filemap.c 67.6 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
#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>
28
#include <linux/backing-dev.h>
L
Linus Torvalds 已提交
29 30
#include <linux/pagevec.h>
#include <linux/blkdev.h>
E
Emil Medve 已提交
31
#include <linux/backing-dev.h>
L
Linus Torvalds 已提交
32 33
#include <linux/security.h>
#include <linux/syscalls.h>
34
#include <linux/cpuset.h>
N
Nick Piggin 已提交
35
#include <linux/hardirq.h> /* for BUG_ON(!in_atomic()) only */
36 37
#include "internal.h"

L
Linus Torvalds 已提交
38 39 40 41 42 43 44
/*
 * 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>

45 46 47 48
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
49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
/*
 * 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)
66 67
 *      ->swap_lock		(exclusive_swap_page, others)
 *        ->mapping->tree_lock
68
 *          ->zone.lock
L
Linus Torvalds 已提交
69
 *
70
 *  ->i_mutex
L
Linus Torvalds 已提交
71 72 73 74
 *    ->i_mmap_lock		(truncate->unmap_mapping_range)
 *
 *  ->mmap_sem
 *    ->i_mmap_lock
75
 *      ->page_table_lock or pte_lock	(various, mainly in memory.c)
L
Linus Torvalds 已提交
76 77 78 79 80
 *        ->mapping->tree_lock	(arch-dependent flush_dcache_mmap_lock)
 *
 *  ->mmap_sem
 *    ->lock_page		(access_process_vm)
 *
81 82
 *  ->i_mutex			(generic_file_buffered_write)
 *    ->mmap_sem		(fault_in_pages_readable->do_page_fault)
L
Linus Torvalds 已提交
83
 *
84
 *  ->i_mutex
L
Linus Torvalds 已提交
85 86 87 88 89 90 91 92 93 94
 *    ->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
95
 *    ->page_table_lock or pte_lock	(anon_vma_prepare and various)
L
Linus Torvalds 已提交
96
 *
97
 *  ->page_table_lock or pte_lock
98
 *    ->swap_lock		(try_to_unmap_one)
L
Linus Torvalds 已提交
99 100 101
 *    ->private_lock		(try_to_unmap_one)
 *    ->tree_lock		(try_to_unmap_one)
 *    ->zone.lru_lock		(follow_page->mark_page_accessed)
102
 *    ->zone.lru_lock		(check_pte_range->isolate_lru_page)
L
Linus Torvalds 已提交
103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124
 *    ->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--;
125
	__dec_zone_page_state(page, NR_FILE_PAGES);
126
	BUG_ON(page_mapped(page));
L
Linus Torvalds 已提交
127 128 129 130 131 132
}

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

M
Matt Mackall 已提交
133
	BUG_ON(!PageLocked(page));
L
Linus Torvalds 已提交
134 135 136 137 138 139 140 141 142 143 144

	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;

145
	page = container_of((unsigned long *)word, struct page, flags);
L
Linus Torvalds 已提交
146 147

	/*
148 149 150 151 152 153 154 155 156 157 158 159 160 161 162
	 * 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
163
	 * ignore it for all cases but swap, where only page_private(page) is
164 165 166
	 * 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
167 168 169 170 171 172 173 174 175 176
	 */
	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;
}

/**
177
 * __filemap_fdatawrite_range - start writeback on mapping dirty pages in range
178 179
 * @mapping:	address space structure to write
 * @start:	offset in bytes where the range starts
180
 * @end:	offset in bytes where the range ends (inclusive)
181
 * @sync_mode:	enable synchronous operation
L
Linus Torvalds 已提交
182
 *
183 184 185
 * 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
186
 * If sync_mode is WB_SYNC_ALL then this is a "data integrity" operation, as
187
 * opposed to a regular memory cleansing writeback.  The difference between
L
Linus Torvalds 已提交
188 189 190
 * these two operations is that if a dirty page/buffer is encountered, it must
 * be waited upon, and not just skipped over.
 */
191 192
int __filemap_fdatawrite_range(struct address_space *mapping, loff_t start,
				loff_t end, int sync_mode)
L
Linus Torvalds 已提交
193 194 195 196 197
{
	int ret;
	struct writeback_control wbc = {
		.sync_mode = sync_mode,
		.nr_to_write = mapping->nrpages * 2,
198 199
		.range_start = start,
		.range_end = end,
L
Linus Torvalds 已提交
200 201 202 203 204 205 206 207 208 209 210 211
	};

	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)
{
212
	return __filemap_fdatawrite_range(mapping, 0, LLONG_MAX, sync_mode);
L
Linus Torvalds 已提交
213 214 215 216 217 218 219 220
}

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

221 222
static int filemap_fdatawrite_range(struct address_space *mapping, loff_t start,
				loff_t end)
L
Linus Torvalds 已提交
223 224 225 226
{
	return __filemap_fdatawrite_range(mapping, start, end, WB_SYNC_ALL);
}

227 228 229 230
/**
 * filemap_flush - mostly a non-blocking flush
 * @mapping:	target address_space
 *
L
Linus Torvalds 已提交
231 232 233 234 235 236 237 238 239
 * 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);

240 241 242 243 244 245
/**
 * 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
246 247 248
 * Wait for writeback to complete against pages indexed by start->end
 * inclusive
 */
249
int wait_on_page_writeback_range(struct address_space *mapping,
L
Linus Torvalds 已提交
250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291
				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;
}

292 293 294 295 296 297 298
/**
 * 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
299 300 301 302
 * 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.
 *
303
 * We need to re-take i_mutex during the generic_osync_inode list walk because
L
Linus Torvalds 已提交
304 305 306
 * it is otherwise livelockable.
 */
int sync_page_range(struct inode *inode, struct address_space *mapping,
307
			loff_t pos, loff_t count)
L
Linus Torvalds 已提交
308 309 310 311 312 313 314 315 316
{
	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) {
317
		mutex_lock(&inode->i_mutex);
L
Linus Torvalds 已提交
318
		ret = generic_osync_inode(inode, mapping, OSYNC_METADATA);
319
		mutex_unlock(&inode->i_mutex);
L
Linus Torvalds 已提交
320 321 322 323 324 325 326
	}
	if (ret == 0)
		ret = wait_on_page_writeback_range(mapping, start, end);
	return ret;
}
EXPORT_SYMBOL(sync_page_range);

327 328 329 330 331 332 333
/**
 * sync_page_range_nolock
 * @inode:	target inode
 * @mapping:	target address_space
 * @pos:	beginning offset in pages to write
 * @count:	number of bytes to write
 *
334
 * Note: Holding i_mutex across sync_page_range_nolock() is not a good idea
L
Linus Torvalds 已提交
335 336 337
 * as it forces O_SYNC writers to different parts of the same file
 * to be serialised right until io completion.
 */
338 339
int sync_page_range_nolock(struct inode *inode, struct address_space *mapping,
			   loff_t pos, loff_t count)
L
Linus Torvalds 已提交
340 341 342 343 344 345 346 347 348 349 350 351 352 353
{
	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;
}
354
EXPORT_SYMBOL(sync_page_range_nolock);
L
Linus Torvalds 已提交
355 356

/**
357
 * filemap_fdatawait - wait for all under-writeback pages to complete
L
Linus Torvalds 已提交
358
 * @mapping: address space structure to wait for
359 360 361
 *
 * Walk the list of under-writeback pages of the given address space
 * and wait for all of them.
L
Linus Torvalds 已提交
362 363 364 365 366 367 368 369 370 371 372 373 374 375 376
 */
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)
{
377
	int err = 0;
L
Linus Torvalds 已提交
378 379

	if (mapping->nrpages) {
380 381 382 383 384 385 386 387 388 389 390 391
		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
392
	}
393
	return err;
L
Linus Torvalds 已提交
394
}
395
EXPORT_SYMBOL(filemap_write_and_wait);
L
Linus Torvalds 已提交
396

397 398 399 400 401 402
/**
 * 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)
 *
403 404 405 406 407
 * 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
408 409 410
int filemap_write_and_wait_range(struct address_space *mapping,
				 loff_t lstart, loff_t lend)
{
411
	int err = 0;
L
Linus Torvalds 已提交
412 413

	if (mapping->nrpages) {
414 415 416 417 418 419 420 421 422 423
		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
424
	}
425
	return err;
L
Linus Torvalds 已提交
426 427
}

428 429 430 431 432 433 434 435
/**
 * 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
436 437 438 439 440 441
 * 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
442
		pgoff_t offset, gfp_t gfp_mask)
L
Linus Torvalds 已提交
443 444 445 446 447 448 449 450 451 452 453 454
{
	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++;
455
			__inc_zone_page_state(page, NR_FILE_PAGES);
L
Linus Torvalds 已提交
456 457 458 459 460 461 462 463 464
		}
		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
465
				pgoff_t offset, gfp_t gfp_mask)
L
Linus Torvalds 已提交
466 467 468 469 470 471 472
{
	int ret = add_to_page_cache(page, mapping, offset, gfp_mask);
	if (ret == 0)
		lru_cache_add(page);
	return ret;
}

473
#ifdef CONFIG_NUMA
474
struct page *__page_cache_alloc(gfp_t gfp)
475 476 477
{
	if (cpuset_do_page_mem_spread()) {
		int n = cpuset_mem_spread_node();
478
		return alloc_pages_node(n, gfp, 0);
479
	}
480
	return alloc_pages(gfp, 0);
481
}
482
EXPORT_SYMBOL(__page_cache_alloc);
483 484
#endif

485 486 487 488 489 490
static int __sleep_on_page_lock(void *word)
{
	io_schedule();
	return 0;
}

L
Linus Torvalds 已提交
491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523
/*
 * 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);

/**
524
 * unlock_page - unlock a locked page
L
Linus Torvalds 已提交
525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546
 * @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);

547 548 549
/**
 * end_page_writeback - end writeback against a page
 * @page: the page
L
Linus Torvalds 已提交
550 551 552 553 554 555 556 557 558 559 560 561
 */
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);

562 563 564
/**
 * __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
565
 *
566
 * Ugly. Running sync_page() in state TASK_UNINTERRUPTIBLE is scary.  If some
L
Linus Torvalds 已提交
567 568 569 570 571 572 573 574 575 576 577 578 579
 * 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);

580 581 582 583 584 585 586 587 588 589 590
/*
 * 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);
}

591 592 593 594 595
/**
 * 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
596 597
 * 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
598
 */
599
struct page * find_get_page(struct address_space *mapping, pgoff_t offset)
L
Linus Torvalds 已提交
600 601 602 603 604 605 606 607 608 609 610 611 612 613
{
	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
614 615
 * @mapping: the address_space to search
 * @offset: the page index
L
Linus Torvalds 已提交
616 617 618 619 620 621 622
 *
 * 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,
623
				pgoff_t offset)
L
Linus Torvalds 已提交
624 625 626 627
{
	struct page *page;

repeat:
N
Nick Piggin 已提交
628
	read_lock_irq(&mapping->tree_lock);
L
Linus Torvalds 已提交
629 630 631 632 633
	page = radix_tree_lookup(&mapping->page_tree, offset);
	if (page) {
		page_cache_get(page);
		if (TestSetPageLocked(page)) {
			read_unlock_irq(&mapping->tree_lock);
634
			__lock_page(page);
L
Linus Torvalds 已提交
635 636

			/* Has the page been truncated while we slept? */
N
Nick Piggin 已提交
637
			if (unlikely(page->mapping != mapping)) {
L
Linus Torvalds 已提交
638 639 640 641
				unlock_page(page);
				page_cache_release(page);
				goto repeat;
			}
N
Nick Piggin 已提交
642 643
			VM_BUG_ON(page->index != offset);
			goto out;
L
Linus Torvalds 已提交
644 645 646
		}
	}
	read_unlock_irq(&mapping->tree_lock);
N
Nick Piggin 已提交
647
out:
L
Linus Torvalds 已提交
648 649 650 651 652 653
	return page;
}
EXPORT_SYMBOL(find_lock_page);

/**
 * find_or_create_page - locate or add a pagecache page
654 655 656
 * @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
657 658 659 660 661 662 663 664 665 666 667 668 669
 *
 * 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,
670
		pgoff_t index, gfp_t gfp_mask)
L
Linus Torvalds 已提交
671
{
N
Nick Piggin 已提交
672
	struct page *page;
L
Linus Torvalds 已提交
673 674 675 676
	int err;
repeat:
	page = find_lock_page(mapping, index);
	if (!page) {
N
Nick Piggin 已提交
677 678 679 680 681 682 683 684 685
		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
686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722
		}
	}
	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;
}

723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753
/**
 * 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;
}
754
EXPORT_SYMBOL(find_get_pages_contig);
755

756 757 758 759 760 761 762 763
/**
 * 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
764
 * Like find_get_pages, except we only return pages which are tagged with
765
 * @tag.   We update @index to index the next page for the traversal.
L
Linus Torvalds 已提交
766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782
 */
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;
}
783
EXPORT_SYMBOL(find_get_pages_tag);
L
Linus Torvalds 已提交
784

785 786 787 788 789
/**
 * grab_cache_page_nowait - returns locked page at given index in given cache
 * @mapping: target address_space
 * @index: the page index
 *
790
 * Same as grab_cache_page(), but do not wait if the page is unavailable.
L
Linus Torvalds 已提交
791 792 793 794 795 796 797 798
 * 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 *
799
grab_cache_page_nowait(struct address_space *mapping, pgoff_t index)
L
Linus Torvalds 已提交
800 801 802 803 804 805 806 807 808
{
	struct page *page = find_get_page(mapping, index);

	if (page) {
		if (!TestSetPageLocked(page))
			return page;
		page_cache_release(page);
		return NULL;
	}
809 810
	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
811 812 813 814 815 816 817
		page_cache_release(page);
		page = NULL;
	}
	return page;
}
EXPORT_SYMBOL(grab_cache_page_nowait);

818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841
/*
 * 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;
}

842 843 844
/**
 * do_generic_mapping_read - generic file read routine
 * @mapping:	address_space to be read
845
 * @ra:		file's readahead state
846 847 848 849 850
 * @filp:	the file to read
 * @ppos:	current file position
 * @desc:	read_descriptor
 * @actor:	read method
 *
L
Linus Torvalds 已提交
851
 * This is a generic file read routine, and uses the
852
 * mapping->a_ops->readpage() function for the actual low-level stuff.
L
Linus Torvalds 已提交
853 854 855 856
 *
 * This is really ugly. But the goto's actually try to clarify some
 * of the logic when it comes to error handling etc.
 *
857 858
 * Note the struct file* is only passed for the use of readpage.
 * It may be NULL.
L
Linus Torvalds 已提交
859 860
 */
void do_generic_mapping_read(struct address_space *mapping,
861
			     struct file_ra_state *ra,
L
Linus Torvalds 已提交
862 863 864 865 866 867
			     struct file *filp,
			     loff_t *ppos,
			     read_descriptor_t *desc,
			     read_actor_t actor)
{
	struct inode *inode = mapping->host;
868 869 870 871
	pgoff_t index;
	pgoff_t last_index;
	pgoff_t prev_index;
	unsigned long offset;      /* offset into pagecache page */
872
	unsigned int prev_offset;
L
Linus Torvalds 已提交
873 874 875
	int error;

	index = *ppos >> PAGE_CACHE_SHIFT;
876 877
	prev_index = ra->prev_pos >> PAGE_CACHE_SHIFT;
	prev_offset = ra->prev_pos & (PAGE_CACHE_SIZE-1);
L
Linus Torvalds 已提交
878 879 880 881 882
	last_index = (*ppos + desc->count + PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT;
	offset = *ppos & ~PAGE_CACHE_MASK;

	for (;;) {
		struct page *page;
883
		pgoff_t end_index;
N
NeilBrown 已提交
884
		loff_t isize;
L
Linus Torvalds 已提交
885 886 887 888 889
		unsigned long nr, ret;

		cond_resched();
find_page:
		page = find_get_page(mapping, index);
890
		if (!page) {
891
			page_cache_sync_readahead(mapping,
892
					ra, filp,
893 894 895 896 897 898
					index, last_index - index);
			page = find_get_page(mapping, index);
			if (unlikely(page == NULL))
				goto no_cached_page;
		}
		if (PageReadahead(page)) {
899
			page_cache_async_readahead(mapping,
900
					ra, filp, page,
901
					index, last_index - index);
L
Linus Torvalds 已提交
902 903 904 905
		}
		if (!PageUptodate(page))
			goto page_not_up_to_date;
page_ok:
N
NeilBrown 已提交
906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931
		/*
		 * 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
932 933 934 935 936 937 938 939 940

		/* 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);

		/*
941 942
		 * When a sequential read accesses a page several times,
		 * only mark it as accessed the first time.
L
Linus Torvalds 已提交
943
		 */
944
		if (prev_index != index || offset != prev_offset)
L
Linus Torvalds 已提交
945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961
			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
962
		prev_offset = offset;
L
Linus Torvalds 已提交
963 964 965 966 967 968 969 970 971 972

		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
973
		/* Did it get truncated before we got the lock? */
L
Linus Torvalds 已提交
974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989
		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);

990 991 992 993 994
		if (unlikely(error)) {
			if (error == AOP_TRUNCATED_PAGE) {
				page_cache_release(page);
				goto find_page;
			}
L
Linus Torvalds 已提交
995
			goto readpage_error;
996
		}
L
Linus Torvalds 已提交
997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010

		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;
1011
				shrink_readahead_size_eio(filp, ra);
L
Linus Torvalds 已提交
1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029
				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
1030 1031 1032 1033
		page = page_cache_alloc_cold(mapping);
		if (!page) {
			desc->error = -ENOMEM;
			goto out;
L
Linus Torvalds 已提交
1034
		}
N
Nick Piggin 已提交
1035
		error = add_to_page_cache_lru(page, mapping,
L
Linus Torvalds 已提交
1036 1037
						index, GFP_KERNEL);
		if (error) {
N
Nick Piggin 已提交
1038
			page_cache_release(page);
L
Linus Torvalds 已提交
1039 1040 1041 1042 1043 1044 1045 1046 1047
			if (error == -EEXIST)
				goto find_page;
			desc->error = error;
			goto out;
		}
		goto readpage;
	}

out:
1048 1049 1050
	ra->prev_pos = prev_index;
	ra->prev_pos <<= PAGE_CACHE_SHIFT;
	ra->prev_pos |= prev_offset;
L
Linus Torvalds 已提交
1051

1052
	*ppos = ((loff_t)index << PAGE_CACHE_SHIFT) + offset;
L
Linus Torvalds 已提交
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 1094 1095
	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;
}

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 1133 1134
/*
 * 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);

1135
/**
H
Henrik Kretzschmar 已提交
1136
 * generic_file_aio_read - generic filesystem read routine
1137 1138 1139
 * @iocb:	kernel I/O control block
 * @iov:	io vector request
 * @nr_segs:	number of segments in the iovec
H
Henrik Kretzschmar 已提交
1140
 * @pos:	current file position
1141
 *
L
Linus Torvalds 已提交
1142 1143 1144 1145
 * This is the "read()" routine for all filesystems
 * that can use the page cache directly.
 */
ssize_t
1146 1147
generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
		unsigned long nr_segs, loff_t pos)
L
Linus Torvalds 已提交
1148 1149 1150 1151 1152
{
	struct file *filp = iocb->ki_filp;
	ssize_t retval;
	unsigned long seg;
	size_t count;
1153
	loff_t *ppos = &iocb->ki_pos;
L
Linus Torvalds 已提交
1154 1155

	count = 0;
1156 1157 1158
	retval = generic_segment_checks(iov, &nr_segs, &count, VERIFY_WRITE);
	if (retval)
		return retval;
L
Linus Torvalds 已提交
1159 1160 1161

	/* coalesce the iovecs and go direct-to-BIO for O_DIRECT */
	if (filp->f_flags & O_DIRECT) {
1162
		loff_t size;
L
Linus Torvalds 已提交
1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177
		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;
		}
1178
		if (likely(retval != 0)) {
1179
			file_accessed(filp);
S
Steven Whitehouse 已提交
1180
			goto out;
1181
		}
L
Linus Torvalds 已提交
1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196
	}

	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;
1197 1198
			if (desc.error) {
				retval = retval ?: desc.error;
L
Linus Torvalds 已提交
1199 1200
				break;
			}
1201 1202
			if (desc.count > 0)
				break;
L
Linus Torvalds 已提交
1203 1204 1205 1206 1207 1208 1209 1210 1211
		}
	}
out:
	return retval;
}
EXPORT_SYMBOL(generic_file_aio_read);

static ssize_t
do_readahead(struct address_space *mapping, struct file *filp,
1212
	     pgoff_t index, unsigned long nr)
L
Linus Torvalds 已提交
1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231
{
	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;
1232 1233
			pgoff_t start = offset >> PAGE_CACHE_SHIFT;
			pgoff_t end = (offset + count - 1) >> PAGE_CACHE_SHIFT;
L
Linus Torvalds 已提交
1234 1235 1236 1237 1238 1239 1240 1241 1242
			unsigned long len = end - start + 1;
			ret = do_readahead(mapping, file, start, len);
		}
		fput(file);
	}
	return ret;
}

#ifdef CONFIG_MMU
1243 1244 1245 1246 1247
/**
 * 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
1248 1249 1250
 * 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.
 */
1251
static int fastcall page_cache_read(struct file * file, pgoff_t offset)
L
Linus Torvalds 已提交
1252 1253 1254
{
	struct address_space *mapping = file->f_mapping;
	struct page *page; 
1255
	int ret;
L
Linus Torvalds 已提交
1256

1257 1258 1259 1260 1261 1262 1263 1264 1265 1266
	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
1267 1268 1269

		page_cache_release(page);

1270 1271 1272
	} while (ret == AOP_TRUNCATED_PAGE);
		
	return ret;
L
Linus Torvalds 已提交
1273 1274 1275 1276
}

#define MMAP_LOTSAMISS  (100)

1277
/**
1278
 * filemap_fault - read in file data for page fault handling
N
Nick Piggin 已提交
1279 1280
 * @vma:	vma in which the fault was taken
 * @vmf:	struct vm_fault containing details of the fault
1281
 *
1282
 * filemap_fault() is invoked via the vma operations vector for a
L
Linus Torvalds 已提交
1283 1284 1285 1286 1287 1288
 * 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
1289
int filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
L
Linus Torvalds 已提交
1290 1291
{
	int error;
1292
	struct file *file = vma->vm_file;
L
Linus Torvalds 已提交
1293 1294 1295 1296
	struct address_space *mapping = file->f_mapping;
	struct file_ra_state *ra = &file->f_ra;
	struct inode *inode = mapping->host;
	struct page *page;
1297 1298
	unsigned long size;
	int did_readaround = 0;
N
Nick Piggin 已提交
1299
	int ret = 0;
L
Linus Torvalds 已提交
1300 1301

	size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
N
Nick Piggin 已提交
1302
	if (vmf->pgoff >= size)
L
Linus Torvalds 已提交
1303 1304 1305
		goto outside_data_content;

	/* If we don't want any read-ahead, don't bother */
1306
	if (VM_RandomReadHint(vma))
L
Linus Torvalds 已提交
1307 1308 1309 1310 1311 1312
		goto no_cached_page;

	/*
	 * Do we have something in the page cache already?
	 */
retry_find:
N
Nick Piggin 已提交
1313
	page = find_lock_page(mapping, vmf->pgoff);
1314 1315 1316 1317 1318
	/*
	 * For sequential accesses, we use the generic readahead logic.
	 */
	if (VM_SequentialReadHint(vma)) {
		if (!page) {
1319
			page_cache_sync_readahead(mapping, ra, file,
1320 1321 1322 1323 1324 1325
							   vmf->pgoff, 1);
			page = find_lock_page(mapping, vmf->pgoff);
			if (!page)
				goto no_cached_page;
		}
		if (PageReadahead(page)) {
1326
			page_cache_async_readahead(mapping, ra, file, page,
1327 1328 1329 1330
							   vmf->pgoff, 1);
		}
	}

L
Linus Torvalds 已提交
1331 1332 1333 1334 1335 1336 1337 1338 1339
	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.
		 */
1340
		if (ra->mmap_miss > MMAP_LOTSAMISS)
L
Linus Torvalds 已提交
1341 1342 1343 1344 1345 1346 1347
			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
1348
			ret = VM_FAULT_MAJOR;
1349
			count_vm_event(PGMAJFAULT);
L
Linus Torvalds 已提交
1350 1351 1352 1353 1354 1355
		}
		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
1356 1357
			if (vmf->pgoff > ra_pages / 2)
				start = vmf->pgoff - ra_pages / 2;
L
Linus Torvalds 已提交
1358 1359
			do_page_cache_readahead(mapping, file, start, ra_pages);
		}
N
Nick Piggin 已提交
1360
		page = find_lock_page(mapping, vmf->pgoff);
L
Linus Torvalds 已提交
1361 1362 1363 1364 1365
		if (!page)
			goto no_cached_page;
	}

	if (!did_readaround)
1366
		ra->mmap_miss--;
L
Linus Torvalds 已提交
1367 1368

	/*
1369 1370
	 * 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
1371
	 */
1372
	if (unlikely(!PageUptodate(page)))
L
Linus Torvalds 已提交
1373 1374
		goto page_not_uptodate;

1375 1376
	/* 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
1377
	if (unlikely(vmf->pgoff >= size)) {
1378
		unlock_page(page);
1379
		page_cache_release(page);
1380 1381 1382
		goto outside_data_content;
	}

L
Linus Torvalds 已提交
1383 1384 1385 1386
	/*
	 * Found the page and have a reference on it.
	 */
	mark_page_accessed(page);
1387
	ra->prev_pos = (loff_t)page->index << PAGE_CACHE_SHIFT;
N
Nick Piggin 已提交
1388
	vmf->page = page;
N
Nick Piggin 已提交
1389
	return ret | VM_FAULT_LOCKED;
L
Linus Torvalds 已提交
1390 1391 1392 1393 1394 1395

outside_data_content:
	/*
	 * An external ptracer can access pages that normally aren't
	 * accessible..
	 */
N
Nick Piggin 已提交
1396 1397 1398
	if (vma->vm_mm == current->mm)
		return VM_FAULT_SIGBUS;

L
Linus Torvalds 已提交
1399 1400 1401 1402 1403 1404
	/* 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
1405
	error = page_cache_read(file, vmf->pgoff);
L
Linus Torvalds 已提交
1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420

	/*
	 * 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
1421 1422
		return VM_FAULT_OOM;
	return VM_FAULT_SIGBUS;
L
Linus Torvalds 已提交
1423 1424

page_not_uptodate:
1425
	/* IO error path */
L
Linus Torvalds 已提交
1426
	if (!did_readaround) {
N
Nick Piggin 已提交
1427
		ret = VM_FAULT_MAJOR;
1428
		count_vm_event(PGMAJFAULT);
L
Linus Torvalds 已提交
1429 1430 1431 1432 1433 1434 1435 1436 1437
	}

	/*
	 * 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);
1438
	error = mapping->a_ops->readpage(file, page);
1439 1440 1441
	page_cache_release(page);

	if (!error || error == AOP_TRUNCATED_PAGE)
1442
		goto retry_find;
L
Linus Torvalds 已提交
1443

1444
	/* Things didn't work out. Return zero to tell the mm layer so. */
1445
	shrink_readahead_size_eio(file, ra);
N
Nick Piggin 已提交
1446
	return VM_FAULT_SIGBUS;
1447 1448 1449
}
EXPORT_SYMBOL(filemap_fault);

L
Linus Torvalds 已提交
1450
struct vm_operations_struct generic_file_vm_ops = {
1451
	.fault		= filemap_fault,
L
Linus Torvalds 已提交
1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463
};

/* 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
1464
	vma->vm_flags |= VM_CAN_NONLINEAR;
L
Linus Torvalds 已提交
1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490
	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);

1491
static struct page *__read_cache_page(struct address_space *mapping,
1492
				pgoff_t index,
L
Linus Torvalds 已提交
1493 1494 1495
				int (*filler)(void *,struct page*),
				void *data)
{
N
Nick Piggin 已提交
1496
	struct page *page;
L
Linus Torvalds 已提交
1497 1498 1499 1500
	int err;
repeat:
	page = find_get_page(mapping, index);
	if (!page) {
N
Nick Piggin 已提交
1501 1502 1503 1504 1505 1506 1507 1508
		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
1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520
			/* 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;
}

1521 1522 1523
/*
 * 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
1524
 */
1525
struct page *read_cache_page_async(struct address_space *mapping,
1526
				pgoff_t index,
L
Linus Torvalds 已提交
1527 1528 1529 1530 1531 1532 1533 1534 1535
				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))
1536
		return page;
L
Linus Torvalds 已提交
1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552
	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);
1553
		return ERR_PTR(err);
L
Linus Torvalds 已提交
1554
	}
1555
out:
1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573
	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,
1574
				pgoff_t index,
1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587
				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
1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598
 out:
	return page;
}
EXPORT_SYMBOL(read_cache_page);

/*
 * The logic we want is
 *
 *	if suid or (sgid and xgrp)
 *		remove privs
 */
1599
int should_remove_suid(struct dentry *dentry)
L
Linus Torvalds 已提交
1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614
{
	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;

1615 1616
	if (unlikely(kill && !capable(CAP_FSETID)))
		return kill;
L
Linus Torvalds 已提交
1617

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

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)
{
1632 1633 1634
	int killsuid = should_remove_suid(dentry);
	int killpriv = security_inode_need_killpriv(dentry);
	int error = 0;
1635

1636 1637 1638 1639 1640 1641
	if (killpriv < 0)
		return killpriv;
	if (killpriv)
		error = security_inode_killpriv(dentry);
	if (!error && killsuid)
		error = __remove_suid(dentry, killsuid);
1642

1643
	return error;
L
Linus Torvalds 已提交
1644 1645 1646
}
EXPORT_SYMBOL(remove_suid);

N
Nick Piggin 已提交
1647
static size_t __iovec_copy_from_user_inatomic(char *vaddr,
L
Linus Torvalds 已提交
1648 1649 1650 1651 1652 1653 1654 1655 1656
			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;
1657
		left = __copy_from_user_inatomic_nocache(vaddr, buf, copy);
L
Linus Torvalds 已提交
1658 1659 1660 1661 1662
		copied += copy;
		bytes -= copy;
		vaddr += copy;
		iov++;

1663
		if (unlikely(left))
L
Linus Torvalds 已提交
1664 1665 1666 1667 1668
			break;
	}
	return copied - left;
}

N
Nick Piggin 已提交
1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695
/*
 * 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
1696
EXPORT_SYMBOL(iov_iter_copy_from_user_atomic);
N
Nick Piggin 已提交
1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722

/*
 * 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
1723
EXPORT_SYMBOL(iov_iter_copy_from_user);
N
Nick Piggin 已提交
1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754

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
1755
EXPORT_SYMBOL(iov_iter_advance);
N
Nick Piggin 已提交
1756

1757 1758 1759 1760 1761 1762 1763 1764 1765 1766
/*
 * 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
1767 1768
{
	char __user *buf = i->iov->iov_base + i->iov_offset;
1769 1770
	bytes = min(bytes, i->iov->iov_len - i->iov_offset);
	return fault_in_pages_readable(buf, bytes);
N
Nick Piggin 已提交
1771
}
N
Nick Piggin 已提交
1772
EXPORT_SYMBOL(iov_iter_fault_in_readable);
N
Nick Piggin 已提交
1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784

/*
 * 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
1785
EXPORT_SYMBOL(iov_iter_single_seg_count);
N
Nick Piggin 已提交
1786

L
Linus Torvalds 已提交
1787 1788 1789
/*
 * Performs necessary checks before doing a write
 *
1790
 * Can adjust writing position or amount of bytes to write.
L
Linus Torvalds 已提交
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 1841 1842 1843 1844 1845 1846 1847 1848
 * 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 {
1849
#ifdef CONFIG_BLOCK
L
Linus Torvalds 已提交
1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860
		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;
1861 1862 1863
#else
		return -EPERM;
#endif
L
Linus Torvalds 已提交
1864 1865 1866 1867 1868
	}
	return 0;
}
EXPORT_SYMBOL(generic_write_checks);

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 1902 1903 1904 1905 1906 1907 1908 1909
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
1910
			unlock_page(page);
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 1946 1947 1948 1949 1950 1951 1952 1953
			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
1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979
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.
1980
	 * i_mutex is held, which protects generic_osync_inode() from
1981
	 * livelocking.  AIO O_DIRECT ops attempt to sync metadata here.
L
Linus Torvalds 已提交
1982
	 */
1983 1984
	if ((written >= 0 || written == -EIOCBQUEUED) &&
	    ((file->f_flags & O_SYNC) || IS_SYNC(inode))) {
1985 1986 1987 1988
		int err = generic_osync_inode(inode, mapping, OSYNC_METADATA);
		if (err < 0)
			written = err;
	}
L
Linus Torvalds 已提交
1989 1990 1991 1992
	return written;
}
EXPORT_SYMBOL(generic_file_direct_write);

N
Nick Piggin 已提交
1993 1994 1995 1996
/*
 * Find or create a page at the given pagecache position. Return the locked
 * page. This function is specifically for buffered writes.
 */
1997
struct page *__grab_cache_page(struct address_space *mapping, pgoff_t index)
N
Nick Piggin 已提交
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017
{
	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;
}
2018
EXPORT_SYMBOL(__grab_cache_page);
N
Nick Piggin 已提交
2019

2020 2021
static ssize_t generic_perform_write_2copy(struct file *file,
				struct iov_iter *i, loff_t pos)
L
Linus Torvalds 已提交
2022
{
A
Andrew Morton 已提交
2023
	struct address_space *mapping = file->f_mapping;
2024
	const struct address_space_operations *a_ops = mapping->a_ops;
2025 2026 2027
	struct inode *inode = mapping->host;
	long status = 0;
	ssize_t written = 0;
L
Linus Torvalds 已提交
2028 2029

	do {
N
Nick Piggin 已提交
2030
		struct page *src_page;
N
Nick Piggin 已提交
2031
		struct page *page;
A
Andrew Morton 已提交
2032 2033
		pgoff_t index;		/* Pagecache index for current page */
		unsigned long offset;	/* Offset into pagecache page */
N
Nick Piggin 已提交
2034
		unsigned long bytes;	/* Bytes to write to page */
A
Andrew Morton 已提交
2035
		size_t copied;		/* Bytes copied from user */
L
Linus Torvalds 已提交
2036

A
Andrew Morton 已提交
2037
		offset = (pos & (PAGE_CACHE_SIZE - 1));
L
Linus Torvalds 已提交
2038
		index = pos >> PAGE_CACHE_SHIFT;
N
Nick Piggin 已提交
2039
		bytes = min_t(unsigned long, PAGE_CACHE_SIZE - offset,
2040
						iov_iter_count(i));
2041

N
Nick Piggin 已提交
2042 2043 2044 2045 2046 2047
		/*
		 * a non-NULL src_page indicates that we're doing the
		 * copy via get_user_pages and kmap.
		 */
		src_page = NULL;

2048 2049 2050 2051 2052
		/*
		 * 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
2053 2054 2055 2056
		 *
		 * 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.
2057
		 */
2058
		if (unlikely(iov_iter_fault_in_readable(i, bytes))) {
N
Nick Piggin 已提交
2059 2060 2061
			status = -EFAULT;
			break;
		}
N
Nick Piggin 已提交
2062 2063

		page = __grab_cache_page(mapping, index);
L
Linus Torvalds 已提交
2064 2065 2066 2067 2068
		if (!page) {
			status = -ENOMEM;
			break;
		}

N
Nick Piggin 已提交
2069 2070 2071 2072 2073
		/*
		 * 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
2074
		if (!PageUptodate(page) && !segment_eq(get_fs(), KERNEL_DS)) {
N
Nick Piggin 已提交
2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088
			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).
			 */
2089
			copied = iov_iter_copy_from_user(src_page, i,
N
Nick Piggin 已提交
2090
								offset, bytes);
N
Nick Piggin 已提交
2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115
			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
2116
		status = a_ops->prepare_write(file, page, offset, offset+bytes);
N
Nick Piggin 已提交
2117 2118
		if (unlikely(status))
			goto fs_write_aop_error;
2119

N
Nick Piggin 已提交
2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134
		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();
2135
			copied = iov_iter_copy_from_user_atomic(page, i,
N
Nick Piggin 已提交
2136
								offset, bytes);
N
Nick Piggin 已提交
2137 2138 2139 2140 2141 2142 2143 2144 2145 2146
			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
2147
		flush_dcache_page(page);
N
Nick Piggin 已提交
2148

L
Linus Torvalds 已提交
2149
		status = a_ops->commit_write(file, page, offset, offset+bytes);
N
Nick Piggin 已提交
2150
		if (unlikely(status < 0))
N
Nick Piggin 已提交
2151 2152
			goto fs_write_aop_error;
		if (unlikely(status > 0)) /* filesystem did partial write */
N
Nick Piggin 已提交
2153 2154 2155 2156 2157 2158 2159
			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
2160

2161
		iov_iter_advance(i, copied);
N
Nick Piggin 已提交
2162
		pos += copied;
2163
		written += copied;
N
Nick Piggin 已提交
2164

L
Linus Torvalds 已提交
2165 2166
		balance_dirty_pages_ratelimited(mapping);
		cond_resched();
N
Nick Piggin 已提交
2167 2168 2169
		continue;

fs_write_aop_error:
N
Nick Piggin 已提交
2170
		unlock_page(page);
N
Nick Piggin 已提交
2171
		page_cache_release(page);
N
Nick Piggin 已提交
2172 2173
		if (src_page)
			page_cache_release(src_page);
N
Nick Piggin 已提交
2174 2175 2176 2177 2178 2179 2180 2181

		/*
		 * 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
2182
		break;
2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194
	} 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
2195 2196 2197 2198 2199 2200 2201
	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;
2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232

	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
2233
		status = a_ops->write_begin(file, mapping, pos, bytes, flags,
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 2284 2285 2286 2287 2288 2289 2290 2291
						&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
2292 2293

	if (likely(status >= 0)) {
2294 2295 2296 2297 2298 2299 2300
		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
2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319
		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);

2320
static ssize_t
L
Linus Torvalds 已提交
2321 2322 2323 2324
__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;
2325
	struct address_space * mapping = file->f_mapping;
L
Linus Torvalds 已提交
2326 2327 2328 2329 2330 2331 2332 2333
	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;
2334 2335 2336
	err = generic_segment_checks(iov, &nr_segs, &ocount, VERIFY_READ);
	if (err)
		return err;
L
Linus Torvalds 已提交
2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353

	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;

2354
	err = remove_suid(file->f_path.dentry);
L
Linus Torvalds 已提交
2355 2356 2357
	if (err)
		goto out;

2358
	file_update_time(file);
L
Linus Torvalds 已提交
2359 2360 2361

	/* coalesce the iovecs and go direct-to-BIO for O_DIRECT */
	if (unlikely(file->f_flags & O_DIRECT)) {
2362 2363 2364 2365 2366
		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
2367 2368 2369 2370 2371 2372 2373 2374
		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;
2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388
		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
2389

2390 2391 2392 2393 2394 2395
		/*
		 * 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
2396 2397 2398 2399
		err = do_sync_mapping_range(file->f_mapping, pos, endbyte,
					    SYNC_FILE_RANGE_WAIT_BEFORE|
					    SYNC_FILE_RANGE_WRITE|
					    SYNC_FILE_RANGE_WAIT_AFTER);
2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414
		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
2415 2416 2417 2418 2419
out:
	current->backing_dev_info = NULL;
	return written ? written : err;
}

2420 2421
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
2422 2423 2424 2425 2426 2427
{
	struct file *file = iocb->ki_filp;
	struct address_space *mapping = file->f_mapping;
	struct inode *inode = mapping->host;
	ssize_t ret;

2428 2429 2430 2431
	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
2432 2433

	if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) {
2434
		ssize_t err;
L
Linus Torvalds 已提交
2435 2436 2437 2438 2439 2440 2441

		err = sync_page_range_nolock(inode, mapping, pos, ret);
		if (err < 0)
			ret = err;
	}
	return ret;
}
2442
EXPORT_SYMBOL(generic_file_aio_write_nolock);
L
Linus Torvalds 已提交
2443

2444 2445
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
2446 2447 2448 2449 2450 2451 2452 2453
{
	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);

2454
	mutex_lock(&inode->i_mutex);
2455 2456
	ret = __generic_file_aio_write_nolock(iocb, iov, nr_segs,
			&iocb->ki_pos);
2457
	mutex_unlock(&inode->i_mutex);
L
Linus Torvalds 已提交
2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470

	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);

/*
2471
 * Called under i_mutex for writes to S_ISREG files.   Returns -EIO if something
L
Linus Torvalds 已提交
2472 2473
 * went wrong during pagecache shootdown.
 */
2474
static ssize_t
L
Linus Torvalds 已提交
2475 2476 2477 2478 2479 2480
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;
2481 2482
	size_t write_len;
	pgoff_t end = 0; /* silence gcc */
L
Linus Torvalds 已提交
2483 2484 2485 2486 2487 2488 2489 2490

	/*
	 * 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);
2491
		end = (offset + write_len - 1) >> PAGE_CACHE_SHIFT;
L
Linus Torvalds 已提交
2492 2493 2494 2495 2496
	       	if (mapping_mapped(mapping))
			unmap_mapping_range(mapping, offset, write_len, 0);
	}

	retval = filemap_write_and_wait(mapping);
2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507
	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
2508
					offset >> PAGE_CACHE_SHIFT, end);
2509 2510
		if (retval)
			goto out;
L
Linus Torvalds 已提交
2511
	}
2512 2513 2514 2515 2516

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

	/*
	 * Finally, try again to invalidate clean pages which might have been
2517 2518 2519 2520 2521
	 * cached by non-direct readahead, or faulted in by get_user_pages()
	 * if the source of the write was an mmap'ed region of the file
	 * we're writing.  Either one is a pretty crazy thing to do,
	 * so we don't support it 100%.  If this invalidation
	 * fails, tough, the write still worked...
2522 2523
	 */
	if (rw == WRITE && mapping->nrpages) {
2524
		invalidate_inode_pages2_range(mapping, offset >> PAGE_CACHE_SHIFT, end);
2525 2526
	}
out:
L
Linus Torvalds 已提交
2527 2528
	return retval;
}
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 2555 2556 2557 2558

/**
 * 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);