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
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
|
*> \brief \b DLAEDA
*
* =========== DOCUMENTATION ===========
*
* Online html documentation available at
* http://www.netlib.org/lapack/explore-html/
*
*> Download DLAEDA + dependencies
*> <a href="http://www.netlib.org/cgi-bin/netlibfiles.tgz?format=tgz&filename=/lapack/lapack_routine/dlaeda.f">
*> [TGZ]</a>
*> <a href="http://www.netlib.org/cgi-bin/netlibfiles.zip?format=zip&filename=/lapack/lapack_routine/dlaeda.f">
*> [ZIP]</a>
*> <a href="http://www.netlib.org/cgi-bin/netlibfiles.txt?format=txt&filename=/lapack/lapack_routine/dlaeda.f">
*> [TXT]</a>
*
* Definition
* ==========
*
* SUBROUTINE DLAEDA( N, TLVLS, CURLVL, CURPBM, PRMPTR, PERM, GIVPTR,
* GIVCOL, GIVNUM, Q, QPTR, Z, ZTEMP, INFO )
*
* .. Scalar Arguments ..
* INTEGER CURLVL, CURPBM, INFO, N, TLVLS
* ..
* .. Array Arguments ..
* INTEGER GIVCOL( 2, * ), GIVPTR( * ), PERM( * ),
* $ PRMPTR( * ), QPTR( * )
* DOUBLE PRECISION GIVNUM( 2, * ), Q( * ), Z( * ), ZTEMP( * )
* ..
*
* Purpose
* =======
*
*>\details \b Purpose:
*>\verbatim
*>
*> DLAEDA computes the Z vector corresponding to the merge step in the
*> CURLVLth step of the merge process with TLVLS steps for the CURPBMth
*> problem.
*>
*>\endverbatim
*
* Arguments
* =========
*
*> \param[in] N
*> \verbatim
*> N is INTEGER
*> The dimension of the symmetric tridiagonal matrix. N >= 0.
*> \endverbatim
*>
*> \param[in] TLVLS
*> \verbatim
*> TLVLS is INTEGER
*> The total number of merging levels in the overall divide and
*> conquer tree.
*> \endverbatim
*>
*> \param[in] CURLVL
*> \verbatim
*> CURLVL is INTEGER
*> The current level in the overall merge routine,
*> 0 <= curlvl <= tlvls.
*> \endverbatim
*>
*> \param[in] CURPBM
*> \verbatim
*> CURPBM is INTEGER
*> The current problem in the current level in the overall
*> merge routine (counting from upper left to lower right).
*> \endverbatim
*>
*> \param[in] PRMPTR
*> \verbatim
*> PRMPTR is INTEGER array, dimension (N lg N)
*> Contains a list of pointers which indicate where in PERM a
*> level's permutation is stored. PRMPTR(i+1) - PRMPTR(i)
*> indicates the size of the permutation and incidentally the
*> size of the full, non-deflated problem.
*> \endverbatim
*>
*> \param[in] PERM
*> \verbatim
*> PERM is INTEGER array, dimension (N lg N)
*> Contains the permutations (from deflation and sorting) to be
*> applied to each eigenblock.
*> \endverbatim
*>
*> \param[in] GIVPTR
*> \verbatim
*> GIVPTR is INTEGER array, dimension (N lg N)
*> Contains a list of pointers which indicate where in GIVCOL a
*> level's Givens rotations are stored. GIVPTR(i+1) - GIVPTR(i)
*> indicates the number of Givens rotations.
*> \endverbatim
*>
*> \param[in] GIVCOL
*> \verbatim
*> GIVCOL is INTEGER array, dimension (2, N lg N)
*> Each pair of numbers indicates a pair of columns to take place
*> in a Givens rotation.
*> \endverbatim
*>
*> \param[in] GIVNUM
*> \verbatim
*> GIVNUM is DOUBLE PRECISION array, dimension (2, N lg N)
*> Each number indicates the S value to be used in the
*> corresponding Givens rotation.
*> \endverbatim
*>
*> \param[in] Q
*> \verbatim
*> Q is DOUBLE PRECISION array, dimension (N**2)
*> Contains the square eigenblocks from previous levels, the
*> starting positions for blocks are given by QPTR.
*> \endverbatim
*>
*> \param[in] QPTR
*> \verbatim
*> QPTR is INTEGER array, dimension (N+2)
*> Contains a list of pointers which indicate where in Q an
*> eigenblock is stored. SQRT( QPTR(i+1) - QPTR(i) ) indicates
*> the size of the block.
*> \endverbatim
*>
*> \param[out] Z
*> \verbatim
*> Z is DOUBLE PRECISION array, dimension (N)
*> On output this vector contains the updating vector (the last
*> row of the first sub-eigenvector matrix and the first row of
*> the second sub-eigenvector matrix).
*> \endverbatim
*>
*> \param[out] ZTEMP
*> \verbatim
*> ZTEMP is DOUBLE PRECISION array, dimension (N)
*> \endverbatim
*>
*> \param[out] INFO
*> \verbatim
*> INFO is INTEGER
*> = 0: successful exit.
*> < 0: if INFO = -i, the i-th argument had an illegal value.
*> \endverbatim
*>
*
* Authors
* =======
*
*> \author Univ. of Tennessee
*> \author Univ. of California Berkeley
*> \author Univ. of Colorado Denver
*> \author NAG Ltd.
*
*> \date November 2011
*
*> \ingroup auxOTHERcomputational
*
*
* Further Details
* ===============
*>\details \b Further \b Details
*> \verbatim
*>
*> Based on contributions by
*> Jeff Rutter, Computer Science Division, University of California
*> at Berkeley, USA
*>
*> \endverbatim
*>
* =====================================================================
SUBROUTINE DLAEDA( N, TLVLS, CURLVL, CURPBM, PRMPTR, PERM, GIVPTR,
$ GIVCOL, GIVNUM, Q, QPTR, Z, ZTEMP, INFO )
*
* -- LAPACK computational routine (version 3.2.2) --
* -- LAPACK is a software package provided by Univ. of Tennessee, --
* -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
* November 2011
*
* .. Scalar Arguments ..
INTEGER CURLVL, CURPBM, INFO, N, TLVLS
* ..
* .. Array Arguments ..
INTEGER GIVCOL( 2, * ), GIVPTR( * ), PERM( * ),
$ PRMPTR( * ), QPTR( * )
DOUBLE PRECISION GIVNUM( 2, * ), Q( * ), Z( * ), ZTEMP( * )
* ..
*
* =====================================================================
*
* .. Parameters ..
DOUBLE PRECISION ZERO, HALF, ONE
PARAMETER ( ZERO = 0.0D0, HALF = 0.5D0, ONE = 1.0D0 )
* ..
* .. Local Scalars ..
INTEGER BSIZ1, BSIZ2, CURR, I, K, MID, PSIZ1, PSIZ2,
$ PTR, ZPTR1
* ..
* .. External Subroutines ..
EXTERNAL DCOPY, DGEMV, DROT, XERBLA
* ..
* .. Intrinsic Functions ..
INTRINSIC DBLE, INT, SQRT
* ..
* .. Executable Statements ..
*
* Test the input parameters.
*
INFO = 0
*
IF( N.LT.0 ) THEN
INFO = -1
END IF
IF( INFO.NE.0 ) THEN
CALL XERBLA( 'DLAEDA', -INFO )
RETURN
END IF
*
* Quick return if possible
*
IF( N.EQ.0 )
$ RETURN
*
* Determine location of first number in second half.
*
MID = N / 2 + 1
*
* Gather last/first rows of appropriate eigenblocks into center of Z
*
PTR = 1
*
* Determine location of lowest level subproblem in the full storage
* scheme
*
CURR = PTR + CURPBM*2**CURLVL + 2**( CURLVL-1 ) - 1
*
* Determine size of these matrices. We add HALF to the value of
* the SQRT in case the machine underestimates one of these square
* roots.
*
BSIZ1 = INT( HALF+SQRT( DBLE( QPTR( CURR+1 )-QPTR( CURR ) ) ) )
BSIZ2 = INT( HALF+SQRT( DBLE( QPTR( CURR+2 )-QPTR( CURR+1 ) ) ) )
DO 10 K = 1, MID - BSIZ1 - 1
Z( K ) = ZERO
10 CONTINUE
CALL DCOPY( BSIZ1, Q( QPTR( CURR )+BSIZ1-1 ), BSIZ1,
$ Z( MID-BSIZ1 ), 1 )
CALL DCOPY( BSIZ2, Q( QPTR( CURR+1 ) ), BSIZ2, Z( MID ), 1 )
DO 20 K = MID + BSIZ2, N
Z( K ) = ZERO
20 CONTINUE
*
* Loop through remaining levels 1 -> CURLVL applying the Givens
* rotations and permutation and then multiplying the center matrices
* against the current Z.
*
PTR = 2**TLVLS + 1
DO 70 K = 1, CURLVL - 1
CURR = PTR + CURPBM*2**( CURLVL-K ) + 2**( CURLVL-K-1 ) - 1
PSIZ1 = PRMPTR( CURR+1 ) - PRMPTR( CURR )
PSIZ2 = PRMPTR( CURR+2 ) - PRMPTR( CURR+1 )
ZPTR1 = MID - PSIZ1
*
* Apply Givens at CURR and CURR+1
*
DO 30 I = GIVPTR( CURR ), GIVPTR( CURR+1 ) - 1
CALL DROT( 1, Z( ZPTR1+GIVCOL( 1, I )-1 ), 1,
$ Z( ZPTR1+GIVCOL( 2, I )-1 ), 1, GIVNUM( 1, I ),
$ GIVNUM( 2, I ) )
30 CONTINUE
DO 40 I = GIVPTR( CURR+1 ), GIVPTR( CURR+2 ) - 1
CALL DROT( 1, Z( MID-1+GIVCOL( 1, I ) ), 1,
$ Z( MID-1+GIVCOL( 2, I ) ), 1, GIVNUM( 1, I ),
$ GIVNUM( 2, I ) )
40 CONTINUE
PSIZ1 = PRMPTR( CURR+1 ) - PRMPTR( CURR )
PSIZ2 = PRMPTR( CURR+2 ) - PRMPTR( CURR+1 )
DO 50 I = 0, PSIZ1 - 1
ZTEMP( I+1 ) = Z( ZPTR1+PERM( PRMPTR( CURR )+I )-1 )
50 CONTINUE
DO 60 I = 0, PSIZ2 - 1
ZTEMP( PSIZ1+I+1 ) = Z( MID+PERM( PRMPTR( CURR+1 )+I )-1 )
60 CONTINUE
*
* Multiply Blocks at CURR and CURR+1
*
* Determine size of these matrices. We add HALF to the value of
* the SQRT in case the machine underestimates one of these
* square roots.
*
BSIZ1 = INT( HALF+SQRT( DBLE( QPTR( CURR+1 )-QPTR( CURR ) ) ) )
BSIZ2 = INT( HALF+SQRT( DBLE( QPTR( CURR+2 )-QPTR( CURR+
$ 1 ) ) ) )
IF( BSIZ1.GT.0 ) THEN
CALL DGEMV( 'T', BSIZ1, BSIZ1, ONE, Q( QPTR( CURR ) ),
$ BSIZ1, ZTEMP( 1 ), 1, ZERO, Z( ZPTR1 ), 1 )
END IF
CALL DCOPY( PSIZ1-BSIZ1, ZTEMP( BSIZ1+1 ), 1, Z( ZPTR1+BSIZ1 ),
$ 1 )
IF( BSIZ2.GT.0 ) THEN
CALL DGEMV( 'T', BSIZ2, BSIZ2, ONE, Q( QPTR( CURR+1 ) ),
$ BSIZ2, ZTEMP( PSIZ1+1 ), 1, ZERO, Z( MID ), 1 )
END IF
CALL DCOPY( PSIZ2-BSIZ2, ZTEMP( PSIZ1+BSIZ2+1 ), 1,
$ Z( MID+BSIZ2 ), 1 )
*
PTR = PTR + 2**( TLVLS-K )
70 CONTINUE
*
RETURN
*
* End of DLAEDA
*
END
|
