Annotation of rpl/lapack/lapack/dlamch.f, revision 1.23
1.10 bertrand 1: *> \brief \b DLAMCH
2: *
3: * =========== DOCUMENTATION ===========
4: *
1.19 bertrand 5: * Online html documentation available at
6: * http://www.netlib.org/lapack/explore-html/
1.10 bertrand 7: *
8: * Definition:
9: * ===========
10: *
11: * DOUBLE PRECISION FUNCTION DLAMCH( CMACH )
1.19 bertrand 12: *
1.22 bertrand 13: * .. Scalar Arguments ..
14: * CHARACTER CMACH
15: * ..
16: *
1.10 bertrand 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
1.22 bertrand 31: *> CMACH is CHARACTER*1
1.10 bertrand 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: *
1.19 bertrand 59: *> \author Univ. of Tennessee
60: *> \author Univ. of California Berkeley
61: *> \author Univ. of Colorado Denver
62: *> \author NAG Ltd.
1.10 bertrand 63: *
1.23 ! bertrand 64:
1.10 bertrand 65: *> \ingroup auxOTHERauxiliary
66: *
67: * =====================================================================
1.1 bertrand 68: DOUBLE PRECISION FUNCTION DLAMCH( CMACH )
69: *
1.23 ! bertrand 70: * -- LAPACK auxiliary routine --
1.7 bertrand 71: * -- LAPACK is a software package provided by Univ. of Tennessee, --
72: * -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
1.1 bertrand 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 ..
1.7 bertrand 85: DOUBLE PRECISION RND, EPS, SFMIN, SMALL, RMACH
1.1 bertrand 86: * ..
87: * .. External Functions ..
88: LOGICAL LSAME
89: EXTERNAL LSAME
90: * ..
1.7 bertrand 91: * .. Intrinsic Functions ..
92: INTRINSIC DIGITS, EPSILON, HUGE, MAXEXPONENT,
93: $ MINEXPONENT, RADIX, TINY
1.1 bertrand 94: * ..
95: * .. Executable Statements ..
96: *
1.7 bertrand 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)
1.1 bertrand 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
1.7 bertrand 122: RMACH = RADIX(ZERO)
1.1 bertrand 123: ELSE IF( LSAME( CMACH, 'P' ) ) THEN
1.7 bertrand 124: RMACH = EPS * RADIX(ZERO)
1.1 bertrand 125: ELSE IF( LSAME( CMACH, 'N' ) ) THEN
1.7 bertrand 126: RMACH = DIGITS(ZERO)
1.1 bertrand 127: ELSE IF( LSAME( CMACH, 'R' ) ) THEN
128: RMACH = RND
129: ELSE IF( LSAME( CMACH, 'M' ) ) THEN
1.7 bertrand 130: RMACH = MINEXPONENT(ZERO)
1.1 bertrand 131: ELSE IF( LSAME( CMACH, 'U' ) ) THEN
1.7 bertrand 132: RMACH = tiny(zero)
1.1 bertrand 133: ELSE IF( LSAME( CMACH, 'L' ) ) THEN
1.7 bertrand 134: RMACH = MAXEXPONENT(ZERO)
1.1 bertrand 135: ELSE IF( LSAME( CMACH, 'O' ) ) THEN
1.7 bertrand 136: RMACH = HUGE(ZERO)
137: ELSE
138: RMACH = ZERO
1.1 bertrand 139: END IF
140: *
141: DLAMCH = RMACH
142: RETURN
143: *
144: * End of DLAMCH
145: *
146: END
147: ************************************************************************
1.10 bertrand 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: *>
1.1 bertrand 168: DOUBLE PRECISION FUNCTION DLAMC3( A, B )
169: *
1.23 ! bertrand 170: * -- LAPACK auxiliary routine --
1.1 bertrand 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: ************************************************************************
CVSweb interface <joel.bertrand@systella.fr>
![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to dlamch.f CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [local] / rpl / lapack / lapack