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
|
/* -*- Mode: C; c-basic-offset:4 ; -*- */
/*
* Copyright (c) 2004-2005 The Trustees of Indiana University and Indiana
* University Research and Technology
* Corporation. All rights reserved.
* Copyright (c) 2004-2017 The University of Tennessee and The University
* of Tennessee Research Foundation. All rights
* reserved.
* Copyright (c) 2004-2005 High Performance Computing Center Stuttgart,
* University of Stuttgart. All rights reserved.
* Copyright (c) 2004-2005 The Regents of the University of California.
* All rights reserved.
* Copyright (c) 2006-2007 University of Houston. All rights reserved.
* Copyright (c) 2007-2012 Cisco Systems, Inc. All rights reserved.
* Copyright (c) 2009 Sun Microsystems, Inc. All rights reserved.
* Copyright (c) 2012 Oak Ridge National Labs. All rights reserved.
* Copyright (c) 2015 Research Organization for Information Science
* and Technology (RIST). All rights reserved.
* Copyright (c) 2018-2022 Triad National Security, LLC. All rights
* reserved.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
#include "ompi_config.h"
#include "ompi/group/group.h"
#include "ompi/constants.h"
#include "mpi.h"
/* define class information */
static void ompi_group_construct(ompi_group_t *);
static void ompi_group_destruct(ompi_group_t *);
static int ompi_group_finalize (void);
OBJ_CLASS_INSTANCE(ompi_group_t,
opal_object_t,
ompi_group_construct,
ompi_group_destruct);
/*
* Table for Fortran <-> C group handle conversion
*/
opal_pointer_array_t ompi_group_f_to_c_table = {{0}};
/*
* Predefined group objects
*/
ompi_predefined_group_t ompi_mpi_group_empty = {{{0}}};
ompi_predefined_group_t ompi_mpi_group_null = {{{0}}};
ompi_predefined_group_t *ompi_mpi_group_empty_addr = &ompi_mpi_group_empty;
ompi_predefined_group_t *ompi_mpi_group_null_addr = &ompi_mpi_group_null;
#if OPAL_ENABLE_FT_MPI
ompi_group_t *ompi_group_all_failed_procs = NULL;
/* Access to ompi_group_all_failed_procs must be serialized as the group will
* be hot-substituted when a failure occurs. A typical use case for this group
* is obtaining a snapshot of the group and operate as-if further updates have
* happened at a later time (i.e., threaded access appears as-if serialized in
* an arbitrary order). This can be done with mutex lock; group pointer copy
* (alias); OBJ_RETAIN the group alias; mutex unlock. You are then responsible
* for releasing the group object you retained when not needed anymore (no
* locking needed there). The lock must not be held when entering progress!
*/
opal_mutex_t ompi_group_afp_mutex = OPAL_MUTEX_STATIC_INIT;
#endif /* OPAL_ENABLE_FT_MPI */
/*
* Allocate a new group structure
*/
ompi_group_t *ompi_group_allocate(ompi_group_t *orig_group, int group_size)
{
/* local variables */
ompi_proc_t **procs = calloc (group_size, sizeof (ompi_proc_t *));
ompi_group_t *new_group;
if (NULL == procs) {
return NULL;
}
new_group = ompi_group_allocate_plist_w_procs (orig_group, procs, group_size);
if (NULL == new_group) {
free (procs);
}
return new_group;
}
ompi_group_t *ompi_group_allocate_plist_w_procs (ompi_group_t *orig_group, ompi_proc_t **procs, int group_size)
{
/* local variables */
ompi_group_t * new_group = NULL;
assert (group_size >= 0);
/* create new group group element */
new_group = OBJ_NEW(ompi_group_t);
if (NULL == new_group) {
return NULL;
}
if (0 > new_group->grp_f_to_c_index) {
OBJ_RELEASE (new_group);
return NULL;
}
/*
* Allocate array of (ompi_proc_t *)'s, one for each
* process in the group.
*/
new_group->grp_proc_pointers = procs;
/* set the group size */
new_group->grp_proc_count = group_size;
/* initialize our rank to MPI_UNDEFINED */
new_group->grp_my_rank = MPI_UNDEFINED;
OMPI_GROUP_SET_DENSE(new_group);
if (NULL != orig_group) {
new_group->grp_instance = orig_group->grp_instance;
} else {
new_group->grp_instance = NULL;
}
ompi_group_increment_proc_count (new_group);
return new_group;
}
ompi_group_t *ompi_group_allocate_sporadic(ompi_group_t *orig_group, int group_size)
{
/* local variables */
ompi_group_t *new_group = NULL;
assert (group_size >= 0);
/* create new group group element */
new_group = OBJ_NEW(ompi_group_t);
if( NULL == new_group) {
goto error_exit;
}
if (0 > new_group->grp_f_to_c_index) {
OBJ_RELEASE(new_group);
new_group = NULL;
goto error_exit;
}
/* allocate array of (grp_sporadic_list )'s */
if (0 < group_size) {
new_group->sparse_data.grp_sporadic.grp_sporadic_list =
(struct ompi_group_sporadic_list_t *)malloc
(sizeof(struct ompi_group_sporadic_list_t ) * group_size);
/* non-empty group */
if ( NULL == new_group->sparse_data.grp_sporadic.grp_sporadic_list) {
/* sporadic list allocation failed */
OBJ_RELEASE (new_group);
new_group = NULL;
goto error_exit;
}
}
/* set the group size */
new_group->grp_proc_count = group_size; /* actually it's the number of
elements in the sporadic list*/
/* initialize our rank to MPI_UNDEFINED */
new_group->grp_my_rank = MPI_UNDEFINED;
new_group->grp_proc_pointers = NULL;
new_group->grp_instance = orig_group->grp_instance;
OMPI_GROUP_SET_SPORADIC(new_group);
error_exit:
return new_group;
}
ompi_group_t *ompi_group_allocate_strided(ompi_group_t *orig_group)
{
ompi_group_t *new_group = NULL;
/* create new group group element */
new_group = OBJ_NEW(ompi_group_t);
if( NULL == new_group ) {
goto error_exit;
}
if (0 > new_group->grp_f_to_c_index) {
OBJ_RELEASE(new_group);
new_group = NULL;
goto error_exit;
}
/* initialize our rank to MPI_UNDEFINED */
new_group->grp_my_rank = MPI_UNDEFINED;
new_group->grp_proc_pointers = NULL;
new_group->grp_instance = orig_group->grp_instance;
OMPI_GROUP_SET_STRIDED(new_group);
new_group->sparse_data.grp_strided.grp_strided_stride = -1;
new_group->sparse_data.grp_strided.grp_strided_offset = -1;
new_group->sparse_data.grp_strided.grp_strided_last_element = -1;
error_exit:
/* return */
return new_group;
}
ompi_group_t *ompi_group_allocate_bmap(ompi_group_t *orig_group , int group_size)
{
ompi_group_t *new_group = NULL;
int orig_group_size;
orig_group_size = orig_group->grp_proc_count;
assert (group_size >= 0);
/* create new group group element */
new_group = OBJ_NEW(ompi_group_t);
if( NULL == new_group) {
goto error_exit;
}
if (0 > new_group->grp_f_to_c_index) {
OBJ_RELEASE(new_group);
new_group = NULL;
goto error_exit;
}
/* allocate the unsigned char list */
new_group->sparse_data.grp_bitmap.grp_bitmap_array = (unsigned char *)malloc
(sizeof(unsigned char) * ompi_group_div_ceil(orig_group_size,BSIZE));
new_group->sparse_data.grp_bitmap.grp_bitmap_array_len =
ompi_group_div_ceil(orig_group_size,BSIZE);
new_group->grp_proc_count = group_size;
/* initialize our rank to MPI_UNDEFINED */
new_group->grp_my_rank = MPI_UNDEFINED;
new_group->grp_proc_pointers = NULL;
new_group->grp_instance = orig_group->grp_instance;
OMPI_GROUP_SET_BITMAP(new_group);
error_exit:
/* return */
return new_group;
}
/**
* @brief Allocate a dense group from a group
*
* @param[in] group group
*
* @returns new group pointer on success
* @returns NULL on error
*
* This function duplicates a group. The new group will have a dense process
* table.
*/
ompi_group_t *ompi_group_flatten (ompi_group_t *group, int max_procs)
{
int proc_count = (max_procs > group->grp_proc_count) ? group->grp_proc_count : max_procs;
size_t proc_pointer_array_size = proc_count * sizeof (group->grp_proc_pointers[0]);
ompi_group_t *new_group = OBJ_NEW(ompi_group_t);;
if (NULL == new_group) {
return NULL;
}
if (0 > new_group->grp_f_to_c_index) {
OBJ_RELEASE (new_group);
return NULL;
}
if (0 != proc_count) {
new_group->grp_proc_pointers = malloc (proc_pointer_array_size);
if (OPAL_UNLIKELY(NULL == new_group->grp_proc_pointers)) {
OBJ_RELEASE(new_group);
return NULL;
}
/*
* Allocate array of (ompi_proc_t *)'s, one for each
* process in the group.
*/
if (!OMPI_GROUP_IS_DENSE(group)) {
for (int i = 0 ; i < proc_count ; i++) {
new_group->grp_proc_pointers[i] = ompi_group_peer_lookup (group, i);
}
} else {
memcpy (new_group->grp_proc_pointers, group->grp_proc_pointers, proc_pointer_array_size);
}
}
/* set the group size */
new_group->grp_proc_count = proc_count;
if (group->grp_my_rank >= max_procs) {
/* initialize our rank to MPI_UNDEFINED */
new_group->grp_my_rank = MPI_UNDEFINED;
} else {
/* rank is the same as in the old group */
new_group->grp_my_rank = group->grp_my_rank;
}
new_group->grp_instance = group->grp_instance;
OMPI_GROUP_SET_DENSE(new_group);
ompi_group_increment_proc_count (new_group);
return new_group;
}
/*
* increment the reference count of the proc structures
*/
void ompi_group_increment_proc_count(ompi_group_t *group)
{
ompi_proc_t * proc_pointer;
for (int proc = 0 ; proc < group->grp_proc_count ; ++proc) {
proc_pointer = ompi_group_peer_lookup_existing (group, proc);
if (proc_pointer) {
OBJ_RETAIN(proc_pointer);
}
}
}
/*
* decrement the reference count of the proc structures
*/
void ompi_group_decrement_proc_count(ompi_group_t *group)
{
ompi_proc_t * proc_pointer;
for (int proc = 0 ; proc < group->grp_proc_count ; ++proc) {
proc_pointer = ompi_group_peer_lookup_existing (group, proc);
if (proc_pointer) {
OBJ_RELEASE(proc_pointer);
}
}
}
/*
* group constructor
*/
static void ompi_group_construct(ompi_group_t *new_group)
{
int ret_val;
/* Note that we do *NOT* increase the refcount on all the included
procs here because that is handled at a different level (e.g.,
the proc counts are not decreased during the destructor,
either). */
/* assign entry in fortran <-> c translation array */
ret_val = opal_pointer_array_add(&ompi_group_f_to_c_table, new_group);
new_group->grp_f_to_c_index = ret_val;
new_group->grp_flags = 0;
/* default the sparse values for groups */
new_group->grp_parent_group_ptr = NULL;
}
/*
* group destructor
*/
static void ompi_group_destruct(ompi_group_t *group)
{
/* Note that we do *NOT* decrease the refcount on all the included
procs here because that is handled at a different level (e.g.,
the proc counts are not increased during the constructor,
either). */
#if OMPI_GROUP_SPARSE
if (OMPI_GROUP_IS_DENSE(group))
/* sparse groups do not increment proc reference counters */
#endif
ompi_group_decrement_proc_count (group);
/* release thegrp_proc_pointers memory */
if (NULL != group->grp_proc_pointers) {
free(group->grp_proc_pointers);
}
if (OMPI_GROUP_IS_SPORADIC(group)) {
if (NULL != group->sparse_data.grp_sporadic.grp_sporadic_list) {
free(group->sparse_data.grp_sporadic.grp_sporadic_list);
}
}
if (OMPI_GROUP_IS_BITMAP(group)) {
if (NULL != group->sparse_data.grp_bitmap.grp_bitmap_array) {
free(group->sparse_data.grp_bitmap.grp_bitmap_array);
}
}
if (NULL != group->grp_parent_group_ptr){
OBJ_RELEASE(group->grp_parent_group_ptr);
}
/* reset the ompi_group_f_to_c_table entry - make sure that the
* entry is in the table */
if (NULL != opal_pointer_array_get_item(&ompi_group_f_to_c_table,
group->grp_f_to_c_index)) {
opal_pointer_array_set_item(&ompi_group_f_to_c_table,
group->grp_f_to_c_index, NULL);
}
}
/*
* Initialize OMPI group infrastructure
*/
int ompi_group_init(void)
{
/* initialize ompi_group_f_to_c_table */
OBJ_CONSTRUCT( &ompi_group_f_to_c_table, opal_pointer_array_t);
if( OPAL_SUCCESS != opal_pointer_array_init(&ompi_group_f_to_c_table, 4,
OMPI_FORTRAN_HANDLE_MAX, 16) ) {
return OMPI_ERROR;
}
/* add MPI_GROUP_NULL to table */
OBJ_CONSTRUCT(&ompi_mpi_group_null, ompi_group_t);
ompi_mpi_group_null.group.grp_proc_count = 0;
ompi_mpi_group_null.group.grp_my_rank = MPI_PROC_NULL;
ompi_mpi_group_null.group.grp_proc_pointers = NULL;
ompi_mpi_group_null.group.grp_flags |= OMPI_GROUP_DENSE;
ompi_mpi_group_null.group.grp_flags |= OMPI_GROUP_INTRINSIC;
/* add MPI_GROUP_EMPTY to table */
OBJ_CONSTRUCT(&ompi_mpi_group_empty, ompi_group_t);
ompi_mpi_group_empty.group.grp_proc_count = 0;
ompi_mpi_group_empty.group.grp_my_rank = MPI_UNDEFINED;
ompi_mpi_group_empty.group.grp_proc_pointers = NULL;
ompi_mpi_group_empty.group.grp_flags |= OMPI_GROUP_DENSE;
ompi_mpi_group_empty.group.grp_flags |= OMPI_GROUP_INTRINSIC;
#if OPAL_ENABLE_FT_MPI
/* Setup global list of failed processes */
ompi_group_all_failed_procs = OBJ_NEW(ompi_group_t);
ompi_group_all_failed_procs->grp_proc_count = 0;
ompi_group_all_failed_procs->grp_my_rank = MPI_UNDEFINED;
ompi_group_all_failed_procs->grp_proc_pointers = NULL;
ompi_group_all_failed_procs->grp_flags |= OMPI_GROUP_DENSE;
ompi_group_all_failed_procs->grp_flags |= OMPI_GROUP_INTRINSIC;
#endif
ompi_mpi_instance_append_finalize (ompi_group_finalize);
return OMPI_SUCCESS;
}
/*
* Clean up group infrastructure
*/
static int ompi_group_finalize (void)
{
ompi_mpi_group_null.group.grp_flags = 0;
OBJ_DESTRUCT(&ompi_mpi_group_null);
ompi_mpi_group_null.group.grp_flags = 0;
OBJ_DESTRUCT(&ompi_mpi_group_empty);
#if OPAL_ENABLE_FT_MPI
if( NULL != ompi_group_all_failed_procs ) {
OBJ_RELEASE(ompi_group_all_failed_procs);
ompi_group_all_failed_procs = NULL;
}
#endif
OBJ_DESTRUCT(&ompi_group_f_to_c_table);
return OMPI_SUCCESS;
}
|