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

NAME

6       DLARZB  -  applies  a real block reflector H or its transpose H**T to a
7       real distributed M-by-N C from the left or the right
8

SYNOPSIS

10       SUBROUTINE DLARZB( SIDE, TRANS, DIRECT, STOREV, M, N, K, L, V, LDV,  T,
11                          LDT, C, LDC, WORK, LDWORK )
12
13           CHARACTER      DIRECT, SIDE, STOREV, TRANS
14
15           INTEGER        K, L, LDC, LDT, LDV, LDWORK, M, N
16
17           DOUBLE         PRECISION  C(  LDC,  *  ), T( LDT, * ), V( LDV, * ),
18                          WORK( LDWORK, * )
19

PURPOSE

21       DLARZB applies a real block reflector H or its transpose H**T to a real
22       distributed  M-by-N   C  from  the  left or the right.  Currently, only
23       STOREV = 'R' and DIRECT = 'B' are supported.
24

ARGUMENTS

26       SIDE    (input) CHARACTER*1
27               = 'L': apply H or H' from the Left
28               = 'R': apply H or H' from the Right
29
30       TRANS   (input) CHARACTER*1
31               = 'N': apply H (No transpose)
32               = 'C': apply H' (Transpose)
33
34       DIRECT  (input) CHARACTER*1
35               Indicates how H is formed from a product of elementary  reflec‐
36               tors  =  'F':  H = H(1) H(2) . . . H(k) (Forward, not supported
37               yet)
38               = 'B': H = H(k) . . . H(2) H(1) (Backward)
39
40       STOREV  (input) CHARACTER*1
41               Indicates how the vectors which define the  elementary  reflec‐
42               tors are stored:
43               = 'C': Columnwise                        (not supported yet)
44               = 'R': Rowwise
45
46       M       (input) INTEGER
47               The number of rows of the matrix C.
48
49       N       (input) INTEGER
50               The number of columns of the matrix C.
51
52       K       (input) INTEGER
53               The  order  of the matrix T (= the number of elementary reflec‐
54               tors whose product defines the block reflector).
55
56       L       (input) INTEGER
57               The number of columns of the matrix V containing the meaningful
58               part  of  the Householder reflectors.  If SIDE = 'L', M >= L >=
59               0, if SIDE = 'R', N >= L >= 0.
60
61       V       (input) DOUBLE PRECISION array, dimension (LDV,NV).
62               If STOREV = 'C', NV = K; if STOREV = 'R', NV = L.
63
64       LDV     (input) INTEGER
65               The leading dimension of the array V.  If STOREV = 'C', LDV  >=
66               L; if STOREV = 'R', LDV >= K.
67
68       T       (input) DOUBLE PRECISION array, dimension (LDT,K)
69               The  triangular  K-by-K  matrix  T in the representation of the
70               block reflector.
71
72       LDT     (input) INTEGER
73               The leading dimension of the array T. LDT >= K.
74
75       C       (input/output) DOUBLE PRECISION array, dimension (LDC,N)
76               On entry, the M-by-N matrix C.  On exit, C  is  overwritten  by
77               H*C or H'*C or C*H or C*H'.
78
79       LDC     (input) INTEGER
80               The leading dimension of the array C. LDC >= max(1,M).
81
82       WORK    (workspace) DOUBLE PRECISION array, dimension (LDWORK,K)
83
84       LDWORK  (input) INTEGER
85               The leading dimension of the array WORK.  If SIDE = 'L', LDWORK
86               >= max(1,N); if SIDE = 'R', LDWORK >= max(1,M).
87

FURTHER DETAILS

89       Based on contributions by
90         A. Petitet, Computer Science Dept., Univ. of Tenn., Knoxville, USA
91
92
93
94 LAPACK routine (version 3.2)    November 2008                       DLARZB(1)
Impressum