SYNOPSIS

Functions/Subroutines

subroutine dlaed6 (KNITER, ORGATI, RHO, D, Z, FINIT, TAU, INFO)

DLAED6 used by sstedc. Computes one Newton step in solution of the secular equation.

Function/Subroutine Documentation

subroutine dlaed6 (integerKNITER, logicalORGATI, double precisionRHO, double precision, dimension( 3 )D, double precision, dimension( 3 )Z, double precisionFINIT, double precisionTAU, integerINFO)

DLAED6 used by sstedc. Computes one Newton step in solution of the secular equation.

Purpose:

 DLAED6 computes the positive or negative root (closest to the origin)
 of
                  z(1)        z(2)        z(3)
 f(x) =   rho + --------- + ---------- + ---------
                 d(1)-x      d(2)-x      d(3)-x

 It is assumed that

       if ORGATI = .true. the root is between d(2) and d(3);
       otherwise it is between d(1) and d(2)

 This routine will be called by DLAED4 when necessary. In most cases,
 the root sought is the smallest in magnitude, though it might not be
 in some extremely rare situations.

Parameters:

KNITER

          KNITER is INTEGER
               Refer to DLAED4 for its significance.

ORGATI

          ORGATI is LOGICAL
               If ORGATI is true, the needed root is between d(2) and
               d(3); otherwise it is between d(1) and d(2).  See
               DLAED4 for further details.

RHO

          RHO is DOUBLE PRECISION
               Refer to the equation f(x) above.

D

          D is DOUBLE PRECISION array, dimension (3)
               D satisfies d(1) < d(2) < d(3).

Z

          Z is DOUBLE PRECISION array, dimension (3)
               Each of the elements in z must be positive.

FINIT

          FINIT is DOUBLE PRECISION
               The value of f at 0. It is more accurate than the one
               evaluated inside this routine (if someone wants to do
               so).

TAU

          TAU is DOUBLE PRECISION
               The root of the equation f(x).

INFO

          INFO is INTEGER
               = 0: successful exit
               > 0: if INFO = 1, failure to converge

Author:

Univ. of Tennessee

Univ. of California Berkeley

Univ. of Colorado Denver

NAG Ltd.

Date:

September 2012

Further Details:

  10/02/03: This version has a few statements commented out for thread
  safety (machine parameters are computed on each entry). SJH.

  05/10/06: Modified from a new version of Ren-Cang Li, use
     Gragg-Thornton-Warner cubic convergent scheme for better stability.

Contributors:

Ren-Cang Li, Computer Science Division, University of California at Berkeley, USA

Definition at line 141 of file dlaed6.f.

Author

Generated automatically by Doxygen for LAPACK from the source code.