projects/linux·blob master · refs

927 lines · 27.4 KiB

1// SPDX-License-Identifier: GPL-2.0-only
2/*
3 * mm/truncate.c - code for taking down pages from address_spaces
4 *
5 * Copyright (C) 2002, Linus Torvalds
6 *
7 * 10Sep2002 Andrew Morton
8 * Initial version.
9 */
10
11#include <linux/kernel.h>
12#include <linux/backing-dev.h>
13#include <linux/dax.h>
14#include <linux/gfp.h>
15#include <linux/mm.h>
16#include <linux/swap.h>
17#include <linux/export.h>
18#include <linux/pagemap.h>
19#include <linux/highmem.h>
20#include <linux/folio_batch.h>
21#include <linux/task_io_accounting_ops.h>
22#include <linux/shmem_fs.h>
23#include <linux/rmap.h>
24#include "internal.h"
25
26static void clear_shadow_entries(struct address_space *mapping,
27 unsigned long start, unsigned long max)
28{
29 XA_STATE(xas, &mapping->i_pages, start);
30 struct folio *folio;
31
32 /* Handled by shmem itself, or for DAX we do nothing. */
33 if (shmem_mapping(mapping) || dax_mapping(mapping))
34 return;
35
36 xas_set_update(&xas, workingset_update_node);
37
38 spin_lock(&mapping->host->i_lock);
39 xas_lock_irq(&xas);
40
41 /* Clear all shadow entries from start to max */
42 xas_for_each(&xas, folio, max) {
43 if (xa_is_value(folio))
44 xas_store(&xas, NULL);
45 }
46
47 xas_unlock_irq(&xas);
48 if (mapping_shrinkable(mapping))
49 inode_lru_list_add(mapping->host);
50 spin_unlock(&mapping->host->i_lock);
51}
52
53/*
54 * Unconditionally remove exceptional entries. Usually called from truncate
55 * path. Note that the folio_batch may be altered by this function by removing
56 * exceptional entries similar to what folio_batch_remove_exceptionals() does.
57 * Please note that indices[] has entries in ascending order as guaranteed by
58 * either find_get_entries() or find_lock_entries().
59 */
60static void truncate_folio_batch_exceptionals(struct address_space *mapping,
61 struct folio_batch *fbatch, pgoff_t *indices)
62{
63 XA_STATE(xas, &mapping->i_pages, indices[0]);
64 int nr = folio_batch_count(fbatch);
65 struct folio *folio;
66 int i, j;
67
68 /* Handled by shmem itself */
69 if (shmem_mapping(mapping))
70 return;
71
72 for (j = 0; j < nr; j++)
73 if (xa_is_value(fbatch->folios[j]))
74 break;
75
76 if (j == nr)
77 return;
78
79 if (dax_mapping(mapping)) {
80 for (i = j; i < nr; i++) {
81 if (xa_is_value(fbatch->folios[i])) {
82 /*
83 * File systems should already have called
84 * dax_break_layout_entry() to remove all DAX
85 * entries while holding a lock to prevent
86 * establishing new entries. Therefore we
87 * shouldn't find any here.
88 */
89 WARN_ON_ONCE(1);
90
91 /*
92 * Delete the mapping so truncate_pagecache()
93 * doesn't loop forever.
94 */
95 dax_delete_mapping_entry(mapping, indices[i]);
96 }
97 }
98 goto out;
99 }
100
101 xas_set(&xas, indices[j]);
102 xas_set_update(&xas, workingset_update_node);
103
104 spin_lock(&mapping->host->i_lock);
105 xas_lock_irq(&xas);
106
107 xas_for_each(&xas, folio, indices[nr-1]) {
108 if (xa_is_value(folio))
109 xas_store(&xas, NULL);
110 }
111
112 xas_unlock_irq(&xas);
113 if (mapping_shrinkable(mapping))
114 inode_lru_list_add(mapping->host);
115 spin_unlock(&mapping->host->i_lock);
116out:
117 folio_batch_remove_exceptionals(fbatch);
118}
119
120/**
121 * folio_invalidate - Invalidate part or all of a folio.
122 * @folio: The folio which is affected.
123 * @offset: start of the range to invalidate
124 * @length: length of the range to invalidate
125 *
126 * folio_invalidate() is called when all or part of the folio has become
127 * invalidated by a truncate operation.
128 *
129 * folio_invalidate() does not have to release all buffers, but it must
130 * ensure that no dirty buffer is left outside @offset and that no I/O
131 * is underway against any of the blocks which are outside the truncation
132 * point. Because the caller is about to free (and possibly reuse) those
133 * blocks on-disk.
134 */
135void folio_invalidate(struct folio *folio, size_t offset, size_t length)
136{
137 const struct address_space_operations *aops = folio->mapping->a_ops;
138
139 if (aops->invalidate_folio)
140 aops->invalidate_folio(folio, offset, length);
141}
142EXPORT_SYMBOL_GPL(folio_invalidate);
143
144/*
145 * If truncate cannot remove the fs-private metadata from the page, the page
146 * becomes orphaned. It will be left on the LRU and may even be mapped into
147 * user pagetables if we're racing with filemap_fault().
148 *
149 * We need to bail out if page->mapping is no longer equal to the original
150 * mapping. This happens a) when the VM reclaimed the page while we waited on
151 * its lock, b) when a concurrent invalidate_mapping_pages got there first and
152 * c) when tmpfs swizzles a page between a tmpfs inode and swapper_space.
153 */
154static void truncate_cleanup_folio(struct folio *folio)
155{
156 if (folio_mapped(folio))
157 unmap_mapping_folio(folio);
158
159 if (folio_needs_release(folio))
160 folio_invalidate(folio, 0, folio_size(folio));
161
162 /*
163 * Some filesystems seem to re-dirty the page even after
164 * the VM has canceled the dirty bit (eg ext3 journaling).
165 * Hence dirty accounting check is placed after invalidation.
166 */
167 folio_cancel_dirty(folio);
168}
169
170int truncate_inode_folio(struct address_space *mapping, struct folio *folio)
171{
172 if (folio->mapping != mapping)
173 return -EIO;
174
175 truncate_cleanup_folio(folio);
176 filemap_remove_folio(folio);
177 return 0;
178}
179
180static int folio_split_or_unmap(struct folio *folio, struct page *split_at,
181 unsigned long min_order)
182{
183 enum ttu_flags ttu_flags =
184 TTU_SYNC |
185 TTU_SPLIT_HUGE_PMD |
186 TTU_IGNORE_MLOCK;
187 int ret;
188
189 ret = folio_split(folio, min_order, split_at, NULL);
190
191 /*
192 * If the split fails, unmap the folio, so it will be refaulted
193 * with PTEs to respect SIGBUS semantics.
194 *
195 * Make an exception for shmem/tmpfs that for long time
196 * intentionally mapped with PMDs across i_size.
197 */
198 if (ret && !shmem_mapping(folio->mapping)) {
199 try_to_unmap(folio, ttu_flags);
200 WARN_ON(folio_mapped(folio));
201 }
202
203 return ret;
204}
205
206/*
207 * Handle partial folios. The folio may be entirely within the
208 * range if a split has raced with us. If not, we zero the part of the
209 * folio that's within the [start, end] range, and then split the folio if
210 * it's large. split_page_range() will discard pages which now lie beyond
211 * i_size, and we rely on the caller to discard pages which lie within a
212 * newly created hole.
213 *
214 * Returns false if splitting failed so the caller can avoid
215 * discarding the entire folio which is stubbornly unsplit.
216 */
217bool truncate_inode_partial_folio(struct folio *folio, loff_t start, loff_t end)
218{
219 loff_t pos = folio_pos(folio);
220 size_t size = folio_size(folio);
221 unsigned int offset, length;
222 struct page *split_at, *split_at2;
223 unsigned int min_order;
224
225 if (pos < start)
226 offset = start - pos;
227 else
228 offset = 0;
229 if (pos + size <= (u64)end)
230 length = size - offset;
231 else
232 length = end + 1 - pos - offset;
233
234 folio_wait_writeback(folio);
235 if (length == size) {
236 truncate_inode_folio(folio->mapping, folio);
237 return true;
238 }
239
240 /*
241 * We may be zeroing pages we're about to discard, but it avoids
242 * doing a complex calculation here, and then doing the zeroing
243 * anyway if the page split fails.
244 */
245 if (!mapping_inaccessible(folio->mapping))
246 folio_zero_range(folio, offset, length);
247
248 if (folio_needs_release(folio))
249 folio_invalidate(folio, offset, length);
250 if (!folio_test_large(folio))
251 return true;
252
253 min_order = mapping_min_folio_order(folio->mapping);
254 split_at = folio_page(folio, PAGE_ALIGN_DOWN(offset) / PAGE_SIZE);
255 if (!folio_split_or_unmap(folio, split_at, min_order)) {
256 /*
257 * try to split at offset + length to make sure folios within
258 * the range can be dropped, especially to avoid memory waste
259 * for shmem truncate
260 */
261 struct folio *folio2;
262
263 if (offset + length == size)
264 goto no_split;
265
266 split_at2 = folio_page(folio,
267 PAGE_ALIGN_DOWN(offset + length) / PAGE_SIZE);
268 folio2 = page_folio(split_at2);
269
270 if (!folio_try_get(folio2))
271 goto no_split;
272
273 if (!folio_test_large(folio2))
274 goto out;
275
276 if (!folio_trylock(folio2))
277 goto out;
278
279 /* make sure folio2 is large and does not change its mapping */
280 if (folio_test_large(folio2) &&
281 folio2->mapping == folio->mapping)
282 folio_split_or_unmap(folio2, split_at2, min_order);
283
284 folio_unlock(folio2);
285out:
286 folio_put(folio2);
287no_split:
288 return true;
289 }
290 if (folio_test_dirty(folio))
291 return false;
292 truncate_inode_folio(folio->mapping, folio);
293 return true;
294}
295
296/*
297 * Used to get rid of pages on hardware memory corruption.
298 */
299int generic_error_remove_folio(struct address_space *mapping,
300 struct folio *folio)
301{
302 if (!mapping)
303 return -EINVAL;
304 /*
305 * Only punch for normal data pages for now.
306 * Handling other types like directories would need more auditing.
307 */
308 if (!S_ISREG(mapping->host->i_mode))
309 return -EIO;
310 return truncate_inode_folio(mapping, folio);
311}
312EXPORT_SYMBOL(generic_error_remove_folio);
313
314/**
315 * mapping_evict_folio() - Remove an unused folio from the page-cache.
316 * @mapping: The mapping this folio belongs to.
317 * @folio: The folio to remove.
318 *
319 * Safely remove one folio from the page cache.
320 * It only drops clean, unused folios.
321 *
322 * Context: Folio must be locked.
323 * Return: The number of pages successfully removed.
324 */
325long mapping_evict_folio(struct address_space *mapping, struct folio *folio)
326{
327 /* The page may have been truncated before it was locked */
328 if (!mapping)
329 return 0;
330 if (folio_test_dirty(folio) || folio_test_writeback(folio))
331 return 0;
332 /* The refcount will be elevated if any page in the folio is mapped */
333 if (folio_ref_count(folio) >
334 folio_nr_pages(folio) + folio_has_private(folio) + 1)
335 return 0;
336 if (!filemap_release_folio(folio, 0))
337 return 0;
338
339 return remove_mapping(mapping, folio);
340}
341
342/**
343 * truncate_inode_pages_range - truncate range of pages specified by start & end byte offsets
344 * @mapping: mapping to truncate
345 * @lstart: offset from which to truncate
346 * @lend: offset to which to truncate (inclusive)
347 *
348 * Truncate the page cache, removing the pages that are between
349 * specified offsets (and zeroing out partial pages
350 * if lstart or lend + 1 is not page aligned).
351 *
352 * Truncate takes two passes - the first pass is nonblocking. It will not
353 * block on page locks and it will not block on writeback. The second pass
354 * will wait. This is to prevent as much IO as possible in the affected region.
355 * The first pass will remove most pages, so the search cost of the second pass
356 * is low.
357 *
358 * We pass down the cache-hot hint to the page freeing code. Even if the
359 * mapping is large, it is probably the case that the final pages are the most
360 * recently touched, and freeing happens in ascending file offset order.
361 *
362 * Note that since ->invalidate_folio() accepts range to invalidate
363 * truncate_inode_pages_range is able to handle cases where lend + 1 is not
364 * page aligned properly.
365 */
366void truncate_inode_pages_range(struct address_space *mapping,
367 loff_t lstart, uoff_t lend)
368{
369 pgoff_t start; /* inclusive */
370 pgoff_t end; /* exclusive */
371 struct folio_batch fbatch;
372 pgoff_t indices[FOLIO_BATCH_SIZE];
373 pgoff_t index;
374 int i;
375 struct folio *folio;
376 bool same_folio;
377
378 if (mapping_empty(mapping))
379 return;
380
381 /*
382 * 'start' and 'end' always covers the range of pages to be fully
383 * truncated. Partial pages are covered with 'partial_start' at the
384 * start of the range and 'partial_end' at the end of the range.
385 * Note that 'end' is exclusive while 'lend' is inclusive.
386 */
387 start = (lstart + PAGE_SIZE - 1) >> PAGE_SHIFT;
388 if (lend == -1)
389 /*
390 * lend == -1 indicates end-of-file so we have to set 'end'
391 * to the highest possible pgoff_t and since the type is
392 * unsigned we're using -1.
393 */
394 end = -1;
395 else
396 end = (lend + 1) >> PAGE_SHIFT;
397
398 folio_batch_init(&fbatch);
399 index = start;
400 while (index < end && find_lock_entries(mapping, &index, end - 1,
401 &fbatch, indices)) {
402 truncate_folio_batch_exceptionals(mapping, &fbatch, indices);
403 for (i = 0; i < folio_batch_count(&fbatch); i++)
404 truncate_cleanup_folio(fbatch.folios[i]);
405 delete_from_page_cache_batch(mapping, &fbatch);
406 for (i = 0; i < folio_batch_count(&fbatch); i++)
407 folio_unlock(fbatch.folios[i]);
408 folio_batch_release(&fbatch);
409 cond_resched();
410 }
411
412 same_folio = (lstart >> PAGE_SHIFT) == (lend >> PAGE_SHIFT);
413 folio = __filemap_get_folio(mapping, lstart >> PAGE_SHIFT, FGP_LOCK, 0);
414 if (!IS_ERR(folio)) {
415 same_folio = lend < folio_next_pos(folio);
416 if (!truncate_inode_partial_folio(folio, lstart, lend)) {
417 start = folio_next_index(folio);
418 if (same_folio)
419 end = folio->index;
420 }
421 folio_unlock(folio);
422 folio_put(folio);
423 folio = NULL;
424 }
425
426 if (!same_folio) {
427 folio = __filemap_get_folio(mapping, lend >> PAGE_SHIFT,
428 FGP_LOCK, 0);
429 if (!IS_ERR(folio)) {
430 if (!truncate_inode_partial_folio(folio, lstart, lend))
431 end = folio->index;
432 folio_unlock(folio);
433 folio_put(folio);
434 }
435 }
436
437 index = start;
438 while (index < end) {
439 cond_resched();
440 if (!find_get_entries(mapping, &index, end - 1, &fbatch,
441 indices)) {
442 /* If all gone from start onwards, we're done */
443 if (index == start)
444 break;
445 /* Otherwise restart to make sure all gone */
446 index = start;
447 continue;
448 }
449
450 for (i = 0; i < folio_batch_count(&fbatch); i++) {
451 struct folio *folio = fbatch.folios[i];
452
453 /* We rely upon deletion not changing folio->index */
454
455 if (xa_is_value(folio))
456 continue;
457
458 folio_lock(folio);
459 VM_BUG_ON_FOLIO(!folio_contains(folio, indices[i]), folio);
460 folio_wait_writeback(folio);
461 truncate_inode_folio(mapping, folio);
462 folio_unlock(folio);
463 }
464 truncate_folio_batch_exceptionals(mapping, &fbatch, indices);
465 folio_batch_release(&fbatch);
466 }
467}
468EXPORT_SYMBOL(truncate_inode_pages_range);
469
470/**
471 * truncate_inode_pages - truncate *all* the pages from an offset
472 * @mapping: mapping to truncate
473 * @lstart: offset from which to truncate
474 *
475 * Called under (and serialised by) inode->i_rwsem and
476 * mapping->invalidate_lock.
477 *
478 * Note: When this function returns, there can be a page in the process of
479 * deletion (inside __filemap_remove_folio()) in the specified range. Thus
480 * mapping->nrpages can be non-zero when this function returns even after
481 * truncation of the whole mapping.
482 */
483void truncate_inode_pages(struct address_space *mapping, loff_t lstart)
484{
485 truncate_inode_pages_range(mapping, lstart, (loff_t)-1);
486}
487EXPORT_SYMBOL(truncate_inode_pages);
488
489/**
490 * truncate_inode_pages_final - truncate *all* pages before inode dies
491 * @mapping: mapping to truncate
492 *
493 * Called under (and serialized by) inode->i_rwsem.
494 *
495 * Filesystems have to use this in the .evict_inode path to inform the
496 * VM that this is the final truncate and the inode is going away.
497 */
498void truncate_inode_pages_final(struct address_space *mapping)
499{
500 /*
501 * Page reclaim can not participate in regular inode lifetime
502 * management (can't call iput()) and thus can race with the
503 * inode teardown. Tell it when the address space is exiting,
504 * so that it does not install eviction information after the
505 * final truncate has begun.
506 */
507 mapping_set_exiting(mapping);
508
509 if (!mapping_empty(mapping)) {
510 /*
511 * As truncation uses a lockless tree lookup, cycle
512 * the tree lock to make sure any ongoing tree
513 * modification that does not see AS_EXITING is
514 * completed before starting the final truncate.
515 */
516 xa_lock_irq(&mapping->i_pages);
517 xa_unlock_irq(&mapping->i_pages);
518 }
519
520 truncate_inode_pages(mapping, 0);
521}
522EXPORT_SYMBOL(truncate_inode_pages_final);
523
524/**
525 * mapping_try_invalidate - Invalidate all the evictable folios of one inode
526 * @mapping: the address_space which holds the folios to invalidate
527 * @start: the offset 'from' which to invalidate
528 * @end: the offset 'to' which to invalidate (inclusive)
529 * @nr_failed: How many folio invalidations failed
530 *
531 * This function is similar to invalidate_mapping_pages(), except that it
532 * returns the number of folios which could not be evicted in @nr_failed.
533 */
534unsigned long mapping_try_invalidate(struct address_space *mapping,
535 pgoff_t start, pgoff_t end, unsigned long *nr_failed)
536{
537 pgoff_t indices[FOLIO_BATCH_SIZE];
538 struct folio_batch fbatch;
539 pgoff_t index = start;
540 unsigned long ret;
541 unsigned long count = 0;
542 int i;
543
544 folio_batch_init(&fbatch);
545 while (find_lock_entries(mapping, &index, end, &fbatch, indices)) {
546 bool xa_has_values = false;
547 int nr = folio_batch_count(&fbatch);
548
549 for (i = 0; i < nr; i++) {
550 struct folio *folio = fbatch.folios[i];
551
552 /* We rely upon deletion not changing folio->index */
553
554 if (xa_is_value(folio)) {
555 xa_has_values = true;
556 count++;
557 continue;
558 }
559
560 ret = mapping_evict_folio(mapping, folio);
561 folio_unlock(folio);
562 /*
563 * Invalidation is a hint that the folio is no longer
564 * of interest and try to speed up its reclaim.
565 */
566 if (!ret) {
567 deactivate_file_folio(folio);
568 /* Likely in the lru cache of a remote CPU */
569 if (nr_failed)
570 (*nr_failed)++;
571 }
572 count += ret;
573 }
574
575 if (xa_has_values)
576 clear_shadow_entries(mapping, indices[0], indices[nr-1]);
577
578 folio_batch_remove_exceptionals(&fbatch);
579 folio_batch_release(&fbatch);
580 cond_resched();
581 }
582 return count;
583}
584
585/**
586 * invalidate_mapping_pages - Invalidate all clean, unlocked cache of one inode
587 * @mapping: the address_space which holds the cache to invalidate
588 * @start: the offset 'from' which to invalidate
589 * @end: the offset 'to' which to invalidate (inclusive)
590 *
591 * This function removes pages that are clean, unmapped and unlocked,
592 * as well as shadow entries. It will not block on IO activity.
593 *
594 * If you want to remove all the pages of one inode, regardless of
595 * their use and writeback state, use truncate_inode_pages().
596 *
597 * Return: The number of indices that had their contents invalidated
598 */
599unsigned long invalidate_mapping_pages(struct address_space *mapping,
600 pgoff_t start, pgoff_t end)
601{
602 return mapping_try_invalidate(mapping, start, end, NULL);
603}
604EXPORT_SYMBOL(invalidate_mapping_pages);
605
606static int folio_launder(struct address_space *mapping, struct folio *folio)
607{
608 if (!folio_test_dirty(folio))
609 return 0;
610 if (folio->mapping != mapping || mapping->a_ops->launder_folio == NULL)
611 return 0;
612 return mapping->a_ops->launder_folio(folio);
613}
614
615/*
616 * This is like mapping_evict_folio(), except it ignores the folio's
617 * refcount. We do this because invalidate_inode_pages2() needs stronger
618 * invalidation guarantees, and cannot afford to leave folios behind because
619 * shrink_folio_list() has a temp ref on them, or because they're transiently
620 * sitting in the folio_add_lru() caches.
621 */
622int folio_unmap_invalidate(struct address_space *mapping, struct folio *folio,
623 gfp_t gfp)
624{
625 void (*free_folio)(struct folio *);
626 int ret;
627
628 VM_BUG_ON_FOLIO(!folio_test_locked(folio), folio);
629
630 if (folio_mapped(folio))
631 unmap_mapping_folio(folio);
632 BUG_ON(folio_mapped(folio));
633
634 ret = folio_launder(mapping, folio);
635 if (ret)
636 return ret;
637 if (folio->mapping != mapping)
638 return -EBUSY;
639 if (!filemap_release_folio(folio, gfp))
640 return -EBUSY;
641
642 spin_lock(&mapping->host->i_lock);
643 xa_lock_irq(&mapping->i_pages);
644 if (folio_test_dirty(folio))
645 goto failed;
646
647 BUG_ON(folio_has_private(folio));
648 __filemap_remove_folio(folio, NULL);
649 xa_unlock_irq(&mapping->i_pages);
650 if (mapping_shrinkable(mapping))
651 inode_lru_list_add(mapping->host);
652 free_folio = mapping->a_ops->free_folio;
653 spin_unlock(&mapping->host->i_lock);
654
655 if (free_folio)
656 free_folio(folio);
657 folio_put_refs(folio, folio_nr_pages(folio));
658 return 1;
659failed:
660 xa_unlock_irq(&mapping->i_pages);
661 spin_unlock(&mapping->host->i_lock);
662 return -EBUSY;
663}
664
665/**
666 * invalidate_inode_pages2_range - remove range of pages from an address_space
667 * @mapping: the address_space
668 * @start: the page offset 'from' which to invalidate
669 * @end: the page offset 'to' which to invalidate (inclusive)
670 *
671 * Any pages which are found to be mapped into pagetables are unmapped prior to
672 * invalidation.
673 *
674 * Return: -EBUSY if any pages could not be invalidated.
675 */
676int invalidate_inode_pages2_range(struct address_space *mapping,
677 pgoff_t start, pgoff_t end)
678{
679 pgoff_t indices[FOLIO_BATCH_SIZE];
680 struct folio_batch fbatch;
681 pgoff_t index;
682 int i;
683 int ret = 0;
684 int ret2 = 0;
685 int did_range_unmap = 0;
686
687 if (mapping_empty(mapping))
688 return 0;
689
690 folio_batch_init(&fbatch);
691 index = start;
692 while (find_get_entries(mapping, &index, end, &fbatch, indices)) {
693 bool xa_has_values = false;
694 int nr = folio_batch_count(&fbatch);
695
696 for (i = 0; i < nr; i++) {
697 struct folio *folio = fbatch.folios[i];
698
699 /* We rely upon deletion not changing folio->index */
700
701 if (xa_is_value(folio)) {
702 xa_has_values = true;
703 if (dax_mapping(mapping) &&
704 !dax_invalidate_mapping_entry_sync(mapping, indices[i]))
705 ret = -EBUSY;
706 continue;
707 }
708
709 if (!did_range_unmap && folio_mapped(folio)) {
710 /*
711 * If folio is mapped, before taking its lock,
712 * zap the rest of the file in one hit.
713 */
714 unmap_mapping_pages(mapping, indices[i],
715 (1 + end - indices[i]), false);
716 did_range_unmap = 1;
717 }
718
719 folio_lock(folio);
720 if (unlikely(folio->mapping != mapping)) {
721 folio_unlock(folio);
722 continue;
723 }
724 VM_BUG_ON_FOLIO(!folio_contains(folio, indices[i]), folio);
725 folio_wait_writeback(folio);
726 ret2 = folio_unmap_invalidate(mapping, folio, GFP_KERNEL);
727 if (ret2 < 0)
728 ret = ret2;
729 folio_unlock(folio);
730 }
731
732 if (xa_has_values)
733 clear_shadow_entries(mapping, indices[0], indices[nr-1]);
734
735 folio_batch_remove_exceptionals(&fbatch);
736 folio_batch_release(&fbatch);
737 cond_resched();
738 }
739 /*
740 * For DAX we invalidate page tables after invalidating page cache. We
741 * could invalidate page tables while invalidating each entry however
742 * that would be expensive. And doing range unmapping before doesn't
743 * work as we have no cheap way to find whether page cache entry didn't
744 * get remapped later.
745 */
746 if (dax_mapping(mapping)) {
747 unmap_mapping_pages(mapping, start, end - start + 1, false);
748 }
749 return ret;
750}
751EXPORT_SYMBOL_GPL(invalidate_inode_pages2_range);
752
753/**
754 * invalidate_inode_pages2 - remove all pages from an address_space
755 * @mapping: the address_space
756 *
757 * Any pages which are found to be mapped into pagetables are unmapped prior to
758 * invalidation.
759 *
760 * Return: -EBUSY if any pages could not be invalidated.
761 */
762int invalidate_inode_pages2(struct address_space *mapping)
763{
764 return invalidate_inode_pages2_range(mapping, 0, -1);
765}
766EXPORT_SYMBOL_GPL(invalidate_inode_pages2);
767
768/**
769 * truncate_pagecache - unmap and remove pagecache that has been truncated
770 * @inode: inode
771 * @newsize: new file size
772 *
773 * inode's new i_size must already be written before truncate_pagecache
774 * is called.
775 *
776 * This function should typically be called before the filesystem
777 * releases resources associated with the freed range (eg. deallocates
778 * blocks). This way, pagecache will always stay logically coherent
779 * with on-disk format, and the filesystem would not have to deal with
780 * situations such as writepage being called for a page that has already
781 * had its underlying blocks deallocated.
782 */
783void truncate_pagecache(struct inode *inode, loff_t newsize)
784{
785 struct address_space *mapping = inode->i_mapping;
786 loff_t holebegin = round_up(newsize, PAGE_SIZE);
787
788 /*
789 * unmap_mapping_range is called twice, first simply for
790 * efficiency so that truncate_inode_pages does fewer
791 * single-page unmaps. However after this first call, and
792 * before truncate_inode_pages finishes, it is possible for
793 * private pages to be COWed, which remain after
794 * truncate_inode_pages finishes, hence the second
795 * unmap_mapping_range call must be made for correctness.
796 */
797 unmap_mapping_range(mapping, holebegin, 0, 1);
798 truncate_inode_pages(mapping, newsize);
799 unmap_mapping_range(mapping, holebegin, 0, 1);
800}
801EXPORT_SYMBOL(truncate_pagecache);
802
803/**
804 * truncate_setsize - update inode and pagecache for a new file size
805 * @inode: inode
806 * @newsize: new file size
807 *
808 * truncate_setsize updates i_size and performs pagecache truncation (if
809 * necessary) to @newsize. It will be typically be called from the filesystem's
810 * setattr function when ATTR_SIZE is passed in.
811 *
812 * Must be called with a lock serializing truncates and writes (generally
813 * i_rwsem but e.g. xfs uses a different lock) and before all filesystem
814 * specific block truncation has been performed.
815 */
816void truncate_setsize(struct inode *inode, loff_t newsize)
817{
818 loff_t oldsize = inode->i_size;
819
820 i_size_write(inode, newsize);
821 if (newsize > oldsize)
822 pagecache_isize_extended(inode, oldsize, newsize);
823 truncate_pagecache(inode, newsize);
824}
825EXPORT_SYMBOL(truncate_setsize);
826
827/**
828 * pagecache_isize_extended - update pagecache after extension of i_size
829 * @inode: inode for which i_size was extended
830 * @from: original inode size
831 * @to: new inode size
832 *
833 * Handle extension of inode size either caused by extending truncate or
834 * by write starting after current i_size. We mark the page straddling
835 * current i_size RO so that page_mkwrite() is called on the first
836 * write access to the page. The filesystem will update its per-block
837 * information before user writes to the page via mmap after the i_size
838 * has been changed.
839 *
840 * The function must be called after i_size is updated so that page fault
841 * coming after we unlock the folio will already see the new i_size.
842 * The function must be called while we still hold i_rwsem - this not only
843 * makes sure i_size is stable but also that userspace cannot observe new
844 * i_size value before we are prepared to store mmap writes at new inode size.
845 */
846void pagecache_isize_extended(struct inode *inode, loff_t from, loff_t to)
847{
848 int bsize = i_blocksize(inode);
849 loff_t rounded_from;
850 struct folio *folio;
851
852 WARN_ON(to > inode->i_size);
853
854 if (from >= to || bsize >= PAGE_SIZE)
855 return;
856 /* Page straddling @from will not have any hole block created? */
857 rounded_from = round_up(from, bsize);
858 if (to <= rounded_from || !(rounded_from & (PAGE_SIZE - 1)))
859 return;
860
861 folio = filemap_lock_folio(inode->i_mapping, from / PAGE_SIZE);
862 /* Folio not cached? Nothing to do */
863 if (IS_ERR(folio))
864 return;
865 /*
866 * See folio_clear_dirty_for_io() for details why folio_mark_dirty()
867 * is needed.
868 */
869 if (folio_mkclean(folio))
870 folio_mark_dirty(folio);
871
872 /*
873 * The post-eof range of the folio must be zeroed before it is exposed
874 * to the file. Writeback normally does this, but since i_size has been
875 * increased we handle it here.
876 */
877 if (folio_test_dirty(folio)) {
878 unsigned int offset, end;
879
880 offset = from - folio_pos(folio);
881 end = min_t(unsigned int, to - folio_pos(folio),
882 folio_size(folio));
883 folio_zero_segment(folio, offset, end);
884 }
885
886 folio_unlock(folio);
887 folio_put(folio);
888}
889EXPORT_SYMBOL(pagecache_isize_extended);
890
891/**
892 * truncate_pagecache_range - unmap and remove pagecache that is hole-punched
893 * @inode: inode
894 * @lstart: offset of beginning of hole
895 * @lend: offset of last byte of hole
896 *
897 * This function should typically be called before the filesystem
898 * releases resources associated with the freed range (eg. deallocates
899 * blocks). This way, pagecache will always stay logically coherent
900 * with on-disk format, and the filesystem would not have to deal with
901 * situations such as writepage being called for a page that has already
902 * had its underlying blocks deallocated.
903 */
904void truncate_pagecache_range(struct inode *inode, loff_t lstart, loff_t lend)
905{
906 struct address_space *mapping = inode->i_mapping;
907 loff_t unmap_start = round_up(lstart, PAGE_SIZE);
908 loff_t unmap_end = round_down(1 + lend, PAGE_SIZE) - 1;
909 /*
910 * This rounding is currently just for example: unmap_mapping_range
911 * expands its hole outwards, whereas we want it to contract the hole
912 * inwards. However, existing callers of truncate_pagecache_range are
913 * doing their own page rounding first. Note that unmap_mapping_range
914 * allows holelen 0 for all, and we allow lend -1 for end of file.
915 */
916
917 /*
918 * Unlike in truncate_pagecache, unmap_mapping_range is called only
919 * once (before truncating pagecache), and without "even_cows" flag:
920 * hole-punching should not remove private COWed pages from the hole.
921 */
922 if ((u64)unmap_end > (u64)unmap_start)
923 unmap_mapping_range(mapping, unmap_start,
924 1 + unmap_end - unmap_start, 0);
925 truncate_inode_pages_range(mapping, lstart, lend);
926}
927EXPORT_SYMBOL(truncate_pagecache_range);