dax.c 39.0 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
/*
 * fs/dax.c - Direct Access filesystem code
 * Copyright (c) 2013-2014 Intel Corporation
 * Author: Matthew Wilcox <matthew.r.wilcox@intel.com>
 * Author: Ross Zwisler <ross.zwisler@linux.intel.com>
 *
 * This program is free software; you can redistribute it and/or modify it
 * under the terms and conditions of the GNU General Public License,
 * version 2, as published by the Free Software Foundation.
 *
 * This program is distributed in the hope it will be useful, but WITHOUT
 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
 * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
 * more details.
 */

#include <linux/atomic.h>
#include <linux/blkdev.h>
#include <linux/buffer_head.h>
20
#include <linux/dax.h>
21 22
#include <linux/fs.h>
#include <linux/genhd.h>
23 24 25
#include <linux/highmem.h>
#include <linux/memcontrol.h>
#include <linux/mm.h>
26
#include <linux/mutex.h>
R
Ross Zwisler 已提交
27
#include <linux/pagevec.h>
28
#include <linux/pmem.h>
29
#include <linux/sched.h>
30
#include <linux/uio.h>
31
#include <linux/vmstat.h>
D
Dan Williams 已提交
32
#include <linux/pfn_t.h>
33
#include <linux/sizes.h>
34
#include <linux/mmu_notifier.h>
35 36
#include <linux/iomap.h>
#include "internal.h"
37

J
Jan Kara 已提交
38 39 40 41
/* We choose 4096 entries - same as per-zone page wait tables */
#define DAX_WAIT_TABLE_BITS 12
#define DAX_WAIT_TABLE_ENTRIES (1 << DAX_WAIT_TABLE_BITS)

42
static wait_queue_head_t wait_table[DAX_WAIT_TABLE_ENTRIES];
J
Jan Kara 已提交
43 44 45 46 47 48 49 50 51 52 53

static int __init init_dax_wait_table(void)
{
	int i;

	for (i = 0; i < DAX_WAIT_TABLE_ENTRIES; i++)
		init_waitqueue_head(wait_table + i);
	return 0;
}
fs_initcall(init_dax_wait_table);

54 55 56 57 58
static long dax_map_atomic(struct block_device *bdev, struct blk_dax_ctl *dax)
{
	struct request_queue *q = bdev->bd_queue;
	long rc = -EIO;

D
Dan Williams 已提交
59
	dax->addr = ERR_PTR(-EIO);
60 61 62 63 64
	if (blk_queue_enter(q, true) != 0)
		return rc;

	rc = bdev_direct_access(bdev, dax);
	if (rc < 0) {
D
Dan Williams 已提交
65
		dax->addr = ERR_PTR(rc);
66 67 68 69 70 71 72 73 74 75 76 77 78 79
		blk_queue_exit(q);
		return rc;
	}
	return rc;
}

static void dax_unmap_atomic(struct block_device *bdev,
		const struct blk_dax_ctl *dax)
{
	if (IS_ERR(dax->addr))
		return;
	blk_queue_exit(bdev->bd_queue);
}

80
static int dax_is_pmd_entry(void *entry)
81
{
82
	return (unsigned long)entry & RADIX_DAX_PMD;
83 84
}

85
static int dax_is_pte_entry(void *entry)
86
{
87
	return !((unsigned long)entry & RADIX_DAX_PMD);
88 89
}

90
static int dax_is_zero_entry(void *entry)
91
{
92
	return (unsigned long)entry & RADIX_DAX_HZP;
93 94
}

95
static int dax_is_empty_entry(void *entry)
96
{
97
	return (unsigned long)entry & RADIX_DAX_EMPTY;
98 99
}

100
struct page *read_dax_sector(struct block_device *bdev, sector_t n)
101
{
102
	struct page *page = alloc_pages(GFP_KERNEL, 0);
103
	struct blk_dax_ctl dax = {
104 105
		.size = PAGE_SIZE,
		.sector = n & ~((((int) PAGE_SIZE) / 512) - 1),
106
	};
107
	long rc;
108

109 110
	if (!page)
		return ERR_PTR(-ENOMEM);
111

112 113 114 115
	rc = dax_map_atomic(bdev, &dax);
	if (rc < 0)
		return ERR_PTR(rc);
	memcpy_from_pmem(page_address(page), dax.addr, PAGE_SIZE);
116
	dax_unmap_atomic(bdev, &dax);
117
	return page;
118
}
119

J
Jan Kara 已提交
120 121 122 123 124
/*
 * DAX radix tree locking
 */
struct exceptional_entry_key {
	struct address_space *mapping;
125
	pgoff_t entry_start;
J
Jan Kara 已提交
126 127 128 129 130 131 132
};

struct wait_exceptional_entry_queue {
	wait_queue_t wait;
	struct exceptional_entry_key key;
};

133 134 135 136 137 138 139 140 141 142
static wait_queue_head_t *dax_entry_waitqueue(struct address_space *mapping,
		pgoff_t index, void *entry, struct exceptional_entry_key *key)
{
	unsigned long hash;

	/*
	 * If 'entry' is a PMD, align the 'index' that we use for the wait
	 * queue to the start of that PMD.  This ensures that all offsets in
	 * the range covered by the PMD map to the same bit lock.
	 */
143
	if (dax_is_pmd_entry(entry))
144 145 146 147 148 149 150 151 152
		index &= ~((1UL << (PMD_SHIFT - PAGE_SHIFT)) - 1);

	key->mapping = mapping;
	key->entry_start = index;

	hash = hash_long((unsigned long)mapping ^ index, DAX_WAIT_TABLE_BITS);
	return wait_table + hash;
}

J
Jan Kara 已提交
153 154 155 156 157 158 159 160
static int wake_exceptional_entry_func(wait_queue_t *wait, unsigned int mode,
				       int sync, void *keyp)
{
	struct exceptional_entry_key *key = keyp;
	struct wait_exceptional_entry_queue *ewait =
		container_of(wait, struct wait_exceptional_entry_queue, wait);

	if (key->mapping != ewait->key.mapping ||
161
	    key->entry_start != ewait->key.entry_start)
J
Jan Kara 已提交
162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186
		return 0;
	return autoremove_wake_function(wait, mode, sync, NULL);
}

/*
 * Check whether the given slot is locked. The function must be called with
 * mapping->tree_lock held
 */
static inline int slot_locked(struct address_space *mapping, void **slot)
{
	unsigned long entry = (unsigned long)
		radix_tree_deref_slot_protected(slot, &mapping->tree_lock);
	return entry & RADIX_DAX_ENTRY_LOCK;
}

/*
 * Mark the given slot is locked. The function must be called with
 * mapping->tree_lock held
 */
static inline void *lock_slot(struct address_space *mapping, void **slot)
{
	unsigned long entry = (unsigned long)
		radix_tree_deref_slot_protected(slot, &mapping->tree_lock);

	entry |= RADIX_DAX_ENTRY_LOCK;
187
	radix_tree_replace_slot(&mapping->page_tree, slot, (void *)entry);
J
Jan Kara 已提交
188 189 190 191 192 193 194 195 196 197 198 199 200
	return (void *)entry;
}

/*
 * Mark the given slot is unlocked. The function must be called with
 * mapping->tree_lock held
 */
static inline void *unlock_slot(struct address_space *mapping, void **slot)
{
	unsigned long entry = (unsigned long)
		radix_tree_deref_slot_protected(slot, &mapping->tree_lock);

	entry &= ~(unsigned long)RADIX_DAX_ENTRY_LOCK;
201
	radix_tree_replace_slot(&mapping->page_tree, slot, (void *)entry);
J
Jan Kara 已提交
202 203 204 205 206 207 208 209 210 211 212 213 214 215 216
	return (void *)entry;
}

/*
 * Lookup entry in radix tree, wait for it to become unlocked if it is
 * exceptional entry and return it. The caller must call
 * put_unlocked_mapping_entry() when he decided not to lock the entry or
 * put_locked_mapping_entry() when he locked the entry and now wants to
 * unlock it.
 *
 * The function must be called with mapping->tree_lock held.
 */
static void *get_unlocked_mapping_entry(struct address_space *mapping,
					pgoff_t index, void ***slotp)
{
217
	void *entry, **slot;
J
Jan Kara 已提交
218
	struct wait_exceptional_entry_queue ewait;
219
	wait_queue_head_t *wq;
J
Jan Kara 已提交
220 221 222 223 224

	init_wait(&ewait.wait);
	ewait.wait.func = wake_exceptional_entry_func;

	for (;;) {
225
		entry = __radix_tree_lookup(&mapping->page_tree, index, NULL,
J
Jan Kara 已提交
226
					  &slot);
227
		if (!entry || !radix_tree_exceptional_entry(entry) ||
J
Jan Kara 已提交
228 229 230
		    !slot_locked(mapping, slot)) {
			if (slotp)
				*slotp = slot;
231
			return entry;
J
Jan Kara 已提交
232
		}
233 234

		wq = dax_entry_waitqueue(mapping, index, entry, &ewait.key);
J
Jan Kara 已提交
235 236 237 238 239 240 241 242 243
		prepare_to_wait_exclusive(wq, &ewait.wait,
					  TASK_UNINTERRUPTIBLE);
		spin_unlock_irq(&mapping->tree_lock);
		schedule();
		finish_wait(wq, &ewait.wait);
		spin_lock_irq(&mapping->tree_lock);
	}
}

244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260
static void dax_unlock_mapping_entry(struct address_space *mapping,
				     pgoff_t index)
{
	void *entry, **slot;

	spin_lock_irq(&mapping->tree_lock);
	entry = __radix_tree_lookup(&mapping->page_tree, index, NULL, &slot);
	if (WARN_ON_ONCE(!entry || !radix_tree_exceptional_entry(entry) ||
			 !slot_locked(mapping, slot))) {
		spin_unlock_irq(&mapping->tree_lock);
		return;
	}
	unlock_slot(mapping, slot);
	spin_unlock_irq(&mapping->tree_lock);
	dax_wake_mapping_entry_waiter(mapping, index, entry, false);
}

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
static void put_locked_mapping_entry(struct address_space *mapping,
				     pgoff_t index, void *entry)
{
	if (!radix_tree_exceptional_entry(entry)) {
		unlock_page(entry);
		put_page(entry);
	} else {
		dax_unlock_mapping_entry(mapping, index);
	}
}

/*
 * Called when we are done with radix tree entry we looked up via
 * get_unlocked_mapping_entry() and which we didn't lock in the end.
 */
static void put_unlocked_mapping_entry(struct address_space *mapping,
				       pgoff_t index, void *entry)
{
	if (!radix_tree_exceptional_entry(entry))
		return;

	/* We have to wake up next waiter for the radix tree entry lock */
	dax_wake_mapping_entry_waiter(mapping, index, entry, false);
}

J
Jan Kara 已提交
286 287 288 289 290 291
/*
 * Find radix tree entry at given index. If it points to a page, return with
 * the page locked. If it points to the exceptional entry, return with the
 * radix tree entry locked. If the radix tree doesn't contain given index,
 * create empty exceptional entry for the index and return with it locked.
 *
292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307
 * When requesting an entry with size RADIX_DAX_PMD, grab_mapping_entry() will
 * either return that locked entry or will return an error.  This error will
 * happen if there are any 4k entries (either zero pages or DAX entries)
 * within the 2MiB range that we are requesting.
 *
 * We always favor 4k entries over 2MiB entries. There isn't a flow where we
 * evict 4k entries in order to 'upgrade' them to a 2MiB entry.  A 2MiB
 * insertion will fail if it finds any 4k entries already in the tree, and a
 * 4k insertion will cause an existing 2MiB entry to be unmapped and
 * downgraded to 4k entries.  This happens for both 2MiB huge zero pages as
 * well as 2MiB empty entries.
 *
 * The exception to this downgrade path is for 2MiB DAX PMD entries that have
 * real storage backing them.  We will leave these real 2MiB DAX entries in
 * the tree, and PTE writes will simply dirty the entire 2MiB DAX entry.
 *
J
Jan Kara 已提交
308 309 310 311
 * Note: Unlike filemap_fault() we don't honor FAULT_FLAG_RETRY flags. For
 * persistent memory the benefit is doubtful. We can add that later if we can
 * show it helps.
 */
312 313
static void *grab_mapping_entry(struct address_space *mapping, pgoff_t index,
		unsigned long size_flag)
J
Jan Kara 已提交
314
{
315
	bool pmd_downgrade = false; /* splitting 2MiB entry into 4k entries? */
316
	void *entry, **slot;
J
Jan Kara 已提交
317 318 319

restart:
	spin_lock_irq(&mapping->tree_lock);
320
	entry = get_unlocked_mapping_entry(mapping, index, &slot);
321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340

	if (entry) {
		if (size_flag & RADIX_DAX_PMD) {
			if (!radix_tree_exceptional_entry(entry) ||
			    dax_is_pte_entry(entry)) {
				put_unlocked_mapping_entry(mapping, index,
						entry);
				entry = ERR_PTR(-EEXIST);
				goto out_unlock;
			}
		} else { /* trying to grab a PTE entry */
			if (radix_tree_exceptional_entry(entry) &&
			    dax_is_pmd_entry(entry) &&
			    (dax_is_zero_entry(entry) ||
			     dax_is_empty_entry(entry))) {
				pmd_downgrade = true;
			}
		}
	}

J
Jan Kara 已提交
341
	/* No entry for given index? Make sure radix tree is big enough. */
342
	if (!entry || pmd_downgrade) {
J
Jan Kara 已提交
343 344
		int err;

345 346 347 348 349 350 351 352
		if (pmd_downgrade) {
			/*
			 * Make sure 'entry' remains valid while we drop
			 * mapping->tree_lock.
			 */
			entry = lock_slot(mapping, slot);
		}

J
Jan Kara 已提交
353
		spin_unlock_irq(&mapping->tree_lock);
354 355 356 357 358 359 360 361 362
		/*
		 * Besides huge zero pages the only other thing that gets
		 * downgraded are empty entries which don't need to be
		 * unmapped.
		 */
		if (pmd_downgrade && dax_is_zero_entry(entry))
			unmap_mapping_range(mapping,
				(index << PAGE_SHIFT) & PMD_MASK, PMD_SIZE, 0);

J
Jan Kara 已提交
363 364
		err = radix_tree_preload(
				mapping_gfp_mask(mapping) & ~__GFP_HIGHMEM);
365 366 367
		if (err) {
			if (pmd_downgrade)
				put_locked_mapping_entry(mapping, index, entry);
J
Jan Kara 已提交
368
			return ERR_PTR(err);
369
		}
J
Jan Kara 已提交
370
		spin_lock_irq(&mapping->tree_lock);
371 372 373 374 375 376 377 378 379 380 381 382

		if (pmd_downgrade) {
			radix_tree_delete(&mapping->page_tree, index);
			mapping->nrexceptional--;
			dax_wake_mapping_entry_waiter(mapping, index, entry,
					true);
		}

		entry = dax_radix_locked_entry(0, size_flag | RADIX_DAX_EMPTY);

		err = __radix_tree_insert(&mapping->page_tree, index,
				dax_radix_order(entry), entry);
J
Jan Kara 已提交
383 384 385
		radix_tree_preload_end();
		if (err) {
			spin_unlock_irq(&mapping->tree_lock);
386 387 388 389 390 391 392
			/*
			 * Someone already created the entry?  This is a
			 * normal failure when inserting PMDs in a range
			 * that already contains PTEs.  In that case we want
			 * to return -EEXIST immediately.
			 */
			if (err == -EEXIST && !(size_flag & RADIX_DAX_PMD))
J
Jan Kara 已提交
393
				goto restart;
394 395 396 397 398 399 400
			/*
			 * Our insertion of a DAX PMD entry failed, most
			 * likely because it collided with a PTE sized entry
			 * at a different index in the PMD range.  We haven't
			 * inserted anything into the radix tree and have no
			 * waiters to wake.
			 */
J
Jan Kara 已提交
401 402 403 404 405
			return ERR_PTR(err);
		}
		/* Good, we have inserted empty locked entry into the tree. */
		mapping->nrexceptional++;
		spin_unlock_irq(&mapping->tree_lock);
406
		return entry;
J
Jan Kara 已提交
407 408
	}
	/* Normal page in radix tree? */
409 410
	if (!radix_tree_exceptional_entry(entry)) {
		struct page *page = entry;
J
Jan Kara 已提交
411 412 413 414 415 416 417 418 419 420 421 422

		get_page(page);
		spin_unlock_irq(&mapping->tree_lock);
		lock_page(page);
		/* Page got truncated? Retry... */
		if (unlikely(page->mapping != mapping)) {
			unlock_page(page);
			put_page(page);
			goto restart;
		}
		return page;
	}
423
	entry = lock_slot(mapping, slot);
424
 out_unlock:
J
Jan Kara 已提交
425
	spin_unlock_irq(&mapping->tree_lock);
426
	return entry;
J
Jan Kara 已提交
427 428
}

429 430 431
/*
 * We do not necessarily hold the mapping->tree_lock when we call this
 * function so it is possible that 'entry' is no longer a valid item in the
432 433 434
 * radix tree.  This is okay because all we really need to do is to find the
 * correct waitqueue where tasks might be waiting for that old 'entry' and
 * wake them.
435
 */
J
Jan Kara 已提交
436
void dax_wake_mapping_entry_waiter(struct address_space *mapping,
437
		pgoff_t index, void *entry, bool wake_all)
J
Jan Kara 已提交
438
{
439 440 441 442
	struct exceptional_entry_key key;
	wait_queue_head_t *wq;

	wq = dax_entry_waitqueue(mapping, index, entry, &key);
J
Jan Kara 已提交
443 444 445 446 447 448 449

	/*
	 * Checking for locked entry and prepare_to_wait_exclusive() happens
	 * under mapping->tree_lock, ditto for entry handling in our callers.
	 * So at this point all tasks that could have seen our entry locked
	 * must be in the waitqueue and the following check will see them.
	 */
450
	if (waitqueue_active(wq))
J
Jan Kara 已提交
451 452 453
		__wake_up(wq, TASK_NORMAL, wake_all ? 0 : 1, &key);
}

454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476
static int __dax_invalidate_mapping_entry(struct address_space *mapping,
					  pgoff_t index, bool trunc)
{
	int ret = 0;
	void *entry;
	struct radix_tree_root *page_tree = &mapping->page_tree;

	spin_lock_irq(&mapping->tree_lock);
	entry = get_unlocked_mapping_entry(mapping, index, NULL);
	if (!entry || !radix_tree_exceptional_entry(entry))
		goto out;
	if (!trunc &&
	    (radix_tree_tag_get(page_tree, index, PAGECACHE_TAG_DIRTY) ||
	     radix_tree_tag_get(page_tree, index, PAGECACHE_TAG_TOWRITE)))
		goto out;
	radix_tree_delete(page_tree, index);
	mapping->nrexceptional--;
	ret = 1;
out:
	put_unlocked_mapping_entry(mapping, index, entry);
	spin_unlock_irq(&mapping->tree_lock);
	return ret;
}
J
Jan Kara 已提交
477 478 479 480 481 482
/*
 * Delete exceptional DAX entry at @index from @mapping. Wait for radix tree
 * entry to get unlocked before deleting it.
 */
int dax_delete_mapping_entry(struct address_space *mapping, pgoff_t index)
{
483
	int ret = __dax_invalidate_mapping_entry(mapping, index, true);
J
Jan Kara 已提交
484 485 486 487 488 489 490 491

	/*
	 * This gets called from truncate / punch_hole path. As such, the caller
	 * must hold locks protecting against concurrent modifications of the
	 * radix tree (usually fs-private i_mmap_sem for writing). Since the
	 * caller has seen exceptional entry for this index, we better find it
	 * at that index as well...
	 */
492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515
	WARN_ON_ONCE(!ret);
	return ret;
}

/*
 * Invalidate exceptional DAX entry if easily possible. This handles DAX
 * entries for invalidate_inode_pages() so we evict the entry only if we can
 * do so without blocking.
 */
int dax_invalidate_mapping_entry(struct address_space *mapping, pgoff_t index)
{
	int ret = 0;
	void *entry, **slot;
	struct radix_tree_root *page_tree = &mapping->page_tree;

	spin_lock_irq(&mapping->tree_lock);
	entry = __radix_tree_lookup(page_tree, index, NULL, &slot);
	if (!entry || !radix_tree_exceptional_entry(entry) ||
	    slot_locked(mapping, slot))
		goto out;
	if (radix_tree_tag_get(page_tree, index, PAGECACHE_TAG_DIRTY) ||
	    radix_tree_tag_get(page_tree, index, PAGECACHE_TAG_TOWRITE))
		goto out;
	radix_tree_delete(page_tree, index);
J
Jan Kara 已提交
516
	mapping->nrexceptional--;
517 518
	ret = 1;
out:
J
Jan Kara 已提交
519
	spin_unlock_irq(&mapping->tree_lock);
520 521 522 523
	if (ret)
		dax_wake_mapping_entry_waiter(mapping, index, entry, true);
	return ret;
}
J
Jan Kara 已提交
524

525 526 527 528 529 530 531
/*
 * Invalidate exceptional DAX entry if it is clean.
 */
int dax_invalidate_mapping_entry_sync(struct address_space *mapping,
				      pgoff_t index)
{
	return __dax_invalidate_mapping_entry(mapping, index, false);
J
Jan Kara 已提交
532 533
}

534 535 536 537 538 539 540 541
/*
 * The user has performed a load from a hole in the file.  Allocating
 * a new page in the file would cause excessive storage usage for
 * workloads with sparse files.  We allocate a page cache page instead.
 * We'll kick it out of the page cache if it's ever written to,
 * otherwise it will simply fall out of the page cache under memory
 * pressure without ever having been dirtied.
 */
J
Jan Kara 已提交
542 543
static int dax_load_hole(struct address_space *mapping, void *entry,
			 struct vm_fault *vmf)
544
{
J
Jan Kara 已提交
545
	struct page *page;
546

J
Jan Kara 已提交
547 548 549 550 551
	/* Hole page already exists? Return it...  */
	if (!radix_tree_exceptional_entry(entry)) {
		vmf->page = entry;
		return VM_FAULT_LOCKED;
	}
552

J
Jan Kara 已提交
553 554 555
	/* This will replace locked radix tree entry with a hole page */
	page = find_or_create_page(mapping, vmf->pgoff,
				   vmf->gfp_mask | __GFP_ZERO);
556
	if (!page)
J
Jan Kara 已提交
557
		return VM_FAULT_OOM;
558 559 560 561
	vmf->page = page;
	return VM_FAULT_LOCKED;
}

562 563
static int copy_user_dax(struct block_device *bdev, sector_t sector, size_t size,
		struct page *to, unsigned long vaddr)
564
{
565
	struct blk_dax_ctl dax = {
566 567
		.sector = sector,
		.size = size,
568
	};
569 570
	void *vto;

571 572
	if (dax_map_atomic(bdev, &dax) < 0)
		return PTR_ERR(dax.addr);
573
	vto = kmap_atomic(to);
574
	copy_user_page(vto, (void __force *)dax.addr, vaddr, to);
575
	kunmap_atomic(vto);
576
	dax_unmap_atomic(bdev, &dax);
577 578 579
	return 0;
}

580 581 582 583 584 585 586
/*
 * By this point grab_mapping_entry() has ensured that we have a locked entry
 * of the appropriate size so we don't have to worry about downgrading PMDs to
 * PTEs.  If we happen to be trying to insert a PTE and there is a PMD
 * already in the tree, we will skip the insertion and just dirty the PMD as
 * appropriate.
 */
J
Jan Kara 已提交
587 588
static void *dax_insert_mapping_entry(struct address_space *mapping,
				      struct vm_fault *vmf,
589 590
				      void *entry, sector_t sector,
				      unsigned long flags)
R
Ross Zwisler 已提交
591 592
{
	struct radix_tree_root *page_tree = &mapping->page_tree;
J
Jan Kara 已提交
593 594 595 596
	int error = 0;
	bool hole_fill = false;
	void *new_entry;
	pgoff_t index = vmf->pgoff;
R
Ross Zwisler 已提交
597

J
Jan Kara 已提交
598
	if (vmf->flags & FAULT_FLAG_WRITE)
599
		__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
R
Ross Zwisler 已提交
600

J
Jan Kara 已提交
601 602 603 604 605 606 607 608 609 610 611 612
	/* Replacing hole page with block mapping? */
	if (!radix_tree_exceptional_entry(entry)) {
		hole_fill = true;
		/*
		 * Unmap the page now before we remove it from page cache below.
		 * The page is locked so it cannot be faulted in again.
		 */
		unmap_mapping_range(mapping, vmf->pgoff << PAGE_SHIFT,
				    PAGE_SIZE, 0);
		error = radix_tree_preload(vmf->gfp_mask & ~__GFP_HIGHMEM);
		if (error)
			return ERR_PTR(error);
613 614 615 616
	} else if (dax_is_zero_entry(entry) && !(flags & RADIX_DAX_HZP)) {
		/* replacing huge zero page with PMD block mapping */
		unmap_mapping_range(mapping,
			(vmf->pgoff << PAGE_SHIFT) & PMD_MASK, PMD_SIZE, 0);
R
Ross Zwisler 已提交
617 618
	}

J
Jan Kara 已提交
619
	spin_lock_irq(&mapping->tree_lock);
620 621
	new_entry = dax_radix_locked_entry(sector, flags);

J
Jan Kara 已提交
622 623 624 625
	if (hole_fill) {
		__delete_from_page_cache(entry, NULL);
		/* Drop pagecache reference */
		put_page(entry);
626 627
		error = __radix_tree_insert(page_tree, index,
				dax_radix_order(new_entry), new_entry);
J
Jan Kara 已提交
628 629
		if (error) {
			new_entry = ERR_PTR(error);
R
Ross Zwisler 已提交
630 631
			goto unlock;
		}
J
Jan Kara 已提交
632
		mapping->nrexceptional++;
633 634 635 636 637 638 639 640 641
	} else if (dax_is_zero_entry(entry) || dax_is_empty_entry(entry)) {
		/*
		 * Only swap our new entry into the radix tree if the current
		 * entry is a zero page or an empty entry.  If a normal PTE or
		 * PMD entry is already in the tree, we leave it alone.  This
		 * means that if we are trying to insert a PTE and the
		 * existing entry is a PMD, we will just leave the PMD in the
		 * tree and dirty it if necessary.
		 */
642
		struct radix_tree_node *node;
J
Jan Kara 已提交
643 644
		void **slot;
		void *ret;
R
Ross Zwisler 已提交
645

646
		ret = __radix_tree_lookup(page_tree, index, &node, &slot);
J
Jan Kara 已提交
647
		WARN_ON_ONCE(ret != entry);
648 649
		__radix_tree_replace(page_tree, node, slot,
				     new_entry, NULL, NULL);
R
Ross Zwisler 已提交
650
	}
J
Jan Kara 已提交
651
	if (vmf->flags & FAULT_FLAG_WRITE)
R
Ross Zwisler 已提交
652 653 654
		radix_tree_tag_set(page_tree, index, PAGECACHE_TAG_DIRTY);
 unlock:
	spin_unlock_irq(&mapping->tree_lock);
J
Jan Kara 已提交
655 656 657 658 659 660 661 662 663 664 665 666
	if (hole_fill) {
		radix_tree_preload_end();
		/*
		 * We don't need hole page anymore, it has been replaced with
		 * locked radix tree entry now.
		 */
		if (mapping->a_ops->freepage)
			mapping->a_ops->freepage(entry);
		unlock_page(entry);
		put_page(entry);
	}
	return new_entry;
R
Ross Zwisler 已提交
667 668
}

669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721
static inline unsigned long
pgoff_address(pgoff_t pgoff, struct vm_area_struct *vma)
{
	unsigned long address;

	address = vma->vm_start + ((pgoff - vma->vm_pgoff) << PAGE_SHIFT);
	VM_BUG_ON_VMA(address < vma->vm_start || address >= vma->vm_end, vma);
	return address;
}

/* Walk all mappings of a given index of a file and writeprotect them */
static void dax_mapping_entry_mkclean(struct address_space *mapping,
				      pgoff_t index, unsigned long pfn)
{
	struct vm_area_struct *vma;
	pte_t *ptep;
	pte_t pte;
	spinlock_t *ptl;
	bool changed;

	i_mmap_lock_read(mapping);
	vma_interval_tree_foreach(vma, &mapping->i_mmap, index, index) {
		unsigned long address;

		cond_resched();

		if (!(vma->vm_flags & VM_SHARED))
			continue;

		address = pgoff_address(index, vma);
		changed = false;
		if (follow_pte(vma->vm_mm, address, &ptep, &ptl))
			continue;
		if (pfn != pte_pfn(*ptep))
			goto unlock;
		if (!pte_dirty(*ptep) && !pte_write(*ptep))
			goto unlock;

		flush_cache_page(vma, address, pfn);
		pte = ptep_clear_flush(vma, address, ptep);
		pte = pte_wrprotect(pte);
		pte = pte_mkclean(pte);
		set_pte_at(vma->vm_mm, address, ptep, pte);
		changed = true;
unlock:
		pte_unmap_unlock(ptep, ptl);

		if (changed)
			mmu_notifier_invalidate_page(vma->vm_mm, address);
	}
	i_mmap_unlock_read(mapping);
}

R
Ross Zwisler 已提交
722 723 724 725 726
static int dax_writeback_one(struct block_device *bdev,
		struct address_space *mapping, pgoff_t index, void *entry)
{
	struct radix_tree_root *page_tree = &mapping->page_tree;
	struct blk_dax_ctl dax;
727
	void *entry2, **slot;
R
Ross Zwisler 已提交
728 729 730
	int ret = 0;

	/*
731 732
	 * A page got tagged dirty in DAX mapping? Something is seriously
	 * wrong.
R
Ross Zwisler 已提交
733
	 */
734 735
	if (WARN_ON(!radix_tree_exceptional_entry(entry)))
		return -EIO;
R
Ross Zwisler 已提交
736

737 738 739 740 741 742 743 744 745 746 747 748
	spin_lock_irq(&mapping->tree_lock);
	entry2 = get_unlocked_mapping_entry(mapping, index, &slot);
	/* Entry got punched out / reallocated? */
	if (!entry2 || !radix_tree_exceptional_entry(entry2))
		goto put_unlocked;
	/*
	 * Entry got reallocated elsewhere? No need to writeback. We have to
	 * compare sectors as we must not bail out due to difference in lockbit
	 * or entry type.
	 */
	if (dax_radix_sector(entry2) != dax_radix_sector(entry))
		goto put_unlocked;
749 750
	if (WARN_ON_ONCE(dax_is_empty_entry(entry) ||
				dax_is_zero_entry(entry))) {
R
Ross Zwisler 已提交
751
		ret = -EIO;
752
		goto put_unlocked;
R
Ross Zwisler 已提交
753 754
	}

755 756 757 758 759 760 761 762 763 764 765 766 767 768 769
	/* Another fsync thread may have already written back this entry */
	if (!radix_tree_tag_get(page_tree, index, PAGECACHE_TAG_TOWRITE))
		goto put_unlocked;
	/* Lock the entry to serialize with page faults */
	entry = lock_slot(mapping, slot);
	/*
	 * We can clear the tag now but we have to be careful so that concurrent
	 * dax_writeback_one() calls for the same index cannot finish before we
	 * actually flush the caches. This is achieved as the calls will look
	 * at the entry only under tree_lock and once they do that they will
	 * see the entry locked and wait for it to unlock.
	 */
	radix_tree_tag_clear(page_tree, index, PAGECACHE_TAG_TOWRITE);
	spin_unlock_irq(&mapping->tree_lock);

770 771 772 773 774 775 776 777 778
	/*
	 * Even if dax_writeback_mapping_range() was given a wbc->range_start
	 * in the middle of a PMD, the 'index' we are given will be aligned to
	 * the start index of the PMD, as will the sector we pull from
	 * 'entry'.  This allows us to flush for PMD_SIZE and not have to
	 * worry about partial PMD writebacks.
	 */
	dax.sector = dax_radix_sector(entry);
	dax.size = PAGE_SIZE << dax_radix_order(entry);
R
Ross Zwisler 已提交
779 780 781 782 783 784

	/*
	 * We cannot hold tree_lock while calling dax_map_atomic() because it
	 * eventually calls cond_resched().
	 */
	ret = dax_map_atomic(bdev, &dax);
785 786
	if (ret < 0) {
		put_locked_mapping_entry(mapping, index, entry);
R
Ross Zwisler 已提交
787
		return ret;
788
	}
R
Ross Zwisler 已提交
789 790 791 792 793 794

	if (WARN_ON_ONCE(ret < dax.size)) {
		ret = -EIO;
		goto unmap;
	}

795
	dax_mapping_entry_mkclean(mapping, index, pfn_t_to_pfn(dax.pfn));
R
Ross Zwisler 已提交
796
	wb_cache_pmem(dax.addr, dax.size);
797 798 799 800 801 802 803 804 805
	/*
	 * After we have flushed the cache, we can clear the dirty tag. There
	 * cannot be new dirty data in the pfn after the flush has completed as
	 * the pfn mappings are writeprotected and fault waits for mapping
	 * entry lock.
	 */
	spin_lock_irq(&mapping->tree_lock);
	radix_tree_tag_clear(page_tree, index, PAGECACHE_TAG_DIRTY);
	spin_unlock_irq(&mapping->tree_lock);
R
Ross Zwisler 已提交
806 807
 unmap:
	dax_unmap_atomic(bdev, &dax);
808
	put_locked_mapping_entry(mapping, index, entry);
R
Ross Zwisler 已提交
809 810
	return ret;

811 812
 put_unlocked:
	put_unlocked_mapping_entry(mapping, index, entry2);
R
Ross Zwisler 已提交
813 814 815 816 817 818 819 820 821
	spin_unlock_irq(&mapping->tree_lock);
	return ret;
}

/*
 * Flush the mapping to the persistent domain within the byte range of [start,
 * end]. This is required by data integrity operations to ensure file data is
 * on persistent storage prior to completion of the operation.
 */
822 823
int dax_writeback_mapping_range(struct address_space *mapping,
		struct block_device *bdev, struct writeback_control *wbc)
R
Ross Zwisler 已提交
824 825
{
	struct inode *inode = mapping->host;
826
	pgoff_t start_index, end_index;
R
Ross Zwisler 已提交
827 828 829 830 831 832 833 834
	pgoff_t indices[PAGEVEC_SIZE];
	struct pagevec pvec;
	bool done = false;
	int i, ret = 0;

	if (WARN_ON_ONCE(inode->i_blkbits != PAGE_SHIFT))
		return -EIO;

835 836 837
	if (!mapping->nrexceptional || wbc->sync_mode != WB_SYNC_ALL)
		return 0;

838 839
	start_index = wbc->range_start >> PAGE_SHIFT;
	end_index = wbc->range_end >> PAGE_SHIFT;
R
Ross Zwisler 已提交
840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867

	tag_pages_for_writeback(mapping, start_index, end_index);

	pagevec_init(&pvec, 0);
	while (!done) {
		pvec.nr = find_get_entries_tag(mapping, start_index,
				PAGECACHE_TAG_TOWRITE, PAGEVEC_SIZE,
				pvec.pages, indices);

		if (pvec.nr == 0)
			break;

		for (i = 0; i < pvec.nr; i++) {
			if (indices[i] > end_index) {
				done = true;
				break;
			}

			ret = dax_writeback_one(bdev, mapping, indices[i],
					pvec.pages[i]);
			if (ret < 0)
				return ret;
		}
	}
	return 0;
}
EXPORT_SYMBOL_GPL(dax_writeback_mapping_range);

J
Jan Kara 已提交
868
static int dax_insert_mapping(struct address_space *mapping,
869 870
		struct block_device *bdev, sector_t sector, size_t size,
		void **entryp, struct vm_area_struct *vma, struct vm_fault *vmf)
871
{
872
	unsigned long vaddr = vmf->address;
873
	struct blk_dax_ctl dax = {
874 875
		.sector = sector,
		.size = size,
876
	};
J
Jan Kara 已提交
877 878
	void *ret;
	void *entry = *entryp;
879

J
Jan Kara 已提交
880 881
	if (dax_map_atomic(bdev, &dax) < 0)
		return PTR_ERR(dax.addr);
882
	dax_unmap_atomic(bdev, &dax);
883

884
	ret = dax_insert_mapping_entry(mapping, vmf, entry, dax.sector, 0);
J
Jan Kara 已提交
885 886
	if (IS_ERR(ret))
		return PTR_ERR(ret);
J
Jan Kara 已提交
887
	*entryp = ret;
R
Ross Zwisler 已提交
888

J
Jan Kara 已提交
889
	return vm_insert_mixed(vma, vaddr, dax.pfn);
890 891
}

892 893 894 895 896 897 898
/**
 * dax_pfn_mkwrite - handle first write to DAX page
 * @vma: The virtual memory area where the fault occurred
 * @vmf: The description of the fault
 */
int dax_pfn_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
{
R
Ross Zwisler 已提交
899
	struct file *file = vma->vm_file;
J
Jan Kara 已提交
900
	struct address_space *mapping = file->f_mapping;
901
	void *entry, **slot;
J
Jan Kara 已提交
902
	pgoff_t index = vmf->pgoff;
903

J
Jan Kara 已提交
904
	spin_lock_irq(&mapping->tree_lock);
905 906 907 908 909 910 911
	entry = get_unlocked_mapping_entry(mapping, index, &slot);
	if (!entry || !radix_tree_exceptional_entry(entry)) {
		if (entry)
			put_unlocked_mapping_entry(mapping, index, entry);
		spin_unlock_irq(&mapping->tree_lock);
		return VM_FAULT_NOPAGE;
	}
J
Jan Kara 已提交
912
	radix_tree_tag_set(&mapping->page_tree, index, PAGECACHE_TAG_DIRTY);
913
	entry = lock_slot(mapping, slot);
J
Jan Kara 已提交
914
	spin_unlock_irq(&mapping->tree_lock);
915 916 917 918 919 920 921
	/*
	 * If we race with somebody updating the PTE and finish_mkwrite_fault()
	 * fails, we don't care. We need to return VM_FAULT_NOPAGE and retry
	 * the fault in either case.
	 */
	finish_mkwrite_fault(vmf);
	put_locked_mapping_entry(mapping, index, entry);
922 923 924 925
	return VM_FAULT_NOPAGE;
}
EXPORT_SYMBOL_GPL(dax_pfn_mkwrite);

926 927 928 929 930 931 932 933 934 935 936 937 938
static bool dax_range_is_aligned(struct block_device *bdev,
				 unsigned int offset, unsigned int length)
{
	unsigned short sector_size = bdev_logical_block_size(bdev);

	if (!IS_ALIGNED(offset, sector_size))
		return false;
	if (!IS_ALIGNED(length, sector_size))
		return false;

	return true;
}

939 940 941 942 943 944 945 946
int __dax_zero_page_range(struct block_device *bdev, sector_t sector,
		unsigned int offset, unsigned int length)
{
	struct blk_dax_ctl dax = {
		.sector		= sector,
		.size		= PAGE_SIZE,
	};

947 948 949 950 951 952 953 954 955 956 957
	if (dax_range_is_aligned(bdev, offset, length)) {
		sector_t start_sector = dax.sector + (offset >> 9);

		return blkdev_issue_zeroout(bdev, start_sector,
				length >> 9, GFP_NOFS, true);
	} else {
		if (dax_map_atomic(bdev, &dax) < 0)
			return PTR_ERR(dax.addr);
		clear_pmem(dax.addr + offset, length);
		dax_unmap_atomic(bdev, &dax);
	}
958 959 960 961
	return 0;
}
EXPORT_SYMBOL_GPL(__dax_zero_page_range);

962
#ifdef CONFIG_FS_IOMAP
963
static sector_t dax_iomap_sector(struct iomap *iomap, loff_t pos)
M
Matthew Wilcox 已提交
964
{
965
	return iomap->blkno + (((pos & PAGE_MASK) - iomap->offset) >> 9);
M
Matthew Wilcox 已提交
966
}
967 968

static loff_t
969
dax_iomap_actor(struct inode *inode, loff_t pos, loff_t length, void *data,
970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992
		struct iomap *iomap)
{
	struct iov_iter *iter = data;
	loff_t end = pos + length, done = 0;
	ssize_t ret = 0;

	if (iov_iter_rw(iter) == READ) {
		end = min(end, i_size_read(inode));
		if (pos >= end)
			return 0;

		if (iomap->type == IOMAP_HOLE || iomap->type == IOMAP_UNWRITTEN)
			return iov_iter_zero(min(length, end - pos), iter);
	}

	if (WARN_ON_ONCE(iomap->type != IOMAP_MAPPED))
		return -EIO;

	while (pos < end) {
		unsigned offset = pos & (PAGE_SIZE - 1);
		struct blk_dax_ctl dax = { 0 };
		ssize_t map_len;

993
		dax.sector = dax_iomap_sector(iomap, pos);
994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024
		dax.size = (length + offset + PAGE_SIZE - 1) & PAGE_MASK;
		map_len = dax_map_atomic(iomap->bdev, &dax);
		if (map_len < 0) {
			ret = map_len;
			break;
		}

		dax.addr += offset;
		map_len -= offset;
		if (map_len > end - pos)
			map_len = end - pos;

		if (iov_iter_rw(iter) == WRITE)
			map_len = copy_from_iter_pmem(dax.addr, map_len, iter);
		else
			map_len = copy_to_iter(dax.addr, map_len, iter);
		dax_unmap_atomic(iomap->bdev, &dax);
		if (map_len <= 0) {
			ret = map_len ? map_len : -EFAULT;
			break;
		}

		pos += map_len;
		length -= map_len;
		done += map_len;
	}

	return done ? done : ret;
}

/**
1025
 * dax_iomap_rw - Perform I/O to a DAX file
1026 1027 1028 1029 1030 1031 1032 1033 1034
 * @iocb:	The control block for this I/O
 * @iter:	The addresses to do I/O from or to
 * @ops:	iomap ops passed from the file system
 *
 * This function performs read and write operations to directly mapped
 * persistent memory.  The callers needs to take care of read/write exclusion
 * and evicting any page cache pages in the region under I/O.
 */
ssize_t
1035
dax_iomap_rw(struct kiocb *iocb, struct iov_iter *iter,
1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064
		struct iomap_ops *ops)
{
	struct address_space *mapping = iocb->ki_filp->f_mapping;
	struct inode *inode = mapping->host;
	loff_t pos = iocb->ki_pos, ret = 0, done = 0;
	unsigned flags = 0;

	if (iov_iter_rw(iter) == WRITE)
		flags |= IOMAP_WRITE;

	/*
	 * Yes, even DAX files can have page cache attached to them:  A zeroed
	 * page is inserted into the pagecache when we have to serve a write
	 * fault on a hole.  It should never be dirtied and can simply be
	 * dropped from the pagecache once we get real data for the page.
	 *
	 * XXX: This is racy against mmap, and there's nothing we can do about
	 * it. We'll eventually need to shift this down even further so that
	 * we can check if we allocated blocks over a hole first.
	 */
	if (mapping->nrpages) {
		ret = invalidate_inode_pages2_range(mapping,
				pos >> PAGE_SHIFT,
				(pos + iov_iter_count(iter) - 1) >> PAGE_SHIFT);
		WARN_ON_ONCE(ret);
	}

	while (iov_iter_count(iter)) {
		ret = iomap_apply(inode, pos, iov_iter_count(iter), flags, ops,
1065
				iter, dax_iomap_actor);
1066 1067 1068 1069 1070 1071 1072 1073 1074
		if (ret <= 0)
			break;
		pos += ret;
		done += ret;
	}

	iocb->ki_pos += done;
	return done ? done : ret;
}
1075
EXPORT_SYMBOL_GPL(dax_iomap_rw);
1076 1077

/**
1078
 * dax_iomap_fault - handle a page fault on a DAX file
1079 1080 1081 1082 1083 1084 1085 1086
 * @vma: The virtual memory area where the fault occurred
 * @vmf: The description of the fault
 * @ops: iomap ops passed from the file system
 *
 * When a page fault occurs, filesystems may call this helper in their fault
 * or mkwrite handler for DAX files. Assumes the caller has done all the
 * necessary locking for the page fault to proceed successfully.
 */
1087
int dax_iomap_fault(struct vm_area_struct *vma, struct vm_fault *vmf,
1088 1089 1090 1091
			struct iomap_ops *ops)
{
	struct address_space *mapping = vma->vm_file->f_mapping;
	struct inode *inode = mapping->host;
1092
	unsigned long vaddr = vmf->address;
1093 1094 1095
	loff_t pos = (loff_t)vmf->pgoff << PAGE_SHIFT;
	sector_t sector;
	struct iomap iomap = { 0 };
J
Jan Kara 已提交
1096
	unsigned flags = IOMAP_FAULT;
1097
	int error, major = 0;
1098
	int vmf_ret = 0;
1099 1100 1101 1102 1103 1104 1105 1106 1107 1108
	void *entry;

	/*
	 * Check whether offset isn't beyond end of file now. Caller is supposed
	 * to hold locks serializing us with truncate / punch hole so this is
	 * a reliable test.
	 */
	if (pos >= i_size_read(inode))
		return VM_FAULT_SIGBUS;

1109
	entry = grab_mapping_entry(mapping, vmf->pgoff, 0);
1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127
	if (IS_ERR(entry)) {
		error = PTR_ERR(entry);
		goto out;
	}

	if ((vmf->flags & FAULT_FLAG_WRITE) && !vmf->cow_page)
		flags |= IOMAP_WRITE;

	/*
	 * Note that we don't bother to use iomap_apply here: DAX required
	 * the file system block size to be equal the page size, which means
	 * that we never have to deal with more than a single extent here.
	 */
	error = ops->iomap_begin(inode, pos, PAGE_SIZE, flags, &iomap);
	if (error)
		goto unlock_entry;
	if (WARN_ON_ONCE(iomap.offset + iomap.length < pos + PAGE_SIZE)) {
		error = -EIO;		/* fs corruption? */
1128
		goto finish_iomap;
1129 1130
	}

1131
	sector = dax_iomap_sector(&iomap, pos);
1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149

	if (vmf->cow_page) {
		switch (iomap.type) {
		case IOMAP_HOLE:
		case IOMAP_UNWRITTEN:
			clear_user_highpage(vmf->cow_page, vaddr);
			break;
		case IOMAP_MAPPED:
			error = copy_user_dax(iomap.bdev, sector, PAGE_SIZE,
					vmf->cow_page, vaddr);
			break;
		default:
			WARN_ON_ONCE(1);
			error = -EIO;
			break;
		}

		if (error)
1150
			goto finish_iomap;
1151 1152 1153 1154 1155

		__SetPageUptodate(vmf->cow_page);
		vmf_ret = finish_fault(vmf);
		if (!vmf_ret)
			vmf_ret = VM_FAULT_DONE_COW;
1156
		goto finish_iomap;
1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170
	}

	switch (iomap.type) {
	case IOMAP_MAPPED:
		if (iomap.flags & IOMAP_F_NEW) {
			count_vm_event(PGMAJFAULT);
			mem_cgroup_count_vm_event(vma->vm_mm, PGMAJFAULT);
			major = VM_FAULT_MAJOR;
		}
		error = dax_insert_mapping(mapping, iomap.bdev, sector,
				PAGE_SIZE, &entry, vma, vmf);
		break;
	case IOMAP_UNWRITTEN:
	case IOMAP_HOLE:
1171
		if (!(vmf->flags & FAULT_FLAG_WRITE)) {
1172
			vmf_ret = dax_load_hole(mapping, entry, vmf);
1173 1174
			break;
		}
1175 1176 1177 1178 1179 1180 1181
		/*FALLTHRU*/
	default:
		WARN_ON_ONCE(1);
		error = -EIO;
		break;
	}

1182 1183
 finish_iomap:
	if (ops->iomap_end) {
1184
		if (error || (vmf_ret & VM_FAULT_ERROR)) {
1185 1186 1187 1188 1189 1190 1191 1192
			/* keep previous error */
			ops->iomap_end(inode, pos, PAGE_SIZE, 0, flags,
					&iomap);
		} else {
			error = ops->iomap_end(inode, pos, PAGE_SIZE,
					PAGE_SIZE, flags, &iomap);
		}
	}
1193
 unlock_entry:
1194
	if (vmf_ret != VM_FAULT_LOCKED || error)
1195
		put_locked_mapping_entry(mapping, vmf->pgoff, entry);
1196 1197 1198 1199 1200 1201
 out:
	if (error == -ENOMEM)
		return VM_FAULT_OOM | major;
	/* -EBUSY is fine, somebody else faulted on the same PTE */
	if (error < 0 && error != -EBUSY)
		return VM_FAULT_SIGBUS | major;
1202
	if (vmf_ret) {
1203
		WARN_ON_ONCE(error); /* -EBUSY from ops->iomap_end? */
1204
		return vmf_ret;
1205
	}
1206 1207
	return VM_FAULT_NOPAGE | major;
}
1208
EXPORT_SYMBOL_GPL(dax_iomap_fault);
1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294

#ifdef CONFIG_FS_DAX_PMD
/*
 * The 'colour' (ie low bits) within a PMD of a page offset.  This comes up
 * more often than one might expect in the below functions.
 */
#define PG_PMD_COLOUR	((PMD_SIZE >> PAGE_SHIFT) - 1)

static int dax_pmd_insert_mapping(struct vm_area_struct *vma, pmd_t *pmd,
		struct vm_fault *vmf, unsigned long address,
		struct iomap *iomap, loff_t pos, bool write, void **entryp)
{
	struct address_space *mapping = vma->vm_file->f_mapping;
	struct block_device *bdev = iomap->bdev;
	struct blk_dax_ctl dax = {
		.sector = dax_iomap_sector(iomap, pos),
		.size = PMD_SIZE,
	};
	long length = dax_map_atomic(bdev, &dax);
	void *ret;

	if (length < 0) /* dax_map_atomic() failed */
		return VM_FAULT_FALLBACK;
	if (length < PMD_SIZE)
		goto unmap_fallback;
	if (pfn_t_to_pfn(dax.pfn) & PG_PMD_COLOUR)
		goto unmap_fallback;
	if (!pfn_t_devmap(dax.pfn))
		goto unmap_fallback;

	dax_unmap_atomic(bdev, &dax);

	ret = dax_insert_mapping_entry(mapping, vmf, *entryp, dax.sector,
			RADIX_DAX_PMD);
	if (IS_ERR(ret))
		return VM_FAULT_FALLBACK;
	*entryp = ret;

	return vmf_insert_pfn_pmd(vma, address, pmd, dax.pfn, write);

 unmap_fallback:
	dax_unmap_atomic(bdev, &dax);
	return VM_FAULT_FALLBACK;
}

static int dax_pmd_load_hole(struct vm_area_struct *vma, pmd_t *pmd,
		struct vm_fault *vmf, unsigned long address,
		struct iomap *iomap, void **entryp)
{
	struct address_space *mapping = vma->vm_file->f_mapping;
	unsigned long pmd_addr = address & PMD_MASK;
	struct page *zero_page;
	spinlock_t *ptl;
	pmd_t pmd_entry;
	void *ret;

	zero_page = mm_get_huge_zero_page(vma->vm_mm);

	if (unlikely(!zero_page))
		return VM_FAULT_FALLBACK;

	ret = dax_insert_mapping_entry(mapping, vmf, *entryp, 0,
			RADIX_DAX_PMD | RADIX_DAX_HZP);
	if (IS_ERR(ret))
		return VM_FAULT_FALLBACK;
	*entryp = ret;

	ptl = pmd_lock(vma->vm_mm, pmd);
	if (!pmd_none(*pmd)) {
		spin_unlock(ptl);
		return VM_FAULT_FALLBACK;
	}

	pmd_entry = mk_pmd(zero_page, vma->vm_page_prot);
	pmd_entry = pmd_mkhuge(pmd_entry);
	set_pmd_at(vma->vm_mm, pmd_addr, pmd, pmd_entry);
	spin_unlock(ptl);
	return VM_FAULT_NOPAGE;
}

int dax_iomap_pmd_fault(struct vm_area_struct *vma, unsigned long address,
		pmd_t *pmd, unsigned int flags, struct iomap_ops *ops)
{
	struct address_space *mapping = vma->vm_file->f_mapping;
	unsigned long pmd_addr = address & PMD_MASK;
	bool write = flags & FAULT_FLAG_WRITE;
J
Jan Kara 已提交
1295
	unsigned int iomap_flags = (write ? IOMAP_WRITE : 0) | IOMAP_FAULT;
1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395
	struct inode *inode = mapping->host;
	int result = VM_FAULT_FALLBACK;
	struct iomap iomap = { 0 };
	pgoff_t max_pgoff, pgoff;
	struct vm_fault vmf;
	void *entry;
	loff_t pos;
	int error;

	/* Fall back to PTEs if we're going to COW */
	if (write && !(vma->vm_flags & VM_SHARED))
		goto fallback;

	/* If the PMD would extend outside the VMA */
	if (pmd_addr < vma->vm_start)
		goto fallback;
	if ((pmd_addr + PMD_SIZE) > vma->vm_end)
		goto fallback;

	/*
	 * Check whether offset isn't beyond end of file now. Caller is
	 * supposed to hold locks serializing us with truncate / punch hole so
	 * this is a reliable test.
	 */
	pgoff = linear_page_index(vma, pmd_addr);
	max_pgoff = (i_size_read(inode) - 1) >> PAGE_SHIFT;

	if (pgoff > max_pgoff)
		return VM_FAULT_SIGBUS;

	/* If the PMD would extend beyond the file size */
	if ((pgoff | PG_PMD_COLOUR) > max_pgoff)
		goto fallback;

	/*
	 * grab_mapping_entry() will make sure we get a 2M empty entry, a DAX
	 * PMD or a HZP entry.  If it can't (because a 4k page is already in
	 * the tree, for instance), it will return -EEXIST and we just fall
	 * back to 4k entries.
	 */
	entry = grab_mapping_entry(mapping, pgoff, RADIX_DAX_PMD);
	if (IS_ERR(entry))
		goto fallback;

	/*
	 * Note that we don't use iomap_apply here.  We aren't doing I/O, only
	 * setting up a mapping, so really we're using iomap_begin() as a way
	 * to look up our filesystem block.
	 */
	pos = (loff_t)pgoff << PAGE_SHIFT;
	error = ops->iomap_begin(inode, pos, PMD_SIZE, iomap_flags, &iomap);
	if (error)
		goto unlock_entry;
	if (iomap.offset + iomap.length < pos + PMD_SIZE)
		goto finish_iomap;

	vmf.pgoff = pgoff;
	vmf.flags = flags;
	vmf.gfp_mask = mapping_gfp_mask(mapping) | __GFP_IO;

	switch (iomap.type) {
	case IOMAP_MAPPED:
		result = dax_pmd_insert_mapping(vma, pmd, &vmf, address,
				&iomap, pos, write, &entry);
		break;
	case IOMAP_UNWRITTEN:
	case IOMAP_HOLE:
		if (WARN_ON_ONCE(write))
			goto finish_iomap;
		result = dax_pmd_load_hole(vma, pmd, &vmf, address, &iomap,
				&entry);
		break;
	default:
		WARN_ON_ONCE(1);
		break;
	}

 finish_iomap:
	if (ops->iomap_end) {
		if (result == VM_FAULT_FALLBACK) {
			ops->iomap_end(inode, pos, PMD_SIZE, 0, iomap_flags,
					&iomap);
		} else {
			error = ops->iomap_end(inode, pos, PMD_SIZE, PMD_SIZE,
					iomap_flags, &iomap);
			if (error)
				result = VM_FAULT_FALLBACK;
		}
	}
 unlock_entry:
	put_locked_mapping_entry(mapping, pgoff, entry);
 fallback:
	if (result == VM_FAULT_FALLBACK) {
		split_huge_pmd(vma, pmd, address);
		count_vm_event(THP_FAULT_FALLBACK);
	}
	return result;
}
EXPORT_SYMBOL_GPL(dax_iomap_pmd_fault);
#endif /* CONFIG_FS_DAX_PMD */
1396
#endif /* CONFIG_FS_IOMAP */