1CHETD2(1)                LAPACK routine (version 3.2)                CHETD2(1)
2
3
4

NAME

6       CHETD2  - reduces a complex Hermitian matrix A to real symmetric tridi‐
7       agonal form T by a unitary similarity transformation
8

SYNOPSIS

10       SUBROUTINE CHETD2( UPLO, N, A, LDA, D, E, TAU, INFO )
11
12           CHARACTER      UPLO
13
14           INTEGER        INFO, LDA, N
15
16           REAL           D( * ), E( * )
17
18           COMPLEX        A( LDA, * ), TAU( * )
19

PURPOSE

21       CHETD2 reduces a complex Hermitian matrix A to real symmetric tridiago‐
22       nal form T by a unitary similarity transformation: Q' * A * Q = T.
23

ARGUMENTS

25       UPLO    (input) CHARACTER*1
26               Specifies  whether  the  upper  or lower triangular part of the
27               Hermitian matrix A is stored:
28               = 'U':  Upper triangular
29               = 'L':  Lower triangular
30
31       N       (input) INTEGER
32               The order of the matrix A.  N >= 0.
33
34       A       (input/output) COMPLEX array, dimension (LDA,N)
35               On entry, the Hermitian matrix A.  If UPLO = 'U',  the  leading
36               n-by-n upper triangular part of A contains the upper triangular
37               part of the matrix A, and the strictly lower triangular part of
38               A  is  not referenced.  If UPLO = 'L', the leading n-by-n lower
39               triangular part of A contains the lower triangular part of  the
40               matrix  A,  and  the strictly upper triangular part of A is not
41               referenced.  On exit, if UPLO = 'U',  the  diagonal  and  first
42               superdiagonal  of  A  are overwritten by the corresponding ele‐
43               ments of the tridiagonal matrix T, and the elements  above  the
44               first  superdiagonal, with the array TAU, represent the unitary
45               matrix Q as a product of elementary reflectors; if UPLO =  'L',
46               the  diagonal  and  first subdiagonal of A are over- written by
47               the corresponding elements of the tridiagonal matrix T, and the
48               elements  below the first subdiagonal, with the array TAU, rep‐
49               resent the unitary matrix Q as a product of elementary  reflec‐
50               tors. See Further Details.  LDA     (input) INTEGER The leading
51               dimension of the array A.  LDA >= max(1,N).
52
53       D       (output) REAL array, dimension (N)
54               The diagonal elements of  the  tridiagonal  matrix  T:  D(i)  =
55               A(i,i).
56
57       E       (output) REAL array, dimension (N-1)
58               The  off-diagonal  elements of the tridiagonal matrix T: E(i) =
59               A(i,i+1) if UPLO = 'U', E(i) = A(i+1,i) if UPLO = 'L'.
60
61       TAU     (output) COMPLEX array, dimension (N-1)
62               The scalar factors of the elementary  reflectors  (see  Further
63               Details).
64
65       INFO    (output) INTEGER
66               = 0:  successful exit
67               < 0:  if INFO = -i, the i-th argument had an illegal value.
68

FURTHER DETAILS

70       If  UPLO  = 'U', the matrix Q is represented as a product of elementary
71       reflectors
72          Q = H(n-1) . . . H(2) H(1).
73       Each H(i) has the form
74          H(i) = I - tau * v * v'
75       where tau is a complex scalar, and v is a complex vector with  v(i+1:n)
76       = 0 and v(i) = 1; v(1:i-1) is stored on exit in
77       A(1:i-1,i+1), and tau in TAU(i).
78       If  UPLO  = 'L', the matrix Q is represented as a product of elementary
79       reflectors
80          Q = H(1) H(2) . . . H(n-1).
81       Each H(i) has the form
82          H(i) = I - tau * v * v'
83       where tau is a complex scalar, and v is a complex vector with v(1:i)  =
84       0  and v(i+1) = 1; v(i+2:n) is stored on exit in A(i+2:n,i), and tau in
85       TAU(i).
86       The contents of A on exit are illustrated  by  the  following  examples
87       with n = 5:
88       if UPLO = 'U':                       if UPLO = 'L':
89         (  d   e   v2  v3  v4 )              (  d                  )
90         (      d   e   v3  v4 )              (  e   d              )
91         (          d   e   v4 )              (  v1  e   d          )
92         (              d   e  )              (  v1  v2  e   d      )
93         (                   d  )              (  v1  v2  v3  e   d  ) where d
94       and e denote diagonal and off-diagonal elements of T, and vi denotes an
95       element of the vector defining H(i).
96
97
98
99 LAPACK routine (version 3.2)    November 2008                       CHETD2(1)
Impressum