1DSTEIN(1)                LAPACK routine (version 3.1)                DSTEIN(1)
2
3
4

NAME

6       DSTEIN - the eigenvectors of a real symmetric tridiagonal matrix T cor‐
7       responding to specified eigenvalues, using inverse iteration
8

SYNOPSIS

10       SUBROUTINE DSTEIN( N, D, E, M, W, IBLOCK, ISPLIT, Z, LDZ, WORK,  IWORK,
11                          IFAIL, INFO )
12
13           INTEGER        INFO, LDZ, M, N
14
15           INTEGER        IBLOCK( * ), IFAIL( * ), ISPLIT( * ), IWORK( * )
16
17           DOUBLE         PRECISION D( * ), E( * ), W( * ), WORK( * ), Z( LDZ,
18                          * )
19

PURPOSE

21       DSTEIN computes the eigenvectors of a real symmetric tridiagonal matrix
22       T corresponding to specified eigenvalues, using inverse iteration.
23
24       The maximum number of iterations allowed for each eigenvector is speci‐
25       fied by an internal parameter MAXITS (currently set to 5).
26
27

ARGUMENTS

29       N       (input) INTEGER
30               The order of the matrix.  N >= 0.
31
32       D       (input) DOUBLE PRECISION array, dimension (N)
33               The n diagonal elements of the tridiagonal matrix T.
34
35       E       (input) DOUBLE PRECISION array, dimension (N-1)
36               The (n-1) subdiagonal elements of the tridiagonal matrix T,  in
37               elements 1 to N-1.
38
39       M       (input) INTEGER
40               The number of eigenvectors to be found.  0 <= M <= N.
41
42       W       (input) DOUBLE PRECISION array, dimension (N)
43               The  first  M  elements  of W contain the eigenvalues for which
44               eigenvectors are to be computed.   The  eigenvalues  should  be
45               grouped by split-off block and ordered from smallest to largest
46               within the block.  ( The output array W from DSTEBZ with  ORDER
47               = 'B' is expected here. )
48
49       IBLOCK  (input) INTEGER array, dimension (N)
50               The  submatrix indices associated with the corresponding eigen‐
51               values in W; IBLOCK(i)=1 if  eigenvalue  W(i)  belongs  to  the
52               first  submatrix from the top, =2 if W(i) belongs to the second
53               submatrix, etc.  ( The  output  array  IBLOCK  from  DSTEBZ  is
54               expected here. )
55
56       ISPLIT  (input) INTEGER array, dimension (N)
57               The  splitting  points,  at which T breaks up into submatrices.
58               The first submatrix consists of rows/columns 1 to ISPLIT( 1  ),
59               the  second  of rows/columns ISPLIT( 1 )+1 through ISPLIT( 2 ),
60               etc.  ( The output array ISPLIT from DSTEBZ is expected here. )
61
62       Z       (output) DOUBLE PRECISION array, dimension (LDZ, M)
63               The computed eigenvectors.  The eigenvector associated with the
64               eigenvalue  W(i) is stored in the i-th column of Z.  Any vector
65               which fails to converge is set to  its  current  iterate  after
66               MAXITS iterations.
67
68       LDZ     (input) INTEGER
69               The leading dimension of the array Z.  LDZ >= max(1,N).
70
71       WORK    (workspace) DOUBLE PRECISION array, dimension (5*N)
72
73       IWORK   (workspace) INTEGER array, dimension (N)
74
75       IFAIL   (output) INTEGER array, dimension (M)
76               On normal exit, all elements of IFAIL are zero.  If one or more
77               eigenvectors fail to converge  after  MAXITS  iterations,  then
78               their indices are stored in array IFAIL.
79
80       INFO    (output) INTEGER
81               = 0: successful exit.
82               < 0: if INFO = -i, the i-th argument had an illegal value
83               >  0:  if  INFO  = i, then i eigenvectors failed to converge in
84               MAXITS iterations.  Their indices are stored in array IFAIL.
85

PARAMETERS

87       MAXITS  INTEGER, default = 5
88               The maximum number of iterations performed.
89
90       EXTRA   INTEGER, default = 2
91               The number of iterations performed after norm growth  criterion
92               is satisfied, should be at least 1.
93
94
95
96 LAPACK routine (version 3.1)    November 2006                       DSTEIN(1)
Impressum