/usr/include/fflas-ffpack/fflas/fflas_fgemm/schedule_winograd.inl is in fflas-ffpack-common 2.2.2-5.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 | /* -*- mode: C++; tab-width: 4; indent-tabs-mode: t; c-basic-offset: 4 -*- */
// vim:sts=4:sw=4:ts=4:noet:sr:cino=>s,f0,{0,g0,(0,\:0,t0,+0,=s
/*
* Copyright (C) 2014 the LinBox group
*
* Written by Clement Pernet <Clement.Pernet@imag.fr>
* Brice Boyer (briceboyer) <boyer.brice@gmail.com>
* Ziad Sultan <ziad.sultan@imag.fr>
*
* ========LICENCE========
* This file is part of the library FFLAS-FFPACK.
*
* FFLAS-FFPACK 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
* ========LICENCE========
*.
*/
/** @file fflas/fflas_fgemm/winograd.inl
* @ingroup MMalgos
* @brief Winograd implementation
* @bib ISSAC09 Scheduling
*/
#ifndef __FFLASFFPACK_fgemm_winograd_INL
#define __FFLASFFPACK_fgemm_winograd_INL
namespace FFLAS { namespace BLAS3 {
template < class Field, class FieldTrait, class Strat, class Param >
inline typename Field::Element_ptr
WinoPar (const Field& F,
const FFLAS_TRANSPOSE ta,
const FFLAS_TRANSPOSE tb,
const size_t mr, const size_t nr, const size_t kr,
const typename Field::Element alpha,
typename Field::ConstElement_ptr A,const size_t lda,
typename Field::ConstElement_ptr B,const size_t ldb,
const typename Field::Element beta,
typename Field::Element_ptr C, const size_t ldc,
// const size_t kmax, const size_t w, const FFLAS_BASE base
MMHelper<Field, MMHelperAlgo::WinogradPar, FieldTrait, ParSeqHelper::Parallel<Strat,Param> > & WH
)
{
FFLASFFPACK_check(F.isZero(beta));
// typedef MMHelper<Field, MMHelperAlgo::WinogradPar, FieldTrait > MMH_t;
typedef MMHelper<Field, MMHelperAlgo::WinogradPar, FieldTrait, ParSeqHelper::Parallel<CuttingStrategy::Recursive,StrategyParameter::TwoDAdaptive> > MMH_t;
const typename MMH_t::DelayedField & DF = WH.delayedField;
typedef typename MMH_t::DelayedField::Element DFElt;
size_t lb, cb, la, ca, ldX2;
// size_t x3rd = std::max(mr,kr);
typename Field::ConstElement_ptr A11=A, A12, A21, A22;
typename Field::ConstElement_ptr B11=B, B12, B21, B22;
typename Field::Element_ptr C11=C, C12=C+nr, C21=C+mr*ldc, C22=C21+nr;
size_t x1rd = std::max(nr,kr);
size_t ldX1;
if (ta == FflasTrans) {
A21 = A + mr;
A12 = A + kr*lda;
A22 = A12 + mr;
la = kr;
ca = mr;
ldX1 = mr;
} else {
A12 = A + kr;
A21 = A + mr*lda;
A22 = A21 + kr;
la = mr;
ca = kr;
ldX1 = x1rd;
}
if (tb == FflasTrans) {
B21 = B + kr;
B12 = B + nr*ldb;
B22 = B12 + kr;
lb = nr;
cb = kr;
ldX2 = kr;
} else {
B12 = B + nr;
B21 = B + kr*ldb;
B22 = B21 + nr;
lb = kr;
ldX2 = cb = nr;
}
// 11 temporary submatrices are required
typename Field::Element_ptr X21 = fflas_new (F, kr, nr);
typename Field::Element_ptr X11 = fflas_new (F,mr,x1rd);
typename Field::Element_ptr X22 = fflas_new (F, kr, nr);
typename Field::Element_ptr X12 = fflas_new (F,mr,x1rd);
typename Field::Element_ptr X23 = fflas_new (F, kr, nr);
typename Field::Element_ptr X13 = fflas_new (F,mr,x1rd);
typename Field::Element_ptr X24 = fflas_new (F, kr, nr);
typename Field::Element_ptr X14 = fflas_new (F,mr,x1rd);
typename Field::Element_ptr X15 = fflas_new (F,mr,x1rd);
typename Field::Element_ptr C_11 = fflas_new (F,mr,nr);
typename Field::Element_ptr CC_11 = fflas_new (F,mr,nr);
SYNCH_GROUP(
// T3 = B22 - B12 in X21 and S3 = A11 - A21 in X11
TASK(MODE(READ(B22, B12) WRITE(X21) CONSTREFERENCE(DF)),
pfsub(DF,lb,cb,B22,ldb,B12,ldb,X21,ldX2, NUM_THREADS););
TASK(MODE(READ(A11, A21) WRITE(X11) CONSTREFERENCE(DF)),
pfsub(DF,la,ca,A11,lda,A21,lda,X11,ldX1, NUM_THREADS););
// T1 = B12 - B11 in X22 and S1 = A21 + A22 in X12
TASK(MODE(READ(B11, B12) WRITE(X22) CONSTREFERENCE(DF)),
pfsub(DF,lb,cb,B12,ldb,B11,ldb,X22,ldX2, NUM_THREADS););
TASK(MODE(READ(A12, A22) WRITE(X12) CONSTREFERENCE(DF)),
pfadd(DF,la,ca,A21,lda,A22,lda,X12,ldX1, NUM_THREADS););
CHECK_DEPENDENCIES;
// T2 = B22 - T1 in X23 and S2 = S1 - A11 in X13
TASK(MODE(READ(B22, X22) READWRITE(X23) CONSTREFERENCE(DF)),
pfsub(DF,lb,cb,B22,ldb,X22,ldX2,X23,ldX2, NUM_THREADS););
TASK(MODE(READ(A11, X12) READWRITE(X13) CONSTREFERENCE(DF)),
// fsub(DF,la,ca,A11,lda,X12,ldX1,X13,ldX1););
pfsub(DF,la,ca,X12,ldX1,A11,lda,X13,ldX1, NUM_THREADS););
/*
fsub(DF,lb,cb,B22,ldb,X2,ldX2,X2,ldX2);
fsubin(DF,la,ca,A11,lda,X1,ldX1););
*/
CHECK_DEPENDENCIES;
// T4 = T2 - B21 in X2 and S4 = A12 -S2 in X1
TASK(MODE(READ(B21, X23) READWRITE(X24) CONSTREFERENCE(DF)),
// fsub(DF,lb,cb,B21,ldb,X23,ldX2,X24,ldX2);
pfsub(DF,lb,cb,X23,ldX2,B21,ldb,X24,ldX2, NUM_THREADS););
TASK(MODE(READ(A12, X13) READWRITE(X14) CONSTREFERENCE(DF)),
pfsub(DF,la,ca,A12,lda,X13,ldX1,X14,ldX1, NUM_THREADS););
/*
fsubin(DF,lb,cb,B21,ldb,X2,ldX2);
fsub(DF,la,ca,A12,lda,X1,ldX1,X1,ldX1););
*/
CHECK_DEPENDENCIES;
// P1 = alpha . A11 * B11 in X1
MMH_t H1(F, WH.recLevel-1, WH.Amin, WH.Amax, WH.Bmin, WH.Bmax, 0, 0);
MMH_t H7(F, WH.recLevel-1, -(WH.Amax-WH.Amin), WH.Amax-WH.Amin, -(WH.Bmax-WH.Bmin), WH.Bmax-WH.Bmin, 0,0);
MMH_t H5(F, WH.recLevel-1, 2*WH.Amin, 2*WH.Amax, -(WH.Bmax-WH.Bmin), WH.Bmax-WH.Bmin, 0, 0);
MMH_t H6(F, WH.recLevel-1, 2*WH.Amin-WH.Amax, 2*WH.Amax-WH.Amin, 2*WH.Bmin-WH.Bmax, 2*WH.Bmax-WH.Bmin, 0, 0);
MMH_t H3(F, WH.recLevel-1, 2*WH.Amin-2*WH.Amax, 2*WH.Amax-2*WH.Amin, WH.Bmin, WH.Bmax, 0, 0);
MMH_t H4(F, WH.recLevel-1, WH.Amin, WH.Amax, 2*WH.Bmin-2*WH.Bmax, 2*WH.Bmax-2*WH.Bmin, 0, 0);
MMH_t H2(F, WH.recLevel-1, WH.Amin, WH.Amax, WH.Bmin, WH.Bmax, 0, 0);
size_t nt = WH.parseq.numthreads();
size_t nt_rec = nt/7;
size_t nt_mod = nt % 7 ;
H1.parseq.set_numthreads(std::max(size_t(1),nt_rec + ((nt_mod-- > 0)?1:0)));
H2.parseq.set_numthreads(std::max(size_t(1),nt_rec + ((nt_mod-- > 0)?1:0)));
H3.parseq.set_numthreads(std::max(size_t(1),nt_rec + ((nt_mod-- > 0)?1:0)));
H4.parseq.set_numthreads(std::max(size_t(1),nt_rec + ((nt_mod-- > 0)?1:0)));
H5.parseq.set_numthreads(std::max(size_t(1),nt_rec + ((nt_mod-- > 0)?1:0)));
H6.parseq.set_numthreads(std::max(size_t(1),nt_rec + ((nt_mod-- > 0)?1:0)));
H7.parseq.set_numthreads(std::max(size_t(1),nt_rec + ((nt_mod-- > 0)?1:0)));
TASK(MODE(READ(A11, B11) WRITE(X15) CONSTREFERENCE(F,H1)),
fgemm (F, ta, tb, mr, nr, kr, alpha, A11, lda, B11, ldb, F.zero, X15, x1rd, H1););
// P7 = alpha . S3 * T3 in C21
TASK(MODE(READ(X11, X21) WRITE(C21) CONSTREFERENCE(F,H7)),
fgemm (F, ta, tb, mr, nr, kr, alpha, X11, ldX1, X21, ldX2, F.zero, C21, ldc, H7););
// P5 = alpha . S1*T1 in C22
TASK(MODE(READ(X12, X22) WRITE(C22) CONSTREFERENCE(F,H5)),
fgemm (F, ta, tb, mr, nr, kr, alpha, X12, ldX1, X22, ldX2, F.zero, C22, ldc, H5););
// P6 = alpha . S2 * T2 in C12
TASK(MODE(READ(X13, X23) WRITE(C12) CONSTREFERENCE(F,H6)),
fgemm (F, ta, tb, mr, nr, kr, alpha, X13, ldX1, X23, ldX2, F.zero, C12, ldc, H6););
// P3 = alpha . S4*B22 in CC_11
TASK(MODE(READ(X14, B22) WRITE(CC_11) CONSTREFERENCE(F,H3)),
fgemm (F, ta, tb, mr, nr, kr, alpha, X14, ldX1, B22, ldb, F.zero, CC_11, nr, H3););
// P4 = alpha . A22 * T4 in C_11
TASK(MODE(READ(A22) WRITE(C_11) READWRITE(X24, X22, X23, X21) CONSTREFERENCE(F,H4)),
fgemm (F, ta, tb, mr, nr, kr, alpha, A22, lda, X24, ldX2, F.zero, C_11, nr, H4);
);
// P2 = alpha . A12 * B21 in C11
TASK(MODE(READ(A12, B21) WRITE(C11) CONSTREFERENCE(F,H2)),
fgemm (F, ta, tb, mr, nr, kr, alpha, A12, lda, B21, ldb, F.zero, C11, ldc, H2););
CHECK_DEPENDENCIES;
DFElt U2Min, U2Max;
DFElt U3Min, U3Max;
DFElt U4Min, U4Max;
DFElt U7Min, U7Max;
DFElt U5Min, U5Max;
// U2 = P1 + P6 in C12 and
// U3 = P7 + U2 in C21 and
// U4 = P5 + U2 in C12 and
// U7 = P5 + U3 in C22 and
// U5 = P3 + U4 in C12
// BIG TASK with 5 Addin function calls
// TASK(MODE(READWRITE(X15, C12) CONSTREFERENCE(F, DF, WH, U2Min, U2Max, H1.Outmin, H1.Outmax, H6.Outmin, H6.Outmax)),
if (Protected::NeedPreAddReduction(U2Min, U2Max, H1.Outmin, H1.Outmax, H6.Outmin, H6.Outmax, WH)){
TASK(MODE(READWRITE(X15) CONSTREFERENCE(F)),
pfreduce (F, mr, x1rd, X15, x1rd, NUM_THREADS);
);
TASK(MODE(READWRITE(C12) CONSTREFERENCE(F)),
pfreduce (F, mr, nr, C12, ldc, NUM_THREADS);
);
CHECK_DEPENDENCIES;
}
TASK(MODE(READWRITE(X15, C12) CONSTREFERENCE(DF)),
pfaddin(DF,mr,nr,X15,x1rd,C12,ldc, NUM_THREADS);
);
CHECK_DEPENDENCIES;
// TASK(MODE(READWRITE(C12, C21) CONSTREFERENCE(F, DF, WH, U3Min, U3Max, U2Min, U2Max)),
if (Protected::NeedPreAddReduction(U3Min, U3Max, U2Min, U2Max, H7.Outmin, H7.Outmax, WH)){
TASK(MODE(READWRITE(C12) CONSTREFERENCE(F)),
pfreduce (F, mr, nr, C12, ldc, NUM_THREADS);
);
TASK(MODE(READWRITE(C21) CONSTREFERENCE(F)),
pfreduce (F, mr, nr, C21, ldc, NUM_THREADS);
);
CHECK_DEPENDENCIES;
}
TASK(MODE(READWRITE(C12, C21) CONSTREFERENCE(DF)),
pfaddin(DF,mr,nr,C12,ldc,C21,ldc, NUM_THREADS);
);
CHECK_DEPENDENCIES;
// TASK(MODE(READWRITE(C12, C22) CONSTREFERENCE(F, DF, WH) VALUE(U4Min, U4Max, U2Min, U2Max)),
if (Protected::NeedPreAddReduction(U4Min, U4Max, U2Min, U2Max, H5.Outmin, H5.Outmax, WH)){
TASK(MODE(READWRITE(C22) CONSTREFERENCE(F)),
pfreduce (F, mr, nr, C22, ldc, NUM_THREADS);
);
TASK(MODE(READWRITE(C12) CONSTREFERENCE(F)),
pfreduce (F, mr, nr, C12, ldc, NUM_THREADS);
);
CHECK_DEPENDENCIES;
}
TASK(MODE(READWRITE(C12, C22) CONSTREFERENCE(DF, WH)),
pfaddin(DF,mr,nr,C22,ldc,C12,ldc, NUM_THREADS);
);
CHECK_DEPENDENCIES;
// TASK(MODE(READWRITE(C22, C21) CONSTREFERENCE(F, DF, WH) VALUE(U3Min, U3Max, U7Min, U7Max)),
if (Protected::NeedPreAddReduction (U7Min,U7Max, U3Min, U3Max, H5.Outmin,H5.Outmax, WH) ){
TASK(MODE(READWRITE(C21) CONSTREFERENCE(F)),
pfreduce (F, mr, nr, C21, ldc, NUM_THREADS);
);
TASK(MODE(READWRITE(C22) CONSTREFERENCE(F)),
pfreduce (F, mr, nr, C22, ldc, NUM_THREADS);
);
CHECK_DEPENDENCIES;
}
TASK(MODE(READWRITE(C22, C21) CONSTREFERENCE(DF, WH)),
pfaddin(DF,mr,nr,C21,ldc,C22,ldc, NUM_THREADS);
);
// TASK(MODE(READWRITE(C12, CC_11) CONSTREFERENCE(F, DF, WH) VALUE(U5Min, U5Max, U4Min, U4Max)),
if (Protected::NeedPreAddReduction (U5Min,U5Max, U4Min, U4Max, H3.Outmin, H3.Outmax, WH) ){
TASK(MODE(READWRITE(C12) CONSTREFERENCE(F)),
pfreduce (F, mr, nr, C12, ldc, NUM_THREADS);
);
TASK(MODE(READWRITE(CC_11) CONSTREFERENCE(F)),
pfreduce (F, mr, nr, CC_11, nr, NUM_THREADS);
);
CHECK_DEPENDENCIES;
}
TASK(MODE(READWRITE(C12, CC_11) CONSTREFERENCE(DF, WH)),
pfaddin(DF,mr,nr,CC_11,nr,C12,ldc, NUM_THREADS);
);
CHECK_DEPENDENCIES;
// U6 = U3 - P4 in C21
DFElt U6Min, U6Max;
// TASK(MODE(READWRITE(C_11, C21) CONSTREFERENCE(F, DF, WH) VALUE(U6Min, U6Max, U3Min, U3Max)),
if (Protected::NeedPreSubReduction (U6Min,U6Max, U3Min, U3Max, H4.Outmin,H4.Outmax, WH) ){
TASK(MODE(READWRITE(CC_11) CONSTREFERENCE(F)),
pfreduce (F, mr, nr, C_11, nr, NUM_THREADS);
);
TASK(MODE(READWRITE(C21) CONSTREFERENCE(F)),
pfreduce (F, mr, nr, C21, ldc, NUM_THREADS);
);
CHECK_DEPENDENCIES
}
TASK(MODE(READWRITE(C_11, C21) CONSTREFERENCE(DF, WH) ),
pfsubin(DF,mr,nr,C_11,nr,C21,ldc, NUM_THREADS);
);
//CHECK_DEPENDENCIES;
// U1 = P2 + P1 in C11
DFElt U1Min, U1Max;
// TASK(MODE(READWRITE(C11, X15/*, X14, X13, X12, X11*/) CONSTREFERENCE(F, DF, WH) VALUE(U1Min, U1Max)),
if (Protected::NeedPreAddReduction (U1Min, U1Max, H1.Outmin, H1.Outmax, H2.Outmin,H2.Outmax, WH) ){
TASK(MODE(READWRITE(X15) CONSTREFERENCE(F)),
pfreduce (F, mr, nr, X15, x1rd, NUM_THREADS);
);
TASK(MODE(READWRITE(C11) CONSTREFERENCE(F)),
pfreduce (F, mr, nr, C11, ldc, NUM_THREADS);
);
CHECK_DEPENDENCIES
}
TASK(MODE(READWRITE(C11, X15) CONSTREFERENCE(DF, WH)),
pfaddin(DF,mr,nr,X15,x1rd,C11,ldc, NUM_THREADS);
);
WH.Outmin = std::min (U1Min, std::min (U5Min, std::min (U6Min, U7Min)));
WH.Outmax = std::max (U1Max, std::max (U5Max, std::max (U6Max, U7Max)));
);
// WAIT;
fflas_delete (CC_11);
fflas_delete (C_11);
fflas_delete (X15);
fflas_delete (X14);
fflas_delete (X24);
fflas_delete (X13);
fflas_delete (X23);
fflas_delete (X12);
fflas_delete (X22);
fflas_delete (X11);
fflas_delete (X21);
return C;
} //wino parallel
template < class Field, class FieldTrait >
inline void Winograd (const Field& F,
const FFLAS_TRANSPOSE ta,
const FFLAS_TRANSPOSE tb,
const size_t mr, const size_t nr, const size_t kr,
const typename Field::Element alpha,
typename Field::ConstElement_ptr A,const size_t lda,
typename Field::ConstElement_ptr B,const size_t ldb,
const typename Field::Element beta,
typename Field::Element_ptr C, const size_t ldc,
// const size_t kmax, const size_t w, const FFLAS_BASE base
MMHelper<Field, MMHelperAlgo::Winograd, FieldTrait> & WH
)
{
FFLASFFPACK_check(F.isZero(beta));
typedef MMHelper<Field, MMHelperAlgo::Winograd, FieldTrait > MMH_t;
typedef typename MMH_t::DelayedField::Element_ptr DFEptr;
typedef typename MMH_t::DelayedField::ConstElement_ptr DFCEptr;
typedef typename MMH_t::DelayedField::Element DFElt;
const typename MMH_t::DelayedField & DF = WH.delayedField;
size_t lb, cb, la, ca, ldX2;
// size_t x3rd = std::max(mr,kr);
typename Field::ConstElement_ptr A11=A, A12, A21, A22;
typename Field::ConstElement_ptr B11=B, B12, B21, B22;
typename Field::Element_ptr C11=C, C12=C+nr, C21=C+mr*ldc, C22=C21+nr;
size_t x1rd = std::max(nr,kr);
size_t ldX1;
if (ta == FflasTrans) {
A21 = A + mr;
A12 = A + kr*lda;
A22 = A12 + mr;
la = kr;
ca = mr;
ldX1 = mr;
} else {
A12 = A + kr;
A21 = A + mr*lda;
A22 = A21 + kr;
la = mr;
ca = kr;
ldX1 = x1rd;
}
if (tb == FflasTrans) {
B21 = B + kr;
B12 = B + nr*ldb;
B22 = B12 + kr;
lb = nr;
cb = kr;
ldX2 = kr;
} else {
B12 = B + nr;
B21 = B + kr*ldb;
B22 = B21 + nr;
lb = kr;
ldX2 = cb = nr;
}
// Two temporary submatrices are required
typename Field::Element_ptr X2 = fflas_new (F, kr, nr);
// T3 = B22 - B12 in X2
fsub(DF,lb,cb, (DFCEptr) B22,ldb, (DFCEptr) B12,ldb, (DFEptr)X2,ldX2);
// S3 = A11 - A21 in X1
typename Field::Element_ptr X1 = fflas_new (F,mr,x1rd);
fsub(DF,la,ca,(DFCEptr)A11,lda,(DFCEptr)A21,lda,(DFEptr)X1,ldX1);
// P7 = alpha . S3 * T3 in C21
MMH_t H7(F, WH.recLevel-1, -(WH.Amax-WH.Amin), WH.Amax-WH.Amin, -(WH.Bmax-WH.Bmin), WH.Bmax-WH.Bmin, 0,0);
fgemm (F, ta, tb, mr, nr, kr, alpha, X1, ldX1, X2, ldX2, F.zero, C21, ldc, H7);
// T1 = B12 - B11 in X2
fsub(DF,lb,cb,(DFCEptr)B12,ldb,(DFCEptr)B11,ldb,(DFEptr)X2,ldX2);
// S1 = A21 + A22 in X1
fadd(DF,la,ca,(DFCEptr)A21,lda,(DFCEptr)A22,lda,(DFEptr)X1,ldX1);
// P5 = alpha . S1*T1 in C22
MMH_t H5(F, WH.recLevel-1, 2*WH.Amin, 2*WH.Amax, -(WH.Bmax-WH.Bmin), WH.Bmax-WH.Bmin, 0, 0);
fgemm (F, ta, tb, mr, nr, kr, alpha, X1, ldX1, X2, ldX2, F.zero, C22, ldc, H5);
// T2 = B22 - T1 in X2
fsub(DF,lb,cb,(DFCEptr)B22,ldb,(DFCEptr)X2,ldX2,(DFEptr)X2,ldX2);
// S2 = S1 - A11 in X1
fsubin(DF,la,ca,(DFCEptr)A11,lda,(DFEptr)X1,ldX1);
// P6 = alpha . S2 * T2 in C12
MMH_t H6(F, WH.recLevel-1, 2*WH.Amin-WH.Amax, 2*WH.Amax-WH.Amin, 2*WH.Bmin-WH.Bmax, 2*WH.Bmax-WH.Bmin, 0, 0);
fgemm (F, ta, tb, mr, nr, kr, alpha, X1, ldX1, X2, ldX2, F.zero, C12, ldc, H6);
// S4 = A12 -S2 in X1
fsub(DF,la,ca,(DFCEptr)A12,lda,(DFCEptr)X1,ldX1,(DFEptr)X1,ldX1);
// P3 = alpha . S4*B22 in C11
MMH_t H3(F, WH.recLevel-1, 2*WH.Amin-2*WH.Amax, 2*WH.Amax-2*WH.Amin, WH.Bmin, WH.Bmax, 0, 0);
fgemm (F, ta, tb, mr, nr, kr, alpha, X1, ldX1, B22, ldb, F.zero, C11, ldc, H3);
// P1 = alpha . A11 * B11 in X1
MMH_t H1(F, WH.recLevel-1, WH.Amin, WH.Amax, WH.Bmin, WH.Bmax, 0, 0);
fgemm (F, ta, tb, mr, nr, kr, alpha, A11, lda, B11, ldb, F.zero, X1, nr, H1);
// U2 = P1 + P6 in C12 and
DFElt U2Min, U2Max;
// This test will be optimized out
if (Protected::NeedPreAddReduction(U2Min, U2Max, H1.Outmin, H1.Outmax, H6.Outmin, H6.Outmax, WH)){
freduce (F, mr, nr, X1, nr);
freduce (F, mr, nr, C12, ldc);
}
faddin(DF,mr,nr,(DFCEptr)X1,nr,(DFEptr)C12,ldc);
// U3 = P7 + U2 in C21 and
DFElt U3Min, U3Max;
// This test will be optimized out
if (Protected::NeedPreAddReduction(U3Min, U3Max, U2Min, U2Max, H7.Outmin, H7.Outmax, WH)){
freduce (F, mr, nr, C12, ldc);
freduce (F, mr, nr, C21, ldc);
}
faddin(DF,mr,nr,(DFCEptr)C12,ldc,(DFEptr)C21,ldc);
// U4 = P5 + U2 in C12 and
DFElt U4Min, U4Max;
// This test will be optimized out
if (Protected::NeedPreAddReduction(U4Min, U4Max, U2Min, U2Max, H5.Outmin, H5.Outmax, WH)){
freduce (F, mr, nr, C22, ldc);
freduce (F, mr, nr, C12, ldc);
}
faddin(DF,mr,nr,(DFCEptr)C22,ldc,(DFEptr)C12,ldc);
// U7 = P5 + U3 in C22 and
DFElt U7Min, U7Max;
// This test will be optimized out
if (Protected::NeedPreAddReduction (U7Min,U7Max, U3Min, U3Max, H5.Outmin,H5.Outmax, WH) ){
freduce (F, mr, nr, C21, ldc);
freduce (F, mr, nr, C22, ldc);
}
faddin(DF,mr,nr,(DFCEptr)C21,ldc,(DFEptr)C22,ldc);
// U5 = P3 + U4 in C12
DFElt U5Min, U5Max;
// This test will be optimized out
if (Protected::NeedPreAddReduction (U5Min,U5Max, U4Min, U4Max, H3.Outmin, H3.Outmax, WH) ){
freduce (F, mr, nr, C12, ldc);
freduce (F, mr, nr, C11, ldc);
}
faddin(DF,mr,nr,(DFCEptr)C11,ldc,(DFEptr)C12,ldc);
// T4 = T2 - B21 in X2
fsubin(DF,lb,cb,(DFCEptr)B21,ldb,(DFEptr)X2,ldX2);
// P4 = alpha . A22 * T4 in C11
MMH_t H4(F, WH.recLevel-1, WH.Amin, WH.Amax, 2*WH.Bmin-2*WH.Bmax, 2*WH.Bmax-2*WH.Bmin, 0, 0);
fgemm (F, ta, tb, mr, nr, kr, alpha, A22, lda, X2, ldX2, F.zero, C11, ldc, H4);
fflas_delete (X2);
// U6 = U3 - P4 in C21
DFElt U6Min, U6Max;
// This test will be optimized out
if (Protected::NeedPreSubReduction (U6Min,U6Max, U3Min, U3Max, H4.Outmin,H4.Outmax, WH) ){
freduce (F, mr, nr, C11, ldc);
freduce (F, mr, nr, C21, ldc);
}
fsubin(DF,mr,nr,(DFCEptr)C11,ldc,(DFEptr)C21,ldc);
// P2 = alpha . A12 * B21 in C11
MMH_t H2(F, WH.recLevel-1, WH.Amin, WH.Amax, WH.Bmin, WH.Bmax, 0, 0);
fgemm (F, ta, tb, mr, nr, kr, alpha, A12, lda, B21, ldb, F.zero, C11, ldc, H2);
// U1 = P2 + P1 in C11
DFElt U1Min, U1Max;
// This test will be optimized out
if (Protected::NeedPreAddReduction (U1Min, U1Max, H1.Outmin, H1.Outmax, H2.Outmin,H2.Outmax, WH) ){
freduce (F, mr, nr, X1, nr);
freduce (F, mr, nr, C11, ldc);
}
faddin(DF,mr,nr,(DFCEptr)X1,nr,(DFEptr)C11,ldc);
fflas_delete (X1);
WH.Outmin = std::min (U1Min, std::min (U5Min, std::min (U6Min, U7Min)));
WH.Outmax = std::max (U1Max, std::max (U5Max, std::max (U6Max, U7Max)));
} // Winograd
} // BLAS3
} // FFLAS
#endif // __FFLASFFPACK_fgemm_winograd_INL
|