?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
USB: handle LPM errors during device suspend correctly
[linux/fpc-iii.git] / fs / f2fs / node.c
blobb418aee09573f6d4aff7c4f821ee4faf86d98b0b
1 /*
2 * fs/f2fs/node.c
3 *
4 * Copyright (c) 2012 Samsung Electronics Co., Ltd.
5 * http://www.samsung.com/
6 *
7 * This program is free software; you can redistribute it and/or modify
8 * it under the terms of the GNU General Public License version 2 as
9 * published by the Free Software Foundation.
10 */
11 #include <linux/fs.h>
12 #include <linux/f2fs_fs.h>
13 #include <linux/mpage.h>
14 #include <linux/backing-dev.h>
15 #include <linux/blkdev.h>
16 #include <linux/pagevec.h>
17 #include <linux/swap.h>
18
19 #include "f2fs.h"
20 #include "node.h"
21 #include "segment.h"
22 #include <trace/events/f2fs.h>
23
24 static struct kmem_cache *nat_entry_slab;
25 static struct kmem_cache *free_nid_slab;
26
27 static void clear_node_page_dirty(struct page *page)
28 {
29 struct address_space *mapping = page->mapping;
30 struct f2fs_sb_info *sbi = F2FS_SB(mapping->host->i_sb);
31 unsigned int long flags;
32
33 if (PageDirty(page)) {
34 spin_lock_irqsave(&mapping->tree_lock, flags);
35 radix_tree_tag_clear(&mapping->page_tree,
36 page_index(page),
37 PAGECACHE_TAG_DIRTY);
38 spin_unlock_irqrestore(&mapping->tree_lock, flags);
39
40 clear_page_dirty_for_io(page);
41 dec_page_count(sbi, F2FS_DIRTY_NODES);
42 }
43 ClearPageUptodate(page);
44 }
45
46 static struct page *get_current_nat_page(struct f2fs_sb_info *sbi, nid_t nid)
47 {
48 pgoff_t index = current_nat_addr(sbi, nid);
49 return get_meta_page(sbi, index);
50 }
51
52 static struct page *get_next_nat_page(struct f2fs_sb_info *sbi, nid_t nid)
53 {
54 struct page *src_page;
55 struct page *dst_page;
56 pgoff_t src_off;
57 pgoff_t dst_off;
58 void *src_addr;
59 void *dst_addr;
60 struct f2fs_nm_info *nm_i = NM_I(sbi);
61
62 src_off = current_nat_addr(sbi, nid);
63 dst_off = next_nat_addr(sbi, src_off);
64
65 /* get current nat block page with lock */
66 src_page = get_meta_page(sbi, src_off);
67
68 /* Dirty src_page means that it is already the new target NAT page. */
69 if (PageDirty(src_page))
70 return src_page;
71
72 dst_page = grab_meta_page(sbi, dst_off);
73
74 src_addr = page_address(src_page);
75 dst_addr = page_address(dst_page);
76 memcpy(dst_addr, src_addr, PAGE_CACHE_SIZE);
77 set_page_dirty(dst_page);
78 f2fs_put_page(src_page, 1);
79
80 set_to_next_nat(nm_i, nid);
81
82 return dst_page;
83 }
84
85 /*
86 * Readahead NAT pages
87 */
88 static void ra_nat_pages(struct f2fs_sb_info *sbi, int nid)
89 {
90 struct address_space *mapping = sbi->meta_inode->i_mapping;
91 struct f2fs_nm_info *nm_i = NM_I(sbi);
92 struct blk_plug plug;
93 struct page *page;
94 pgoff_t index;
95 int i;
96
97 blk_start_plug(&plug);
98
99 for (i = 0; i < FREE_NID_PAGES; i++, nid += NAT_ENTRY_PER_BLOCK) {
100 if (nid >= nm_i->max_nid)
101 nid = 0;
102 index = current_nat_addr(sbi, nid);
103
104 page = grab_cache_page(mapping, index);
105 if (!page)
106 continue;
107 if (PageUptodate(page)) {
108 f2fs_put_page(page, 1);
109 continue;
110 }
111 if (f2fs_readpage(sbi, page, index, READ))
112 continue;
113
114 f2fs_put_page(page, 0);
115 }
116 blk_finish_plug(&plug);
117 }
118
119 static struct nat_entry *__lookup_nat_cache(struct f2fs_nm_info *nm_i, nid_t n)
120 {
121 return radix_tree_lookup(&nm_i->nat_root, n);
122 }
123
124 static unsigned int __gang_lookup_nat_cache(struct f2fs_nm_info *nm_i,
125 nid_t start, unsigned int nr, struct nat_entry **ep)
126 {
127 return radix_tree_gang_lookup(&nm_i->nat_root, (void **)ep, start, nr);
128 }
129
130 static void __del_from_nat_cache(struct f2fs_nm_info *nm_i, struct nat_entry *e)
131 {
132 list_del(&e->list);
133 radix_tree_delete(&nm_i->nat_root, nat_get_nid(e));
134 nm_i->nat_cnt--;
135 kmem_cache_free(nat_entry_slab, e);
136 }
137
138 int is_checkpointed_node(struct f2fs_sb_info *sbi, nid_t nid)
139 {
140 struct f2fs_nm_info *nm_i = NM_I(sbi);
141 struct nat_entry *e;
142 int is_cp = 1;
143
144 read_lock(&nm_i->nat_tree_lock);
145 e = __lookup_nat_cache(nm_i, nid);
146 if (e && !e->checkpointed)
147 is_cp = 0;
148 read_unlock(&nm_i->nat_tree_lock);
149 return is_cp;
150 }
151
152 static struct nat_entry *grab_nat_entry(struct f2fs_nm_info *nm_i, nid_t nid)
153 {
154 struct nat_entry *new;
155
156 new = kmem_cache_alloc(nat_entry_slab, GFP_ATOMIC);
157 if (!new)
158 return NULL;
159 if (radix_tree_insert(&nm_i->nat_root, nid, new)) {
160 kmem_cache_free(nat_entry_slab, new);
161 return NULL;
162 }
163 memset(new, 0, sizeof(struct nat_entry));
164 nat_set_nid(new, nid);
165 list_add_tail(&new->list, &nm_i->nat_entries);
166 nm_i->nat_cnt++;
167 return new;
168 }
169
170 static void cache_nat_entry(struct f2fs_nm_info *nm_i, nid_t nid,
171 struct f2fs_nat_entry *ne)
172 {
173 struct nat_entry *e;
174 retry:
175 write_lock(&nm_i->nat_tree_lock);
176 e = __lookup_nat_cache(nm_i, nid);
177 if (!e) {
178 e = grab_nat_entry(nm_i, nid);
179 if (!e) {
180 write_unlock(&nm_i->nat_tree_lock);
181 goto retry;
182 }
183 nat_set_blkaddr(e, le32_to_cpu(ne->block_addr));
184 nat_set_ino(e, le32_to_cpu(ne->ino));
185 nat_set_version(e, ne->version);
186 e->checkpointed = true;
187 }
188 write_unlock(&nm_i->nat_tree_lock);
189 }
190
191 static void set_node_addr(struct f2fs_sb_info *sbi, struct node_info *ni,
192 block_t new_blkaddr)
193 {
194 struct f2fs_nm_info *nm_i = NM_I(sbi);
195 struct nat_entry *e;
196 retry:
197 write_lock(&nm_i->nat_tree_lock);
198 e = __lookup_nat_cache(nm_i, ni->nid);
199 if (!e) {
200 e = grab_nat_entry(nm_i, ni->nid);
201 if (!e) {
202 write_unlock(&nm_i->nat_tree_lock);
203 goto retry;
204 }
205 e->ni = *ni;
206 e->checkpointed = true;
207 BUG_ON(ni->blk_addr == NEW_ADDR);
208 } else if (new_blkaddr == NEW_ADDR) {
209 /*
210 * when nid is reallocated,
211 * previous nat entry can be remained in nat cache.
212 * So, reinitialize it with new information.
213 */
214 e->ni = *ni;
215 BUG_ON(ni->blk_addr != NULL_ADDR);
216 }
217
218 if (new_blkaddr == NEW_ADDR)
219 e->checkpointed = false;
220
221 /* sanity check */
222 BUG_ON(nat_get_blkaddr(e) != ni->blk_addr);
223 BUG_ON(nat_get_blkaddr(e) == NULL_ADDR &&
224 new_blkaddr == NULL_ADDR);
225 BUG_ON(nat_get_blkaddr(e) == NEW_ADDR &&
226 new_blkaddr == NEW_ADDR);
227 BUG_ON(nat_get_blkaddr(e) != NEW_ADDR &&
228 nat_get_blkaddr(e) != NULL_ADDR &&
229 new_blkaddr == NEW_ADDR);
230
231 /* increament version no as node is removed */
232 if (nat_get_blkaddr(e) != NEW_ADDR && new_blkaddr == NULL_ADDR) {
233 unsigned char version = nat_get_version(e);
234 nat_set_version(e, inc_node_version(version));
235 }
236
237 /* change address */
238 nat_set_blkaddr(e, new_blkaddr);
239 __set_nat_cache_dirty(nm_i, e);
240 write_unlock(&nm_i->nat_tree_lock);
241 }
242
243 static int try_to_free_nats(struct f2fs_sb_info *sbi, int nr_shrink)
244 {
245 struct f2fs_nm_info *nm_i = NM_I(sbi);
246
247 if (nm_i->nat_cnt <= NM_WOUT_THRESHOLD)
248 return 0;
249
250 write_lock(&nm_i->nat_tree_lock);
251 while (nr_shrink && !list_empty(&nm_i->nat_entries)) {
252 struct nat_entry *ne;
253 ne = list_first_entry(&nm_i->nat_entries,
254 struct nat_entry, list);
255 __del_from_nat_cache(nm_i, ne);
256 nr_shrink--;
257 }
258 write_unlock(&nm_i->nat_tree_lock);
259 return nr_shrink;
260 }
261
262 /*
263 * This function returns always success
264 */
265 void get_node_info(struct f2fs_sb_info *sbi, nid_t nid, struct node_info *ni)
266 {
267 struct f2fs_nm_info *nm_i = NM_I(sbi);
268 struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
269 struct f2fs_summary_block *sum = curseg->sum_blk;
270 nid_t start_nid = START_NID(nid);
271 struct f2fs_nat_block *nat_blk;
272 struct page *page = NULL;
273 struct f2fs_nat_entry ne;
274 struct nat_entry *e;
275 int i;
276
277 memset(&ne, 0, sizeof(struct f2fs_nat_entry));
278 ni->nid = nid;
279
280 /* Check nat cache */
281 read_lock(&nm_i->nat_tree_lock);
282 e = __lookup_nat_cache(nm_i, nid);
283 if (e) {
284 ni->ino = nat_get_ino(e);
285 ni->blk_addr = nat_get_blkaddr(e);
286 ni->version = nat_get_version(e);
287 }
288 read_unlock(&nm_i->nat_tree_lock);
289 if (e)
290 return;
291
292 /* Check current segment summary */
293 mutex_lock(&curseg->curseg_mutex);
294 i = lookup_journal_in_cursum(sum, NAT_JOURNAL, nid, 0);
295 if (i >= 0) {
296 ne = nat_in_journal(sum, i);
297 node_info_from_raw_nat(ni, &ne);
298 }
299 mutex_unlock(&curseg->curseg_mutex);
300 if (i >= 0)
301 goto cache;
302
303 /* Fill node_info from nat page */
304 page = get_current_nat_page(sbi, start_nid);
305 nat_blk = (struct f2fs_nat_block *)page_address(page);
306 ne = nat_blk->entries[nid - start_nid];
307 node_info_from_raw_nat(ni, &ne);
308 f2fs_put_page(page, 1);
309 cache:
310 /* cache nat entry */
311 cache_nat_entry(NM_I(sbi), nid, &ne);
312 }
313
314 /*
315 * The maximum depth is four.
316 * Offset[0] will have raw inode offset.
317 */
318 static int get_node_path(long block, int offset[4], unsigned int noffset[4])
319 {
320 const long direct_index = ADDRS_PER_INODE;
321 const long direct_blks = ADDRS_PER_BLOCK;
322 const long dptrs_per_blk = NIDS_PER_BLOCK;
323 const long indirect_blks = ADDRS_PER_BLOCK * NIDS_PER_BLOCK;
324 const long dindirect_blks = indirect_blks * NIDS_PER_BLOCK;
325 int n = 0;
326 int level = 0;
327
328 noffset[0] = 0;
329
330 if (block < direct_index) {
331 offset[n] = block;
332 goto got;
333 }
334 block -= direct_index;
335 if (block < direct_blks) {
336 offset[n++] = NODE_DIR1_BLOCK;
337 noffset[n] = 1;
338 offset[n] = block;
339 level = 1;
340 goto got;
341 }
342 block -= direct_blks;
343 if (block < direct_blks) {
344 offset[n++] = NODE_DIR2_BLOCK;
345 noffset[n] = 2;
346 offset[n] = block;
347 level = 1;
348 goto got;
349 }
350 block -= direct_blks;
351 if (block < indirect_blks) {
352 offset[n++] = NODE_IND1_BLOCK;
353 noffset[n] = 3;
354 offset[n++] = block / direct_blks;
355 noffset[n] = 4 + offset[n - 1];
356 offset[n] = block % direct_blks;
357 level = 2;
358 goto got;
359 }
360 block -= indirect_blks;
361 if (block < indirect_blks) {
362 offset[n++] = NODE_IND2_BLOCK;
363 noffset[n] = 4 + dptrs_per_blk;
364 offset[n++] = block / direct_blks;
365 noffset[n] = 5 + dptrs_per_blk + offset[n - 1];
366 offset[n] = block % direct_blks;
367 level = 2;
368 goto got;
369 }
370 block -= indirect_blks;
371 if (block < dindirect_blks) {
372 offset[n++] = NODE_DIND_BLOCK;
373 noffset[n] = 5 + (dptrs_per_blk * 2);
374 offset[n++] = block / indirect_blks;
375 noffset[n] = 6 + (dptrs_per_blk * 2) +
376 offset[n - 1] * (dptrs_per_blk + 1);
377 offset[n++] = (block / direct_blks) % dptrs_per_blk;
378 noffset[n] = 7 + (dptrs_per_blk * 2) +
379 offset[n - 2] * (dptrs_per_blk + 1) +
380 offset[n - 1];
381 offset[n] = block % direct_blks;
382 level = 3;
383 goto got;
384 } else {
385 BUG();
386 }
387 got:
388 return level;
389 }
390
391 /*
392 * Caller should call f2fs_put_dnode(dn).
393 * Also, it should grab and release a mutex by calling mutex_lock_op() and
394 * mutex_unlock_op() only if ro is not set RDONLY_NODE.
395 * In the case of RDONLY_NODE, we don't need to care about mutex.
396 */
397 int get_dnode_of_data(struct dnode_of_data *dn, pgoff_t index, int mode)
398 {
399 struct f2fs_sb_info *sbi = F2FS_SB(dn->inode->i_sb);
400 struct page *npage[4];
401 struct page *parent;
402 int offset[4];
403 unsigned int noffset[4];
404 nid_t nids[4];
405 int level, i;
406 int err = 0;
407
408 level = get_node_path(index, offset, noffset);
409
410 nids[0] = dn->inode->i_ino;
411 npage[0] = dn->inode_page;
412
413 if (!npage[0]) {
414 npage[0] = get_node_page(sbi, nids[0]);
415 if (IS_ERR(npage[0]))
416 return PTR_ERR(npage[0]);
417 }
418 parent = npage[0];
419 if (level != 0)
420 nids[1] = get_nid(parent, offset[0], true);
421 dn->inode_page = npage[0];
422 dn->inode_page_locked = true;
423
424 /* get indirect or direct nodes */
425 for (i = 1; i <= level; i++) {
426 bool done = false;
427
428 if (!nids[i] && mode == ALLOC_NODE) {
429 /* alloc new node */
430 if (!alloc_nid(sbi, &(nids[i]))) {
431 err = -ENOSPC;
432 goto release_pages;
433 }
434
435 dn->nid = nids[i];
436 npage[i] = new_node_page(dn, noffset[i], NULL);
437 if (IS_ERR(npage[i])) {
438 alloc_nid_failed(sbi, nids[i]);
439 err = PTR_ERR(npage[i]);
440 goto release_pages;
441 }
442
443 set_nid(parent, offset[i - 1], nids[i], i == 1);
444 alloc_nid_done(sbi, nids[i]);
445 done = true;
446 } else if (mode == LOOKUP_NODE_RA && i == level && level > 1) {
447 npage[i] = get_node_page_ra(parent, offset[i - 1]);
448 if (IS_ERR(npage[i])) {
449 err = PTR_ERR(npage[i]);
450 goto release_pages;
451 }
452 done = true;
453 }
454 if (i == 1) {
455 dn->inode_page_locked = false;
456 unlock_page(parent);
457 } else {
458 f2fs_put_page(parent, 1);
459 }
460
461 if (!done) {
462 npage[i] = get_node_page(sbi, nids[i]);
463 if (IS_ERR(npage[i])) {
464 err = PTR_ERR(npage[i]);
465 f2fs_put_page(npage[0], 0);
466 goto release_out;
467 }
468 }
469 if (i < level) {
470 parent = npage[i];
471 nids[i + 1] = get_nid(parent, offset[i], false);
472 }
473 }
474 dn->nid = nids[level];
475 dn->ofs_in_node = offset[level];
476 dn->node_page = npage[level];
477 dn->data_blkaddr = datablock_addr(dn->node_page, dn->ofs_in_node);
478 return 0;
479
480 release_pages:
481 f2fs_put_page(parent, 1);
482 if (i > 1)
483 f2fs_put_page(npage[0], 0);
484 release_out:
485 dn->inode_page = NULL;
486 dn->node_page = NULL;
487 return err;
488 }
489
490 static void truncate_node(struct dnode_of_data *dn)
491 {
492 struct f2fs_sb_info *sbi = F2FS_SB(dn->inode->i_sb);
493 struct node_info ni;
494
495 get_node_info(sbi, dn->nid, &ni);
496 if (dn->inode->i_blocks == 0) {
497 BUG_ON(ni.blk_addr != NULL_ADDR);
498 goto invalidate;
499 }
500 BUG_ON(ni.blk_addr == NULL_ADDR);
501
502 /* Deallocate node address */
503 invalidate_blocks(sbi, ni.blk_addr);
504 dec_valid_node_count(sbi, dn->inode, 1);
505 set_node_addr(sbi, &ni, NULL_ADDR);
506
507 if (dn->nid == dn->inode->i_ino) {
508 remove_orphan_inode(sbi, dn->nid);
509 dec_valid_inode_count(sbi);
510 } else {
511 sync_inode_page(dn);
512 }
513 invalidate:
514 clear_node_page_dirty(dn->node_page);
515 F2FS_SET_SB_DIRT(sbi);
516
517 f2fs_put_page(dn->node_page, 1);
518 dn->node_page = NULL;
519 trace_f2fs_truncate_node(dn->inode, dn->nid, ni.blk_addr);
520 }
521
522 static int truncate_dnode(struct dnode_of_data *dn)
523 {
524 struct f2fs_sb_info *sbi = F2FS_SB(dn->inode->i_sb);
525 struct page *page;
526
527 if (dn->nid == 0)
528 return 1;
529
530 /* get direct node */
531 page = get_node_page(sbi, dn->nid);
532 if (IS_ERR(page) && PTR_ERR(page) == -ENOENT)
533 return 1;
534 else if (IS_ERR(page))
535 return PTR_ERR(page);
536
537 /* Make dnode_of_data for parameter */
538 dn->node_page = page;
539 dn->ofs_in_node = 0;
540 truncate_data_blocks(dn);
541 truncate_node(dn);
542 return 1;
543 }
544
545 static int truncate_nodes(struct dnode_of_data *dn, unsigned int nofs,
546 int ofs, int depth)
547 {
548 struct f2fs_sb_info *sbi = F2FS_SB(dn->inode->i_sb);
549 struct dnode_of_data rdn = *dn;
550 struct page *page;
551 struct f2fs_node *rn;
552 nid_t child_nid;
553 unsigned int child_nofs;
554 int freed = 0;
555 int i, ret;
556
557 if (dn->nid == 0)
558 return NIDS_PER_BLOCK + 1;
559
560 trace_f2fs_truncate_nodes_enter(dn->inode, dn->nid, dn->data_blkaddr);
561
562 page = get_node_page(sbi, dn->nid);
563 if (IS_ERR(page)) {
564 trace_f2fs_truncate_nodes_exit(dn->inode, PTR_ERR(page));
565 return PTR_ERR(page);
566 }
567
568 rn = (struct f2fs_node *)page_address(page);
569 if (depth < 3) {
570 for (i = ofs; i < NIDS_PER_BLOCK; i++, freed++) {
571 child_nid = le32_to_cpu(rn->in.nid[i]);
572 if (child_nid == 0)
573 continue;
574 rdn.nid = child_nid;
575 ret = truncate_dnode(&rdn);
576 if (ret < 0)
577 goto out_err;
578 set_nid(page, i, 0, false);
579 }
580 } else {
581 child_nofs = nofs + ofs * (NIDS_PER_BLOCK + 1) + 1;
582 for (i = ofs; i < NIDS_PER_BLOCK; i++) {
583 child_nid = le32_to_cpu(rn->in.nid[i]);
584 if (child_nid == 0) {
585 child_nofs += NIDS_PER_BLOCK + 1;
586 continue;
587 }
588 rdn.nid = child_nid;
589 ret = truncate_nodes(&rdn, child_nofs, 0, depth - 1);
590 if (ret == (NIDS_PER_BLOCK + 1)) {
591 set_nid(page, i, 0, false);
592 child_nofs += ret;
593 } else if (ret < 0 && ret != -ENOENT) {
594 goto out_err;
595 }
596 }
597 freed = child_nofs;
598 }
599
600 if (!ofs) {
601 /* remove current indirect node */
602 dn->node_page = page;
603 truncate_node(dn);
604 freed++;
605 } else {
606 f2fs_put_page(page, 1);
607 }
608 trace_f2fs_truncate_nodes_exit(dn->inode, freed);
609 return freed;
610
611 out_err:
612 f2fs_put_page(page, 1);
613 trace_f2fs_truncate_nodes_exit(dn->inode, ret);
614 return ret;
615 }
616
617 static int truncate_partial_nodes(struct dnode_of_data *dn,
618 struct f2fs_inode *ri, int *offset, int depth)
619 {
620 struct f2fs_sb_info *sbi = F2FS_SB(dn->inode->i_sb);
621 struct page *pages[2];
622 nid_t nid[3];
623 nid_t child_nid;
624 int err = 0;
625 int i;
626 int idx = depth - 2;
627
628 nid[0] = le32_to_cpu(ri->i_nid[offset[0] - NODE_DIR1_BLOCK]);
629 if (!nid[0])
630 return 0;
631
632 /* get indirect nodes in the path */
633 for (i = 0; i < depth - 1; i++) {
634 /* refernece count'll be increased */
635 pages[i] = get_node_page(sbi, nid[i]);
636 if (IS_ERR(pages[i])) {
637 depth = i + 1;
638 err = PTR_ERR(pages[i]);
639 goto fail;
640 }
641 nid[i + 1] = get_nid(pages[i], offset[i + 1], false);
642 }
643
644 /* free direct nodes linked to a partial indirect node */
645 for (i = offset[depth - 1]; i < NIDS_PER_BLOCK; i++) {
646 child_nid = get_nid(pages[idx], i, false);
647 if (!child_nid)
648 continue;
649 dn->nid = child_nid;
650 err = truncate_dnode(dn);
651 if (err < 0)
652 goto fail;
653 set_nid(pages[idx], i, 0, false);
654 }
655
656 if (offset[depth - 1] == 0) {
657 dn->node_page = pages[idx];
658 dn->nid = nid[idx];
659 truncate_node(dn);
660 } else {
661 f2fs_put_page(pages[idx], 1);
662 }
663 offset[idx]++;
664 offset[depth - 1] = 0;
665 fail:
666 for (i = depth - 3; i >= 0; i--)
667 f2fs_put_page(pages[i], 1);
668
669 trace_f2fs_truncate_partial_nodes(dn->inode, nid, depth, err);
670
671 return err;
672 }
673
674 /*
675 * All the block addresses of data and nodes should be nullified.
676 */
677 int truncate_inode_blocks(struct inode *inode, pgoff_t from)
678 {
679 struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
680 struct address_space *node_mapping = sbi->node_inode->i_mapping;
681 int err = 0, cont = 1;
682 int level, offset[4], noffset[4];
683 unsigned int nofs = 0;
684 struct f2fs_node *rn;
685 struct dnode_of_data dn;
686 struct page *page;
687
688 trace_f2fs_truncate_inode_blocks_enter(inode, from);
689
690 level = get_node_path(from, offset, noffset);
691 restart:
692 page = get_node_page(sbi, inode->i_ino);
693 if (IS_ERR(page)) {
694 trace_f2fs_truncate_inode_blocks_exit(inode, PTR_ERR(page));
695 return PTR_ERR(page);
696 }
697
698 set_new_dnode(&dn, inode, page, NULL, 0);
699 unlock_page(page);
700
701 rn = page_address(page);
702 switch (level) {
703 case 0:
704 case 1:
705 nofs = noffset[1];
706 break;
707 case 2:
708 nofs = noffset[1];
709 if (!offset[level - 1])
710 goto skip_partial;
711 err = truncate_partial_nodes(&dn, &rn->i, offset, level);
712 if (err < 0 && err != -ENOENT)
713 goto fail;
714 nofs += 1 + NIDS_PER_BLOCK;
715 break;
716 case 3:
717 nofs = 5 + 2 * NIDS_PER_BLOCK;
718 if (!offset[level - 1])
719 goto skip_partial;
720 err = truncate_partial_nodes(&dn, &rn->i, offset, level);
721 if (err < 0 && err != -ENOENT)
722 goto fail;
723 break;
724 default:
725 BUG();
726 }
727
728 skip_partial:
729 while (cont) {
730 dn.nid = le32_to_cpu(rn->i.i_nid[offset[0] - NODE_DIR1_BLOCK]);
731 switch (offset[0]) {
732 case NODE_DIR1_BLOCK:
733 case NODE_DIR2_BLOCK:
734 err = truncate_dnode(&dn);
735 break;
736
737 case NODE_IND1_BLOCK:
738 case NODE_IND2_BLOCK:
739 err = truncate_nodes(&dn, nofs, offset[1], 2);
740 break;
741
742 case NODE_DIND_BLOCK:
743 err = truncate_nodes(&dn, nofs, offset[1], 3);
744 cont = 0;
745 break;
746
747 default:
748 BUG();
749 }
750 if (err < 0 && err != -ENOENT)
751 goto fail;
752 if (offset[1] == 0 &&
753 rn->i.i_nid[offset[0] - NODE_DIR1_BLOCK]) {
754 lock_page(page);
755 if (page->mapping != node_mapping) {
756 f2fs_put_page(page, 1);
757 goto restart;
758 }
759 wait_on_page_writeback(page);
760 rn->i.i_nid[offset[0] - NODE_DIR1_BLOCK] = 0;
761 set_page_dirty(page);
762 unlock_page(page);
763 }
764 offset[1] = 0;
765 offset[0]++;
766 nofs += err;
767 }
768 fail:
769 f2fs_put_page(page, 0);
770 trace_f2fs_truncate_inode_blocks_exit(inode, err);
771 return err > 0 ? 0 : err;
772 }
773
774 /*
775 * Caller should grab and release a mutex by calling mutex_lock_op() and
776 * mutex_unlock_op().
777 */
778 int remove_inode_page(struct inode *inode)
779 {
780 struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
781 struct page *page;
782 nid_t ino = inode->i_ino;
783 struct dnode_of_data dn;
784
785 page = get_node_page(sbi, ino);
786 if (IS_ERR(page))
787 return PTR_ERR(page);
788
789 if (F2FS_I(inode)->i_xattr_nid) {
790 nid_t nid = F2FS_I(inode)->i_xattr_nid;
791 struct page *npage = get_node_page(sbi, nid);
792
793 if (IS_ERR(npage))
794 return PTR_ERR(npage);
795
796 F2FS_I(inode)->i_xattr_nid = 0;
797 set_new_dnode(&dn, inode, page, npage, nid);
798 dn.inode_page_locked = 1;
799 truncate_node(&dn);
800 }
801
802 /* 0 is possible, after f2fs_new_inode() is failed */
803 BUG_ON(inode->i_blocks != 0 && inode->i_blocks != 1);
804 set_new_dnode(&dn, inode, page, page, ino);
805 truncate_node(&dn);
806 return 0;
807 }
808
809 struct page *new_inode_page(struct inode *inode, const struct qstr *name)
810 {
811 struct dnode_of_data dn;
812
813 /* allocate inode page for new inode */
814 set_new_dnode(&dn, inode, NULL, NULL, inode->i_ino);
815
816 /* caller should f2fs_put_page(page, 1); */
817 return new_node_page(&dn, 0, NULL);
818 }
819
820 struct page *new_node_page(struct dnode_of_data *dn,
821 unsigned int ofs, struct page *ipage)
822 {
823 struct f2fs_sb_info *sbi = F2FS_SB(dn->inode->i_sb);
824 struct address_space *mapping = sbi->node_inode->i_mapping;
825 struct node_info old_ni, new_ni;
826 struct page *page;
827 int err;
828
829 if (is_inode_flag_set(F2FS_I(dn->inode), FI_NO_ALLOC))
830 return ERR_PTR(-EPERM);
831
832 page = grab_cache_page(mapping, dn->nid);
833 if (!page)
834 return ERR_PTR(-ENOMEM);
835
836 get_node_info(sbi, dn->nid, &old_ni);
837
838 SetPageUptodate(page);
839 fill_node_footer(page, dn->nid, dn->inode->i_ino, ofs, true);
840
841 /* Reinitialize old_ni with new node page */
842 BUG_ON(old_ni.blk_addr != NULL_ADDR);
843 new_ni = old_ni;
844 new_ni.ino = dn->inode->i_ino;
845
846 if (!inc_valid_node_count(sbi, dn->inode, 1)) {
847 err = -ENOSPC;
848 goto fail;
849 }
850 set_node_addr(sbi, &new_ni, NEW_ADDR);
851 set_cold_node(dn->inode, page);
852
853 dn->node_page = page;
854 if (ipage)
855 update_inode(dn->inode, ipage);
856 else
857 sync_inode_page(dn);
858 set_page_dirty(page);
859 if (ofs == 0)
860 inc_valid_inode_count(sbi);
861
862 return page;
863
864 fail:
865 clear_node_page_dirty(page);
866 f2fs_put_page(page, 1);
867 return ERR_PTR(err);
868 }
869
870 /*
871 * Caller should do after getting the following values.
872 * 0: f2fs_put_page(page, 0)
873 * LOCKED_PAGE: f2fs_put_page(page, 1)
874 * error: nothing
875 */
876 static int read_node_page(struct page *page, int type)
877 {
878 struct f2fs_sb_info *sbi = F2FS_SB(page->mapping->host->i_sb);
879 struct node_info ni;
880
881 get_node_info(sbi, page->index, &ni);
882
883 if (ni.blk_addr == NULL_ADDR) {
884 f2fs_put_page(page, 1);
885 return -ENOENT;
886 }
887
888 if (PageUptodate(page))
889 return LOCKED_PAGE;
890
891 return f2fs_readpage(sbi, page, ni.blk_addr, type);
892 }
893
894 /*
895 * Readahead a node page
896 */
897 void ra_node_page(struct f2fs_sb_info *sbi, nid_t nid)
898 {
899 struct address_space *mapping = sbi->node_inode->i_mapping;
900 struct page *apage;
901 int err;
902
903 apage = find_get_page(mapping, nid);
904 if (apage && PageUptodate(apage)) {
905 f2fs_put_page(apage, 0);
906 return;
907 }
908 f2fs_put_page(apage, 0);
909
910 apage = grab_cache_page(mapping, nid);
911 if (!apage)
912 return;
913
914 err = read_node_page(apage, READA);
915 if (err == 0)
916 f2fs_put_page(apage, 0);
917 else if (err == LOCKED_PAGE)
918 f2fs_put_page(apage, 1);
919 return;
920 }
921
922 struct page *get_node_page(struct f2fs_sb_info *sbi, pgoff_t nid)
923 {
924 struct address_space *mapping = sbi->node_inode->i_mapping;
925 struct page *page;
926 int err;
927 repeat:
928 page = grab_cache_page(mapping, nid);
929 if (!page)
930 return ERR_PTR(-ENOMEM);
931
932 err = read_node_page(page, READ_SYNC);
933 if (err < 0)
934 return ERR_PTR(err);
935 else if (err == LOCKED_PAGE)
936 goto got_it;
937
938 lock_page(page);
939 if (!PageUptodate(page)) {
940 f2fs_put_page(page, 1);
941 return ERR_PTR(-EIO);
942 }
943 if (page->mapping != mapping) {
944 f2fs_put_page(page, 1);
945 goto repeat;
946 }
947 got_it:
948 BUG_ON(nid != nid_of_node(page));
949 mark_page_accessed(page);
950 return page;
951 }
952
953 /*
954 * Return a locked page for the desired node page.
955 * And, readahead MAX_RA_NODE number of node pages.
956 */
957 struct page *get_node_page_ra(struct page *parent, int start)
958 {
959 struct f2fs_sb_info *sbi = F2FS_SB(parent->mapping->host->i_sb);
960 struct address_space *mapping = sbi->node_inode->i_mapping;
961 struct blk_plug plug;
962 struct page *page;
963 int err, i, end;
964 nid_t nid;
965
966 /* First, try getting the desired direct node. */
967 nid = get_nid(parent, start, false);
968 if (!nid)
969 return ERR_PTR(-ENOENT);
970 repeat:
971 page = grab_cache_page(mapping, nid);
972 if (!page)
973 return ERR_PTR(-ENOMEM);
974
975 err = read_node_page(page, READ_SYNC);
976 if (err < 0)
977 return ERR_PTR(err);
978 else if (err == LOCKED_PAGE)
979 goto page_hit;
980
981 blk_start_plug(&plug);
982
983 /* Then, try readahead for siblings of the desired node */
984 end = start + MAX_RA_NODE;
985 end = min(end, NIDS_PER_BLOCK);
986 for (i = start + 1; i < end; i++) {
987 nid = get_nid(parent, i, false);
988 if (!nid)
989 continue;
990 ra_node_page(sbi, nid);
991 }
992
993 blk_finish_plug(&plug);
994
995 lock_page(page);
996 if (page->mapping != mapping) {
997 f2fs_put_page(page, 1);
998 goto repeat;
999 }
1000 page_hit:
1001 if (!PageUptodate(page)) {
1002 f2fs_put_page(page, 1);
1003 return ERR_PTR(-EIO);
1004 }
1005 mark_page_accessed(page);
1006 return page;
1007 }
1008
1009 void sync_inode_page(struct dnode_of_data *dn)
1010 {
1011 if (IS_INODE(dn->node_page) || dn->inode_page == dn->node_page) {
1012 update_inode(dn->inode, dn->node_page);
1013 } else if (dn->inode_page) {
1014 if (!dn->inode_page_locked)
1015 lock_page(dn->inode_page);
1016 update_inode(dn->inode, dn->inode_page);
1017 if (!dn->inode_page_locked)
1018 unlock_page(dn->inode_page);
1019 } else {
1020 update_inode_page(dn->inode);
1021 }
1022 }
1023
1024 int sync_node_pages(struct f2fs_sb_info *sbi, nid_t ino,
1025 struct writeback_control *wbc)
1026 {
1027 struct address_space *mapping = sbi->node_inode->i_mapping;
1028 pgoff_t index, end;
1029 struct pagevec pvec;
1030 int step = ino ? 2 : 0;
1031 int nwritten = 0, wrote = 0;
1032
1033 pagevec_init(&pvec, 0);
1034
1035 next_step:
1036 index = 0;
1037 end = LONG_MAX;
1038
1039 while (index <= end) {
1040 int i, nr_pages;
1041 nr_pages = pagevec_lookup_tag(&pvec, mapping, &index,
1042 PAGECACHE_TAG_DIRTY,
1043 min(end - index, (pgoff_t)PAGEVEC_SIZE-1) + 1);
1044 if (nr_pages == 0)
1045 break;
1046
1047 for (i = 0; i < nr_pages; i++) {
1048 struct page *page = pvec.pages[i];
1049
1050 /*
1051 * flushing sequence with step:
1052 * 0. indirect nodes
1053 * 1. dentry dnodes
1054 * 2. file dnodes
1055 */
1056 if (step == 0 && IS_DNODE(page))
1057 continue;
1058 if (step == 1 && (!IS_DNODE(page) ||
1059 is_cold_node(page)))
1060 continue;
1061 if (step == 2 && (!IS_DNODE(page) ||
1062 !is_cold_node(page)))
1063 continue;
1064
1065 /*
1066 * If an fsync mode,
1067 * we should not skip writing node pages.
1068 */
1069 if (ino && ino_of_node(page) == ino)
1070 lock_page(page);
1071 else if (!trylock_page(page))
1072 continue;
1073
1074 if (unlikely(page->mapping != mapping)) {
1075 continue_unlock:
1076 unlock_page(page);
1077 continue;
1078 }
1079 if (ino && ino_of_node(page) != ino)
1080 goto continue_unlock;
1081
1082 if (!PageDirty(page)) {
1083 /* someone wrote it for us */
1084 goto continue_unlock;
1085 }
1086
1087 if (!clear_page_dirty_for_io(page))
1088 goto continue_unlock;
1089
1090 /* called by fsync() */
1091 if (ino && IS_DNODE(page)) {
1092 int mark = !is_checkpointed_node(sbi, ino);
1093 set_fsync_mark(page, 1);
1094 if (IS_INODE(page))
1095 set_dentry_mark(page, mark);
1096 nwritten++;
1097 } else {
1098 set_fsync_mark(page, 0);
1099 set_dentry_mark(page, 0);
1100 }
1101 mapping->a_ops->writepage(page, wbc);
1102 wrote++;
1103
1104 if (--wbc->nr_to_write == 0)
1105 break;
1106 }
1107 pagevec_release(&pvec);
1108 cond_resched();
1109
1110 if (wbc->nr_to_write == 0) {
1111 step = 2;
1112 break;
1113 }
1114 }
1115
1116 if (step < 2) {
1117 step++;
1118 goto next_step;
1119 }
1120
1121 if (wrote)
1122 f2fs_submit_bio(sbi, NODE, wbc->sync_mode == WB_SYNC_ALL);
1123
1124 return nwritten;
1125 }
1126
1127 static int f2fs_write_node_page(struct page *page,
1128 struct writeback_control *wbc)
1129 {
1130 struct f2fs_sb_info *sbi = F2FS_SB(page->mapping->host->i_sb);
1131 nid_t nid;
1132 block_t new_addr;
1133 struct node_info ni;
1134
1135 wait_on_page_writeback(page);
1136
1137 /* get old block addr of this node page */
1138 nid = nid_of_node(page);
1139 BUG_ON(page->index != nid);
1140
1141 get_node_info(sbi, nid, &ni);
1142
1143 /* This page is already truncated */
1144 if (ni.blk_addr == NULL_ADDR) {
1145 dec_page_count(sbi, F2FS_DIRTY_NODES);
1146 unlock_page(page);
1147 return 0;
1148 }
1149
1150 if (wbc->for_reclaim) {
1151 dec_page_count(sbi, F2FS_DIRTY_NODES);
1152 wbc->pages_skipped++;
1153 set_page_dirty(page);
1154 return AOP_WRITEPAGE_ACTIVATE;
1155 }
1156
1157 mutex_lock(&sbi->node_write);
1158 set_page_writeback(page);
1159 write_node_page(sbi, page, nid, ni.blk_addr, &new_addr);
1160 set_node_addr(sbi, &ni, new_addr);
1161 dec_page_count(sbi, F2FS_DIRTY_NODES);
1162 mutex_unlock(&sbi->node_write);
1163 unlock_page(page);
1164 return 0;
1165 }
1166
1167 /*
1168 * It is very important to gather dirty pages and write at once, so that we can
1169 * submit a big bio without interfering other data writes.
1170 * Be default, 512 pages (2MB), a segment size, is quite reasonable.
1171 */
1172 #define COLLECT_DIRTY_NODES 512
1173 static int f2fs_write_node_pages(struct address_space *mapping,
1174 struct writeback_control *wbc)
1175 {
1176 struct f2fs_sb_info *sbi = F2FS_SB(mapping->host->i_sb);
1177 long nr_to_write = wbc->nr_to_write;
1178
1179 /* First check balancing cached NAT entries */
1180 if (try_to_free_nats(sbi, NAT_ENTRY_PER_BLOCK)) {
1181 f2fs_sync_fs(sbi->sb, true);
1182 return 0;
1183 }
1184
1185 /* collect a number of dirty node pages and write together */
1186 if (get_pages(sbi, F2FS_DIRTY_NODES) < COLLECT_DIRTY_NODES)
1187 return 0;
1188
1189 /* if mounting is failed, skip writing node pages */
1190 wbc->nr_to_write = max_hw_blocks(sbi);
1191 sync_node_pages(sbi, 0, wbc);
1192 wbc->nr_to_write = nr_to_write - (max_hw_blocks(sbi) - wbc->nr_to_write);
1193 return 0;
1194 }
1195
1196 static int f2fs_set_node_page_dirty(struct page *page)
1197 {
1198 struct address_space *mapping = page->mapping;
1199 struct f2fs_sb_info *sbi = F2FS_SB(mapping->host->i_sb);
1200
1201 SetPageUptodate(page);
1202 if (!PageDirty(page)) {
1203 __set_page_dirty_nobuffers(page);
1204 inc_page_count(sbi, F2FS_DIRTY_NODES);
1205 SetPagePrivate(page);
1206 return 1;
1207 }
1208 return 0;
1209 }
1210
1211 static void f2fs_invalidate_node_page(struct page *page, unsigned int offset,
1212 unsigned int length)
1213 {
1214 struct inode *inode = page->mapping->host;
1215 struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
1216 if (PageDirty(page))
1217 dec_page_count(sbi, F2FS_DIRTY_NODES);
1218 ClearPagePrivate(page);
1219 }
1220
1221 static int f2fs_release_node_page(struct page *page, gfp_t wait)
1222 {
1223 ClearPagePrivate(page);
1224 return 1;
1225 }
1226
1227 /*
1228 * Structure of the f2fs node operations
1229 */
1230 const struct address_space_operations f2fs_node_aops = {
1231 .writepage = f2fs_write_node_page,
1232 .writepages = f2fs_write_node_pages,
1233 .set_page_dirty = f2fs_set_node_page_dirty,
1234 .invalidatepage = f2fs_invalidate_node_page,
1235 .releasepage = f2fs_release_node_page,
1236 };
1237
1238 static struct free_nid *__lookup_free_nid_list(nid_t n, struct list_head *head)
1239 {
1240 struct list_head *this;
1241 struct free_nid *i;
1242 list_for_each(this, head) {
1243 i = list_entry(this, struct free_nid, list);
1244 if (i->nid == n)
1245 return i;
1246 }
1247 return NULL;
1248 }
1249
1250 static void __del_from_free_nid_list(struct free_nid *i)
1251 {
1252 list_del(&i->list);
1253 kmem_cache_free(free_nid_slab, i);
1254 }
1255
1256 static int add_free_nid(struct f2fs_nm_info *nm_i, nid_t nid, bool build)
1257 {
1258 struct free_nid *i;
1259 struct nat_entry *ne;
1260 bool allocated = false;
1261
1262 if (nm_i->fcnt > 2 * MAX_FREE_NIDS)
1263 return -1;
1264
1265 /* 0 nid should not be used */
1266 if (nid == 0)
1267 return 0;
1268
1269 if (!build)
1270 goto retry;
1271
1272 /* do not add allocated nids */
1273 read_lock(&nm_i->nat_tree_lock);
1274 ne = __lookup_nat_cache(nm_i, nid);
1275 if (ne && nat_get_blkaddr(ne) != NULL_ADDR)
1276 allocated = true;
1277 read_unlock(&nm_i->nat_tree_lock);
1278 if (allocated)
1279 return 0;
1280 retry:
1281 i = kmem_cache_alloc(free_nid_slab, GFP_NOFS);
1282 if (!i) {
1283 cond_resched();
1284 goto retry;
1285 }
1286 i->nid = nid;
1287 i->state = NID_NEW;
1288
1289 spin_lock(&nm_i->free_nid_list_lock);
1290 if (__lookup_free_nid_list(nid, &nm_i->free_nid_list)) {
1291 spin_unlock(&nm_i->free_nid_list_lock);
1292 kmem_cache_free(free_nid_slab, i);
1293 return 0;
1294 }
1295 list_add_tail(&i->list, &nm_i->free_nid_list);
1296 nm_i->fcnt++;
1297 spin_unlock(&nm_i->free_nid_list_lock);
1298 return 1;
1299 }
1300
1301 static void remove_free_nid(struct f2fs_nm_info *nm_i, nid_t nid)
1302 {
1303 struct free_nid *i;
1304 spin_lock(&nm_i->free_nid_list_lock);
1305 i = __lookup_free_nid_list(nid, &nm_i->free_nid_list);
1306 if (i && i->state == NID_NEW) {
1307 __del_from_free_nid_list(i);
1308 nm_i->fcnt--;
1309 }
1310 spin_unlock(&nm_i->free_nid_list_lock);
1311 }
1312
1313 static void scan_nat_page(struct f2fs_nm_info *nm_i,
1314 struct page *nat_page, nid_t start_nid)
1315 {
1316 struct f2fs_nat_block *nat_blk = page_address(nat_page);
1317 block_t blk_addr;
1318 int i;
1319
1320 i = start_nid % NAT_ENTRY_PER_BLOCK;
1321
1322 for (; i < NAT_ENTRY_PER_BLOCK; i++, start_nid++) {
1323
1324 if (start_nid >= nm_i->max_nid)
1325 break;
1326
1327 blk_addr = le32_to_cpu(nat_blk->entries[i].block_addr);
1328 BUG_ON(blk_addr == NEW_ADDR);
1329 if (blk_addr == NULL_ADDR) {
1330 if (add_free_nid(nm_i, start_nid, true) < 0)
1331 break;
1332 }
1333 }
1334 }
1335
1336 static void build_free_nids(struct f2fs_sb_info *sbi)
1337 {
1338 struct f2fs_nm_info *nm_i = NM_I(sbi);
1339 struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
1340 struct f2fs_summary_block *sum = curseg->sum_blk;
1341 int i = 0;
1342 nid_t nid = nm_i->next_scan_nid;
1343
1344 /* Enough entries */
1345 if (nm_i->fcnt > NAT_ENTRY_PER_BLOCK)
1346 return;
1347
1348 /* readahead nat pages to be scanned */
1349 ra_nat_pages(sbi, nid);
1350
1351 while (1) {
1352 struct page *page = get_current_nat_page(sbi, nid);
1353
1354 scan_nat_page(nm_i, page, nid);
1355 f2fs_put_page(page, 1);
1356
1357 nid += (NAT_ENTRY_PER_BLOCK - (nid % NAT_ENTRY_PER_BLOCK));
1358 if (nid >= nm_i->max_nid)
1359 nid = 0;
1360
1361 if (i++ == FREE_NID_PAGES)
1362 break;
1363 }
1364
1365 /* go to the next free nat pages to find free nids abundantly */
1366 nm_i->next_scan_nid = nid;
1367
1368 /* find free nids from current sum_pages */
1369 mutex_lock(&curseg->curseg_mutex);
1370 for (i = 0; i < nats_in_cursum(sum); i++) {
1371 block_t addr = le32_to_cpu(nat_in_journal(sum, i).block_addr);
1372 nid = le32_to_cpu(nid_in_journal(sum, i));
1373 if (addr == NULL_ADDR)
1374 add_free_nid(nm_i, nid, true);
1375 else
1376 remove_free_nid(nm_i, nid);
1377 }
1378 mutex_unlock(&curseg->curseg_mutex);
1379 }
1380
1381 /*
1382 * If this function returns success, caller can obtain a new nid
1383 * from second parameter of this function.
1384 * The returned nid could be used ino as well as nid when inode is created.
1385 */
1386 bool alloc_nid(struct f2fs_sb_info *sbi, nid_t *nid)
1387 {
1388 struct f2fs_nm_info *nm_i = NM_I(sbi);
1389 struct free_nid *i = NULL;
1390 struct list_head *this;
1391 retry:
1392 if (sbi->total_valid_node_count + 1 >= nm_i->max_nid)
1393 return false;
1394
1395 spin_lock(&nm_i->free_nid_list_lock);
1396
1397 /* We should not use stale free nids created by build_free_nids */
1398 if (nm_i->fcnt && !sbi->on_build_free_nids) {
1399 BUG_ON(list_empty(&nm_i->free_nid_list));
1400 list_for_each(this, &nm_i->free_nid_list) {
1401 i = list_entry(this, struct free_nid, list);
1402 if (i->state == NID_NEW)
1403 break;
1404 }
1405
1406 BUG_ON(i->state != NID_NEW);
1407 *nid = i->nid;
1408 i->state = NID_ALLOC;
1409 nm_i->fcnt--;
1410 spin_unlock(&nm_i->free_nid_list_lock);
1411 return true;
1412 }
1413 spin_unlock(&nm_i->free_nid_list_lock);
1414
1415 /* Let's scan nat pages and its caches to get free nids */
1416 mutex_lock(&nm_i->build_lock);
1417 sbi->on_build_free_nids = 1;
1418 build_free_nids(sbi);
1419 sbi->on_build_free_nids = 0;
1420 mutex_unlock(&nm_i->build_lock);
1421 goto retry;
1422 }
1423
1424 /*
1425 * alloc_nid() should be called prior to this function.
1426 */
1427 void alloc_nid_done(struct f2fs_sb_info *sbi, nid_t nid)
1428 {
1429 struct f2fs_nm_info *nm_i = NM_I(sbi);
1430 struct free_nid *i;
1431
1432 spin_lock(&nm_i->free_nid_list_lock);
1433 i = __lookup_free_nid_list(nid, &nm_i->free_nid_list);
1434 BUG_ON(!i || i->state != NID_ALLOC);
1435 __del_from_free_nid_list(i);
1436 spin_unlock(&nm_i->free_nid_list_lock);
1437 }
1438
1439 /*
1440 * alloc_nid() should be called prior to this function.
1441 */
1442 void alloc_nid_failed(struct f2fs_sb_info *sbi, nid_t nid)
1443 {
1444 struct f2fs_nm_info *nm_i = NM_I(sbi);
1445 struct free_nid *i;
1446
1447 spin_lock(&nm_i->free_nid_list_lock);
1448 i = __lookup_free_nid_list(nid, &nm_i->free_nid_list);
1449 BUG_ON(!i || i->state != NID_ALLOC);
1450 if (nm_i->fcnt > 2 * MAX_FREE_NIDS) {
1451 __del_from_free_nid_list(i);
1452 } else {
1453 i->state = NID_NEW;
1454 nm_i->fcnt++;
1455 }
1456 spin_unlock(&nm_i->free_nid_list_lock);
1457 }
1458
1459 void recover_node_page(struct f2fs_sb_info *sbi, struct page *page,
1460 struct f2fs_summary *sum, struct node_info *ni,
1461 block_t new_blkaddr)
1462 {
1463 rewrite_node_page(sbi, page, sum, ni->blk_addr, new_blkaddr);
1464 set_node_addr(sbi, ni, new_blkaddr);
1465 clear_node_page_dirty(page);
1466 }
1467
1468 int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page)
1469 {
1470 struct address_space *mapping = sbi->node_inode->i_mapping;
1471 struct f2fs_node *src, *dst;
1472 nid_t ino = ino_of_node(page);
1473 struct node_info old_ni, new_ni;
1474 struct page *ipage;
1475
1476 ipage = grab_cache_page(mapping, ino);
1477 if (!ipage)
1478 return -ENOMEM;
1479
1480 /* Should not use this inode from free nid list */
1481 remove_free_nid(NM_I(sbi), ino);
1482
1483 get_node_info(sbi, ino, &old_ni);
1484 SetPageUptodate(ipage);
1485 fill_node_footer(ipage, ino, ino, 0, true);
1486
1487 src = (struct f2fs_node *)page_address(page);
1488 dst = (struct f2fs_node *)page_address(ipage);
1489
1490 memcpy(dst, src, (unsigned long)&src->i.i_ext - (unsigned long)&src->i);
1491 dst->i.i_size = 0;
1492 dst->i.i_blocks = cpu_to_le64(1);
1493 dst->i.i_links = cpu_to_le32(1);
1494 dst->i.i_xattr_nid = 0;
1495
1496 new_ni = old_ni;
1497 new_ni.ino = ino;
1498
1499 if (!inc_valid_node_count(sbi, NULL, 1))
1500 WARN_ON(1);
1501 set_node_addr(sbi, &new_ni, NEW_ADDR);
1502 inc_valid_inode_count(sbi);
1503 f2fs_put_page(ipage, 1);
1504 return 0;
1505 }
1506
1507 int restore_node_summary(struct f2fs_sb_info *sbi,
1508 unsigned int segno, struct f2fs_summary_block *sum)
1509 {
1510 struct f2fs_node *rn;
1511 struct f2fs_summary *sum_entry;
1512 struct page *page;
1513 block_t addr;
1514 int i, last_offset;
1515
1516 /* alloc temporal page for read node */
1517 page = alloc_page(GFP_NOFS | __GFP_ZERO);
1518 if (IS_ERR(page))
1519 return PTR_ERR(page);
1520 lock_page(page);
1521
1522 /* scan the node segment */
1523 last_offset = sbi->blocks_per_seg;
1524 addr = START_BLOCK(sbi, segno);
1525 sum_entry = &sum->entries[0];
1526
1527 for (i = 0; i < last_offset; i++, sum_entry++) {
1528 /*
1529 * In order to read next node page,
1530 * we must clear PageUptodate flag.
1531 */
1532 ClearPageUptodate(page);
1533
1534 if (f2fs_readpage(sbi, page, addr, READ_SYNC))
1535 goto out;
1536
1537 lock_page(page);
1538 rn = (struct f2fs_node *)page_address(page);
1539 sum_entry->nid = rn->footer.nid;
1540 sum_entry->version = 0;
1541 sum_entry->ofs_in_node = 0;
1542 addr++;
1543 }
1544 unlock_page(page);
1545 out:
1546 __free_pages(page, 0);
1547 return 0;
1548 }
1549
1550 static bool flush_nats_in_journal(struct f2fs_sb_info *sbi)
1551 {
1552 struct f2fs_nm_info *nm_i = NM_I(sbi);
1553 struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
1554 struct f2fs_summary_block *sum = curseg->sum_blk;
1555 int i;
1556
1557 mutex_lock(&curseg->curseg_mutex);
1558
1559 if (nats_in_cursum(sum) < NAT_JOURNAL_ENTRIES) {
1560 mutex_unlock(&curseg->curseg_mutex);
1561 return false;
1562 }
1563
1564 for (i = 0; i < nats_in_cursum(sum); i++) {
1565 struct nat_entry *ne;
1566 struct f2fs_nat_entry raw_ne;
1567 nid_t nid = le32_to_cpu(nid_in_journal(sum, i));
1568
1569 raw_ne = nat_in_journal(sum, i);
1570 retry:
1571 write_lock(&nm_i->nat_tree_lock);
1572 ne = __lookup_nat_cache(nm_i, nid);
1573 if (ne) {
1574 __set_nat_cache_dirty(nm_i, ne);
1575 write_unlock(&nm_i->nat_tree_lock);
1576 continue;
1577 }
1578 ne = grab_nat_entry(nm_i, nid);
1579 if (!ne) {
1580 write_unlock(&nm_i->nat_tree_lock);
1581 goto retry;
1582 }
1583 nat_set_blkaddr(ne, le32_to_cpu(raw_ne.block_addr));
1584 nat_set_ino(ne, le32_to_cpu(raw_ne.ino));
1585 nat_set_version(ne, raw_ne.version);
1586 __set_nat_cache_dirty(nm_i, ne);
1587 write_unlock(&nm_i->nat_tree_lock);
1588 }
1589 update_nats_in_cursum(sum, -i);
1590 mutex_unlock(&curseg->curseg_mutex);
1591 return true;
1592 }
1593
1594 /*
1595 * This function is called during the checkpointing process.
1596 */
1597 void flush_nat_entries(struct f2fs_sb_info *sbi)
1598 {
1599 struct f2fs_nm_info *nm_i = NM_I(sbi);
1600 struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
1601 struct f2fs_summary_block *sum = curseg->sum_blk;
1602 struct list_head *cur, *n;
1603 struct page *page = NULL;
1604 struct f2fs_nat_block *nat_blk = NULL;
1605 nid_t start_nid = 0, end_nid = 0;
1606 bool flushed;
1607
1608 flushed = flush_nats_in_journal(sbi);
1609
1610 if (!flushed)
1611 mutex_lock(&curseg->curseg_mutex);
1612
1613 /* 1) flush dirty nat caches */
1614 list_for_each_safe(cur, n, &nm_i->dirty_nat_entries) {
1615 struct nat_entry *ne;
1616 nid_t nid;
1617 struct f2fs_nat_entry raw_ne;
1618 int offset = -1;
1619 block_t new_blkaddr;
1620
1621 ne = list_entry(cur, struct nat_entry, list);
1622 nid = nat_get_nid(ne);
1623
1624 if (nat_get_blkaddr(ne) == NEW_ADDR)
1625 continue;
1626 if (flushed)
1627 goto to_nat_page;
1628
1629 /* if there is room for nat enries in curseg->sumpage */
1630 offset = lookup_journal_in_cursum(sum, NAT_JOURNAL, nid, 1);
1631 if (offset >= 0) {
1632 raw_ne = nat_in_journal(sum, offset);
1633 goto flush_now;
1634 }
1635 to_nat_page:
1636 if (!page || (start_nid > nid || nid > end_nid)) {
1637 if (page) {
1638 f2fs_put_page(page, 1);
1639 page = NULL;
1640 }
1641 start_nid = START_NID(nid);
1642 end_nid = start_nid + NAT_ENTRY_PER_BLOCK - 1;
1643
1644 /*
1645 * get nat block with dirty flag, increased reference
1646 * count, mapped and lock
1647 */
1648 page = get_next_nat_page(sbi, start_nid);
1649 nat_blk = page_address(page);
1650 }
1651
1652 BUG_ON(!nat_blk);
1653 raw_ne = nat_blk->entries[nid - start_nid];
1654 flush_now:
1655 new_blkaddr = nat_get_blkaddr(ne);
1656
1657 raw_ne.ino = cpu_to_le32(nat_get_ino(ne));
1658 raw_ne.block_addr = cpu_to_le32(new_blkaddr);
1659 raw_ne.version = nat_get_version(ne);
1660
1661 if (offset < 0) {
1662 nat_blk->entries[nid - start_nid] = raw_ne;
1663 } else {
1664 nat_in_journal(sum, offset) = raw_ne;
1665 nid_in_journal(sum, offset) = cpu_to_le32(nid);
1666 }
1667
1668 if (nat_get_blkaddr(ne) == NULL_ADDR &&
1669 add_free_nid(NM_I(sbi), nid, false) <= 0) {
1670 write_lock(&nm_i->nat_tree_lock);
1671 __del_from_nat_cache(nm_i, ne);
1672 write_unlock(&nm_i->nat_tree_lock);
1673 } else {
1674 write_lock(&nm_i->nat_tree_lock);
1675 __clear_nat_cache_dirty(nm_i, ne);
1676 ne->checkpointed = true;
1677 write_unlock(&nm_i->nat_tree_lock);
1678 }
1679 }
1680 if (!flushed)
1681 mutex_unlock(&curseg->curseg_mutex);
1682 f2fs_put_page(page, 1);
1683
1684 /* 2) shrink nat caches if necessary */
1685 try_to_free_nats(sbi, nm_i->nat_cnt - NM_WOUT_THRESHOLD);
1686 }
1687
1688 static int init_node_manager(struct f2fs_sb_info *sbi)
1689 {
1690 struct f2fs_super_block *sb_raw = F2FS_RAW_SUPER(sbi);
1691 struct f2fs_nm_info *nm_i = NM_I(sbi);
1692 unsigned char *version_bitmap;
1693 unsigned int nat_segs, nat_blocks;
1694
1695 nm_i->nat_blkaddr = le32_to_cpu(sb_raw->nat_blkaddr);
1696
1697 /* segment_count_nat includes pair segment so divide to 2. */
1698 nat_segs = le32_to_cpu(sb_raw->segment_count_nat) >> 1;
1699 nat_blocks = nat_segs << le32_to_cpu(sb_raw->log_blocks_per_seg);
1700 nm_i->max_nid = NAT_ENTRY_PER_BLOCK * nat_blocks;
1701 nm_i->fcnt = 0;
1702 nm_i->nat_cnt = 0;
1703
1704 INIT_LIST_HEAD(&nm_i->free_nid_list);
1705 INIT_RADIX_TREE(&nm_i->nat_root, GFP_ATOMIC);
1706 INIT_LIST_HEAD(&nm_i->nat_entries);
1707 INIT_LIST_HEAD(&nm_i->dirty_nat_entries);
1708
1709 mutex_init(&nm_i->build_lock);
1710 spin_lock_init(&nm_i->free_nid_list_lock);
1711 rwlock_init(&nm_i->nat_tree_lock);
1712
1713 nm_i->next_scan_nid = le32_to_cpu(sbi->ckpt->next_free_nid);
1714 nm_i->bitmap_size = __bitmap_size(sbi, NAT_BITMAP);
1715 version_bitmap = __bitmap_ptr(sbi, NAT_BITMAP);
1716 if (!version_bitmap)
1717 return -EFAULT;
1718
1719 nm_i->nat_bitmap = kmemdup(version_bitmap, nm_i->bitmap_size,
1720 GFP_KERNEL);
1721 if (!nm_i->nat_bitmap)
1722 return -ENOMEM;
1723 return 0;
1724 }
1725
1726 int build_node_manager(struct f2fs_sb_info *sbi)
1727 {
1728 int err;
1729
1730 sbi->nm_info = kzalloc(sizeof(struct f2fs_nm_info), GFP_KERNEL);
1731 if (!sbi->nm_info)
1732 return -ENOMEM;
1733
1734 err = init_node_manager(sbi);
1735 if (err)
1736 return err;
1737
1738 build_free_nids(sbi);
1739 return 0;
1740 }
1741
1742 void destroy_node_manager(struct f2fs_sb_info *sbi)
1743 {
1744 struct f2fs_nm_info *nm_i = NM_I(sbi);
1745 struct free_nid *i, *next_i;
1746 struct nat_entry *natvec[NATVEC_SIZE];
1747 nid_t nid = 0;
1748 unsigned int found;
1749
1750 if (!nm_i)
1751 return;
1752
1753 /* destroy free nid list */
1754 spin_lock(&nm_i->free_nid_list_lock);
1755 list_for_each_entry_safe(i, next_i, &nm_i->free_nid_list, list) {
1756 BUG_ON(i->state == NID_ALLOC);
1757 __del_from_free_nid_list(i);
1758 nm_i->fcnt--;
1759 }
1760 BUG_ON(nm_i->fcnt);
1761 spin_unlock(&nm_i->free_nid_list_lock);
1762
1763 /* destroy nat cache */
1764 write_lock(&nm_i->nat_tree_lock);
1765 while ((found = __gang_lookup_nat_cache(nm_i,
1766 nid, NATVEC_SIZE, natvec))) {
1767 unsigned idx;
1768 for (idx = 0; idx < found; idx++) {
1769 struct nat_entry *e = natvec[idx];
1770 nid = nat_get_nid(e) + 1;
1771 __del_from_nat_cache(nm_i, e);
1772 }
1773 }
1774 BUG_ON(nm_i->nat_cnt);
1775 write_unlock(&nm_i->nat_tree_lock);
1776
1777 kfree(nm_i->nat_bitmap);
1778 sbi->nm_info = NULL;
1779 kfree(nm_i);
1780 }
1781
1782 int __init create_node_manager_caches(void)
1783 {
1784 nat_entry_slab = f2fs_kmem_cache_create("nat_entry",
1785 sizeof(struct nat_entry), NULL);
1786 if (!nat_entry_slab)
1787 return -ENOMEM;
1788
1789 free_nid_slab = f2fs_kmem_cache_create("free_nid",
1790 sizeof(struct free_nid), NULL);
1791 if (!free_nid_slab) {
1792 kmem_cache_destroy(nat_entry_slab);
1793 return -ENOMEM;
1794 }
1795 return 0;
1796 }
1797
1798 void destroy_node_manager_caches(void)
1799 {
1800 kmem_cache_destroy(free_nid_slab);
1801 kmem_cache_destroy(nat_entry_slab);
1802 }
Linux with added FPC-III support