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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939
|
//------------------------------------------------------------------------------
// GB_jit_kernel_proto.h: prototypes for all JIT kernels
//------------------------------------------------------------------------------
// SuiteSparse:GraphBLAS, Timothy A. Davis, (c) 2017-2025, All Rights Reserved.
// SPDX-License-Identifier: Apache-2.0
//------------------------------------------------------------------------------
#ifndef GB_JIT_KERNEL_PROTO_H
#define GB_JIT_KERNEL_PROTO_H
//------------------------------------------------------------------------------
// JIT kernel prototypes
//------------------------------------------------------------------------------
#define GB_JIT_QUERY_PROTO(query_func) \
bool query_func (uint64_t *hash, int v [3], const char *defn [5], \
void *id, void *term, size_t id_size, size_t term_size)
#define GB_JIT_KERNEL_USER_OP_PROTO(GB_jit_kernel_user_op) \
GrB_Info GB_jit_kernel_user_op (void **user_function, char **defn)
#define GB_JIT_KERNEL_USER_TYPE_PROTO(GB_jit_kernel_user_type) \
GrB_Info GB_jit_kernel_user_type (size_t *user_type_size, char **defn)
#define GB_JIT_KERNEL_ADD_PROTO(GB_jit_kernel_add) \
GrB_Info GB_jit_kernel_add \
( \
GrB_Matrix C, \
const GrB_Matrix M, \
const GrB_Matrix A, \
const GrB_Matrix B, \
const bool Ch_is_Mh, \
const int64_t *restrict C_to_M, \
const int64_t *restrict C_to_A, \
const int64_t *restrict C_to_B, \
const GB_task_struct *restrict TaskList, \
const int C_ntasks, \
const int C_nthreads, \
const int64_t *M_ek_slicing, \
const int M_nthreads, \
const int M_ntasks, \
const int64_t *A_ek_slicing, \
const int A_nthreads, \
const int A_ntasks, \
const int64_t *B_ek_slicing, \
const int B_nthreads, \
const int B_ntasks, \
const bool M_is_A, \
const bool M_is_B, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_APPLY_BIND1ST_PROTO(GB_jit_kernel_apply_bind1st) \
GrB_Info GB_jit_kernel_apply_bind1st \
( \
GB_void *Cx_output, /* Cx and Bx may be aliased */ \
const GB_void *x_input, \
const GB_void *Bx_input, \
const int8_t *restrict Bb, \
const int64_t bnz, \
const int nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_APPLY_BIND2ND_PROTO(GB_jit_kernel_apply_bind2nd) \
GrB_Info GB_jit_kernel_apply_bind2nd \
( \
GB_void *Cx_output, /* Cx and Ax may be aliased */ \
const GB_void *Ax_input, \
const GB_void *y_input, \
const int8_t *restrict Ab, \
const int64_t anz, \
const int nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_APPLY_UNOP_PROTO(GB_jit_kernel_apply_unop) \
GrB_Info GB_jit_kernel_apply_unop \
( \
GB_void *Cx_out, /* Cx and Ax may be aliased */ \
const GrB_Matrix A, \
const void *ythunk, /* for idx unops (op->ytype scalar) */ \
const int64_t *A_ek_slicing, \
const int A_ntasks, \
const int A_nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_AXB_DOT2_PROTO(GB_jit_kernel_AxB_dot2) \
GrB_Info GB_jit_kernel_AxB_dot2 \
( \
GrB_Matrix C, \
const GrB_Matrix M, \
const GrB_Matrix A, \
const int64_t *restrict A_slice, \
const GrB_Matrix B, \
const int64_t *restrict B_slice, \
const int nthreads, \
const int naslice, \
const int nbslice, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_AXB_DOT2N_PROTO(GB_jit_kernel_AxB_dot2n) \
GrB_Info GB_jit_kernel_AxB_dot2n \
( \
GrB_Matrix C, \
const GrB_Matrix M, \
const GrB_Matrix A, \
const int64_t *restrict A_slice, \
const GrB_Matrix B, \
const int64_t *restrict B_slice, \
const int nthreads, \
const int naslice, \
const int nbslice, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_AXB_DOT3_PROTO(GB_jit_kernel_AxB_dot3) \
GrB_Info GB_jit_kernel_AxB_dot3 \
( \
GrB_Matrix C, \
const GrB_Matrix M, \
const GrB_Matrix A, \
const GrB_Matrix B, \
const GB_task_struct *restrict TaskList, \
const int ntasks, \
const int nthreads, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_AXB_DOT4_PROTO(GB_jit_kernel_AxB_dot4) \
GrB_Info GB_jit_kernel_AxB_dot4 \
( \
GrB_Matrix C, \
const GrB_Matrix A, \
const GrB_Matrix B, \
const int64_t *restrict A_slice, \
const int64_t *restrict B_slice, \
const int naslice, \
const int nbslice, \
const int nthreads, \
GB_Werk Werk, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_AXB_SAXBIT_PROTO(GB_jit_kernel_AxB_saxbit) \
GrB_Info GB_jit_kernel_AxB_saxbit \
( \
GrB_Matrix C, \
const GrB_Matrix M, \
const GrB_Matrix A, \
const GrB_Matrix B, \
const int ntasks, \
const int nthreads, \
const int nfine_tasks_per_vector, \
const bool use_coarse_tasks, \
const bool use_atomics, \
const int64_t *M_ek_slicing, \
const int M_nthreads, \
const int M_ntasks, \
const int64_t *restrict A_slice, \
const int64_t *restrict H_slice, \
GB_void *restrict Wcx, \
int8_t *restrict Wf, \
const int nthreads_max, \
double chunk, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_AXB_SAXPY3_PROTO(GB_jit_kernel_AxB_saxpy3) \
GrB_Info GB_jit_kernel_AxB_saxpy3 \
( \
GrB_Matrix C, /* C<any M>=A*B, C sparse or hypersparse */ \
const GrB_Matrix M, \
const bool M_in_place, \
const GrB_Matrix A, \
const GrB_Matrix B, \
GB_saxpy3task_struct *restrict SaxpyTasks, \
const int ntasks, \
const int nfine, \
const int nthreads, \
const int do_sort, \
const int nthreads_max, \
double chunk, \
GB_Werk Werk, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_AXB_SAXPY4_PROTO(GB_jit_kernel_AxB_saxpy4) \
GrB_Info GB_jit_kernel_AxB_saxpy4 \
( \
GrB_Matrix C, \
const GrB_Matrix A, \
const GrB_Matrix B, \
const int ntasks, \
const int nthreads, \
const int nfine_tasks_per_vector, \
const bool use_coarse_tasks, \
const bool use_atomics, \
const int64_t *restrict A_slice, \
const int64_t *restrict H_slice, \
GB_void *restrict Wcx, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_AXB_SAXPY5_PROTO(GB_jit_kernel_AxB_saxpy5) \
GrB_Info GB_jit_kernel_AxB_saxpy5 \
( \
GrB_Matrix C, \
const GrB_Matrix A, \
const GrB_Matrix B, \
const int ntasks, \
const int nthreads, \
const int64_t *restrict B_slice, \
bool cpu_has_avx2, \
bool cpu_has_avx512f, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_BUILD_PROTO(GB_jit_kernel_build) \
GrB_Info GB_jit_kernel_build \
( \
GB_void *restrict Tx_void, \
void *restrict Ti_void, \
const GB_void *restrict Sx_void, \
const int64_t nvals, \
const void *restrict I_work_void, \
const void *restrict K_work_void, \
const int64_t duplicate_entry_input, \
const int64_t *restrict tstart_slice, \
const int64_t *restrict tnz_slice, \
const int nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_COLSCALE_PROTO(GB_jit_kernel_colscale) \
GrB_Info GB_jit_kernel_colscale \
( \
GrB_Matrix C, \
const GrB_Matrix A, \
const GrB_Matrix D, \
const int64_t *A_ek_slicing, \
const int A_ntasks, \
const int A_nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_CONCAT_BITMAP_PROTO(GB_jit_kernel_concat_bitmap) \
GrB_Info GB_jit_kernel_concat_bitmap \
( \
GrB_Matrix C, \
const int64_t cistart, \
const int64_t cvstart, \
const GrB_Matrix A, \
const int nthreads_max, \
double chunk, \
GB_Werk Werk, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_CONCAT_FULL_PROTO(GB_jit_kernel_concat_full) \
GrB_Info GB_jit_kernel_concat_full \
( \
GrB_Matrix C, \
const int64_t cistart, \
const int64_t cvstart, \
const GrB_Matrix A, \
const int A_nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_CONCAT_SPARSE_PROTO(GB_jit_kernel_concat_sparse) \
GrB_Info GB_jit_kernel_concat_sparse \
( \
GrB_Matrix C, \
const int64_t cistart, \
const GrB_Matrix A, \
const void *W_parameter, \
const int64_t *A_ek_slicing, \
const int A_ntasks, \
const int A_nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_CONVERT_S2B_PROTO(GB_jit_kernel_convert_s2b) \
GrB_Info GB_jit_kernel_convert_s2b \
( \
GB_void *restrict Cx_new, \
int8_t *restrict Cb, \
const GrB_Matrix A, \
const int64_t *A_ek_slicing, \
const int A_ntasks, \
const int A_nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_EMULT_02_PROTO(GB_jit_kernel_emult_02) \
GrB_Info GB_jit_kernel_emult_02 \
( \
GrB_Matrix C, \
const GrB_Matrix M, \
const bool Mask_struct, \
const bool Mask_comp, \
const GrB_Matrix A, \
const GrB_Matrix B, \
const uint64_t *restrict Cp_kfirst, \
const int64_t *A_ek_slicing, \
const int A_ntasks, \
const int A_nthreads, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_EMULT_03_PROTO(GB_jit_kernel_emult_03) \
GrB_Info GB_jit_kernel_emult_03 \
( \
GrB_Matrix C, \
const GrB_Matrix M, \
const bool Mask_struct, \
const bool Mask_comp, \
const GrB_Matrix A, \
const GrB_Matrix B, \
const uint64_t *restrict Cp_kfirst, \
const int64_t *B_ek_slicing, \
const int B_ntasks, \
const int B_nthreads, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_EMULT_04_PROTO(GB_jit_kernel_emult_04) \
GrB_Info GB_jit_kernel_emult_04 \
( \
GrB_Matrix C, \
const GrB_Matrix M, \
const bool Mask_struct, \
const GrB_Matrix A, \
const GrB_Matrix B, \
const uint64_t *restrict Cp_kfirst, \
const int64_t *M_ek_slicing, \
const int M_ntasks, \
const int M_nthreads, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_EMULT_08_PROTO(GB_jit_kernel_emult_08) \
GrB_Info GB_jit_kernel_emult_08 \
( \
GrB_Matrix C, \
const GrB_Matrix M, \
const bool Mask_struct, \
const bool Mask_comp, \
const GrB_Matrix A, \
const GrB_Matrix B, \
const int64_t *restrict C_to_M, \
const int64_t *restrict C_to_A, \
const int64_t *restrict C_to_B, \
const GB_task_struct *restrict TaskList, \
const int C_ntasks, \
const int C_nthreads, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_EMULT_BITMAP_PROTO(GB_jit_kernel_emult_bitmap) \
GrB_Info GB_jit_kernel_emult_bitmap \
( \
GrB_Matrix C, \
const GrB_Matrix M, \
const bool Mask_struct, \
const bool Mask_comp, \
const GrB_Matrix A, \
const GrB_Matrix B, \
const int64_t *M_ek_slicing, \
const int M_ntasks, \
const int M_nthreads, \
const int C_nthreads, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_EWISE_FULLA_PROTO(GB_jit_kernel_ewise_fulla) \
GrB_Info GB_jit_kernel_ewise_fulla \
( \
GrB_Matrix C, \
const GrB_Matrix A, \
const GrB_Matrix B, \
const int nthreads, \
const bool A_is_B, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_EWISE_FULLN_PROTO(GB_jit_kernel_ewise_fulln) \
GrB_Info GB_jit_kernel_ewise_fulln \
( \
GrB_Matrix C, \
const GrB_Matrix A, \
const GrB_Matrix B, \
const int nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_REDUCE_PROTO(GB_jit_kernel_reduce) \
GrB_Info GB_jit_kernel_reduce \
( \
GB_void *result, \
const GrB_Matrix A, \
GB_void *restrict Workspace, \
bool *restrict F, \
const int ntasks, \
const int nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_ROWSCALE_PROTO(GB_jit_kernel_rowscale) \
GrB_Info GB_jit_kernel_rowscale \
( \
GrB_Matrix C, \
const GrB_Matrix D, \
const GrB_Matrix B, \
const int nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_SELECT_BITMAP_PROTO(GB_jit_kernel_select_bitmap) \
GrB_Info GB_jit_kernel_select_bitmap \
( \
GrB_Matrix C, \
GrB_Matrix A, \
const GB_void *restrict ythunk, \
const int nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_SELECT_PHASE1_PROTO(GB_jit_kernel_select_phase1) \
GrB_Info GB_jit_kernel_select_phase1 \
( \
GrB_Matrix C, \
uint64_t *restrict Wfirst, \
uint64_t *restrict Wlast, \
const GrB_Matrix A, \
const GB_void *restrict ythunk, \
const int64_t *A_ek_slicing, \
const int A_ntasks, \
const int A_nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_SELECT_PHASE2_PROTO(GB_jit_kernel_select_phase2) \
GrB_Info GB_jit_kernel_select_phase2 \
( \
GrB_Matrix C, \
const uint64_t *restrict Cp_kfirst, \
const GrB_Matrix A, \
const GB_void *restrict ythunk, \
const int64_t *A_ek_slicing, \
const int A_ntasks, \
const int A_nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_SPLIT_BITMAP_PROTO(GB_jit_kernel_split_bitmap) \
GrB_Info GB_jit_kernel_split_bitmap \
( \
GrB_Matrix C, \
GrB_Matrix A, \
const int64_t avstart, \
const int64_t aistart, \
const int C_nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_SPLIT_FULL_PROTO(GB_jit_kernel_split_full) \
GrB_Info GB_jit_kernel_split_full \
( \
GrB_Matrix C, \
GrB_Matrix A, \
const int64_t avstart, \
const int64_t aistart, \
const int C_nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_SPLIT_SPARSE_PROTO(GB_jit_kernel_split_sparse) \
GrB_Info GB_jit_kernel_split_sparse \
( \
GrB_Matrix C, \
GrB_Matrix A, \
const int64_t akstart, \
const int64_t aistart, \
const void *Wp_workspace, \
const int64_t *C_ek_slicing, \
const int C_ntasks, \
const int C_nthreads, \
const GB_callback_struct *restrict my_callback \
)
// used for all subassign kernels:
#define GB_JIT_KERNEL_SUBASSIGN_PROTO(GB_jit_kernel_subassign_any) \
GrB_Info GB_jit_kernel_subassign_any \
( \
GrB_Matrix C, \
const bool C_replace, \
const void *I_input, \
const int64_t ni, \
const int64_t nI, \
const int Ikind, \
const int64_t Icolon [3], \
const void *J_input, \
const int64_t nj, \
const int64_t nJ, \
const int Jkind, \
const int64_t Jcolon [3], \
const GrB_Matrix M, \
const bool Mask_comp, \
const bool Mask_struct, \
const GrB_BinaryOp accum, \
const GrB_Matrix A, \
const void *scalar, \
const GrB_Type scalar_type, \
const GrB_Matrix S, \
const int assign_kind, \
GB_Werk Werk, \
const int nthreads_max, \
double chunk, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_TRANS_BIND1ST_PROTO(GB_jit_kernel_trans_bind1st) \
GrB_Info GB_jit_kernel_trans_bind1st \
( \
GrB_Matrix C, \
const GB_void *x_input, \
const GrB_Matrix A, \
void **Workspaces, \
const int64_t *restrict A_slice, \
const int nworkspaces, \
const int nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_TRANS_BIND2ND_PROTO(GB_jit_kernel_trans_bind2nd) \
GrB_Info GB_jit_kernel_trans_bind2nd \
( \
GrB_Matrix C, \
const GrB_Matrix A, \
const GB_void *y_input, \
void **Workspaces, \
const int64_t *restrict A_slice, \
const int nworkspaces, \
const int nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_TRANS_UNOP_PROTO(GB_jit_kernel_trans_unop) \
GrB_Info GB_jit_kernel_trans_unop \
( \
GrB_Matrix C, \
const GrB_Matrix A, \
void **Workspaces, \
const int64_t *restrict A_slice, \
const int nworkspaces, \
const int nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_UNION_PROTO(GB_jit_kernel_union) \
GrB_Info GB_jit_kernel_union \
( \
GrB_Matrix C, \
const GrB_Matrix M, \
const GrB_Matrix A, \
const GrB_Matrix B, \
const GB_void *alpha_scalar_in, \
const GB_void *beta_scalar_in, \
const bool Ch_is_Mh, \
const int64_t *restrict C_to_M, \
const int64_t *restrict C_to_A, \
const int64_t *restrict C_to_B, \
const GB_task_struct *restrict TaskList, \
const int C_ntasks, \
const int C_nthreads, \
const int64_t *M_ek_slicing, \
const int M_nthreads, \
const int M_ntasks, \
const int64_t *A_ek_slicing, \
const int A_nthreads, \
const int A_ntasks, \
const int64_t *B_ek_slicing, \
const int B_nthreads, \
const int B_ntasks, \
const bool M_is_A, \
const bool M_is_B, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_MASKER_PHASE1_PROTO(GB_jit_kernel_masker1) \
GrB_Info GB_jit_kernel_masker1 \
( \
void *Rp_parameter, \
int64_t *Rnvec_nonempty, \
GB_task_struct *restrict TaskList, \
const int R_ntasks, \
const int R_nthreads, \
const int64_t Rnvec, \
const void *Rh_parameter, \
const int64_t *restrict R_to_M, \
const int64_t *restrict R_to_C, \
const int64_t *restrict R_to_Z, \
const GrB_Matrix M, \
const bool Mask_comp, \
const bool Mask_struct, \
const GrB_Matrix C, \
const GrB_Matrix Z, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_MASKER_PHASE2_PROTO(GB_jit_kernel_masker2) \
GrB_Info GB_jit_kernel_masker2 \
( \
GrB_Matrix R, \
const GB_task_struct *restrict TaskList, \
const int R_ntasks, \
const int R_nthreads, \
const int64_t *restrict R_to_M, \
const int64_t *restrict R_to_C, \
const int64_t *restrict R_to_Z, \
const GrB_Matrix M, \
const bool Mask_comp, \
const bool Mask_struct, \
const GrB_Matrix C, \
const GrB_Matrix Z, \
const int64_t *restrict C_ek_slicing, \
const int C_nthreads, \
const int C_ntasks, \
const int64_t *restrict M_ek_slicing, \
const int M_nthreads, \
const int M_ntasks, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_SUBREF_SPARSE_PROTO(GB_jit_kernel_subref_sparse) \
GrB_Info GB_jit_kernel_subref_sparse \
( \
GrB_Matrix C, \
const GB_task_struct *restrict TaskList, \
const int ntasks, \
const int nthreads, \
const bool post_sort, \
const void *Ihead_input, \
const void *Inext_input, \
const void *Ap_start_input, \
const void *Ap_end_input, \
const int64_t nI, \
const int64_t Icolon [3], \
const GrB_Matrix A, \
const void *I_input, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_BITMAP_SUBREF_PROTO(GB_jit_kernel_bitmap_subref) \
GrB_Info GB_jit_kernel_bitmap_subref \
( \
GrB_Matrix C, \
GrB_Matrix A, \
const void *I_input, \
const int64_t nI, \
const int64_t Icolon [3], \
const void *J_input, \
const int64_t nJ, \
const int64_t Jcolon [3], \
GB_Werk Werk, \
const int nthreads_max, \
double chunk, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_ISO_EXPAND_PROTO(GB_jit_kernel_iso_expand) \
GrB_Info GB_jit_kernel_iso_expand \
( \
void *restrict X, \
const int64_t n, \
const void *restrict scalar, \
const int nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_UNJUMBLE_PROTO(GB_jit_kernel_unjumble) \
GrB_Info GB_jit_kernel_unjumble \
( \
const GrB_Matrix A, \
const int64_t *A_slice, \
const int ntasks, \
const int nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_CONVERT_B2S_PROTO(GB_jit_kernel_convert_b2s) \
GrB_Info GB_jit_kernel_convert_b2s \
( \
const void *Cp_input, \
void *Ci_input, \
void *Cj_input, \
void *Cx_new, \
const GrB_Matrix A, \
const void *W_input, \
const int nthreads, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_KRONER_PROTO(GB_jit_kernel_kroner) \
GrB_Info GB_jit_kernel_kroner \
( \
GrB_Matrix C, \
const GrB_Matrix A, \
const GrB_Matrix B, \
const int nthreads, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_KERNEL_SORT_PROTO(GB_jit_kernel_sort) \
GrB_Info GB_jit_kernel_sort \
( \
GrB_Matrix C, \
int nthreads, \
GB_Werk Werk, \
const GB_callback_struct *restrict my_callback \
)
//------------------------------------------------------------------------------
// CUDA JIT prototypes
//------------------------------------------------------------------------------
#define GB_JIT_CUDA_KERNEL_REDUCE_PROTO(GB_jit_kernel_reduce) \
GrB_Info GB_jit_kernel_reduce \
( \
GB_void *zscalar, \
GrB_Matrix V, \
const GrB_Matrix A, \
cudaStream_t stream, \
int32_t gridsz, \
int32_t blocksz, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_CUDA_KERNEL_ROWSCALE_PROTO(GB_jit_kernel_rowscale) \
GrB_Info GB_jit_kernel_rowscale \
( \
GrB_Matrix C, \
GrB_Matrix D, \
GrB_Matrix B, \
cudaStream_t stream, \
int32_t gridsz, \
int32_t blocksz, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_CUDA_KERNEL_COLSCALE_PROTO(GB_jit_kernel_colscale) \
GrB_Info GB_jit_kernel_colscale \
( \
GrB_Matrix C, \
GrB_Matrix A, \
GrB_Matrix D, \
cudaStream_t stream, \
int32_t gridsz, \
int32_t blocksz, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_CUDA_KERNEL_APPLY_BIND1ST_PROTO(GB_jit_kernel_apply_bind1st) \
GrB_Info GB_jit_kernel_apply_bind1st \
( \
GB_void *Cx, \
const GB_void *scalarx, \
GrB_Matrix B, \
cudaStream_t stream, \
int32_t gridsz, \
int32_t blocksz, \
const GB_callback_struct *restrict my_callback \
) \
#define GB_JIT_CUDA_KERNEL_APPLY_BIND2ND_PROTO(GB_jit_kernel_apply_bind2nd) \
GrB_Info GB_jit_kernel_apply_bind2nd \
( \
GB_void *Cx, \
GrB_Matrix A, \
const GB_void *scalarx, \
cudaStream_t stream, \
int32_t gridsz, \
int32_t blocksz, \
const GB_callback_struct *restrict my_callback \
) \
#define GB_JIT_CUDA_KERNEL_APPLY_UNOP_PROTO(GB_jit_kernel_apply_unop) \
GrB_Info GB_jit_kernel_apply_unop \
( \
GB_void *Cx, \
GrB_Matrix A, \
const GB_void *ythunk, \
cudaStream_t stream, \
int32_t gridsz, \
int32_t blocksz, \
const GB_callback_struct *restrict my_callback \
) \
#define GB_JIT_CUDA_KERNEL_SELECT_BITMAP_PROTO(GB_jit_kernel_select_bitmap) \
GrB_Info GB_jit_kernel_select_bitmap \
( \
GrB_Matrix C, \
GrB_Matrix A, \
const GB_void *ythunk, \
cudaStream_t stream, \
int32_t gridsz, \
int32_t blocksz, \
const GB_callback_struct *restrict my_callback \
) \
#define GB_JIT_CUDA_KERNEL_SELECT_SPARSE_PROTO(GB_jit_kernel_select_sparse) \
GrB_Info GB_jit_kernel_select_sparse \
( \
GrB_Matrix C, \
GrB_Matrix A, \
const GB_void *ythunk, \
cudaStream_t stream, \
int32_t gridsz, \
int32_t blocksz, \
const GB_callback_struct *restrict my_callback \
)
#define GB_JIT_CUDA_KERNEL_DOT3_PROTO(GB_jit_kernel_AxB_dot3) \
GrB_Info GB_jit_kernel_AxB_dot3 \
( \
GrB_Matrix C, \
const GrB_Matrix M, \
const GrB_Matrix A, \
const GrB_Matrix B, \
cudaStream_t stream, \
int device, \
int number_of_sms, \
const void *theta, \
const GB_callback_struct *restrict my_callback \
)
//------------------------------------------------------------------------------
// shorthand macros for GB_prejit.c:
//------------------------------------------------------------------------------
#define JIT_DOT2(g) GB_JIT_KERNEL_AXB_DOT2_PROTO(g) ;
#define JIT_DO2N(g) GB_JIT_KERNEL_AXB_DOT2N_PROTO(g) ;
#define JIT_DOT3(g) GB_JIT_KERNEL_AXB_DOT3_PROTO(g) ;
#define JIT_DOT4(g) GB_JIT_KERNEL_AXB_DOT4_PROTO(g) ;
#define JIT_SAXB(g) GB_JIT_KERNEL_AXB_SAXBIT_PROTO(g) ;
#define JIT_SAX3(g) GB_JIT_KERNEL_AXB_SAXPY3_PROTO(g) ;
#define JIT_SAX4(g) GB_JIT_KERNEL_AXB_SAXPY4_PROTO(g) ;
#define JIT_SAX5(g) GB_JIT_KERNEL_AXB_SAXPY5_PROTO(g) ;
#define JIT_ADD(g) GB_JIT_KERNEL_ADD_PROTO(g) ;
#define JIT_AP1(g) GB_JIT_KERNEL_APPLY_BIND1ST_PROTO(g) ;
#define JIT_AP2(g) GB_JIT_KERNEL_APPLY_BIND2ND_PROTO(g) ;
#define JIT_AP0(g) GB_JIT_KERNEL_APPLY_UNOP_PROTO(g) ;
#define JIT_BLD(g) GB_JIT_KERNEL_BUILD_PROTO(g) ;
#define JIT_COLS(g) GB_JIT_KERNEL_COLSCALE_PROTO(g) ;
#define JIT_CONB(g) GB_JIT_KERNEL_CONCAT_BITMAP_PROTO(g) ;
#define JIT_CONF(g) GB_JIT_KERNEL_CONCAT_FULL_PROTO(g) ;
#define JIT_CONS(g) GB_JIT_KERNEL_CONCAT_SPARSE_PROTO(g) ;
#define JIT_CS2B(g) GB_JIT_KERNEL_CONVERT_S2B_PROTO(g) ;
#define JIT_CB2S(g) GB_JIT_KERNEL_CONVERT_B2S_PROTO(g) ;
#define JIT_EM2(g) GB_JIT_KERNEL_EMULT_02_PROTO(g) ;
#define JIT_EM3(g) GB_JIT_KERNEL_EMULT_03_PROTO(g) ;
#define JIT_EM4(g) GB_JIT_KERNEL_EMULT_04_PROTO(g) ;
#define JIT_EM8(g) GB_JIT_KERNEL_EMULT_08_PROTO(g) ;
#define JIT_EMB(g) GB_JIT_KERNEL_EMULT_BITMAP_PROTO(g) ;
#define JIT_EWFA(g) GB_JIT_KERNEL_EWISE_FULLA_PROTO(g) ;
#define JIT_EWFN(g) GB_JIT_KERNEL_EWISE_FULLN_PROTO(g) ;
#define JIT_RED(g) GB_JIT_KERNEL_REDUCE_PROTO(g) ;
#define JIT_ROWS(g) GB_JIT_KERNEL_ROWSCALE_PROTO(g) ;
#define JIT_SELB(g) GB_JIT_KERNEL_SELECT_BITMAP_PROTO(g) ;
#define JIT_SEL1(g) GB_JIT_KERNEL_SELECT_PHASE1_PROTO(g) ;
#define JIT_SEL2(g) GB_JIT_KERNEL_SELECT_PHASE2_PROTO(g) ;
#define JIT_SPB(g) GB_JIT_KERNEL_SPLIT_BITMAP_PROTO(g) ;
#define JIT_SPF(g) GB_JIT_KERNEL_SPLIT_FULL_PROTO(g) ;
#define JIT_SPS(g) GB_JIT_KERNEL_SPLIT_SPARSE_PROTO(g) ;
#define JIT_SUB(g) GB_JIT_KERNEL_SUBASSIGN_PROTO(g) ;
#define JIT_TR1(g) GB_JIT_KERNEL_TRANS_BIND1ST_PROTO(g) ;
#define JIT_TR2(g) GB_JIT_KERNEL_TRANS_BIND2ND_PROTO(g) ;
#define JIT_TR0(g) GB_JIT_KERNEL_TRANS_UNOP_PROTO(g) ;
#define JIT_UNI(g) GB_JIT_KERNEL_UNION_PROTO(g) ;
#define JIT_UOP(g) GB_JIT_KERNEL_USER_OP_PROTO(g) ;
#define JIT_UTYP(g) GB_JIT_KERNEL_USER_TYPE_PROTO(g) ;
#define JIT_MAS1(g) GB_JIT_KERNEL_MASKER_PHASE1_PROTO(g) ;
#define JIT_MAS2(g) GB_JIT_KERNEL_MASKER_PHASE2_PROTO(g) ;
#define JIT_SREF(g) GB_JIT_KERNEL_SUBREF_SPARSE_PROTO(g) ;
#define JIT_BREF(g) GB_JIT_KERNEL_BITMAP_SUBREF_PROTO(g) ;
#define JIT_ISOE(g) GB_JIT_KERNEL_ISO_EXPAND_PROTO(g) ;
#define JIT_UNJU(g) GB_JIT_KERNEL_UNJUMBLE_PROTO(g) ;
#define JIT_SORT(g) GB_JIT_KERNEL_SORT_PROTO(g) ;
#define JIT_KRON(g) GB_JIT_KERNEL_KRONER_PROTO(g) ;
#define JIT_Q(q) GB_JIT_QUERY_PROTO(q) ;
//------------------------------------------------------------------------------
// shorthand macros for GB_cuda_prejit.c:
//------------------------------------------------------------------------------
#define JIT_CUDA_RED(g) GB_JIT_CUDA_KERNEL_REDUCE_PROTO(g) ;
#define JIT_CUDA_DOT3(g) GB_JIT_CUDA_KERNEL_DOT3_PROTO(g) ;
//------------------------------------------------------------------------------
// GB_GET_CALLBACK: get function pointer from the callback struct for JIT kernel
//------------------------------------------------------------------------------
#undef GB_GET_CALLBACK
#ifdef GB_JIT_RUNTIME
// JIT kernels (CPU and CUDA) require the function pointers
#define GB_GET_CALLBACKS \
GB_abort = my_callback->GB_abort_func
#define GB_GET_CALLBACK(function) \
function ## _f function = my_callback->function ## _func
#else
// PreJIT kernels link against -lgraphblas and do not need function pointers
#define GB_GET_CALLBACKS
#define GB_GET_CALLBACK(function)
#endif
#endif
|