Home Browse Files
       1
       2
       3
       4
       5
       6
       7
       8
       9
      10
      11
      12
      13
      14
      15
      16
      17
      18
      19
      20
      21
      22
      23
      24
      25
      26
      27
      28
      29
      30
      31
      32
      33
      34
      35
      36
      37
      38
      39
      40
      41
      42
      43
      44
      45
      46
      47
      48
      49
      50
      51
      52
      53
      54
      55
      56
      57
      58
      59
      60
      61
      62
      63
      64
      65
      66
      67
      68
      69
      70
      71
      72
      73
      74
      75
      76
      77
      78
      79
      80
      81
      82
      83
      84
      85
      86
      87
      88
      89
      90
      91
      92
      93
      94
      95
      96
      97
      98
      99
     100
     101
     102
     103
     104
     105
     106
     107
     108
     109
     110
     111
     112
     113
     114
     115
     116
     117
     118
     119
     120
     121
     122
      SUBROUTINE DLAS2( F, G, H, SSMIN, SSMAX )
*
*  -- LAPACK auxiliary routine (version 3.2) --
*  -- LAPACK is a software package provided by Univ. of Tennessee,    --
*  -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
*     November 2006
*
*     .. Scalar Arguments ..
      DOUBLE PRECISION   F, G, H, SSMAX, SSMIN
*     ..
*
*  Purpose
*  =======
*
*  DLAS2  computes the singular values of the 2-by-2 matrix
*     [  F   G  ]
*     [  0   H  ].
*  On return, SSMIN is the smaller singular value and SSMAX is the
*  larger singular value.
*
*  Arguments
*  =========
*
*  F       (input) DOUBLE PRECISION
*          The (1,1) element of the 2-by-2 matrix.
*
*  G       (input) DOUBLE PRECISION
*          The (1,2) element of the 2-by-2 matrix.
*
*  H       (input) DOUBLE PRECISION
*          The (2,2) element of the 2-by-2 matrix.
*
*  SSMIN   (output) DOUBLE PRECISION
*          The smaller singular value.
*
*  SSMAX   (output) DOUBLE PRECISION
*          The larger singular value.
*
*  Further Details
*  ===============
*
*  Barring over/underflow, all output quantities are correct to within
*  a few units in the last place (ulps), even in the absence of a guard
*  digit in addition/subtraction.
*
*  In IEEE arithmetic, the code works correctly if one matrix element is
*  infinite.
*
*  Overflow will not occur unless the largest singular value itself
*  overflows, or is within a few ulps of overflow. (On machines with
*  partial overflow, like the Cray, overflow may occur if the largest
*  singular value is within a factor of 2 of overflow.)
*
*  Underflow is harmless if underflow is gradual. Otherwise, results
*  may correspond to a matrix modified by perturbations of size near
*  the underflow threshold.
*
*  ====================================================================
*
*     .. Parameters ..
      DOUBLE PRECISION   ZERO
      PARAMETER          ( ZERO = 0.0D0 )
      DOUBLE PRECISION   ONE
      PARAMETER          ( ONE = 1.0D0 )
      DOUBLE PRECISION   TWO
      PARAMETER          ( TWO = 2.0D0 )
*     ..
*     .. Local Scalars ..
      DOUBLE PRECISION   AS, AT, AU, C, FA, FHMN, FHMX, GA, HA
*     ..
*     .. Intrinsic Functions ..
      INTRINSIC          ABS, MAX, MIN, SQRT
*     ..
*     .. Executable Statements ..
*
      FA = ABS( F )
      GA = ABS( G )
      HA = ABS( H )
      FHMN = MIN( FA, HA )
      FHMX = MAX( FA, HA )
      IF( FHMN.EQ.ZERO ) THEN
         SSMIN = ZERO
         IF( FHMX.EQ.ZERO ) THEN
            SSMAX = GA
         ELSE
            SSMAX = MAX( FHMX, GA )*SQRT( ONE+
     $              ( MIN( FHMX, GA ) / MAX( FHMX, GA ) )**2 )
         END IF
      ELSE
         IF( GA.LT.FHMX ) THEN
            AS = ONE + FHMN / FHMX
            AT = ( FHMX-FHMN ) / FHMX
            AU = ( GA / FHMX )**2
            C = TWO / ( SQRT( AS*AS+AU )+SQRT( AT*AT+AU ) )
            SSMIN = FHMN*C
            SSMAX = FHMX / C
         ELSE
            AU = FHMX / GA
            IF( AU.EQ.ZERO ) THEN
*
*              Avoid possible harmful underflow if exponent range
*              asymmetric (true SSMIN may not underflow even if
*              AU underflows)
*
               SSMIN = ( FHMN*FHMX ) / GA
               SSMAX = GA
            ELSE
               AS = ONE + FHMN / FHMX
               AT = ( FHMX-FHMN ) / FHMX
               C = ONE / ( SQRT( ONE+( AS*AU )**2 )+
     $             SQRT( ONE+( AT*AU )**2 ) )
               SSMIN = ( FHMN*C )*AU
               SSMIN = SSMIN + SSMIN
               SSMAX = GA / ( C+C )
            END IF
         END IF
      END IF
      RETURN
*
*     End of DLAS2
*
      END





































































































Home