1 // SPDX-License-Identifier: GPL-2.0-only 2 /* Copyright (c) 2011-2014 PLUMgrid, http://plumgrid.com 3 * Copyright (c) 2016 Facebook 4 */ 5 #include <linux/bpf.h> 6 #include <linux/btf.h> 7 #include <linux/jhash.h> 8 #include <linux/filter.h> 9 #include <linux/rculist_nulls.h> 10 #include <linux/random.h> 11 #include <uapi/linux/btf.h> 12 #include <linux/rcupdate_trace.h> 13 #include "percpu_freelist.h" 14 #include "bpf_lru_list.h" 15 #include "map_in_map.h" 16 17 #define HTAB_CREATE_FLAG_MASK \ 18 (BPF_F_NO_PREALLOC | BPF_F_NO_COMMON_LRU | BPF_F_NUMA_NODE | \ 19 BPF_F_ACCESS_MASK | BPF_F_ZERO_SEED) 20 21 #define BATCH_OPS(_name) \ 22 .map_lookup_batch = \ 23 _name##_map_lookup_batch, \ 24 .map_lookup_and_delete_batch = \ 25 _name##_map_lookup_and_delete_batch, \ 26 .map_update_batch = \ 27 generic_map_update_batch, \ 28 .map_delete_batch = \ 29 generic_map_delete_batch 30 31 /* 32 * The bucket lock has two protection scopes: 33 * 34 * 1) Serializing concurrent operations from BPF programs on differrent 35 * CPUs 36 * 37 * 2) Serializing concurrent operations from BPF programs and sys_bpf() 38 * 39 * BPF programs can execute in any context including perf, kprobes and 40 * tracing. As there are almost no limits where perf, kprobes and tracing 41 * can be invoked from the lock operations need to be protected against 42 * deadlocks. Deadlocks can be caused by recursion and by an invocation in 43 * the lock held section when functions which acquire this lock are invoked 44 * from sys_bpf(). BPF recursion is prevented by incrementing the per CPU 45 * variable bpf_prog_active, which prevents BPF programs attached to perf 46 * events, kprobes and tracing to be invoked before the prior invocation 47 * from one of these contexts completed. sys_bpf() uses the same mechanism 48 * by pinning the task to the current CPU and incrementing the recursion 49 * protection accross the map operation. 50 * 51 * This has subtle implications on PREEMPT_RT. PREEMPT_RT forbids certain 52 * operations like memory allocations (even with GFP_ATOMIC) from atomic 53 * contexts. This is required because even with GFP_ATOMIC the memory 54 * allocator calls into code pathes which acquire locks with long held lock 55 * sections. To ensure the deterministic behaviour these locks are regular 56 * spinlocks, which are converted to 'sleepable' spinlocks on RT. The only 57 * true atomic contexts on an RT kernel are the low level hardware 58 * handling, scheduling, low level interrupt handling, NMIs etc. None of 59 * these contexts should ever do memory allocations. 60 * 61 * As regular device interrupt handlers and soft interrupts are forced into 62 * thread context, the existing code which does 63 * spin_lock*(); alloc(GPF_ATOMIC); spin_unlock*(); 64 * just works. 65 * 66 * In theory the BPF locks could be converted to regular spinlocks as well, 67 * but the bucket locks and percpu_freelist locks can be taken from 68 * arbitrary contexts (perf, kprobes, tracepoints) which are required to be 69 * atomic contexts even on RT. These mechanisms require preallocated maps, 70 * so there is no need to invoke memory allocations within the lock held 71 * sections. 72 * 73 * BPF maps which need dynamic allocation are only used from (forced) 74 * thread context on RT and can therefore use regular spinlocks which in 75 * turn allows to invoke memory allocations from the lock held section. 76 * 77 * On a non RT kernel this distinction is neither possible nor required. 78 * spinlock maps to raw_spinlock and the extra code is optimized out by the 79 * compiler. 80 */ 81 struct bucket { 82 struct hlist_nulls_head head; 83 union { 84 raw_spinlock_t raw_lock; 85 spinlock_t lock; 86 }; 87 }; 88 89 struct bpf_htab { 90 struct bpf_map map; 91 struct bucket *buckets; 92 void *elems; 93 union { 94 struct pcpu_freelist freelist; 95 struct bpf_lru lru; 96 }; 97 struct htab_elem *__percpu *extra_elems; 98 atomic_t count; /* number of elements in this hashtable */ 99 u32 n_buckets; /* number of hash buckets */ 100 u32 elem_size; /* size of each element in bytes */ 101 u32 hashrnd; 102 }; 103 104 /* each htab element is struct htab_elem + key + value */ 105 struct htab_elem { 106 union { 107 struct hlist_nulls_node hash_node; 108 struct { 109 void *padding; 110 union { 111 struct bpf_htab *htab; 112 struct pcpu_freelist_node fnode; 113 struct htab_elem *batch_flink; 114 }; 115 }; 116 }; 117 union { 118 struct rcu_head rcu; 119 struct bpf_lru_node lru_node; 120 }; 121 u32 hash; 122 char key[] __aligned(8); 123 }; 124 125 static inline bool htab_is_prealloc(const struct bpf_htab *htab) 126 { 127 return !(htab->map.map_flags & BPF_F_NO_PREALLOC); 128 } 129 130 static inline bool htab_use_raw_lock(const struct bpf_htab *htab) 131 { 132 return (!IS_ENABLED(CONFIG_PREEMPT_RT) || htab_is_prealloc(htab)); 133 } 134 135 static void htab_init_buckets(struct bpf_htab *htab) 136 { 137 unsigned i; 138 139 for (i = 0; i < htab->n_buckets; i++) { 140 INIT_HLIST_NULLS_HEAD(&htab->buckets[i].head, i); 141 if (htab_use_raw_lock(htab)) 142 raw_spin_lock_init(&htab->buckets[i].raw_lock); 143 else 144 spin_lock_init(&htab->buckets[i].lock); 145 } 146 } 147 148 static inline unsigned long htab_lock_bucket(const struct bpf_htab *htab, 149 struct bucket *b) 150 { 151 unsigned long flags; 152 153 if (htab_use_raw_lock(htab)) 154 raw_spin_lock_irqsave(&b->raw_lock, flags); 155 else 156 spin_lock_irqsave(&b->lock, flags); 157 return flags; 158 } 159 160 static inline void htab_unlock_bucket(const struct bpf_htab *htab, 161 struct bucket *b, 162 unsigned long flags) 163 { 164 if (htab_use_raw_lock(htab)) 165 raw_spin_unlock_irqrestore(&b->raw_lock, flags); 166 else 167 spin_unlock_irqrestore(&b->lock, flags); 168 } 169 170 static bool htab_lru_map_delete_node(void *arg, struct bpf_lru_node *node); 171 172 static bool htab_is_lru(const struct bpf_htab *htab) 173 { 174 return htab->map.map_type == BPF_MAP_TYPE_LRU_HASH || 175 htab->map.map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH; 176 } 177 178 static bool htab_is_percpu(const struct bpf_htab *htab) 179 { 180 return htab->map.map_type == BPF_MAP_TYPE_PERCPU_HASH || 181 htab->map.map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH; 182 } 183 184 static inline void htab_elem_set_ptr(struct htab_elem *l, u32 key_size, 185 void __percpu *pptr) 186 { 187 *(void __percpu **)(l->key + key_size) = pptr; 188 } 189 190 static inline void __percpu *htab_elem_get_ptr(struct htab_elem *l, u32 key_size) 191 { 192 return *(void __percpu **)(l->key + key_size); 193 } 194 195 static void *fd_htab_map_get_ptr(const struct bpf_map *map, struct htab_elem *l) 196 { 197 return *(void **)(l->key + roundup(map->key_size, 8)); 198 } 199 200 static struct htab_elem *get_htab_elem(struct bpf_htab *htab, int i) 201 { 202 return (struct htab_elem *) (htab->elems + i * htab->elem_size); 203 } 204 205 static void htab_free_elems(struct bpf_htab *htab) 206 { 207 int i; 208 209 if (!htab_is_percpu(htab)) 210 goto free_elems; 211 212 for (i = 0; i < htab->map.max_entries; i++) { 213 void __percpu *pptr; 214 215 pptr = htab_elem_get_ptr(get_htab_elem(htab, i), 216 htab->map.key_size); 217 free_percpu(pptr); 218 cond_resched(); 219 } 220 free_elems: 221 bpf_map_area_free(htab->elems); 222 } 223 224 /* The LRU list has a lock (lru_lock). Each htab bucket has a lock 225 * (bucket_lock). If both locks need to be acquired together, the lock 226 * order is always lru_lock -> bucket_lock and this only happens in 227 * bpf_lru_list.c logic. For example, certain code path of 228 * bpf_lru_pop_free(), which is called by function prealloc_lru_pop(), 229 * will acquire lru_lock first followed by acquiring bucket_lock. 230 * 231 * In hashtab.c, to avoid deadlock, lock acquisition of 232 * bucket_lock followed by lru_lock is not allowed. In such cases, 233 * bucket_lock needs to be released first before acquiring lru_lock. 234 */ 235 static struct htab_elem *prealloc_lru_pop(struct bpf_htab *htab, void *key, 236 u32 hash) 237 { 238 struct bpf_lru_node *node = bpf_lru_pop_free(&htab->lru, hash); 239 struct htab_elem *l; 240 241 if (node) { 242 l = container_of(node, struct htab_elem, lru_node); 243 memcpy(l->key, key, htab->map.key_size); 244 return l; 245 } 246 247 return NULL; 248 } 249 250 static int prealloc_init(struct bpf_htab *htab) 251 { 252 u32 num_entries = htab->map.max_entries; 253 int err = -ENOMEM, i; 254 255 if (!htab_is_percpu(htab) && !htab_is_lru(htab)) 256 num_entries += num_possible_cpus(); 257 258 htab->elems = bpf_map_area_alloc(htab->elem_size * num_entries, 259 htab->map.numa_node); 260 if (!htab->elems) 261 return -ENOMEM; 262 263 if (!htab_is_percpu(htab)) 264 goto skip_percpu_elems; 265 266 for (i = 0; i < num_entries; i++) { 267 u32 size = round_up(htab->map.value_size, 8); 268 void __percpu *pptr; 269 270 pptr = __alloc_percpu_gfp(size, 8, GFP_USER | __GFP_NOWARN); 271 if (!pptr) 272 goto free_elems; 273 htab_elem_set_ptr(get_htab_elem(htab, i), htab->map.key_size, 274 pptr); 275 cond_resched(); 276 } 277 278 skip_percpu_elems: 279 if (htab_is_lru(htab)) 280 err = bpf_lru_init(&htab->lru, 281 htab->map.map_flags & BPF_F_NO_COMMON_LRU, 282 offsetof(struct htab_elem, hash) - 283 offsetof(struct htab_elem, lru_node), 284 htab_lru_map_delete_node, 285 htab); 286 else 287 err = pcpu_freelist_init(&htab->freelist); 288 289 if (err) 290 goto free_elems; 291 292 if (htab_is_lru(htab)) 293 bpf_lru_populate(&htab->lru, htab->elems, 294 offsetof(struct htab_elem, lru_node), 295 htab->elem_size, num_entries); 296 else 297 pcpu_freelist_populate(&htab->freelist, 298 htab->elems + offsetof(struct htab_elem, fnode), 299 htab->elem_size, num_entries); 300 301 return 0; 302 303 free_elems: 304 htab_free_elems(htab); 305 return err; 306 } 307 308 static void prealloc_destroy(struct bpf_htab *htab) 309 { 310 htab_free_elems(htab); 311 312 if (htab_is_lru(htab)) 313 bpf_lru_destroy(&htab->lru); 314 else 315 pcpu_freelist_destroy(&htab->freelist); 316 } 317 318 static int alloc_extra_elems(struct bpf_htab *htab) 319 { 320 struct htab_elem *__percpu *pptr, *l_new; 321 struct pcpu_freelist_node *l; 322 int cpu; 323 324 pptr = __alloc_percpu_gfp(sizeof(struct htab_elem *), 8, 325 GFP_USER | __GFP_NOWARN); 326 if (!pptr) 327 return -ENOMEM; 328 329 for_each_possible_cpu(cpu) { 330 l = pcpu_freelist_pop(&htab->freelist); 331 /* pop will succeed, since prealloc_init() 332 * preallocated extra num_possible_cpus elements 333 */ 334 l_new = container_of(l, struct htab_elem, fnode); 335 *per_cpu_ptr(pptr, cpu) = l_new; 336 } 337 htab->extra_elems = pptr; 338 return 0; 339 } 340 341 /* Called from syscall */ 342 static int htab_map_alloc_check(union bpf_attr *attr) 343 { 344 bool percpu = (attr->map_type == BPF_MAP_TYPE_PERCPU_HASH || 345 attr->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH); 346 bool lru = (attr->map_type == BPF_MAP_TYPE_LRU_HASH || 347 attr->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH); 348 /* percpu_lru means each cpu has its own LRU list. 349 * it is different from BPF_MAP_TYPE_PERCPU_HASH where 350 * the map's value itself is percpu. percpu_lru has 351 * nothing to do with the map's value. 352 */ 353 bool percpu_lru = (attr->map_flags & BPF_F_NO_COMMON_LRU); 354 bool prealloc = !(attr->map_flags & BPF_F_NO_PREALLOC); 355 bool zero_seed = (attr->map_flags & BPF_F_ZERO_SEED); 356 int numa_node = bpf_map_attr_numa_node(attr); 357 358 BUILD_BUG_ON(offsetof(struct htab_elem, htab) != 359 offsetof(struct htab_elem, hash_node.pprev)); 360 BUILD_BUG_ON(offsetof(struct htab_elem, fnode.next) != 361 offsetof(struct htab_elem, hash_node.pprev)); 362 363 if (lru && !bpf_capable()) 364 /* LRU implementation is much complicated than other 365 * maps. Hence, limit to CAP_BPF. 366 */ 367 return -EPERM; 368 369 if (zero_seed && !capable(CAP_SYS_ADMIN)) 370 /* Guard against local DoS, and discourage production use. */ 371 return -EPERM; 372 373 if (attr->map_flags & ~HTAB_CREATE_FLAG_MASK || 374 !bpf_map_flags_access_ok(attr->map_flags)) 375 return -EINVAL; 376 377 if (!lru && percpu_lru) 378 return -EINVAL; 379 380 if (lru && !prealloc) 381 return -ENOTSUPP; 382 383 if (numa_node != NUMA_NO_NODE && (percpu || percpu_lru)) 384 return -EINVAL; 385 386 /* check sanity of attributes. 387 * value_size == 0 may be allowed in the future to use map as a set 388 */ 389 if (attr->max_entries == 0 || attr->key_size == 0 || 390 attr->value_size == 0) 391 return -EINVAL; 392 393 if (attr->key_size > MAX_BPF_STACK) 394 /* eBPF programs initialize keys on stack, so they cannot be 395 * larger than max stack size 396 */ 397 return -E2BIG; 398 399 if (attr->value_size >= KMALLOC_MAX_SIZE - 400 MAX_BPF_STACK - sizeof(struct htab_elem)) 401 /* if value_size is bigger, the user space won't be able to 402 * access the elements via bpf syscall. This check also makes 403 * sure that the elem_size doesn't overflow and it's 404 * kmalloc-able later in htab_map_update_elem() 405 */ 406 return -E2BIG; 407 408 return 0; 409 } 410 411 static struct bpf_map *htab_map_alloc(union bpf_attr *attr) 412 { 413 bool percpu = (attr->map_type == BPF_MAP_TYPE_PERCPU_HASH || 414 attr->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH); 415 bool lru = (attr->map_type == BPF_MAP_TYPE_LRU_HASH || 416 attr->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH); 417 /* percpu_lru means each cpu has its own LRU list. 418 * it is different from BPF_MAP_TYPE_PERCPU_HASH where 419 * the map's value itself is percpu. percpu_lru has 420 * nothing to do with the map's value. 421 */ 422 bool percpu_lru = (attr->map_flags & BPF_F_NO_COMMON_LRU); 423 bool prealloc = !(attr->map_flags & BPF_F_NO_PREALLOC); 424 struct bpf_htab *htab; 425 u64 cost; 426 int err; 427 428 htab = kzalloc(sizeof(*htab), GFP_USER); 429 if (!htab) 430 return ERR_PTR(-ENOMEM); 431 432 bpf_map_init_from_attr(&htab->map, attr); 433 434 if (percpu_lru) { 435 /* ensure each CPU's lru list has >=1 elements. 436 * since we are at it, make each lru list has the same 437 * number of elements. 438 */ 439 htab->map.max_entries = roundup(attr->max_entries, 440 num_possible_cpus()); 441 if (htab->map.max_entries < attr->max_entries) 442 htab->map.max_entries = rounddown(attr->max_entries, 443 num_possible_cpus()); 444 } 445 446 /* hash table size must be power of 2 */ 447 htab->n_buckets = roundup_pow_of_two(htab->map.max_entries); 448 449 htab->elem_size = sizeof(struct htab_elem) + 450 round_up(htab->map.key_size, 8); 451 if (percpu) 452 htab->elem_size += sizeof(void *); 453 else 454 htab->elem_size += round_up(htab->map.value_size, 8); 455 456 err = -E2BIG; 457 /* prevent zero size kmalloc and check for u32 overflow */ 458 if (htab->n_buckets == 0 || 459 htab->n_buckets > U32_MAX / sizeof(struct bucket)) 460 goto free_htab; 461 462 cost = (u64) htab->n_buckets * sizeof(struct bucket) + 463 (u64) htab->elem_size * htab->map.max_entries; 464 465 if (percpu) 466 cost += (u64) round_up(htab->map.value_size, 8) * 467 num_possible_cpus() * htab->map.max_entries; 468 else 469 cost += (u64) htab->elem_size * num_possible_cpus(); 470 471 /* if map size is larger than memlock limit, reject it */ 472 err = bpf_map_charge_init(&htab->map.memory, cost); 473 if (err) 474 goto free_htab; 475 476 err = -ENOMEM; 477 htab->buckets = bpf_map_area_alloc(htab->n_buckets * 478 sizeof(struct bucket), 479 htab->map.numa_node); 480 if (!htab->buckets) 481 goto free_charge; 482 483 if (htab->map.map_flags & BPF_F_ZERO_SEED) 484 htab->hashrnd = 0; 485 else 486 htab->hashrnd = get_random_int(); 487 488 htab_init_buckets(htab); 489 490 if (prealloc) { 491 err = prealloc_init(htab); 492 if (err) 493 goto free_buckets; 494 495 if (!percpu && !lru) { 496 /* lru itself can remove the least used element, so 497 * there is no need for an extra elem during map_update. 498 */ 499 err = alloc_extra_elems(htab); 500 if (err) 501 goto free_prealloc; 502 } 503 } 504 505 return &htab->map; 506 507 free_prealloc: 508 prealloc_destroy(htab); 509 free_buckets: 510 bpf_map_area_free(htab->buckets); 511 free_charge: 512 bpf_map_charge_finish(&htab->map.memory); 513 free_htab: 514 kfree(htab); 515 return ERR_PTR(err); 516 } 517 518 static inline u32 htab_map_hash(const void *key, u32 key_len, u32 hashrnd) 519 { 520 return jhash(key, key_len, hashrnd); 521 } 522 523 static inline struct bucket *__select_bucket(struct bpf_htab *htab, u32 hash) 524 { 525 return &htab->buckets[hash & (htab->n_buckets - 1)]; 526 } 527 528 static inline struct hlist_nulls_head *select_bucket(struct bpf_htab *htab, u32 hash) 529 { 530 return &__select_bucket(htab, hash)->head; 531 } 532 533 /* this lookup function can only be called with bucket lock taken */ 534 static struct htab_elem *lookup_elem_raw(struct hlist_nulls_head *head, u32 hash, 535 void *key, u32 key_size) 536 { 537 struct hlist_nulls_node *n; 538 struct htab_elem *l; 539 540 hlist_nulls_for_each_entry_rcu(l, n, head, hash_node) 541 if (l->hash == hash && !memcmp(&l->key, key, key_size)) 542 return l; 543 544 return NULL; 545 } 546 547 /* can be called without bucket lock. it will repeat the loop in 548 * the unlikely event when elements moved from one bucket into another 549 * while link list is being walked 550 */ 551 static struct htab_elem *lookup_nulls_elem_raw(struct hlist_nulls_head *head, 552 u32 hash, void *key, 553 u32 key_size, u32 n_buckets) 554 { 555 struct hlist_nulls_node *n; 556 struct htab_elem *l; 557 558 again: 559 hlist_nulls_for_each_entry_rcu(l, n, head, hash_node) 560 if (l->hash == hash && !memcmp(&l->key, key, key_size)) 561 return l; 562 563 if (unlikely(get_nulls_value(n) != (hash & (n_buckets - 1)))) 564 goto again; 565 566 return NULL; 567 } 568 569 /* Called from syscall or from eBPF program directly, so 570 * arguments have to match bpf_map_lookup_elem() exactly. 571 * The return value is adjusted by BPF instructions 572 * in htab_map_gen_lookup(). 573 */ 574 static void *__htab_map_lookup_elem(struct bpf_map *map, void *key) 575 { 576 struct bpf_htab *htab = container_of(map, struct bpf_htab, map); 577 struct hlist_nulls_head *head; 578 struct htab_elem *l; 579 u32 hash, key_size; 580 581 WARN_ON_ONCE(!rcu_read_lock_held() && !rcu_read_lock_trace_held()); 582 583 key_size = map->key_size; 584 585 hash = htab_map_hash(key, key_size, htab->hashrnd); 586 587 head = select_bucket(htab, hash); 588 589 l = lookup_nulls_elem_raw(head, hash, key, key_size, htab->n_buckets); 590 591 return l; 592 } 593 594 static void *htab_map_lookup_elem(struct bpf_map *map, void *key) 595 { 596 struct htab_elem *l = __htab_map_lookup_elem(map, key); 597 598 if (l) 599 return l->key + round_up(map->key_size, 8); 600 601 return NULL; 602 } 603 604 /* inline bpf_map_lookup_elem() call. 605 * Instead of: 606 * bpf_prog 607 * bpf_map_lookup_elem 608 * map->ops->map_lookup_elem 609 * htab_map_lookup_elem 610 * __htab_map_lookup_elem 611 * do: 612 * bpf_prog 613 * __htab_map_lookup_elem 614 */ 615 static int htab_map_gen_lookup(struct bpf_map *map, struct bpf_insn *insn_buf) 616 { 617 struct bpf_insn *insn = insn_buf; 618 const int ret = BPF_REG_0; 619 620 BUILD_BUG_ON(!__same_type(&__htab_map_lookup_elem, 621 (void *(*)(struct bpf_map *map, void *key))NULL)); 622 *insn++ = BPF_EMIT_CALL(BPF_CAST_CALL(__htab_map_lookup_elem)); 623 *insn++ = BPF_JMP_IMM(BPF_JEQ, ret, 0, 1); 624 *insn++ = BPF_ALU64_IMM(BPF_ADD, ret, 625 offsetof(struct htab_elem, key) + 626 round_up(map->key_size, 8)); 627 return insn - insn_buf; 628 } 629 630 static __always_inline void *__htab_lru_map_lookup_elem(struct bpf_map *map, 631 void *key, const bool mark) 632 { 633 struct htab_elem *l = __htab_map_lookup_elem(map, key); 634 635 if (l) { 636 if (mark) 637 bpf_lru_node_set_ref(&l->lru_node); 638 return l->key + round_up(map->key_size, 8); 639 } 640 641 return NULL; 642 } 643 644 static void *htab_lru_map_lookup_elem(struct bpf_map *map, void *key) 645 { 646 return __htab_lru_map_lookup_elem(map, key, true); 647 } 648 649 static void *htab_lru_map_lookup_elem_sys(struct bpf_map *map, void *key) 650 { 651 return __htab_lru_map_lookup_elem(map, key, false); 652 } 653 654 static int htab_lru_map_gen_lookup(struct bpf_map *map, 655 struct bpf_insn *insn_buf) 656 { 657 struct bpf_insn *insn = insn_buf; 658 const int ret = BPF_REG_0; 659 const int ref_reg = BPF_REG_1; 660 661 BUILD_BUG_ON(!__same_type(&__htab_map_lookup_elem, 662 (void *(*)(struct bpf_map *map, void *key))NULL)); 663 *insn++ = BPF_EMIT_CALL(BPF_CAST_CALL(__htab_map_lookup_elem)); 664 *insn++ = BPF_JMP_IMM(BPF_JEQ, ret, 0, 4); 665 *insn++ = BPF_LDX_MEM(BPF_B, ref_reg, ret, 666 offsetof(struct htab_elem, lru_node) + 667 offsetof(struct bpf_lru_node, ref)); 668 *insn++ = BPF_JMP_IMM(BPF_JNE, ref_reg, 0, 1); 669 *insn++ = BPF_ST_MEM(BPF_B, ret, 670 offsetof(struct htab_elem, lru_node) + 671 offsetof(struct bpf_lru_node, ref), 672 1); 673 *insn++ = BPF_ALU64_IMM(BPF_ADD, ret, 674 offsetof(struct htab_elem, key) + 675 round_up(map->key_size, 8)); 676 return insn - insn_buf; 677 } 678 679 /* It is called from the bpf_lru_list when the LRU needs to delete 680 * older elements from the htab. 681 */ 682 static bool htab_lru_map_delete_node(void *arg, struct bpf_lru_node *node) 683 { 684 struct bpf_htab *htab = (struct bpf_htab *)arg; 685 struct htab_elem *l = NULL, *tgt_l; 686 struct hlist_nulls_head *head; 687 struct hlist_nulls_node *n; 688 unsigned long flags; 689 struct bucket *b; 690 691 tgt_l = container_of(node, struct htab_elem, lru_node); 692 b = __select_bucket(htab, tgt_l->hash); 693 head = &b->head; 694 695 flags = htab_lock_bucket(htab, b); 696 697 hlist_nulls_for_each_entry_rcu(l, n, head, hash_node) 698 if (l == tgt_l) { 699 hlist_nulls_del_rcu(&l->hash_node); 700 break; 701 } 702 703 htab_unlock_bucket(htab, b, flags); 704 705 return l == tgt_l; 706 } 707 708 /* Called from syscall */ 709 static int htab_map_get_next_key(struct bpf_map *map, void *key, void *next_key) 710 { 711 struct bpf_htab *htab = container_of(map, struct bpf_htab, map); 712 struct hlist_nulls_head *head; 713 struct htab_elem *l, *next_l; 714 u32 hash, key_size; 715 int i = 0; 716 717 WARN_ON_ONCE(!rcu_read_lock_held()); 718 719 key_size = map->key_size; 720 721 if (!key) 722 goto find_first_elem; 723 724 hash = htab_map_hash(key, key_size, htab->hashrnd); 725 726 head = select_bucket(htab, hash); 727 728 /* lookup the key */ 729 l = lookup_nulls_elem_raw(head, hash, key, key_size, htab->n_buckets); 730 731 if (!l) 732 goto find_first_elem; 733 734 /* key was found, get next key in the same bucket */ 735 next_l = hlist_nulls_entry_safe(rcu_dereference_raw(hlist_nulls_next_rcu(&l->hash_node)), 736 struct htab_elem, hash_node); 737 738 if (next_l) { 739 /* if next elem in this hash list is non-zero, just return it */ 740 memcpy(next_key, next_l->key, key_size); 741 return 0; 742 } 743 744 /* no more elements in this hash list, go to the next bucket */ 745 i = hash & (htab->n_buckets - 1); 746 i++; 747 748 find_first_elem: 749 /* iterate over buckets */ 750 for (; i < htab->n_buckets; i++) { 751 head = select_bucket(htab, i); 752 753 /* pick first element in the bucket */ 754 next_l = hlist_nulls_entry_safe(rcu_dereference_raw(hlist_nulls_first_rcu(head)), 755 struct htab_elem, hash_node); 756 if (next_l) { 757 /* if it's not empty, just return it */ 758 memcpy(next_key, next_l->key, key_size); 759 return 0; 760 } 761 } 762 763 /* iterated over all buckets and all elements */ 764 return -ENOENT; 765 } 766 767 static void htab_elem_free(struct bpf_htab *htab, struct htab_elem *l) 768 { 769 if (htab->map.map_type == BPF_MAP_TYPE_PERCPU_HASH) 770 free_percpu(htab_elem_get_ptr(l, htab->map.key_size)); 771 kfree(l); 772 } 773 774 static void htab_elem_free_rcu(struct rcu_head *head) 775 { 776 struct htab_elem *l = container_of(head, struct htab_elem, rcu); 777 struct bpf_htab *htab = l->htab; 778 779 htab_elem_free(htab, l); 780 } 781 782 static void htab_put_fd_value(struct bpf_htab *htab, struct htab_elem *l) 783 { 784 struct bpf_map *map = &htab->map; 785 void *ptr; 786 787 if (map->ops->map_fd_put_ptr) { 788 ptr = fd_htab_map_get_ptr(map, l); 789 map->ops->map_fd_put_ptr(ptr); 790 } 791 } 792 793 static void free_htab_elem(struct bpf_htab *htab, struct htab_elem *l) 794 { 795 htab_put_fd_value(htab, l); 796 797 if (htab_is_prealloc(htab)) { 798 __pcpu_freelist_push(&htab->freelist, &l->fnode); 799 } else { 800 atomic_dec(&htab->count); 801 l->htab = htab; 802 call_rcu(&l->rcu, htab_elem_free_rcu); 803 } 804 } 805 806 static void pcpu_copy_value(struct bpf_htab *htab, void __percpu *pptr, 807 void *value, bool onallcpus) 808 { 809 if (!onallcpus) { 810 /* copy true value_size bytes */ 811 memcpy(this_cpu_ptr(pptr), value, htab->map.value_size); 812 } else { 813 u32 size = round_up(htab->map.value_size, 8); 814 int off = 0, cpu; 815 816 for_each_possible_cpu(cpu) { 817 bpf_long_memcpy(per_cpu_ptr(pptr, cpu), 818 value + off, size); 819 off += size; 820 } 821 } 822 } 823 824 static void pcpu_init_value(struct bpf_htab *htab, void __percpu *pptr, 825 void *value, bool onallcpus) 826 { 827 /* When using prealloc and not setting the initial value on all cpus, 828 * zero-fill element values for other cpus (just as what happens when 829 * not using prealloc). Otherwise, bpf program has no way to ensure 830 * known initial values for cpus other than current one 831 * (onallcpus=false always when coming from bpf prog). 832 */ 833 if (htab_is_prealloc(htab) && !onallcpus) { 834 u32 size = round_up(htab->map.value_size, 8); 835 int current_cpu = raw_smp_processor_id(); 836 int cpu; 837 838 for_each_possible_cpu(cpu) { 839 if (cpu == current_cpu) 840 bpf_long_memcpy(per_cpu_ptr(pptr, cpu), value, 841 size); 842 else 843 memset(per_cpu_ptr(pptr, cpu), 0, size); 844 } 845 } else { 846 pcpu_copy_value(htab, pptr, value, onallcpus); 847 } 848 } 849 850 static bool fd_htab_map_needs_adjust(const struct bpf_htab *htab) 851 { 852 return htab->map.map_type == BPF_MAP_TYPE_HASH_OF_MAPS && 853 BITS_PER_LONG == 64; 854 } 855 856 static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key, 857 void *value, u32 key_size, u32 hash, 858 bool percpu, bool onallcpus, 859 struct htab_elem *old_elem) 860 { 861 u32 size = htab->map.value_size; 862 bool prealloc = htab_is_prealloc(htab); 863 struct htab_elem *l_new, **pl_new; 864 void __percpu *pptr; 865 866 if (prealloc) { 867 if (old_elem) { 868 /* if we're updating the existing element, 869 * use per-cpu extra elems to avoid freelist_pop/push 870 */ 871 pl_new = this_cpu_ptr(htab->extra_elems); 872 l_new = *pl_new; 873 htab_put_fd_value(htab, old_elem); 874 *pl_new = old_elem; 875 } else { 876 struct pcpu_freelist_node *l; 877 878 l = __pcpu_freelist_pop(&htab->freelist); 879 if (!l) 880 return ERR_PTR(-E2BIG); 881 l_new = container_of(l, struct htab_elem, fnode); 882 } 883 } else { 884 if (atomic_inc_return(&htab->count) > htab->map.max_entries) 885 if (!old_elem) { 886 /* when map is full and update() is replacing 887 * old element, it's ok to allocate, since 888 * old element will be freed immediately. 889 * Otherwise return an error 890 */ 891 l_new = ERR_PTR(-E2BIG); 892 goto dec_count; 893 } 894 l_new = kmalloc_node(htab->elem_size, GFP_ATOMIC | __GFP_NOWARN, 895 htab->map.numa_node); 896 if (!l_new) { 897 l_new = ERR_PTR(-ENOMEM); 898 goto dec_count; 899 } 900 check_and_init_map_lock(&htab->map, 901 l_new->key + round_up(key_size, 8)); 902 } 903 904 memcpy(l_new->key, key, key_size); 905 if (percpu) { 906 size = round_up(size, 8); 907 if (prealloc) { 908 pptr = htab_elem_get_ptr(l_new, key_size); 909 } else { 910 /* alloc_percpu zero-fills */ 911 pptr = __alloc_percpu_gfp(size, 8, 912 GFP_ATOMIC | __GFP_NOWARN); 913 if (!pptr) { 914 kfree(l_new); 915 l_new = ERR_PTR(-ENOMEM); 916 goto dec_count; 917 } 918 } 919 920 pcpu_init_value(htab, pptr, value, onallcpus); 921 922 if (!prealloc) 923 htab_elem_set_ptr(l_new, key_size, pptr); 924 } else if (fd_htab_map_needs_adjust(htab)) { 925 size = round_up(size, 8); 926 memcpy(l_new->key + round_up(key_size, 8), value, size); 927 } else { 928 copy_map_value(&htab->map, 929 l_new->key + round_up(key_size, 8), 930 value); 931 } 932 933 l_new->hash = hash; 934 return l_new; 935 dec_count: 936 atomic_dec(&htab->count); 937 return l_new; 938 } 939 940 static int check_flags(struct bpf_htab *htab, struct htab_elem *l_old, 941 u64 map_flags) 942 { 943 if (l_old && (map_flags & ~BPF_F_LOCK) == BPF_NOEXIST) 944 /* elem already exists */ 945 return -EEXIST; 946 947 if (!l_old && (map_flags & ~BPF_F_LOCK) == BPF_EXIST) 948 /* elem doesn't exist, cannot update it */ 949 return -ENOENT; 950 951 return 0; 952 } 953 954 /* Called from syscall or from eBPF program */ 955 static int htab_map_update_elem(struct bpf_map *map, void *key, void *value, 956 u64 map_flags) 957 { 958 struct bpf_htab *htab = container_of(map, struct bpf_htab, map); 959 struct htab_elem *l_new = NULL, *l_old; 960 struct hlist_nulls_head *head; 961 unsigned long flags; 962 struct bucket *b; 963 u32 key_size, hash; 964 int ret; 965 966 if (unlikely((map_flags & ~BPF_F_LOCK) > BPF_EXIST)) 967 /* unknown flags */ 968 return -EINVAL; 969 970 WARN_ON_ONCE(!rcu_read_lock_held() && !rcu_read_lock_trace_held()); 971 972 key_size = map->key_size; 973 974 hash = htab_map_hash(key, key_size, htab->hashrnd); 975 976 b = __select_bucket(htab, hash); 977 head = &b->head; 978 979 if (unlikely(map_flags & BPF_F_LOCK)) { 980 if (unlikely(!map_value_has_spin_lock(map))) 981 return -EINVAL; 982 /* find an element without taking the bucket lock */ 983 l_old = lookup_nulls_elem_raw(head, hash, key, key_size, 984 htab->n_buckets); 985 ret = check_flags(htab, l_old, map_flags); 986 if (ret) 987 return ret; 988 if (l_old) { 989 /* grab the element lock and update value in place */ 990 copy_map_value_locked(map, 991 l_old->key + round_up(key_size, 8), 992 value, false); 993 return 0; 994 } 995 /* fall through, grab the bucket lock and lookup again. 996 * 99.9% chance that the element won't be found, 997 * but second lookup under lock has to be done. 998 */ 999 } 1000 1001 flags = htab_lock_bucket(htab, b); 1002 1003 l_old = lookup_elem_raw(head, hash, key, key_size); 1004 1005 ret = check_flags(htab, l_old, map_flags); 1006 if (ret) 1007 goto err; 1008 1009 if (unlikely(l_old && (map_flags & BPF_F_LOCK))) { 1010 /* first lookup without the bucket lock didn't find the element, 1011 * but second lookup with the bucket lock found it. 1012 * This case is highly unlikely, but has to be dealt with: 1013 * grab the element lock in addition to the bucket lock 1014 * and update element in place 1015 */ 1016 copy_map_value_locked(map, 1017 l_old->key + round_up(key_size, 8), 1018 value, false); 1019 ret = 0; 1020 goto err; 1021 } 1022 1023 l_new = alloc_htab_elem(htab, key, value, key_size, hash, false, false, 1024 l_old); 1025 if (IS_ERR(l_new)) { 1026 /* all pre-allocated elements are in use or memory exhausted */ 1027 ret = PTR_ERR(l_new); 1028 goto err; 1029 } 1030 1031 /* add new element to the head of the list, so that 1032 * concurrent search will find it before old elem 1033 */ 1034 hlist_nulls_add_head_rcu(&l_new->hash_node, head); 1035 if (l_old) { 1036 hlist_nulls_del_rcu(&l_old->hash_node); 1037 if (!htab_is_prealloc(htab)) 1038 free_htab_elem(htab, l_old); 1039 } 1040 ret = 0; 1041 err: 1042 htab_unlock_bucket(htab, b, flags); 1043 return ret; 1044 } 1045 1046 static int htab_lru_map_update_elem(struct bpf_map *map, void *key, void *value, 1047 u64 map_flags) 1048 { 1049 struct bpf_htab *htab = container_of(map, struct bpf_htab, map); 1050 struct htab_elem *l_new, *l_old = NULL; 1051 struct hlist_nulls_head *head; 1052 unsigned long flags; 1053 struct bucket *b; 1054 u32 key_size, hash; 1055 int ret; 1056 1057 if (unlikely(map_flags > BPF_EXIST)) 1058 /* unknown flags */ 1059 return -EINVAL; 1060 1061 WARN_ON_ONCE(!rcu_read_lock_held() && !rcu_read_lock_trace_held()); 1062 1063 key_size = map->key_size; 1064 1065 hash = htab_map_hash(key, key_size, htab->hashrnd); 1066 1067 b = __select_bucket(htab, hash); 1068 head = &b->head; 1069 1070 /* For LRU, we need to alloc before taking bucket's 1071 * spinlock because getting free nodes from LRU may need 1072 * to remove older elements from htab and this removal 1073 * operation will need a bucket lock. 1074 */ 1075 l_new = prealloc_lru_pop(htab, key, hash); 1076 if (!l_new) 1077 return -ENOMEM; 1078 memcpy(l_new->key + round_up(map->key_size, 8), value, map->value_size); 1079 1080 flags = htab_lock_bucket(htab, b); 1081 1082 l_old = lookup_elem_raw(head, hash, key, key_size); 1083 1084 ret = check_flags(htab, l_old, map_flags); 1085 if (ret) 1086 goto err; 1087 1088 /* add new element to the head of the list, so that 1089 * concurrent search will find it before old elem 1090 */ 1091 hlist_nulls_add_head_rcu(&l_new->hash_node, head); 1092 if (l_old) { 1093 bpf_lru_node_set_ref(&l_new->lru_node); 1094 hlist_nulls_del_rcu(&l_old->hash_node); 1095 } 1096 ret = 0; 1097 1098 err: 1099 htab_unlock_bucket(htab, b, flags); 1100 1101 if (ret) 1102 bpf_lru_push_free(&htab->lru, &l_new->lru_node); 1103 else if (l_old) 1104 bpf_lru_push_free(&htab->lru, &l_old->lru_node); 1105 1106 return ret; 1107 } 1108 1109 static int __htab_percpu_map_update_elem(struct bpf_map *map, void *key, 1110 void *value, u64 map_flags, 1111 bool onallcpus) 1112 { 1113 struct bpf_htab *htab = container_of(map, struct bpf_htab, map); 1114 struct htab_elem *l_new = NULL, *l_old; 1115 struct hlist_nulls_head *head; 1116 unsigned long flags; 1117 struct bucket *b; 1118 u32 key_size, hash; 1119 int ret; 1120 1121 if (unlikely(map_flags > BPF_EXIST)) 1122 /* unknown flags */ 1123 return -EINVAL; 1124 1125 WARN_ON_ONCE(!rcu_read_lock_held()); 1126 1127 key_size = map->key_size; 1128 1129 hash = htab_map_hash(key, key_size, htab->hashrnd); 1130 1131 b = __select_bucket(htab, hash); 1132 head = &b->head; 1133 1134 flags = htab_lock_bucket(htab, b); 1135 1136 l_old = lookup_elem_raw(head, hash, key, key_size); 1137 1138 ret = check_flags(htab, l_old, map_flags); 1139 if (ret) 1140 goto err; 1141 1142 if (l_old) { 1143 /* per-cpu hash map can update value in-place */ 1144 pcpu_copy_value(htab, htab_elem_get_ptr(l_old, key_size), 1145 value, onallcpus); 1146 } else { 1147 l_new = alloc_htab_elem(htab, key, value, key_size, 1148 hash, true, onallcpus, NULL); 1149 if (IS_ERR(l_new)) { 1150 ret = PTR_ERR(l_new); 1151 goto err; 1152 } 1153 hlist_nulls_add_head_rcu(&l_new->hash_node, head); 1154 } 1155 ret = 0; 1156 err: 1157 htab_unlock_bucket(htab, b, flags); 1158 return ret; 1159 } 1160 1161 static int __htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key, 1162 void *value, u64 map_flags, 1163 bool onallcpus) 1164 { 1165 struct bpf_htab *htab = container_of(map, struct bpf_htab, map); 1166 struct htab_elem *l_new = NULL, *l_old; 1167 struct hlist_nulls_head *head; 1168 unsigned long flags; 1169 struct bucket *b; 1170 u32 key_size, hash; 1171 int ret; 1172 1173 if (unlikely(map_flags > BPF_EXIST)) 1174 /* unknown flags */ 1175 return -EINVAL; 1176 1177 WARN_ON_ONCE(!rcu_read_lock_held()); 1178 1179 key_size = map->key_size; 1180 1181 hash = htab_map_hash(key, key_size, htab->hashrnd); 1182 1183 b = __select_bucket(htab, hash); 1184 head = &b->head; 1185 1186 /* For LRU, we need to alloc before taking bucket's 1187 * spinlock because LRU's elem alloc may need 1188 * to remove older elem from htab and this removal 1189 * operation will need a bucket lock. 1190 */ 1191 if (map_flags != BPF_EXIST) { 1192 l_new = prealloc_lru_pop(htab, key, hash); 1193 if (!l_new) 1194 return -ENOMEM; 1195 } 1196 1197 flags = htab_lock_bucket(htab, b); 1198 1199 l_old = lookup_elem_raw(head, hash, key, key_size); 1200 1201 ret = check_flags(htab, l_old, map_flags); 1202 if (ret) 1203 goto err; 1204 1205 if (l_old) { 1206 bpf_lru_node_set_ref(&l_old->lru_node); 1207 1208 /* per-cpu hash map can update value in-place */ 1209 pcpu_copy_value(htab, htab_elem_get_ptr(l_old, key_size), 1210 value, onallcpus); 1211 } else { 1212 pcpu_init_value(htab, htab_elem_get_ptr(l_new, key_size), 1213 value, onallcpus); 1214 hlist_nulls_add_head_rcu(&l_new->hash_node, head); 1215 l_new = NULL; 1216 } 1217 ret = 0; 1218 err: 1219 htab_unlock_bucket(htab, b, flags); 1220 if (l_new) 1221 bpf_lru_push_free(&htab->lru, &l_new->lru_node); 1222 return ret; 1223 } 1224 1225 static int htab_percpu_map_update_elem(struct bpf_map *map, void *key, 1226 void *value, u64 map_flags) 1227 { 1228 return __htab_percpu_map_update_elem(map, key, value, map_flags, false); 1229 } 1230 1231 static int htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key, 1232 void *value, u64 map_flags) 1233 { 1234 return __htab_lru_percpu_map_update_elem(map, key, value, map_flags, 1235 false); 1236 } 1237 1238 /* Called from syscall or from eBPF program */ 1239 static int htab_map_delete_elem(struct bpf_map *map, void *key) 1240 { 1241 struct bpf_htab *htab = container_of(map, struct bpf_htab, map); 1242 struct hlist_nulls_head *head; 1243 struct bucket *b; 1244 struct htab_elem *l; 1245 unsigned long flags; 1246 u32 hash, key_size; 1247 int ret = -ENOENT; 1248 1249 WARN_ON_ONCE(!rcu_read_lock_held() && !rcu_read_lock_trace_held()); 1250 1251 key_size = map->key_size; 1252 1253 hash = htab_map_hash(key, key_size, htab->hashrnd); 1254 b = __select_bucket(htab, hash); 1255 head = &b->head; 1256 1257 flags = htab_lock_bucket(htab, b); 1258 1259 l = lookup_elem_raw(head, hash, key, key_size); 1260 1261 if (l) { 1262 hlist_nulls_del_rcu(&l->hash_node); 1263 free_htab_elem(htab, l); 1264 ret = 0; 1265 } 1266 1267 htab_unlock_bucket(htab, b, flags); 1268 return ret; 1269 } 1270 1271 static int htab_lru_map_delete_elem(struct bpf_map *map, void *key) 1272 { 1273 struct bpf_htab *htab = container_of(map, struct bpf_htab, map); 1274 struct hlist_nulls_head *head; 1275 struct bucket *b; 1276 struct htab_elem *l; 1277 unsigned long flags; 1278 u32 hash, key_size; 1279 int ret = -ENOENT; 1280 1281 WARN_ON_ONCE(!rcu_read_lock_held() && !rcu_read_lock_trace_held()); 1282 1283 key_size = map->key_size; 1284 1285 hash = htab_map_hash(key, key_size, htab->hashrnd); 1286 b = __select_bucket(htab, hash); 1287 head = &b->head; 1288 1289 flags = htab_lock_bucket(htab, b); 1290 1291 l = lookup_elem_raw(head, hash, key, key_size); 1292 1293 if (l) { 1294 hlist_nulls_del_rcu(&l->hash_node); 1295 ret = 0; 1296 } 1297 1298 htab_unlock_bucket(htab, b, flags); 1299 if (l) 1300 bpf_lru_push_free(&htab->lru, &l->lru_node); 1301 return ret; 1302 } 1303 1304 static void delete_all_elements(struct bpf_htab *htab) 1305 { 1306 int i; 1307 1308 for (i = 0; i < htab->n_buckets; i++) { 1309 struct hlist_nulls_head *head = select_bucket(htab, i); 1310 struct hlist_nulls_node *n; 1311 struct htab_elem *l; 1312 1313 hlist_nulls_for_each_entry_safe(l, n, head, hash_node) { 1314 hlist_nulls_del_rcu(&l->hash_node); 1315 htab_elem_free(htab, l); 1316 } 1317 } 1318 } 1319 1320 /* Called when map->refcnt goes to zero, either from workqueue or from syscall */ 1321 static void htab_map_free(struct bpf_map *map) 1322 { 1323 struct bpf_htab *htab = container_of(map, struct bpf_htab, map); 1324 1325 /* bpf_free_used_maps() or close(map_fd) will trigger this map_free callback. 1326 * bpf_free_used_maps() is called after bpf prog is no longer executing. 1327 * There is no need to synchronize_rcu() here to protect map elements. 1328 */ 1329 1330 /* some of free_htab_elem() callbacks for elements of this map may 1331 * not have executed. Wait for them. 1332 */ 1333 rcu_barrier(); 1334 if (!htab_is_prealloc(htab)) 1335 delete_all_elements(htab); 1336 else 1337 prealloc_destroy(htab); 1338 1339 free_percpu(htab->extra_elems); 1340 bpf_map_area_free(htab->buckets); 1341 kfree(htab); 1342 } 1343 1344 static void htab_map_seq_show_elem(struct bpf_map *map, void *key, 1345 struct seq_file *m) 1346 { 1347 void *value; 1348 1349 rcu_read_lock(); 1350 1351 value = htab_map_lookup_elem(map, key); 1352 if (!value) { 1353 rcu_read_unlock(); 1354 return; 1355 } 1356 1357 btf_type_seq_show(map->btf, map->btf_key_type_id, key, m); 1358 seq_puts(m, ": "); 1359 btf_type_seq_show(map->btf, map->btf_value_type_id, value, m); 1360 seq_puts(m, "\n"); 1361 1362 rcu_read_unlock(); 1363 } 1364 1365 static int 1366 __htab_map_lookup_and_delete_batch(struct bpf_map *map, 1367 const union bpf_attr *attr, 1368 union bpf_attr __user *uattr, 1369 bool do_delete, bool is_lru_map, 1370 bool is_percpu) 1371 { 1372 struct bpf_htab *htab = container_of(map, struct bpf_htab, map); 1373 u32 bucket_cnt, total, key_size, value_size, roundup_key_size; 1374 void *keys = NULL, *values = NULL, *value, *dst_key, *dst_val; 1375 void __user *uvalues = u64_to_user_ptr(attr->batch.values); 1376 void __user *ukeys = u64_to_user_ptr(attr->batch.keys); 1377 void *ubatch = u64_to_user_ptr(attr->batch.in_batch); 1378 u32 batch, max_count, size, bucket_size; 1379 struct htab_elem *node_to_free = NULL; 1380 u64 elem_map_flags, map_flags; 1381 struct hlist_nulls_head *head; 1382 struct hlist_nulls_node *n; 1383 unsigned long flags = 0; 1384 bool locked = false; 1385 struct htab_elem *l; 1386 struct bucket *b; 1387 int ret = 0; 1388 1389 elem_map_flags = attr->batch.elem_flags; 1390 if ((elem_map_flags & ~BPF_F_LOCK) || 1391 ((elem_map_flags & BPF_F_LOCK) && !map_value_has_spin_lock(map))) 1392 return -EINVAL; 1393 1394 map_flags = attr->batch.flags; 1395 if (map_flags) 1396 return -EINVAL; 1397 1398 max_count = attr->batch.count; 1399 if (!max_count) 1400 return 0; 1401 1402 if (put_user(0, &uattr->batch.count)) 1403 return -EFAULT; 1404 1405 batch = 0; 1406 if (ubatch && copy_from_user(&batch, ubatch, sizeof(batch))) 1407 return -EFAULT; 1408 1409 if (batch >= htab->n_buckets) 1410 return -ENOENT; 1411 1412 key_size = htab->map.key_size; 1413 roundup_key_size = round_up(htab->map.key_size, 8); 1414 value_size = htab->map.value_size; 1415 size = round_up(value_size, 8); 1416 if (is_percpu) 1417 value_size = size * num_possible_cpus(); 1418 total = 0; 1419 /* while experimenting with hash tables with sizes ranging from 10 to 1420 * 1000, it was observed that a bucket can have upto 5 entries. 1421 */ 1422 bucket_size = 5; 1423 1424 alloc: 1425 /* We cannot do copy_from_user or copy_to_user inside 1426 * the rcu_read_lock. Allocate enough space here. 1427 */ 1428 keys = kvmalloc(key_size * bucket_size, GFP_USER | __GFP_NOWARN); 1429 values = kvmalloc(value_size * bucket_size, GFP_USER | __GFP_NOWARN); 1430 if (!keys || !values) { 1431 ret = -ENOMEM; 1432 goto after_loop; 1433 } 1434 1435 again: 1436 bpf_disable_instrumentation(); 1437 rcu_read_lock(); 1438 again_nocopy: 1439 dst_key = keys; 1440 dst_val = values; 1441 b = &htab->buckets[batch]; 1442 head = &b->head; 1443 /* do not grab the lock unless need it (bucket_cnt > 0). */ 1444 if (locked) 1445 flags = htab_lock_bucket(htab, b); 1446 1447 bucket_cnt = 0; 1448 hlist_nulls_for_each_entry_rcu(l, n, head, hash_node) 1449 bucket_cnt++; 1450 1451 if (bucket_cnt && !locked) { 1452 locked = true; 1453 goto again_nocopy; 1454 } 1455 1456 if (bucket_cnt > (max_count - total)) { 1457 if (total == 0) 1458 ret = -ENOSPC; 1459 /* Note that since bucket_cnt > 0 here, it is implicit 1460 * that the locked was grabbed, so release it. 1461 */ 1462 htab_unlock_bucket(htab, b, flags); 1463 rcu_read_unlock(); 1464 bpf_enable_instrumentation(); 1465 goto after_loop; 1466 } 1467 1468 if (bucket_cnt > bucket_size) { 1469 bucket_size = bucket_cnt; 1470 /* Note that since bucket_cnt > 0 here, it is implicit 1471 * that the locked was grabbed, so release it. 1472 */ 1473 htab_unlock_bucket(htab, b, flags); 1474 rcu_read_unlock(); 1475 bpf_enable_instrumentation(); 1476 kvfree(keys); 1477 kvfree(values); 1478 goto alloc; 1479 } 1480 1481 /* Next block is only safe to run if you have grabbed the lock */ 1482 if (!locked) 1483 goto next_batch; 1484 1485 hlist_nulls_for_each_entry_safe(l, n, head, hash_node) { 1486 memcpy(dst_key, l->key, key_size); 1487 1488 if (is_percpu) { 1489 int off = 0, cpu; 1490 void __percpu *pptr; 1491 1492 pptr = htab_elem_get_ptr(l, map->key_size); 1493 for_each_possible_cpu(cpu) { 1494 bpf_long_memcpy(dst_val + off, 1495 per_cpu_ptr(pptr, cpu), size); 1496 off += size; 1497 } 1498 } else { 1499 value = l->key + roundup_key_size; 1500 if (elem_map_flags & BPF_F_LOCK) 1501 copy_map_value_locked(map, dst_val, value, 1502 true); 1503 else 1504 copy_map_value(map, dst_val, value); 1505 check_and_init_map_lock(map, dst_val); 1506 } 1507 if (do_delete) { 1508 hlist_nulls_del_rcu(&l->hash_node); 1509 1510 /* bpf_lru_push_free() will acquire lru_lock, which 1511 * may cause deadlock. See comments in function 1512 * prealloc_lru_pop(). Let us do bpf_lru_push_free() 1513 * after releasing the bucket lock. 1514 */ 1515 if (is_lru_map) { 1516 l->batch_flink = node_to_free; 1517 node_to_free = l; 1518 } else { 1519 free_htab_elem(htab, l); 1520 } 1521 } 1522 dst_key += key_size; 1523 dst_val += value_size; 1524 } 1525 1526 htab_unlock_bucket(htab, b, flags); 1527 locked = false; 1528 1529 while (node_to_free) { 1530 l = node_to_free; 1531 node_to_free = node_to_free->batch_flink; 1532 bpf_lru_push_free(&htab->lru, &l->lru_node); 1533 } 1534 1535 next_batch: 1536 /* If we are not copying data, we can go to next bucket and avoid 1537 * unlocking the rcu. 1538 */ 1539 if (!bucket_cnt && (batch + 1 < htab->n_buckets)) { 1540 batch++; 1541 goto again_nocopy; 1542 } 1543 1544 rcu_read_unlock(); 1545 bpf_enable_instrumentation(); 1546 if (bucket_cnt && (copy_to_user(ukeys + total * key_size, keys, 1547 key_size * bucket_cnt) || 1548 copy_to_user(uvalues + total * value_size, values, 1549 value_size * bucket_cnt))) { 1550 ret = -EFAULT; 1551 goto after_loop; 1552 } 1553 1554 total += bucket_cnt; 1555 batch++; 1556 if (batch >= htab->n_buckets) { 1557 ret = -ENOENT; 1558 goto after_loop; 1559 } 1560 goto again; 1561 1562 after_loop: 1563 if (ret == -EFAULT) 1564 goto out; 1565 1566 /* copy # of entries and next batch */ 1567 ubatch = u64_to_user_ptr(attr->batch.out_batch); 1568 if (copy_to_user(ubatch, &batch, sizeof(batch)) || 1569 put_user(total, &uattr->batch.count)) 1570 ret = -EFAULT; 1571 1572 out: 1573 kvfree(keys); 1574 kvfree(values); 1575 return ret; 1576 } 1577 1578 static int 1579 htab_percpu_map_lookup_batch(struct bpf_map *map, const union bpf_attr *attr, 1580 union bpf_attr __user *uattr) 1581 { 1582 return __htab_map_lookup_and_delete_batch(map, attr, uattr, false, 1583 false, true); 1584 } 1585 1586 static int 1587 htab_percpu_map_lookup_and_delete_batch(struct bpf_map *map, 1588 const union bpf_attr *attr, 1589 union bpf_attr __user *uattr) 1590 { 1591 return __htab_map_lookup_and_delete_batch(map, attr, uattr, true, 1592 false, true); 1593 } 1594 1595 static int 1596 htab_map_lookup_batch(struct bpf_map *map, const union bpf_attr *attr, 1597 union bpf_attr __user *uattr) 1598 { 1599 return __htab_map_lookup_and_delete_batch(map, attr, uattr, false, 1600 false, false); 1601 } 1602 1603 static int 1604 htab_map_lookup_and_delete_batch(struct bpf_map *map, 1605 const union bpf_attr *attr, 1606 union bpf_attr __user *uattr) 1607 { 1608 return __htab_map_lookup_and_delete_batch(map, attr, uattr, true, 1609 false, false); 1610 } 1611 1612 static int 1613 htab_lru_percpu_map_lookup_batch(struct bpf_map *map, 1614 const union bpf_attr *attr, 1615 union bpf_attr __user *uattr) 1616 { 1617 return __htab_map_lookup_and_delete_batch(map, attr, uattr, false, 1618 true, true); 1619 } 1620 1621 static int 1622 htab_lru_percpu_map_lookup_and_delete_batch(struct bpf_map *map, 1623 const union bpf_attr *attr, 1624 union bpf_attr __user *uattr) 1625 { 1626 return __htab_map_lookup_and_delete_batch(map, attr, uattr, true, 1627 true, true); 1628 } 1629 1630 static int 1631 htab_lru_map_lookup_batch(struct bpf_map *map, const union bpf_attr *attr, 1632 union bpf_attr __user *uattr) 1633 { 1634 return __htab_map_lookup_and_delete_batch(map, attr, uattr, false, 1635 true, false); 1636 } 1637 1638 static int 1639 htab_lru_map_lookup_and_delete_batch(struct bpf_map *map, 1640 const union bpf_attr *attr, 1641 union bpf_attr __user *uattr) 1642 { 1643 return __htab_map_lookup_and_delete_batch(map, attr, uattr, true, 1644 true, false); 1645 } 1646 1647 struct bpf_iter_seq_hash_map_info { 1648 struct bpf_map *map; 1649 struct bpf_htab *htab; 1650 void *percpu_value_buf; // non-zero means percpu hash 1651 u32 bucket_id; 1652 u32 skip_elems; 1653 }; 1654 1655 static struct htab_elem * 1656 bpf_hash_map_seq_find_next(struct bpf_iter_seq_hash_map_info *info, 1657 struct htab_elem *prev_elem) 1658 { 1659 const struct bpf_htab *htab = info->htab; 1660 u32 skip_elems = info->skip_elems; 1661 u32 bucket_id = info->bucket_id; 1662 struct hlist_nulls_head *head; 1663 struct hlist_nulls_node *n; 1664 struct htab_elem *elem; 1665 struct bucket *b; 1666 u32 i, count; 1667 1668 if (bucket_id >= htab->n_buckets) 1669 return NULL; 1670 1671 /* try to find next elem in the same bucket */ 1672 if (prev_elem) { 1673 /* no update/deletion on this bucket, prev_elem should be still valid 1674 * and we won't skip elements. 1675 */ 1676 n = rcu_dereference_raw(hlist_nulls_next_rcu(&prev_elem->hash_node)); 1677 elem = hlist_nulls_entry_safe(n, struct htab_elem, hash_node); 1678 if (elem) 1679 return elem; 1680 1681 /* not found, unlock and go to the next bucket */ 1682 b = &htab->buckets[bucket_id++]; 1683 rcu_read_unlock(); 1684 skip_elems = 0; 1685 } 1686 1687 for (i = bucket_id; i < htab->n_buckets; i++) { 1688 b = &htab->buckets[i]; 1689 rcu_read_lock(); 1690 1691 count = 0; 1692 head = &b->head; 1693 hlist_nulls_for_each_entry_rcu(elem, n, head, hash_node) { 1694 if (count >= skip_elems) { 1695 info->bucket_id = i; 1696 info->skip_elems = count; 1697 return elem; 1698 } 1699 count++; 1700 } 1701 1702 rcu_read_unlock(); 1703 skip_elems = 0; 1704 } 1705 1706 info->bucket_id = i; 1707 info->skip_elems = 0; 1708 return NULL; 1709 } 1710 1711 static void *bpf_hash_map_seq_start(struct seq_file *seq, loff_t *pos) 1712 { 1713 struct bpf_iter_seq_hash_map_info *info = seq->private; 1714 struct htab_elem *elem; 1715 1716 elem = bpf_hash_map_seq_find_next(info, NULL); 1717 if (!elem) 1718 return NULL; 1719 1720 if (*pos == 0) 1721 ++*pos; 1722 return elem; 1723 } 1724 1725 static void *bpf_hash_map_seq_next(struct seq_file *seq, void *v, loff_t *pos) 1726 { 1727 struct bpf_iter_seq_hash_map_info *info = seq->private; 1728 1729 ++*pos; 1730 ++info->skip_elems; 1731 return bpf_hash_map_seq_find_next(info, v); 1732 } 1733 1734 static int __bpf_hash_map_seq_show(struct seq_file *seq, struct htab_elem *elem) 1735 { 1736 struct bpf_iter_seq_hash_map_info *info = seq->private; 1737 u32 roundup_key_size, roundup_value_size; 1738 struct bpf_iter__bpf_map_elem ctx = {}; 1739 struct bpf_map *map = info->map; 1740 struct bpf_iter_meta meta; 1741 int ret = 0, off = 0, cpu; 1742 struct bpf_prog *prog; 1743 void __percpu *pptr; 1744 1745 meta.seq = seq; 1746 prog = bpf_iter_get_info(&meta, elem == NULL); 1747 if (prog) { 1748 ctx.meta = &meta; 1749 ctx.map = info->map; 1750 if (elem) { 1751 roundup_key_size = round_up(map->key_size, 8); 1752 ctx.key = elem->key; 1753 if (!info->percpu_value_buf) { 1754 ctx.value = elem->key + roundup_key_size; 1755 } else { 1756 roundup_value_size = round_up(map->value_size, 8); 1757 pptr = htab_elem_get_ptr(elem, map->key_size); 1758 for_each_possible_cpu(cpu) { 1759 bpf_long_memcpy(info->percpu_value_buf + off, 1760 per_cpu_ptr(pptr, cpu), 1761 roundup_value_size); 1762 off += roundup_value_size; 1763 } 1764 ctx.value = info->percpu_value_buf; 1765 } 1766 } 1767 ret = bpf_iter_run_prog(prog, &ctx); 1768 } 1769 1770 return ret; 1771 } 1772 1773 static int bpf_hash_map_seq_show(struct seq_file *seq, void *v) 1774 { 1775 return __bpf_hash_map_seq_show(seq, v); 1776 } 1777 1778 static void bpf_hash_map_seq_stop(struct seq_file *seq, void *v) 1779 { 1780 if (!v) 1781 (void)__bpf_hash_map_seq_show(seq, NULL); 1782 else 1783 rcu_read_unlock(); 1784 } 1785 1786 static int bpf_iter_init_hash_map(void *priv_data, 1787 struct bpf_iter_aux_info *aux) 1788 { 1789 struct bpf_iter_seq_hash_map_info *seq_info = priv_data; 1790 struct bpf_map *map = aux->map; 1791 void *value_buf; 1792 u32 buf_size; 1793 1794 if (map->map_type == BPF_MAP_TYPE_PERCPU_HASH || 1795 map->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH) { 1796 buf_size = round_up(map->value_size, 8) * num_possible_cpus(); 1797 value_buf = kmalloc(buf_size, GFP_USER | __GFP_NOWARN); 1798 if (!value_buf) 1799 return -ENOMEM; 1800 1801 seq_info->percpu_value_buf = value_buf; 1802 } 1803 1804 seq_info->map = map; 1805 seq_info->htab = container_of(map, struct bpf_htab, map); 1806 return 0; 1807 } 1808 1809 static void bpf_iter_fini_hash_map(void *priv_data) 1810 { 1811 struct bpf_iter_seq_hash_map_info *seq_info = priv_data; 1812 1813 kfree(seq_info->percpu_value_buf); 1814 } 1815 1816 static const struct seq_operations bpf_hash_map_seq_ops = { 1817 .start = bpf_hash_map_seq_start, 1818 .next = bpf_hash_map_seq_next, 1819 .stop = bpf_hash_map_seq_stop, 1820 .show = bpf_hash_map_seq_show, 1821 }; 1822 1823 static const struct bpf_iter_seq_info iter_seq_info = { 1824 .seq_ops = &bpf_hash_map_seq_ops, 1825 .init_seq_private = bpf_iter_init_hash_map, 1826 .fini_seq_private = bpf_iter_fini_hash_map, 1827 .seq_priv_size = sizeof(struct bpf_iter_seq_hash_map_info), 1828 }; 1829 1830 static int htab_map_btf_id; 1831 const struct bpf_map_ops htab_map_ops = { 1832 .map_meta_equal = bpf_map_meta_equal, 1833 .map_alloc_check = htab_map_alloc_check, 1834 .map_alloc = htab_map_alloc, 1835 .map_free = htab_map_free, 1836 .map_get_next_key = htab_map_get_next_key, 1837 .map_lookup_elem = htab_map_lookup_elem, 1838 .map_update_elem = htab_map_update_elem, 1839 .map_delete_elem = htab_map_delete_elem, 1840 .map_gen_lookup = htab_map_gen_lookup, 1841 .map_seq_show_elem = htab_map_seq_show_elem, 1842 BATCH_OPS(htab), 1843 .map_btf_name = "bpf_htab", 1844 .map_btf_id = &htab_map_btf_id, 1845 .iter_seq_info = &iter_seq_info, 1846 }; 1847 1848 static int htab_lru_map_btf_id; 1849 const struct bpf_map_ops htab_lru_map_ops = { 1850 .map_meta_equal = bpf_map_meta_equal, 1851 .map_alloc_check = htab_map_alloc_check, 1852 .map_alloc = htab_map_alloc, 1853 .map_free = htab_map_free, 1854 .map_get_next_key = htab_map_get_next_key, 1855 .map_lookup_elem = htab_lru_map_lookup_elem, 1856 .map_lookup_elem_sys_only = htab_lru_map_lookup_elem_sys, 1857 .map_update_elem = htab_lru_map_update_elem, 1858 .map_delete_elem = htab_lru_map_delete_elem, 1859 .map_gen_lookup = htab_lru_map_gen_lookup, 1860 .map_seq_show_elem = htab_map_seq_show_elem, 1861 BATCH_OPS(htab_lru), 1862 .map_btf_name = "bpf_htab", 1863 .map_btf_id = &htab_lru_map_btf_id, 1864 .iter_seq_info = &iter_seq_info, 1865 }; 1866 1867 /* Called from eBPF program */ 1868 static void *htab_percpu_map_lookup_elem(struct bpf_map *map, void *key) 1869 { 1870 struct htab_elem *l = __htab_map_lookup_elem(map, key); 1871 1872 if (l) 1873 return this_cpu_ptr(htab_elem_get_ptr(l, map->key_size)); 1874 else 1875 return NULL; 1876 } 1877 1878 static void *htab_lru_percpu_map_lookup_elem(struct bpf_map *map, void *key) 1879 { 1880 struct htab_elem *l = __htab_map_lookup_elem(map, key); 1881 1882 if (l) { 1883 bpf_lru_node_set_ref(&l->lru_node); 1884 return this_cpu_ptr(htab_elem_get_ptr(l, map->key_size)); 1885 } 1886 1887 return NULL; 1888 } 1889 1890 int bpf_percpu_hash_copy(struct bpf_map *map, void *key, void *value) 1891 { 1892 struct htab_elem *l; 1893 void __percpu *pptr; 1894 int ret = -ENOENT; 1895 int cpu, off = 0; 1896 u32 size; 1897 1898 /* per_cpu areas are zero-filled and bpf programs can only 1899 * access 'value_size' of them, so copying rounded areas 1900 * will not leak any kernel data 1901 */ 1902 size = round_up(map->value_size, 8); 1903 rcu_read_lock(); 1904 l = __htab_map_lookup_elem(map, key); 1905 if (!l) 1906 goto out; 1907 /* We do not mark LRU map element here in order to not mess up 1908 * eviction heuristics when user space does a map walk. 1909 */ 1910 pptr = htab_elem_get_ptr(l, map->key_size); 1911 for_each_possible_cpu(cpu) { 1912 bpf_long_memcpy(value + off, 1913 per_cpu_ptr(pptr, cpu), size); 1914 off += size; 1915 } 1916 ret = 0; 1917 out: 1918 rcu_read_unlock(); 1919 return ret; 1920 } 1921 1922 int bpf_percpu_hash_update(struct bpf_map *map, void *key, void *value, 1923 u64 map_flags) 1924 { 1925 struct bpf_htab *htab = container_of(map, struct bpf_htab, map); 1926 int ret; 1927 1928 rcu_read_lock(); 1929 if (htab_is_lru(htab)) 1930 ret = __htab_lru_percpu_map_update_elem(map, key, value, 1931 map_flags, true); 1932 else 1933 ret = __htab_percpu_map_update_elem(map, key, value, map_flags, 1934 true); 1935 rcu_read_unlock(); 1936 1937 return ret; 1938 } 1939 1940 static void htab_percpu_map_seq_show_elem(struct bpf_map *map, void *key, 1941 struct seq_file *m) 1942 { 1943 struct htab_elem *l; 1944 void __percpu *pptr; 1945 int cpu; 1946 1947 rcu_read_lock(); 1948 1949 l = __htab_map_lookup_elem(map, key); 1950 if (!l) { 1951 rcu_read_unlock(); 1952 return; 1953 } 1954 1955 btf_type_seq_show(map->btf, map->btf_key_type_id, key, m); 1956 seq_puts(m, ": {\n"); 1957 pptr = htab_elem_get_ptr(l, map->key_size); 1958 for_each_possible_cpu(cpu) { 1959 seq_printf(m, "\tcpu%d: ", cpu); 1960 btf_type_seq_show(map->btf, map->btf_value_type_id, 1961 per_cpu_ptr(pptr, cpu), m); 1962 seq_puts(m, "\n"); 1963 } 1964 seq_puts(m, "}\n"); 1965 1966 rcu_read_unlock(); 1967 } 1968 1969 static int htab_percpu_map_btf_id; 1970 const struct bpf_map_ops htab_percpu_map_ops = { 1971 .map_meta_equal = bpf_map_meta_equal, 1972 .map_alloc_check = htab_map_alloc_check, 1973 .map_alloc = htab_map_alloc, 1974 .map_free = htab_map_free, 1975 .map_get_next_key = htab_map_get_next_key, 1976 .map_lookup_elem = htab_percpu_map_lookup_elem, 1977 .map_update_elem = htab_percpu_map_update_elem, 1978 .map_delete_elem = htab_map_delete_elem, 1979 .map_seq_show_elem = htab_percpu_map_seq_show_elem, 1980 BATCH_OPS(htab_percpu), 1981 .map_btf_name = "bpf_htab", 1982 .map_btf_id = &htab_percpu_map_btf_id, 1983 .iter_seq_info = &iter_seq_info, 1984 }; 1985 1986 static int htab_lru_percpu_map_btf_id; 1987 const struct bpf_map_ops htab_lru_percpu_map_ops = { 1988 .map_meta_equal = bpf_map_meta_equal, 1989 .map_alloc_check = htab_map_alloc_check, 1990 .map_alloc = htab_map_alloc, 1991 .map_free = htab_map_free, 1992 .map_get_next_key = htab_map_get_next_key, 1993 .map_lookup_elem = htab_lru_percpu_map_lookup_elem, 1994 .map_update_elem = htab_lru_percpu_map_update_elem, 1995 .map_delete_elem = htab_lru_map_delete_elem, 1996 .map_seq_show_elem = htab_percpu_map_seq_show_elem, 1997 BATCH_OPS(htab_lru_percpu), 1998 .map_btf_name = "bpf_htab", 1999 .map_btf_id = &htab_lru_percpu_map_btf_id, 2000 .iter_seq_info = &iter_seq_info, 2001 }; 2002 2003 static int fd_htab_map_alloc_check(union bpf_attr *attr) 2004 { 2005 if (attr->value_size != sizeof(u32)) 2006 return -EINVAL; 2007 return htab_map_alloc_check(attr); 2008 } 2009 2010 static void fd_htab_map_free(struct bpf_map *map) 2011 { 2012 struct bpf_htab *htab = container_of(map, struct bpf_htab, map); 2013 struct hlist_nulls_node *n; 2014 struct hlist_nulls_head *head; 2015 struct htab_elem *l; 2016 int i; 2017 2018 for (i = 0; i < htab->n_buckets; i++) { 2019 head = select_bucket(htab, i); 2020 2021 hlist_nulls_for_each_entry_safe(l, n, head, hash_node) { 2022 void *ptr = fd_htab_map_get_ptr(map, l); 2023 2024 map->ops->map_fd_put_ptr(ptr); 2025 } 2026 } 2027 2028 htab_map_free(map); 2029 } 2030 2031 /* only called from syscall */ 2032 int bpf_fd_htab_map_lookup_elem(struct bpf_map *map, void *key, u32 *value) 2033 { 2034 void **ptr; 2035 int ret = 0; 2036 2037 if (!map->ops->map_fd_sys_lookup_elem) 2038 return -ENOTSUPP; 2039 2040 rcu_read_lock(); 2041 ptr = htab_map_lookup_elem(map, key); 2042 if (ptr) 2043 *value = map->ops->map_fd_sys_lookup_elem(READ_ONCE(*ptr)); 2044 else 2045 ret = -ENOENT; 2046 rcu_read_unlock(); 2047 2048 return ret; 2049 } 2050 2051 /* only called from syscall */ 2052 int bpf_fd_htab_map_update_elem(struct bpf_map *map, struct file *map_file, 2053 void *key, void *value, u64 map_flags) 2054 { 2055 void *ptr; 2056 int ret; 2057 u32 ufd = *(u32 *)value; 2058 2059 ptr = map->ops->map_fd_get_ptr(map, map_file, ufd); 2060 if (IS_ERR(ptr)) 2061 return PTR_ERR(ptr); 2062 2063 ret = htab_map_update_elem(map, key, &ptr, map_flags); 2064 if (ret) 2065 map->ops->map_fd_put_ptr(ptr); 2066 2067 return ret; 2068 } 2069 2070 static struct bpf_map *htab_of_map_alloc(union bpf_attr *attr) 2071 { 2072 struct bpf_map *map, *inner_map_meta; 2073 2074 inner_map_meta = bpf_map_meta_alloc(attr->inner_map_fd); 2075 if (IS_ERR(inner_map_meta)) 2076 return inner_map_meta; 2077 2078 map = htab_map_alloc(attr); 2079 if (IS_ERR(map)) { 2080 bpf_map_meta_free(inner_map_meta); 2081 return map; 2082 } 2083 2084 map->inner_map_meta = inner_map_meta; 2085 2086 return map; 2087 } 2088 2089 static void *htab_of_map_lookup_elem(struct bpf_map *map, void *key) 2090 { 2091 struct bpf_map **inner_map = htab_map_lookup_elem(map, key); 2092 2093 if (!inner_map) 2094 return NULL; 2095 2096 return READ_ONCE(*inner_map); 2097 } 2098 2099 static int htab_of_map_gen_lookup(struct bpf_map *map, 2100 struct bpf_insn *insn_buf) 2101 { 2102 struct bpf_insn *insn = insn_buf; 2103 const int ret = BPF_REG_0; 2104 2105 BUILD_BUG_ON(!__same_type(&__htab_map_lookup_elem, 2106 (void *(*)(struct bpf_map *map, void *key))NULL)); 2107 *insn++ = BPF_EMIT_CALL(BPF_CAST_CALL(__htab_map_lookup_elem)); 2108 *insn++ = BPF_JMP_IMM(BPF_JEQ, ret, 0, 2); 2109 *insn++ = BPF_ALU64_IMM(BPF_ADD, ret, 2110 offsetof(struct htab_elem, key) + 2111 round_up(map->key_size, 8)); 2112 *insn++ = BPF_LDX_MEM(BPF_DW, ret, ret, 0); 2113 2114 return insn - insn_buf; 2115 } 2116 2117 static void htab_of_map_free(struct bpf_map *map) 2118 { 2119 bpf_map_meta_free(map->inner_map_meta); 2120 fd_htab_map_free(map); 2121 } 2122 2123 static int htab_of_maps_map_btf_id; 2124 const struct bpf_map_ops htab_of_maps_map_ops = { 2125 .map_alloc_check = fd_htab_map_alloc_check, 2126 .map_alloc = htab_of_map_alloc, 2127 .map_free = htab_of_map_free, 2128 .map_get_next_key = htab_map_get_next_key, 2129 .map_lookup_elem = htab_of_map_lookup_elem, 2130 .map_delete_elem = htab_map_delete_elem, 2131 .map_fd_get_ptr = bpf_map_fd_get_ptr, 2132 .map_fd_put_ptr = bpf_map_fd_put_ptr, 2133 .map_fd_sys_lookup_elem = bpf_map_fd_sys_lookup_elem, 2134 .map_gen_lookup = htab_of_map_gen_lookup, 2135 .map_check_btf = map_check_no_btf, 2136 .map_btf_name = "bpf_htab", 2137 .map_btf_id = &htab_of_maps_map_btf_id, 2138 }; 2139