1 /* SPDX-License-Identifier: GPL-2.0 */ 2 #ifndef _BCACHEFS_BTREE_UPDATE_INTERIOR_H 3 #define _BCACHEFS_BTREE_UPDATE_INTERIOR_H 4 5 #include "btree_cache.h" 6 #include "btree_locking.h" 7 #include "btree_update.h" 8 9 void __bch2_btree_calc_format(struct bkey_format_state *, struct btree *); 10 bool bch2_btree_node_format_fits(struct bch_fs *c, struct btree *, 11 struct bkey_format *); 12 13 #define BTREE_UPDATE_NODES_MAX ((BTREE_MAX_DEPTH - 2) * 2 + GC_MERGE_NODES) 14 15 #define BTREE_UPDATE_JOURNAL_RES (BTREE_UPDATE_NODES_MAX * (BKEY_BTREE_PTR_U64s_MAX + 1)) 16 17 /* 18 * Tracks an in progress split/rewrite of a btree node and the update to the 19 * parent node: 20 * 21 * When we split/rewrite a node, we do all the updates in memory without 22 * waiting for any writes to complete - we allocate the new node(s) and update 23 * the parent node, possibly recursively up to the root. 24 * 25 * The end result is that we have one or more new nodes being written - 26 * possibly several, if there were multiple splits - and then a write (updating 27 * an interior node) which will make all these new nodes visible. 28 * 29 * Additionally, as we split/rewrite nodes we free the old nodes - but the old 30 * nodes can't be freed (their space on disk can't be reclaimed) until the 31 * update to the interior node that makes the new node visible completes - 32 * until then, the old nodes are still reachable on disk. 33 * 34 */ 35 struct btree_update { 36 struct closure cl; 37 struct bch_fs *c; 38 u64 start_time; 39 40 struct list_head list; 41 struct list_head unwritten_list; 42 43 /* What kind of update are we doing? */ 44 enum { 45 BTREE_INTERIOR_NO_UPDATE, 46 BTREE_INTERIOR_UPDATING_NODE, 47 BTREE_INTERIOR_UPDATING_ROOT, 48 BTREE_INTERIOR_UPDATING_AS, 49 } mode; 50 51 unsigned nodes_written:1; 52 unsigned took_gc_lock:1; 53 54 enum btree_id btree_id; 55 unsigned update_level; 56 57 struct disk_reservation disk_res; 58 59 /* 60 * BTREE_INTERIOR_UPDATING_NODE: 61 * The update that made the new nodes visible was a regular update to an 62 * existing interior node - @b. We can't write out the update to @b 63 * until the new nodes we created are finished writing, so we block @b 64 * from writing by putting this btree_interior update on the 65 * @b->write_blocked list with @write_blocked_list: 66 */ 67 struct btree *b; 68 struct list_head write_blocked_list; 69 70 /* 71 * We may be freeing nodes that were dirty, and thus had journal entries 72 * pinned: we need to transfer the oldest of those pins to the 73 * btree_update operation, and release it when the new node(s) 74 * are all persistent and reachable: 75 */ 76 struct journal_entry_pin journal; 77 78 /* Preallocated nodes we reserve when we start the update: */ 79 struct prealloc_nodes { 80 struct btree *b[BTREE_UPDATE_NODES_MAX]; 81 unsigned nr; 82 } prealloc_nodes[2]; 83 84 /* Nodes being freed: */ 85 struct keylist old_keys; 86 u64 _old_keys[BTREE_UPDATE_NODES_MAX * 87 BKEY_BTREE_PTR_U64s_MAX]; 88 89 /* Nodes being added: */ 90 struct keylist new_keys; 91 u64 _new_keys[BTREE_UPDATE_NODES_MAX * 92 BKEY_BTREE_PTR_U64s_MAX]; 93 94 /* New nodes, that will be made reachable by this update: */ 95 struct btree *new_nodes[BTREE_UPDATE_NODES_MAX]; 96 unsigned nr_new_nodes; 97 98 struct btree *old_nodes[BTREE_UPDATE_NODES_MAX]; 99 __le64 old_nodes_seq[BTREE_UPDATE_NODES_MAX]; 100 unsigned nr_old_nodes; 101 102 open_bucket_idx_t open_buckets[BTREE_UPDATE_NODES_MAX * 103 BCH_REPLICAS_MAX]; 104 open_bucket_idx_t nr_open_buckets; 105 106 unsigned journal_u64s; 107 u64 journal_entries[BTREE_UPDATE_JOURNAL_RES]; 108 109 /* Only here to reduce stack usage on recursive splits: */ 110 struct keylist parent_keys; 111 /* 112 * Enough room for btree_split's keys without realloc - btree node 113 * pointers never have crc/compression info, so we only need to acount 114 * for the pointers for three keys 115 */ 116 u64 inline_keys[BKEY_BTREE_PTR_U64s_MAX * 3]; 117 }; 118 119 struct btree *__bch2_btree_node_alloc_replacement(struct btree_update *, 120 struct btree_trans *, 121 struct btree *, 122 struct bkey_format); 123 124 int bch2_btree_split_leaf(struct btree_trans *, struct btree_path *, unsigned); 125 126 int __bch2_foreground_maybe_merge(struct btree_trans *, struct btree_path *, 127 unsigned, unsigned, enum btree_node_sibling); 128 129 static inline int bch2_foreground_maybe_merge_sibling(struct btree_trans *trans, 130 struct btree_path *path, 131 unsigned level, unsigned flags, 132 enum btree_node_sibling sib) 133 { 134 struct btree *b; 135 136 EBUG_ON(!btree_node_locked(path, level)); 137 138 b = path->l[level].b; 139 if (b->sib_u64s[sib] > trans->c->btree_foreground_merge_threshold) 140 return 0; 141 142 return __bch2_foreground_maybe_merge(trans, path, level, flags, sib); 143 } 144 145 static inline int bch2_foreground_maybe_merge(struct btree_trans *trans, 146 struct btree_path *path, 147 unsigned level, 148 unsigned flags) 149 { 150 return bch2_foreground_maybe_merge_sibling(trans, path, level, flags, 151 btree_prev_sib) ?: 152 bch2_foreground_maybe_merge_sibling(trans, path, level, flags, 153 btree_next_sib); 154 } 155 156 int bch2_btree_node_rewrite(struct btree_trans *, struct btree_iter *, 157 struct btree *, unsigned); 158 void bch2_btree_node_rewrite_async(struct bch_fs *, struct btree *); 159 int bch2_btree_node_update_key(struct btree_trans *, struct btree_iter *, 160 struct btree *, struct bkey_i *, 161 unsigned, bool); 162 int bch2_btree_node_update_key_get_iter(struct btree_trans *, struct btree *, 163 struct bkey_i *, unsigned, bool); 164 165 void bch2_btree_set_root_for_read(struct bch_fs *, struct btree *); 166 void bch2_btree_root_alloc(struct bch_fs *, enum btree_id); 167 168 static inline unsigned btree_update_reserve_required(struct bch_fs *c, 169 struct btree *b) 170 { 171 unsigned depth = btree_node_root(c, b)->c.level + 1; 172 173 /* 174 * Number of nodes we might have to allocate in a worst case btree 175 * split operation - we split all the way up to the root, then allocate 176 * a new root, unless we're already at max depth: 177 */ 178 if (depth < BTREE_MAX_DEPTH) 179 return (depth - b->c.level) * 2 + 1; 180 else 181 return (depth - b->c.level) * 2 - 1; 182 } 183 184 static inline void btree_node_reset_sib_u64s(struct btree *b) 185 { 186 b->sib_u64s[0] = b->nr.live_u64s; 187 b->sib_u64s[1] = b->nr.live_u64s; 188 } 189 190 static inline void *btree_data_end(struct bch_fs *c, struct btree *b) 191 { 192 return (void *) b->data + btree_bytes(c); 193 } 194 195 static inline struct bkey_packed *unwritten_whiteouts_start(struct bch_fs *c, 196 struct btree *b) 197 { 198 return (void *) ((u64 *) btree_data_end(c, b) - b->whiteout_u64s); 199 } 200 201 static inline struct bkey_packed *unwritten_whiteouts_end(struct bch_fs *c, 202 struct btree *b) 203 { 204 return btree_data_end(c, b); 205 } 206 207 static inline void *write_block(struct btree *b) 208 { 209 return (void *) b->data + (b->written << 9); 210 } 211 212 static inline bool __btree_addr_written(struct btree *b, void *p) 213 { 214 return p < write_block(b); 215 } 216 217 static inline bool bset_written(struct btree *b, struct bset *i) 218 { 219 return __btree_addr_written(b, i); 220 } 221 222 static inline bool bkey_written(struct btree *b, struct bkey_packed *k) 223 { 224 return __btree_addr_written(b, k); 225 } 226 227 static inline ssize_t __bch_btree_u64s_remaining(struct bch_fs *c, 228 struct btree *b, 229 void *end) 230 { 231 ssize_t used = bset_byte_offset(b, end) / sizeof(u64) + 232 b->whiteout_u64s; 233 ssize_t total = c->opts.btree_node_size >> 3; 234 235 /* Always leave one extra u64 for bch2_varint_decode: */ 236 used++; 237 238 return total - used; 239 } 240 241 static inline size_t bch_btree_keys_u64s_remaining(struct bch_fs *c, 242 struct btree *b) 243 { 244 ssize_t remaining = __bch_btree_u64s_remaining(c, b, 245 btree_bkey_last(b, bset_tree_last(b))); 246 247 BUG_ON(remaining < 0); 248 249 if (bset_written(b, btree_bset_last(b))) 250 return 0; 251 252 return remaining; 253 } 254 255 #define BTREE_WRITE_SET_U64s_BITS 9 256 257 static inline unsigned btree_write_set_buffer(struct btree *b) 258 { 259 /* 260 * Could buffer up larger amounts of keys for btrees with larger keys, 261 * pending benchmarking: 262 */ 263 return 8 << BTREE_WRITE_SET_U64s_BITS; 264 } 265 266 static inline struct btree_node_entry *want_new_bset(struct bch_fs *c, 267 struct btree *b) 268 { 269 struct bset_tree *t = bset_tree_last(b); 270 struct btree_node_entry *bne = max(write_block(b), 271 (void *) btree_bkey_last(b, bset_tree_last(b))); 272 ssize_t remaining_space = 273 __bch_btree_u64s_remaining(c, b, bne->keys.start); 274 275 if (unlikely(bset_written(b, bset(b, t)))) { 276 if (remaining_space > (ssize_t) (block_bytes(c) >> 3)) 277 return bne; 278 } else { 279 if (unlikely(bset_u64s(t) * sizeof(u64) > btree_write_set_buffer(b)) && 280 remaining_space > (ssize_t) (btree_write_set_buffer(b) >> 3)) 281 return bne; 282 } 283 284 return NULL; 285 } 286 287 static inline void push_whiteout(struct bch_fs *c, struct btree *b, 288 struct bpos pos) 289 { 290 struct bkey_packed k; 291 292 BUG_ON(bch_btree_keys_u64s_remaining(c, b) < BKEY_U64s); 293 EBUG_ON(btree_node_just_written(b)); 294 295 if (!bkey_pack_pos(&k, pos, b)) { 296 struct bkey *u = (void *) &k; 297 298 bkey_init(u); 299 u->p = pos; 300 } 301 302 k.needs_whiteout = true; 303 304 b->whiteout_u64s += k.u64s; 305 bkey_p_copy(unwritten_whiteouts_start(c, b), &k); 306 } 307 308 /* 309 * write lock must be held on @b (else the dirty bset that we were going to 310 * insert into could be written out from under us) 311 */ 312 static inline bool bch2_btree_node_insert_fits(struct bch_fs *c, 313 struct btree *b, unsigned u64s) 314 { 315 if (unlikely(btree_node_need_rewrite(b))) 316 return false; 317 318 return u64s <= bch_btree_keys_u64s_remaining(c, b); 319 } 320 321 void bch2_btree_updates_to_text(struct printbuf *, struct bch_fs *); 322 323 bool bch2_btree_interior_updates_flush(struct bch_fs *); 324 325 void bch2_journal_entry_to_btree_root(struct bch_fs *, struct jset_entry *); 326 struct jset_entry *bch2_btree_roots_to_journal_entries(struct bch_fs *, 327 struct jset_entry *, unsigned long); 328 329 void bch2_do_pending_node_rewrites(struct bch_fs *); 330 void bch2_free_pending_node_rewrites(struct bch_fs *); 331 332 void bch2_fs_btree_interior_update_exit(struct bch_fs *); 333 void bch2_fs_btree_interior_update_init_early(struct bch_fs *); 334 int bch2_fs_btree_interior_update_init(struct bch_fs *); 335 336 #endif /* _BCACHEFS_BTREE_UPDATE_INTERIOR_H */ 337