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
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
|
/******************************************************************************
* SOFA, Simulation Open-Framework Architecture, version 1.0 beta 4 *
* (c) 2006-2009 MGH, INRIA, USTL, UJF, CNRS *
* *
* This library is free software; you can redistribute it and/or modify it *
* under the terms of the GNU Lesser General Public License as published by *
* the Free Software Foundation; either version 2.1 of the License, or (at *
* your option) any later version. *
* *
* This library is distributed in the hope that it will be useful, but WITHOUT *
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or *
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License *
* for more details. *
* *
* You should have received a copy of the GNU Lesser General Public License *
* along with this library; if not, write to the Free Software Foundation, *
* Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. *
*******************************************************************************
* SOFA :: Modules *
* *
* Authors: The SOFA Team and external contributors (see Authors.txt) *
* *
* Contact information: contact@sofa-framework.org *
******************************************************************************/
#ifndef SOFA_GPU_CUDA_CUDASPRINGFORCEFIELD_INL
#define SOFA_GPU_CUDA_CUDASPRINGFORCEFIELD_INL
#include "CudaSpringForceField.h"
#include <sofa/component/forcefield/SpringForceField.inl>
#include <sofa/component/forcefield/StiffSpringForceField.inl>
#include <sofa/component/forcefield/MeshSpringForceField.inl>
#include <sofa/component/forcefield/TriangleBendingSprings.inl>
#include <sofa/component/forcefield/QuadBendingSprings.inl>
namespace sofa
{
namespace gpu
{
namespace cuda
{
extern "C"
{
void SpringForceFieldCuda3f_addForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* x, const void* v);
void SpringForceFieldCuda3f_addExternalForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* x1, const void* v1, const void* x2, const void* v2);
void StiffSpringForceFieldCuda3f_addForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* x, const void* v, void* dfdx);
void StiffSpringForceFieldCuda3f_addExternalForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* x1, const void* v1, const void* x2, const void* v2, void* dfdx);
void StiffSpringForceFieldCuda3f_addDForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* dx, const void* x, const void* dfdx, double factor);
void StiffSpringForceFieldCuda3f_addExternalDForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* dx1, const void* x1, const void* dx2, const void* x2, const void* dfdx, double factor);
void SpringForceFieldCuda3f1_addForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* x, const void* v);
void SpringForceFieldCuda3f1_addExternalForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* x1, const void* v1, const void* x2, const void* v2);
void StiffSpringForceFieldCuda3f1_addForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* x, const void* v, void* dfdx);
void StiffSpringForceFieldCuda3f1_addExternalForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* x1, const void* v1, const void* x2, const void* v2, void* dfdx);
void StiffSpringForceFieldCuda3f1_addDForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* dx, const void* x, const void* dfdx, double factor);
void StiffSpringForceFieldCuda3f1_addExternalDForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* dx1, const void* x1, const void* dx2, const void* x2, const void* dfdx, double factor);
#ifdef SOFA_GPU_CUDA_DOUBLE
void SpringForceFieldCuda3d_addForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* x, const void* v);
void SpringForceFieldCuda3d_addExternalForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* x1, const void* v1, const void* x2, const void* v2);
void StiffSpringForceFieldCuda3d_addForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* x, const void* v, void* dfdx);
void StiffSpringForceFieldCuda3d_addExternalForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* x1, const void* v1, const void* x2, const void* v2, void* dfdx);
void StiffSpringForceFieldCuda3d_addDForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* dx, const void* x, const void* dfdx, double factor);
void StiffSpringForceFieldCuda3d_addExternalDForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* dx1, const void* x1, const void* dx2, const void* x2, const void* dfdx, double factor);
void SpringForceFieldCuda3d1_addForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* x, const void* v);
void SpringForceFieldCuda3d1_addExternalForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* x1, const void* v1, const void* x2, const void* v2);
void StiffSpringForceFieldCuda3d1_addForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* x, const void* v, void* dfdx);
void StiffSpringForceFieldCuda3d1_addExternalForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* x1, const void* v1, const void* x2, const void* v2, void* dfdx);
void StiffSpringForceFieldCuda3d1_addDForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* dx, const void* x, const void* dfdx, double factor);
void StiffSpringForceFieldCuda3d1_addExternalDForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* dx1, const void* x1, const void* dx2, const void* x2, const void* dfdx, double factor);
#endif // SOFA_GPU_CUDA_DOUBLE
} // extern "C"
template<>
class CudaKernelsSpringForceField<CudaVec3fTypes>
{
public:
static void addForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* x, const void* v)
{ SpringForceFieldCuda3f_addForce(nbVertex, nbSpringPerVertex, springs, f, x, v); }
static void addExternalForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* x1, const void* v1, const void* x2, const void* v2)
{ SpringForceFieldCuda3f_addExternalForce(nbVertex, nbSpringPerVertex, springs, f1, x1, v1, x2, v2); }
static void addForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* x, const void* v, void* dfdx)
{ StiffSpringForceFieldCuda3f_addForce(nbVertex, nbSpringPerVertex, springs, f, x, v, dfdx); }
static void addExternalForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* x1, const void* v1, const void* x2, const void* v2, void* dfdx)
{ StiffSpringForceFieldCuda3f_addExternalForce(nbVertex, nbSpringPerVertex, springs, f1, x1, v1, x2, v2, dfdx); }
static void addDForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* dx, const void* x, const void* dfdx, double factor)
{ StiffSpringForceFieldCuda3f_addDForce(nbVertex, nbSpringPerVertex, springs, f, dx, x, dfdx, factor); }
static void addExternalDForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* dx1, const void* x1, const void* dx2, const void* x2, const void* dfdx, double factor)
{ StiffSpringForceFieldCuda3f_addExternalDForce(nbVertex, nbSpringPerVertex, springs, f1, dx1, x1, dx2, x2, dfdx, factor); }
};
template<>
class CudaKernelsSpringForceField<CudaVec3f1Types>
{
public:
static void addForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* x, const void* v)
{ SpringForceFieldCuda3f1_addForce(nbVertex, nbSpringPerVertex, springs, f, x, v); }
static void addExternalForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* x1, const void* v1, const void* x2, const void* v2)
{ SpringForceFieldCuda3f1_addExternalForce(nbVertex, nbSpringPerVertex, springs, f1, x1, v1, x2, v2); }
static void addForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* x, const void* v, void* dfdx)
{ StiffSpringForceFieldCuda3f1_addForce(nbVertex, nbSpringPerVertex, springs, f, x, v, dfdx); }
static void addExternalForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* x1, const void* v1, const void* x2, const void* v2, void* dfdx)
{ StiffSpringForceFieldCuda3f1_addExternalForce(nbVertex, nbSpringPerVertex, springs, f1, x1, v1, x2, v2, dfdx); }
static void addDForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* dx, const void* x, const void* dfdx, double factor)
{ StiffSpringForceFieldCuda3f1_addDForce(nbVertex, nbSpringPerVertex, springs, f, dx, x, dfdx, factor); }
static void addExternalDForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* dx1, const void* x1, const void* dx2, const void* x2, const void* dfdx, double factor)
{ StiffSpringForceFieldCuda3f1_addExternalDForce(nbVertex, nbSpringPerVertex, springs, f1, dx1, x1, dx2, x2, dfdx, factor); }
};
#ifdef SOFA_GPU_CUDA_DOUBLE
template<>
class CudaKernelsSpringForceField<CudaVec3dTypes>
{
public:
static void addForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* x, const void* v)
{ SpringForceFieldCuda3d_addForce(nbVertex, nbSpringPerVertex, springs, f, x, v); }
static void addExternalForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* x1, const void* v1, const void* x2, const void* v2)
{ SpringForceFieldCuda3d_addExternalForce(nbVertex, nbSpringPerVertex, springs, f1, x1, v1, x2, v2); }
static void addForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* x, const void* v, void* dfdx)
{ StiffSpringForceFieldCuda3d_addForce(nbVertex, nbSpringPerVertex, springs, f, x, v, dfdx); }
static void addExternalForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* x1, const void* v1, const void* x2, const void* v2, void* dfdx)
{ StiffSpringForceFieldCuda3d_addExternalForce(nbVertex, nbSpringPerVertex, springs, f1, x1, v1, x2, v2, dfdx); }
static void addDForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* dx, const void* x, const void* dfdx, double factor)
{ StiffSpringForceFieldCuda3d_addDForce(nbVertex, nbSpringPerVertex, springs, f, dx, x, dfdx, factor); }
static void addExternalDForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* dx1, const void* x1, const void* dx2, const void* x2, const void* dfdx, double factor)
{ StiffSpringForceFieldCuda3d_addExternalDForce(nbVertex, nbSpringPerVertex, springs, f1, dx1, x1, dx2, x2, dfdx, factor); }
};
template<>
class CudaKernelsSpringForceField<CudaVec3d1Types>
{
public:
static void addForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* x, const void* v)
{ SpringForceFieldCuda3d1_addForce(nbVertex, nbSpringPerVertex, springs, f, x, v); }
static void addExternalForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* x1, const void* v1, const void* x2, const void* v2)
{ SpringForceFieldCuda3d1_addExternalForce(nbVertex, nbSpringPerVertex, springs, f1, x1, v1, x2, v2); }
static void addForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* x, const void* v, void* dfdx)
{ StiffSpringForceFieldCuda3d1_addForce(nbVertex, nbSpringPerVertex, springs, f, x, v, dfdx); }
static void addExternalForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* x1, const void* v1, const void* x2, const void* v2, void* dfdx)
{ StiffSpringForceFieldCuda3d1_addExternalForce(nbVertex, nbSpringPerVertex, springs, f1, x1, v1, x2, v2, dfdx); }
static void addDForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f, const void* dx, const void* x, const void* dfdx, double factor)
{ StiffSpringForceFieldCuda3d1_addDForce(nbVertex, nbSpringPerVertex, springs, f, dx, x, dfdx, factor); }
static void addExternalDForce(unsigned int nbVertex, unsigned int nbSpringPerVertex, const void* springs, void* f1, const void* dx1, const void* x1, const void* dx2, const void* x2, const void* dfdx, double factor)
{ StiffSpringForceFieldCuda3d1_addExternalDForce(nbVertex, nbSpringPerVertex, springs, f1, dx1, x1, dx2, x2, dfdx, factor); }
};
#endif // SOFA_GPU_CUDA_DOUBLE
} // namespace cuda
} // namespace gpu
namespace component
{
namespace forcefield
{
using namespace gpu::cuda;
template<class TCoord, class TDeriv, class TReal>
void SpringForceFieldInternalData< gpu::cuda::CudaVectorTypes<TCoord,TDeriv,TReal> >::init(Main* m, bool stiff)
{
Data& data = m->data;
m->Inherit::init();
const sofa::helper::vector<Spring>& springs = m->springs.getValue();
if (!springs.empty())
{
bool external = (m->mstate1!=m->mstate2);
if (external)
{
std::map<int,int> nsprings1;
std::map<int,int> nsprings2;
for (unsigned int i=0;i<springs.size();i++)
{
nsprings1[springs[i].m1]++;
nsprings2[springs[i].m2]++;
}
int nmax1 = 0;
for (std::map<int,int>::const_iterator it = nsprings1.begin(); it != nsprings1.end(); ++it)
if (it->second > nmax1)
nmax1 = it->second;
data.springs1.init(nsprings1.begin()->first, nsprings1.rbegin()->first - nsprings1.begin()->first + 1, nmax1);
int nmax2 = 0;
for (std::map<int,int>::const_iterator it = nsprings2.begin(); it != nsprings2.end(); ++it)
if (it->second > nmax2)
nmax2 = it->second;
data.springs2.init(nsprings2.begin()->first, nsprings2.rbegin()->first - nsprings2.begin()->first + 1, nmax2);
nsprings1.clear();
nsprings2.clear();
for (unsigned int i=0;i<springs.size();i++)
{
int m1 = springs[i].m1 - data.springs1.vertex0;
int m2 = springs[i].m2 - data.springs2.vertex0;
data.springs1.set(m1, nsprings1[m1]++, m2,
(float)springs[i].initpos,
(float)springs[i].ks,
(float)springs[i].kd);
data.springs2.set(m2, nsprings2[m2]++, m1,
(float)springs[i].initpos,
(float)springs[i].ks,
(float)springs[i].kd);
}
}
else
{
std::map<int,int> nsprings;
for (unsigned int i=0;i<springs.size();i++)
{
nsprings[springs[i].m1]++;
nsprings[springs[i].m2]++;
}
int nmax = 0;
for (std::map<int,int>::const_iterator it = nsprings.begin(); it != nsprings.end(); ++it)
if (it->second > nmax)
nmax = it->second;
data.springs1.init(nsprings.begin()->first, nsprings.rbegin()->first - nsprings.begin()->first + 1, nmax);
std::cout << "CUDA SpringForceField: "<<springs.size()<<" springs, "<<data.springs1.nbVertex<<" attached points, max "<<data.springs1.nbSpringPerVertex<<" springs per point."<<std::endl;
nsprings.clear();
for (unsigned int i=0;i<springs.size();i++)
{
int m1 = springs[i].m1 - data.springs1.vertex0;
int m2 = springs[i].m2 - data.springs1.vertex0;
data.springs1.set(m1, nsprings[m1]++, m2,
(float)springs[i].initpos,
(float)springs[i].ks,
(float)springs[i].kd);
data.springs1.set(m2, nsprings[m2]++, m1,
(float)springs[i].initpos,
(float)springs[i].ks,
(float)springs[i].kd);
}
}
}
if (stiff)
{
data.springs1.dfdx.resize(data.springs1.springs.size());
data.springs2.dfdx.resize(data.springs2.springs.size());
}
}
// -- InteractionForceField interface
template<class TCoord, class TDeriv, class TReal>
void SpringForceFieldInternalData< gpu::cuda::CudaVectorTypes<TCoord,TDeriv,TReal> >::addForce(Main* m, bool stiff, VecDeriv& f1, VecDeriv& f2, const VecCoord& x1, const VecCoord& x2, const VecDeriv& v1, const VecDeriv& v2)
{
Data& data = m->data;
if (m->mstate1 == m->mstate2)
{
VecDeriv& f = f1;
const VecCoord& x = x1;
const VecDeriv& v = v1;
f.resize(x.size());
int d = data.springs1.vertex0;
if (data.springs1.nbSpringPerVertex > 0)
{
if (!stiff)
Kernels::addForce(data.springs1.nbVertex,
data.springs1.nbSpringPerVertex,
data.springs1.springs.deviceRead(),
( Deriv*)f.deviceWrite() + d,
(const Coord*)x.deviceRead() + d,
(const Deriv*)v.deviceRead() + d);
else
Kernels::addForce(data.springs1.nbVertex,
data.springs1.nbSpringPerVertex,
data.springs1.springs.deviceRead(),
( Deriv*)f.deviceWrite() + d,
(const Coord*)x.deviceRead() + d,
(const Deriv*)v.deviceRead() + d,
data.springs1.dfdx.deviceWrite());
}
}
else
{
f1.resize(x1.size());
f2.resize(x2.size());
int d1 = data.springs1.vertex0;
int d2 = data.springs2.vertex0;
if (data.springs1.nbSpringPerVertex > 0)
{
if (!stiff)
Kernels::addExternalForce(data.springs1.nbVertex,
data.springs1.nbSpringPerVertex,
data.springs1.springs.deviceRead(),
( Deriv*)f1.deviceWrite() + d1,
(const Coord*)x1.deviceRead() + d1,
(const Deriv*)v1.deviceRead() + d1,
(const Coord*)x2.deviceRead() + d2,
(const Deriv*)v2.deviceRead() + d2);
else
Kernels::addExternalForce(data.springs1.nbVertex,
data.springs1.nbSpringPerVertex,
data.springs1.springs.deviceRead(),
( Deriv*)f1.deviceWrite() + d1,
(const Coord*)x1.deviceRead() + d1,
(const Deriv*)v1.deviceRead() + d1,
(const Coord*)x2.deviceRead() + d2,
(const Deriv*)v2.deviceRead() + d2,
data.springs1.dfdx.deviceWrite());
}
if (data.springs2.nbSpringPerVertex > 0)
{
if (!stiff)
Kernels::addExternalForce(data.springs2.nbVertex,
data.springs2.nbSpringPerVertex,
data.springs2.springs.deviceRead(),
( Deriv*)f2.deviceWrite() + d2,
(const Coord*)x2.deviceRead() + d2,
(const Deriv*)v2.deviceRead() + d2,
(const Coord*)x1.deviceRead() + d1,
(const Deriv*)v1.deviceRead() + d1);
else
Kernels::addExternalForce(data.springs2.nbVertex,
data.springs2.nbSpringPerVertex,
data.springs2.springs.deviceRead(),
( Deriv*)f2.deviceWrite() + d2,
(const Coord*)x2.deviceRead() + d2,
(const Deriv*)v2.deviceRead() + d2,
(const Coord*)x1.deviceRead() + d1,
(const Deriv*)v1.deviceRead() + d1,
data.springs2.dfdx.deviceWrite());
}
}
}
template<class TCoord, class TDeriv, class TReal>
void SpringForceFieldInternalData< gpu::cuda::CudaVectorTypes<TCoord,TDeriv,TReal> >::addDForce(Main* m, bool stiff, VecDeriv& df1, VecDeriv& df2, const VecDeriv& dx1, const VecDeriv& dx2, double kFactor, double /*bFactor*/)
{
if (!stiff) return;
Data& data = m->data;
if (m->mstate1 == m->mstate2)
{
VecDeriv& df = df1;
const VecDeriv& dx = dx1;
const VecCoord& x = *m->mstate1->getX();
df.resize(x.size());
int d = data.springs1.vertex0;
if (data.springs1.nbSpringPerVertex > 0)
{
Kernels::addDForce(data.springs1.nbVertex,
data.springs1.nbSpringPerVertex,
data.springs1.springs.deviceRead(),
( Deriv*)df.deviceWrite() + d,
(const Deriv*)dx.deviceRead() + d,
(const Coord*)x.deviceRead() + d,
data.springs1.dfdx.deviceRead(),
kFactor);
}
}
else
{
const VecCoord& x1 = *m->mstate1->getX();
const VecCoord& x2 = *m->mstate2->getX();
df1.resize(x1.size());
df2.resize(x2.size());
int d1 = data.springs1.vertex0;
int d2 = data.springs2.vertex0;
if (data.springs1.nbSpringPerVertex > 0)
{
Kernels::addExternalDForce(data.springs1.nbVertex,
data.springs1.nbSpringPerVertex,
data.springs1.springs.deviceRead(),
( Deriv*)df1.deviceWrite() + d1,
(const Coord*)x1.deviceRead() + d1,
(const Deriv*)dx1.deviceRead() + d1,
(const Coord*)x2.deviceRead() + d2,
(const Deriv*)dx2.deviceRead() + d2,
data.springs1.dfdx.deviceRead(),
kFactor);
}
if (data.springs2.nbSpringPerVertex > 0)
{
Kernels::addExternalDForce(data.springs2.nbVertex,
data.springs2.nbSpringPerVertex,
data.springs2.springs.deviceRead(),
( Deriv*)df2.deviceWrite() + d2,
(const Deriv*)dx2.deviceRead() + d2,
(const Coord*)x2.deviceRead() + d2,
(const Deriv*)dx1.deviceRead() + d1,
(const Coord*)x1.deviceRead() + d1,
data.springs2.dfdx.deviceRead(),
kFactor);
}
}
}
// I know using macros is bad design but this is the only way not to repeat the code for all CUDA types
#define CudaSpringForceField_ImplMethods(T) \
template<> void SpringForceField< T >::init() \
{ data.init(this, false); } \
template<> void SpringForceField< T >::addForce(VecDeriv& f1, VecDeriv& f2, const VecCoord& x1, const VecCoord& x2, const VecDeriv& v1, const VecDeriv& v2) \
{ data.addForce(this, false, f1, f2, x1, x2, v1, v2); } \
template<> void StiffSpringForceField< T >::init() \
{ data.init(this, true); } \
template<> void StiffSpringForceField< T >::addForce(VecDeriv& f1, VecDeriv& f2, const VecCoord& x1, const VecCoord& x2, const VecDeriv& v1, const VecDeriv& v2) \
{ data.addForce(this, true, f1, f2, x1, x2, v1, v2); } \
template<> void StiffSpringForceField< T >::addDForce(VecDeriv& df1, VecDeriv& df2, const VecDeriv& dx1, const VecDeriv& dx2, double kFactor, double bFactor) \
{ data.addDForce(this, true, df1, df2, dx1, dx2, kFactor, bFactor); }
CudaSpringForceField_ImplMethods(gpu::cuda::CudaVec3fTypes);
CudaSpringForceField_ImplMethods(gpu::cuda::CudaVec3f1Types);
#ifdef SOFA_GPU_CUDA_DOUBLE
CudaSpringForceField_ImplMethods(gpu::cuda::CudaVec3dTypes);
CudaSpringForceField_ImplMethods(gpu::cuda::CudaVec3d1Types);
#endif // SOFA_GPU_CUDA_DOUBLE
#undef CudaSpringForceField_ImplMethods
} // namespace forcefield
} // namespace component
} // namespace sofa
#endif
|
