Return to dlamch.f CVS log

Up to [local] / rpl / lapack / lapack

Première mise à jour de lapack et blas.

    1: *> \brief \b DLAMCH
    2: *
    3: *  =========== DOCUMENTATION ===========
    4: *
    5: * Online html documentation available at
    6: *            http://www.netlib.org/lapack/explore-html/
    7: *
    8: *  Definition:
    9: *  ===========
   10: *
   11: *      DOUBLE PRECISION FUNCTION DLAMCH( CMACH )
   12: *
   13: *     .. Scalar Arguments ..
   14: *     CHARACTER          CMACH
   15: *     ..
   16: *
   17: *
   18: *> \par Purpose:
   19: *  =============
   20: *>
   21: *> \verbatim
   22: *>
   23: *> DLAMCH determines double precision machine parameters.
   24: *> \endverbatim
   25: *
   26: *  Arguments:
   27: *  ==========
   28: *
   29: *> \param[in] CMACH
   30: *> \verbatim
   31: *>          CMACH is CHARACTER*1
   32: *>          Specifies the value to be returned by DLAMCH:
   33: *>          = 'E' or 'e',   DLAMCH := eps
   34: *>          = 'S' or 's ,   DLAMCH := sfmin
   35: *>          = 'B' or 'b',   DLAMCH := base
   36: *>          = 'P' or 'p',   DLAMCH := eps*base
   37: *>          = 'N' or 'n',   DLAMCH := t
   38: *>          = 'R' or 'r',   DLAMCH := rnd
   39: *>          = 'M' or 'm',   DLAMCH := emin
   40: *>          = 'U' or 'u',   DLAMCH := rmin
   41: *>          = 'L' or 'l',   DLAMCH := emax
   42: *>          = 'O' or 'o',   DLAMCH := rmax
   43: *>          where
   44: *>          eps   = relative machine precision
   45: *>          sfmin = safe minimum, such that 1/sfmin does not overflow
   46: *>          base  = base of the machine
   47: *>          prec  = eps*base
   48: *>          t     = number of (base) digits in the mantissa
   49: *>          rnd   = 1.0 when rounding occurs in addition, 0.0 otherwise
   50: *>          emin  = minimum exponent before (gradual) underflow
   51: *>          rmin  = underflow threshold - base**(emin-1)
   52: *>          emax  = largest exponent before overflow
   53: *>          rmax  = overflow threshold  - (base**emax)*(1-eps)
   54: *> \endverbatim
   55: *
   56: *  Authors:
   57: *  ========
   58: *
   59: *> \author Univ. of Tennessee
   60: *> \author Univ. of California Berkeley
   61: *> \author Univ. of Colorado Denver
   62: *> \author NAG Ltd.
   63: *
   64: 
   65: *> \ingroup auxOTHERauxiliary
   66: *
   67: *  =====================================================================
   68:       DOUBLE PRECISION FUNCTION DLAMCH( CMACH )
   69: *
   70: *  -- LAPACK auxiliary routine --
   71: *  -- LAPACK is a software package provided by Univ. of Tennessee,    --
   72: *  -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
   73: *
   74: *     .. Scalar Arguments ..
   75:       CHARACTER          CMACH
   76: *     ..
   77: *
   78: * =====================================================================
   79: *
   80: *     .. Parameters ..
   81:       DOUBLE PRECISION   ONE, ZERO
   82:       PARAMETER          ( ONE = 1.0D+0, ZERO = 0.0D+0 )
   83: *     ..
   84: *     .. Local Scalars ..
   85:       DOUBLE PRECISION   RND, EPS, SFMIN, SMALL, RMACH
   86: *     ..
   87: *     .. External Functions ..
   88:       LOGICAL            LSAME
   89:       EXTERNAL           LSAME
   90: *     ..
   91: *     .. Intrinsic Functions ..
   92:       INTRINSIC          DIGITS, EPSILON, HUGE, MAXEXPONENT,
   93:      $                   MINEXPONENT, RADIX, TINY
   94: *     ..
   95: *     .. Executable Statements ..
   96: *
   97: *
   98: *     Assume rounding, not chopping. Always.
   99: *
  100:       RND = ONE
  101: *
  102:       IF( ONE.EQ.RND ) THEN
  103:          EPS = EPSILON(ZERO) * 0.5
  104:       ELSE
  105:          EPS = EPSILON(ZERO)
  106:       END IF
  107: *
  108:       IF( LSAME( CMACH, 'E' ) ) THEN
  109:          RMACH = EPS
  110:       ELSE IF( LSAME( CMACH, 'S' ) ) THEN
  111:          SFMIN = TINY(ZERO)
  112:          SMALL = ONE / HUGE(ZERO)
  113:          IF( SMALL.GE.SFMIN ) THEN
  114: *
  115: *           Use SMALL plus a bit, to avoid the possibility of rounding
  116: *           causing overflow when computing  1/sfmin.
  117: *
  118:             SFMIN = SMALL*( ONE+EPS )
  119:          END IF
  120:          RMACH = SFMIN
  121:       ELSE IF( LSAME( CMACH, 'B' ) ) THEN
  122:          RMACH = RADIX(ZERO)
  123:       ELSE IF( LSAME( CMACH, 'P' ) ) THEN
  124:          RMACH = EPS * RADIX(ZERO)
  125:       ELSE IF( LSAME( CMACH, 'N' ) ) THEN
  126:          RMACH = DIGITS(ZERO)
  127:       ELSE IF( LSAME( CMACH, 'R' ) ) THEN
  128:          RMACH = RND
  129:       ELSE IF( LSAME( CMACH, 'M' ) ) THEN
  130:          RMACH = MINEXPONENT(ZERO)
  131:       ELSE IF( LSAME( CMACH, 'U' ) ) THEN
  132:          RMACH = tiny(zero)
  133:       ELSE IF( LSAME( CMACH, 'L' ) ) THEN
  134:          RMACH = MAXEXPONENT(ZERO)
  135:       ELSE IF( LSAME( CMACH, 'O' ) ) THEN
  136:          RMACH = HUGE(ZERO)
  137:       ELSE
  138:          RMACH = ZERO
  139:       END IF
  140: *
  141:       DLAMCH = RMACH
  142:       RETURN
  143: *
  144: *     End of DLAMCH
  145: *
  146:       END
  147: ************************************************************************
  148: *> \brief \b DLAMC3
  149: *> \details
  150: *> \b Purpose:
  151: *> \verbatim
  152: *> DLAMC3  is intended to force  A  and  B  to be stored prior to doing
  153: *> the addition of  A  and  B ,  for use in situations where optimizers
  154: *> might hold one of these in a register.
  155: *> \endverbatim
  156: *> \author LAPACK is a software package provided by Univ. of Tennessee, Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..
  157: *> \param[in] A
  158: *> \verbatim
  159: *>          A is a DOUBLE PRECISION
  160: *> \endverbatim
  161: *>
  162: *> \param[in] B
  163: *> \verbatim
  164: *>          B is a DOUBLE PRECISION
  165: *>          The values A and B.
  166: *> \endverbatim
  167: *>
  168:       DOUBLE PRECISION FUNCTION DLAMC3( A, B )
  169: *
  170: *  -- LAPACK auxiliary routine --
  171: *     Univ. of Tennessee, Univ. of California Berkeley and NAG Ltd..
  172: *
  173: *     .. Scalar Arguments ..
  174:       DOUBLE PRECISION   A, B
  175: *     ..
  176: * =====================================================================
  177: *
  178: *     .. Executable Statements ..
  179: *
  180:       DLAMC3 = A + B
  181: *
  182:       RETURN
  183: *
  184: *     End of DLAMC3
  185: *
  186:       END
  187: *
  188: ************************************************************************