Joern Engel | 5db53f3 | 2009-11-20 20:13:39 +0100 | [diff] [blame^] | 1 | /* |
| 2 | * fs/logfs/gc.c - garbage collection code |
| 3 | * |
| 4 | * As should be obvious for Linux kernel code, license is GPLv2 |
| 5 | * |
| 6 | * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> |
| 7 | */ |
| 8 | #include "logfs.h" |
| 9 | #include <linux/sched.h> |
| 10 | |
| 11 | /* |
| 12 | * Wear leveling needs to kick in when the difference between low erase |
| 13 | * counts and high erase counts gets too big. A good value for "too big" |
| 14 | * may be somewhat below 10% of maximum erase count for the device. |
| 15 | * Why not 397, to pick a nice round number with no specific meaning? :) |
| 16 | * |
| 17 | * WL_RATELIMIT is the minimum time between two wear level events. A huge |
| 18 | * number of segments may fulfil the requirements for wear leveling at the |
| 19 | * same time. If that happens we don't want to cause a latency from hell, |
| 20 | * but just gently pick one segment every so often and minimize overhead. |
| 21 | */ |
| 22 | #define WL_DELTA 397 |
| 23 | #define WL_RATELIMIT 100 |
| 24 | #define MAX_OBJ_ALIASES 2600 |
| 25 | #define SCAN_RATIO 512 /* number of scanned segments per gc'd segment */ |
| 26 | #define LIST_SIZE 64 /* base size of candidate lists */ |
| 27 | #define SCAN_ROUNDS 128 /* maximum number of complete medium scans */ |
| 28 | #define SCAN_ROUNDS_HIGH 4 /* maximum number of higher-level scans */ |
| 29 | |
| 30 | static int no_free_segments(struct super_block *sb) |
| 31 | { |
| 32 | struct logfs_super *super = logfs_super(sb); |
| 33 | |
| 34 | return super->s_free_list.count; |
| 35 | } |
| 36 | |
| 37 | /* journal has distance -1, top-most ifile layer distance 0 */ |
| 38 | static u8 root_distance(struct super_block *sb, gc_level_t __gc_level) |
| 39 | { |
| 40 | struct logfs_super *super = logfs_super(sb); |
| 41 | u8 gc_level = (__force u8)__gc_level; |
| 42 | |
| 43 | switch (gc_level) { |
| 44 | case 0: /* fall through */ |
| 45 | case 1: /* fall through */ |
| 46 | case 2: /* fall through */ |
| 47 | case 3: |
| 48 | /* file data or indirect blocks */ |
| 49 | return super->s_ifile_levels + super->s_iblock_levels - gc_level; |
| 50 | case 6: /* fall through */ |
| 51 | case 7: /* fall through */ |
| 52 | case 8: /* fall through */ |
| 53 | case 9: |
| 54 | /* inode file data or indirect blocks */ |
| 55 | return super->s_ifile_levels - (gc_level - 6); |
| 56 | default: |
| 57 | printk(KERN_ERR"LOGFS: segment of unknown level %x found\n", |
| 58 | gc_level); |
| 59 | WARN_ON(1); |
| 60 | return super->s_ifile_levels + super->s_iblock_levels; |
| 61 | } |
| 62 | } |
| 63 | |
| 64 | static int segment_is_reserved(struct super_block *sb, u32 segno) |
| 65 | { |
| 66 | struct logfs_super *super = logfs_super(sb); |
| 67 | struct logfs_area *area; |
| 68 | void *reserved; |
| 69 | int i; |
| 70 | |
| 71 | /* Some segments are reserved. Just pretend they were all valid */ |
| 72 | reserved = btree_lookup32(&super->s_reserved_segments, segno); |
| 73 | if (reserved) |
| 74 | return 1; |
| 75 | |
| 76 | /* Currently open segments */ |
| 77 | for_each_area(i) { |
| 78 | area = super->s_area[i]; |
| 79 | if (area->a_is_open && area->a_segno == segno) |
| 80 | return 1; |
| 81 | } |
| 82 | |
| 83 | return 0; |
| 84 | } |
| 85 | |
| 86 | static void logfs_mark_segment_bad(struct super_block *sb, u32 segno) |
| 87 | { |
| 88 | BUG(); |
| 89 | } |
| 90 | |
| 91 | /* |
| 92 | * Returns the bytes consumed by valid objects in this segment. Object headers |
| 93 | * are counted, the segment header is not. |
| 94 | */ |
| 95 | static u32 logfs_valid_bytes(struct super_block *sb, u32 segno, u32 *ec, |
| 96 | gc_level_t *gc_level) |
| 97 | { |
| 98 | struct logfs_segment_entry se; |
| 99 | u32 ec_level; |
| 100 | |
| 101 | logfs_get_segment_entry(sb, segno, &se); |
| 102 | if (se.ec_level == cpu_to_be32(BADSEG) || |
| 103 | se.valid == cpu_to_be32(RESERVED)) |
| 104 | return RESERVED; |
| 105 | |
| 106 | ec_level = be32_to_cpu(se.ec_level); |
| 107 | *ec = ec_level >> 4; |
| 108 | *gc_level = GC_LEVEL(ec_level & 0xf); |
| 109 | return be32_to_cpu(se.valid); |
| 110 | } |
| 111 | |
| 112 | static void logfs_cleanse_block(struct super_block *sb, u64 ofs, u64 ino, |
| 113 | u64 bix, gc_level_t gc_level) |
| 114 | { |
| 115 | struct inode *inode; |
| 116 | int err, cookie; |
| 117 | |
| 118 | inode = logfs_safe_iget(sb, ino, &cookie); |
| 119 | err = logfs_rewrite_block(inode, bix, ofs, gc_level, 0); |
| 120 | BUG_ON(err); |
| 121 | logfs_safe_iput(inode, cookie); |
| 122 | } |
| 123 | |
| 124 | static u32 logfs_gc_segment(struct super_block *sb, u32 segno, u8 dist) |
| 125 | { |
| 126 | struct logfs_super *super = logfs_super(sb); |
| 127 | struct logfs_segment_header sh; |
| 128 | struct logfs_object_header oh; |
| 129 | u64 ofs, ino, bix; |
| 130 | u32 seg_ofs, logical_segno, cleaned = 0; |
| 131 | int err, len, valid; |
| 132 | gc_level_t gc_level; |
| 133 | |
| 134 | LOGFS_BUG_ON(segment_is_reserved(sb, segno), sb); |
| 135 | |
| 136 | btree_insert32(&super->s_reserved_segments, segno, (void *)1, GFP_NOFS); |
| 137 | err = wbuf_read(sb, dev_ofs(sb, segno, 0), sizeof(sh), &sh); |
| 138 | BUG_ON(err); |
| 139 | gc_level = GC_LEVEL(sh.level); |
| 140 | logical_segno = be32_to_cpu(sh.segno); |
| 141 | if (sh.crc != logfs_crc32(&sh, sizeof(sh), 4)) { |
| 142 | logfs_mark_segment_bad(sb, segno); |
| 143 | cleaned = -1; |
| 144 | goto out; |
| 145 | } |
| 146 | |
| 147 | for (seg_ofs = LOGFS_SEGMENT_HEADERSIZE; |
| 148 | seg_ofs + sizeof(oh) < super->s_segsize; ) { |
| 149 | ofs = dev_ofs(sb, logical_segno, seg_ofs); |
| 150 | err = wbuf_read(sb, dev_ofs(sb, segno, seg_ofs), sizeof(oh), |
| 151 | &oh); |
| 152 | BUG_ON(err); |
| 153 | |
| 154 | if (!memchr_inv(&oh, 0xff, sizeof(oh))) |
| 155 | break; |
| 156 | |
| 157 | if (oh.crc != logfs_crc32(&oh, sizeof(oh) - 4, 4)) { |
| 158 | logfs_mark_segment_bad(sb, segno); |
| 159 | cleaned = super->s_segsize - 1; |
| 160 | goto out; |
| 161 | } |
| 162 | |
| 163 | ino = be64_to_cpu(oh.ino); |
| 164 | bix = be64_to_cpu(oh.bix); |
| 165 | len = sizeof(oh) + be16_to_cpu(oh.len); |
| 166 | valid = logfs_is_valid_block(sb, ofs, ino, bix, gc_level); |
| 167 | if (valid == 1) { |
| 168 | logfs_cleanse_block(sb, ofs, ino, bix, gc_level); |
| 169 | cleaned += len; |
| 170 | } else if (valid == 2) { |
| 171 | /* Will be invalid upon journal commit */ |
| 172 | cleaned += len; |
| 173 | } |
| 174 | seg_ofs += len; |
| 175 | } |
| 176 | out: |
| 177 | btree_remove32(&super->s_reserved_segments, segno); |
| 178 | return cleaned; |
| 179 | } |
| 180 | |
| 181 | static struct gc_candidate *add_list(struct gc_candidate *cand, |
| 182 | struct candidate_list *list) |
| 183 | { |
| 184 | struct rb_node **p = &list->rb_tree.rb_node; |
| 185 | struct rb_node *parent = NULL; |
| 186 | struct gc_candidate *cur; |
| 187 | int comp; |
| 188 | |
| 189 | cand->list = list; |
| 190 | while (*p) { |
| 191 | parent = *p; |
| 192 | cur = rb_entry(parent, struct gc_candidate, rb_node); |
| 193 | |
| 194 | if (list->sort_by_ec) |
| 195 | comp = cand->erase_count < cur->erase_count; |
| 196 | else |
| 197 | comp = cand->valid < cur->valid; |
| 198 | |
| 199 | if (comp) |
| 200 | p = &parent->rb_left; |
| 201 | else |
| 202 | p = &parent->rb_right; |
| 203 | } |
| 204 | rb_link_node(&cand->rb_node, parent, p); |
| 205 | rb_insert_color(&cand->rb_node, &list->rb_tree); |
| 206 | |
| 207 | if (list->count <= list->maxcount) { |
| 208 | list->count++; |
| 209 | return NULL; |
| 210 | } |
| 211 | cand = rb_entry(rb_last(&list->rb_tree), struct gc_candidate, rb_node); |
| 212 | rb_erase(&cand->rb_node, &list->rb_tree); |
| 213 | cand->list = NULL; |
| 214 | return cand; |
| 215 | } |
| 216 | |
| 217 | static void remove_from_list(struct gc_candidate *cand) |
| 218 | { |
| 219 | struct candidate_list *list = cand->list; |
| 220 | |
| 221 | rb_erase(&cand->rb_node, &list->rb_tree); |
| 222 | list->count--; |
| 223 | } |
| 224 | |
| 225 | static void free_candidate(struct super_block *sb, struct gc_candidate *cand) |
| 226 | { |
| 227 | struct logfs_super *super = logfs_super(sb); |
| 228 | |
| 229 | btree_remove32(&super->s_cand_tree, cand->segno); |
| 230 | kfree(cand); |
| 231 | } |
| 232 | |
| 233 | u32 get_best_cand(struct super_block *sb, struct candidate_list *list, u32 *ec) |
| 234 | { |
| 235 | struct gc_candidate *cand; |
| 236 | u32 segno; |
| 237 | |
| 238 | BUG_ON(list->count == 0); |
| 239 | |
| 240 | cand = rb_entry(rb_first(&list->rb_tree), struct gc_candidate, rb_node); |
| 241 | remove_from_list(cand); |
| 242 | segno = cand->segno; |
| 243 | if (ec) |
| 244 | *ec = cand->erase_count; |
| 245 | free_candidate(sb, cand); |
| 246 | return segno; |
| 247 | } |
| 248 | |
| 249 | /* |
| 250 | * We have several lists to manage segments with. The reserve_list is used to |
| 251 | * deal with bad blocks. We try to keep the best (lowest ec) segments on this |
| 252 | * list. |
| 253 | * The free_list contains free segments for normal usage. It usually gets the |
| 254 | * second pick after the reserve_list. But when the free_list is running short |
| 255 | * it is more important to keep the free_list full than to keep a reserve. |
| 256 | * |
| 257 | * Segments that are not free are put onto a per-level low_list. If we have |
| 258 | * to run garbage collection, we pick a candidate from there. All segments on |
| 259 | * those lists should have at least some free space so GC will make progress. |
| 260 | * |
| 261 | * And last we have the ec_list, which is used to pick segments for wear |
| 262 | * leveling. |
| 263 | * |
| 264 | * If all appropriate lists are full, we simply free the candidate and forget |
| 265 | * about that segment for a while. We have better candidates for each purpose. |
| 266 | */ |
| 267 | static void __add_candidate(struct super_block *sb, struct gc_candidate *cand) |
| 268 | { |
| 269 | struct logfs_super *super = logfs_super(sb); |
| 270 | u32 full = super->s_segsize - LOGFS_SEGMENT_RESERVE; |
| 271 | |
| 272 | if (cand->valid == 0) { |
| 273 | /* 100% free segments */ |
| 274 | log_gc_noisy("add reserve segment %x (ec %x) at %llx\n", |
| 275 | cand->segno, cand->erase_count, |
| 276 | dev_ofs(sb, cand->segno, 0)); |
| 277 | cand = add_list(cand, &super->s_reserve_list); |
| 278 | if (cand) { |
| 279 | log_gc_noisy("add free segment %x (ec %x) at %llx\n", |
| 280 | cand->segno, cand->erase_count, |
| 281 | dev_ofs(sb, cand->segno, 0)); |
| 282 | cand = add_list(cand, &super->s_free_list); |
| 283 | } |
| 284 | } else { |
| 285 | /* good candidates for Garbage Collection */ |
| 286 | if (cand->valid < full) |
| 287 | cand = add_list(cand, &super->s_low_list[cand->dist]); |
| 288 | /* good candidates for wear leveling, |
| 289 | * segments that were recently written get ignored */ |
| 290 | if (cand) |
| 291 | cand = add_list(cand, &super->s_ec_list); |
| 292 | } |
| 293 | if (cand) |
| 294 | free_candidate(sb, cand); |
| 295 | } |
| 296 | |
| 297 | static int add_candidate(struct super_block *sb, u32 segno, u32 valid, u32 ec, |
| 298 | u8 dist) |
| 299 | { |
| 300 | struct logfs_super *super = logfs_super(sb); |
| 301 | struct gc_candidate *cand; |
| 302 | |
| 303 | cand = kmalloc(sizeof(*cand), GFP_NOFS); |
| 304 | if (!cand) |
| 305 | return -ENOMEM; |
| 306 | |
| 307 | cand->segno = segno; |
| 308 | cand->valid = valid; |
| 309 | cand->erase_count = ec; |
| 310 | cand->dist = dist; |
| 311 | |
| 312 | btree_insert32(&super->s_cand_tree, segno, cand, GFP_NOFS); |
| 313 | __add_candidate(sb, cand); |
| 314 | return 0; |
| 315 | } |
| 316 | |
| 317 | static void remove_segment_from_lists(struct super_block *sb, u32 segno) |
| 318 | { |
| 319 | struct logfs_super *super = logfs_super(sb); |
| 320 | struct gc_candidate *cand; |
| 321 | |
| 322 | cand = btree_lookup32(&super->s_cand_tree, segno); |
| 323 | if (cand) { |
| 324 | remove_from_list(cand); |
| 325 | free_candidate(sb, cand); |
| 326 | } |
| 327 | } |
| 328 | |
| 329 | static void scan_segment(struct super_block *sb, u32 segno) |
| 330 | { |
| 331 | u32 valid, ec = 0; |
| 332 | gc_level_t gc_level = 0; |
| 333 | u8 dist; |
| 334 | |
| 335 | if (segment_is_reserved(sb, segno)) |
| 336 | return; |
| 337 | |
| 338 | remove_segment_from_lists(sb, segno); |
| 339 | valid = logfs_valid_bytes(sb, segno, &ec, &gc_level); |
| 340 | if (valid == RESERVED) |
| 341 | return; |
| 342 | |
| 343 | dist = root_distance(sb, gc_level); |
| 344 | add_candidate(sb, segno, valid, ec, dist); |
| 345 | } |
| 346 | |
| 347 | static struct gc_candidate *first_in_list(struct candidate_list *list) |
| 348 | { |
| 349 | if (list->count == 0) |
| 350 | return NULL; |
| 351 | return rb_entry(rb_first(&list->rb_tree), struct gc_candidate, rb_node); |
| 352 | } |
| 353 | |
| 354 | /* |
| 355 | * Find the best segment for garbage collection. Main criterion is |
| 356 | * the segment requiring the least effort to clean. Secondary |
| 357 | * criterion is to GC on the lowest level available. |
| 358 | * |
| 359 | * So we search the least effort segment on the lowest level first, |
| 360 | * then move up and pick another segment iff is requires significantly |
| 361 | * less effort. Hence the LOGFS_MAX_OBJECTSIZE in the comparison. |
| 362 | */ |
| 363 | static struct gc_candidate *get_candidate(struct super_block *sb) |
| 364 | { |
| 365 | struct logfs_super *super = logfs_super(sb); |
| 366 | int i, max_dist; |
| 367 | struct gc_candidate *cand = NULL, *this; |
| 368 | |
| 369 | max_dist = min(no_free_segments(sb), LOGFS_NO_AREAS); |
| 370 | |
| 371 | for (i = max_dist; i >= 0; i--) { |
| 372 | this = first_in_list(&super->s_low_list[i]); |
| 373 | if (!this) |
| 374 | continue; |
| 375 | if (!cand) |
| 376 | cand = this; |
| 377 | if (this->valid + LOGFS_MAX_OBJECTSIZE <= cand->valid) |
| 378 | cand = this; |
| 379 | } |
| 380 | return cand; |
| 381 | } |
| 382 | |
| 383 | static int __logfs_gc_once(struct super_block *sb, struct gc_candidate *cand) |
| 384 | { |
| 385 | struct logfs_super *super = logfs_super(sb); |
| 386 | gc_level_t gc_level; |
| 387 | u32 cleaned, valid, segno, ec; |
| 388 | u8 dist; |
| 389 | |
| 390 | if (!cand) { |
| 391 | log_gc("GC attempted, but no candidate found\n"); |
| 392 | return 0; |
| 393 | } |
| 394 | |
| 395 | segno = cand->segno; |
| 396 | dist = cand->dist; |
| 397 | valid = logfs_valid_bytes(sb, segno, &ec, &gc_level); |
| 398 | free_candidate(sb, cand); |
| 399 | log_gc("GC segment #%02x at %llx, %x required, %x free, %x valid, %llx free\n", |
| 400 | segno, (u64)segno << super->s_segshift, |
| 401 | dist, no_free_segments(sb), valid, |
| 402 | super->s_free_bytes); |
| 403 | cleaned = logfs_gc_segment(sb, segno, dist); |
| 404 | log_gc("GC segment #%02x complete - now %x valid\n", segno, |
| 405 | valid - cleaned); |
| 406 | BUG_ON(cleaned != valid); |
| 407 | return 1; |
| 408 | } |
| 409 | |
| 410 | static int logfs_gc_once(struct super_block *sb) |
| 411 | { |
| 412 | struct gc_candidate *cand; |
| 413 | |
| 414 | cand = get_candidate(sb); |
| 415 | if (cand) |
| 416 | remove_from_list(cand); |
| 417 | return __logfs_gc_once(sb, cand); |
| 418 | } |
| 419 | |
| 420 | /* returns 1 if a wrap occurs, 0 otherwise */ |
| 421 | static int logfs_scan_some(struct super_block *sb) |
| 422 | { |
| 423 | struct logfs_super *super = logfs_super(sb); |
| 424 | u32 segno; |
| 425 | int i, ret = 0; |
| 426 | |
| 427 | segno = super->s_sweeper; |
| 428 | for (i = SCAN_RATIO; i > 0; i--) { |
| 429 | segno++; |
| 430 | if (segno >= super->s_no_segs) { |
| 431 | segno = 0; |
| 432 | ret = 1; |
| 433 | /* Break out of the loop. We want to read a single |
| 434 | * block from the segment size on next invocation if |
| 435 | * SCAN_RATIO is set to match block size |
| 436 | */ |
| 437 | break; |
| 438 | } |
| 439 | |
| 440 | scan_segment(sb, segno); |
| 441 | } |
| 442 | super->s_sweeper = segno; |
| 443 | return ret; |
| 444 | } |
| 445 | |
| 446 | /* |
| 447 | * In principle, this function should loop forever, looking for GC candidates |
| 448 | * and moving data. LogFS is designed in such a way that this loop is |
| 449 | * guaranteed to terminate. |
| 450 | * |
| 451 | * Limiting the loop to some iterations serves purely to catch cases when |
| 452 | * these guarantees have failed. An actual endless loop is an obvious bug |
| 453 | * and should be reported as such. |
| 454 | */ |
| 455 | static void __logfs_gc_pass(struct super_block *sb, int target) |
| 456 | { |
| 457 | struct logfs_super *super = logfs_super(sb); |
| 458 | struct logfs_block *block; |
| 459 | int round, progress, last_progress = 0; |
| 460 | |
| 461 | if (no_free_segments(sb) >= target && |
| 462 | super->s_no_object_aliases < MAX_OBJ_ALIASES) |
| 463 | return; |
| 464 | |
| 465 | log_gc("__logfs_gc_pass(%x)\n", target); |
| 466 | for (round = 0; round < SCAN_ROUNDS; ) { |
| 467 | if (no_free_segments(sb) >= target) |
| 468 | goto write_alias; |
| 469 | |
| 470 | /* Sync in-memory state with on-medium state in case they |
| 471 | * diverged */ |
| 472 | logfs_write_anchor(super->s_master_inode); |
| 473 | round += logfs_scan_some(sb); |
| 474 | if (no_free_segments(sb) >= target) |
| 475 | goto write_alias; |
| 476 | progress = logfs_gc_once(sb); |
| 477 | if (progress) |
| 478 | last_progress = round; |
| 479 | else if (round - last_progress > 2) |
| 480 | break; |
| 481 | continue; |
| 482 | |
| 483 | /* |
| 484 | * The goto logic is nasty, I just don't know a better way to |
| 485 | * code it. GC is supposed to ensure two things: |
| 486 | * 1. Enough free segments are available. |
| 487 | * 2. The number of aliases is bounded. |
| 488 | * When 1. is achieved, we take a look at 2. and write back |
| 489 | * some alias-containing blocks, if necessary. However, after |
| 490 | * each such write we need to go back to 1., as writes can |
| 491 | * consume free segments. |
| 492 | */ |
| 493 | write_alias: |
| 494 | if (super->s_no_object_aliases < MAX_OBJ_ALIASES) |
| 495 | return; |
| 496 | if (list_empty(&super->s_object_alias)) { |
| 497 | /* All aliases are still in btree */ |
| 498 | return; |
| 499 | } |
| 500 | log_gc("Write back one alias\n"); |
| 501 | block = list_entry(super->s_object_alias.next, |
| 502 | struct logfs_block, alias_list); |
| 503 | block->ops->write_block(block); |
| 504 | /* |
| 505 | * To round off the nasty goto logic, we reset round here. It |
| 506 | * is a safety-net for GC not making any progress and limited |
| 507 | * to something reasonably small. If incremented it for every |
| 508 | * single alias, the loop could terminate rather quickly. |
| 509 | */ |
| 510 | round = 0; |
| 511 | } |
| 512 | LOGFS_BUG(sb); |
| 513 | } |
| 514 | |
| 515 | static int wl_ratelimit(struct super_block *sb, u64 *next_event) |
| 516 | { |
| 517 | struct logfs_super *super = logfs_super(sb); |
| 518 | |
| 519 | if (*next_event < super->s_gec) { |
| 520 | *next_event = super->s_gec + WL_RATELIMIT; |
| 521 | return 0; |
| 522 | } |
| 523 | return 1; |
| 524 | } |
| 525 | |
| 526 | static void logfs_wl_pass(struct super_block *sb) |
| 527 | { |
| 528 | struct logfs_super *super = logfs_super(sb); |
| 529 | struct gc_candidate *wl_cand, *free_cand; |
| 530 | |
| 531 | if (wl_ratelimit(sb, &super->s_wl_gec_ostore)) |
| 532 | return; |
| 533 | |
| 534 | wl_cand = first_in_list(&super->s_ec_list); |
| 535 | if (!wl_cand) |
| 536 | return; |
| 537 | free_cand = first_in_list(&super->s_free_list); |
| 538 | if (!free_cand) |
| 539 | return; |
| 540 | |
| 541 | if (wl_cand->erase_count < free_cand->erase_count + WL_DELTA) { |
| 542 | remove_from_list(wl_cand); |
| 543 | __logfs_gc_once(sb, wl_cand); |
| 544 | } |
| 545 | } |
| 546 | |
| 547 | /* |
| 548 | * The journal needs wear leveling as well. But moving the journal is an |
| 549 | * expensive operation so we try to avoid it as much as possible. And if we |
| 550 | * have to do it, we move the whole journal, not individual segments. |
| 551 | * |
| 552 | * Ratelimiting is not strictly necessary here, it mainly serves to avoid the |
| 553 | * calculations. First we check whether moving the journal would be a |
| 554 | * significant improvement. That means that a) the current journal segments |
| 555 | * have more wear than the future journal segments and b) the current journal |
| 556 | * segments have more wear than normal ostore segments. |
| 557 | * Rationale for b) is that we don't have to move the journal if it is aging |
| 558 | * less than the ostore, even if the reserve segments age even less (they are |
| 559 | * excluded from wear leveling, after all). |
| 560 | * Next we check that the superblocks have less wear than the journal. Since |
| 561 | * moving the journal requires writing the superblocks, we have to protect the |
| 562 | * superblocks even more than the journal. |
| 563 | * |
| 564 | * Also we double the acceptable wear difference, compared to ostore wear |
| 565 | * leveling. Journal data is read and rewritten rapidly, comparatively. So |
| 566 | * soft errors have much less time to accumulate and we allow the journal to |
| 567 | * be a bit worse than the ostore. |
| 568 | */ |
| 569 | static void logfs_journal_wl_pass(struct super_block *sb) |
| 570 | { |
| 571 | struct logfs_super *super = logfs_super(sb); |
| 572 | struct gc_candidate *cand; |
| 573 | u32 min_journal_ec = -1, max_reserve_ec = 0; |
| 574 | int i; |
| 575 | |
| 576 | if (wl_ratelimit(sb, &super->s_wl_gec_journal)) |
| 577 | return; |
| 578 | |
| 579 | if (super->s_reserve_list.count < super->s_no_journal_segs) { |
| 580 | /* Reserve is not full enough to move complete journal */ |
| 581 | return; |
| 582 | } |
| 583 | |
| 584 | journal_for_each(i) |
| 585 | if (super->s_journal_seg[i]) |
| 586 | min_journal_ec = min(min_journal_ec, |
| 587 | super->s_journal_ec[i]); |
| 588 | cand = rb_entry(rb_first(&super->s_free_list.rb_tree), |
| 589 | struct gc_candidate, rb_node); |
| 590 | max_reserve_ec = cand->erase_count; |
| 591 | for (i = 0; i < 2; i++) { |
| 592 | struct logfs_segment_entry se; |
| 593 | u32 segno = seg_no(sb, super->s_sb_ofs[i]); |
| 594 | u32 ec; |
| 595 | |
| 596 | logfs_get_segment_entry(sb, segno, &se); |
| 597 | ec = be32_to_cpu(se.ec_level) >> 4; |
| 598 | max_reserve_ec = max(max_reserve_ec, ec); |
| 599 | } |
| 600 | |
| 601 | if (min_journal_ec > max_reserve_ec + 2 * WL_DELTA) { |
| 602 | do_logfs_journal_wl_pass(sb); |
| 603 | } |
| 604 | } |
| 605 | |
| 606 | void logfs_gc_pass(struct super_block *sb) |
| 607 | { |
| 608 | struct logfs_super *super = logfs_super(sb); |
| 609 | |
| 610 | //BUG_ON(mutex_trylock(&logfs_super(sb)->s_w_mutex)); |
| 611 | /* Write journal before free space is getting saturated with dirty |
| 612 | * objects. |
| 613 | */ |
| 614 | if (super->s_dirty_used_bytes + super->s_dirty_free_bytes |
| 615 | + LOGFS_MAX_OBJECTSIZE >= super->s_free_bytes) |
| 616 | logfs_write_anchor(super->s_master_inode); |
| 617 | __logfs_gc_pass(sb, logfs_super(sb)->s_total_levels); |
| 618 | logfs_wl_pass(sb); |
| 619 | logfs_journal_wl_pass(sb); |
| 620 | } |
| 621 | |
| 622 | static int check_area(struct super_block *sb, int i) |
| 623 | { |
| 624 | struct logfs_super *super = logfs_super(sb); |
| 625 | struct logfs_area *area = super->s_area[i]; |
| 626 | struct logfs_object_header oh; |
| 627 | u32 segno = area->a_segno; |
| 628 | u32 ofs = area->a_used_bytes; |
| 629 | __be32 crc; |
| 630 | int err; |
| 631 | |
| 632 | if (!area->a_is_open) |
| 633 | return 0; |
| 634 | |
| 635 | for (ofs = area->a_used_bytes; |
| 636 | ofs <= super->s_segsize - sizeof(oh); |
| 637 | ofs += (u32)be16_to_cpu(oh.len) + sizeof(oh)) { |
| 638 | err = wbuf_read(sb, dev_ofs(sb, segno, ofs), sizeof(oh), &oh); |
| 639 | if (err) |
| 640 | return err; |
| 641 | |
| 642 | if (!memchr_inv(&oh, 0xff, sizeof(oh))) |
| 643 | break; |
| 644 | |
| 645 | crc = logfs_crc32(&oh, sizeof(oh) - 4, 4); |
| 646 | if (crc != oh.crc) { |
| 647 | printk(KERN_INFO "interrupted header at %llx\n", |
| 648 | dev_ofs(sb, segno, ofs)); |
| 649 | return 0; |
| 650 | } |
| 651 | } |
| 652 | if (ofs != area->a_used_bytes) { |
| 653 | printk(KERN_INFO "%x bytes unaccounted data found at %llx\n", |
| 654 | ofs - area->a_used_bytes, |
| 655 | dev_ofs(sb, segno, area->a_used_bytes)); |
| 656 | area->a_used_bytes = ofs; |
| 657 | } |
| 658 | return 0; |
| 659 | } |
| 660 | |
| 661 | int logfs_check_areas(struct super_block *sb) |
| 662 | { |
| 663 | int i, err; |
| 664 | |
| 665 | for_each_area(i) { |
| 666 | err = check_area(sb, i); |
| 667 | if (err) |
| 668 | return err; |
| 669 | } |
| 670 | return 0; |
| 671 | } |
| 672 | |
| 673 | static void logfs_init_candlist(struct candidate_list *list, int maxcount, |
| 674 | int sort_by_ec) |
| 675 | { |
| 676 | list->count = 0; |
| 677 | list->maxcount = maxcount; |
| 678 | list->sort_by_ec = sort_by_ec; |
| 679 | list->rb_tree = RB_ROOT; |
| 680 | } |
| 681 | |
| 682 | int logfs_init_gc(struct super_block *sb) |
| 683 | { |
| 684 | struct logfs_super *super = logfs_super(sb); |
| 685 | int i; |
| 686 | |
| 687 | btree_init_mempool32(&super->s_cand_tree, super->s_btree_pool); |
| 688 | logfs_init_candlist(&super->s_free_list, LIST_SIZE + SCAN_RATIO, 1); |
| 689 | logfs_init_candlist(&super->s_reserve_list, |
| 690 | super->s_bad_seg_reserve, 1); |
| 691 | for_each_area(i) |
| 692 | logfs_init_candlist(&super->s_low_list[i], LIST_SIZE, 0); |
| 693 | logfs_init_candlist(&super->s_ec_list, LIST_SIZE, 1); |
| 694 | return 0; |
| 695 | } |
| 696 | |
| 697 | static void logfs_cleanup_list(struct super_block *sb, |
| 698 | struct candidate_list *list) |
| 699 | { |
| 700 | struct gc_candidate *cand; |
| 701 | |
| 702 | while (list->count) { |
| 703 | cand = rb_entry(list->rb_tree.rb_node, struct gc_candidate, |
| 704 | rb_node); |
| 705 | remove_from_list(cand); |
| 706 | free_candidate(sb, cand); |
| 707 | } |
| 708 | BUG_ON(list->rb_tree.rb_node); |
| 709 | } |
| 710 | |
| 711 | void logfs_cleanup_gc(struct super_block *sb) |
| 712 | { |
| 713 | struct logfs_super *super = logfs_super(sb); |
| 714 | int i; |
| 715 | |
| 716 | if (!super->s_free_list.count) |
| 717 | return; |
| 718 | |
| 719 | /* |
| 720 | * FIXME: The btree may still contain a single empty node. So we |
| 721 | * call the grim visitor to clean up that mess. Btree code should |
| 722 | * do it for us, really. |
| 723 | */ |
| 724 | btree_grim_visitor32(&super->s_cand_tree, 0, NULL); |
| 725 | logfs_cleanup_list(sb, &super->s_free_list); |
| 726 | logfs_cleanup_list(sb, &super->s_reserve_list); |
| 727 | for_each_area(i) |
| 728 | logfs_cleanup_list(sb, &super->s_low_list[i]); |
| 729 | logfs_cleanup_list(sb, &super->s_ec_list); |
| 730 | } |