1 /* SPDX-License-Identifier: BSD-3-Clause 2 * Copyright(c) 2010-2014 Intel Corporation 3 */ 4 #include <string.h> 5 #include <stdint.h> 6 #include <errno.h> 7 #include <stdarg.h> 8 #include <stdio.h> 9 #include <sys/queue.h> 10 11 #include <rte_log.h> 12 #include <rte_branch_prediction.h> 13 #include <rte_common.h> 14 #include <rte_memory.h> 15 #include <rte_malloc.h> 16 #include <rte_memcpy.h> 17 #include <rte_eal.h> 18 #include <rte_eal_memconfig.h> 19 #include <rte_per_lcore.h> 20 #include <rte_string_fns.h> 21 #include <rte_errno.h> 22 #include <rte_rwlock.h> 23 #include <rte_spinlock.h> 24 #include <rte_hash.h> 25 #include <assert.h> 26 #include <rte_jhash.h> 27 #include <rte_tailq.h> 28 29 #include "rte_lpm6.h" 30 31 #define RTE_LPM6_TBL24_NUM_ENTRIES (1 << 24) 32 #define RTE_LPM6_TBL8_GROUP_NUM_ENTRIES 256 33 #define RTE_LPM6_TBL8_MAX_NUM_GROUPS (1 << 21) 34 35 #define RTE_LPM6_VALID_EXT_ENTRY_BITMASK 0xA0000000 36 #define RTE_LPM6_LOOKUP_SUCCESS 0x20000000 37 #define RTE_LPM6_TBL8_BITMASK 0x001FFFFF 38 39 #define ADD_FIRST_BYTE 3 40 #define LOOKUP_FIRST_BYTE 4 41 #define BYTE_SIZE 8 42 #define BYTES2_SIZE 16 43 44 #define RULE_HASH_TABLE_EXTRA_SPACE 64 45 #define TBL24_IND UINT32_MAX 46 47 #define lpm6_tbl8_gindex next_hop 48 49 /** Flags for setting an entry as valid/invalid. */ 50 enum valid_flag { 51 INVALID = 0, 52 VALID 53 }; 54 55 TAILQ_HEAD(rte_lpm6_list, rte_tailq_entry); 56 57 static struct rte_tailq_elem rte_lpm6_tailq = { 58 .name = "RTE_LPM6", 59 }; 60 EAL_REGISTER_TAILQ(rte_lpm6_tailq) 61 62 /** Tbl entry structure. It is the same for both tbl24 and tbl8 */ 63 struct rte_lpm6_tbl_entry { 64 uint32_t next_hop: 21; /**< Next hop / next table to be checked. */ 65 uint32_t depth :8; /**< Rule depth. */ 66 67 /* Flags. */ 68 uint32_t valid :1; /**< Validation flag. */ 69 uint32_t valid_group :1; /**< Group validation flag. */ 70 uint32_t ext_entry :1; /**< External entry. */ 71 }; 72 73 /** Rules tbl entry structure. */ 74 struct rte_lpm6_rule { 75 uint8_t ip[RTE_LPM6_IPV6_ADDR_SIZE]; /**< Rule IP address. */ 76 uint32_t next_hop; /**< Rule next hop. */ 77 uint8_t depth; /**< Rule depth. */ 78 }; 79 80 /** Rules tbl entry key. */ 81 struct rte_lpm6_rule_key { 82 uint8_t ip[RTE_LPM6_IPV6_ADDR_SIZE]; /**< Rule IP address. */ 83 uint8_t depth; /**< Rule depth. */ 84 }; 85 86 /* Header of tbl8 */ 87 struct rte_lpm_tbl8_hdr { 88 uint32_t owner_tbl_ind; /**< owner table: TBL24_IND if owner is tbl24, 89 * otherwise index of tbl8 90 */ 91 uint32_t owner_entry_ind; /**< index of the owner table entry where 92 * pointer to the tbl8 is stored 93 */ 94 uint32_t ref_cnt; /**< table reference counter */ 95 }; 96 97 /** LPM6 structure. */ 98 struct rte_lpm6 { 99 /* LPM metadata. */ 100 char name[RTE_LPM6_NAMESIZE]; /**< Name of the lpm. */ 101 uint32_t max_rules; /**< Max number of rules. */ 102 uint32_t used_rules; /**< Used rules so far. */ 103 uint32_t number_tbl8s; /**< Number of tbl8s to allocate. */ 104 105 /* LPM Tables. */ 106 struct rte_hash *rules_tbl; /**< LPM rules. */ 107 struct rte_lpm6_tbl_entry tbl24[RTE_LPM6_TBL24_NUM_ENTRIES] 108 __rte_cache_aligned; /**< LPM tbl24 table. */ 109 110 uint32_t *tbl8_pool; /**< pool of indexes of free tbl8s */ 111 uint32_t tbl8_pool_pos; /**< current position in the tbl8 pool */ 112 113 struct rte_lpm_tbl8_hdr *tbl8_hdrs; /* array of tbl8 headers */ 114 115 struct rte_lpm6_tbl_entry tbl8[0] 116 __rte_cache_aligned; /**< LPM tbl8 table. */ 117 }; 118 119 /* 120 * Takes an array of uint8_t (IPv6 address) and masks it using the depth. 121 * It leaves untouched one bit per unit in the depth variable 122 * and set the rest to 0. 123 */ 124 static inline void 125 ip6_mask_addr(uint8_t *ip, uint8_t depth) 126 { 127 int16_t part_depth, mask; 128 int i; 129 130 part_depth = depth; 131 132 for (i = 0; i < RTE_LPM6_IPV6_ADDR_SIZE; i++) { 133 if (part_depth < BYTE_SIZE && part_depth >= 0) { 134 mask = (uint16_t)(~(UINT8_MAX >> part_depth)); 135 ip[i] = (uint8_t)(ip[i] & mask); 136 } else if (part_depth < 0) 137 ip[i] = 0; 138 139 part_depth -= BYTE_SIZE; 140 } 141 } 142 143 /* copy ipv6 address */ 144 static inline void 145 ip6_copy_addr(uint8_t *dst, const uint8_t *src) 146 { 147 rte_memcpy(dst, src, RTE_LPM6_IPV6_ADDR_SIZE); 148 } 149 150 /* 151 * LPM6 rule hash function 152 * 153 * It's used as a hash function for the rte_hash 154 * containing rules 155 */ 156 static inline uint32_t 157 rule_hash(const void *data, __rte_unused uint32_t data_len, 158 uint32_t init_val) 159 { 160 return rte_jhash(data, sizeof(struct rte_lpm6_rule_key), init_val); 161 } 162 163 /* 164 * Init pool of free tbl8 indexes 165 */ 166 static void 167 tbl8_pool_init(struct rte_lpm6 *lpm) 168 { 169 uint32_t i; 170 171 /* put entire range of indexes to the tbl8 pool */ 172 for (i = 0; i < lpm->number_tbl8s; i++) 173 lpm->tbl8_pool[i] = i; 174 175 lpm->tbl8_pool_pos = 0; 176 } 177 178 /* 179 * Get an index of a free tbl8 from the pool 180 */ 181 static inline uint32_t 182 tbl8_get(struct rte_lpm6 *lpm, uint32_t *tbl8_ind) 183 { 184 if (lpm->tbl8_pool_pos == lpm->number_tbl8s) 185 /* no more free tbl8 */ 186 return -ENOSPC; 187 188 /* next index */ 189 *tbl8_ind = lpm->tbl8_pool[lpm->tbl8_pool_pos++]; 190 return 0; 191 } 192 193 /* 194 * Put an index of a free tbl8 back to the pool 195 */ 196 static inline uint32_t 197 tbl8_put(struct rte_lpm6 *lpm, uint32_t tbl8_ind) 198 { 199 if (lpm->tbl8_pool_pos == 0) 200 /* pool is full */ 201 return -ENOSPC; 202 203 lpm->tbl8_pool[--lpm->tbl8_pool_pos] = tbl8_ind; 204 return 0; 205 } 206 207 /* 208 * Returns number of tbl8s available in the pool 209 */ 210 static inline uint32_t 211 tbl8_available(struct rte_lpm6 *lpm) 212 { 213 return lpm->number_tbl8s - lpm->tbl8_pool_pos; 214 } 215 216 /* 217 * Init a rule key. 218 * note that ip must be already masked 219 */ 220 static inline void 221 rule_key_init(struct rte_lpm6_rule_key *key, uint8_t *ip, uint8_t depth) 222 { 223 ip6_copy_addr(key->ip, ip); 224 key->depth = depth; 225 } 226 227 /* 228 * Rebuild the entire LPM tree by reinserting all rules 229 */ 230 static void 231 rebuild_lpm(struct rte_lpm6 *lpm) 232 { 233 uint64_t next_hop; 234 struct rte_lpm6_rule_key *rule_key; 235 uint32_t iter = 0; 236 237 while (rte_hash_iterate(lpm->rules_tbl, (void *) &rule_key, 238 (void **) &next_hop, &iter) >= 0) 239 rte_lpm6_add(lpm, rule_key->ip, rule_key->depth, 240 (uint32_t) next_hop); 241 } 242 243 /* 244 * Allocates memory for LPM object 245 */ 246 struct rte_lpm6 * 247 rte_lpm6_create(const char *name, int socket_id, 248 const struct rte_lpm6_config *config) 249 { 250 char mem_name[RTE_LPM6_NAMESIZE]; 251 struct rte_lpm6 *lpm = NULL; 252 struct rte_tailq_entry *te; 253 uint64_t mem_size; 254 struct rte_lpm6_list *lpm_list; 255 struct rte_hash *rules_tbl = NULL; 256 uint32_t *tbl8_pool = NULL; 257 struct rte_lpm_tbl8_hdr *tbl8_hdrs = NULL; 258 259 lpm_list = RTE_TAILQ_CAST(rte_lpm6_tailq.head, rte_lpm6_list); 260 261 RTE_BUILD_BUG_ON(sizeof(struct rte_lpm6_tbl_entry) != sizeof(uint32_t)); 262 263 /* Check user arguments. */ 264 if ((name == NULL) || (socket_id < -1) || (config == NULL) || 265 (config->max_rules == 0) || 266 config->number_tbl8s > RTE_LPM6_TBL8_MAX_NUM_GROUPS) { 267 rte_errno = EINVAL; 268 return NULL; 269 } 270 271 /* create rules hash table */ 272 snprintf(mem_name, sizeof(mem_name), "LRH_%s", name); 273 struct rte_hash_parameters rule_hash_tbl_params = { 274 .entries = config->max_rules * 1.2 + 275 RULE_HASH_TABLE_EXTRA_SPACE, 276 .key_len = sizeof(struct rte_lpm6_rule_key), 277 .hash_func = rule_hash, 278 .hash_func_init_val = 0, 279 .name = mem_name, 280 .reserved = 0, 281 .socket_id = socket_id, 282 .extra_flag = 0 283 }; 284 285 rules_tbl = rte_hash_create(&rule_hash_tbl_params); 286 if (rules_tbl == NULL) { 287 RTE_LOG(ERR, LPM, "LPM rules hash table allocation failed: %s (%d)", 288 rte_strerror(rte_errno), rte_errno); 289 goto fail_wo_unlock; 290 } 291 292 /* allocate tbl8 indexes pool */ 293 tbl8_pool = rte_malloc(NULL, 294 sizeof(uint32_t) * config->number_tbl8s, 295 RTE_CACHE_LINE_SIZE); 296 if (tbl8_pool == NULL) { 297 RTE_LOG(ERR, LPM, "LPM tbl8 pool allocation failed: %s (%d)", 298 rte_strerror(rte_errno), rte_errno); 299 rte_errno = ENOMEM; 300 goto fail_wo_unlock; 301 } 302 303 /* allocate tbl8 headers */ 304 tbl8_hdrs = rte_malloc(NULL, 305 sizeof(struct rte_lpm_tbl8_hdr) * config->number_tbl8s, 306 RTE_CACHE_LINE_SIZE); 307 if (tbl8_hdrs == NULL) { 308 RTE_LOG(ERR, LPM, "LPM tbl8 headers allocation failed: %s (%d)", 309 rte_strerror(rte_errno), rte_errno); 310 rte_errno = ENOMEM; 311 goto fail_wo_unlock; 312 } 313 314 snprintf(mem_name, sizeof(mem_name), "LPM_%s", name); 315 316 /* Determine the amount of memory to allocate. */ 317 mem_size = sizeof(*lpm) + (sizeof(lpm->tbl8[0]) * 318 RTE_LPM6_TBL8_GROUP_NUM_ENTRIES * config->number_tbl8s); 319 320 rte_mcfg_tailq_write_lock(); 321 322 /* Guarantee there's no existing */ 323 TAILQ_FOREACH(te, lpm_list, next) { 324 lpm = (struct rte_lpm6 *) te->data; 325 if (strncmp(name, lpm->name, RTE_LPM6_NAMESIZE) == 0) 326 break; 327 } 328 lpm = NULL; 329 if (te != NULL) { 330 rte_errno = EEXIST; 331 goto fail; 332 } 333 334 /* allocate tailq entry */ 335 te = rte_zmalloc("LPM6_TAILQ_ENTRY", sizeof(*te), 0); 336 if (te == NULL) { 337 RTE_LOG(ERR, LPM, "Failed to allocate tailq entry!\n"); 338 rte_errno = ENOMEM; 339 goto fail; 340 } 341 342 /* Allocate memory to store the LPM data structures. */ 343 lpm = rte_zmalloc_socket(mem_name, (size_t)mem_size, 344 RTE_CACHE_LINE_SIZE, socket_id); 345 346 if (lpm == NULL) { 347 RTE_LOG(ERR, LPM, "LPM memory allocation failed\n"); 348 rte_free(te); 349 rte_errno = ENOMEM; 350 goto fail; 351 } 352 353 /* Save user arguments. */ 354 lpm->max_rules = config->max_rules; 355 lpm->number_tbl8s = config->number_tbl8s; 356 strlcpy(lpm->name, name, sizeof(lpm->name)); 357 lpm->rules_tbl = rules_tbl; 358 lpm->tbl8_pool = tbl8_pool; 359 lpm->tbl8_hdrs = tbl8_hdrs; 360 361 /* init the stack */ 362 tbl8_pool_init(lpm); 363 364 te->data = (void *) lpm; 365 366 TAILQ_INSERT_TAIL(lpm_list, te, next); 367 rte_mcfg_tailq_write_unlock(); 368 return lpm; 369 370 fail: 371 rte_mcfg_tailq_write_unlock(); 372 373 fail_wo_unlock: 374 rte_free(tbl8_hdrs); 375 rte_free(tbl8_pool); 376 rte_hash_free(rules_tbl); 377 378 return NULL; 379 } 380 381 /* 382 * Find an existing lpm table and return a pointer to it. 383 */ 384 struct rte_lpm6 * 385 rte_lpm6_find_existing(const char *name) 386 { 387 struct rte_lpm6 *l = NULL; 388 struct rte_tailq_entry *te; 389 struct rte_lpm6_list *lpm_list; 390 391 lpm_list = RTE_TAILQ_CAST(rte_lpm6_tailq.head, rte_lpm6_list); 392 393 rte_mcfg_tailq_read_lock(); 394 TAILQ_FOREACH(te, lpm_list, next) { 395 l = (struct rte_lpm6 *) te->data; 396 if (strncmp(name, l->name, RTE_LPM6_NAMESIZE) == 0) 397 break; 398 } 399 rte_mcfg_tailq_read_unlock(); 400 401 if (te == NULL) { 402 rte_errno = ENOENT; 403 return NULL; 404 } 405 406 return l; 407 } 408 409 /* 410 * Deallocates memory for given LPM table. 411 */ 412 void 413 rte_lpm6_free(struct rte_lpm6 *lpm) 414 { 415 struct rte_lpm6_list *lpm_list; 416 struct rte_tailq_entry *te; 417 418 /* Check user arguments. */ 419 if (lpm == NULL) 420 return; 421 422 lpm_list = RTE_TAILQ_CAST(rte_lpm6_tailq.head, rte_lpm6_list); 423 424 rte_mcfg_tailq_write_lock(); 425 426 /* find our tailq entry */ 427 TAILQ_FOREACH(te, lpm_list, next) { 428 if (te->data == (void *) lpm) 429 break; 430 } 431 432 if (te != NULL) 433 TAILQ_REMOVE(lpm_list, te, next); 434 435 rte_mcfg_tailq_write_unlock(); 436 437 rte_free(lpm->tbl8_hdrs); 438 rte_free(lpm->tbl8_pool); 439 rte_hash_free(lpm->rules_tbl); 440 rte_free(lpm); 441 rte_free(te); 442 } 443 444 /* Find a rule */ 445 static inline int 446 rule_find_with_key(struct rte_lpm6 *lpm, 447 const struct rte_lpm6_rule_key *rule_key, 448 uint32_t *next_hop) 449 { 450 uint64_t hash_val; 451 int ret; 452 453 /* lookup for a rule */ 454 ret = rte_hash_lookup_data(lpm->rules_tbl, (const void *) rule_key, 455 (void **) &hash_val); 456 if (ret >= 0) { 457 *next_hop = (uint32_t) hash_val; 458 return 1; 459 } 460 461 return 0; 462 } 463 464 /* Find a rule */ 465 static int 466 rule_find(struct rte_lpm6 *lpm, uint8_t *ip, uint8_t depth, 467 uint32_t *next_hop) 468 { 469 struct rte_lpm6_rule_key rule_key; 470 471 /* init a rule key */ 472 rule_key_init(&rule_key, ip, depth); 473 474 return rule_find_with_key(lpm, &rule_key, next_hop); 475 } 476 477 /* 478 * Checks if a rule already exists in the rules table and updates 479 * the nexthop if so. Otherwise it adds a new rule if enough space is available. 480 * 481 * Returns: 482 * 0 - next hop of existed rule is updated 483 * 1 - new rule successfully added 484 * <0 - error 485 */ 486 static inline int 487 rule_add(struct rte_lpm6 *lpm, uint8_t *ip, uint8_t depth, uint32_t next_hop) 488 { 489 int ret, rule_exist; 490 struct rte_lpm6_rule_key rule_key; 491 uint32_t unused; 492 493 /* init a rule key */ 494 rule_key_init(&rule_key, ip, depth); 495 496 /* Scan through rule list to see if rule already exists. */ 497 rule_exist = rule_find_with_key(lpm, &rule_key, &unused); 498 499 /* 500 * If rule does not exist check if there is space to add a new rule to 501 * this rule group. If there is no space return error. 502 */ 503 if (!rule_exist && lpm->used_rules == lpm->max_rules) 504 return -ENOSPC; 505 506 /* add the rule or update rules next hop */ 507 ret = rte_hash_add_key_data(lpm->rules_tbl, &rule_key, 508 (void *)(uintptr_t) next_hop); 509 if (ret < 0) 510 return ret; 511 512 /* Increment the used rules counter for this rule group. */ 513 if (!rule_exist) { 514 lpm->used_rules++; 515 return 1; 516 } 517 518 return 0; 519 } 520 521 /* 522 * Function that expands a rule across the data structure when a less-generic 523 * one has been added before. It assures that every possible combination of bits 524 * in the IP address returns a match. 525 */ 526 static void 527 expand_rule(struct rte_lpm6 *lpm, uint32_t tbl8_gindex, uint8_t old_depth, 528 uint8_t new_depth, uint32_t next_hop, uint8_t valid) 529 { 530 uint32_t tbl8_group_end, tbl8_gindex_next, j; 531 532 tbl8_group_end = tbl8_gindex + RTE_LPM6_TBL8_GROUP_NUM_ENTRIES; 533 534 struct rte_lpm6_tbl_entry new_tbl8_entry = { 535 .valid = valid, 536 .valid_group = valid, 537 .depth = new_depth, 538 .next_hop = next_hop, 539 .ext_entry = 0, 540 }; 541 542 for (j = tbl8_gindex; j < tbl8_group_end; j++) { 543 if (!lpm->tbl8[j].valid || (lpm->tbl8[j].ext_entry == 0 544 && lpm->tbl8[j].depth <= old_depth)) { 545 546 lpm->tbl8[j] = new_tbl8_entry; 547 548 } else if (lpm->tbl8[j].ext_entry == 1) { 549 550 tbl8_gindex_next = lpm->tbl8[j].lpm6_tbl8_gindex 551 * RTE_LPM6_TBL8_GROUP_NUM_ENTRIES; 552 expand_rule(lpm, tbl8_gindex_next, old_depth, new_depth, 553 next_hop, valid); 554 } 555 } 556 } 557 558 /* 559 * Init a tbl8 header 560 */ 561 static inline void 562 init_tbl8_header(struct rte_lpm6 *lpm, uint32_t tbl_ind, 563 uint32_t owner_tbl_ind, uint32_t owner_entry_ind) 564 { 565 struct rte_lpm_tbl8_hdr *tbl_hdr = &lpm->tbl8_hdrs[tbl_ind]; 566 tbl_hdr->owner_tbl_ind = owner_tbl_ind; 567 tbl_hdr->owner_entry_ind = owner_entry_ind; 568 tbl_hdr->ref_cnt = 0; 569 } 570 571 /* 572 * Calculate index to the table based on the number and position 573 * of the bytes being inspected in this step. 574 */ 575 static uint32_t 576 get_bitshift(const uint8_t *ip, uint8_t first_byte, uint8_t bytes) 577 { 578 uint32_t entry_ind, i; 579 int8_t bitshift; 580 581 entry_ind = 0; 582 for (i = first_byte; i < (uint32_t)(first_byte + bytes); i++) { 583 bitshift = (int8_t)((bytes - i)*BYTE_SIZE); 584 585 if (bitshift < 0) 586 bitshift = 0; 587 entry_ind = entry_ind | ip[i-1] << bitshift; 588 } 589 590 return entry_ind; 591 } 592 593 /* 594 * Simulate adding a new route to the LPM counting number 595 * of new tables that will be needed 596 * 597 * It returns 0 on success, or 1 if 598 * the process needs to be continued by calling the function again. 599 */ 600 static inline int 601 simulate_add_step(struct rte_lpm6 *lpm, struct rte_lpm6_tbl_entry *tbl, 602 struct rte_lpm6_tbl_entry **next_tbl, const uint8_t *ip, 603 uint8_t bytes, uint8_t first_byte, uint8_t depth, 604 uint32_t *need_tbl_nb) 605 { 606 uint32_t entry_ind; 607 uint8_t bits_covered; 608 uint32_t next_tbl_ind; 609 610 /* 611 * Calculate index to the table based on the number and position 612 * of the bytes being inspected in this step. 613 */ 614 entry_ind = get_bitshift(ip, first_byte, bytes); 615 616 /* Number of bits covered in this step */ 617 bits_covered = (uint8_t)((bytes+first_byte-1)*BYTE_SIZE); 618 619 if (depth <= bits_covered) { 620 *need_tbl_nb = 0; 621 return 0; 622 } 623 624 if (tbl[entry_ind].valid == 0 || tbl[entry_ind].ext_entry == 0) { 625 /* from this point on a new table is needed on each level 626 * that is not covered yet 627 */ 628 depth -= bits_covered; 629 uint32_t cnt = depth >> 3; /* depth / BYTE_SIZE */ 630 if (depth & 7) /* 0b00000111 */ 631 /* if depth % 8 > 0 then one more table is needed 632 * for those last bits 633 */ 634 cnt++; 635 636 *need_tbl_nb = cnt; 637 return 0; 638 } 639 640 next_tbl_ind = tbl[entry_ind].lpm6_tbl8_gindex; 641 *next_tbl = &(lpm->tbl8[next_tbl_ind * 642 RTE_LPM6_TBL8_GROUP_NUM_ENTRIES]); 643 *need_tbl_nb = 0; 644 return 1; 645 } 646 647 /* 648 * Partially adds a new route to the data structure (tbl24+tbl8s). 649 * It returns 0 on success, a negative number on failure, or 1 if 650 * the process needs to be continued by calling the function again. 651 */ 652 static inline int 653 add_step(struct rte_lpm6 *lpm, struct rte_lpm6_tbl_entry *tbl, 654 uint32_t tbl_ind, struct rte_lpm6_tbl_entry **next_tbl, 655 uint32_t *next_tbl_ind, uint8_t *ip, uint8_t bytes, 656 uint8_t first_byte, uint8_t depth, uint32_t next_hop, 657 uint8_t is_new_rule) 658 { 659 uint32_t entry_ind, tbl_range, tbl8_group_start, tbl8_group_end, i; 660 uint32_t tbl8_gindex; 661 uint8_t bits_covered; 662 int ret; 663 664 /* 665 * Calculate index to the table based on the number and position 666 * of the bytes being inspected in this step. 667 */ 668 entry_ind = get_bitshift(ip, first_byte, bytes); 669 670 /* Number of bits covered in this step */ 671 bits_covered = (uint8_t)((bytes+first_byte-1)*BYTE_SIZE); 672 673 /* 674 * If depth if smaller than this number (ie this is the last step) 675 * expand the rule across the relevant positions in the table. 676 */ 677 if (depth <= bits_covered) { 678 tbl_range = 1 << (bits_covered - depth); 679 680 for (i = entry_ind; i < (entry_ind + tbl_range); i++) { 681 if (!tbl[i].valid || (tbl[i].ext_entry == 0 && 682 tbl[i].depth <= depth)) { 683 684 struct rte_lpm6_tbl_entry new_tbl_entry = { 685 .next_hop = next_hop, 686 .depth = depth, 687 .valid = VALID, 688 .valid_group = VALID, 689 .ext_entry = 0, 690 }; 691 692 tbl[i] = new_tbl_entry; 693 694 } else if (tbl[i].ext_entry == 1) { 695 696 /* 697 * If tbl entry is valid and extended calculate the index 698 * into next tbl8 and expand the rule across the data structure. 699 */ 700 tbl8_gindex = tbl[i].lpm6_tbl8_gindex * 701 RTE_LPM6_TBL8_GROUP_NUM_ENTRIES; 702 expand_rule(lpm, tbl8_gindex, depth, depth, 703 next_hop, VALID); 704 } 705 } 706 707 /* update tbl8 rule reference counter */ 708 if (tbl_ind != TBL24_IND && is_new_rule) 709 lpm->tbl8_hdrs[tbl_ind].ref_cnt++; 710 711 return 0; 712 } 713 /* 714 * If this is not the last step just fill one position 715 * and calculate the index to the next table. 716 */ 717 else { 718 /* If it's invalid a new tbl8 is needed */ 719 if (!tbl[entry_ind].valid) { 720 /* get a new table */ 721 ret = tbl8_get(lpm, &tbl8_gindex); 722 if (ret != 0) 723 return -ENOSPC; 724 725 /* invalidate all new tbl8 entries */ 726 tbl8_group_start = tbl8_gindex * 727 RTE_LPM6_TBL8_GROUP_NUM_ENTRIES; 728 memset(&lpm->tbl8[tbl8_group_start], 0, 729 RTE_LPM6_TBL8_GROUP_NUM_ENTRIES * 730 sizeof(struct rte_lpm6_tbl_entry)); 731 732 /* init the new table's header: 733 * save the reference to the owner table 734 */ 735 init_tbl8_header(lpm, tbl8_gindex, tbl_ind, entry_ind); 736 737 /* reference to a new tbl8 */ 738 struct rte_lpm6_tbl_entry new_tbl_entry = { 739 .lpm6_tbl8_gindex = tbl8_gindex, 740 .depth = 0, 741 .valid = VALID, 742 .valid_group = VALID, 743 .ext_entry = 1, 744 }; 745 746 tbl[entry_ind] = new_tbl_entry; 747 748 /* update the current table's reference counter */ 749 if (tbl_ind != TBL24_IND) 750 lpm->tbl8_hdrs[tbl_ind].ref_cnt++; 751 } 752 /* 753 * If it's valid but not extended the rule that was stored 754 * here needs to be moved to the next table. 755 */ 756 else if (tbl[entry_ind].ext_entry == 0) { 757 /* get a new tbl8 index */ 758 ret = tbl8_get(lpm, &tbl8_gindex); 759 if (ret != 0) 760 return -ENOSPC; 761 762 tbl8_group_start = tbl8_gindex * 763 RTE_LPM6_TBL8_GROUP_NUM_ENTRIES; 764 tbl8_group_end = tbl8_group_start + 765 RTE_LPM6_TBL8_GROUP_NUM_ENTRIES; 766 767 struct rte_lpm6_tbl_entry tbl_entry = { 768 .next_hop = tbl[entry_ind].next_hop, 769 .depth = tbl[entry_ind].depth, 770 .valid = VALID, 771 .valid_group = VALID, 772 .ext_entry = 0 773 }; 774 775 /* Populate new tbl8 with tbl value. */ 776 for (i = tbl8_group_start; i < tbl8_group_end; i++) 777 lpm->tbl8[i] = tbl_entry; 778 779 /* init the new table's header: 780 * save the reference to the owner table 781 */ 782 init_tbl8_header(lpm, tbl8_gindex, tbl_ind, entry_ind); 783 784 /* 785 * Update tbl entry to point to new tbl8 entry. Note: The 786 * ext_flag and tbl8_index need to be updated simultaneously, 787 * so assign whole structure in one go. 788 */ 789 struct rte_lpm6_tbl_entry new_tbl_entry = { 790 .lpm6_tbl8_gindex = tbl8_gindex, 791 .depth = 0, 792 .valid = VALID, 793 .valid_group = VALID, 794 .ext_entry = 1, 795 }; 796 797 tbl[entry_ind] = new_tbl_entry; 798 799 /* update the current table's reference counter */ 800 if (tbl_ind != TBL24_IND) 801 lpm->tbl8_hdrs[tbl_ind].ref_cnt++; 802 } 803 804 *next_tbl_ind = tbl[entry_ind].lpm6_tbl8_gindex; 805 *next_tbl = &(lpm->tbl8[*next_tbl_ind * 806 RTE_LPM6_TBL8_GROUP_NUM_ENTRIES]); 807 } 808 809 return 1; 810 } 811 812 /* 813 * Simulate adding a route to LPM 814 * 815 * Returns: 816 * 0 on success 817 * -ENOSPC not enough tbl8 left 818 */ 819 static int 820 simulate_add(struct rte_lpm6 *lpm, const uint8_t *masked_ip, uint8_t depth) 821 { 822 struct rte_lpm6_tbl_entry *tbl; 823 struct rte_lpm6_tbl_entry *tbl_next = NULL; 824 int ret, i; 825 826 /* number of new tables needed for a step */ 827 uint32_t need_tbl_nb; 828 /* total number of new tables needed */ 829 uint32_t total_need_tbl_nb; 830 831 /* Inspect the first three bytes through tbl24 on the first step. */ 832 ret = simulate_add_step(lpm, lpm->tbl24, &tbl_next, masked_ip, 833 ADD_FIRST_BYTE, 1, depth, &need_tbl_nb); 834 total_need_tbl_nb = need_tbl_nb; 835 /* 836 * Inspect one by one the rest of the bytes until 837 * the process is completed. 838 */ 839 for (i = ADD_FIRST_BYTE; i < RTE_LPM6_IPV6_ADDR_SIZE && ret == 1; i++) { 840 tbl = tbl_next; 841 ret = simulate_add_step(lpm, tbl, &tbl_next, masked_ip, 1, 842 (uint8_t)(i + 1), depth, &need_tbl_nb); 843 total_need_tbl_nb += need_tbl_nb; 844 } 845 846 if (tbl8_available(lpm) < total_need_tbl_nb) 847 /* not enough tbl8 to add a rule */ 848 return -ENOSPC; 849 850 return 0; 851 } 852 853 /* 854 * Add a route 855 */ 856 int 857 rte_lpm6_add(struct rte_lpm6 *lpm, const uint8_t *ip, uint8_t depth, 858 uint32_t next_hop) 859 { 860 struct rte_lpm6_tbl_entry *tbl; 861 struct rte_lpm6_tbl_entry *tbl_next = NULL; 862 /* init to avoid compiler warning */ 863 uint32_t tbl_next_num = 123456; 864 int status; 865 uint8_t masked_ip[RTE_LPM6_IPV6_ADDR_SIZE]; 866 int i; 867 868 /* Check user arguments. */ 869 if ((lpm == NULL) || (depth < 1) || (depth > RTE_LPM6_MAX_DEPTH)) 870 return -EINVAL; 871 872 /* Copy the IP and mask it to avoid modifying user's input data. */ 873 ip6_copy_addr(masked_ip, ip); 874 ip6_mask_addr(masked_ip, depth); 875 876 /* Simulate adding a new route */ 877 int ret = simulate_add(lpm, masked_ip, depth); 878 if (ret < 0) 879 return ret; 880 881 /* Add the rule to the rule table. */ 882 int is_new_rule = rule_add(lpm, masked_ip, depth, next_hop); 883 /* If there is no space available for new rule return error. */ 884 if (is_new_rule < 0) 885 return is_new_rule; 886 887 /* Inspect the first three bytes through tbl24 on the first step. */ 888 tbl = lpm->tbl24; 889 status = add_step(lpm, tbl, TBL24_IND, &tbl_next, &tbl_next_num, 890 masked_ip, ADD_FIRST_BYTE, 1, depth, next_hop, 891 is_new_rule); 892 assert(status >= 0); 893 894 /* 895 * Inspect one by one the rest of the bytes until 896 * the process is completed. 897 */ 898 for (i = ADD_FIRST_BYTE; i < RTE_LPM6_IPV6_ADDR_SIZE && status == 1; i++) { 899 tbl = tbl_next; 900 status = add_step(lpm, tbl, tbl_next_num, &tbl_next, 901 &tbl_next_num, masked_ip, 1, (uint8_t)(i + 1), 902 depth, next_hop, is_new_rule); 903 assert(status >= 0); 904 } 905 906 return status; 907 } 908 909 /* 910 * Takes a pointer to a table entry and inspect one level. 911 * The function returns 0 on lookup success, ENOENT if no match was found 912 * or 1 if the process needs to be continued by calling the function again. 913 */ 914 static inline int 915 lookup_step(const struct rte_lpm6 *lpm, const struct rte_lpm6_tbl_entry *tbl, 916 const struct rte_lpm6_tbl_entry **tbl_next, const uint8_t *ip, 917 uint8_t first_byte, uint32_t *next_hop) 918 { 919 uint32_t tbl8_index, tbl_entry; 920 921 /* Take the integer value from the pointer. */ 922 tbl_entry = *(const uint32_t *)tbl; 923 924 /* If it is valid and extended we calculate the new pointer to return. */ 925 if ((tbl_entry & RTE_LPM6_VALID_EXT_ENTRY_BITMASK) == 926 RTE_LPM6_VALID_EXT_ENTRY_BITMASK) { 927 928 tbl8_index = ip[first_byte-1] + 929 ((tbl_entry & RTE_LPM6_TBL8_BITMASK) * 930 RTE_LPM6_TBL8_GROUP_NUM_ENTRIES); 931 932 *tbl_next = &lpm->tbl8[tbl8_index]; 933 934 return 1; 935 } else { 936 /* If not extended then we can have a match. */ 937 *next_hop = ((uint32_t)tbl_entry & RTE_LPM6_TBL8_BITMASK); 938 return (tbl_entry & RTE_LPM6_LOOKUP_SUCCESS) ? 0 : -ENOENT; 939 } 940 } 941 942 /* 943 * Looks up an IP 944 */ 945 int 946 rte_lpm6_lookup(const struct rte_lpm6 *lpm, const uint8_t *ip, 947 uint32_t *next_hop) 948 { 949 const struct rte_lpm6_tbl_entry *tbl; 950 const struct rte_lpm6_tbl_entry *tbl_next = NULL; 951 int status; 952 uint8_t first_byte; 953 uint32_t tbl24_index; 954 955 /* DEBUG: Check user input arguments. */ 956 if ((lpm == NULL) || (ip == NULL) || (next_hop == NULL)) 957 return -EINVAL; 958 959 first_byte = LOOKUP_FIRST_BYTE; 960 tbl24_index = (ip[0] << BYTES2_SIZE) | (ip[1] << BYTE_SIZE) | ip[2]; 961 962 /* Calculate pointer to the first entry to be inspected */ 963 tbl = &lpm->tbl24[tbl24_index]; 964 965 do { 966 /* Continue inspecting following levels until success or failure */ 967 status = lookup_step(lpm, tbl, &tbl_next, ip, first_byte++, next_hop); 968 tbl = tbl_next; 969 } while (status == 1); 970 971 return status; 972 } 973 974 /* 975 * Looks up a group of IP addresses 976 */ 977 int 978 rte_lpm6_lookup_bulk_func(const struct rte_lpm6 *lpm, 979 uint8_t ips[][RTE_LPM6_IPV6_ADDR_SIZE], 980 int32_t *next_hops, unsigned int n) 981 { 982 unsigned int i; 983 const struct rte_lpm6_tbl_entry *tbl; 984 const struct rte_lpm6_tbl_entry *tbl_next = NULL; 985 uint32_t tbl24_index, next_hop; 986 uint8_t first_byte; 987 int status; 988 989 /* DEBUG: Check user input arguments. */ 990 if ((lpm == NULL) || (ips == NULL) || (next_hops == NULL)) 991 return -EINVAL; 992 993 for (i = 0; i < n; i++) { 994 first_byte = LOOKUP_FIRST_BYTE; 995 tbl24_index = (ips[i][0] << BYTES2_SIZE) | 996 (ips[i][1] << BYTE_SIZE) | ips[i][2]; 997 998 /* Calculate pointer to the first entry to be inspected */ 999 tbl = &lpm->tbl24[tbl24_index]; 1000 1001 do { 1002 /* Continue inspecting following levels 1003 * until success or failure 1004 */ 1005 status = lookup_step(lpm, tbl, &tbl_next, ips[i], 1006 first_byte++, &next_hop); 1007 tbl = tbl_next; 1008 } while (status == 1); 1009 1010 if (status < 0) 1011 next_hops[i] = -1; 1012 else 1013 next_hops[i] = (int32_t)next_hop; 1014 } 1015 1016 return 0; 1017 } 1018 1019 /* 1020 * Look for a rule in the high-level rules table 1021 */ 1022 int 1023 rte_lpm6_is_rule_present(struct rte_lpm6 *lpm, const uint8_t *ip, uint8_t depth, 1024 uint32_t *next_hop) 1025 { 1026 uint8_t masked_ip[RTE_LPM6_IPV6_ADDR_SIZE]; 1027 1028 /* Check user arguments. */ 1029 if ((lpm == NULL) || next_hop == NULL || ip == NULL || 1030 (depth < 1) || (depth > RTE_LPM6_MAX_DEPTH)) 1031 return -EINVAL; 1032 1033 /* Copy the IP and mask it to avoid modifying user's input data. */ 1034 ip6_copy_addr(masked_ip, ip); 1035 ip6_mask_addr(masked_ip, depth); 1036 1037 return rule_find(lpm, masked_ip, depth, next_hop); 1038 } 1039 1040 /* 1041 * Delete a rule from the rule table. 1042 * NOTE: Valid range for depth parameter is 1 .. 128 inclusive. 1043 * return 1044 * 0 on success 1045 * <0 on failure 1046 */ 1047 static inline int 1048 rule_delete(struct rte_lpm6 *lpm, uint8_t *ip, uint8_t depth) 1049 { 1050 int ret; 1051 struct rte_lpm6_rule_key rule_key; 1052 1053 /* init rule key */ 1054 rule_key_init(&rule_key, ip, depth); 1055 1056 /* delete the rule */ 1057 ret = rte_hash_del_key(lpm->rules_tbl, (void *) &rule_key); 1058 if (ret >= 0) 1059 lpm->used_rules--; 1060 1061 return ret; 1062 } 1063 1064 /* 1065 * Deletes a group of rules 1066 * 1067 * Note that the function rebuilds the lpm table, 1068 * rather than doing incremental updates like 1069 * the regular delete function 1070 */ 1071 int 1072 rte_lpm6_delete_bulk_func(struct rte_lpm6 *lpm, 1073 uint8_t ips[][RTE_LPM6_IPV6_ADDR_SIZE], uint8_t *depths, 1074 unsigned n) 1075 { 1076 uint8_t masked_ip[RTE_LPM6_IPV6_ADDR_SIZE]; 1077 unsigned i; 1078 1079 /* Check input arguments. */ 1080 if ((lpm == NULL) || (ips == NULL) || (depths == NULL)) 1081 return -EINVAL; 1082 1083 for (i = 0; i < n; i++) { 1084 ip6_copy_addr(masked_ip, ips[i]); 1085 ip6_mask_addr(masked_ip, depths[i]); 1086 rule_delete(lpm, masked_ip, depths[i]); 1087 } 1088 1089 /* 1090 * Set all the table entries to 0 (ie delete every rule 1091 * from the data structure. 1092 */ 1093 memset(lpm->tbl24, 0, sizeof(lpm->tbl24)); 1094 memset(lpm->tbl8, 0, sizeof(lpm->tbl8[0]) 1095 * RTE_LPM6_TBL8_GROUP_NUM_ENTRIES * lpm->number_tbl8s); 1096 tbl8_pool_init(lpm); 1097 1098 /* 1099 * Add every rule again (except for the ones that were removed from 1100 * the rules table). 1101 */ 1102 rebuild_lpm(lpm); 1103 1104 return 0; 1105 } 1106 1107 /* 1108 * Delete all rules from the LPM table. 1109 */ 1110 void 1111 rte_lpm6_delete_all(struct rte_lpm6 *lpm) 1112 { 1113 /* Zero used rules counter. */ 1114 lpm->used_rules = 0; 1115 1116 /* Zero tbl24. */ 1117 memset(lpm->tbl24, 0, sizeof(lpm->tbl24)); 1118 1119 /* Zero tbl8. */ 1120 memset(lpm->tbl8, 0, sizeof(lpm->tbl8[0]) * 1121 RTE_LPM6_TBL8_GROUP_NUM_ENTRIES * lpm->number_tbl8s); 1122 1123 /* init pool of free tbl8 indexes */ 1124 tbl8_pool_init(lpm); 1125 1126 /* Delete all rules form the rules table. */ 1127 rte_hash_reset(lpm->rules_tbl); 1128 } 1129 1130 /* 1131 * Convert a depth to a one byte long mask 1132 * Example: 4 will be converted to 0xF0 1133 */ 1134 static uint8_t __attribute__((pure)) 1135 depth_to_mask_1b(uint8_t depth) 1136 { 1137 /* To calculate a mask start with a 1 on the left hand side and right 1138 * shift while populating the left hand side with 1's 1139 */ 1140 return (signed char)0x80 >> (depth - 1); 1141 } 1142 1143 /* 1144 * Find a less specific rule 1145 */ 1146 static int 1147 rule_find_less_specific(struct rte_lpm6 *lpm, uint8_t *ip, uint8_t depth, 1148 struct rte_lpm6_rule *rule) 1149 { 1150 int ret; 1151 uint32_t next_hop; 1152 uint8_t mask; 1153 struct rte_lpm6_rule_key rule_key; 1154 1155 if (depth == 1) 1156 return 0; 1157 1158 rule_key_init(&rule_key, ip, depth); 1159 1160 while (depth > 1) { 1161 depth--; 1162 1163 /* each iteration zero one more bit of the key */ 1164 mask = depth & 7; /* depth % BYTE_SIZE */ 1165 if (mask > 0) 1166 mask = depth_to_mask_1b(mask); 1167 1168 rule_key.depth = depth; 1169 rule_key.ip[depth >> 3] &= mask; 1170 1171 ret = rule_find_with_key(lpm, &rule_key, &next_hop); 1172 if (ret) { 1173 rule->depth = depth; 1174 ip6_copy_addr(rule->ip, rule_key.ip); 1175 rule->next_hop = next_hop; 1176 return 1; 1177 } 1178 } 1179 1180 return 0; 1181 } 1182 1183 /* 1184 * Find range of tbl8 cells occupied by a rule 1185 */ 1186 static void 1187 rule_find_range(struct rte_lpm6 *lpm, const uint8_t *ip, uint8_t depth, 1188 struct rte_lpm6_tbl_entry **from, 1189 struct rte_lpm6_tbl_entry **to, 1190 uint32_t *out_tbl_ind) 1191 { 1192 uint32_t ind; 1193 uint32_t first_3bytes = (uint32_t)ip[0] << 16 | ip[1] << 8 | ip[2]; 1194 1195 if (depth <= 24) { 1196 /* rule is within the top level */ 1197 ind = first_3bytes; 1198 *from = &lpm->tbl24[ind]; 1199 ind += (1 << (24 - depth)) - 1; 1200 *to = &lpm->tbl24[ind]; 1201 *out_tbl_ind = TBL24_IND; 1202 } else { 1203 /* top level entry */ 1204 struct rte_lpm6_tbl_entry *tbl = &lpm->tbl24[first_3bytes]; 1205 assert(tbl->ext_entry == 1); 1206 /* first tbl8 */ 1207 uint32_t tbl_ind = tbl->lpm6_tbl8_gindex; 1208 tbl = &lpm->tbl8[tbl_ind * 1209 RTE_LPM6_TBL8_GROUP_NUM_ENTRIES]; 1210 /* current ip byte, the top level is already behind */ 1211 uint8_t byte = 3; 1212 /* minus top level */ 1213 depth -= 24; 1214 1215 /* iterate through levels (tbl8s) 1216 * until we reach the last one 1217 */ 1218 while (depth > 8) { 1219 tbl += ip[byte]; 1220 assert(tbl->ext_entry == 1); 1221 /* go to the next level/tbl8 */ 1222 tbl_ind = tbl->lpm6_tbl8_gindex; 1223 tbl = &lpm->tbl8[tbl_ind * 1224 RTE_LPM6_TBL8_GROUP_NUM_ENTRIES]; 1225 byte += 1; 1226 depth -= 8; 1227 } 1228 1229 /* last level/tbl8 */ 1230 ind = ip[byte] & depth_to_mask_1b(depth); 1231 *from = &tbl[ind]; 1232 ind += (1 << (8 - depth)) - 1; 1233 *to = &tbl[ind]; 1234 *out_tbl_ind = tbl_ind; 1235 } 1236 } 1237 1238 /* 1239 * Remove a table from the LPM tree 1240 */ 1241 static void 1242 remove_tbl(struct rte_lpm6 *lpm, struct rte_lpm_tbl8_hdr *tbl_hdr, 1243 uint32_t tbl_ind, struct rte_lpm6_rule *lsp_rule) 1244 { 1245 struct rte_lpm6_tbl_entry *owner_entry; 1246 1247 if (tbl_hdr->owner_tbl_ind == TBL24_IND) 1248 owner_entry = &lpm->tbl24[tbl_hdr->owner_entry_ind]; 1249 else { 1250 uint32_t owner_tbl_ind = tbl_hdr->owner_tbl_ind; 1251 owner_entry = &lpm->tbl8[ 1252 owner_tbl_ind * RTE_LPM6_TBL8_GROUP_NUM_ENTRIES + 1253 tbl_hdr->owner_entry_ind]; 1254 1255 struct rte_lpm_tbl8_hdr *owner_tbl_hdr = 1256 &lpm->tbl8_hdrs[owner_tbl_ind]; 1257 if (--owner_tbl_hdr->ref_cnt == 0) 1258 remove_tbl(lpm, owner_tbl_hdr, owner_tbl_ind, lsp_rule); 1259 } 1260 1261 assert(owner_entry->ext_entry == 1); 1262 1263 /* unlink the table */ 1264 if (lsp_rule != NULL) { 1265 struct rte_lpm6_tbl_entry new_tbl_entry = { 1266 .next_hop = lsp_rule->next_hop, 1267 .depth = lsp_rule->depth, 1268 .valid = VALID, 1269 .valid_group = VALID, 1270 .ext_entry = 0 1271 }; 1272 1273 *owner_entry = new_tbl_entry; 1274 } else { 1275 struct rte_lpm6_tbl_entry new_tbl_entry = { 1276 .next_hop = 0, 1277 .depth = 0, 1278 .valid = INVALID, 1279 .valid_group = INVALID, 1280 .ext_entry = 0 1281 }; 1282 1283 *owner_entry = new_tbl_entry; 1284 } 1285 1286 /* return the table to the pool */ 1287 tbl8_put(lpm, tbl_ind); 1288 } 1289 1290 /* 1291 * Deletes a rule 1292 */ 1293 int 1294 rte_lpm6_delete(struct rte_lpm6 *lpm, const uint8_t *ip, uint8_t depth) 1295 { 1296 uint8_t masked_ip[RTE_LPM6_IPV6_ADDR_SIZE]; 1297 struct rte_lpm6_rule lsp_rule_obj; 1298 struct rte_lpm6_rule *lsp_rule; 1299 int ret; 1300 uint32_t tbl_ind; 1301 struct rte_lpm6_tbl_entry *from, *to; 1302 1303 /* Check input arguments. */ 1304 if ((lpm == NULL) || (depth < 1) || (depth > RTE_LPM6_MAX_DEPTH)) 1305 return -EINVAL; 1306 1307 /* Copy the IP and mask it to avoid modifying user's input data. */ 1308 ip6_copy_addr(masked_ip, ip); 1309 ip6_mask_addr(masked_ip, depth); 1310 1311 /* Delete the rule from the rule table. */ 1312 ret = rule_delete(lpm, masked_ip, depth); 1313 if (ret < 0) 1314 return -ENOENT; 1315 1316 /* find rule cells */ 1317 rule_find_range(lpm, masked_ip, depth, &from, &to, &tbl_ind); 1318 1319 /* find a less specific rule (a rule with smaller depth) 1320 * note: masked_ip will be modified, don't use it anymore 1321 */ 1322 ret = rule_find_less_specific(lpm, masked_ip, depth, 1323 &lsp_rule_obj); 1324 lsp_rule = ret ? &lsp_rule_obj : NULL; 1325 1326 /* decrement the table rule counter, 1327 * note that tbl24 doesn't have a header 1328 */ 1329 if (tbl_ind != TBL24_IND) { 1330 struct rte_lpm_tbl8_hdr *tbl_hdr = &lpm->tbl8_hdrs[tbl_ind]; 1331 if (--tbl_hdr->ref_cnt == 0) { 1332 /* remove the table */ 1333 remove_tbl(lpm, tbl_hdr, tbl_ind, lsp_rule); 1334 return 0; 1335 } 1336 } 1337 1338 /* iterate rule cells */ 1339 for (; from <= to; from++) 1340 if (from->ext_entry == 1) { 1341 /* reference to a more specific space 1342 * of the prefix/rule. Entries in a more 1343 * specific space that are not used by 1344 * a more specific prefix must be occupied 1345 * by the prefix 1346 */ 1347 if (lsp_rule != NULL) 1348 expand_rule(lpm, 1349 from->lpm6_tbl8_gindex * 1350 RTE_LPM6_TBL8_GROUP_NUM_ENTRIES, 1351 depth, lsp_rule->depth, 1352 lsp_rule->next_hop, VALID); 1353 else 1354 /* since the prefix has no less specific prefix, 1355 * its more specific space must be invalidated 1356 */ 1357 expand_rule(lpm, 1358 from->lpm6_tbl8_gindex * 1359 RTE_LPM6_TBL8_GROUP_NUM_ENTRIES, 1360 depth, 0, 0, INVALID); 1361 } else if (from->depth == depth) { 1362 /* entry is not a reference and belongs to the prefix */ 1363 if (lsp_rule != NULL) { 1364 struct rte_lpm6_tbl_entry new_tbl_entry = { 1365 .next_hop = lsp_rule->next_hop, 1366 .depth = lsp_rule->depth, 1367 .valid = VALID, 1368 .valid_group = VALID, 1369 .ext_entry = 0 1370 }; 1371 1372 *from = new_tbl_entry; 1373 } else { 1374 struct rte_lpm6_tbl_entry new_tbl_entry = { 1375 .next_hop = 0, 1376 .depth = 0, 1377 .valid = INVALID, 1378 .valid_group = INVALID, 1379 .ext_entry = 0 1380 }; 1381 1382 *from = new_tbl_entry; 1383 } 1384 } 1385 1386 return 0; 1387 } 1388