LAPACK
3.4.2
LAPACK: Linear Algebra PACKage
Main Page
Modules
Files
File List
File Members
All
Files
Functions
Groups
xlaenv.f
Go to the documentation of this file.
1
*> \brief \b XLAENV
2
*
3
* =========== DOCUMENTATION ===========
4
*
5
* Online html documentation available at
6
* http://www.netlib.org/lapack/explore-html/
7
*
8
* Definition:
9
* ===========
10
*
11
* SUBROUTINE XLAENV( ISPEC, NVALUE )
12
*
13
* .. Scalar Arguments ..
14
* INTEGER ISPEC, NVALUE
15
* ..
16
*
17
*
18
*> \par Purpose:
19
* =============
20
*>
21
*> \verbatim
22
*>
23
*> XLAENV sets certain machine- and problem-dependent quantities
24
*> which will later be retrieved by ILAENV.
25
*> \endverbatim
26
*
27
* Arguments:
28
* ==========
29
*
30
*> \param[in] ISPEC
31
*> \verbatim
32
*> ISPEC is INTEGER
33
*> Specifies the parameter to be set in the COMMON array IPARMS.
34
*> = 1: the optimal blocksize; if this value is 1, an unblocked
35
*> algorithm will give the best performance.
36
*> = 2: the minimum block size for which the block routine
37
*> should be used; if the usable block size is less than
38
*> this value, an unblocked routine should be used.
39
*> = 3: the crossover point (in a block routine, for N less
40
*> than this value, an unblocked routine should be used)
41
*> = 4: the number of shifts, used in the nonsymmetric
42
*> eigenvalue routines
43
*> = 5: the minimum column dimension for blocking to be used;
44
*> rectangular blocks must have dimension at least k by m,
45
*> where k is given by ILAENV(2,...) and m by ILAENV(5,...)
46
*> = 6: the crossover point for the SVD (when reducing an m by n
47
*> matrix to bidiagonal form, if max(m,n)/min(m,n) exceeds
48
*> this value, a QR factorization is used first to reduce
49
*> the matrix to a triangular form)
50
*> = 7: the number of processors
51
*> = 8: another crossover point, for the multishift QR and QZ
52
*> methods for nonsymmetric eigenvalue problems.
53
*> = 9: maximum size of the subproblems at the bottom of the
54
*> computation tree in the divide-and-conquer algorithm
55
*> (used by xGELSD and xGESDD)
56
*> =10: ieee NaN arithmetic can be trusted not to trap
57
*> =11: infinity arithmetic can be trusted not to trap
58
*> 12 <= ISPEC <= 16:
59
*> xHSEQR or one of its subroutines,
60
*> see IPARMQ for detailed explanation
61
*> \endverbatim
62
*>
63
*> \param[in] NVALUE
64
*> \verbatim
65
*> NVALUE is INTEGER
66
*> The value of the parameter specified by ISPEC.
67
*> \endverbatim
68
*
69
* Authors:
70
* ========
71
*
72
*> \author Univ. of Tennessee
73
*> \author Univ. of California Berkeley
74
*> \author Univ. of Colorado Denver
75
*> \author NAG Ltd.
76
*
77
*> \date November 2011
78
*
79
*> \ingroup aux_eig
80
*
81
* =====================================================================
82
SUBROUTINE
xlaenv
( ISPEC, NVALUE )
83
*
84
* -- LAPACK test routine (version 3.4.0) --
85
* -- LAPACK is a software package provided by Univ. of Tennessee, --
86
* -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
87
* November 2011
88
*
89
* .. Scalar Arguments ..
90
INTEGER
ispec, nvalue
91
* ..
92
*
93
* =====================================================================
94
*
95
* .. Arrays in Common ..
96
INTEGER
iparms( 100 )
97
* ..
98
* .. Common blocks ..
99
common / claenv / iparms
100
* ..
101
* .. Save statement ..
102
SAVE
/ claenv /
103
* ..
104
* .. Executable Statements ..
105
*
106
IF
( ispec.GE.1 .AND. ispec.LE.16 )
THEN
107
iparms( ispec ) = nvalue
108
END IF
109
*
110
return
111
*
112
* End of XLAENV
113
*
114
END
TESTING
EIG
xlaenv.f
Generated on Tue Sep 25 2012 16:29:02 for LAPACK by
1.8.1.1