddInitLinear ( DdManager table  ) 

Initializes the linear transform matrix.

Returns:
1 if successful; 0 otherwise.
Side effects
none
int cuddLinearAndSifting ( DdManager table,
int  lower,
int  upper 
)

BDD reduction based on combination of sifting and linear transformations.

Assumes that no dead nodes are present.

  1. Order all the variables according to the number of entries in each unique table.
  2. Sift the variable up and down, remembering each time the total size of the DD heap. At each position, linear transformation of the two adjacent variables is tried and is accepted if it reduces the size of the DD.
  3. Select the best permutation.
  4. Repeat 3 and 4 for all variables.
Returns:
1 if successful; 0 otherwise.
Side effects
None
int cuddLinearInPlace ( DdManager table,
int  x,
int  y 
)

Linearly combines two adjacent variables.

Specifically, replaces the top variable with the exclusive nor of the two variables. It assumes that no dead nodes are present on entry to this procedure. The procedure then guarantees that no dead nodes will be present when it terminates. cuddLinearInPlace assumes that x < y.

Returns:
the number of keys in the table if successful; 0 otherwise.
Side effects
The two subtables corrresponding to variables x and y are modified. The global counters of the unique table are also affected.
See also:
cuddSwapInPlace
int cuddResizeLinear ( DdManager table  ) 

Resizes the linear transform matrix.

Returns:
1 if successful; 0 otherwise.
Side effects
none
void cuddUpdateInteractionMatrix ( DdManager table,
int  xindex,
int  yindex 
)

Updates the interaction matrix.

Side effects
none
static void cuddXorLinear ( DdManager table,
int  x,
int  y 
) [static]

XORs two rows of the linear transform matrix.

Replaces the first row with the result.

Side effects
none
static int ddLinearAndSiftingAux ( DdManager table,
int  x,
int  xLow,
int  xHigh 
) [static]

Given xLow <= x <= xHigh moves x up and down between the boundaries.

At each step a linear transformation is tried, and, if it decreases the size of the DD, it is accepted. Finds the best position and does the required changes.

Returns:
1 if successful; 0 otherwise.
Side effects
None
static int ddLinearAndSiftingBackward ( DdManager table,
int  size,
Move moves 
) [static]

Given a set of moves, returns the DD heap to the order giving the minimum size.

In case of ties, returns to the closest position giving the minimum size.

Returns:
1 in case of success; 0 otherwise.
Side effects
None
static Move* ddLinearAndSiftingDown ( DdManager table,
int  x,
int  xHigh,
Move prevMoves 
) [static]

Sifts a variable down and applies linear transformations.

Moves x down until either it reaches the bound (xHigh) or the size of the DD heap increases too much.

Returns:
the set of moves in case of success; NULL if memory is full.
Side effects
None
static Move* ddLinearAndSiftingUp ( DdManager table,
int  y,
int  xLow,
Move prevMoves 
) [static]

Sifts a variable up and applies linear transformations.

Moves y up until either it reaches the bound (xLow) or the size of the DD heap increases too much.

Returns:
the set of moves in case of success; NULL if memory is full.
Side effects
None
static int ddLinearUniqueCompare ( void const *  ptrX,
void const *  ptrY 
) [static]

Comparison function used by qsort.

Comparison function used by qsort to order the variables according to the number of keys in the subtables.

Returns:
the difference in number of keys between the two variables being compared.
Side effects
None
static Move* ddUndoMoves ( DdManager table,
Move moves 
) [static]

Given a set of moves, returns the DD heap to the order in effect before the moves.

Returns:
1 in case of success; 0 otherwise.
Side effects
None
 All Data Structures Files Functions Variables Typedefs Enumerations Defines

Generated on 31 Dec 2015 for cudd by  doxygen 1.6.1