LAPACK: SRC/cunghr.f Source File

Go to the documentation of this file.
00001       SUBROUTINE CUNGHR( N, ILO, IHI, A, LDA, TAU, WORK, LWORK, INFO )
00002 *
00003 *  -- LAPACK routine (version 3.2) --
00004 *  -- LAPACK is a software package provided by Univ. of Tennessee,    --
00005 *  -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
00006 *     November 2006
00007 *
00008 *     .. Scalar Arguments ..
00009       INTEGER            IHI, ILO, INFO, LDA, LWORK, N
00010 *     ..
00011 *     .. Array Arguments ..
00012       COMPLEX            A( LDA, * ), TAU( * ), WORK( * )
00013 *     ..
00014 *
00015 *  Purpose
00016 *  =======
00017 *
00018 *  CUNGHR generates a complex unitary matrix Q which is defined as the
00019 *  product of IHI-ILO elementary reflectors of order N, as returned by
00020 *  CGEHRD:
00021 *
00022 *  Q = H(ilo) H(ilo+1) . . . H(ihi-1).
00023 *
00024 *  Arguments
00025 *  =========
00026 *
00027 *  N       (input) INTEGER
00028 *          The order of the matrix Q. N >= 0.
00029 *
00030 *  ILO     (input) INTEGER
00031 *  IHI     (input) INTEGER
00032 *          ILO and IHI must have the same values as in the previous call
00033 *          of CGEHRD. Q is equal to the unit matrix except in the
00034 *          submatrix Q(ilo+1:ihi,ilo+1:ihi).
00035 *          1 <= ILO <= IHI <= N, if N > 0; ILO=1 and IHI=0, if N=0.
00036 *
00037 *  A       (input/output) COMPLEX array, dimension (LDA,N)
00038 *          On entry, the vectors which define the elementary reflectors,
00039 *          as returned by CGEHRD.
00040 *          On exit, the N-by-N unitary matrix Q.
00041 *
00042 *  LDA     (input) INTEGER
00043 *          The leading dimension of the array A. LDA >= max(1,N).
00044 *
00045 *  TAU     (input) COMPLEX array, dimension (N-1)
00046 *          TAU(i) must contain the scalar factor of the elementary
00047 *          reflector H(i), as returned by CGEHRD.
00048 *
00049 *  WORK    (workspace/output) COMPLEX array, dimension (MAX(1,LWORK))
00050 *          On exit, if INFO = 0, WORK(1) returns the optimal LWORK.
00051 *
00052 *  LWORK   (input) INTEGER
00053 *          The dimension of the array WORK. LWORK >= IHI-ILO.
00054 *          For optimum performance LWORK >= (IHI-ILO)*NB, where NB is
00055 *          the optimal blocksize.
00056 *
00057 *          If LWORK = -1, then a workspace query is assumed; the routine
00058 *          only calculates the optimal size of the WORK array, returns
00059 *          this value as the first entry of the WORK array, and no error
00060 *          message related to LWORK is issued by XERBLA.
00061 *
00062 *  INFO    (output) INTEGER
00063 *          = 0:  successful exit
00064 *          < 0:  if INFO = -i, the i-th argument had an illegal value
00065 *
00066 *  =====================================================================
00067 *
00068 *     .. Parameters ..
00069       COMPLEX            ZERO, ONE
00070       PARAMETER          ( ZERO = ( 0.0E+0, 0.0E+0 ),
00071      $                   ONE = ( 1.0E+0, 0.0E+0 ) )
00072 *     ..
00073 *     .. Local Scalars ..
00074       LOGICAL            LQUERY
00075       INTEGER            I, IINFO, J, LWKOPT, NB, NH
00076 *     ..
00077 *     .. External Subroutines ..
00078       EXTERNAL           CUNGQR, XERBLA
00079 *     ..
00080 *     .. External Functions ..
00081       INTEGER            ILAENV
00082       EXTERNAL           ILAENV
00083 *     ..
00084 *     .. Intrinsic Functions ..
00085       INTRINSIC          MAX, MIN
00086 *     ..
00087 *     .. Executable Statements ..
00088 *
00089 *     Test the input arguments
00090 *
00091       INFO = 0
00092       NH = IHI - ILO
00093       LQUERY = ( LWORK.EQ.-1 )
00094       IF( N.LT.0 ) THEN
00095          INFO = -1
00096       ELSE IF( ILO.LT.1 .OR. ILO.GT.MAX( 1, N ) ) THEN
00097          INFO = -2
00098       ELSE IF( IHI.LT.MIN( ILO, N ) .OR. IHI.GT.N ) THEN
00099          INFO = -3
00100       ELSE IF( LDA.LT.MAX( 1, N ) ) THEN
00101          INFO = -5
00102       ELSE IF( LWORK.LT.MAX( 1, NH ) .AND. .NOT.LQUERY ) THEN
00103          INFO = -8
00104       END IF
00105 *
00106       IF( INFO.EQ.0 ) THEN
00107          NB = ILAENV( 1, 'CUNGQR', ' ', NH, NH, NH, -1 )
00108          LWKOPT = MAX( 1, NH )*NB
00109          WORK( 1 ) = LWKOPT
00110       END IF
00111 *
00112       IF( INFO.NE.0 ) THEN
00113          CALL XERBLA( 'CUNGHR', -INFO )
00114          RETURN
00115       ELSE IF( LQUERY ) THEN
00116          RETURN
00117       END IF
00118 *
00119 *     Quick return if possible
00120 *
00121       IF( N.EQ.0 ) THEN
00122          WORK( 1 ) = 1
00123          RETURN
00124       END IF
00125 *
00126 *     Shift the vectors which define the elementary reflectors one
00127 *     column to the right, and set the first ilo and the last n-ihi
00128 *     rows and columns to those of the unit matrix
00129 *
00130       DO 40 J = IHI, ILO + 1, -1
00131          DO 10 I = 1, J - 1
00132             A( I, J ) = ZERO
00133    10    CONTINUE
00134          DO 20 I = J + 1, IHI
00135             A( I, J ) = A( I, J-1 )
00136    20    CONTINUE
00137          DO 30 I = IHI + 1, N
00138             A( I, J ) = ZERO
00139    30    CONTINUE
00140    40 CONTINUE
00141       DO 60 J = 1, ILO
00142          DO 50 I = 1, N
00143             A( I, J ) = ZERO
00144    50    CONTINUE
00145          A( J, J ) = ONE
00146    60 CONTINUE
00147       DO 80 J = IHI + 1, N
00148          DO 70 I = 1, N
00149             A( I, J ) = ZERO
00150    70    CONTINUE
00151          A( J, J ) = ONE
00152    80 CONTINUE
00153 *
00154       IF( NH.GT.0 ) THEN
00155 *
00156 *        Generate Q(ilo+1:ihi,ilo+1:ihi)
00157 *
00158          CALL CUNGQR( NH, NH, NH, A( ILO+1, ILO+1 ), LDA, TAU( ILO ),
00159      $                WORK, LWORK, IINFO )
00160       END IF
00161       WORK( 1 ) = LWKOPT
00162       RETURN
00163 *
00164 *     End of CUNGHR
00165 *
00166       END