Lines Matching refs:scratch
20 private CeedScalar *restrict V, local CeedScalar *restrict scratch) { in ContractX1d() argument
23 scratch[item_id_x] = *U; in ContractX1d()
29 *V += B[i + item_id_x * P_1D] * scratch[i]; // Contract x direction in ContractX1d()
39 … private CeedScalar *restrict V, local CeedScalar *restrict scratch) { in ContractTransposeX1d() argument
42 scratch[item_id_x] = *U; in ContractTransposeX1d()
48 *V += B[item_id_x + i * P_1D] * scratch[i]; // Contract x direction in ContractTransposeX1d()
58 …t CeedScalar *restrict s_B, private CeedScalar *restrict r_V, local CeedScalar *restrict scratch) { in Interp1d() argument
60 ContractX1d(P_1D, Q_1D, r_U + comp, s_B, r_V + comp, scratch); in Interp1d()
68 …t CeedScalar *restrict s_B, private CeedScalar *restrict r_V, local CeedScalar *restrict scratch) { in InterpTranspose1d() argument
70 ContractTransposeX1d(P_1D, Q_1D, r_U + comp, s_B, r_V + comp, scratch); in InterpTranspose1d()
78 …t CeedScalar *restrict s_G, private CeedScalar *restrict r_V, local CeedScalar *restrict scratch) { in Grad1d() argument
80 ContractX1d(P_1D, Q_1D, r_U + comp, s_G, r_V + comp, scratch); in Grad1d()
88 …t CeedScalar *restrict s_G, private CeedScalar *restrict r_V, local CeedScalar *restrict scratch) { in GradTranspose1d() argument
90 ContractTransposeX1d(P_1D, Q_1D, r_U + comp, s_G, r_V + comp, scratch); in GradTranspose1d()
110 private CeedScalar *restrict V, local CeedScalar *restrict scratch) { in ContractX2d() argument
114 scratch[item_id_x + item_id_y * T_1D] = *U; in ContractX2d()
120 *V += B[i + item_id_x * P_1D] * scratch[i + item_id_y * T_1D]; // Contract x direction in ContractX2d()
130 private CeedScalar *restrict V, local CeedScalar *restrict scratch) { in ContractY2d() argument
134 scratch[item_id_x + item_id_y * T_1D] = *U; in ContractY2d()
140 *V += B[i + item_id_y * P_1D] * scratch[item_id_x + i * T_1D]; // Contract y direction in ContractY2d()
150 … private CeedScalar *restrict V, local CeedScalar *restrict scratch) { in ContractTransposeY2d() argument
154 scratch[item_id_x + item_id_y * T_1D] = *U; in ContractTransposeY2d()
160 *V += B[item_id_y + i * P_1D] * scratch[item_id_x + i * T_1D]; // Contract y direction in ContractTransposeY2d()
170 … private CeedScalar *restrict V, local CeedScalar *restrict scratch) { in ContractTransposeX2d() argument
174 scratch[item_id_x + item_id_y * T_1D] = *U; in ContractTransposeX2d()
180 *V += B[item_id_x + i * P_1D] * scratch[i + item_id_y * T_1D]; // Contract x direction in ContractTransposeX2d()
190 … private CeedScalar *restrict V, local CeedScalar *restrict scratch) { in ContractTransposeAddX2d() argument
194 scratch[item_id_x + item_id_y * T_1D] = *U; in ContractTransposeAddX2d()
199 *V += B[item_id_x + i * P_1D] * scratch[i + item_id_y * T_1D]; // Contract x direction in ContractTransposeAddX2d()
209 …t CeedScalar *restrict s_B, private CeedScalar *restrict r_V, local CeedScalar *restrict scratch) { in InterpTensor2d() argument
213 ContractX2d(P_1D, Q_1D, r_U + comp, s_B, r_t, scratch); in InterpTensor2d()
214 ContractY2d(P_1D, Q_1D, r_t, s_B, r_V + comp, scratch); in InterpTensor2d()
222 …t CeedScalar *restrict s_B, private CeedScalar *restrict r_V, local CeedScalar *restrict scratch) { in InterpTransposeTensor2d() argument
226 ContractTransposeY2d(P_1D, Q_1D, r_U + comp, s_B, r_t, scratch); in InterpTransposeTensor2d()
227 ContractTransposeX2d(P_1D, Q_1D, r_t, s_B, r_V + comp, scratch); in InterpTransposeTensor2d()
236 local CeedScalar *restrict scratch) { in GradTensor2d() argument
240 ContractX2d(P_1D, Q_1D, r_U + comp, s_G, r_t, scratch); in GradTensor2d()
241 ContractY2d(P_1D, Q_1D, r_t, s_B, r_V + comp + 0 * NUM_COMP, scratch); in GradTensor2d()
242 ContractX2d(P_1D, Q_1D, r_U + comp, s_B, r_t, scratch); in GradTensor2d()
243 ContractY2d(P_1D, Q_1D, r_t, s_G, r_V + comp + 1 * NUM_COMP, scratch); in GradTensor2d()
252 local CeedScalar *restrict scratch) { in GradTransposeTensor2d() argument
256 ContractTransposeY2d(P_1D, Q_1D, r_U + comp + 0 * NUM_COMP, s_B, r_t, scratch); in GradTransposeTensor2d()
257 ContractTransposeX2d(P_1D, Q_1D, r_t, s_G, r_V + comp, scratch); in GradTransposeTensor2d()
258 ContractTransposeY2d(P_1D, Q_1D, r_U + comp + 1 * NUM_COMP, s_G, r_t, scratch); in GradTransposeTensor2d()
259 ContractTransposeAddX2d(P_1D, Q_1D, r_t, s_B, r_V + comp, scratch); in GradTransposeTensor2d()
281 private CeedScalar *restrict V, local CeedScalar *restrict scratch) { in ContractX3d() argument
291 scratch[item_id_x + item_id_y * T_1D] = U[k]; in ContractX3d()
297 V[k] += r_B[i] * scratch[i + item_id_y * T_1D]; // Contract x direction in ContractX3d()
308 private CeedScalar *restrict V, local CeedScalar *restrict scratch) { in ContractY3d() argument
318 scratch[item_id_x + item_id_y * T_1D] = U[k]; in ContractY3d()
324 V[k] += r_B[i] * scratch[item_id_x + i * T_1D]; // Contract y direction in ContractY3d()
335 private CeedScalar *restrict V, local CeedScalar *restrict scratch) { in ContractZ3d() argument
353 … private CeedScalar *restrict V, local CeedScalar *restrict scratch) { in ContractTransposeZ3d() argument
371 … private CeedScalar *restrict V, local CeedScalar *restrict scratch) { in ContractTransposeY3d() argument
381 scratch[item_id_x + item_id_y * T_1D] = U[k]; in ContractTransposeY3d()
387 V[k] += r_B[i] * scratch[item_id_x + i * T_1D]; // Contract y direction in ContractTransposeY3d()
398 … private CeedScalar *restrict V, local CeedScalar *restrict scratch) { in ContractTransposeAddY3d() argument
408 scratch[item_id_x + item_id_y * T_1D] = U[k]; in ContractTransposeAddY3d()
412 V[k] += r_B[i] * scratch[item_id_x + i * T_1D]; // Contract y direction in ContractTransposeAddY3d()
423 … private CeedScalar *restrict V, local CeedScalar *restrict scratch) { in ContractTransposeX3d() argument
433 scratch[item_id_x + item_id_y * T_1D] = U[k]; in ContractTransposeX3d()
438 V[k] += r_B[i] * scratch[i + item_id_y * T_1D]; // Contract x direction in ContractTransposeX3d()
449 … private CeedScalar *restrict V, local CeedScalar *restrict scratch) { in ContractTransposeAddX3d() argument
459 scratch[item_id_x + item_id_y * T_1D] = U[k]; in ContractTransposeAddX3d()
464 V[k] += r_B[i] * scratch[i + item_id_y * T_1D]; // Contract x direction in ContractTransposeAddX3d()
475 …t CeedScalar *restrict s_B, private CeedScalar *restrict r_V, local CeedScalar *restrict scratch) { in InterpTensor3d() argument
480 ContractX3d(P_1D, Q_1D, r_U + comp * P_1D, s_B, r_t1, scratch); in InterpTensor3d()
481 ContractY3d(P_1D, Q_1D, r_t1, s_B, r_t2, scratch); in InterpTensor3d()
482 ContractZ3d(P_1D, Q_1D, r_t2, s_B, r_V + comp * Q_1D, scratch); in InterpTensor3d()
490 …t CeedScalar *restrict s_B, private CeedScalar *restrict r_V, local CeedScalar *restrict scratch) { in InterpTransposeTensor3d() argument
495 ContractTransposeZ3d(P_1D, Q_1D, r_U + comp * Q_1D, s_B, r_t1, scratch); in InterpTransposeTensor3d()
496 ContractTransposeY3d(P_1D, Q_1D, r_t1, s_B, r_t2, scratch); in InterpTransposeTensor3d()
497 ContractTransposeX3d(P_1D, Q_1D, r_t2, s_B, r_V + comp * P_1D, scratch); in InterpTransposeTensor3d()
506 local CeedScalar *restrict scratch) { in GradTensor3d() argument
511 ContractX3d(P_1D, Q_1D, r_U + comp * P_1D, s_G, r_t1, scratch); in GradTensor3d()
512 ContractY3d(P_1D, Q_1D, r_t1, s_B, r_t2, scratch); in GradTensor3d()
513 ContractZ3d(P_1D, Q_1D, r_t2, s_B, r_V + comp * Q_1D + 0 * NUM_COMP * Q_1D, scratch); in GradTensor3d()
514 ContractX3d(P_1D, Q_1D, r_U + comp * P_1D, s_B, r_t1, scratch); in GradTensor3d()
515 ContractY3d(P_1D, Q_1D, r_t1, s_G, r_t2, scratch); in GradTensor3d()
516 ContractZ3d(P_1D, Q_1D, r_t2, s_B, r_V + comp * Q_1D + 1 * NUM_COMP * Q_1D, scratch); in GradTensor3d()
517 ContractX3d(P_1D, Q_1D, r_U + comp * P_1D, s_B, r_t1, scratch); in GradTensor3d()
518 ContractY3d(P_1D, Q_1D, r_t1, s_B, r_t2, scratch); in GradTensor3d()
519 ContractZ3d(P_1D, Q_1D, r_t2, s_G, r_V + comp * Q_1D + 2 * NUM_COMP * Q_1D, scratch); in GradTensor3d()
528 local CeedScalar *restrict scratch) { in GradTransposeTensor3d() argument
533 ContractTransposeZ3d(P_1D, Q_1D, r_U + comp * Q_1D + 0 * NUM_COMP * Q_1D, s_B, r_t1, scratch); in GradTransposeTensor3d()
534 ContractTransposeY3d(P_1D, Q_1D, r_t1, s_B, r_t2, scratch); in GradTransposeTensor3d()
535 ContractTransposeX3d(P_1D, Q_1D, r_t2, s_G, r_V + comp * P_1D, scratch); in GradTransposeTensor3d()
536 ContractTransposeZ3d(P_1D, Q_1D, r_U + comp * Q_1D + 1 * NUM_COMP * Q_1D, s_B, r_t1, scratch); in GradTransposeTensor3d()
537 ContractTransposeY3d(P_1D, Q_1D, r_t1, s_G, r_t2, scratch); in GradTransposeTensor3d()
538 ContractTransposeAddX3d(P_1D, Q_1D, r_t2, s_B, r_V + comp * P_1D, scratch); in GradTransposeTensor3d()
539 ContractTransposeZ3d(P_1D, Q_1D, r_U + comp * Q_1D + 2 * NUM_COMP * Q_1D, s_G, r_t1, scratch); in GradTransposeTensor3d()
540 ContractTransposeY3d(P_1D, Q_1D, r_t1, s_B, r_t2, scratch); in GradTransposeTensor3d()
541 ContractTransposeAddX3d(P_1D, Q_1D, r_t2, s_B, r_V + comp * P_1D, scratch); in GradTransposeTensor3d()
550 local CeedScalar *restrict scratch) { in GradTensorCollocated3d() argument
555 ContractX3d(P_1D, Q_1D, r_U + comp * P_1D, s_B, r_t1, scratch); in GradTensorCollocated3d()
556 ContractY3d(P_1D, Q_1D, r_t1, s_B, r_t2, scratch); in GradTensorCollocated3d()
557 ContractZ3d(P_1D, Q_1D, r_t2, s_B, r_t1, scratch); in GradTensorCollocated3d()
558 ContractX3d(Q_1D, Q_1D, r_t1, s_G, r_V + comp * Q_1D + 0 * NUM_COMP * Q_1D, scratch); in GradTensorCollocated3d()
559 ContractY3d(Q_1D, Q_1D, r_t1, s_G, r_V + comp * Q_1D + 1 * NUM_COMP * Q_1D, scratch); in GradTensorCollocated3d()
560 ContractZ3d(Q_1D, Q_1D, r_t1, s_G, r_V + comp * Q_1D + 2 * NUM_COMP * Q_1D, scratch); in GradTensorCollocated3d()
569 … private CeedScalar *restrict r_V, local CeedScalar *restrict scratch) { in GradTransposeTensorCollocated3d() argument
574 ContractTransposeZ3d(Q_1D, Q_1D, r_U + comp * Q_1D + 2 * NUM_COMP * Q_1D, s_G, r_t2, scratch); in GradTransposeTensorCollocated3d()
575 … ContractTransposeAddY3d(Q_1D, Q_1D, r_U + comp * Q_1D + 1 * NUM_COMP * Q_1D, s_G, r_t2, scratch); in GradTransposeTensorCollocated3d()
576 … ContractTransposeAddX3d(Q_1D, Q_1D, r_U + comp * Q_1D + 0 * NUM_COMP * Q_1D, s_G, r_t2, scratch); in GradTransposeTensorCollocated3d()
577 ContractTransposeZ3d(P_1D, Q_1D, r_t2, s_B, r_t1, scratch); in GradTransposeTensorCollocated3d()
578 ContractTransposeY3d(P_1D, Q_1D, r_t1, s_B, r_t2, scratch); in GradTransposeTensorCollocated3d()
579 ContractTransposeX3d(P_1D, Q_1D, r_t2, s_B, r_V + comp * P_1D, scratch); in GradTransposeTensorCollocated3d()