filemap.c 67.4 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
/*
 *	linux/mm/filemap.c
 *
 * Copyright (C) 1994-1999  Linus Torvalds
 */

/*
 * This file handles the generic file mmap semantics used by
 * most "normal" filesystems (but you don't /have/ to use this:
 * the NFS filesystem used to do this differently, for example)
 */
#include <linux/module.h>
#include <linux/slab.h>
#include <linux/compiler.h>
#include <linux/fs.h>
16
#include <linux/uaccess.h>
L
Linus Torvalds 已提交
17
#include <linux/aio.h>
18
#include <linux/capability.h>
L
Linus Torvalds 已提交
19 20 21 22 23 24 25 26 27 28 29 30 31
#include <linux/kernel_stat.h>
#include <linux/mm.h>
#include <linux/swap.h>
#include <linux/mman.h>
#include <linux/pagemap.h>
#include <linux/file.h>
#include <linux/uio.h>
#include <linux/hash.h>
#include <linux/writeback.h>
#include <linux/pagevec.h>
#include <linux/blkdev.h>
#include <linux/security.h>
#include <linux/syscalls.h>
32
#include <linux/cpuset.h>
N
Nick Piggin 已提交
33
#include <linux/hardirq.h> /* for BUG_ON(!in_atomic()) only */
34 35
#include "internal.h"

L
Linus Torvalds 已提交
36 37 38 39 40 41 42
/*
 * FIXME: remove all knowledge of the buffer layer from the core VM
 */
#include <linux/buffer_head.h> /* for generic_osync_inode */

#include <asm/mman.h>

43 44 45 46
static ssize_t
generic_file_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov,
	loff_t offset, unsigned long nr_segs);

L
Linus Torvalds 已提交
47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63
/*
 * Shared mappings implemented 30.11.1994. It's not fully working yet,
 * though.
 *
 * Shared mappings now work. 15.8.1995  Bruno.
 *
 * finished 'unifying' the page and buffer cache and SMP-threaded the
 * page-cache, 21.05.1999, Ingo Molnar <mingo@redhat.com>
 *
 * SMP-threaded pagemap-LRU 1999, Andrea Arcangeli <andrea@suse.de>
 */

/*
 * Lock ordering:
 *
 *  ->i_mmap_lock		(vmtruncate)
 *    ->private_lock		(__free_pte->__set_page_dirty_buffers)
64 65
 *      ->swap_lock		(exclusive_swap_page, others)
 *        ->mapping->tree_lock
66
 *          ->zone.lock
L
Linus Torvalds 已提交
67
 *
68
 *  ->i_mutex
L
Linus Torvalds 已提交
69 70 71 72
 *    ->i_mmap_lock		(truncate->unmap_mapping_range)
 *
 *  ->mmap_sem
 *    ->i_mmap_lock
73
 *      ->page_table_lock or pte_lock	(various, mainly in memory.c)
L
Linus Torvalds 已提交
74 75 76 77 78
 *        ->mapping->tree_lock	(arch-dependent flush_dcache_mmap_lock)
 *
 *  ->mmap_sem
 *    ->lock_page		(access_process_vm)
 *
79 80
 *  ->i_mutex			(generic_file_buffered_write)
 *    ->mmap_sem		(fault_in_pages_readable->do_page_fault)
L
Linus Torvalds 已提交
81
 *
82
 *  ->i_mutex
L
Linus Torvalds 已提交
83 84 85 86 87 88 89 90 91 92
 *    ->i_alloc_sem             (various)
 *
 *  ->inode_lock
 *    ->sb_lock			(fs/fs-writeback.c)
 *    ->mapping->tree_lock	(__sync_single_inode)
 *
 *  ->i_mmap_lock
 *    ->anon_vma.lock		(vma_adjust)
 *
 *  ->anon_vma.lock
93
 *    ->page_table_lock or pte_lock	(anon_vma_prepare and various)
L
Linus Torvalds 已提交
94
 *
95
 *  ->page_table_lock or pte_lock
96
 *    ->swap_lock		(try_to_unmap_one)
L
Linus Torvalds 已提交
97 98 99
 *    ->private_lock		(try_to_unmap_one)
 *    ->tree_lock		(try_to_unmap_one)
 *    ->zone.lru_lock		(follow_page->mark_page_accessed)
100
 *    ->zone.lru_lock		(check_pte_range->isolate_lru_page)
L
Linus Torvalds 已提交
101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122
 *    ->private_lock		(page_remove_rmap->set_page_dirty)
 *    ->tree_lock		(page_remove_rmap->set_page_dirty)
 *    ->inode_lock		(page_remove_rmap->set_page_dirty)
 *    ->inode_lock		(zap_pte_range->set_page_dirty)
 *    ->private_lock		(zap_pte_range->__set_page_dirty_buffers)
 *
 *  ->task->proc_lock
 *    ->dcache_lock		(proc_pid_lookup)
 */

/*
 * Remove a page from the page cache and free it. Caller has to make
 * sure the page is locked and that nobody else uses it - or that usage
 * is safe.  The caller must hold a write_lock on the mapping's tree_lock.
 */
void __remove_from_page_cache(struct page *page)
{
	struct address_space *mapping = page->mapping;

	radix_tree_delete(&mapping->page_tree, page->index);
	page->mapping = NULL;
	mapping->nrpages--;
123
	__dec_zone_page_state(page, NR_FILE_PAGES);
124
	BUG_ON(page_mapped(page));
L
Linus Torvalds 已提交
125 126 127 128 129 130
}

void remove_from_page_cache(struct page *page)
{
	struct address_space *mapping = page->mapping;

M
Matt Mackall 已提交
131
	BUG_ON(!PageLocked(page));
L
Linus Torvalds 已提交
132 133 134 135 136 137 138 139 140 141 142

	write_lock_irq(&mapping->tree_lock);
	__remove_from_page_cache(page);
	write_unlock_irq(&mapping->tree_lock);
}

static int sync_page(void *word)
{
	struct address_space *mapping;
	struct page *page;

143
	page = container_of((unsigned long *)word, struct page, flags);
L
Linus Torvalds 已提交
144 145

	/*
146 147 148 149 150 151 152 153 154 155 156 157 158 159 160
	 * page_mapping() is being called without PG_locked held.
	 * Some knowledge of the state and use of the page is used to
	 * reduce the requirements down to a memory barrier.
	 * The danger here is of a stale page_mapping() return value
	 * indicating a struct address_space different from the one it's
	 * associated with when it is associated with one.
	 * After smp_mb(), it's either the correct page_mapping() for
	 * the page, or an old page_mapping() and the page's own
	 * page_mapping() has gone NULL.
	 * The ->sync_page() address_space operation must tolerate
	 * page_mapping() going NULL. By an amazing coincidence,
	 * this comes about because none of the users of the page
	 * in the ->sync_page() methods make essential use of the
	 * page_mapping(), merely passing the page down to the backing
	 * device's unplug functions when it's non-NULL, which in turn
H
Hugh Dickins 已提交
161
	 * ignore it for all cases but swap, where only page_private(page) is
162 163 164
	 * of interest. When page_mapping() does go NULL, the entire
	 * call stack gracefully ignores the page and returns.
	 * -- wli
L
Linus Torvalds 已提交
165 166 167 168 169 170 171 172 173 174
	 */
	smp_mb();
	mapping = page_mapping(page);
	if (mapping && mapping->a_ops && mapping->a_ops->sync_page)
		mapping->a_ops->sync_page(page);
	io_schedule();
	return 0;
}

/**
175
 * __filemap_fdatawrite_range - start writeback on mapping dirty pages in range
176 177
 * @mapping:	address space structure to write
 * @start:	offset in bytes where the range starts
178
 * @end:	offset in bytes where the range ends (inclusive)
179
 * @sync_mode:	enable synchronous operation
L
Linus Torvalds 已提交
180
 *
181 182 183
 * Start writeback against all of a mapping's dirty pages that lie
 * within the byte offsets <start, end> inclusive.
 *
L
Linus Torvalds 已提交
184
 * If sync_mode is WB_SYNC_ALL then this is a "data integrity" operation, as
185
 * opposed to a regular memory cleansing writeback.  The difference between
L
Linus Torvalds 已提交
186 187 188
 * these two operations is that if a dirty page/buffer is encountered, it must
 * be waited upon, and not just skipped over.
 */
189 190
int __filemap_fdatawrite_range(struct address_space *mapping, loff_t start,
				loff_t end, int sync_mode)
L
Linus Torvalds 已提交
191 192 193 194 195
{
	int ret;
	struct writeback_control wbc = {
		.sync_mode = sync_mode,
		.nr_to_write = mapping->nrpages * 2,
196 197
		.range_start = start,
		.range_end = end,
L
Linus Torvalds 已提交
198 199 200 201 202 203 204 205 206 207 208 209
	};

	if (!mapping_cap_writeback_dirty(mapping))
		return 0;

	ret = do_writepages(mapping, &wbc);
	return ret;
}

static inline int __filemap_fdatawrite(struct address_space *mapping,
	int sync_mode)
{
210
	return __filemap_fdatawrite_range(mapping, 0, LLONG_MAX, sync_mode);
L
Linus Torvalds 已提交
211 212 213 214 215 216 217 218
}

int filemap_fdatawrite(struct address_space *mapping)
{
	return __filemap_fdatawrite(mapping, WB_SYNC_ALL);
}
EXPORT_SYMBOL(filemap_fdatawrite);

219 220
static int filemap_fdatawrite_range(struct address_space *mapping, loff_t start,
				loff_t end)
L
Linus Torvalds 已提交
221 222 223 224
{
	return __filemap_fdatawrite_range(mapping, start, end, WB_SYNC_ALL);
}

225 226 227 228
/**
 * filemap_flush - mostly a non-blocking flush
 * @mapping:	target address_space
 *
L
Linus Torvalds 已提交
229 230 231 232 233 234 235 236 237
 * This is a mostly non-blocking flush.  Not suitable for data-integrity
 * purposes - I/O may not be started against all dirty pages.
 */
int filemap_flush(struct address_space *mapping)
{
	return __filemap_fdatawrite(mapping, WB_SYNC_NONE);
}
EXPORT_SYMBOL(filemap_flush);

238 239 240 241 242 243
/**
 * wait_on_page_writeback_range - wait for writeback to complete
 * @mapping:	target address_space
 * @start:	beginning page index
 * @end:	ending page index
 *
L
Linus Torvalds 已提交
244 245 246
 * Wait for writeback to complete against pages indexed by start->end
 * inclusive
 */
247
int wait_on_page_writeback_range(struct address_space *mapping,
L
Linus Torvalds 已提交
248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289
				pgoff_t start, pgoff_t end)
{
	struct pagevec pvec;
	int nr_pages;
	int ret = 0;
	pgoff_t index;

	if (end < start)
		return 0;

	pagevec_init(&pvec, 0);
	index = start;
	while ((index <= end) &&
			(nr_pages = pagevec_lookup_tag(&pvec, mapping, &index,
			PAGECACHE_TAG_WRITEBACK,
			min(end - index, (pgoff_t)PAGEVEC_SIZE-1) + 1)) != 0) {
		unsigned i;

		for (i = 0; i < nr_pages; i++) {
			struct page *page = pvec.pages[i];

			/* until radix tree lookup accepts end_index */
			if (page->index > end)
				continue;

			wait_on_page_writeback(page);
			if (PageError(page))
				ret = -EIO;
		}
		pagevec_release(&pvec);
		cond_resched();
	}

	/* Check for outstanding write errors */
	if (test_and_clear_bit(AS_ENOSPC, &mapping->flags))
		ret = -ENOSPC;
	if (test_and_clear_bit(AS_EIO, &mapping->flags))
		ret = -EIO;

	return ret;
}

290 291 292 293 294 295 296
/**
 * sync_page_range - write and wait on all pages in the passed range
 * @inode:	target inode
 * @mapping:	target address_space
 * @pos:	beginning offset in pages to write
 * @count:	number of bytes to write
 *
L
Linus Torvalds 已提交
297 298 299 300
 * Write and wait upon all the pages in the passed range.  This is a "data
 * integrity" operation.  It waits upon in-flight writeout before starting and
 * waiting upon new writeout.  If there was an IO error, return it.
 *
301
 * We need to re-take i_mutex during the generic_osync_inode list walk because
L
Linus Torvalds 已提交
302 303 304
 * it is otherwise livelockable.
 */
int sync_page_range(struct inode *inode, struct address_space *mapping,
305
			loff_t pos, loff_t count)
L
Linus Torvalds 已提交
306 307 308 309 310 311 312 313 314
{
	pgoff_t start = pos >> PAGE_CACHE_SHIFT;
	pgoff_t end = (pos + count - 1) >> PAGE_CACHE_SHIFT;
	int ret;

	if (!mapping_cap_writeback_dirty(mapping) || !count)
		return 0;
	ret = filemap_fdatawrite_range(mapping, pos, pos + count - 1);
	if (ret == 0) {
315
		mutex_lock(&inode->i_mutex);
L
Linus Torvalds 已提交
316
		ret = generic_osync_inode(inode, mapping, OSYNC_METADATA);
317
		mutex_unlock(&inode->i_mutex);
L
Linus Torvalds 已提交
318 319 320 321 322 323 324
	}
	if (ret == 0)
		ret = wait_on_page_writeback_range(mapping, start, end);
	return ret;
}
EXPORT_SYMBOL(sync_page_range);

325 326 327 328 329 330 331
/**
 * sync_page_range_nolock
 * @inode:	target inode
 * @mapping:	target address_space
 * @pos:	beginning offset in pages to write
 * @count:	number of bytes to write
 *
332
 * Note: Holding i_mutex across sync_page_range_nolock() is not a good idea
L
Linus Torvalds 已提交
333 334 335
 * as it forces O_SYNC writers to different parts of the same file
 * to be serialised right until io completion.
 */
336 337
int sync_page_range_nolock(struct inode *inode, struct address_space *mapping,
			   loff_t pos, loff_t count)
L
Linus Torvalds 已提交
338 339 340 341 342 343 344 345 346 347 348 349 350 351
{
	pgoff_t start = pos >> PAGE_CACHE_SHIFT;
	pgoff_t end = (pos + count - 1) >> PAGE_CACHE_SHIFT;
	int ret;

	if (!mapping_cap_writeback_dirty(mapping) || !count)
		return 0;
	ret = filemap_fdatawrite_range(mapping, pos, pos + count - 1);
	if (ret == 0)
		ret = generic_osync_inode(inode, mapping, OSYNC_METADATA);
	if (ret == 0)
		ret = wait_on_page_writeback_range(mapping, start, end);
	return ret;
}
352
EXPORT_SYMBOL(sync_page_range_nolock);
L
Linus Torvalds 已提交
353 354

/**
355
 * filemap_fdatawait - wait for all under-writeback pages to complete
L
Linus Torvalds 已提交
356
 * @mapping: address space structure to wait for
357 358 359
 *
 * Walk the list of under-writeback pages of the given address space
 * and wait for all of them.
L
Linus Torvalds 已提交
360 361 362 363 364 365 366 367 368 369 370 371 372 373 374
 */
int filemap_fdatawait(struct address_space *mapping)
{
	loff_t i_size = i_size_read(mapping->host);

	if (i_size == 0)
		return 0;

	return wait_on_page_writeback_range(mapping, 0,
				(i_size - 1) >> PAGE_CACHE_SHIFT);
}
EXPORT_SYMBOL(filemap_fdatawait);

int filemap_write_and_wait(struct address_space *mapping)
{
375
	int err = 0;
L
Linus Torvalds 已提交
376 377

	if (mapping->nrpages) {
378 379 380 381 382 383 384 385 386 387 388 389
		err = filemap_fdatawrite(mapping);
		/*
		 * Even if the above returned error, the pages may be
		 * written partially (e.g. -ENOSPC), so we wait for it.
		 * But the -EIO is special case, it may indicate the worst
		 * thing (e.g. bug) happened, so we avoid waiting for it.
		 */
		if (err != -EIO) {
			int err2 = filemap_fdatawait(mapping);
			if (!err)
				err = err2;
		}
L
Linus Torvalds 已提交
390
	}
391
	return err;
L
Linus Torvalds 已提交
392
}
393
EXPORT_SYMBOL(filemap_write_and_wait);
L
Linus Torvalds 已提交
394

395 396 397 398 399 400
/**
 * filemap_write_and_wait_range - write out & wait on a file range
 * @mapping:	the address_space for the pages
 * @lstart:	offset in bytes where the range starts
 * @lend:	offset in bytes where the range ends (inclusive)
 *
401 402 403 404 405
 * Write out and wait upon file offsets lstart->lend, inclusive.
 *
 * Note that `lend' is inclusive (describes the last byte to be written) so
 * that this function can be used to write to the very end-of-file (end = -1).
 */
L
Linus Torvalds 已提交
406 407 408
int filemap_write_and_wait_range(struct address_space *mapping,
				 loff_t lstart, loff_t lend)
{
409
	int err = 0;
L
Linus Torvalds 已提交
410 411

	if (mapping->nrpages) {
412 413 414 415 416 417 418 419 420 421
		err = __filemap_fdatawrite_range(mapping, lstart, lend,
						 WB_SYNC_ALL);
		/* See comment of filemap_write_and_wait() */
		if (err != -EIO) {
			int err2 = wait_on_page_writeback_range(mapping,
						lstart >> PAGE_CACHE_SHIFT,
						lend >> PAGE_CACHE_SHIFT);
			if (!err)
				err = err2;
		}
L
Linus Torvalds 已提交
422
	}
423
	return err;
L
Linus Torvalds 已提交
424 425
}

426 427 428 429 430 431 432 433
/**
 * add_to_page_cache - add newly allocated pagecache pages
 * @page:	page to add
 * @mapping:	the page's address_space
 * @offset:	page index
 * @gfp_mask:	page allocation mode
 *
 * This function is used to add newly allocated pagecache pages;
L
Linus Torvalds 已提交
434 435 436 437 438 439
 * the page is new, so we can just run SetPageLocked() against it.
 * The other page state flags were set by rmqueue().
 *
 * This function does not add the page to the LRU.  The caller must do that.
 */
int add_to_page_cache(struct page *page, struct address_space *mapping,
A
Al Viro 已提交
440
		pgoff_t offset, gfp_t gfp_mask)
L
Linus Torvalds 已提交
441 442 443 444 445 446 447 448 449 450 451 452
{
	int error = radix_tree_preload(gfp_mask & ~__GFP_HIGHMEM);

	if (error == 0) {
		write_lock_irq(&mapping->tree_lock);
		error = radix_tree_insert(&mapping->page_tree, offset, page);
		if (!error) {
			page_cache_get(page);
			SetPageLocked(page);
			page->mapping = mapping;
			page->index = offset;
			mapping->nrpages++;
453
			__inc_zone_page_state(page, NR_FILE_PAGES);
L
Linus Torvalds 已提交
454 455 456 457 458 459 460 461 462
		}
		write_unlock_irq(&mapping->tree_lock);
		radix_tree_preload_end();
	}
	return error;
}
EXPORT_SYMBOL(add_to_page_cache);

int add_to_page_cache_lru(struct page *page, struct address_space *mapping,
A
Al Viro 已提交
463
				pgoff_t offset, gfp_t gfp_mask)
L
Linus Torvalds 已提交
464 465 466 467 468 469 470
{
	int ret = add_to_page_cache(page, mapping, offset, gfp_mask);
	if (ret == 0)
		lru_cache_add(page);
	return ret;
}

471
#ifdef CONFIG_NUMA
472
struct page *__page_cache_alloc(gfp_t gfp)
473 474 475
{
	if (cpuset_do_page_mem_spread()) {
		int n = cpuset_mem_spread_node();
476
		return alloc_pages_node(n, gfp, 0);
477
	}
478
	return alloc_pages(gfp, 0);
479
}
480
EXPORT_SYMBOL(__page_cache_alloc);
481 482
#endif

483 484 485 486 487 488
static int __sleep_on_page_lock(void *word)
{
	io_schedule();
	return 0;
}

L
Linus Torvalds 已提交
489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521
/*
 * In order to wait for pages to become available there must be
 * waitqueues associated with pages. By using a hash table of
 * waitqueues where the bucket discipline is to maintain all
 * waiters on the same queue and wake all when any of the pages
 * become available, and for the woken contexts to check to be
 * sure the appropriate page became available, this saves space
 * at a cost of "thundering herd" phenomena during rare hash
 * collisions.
 */
static wait_queue_head_t *page_waitqueue(struct page *page)
{
	const struct zone *zone = page_zone(page);

	return &zone->wait_table[hash_ptr(page, zone->wait_table_bits)];
}

static inline void wake_up_page(struct page *page, int bit)
{
	__wake_up_bit(page_waitqueue(page), &page->flags, bit);
}

void fastcall wait_on_page_bit(struct page *page, int bit_nr)
{
	DEFINE_WAIT_BIT(wait, &page->flags, bit_nr);

	if (test_bit(bit_nr, &page->flags))
		__wait_on_bit(page_waitqueue(page), &wait, sync_page,
							TASK_UNINTERRUPTIBLE);
}
EXPORT_SYMBOL(wait_on_page_bit);

/**
522
 * unlock_page - unlock a locked page
L
Linus Torvalds 已提交
523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544
 * @page: the page
 *
 * Unlocks the page and wakes up sleepers in ___wait_on_page_locked().
 * Also wakes sleepers in wait_on_page_writeback() because the wakeup
 * mechananism between PageLocked pages and PageWriteback pages is shared.
 * But that's OK - sleepers in wait_on_page_writeback() just go back to sleep.
 *
 * The first mb is necessary to safely close the critical section opened by the
 * TestSetPageLocked(), the second mb is necessary to enforce ordering between
 * the clear_bit and the read of the waitqueue (to avoid SMP races with a
 * parallel wait_on_page_locked()).
 */
void fastcall unlock_page(struct page *page)
{
	smp_mb__before_clear_bit();
	if (!TestClearPageLocked(page))
		BUG();
	smp_mb__after_clear_bit(); 
	wake_up_page(page, PG_locked);
}
EXPORT_SYMBOL(unlock_page);

545 546 547
/**
 * end_page_writeback - end writeback against a page
 * @page: the page
L
Linus Torvalds 已提交
548 549 550 551 552 553 554 555 556 557 558 559
 */
void end_page_writeback(struct page *page)
{
	if (!TestClearPageReclaim(page) || rotate_reclaimable_page(page)) {
		if (!test_clear_page_writeback(page))
			BUG();
	}
	smp_mb__after_clear_bit();
	wake_up_page(page, PG_writeback);
}
EXPORT_SYMBOL(end_page_writeback);

560 561 562
/**
 * __lock_page - get a lock on the page, assuming we need to sleep to get it
 * @page: the page to lock
L
Linus Torvalds 已提交
563
 *
564
 * Ugly. Running sync_page() in state TASK_UNINTERRUPTIBLE is scary.  If some
L
Linus Torvalds 已提交
565 566 567 568 569 570 571 572 573 574 575 576 577
 * random driver's requestfn sets TASK_RUNNING, we could busywait.  However
 * chances are that on the second loop, the block layer's plug list is empty,
 * so sync_page() will then return in state TASK_UNINTERRUPTIBLE.
 */
void fastcall __lock_page(struct page *page)
{
	DEFINE_WAIT_BIT(wait, &page->flags, PG_locked);

	__wait_on_bit_lock(page_waitqueue(page), &wait, sync_page,
							TASK_UNINTERRUPTIBLE);
}
EXPORT_SYMBOL(__lock_page);

578 579 580 581 582 583 584 585 586 587 588
/*
 * Variant of lock_page that does not require the caller to hold a reference
 * on the page's mapping.
 */
void fastcall __lock_page_nosync(struct page *page)
{
	DEFINE_WAIT_BIT(wait, &page->flags, PG_locked);
	__wait_on_bit_lock(page_waitqueue(page), &wait, __sleep_on_page_lock,
							TASK_UNINTERRUPTIBLE);
}

589 590 591 592 593
/**
 * find_get_page - find and get a page reference
 * @mapping: the address_space to search
 * @offset: the page index
 *
N
Nick Piggin 已提交
594 595
 * Is there a pagecache struct page at the given (mapping, offset) tuple?
 * If yes, increment its refcount and return it; if no, return NULL.
L
Linus Torvalds 已提交
596
 */
597
struct page * find_get_page(struct address_space *mapping, pgoff_t offset)
L
Linus Torvalds 已提交
598 599 600 601 602 603 604 605 606 607 608 609 610 611
{
	struct page *page;

	read_lock_irq(&mapping->tree_lock);
	page = radix_tree_lookup(&mapping->page_tree, offset);
	if (page)
		page_cache_get(page);
	read_unlock_irq(&mapping->tree_lock);
	return page;
}
EXPORT_SYMBOL(find_get_page);

/**
 * find_lock_page - locate, pin and lock a pagecache page
612 613
 * @mapping: the address_space to search
 * @offset: the page index
L
Linus Torvalds 已提交
614 615 616 617 618 619 620
 *
 * Locates the desired pagecache page, locks it, increments its reference
 * count and returns its address.
 *
 * Returns zero if the page was not present. find_lock_page() may sleep.
 */
struct page *find_lock_page(struct address_space *mapping,
621
				pgoff_t offset)
L
Linus Torvalds 已提交
622 623 624 625
{
	struct page *page;

repeat:
N
Nick Piggin 已提交
626
	read_lock_irq(&mapping->tree_lock);
L
Linus Torvalds 已提交
627 628 629 630 631
	page = radix_tree_lookup(&mapping->page_tree, offset);
	if (page) {
		page_cache_get(page);
		if (TestSetPageLocked(page)) {
			read_unlock_irq(&mapping->tree_lock);
632
			__lock_page(page);
L
Linus Torvalds 已提交
633 634

			/* Has the page been truncated while we slept? */
N
Nick Piggin 已提交
635
			if (unlikely(page->mapping != mapping)) {
L
Linus Torvalds 已提交
636 637 638 639
				unlock_page(page);
				page_cache_release(page);
				goto repeat;
			}
N
Nick Piggin 已提交
640 641
			VM_BUG_ON(page->index != offset);
			goto out;
L
Linus Torvalds 已提交
642 643 644
		}
	}
	read_unlock_irq(&mapping->tree_lock);
N
Nick Piggin 已提交
645
out:
L
Linus Torvalds 已提交
646 647 648 649 650 651
	return page;
}
EXPORT_SYMBOL(find_lock_page);

/**
 * find_or_create_page - locate or add a pagecache page
652 653 654
 * @mapping: the page's address_space
 * @index: the page's index into the mapping
 * @gfp_mask: page allocation mode
L
Linus Torvalds 已提交
655 656 657 658 659 660 661 662 663 664 665 666 667
 *
 * Locates a page in the pagecache.  If the page is not present, a new page
 * is allocated using @gfp_mask and is added to the pagecache and to the VM's
 * LRU list.  The returned page is locked and has its reference count
 * incremented.
 *
 * find_or_create_page() may sleep, even if @gfp_flags specifies an atomic
 * allocation!
 *
 * find_or_create_page() returns the desired page's address, or zero on
 * memory exhaustion.
 */
struct page *find_or_create_page(struct address_space *mapping,
668
		pgoff_t index, gfp_t gfp_mask)
L
Linus Torvalds 已提交
669
{
N
Nick Piggin 已提交
670
	struct page *page;
L
Linus Torvalds 已提交
671 672 673 674
	int err;
repeat:
	page = find_lock_page(mapping, index);
	if (!page) {
N
Nick Piggin 已提交
675 676 677 678 679 680 681 682 683
		page = __page_cache_alloc(gfp_mask);
		if (!page)
			return NULL;
		err = add_to_page_cache_lru(page, mapping, index, gfp_mask);
		if (unlikely(err)) {
			page_cache_release(page);
			page = NULL;
			if (err == -EEXIST)
				goto repeat;
L
Linus Torvalds 已提交
684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720
		}
	}
	return page;
}
EXPORT_SYMBOL(find_or_create_page);

/**
 * find_get_pages - gang pagecache lookup
 * @mapping:	The address_space to search
 * @start:	The starting page index
 * @nr_pages:	The maximum number of pages
 * @pages:	Where the resulting pages are placed
 *
 * find_get_pages() will search for and return a group of up to
 * @nr_pages pages in the mapping.  The pages are placed at @pages.
 * find_get_pages() takes a reference against the returned pages.
 *
 * The search returns a group of mapping-contiguous pages with ascending
 * indexes.  There may be holes in the indices due to not-present pages.
 *
 * find_get_pages() returns the number of pages which were found.
 */
unsigned find_get_pages(struct address_space *mapping, pgoff_t start,
			    unsigned int nr_pages, struct page **pages)
{
	unsigned int i;
	unsigned int ret;

	read_lock_irq(&mapping->tree_lock);
	ret = radix_tree_gang_lookup(&mapping->page_tree,
				(void **)pages, start, nr_pages);
	for (i = 0; i < ret; i++)
		page_cache_get(pages[i]);
	read_unlock_irq(&mapping->tree_lock);
	return ret;
}

721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751
/**
 * find_get_pages_contig - gang contiguous pagecache lookup
 * @mapping:	The address_space to search
 * @index:	The starting page index
 * @nr_pages:	The maximum number of pages
 * @pages:	Where the resulting pages are placed
 *
 * find_get_pages_contig() works exactly like find_get_pages(), except
 * that the returned number of pages are guaranteed to be contiguous.
 *
 * find_get_pages_contig() returns the number of pages which were found.
 */
unsigned find_get_pages_contig(struct address_space *mapping, pgoff_t index,
			       unsigned int nr_pages, struct page **pages)
{
	unsigned int i;
	unsigned int ret;

	read_lock_irq(&mapping->tree_lock);
	ret = radix_tree_gang_lookup(&mapping->page_tree,
				(void **)pages, index, nr_pages);
	for (i = 0; i < ret; i++) {
		if (pages[i]->mapping == NULL || pages[i]->index != index)
			break;

		page_cache_get(pages[i]);
		index++;
	}
	read_unlock_irq(&mapping->tree_lock);
	return i;
}
752
EXPORT_SYMBOL(find_get_pages_contig);
753

754 755 756 757 758 759 760 761
/**
 * find_get_pages_tag - find and return pages that match @tag
 * @mapping:	the address_space to search
 * @index:	the starting page index
 * @tag:	the tag index
 * @nr_pages:	the maximum number of pages
 * @pages:	where the resulting pages are placed
 *
L
Linus Torvalds 已提交
762
 * Like find_get_pages, except we only return pages which are tagged with
763
 * @tag.   We update @index to index the next page for the traversal.
L
Linus Torvalds 已提交
764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780
 */
unsigned find_get_pages_tag(struct address_space *mapping, pgoff_t *index,
			int tag, unsigned int nr_pages, struct page **pages)
{
	unsigned int i;
	unsigned int ret;

	read_lock_irq(&mapping->tree_lock);
	ret = radix_tree_gang_lookup_tag(&mapping->page_tree,
				(void **)pages, *index, nr_pages, tag);
	for (i = 0; i < ret; i++)
		page_cache_get(pages[i]);
	if (ret)
		*index = pages[ret - 1]->index + 1;
	read_unlock_irq(&mapping->tree_lock);
	return ret;
}
781
EXPORT_SYMBOL(find_get_pages_tag);
L
Linus Torvalds 已提交
782

783 784 785 786 787
/**
 * grab_cache_page_nowait - returns locked page at given index in given cache
 * @mapping: target address_space
 * @index: the page index
 *
788
 * Same as grab_cache_page(), but do not wait if the page is unavailable.
L
Linus Torvalds 已提交
789 790 791 792 793 794 795 796
 * This is intended for speculative data generators, where the data can
 * be regenerated if the page couldn't be grabbed.  This routine should
 * be safe to call while holding the lock for another page.
 *
 * Clear __GFP_FS when allocating the page to avoid recursion into the fs
 * and deadlock against the caller's locked page.
 */
struct page *
797
grab_cache_page_nowait(struct address_space *mapping, pgoff_t index)
L
Linus Torvalds 已提交
798 799 800 801 802 803 804 805 806
{
	struct page *page = find_get_page(mapping, index);

	if (page) {
		if (!TestSetPageLocked(page))
			return page;
		page_cache_release(page);
		return NULL;
	}
807 808
	page = __page_cache_alloc(mapping_gfp_mask(mapping) & ~__GFP_FS);
	if (page && add_to_page_cache_lru(page, mapping, index, GFP_KERNEL)) {
L
Linus Torvalds 已提交
809 810 811 812 813 814 815
		page_cache_release(page);
		page = NULL;
	}
	return page;
}
EXPORT_SYMBOL(grab_cache_page_nowait);

816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839
/*
 * CD/DVDs are error prone. When a medium error occurs, the driver may fail
 * a _large_ part of the i/o request. Imagine the worst scenario:
 *
 *      ---R__________________________________________B__________
 *         ^ reading here                             ^ bad block(assume 4k)
 *
 * read(R) => miss => readahead(R...B) => media error => frustrating retries
 * => failing the whole request => read(R) => read(R+1) =>
 * readahead(R+1...B+1) => bang => read(R+2) => read(R+3) =>
 * readahead(R+3...B+2) => bang => read(R+3) => read(R+4) =>
 * readahead(R+4...B+3) => bang => read(R+4) => read(R+5) => ......
 *
 * It is going insane. Fix it by quickly scaling down the readahead size.
 */
static void shrink_readahead_size_eio(struct file *filp,
					struct file_ra_state *ra)
{
	if (!ra->ra_pages)
		return;

	ra->ra_pages /= 4;
}

840 841 842
/**
 * do_generic_mapping_read - generic file read routine
 * @mapping:	address_space to be read
S
Stephen Hemminger 已提交
843
 * @_ra:	file's readahead state
844 845 846 847 848
 * @filp:	the file to read
 * @ppos:	current file position
 * @desc:	read_descriptor
 * @actor:	read method
 *
L
Linus Torvalds 已提交
849
 * This is a generic file read routine, and uses the
850
 * mapping->a_ops->readpage() function for the actual low-level stuff.
L
Linus Torvalds 已提交
851 852 853 854
 *
 * This is really ugly. But the goto's actually try to clarify some
 * of the logic when it comes to error handling etc.
 *
855 856
 * Note the struct file* is only passed for the use of readpage.
 * It may be NULL.
L
Linus Torvalds 已提交
857 858
 */
void do_generic_mapping_read(struct address_space *mapping,
859
			     struct file_ra_state *ra,
L
Linus Torvalds 已提交
860 861 862 863 864 865
			     struct file *filp,
			     loff_t *ppos,
			     read_descriptor_t *desc,
			     read_actor_t actor)
{
	struct inode *inode = mapping->host;
866 867 868 869
	pgoff_t index;
	pgoff_t last_index;
	pgoff_t prev_index;
	unsigned long offset;      /* offset into pagecache page */
870
	unsigned int prev_offset;
L
Linus Torvalds 已提交
871 872 873
	int error;

	index = *ppos >> PAGE_CACHE_SHIFT;
874 875
	prev_index = ra->prev_pos >> PAGE_CACHE_SHIFT;
	prev_offset = ra->prev_pos & (PAGE_CACHE_SIZE-1);
L
Linus Torvalds 已提交
876 877 878 879 880
	last_index = (*ppos + desc->count + PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT;
	offset = *ppos & ~PAGE_CACHE_MASK;

	for (;;) {
		struct page *page;
881
		pgoff_t end_index;
N
NeilBrown 已提交
882
		loff_t isize;
L
Linus Torvalds 已提交
883 884 885 886 887
		unsigned long nr, ret;

		cond_resched();
find_page:
		page = find_get_page(mapping, index);
888
		if (!page) {
889
			page_cache_sync_readahead(mapping,
890
					ra, filp,
891 892 893 894 895 896
					index, last_index - index);
			page = find_get_page(mapping, index);
			if (unlikely(page == NULL))
				goto no_cached_page;
		}
		if (PageReadahead(page)) {
897
			page_cache_async_readahead(mapping,
898
					ra, filp, page,
899
					index, last_index - index);
L
Linus Torvalds 已提交
900 901 902 903
		}
		if (!PageUptodate(page))
			goto page_not_up_to_date;
page_ok:
N
NeilBrown 已提交
904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929
		/*
		 * i_size must be checked after we know the page is Uptodate.
		 *
		 * Checking i_size after the check allows us to calculate
		 * the correct value for "nr", which means the zero-filled
		 * part of the page is not copied back to userspace (unless
		 * another truncate extends the file - this is desired though).
		 */

		isize = i_size_read(inode);
		end_index = (isize - 1) >> PAGE_CACHE_SHIFT;
		if (unlikely(!isize || index > end_index)) {
			page_cache_release(page);
			goto out;
		}

		/* nr is the maximum number of bytes to copy from this page */
		nr = PAGE_CACHE_SIZE;
		if (index == end_index) {
			nr = ((isize - 1) & ~PAGE_CACHE_MASK) + 1;
			if (nr <= offset) {
				page_cache_release(page);
				goto out;
			}
		}
		nr = nr - offset;
L
Linus Torvalds 已提交
930 931 932 933 934 935 936 937 938

		/* If users can be writing to this page using arbitrary
		 * virtual addresses, take care about potential aliasing
		 * before reading the page on the kernel side.
		 */
		if (mapping_writably_mapped(mapping))
			flush_dcache_page(page);

		/*
939 940
		 * When a sequential read accesses a page several times,
		 * only mark it as accessed the first time.
L
Linus Torvalds 已提交
941
		 */
942
		if (prev_index != index || offset != prev_offset)
L
Linus Torvalds 已提交
943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959
			mark_page_accessed(page);
		prev_index = index;

		/*
		 * Ok, we have the page, and it's up-to-date, so
		 * now we can copy it to user space...
		 *
		 * The actor routine returns how many bytes were actually used..
		 * NOTE! This may not be the same as how much of a user buffer
		 * we filled up (we may be padding etc), so we can only update
		 * "pos" here (the actor routine has to update the user buffer
		 * pointers and the remaining count).
		 */
		ret = actor(desc, page, offset, nr);
		offset += ret;
		index += offset >> PAGE_CACHE_SHIFT;
		offset &= ~PAGE_CACHE_MASK;
J
Jan Kara 已提交
960
		prev_offset = offset;
L
Linus Torvalds 已提交
961 962 963 964 965 966 967 968 969 970

		page_cache_release(page);
		if (ret == nr && desc->count)
			continue;
		goto out;

page_not_up_to_date:
		/* Get exclusive access to the page ... */
		lock_page(page);

N
Nick Piggin 已提交
971
		/* Did it get truncated before we got the lock? */
L
Linus Torvalds 已提交
972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987
		if (!page->mapping) {
			unlock_page(page);
			page_cache_release(page);
			continue;
		}

		/* Did somebody else fill it already? */
		if (PageUptodate(page)) {
			unlock_page(page);
			goto page_ok;
		}

readpage:
		/* Start the actual read. The read will unlock the page. */
		error = mapping->a_ops->readpage(filp, page);

988 989 990 991 992
		if (unlikely(error)) {
			if (error == AOP_TRUNCATED_PAGE) {
				page_cache_release(page);
				goto find_page;
			}
L
Linus Torvalds 已提交
993
			goto readpage_error;
994
		}
L
Linus Torvalds 已提交
995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008

		if (!PageUptodate(page)) {
			lock_page(page);
			if (!PageUptodate(page)) {
				if (page->mapping == NULL) {
					/*
					 * invalidate_inode_pages got it
					 */
					unlock_page(page);
					page_cache_release(page);
					goto find_page;
				}
				unlock_page(page);
				error = -EIO;
1009
				shrink_readahead_size_eio(filp, ra);
L
Linus Torvalds 已提交
1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027
				goto readpage_error;
			}
			unlock_page(page);
		}

		goto page_ok;

readpage_error:
		/* UHHUH! A synchronous read error occurred. Report it */
		desc->error = error;
		page_cache_release(page);
		goto out;

no_cached_page:
		/*
		 * Ok, it wasn't cached, so we need to create a new
		 * page..
		 */
N
Nick Piggin 已提交
1028 1029 1030 1031
		page = page_cache_alloc_cold(mapping);
		if (!page) {
			desc->error = -ENOMEM;
			goto out;
L
Linus Torvalds 已提交
1032
		}
N
Nick Piggin 已提交
1033
		error = add_to_page_cache_lru(page, mapping,
L
Linus Torvalds 已提交
1034 1035
						index, GFP_KERNEL);
		if (error) {
N
Nick Piggin 已提交
1036
			page_cache_release(page);
L
Linus Torvalds 已提交
1037 1038 1039 1040 1041 1042 1043 1044 1045
			if (error == -EEXIST)
				goto find_page;
			desc->error = error;
			goto out;
		}
		goto readpage;
	}

out:
1046 1047 1048
	ra->prev_pos = prev_index;
	ra->prev_pos <<= PAGE_CACHE_SHIFT;
	ra->prev_pos |= prev_offset;
L
Linus Torvalds 已提交
1049

1050
	*ppos = ((loff_t)index << PAGE_CACHE_SHIFT) + offset;
L
Linus Torvalds 已提交
1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093
	if (filp)
		file_accessed(filp);
}
EXPORT_SYMBOL(do_generic_mapping_read);

int file_read_actor(read_descriptor_t *desc, struct page *page,
			unsigned long offset, unsigned long size)
{
	char *kaddr;
	unsigned long left, count = desc->count;

	if (size > count)
		size = count;

	/*
	 * Faults on the destination of a read are common, so do it before
	 * taking the kmap.
	 */
	if (!fault_in_pages_writeable(desc->arg.buf, size)) {
		kaddr = kmap_atomic(page, KM_USER0);
		left = __copy_to_user_inatomic(desc->arg.buf,
						kaddr + offset, size);
		kunmap_atomic(kaddr, KM_USER0);
		if (left == 0)
			goto success;
	}

	/* Do it the slow way */
	kaddr = kmap(page);
	left = __copy_to_user(desc->arg.buf, kaddr + offset, size);
	kunmap(page);

	if (left) {
		size -= left;
		desc->error = -EFAULT;
	}
success:
	desc->count = count - size;
	desc->written += size;
	desc->arg.buf += size;
	return size;
}

1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132
/*
 * Performs necessary checks before doing a write
 * @iov:	io vector request
 * @nr_segs:	number of segments in the iovec
 * @count:	number of bytes to write
 * @access_flags: type of access: %VERIFY_READ or %VERIFY_WRITE
 *
 * Adjust number of segments and amount of bytes to write (nr_segs should be
 * properly initialized first). Returns appropriate error code that caller
 * should return or zero in case that write should be allowed.
 */
int generic_segment_checks(const struct iovec *iov,
			unsigned long *nr_segs, size_t *count, int access_flags)
{
	unsigned long   seg;
	size_t cnt = 0;
	for (seg = 0; seg < *nr_segs; seg++) {
		const struct iovec *iv = &iov[seg];

		/*
		 * If any segment has a negative length, or the cumulative
		 * length ever wraps negative then return -EINVAL.
		 */
		cnt += iv->iov_len;
		if (unlikely((ssize_t)(cnt|iv->iov_len) < 0))
			return -EINVAL;
		if (access_ok(access_flags, iv->iov_base, iv->iov_len))
			continue;
		if (seg == 0)
			return -EFAULT;
		*nr_segs = seg;
		cnt -= iv->iov_len;	/* This segment is no good */
		break;
	}
	*count = cnt;
	return 0;
}
EXPORT_SYMBOL(generic_segment_checks);

1133
/**
H
Henrik Kretzschmar 已提交
1134
 * generic_file_aio_read - generic filesystem read routine
1135 1136 1137
 * @iocb:	kernel I/O control block
 * @iov:	io vector request
 * @nr_segs:	number of segments in the iovec
H
Henrik Kretzschmar 已提交
1138
 * @pos:	current file position
1139
 *
L
Linus Torvalds 已提交
1140 1141 1142 1143
 * This is the "read()" routine for all filesystems
 * that can use the page cache directly.
 */
ssize_t
1144 1145
generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
		unsigned long nr_segs, loff_t pos)
L
Linus Torvalds 已提交
1146 1147 1148 1149 1150
{
	struct file *filp = iocb->ki_filp;
	ssize_t retval;
	unsigned long seg;
	size_t count;
1151
	loff_t *ppos = &iocb->ki_pos;
L
Linus Torvalds 已提交
1152 1153

	count = 0;
1154 1155 1156
	retval = generic_segment_checks(iov, &nr_segs, &count, VERIFY_WRITE);
	if (retval)
		return retval;
L
Linus Torvalds 已提交
1157 1158 1159

	/* coalesce the iovecs and go direct-to-BIO for O_DIRECT */
	if (filp->f_flags & O_DIRECT) {
1160
		loff_t size;
L
Linus Torvalds 已提交
1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175
		struct address_space *mapping;
		struct inode *inode;

		mapping = filp->f_mapping;
		inode = mapping->host;
		retval = 0;
		if (!count)
			goto out; /* skip atime */
		size = i_size_read(inode);
		if (pos < size) {
			retval = generic_file_direct_IO(READ, iocb,
						iov, pos, nr_segs);
			if (retval > 0)
				*ppos = pos + retval;
		}
1176
		if (likely(retval != 0)) {
1177
			file_accessed(filp);
S
Steven Whitehouse 已提交
1178
			goto out;
1179
		}
L
Linus Torvalds 已提交
1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194
	}

	retval = 0;
	if (count) {
		for (seg = 0; seg < nr_segs; seg++) {
			read_descriptor_t desc;

			desc.written = 0;
			desc.arg.buf = iov[seg].iov_base;
			desc.count = iov[seg].iov_len;
			if (desc.count == 0)
				continue;
			desc.error = 0;
			do_generic_file_read(filp,ppos,&desc,file_read_actor);
			retval += desc.written;
1195 1196
			if (desc.error) {
				retval = retval ?: desc.error;
L
Linus Torvalds 已提交
1197 1198
				break;
			}
1199 1200
			if (desc.count > 0)
				break;
L
Linus Torvalds 已提交
1201 1202 1203 1204 1205 1206 1207 1208 1209
		}
	}
out:
	return retval;
}
EXPORT_SYMBOL(generic_file_aio_read);

static ssize_t
do_readahead(struct address_space *mapping, struct file *filp,
1210
	     pgoff_t index, unsigned long nr)
L
Linus Torvalds 已提交
1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229
{
	if (!mapping || !mapping->a_ops || !mapping->a_ops->readpage)
		return -EINVAL;

	force_page_cache_readahead(mapping, filp, index,
					max_sane_readahead(nr));
	return 0;
}

asmlinkage ssize_t sys_readahead(int fd, loff_t offset, size_t count)
{
	ssize_t ret;
	struct file *file;

	ret = -EBADF;
	file = fget(fd);
	if (file) {
		if (file->f_mode & FMODE_READ) {
			struct address_space *mapping = file->f_mapping;
1230 1231
			pgoff_t start = offset >> PAGE_CACHE_SHIFT;
			pgoff_t end = (offset + count - 1) >> PAGE_CACHE_SHIFT;
L
Linus Torvalds 已提交
1232 1233 1234 1235 1236 1237 1238 1239 1240
			unsigned long len = end - start + 1;
			ret = do_readahead(mapping, file, start, len);
		}
		fput(file);
	}
	return ret;
}

#ifdef CONFIG_MMU
1241 1242 1243 1244 1245
/**
 * page_cache_read - adds requested page to the page cache if not already there
 * @file:	file to read
 * @offset:	page index
 *
L
Linus Torvalds 已提交
1246 1247 1248
 * This adds the requested page to the page cache if it isn't already there,
 * and schedules an I/O to read in its contents from disk.
 */
1249
static int fastcall page_cache_read(struct file * file, pgoff_t offset)
L
Linus Torvalds 已提交
1250 1251 1252
{
	struct address_space *mapping = file->f_mapping;
	struct page *page; 
1253
	int ret;
L
Linus Torvalds 已提交
1254

1255 1256 1257 1258 1259 1260 1261 1262 1263 1264
	do {
		page = page_cache_alloc_cold(mapping);
		if (!page)
			return -ENOMEM;

		ret = add_to_page_cache_lru(page, mapping, offset, GFP_KERNEL);
		if (ret == 0)
			ret = mapping->a_ops->readpage(file, page);
		else if (ret == -EEXIST)
			ret = 0; /* losing race to add is OK */
L
Linus Torvalds 已提交
1265 1266 1267

		page_cache_release(page);

1268 1269 1270
	} while (ret == AOP_TRUNCATED_PAGE);
		
	return ret;
L
Linus Torvalds 已提交
1271 1272 1273 1274
}

#define MMAP_LOTSAMISS  (100)

1275
/**
1276
 * filemap_fault - read in file data for page fault handling
N
Nick Piggin 已提交
1277 1278
 * @vma:	vma in which the fault was taken
 * @vmf:	struct vm_fault containing details of the fault
1279
 *
1280
 * filemap_fault() is invoked via the vma operations vector for a
L
Linus Torvalds 已提交
1281 1282 1283 1284 1285 1286
 * mapped memory region to read in file data during a page fault.
 *
 * The goto's are kind of ugly, but this streamlines the normal case of having
 * it in the page cache, and handles the special cases reasonably without
 * having a lot of duplicated code.
 */
N
Nick Piggin 已提交
1287
int filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
L
Linus Torvalds 已提交
1288 1289
{
	int error;
1290
	struct file *file = vma->vm_file;
L
Linus Torvalds 已提交
1291 1292 1293 1294
	struct address_space *mapping = file->f_mapping;
	struct file_ra_state *ra = &file->f_ra;
	struct inode *inode = mapping->host;
	struct page *page;
1295 1296
	unsigned long size;
	int did_readaround = 0;
N
Nick Piggin 已提交
1297
	int ret = 0;
L
Linus Torvalds 已提交
1298 1299

	size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
N
Nick Piggin 已提交
1300
	if (vmf->pgoff >= size)
L
Linus Torvalds 已提交
1301 1302 1303
		goto outside_data_content;

	/* If we don't want any read-ahead, don't bother */
1304
	if (VM_RandomReadHint(vma))
L
Linus Torvalds 已提交
1305 1306 1307 1308 1309 1310
		goto no_cached_page;

	/*
	 * Do we have something in the page cache already?
	 */
retry_find:
N
Nick Piggin 已提交
1311
	page = find_lock_page(mapping, vmf->pgoff);
1312 1313 1314 1315 1316
	/*
	 * For sequential accesses, we use the generic readahead logic.
	 */
	if (VM_SequentialReadHint(vma)) {
		if (!page) {
1317
			page_cache_sync_readahead(mapping, ra, file,
1318 1319 1320 1321 1322 1323
							   vmf->pgoff, 1);
			page = find_lock_page(mapping, vmf->pgoff);
			if (!page)
				goto no_cached_page;
		}
		if (PageReadahead(page)) {
1324
			page_cache_async_readahead(mapping, ra, file, page,
1325 1326 1327 1328
							   vmf->pgoff, 1);
		}
	}

L
Linus Torvalds 已提交
1329 1330 1331 1332 1333 1334 1335 1336 1337
	if (!page) {
		unsigned long ra_pages;

		ra->mmap_miss++;

		/*
		 * Do we miss much more than hit in this file? If so,
		 * stop bothering with read-ahead. It will only hurt.
		 */
1338
		if (ra->mmap_miss > MMAP_LOTSAMISS)
L
Linus Torvalds 已提交
1339 1340 1341 1342 1343 1344 1345
			goto no_cached_page;

		/*
		 * To keep the pgmajfault counter straight, we need to
		 * check did_readaround, as this is an inner loop.
		 */
		if (!did_readaround) {
N
Nick Piggin 已提交
1346
			ret = VM_FAULT_MAJOR;
1347
			count_vm_event(PGMAJFAULT);
L
Linus Torvalds 已提交
1348 1349 1350 1351 1352 1353
		}
		did_readaround = 1;
		ra_pages = max_sane_readahead(file->f_ra.ra_pages);
		if (ra_pages) {
			pgoff_t start = 0;

N
Nick Piggin 已提交
1354 1355
			if (vmf->pgoff > ra_pages / 2)
				start = vmf->pgoff - ra_pages / 2;
L
Linus Torvalds 已提交
1356 1357
			do_page_cache_readahead(mapping, file, start, ra_pages);
		}
N
Nick Piggin 已提交
1358
		page = find_lock_page(mapping, vmf->pgoff);
L
Linus Torvalds 已提交
1359 1360 1361 1362 1363
		if (!page)
			goto no_cached_page;
	}

	if (!did_readaround)
1364
		ra->mmap_miss--;
L
Linus Torvalds 已提交
1365 1366

	/*
1367 1368
	 * We have a locked page in the page cache, now we need to check
	 * that it's up-to-date. If not, it is going to be due to an error.
L
Linus Torvalds 已提交
1369
	 */
1370
	if (unlikely(!PageUptodate(page)))
L
Linus Torvalds 已提交
1371 1372
		goto page_not_uptodate;

1373 1374
	/* Must recheck i_size under page lock */
	size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
N
Nick Piggin 已提交
1375
	if (unlikely(vmf->pgoff >= size)) {
1376
		unlock_page(page);
1377
		page_cache_release(page);
1378 1379 1380
		goto outside_data_content;
	}

L
Linus Torvalds 已提交
1381 1382 1383 1384
	/*
	 * Found the page and have a reference on it.
	 */
	mark_page_accessed(page);
1385
	ra->prev_pos = (loff_t)page->index << PAGE_CACHE_SHIFT;
N
Nick Piggin 已提交
1386
	vmf->page = page;
N
Nick Piggin 已提交
1387
	return ret | VM_FAULT_LOCKED;
L
Linus Torvalds 已提交
1388 1389 1390 1391 1392 1393

outside_data_content:
	/*
	 * An external ptracer can access pages that normally aren't
	 * accessible..
	 */
N
Nick Piggin 已提交
1394 1395 1396
	if (vma->vm_mm == current->mm)
		return VM_FAULT_SIGBUS;

L
Linus Torvalds 已提交
1397 1398 1399 1400 1401 1402
	/* Fall through to the non-read-ahead case */
no_cached_page:
	/*
	 * We're only likely to ever get here if MADV_RANDOM is in
	 * effect.
	 */
N
Nick Piggin 已提交
1403
	error = page_cache_read(file, vmf->pgoff);
L
Linus Torvalds 已提交
1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418

	/*
	 * The page we want has now been added to the page cache.
	 * In the unlikely event that someone removed it in the
	 * meantime, we'll just come back here and read it again.
	 */
	if (error >= 0)
		goto retry_find;

	/*
	 * An error return from page_cache_read can result if the
	 * system is low on memory, or a problem occurs while trying
	 * to schedule I/O.
	 */
	if (error == -ENOMEM)
N
Nick Piggin 已提交
1419 1420
		return VM_FAULT_OOM;
	return VM_FAULT_SIGBUS;
L
Linus Torvalds 已提交
1421 1422

page_not_uptodate:
1423
	/* IO error path */
L
Linus Torvalds 已提交
1424
	if (!did_readaround) {
N
Nick Piggin 已提交
1425
		ret = VM_FAULT_MAJOR;
1426
		count_vm_event(PGMAJFAULT);
L
Linus Torvalds 已提交
1427 1428 1429 1430 1431 1432 1433 1434 1435
	}

	/*
	 * Umm, take care of errors if the page isn't up-to-date.
	 * Try to re-read it _once_. We do this synchronously,
	 * because there really aren't any performance issues here
	 * and we need to check for errors.
	 */
	ClearPageError(page);
1436
	error = mapping->a_ops->readpage(file, page);
1437 1438 1439
	page_cache_release(page);

	if (!error || error == AOP_TRUNCATED_PAGE)
1440
		goto retry_find;
L
Linus Torvalds 已提交
1441

1442
	/* Things didn't work out. Return zero to tell the mm layer so. */
1443
	shrink_readahead_size_eio(file, ra);
N
Nick Piggin 已提交
1444
	return VM_FAULT_SIGBUS;
1445 1446 1447
}
EXPORT_SYMBOL(filemap_fault);

L
Linus Torvalds 已提交
1448
struct vm_operations_struct generic_file_vm_ops = {
1449
	.fault		= filemap_fault,
L
Linus Torvalds 已提交
1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461
};

/* This is used for a general mmap of a disk file */

int generic_file_mmap(struct file * file, struct vm_area_struct * vma)
{
	struct address_space *mapping = file->f_mapping;

	if (!mapping->a_ops->readpage)
		return -ENOEXEC;
	file_accessed(file);
	vma->vm_ops = &generic_file_vm_ops;
N
Nick Piggin 已提交
1462
	vma->vm_flags |= VM_CAN_NONLINEAR;
L
Linus Torvalds 已提交
1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488
	return 0;
}

/*
 * This is for filesystems which do not implement ->writepage.
 */
int generic_file_readonly_mmap(struct file *file, struct vm_area_struct *vma)
{
	if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_MAYWRITE))
		return -EINVAL;
	return generic_file_mmap(file, vma);
}
#else
int generic_file_mmap(struct file * file, struct vm_area_struct * vma)
{
	return -ENOSYS;
}
int generic_file_readonly_mmap(struct file * file, struct vm_area_struct * vma)
{
	return -ENOSYS;
}
#endif /* CONFIG_MMU */

EXPORT_SYMBOL(generic_file_mmap);
EXPORT_SYMBOL(generic_file_readonly_mmap);

1489
static struct page *__read_cache_page(struct address_space *mapping,
1490
				pgoff_t index,
L
Linus Torvalds 已提交
1491 1492 1493
				int (*filler)(void *,struct page*),
				void *data)
{
N
Nick Piggin 已提交
1494
	struct page *page;
L
Linus Torvalds 已提交
1495 1496 1497 1498
	int err;
repeat:
	page = find_get_page(mapping, index);
	if (!page) {
N
Nick Piggin 已提交
1499 1500 1501 1502 1503 1504 1505 1506
		page = page_cache_alloc_cold(mapping);
		if (!page)
			return ERR_PTR(-ENOMEM);
		err = add_to_page_cache_lru(page, mapping, index, GFP_KERNEL);
		if (unlikely(err)) {
			page_cache_release(page);
			if (err == -EEXIST)
				goto repeat;
L
Linus Torvalds 已提交
1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518
			/* Presumably ENOMEM for radix tree node */
			return ERR_PTR(err);
		}
		err = filler(data, page);
		if (err < 0) {
			page_cache_release(page);
			page = ERR_PTR(err);
		}
	}
	return page;
}

1519 1520 1521
/*
 * Same as read_cache_page, but don't wait for page to become unlocked
 * after submitting it to the filler.
L
Linus Torvalds 已提交
1522
 */
1523
struct page *read_cache_page_async(struct address_space *mapping,
1524
				pgoff_t index,
L
Linus Torvalds 已提交
1525 1526 1527 1528 1529 1530 1531 1532 1533
				int (*filler)(void *,struct page*),
				void *data)
{
	struct page *page;
	int err;

retry:
	page = __read_cache_page(mapping, index, filler, data);
	if (IS_ERR(page))
1534
		return page;
L
Linus Torvalds 已提交
1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550
	if (PageUptodate(page))
		goto out;

	lock_page(page);
	if (!page->mapping) {
		unlock_page(page);
		page_cache_release(page);
		goto retry;
	}
	if (PageUptodate(page)) {
		unlock_page(page);
		goto out;
	}
	err = filler(data, page);
	if (err < 0) {
		page_cache_release(page);
1551
		return ERR_PTR(err);
L
Linus Torvalds 已提交
1552
	}
1553
out:
1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571
	mark_page_accessed(page);
	return page;
}
EXPORT_SYMBOL(read_cache_page_async);

/**
 * read_cache_page - read into page cache, fill it if needed
 * @mapping:	the page's address_space
 * @index:	the page index
 * @filler:	function to perform the read
 * @data:	destination for read data
 *
 * Read into the page cache. If a page already exists, and PageUptodate() is
 * not set, try to fill the page then wait for it to become unlocked.
 *
 * If the page does not get brought uptodate, return -EIO.
 */
struct page *read_cache_page(struct address_space *mapping,
1572
				pgoff_t index,
1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585
				int (*filler)(void *,struct page*),
				void *data)
{
	struct page *page;

	page = read_cache_page_async(mapping, index, filler, data);
	if (IS_ERR(page))
		goto out;
	wait_on_page_locked(page);
	if (!PageUptodate(page)) {
		page_cache_release(page);
		page = ERR_PTR(-EIO);
	}
L
Linus Torvalds 已提交
1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596
 out:
	return page;
}
EXPORT_SYMBOL(read_cache_page);

/*
 * The logic we want is
 *
 *	if suid or (sgid and xgrp)
 *		remove privs
 */
1597
int should_remove_suid(struct dentry *dentry)
L
Linus Torvalds 已提交
1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612
{
	mode_t mode = dentry->d_inode->i_mode;
	int kill = 0;

	/* suid always must be killed */
	if (unlikely(mode & S_ISUID))
		kill = ATTR_KILL_SUID;

	/*
	 * sgid without any exec bits is just a mandatory locking mark; leave
	 * it alone.  If some exec bits are set, it's a real sgid; kill it.
	 */
	if (unlikely((mode & S_ISGID) && (mode & S_IXGRP)))
		kill |= ATTR_KILL_SGID;

1613 1614
	if (unlikely(kill && !capable(CAP_FSETID)))
		return kill;
L
Linus Torvalds 已提交
1615

1616 1617
	return 0;
}
M
Mark Fasheh 已提交
1618
EXPORT_SYMBOL(should_remove_suid);
1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635

int __remove_suid(struct dentry *dentry, int kill)
{
	struct iattr newattrs;

	newattrs.ia_valid = ATTR_FORCE | kill;
	return notify_change(dentry, &newattrs);
}

int remove_suid(struct dentry *dentry)
{
	int kill = should_remove_suid(dentry);

	if (unlikely(kill))
		return __remove_suid(dentry, kill);

	return 0;
L
Linus Torvalds 已提交
1636 1637 1638
}
EXPORT_SYMBOL(remove_suid);

N
Nick Piggin 已提交
1639
static size_t __iovec_copy_from_user_inatomic(char *vaddr,
L
Linus Torvalds 已提交
1640 1641 1642 1643 1644 1645 1646 1647 1648
			const struct iovec *iov, size_t base, size_t bytes)
{
	size_t copied = 0, left = 0;

	while (bytes) {
		char __user *buf = iov->iov_base + base;
		int copy = min(bytes, iov->iov_len - base);

		base = 0;
1649
		left = __copy_from_user_inatomic_nocache(vaddr, buf, copy);
L
Linus Torvalds 已提交
1650 1651 1652 1653 1654
		copied += copy;
		bytes -= copy;
		vaddr += copy;
		iov++;

1655
		if (unlikely(left))
L
Linus Torvalds 已提交
1656 1657 1658 1659 1660
			break;
	}
	return copied - left;
}

N
Nick Piggin 已提交
1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687
/*
 * Copy as much as we can into the page and return the number of bytes which
 * were sucessfully copied.  If a fault is encountered then return the number of
 * bytes which were copied.
 */
size_t iov_iter_copy_from_user_atomic(struct page *page,
		struct iov_iter *i, unsigned long offset, size_t bytes)
{
	char *kaddr;
	size_t copied;

	BUG_ON(!in_atomic());
	kaddr = kmap_atomic(page, KM_USER0);
	if (likely(i->nr_segs == 1)) {
		int left;
		char __user *buf = i->iov->iov_base + i->iov_offset;
		left = __copy_from_user_inatomic_nocache(kaddr + offset,
							buf, bytes);
		copied = bytes - left;
	} else {
		copied = __iovec_copy_from_user_inatomic(kaddr + offset,
						i->iov, i->iov_offset, bytes);
	}
	kunmap_atomic(kaddr, KM_USER0);

	return copied;
}
N
Nick Piggin 已提交
1688
EXPORT_SYMBOL(iov_iter_copy_from_user_atomic);
N
Nick Piggin 已提交
1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714

/*
 * This has the same sideeffects and return value as
 * iov_iter_copy_from_user_atomic().
 * The difference is that it attempts to resolve faults.
 * Page must not be locked.
 */
size_t iov_iter_copy_from_user(struct page *page,
		struct iov_iter *i, unsigned long offset, size_t bytes)
{
	char *kaddr;
	size_t copied;

	kaddr = kmap(page);
	if (likely(i->nr_segs == 1)) {
		int left;
		char __user *buf = i->iov->iov_base + i->iov_offset;
		left = __copy_from_user_nocache(kaddr + offset, buf, bytes);
		copied = bytes - left;
	} else {
		copied = __iovec_copy_from_user_inatomic(kaddr + offset,
						i->iov, i->iov_offset, bytes);
	}
	kunmap(page);
	return copied;
}
N
Nick Piggin 已提交
1715
EXPORT_SYMBOL(iov_iter_copy_from_user);
N
Nick Piggin 已提交
1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746

static void __iov_iter_advance_iov(struct iov_iter *i, size_t bytes)
{
	if (likely(i->nr_segs == 1)) {
		i->iov_offset += bytes;
	} else {
		const struct iovec *iov = i->iov;
		size_t base = i->iov_offset;

		while (bytes) {
			int copy = min(bytes, iov->iov_len - base);

			bytes -= copy;
			base += copy;
			if (iov->iov_len == base) {
				iov++;
				base = 0;
			}
		}
		i->iov = iov;
		i->iov_offset = base;
	}
}

void iov_iter_advance(struct iov_iter *i, size_t bytes)
{
	BUG_ON(i->count < bytes);

	__iov_iter_advance_iov(i, bytes);
	i->count -= bytes;
}
N
Nick Piggin 已提交
1747
EXPORT_SYMBOL(iov_iter_advance);
N
Nick Piggin 已提交
1748

1749 1750 1751 1752 1753 1754 1755 1756 1757 1758
/*
 * Fault in the first iovec of the given iov_iter, to a maximum length
 * of bytes. Returns 0 on success, or non-zero if the memory could not be
 * accessed (ie. because it is an invalid address).
 *
 * writev-intensive code may want this to prefault several iovecs -- that
 * would be possible (callers must not rely on the fact that _only_ the
 * first iovec will be faulted with the current implementation).
 */
int iov_iter_fault_in_readable(struct iov_iter *i, size_t bytes)
N
Nick Piggin 已提交
1759 1760
{
	char __user *buf = i->iov->iov_base + i->iov_offset;
1761 1762
	bytes = min(bytes, i->iov->iov_len - i->iov_offset);
	return fault_in_pages_readable(buf, bytes);
N
Nick Piggin 已提交
1763
}
N
Nick Piggin 已提交
1764
EXPORT_SYMBOL(iov_iter_fault_in_readable);
N
Nick Piggin 已提交
1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776

/*
 * Return the count of just the current iov_iter segment.
 */
size_t iov_iter_single_seg_count(struct iov_iter *i)
{
	const struct iovec *iov = i->iov;
	if (i->nr_segs == 1)
		return i->count;
	else
		return min(i->count, iov->iov_len - i->iov_offset);
}
N
Nick Piggin 已提交
1777
EXPORT_SYMBOL(iov_iter_single_seg_count);
N
Nick Piggin 已提交
1778

L
Linus Torvalds 已提交
1779 1780 1781
/*
 * Performs necessary checks before doing a write
 *
1782
 * Can adjust writing position or amount of bytes to write.
L
Linus Torvalds 已提交
1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840
 * Returns appropriate error code that caller should return or
 * zero in case that write should be allowed.
 */
inline int generic_write_checks(struct file *file, loff_t *pos, size_t *count, int isblk)
{
	struct inode *inode = file->f_mapping->host;
	unsigned long limit = current->signal->rlim[RLIMIT_FSIZE].rlim_cur;

        if (unlikely(*pos < 0))
                return -EINVAL;

	if (!isblk) {
		/* FIXME: this is for backwards compatibility with 2.4 */
		if (file->f_flags & O_APPEND)
                        *pos = i_size_read(inode);

		if (limit != RLIM_INFINITY) {
			if (*pos >= limit) {
				send_sig(SIGXFSZ, current, 0);
				return -EFBIG;
			}
			if (*count > limit - (typeof(limit))*pos) {
				*count = limit - (typeof(limit))*pos;
			}
		}
	}

	/*
	 * LFS rule
	 */
	if (unlikely(*pos + *count > MAX_NON_LFS &&
				!(file->f_flags & O_LARGEFILE))) {
		if (*pos >= MAX_NON_LFS) {
			return -EFBIG;
		}
		if (*count > MAX_NON_LFS - (unsigned long)*pos) {
			*count = MAX_NON_LFS - (unsigned long)*pos;
		}
	}

	/*
	 * Are we about to exceed the fs block limit ?
	 *
	 * If we have written data it becomes a short write.  If we have
	 * exceeded without writing data we send a signal and return EFBIG.
	 * Linus frestrict idea will clean these up nicely..
	 */
	if (likely(!isblk)) {
		if (unlikely(*pos >= inode->i_sb->s_maxbytes)) {
			if (*count || *pos > inode->i_sb->s_maxbytes) {
				return -EFBIG;
			}
			/* zero-length writes at ->s_maxbytes are OK */
		}

		if (unlikely(*pos + *count > inode->i_sb->s_maxbytes))
			*count = inode->i_sb->s_maxbytes - *pos;
	} else {
1841
#ifdef CONFIG_BLOCK
L
Linus Torvalds 已提交
1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852
		loff_t isize;
		if (bdev_read_only(I_BDEV(inode)))
			return -EPERM;
		isize = i_size_read(inode);
		if (*pos >= isize) {
			if (*count || *pos > isize)
				return -ENOSPC;
		}

		if (*pos + *count > isize)
			*count = isize - *pos;
1853 1854 1855
#else
		return -EPERM;
#endif
L
Linus Torvalds 已提交
1856 1857 1858 1859 1860
	}
	return 0;
}
EXPORT_SYMBOL(generic_write_checks);

1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901
int pagecache_write_begin(struct file *file, struct address_space *mapping,
				loff_t pos, unsigned len, unsigned flags,
				struct page **pagep, void **fsdata)
{
	const struct address_space_operations *aops = mapping->a_ops;

	if (aops->write_begin) {
		return aops->write_begin(file, mapping, pos, len, flags,
							pagep, fsdata);
	} else {
		int ret;
		pgoff_t index = pos >> PAGE_CACHE_SHIFT;
		unsigned offset = pos & (PAGE_CACHE_SIZE - 1);
		struct inode *inode = mapping->host;
		struct page *page;
again:
		page = __grab_cache_page(mapping, index);
		*pagep = page;
		if (!page)
			return -ENOMEM;

		if (flags & AOP_FLAG_UNINTERRUPTIBLE && !PageUptodate(page)) {
			/*
			 * There is no way to resolve a short write situation
			 * for a !Uptodate page (except by double copying in
			 * the caller done by generic_perform_write_2copy).
			 *
			 * Instead, we have to bring it uptodate here.
			 */
			ret = aops->readpage(file, page);
			page_cache_release(page);
			if (ret) {
				if (ret == AOP_TRUNCATED_PAGE)
					goto again;
				return ret;
			}
			goto again;
		}

		ret = aops->prepare_write(file, page, offset, offset+len);
		if (ret) {
N
Nick Piggin 已提交
1902
			unlock_page(page);
1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945
			page_cache_release(page);
			if (pos + len > inode->i_size)
				vmtruncate(inode, inode->i_size);
		}
		return ret;
	}
}
EXPORT_SYMBOL(pagecache_write_begin);

int pagecache_write_end(struct file *file, struct address_space *mapping,
				loff_t pos, unsigned len, unsigned copied,
				struct page *page, void *fsdata)
{
	const struct address_space_operations *aops = mapping->a_ops;
	int ret;

	if (aops->write_end) {
		mark_page_accessed(page);
		ret = aops->write_end(file, mapping, pos, len, copied,
							page, fsdata);
	} else {
		unsigned offset = pos & (PAGE_CACHE_SIZE - 1);
		struct inode *inode = mapping->host;

		flush_dcache_page(page);
		ret = aops->commit_write(file, page, offset, offset+len);
		unlock_page(page);
		mark_page_accessed(page);
		page_cache_release(page);

		if (ret < 0) {
			if (pos + len > inode->i_size)
				vmtruncate(inode, inode->i_size);
		} else if (ret > 0)
			ret = min_t(size_t, copied, ret);
		else
			ret = copied;
	}

	return ret;
}
EXPORT_SYMBOL(pagecache_write_end);

L
Linus Torvalds 已提交
1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971
ssize_t
generic_file_direct_write(struct kiocb *iocb, const struct iovec *iov,
		unsigned long *nr_segs, loff_t pos, loff_t *ppos,
		size_t count, size_t ocount)
{
	struct file	*file = iocb->ki_filp;
	struct address_space *mapping = file->f_mapping;
	struct inode	*inode = mapping->host;
	ssize_t		written;

	if (count != ocount)
		*nr_segs = iov_shorten((struct iovec *)iov, *nr_segs, count);

	written = generic_file_direct_IO(WRITE, iocb, iov, pos, *nr_segs);
	if (written > 0) {
		loff_t end = pos + written;
		if (end > i_size_read(inode) && !S_ISBLK(inode->i_mode)) {
			i_size_write(inode,  end);
			mark_inode_dirty(inode);
		}
		*ppos = end;
	}

	/*
	 * Sync the fs metadata but not the minor inode changes and
	 * of course not the data as we did direct DMA for the IO.
1972
	 * i_mutex is held, which protects generic_osync_inode() from
1973
	 * livelocking.  AIO O_DIRECT ops attempt to sync metadata here.
L
Linus Torvalds 已提交
1974
	 */
1975 1976
	if ((written >= 0 || written == -EIOCBQUEUED) &&
	    ((file->f_flags & O_SYNC) || IS_SYNC(inode))) {
1977 1978 1979 1980
		int err = generic_osync_inode(inode, mapping, OSYNC_METADATA);
		if (err < 0)
			written = err;
	}
L
Linus Torvalds 已提交
1981 1982 1983 1984
	return written;
}
EXPORT_SYMBOL(generic_file_direct_write);

N
Nick Piggin 已提交
1985 1986 1987 1988
/*
 * Find or create a page at the given pagecache position. Return the locked
 * page. This function is specifically for buffered writes.
 */
1989
struct page *__grab_cache_page(struct address_space *mapping, pgoff_t index)
N
Nick Piggin 已提交
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
{
	int status;
	struct page *page;
repeat:
	page = find_lock_page(mapping, index);
	if (likely(page))
		return page;

	page = page_cache_alloc(mapping);
	if (!page)
		return NULL;
	status = add_to_page_cache_lru(page, mapping, index, GFP_KERNEL);
	if (unlikely(status)) {
		page_cache_release(page);
		if (status == -EEXIST)
			goto repeat;
		return NULL;
	}
	return page;
}
2010
EXPORT_SYMBOL(__grab_cache_page);
N
Nick Piggin 已提交
2011

2012 2013
static ssize_t generic_perform_write_2copy(struct file *file,
				struct iov_iter *i, loff_t pos)
L
Linus Torvalds 已提交
2014
{
A
Andrew Morton 已提交
2015
	struct address_space *mapping = file->f_mapping;
2016
	const struct address_space_operations *a_ops = mapping->a_ops;
2017 2018 2019
	struct inode *inode = mapping->host;
	long status = 0;
	ssize_t written = 0;
L
Linus Torvalds 已提交
2020 2021

	do {
N
Nick Piggin 已提交
2022
		struct page *src_page;
N
Nick Piggin 已提交
2023
		struct page *page;
A
Andrew Morton 已提交
2024 2025
		pgoff_t index;		/* Pagecache index for current page */
		unsigned long offset;	/* Offset into pagecache page */
N
Nick Piggin 已提交
2026
		unsigned long bytes;	/* Bytes to write to page */
A
Andrew Morton 已提交
2027
		size_t copied;		/* Bytes copied from user */
L
Linus Torvalds 已提交
2028

A
Andrew Morton 已提交
2029
		offset = (pos & (PAGE_CACHE_SIZE - 1));
L
Linus Torvalds 已提交
2030
		index = pos >> PAGE_CACHE_SHIFT;
N
Nick Piggin 已提交
2031
		bytes = min_t(unsigned long, PAGE_CACHE_SIZE - offset,
2032
						iov_iter_count(i));
2033

N
Nick Piggin 已提交
2034 2035 2036 2037 2038 2039
		/*
		 * a non-NULL src_page indicates that we're doing the
		 * copy via get_user_pages and kmap.
		 */
		src_page = NULL;

2040 2041 2042 2043 2044
		/*
		 * Bring in the user page that we will copy from _first_.
		 * Otherwise there's a nasty deadlock on copying from the
		 * same page as we're writing to, without it being marked
		 * up-to-date.
N
Nick Piggin 已提交
2045 2046 2047 2048
		 *
		 * Not only is this an optimisation, but it is also required
		 * to check that the address is actually valid, when atomic
		 * usercopies are used, below.
2049
		 */
2050
		if (unlikely(iov_iter_fault_in_readable(i, bytes))) {
N
Nick Piggin 已提交
2051 2052 2053
			status = -EFAULT;
			break;
		}
N
Nick Piggin 已提交
2054 2055

		page = __grab_cache_page(mapping, index);
L
Linus Torvalds 已提交
2056 2057 2058 2059 2060
		if (!page) {
			status = -ENOMEM;
			break;
		}

N
Nick Piggin 已提交
2061 2062 2063 2064 2065
		/*
		 * non-uptodate pages cannot cope with short copies, and we
		 * cannot take a pagefault with the destination page locked.
		 * So pin the source page to copy it.
		 */
N
Nick Piggin 已提交
2066
		if (!PageUptodate(page) && !segment_eq(get_fs(), KERNEL_DS)) {
N
Nick Piggin 已提交
2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080
			unlock_page(page);

			src_page = alloc_page(GFP_KERNEL);
			if (!src_page) {
				page_cache_release(page);
				status = -ENOMEM;
				break;
			}

			/*
			 * Cannot get_user_pages with a page locked for the
			 * same reason as we can't take a page fault with a
			 * page locked (as explained below).
			 */
2081
			copied = iov_iter_copy_from_user(src_page, i,
N
Nick Piggin 已提交
2082
								offset, bytes);
N
Nick Piggin 已提交
2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107
			if (unlikely(copied == 0)) {
				status = -EFAULT;
				page_cache_release(page);
				page_cache_release(src_page);
				break;
			}
			bytes = copied;

			lock_page(page);
			/*
			 * Can't handle the page going uptodate here, because
			 * that means we would use non-atomic usercopies, which
			 * zero out the tail of the page, which can cause
			 * zeroes to become transiently visible. We could just
			 * use a non-zeroing copy, but the APIs aren't too
			 * consistent.
			 */
			if (unlikely(!page->mapping || PageUptodate(page))) {
				unlock_page(page);
				page_cache_release(page);
				page_cache_release(src_page);
				continue;
			}
		}

L
Linus Torvalds 已提交
2108
		status = a_ops->prepare_write(file, page, offset, offset+bytes);
N
Nick Piggin 已提交
2109 2110
		if (unlikely(status))
			goto fs_write_aop_error;
2111

N
Nick Piggin 已提交
2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126
		if (!src_page) {
			/*
			 * Must not enter the pagefault handler here, because
			 * we hold the page lock, so we might recursively
			 * deadlock on the same lock, or get an ABBA deadlock
			 * against a different lock, or against the mmap_sem
			 * (which nests outside the page lock).  So increment
			 * preempt count, and use _atomic usercopies.
			 *
			 * The page is uptodate so we are OK to encounter a
			 * short copy: if unmodified parts of the page are
			 * marked dirty and written out to disk, it doesn't
			 * really matter.
			 */
			pagefault_disable();
2127
			copied = iov_iter_copy_from_user_atomic(page, i,
N
Nick Piggin 已提交
2128
								offset, bytes);
N
Nick Piggin 已提交
2129 2130 2131 2132 2133 2134 2135 2136 2137 2138
			pagefault_enable();
		} else {
			void *src, *dst;
			src = kmap_atomic(src_page, KM_USER0);
			dst = kmap_atomic(page, KM_USER1);
			memcpy(dst + offset, src + offset, bytes);
			kunmap_atomic(dst, KM_USER1);
			kunmap_atomic(src, KM_USER0);
			copied = bytes;
		}
L
Linus Torvalds 已提交
2139
		flush_dcache_page(page);
N
Nick Piggin 已提交
2140

L
Linus Torvalds 已提交
2141
		status = a_ops->commit_write(file, page, offset, offset+bytes);
N
Nick Piggin 已提交
2142
		if (unlikely(status < 0))
N
Nick Piggin 已提交
2143 2144
			goto fs_write_aop_error;
		if (unlikely(status > 0)) /* filesystem did partial write */
N
Nick Piggin 已提交
2145 2146 2147 2148 2149 2150 2151
			copied = min_t(size_t, copied, status);

		unlock_page(page);
		mark_page_accessed(page);
		page_cache_release(page);
		if (src_page)
			page_cache_release(src_page);
N
Nick Piggin 已提交
2152

2153
		iov_iter_advance(i, copied);
N
Nick Piggin 已提交
2154
		pos += copied;
2155
		written += copied;
N
Nick Piggin 已提交
2156

L
Linus Torvalds 已提交
2157 2158
		balance_dirty_pages_ratelimited(mapping);
		cond_resched();
N
Nick Piggin 已提交
2159 2160 2161
		continue;

fs_write_aop_error:
N
Nick Piggin 已提交
2162
		unlock_page(page);
N
Nick Piggin 已提交
2163
		page_cache_release(page);
N
Nick Piggin 已提交
2164 2165
		if (src_page)
			page_cache_release(src_page);
N
Nick Piggin 已提交
2166 2167 2168 2169 2170 2171 2172 2173

		/*
		 * prepare_write() may have instantiated a few blocks
		 * outside i_size.  Trim these off again. Don't need
		 * i_size_read because we hold i_mutex.
		 */
		if (pos + bytes > inode->i_size)
			vmtruncate(inode, inode->i_size);
N
Nick Piggin 已提交
2174
		break;
2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186
	} while (iov_iter_count(i));

	return written ? written : status;
}

static ssize_t generic_perform_write(struct file *file,
				struct iov_iter *i, loff_t pos)
{
	struct address_space *mapping = file->f_mapping;
	const struct address_space_operations *a_ops = mapping->a_ops;
	long status = 0;
	ssize_t written = 0;
N
Nick Piggin 已提交
2187 2188 2189 2190 2191 2192 2193
	unsigned int flags = 0;

	/*
	 * Copies from kernel address space cannot fail (NFSD is a big user).
	 */
	if (segment_eq(get_fs(), KERNEL_DS))
		flags |= AOP_FLAG_UNINTERRUPTIBLE;
2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224

	do {
		struct page *page;
		pgoff_t index;		/* Pagecache index for current page */
		unsigned long offset;	/* Offset into pagecache page */
		unsigned long bytes;	/* Bytes to write to page */
		size_t copied;		/* Bytes copied from user */
		void *fsdata;

		offset = (pos & (PAGE_CACHE_SIZE - 1));
		index = pos >> PAGE_CACHE_SHIFT;
		bytes = min_t(unsigned long, PAGE_CACHE_SIZE - offset,
						iov_iter_count(i));

again:

		/*
		 * Bring in the user page that we will copy from _first_.
		 * Otherwise there's a nasty deadlock on copying from the
		 * same page as we're writing to, without it being marked
		 * up-to-date.
		 *
		 * Not only is this an optimisation, but it is also required
		 * to check that the address is actually valid, when atomic
		 * usercopies are used, below.
		 */
		if (unlikely(iov_iter_fault_in_readable(i, bytes))) {
			status = -EFAULT;
			break;
		}

N
Nick Piggin 已提交
2225
		status = a_ops->write_begin(file, mapping, pos, bytes, flags,
2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283
						&page, &fsdata);
		if (unlikely(status))
			break;

		pagefault_disable();
		copied = iov_iter_copy_from_user_atomic(page, i, offset, bytes);
		pagefault_enable();
		flush_dcache_page(page);

		status = a_ops->write_end(file, mapping, pos, bytes, copied,
						page, fsdata);
		if (unlikely(status < 0))
			break;
		copied = status;

		cond_resched();

		if (unlikely(copied == 0)) {
			/*
			 * If we were unable to copy any data at all, we must
			 * fall back to a single segment length write.
			 *
			 * If we didn't fallback here, we could livelock
			 * because not all segments in the iov can be copied at
			 * once without a pagefault.
			 */
			bytes = min_t(unsigned long, PAGE_CACHE_SIZE - offset,
						iov_iter_single_seg_count(i));
			goto again;
		}
		iov_iter_advance(i, copied);
		pos += copied;
		written += copied;

		balance_dirty_pages_ratelimited(mapping);

	} while (iov_iter_count(i));

	return written ? written : status;
}

ssize_t
generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov,
		unsigned long nr_segs, loff_t pos, loff_t *ppos,
		size_t count, ssize_t written)
{
	struct file *file = iocb->ki_filp;
	struct address_space *mapping = file->f_mapping;
	const struct address_space_operations *a_ops = mapping->a_ops;
	struct inode *inode = mapping->host;
	ssize_t status;
	struct iov_iter i;

	iov_iter_init(&i, iov, nr_segs, count, written);
	if (a_ops->write_begin)
		status = generic_perform_write(file, &i, pos);
	else
		status = generic_perform_write_2copy(file, &i, pos);
L
Linus Torvalds 已提交
2284 2285

	if (likely(status >= 0)) {
2286 2287 2288 2289 2290 2291 2292
		written += status;
		*ppos = pos + status;

		/*
		 * For now, when the user asks for O_SYNC, we'll actually give
		 * O_DSYNC
		 */
L
Linus Torvalds 已提交
2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311
		if (unlikely((file->f_flags & O_SYNC) || IS_SYNC(inode))) {
			if (!a_ops->writepage || !is_sync_kiocb(iocb))
				status = generic_osync_inode(inode, mapping,
						OSYNC_METADATA|OSYNC_DATA);
		}
  	}
	
	/*
	 * If we get here for O_DIRECT writes then we must have fallen through
	 * to buffered writes (block instantiation inside i_size).  So we sync
	 * the file data here, to try to honour O_DIRECT expectations.
	 */
	if (unlikely(file->f_flags & O_DIRECT) && written)
		status = filemap_write_and_wait(mapping);

	return written ? written : status;
}
EXPORT_SYMBOL(generic_file_buffered_write);

2312
static ssize_t
L
Linus Torvalds 已提交
2313 2314 2315 2316
__generic_file_aio_write_nolock(struct kiocb *iocb, const struct iovec *iov,
				unsigned long nr_segs, loff_t *ppos)
{
	struct file *file = iocb->ki_filp;
2317
	struct address_space * mapping = file->f_mapping;
L
Linus Torvalds 已提交
2318 2319 2320 2321 2322 2323 2324 2325
	size_t ocount;		/* original count */
	size_t count;		/* after file limit checks */
	struct inode 	*inode = mapping->host;
	loff_t		pos;
	ssize_t		written;
	ssize_t		err;

	ocount = 0;
2326 2327 2328
	err = generic_segment_checks(iov, &nr_segs, &ocount, VERIFY_READ);
	if (err)
		return err;
L
Linus Torvalds 已提交
2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345

	count = ocount;
	pos = *ppos;

	vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);

	/* We can write back this queue in page reclaim */
	current->backing_dev_info = mapping->backing_dev_info;
	written = 0;

	err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode));
	if (err)
		goto out;

	if (count == 0)
		goto out;

2346
	err = remove_suid(file->f_path.dentry);
L
Linus Torvalds 已提交
2347 2348 2349
	if (err)
		goto out;

2350
	file_update_time(file);
L
Linus Torvalds 已提交
2351 2352 2353

	/* coalesce the iovecs and go direct-to-BIO for O_DIRECT */
	if (unlikely(file->f_flags & O_DIRECT)) {
2354 2355 2356 2357 2358
		loff_t endbyte;
		ssize_t written_buffered;

		written = generic_file_direct_write(iocb, iov, &nr_segs, pos,
							ppos, count, ocount);
L
Linus Torvalds 已提交
2359 2360 2361 2362 2363 2364 2365 2366
		if (written < 0 || written == count)
			goto out;
		/*
		 * direct-io write to a hole: fall through to buffered I/O
		 * for completing the rest of the request.
		 */
		pos += written;
		count -= written;
2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380
		written_buffered = generic_file_buffered_write(iocb, iov,
						nr_segs, pos, ppos, count,
						written);
		/*
		 * If generic_file_buffered_write() retuned a synchronous error
		 * then we want to return the number of bytes which were
		 * direct-written, or the error code if that was zero.  Note
		 * that this differs from normal direct-io semantics, which
		 * will return -EFOO even if some bytes were written.
		 */
		if (written_buffered < 0) {
			err = written_buffered;
			goto out;
		}
L
Linus Torvalds 已提交
2381

2382 2383 2384 2385 2386 2387
		/*
		 * We need to ensure that the page cache pages are written to
		 * disk and invalidated to preserve the expected O_DIRECT
		 * semantics.
		 */
		endbyte = pos + written_buffered - written - 1;
M
Mark Fasheh 已提交
2388 2389 2390 2391
		err = do_sync_mapping_range(file->f_mapping, pos, endbyte,
					    SYNC_FILE_RANGE_WAIT_BEFORE|
					    SYNC_FILE_RANGE_WRITE|
					    SYNC_FILE_RANGE_WAIT_AFTER);
2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406
		if (err == 0) {
			written = written_buffered;
			invalidate_mapping_pages(mapping,
						 pos >> PAGE_CACHE_SHIFT,
						 endbyte >> PAGE_CACHE_SHIFT);
		} else {
			/*
			 * We don't know how much we wrote, so just return
			 * the number of bytes which were direct-written
			 */
		}
	} else {
		written = generic_file_buffered_write(iocb, iov, nr_segs,
				pos, ppos, count, written);
	}
L
Linus Torvalds 已提交
2407 2408 2409 2410 2411
out:
	current->backing_dev_info = NULL;
	return written ? written : err;
}

2412 2413
ssize_t generic_file_aio_write_nolock(struct kiocb *iocb,
		const struct iovec *iov, unsigned long nr_segs, loff_t pos)
L
Linus Torvalds 已提交
2414 2415 2416 2417 2418 2419
{
	struct file *file = iocb->ki_filp;
	struct address_space *mapping = file->f_mapping;
	struct inode *inode = mapping->host;
	ssize_t ret;

2420 2421 2422 2423
	BUG_ON(iocb->ki_pos != pos);

	ret = __generic_file_aio_write_nolock(iocb, iov, nr_segs,
			&iocb->ki_pos);
L
Linus Torvalds 已提交
2424 2425

	if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) {
2426
		ssize_t err;
L
Linus Torvalds 已提交
2427 2428 2429 2430 2431 2432 2433

		err = sync_page_range_nolock(inode, mapping, pos, ret);
		if (err < 0)
			ret = err;
	}
	return ret;
}
2434
EXPORT_SYMBOL(generic_file_aio_write_nolock);
L
Linus Torvalds 已提交
2435

2436 2437
ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
		unsigned long nr_segs, loff_t pos)
L
Linus Torvalds 已提交
2438 2439 2440 2441 2442 2443 2444 2445
{
	struct file *file = iocb->ki_filp;
	struct address_space *mapping = file->f_mapping;
	struct inode *inode = mapping->host;
	ssize_t ret;

	BUG_ON(iocb->ki_pos != pos);

2446
	mutex_lock(&inode->i_mutex);
2447 2448
	ret = __generic_file_aio_write_nolock(iocb, iov, nr_segs,
			&iocb->ki_pos);
2449
	mutex_unlock(&inode->i_mutex);
L
Linus Torvalds 已提交
2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462

	if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) {
		ssize_t err;

		err = sync_page_range(inode, mapping, pos, ret);
		if (err < 0)
			ret = err;
	}
	return ret;
}
EXPORT_SYMBOL(generic_file_aio_write);

/*
2463
 * Called under i_mutex for writes to S_ISREG files.   Returns -EIO if something
L
Linus Torvalds 已提交
2464 2465
 * went wrong during pagecache shootdown.
 */
2466
static ssize_t
L
Linus Torvalds 已提交
2467 2468 2469 2470 2471 2472
generic_file_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov,
	loff_t offset, unsigned long nr_segs)
{
	struct file *file = iocb->ki_filp;
	struct address_space *mapping = file->f_mapping;
	ssize_t retval;
2473 2474
	size_t write_len;
	pgoff_t end = 0; /* silence gcc */
L
Linus Torvalds 已提交
2475 2476 2477 2478 2479 2480 2481 2482

	/*
	 * If it's a write, unmap all mmappings of the file up-front.  This
	 * will cause any pte dirty bits to be propagated into the pageframes
	 * for the subsequent filemap_write_and_wait().
	 */
	if (rw == WRITE) {
		write_len = iov_length(iov, nr_segs);
2483
		end = (offset + write_len - 1) >> PAGE_CACHE_SHIFT;
L
Linus Torvalds 已提交
2484 2485 2486 2487 2488
	       	if (mapping_mapped(mapping))
			unmap_mapping_range(mapping, offset, write_len, 0);
	}

	retval = filemap_write_and_wait(mapping);
2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499
	if (retval)
		goto out;

	/*
	 * After a write we want buffered reads to be sure to go to disk to get
	 * the new data.  We invalidate clean cached page from the region we're
	 * about to write.  We do this *before* the write so that we can return
	 * -EIO without clobbering -EIOCBQUEUED from ->direct_IO().
	 */
	if (rw == WRITE && mapping->nrpages) {
		retval = invalidate_inode_pages2_range(mapping,
L
Linus Torvalds 已提交
2500
					offset >> PAGE_CACHE_SHIFT, end);
2501 2502
		if (retval)
			goto out;
L
Linus Torvalds 已提交
2503
	}
2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522

	retval = mapping->a_ops->direct_IO(rw, iocb, iov, offset, nr_segs);
	if (retval)
		goto out;

	/*
	 * Finally, try again to invalidate clean pages which might have been
	 * faulted in by get_user_pages() if the source of the write was an
	 * mmap()ed region of the file we're writing.  That's a pretty crazy
	 * thing to do, so we don't support it 100%.  If this invalidation
	 * fails and we have -EIOCBQUEUED we ignore the failure.
	 */
	if (rw == WRITE && mapping->nrpages) {
		int err = invalidate_inode_pages2_range(mapping,
					      offset >> PAGE_CACHE_SHIFT, end);
		if (err && retval >= 0)
			retval = err;
	}
out:
L
Linus Torvalds 已提交
2523 2524
	return retval;
}
2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554

/**
 * try_to_release_page() - release old fs-specific metadata on a page
 *
 * @page: the page which the kernel is trying to free
 * @gfp_mask: memory allocation flags (and I/O mode)
 *
 * The address_space is to try to release any data against the page
 * (presumably at page->private).  If the release was successful, return `1'.
 * Otherwise return zero.
 *
 * The @gfp_mask argument specifies whether I/O may be performed to release
 * this page (__GFP_IO), and whether the call may block (__GFP_WAIT).
 *
 * NOTE: @gfp_mask may go away, and this function may become non-blocking.
 */
int try_to_release_page(struct page *page, gfp_t gfp_mask)
{
	struct address_space * const mapping = page->mapping;

	BUG_ON(!PageLocked(page));
	if (PageWriteback(page))
		return 0;

	if (mapping && mapping->a_ops->releasepage)
		return mapping->a_ops->releasepage(page, gfp_mask);
	return try_to_free_buffers(page);
}

EXPORT_SYMBOL(try_to_release_page);