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
|
//------------------------------------------------------------------------------
// GB_select_sparse: select entries from a matrix (C is sparse/hypersparse)
//------------------------------------------------------------------------------
// SuiteSparse:GraphBLAS, Timothy A. Davis, (c) 2017-2025, All Rights Reserved.
// SPDX-License-Identifier: Apache-2.0
//------------------------------------------------------------------------------
#include "select/GB_select.h"
#ifndef GBCOMPACT
#include "FactoryKernels/GB_sel__include.h"
#endif
#include "scalar/GB_Scalar_wrap.h"
#include "jitifyer/GB_stringify.h"
#include "slice/factory/GB_ek_slice_merge.h"
#define GB_FREE_WORKSPACE \
{ \
GB_FREE_MEMORY (&Zp, Zp_size) ; \
GB_WERK_POP (Work, uint64_t) ; \
GB_WERK_POP (A_ek_slicing, int64_t) ; \
}
#define GB_FREE_ALL \
{ \
GB_phybix_free (C) ; \
GB_FREE_WORKSPACE ; \
}
GrB_Info GB_select_sparse
(
GrB_Matrix C, // output matrix; empty header on input
const bool C_iso, // if true, construct C as iso
const GrB_IndexUnaryOp op,
const bool flipij, // if true, flip i and j for the op
const GrB_Matrix A, // input matrix
const int64_t ithunk, // input scalar, cast to int64_t
const GB_void *restrict athunk, // same input scalar, but cast to A->type
const GB_void *restrict ythunk, // same input scalar, but cast to op->ytype
GB_Werk Werk
)
{
//--------------------------------------------------------------------------
// check inputs
//--------------------------------------------------------------------------
// C is always an empty header on input. A is never bitmap. It is
// sparse/hypersparse, with one exception: for the DIAG operator, A may be
// sparse, hypersparse, or full.
ASSERT (C != NULL && (C->header_size == 0 || GBNSTATIC)) ;
ASSERT_MATRIX_OK (A, "A input for GB_select_sparse", GB0) ;
ASSERT_INDEXUNARYOP_OK (op, "op for GB_select_sparse", GB0) ;
ASSERT (!GB_IS_BITMAP (A)) ;
ASSERT (GB_IS_SPARSE (A) || GB_IS_HYPERSPARSE (A) || GB_IS_FULL (A)) ;
ASSERT (GB_IMPLIES (op->opcode != GB_DIAG_idxunop_code,
GB_IS_SPARSE (A) || GB_IS_HYPERSPARSE (A))) ;
//--------------------------------------------------------------------------
// declare workspace
//--------------------------------------------------------------------------
GrB_Info info ;
void *Zp = NULL ; size_t Zp_size = 0 ;
GB_WERK_DECLARE (Work, uint64_t) ;
GB_WERK_DECLARE (A_ek_slicing, int64_t) ;
GB_Opcode opcode = op->opcode ;
const bool A_iso = A->iso ;
const GB_Type_code acode = A->type->code ;
//--------------------------------------------------------------------------
// determine the max number of threads to use
//--------------------------------------------------------------------------
int nthreads_max = GB_Context_nthreads_max ( ) ;
double chunk = GB_Context_chunk ( ) ;
//--------------------------------------------------------------------------
// get A: sparse, hypersparse, or full
//--------------------------------------------------------------------------
int64_t anvec = A->nvec ;
bool A_is_hyper = GB_IS_HYPERSPARSE (A) ;
//--------------------------------------------------------------------------
// create the C matrix
//--------------------------------------------------------------------------
int csparsity = (A_is_hyper) ? GxB_HYPERSPARSE : GxB_SPARSE ;
int64_t anz = GB_nnz (A) ;
// determine the p_is_32, j_is_32, and i_is_32 settings for the new matrix
bool Cp_is_32, Cj_is_32, Ci_is_32 ;
GB_determine_pji_is_32 (&Cp_is_32, &Cj_is_32, &Ci_is_32,
csparsity, anz, A->vlen, A->vdim, Werk) ;
GB_OK (GB_new (&C, // sparse or hyper (from A), existing header
A->type, A->vlen, A->vdim, GB_ph_calloc, A->is_csc,
csparsity, A->hyper_switch, A->plen, Cp_is_32, Cj_is_32, Ci_is_32)) ;
ASSERT (csparsity == GB_sparsity (C)) ;
ASSERT (Cp_is_32 == C->p_is_32) ;
ASSERT (Cj_is_32 == C->j_is_32) ;
ASSERT (Ci_is_32 == C->i_is_32) ;
Cp_is_32 = C->p_is_32 ;
Cj_is_32 = C->j_is_32 ;
Ci_is_32 = C->i_is_32 ;
bool Aj_is_32 = A->j_is_32 ;
GB_Type_code ajcode = Aj_is_32 ? GB_UINT32_code : GB_UINT64_code ;
GB_Type_code cjcode = Cj_is_32 ? GB_UINT32_code : GB_UINT64_code ;
size_t cpsize = Cp_is_32 ? sizeof (uint32_t) : sizeof (uint64_t) ;
if (A_is_hyper)
{
// C->h is a deep copy of A->h
GB_cast_int (C->h, cjcode, A->h, ajcode, A->nvec, nthreads_max) ;
}
C->nvec = A->nvec ;
C->nvals = 0 ;
C->magic = GB_MAGIC ;
// C->Y is not yet constructed
ASSERT (C->Y == NULL) ;
ASSERT_MATRIX_OK (C, "C initialized as empty for GB_selector", GB0) ;
ASSERT (C->i == NULL) ;
ASSERT (C->x == NULL) ;
C->iso = C_iso ;
//--------------------------------------------------------------------------
// slice the entries for each task
//--------------------------------------------------------------------------
int A_ntasks, A_nthreads ;
double work = 8*anvec + ((opcode == GB_DIAG_idxunop_code) ? 0 : anz) ;
GB_SLICE_MATRIX_WORK2 (A, 8, work, anz) ;
//--------------------------------------------------------------------------
// allocate workspace for each task
//--------------------------------------------------------------------------
GB_WERK_PUSH (Work, 3*A_ntasks, uint64_t) ;
if (Work == NULL)
{
// out of memory
GB_FREE_ALL ;
return (GrB_OUT_OF_MEMORY) ;
}
uint64_t *restrict Wfirst = Work ;
uint64_t *restrict Wlast = Work + A_ntasks ;
uint64_t *restrict Cp_kfirst = Work + A_ntasks * 2 ;
//--------------------------------------------------------------------------
// allocate workspace for phase1
//--------------------------------------------------------------------------
// phase1 counts the number of live entries in each vector of A. The
// result is computed in Cp, where Cp [k] is the number of live entries in
// the kth vector of A. Zp [k] is the location of the A(i,k) entry, for
// positional operators.
bool op_is_positional = GB_IS_INDEXUNARYOP_CODE_POSITIONAL (opcode) ;
if (op_is_positional)
{
// allocate Zp
Zp = GB_MALLOC_MEMORY (C->plen + 1, cpsize, &Zp_size) ;
if (Zp == NULL)
{
// out of memory
GB_FREE_ALL ;
return (GrB_OUT_OF_MEMORY) ;
}
}
//==========================================================================
// phase1: count the live entries in each column
//==========================================================================
info = GrB_NO_VALUE ;
if (op_is_positional || opcode == GB_NONZOMBIE_idxunop_code)
{
//----------------------------------------------------------------------
// positional ops or nonzombie phase1 do not depend on the values
//----------------------------------------------------------------------
// no JIT worker needed for these operators
GB_OK (GB_select_positional_phase1 (C, Zp, Wfirst, Wlast, A, ithunk,
op, A_ek_slicing, A_ntasks, A_nthreads)) ;
}
else
{
//----------------------------------------------------------------------
// entry selectors depend on the values in phase1
//----------------------------------------------------------------------
ASSERT (!A_iso || opcode == GB_USER_idxunop_code) ;
ASSERT ((opcode >= GB_VALUENE_idxunop_code
&& opcode <= GB_VALUELE_idxunop_code)
|| (opcode == GB_USER_idxunop_code)) ;
#ifndef GBCOMPACT
GB_IF_FACTORY_KERNELS_ENABLED
{
//------------------------------------------------------------------
// via the factory kernel (includes user-defined ops)
//------------------------------------------------------------------
// define the worker for the switch factory
#define GB_sel1(opname,aname) GB (_sel_phase1_ ## opname ## aname)
#define GB_SEL_WORKER(opname,aname) \
{ \
info = GB_sel1 (opname, aname) (C, Wfirst, Wlast, A, \
ythunk, A_ek_slicing, A_ntasks, A_nthreads) ; \
} \
break ;
// launch the switch factory
#include "select/factory/GB_select_entry_factory.c"
#undef GB_SEL_WORKER
}
#endif
//----------------------------------------------------------------------
// via the JIT or PreJIT kernel
//----------------------------------------------------------------------
if (info == GrB_NO_VALUE)
{
info = GB_select_phase1_jit (C, Wfirst, Wlast, A, ythunk, op,
flipij, A_ek_slicing, A_ntasks, A_nthreads) ;
}
//----------------------------------------------------------------------
// via the generic kernel
//----------------------------------------------------------------------
if (info == GrB_NO_VALUE)
{
// generic entry selector, phase1
GBURBLE ("(generic select) ") ;
info = GB_select_generic_phase1 (C, Wfirst, Wlast, A, flipij,
ythunk, op, A_ek_slicing, A_ntasks, A_nthreads) ;
}
}
GB_OK (info) ; // check for out-of-memory or other failures in phase1
//==========================================================================
// phase1b: cumulative sum and allocate C
//==========================================================================
//--------------------------------------------------------------------------
// finalize Cp, cumulative sum of Cp, and compute Cp_kfirst
//--------------------------------------------------------------------------
GB_Cp_DECLARE (Cp, ) ; GB_Cp_PTR (Cp, C) ;
if (!op_is_positional)
{
// GB_select_positional_phase1 finalizes Cp in the
// select/factory/GB_select_positional_phase1_template.c. This phase
// is only needed for entry-style selectors, done by
// select/template/GB_select_entry_phase1_template.c:
GB_ek_slice_merge1 (Cp, Cp_is_32, Wfirst, Wlast, A_ek_slicing,
A_ntasks) ;
}
int64_t nvec_nonempty ;
GB_cumsum (Cp, Cp_is_32, anvec, &nvec_nonempty, A_nthreads, Werk) ;
GB_nvec_nonempty_set (C, nvec_nonempty) ;
GB_ek_slice_merge2 (Cp_kfirst, Cp, Cp_is_32, Wfirst, Wlast, A_ek_slicing,
A_ntasks) ;
//--------------------------------------------------------------------------
// allocate new space for the compacted C->i and C->x
//--------------------------------------------------------------------------
uint64_t cnz = GB_IGET (Cp, anvec) ;
GB_OK (GB_bix_alloc (C, cnz, csparsity, false, true, C_iso)) ;
C->jumbled = A->jumbled ;
C->nvals = cnz ;
ASSERT (C->iso == C_iso) ;
//--------------------------------------------------------------------------
// set the iso value of C
//--------------------------------------------------------------------------
if (C_iso)
{
// The pattern of C is computed by the worker below.
GB_select_iso (C->x, opcode, athunk, A->x, A->type->size) ;
}
//==========================================================================
// phase2: select the entries
//==========================================================================
info = GrB_NO_VALUE ;
if (op_is_positional || (opcode == GB_NONZOMBIE_idxunop_code && A_iso))
{
//----------------------------------------------------------------------
// positional ops do not depend on the values
//----------------------------------------------------------------------
// no JIT worker needed for these operators
info = GB_select_positional_phase2 (C, Zp, Cp_kfirst, A, flipij,
ithunk, op, A_ek_slicing, A_ntasks, A_nthreads) ;
}
else
{
//----------------------------------------------------------------------
// entry selectors depend on the values in phase2
//----------------------------------------------------------------------
ASSERT (!A_iso || opcode == GB_USER_idxunop_code) ;
ASSERT ((opcode >= GB_VALUENE_idxunop_code &&
opcode <= GB_VALUELE_idxunop_code)
|| (opcode == GB_NONZOMBIE_idxunop_code && !A_iso)
|| (opcode == GB_USER_idxunop_code)) ;
#ifndef GBCOMPACT
GB_IF_FACTORY_KERNELS_ENABLED
{
//------------------------------------------------------------------
// via the factory kernel
//------------------------------------------------------------------
// define the worker for the switch factory
#define GB_SELECT_PHASE2
#define GB_sel2(opname,aname) GB (_sel_phase2_ ## opname ## aname)
#define GB_SEL_WORKER(opname,aname) \
{ \
info = GB_sel2 (opname, aname) (C, Cp_kfirst, A, ythunk, \
A_ek_slicing, A_ntasks, A_nthreads) ; \
} \
break ;
// launch the switch factory
#include "select/factory/GB_select_entry_factory.c"
}
#endif
//----------------------------------------------------------------------
// via the JIT or PreJIT kernel
//----------------------------------------------------------------------
if (info == GrB_NO_VALUE)
{
info = GB_select_phase2_jit (C, Cp_kfirst, A, flipij, ythunk, op,
A_ek_slicing, A_ntasks, A_nthreads) ;
}
//----------------------------------------------------------------------
// via the generic kernel
//----------------------------------------------------------------------
if (info == GrB_NO_VALUE)
{
// generic entry selector, phase2
info = GB_select_generic_phase2 (C, Cp_kfirst, A, flipij, ythunk,
op, A_ek_slicing, A_ntasks, A_nthreads) ;
}
}
GB_OK (info) ; // phase2 cannot fail, but check just in case
//==========================================================================
// finalize the result, free workspace, and return result
//==========================================================================
GB_FREE_WORKSPACE ;
ASSERT_MATRIX_OK (C, "C before hyper_prune for GB_selector", GB0) ;
GB_OK (GB_hyper_prune (C, Werk)) ;
ASSERT_MATRIX_OK (C, "C output for GB_selector", GB0) ;
return (GrB_SUCCESS) ;
}
|