| 1 | /* |
| 2 | * SPDX-License-Identifier: LGPL-2.1-or-later |
| 3 | * |
| 4 | * Copyright 2010-2011 Mathieu Desnoyers <mathieu.desnoyers@efficios.com> |
| 5 | * Copyright 2011 Lai Jiangshan <laijs@cn.fujitsu.com> |
| 6 | * |
| 7 | * Userspace RCU library - Lock-Free Resizable RCU Hash Table |
| 8 | */ |
| 9 | |
| 10 | /* |
| 11 | * Based on the following articles: |
| 12 | * - Ori Shalev and Nir Shavit. Split-ordered lists: Lock-free |
| 13 | * extensible hash tables. J. ACM 53, 3 (May 2006), 379-405. |
| 14 | * - Michael, M. M. High performance dynamic lock-free hash tables |
| 15 | * and list-based sets. In Proceedings of the fourteenth annual ACM |
| 16 | * symposium on Parallel algorithms and architectures, ACM Press, |
| 17 | * (2002), 73-82. |
| 18 | * |
| 19 | * Some specificities of this Lock-Free Resizable RCU Hash Table |
| 20 | * implementation: |
| 21 | * |
| 22 | * - RCU read-side critical section allows readers to perform hash |
| 23 | * table lookups, as well as traversals, and use the returned objects |
| 24 | * safely by allowing memory reclaim to take place only after a grace |
| 25 | * period. |
| 26 | * - Add and remove operations are lock-free, and do not need to |
| 27 | * allocate memory. They need to be executed within RCU read-side |
| 28 | * critical section to ensure the objects they read are valid and to |
| 29 | * deal with the cmpxchg ABA problem. |
| 30 | * - add and add_unique operations are supported. add_unique checks if |
| 31 | * the node key already exists in the hash table. It ensures not to |
| 32 | * populate a duplicate key if the node key already exists in the hash |
| 33 | * table. |
| 34 | * - The resize operation executes concurrently with |
| 35 | * add/add_unique/add_replace/remove/lookup/traversal. |
| 36 | * - Hash table nodes are contained within a split-ordered list. This |
| 37 | * list is ordered by incrementing reversed-bits-hash value. |
| 38 | * - An index of bucket nodes is kept. These bucket nodes are the hash |
| 39 | * table "buckets". These buckets are internal nodes that allow to |
| 40 | * perform a fast hash lookup, similarly to a skip list. These |
| 41 | * buckets are chained together in the split-ordered list, which |
| 42 | * allows recursive expansion by inserting new buckets between the |
| 43 | * existing buckets. The split-ordered list allows adding new buckets |
| 44 | * between existing buckets as the table needs to grow. |
| 45 | * - The resize operation for small tables only allows expanding the |
| 46 | * hash table. It is triggered automatically by detecting long chains |
| 47 | * in the add operation. |
| 48 | * - The resize operation for larger tables (and available through an |
| 49 | * API) allows both expanding and shrinking the hash table. |
| 50 | * - Split-counters are used to keep track of the number of |
| 51 | * nodes within the hash table for automatic resize triggering. |
| 52 | * - Resize operation initiated by long chain detection is executed by a |
| 53 | * worker thread, which keeps lock-freedom of add and remove. |
| 54 | * - Resize operations are protected by a mutex. |
| 55 | * - The removal operation is split in two parts: first, a "removed" |
| 56 | * flag is set in the next pointer within the node to remove. Then, |
| 57 | * a "garbage collection" is performed in the bucket containing the |
| 58 | * removed node (from the start of the bucket up to the removed node). |
| 59 | * All encountered nodes with "removed" flag set in their next |
| 60 | * pointers are removed from the linked-list. If the cmpxchg used for |
| 61 | * removal fails (due to concurrent garbage-collection or concurrent |
| 62 | * add), we retry from the beginning of the bucket. This ensures that |
| 63 | * the node with "removed" flag set is removed from the hash table |
| 64 | * (not visible to lookups anymore) before the RCU read-side critical |
| 65 | * section held across removal ends. Furthermore, this ensures that |
| 66 | * the node with "removed" flag set is removed from the linked-list |
| 67 | * before its memory is reclaimed. After setting the "removal" flag, |
| 68 | * only the thread which removal is the first to set the "removal |
| 69 | * owner" flag (with an xchg) into a node's next pointer is considered |
| 70 | * to have succeeded its removal (and thus owns the node to reclaim). |
| 71 | * Because we garbage-collect starting from an invariant node (the |
| 72 | * start-of-bucket bucket node) up to the "removed" node (or find a |
| 73 | * reverse-hash that is higher), we are sure that a successful |
| 74 | * traversal of the chain leads to a chain that is present in the |
| 75 | * linked-list (the start node is never removed) and that it does not |
| 76 | * contain the "removed" node anymore, even if concurrent delete/add |
| 77 | * operations are changing the structure of the list concurrently. |
| 78 | * - The add operations perform garbage collection of buckets if they |
| 79 | * encounter nodes with removed flag set in the bucket where they want |
| 80 | * to add their new node. This ensures lock-freedom of add operation by |
| 81 | * helping the remover unlink nodes from the list rather than to wait |
| 82 | * for it do to so. |
| 83 | * - There are three memory backends for the hash table buckets: the |
| 84 | * "order table", the "chunks", and the "mmap". |
| 85 | * - These bucket containers contain a compact version of the hash table |
| 86 | * nodes. |
| 87 | * - The RCU "order table": |
| 88 | * - has a first level table indexed by log2(hash index) which is |
| 89 | * copied and expanded by the resize operation. This order table |
| 90 | * allows finding the "bucket node" tables. |
| 91 | * - There is one bucket node table per hash index order. The size of |
| 92 | * each bucket node table is half the number of hashes contained in |
| 93 | * this order (except for order 0). |
| 94 | * - The RCU "chunks" is best suited for close interaction with a page |
| 95 | * allocator. It uses a linear array as index to "chunks" containing |
| 96 | * each the same number of buckets. |
| 97 | * - The RCU "mmap" memory backend uses a single memory map to hold |
| 98 | * all buckets. |
| 99 | * - synchronize_rcu is used to garbage-collect the old bucket node table. |
| 100 | * |
| 101 | * Ordering Guarantees: |
| 102 | * |
| 103 | * To discuss these guarantees, we first define "read" operation as any |
| 104 | * of the the basic lttng_ust_lfht_lookup, lttng_ust_lfht_next_duplicate, |
| 105 | * lttng_ust_lfht_first, lttng_ust_lfht_next operation, as well as |
| 106 | * lttng_ust_lfht_add_unique (failure). |
| 107 | * |
| 108 | * We define "read traversal" operation as any of the following |
| 109 | * group of operations |
| 110 | * - lttng_ust_lfht_lookup followed by iteration with lttng_ust_lfht_next_duplicate |
| 111 | * (and/or lttng_ust_lfht_next, although less common). |
| 112 | * - lttng_ust_lfht_add_unique (failure) followed by iteration with |
| 113 | * lttng_ust_lfht_next_duplicate (and/or lttng_ust_lfht_next, although less |
| 114 | * common). |
| 115 | * - lttng_ust_lfht_first followed iteration with lttng_ust_lfht_next (and/or |
| 116 | * lttng_ust_lfht_next_duplicate, although less common). |
| 117 | * |
| 118 | * We define "write" operations as any of lttng_ust_lfht_add, lttng_ust_lfht_replace, |
| 119 | * lttng_ust_lfht_add_unique (success), lttng_ust_lfht_add_replace, lttng_ust_lfht_del. |
| 120 | * |
| 121 | * When lttng_ust_lfht_add_unique succeeds (returns the node passed as |
| 122 | * parameter), it acts as a "write" operation. When lttng_ust_lfht_add_unique |
| 123 | * fails (returns a node different from the one passed as parameter), it |
| 124 | * acts as a "read" operation. A lttng_ust_lfht_add_unique failure is a |
| 125 | * lttng_ust_lfht_lookup "read" operation, therefore, any ordering guarantee |
| 126 | * referring to "lookup" imply any of "lookup" or lttng_ust_lfht_add_unique |
| 127 | * (failure). |
| 128 | * |
| 129 | * We define "prior" and "later" node as nodes observable by reads and |
| 130 | * read traversals respectively before and after a write or sequence of |
| 131 | * write operations. |
| 132 | * |
| 133 | * Hash-table operations are often cascaded, for example, the pointer |
| 134 | * returned by a lttng_ust_lfht_lookup() might be passed to a lttng_ust_lfht_next(), |
| 135 | * whose return value might in turn be passed to another hash-table |
| 136 | * operation. This entire cascaded series of operations must be enclosed |
| 137 | * by a pair of matching rcu_read_lock() and rcu_read_unlock() |
| 138 | * operations. |
| 139 | * |
| 140 | * The following ordering guarantees are offered by this hash table: |
| 141 | * |
| 142 | * A.1) "read" after "write": if there is ordering between a write and a |
| 143 | * later read, then the read is guaranteed to see the write or some |
| 144 | * later write. |
| 145 | * A.2) "read traversal" after "write": given that there is dependency |
| 146 | * ordering between reads in a "read traversal", if there is |
| 147 | * ordering between a write and the first read of the traversal, |
| 148 | * then the "read traversal" is guaranteed to see the write or |
| 149 | * some later write. |
| 150 | * B.1) "write" after "read": if there is ordering between a read and a |
| 151 | * later write, then the read will never see the write. |
| 152 | * B.2) "write" after "read traversal": given that there is dependency |
| 153 | * ordering between reads in a "read traversal", if there is |
| 154 | * ordering between the last read of the traversal and a later |
| 155 | * write, then the "read traversal" will never see the write. |
| 156 | * C) "write" while "read traversal": if a write occurs during a "read |
| 157 | * traversal", the traversal may, or may not, see the write. |
| 158 | * D.1) "write" after "write": if there is ordering between a write and |
| 159 | * a later write, then the later write is guaranteed to see the |
| 160 | * effects of the first write. |
| 161 | * D.2) Concurrent "write" pairs: The system will assign an arbitrary |
| 162 | * order to any pair of concurrent conflicting writes. |
| 163 | * Non-conflicting writes (for example, to different keys) are |
| 164 | * unordered. |
| 165 | * E) If a grace period separates a "del" or "replace" operation |
| 166 | * and a subsequent operation, then that subsequent operation is |
| 167 | * guaranteed not to see the removed item. |
| 168 | * F) Uniqueness guarantee: given a hash table that does not contain |
| 169 | * duplicate items for a given key, there will only be one item in |
| 170 | * the hash table after an arbitrary sequence of add_unique and/or |
| 171 | * add_replace operations. Note, however, that a pair of |
| 172 | * concurrent read operations might well access two different items |
| 173 | * with that key. |
| 174 | * G.1) If a pair of lookups for a given key are ordered (e.g. by a |
| 175 | * memory barrier), then the second lookup will return the same |
| 176 | * node as the previous lookup, or some later node. |
| 177 | * G.2) A "read traversal" that starts after the end of a prior "read |
| 178 | * traversal" (ordered by memory barriers) is guaranteed to see the |
| 179 | * same nodes as the previous traversal, or some later nodes. |
| 180 | * G.3) Concurrent "read" pairs: concurrent reads are unordered. For |
| 181 | * example, if a pair of reads to the same key run concurrently |
| 182 | * with an insertion of that same key, the reads remain unordered |
| 183 | * regardless of their return values. In other words, you cannot |
| 184 | * rely on the values returned by the reads to deduce ordering. |
| 185 | * |
| 186 | * Progress guarantees: |
| 187 | * |
| 188 | * * Reads are wait-free. These operations always move forward in the |
| 189 | * hash table linked list, and this list has no loop. |
| 190 | * * Writes are lock-free. Any retry loop performed by a write operation |
| 191 | * is triggered by progress made within another update operation. |
| 192 | * |
| 193 | * Bucket node tables: |
| 194 | * |
| 195 | * hash table hash table the last all bucket node tables |
| 196 | * order size bucket node 0 1 2 3 4 5 6(index) |
| 197 | * table size |
| 198 | * 0 1 1 1 |
| 199 | * 1 2 1 1 1 |
| 200 | * 2 4 2 1 1 2 |
| 201 | * 3 8 4 1 1 2 4 |
| 202 | * 4 16 8 1 1 2 4 8 |
| 203 | * 5 32 16 1 1 2 4 8 16 |
| 204 | * 6 64 32 1 1 2 4 8 16 32 |
| 205 | * |
| 206 | * When growing/shrinking, we only focus on the last bucket node table |
| 207 | * which size is (!order ? 1 : (1 << (order -1))). |
| 208 | * |
| 209 | * Example for growing/shrinking: |
| 210 | * grow hash table from order 5 to 6: init the index=6 bucket node table |
| 211 | * shrink hash table from order 6 to 5: fini the index=6 bucket node table |
| 212 | * |
| 213 | * A bit of ascii art explanation: |
| 214 | * |
| 215 | * The order index is the off-by-one compared to the actual power of 2 |
| 216 | * because we use index 0 to deal with the 0 special-case. |
| 217 | * |
| 218 | * This shows the nodes for a small table ordered by reversed bits: |
| 219 | * |
| 220 | * bits reverse |
| 221 | * 0 000 000 |
| 222 | * 4 100 001 |
| 223 | * 2 010 010 |
| 224 | * 6 110 011 |
| 225 | * 1 001 100 |
| 226 | * 5 101 101 |
| 227 | * 3 011 110 |
| 228 | * 7 111 111 |
| 229 | * |
| 230 | * This shows the nodes in order of non-reversed bits, linked by |
| 231 | * reversed-bit order. |
| 232 | * |
| 233 | * order bits reverse |
| 234 | * 0 0 000 000 |
| 235 | * 1 | 1 001 100 <- |
| 236 | * 2 | | 2 010 010 <- | |
| 237 | * | | | 3 011 110 | <- | |
| 238 | * 3 -> | | | 4 100 001 | | |
| 239 | * -> | | 5 101 101 | |
| 240 | * -> | 6 110 011 |
| 241 | * -> 7 111 111 |
| 242 | */ |
| 243 | |
| 244 | /* |
| 245 | * Note on port to lttng-ust: auto-resize and accounting features are |
| 246 | * removed. |
| 247 | */ |
| 248 | |
| 249 | #define _LGPL_SOURCE |
| 250 | #include <stdlib.h> |
| 251 | #include <errno.h> |
| 252 | #include <assert.h> |
| 253 | #include <stdio.h> |
| 254 | #include <stdint.h> |
| 255 | #include <string.h> |
| 256 | #include <sched.h> |
| 257 | #include <unistd.h> |
| 258 | |
| 259 | #include <lttng/ust-arch.h> |
| 260 | #include <lttng/urcu/pointer.h> |
| 261 | #include <urcu/arch.h> |
| 262 | #include <urcu/uatomic.h> |
| 263 | #include <urcu/compiler.h> |
| 264 | #include "rculfhash.h" |
| 265 | #include "rculfhash-internal.h" |
| 266 | #include <stdio.h> |
| 267 | #include <pthread.h> |
| 268 | #include <signal.h> |
| 269 | |
| 270 | /* |
| 271 | * Split-counters lazily update the global counter each 1024 |
| 272 | * addition/removal. It automatically keeps track of resize required. |
| 273 | * We use the bucket length as indicator for need to expand for small |
| 274 | * tables and machines lacking per-cpu data support. |
| 275 | */ |
| 276 | #define COUNT_COMMIT_ORDER 10 |
| 277 | |
| 278 | /* |
| 279 | * Define the minimum table size. |
| 280 | */ |
| 281 | #define MIN_TABLE_ORDER 0 |
| 282 | #define MIN_TABLE_SIZE (1UL << MIN_TABLE_ORDER) |
| 283 | |
| 284 | /* |
| 285 | * Minimum number of bucket nodes to touch per thread to parallelize grow/shrink. |
| 286 | */ |
| 287 | #define MIN_PARTITION_PER_THREAD_ORDER 12 |
| 288 | #define MIN_PARTITION_PER_THREAD (1UL << MIN_PARTITION_PER_THREAD_ORDER) |
| 289 | |
| 290 | /* |
| 291 | * The removed flag needs to be updated atomically with the pointer. |
| 292 | * It indicates that no node must attach to the node scheduled for |
| 293 | * removal, and that node garbage collection must be performed. |
| 294 | * The bucket flag does not require to be updated atomically with the |
| 295 | * pointer, but it is added as a pointer low bit flag to save space. |
| 296 | * The "removal owner" flag is used to detect which of the "del" |
| 297 | * operation that has set the "removed flag" gets to return the removed |
| 298 | * node to its caller. Note that the replace operation does not need to |
| 299 | * iteract with the "removal owner" flag, because it validates that |
| 300 | * the "removed" flag is not set before performing its cmpxchg. |
| 301 | */ |
| 302 | #define REMOVED_FLAG (1UL << 0) |
| 303 | #define BUCKET_FLAG (1UL << 1) |
| 304 | #define REMOVAL_OWNER_FLAG (1UL << 2) |
| 305 | #define FLAGS_MASK ((1UL << 3) - 1) |
| 306 | |
| 307 | /* Value of the end pointer. Should not interact with flags. */ |
| 308 | #define END_VALUE NULL |
| 309 | |
| 310 | /* |
| 311 | * ht_items_count: Split-counters counting the number of node addition |
| 312 | * and removal in the table. Only used if the LTTNG_UST_LFHT_ACCOUNTING flag |
| 313 | * is set at hash table creation. |
| 314 | * |
| 315 | * These are free-running counters, never reset to zero. They count the |
| 316 | * number of add/remove, and trigger every (1 << COUNT_COMMIT_ORDER) |
| 317 | * operations to update the global counter. We choose a power-of-2 value |
| 318 | * for the trigger to deal with 32 or 64-bit overflow of the counter. |
| 319 | */ |
| 320 | struct ht_items_count { |
| 321 | unsigned long add, del; |
| 322 | } __attribute__((aligned(CAA_CACHE_LINE_SIZE))); |
| 323 | |
| 324 | #ifdef CONFIG_LTTNG_UST_LFHT_ITER_DEBUG |
| 325 | |
| 326 | static |
| 327 | void lttng_ust_lfht_iter_debug_set_ht(struct lttng_ust_lfht *ht, struct lttng_ust_lfht_iter *iter) |
| 328 | { |
| 329 | iter->lfht = ht; |
| 330 | } |
| 331 | |
| 332 | #define lttng_ust_lfht_iter_debug_assert(...) assert(__VA_ARGS__) |
| 333 | |
| 334 | #else |
| 335 | |
| 336 | static |
| 337 | void lttng_ust_lfht_iter_debug_set_ht(struct lttng_ust_lfht *ht, struct lttng_ust_lfht_iter *iter) |
| 338 | { |
| 339 | } |
| 340 | |
| 341 | #define lttng_ust_lfht_iter_debug_assert(...) |
| 342 | |
| 343 | #endif |
| 344 | |
| 345 | /* |
| 346 | * Algorithm to reverse bits in a word by lookup table, extended to |
| 347 | * 64-bit words. |
| 348 | * Source: |
| 349 | * http://graphics.stanford.edu/~seander/bithacks.html#BitReverseTable |
| 350 | * Originally from Public Domain. |
| 351 | */ |
| 352 | |
| 353 | static const uint8_t BitReverseTable256[256] = |
| 354 | { |
| 355 | #define R2(n) (n), (n) + 2*64, (n) + 1*64, (n) + 3*64 |
| 356 | #define R4(n) R2(n), R2((n) + 2*16), R2((n) + 1*16), R2((n) + 3*16) |
| 357 | #define R6(n) R4(n), R4((n) + 2*4 ), R4((n) + 1*4 ), R4((n) + 3*4 ) |
| 358 | R6(0), R6(2), R6(1), R6(3) |
| 359 | }; |
| 360 | #undef R2 |
| 361 | #undef R4 |
| 362 | #undef R6 |
| 363 | |
| 364 | static |
| 365 | uint8_t bit_reverse_u8(uint8_t v) |
| 366 | { |
| 367 | return BitReverseTable256[v]; |
| 368 | } |
| 369 | |
| 370 | #if (CAA_BITS_PER_LONG == 32) |
| 371 | static |
| 372 | uint32_t bit_reverse_u32(uint32_t v) |
| 373 | { |
| 374 | return ((uint32_t) bit_reverse_u8(v) << 24) | |
| 375 | ((uint32_t) bit_reverse_u8(v >> 8) << 16) | |
| 376 | ((uint32_t) bit_reverse_u8(v >> 16) << 8) | |
| 377 | ((uint32_t) bit_reverse_u8(v >> 24)); |
| 378 | } |
| 379 | #else |
| 380 | static |
| 381 | uint64_t bit_reverse_u64(uint64_t v) |
| 382 | { |
| 383 | return ((uint64_t) bit_reverse_u8(v) << 56) | |
| 384 | ((uint64_t) bit_reverse_u8(v >> 8) << 48) | |
| 385 | ((uint64_t) bit_reverse_u8(v >> 16) << 40) | |
| 386 | ((uint64_t) bit_reverse_u8(v >> 24) << 32) | |
| 387 | ((uint64_t) bit_reverse_u8(v >> 32) << 24) | |
| 388 | ((uint64_t) bit_reverse_u8(v >> 40) << 16) | |
| 389 | ((uint64_t) bit_reverse_u8(v >> 48) << 8) | |
| 390 | ((uint64_t) bit_reverse_u8(v >> 56)); |
| 391 | } |
| 392 | #endif |
| 393 | |
| 394 | static |
| 395 | unsigned long bit_reverse_ulong(unsigned long v) |
| 396 | { |
| 397 | #if (CAA_BITS_PER_LONG == 32) |
| 398 | return bit_reverse_u32(v); |
| 399 | #else |
| 400 | return bit_reverse_u64(v); |
| 401 | #endif |
| 402 | } |
| 403 | |
| 404 | /* |
| 405 | * fls: returns the position of the most significant bit. |
| 406 | * Returns 0 if no bit is set, else returns the position of the most |
| 407 | * significant bit (from 1 to 32 on 32-bit, from 1 to 64 on 64-bit). |
| 408 | */ |
| 409 | #if defined(LTTNG_UST_ARCH_X86) |
| 410 | static inline |
| 411 | unsigned int fls_u32(uint32_t x) |
| 412 | { |
| 413 | int r; |
| 414 | |
| 415 | __asm__ ("bsrl %1,%0\n\t" |
| 416 | "jnz 1f\n\t" |
| 417 | "movl $-1,%0\n\t" |
| 418 | "1:\n\t" |
| 419 | : "=r" (r) : "rm" (x)); |
| 420 | return r + 1; |
| 421 | } |
| 422 | #define HAS_FLS_U32 |
| 423 | #endif |
| 424 | |
| 425 | #if defined(LTTNG_UST_ARCH_AMD64) |
| 426 | static inline |
| 427 | unsigned int fls_u64(uint64_t x) |
| 428 | { |
| 429 | long r; |
| 430 | |
| 431 | __asm__ ("bsrq %1,%0\n\t" |
| 432 | "jnz 1f\n\t" |
| 433 | "movq $-1,%0\n\t" |
| 434 | "1:\n\t" |
| 435 | : "=r" (r) : "rm" (x)); |
| 436 | return r + 1; |
| 437 | } |
| 438 | #define HAS_FLS_U64 |
| 439 | #endif |
| 440 | |
| 441 | #ifndef HAS_FLS_U64 |
| 442 | static __attribute__((unused)) |
| 443 | unsigned int fls_u64(uint64_t x) |
| 444 | { |
| 445 | unsigned int r = 64; |
| 446 | |
| 447 | if (!x) |
| 448 | return 0; |
| 449 | |
| 450 | if (!(x & 0xFFFFFFFF00000000ULL)) { |
| 451 | x <<= 32; |
| 452 | r -= 32; |
| 453 | } |
| 454 | if (!(x & 0xFFFF000000000000ULL)) { |
| 455 | x <<= 16; |
| 456 | r -= 16; |
| 457 | } |
| 458 | if (!(x & 0xFF00000000000000ULL)) { |
| 459 | x <<= 8; |
| 460 | r -= 8; |
| 461 | } |
| 462 | if (!(x & 0xF000000000000000ULL)) { |
| 463 | x <<= 4; |
| 464 | r -= 4; |
| 465 | } |
| 466 | if (!(x & 0xC000000000000000ULL)) { |
| 467 | x <<= 2; |
| 468 | r -= 2; |
| 469 | } |
| 470 | if (!(x & 0x8000000000000000ULL)) { |
| 471 | x <<= 1; |
| 472 | r -= 1; |
| 473 | } |
| 474 | return r; |
| 475 | } |
| 476 | #endif |
| 477 | |
| 478 | #ifndef HAS_FLS_U32 |
| 479 | static __attribute__((unused)) |
| 480 | unsigned int fls_u32(uint32_t x) |
| 481 | { |
| 482 | unsigned int r = 32; |
| 483 | |
| 484 | if (!x) |
| 485 | return 0; |
| 486 | if (!(x & 0xFFFF0000U)) { |
| 487 | x <<= 16; |
| 488 | r -= 16; |
| 489 | } |
| 490 | if (!(x & 0xFF000000U)) { |
| 491 | x <<= 8; |
| 492 | r -= 8; |
| 493 | } |
| 494 | if (!(x & 0xF0000000U)) { |
| 495 | x <<= 4; |
| 496 | r -= 4; |
| 497 | } |
| 498 | if (!(x & 0xC0000000U)) { |
| 499 | x <<= 2; |
| 500 | r -= 2; |
| 501 | } |
| 502 | if (!(x & 0x80000000U)) { |
| 503 | x <<= 1; |
| 504 | r -= 1; |
| 505 | } |
| 506 | return r; |
| 507 | } |
| 508 | #endif |
| 509 | |
| 510 | unsigned int lttng_ust_lfht_fls_ulong(unsigned long x) |
| 511 | { |
| 512 | #if (CAA_BITS_PER_LONG == 32) |
| 513 | return fls_u32(x); |
| 514 | #else |
| 515 | return fls_u64(x); |
| 516 | #endif |
| 517 | } |
| 518 | |
| 519 | /* |
| 520 | * Return the minimum order for which x <= (1UL << order). |
| 521 | * Return -1 if x is 0. |
| 522 | */ |
| 523 | int lttng_ust_lfht_get_count_order_u32(uint32_t x) |
| 524 | { |
| 525 | if (!x) |
| 526 | return -1; |
| 527 | |
| 528 | return fls_u32(x - 1); |
| 529 | } |
| 530 | |
| 531 | /* |
| 532 | * Return the minimum order for which x <= (1UL << order). |
| 533 | * Return -1 if x is 0. |
| 534 | */ |
| 535 | int lttng_ust_lfht_get_count_order_ulong(unsigned long x) |
| 536 | { |
| 537 | if (!x) |
| 538 | return -1; |
| 539 | |
| 540 | return lttng_ust_lfht_fls_ulong(x - 1); |
| 541 | } |
| 542 | |
| 543 | static |
| 544 | struct lttng_ust_lfht_node *clear_flag(struct lttng_ust_lfht_node *node) |
| 545 | { |
| 546 | return (struct lttng_ust_lfht_node *) (((unsigned long) node) & ~FLAGS_MASK); |
| 547 | } |
| 548 | |
| 549 | static |
| 550 | int is_removed(const struct lttng_ust_lfht_node *node) |
| 551 | { |
| 552 | return ((unsigned long) node) & REMOVED_FLAG; |
| 553 | } |
| 554 | |
| 555 | static |
| 556 | int is_bucket(struct lttng_ust_lfht_node *node) |
| 557 | { |
| 558 | return ((unsigned long) node) & BUCKET_FLAG; |
| 559 | } |
| 560 | |
| 561 | static |
| 562 | struct lttng_ust_lfht_node *flag_bucket(struct lttng_ust_lfht_node *node) |
| 563 | { |
| 564 | return (struct lttng_ust_lfht_node *) (((unsigned long) node) | BUCKET_FLAG); |
| 565 | } |
| 566 | |
| 567 | static |
| 568 | int is_removal_owner(struct lttng_ust_lfht_node *node) |
| 569 | { |
| 570 | return ((unsigned long) node) & REMOVAL_OWNER_FLAG; |
| 571 | } |
| 572 | |
| 573 | static |
| 574 | struct lttng_ust_lfht_node *flag_removal_owner(struct lttng_ust_lfht_node *node) |
| 575 | { |
| 576 | return (struct lttng_ust_lfht_node *) (((unsigned long) node) | REMOVAL_OWNER_FLAG); |
| 577 | } |
| 578 | |
| 579 | static |
| 580 | struct lttng_ust_lfht_node *flag_removed_or_removal_owner(struct lttng_ust_lfht_node *node) |
| 581 | { |
| 582 | return (struct lttng_ust_lfht_node *) (((unsigned long) node) | REMOVED_FLAG | REMOVAL_OWNER_FLAG); |
| 583 | } |
| 584 | |
| 585 | static |
| 586 | struct lttng_ust_lfht_node *get_end(void) |
| 587 | { |
| 588 | return (struct lttng_ust_lfht_node *) END_VALUE; |
| 589 | } |
| 590 | |
| 591 | static |
| 592 | int is_end(struct lttng_ust_lfht_node *node) |
| 593 | { |
| 594 | return clear_flag(node) == (struct lttng_ust_lfht_node *) END_VALUE; |
| 595 | } |
| 596 | |
| 597 | static |
| 598 | void lttng_ust_lfht_alloc_bucket_table(struct lttng_ust_lfht *ht, unsigned long order) |
| 599 | { |
| 600 | return ht->mm->alloc_bucket_table(ht, order); |
| 601 | } |
| 602 | |
| 603 | /* |
| 604 | * lttng_ust_lfht_free_bucket_table() should be called with decreasing order. |
| 605 | * When lttng_ust_lfht_free_bucket_table(0) is called, it means the whole |
| 606 | * lfht is destroyed. |
| 607 | */ |
| 608 | static |
| 609 | void lttng_ust_lfht_free_bucket_table(struct lttng_ust_lfht *ht, unsigned long order) |
| 610 | { |
| 611 | return ht->mm->free_bucket_table(ht, order); |
| 612 | } |
| 613 | |
| 614 | static inline |
| 615 | struct lttng_ust_lfht_node *bucket_at(struct lttng_ust_lfht *ht, unsigned long index) |
| 616 | { |
| 617 | return ht->bucket_at(ht, index); |
| 618 | } |
| 619 | |
| 620 | static inline |
| 621 | struct lttng_ust_lfht_node *lookup_bucket(struct lttng_ust_lfht *ht, unsigned long size, |
| 622 | unsigned long hash) |
| 623 | { |
| 624 | assert(size > 0); |
| 625 | return bucket_at(ht, hash & (size - 1)); |
| 626 | } |
| 627 | |
| 628 | /* |
| 629 | * Remove all logically deleted nodes from a bucket up to a certain node key. |
| 630 | */ |
| 631 | static |
| 632 | void _lttng_ust_lfht_gc_bucket(struct lttng_ust_lfht_node *bucket, struct lttng_ust_lfht_node *node) |
| 633 | { |
| 634 | struct lttng_ust_lfht_node *iter_prev, *iter, *next, *new_next; |
| 635 | |
| 636 | assert(!is_bucket(bucket)); |
| 637 | assert(!is_removed(bucket)); |
| 638 | assert(!is_removal_owner(bucket)); |
| 639 | assert(!is_bucket(node)); |
| 640 | assert(!is_removed(node)); |
| 641 | assert(!is_removal_owner(node)); |
| 642 | for (;;) { |
| 643 | iter_prev = bucket; |
| 644 | /* We can always skip the bucket node initially */ |
| 645 | iter = lttng_ust_rcu_dereference(iter_prev->next); |
| 646 | assert(!is_removed(iter)); |
| 647 | assert(!is_removal_owner(iter)); |
| 648 | assert(iter_prev->reverse_hash <= node->reverse_hash); |
| 649 | /* |
| 650 | * We should never be called with bucket (start of chain) |
| 651 | * and logically removed node (end of path compression |
| 652 | * marker) being the actual same node. This would be a |
| 653 | * bug in the algorithm implementation. |
| 654 | */ |
| 655 | assert(bucket != node); |
| 656 | for (;;) { |
| 657 | if (caa_unlikely(is_end(iter))) |
| 658 | return; |
| 659 | if (caa_likely(clear_flag(iter)->reverse_hash > node->reverse_hash)) |
| 660 | return; |
| 661 | next = lttng_ust_rcu_dereference(clear_flag(iter)->next); |
| 662 | if (caa_likely(is_removed(next))) |
| 663 | break; |
| 664 | iter_prev = clear_flag(iter); |
| 665 | iter = next; |
| 666 | } |
| 667 | assert(!is_removed(iter)); |
| 668 | assert(!is_removal_owner(iter)); |
| 669 | if (is_bucket(iter)) |
| 670 | new_next = flag_bucket(clear_flag(next)); |
| 671 | else |
| 672 | new_next = clear_flag(next); |
| 673 | (void) uatomic_cmpxchg(&iter_prev->next, iter, new_next); |
| 674 | } |
| 675 | } |
| 676 | |
| 677 | static |
| 678 | int _lttng_ust_lfht_replace(struct lttng_ust_lfht *ht, unsigned long size, |
| 679 | struct lttng_ust_lfht_node *old_node, |
| 680 | struct lttng_ust_lfht_node *old_next, |
| 681 | struct lttng_ust_lfht_node *new_node) |
| 682 | { |
| 683 | struct lttng_ust_lfht_node *bucket, *ret_next; |
| 684 | |
| 685 | if (!old_node) /* Return -ENOENT if asked to replace NULL node */ |
| 686 | return -ENOENT; |
| 687 | |
| 688 | assert(!is_removed(old_node)); |
| 689 | assert(!is_removal_owner(old_node)); |
| 690 | assert(!is_bucket(old_node)); |
| 691 | assert(!is_removed(new_node)); |
| 692 | assert(!is_removal_owner(new_node)); |
| 693 | assert(!is_bucket(new_node)); |
| 694 | assert(new_node != old_node); |
| 695 | for (;;) { |
| 696 | /* Insert after node to be replaced */ |
| 697 | if (is_removed(old_next)) { |
| 698 | /* |
| 699 | * Too late, the old node has been removed under us |
| 700 | * between lookup and replace. Fail. |
| 701 | */ |
| 702 | return -ENOENT; |
| 703 | } |
| 704 | assert(old_next == clear_flag(old_next)); |
| 705 | assert(new_node != old_next); |
| 706 | /* |
| 707 | * REMOVAL_OWNER flag is _NEVER_ set before the REMOVED |
| 708 | * flag. It is either set atomically at the same time |
| 709 | * (replace) or after (del). |
| 710 | */ |
| 711 | assert(!is_removal_owner(old_next)); |
| 712 | new_node->next = old_next; |
| 713 | /* |
| 714 | * Here is the whole trick for lock-free replace: we add |
| 715 | * the replacement node _after_ the node we want to |
| 716 | * replace by atomically setting its next pointer at the |
| 717 | * same time we set its removal flag. Given that |
| 718 | * the lookups/get next use an iterator aware of the |
| 719 | * next pointer, they will either skip the old node due |
| 720 | * to the removal flag and see the new node, or use |
| 721 | * the old node, but will not see the new one. |
| 722 | * This is a replacement of a node with another node |
| 723 | * that has the same value: we are therefore not |
| 724 | * removing a value from the hash table. We set both the |
| 725 | * REMOVED and REMOVAL_OWNER flags atomically so we own |
| 726 | * the node after successful cmpxchg. |
| 727 | */ |
| 728 | ret_next = uatomic_cmpxchg(&old_node->next, |
| 729 | old_next, flag_removed_or_removal_owner(new_node)); |
| 730 | if (ret_next == old_next) |
| 731 | break; /* We performed the replacement. */ |
| 732 | old_next = ret_next; |
| 733 | } |
| 734 | |
| 735 | /* |
| 736 | * Ensure that the old node is not visible to readers anymore: |
| 737 | * lookup for the node, and remove it (along with any other |
| 738 | * logically removed node) if found. |
| 739 | */ |
| 740 | bucket = lookup_bucket(ht, size, bit_reverse_ulong(old_node->reverse_hash)); |
| 741 | _lttng_ust_lfht_gc_bucket(bucket, new_node); |
| 742 | |
| 743 | assert(is_removed(CMM_LOAD_SHARED(old_node->next))); |
| 744 | return 0; |
| 745 | } |
| 746 | |
| 747 | /* |
| 748 | * A non-NULL unique_ret pointer uses the "add unique" (or uniquify) add |
| 749 | * mode. A NULL unique_ret allows creation of duplicate keys. |
| 750 | */ |
| 751 | static |
| 752 | void _lttng_ust_lfht_add(struct lttng_ust_lfht *ht, |
| 753 | unsigned long hash, |
| 754 | lttng_ust_lfht_match_fct match, |
| 755 | const void *key, |
| 756 | unsigned long size, |
| 757 | struct lttng_ust_lfht_node *node, |
| 758 | struct lttng_ust_lfht_iter *unique_ret, |
| 759 | int bucket_flag) |
| 760 | { |
| 761 | struct lttng_ust_lfht_node *iter_prev, *iter, *next, *new_node, *new_next, |
| 762 | *return_node; |
| 763 | struct lttng_ust_lfht_node *bucket; |
| 764 | |
| 765 | assert(!is_bucket(node)); |
| 766 | assert(!is_removed(node)); |
| 767 | assert(!is_removal_owner(node)); |
| 768 | bucket = lookup_bucket(ht, size, hash); |
| 769 | for (;;) { |
| 770 | /* |
| 771 | * iter_prev points to the non-removed node prior to the |
| 772 | * insert location. |
| 773 | */ |
| 774 | iter_prev = bucket; |
| 775 | /* We can always skip the bucket node initially */ |
| 776 | iter = lttng_ust_rcu_dereference(iter_prev->next); |
| 777 | assert(iter_prev->reverse_hash <= node->reverse_hash); |
| 778 | for (;;) { |
| 779 | if (caa_unlikely(is_end(iter))) |
| 780 | goto insert; |
| 781 | if (caa_likely(clear_flag(iter)->reverse_hash > node->reverse_hash)) |
| 782 | goto insert; |
| 783 | |
| 784 | /* bucket node is the first node of the identical-hash-value chain */ |
| 785 | if (bucket_flag && clear_flag(iter)->reverse_hash == node->reverse_hash) |
| 786 | goto insert; |
| 787 | |
| 788 | next = lttng_ust_rcu_dereference(clear_flag(iter)->next); |
| 789 | if (caa_unlikely(is_removed(next))) |
| 790 | goto gc_node; |
| 791 | |
| 792 | /* uniquely add */ |
| 793 | if (unique_ret |
| 794 | && !is_bucket(next) |
| 795 | && clear_flag(iter)->reverse_hash == node->reverse_hash) { |
| 796 | struct lttng_ust_lfht_iter d_iter = { |
| 797 | .node = node, |
| 798 | .next = iter, |
| 799 | #ifdef CONFIG_LTTNG_UST_LFHT_ITER_DEBUG |
| 800 | .lfht = ht, |
| 801 | #endif |
| 802 | }; |
| 803 | |
| 804 | /* |
| 805 | * uniquely adding inserts the node as the first |
| 806 | * node of the identical-hash-value node chain. |
| 807 | * |
| 808 | * This semantic ensures no duplicated keys |
| 809 | * should ever be observable in the table |
| 810 | * (including traversing the table node by |
| 811 | * node by forward iterations) |
| 812 | */ |
| 813 | lttng_ust_lfht_next_duplicate(ht, match, key, &d_iter); |
| 814 | if (!d_iter.node) |
| 815 | goto insert; |
| 816 | |
| 817 | *unique_ret = d_iter; |
| 818 | return; |
| 819 | } |
| 820 | |
| 821 | iter_prev = clear_flag(iter); |
| 822 | iter = next; |
| 823 | } |
| 824 | |
| 825 | insert: |
| 826 | assert(node != clear_flag(iter)); |
| 827 | assert(!is_removed(iter_prev)); |
| 828 | assert(!is_removal_owner(iter_prev)); |
| 829 | assert(!is_removed(iter)); |
| 830 | assert(!is_removal_owner(iter)); |
| 831 | assert(iter_prev != node); |
| 832 | if (!bucket_flag) |
| 833 | node->next = clear_flag(iter); |
| 834 | else |
| 835 | node->next = flag_bucket(clear_flag(iter)); |
| 836 | if (is_bucket(iter)) |
| 837 | new_node = flag_bucket(node); |
| 838 | else |
| 839 | new_node = node; |
| 840 | if (uatomic_cmpxchg(&iter_prev->next, iter, |
| 841 | new_node) != iter) { |
| 842 | continue; /* retry */ |
| 843 | } else { |
| 844 | return_node = node; |
| 845 | goto end; |
| 846 | } |
| 847 | |
| 848 | gc_node: |
| 849 | assert(!is_removed(iter)); |
| 850 | assert(!is_removal_owner(iter)); |
| 851 | if (is_bucket(iter)) |
| 852 | new_next = flag_bucket(clear_flag(next)); |
| 853 | else |
| 854 | new_next = clear_flag(next); |
| 855 | (void) uatomic_cmpxchg(&iter_prev->next, iter, new_next); |
| 856 | /* retry */ |
| 857 | } |
| 858 | end: |
| 859 | if (unique_ret) { |
| 860 | unique_ret->node = return_node; |
| 861 | /* unique_ret->next left unset, never used. */ |
| 862 | } |
| 863 | } |
| 864 | |
| 865 | static |
| 866 | int _lttng_ust_lfht_del(struct lttng_ust_lfht *ht, unsigned long size, |
| 867 | struct lttng_ust_lfht_node *node) |
| 868 | { |
| 869 | struct lttng_ust_lfht_node *bucket, *next; |
| 870 | |
| 871 | if (!node) /* Return -ENOENT if asked to delete NULL node */ |
| 872 | return -ENOENT; |
| 873 | |
| 874 | /* logically delete the node */ |
| 875 | assert(!is_bucket(node)); |
| 876 | assert(!is_removed(node)); |
| 877 | assert(!is_removal_owner(node)); |
| 878 | |
| 879 | /* |
| 880 | * We are first checking if the node had previously been |
| 881 | * logically removed (this check is not atomic with setting the |
| 882 | * logical removal flag). Return -ENOENT if the node had |
| 883 | * previously been removed. |
| 884 | */ |
| 885 | next = CMM_LOAD_SHARED(node->next); /* next is not dereferenced */ |
| 886 | if (caa_unlikely(is_removed(next))) |
| 887 | return -ENOENT; |
| 888 | assert(!is_bucket(next)); |
| 889 | /* |
| 890 | * The del operation semantic guarantees a full memory barrier |
| 891 | * before the uatomic_or atomic commit of the deletion flag. |
| 892 | */ |
| 893 | cmm_smp_mb__before_uatomic_or(); |
| 894 | /* |
| 895 | * We set the REMOVED_FLAG unconditionally. Note that there may |
| 896 | * be more than one concurrent thread setting this flag. |
| 897 | * Knowing which wins the race will be known after the garbage |
| 898 | * collection phase, stay tuned! |
| 899 | */ |
| 900 | uatomic_or(&node->next, REMOVED_FLAG); |
| 901 | /* We performed the (logical) deletion. */ |
| 902 | |
| 903 | /* |
| 904 | * Ensure that the node is not visible to readers anymore: lookup for |
| 905 | * the node, and remove it (along with any other logically removed node) |
| 906 | * if found. |
| 907 | */ |
| 908 | bucket = lookup_bucket(ht, size, bit_reverse_ulong(node->reverse_hash)); |
| 909 | _lttng_ust_lfht_gc_bucket(bucket, node); |
| 910 | |
| 911 | assert(is_removed(CMM_LOAD_SHARED(node->next))); |
| 912 | /* |
| 913 | * Last phase: atomically exchange node->next with a version |
| 914 | * having "REMOVAL_OWNER_FLAG" set. If the returned node->next |
| 915 | * pointer did _not_ have "REMOVAL_OWNER_FLAG" set, we now own |
| 916 | * the node and win the removal race. |
| 917 | * It is interesting to note that all "add" paths are forbidden |
| 918 | * to change the next pointer starting from the point where the |
| 919 | * REMOVED_FLAG is set, so here using a read, followed by a |
| 920 | * xchg() suffice to guarantee that the xchg() will ever only |
| 921 | * set the "REMOVAL_OWNER_FLAG" (or change nothing if the flag |
| 922 | * was already set). |
| 923 | */ |
| 924 | if (!is_removal_owner(uatomic_xchg(&node->next, |
| 925 | flag_removal_owner(node->next)))) |
| 926 | return 0; |
| 927 | else |
| 928 | return -ENOENT; |
| 929 | } |
| 930 | |
| 931 | /* |
| 932 | * Never called with size < 1. |
| 933 | */ |
| 934 | static |
| 935 | void lttng_ust_lfht_create_bucket(struct lttng_ust_lfht *ht, unsigned long size) |
| 936 | { |
| 937 | struct lttng_ust_lfht_node *prev, *node; |
| 938 | unsigned long order, len, i; |
| 939 | int bucket_order; |
| 940 | |
| 941 | lttng_ust_lfht_alloc_bucket_table(ht, 0); |
| 942 | |
| 943 | dbg_printf("create bucket: order 0 index 0 hash 0\n"); |
| 944 | node = bucket_at(ht, 0); |
| 945 | node->next = flag_bucket(get_end()); |
| 946 | node->reverse_hash = 0; |
| 947 | |
| 948 | bucket_order = lttng_ust_lfht_get_count_order_ulong(size); |
| 949 | assert(bucket_order >= 0); |
| 950 | |
| 951 | for (order = 1; order < (unsigned long) bucket_order + 1; order++) { |
| 952 | len = 1UL << (order - 1); |
| 953 | lttng_ust_lfht_alloc_bucket_table(ht, order); |
| 954 | |
| 955 | for (i = 0; i < len; i++) { |
| 956 | /* |
| 957 | * Now, we are trying to init the node with the |
| 958 | * hash=(len+i) (which is also a bucket with the |
| 959 | * index=(len+i)) and insert it into the hash table, |
| 960 | * so this node has to be inserted after the bucket |
| 961 | * with the index=(len+i)&(len-1)=i. And because there |
| 962 | * is no other non-bucket node nor bucket node with |
| 963 | * larger index/hash inserted, so the bucket node |
| 964 | * being inserted should be inserted directly linked |
| 965 | * after the bucket node with index=i. |
| 966 | */ |
| 967 | prev = bucket_at(ht, i); |
| 968 | node = bucket_at(ht, len + i); |
| 969 | |
| 970 | dbg_printf("create bucket: order %lu index %lu hash %lu\n", |
| 971 | order, len + i, len + i); |
| 972 | node->reverse_hash = bit_reverse_ulong(len + i); |
| 973 | |
| 974 | /* insert after prev */ |
| 975 | assert(is_bucket(prev->next)); |
| 976 | node->next = prev->next; |
| 977 | prev->next = flag_bucket(node); |
| 978 | } |
| 979 | } |
| 980 | } |
| 981 | |
| 982 | #if (CAA_BITS_PER_LONG > 32) |
| 983 | /* |
| 984 | * For 64-bit architectures, with max number of buckets small enough not to |
| 985 | * use the entire 64-bit memory mapping space (and allowing a fair number of |
| 986 | * hash table instances), use the mmap allocator, which is faster. Otherwise, |
| 987 | * fallback to the order allocator. |
| 988 | */ |
| 989 | static |
| 990 | const struct lttng_ust_lfht_mm_type *get_mm_type(unsigned long max_nr_buckets) |
| 991 | { |
| 992 | if (max_nr_buckets && max_nr_buckets <= (1ULL << 32)) |
| 993 | return <tng_ust_lfht_mm_mmap; |
| 994 | else |
| 995 | return <tng_ust_lfht_mm_order; |
| 996 | } |
| 997 | #else |
| 998 | /* |
| 999 | * For 32-bit architectures, use the order allocator. |
| 1000 | */ |
| 1001 | static |
| 1002 | const struct lttng_ust_lfht_mm_type *get_mm_type(unsigned long max_nr_buckets) |
| 1003 | { |
| 1004 | return <tng_ust_lfht_mm_order; |
| 1005 | } |
| 1006 | #endif |
| 1007 | |
| 1008 | struct lttng_ust_lfht *lttng_ust_lfht_new(unsigned long init_size, |
| 1009 | unsigned long min_nr_alloc_buckets, |
| 1010 | unsigned long max_nr_buckets, |
| 1011 | int flags, |
| 1012 | const struct lttng_ust_lfht_mm_type *mm) |
| 1013 | { |
| 1014 | struct lttng_ust_lfht *ht; |
| 1015 | unsigned long order; |
| 1016 | |
| 1017 | /* min_nr_alloc_buckets must be power of two */ |
| 1018 | if (!min_nr_alloc_buckets || (min_nr_alloc_buckets & (min_nr_alloc_buckets - 1))) |
| 1019 | return NULL; |
| 1020 | |
| 1021 | /* init_size must be power of two */ |
| 1022 | if (!init_size || (init_size & (init_size - 1))) |
| 1023 | return NULL; |
| 1024 | |
| 1025 | /* |
| 1026 | * Memory management plugin default. |
| 1027 | */ |
| 1028 | if (!mm) |
| 1029 | mm = get_mm_type(max_nr_buckets); |
| 1030 | |
| 1031 | /* max_nr_buckets == 0 for order based mm means infinite */ |
| 1032 | if (mm == <tng_ust_lfht_mm_order && !max_nr_buckets) |
| 1033 | max_nr_buckets = 1UL << (MAX_TABLE_ORDER - 1); |
| 1034 | |
| 1035 | /* max_nr_buckets must be power of two */ |
| 1036 | if (!max_nr_buckets || (max_nr_buckets & (max_nr_buckets - 1))) |
| 1037 | return NULL; |
| 1038 | |
| 1039 | if (flags & LTTNG_UST_LFHT_AUTO_RESIZE) |
| 1040 | return NULL; |
| 1041 | |
| 1042 | min_nr_alloc_buckets = max(min_nr_alloc_buckets, MIN_TABLE_SIZE); |
| 1043 | init_size = max(init_size, MIN_TABLE_SIZE); |
| 1044 | max_nr_buckets = max(max_nr_buckets, min_nr_alloc_buckets); |
| 1045 | init_size = min(init_size, max_nr_buckets); |
| 1046 | |
| 1047 | ht = mm->alloc_lttng_ust_lfht(min_nr_alloc_buckets, max_nr_buckets); |
| 1048 | assert(ht); |
| 1049 | assert(ht->mm == mm); |
| 1050 | assert(ht->bucket_at == mm->bucket_at); |
| 1051 | |
| 1052 | ht->flags = flags; |
| 1053 | /* this mutex should not nest in read-side C.S. */ |
| 1054 | pthread_mutex_init(&ht->resize_mutex, NULL); |
| 1055 | order = lttng_ust_lfht_get_count_order_ulong(init_size); |
| 1056 | ht->resize_target = 1UL << order; |
| 1057 | lttng_ust_lfht_create_bucket(ht, 1UL << order); |
| 1058 | ht->size = 1UL << order; |
| 1059 | return ht; |
| 1060 | } |
| 1061 | |
| 1062 | void lttng_ust_lfht_lookup(struct lttng_ust_lfht *ht, unsigned long hash, |
| 1063 | lttng_ust_lfht_match_fct match, const void *key, |
| 1064 | struct lttng_ust_lfht_iter *iter) |
| 1065 | { |
| 1066 | struct lttng_ust_lfht_node *node, *next, *bucket; |
| 1067 | unsigned long reverse_hash, size; |
| 1068 | |
| 1069 | lttng_ust_lfht_iter_debug_set_ht(ht, iter); |
| 1070 | |
| 1071 | reverse_hash = bit_reverse_ulong(hash); |
| 1072 | |
| 1073 | size = lttng_ust_rcu_dereference(ht->size); |
| 1074 | bucket = lookup_bucket(ht, size, hash); |
| 1075 | /* We can always skip the bucket node initially */ |
| 1076 | node = lttng_ust_rcu_dereference(bucket->next); |
| 1077 | node = clear_flag(node); |
| 1078 | for (;;) { |
| 1079 | if (caa_unlikely(is_end(node))) { |
| 1080 | node = next = NULL; |
| 1081 | break; |
| 1082 | } |
| 1083 | if (caa_unlikely(node->reverse_hash > reverse_hash)) { |
| 1084 | node = next = NULL; |
| 1085 | break; |
| 1086 | } |
| 1087 | next = lttng_ust_rcu_dereference(node->next); |
| 1088 | assert(node == clear_flag(node)); |
| 1089 | if (caa_likely(!is_removed(next)) |
| 1090 | && !is_bucket(next) |
| 1091 | && node->reverse_hash == reverse_hash |
| 1092 | && caa_likely(match(node, key))) { |
| 1093 | break; |
| 1094 | } |
| 1095 | node = clear_flag(next); |
| 1096 | } |
| 1097 | assert(!node || !is_bucket(CMM_LOAD_SHARED(node->next))); |
| 1098 | iter->node = node; |
| 1099 | iter->next = next; |
| 1100 | } |
| 1101 | |
| 1102 | void lttng_ust_lfht_next_duplicate(struct lttng_ust_lfht *ht, lttng_ust_lfht_match_fct match, |
| 1103 | const void *key, struct lttng_ust_lfht_iter *iter) |
| 1104 | { |
| 1105 | struct lttng_ust_lfht_node *node, *next; |
| 1106 | unsigned long reverse_hash; |
| 1107 | |
| 1108 | lttng_ust_lfht_iter_debug_assert(ht == iter->lfht); |
| 1109 | node = iter->node; |
| 1110 | reverse_hash = node->reverse_hash; |
| 1111 | next = iter->next; |
| 1112 | node = clear_flag(next); |
| 1113 | |
| 1114 | for (;;) { |
| 1115 | if (caa_unlikely(is_end(node))) { |
| 1116 | node = next = NULL; |
| 1117 | break; |
| 1118 | } |
| 1119 | if (caa_unlikely(node->reverse_hash > reverse_hash)) { |
| 1120 | node = next = NULL; |
| 1121 | break; |
| 1122 | } |
| 1123 | next = lttng_ust_rcu_dereference(node->next); |
| 1124 | if (caa_likely(!is_removed(next)) |
| 1125 | && !is_bucket(next) |
| 1126 | && caa_likely(match(node, key))) { |
| 1127 | break; |
| 1128 | } |
| 1129 | node = clear_flag(next); |
| 1130 | } |
| 1131 | assert(!node || !is_bucket(CMM_LOAD_SHARED(node->next))); |
| 1132 | iter->node = node; |
| 1133 | iter->next = next; |
| 1134 | } |
| 1135 | |
| 1136 | void lttng_ust_lfht_next(struct lttng_ust_lfht *ht, struct lttng_ust_lfht_iter *iter) |
| 1137 | { |
| 1138 | struct lttng_ust_lfht_node *node, *next; |
| 1139 | |
| 1140 | lttng_ust_lfht_iter_debug_assert(ht == iter->lfht); |
| 1141 | node = clear_flag(iter->next); |
| 1142 | for (;;) { |
| 1143 | if (caa_unlikely(is_end(node))) { |
| 1144 | node = next = NULL; |
| 1145 | break; |
| 1146 | } |
| 1147 | next = lttng_ust_rcu_dereference(node->next); |
| 1148 | if (caa_likely(!is_removed(next)) |
| 1149 | && !is_bucket(next)) { |
| 1150 | break; |
| 1151 | } |
| 1152 | node = clear_flag(next); |
| 1153 | } |
| 1154 | assert(!node || !is_bucket(CMM_LOAD_SHARED(node->next))); |
| 1155 | iter->node = node; |
| 1156 | iter->next = next; |
| 1157 | } |
| 1158 | |
| 1159 | void lttng_ust_lfht_first(struct lttng_ust_lfht *ht, struct lttng_ust_lfht_iter *iter) |
| 1160 | { |
| 1161 | lttng_ust_lfht_iter_debug_set_ht(ht, iter); |
| 1162 | /* |
| 1163 | * Get next after first bucket node. The first bucket node is the |
| 1164 | * first node of the linked list. |
| 1165 | */ |
| 1166 | iter->next = bucket_at(ht, 0)->next; |
| 1167 | lttng_ust_lfht_next(ht, iter); |
| 1168 | } |
| 1169 | |
| 1170 | void lttng_ust_lfht_add(struct lttng_ust_lfht *ht, unsigned long hash, |
| 1171 | struct lttng_ust_lfht_node *node) |
| 1172 | { |
| 1173 | unsigned long size; |
| 1174 | |
| 1175 | node->reverse_hash = bit_reverse_ulong(hash); |
| 1176 | size = lttng_ust_rcu_dereference(ht->size); |
| 1177 | _lttng_ust_lfht_add(ht, hash, NULL, NULL, size, node, NULL, 0); |
| 1178 | } |
| 1179 | |
| 1180 | struct lttng_ust_lfht_node *lttng_ust_lfht_add_unique(struct lttng_ust_lfht *ht, |
| 1181 | unsigned long hash, |
| 1182 | lttng_ust_lfht_match_fct match, |
| 1183 | const void *key, |
| 1184 | struct lttng_ust_lfht_node *node) |
| 1185 | { |
| 1186 | unsigned long size; |
| 1187 | struct lttng_ust_lfht_iter iter; |
| 1188 | |
| 1189 | node->reverse_hash = bit_reverse_ulong(hash); |
| 1190 | size = lttng_ust_rcu_dereference(ht->size); |
| 1191 | _lttng_ust_lfht_add(ht, hash, match, key, size, node, &iter, 0); |
| 1192 | return iter.node; |
| 1193 | } |
| 1194 | |
| 1195 | struct lttng_ust_lfht_node *lttng_ust_lfht_add_replace(struct lttng_ust_lfht *ht, |
| 1196 | unsigned long hash, |
| 1197 | lttng_ust_lfht_match_fct match, |
| 1198 | const void *key, |
| 1199 | struct lttng_ust_lfht_node *node) |
| 1200 | { |
| 1201 | unsigned long size; |
| 1202 | struct lttng_ust_lfht_iter iter; |
| 1203 | |
| 1204 | node->reverse_hash = bit_reverse_ulong(hash); |
| 1205 | size = lttng_ust_rcu_dereference(ht->size); |
| 1206 | for (;;) { |
| 1207 | _lttng_ust_lfht_add(ht, hash, match, key, size, node, &iter, 0); |
| 1208 | if (iter.node == node) { |
| 1209 | return NULL; |
| 1210 | } |
| 1211 | |
| 1212 | if (!_lttng_ust_lfht_replace(ht, size, iter.node, iter.next, node)) |
| 1213 | return iter.node; |
| 1214 | } |
| 1215 | } |
| 1216 | |
| 1217 | int lttng_ust_lfht_replace(struct lttng_ust_lfht *ht, |
| 1218 | struct lttng_ust_lfht_iter *old_iter, |
| 1219 | unsigned long hash, |
| 1220 | lttng_ust_lfht_match_fct match, |
| 1221 | const void *key, |
| 1222 | struct lttng_ust_lfht_node *new_node) |
| 1223 | { |
| 1224 | unsigned long size; |
| 1225 | |
| 1226 | new_node->reverse_hash = bit_reverse_ulong(hash); |
| 1227 | if (!old_iter->node) |
| 1228 | return -ENOENT; |
| 1229 | if (caa_unlikely(old_iter->node->reverse_hash != new_node->reverse_hash)) |
| 1230 | return -EINVAL; |
| 1231 | if (caa_unlikely(!match(old_iter->node, key))) |
| 1232 | return -EINVAL; |
| 1233 | size = lttng_ust_rcu_dereference(ht->size); |
| 1234 | return _lttng_ust_lfht_replace(ht, size, old_iter->node, old_iter->next, |
| 1235 | new_node); |
| 1236 | } |
| 1237 | |
| 1238 | int lttng_ust_lfht_del(struct lttng_ust_lfht *ht, struct lttng_ust_lfht_node *node) |
| 1239 | { |
| 1240 | unsigned long size; |
| 1241 | |
| 1242 | size = lttng_ust_rcu_dereference(ht->size); |
| 1243 | return _lttng_ust_lfht_del(ht, size, node); |
| 1244 | } |
| 1245 | |
| 1246 | int lttng_ust_lfht_is_node_deleted(const struct lttng_ust_lfht_node *node) |
| 1247 | { |
| 1248 | return is_removed(CMM_LOAD_SHARED(node->next)); |
| 1249 | } |
| 1250 | |
| 1251 | static |
| 1252 | int lttng_ust_lfht_delete_bucket(struct lttng_ust_lfht *ht) |
| 1253 | { |
| 1254 | struct lttng_ust_lfht_node *node; |
| 1255 | unsigned long order, i, size; |
| 1256 | |
| 1257 | /* Check that the table is empty */ |
| 1258 | node = bucket_at(ht, 0); |
| 1259 | do { |
| 1260 | node = clear_flag(node)->next; |
| 1261 | if (!is_bucket(node)) |
| 1262 | return -EPERM; |
| 1263 | assert(!is_removed(node)); |
| 1264 | assert(!is_removal_owner(node)); |
| 1265 | } while (!is_end(node)); |
| 1266 | /* |
| 1267 | * size accessed without lttng_ust_rcu_dereference because hash table is |
| 1268 | * being destroyed. |
| 1269 | */ |
| 1270 | size = ht->size; |
| 1271 | /* Internal sanity check: all nodes left should be buckets */ |
| 1272 | for (i = 0; i < size; i++) { |
| 1273 | node = bucket_at(ht, i); |
| 1274 | dbg_printf("delete bucket: index %lu expected hash %lu hash %lu\n", |
| 1275 | i, i, bit_reverse_ulong(node->reverse_hash)); |
| 1276 | assert(is_bucket(node->next)); |
| 1277 | } |
| 1278 | |
| 1279 | for (order = lttng_ust_lfht_get_count_order_ulong(size); (long)order >= 0; order--) |
| 1280 | lttng_ust_lfht_free_bucket_table(ht, order); |
| 1281 | |
| 1282 | return 0; |
| 1283 | } |
| 1284 | |
| 1285 | /* |
| 1286 | * Should only be called when no more concurrent readers nor writers can |
| 1287 | * possibly access the table. |
| 1288 | */ |
| 1289 | int lttng_ust_lfht_destroy(struct lttng_ust_lfht *ht) |
| 1290 | { |
| 1291 | int ret; |
| 1292 | |
| 1293 | ret = lttng_ust_lfht_delete_bucket(ht); |
| 1294 | if (ret) |
| 1295 | return ret; |
| 1296 | ret = pthread_mutex_destroy(&ht->resize_mutex); |
| 1297 | if (ret) |
| 1298 | ret = -EBUSY; |
| 1299 | poison_free(ht); |
| 1300 | return ret; |
| 1301 | } |