Google
 

Trailing-Edge - PDP-10 Archives - decuslib10-02 - 43,50145/factr.doc
There are 2 other files named factr.doc in the archive. Click here to see a list.
SUBROUTINE FACTR

PURPOSE
   FACTORIZATION OF THE MATRIX A INTO A PRODUCT OF A LOWER
   TRIANGULAR MATRIX L AND AN UPPER TRIANGULAR MATRIX U.  L HAS
   UNIT DIAGONAL WHICH IS NOT STORED.

USAGE
   CALL FACTR(A,PER,N,IA,IER)

DESCRIPTION OF PARAMETERS
   A	  MATRIX A
   PER	  ONE DIMENSIONAL ARRAY WHERE PERMUTATIONS OF ROWS OF
	  THE MATRIX ARE STORED
	  DIMENSION OF PER MUST BE GREATER THAN OR EQUAL TO N
   N	  ORDER OF THE MATRIX A
   IA	  SIZE OF THE FIRST DIMENSION ASSIGNED TO THE ARRAY A
	  IN THE CALLING PROGRAM WHEN THE MATRIX IS IN DOUBLE
	  SUBSCRIPTED DATA STORAGE MODE.  IA=N WHEN THE MATRIX
	  IS IN SSP VECTOR STORAGE MODE.
   IER	  ERROR INDICATOR WHICH IS ZERO IF THERE IS NO ERROR,
	  AND IS THREE IF THE PROCEDURE FAILS.

REMARKS
   THE ORIGINAL MATRIX, A,IS REPLACED BY THE TRIANGULAR FACTORS

SUBROUTINES AND FUNCTION SUBPROGRAMS REQUIRED
   NONE

METHOD
   SUCCESSIVE COMPUTATION OF THE COLUMNS OF L AND THE
   CORRESPONDING ROWS OF U.

REFERENCES
   J. H. WILKINSON - THE ALGEBRAIC EIGENVALUE PROBLEM -
   CLARENDON PRESS, OXFORD, 1965. H. J. BOWDLER, R. S. MARTIN,
   G. PETERS, AND J. H. WILKINSON - 'SOLUTION OF REAL AND
   COMPLEX SYSTEMS OF LINEAR EQUATIONS', NUMERISCHE MATHEMATIK,
   VOL. 8, NO. 3, 1966, P. 217-234.