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
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
|
/*
* Copyright (c) 2011, Michael Lehn
*
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* 1) Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2) Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in
* the documentation and/or other materials provided with the
* distribution.
* 3) Neither the name of the FLENS development group nor the names of
* its contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
/* Based on
*
SUBROUTINE DLASCL( TYPE, KL, KU, CFROM, CTO, M, N, A, LDA, INFO )
*
* -- LAPACK auxiliary routine (version 3.3.0) --
* -- LAPACK is a software package provided by Univ. of Tennessee, --
* -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
* November 2010
*/
#ifndef FLENS_LAPACK_LA_LASCL_TCC
#define FLENS_LAPACK_LA_LASCL_TCC 1
#include <cxxstd/cmath.h>
#include <flens/lapack/lapack.h>
#include <flens/matrixtypes/matrixtypes.h>
#include <flens/vectortypes/vectortypes.h>
////////
////////
////////
////////
////////
////////
////////
////////
//////// TODO: Completely redesign this crap!
////////
////////
////////
//////// ... but it works.
////////
////////
////////
////////
////////
namespace flens { namespace lapack {
//== generic lapack implementation =============================================
namespace generic {
template <typename Int, typename T, typename MA>
typename RestrictTo<IsSame<typename MA::ElementType, T>::value
|| IsSame<typename ComplexTrait<typename MA::ElementType
>::PrimitiveType, T>::value,
void>::Type
lascl_impl(LASCL::Type type,
Int kl,
Int ku,
const T &cFrom,
const T &cTo,
MA &A)
{
using namespace LASCL;
typedef typename MA::IndexType IndexType;
using std::abs;
using std::isnan;
using std::max;
using std::min;
const IndexType m = A.numRows();
const IndexType n = A.numCols();
const T Zero = 0;
const T One = 1;
if ((cFrom==Zero) || isnan(cFrom)) {
ASSERT(0);
} else if (isnan(cTo)) {
ASSERT(0);
} else if (type==SymmetricLowerBand) {
ASSERT(A.numRows()==A.numCols());
} else if (type==SymmetricUpperBand) {
ASSERT(A.numRows()==A.numCols());
}
//
// Quick return if possible
//
if ((n==0) || (m==0)) {
return;
}
//
// Get machine parameters
//
const T smallNum = lamch<T>(SafeMin);
const T bigNum = T(1) / smallNum;
//
// Make copies of cFrom, cTo
//
T cFromC = cFrom;
T cToC = cTo;
bool done = false;
do {
T cFrom1 = cFromC*smallNum;
T cTo1;
T mul;
if (cFrom1==cFromC) {
// cFromC is an inf. Multiply by a correctly signed zero for
// finite cToC, or a NaN if cToC is infinite.
mul = cToC / cFromC;
done = true;
cTo1 = cToC;
} else {
cTo1 = cToC / bigNum;
if (cTo1==cToC) {
// cToC is either 0 or an inf. In both cases, cToC itself
// serves as the correct multiplication factor.
mul = cToC;
done = true;
cFromC = One;
} else if (abs(cFrom1)>abs(cToC) && cToC!=Zero) {
mul = smallNum;
done = false;
cFromC = cFrom1;
} else if (abs(cTo1)>abs(cFromC)) {
mul = bigNum;
done = false;
cToC = cTo1;
} else {
mul = cToC / cFromC;
done = true;
}
}
if (type==FullMatrix) {
//
// Full matrix
//
for (IndexType j=1; j<=n; ++j) {
for (IndexType i=1; i<=m; ++i) {
A(i,j) *= mul;
}
}
} else if (type==LowerTriangular) {
//
// Lower triangular matrix
//
for (IndexType j=1; j<=n; ++j) {
for (IndexType i=j; i<=m; ++i) {
A(i,j) *= mul;
}
}
} else if (type==UpperTriangular) {
//
// Upper triangular matrix
//
for (IndexType j=1; j<=n; ++j) {
for (IndexType i=1; i<=min(j,m); ++i) {
A(i,j) *= mul;
}
}
} else if (type==UpperHessenberg) {
//
// Upper Hessenberg matrix
//
for (IndexType j=1; j<=n; ++j) {
for (IndexType i=1; i<=min(j+1,m); ++i) {
A(i,j) *= mul;
}
}
} else if (type==SymmetricLowerBand) {
//
// Lower half of a symmetric band matrix
//
// TODO: this only works for the internal fullstorage of a
// band matrix not for the external element access
// of SbMatrix
const IndexType k3 = kl + 1;
const IndexType k4 = n + 1;
for (IndexType j=1; j<=n; ++j) {
for (IndexType i=1; i<=min(k3, k4-j); ++i) {
A(i,j) *= mul;
}
}
} else if (type==SymmetricUpperBand) {
//
// Upper half of a symmetric band matrix
//
// TODO: this only works for the internal fullstorage of a
// band matrix not for the external element access
// of SbMatrix
const IndexType k1 = ku + 2;
const IndexType k3 = ku + 1;
for (IndexType j=1; j<=n; ++j) {
for (IndexType i=max(k1-j,IndexType(1)); i<=k3; ++i) {
A(i,j) *= mul;
}
}
} else if (type==GeneralBand) {
//
// Band matrix
//
// TODO: this only works for the internal fullstorage of a
// band matrix not for the external element access
// of GeMatrix
const IndexType k1 = kl + ku + 2;
const IndexType k2 = kl + 1;
const IndexType k3 = 2*kl + ku + 1;
const IndexType k4 = kl + ku + 1 + m;
for (IndexType j=1; j<=n; ++j) {
for (IndexType i=max(k1-j,k2); i<=min(k3,k4-j); ++i) {
A(i,j) *= mul;
}
}
} else {
ASSERT(0);
}
} while (!done);
}
} // namespace generic
//== interface for native lapack ===============================================
#ifdef USE_CXXLAPACK
namespace external {
template <typename Int, typename T, typename MA>
void
lascl_impl(LASCL::Type type,
Int kl,
Int ku,
const T &cFrom,
const T &cTo,
MA &A)
{
typedef typename MA::IndexType IndexType;
cxxlapack::lascl<IndexType>(getF77Char(type),
kl,
ku,
cFrom,
cTo,
A.numRows(),
A.numCols(),
A.data(),
A.leadingDimension());
}
} // namespace external
#endif // USE_CXXLAPACK
//== public interface ==========================================================
template <typename Int, typename T, typename MA>
typename RestrictTo<IsMatrix<MA>::value,
void>::Type
lascl(LASCL::Type type, Int kl, Int ku, const T &cFrom, const T &cTo, MA &&A)
{
LAPACK_DEBUG_OUT("lascl");
# ifdef CHECK_CXXLAPACK
typedef typename RemoveRef<MA>::Type MatrixA;
typename MatrixA::NoView A_ = A;
# endif
LAPACK_SELECT::lascl_impl(type, kl, ku, cFrom, cTo, A);
# ifdef CHECK_CXXLAPACK
external::lascl_impl(type, kl, ku, cFrom, cTo, A_);
if (! isIdentical(A, A_, " A", "A_")) {
std::cerr << "CXXLAPACK: A = " << A << std::endl;
std::cerr << "F77LAPACK: A_ = " << A_ << std::endl;
ASSERT(0);
}
# endif
}
//-- convert vector to matrix --------------------------------------------------
template <typename Int, typename T, typename VX>
typename RestrictTo<IsDenseVector<VX>::value,
void>::Type
lascl(LASCL::Type type, Int kl, Int ku, const T &cFrom, const T &cTo, VX &&x)
{
using namespace LASCL;
typedef typename RemoveRef<VX>::Type VectorX;
typedef typename VectorX::ElementType TX;
typedef FullStorage<TX, ColMajor> FS;
typename GeMatrix<FS>::View A(x.length(), 1, x);
ASSERT(type==FullMatrix);
lascl(type, kl, ku, cFrom, cTo, A);
}
//-- convert scalar to matrix --------------------------------------------------
template <typename Int, typename T, typename ALPHA>
typename RestrictTo<!IsMatrix<ALPHA>::value
&& !IsVector<ALPHA>::value,
void>::Type
lascl(LASCL::Type type, Int kl, Int ku, const T &cFrom, const T &cTo,
ALPHA &&alpha)
{
using namespace LASCL;
typedef typename RemoveRef<ALPHA>::Type Alpha;
typedef typename GeMatrixView<Alpha>::Engine ViewEngine;
GeMatrixView<Alpha> A = ViewEngine(1, 1, 1, &alpha);
ASSERT(type==FullMatrix);
lascl(type, kl, ku, cFrom, cTo, A);
}
} } // namespace lapack, flens
#endif // FLENS_LAPACK_LA_LASCL_TCC
|