1CLAHQR(1)           LAPACK auxiliary routine (version 3.1)           CLAHQR(1)
2
3
4

NAME

6       CLAHQR  - i an auxiliary routine called by CHSEQR to update the  eigen‐
7       values and Schur decomposition already computed by CHSEQR, by   dealing
8       with the Hessenberg submatrix in rows and columns ILO to  IHI
9

SYNOPSIS

11       SUBROUTINE CLAHQR( WANTT, WANTZ, N, ILO, IHI, H, LDH, W, ILOZ, IHIZ, Z,
12                          LDZ, INFO )
13
14           INTEGER        IHI, IHIZ, ILO, ILOZ, INFO, LDH, LDZ, N
15
16           LOGICAL        WANTT, WANTZ
17
18           COMPLEX        H( LDH, * ), W( * ), Z( LDZ, * )
19

PURPOSE

21          CLAHQR is an auxiliary routine called by CHSEQR to update the
22          eigenvalues and Schur decomposition already computed by CHSEQR, by
23          dealing with the Hessenberg submatrix in rows and columns ILO to
24          IHI.
25
26

ARGUMENTS

28       WANTT   (input) LOGICAL
29               = .TRUE. : the full Schur form T is required;
30               = .FALSE.: only eigenvalues are required.
31
32       WANTZ   (input) LOGICAL
33               = .TRUE. : the matrix of Schur vectors Z is required;
34               = .FALSE.: Schur vectors are not required.
35
36       N       (input) INTEGER
37               The order of the matrix H.  N >= 0.
38
39       ILO     (input) INTEGER
40               IHI     (input) INTEGER It is assumed that H is  already  upper
41               triangular in rows and columns IHI+1:N, and that H(ILO,ILO-1) =
42               0 (unless ILO = 1).  CLAHQR works primarily with the Hessenberg
43               submatrix in rows and columns ILO to IHI, but applies transfor‐
44               mations to  all  of  H  if  WANTT  is  .TRUE..   1  <=  ILO  <=
45               max(1,IHI); IHI <= N.
46
47       H       (input/output) COMPLEX array, dimension (LDH,N)
48               On  entry,  the upper Hessenberg matrix H.  On exit, if INFO is
49               zero and if WANTT is .TRUE., then H is upper triangular in rows
50               and  columns ILO:IHI.  If INFO is zero and if WANTT is .FALSE.,
51               then the contents of H are unspecified  on  exit.   The  output
52               state  of H in case INF is positive is below under the descrip‐
53               tion of INFO.
54
55       LDH     (input) INTEGER
56               The leading dimension of the array H. LDH >= max(1,N).
57
58       W       (output) COMPLEX array, dimension (N)
59               The computed eigenvalues ILO to IHI are stored  in  the  corre‐
60               sponding elements of W. If WANTT is .TRUE., the eigenvalues are
61               stored in the same order as on the diagonal of the  Schur  form
62               returned in H, with W(i) = H(i,i).
63
64       ILOZ    (input) INTEGER
65               IHIZ     (input)  INTEGER Specify the rows of Z to which trans‐
66               formations must be applied if WANTZ is .TRUE..  1  <=  ILOZ  <=
67               ILO; IHI <= IHIZ <= N.
68
69       Z       (input/output) COMPLEX array, dimension (LDZ,N)
70               If  WANTZ is .TRUE., on entry Z must contain the current matrix
71               Z of transformations accumulated by CHSEQR, and on exit  Z  has
72               been updated; transformations are applied only to the submatrix
73               Z(ILOZ:IHIZ,ILO:IHI).  If WANTZ is .FALSE.,  Z  is  not  refer‐
74               enced.
75
76       LDZ     (input) INTEGER
77               The leading dimension of the array Z. LDZ >= max(1,N).
78
79       INFO    (output) INTEGER
80               =   0: successful exit
81               eigenvalues  ILO  to IHI in a total of 30 iterations per eigen‐
82               value; elements i+1:ihi of W contain  those  eigenvalues  which
83               have been successfully computed.
84
85               If  INFO .GT. 0 and WANTT is .FALSE., then on exit, the remain‐
86               ing unconverged eigenvalues are the eigenvalues  of  the  upper
87               Hessenberg  matrix  rows  and  columns  ILO thorugh INFO of the
88               final, output value of H.
89
90               If  INFO  .GT.  0  and  WANTT  is  .TRUE.,  then  on  exit  (*)
91               (initial  value  of  H)*U  = U*(final value of H) where U is an
92               orthognal matrix.    The final value of H is  upper  Hessenberg
93               and triangular in rows and columns INFO+1 through IHI.
94
95               If  INFO  .GT. 0 and WANTZ is .TRUE., then on exit (final value
96               of Z)  = (initial value of  Z)*U  where  U  is  the  orthogonal
97               matrix in (*) (regardless of the value of WANTT.)
98

FURTHER DETAILS

100          02-96 Based on modifications by
101          David Day, Sandia National Laboratory, USA
102
103          12-04 Further modifications by
104          Ralph Byers, University of Kansas, USA
105
106            This is a modified version of CLAHQR from LAPACK version 3.0.
107            It is (1) more robust against overflow and underflow and
108            (2) adopts the more conservative Ahues & Tisseur stopping
109            criterion (LAWN 122, 1997).
110
111
112
113
114 LAPACK auxiliary routine (versionNo3v.e1m)ber 2006                       CLAHQR(1)
Impressum