Newer
Older

Camille Coti
committed
#include "profiling.h"
#include "tensormatrix.h"
#include "products.h"
namespace gi = GiNaC;
/*******************************************************************************
* Big multiplication *
*******************************************************************************/
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
gi::ex Tens = 0;
int a1, a2, a3, a4, a5, a6;
int A1, A2, A3, A4, A5, A6;
gi::ex W1, W2, W3, W4, W5, W6, W7;
gi::ex Z1, Z2, Z6, t5, tE, t1, t12, t123, t126, t13, t134, t14, t16, t2, t23, t24, t26, t3, t4, X7Y5;
gi::ex TE, T1, T2, T3, T4, T5, T12, T13, T14, T16, T23, T24, T26, T123, T126, T134;
TE = T1 = T2 = T3 = T4 = T5 = T12 = T13 = T14 = T16 = T23 = T24 = T26 = T123 = T126 = T134 = 0;
int N = size/2;
for( a4 = 0 ; a4 < N ; a4++ ) {
A4 = a4 + N;
for( a2 = 0 ; a2 < N ; a2++ ) {
A2 = a2 + N;
for( a6 = 0 ; a6 < N ; a6++ ) {
A6 = a6 + N;
W1 = T[a4][a2][a6];
W2 = T[a4][A2][a6];
W3 = T[a4][a2][A6];
W4 = T[A4][A2][a6];
W5 = T[a4][A2][A6];
W6 = T[A4][a2][A6];
W7 = T[A4][A2][A6];
for( a1 = 0 ; a1 < N ; a1++ ) {
A1 = a1 + N;
for( a5 = 0 ; a5 < N ; a5++ ) {
A5 = a5 + N;
Z1 = T[a1][a5][a6];
Z2 = T[A1][a5][a6];
Z6 = T[A1][a5][A6];
t5 = W3*T[a1][A5][a6];
tE = W4*T[A1][A5][A6];
t1 = W3*Z2;
t13 = t1;
t2 = W5*Z1;
t23 = t2;
t3 = W3*Z1;
t4 = W6*Z1;
t12 = W5*Z2;
t14 = W6*Z2;
t134 = t14 ;
t16 = W1*Z6;
t24 = W7*Z1;
t26 = W2*T[a1][a5][A6];
t123 = W5*Z2;
t126 = W2*Z6;
for( a3 = 0 ; a3 < N ; a3++ ) {
A3 = a3 + N;
TE = TE + tE*T[a1][a2][a3]*T[a4][a5][A3];
T5 = T5 + t5*T[A1][A2][A3]*T[A4][a5][a3];
X7Y5 = T[a1][A2][A3]*T[A4][A5][a3];
T1 = T1 + t1*X7Y5;
T16 = T16 + t16*X7Y5;
T2 = T2 + t2*T[A1][a2][A3]*T[A4][A5][a3];
T3 = T3 + t3*T[A1][A2][a3]*T[A4][A5][A3];
T4 = T4 + t4*T[A1][A2][A3]*T[a4][A5][a3];
T12 = T12 + t12*T[a1][a2][A3]*T[A4][A5][a3];
T13 = T13 + t13*T[a1][A2][a3]*T[A4][A5][A3];
T14 = T14 + t14*T[a1][A2][A3]*T[a4][A5][a3];
T23 = T23 + t23*T[A1][a2][a3]*T[A4][A5][A3];
T24 = T24 + t24*T[A1][a2][A3]*T[a4][A5][a3];
T26 = T26 + t26*T[A1][a2][A3]*T[A4][A5][a3];
T123 = T123 + t123*T[a1][a2][a3]*T[A4][A5][A3];
T126 = T126 + t126*T[a1][a2][A3]*T[A4][A5][a3];
T134 = T134 + t134*T[a1][A2][a3]*T[a4][A5][A3];
}
}
}
}
}
}
Tens = 4*(TE+T12+T13+T14+T16+T23+T24+T26 - (T1 + T2 + T3 + T4 + T5 +T123 + T126 + T134));
return Tens;
}
/*******************************************************************************
* 1-level decomposition *
*******************************************************************************/
gi::ex multiply_1level( tensor3D_t& T, int size ) { // simpler: same dimension everywhere

Camille Coti
committed
double time;
#ifdef TAUPROF
void* ptr;
TAU_PROFILER_CREATE( ptr, "b","", TAU_USER );
#endif
int N = size/2;
for( a4 = 0 ; a4 < N ; a4++ ) {
for( a2 = 0 ; a2 < N ; a2++ ) {
#ifdef TAUPROF
TAU_START( "b" );
TAU_PROFILER_START( ptr );
#endif
/* Beyond this point, a2 and a3 are only used in the simplectic matrix */
for( a1 = 0 ; a1 < N ; a1++ ){
Tn = one_level1_product( &T, size, a4, a2, a1 );
#ifdef TAUPROF
TAU_STOP( "b" );
// time = getTimeSpent( ptr );
long calls, childcalls;
double incl[TAU_MAX_COUNTERS], excl[TAU_MAX_COUNTERS];
const char **counters;
int numcounters, i, j;
TAU_PROFILER_GET_CALLS(ptr, &calls);
TAU_PROFILER_GET_CHILD_CALLS(ptr, &childcalls);
TAU_PROFILER_GET_INCLUSIVE_VALUES(ptr, &incl);
TAU_PROFILER_GET_EXCLUSIVE_VALUES(ptr, &excl);
TAU_PROFILER_GET_COUNTER_INFO(&counters, &numcounters);
printf("Calls = %ld, child = %ld\n", calls, childcalls);
printf("numcounters = %d\n", numcounters);
for (j = 0; j < numcounters ; j++) {
printf(">>>");
printf("counter [%d] = %s\n", j, counters[j]);
printf(" excl [%d] = %g, incl [%d] = %g\n", j, excl[j], j, incl[j]);
}
// time = getTimeSpent( "b" );

Camille Coti
committed
#endif // TAUPROF

Camille Coti
committed
gi::ex one_level1_product( tensor3D_t* T, int size, int a4, int a2, int a1 ){
gi::ex W1, W2, W3, W4, W5, W6, W7;
gi::ex Z1, Z2, Z6, t5, tE, t1, t12, t123, t126, t13, t134, t14, t16, t2, t23, t24, t26, t3, t4, X7Y5;
gi::ex TE, T1, T2, T3, T4, T5, T12, T13, T14, T16, T23, T24, T26, T123, T126, T134;

Camille Coti
committed
const char timerB[] = "B";
int a3, a5, a6;
int A1, A2, A3, A4, A5, A6;
TE = T1 = T2 = T3 = T4 = T5 = T12 = T13 = T14 = T16 = T23 = T24 = T26 = T123 = T126 = T134 = 0;

Camille Coti
committed
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
int N = size/2;
A1 = a1 + N;
A4 = a4 + N;
A2 = a2 + N;
for( a6 = 0 ; a6 < N ; a6++ ) {
A6 = a6 + N;
W1 = (*T)[a4][a2][a6];
W2 = (*T)[a4][A2][a6];
W3 = (*T)[a4][a2][A6];
W4 = (*T)[A4][A2][a6];
W5 = (*T)[a4][A2][A6];
W6 = (*T)[A4][a2][A6];
W7 = (*T)[A4][A2][A6];
for( a5 = 0 ; a5 < N ; a5++ ) {
A5 = a5 + N;
Z1 = (*T)[a1][a5][a6];
Z2 = (*T)[A1][a5][a6];
Z6 = (*T)[A1][a5][A6];
t5 = W3*(*T)[a1][A5][a6];
tE = W4*(*T)[A1][A5][A6];
t1 = W3*Z2;
t13 = t1;
t2 = W5*Z1;
t23 = t2;
t3 = W3*Z1;
t4 = W6*Z1;
t12 = W5*Z2;
t14 = W6*Z2;
t134 = t14 ;
t16 = W1*Z6;
t24 = W7*Z1;
t26 = W2*(*T)[a1][a5][A6];
t123 = W5*Z2;
t126 = W2*Z6;
for( a3 = 0 ; a3 < N ; a3++ ) {
A3 = a3 + N;
TE = TE + tE*(*T)[a1][a2][a3]*(*T)[a4][a5][A3];
T5 = T5 + t5*(*T)[A1][A2][A3]*(*T)[A4][a5][a3];
X7Y5 = (*T)[a1][A2][A3]*(*T)[A4][A5][a3];
T1 = T1 + t1*X7Y5;
T16 = T16 + t16*X7Y5;
T2 = T2 + t2*(*T)[A1][a2][A3]*(*T)[A4][A5][a3];
T3 = T3 + t3*(*T)[A1][A2][a3]*(*T)[A4][A5][A3];
T4 = T4 + t4*(*T)[A1][A2][A3]*(*T)[a4][A5][a3];
T12 = T12 + t12*(*T)[a1][a2][A3]*(*T)[A4][A5][a3];
T13 = T13 + t13*(*T)[a1][A2][a3]*(*T)[A4][A5][A3];
T14 = T14 + t14*(*T)[a1][A2][A3]*(*T)[a4][A5][a3];
T23 = T23 + t23*(*T)[A1][a2][a3]*(*T)[A4][A5][A3];
T24 = T24 + t24*(*T)[A1][a2][A3]*(*T)[a4][A5][a3];
T26 = T26 + t26*(*T)[A1][a2][A3]*(*T)[A4][A5][a3];
T123 = T123 + t123*(*T)[a1][a2][a3]*(*T)[A4][A5][A3];
T126 = T126 + t126*(*T)[a1][a2][A3]*(*T)[A4][A5][a3];
T134 = T134 + t134*(*T)[a1][A2][a3]*(*T)[a4][A5][A3];
Tens = 4*(TE+T12+T13+T14+T16+T23+T24+T26 - (T1 + T2 + T3 + T4 + T5 +T123 + T126 + T134));
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
#if 0
std::ostringstream oss;
oss << "output_" << getpid();
std::ofstream fd;
fd.open( oss.str(), std::ios::app );
fd << "T" << A << "=" << Tens << ";";
// fd << " with " << A << " " << a1 << " " << a2 << " " << a3 << " " << b1;
fd << std::endl;
fd.close();
#endif
return Tens;
}
gi::ex one_level1_product( tensor3D_t* T, int size, int a4 ){
gi::ex Tens = 0;
gi::ex Ti0, Ti1, Ti2;
gi::ex W1, W2, W3, W4, W5, W6, W7;
gi::ex Z1, Z2, Z6, t5, tE, t1, t12, t123, t126, t13, t134, t14, t16, t2, t23, t24, t26, t3, t4, X7Y5;
gi::ex TE, T1, T2, T3, T4, T5, T12, T13, T14, T16, T23, T24, T26, T123, T126, T134;
const char timerB[] = "B";
int a1, a2, a3, a5, a6;
int A1, A2, A3, A4, A5, A6;
TE = T1 = T2 = T3 = T4 = T5 = T12 = T13 = T14 = T16 = T23 = T24 = T26 = T123 = T126 = T134 = 0;
Ti0 = Ti1 = Ti2 = 0;
int N = size/2;
A4 = a4 + N;
for( a2 = 0 ; a2 < N ; a2++ ) {
A2 = a2 + N;
Ti2 = 0;
for( a6 = 0 ; a6 < N ; a6++ ) {
A6 = a6 + N;
W1 = (*T)[a4][a2][a6];
W2 = (*T)[a4][A2][a6];
W3 = (*T)[a4][a2][A6];
W4 = (*T)[A4][A2][a6];
W5 = (*T)[a4][A2][A6];
W6 = (*T)[A4][a2][A6];
W7 = (*T)[A4][A2][A6];
Ti1 = 0;
for( a1 = 0 ; a1 < N ; a1++ ) {
A1 = a1 + N;
Ti0 = TE = T12 = T13 = T14 = T16 = T23 = T24 = T26 = T1 = T2 = T3 = T4 = T5 = T123 = T126 = T134 = 0;
for( a5 = 0 ; a5 < N ; a5++ ) {
A5 = a5 + N;
Z1 = (*T)[a1][a5][a6];
Z2 = (*T)[A1][a5][a6];
Z6 = (*T)[A1][a5][A6];
t5 = W3*(*T)[a1][A5][a6];
tE = W4*(*T)[A1][A5][A6];
t1 = W3*Z2;
t13 = t1;
t2 = W5*Z1;
t23 = t2;
t3 = W3*Z1;
t4 = W6*Z1;
t12 = W5*Z2;
t14 = W6*Z2;
t134 = t14 ;
t16 = W1*Z6;
t24 = W7*Z1;
t26 = W2*(*T)[a1][a5][A6];
t123 = W5*Z2;
t126 = W2*Z6;
for( a3 = 0 ; a3 < N ; a3++ ) {
A3 = a3 + N;
TE = TE + tE*(*T)[a1][a2][a3]*(*T)[a4][a5][A3];
T5 = T5 + t5*(*T)[A1][A2][A3]*(*T)[A4][a5][a3];
X7Y5 = (*T)[a1][A2][A3]*(*T)[A4][A5][a3];
T1 = T1 + t1*X7Y5;
T16 = T16 + t16*X7Y5;
T2 = T2 + t2*(*T)[A1][a2][A3]*(*T)[A4][A5][a3];
T3 = T3 + t3*(*T)[A1][A2][a3]*(*T)[A4][A5][A3];
T4 = T4 + t4*(*T)[A1][A2][A3]*(*T)[a4][A5][a3];
T12 = T12 + t12*(*T)[a1][a2][A3]*(*T)[A4][A5][a3];
T13 = T13 + t13*(*T)[a1][A2][a3]*(*T)[A4][A5][A3];
T14 = T14 + t14*(*T)[a1][A2][A3]*(*T)[a4][A5][a3];
T23 = T23 + t23*(*T)[A1][a2][a3]*(*T)[A4][A5][A3];
T24 = T24 + t24*(*T)[A1][a2][A3]*(*T)[a4][A5][a3];
T26 = T26 + t26*(*T)[A1][a2][A3]*(*T)[A4][A5][a3];
T123 = T123 + t123*(*T)[a1][a2][a3]*(*T)[A4][A5][A3];
T126 = T126 + t126*(*T)[a1][a2][A3]*(*T)[A4][A5][a3];
T134 = T134 + t134*(*T)[a1][A2][a3]*(*T)[a4][A5][A3];
}
Ti0 += ( 4*(TE+T12+T13+T14+T16+T23+T24+T26 - (T1 + T2 + T3 + T4 + T5 +T123 + T126 + T134)) );
}
Ti1 += Ti0;
}
Ti2 += Ti1;
}
Tens += Ti2;
}
#if 0
std::ostringstream oss;
oss << "output_" << getpid();
std::ofstream fd;
fd.open( oss.str(), std::ios::app );
fd << "T" << A << "=" << Tens << ";";
// fd << " with " << A << " " << a1 << " " << a2 << " " << a3 << " " << b1;
fd << std::endl;
fd.close();
#endif
return Tens;
}
gi::ex one_level2_product( tensor3D_t* T, int size, int a4, int a2, int a1, int a6 ){
int a3, a5;
int A1, A2, A3, A4, A5, A6;
gi::ex W1, W2, W3, W4, W5, W6, W7;
gi::ex Z1, Z2, Z6, t5, tE, t1, t12, t123, t126, t13, t134, t14, t16, t2, t23, t24, t26, t3, t4, X7Y5;
gi::ex TE, T1, T2, T3, T4, T5, T12, T13, T14, T16, T23, T24, T26, T123, T126, T134;
TE = T1 = T2 = T3 = T4 = T5 = T12 = T13 = T14 = T16 = T23 = T24 = T26 = T123 = T126 = T134 = 0;
int N = size/2;
A1 = a1 + N;
A4 = a4 + N;
A2 = a2 + N;
A6 = a6 + N;
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
W1 = (*T)[a4][a2][a6];
W2 = (*T)[a4][A2][a6];
W3 = (*T)[a4][a2][A6];
W4 = (*T)[A4][A2][a6];
W5 = (*T)[a4][A2][A6];
W6 = (*T)[A4][a2][A6];
W7 = (*T)[A4][A2][A6];
for( a5 = 0 ; a5 < N ; a5++ ) {
A5 = a5 + N;
Z1 = (*T)[a1][a5][a6];
Z2 = (*T)[A1][a5][a6];
Z6 = (*T)[A1][a5][A6];
t5 = W3*(*T)[a1][A5][a6];
tE = W4*(*T)[A1][A5][A6];
t1 = W3*Z2;
t13 = t1;
t2 = W5*Z1;
t23 = t2;
t3 = W3*Z1;
t4 = W6*Z1;
t12 = W5*Z2;
t14 = W6*Z2;
t134 = t14 ;
t16 = W1*Z6;
t24 = W7*Z1;
t26 = W2*(*T)[a1][a5][A6];
t123 = W5*Z2;
t126 = W2*Z6;
for( a3 = 0 ; a3 < N ; a3++ ) {
A3 = a3 + N;
TE = TE + tE*(*T)[a1][a2][a3]*(*T)[a4][a5][A3];
T5 = T5 + t5*(*T)[A1][A2][A3]*(*T)[A4][a5][a3];
X7Y5 = (*T)[a1][A2][A3]*(*T)[A4][A5][a3];
T1 = T1 + t1*X7Y5;
T16 = T16 + t16*X7Y5;
T2 = T2 + t2*(*T)[A1][a2][A3]*(*T)[A4][A5][a3];
T3 = T3 + t3*(*T)[A1][A2][a3]*(*T)[A4][A5][A3];
T4 = T4 + t4*(*T)[A1][A2][A3]*(*T)[a4][A5][a3];
T12 = T12 + t12*(*T)[a1][a2][A3]*(*T)[A4][A5][a3];
T13 = T13 + t13*(*T)[a1][A2][a3]*(*T)[A4][A5][A3];
T14 = T14 + t14*(*T)[a1][A2][A3]*(*T)[a4][A5][a3];
T23 = T23 + t23*(*T)[A1][a2][a3]*(*T)[A4][A5][A3];
T24 = T24 + t24*(*T)[A1][a2][A3]*(*T)[a4][A5][a3];
T26 = T26 + t26*(*T)[A1][a2][A3]*(*T)[A4][A5][a3];
T123 = T123 + t123*(*T)[a1][a2][a3]*(*T)[A4][A5][A3];
T126 = T126 + t126*(*T)[a1][a2][A3]*(*T)[A4][A5][a3];
T134 = T134 + t134*(*T)[a1][A2][a3]*(*T)[a4][A5][A3];
Tens = 4*(TE+T12+T13+T14+T16+T23+T24+T26 - (T1 + T2 + T3 + T4 + T5 +T123 + T126 + T134));
return Tens;
}
/*******************************************************************************
* 2-level decomposition *
*******************************************************************************/
gi::ex multiply_2levels( tensor3D_t& T, int size ) { // simpler: same dimension everywhere
for( a4 = 0 ; a4 < N ; a4++ ) {
for( a2 = 0 ; a2 < N ; a2++ ) {
Tens += two_level1_product( &T, size, a2, a4 );
gi::ex two_level1_product( tensor3D_t* T, int size, int a4, int a2 ){
int a1, a6;
int N = size/2;
for( a6 = 0 ; a6 < N ; a6++ ) {
for( a1 = 0 ; a1 < N ; a1++ ) {
Tens += two_level2_product( T, size, a2, a4, a6, a1 );
gi::ex two_level2_product( tensor3D_t* T, int size, int a4, int a2, int a1, int a6 ) {
gi::ex Tens = 0;
int a3, a5;
int A1, A2, A3, A4, A5, A6;
gi::ex W1, W2, W3, W4, W5, W6, W7;
gi::ex Z1, Z2, Z6, t5, tE, t1, t12, t123, t126, t13, t134, t14, t16, t2, t23, t24, t26, t3, t4, X7Y5;
gi::ex TE, T1, T2, T3, T4, T5, T12, T13, T14, T16, T23, T24, T26, T123, T126, T134;
TE = T1 = T2 = T3 = T4 = T5 = T12 = T13 = T14 = T16 = T23 = T24 = T26 = T123 = T126 = T134 = 0;
int N = size/2;
A4 = a4 + N;
A2 = a2 + N;
A1 = a1 + N;
A6 = a6 + N;
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
W1 = (*T)[a4][a2][a6];
W2 = (*T)[a4][A2][a6];
W3 = (*T)[a4][a2][A6];
W4 = (*T)[A4][A2][a6];
W5 = (*T)[a4][A2][A6];
W6 = (*T)[A4][a2][A6];
W7 = (*T)[A4][A2][A6];
for( a5 = 0 ; a5 < N ; a5++ ) {
A5 = a5 + N;
Z1 = (*T)[a1][a5][a6];
Z2 = (*T)[A1][a5][a6];
Z6 = (*T)[A1][a5][A6];
t5 = W3*(*T)[a1][A5][a6];
tE = W4*(*T)[A1][A5][A6];
t1 = W3*Z2;
t13 = t1;
t2 = W5*Z1;
t23 = t2;
t3 = W3*Z1;
t4 = W6*Z1;
t12 = W5*Z2;
t14 = W6*Z2;
t134 = t14 ;
t16 = W1*Z6;
t24 = W7*Z1;
t26 = W2*(*T)[a1][a5][A6];
t123 = W5*Z2;
t126 = W2*Z6;
for( a3 = 0 ; a3 < N ; a3++ ) {
A3 = a3 + N;
TE = TE + tE*(*T)[a1][a2][a3]*(*T)[a4][a5][A3];
T5 = T5 + t5*(*T)[A1][A2][A3]*(*T)[A4][a5][a3];
X7Y5 = (*T)[a1][A2][A3]*(*T)[A4][A5][a3];
T1 = T1 + t1*X7Y5;
T16 = T16 + t16*X7Y5;
T2 = T2 + t2*(*T)[A1][a2][A3]*(*T)[A4][A5][a3];
T3 = T3 + t3*(*T)[A1][A2][a3]*(*T)[A4][A5][A3];
T4 = T4 + t4*(*T)[A1][A2][A3]*(*T)[a4][A5][a3];
T12 = T12 + t12*(*T)[a1][a2][A3]*(*T)[A4][A5][a3];
T13 = T13 + t13*(*T)[a1][A2][a3]*(*T)[A4][A5][A3];
T14 = T14 + t14*(*T)[a1][A2][A3]*(*T)[a4][A5][a3];
T23 = T23 + t23*(*T)[A1][a2][a3]*(*T)[A4][A5][A3];
T24 = T24 + t24*(*T)[A1][a2][A3]*(*T)[a4][A5][a3];
T26 = T26 + t26*(*T)[A1][a2][A3]*(*T)[A4][A5][a3];
T123 = T123 + t123*(*T)[a1][a2][a3]*(*T)[A4][A5][A3];
T126 = T126 + t126*(*T)[a1][a2][A3]*(*T)[A4][A5][a3];
T134 = T134 + t134*(*T)[a1][A2][a3]*(*T)[a4][A5][A3];
}
Tens = 4*(TE+T12+T13+T14+T16+T23+T24+T26 - (T1 + T2 + T3 + T4 + T5 +T123 + T126 + T134));
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
gi::ex two_level2_product( tensor3D_t* T, int size, int a4, int a2 ) {
gi::ex Tens = 0;
int a3, a5, a1, a6;
int A1, A2, A3, A4, A5, A6;
gi::ex W1, W2, W3, W4, W5, W6, W7;
gi::ex Z1, Z2, Z6, t5, tE, t1, t12, t123, t126, t13, t134, t14, t16, t2, t23, t24, t26, t3, t4, X7Y5;
gi::ex TE, T1, T2, T3, T4, T5, T12, T13, T14, T16, T23, T24, T26, T123, T126, T134;
TE = T1 = T2 = T3 = T4 = T5 = T12 = T13 = T14 = T16 = T23 = T24 = T26 = T123 = T126 = T134 = 0;
gi::ex Ti0, Ti1;
int N = size/2;
A4 = a4 + N;
A2 = a2 + N;
for( a6 = 0 ; a6 < N ; a6++ ) {
A6 = a6 + N;
W1 = (*T)[a4][a2][a6];
W2 = (*T)[a4][A2][a6];
W3 = (*T)[a4][a2][A6];
W4 = (*T)[A4][A2][a6];
W5 = (*T)[a4][A2][A6];
W6 = (*T)[A4][a2][A6];
W7 = (*T)[A4][A2][A6];
Ti1 = 0;
for( a1 = 0 ; a1 < N ; a1++ ) {
A1 = a1 + N;
Ti0 = TE = T12 = T13 = T14 = T16 = T23 = T24 = T26 = T1 = T2 = T3 = T4 = T5 = T123 = T126 = T134 = 0;
for( a5 = 0 ; a5 < N ; a5++ ) {
A5 = a5 + N;
Z1 = (*T)[a1][a5][a6];
Z2 = (*T)[A1][a5][a6];
Z6 = (*T)[A1][a5][A6];
t5 = W3*(*T)[a1][A5][a6];
tE = W4*(*T)[A1][A5][A6];
t1 = W3*Z2;
t13 = t1;
t2 = W5*Z1;
t23 = t2;
t3 = W3*Z1;
t4 = W6*Z1;
t12 = W5*Z2;
t14 = W6*Z2;
t134 = t14 ;
t16 = W1*Z6;
t24 = W7*Z1;
t26 = W2*(*T)[a1][a5][A6];
t123 = W5*Z2;
t126 = W2*Z6;
for( a3 = 0 ; a3 < N ; a3++ ) {
A3 = a3 + N;
TE = TE + tE*(*T)[a1][a2][a3]*(*T)[a4][a5][A3];
T5 = T5 + t5*(*T)[A1][A2][A3]*(*T)[A4][a5][a3];
X7Y5 = (*T)[a1][A2][A3]*(*T)[A4][A5][a3];
T1 = T1 + t1*X7Y5;
T16 = T16 + t16*X7Y5;
T2 = T2 + t2*(*T)[A1][a2][A3]*(*T)[A4][A5][a3];
T3 = T3 + t3*(*T)[A1][A2][a3]*(*T)[A4][A5][A3];
T4 = T4 + t4*(*T)[A1][A2][A3]*(*T)[a4][A5][a3];
T12 = T12 + t12*(*T)[a1][a2][A3]*(*T)[A4][A5][a3];
T13 = T13 + t13*(*T)[a1][A2][a3]*(*T)[A4][A5][A3];
T14 = T14 + t14*(*T)[a1][A2][A3]*(*T)[a4][A5][a3];
T23 = T23 + t23*(*T)[A1][a2][a3]*(*T)[A4][A5][A3];
T24 = T24 + t24*(*T)[A1][a2][A3]*(*T)[a4][A5][a3];
T26 = T26 + t26*(*T)[A1][a2][A3]*(*T)[A4][A5][a3];
T123 = T123 + t123*(*T)[a1][a2][a3]*(*T)[A4][A5][A3];
T126 = T126 + t126*(*T)[a1][a2][A3]*(*T)[A4][A5][a3];
T134 = T134 + t134*(*T)[a1][A2][a3]*(*T)[a4][A5][A3];
}
Ti0 += ( 4*(TE+T12+T13+T14+T16+T23+T24+T26 - (T1 + T2 + T3 + T4 + T5 +T123 + T126 + T134)) );
}
Ti1 += Ti0;
}
Tens += Ti1;
}
return Tens;
}