memory.c 115.7 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47
/*
 *  linux/mm/memory.c
 *
 *  Copyright (C) 1991, 1992, 1993, 1994  Linus Torvalds
 */

/*
 * demand-loading started 01.12.91 - seems it is high on the list of
 * things wanted, and it should be easy to implement. - Linus
 */

/*
 * Ok, demand-loading was easy, shared pages a little bit tricker. Shared
 * pages started 02.12.91, seems to work. - Linus.
 *
 * Tested sharing by executing about 30 /bin/sh: under the old kernel it
 * would have taken more than the 6M I have free, but it worked well as
 * far as I could see.
 *
 * Also corrected some "invalidate()"s - I wasn't doing enough of them.
 */

/*
 * Real VM (paging to/from disk) started 18.12.91. Much more work and
 * thought has to go into this. Oh, well..
 * 19.12.91  -  works, somewhat. Sometimes I get faults, don't know why.
 *		Found it. Everything seems to work now.
 * 20.12.91  -  Ok, making the swap-device changeable like the root.
 */

/*
 * 05.04.94  -  Multi-page memory management added for v1.1.
 * 		Idea by Alex Bligh (alex@cconcepts.co.uk)
 *
 * 16.07.99  -  Support of BIGMEM added by Gerhard Wichert, Siemens AG
 *		(Gerhard.Wichert@pdb.siemens.de)
 *
 * Aug/Sep 2004 Changed to four level page tables (Andi Kleen)
 */

#include <linux/kernel_stat.h>
#include <linux/mm.h>
#include <linux/hugetlb.h>
#include <linux/mman.h>
#include <linux/swap.h>
#include <linux/highmem.h>
#include <linux/pagemap.h>
H
Hugh Dickins 已提交
48
#include <linux/ksm.h>
L
Linus Torvalds 已提交
49
#include <linux/rmap.h>
50
#include <linux/export.h>
51
#include <linux/delayacct.h>
L
Linus Torvalds 已提交
52
#include <linux/init.h>
P
Peter Zijlstra 已提交
53
#include <linux/writeback.h>
54
#include <linux/memcontrol.h>
A
Andrea Arcangeli 已提交
55
#include <linux/mmu_notifier.h>
56 57 58
#include <linux/kallsyms.h>
#include <linux/swapops.h>
#include <linux/elf.h>
59
#include <linux/gfp.h>
60
#include <linux/migrate.h>
A
Andy Shevchenko 已提交
61
#include <linux/string.h>
L
Linus Torvalds 已提交
62

A
Alexey Dobriyan 已提交
63
#include <asm/io.h>
L
Linus Torvalds 已提交
64 65 66 67 68 69
#include <asm/pgalloc.h>
#include <asm/uaccess.h>
#include <asm/tlb.h>
#include <asm/tlbflush.h>
#include <asm/pgtable.h>

70 71
#include "internal.h"

72 73
#ifdef LAST_CPUPID_NOT_IN_PAGE_FLAGS
#warning Unfortunate NUMA and NUMA Balancing config, growing page-frame for last_cpupid.
74 75
#endif

A
Andy Whitcroft 已提交
76
#ifndef CONFIG_NEED_MULTIPLE_NODES
L
Linus Torvalds 已提交
77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
/* use the per-pgdat data instead for discontigmem - mbligh */
unsigned long max_mapnr;
struct page *mem_map;

EXPORT_SYMBOL(max_mapnr);
EXPORT_SYMBOL(mem_map);
#endif

/*
 * A number of key systems in x86 including ioremap() rely on the assumption
 * that high_memory defines the upper bound on direct map memory, then end
 * of ZONE_NORMAL.  Under CONFIG_DISCONTIG this means that max_low_pfn and
 * highstart_pfn must be the same; there must be no gap between ZONE_NORMAL
 * and ZONE_HIGHMEM.
 */
void * high_memory;

EXPORT_SYMBOL(high_memory);

96 97 98 99 100 101 102 103 104 105 106 107
/*
 * Randomize the address space (stacks, mmaps, brk, etc.).
 *
 * ( When CONFIG_COMPAT_BRK=y we exclude brk from randomization,
 *   as ancient (libc5 based) binaries can segfault. )
 */
int randomize_va_space __read_mostly =
#ifdef CONFIG_COMPAT_BRK
					1;
#else
					2;
#endif
108 109 110 111

static int __init disable_randmaps(char *s)
{
	randomize_va_space = 0;
112
	return 1;
113 114 115
}
__setup("norandmaps", disable_randmaps);

H
Hugh Dickins 已提交
116
unsigned long zero_pfn __read_mostly;
H
Hugh Dickins 已提交
117
unsigned long highest_memmap_pfn __read_mostly;
H
Hugh Dickins 已提交
118 119 120 121 122 123 124 125 126 127

/*
 * CONFIG_MMU architectures set up ZERO_PAGE in their paging_init()
 */
static int __init init_zero_pfn(void)
{
	zero_pfn = page_to_pfn(ZERO_PAGE(0));
	return 0;
}
core_initcall(init_zero_pfn);
128

K
KAMEZAWA Hiroyuki 已提交
129

130 131
#if defined(SPLIT_RSS_COUNTING)

132
void sync_mm_rss(struct mm_struct *mm)
133 134 135 136
{
	int i;

	for (i = 0; i < NR_MM_COUNTERS; i++) {
137 138 139
		if (current->rss_stat.count[i]) {
			add_mm_counter(mm, i, current->rss_stat.count[i]);
			current->rss_stat.count[i] = 0;
140 141
		}
	}
142
	current->rss_stat.events = 0;
143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163
}

static void add_mm_counter_fast(struct mm_struct *mm, int member, int val)
{
	struct task_struct *task = current;

	if (likely(task->mm == mm))
		task->rss_stat.count[member] += val;
	else
		add_mm_counter(mm, member, val);
}
#define inc_mm_counter_fast(mm, member) add_mm_counter_fast(mm, member, 1)
#define dec_mm_counter_fast(mm, member) add_mm_counter_fast(mm, member, -1)

/* sync counter once per 64 page faults */
#define TASK_RSS_EVENTS_THRESH	(64)
static void check_sync_rss_stat(struct task_struct *task)
{
	if (unlikely(task != current))
		return;
	if (unlikely(task->rss_stat.events++ > TASK_RSS_EVENTS_THRESH))
164
		sync_mm_rss(task->mm);
165
}
166
#else /* SPLIT_RSS_COUNTING */
167 168 169 170 171 172 173 174

#define inc_mm_counter_fast(mm, member) inc_mm_counter(mm, member)
#define dec_mm_counter_fast(mm, member) dec_mm_counter(mm, member)

static void check_sync_rss_stat(struct task_struct *task)
{
}

175 176 177 178 179 180 181 182 183 184 185 186 187 188
#endif /* SPLIT_RSS_COUNTING */

#ifdef HAVE_GENERIC_MMU_GATHER

static int tlb_next_batch(struct mmu_gather *tlb)
{
	struct mmu_gather_batch *batch;

	batch = tlb->active;
	if (batch->next) {
		tlb->active = batch->next;
		return 1;
	}

189 190 191
	if (tlb->batch_count == MAX_GATHER_BATCH_COUNT)
		return 0;

192 193 194 195
	batch = (void *)__get_free_pages(GFP_NOWAIT | __GFP_NOWARN, 0);
	if (!batch)
		return 0;

196
	tlb->batch_count++;
197 198 199 200 201 202 203 204 205 206 207 208 209 210 211
	batch->next = NULL;
	batch->nr   = 0;
	batch->max  = MAX_GATHER_BATCH;

	tlb->active->next = batch;
	tlb->active = batch;

	return 1;
}

/* tlb_gather_mmu
 *	Called to initialize an (on-stack) mmu_gather structure for page-table
 *	tear-down from @mm. The @fullmm argument is used when @mm is without
 *	users and we're going to destroy the full address space (exit/execve).
 */
212
void tlb_gather_mmu(struct mmu_gather *tlb, struct mm_struct *mm, unsigned long start, unsigned long end)
213 214 215
{
	tlb->mm = mm;

216 217
	/* Is it from 0 to ~0? */
	tlb->fullmm     = !(start | (end+1));
218
	tlb->need_flush_all = 0;
219 220
	tlb->start	= start;
	tlb->end	= end;
221 222 223 224 225
	tlb->need_flush = 0;
	tlb->local.next = NULL;
	tlb->local.nr   = 0;
	tlb->local.max  = ARRAY_SIZE(tlb->__pages);
	tlb->active     = &tlb->local;
226
	tlb->batch_count = 0;
227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242

#ifdef CONFIG_HAVE_RCU_TABLE_FREE
	tlb->batch = NULL;
#endif
}

void tlb_flush_mmu(struct mmu_gather *tlb)
{
	struct mmu_gather_batch *batch;

	if (!tlb->need_flush)
		return;
	tlb->need_flush = 0;
	tlb_flush(tlb);
#ifdef CONFIG_HAVE_RCU_TABLE_FREE
	tlb_table_flush(tlb);
243 244
#endif

245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281
	for (batch = &tlb->local; batch; batch = batch->next) {
		free_pages_and_swap_cache(batch->pages, batch->nr);
		batch->nr = 0;
	}
	tlb->active = &tlb->local;
}

/* tlb_finish_mmu
 *	Called at the end of the shootdown operation to free up any resources
 *	that were required.
 */
void tlb_finish_mmu(struct mmu_gather *tlb, unsigned long start, unsigned long end)
{
	struct mmu_gather_batch *batch, *next;

	tlb_flush_mmu(tlb);

	/* keep the page table cache within bounds */
	check_pgt_cache();

	for (batch = tlb->local.next; batch; batch = next) {
		next = batch->next;
		free_pages((unsigned long)batch, 0);
	}
	tlb->local.next = NULL;
}

/* __tlb_remove_page
 *	Must perform the equivalent to __free_pte(pte_get_and_clear(ptep)), while
 *	handling the additional races in SMP caused by other CPUs caching valid
 *	mappings in their TLBs. Returns the number of free page slots left.
 *	When out of page slots we must call tlb_flush_mmu().
 */
int __tlb_remove_page(struct mmu_gather *tlb, struct page *page)
{
	struct mmu_gather_batch *batch;

S
Shaohua Li 已提交
282
	VM_BUG_ON(!tlb->need_flush);
283 284 285 286 287 288

	batch = tlb->active;
	batch->pages[batch->nr++] = page;
	if (batch->nr == batch->max) {
		if (!tlb_next_batch(tlb))
			return 0;
289
		batch = tlb->active;
290 291 292 293 294 295 296 297
	}
	VM_BUG_ON(batch->nr > batch->max);

	return batch->max - batch->nr;
}

#endif /* HAVE_GENERIC_MMU_GATHER */

298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372
#ifdef CONFIG_HAVE_RCU_TABLE_FREE

/*
 * See the comment near struct mmu_table_batch.
 */

static void tlb_remove_table_smp_sync(void *arg)
{
	/* Simply deliver the interrupt */
}

static void tlb_remove_table_one(void *table)
{
	/*
	 * This isn't an RCU grace period and hence the page-tables cannot be
	 * assumed to be actually RCU-freed.
	 *
	 * It is however sufficient for software page-table walkers that rely on
	 * IRQ disabling. See the comment near struct mmu_table_batch.
	 */
	smp_call_function(tlb_remove_table_smp_sync, NULL, 1);
	__tlb_remove_table(table);
}

static void tlb_remove_table_rcu(struct rcu_head *head)
{
	struct mmu_table_batch *batch;
	int i;

	batch = container_of(head, struct mmu_table_batch, rcu);

	for (i = 0; i < batch->nr; i++)
		__tlb_remove_table(batch->tables[i]);

	free_page((unsigned long)batch);
}

void tlb_table_flush(struct mmu_gather *tlb)
{
	struct mmu_table_batch **batch = &tlb->batch;

	if (*batch) {
		call_rcu_sched(&(*batch)->rcu, tlb_remove_table_rcu);
		*batch = NULL;
	}
}

void tlb_remove_table(struct mmu_gather *tlb, void *table)
{
	struct mmu_table_batch **batch = &tlb->batch;

	tlb->need_flush = 1;

	/*
	 * When there's less then two users of this mm there cannot be a
	 * concurrent page-table walk.
	 */
	if (atomic_read(&tlb->mm->mm_users) < 2) {
		__tlb_remove_table(table);
		return;
	}

	if (*batch == NULL) {
		*batch = (struct mmu_table_batch *)__get_free_page(GFP_NOWAIT | __GFP_NOWARN);
		if (*batch == NULL) {
			tlb_remove_table_one(table);
			return;
		}
		(*batch)->nr = 0;
	}
	(*batch)->tables[(*batch)->nr++] = table;
	if ((*batch)->nr == MAX_TABLE_BATCH)
		tlb_table_flush(tlb);
}

373
#endif /* CONFIG_HAVE_RCU_TABLE_FREE */
374

L
Linus Torvalds 已提交
375 376 377 378
/*
 * Note: this doesn't free the actual pages themselves. That
 * has been handled earlier when unmapping all the memory regions.
 */
379 380
static void free_pte_range(struct mmu_gather *tlb, pmd_t *pmd,
			   unsigned long addr)
L
Linus Torvalds 已提交
381
{
382
	pgtable_t token = pmd_pgtable(*pmd);
383
	pmd_clear(pmd);
384
	pte_free_tlb(tlb, token, addr);
385
	atomic_long_dec(&tlb->mm->nr_ptes);
L
Linus Torvalds 已提交
386 387
}

388 389 390
static inline void free_pmd_range(struct mmu_gather *tlb, pud_t *pud,
				unsigned long addr, unsigned long end,
				unsigned long floor, unsigned long ceiling)
L
Linus Torvalds 已提交
391 392 393
{
	pmd_t *pmd;
	unsigned long next;
394
	unsigned long start;
L
Linus Torvalds 已提交
395

396
	start = addr;
L
Linus Torvalds 已提交
397 398 399 400 401
	pmd = pmd_offset(pud, addr);
	do {
		next = pmd_addr_end(addr, end);
		if (pmd_none_or_clear_bad(pmd))
			continue;
402
		free_pte_range(tlb, pmd, addr);
L
Linus Torvalds 已提交
403 404
	} while (pmd++, addr = next, addr != end);

405 406 407 408 409 410 411
	start &= PUD_MASK;
	if (start < floor)
		return;
	if (ceiling) {
		ceiling &= PUD_MASK;
		if (!ceiling)
			return;
L
Linus Torvalds 已提交
412
	}
413 414 415 416 417
	if (end - 1 > ceiling - 1)
		return;

	pmd = pmd_offset(pud, start);
	pud_clear(pud);
418
	pmd_free_tlb(tlb, pmd, start);
L
Linus Torvalds 已提交
419 420
}

421 422 423
static inline void free_pud_range(struct mmu_gather *tlb, pgd_t *pgd,
				unsigned long addr, unsigned long end,
				unsigned long floor, unsigned long ceiling)
L
Linus Torvalds 已提交
424 425 426
{
	pud_t *pud;
	unsigned long next;
427
	unsigned long start;
L
Linus Torvalds 已提交
428

429
	start = addr;
L
Linus Torvalds 已提交
430 431 432 433 434
	pud = pud_offset(pgd, addr);
	do {
		next = pud_addr_end(addr, end);
		if (pud_none_or_clear_bad(pud))
			continue;
435
		free_pmd_range(tlb, pud, addr, next, floor, ceiling);
L
Linus Torvalds 已提交
436 437
	} while (pud++, addr = next, addr != end);

438 439 440 441 442 443 444
	start &= PGDIR_MASK;
	if (start < floor)
		return;
	if (ceiling) {
		ceiling &= PGDIR_MASK;
		if (!ceiling)
			return;
L
Linus Torvalds 已提交
445
	}
446 447 448 449 450
	if (end - 1 > ceiling - 1)
		return;

	pud = pud_offset(pgd, start);
	pgd_clear(pgd);
451
	pud_free_tlb(tlb, pud, start);
L
Linus Torvalds 已提交
452 453 454
}

/*
455
 * This function frees user-level page tables of a process.
L
Linus Torvalds 已提交
456
 */
457
void free_pgd_range(struct mmu_gather *tlb,
458 459
			unsigned long addr, unsigned long end,
			unsigned long floor, unsigned long ceiling)
L
Linus Torvalds 已提交
460 461 462
{
	pgd_t *pgd;
	unsigned long next;
463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488

	/*
	 * The next few lines have given us lots of grief...
	 *
	 * Why are we testing PMD* at this top level?  Because often
	 * there will be no work to do at all, and we'd prefer not to
	 * go all the way down to the bottom just to discover that.
	 *
	 * Why all these "- 1"s?  Because 0 represents both the bottom
	 * of the address space and the top of it (using -1 for the
	 * top wouldn't help much: the masks would do the wrong thing).
	 * The rule is that addr 0 and floor 0 refer to the bottom of
	 * the address space, but end 0 and ceiling 0 refer to the top
	 * Comparisons need to use "end - 1" and "ceiling - 1" (though
	 * that end 0 case should be mythical).
	 *
	 * Wherever addr is brought up or ceiling brought down, we must
	 * be careful to reject "the opposite 0" before it confuses the
	 * subsequent tests.  But what about where end is brought down
	 * by PMD_SIZE below? no, end can't go down to 0 there.
	 *
	 * Whereas we round start (addr) and ceiling down, by different
	 * masks at different levels, in order to test whether a table
	 * now has no other vmas using it, so can be freed, we don't
	 * bother to round floor or end up - the tests don't need that.
	 */
L
Linus Torvalds 已提交
489

490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505
	addr &= PMD_MASK;
	if (addr < floor) {
		addr += PMD_SIZE;
		if (!addr)
			return;
	}
	if (ceiling) {
		ceiling &= PMD_MASK;
		if (!ceiling)
			return;
	}
	if (end - 1 > ceiling - 1)
		end -= PMD_SIZE;
	if (addr > end - 1)
		return;

506
	pgd = pgd_offset(tlb->mm, addr);
L
Linus Torvalds 已提交
507 508 509 510
	do {
		next = pgd_addr_end(addr, end);
		if (pgd_none_or_clear_bad(pgd))
			continue;
511
		free_pud_range(tlb, pgd, addr, next, floor, ceiling);
L
Linus Torvalds 已提交
512
	} while (pgd++, addr = next, addr != end);
513 514
}

515
void free_pgtables(struct mmu_gather *tlb, struct vm_area_struct *vma,
516
		unsigned long floor, unsigned long ceiling)
517 518 519 520 521
{
	while (vma) {
		struct vm_area_struct *next = vma->vm_next;
		unsigned long addr = vma->vm_start;

522
		/*
N
npiggin@suse.de 已提交
523 524
		 * Hide vma from rmap and truncate_pagecache before freeing
		 * pgtables
525
		 */
526
		unlink_anon_vmas(vma);
527 528
		unlink_file_vma(vma);

529
		if (is_vm_hugetlb_page(vma)) {
530
			hugetlb_free_pgd_range(tlb, addr, vma->vm_end,
531
				floor, next? next->vm_start: ceiling);
532 533 534 535 536
		} else {
			/*
			 * Optimization: gather nearby vmas into one call down
			 */
			while (next && next->vm_start <= vma->vm_end + PMD_SIZE
537
			       && !is_vm_hugetlb_page(next)) {
538 539
				vma = next;
				next = vma->vm_next;
540
				unlink_anon_vmas(vma);
541
				unlink_file_vma(vma);
542 543 544 545
			}
			free_pgd_range(tlb, addr, vma->vm_end,
				floor, next? next->vm_start: ceiling);
		}
546 547
		vma = next;
	}
L
Linus Torvalds 已提交
548 549
}

550 551
int __pte_alloc(struct mm_struct *mm, struct vm_area_struct *vma,
		pmd_t *pmd, unsigned long address)
L
Linus Torvalds 已提交
552
{
553
	spinlock_t *ptl;
554
	pgtable_t new = pte_alloc_one(mm, address);
555
	int wait_split_huge_page;
556 557 558
	if (!new)
		return -ENOMEM;

559 560 561 562 563 564 565 566 567 568 569 570 571 572 573
	/*
	 * Ensure all pte setup (eg. pte page lock and page clearing) are
	 * visible before the pte is made visible to other CPUs by being
	 * put into page tables.
	 *
	 * The other side of the story is the pointer chasing in the page
	 * table walking code (when walking the page table without locking;
	 * ie. most of the time). Fortunately, these data accesses consist
	 * of a chain of data-dependent loads, meaning most CPUs (alpha
	 * being the notable exception) will already guarantee loads are
	 * seen in-order. See the alpha page table accessors for the
	 * smp_read_barrier_depends() barriers in page table walking code.
	 */
	smp_wmb(); /* Could be smp_wmb__xxx(before|after)_spin_lock */

574
	ptl = pmd_lock(mm, pmd);
575 576
	wait_split_huge_page = 0;
	if (likely(pmd_none(*pmd))) {	/* Has another populated it ? */
577
		atomic_long_inc(&mm->nr_ptes);
L
Linus Torvalds 已提交
578
		pmd_populate(mm, pmd, new);
579
		new = NULL;
580 581
	} else if (unlikely(pmd_trans_splitting(*pmd)))
		wait_split_huge_page = 1;
582
	spin_unlock(ptl);
583 584
	if (new)
		pte_free(mm, new);
585 586
	if (wait_split_huge_page)
		wait_split_huge_page(vma->anon_vma, pmd);
587
	return 0;
L
Linus Torvalds 已提交
588 589
}

590
int __pte_alloc_kernel(pmd_t *pmd, unsigned long address)
L
Linus Torvalds 已提交
591
{
592 593 594 595
	pte_t *new = pte_alloc_one_kernel(&init_mm, address);
	if (!new)
		return -ENOMEM;

596 597
	smp_wmb(); /* See comment in __pte_alloc */

598
	spin_lock(&init_mm.page_table_lock);
599
	if (likely(pmd_none(*pmd))) {	/* Has another populated it ? */
600
		pmd_populate_kernel(&init_mm, pmd, new);
601
		new = NULL;
602 603
	} else
		VM_BUG_ON(pmd_trans_splitting(*pmd));
604
	spin_unlock(&init_mm.page_table_lock);
605 606
	if (new)
		pte_free_kernel(&init_mm, new);
607
	return 0;
L
Linus Torvalds 已提交
608 609
}

K
KAMEZAWA Hiroyuki 已提交
610 611 612 613 614 615
static inline void init_rss_vec(int *rss)
{
	memset(rss, 0, sizeof(int) * NR_MM_COUNTERS);
}

static inline void add_mm_rss_vec(struct mm_struct *mm, int *rss)
616
{
K
KAMEZAWA Hiroyuki 已提交
617 618
	int i;

619
	if (current->mm == mm)
620
		sync_mm_rss(mm);
K
KAMEZAWA Hiroyuki 已提交
621 622 623
	for (i = 0; i < NR_MM_COUNTERS; i++)
		if (rss[i])
			add_mm_counter(mm, i, rss[i]);
624 625
}

N
Nick Piggin 已提交
626
/*
627 628 629
 * This function is called to print an error when a bad pte
 * is found. For example, we might have a PFN-mapped pte in
 * a region that doesn't allow it.
N
Nick Piggin 已提交
630 631 632
 *
 * The calling function must still handle the error.
 */
633 634
static void print_bad_pte(struct vm_area_struct *vma, unsigned long addr,
			  pte_t pte, struct page *page)
N
Nick Piggin 已提交
635
{
636 637 638 639 640
	pgd_t *pgd = pgd_offset(vma->vm_mm, addr);
	pud_t *pud = pud_offset(pgd, addr);
	pmd_t *pmd = pmd_offset(pud, addr);
	struct address_space *mapping;
	pgoff_t index;
641 642 643 644 645 646 647 648 649 650 651 652 653 654
	static unsigned long resume;
	static unsigned long nr_shown;
	static unsigned long nr_unshown;

	/*
	 * Allow a burst of 60 reports, then keep quiet for that minute;
	 * or allow a steady drip of one report per second.
	 */
	if (nr_shown == 60) {
		if (time_before(jiffies, resume)) {
			nr_unshown++;
			return;
		}
		if (nr_unshown) {
655 656
			printk(KERN_ALERT
				"BUG: Bad page map: %lu messages suppressed\n",
657 658 659 660 661 662 663
				nr_unshown);
			nr_unshown = 0;
		}
		nr_shown = 0;
	}
	if (nr_shown++ == 0)
		resume = jiffies + 60 * HZ;
664 665 666 667

	mapping = vma->vm_file ? vma->vm_file->f_mapping : NULL;
	index = linear_page_index(vma, addr);

668 669
	printk(KERN_ALERT
		"BUG: Bad page map in process %s  pte:%08llx pmd:%08llx\n",
670 671
		current->comm,
		(long long)pte_val(pte), (long long)pmd_val(*pmd));
672 673
	if (page)
		dump_page(page);
674
	printk(KERN_ALERT
675 676 677 678 679 680
		"addr:%p vm_flags:%08lx anon_vma:%p mapping:%p index:%lx\n",
		(void *)addr, vma->vm_flags, vma->anon_vma, mapping, index);
	/*
	 * Choose text because data symbols depend on CONFIG_KALLSYMS_ALL=y
	 */
	if (vma->vm_ops)
J
Joe Perches 已提交
681 682
		printk(KERN_ALERT "vma->vm_ops->fault: %pSR\n",
		       vma->vm_ops->fault);
A
Al Viro 已提交
683
	if (vma->vm_file)
J
Joe Perches 已提交
684 685
		printk(KERN_ALERT "vma->vm_file->f_op->mmap: %pSR\n",
		       vma->vm_file->f_op->mmap);
N
Nick Piggin 已提交
686
	dump_stack();
687
	add_taint(TAINT_BAD_PAGE, LOCKDEP_NOW_UNRELIABLE);
N
Nick Piggin 已提交
688 689
}

690
static inline bool is_cow_mapping(vm_flags_t flags)
691 692 693 694
{
	return (flags & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE;
}

H
Hugh Dickins 已提交
695
/*
N
Nick Piggin 已提交
696
 * vm_normal_page -- This function gets the "struct page" associated with a pte.
697
 *
N
Nick Piggin 已提交
698 699 700
 * "Special" mappings do not wish to be associated with a "struct page" (either
 * it doesn't exist, or it exists but they don't want to touch it). In this
 * case, NULL is returned here. "Normal" mappings do have a struct page.
J
Jared Hulbert 已提交
701
 *
N
Nick Piggin 已提交
702 703 704 705 706 707 708 709
 * There are 2 broad cases. Firstly, an architecture may define a pte_special()
 * pte bit, in which case this function is trivial. Secondly, an architecture
 * may not have a spare pte bit, which requires a more complicated scheme,
 * described below.
 *
 * A raw VM_PFNMAP mapping (ie. one that is not COWed) is always considered a
 * special mapping (even if there are underlying and valid "struct pages").
 * COWed pages of a VM_PFNMAP are always normal.
710
 *
J
Jared Hulbert 已提交
711 712
 * The way we recognize COWed pages within VM_PFNMAP mappings is through the
 * rules set up by "remap_pfn_range()": the vma will have the VM_PFNMAP bit
N
Nick Piggin 已提交
713 714
 * set, and the vm_pgoff will point to the first PFN mapped: thus every special
 * mapping will always honor the rule
715 716 717
 *
 *	pfn_of_page == vma->vm_pgoff + ((addr - vma->vm_start) >> PAGE_SHIFT)
 *
N
Nick Piggin 已提交
718 719 720 721 722 723
 * And for normal mappings this is false.
 *
 * This restricts such mappings to be a linear translation from virtual address
 * to pfn. To get around this restriction, we allow arbitrary mappings so long
 * as the vma is not a COW mapping; in that case, we know that all ptes are
 * special (because none can have been COWed).
J
Jared Hulbert 已提交
724 725
 *
 *
N
Nick Piggin 已提交
726
 * In order to support COW of arbitrary special mappings, we have VM_MIXEDMAP.
J
Jared Hulbert 已提交
727 728 729 730 731 732 733 734 735
 *
 * VM_MIXEDMAP mappings can likewise contain memory with or without "struct
 * page" backing, however the difference is that _all_ pages with a struct
 * page (that is, those where pfn_valid is true) are refcounted and considered
 * normal pages by the VM. The disadvantage is that pages are refcounted
 * (which can be slower and simply not an option for some PFNMAP users). The
 * advantage is that we don't have to follow the strict linearity rule of
 * PFNMAP mappings in order to support COWable mappings.
 *
H
Hugh Dickins 已提交
736
 */
N
Nick Piggin 已提交
737 738 739 740 741 742 743
#ifdef __HAVE_ARCH_PTE_SPECIAL
# define HAVE_PTE_SPECIAL 1
#else
# define HAVE_PTE_SPECIAL 0
#endif
struct page *vm_normal_page(struct vm_area_struct *vma, unsigned long addr,
				pte_t pte)
H
Hugh Dickins 已提交
744
{
745
	unsigned long pfn = pte_pfn(pte);
N
Nick Piggin 已提交
746 747

	if (HAVE_PTE_SPECIAL) {
748 749
		if (likely(!pte_special(pte)))
			goto check_pfn;
H
Hugh Dickins 已提交
750 751
		if (vma->vm_flags & (VM_PFNMAP | VM_MIXEDMAP))
			return NULL;
H
Hugh Dickins 已提交
752
		if (!is_zero_pfn(pfn))
753
			print_bad_pte(vma, addr, pte, NULL);
N
Nick Piggin 已提交
754 755 756 757 758
		return NULL;
	}

	/* !HAVE_PTE_SPECIAL case follows: */

J
Jared Hulbert 已提交
759 760 761 762 763 764
	if (unlikely(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP))) {
		if (vma->vm_flags & VM_MIXEDMAP) {
			if (!pfn_valid(pfn))
				return NULL;
			goto out;
		} else {
N
Nick Piggin 已提交
765 766
			unsigned long off;
			off = (addr - vma->vm_start) >> PAGE_SHIFT;
J
Jared Hulbert 已提交
767 768 769 770 771
			if (pfn == vma->vm_pgoff + off)
				return NULL;
			if (!is_cow_mapping(vma->vm_flags))
				return NULL;
		}
772 773
	}

H
Hugh Dickins 已提交
774 775
	if (is_zero_pfn(pfn))
		return NULL;
776 777 778 779 780
check_pfn:
	if (unlikely(pfn > highest_memmap_pfn)) {
		print_bad_pte(vma, addr, pte, NULL);
		return NULL;
	}
781 782

	/*
N
Nick Piggin 已提交
783 784
	 * NOTE! We still have PageReserved() pages in the page tables.
	 * eg. VDSO mappings can cause them to exist.
785
	 */
J
Jared Hulbert 已提交
786
out:
787
	return pfn_to_page(pfn);
H
Hugh Dickins 已提交
788 789
}

L
Linus Torvalds 已提交
790 791 792 793 794 795
/*
 * copy one vm_area from one task to the other. Assumes the page tables
 * already present in the new task to be cleared in the whole range
 * covered by this vma.
 */

H
Hugh Dickins 已提交
796
static inline unsigned long
L
Linus Torvalds 已提交
797
copy_one_pte(struct mm_struct *dst_mm, struct mm_struct *src_mm,
N
Nick Piggin 已提交
798
		pte_t *dst_pte, pte_t *src_pte, struct vm_area_struct *vma,
H
Hugh Dickins 已提交
799
		unsigned long addr, int *rss)
L
Linus Torvalds 已提交
800
{
N
Nick Piggin 已提交
801
	unsigned long vm_flags = vma->vm_flags;
L
Linus Torvalds 已提交
802 803 804 805 806 807
	pte_t pte = *src_pte;
	struct page *page;

	/* pte contains position in swap or file, so copy. */
	if (unlikely(!pte_present(pte))) {
		if (!pte_file(pte)) {
808 809
			swp_entry_t entry = pte_to_swp_entry(pte);

H
Hugh Dickins 已提交
810 811 812
			if (swap_duplicate(entry) < 0)
				return entry.val;

L
Linus Torvalds 已提交
813 814 815
			/* make sure dst_mm is on swapoff's mmlist. */
			if (unlikely(list_empty(&dst_mm->mmlist))) {
				spin_lock(&mmlist_lock);
816 817 818
				if (list_empty(&dst_mm->mmlist))
					list_add(&dst_mm->mmlist,
						 &src_mm->mmlist);
L
Linus Torvalds 已提交
819 820
				spin_unlock(&mmlist_lock);
			}
K
KAMEZAWA Hiroyuki 已提交
821 822
			if (likely(!non_swap_entry(entry)))
				rss[MM_SWAPENTS]++;
823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838
			else if (is_migration_entry(entry)) {
				page = migration_entry_to_page(entry);

				if (PageAnon(page))
					rss[MM_ANONPAGES]++;
				else
					rss[MM_FILEPAGES]++;

				if (is_write_migration_entry(entry) &&
				    is_cow_mapping(vm_flags)) {
					/*
					 * COW mappings require pages in both
					 * parent and child to be set to read.
					 */
					make_migration_entry_read(&entry);
					pte = swp_entry_to_pte(entry);
839 840
					if (pte_swp_soft_dirty(*src_pte))
						pte = pte_swp_mksoft_dirty(pte);
841 842
					set_pte_at(src_mm, addr, src_pte, pte);
				}
843
			}
L
Linus Torvalds 已提交
844
		}
845
		goto out_set_pte;
L
Linus Torvalds 已提交
846 847 848 849 850 851
	}

	/*
	 * If it's a COW mapping, write protect it both
	 * in the parent and the child
	 */
852
	if (is_cow_mapping(vm_flags)) {
L
Linus Torvalds 已提交
853
		ptep_set_wrprotect(src_mm, addr, src_pte);
854
		pte = pte_wrprotect(pte);
L
Linus Torvalds 已提交
855 856 857 858 859 860 861 862 863
	}

	/*
	 * If it's a shared mapping, mark it clean in
	 * the child
	 */
	if (vm_flags & VM_SHARED)
		pte = pte_mkclean(pte);
	pte = pte_mkold(pte);
864 865 866 867

	page = vm_normal_page(vma, addr, pte);
	if (page) {
		get_page(page);
H
Hugh Dickins 已提交
868
		page_dup_rmap(page);
K
KAMEZAWA Hiroyuki 已提交
869 870 871 872
		if (PageAnon(page))
			rss[MM_ANONPAGES]++;
		else
			rss[MM_FILEPAGES]++;
873
	}
874 875 876

out_set_pte:
	set_pte_at(dst_mm, addr, dst_pte, pte);
H
Hugh Dickins 已提交
877
	return 0;
L
Linus Torvalds 已提交
878 879
}

880 881 882
int copy_pte_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
		   pmd_t *dst_pmd, pmd_t *src_pmd, struct vm_area_struct *vma,
		   unsigned long addr, unsigned long end)
L
Linus Torvalds 已提交
883
{
884
	pte_t *orig_src_pte, *orig_dst_pte;
L
Linus Torvalds 已提交
885
	pte_t *src_pte, *dst_pte;
H
Hugh Dickins 已提交
886
	spinlock_t *src_ptl, *dst_ptl;
887
	int progress = 0;
K
KAMEZAWA Hiroyuki 已提交
888
	int rss[NR_MM_COUNTERS];
H
Hugh Dickins 已提交
889
	swp_entry_t entry = (swp_entry_t){0};
L
Linus Torvalds 已提交
890 891

again:
K
KAMEZAWA Hiroyuki 已提交
892 893
	init_rss_vec(rss);

H
Hugh Dickins 已提交
894
	dst_pte = pte_alloc_map_lock(dst_mm, dst_pmd, addr, &dst_ptl);
L
Linus Torvalds 已提交
895 896
	if (!dst_pte)
		return -ENOMEM;
P
Peter Zijlstra 已提交
897
	src_pte = pte_offset_map(src_pmd, addr);
H
Hugh Dickins 已提交
898
	src_ptl = pte_lockptr(src_mm, src_pmd);
I
Ingo Molnar 已提交
899
	spin_lock_nested(src_ptl, SINGLE_DEPTH_NESTING);
900 901
	orig_src_pte = src_pte;
	orig_dst_pte = dst_pte;
902
	arch_enter_lazy_mmu_mode();
L
Linus Torvalds 已提交
903 904 905 906 907 908

	do {
		/*
		 * We are holding two locks at this point - either of them
		 * could generate latencies in another task on another CPU.
		 */
909 910 911
		if (progress >= 32) {
			progress = 0;
			if (need_resched() ||
N
Nick Piggin 已提交
912
			    spin_needbreak(src_ptl) || spin_needbreak(dst_ptl))
913 914
				break;
		}
L
Linus Torvalds 已提交
915 916 917 918
		if (pte_none(*src_pte)) {
			progress++;
			continue;
		}
H
Hugh Dickins 已提交
919 920 921 922
		entry.val = copy_one_pte(dst_mm, src_mm, dst_pte, src_pte,
							vma, addr, rss);
		if (entry.val)
			break;
L
Linus Torvalds 已提交
923 924 925
		progress += 8;
	} while (dst_pte++, src_pte++, addr += PAGE_SIZE, addr != end);

926
	arch_leave_lazy_mmu_mode();
H
Hugh Dickins 已提交
927
	spin_unlock(src_ptl);
P
Peter Zijlstra 已提交
928
	pte_unmap(orig_src_pte);
K
KAMEZAWA Hiroyuki 已提交
929
	add_mm_rss_vec(dst_mm, rss);
930
	pte_unmap_unlock(orig_dst_pte, dst_ptl);
H
Hugh Dickins 已提交
931
	cond_resched();
H
Hugh Dickins 已提交
932 933 934 935 936 937

	if (entry.val) {
		if (add_swap_count_continuation(entry, GFP_KERNEL) < 0)
			return -ENOMEM;
		progress = 0;
	}
L
Linus Torvalds 已提交
938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955
	if (addr != end)
		goto again;
	return 0;
}

static inline int copy_pmd_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
		pud_t *dst_pud, pud_t *src_pud, struct vm_area_struct *vma,
		unsigned long addr, unsigned long end)
{
	pmd_t *src_pmd, *dst_pmd;
	unsigned long next;

	dst_pmd = pmd_alloc(dst_mm, dst_pud, addr);
	if (!dst_pmd)
		return -ENOMEM;
	src_pmd = pmd_offset(src_pud, addr);
	do {
		next = pmd_addr_end(addr, end);
956 957
		if (pmd_trans_huge(*src_pmd)) {
			int err;
958
			VM_BUG_ON(next-addr != HPAGE_PMD_SIZE);
959 960 961 962 963 964 965 966
			err = copy_huge_pmd(dst_mm, src_mm,
					    dst_pmd, src_pmd, addr, vma);
			if (err == -ENOMEM)
				return -ENOMEM;
			if (!err)
				continue;
			/* fall through */
		}
L
Linus Torvalds 已提交
967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004
		if (pmd_none_or_clear_bad(src_pmd))
			continue;
		if (copy_pte_range(dst_mm, src_mm, dst_pmd, src_pmd,
						vma, addr, next))
			return -ENOMEM;
	} while (dst_pmd++, src_pmd++, addr = next, addr != end);
	return 0;
}

static inline int copy_pud_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
		pgd_t *dst_pgd, pgd_t *src_pgd, struct vm_area_struct *vma,
		unsigned long addr, unsigned long end)
{
	pud_t *src_pud, *dst_pud;
	unsigned long next;

	dst_pud = pud_alloc(dst_mm, dst_pgd, addr);
	if (!dst_pud)
		return -ENOMEM;
	src_pud = pud_offset(src_pgd, addr);
	do {
		next = pud_addr_end(addr, end);
		if (pud_none_or_clear_bad(src_pud))
			continue;
		if (copy_pmd_range(dst_mm, src_mm, dst_pud, src_pud,
						vma, addr, next))
			return -ENOMEM;
	} while (dst_pud++, src_pud++, addr = next, addr != end);
	return 0;
}

int copy_page_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
		struct vm_area_struct *vma)
{
	pgd_t *src_pgd, *dst_pgd;
	unsigned long next;
	unsigned long addr = vma->vm_start;
	unsigned long end = vma->vm_end;
1005 1006 1007
	unsigned long mmun_start;	/* For mmu_notifiers */
	unsigned long mmun_end;		/* For mmu_notifiers */
	bool is_cow;
A
Andrea Arcangeli 已提交
1008
	int ret;
L
Linus Torvalds 已提交
1009

1010 1011 1012 1013 1014 1015
	/*
	 * Don't copy ptes where a page fault will fill them correctly.
	 * Fork becomes much lighter when there are big shared or private
	 * readonly mappings. The tradeoff is that copy_page_range is more
	 * efficient than faulting.
	 */
1016 1017
	if (!(vma->vm_flags & (VM_HUGETLB | VM_NONLINEAR |
			       VM_PFNMAP | VM_MIXEDMAP))) {
1018 1019 1020 1021
		if (!vma->anon_vma)
			return 0;
	}

L
Linus Torvalds 已提交
1022 1023 1024
	if (is_vm_hugetlb_page(vma))
		return copy_hugetlb_page_range(dst_mm, src_mm, vma);

1025
	if (unlikely(vma->vm_flags & VM_PFNMAP)) {
1026 1027 1028 1029
		/*
		 * We do not free on error cases below as remove_vma
		 * gets called on error from higher level routine
		 */
1030
		ret = track_pfn_copy(vma);
1031 1032 1033 1034
		if (ret)
			return ret;
	}

A
Andrea Arcangeli 已提交
1035 1036 1037 1038 1039 1040
	/*
	 * We need to invalidate the secondary MMU mappings only when
	 * there could be a permission downgrade on the ptes of the
	 * parent mm. And a permission downgrade will only happen if
	 * is_cow_mapping() returns true.
	 */
1041 1042 1043 1044 1045 1046
	is_cow = is_cow_mapping(vma->vm_flags);
	mmun_start = addr;
	mmun_end   = end;
	if (is_cow)
		mmu_notifier_invalidate_range_start(src_mm, mmun_start,
						    mmun_end);
A
Andrea Arcangeli 已提交
1047 1048

	ret = 0;
L
Linus Torvalds 已提交
1049 1050 1051 1052 1053 1054
	dst_pgd = pgd_offset(dst_mm, addr);
	src_pgd = pgd_offset(src_mm, addr);
	do {
		next = pgd_addr_end(addr, end);
		if (pgd_none_or_clear_bad(src_pgd))
			continue;
A
Andrea Arcangeli 已提交
1055 1056 1057 1058 1059
		if (unlikely(copy_pud_range(dst_mm, src_mm, dst_pgd, src_pgd,
					    vma, addr, next))) {
			ret = -ENOMEM;
			break;
		}
L
Linus Torvalds 已提交
1060
	} while (dst_pgd++, src_pgd++, addr = next, addr != end);
A
Andrea Arcangeli 已提交
1061

1062 1063
	if (is_cow)
		mmu_notifier_invalidate_range_end(src_mm, mmun_start, mmun_end);
A
Andrea Arcangeli 已提交
1064
	return ret;
L
Linus Torvalds 已提交
1065 1066
}

1067
static unsigned long zap_pte_range(struct mmu_gather *tlb,
N
Nick Piggin 已提交
1068
				struct vm_area_struct *vma, pmd_t *pmd,
L
Linus Torvalds 已提交
1069
				unsigned long addr, unsigned long end,
1070
				struct zap_details *details)
L
Linus Torvalds 已提交
1071
{
N
Nick Piggin 已提交
1072
	struct mm_struct *mm = tlb->mm;
P
Peter Zijlstra 已提交
1073
	int force_flush = 0;
K
KAMEZAWA Hiroyuki 已提交
1074
	int rss[NR_MM_COUNTERS];
1075
	spinlock_t *ptl;
1076
	pte_t *start_pte;
1077
	pte_t *pte;
K
KAMEZAWA Hiroyuki 已提交
1078

P
Peter Zijlstra 已提交
1079
again:
1080
	init_rss_vec(rss);
1081 1082
	start_pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
	pte = start_pte;
1083
	arch_enter_lazy_mmu_mode();
L
Linus Torvalds 已提交
1084 1085
	do {
		pte_t ptent = *pte;
1086
		if (pte_none(ptent)) {
L
Linus Torvalds 已提交
1087
			continue;
1088
		}
1089

L
Linus Torvalds 已提交
1090
		if (pte_present(ptent)) {
H
Hugh Dickins 已提交
1091
			struct page *page;
1092

1093
			page = vm_normal_page(vma, addr, ptent);
L
Linus Torvalds 已提交
1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111
			if (unlikely(details) && page) {
				/*
				 * unmap_shared_mapping_pages() wants to
				 * invalidate cache without truncating:
				 * unmap shared but keep private pages.
				 */
				if (details->check_mapping &&
				    details->check_mapping != page->mapping)
					continue;
				/*
				 * Each page->index must be checked when
				 * invalidating or truncating nonlinear.
				 */
				if (details->nonlinear_vma &&
				    (page->index < details->first_index ||
				     page->index > details->last_index))
					continue;
			}
N
Nick Piggin 已提交
1112
			ptent = ptep_get_and_clear_full(mm, addr, pte,
1113
							tlb->fullmm);
L
Linus Torvalds 已提交
1114 1115 1116 1117 1118
			tlb_remove_tlb_entry(tlb, pte, addr);
			if (unlikely(!page))
				continue;
			if (unlikely(details) && details->nonlinear_vma
			    && linear_page_index(details->nonlinear_vma,
1119 1120 1121 1122 1123 1124
						addr) != page->index) {
				pte_t ptfile = pgoff_to_pte(page->index);
				if (pte_soft_dirty(ptent))
					pte_file_mksoft_dirty(ptfile);
				set_pte_at(mm, addr, pte, ptfile);
			}
L
Linus Torvalds 已提交
1125
			if (PageAnon(page))
K
KAMEZAWA Hiroyuki 已提交
1126
				rss[MM_ANONPAGES]--;
1127 1128 1129
			else {
				if (pte_dirty(ptent))
					set_page_dirty(page);
1130
				if (pte_young(ptent) &&
1131
				    likely(!(vma->vm_flags & VM_SEQ_READ)))
1132
					mark_page_accessed(page);
K
KAMEZAWA Hiroyuki 已提交
1133
				rss[MM_FILEPAGES]--;
1134
			}
1135
			page_remove_rmap(page);
1136 1137
			if (unlikely(page_mapcount(page) < 0))
				print_bad_pte(vma, addr, ptent, page);
P
Peter Zijlstra 已提交
1138 1139 1140
			force_flush = !__tlb_remove_page(tlb, page);
			if (force_flush)
				break;
L
Linus Torvalds 已提交
1141 1142 1143 1144 1145 1146 1147 1148
			continue;
		}
		/*
		 * If details->check_mapping, we leave swap entries;
		 * if details->nonlinear_vma, we leave file entries.
		 */
		if (unlikely(details))
			continue;
1149 1150 1151
		if (pte_file(ptent)) {
			if (unlikely(!(vma->vm_flags & VM_NONLINEAR)))
				print_bad_pte(vma, addr, ptent, NULL);
K
KAMEZAWA Hiroyuki 已提交
1152 1153 1154 1155 1156
		} else {
			swp_entry_t entry = pte_to_swp_entry(ptent);

			if (!non_swap_entry(entry))
				rss[MM_SWAPENTS]--;
1157 1158 1159 1160 1161 1162 1163 1164 1165 1166
			else if (is_migration_entry(entry)) {
				struct page *page;

				page = migration_entry_to_page(entry);

				if (PageAnon(page))
					rss[MM_ANONPAGES]--;
				else
					rss[MM_FILEPAGES]--;
			}
K
KAMEZAWA Hiroyuki 已提交
1167 1168 1169
			if (unlikely(!free_swap_and_cache(entry)))
				print_bad_pte(vma, addr, ptent, NULL);
		}
1170
		pte_clear_not_present_full(mm, addr, pte, tlb->fullmm);
1171
	} while (pte++, addr += PAGE_SIZE, addr != end);
1172

K
KAMEZAWA Hiroyuki 已提交
1173
	add_mm_rss_vec(mm, rss);
1174
	arch_leave_lazy_mmu_mode();
1175
	pte_unmap_unlock(start_pte, ptl);
1176

P
Peter Zijlstra 已提交
1177 1178 1179 1180 1181 1182
	/*
	 * mmu_gather ran out of room to batch pages, we break out of
	 * the PTE lock to avoid doing the potential expensive TLB invalidate
	 * and page-free while holding it.
	 */
	if (force_flush) {
1183 1184
		unsigned long old_end;

P
Peter Zijlstra 已提交
1185
		force_flush = 0;
1186

1187 1188 1189 1190 1191 1192
		/*
		 * Flush the TLB just for the previous segment,
		 * then update the range to be the remaining
		 * TLB range.
		 */
		old_end = tlb->end;
1193
		tlb->end = addr;
1194

P
Peter Zijlstra 已提交
1195
		tlb_flush_mmu(tlb);
1196 1197 1198 1199 1200

		tlb->start = addr;
		tlb->end = old_end;

		if (addr != end)
P
Peter Zijlstra 已提交
1201 1202 1203
			goto again;
	}

1204
	return addr;
L
Linus Torvalds 已提交
1205 1206
}

1207
static inline unsigned long zap_pmd_range(struct mmu_gather *tlb,
N
Nick Piggin 已提交
1208
				struct vm_area_struct *vma, pud_t *pud,
L
Linus Torvalds 已提交
1209
				unsigned long addr, unsigned long end,
1210
				struct zap_details *details)
L
Linus Torvalds 已提交
1211 1212 1213 1214 1215 1216 1217
{
	pmd_t *pmd;
	unsigned long next;

	pmd = pmd_offset(pud, addr);
	do {
		next = pmd_addr_end(addr, end);
1218
		if (pmd_trans_huge(*pmd)) {
1219
			if (next - addr != HPAGE_PMD_SIZE) {
1220 1221 1222 1223 1224 1225 1226 1227 1228
#ifdef CONFIG_DEBUG_VM
				if (!rwsem_is_locked(&tlb->mm->mmap_sem)) {
					pr_err("%s: mmap_sem is unlocked! addr=0x%lx end=0x%lx vma->vm_start=0x%lx vma->vm_end=0x%lx\n",
						__func__, addr, end,
						vma->vm_start,
						vma->vm_end);
					BUG();
				}
#endif
1229
				split_huge_page_pmd(vma, addr, pmd);
S
Shaohua Li 已提交
1230
			} else if (zap_huge_pmd(tlb, vma, pmd, addr))
1231
				goto next;
1232 1233
			/* fall through */
		}
1234 1235 1236 1237 1238 1239 1240 1241 1242
		/*
		 * Here there can be other concurrent MADV_DONTNEED or
		 * trans huge page faults running, and if the pmd is
		 * none or trans huge it can change under us. This is
		 * because MADV_DONTNEED holds the mmap_sem in read
		 * mode.
		 */
		if (pmd_none_or_trans_huge_or_clear_bad(pmd))
			goto next;
1243
		next = zap_pte_range(tlb, vma, pmd, addr, next, details);
1244
next:
1245 1246
		cond_resched();
	} while (pmd++, addr = next, addr != end);
1247 1248

	return addr;
L
Linus Torvalds 已提交
1249 1250
}

1251
static inline unsigned long zap_pud_range(struct mmu_gather *tlb,
N
Nick Piggin 已提交
1252
				struct vm_area_struct *vma, pgd_t *pgd,
L
Linus Torvalds 已提交
1253
				unsigned long addr, unsigned long end,
1254
				struct zap_details *details)
L
Linus Torvalds 已提交
1255 1256 1257 1258 1259 1260 1261
{
	pud_t *pud;
	unsigned long next;

	pud = pud_offset(pgd, addr);
	do {
		next = pud_addr_end(addr, end);
1262
		if (pud_none_or_clear_bad(pud))
L
Linus Torvalds 已提交
1263
			continue;
1264 1265
		next = zap_pmd_range(tlb, vma, pud, addr, next, details);
	} while (pud++, addr = next, addr != end);
1266 1267

	return addr;
L
Linus Torvalds 已提交
1268 1269
}

A
Al Viro 已提交
1270 1271 1272 1273
static void unmap_page_range(struct mmu_gather *tlb,
			     struct vm_area_struct *vma,
			     unsigned long addr, unsigned long end,
			     struct zap_details *details)
L
Linus Torvalds 已提交
1274 1275 1276 1277 1278 1279 1280 1281
{
	pgd_t *pgd;
	unsigned long next;

	if (details && !details->check_mapping && !details->nonlinear_vma)
		details = NULL;

	BUG_ON(addr >= end);
1282
	mem_cgroup_uncharge_start();
L
Linus Torvalds 已提交
1283 1284 1285 1286
	tlb_start_vma(tlb, vma);
	pgd = pgd_offset(vma->vm_mm, addr);
	do {
		next = pgd_addr_end(addr, end);
1287
		if (pgd_none_or_clear_bad(pgd))
L
Linus Torvalds 已提交
1288
			continue;
1289 1290
		next = zap_pud_range(tlb, vma, pgd, addr, next, details);
	} while (pgd++, addr = next, addr != end);
L
Linus Torvalds 已提交
1291
	tlb_end_vma(tlb, vma);
1292
	mem_cgroup_uncharge_end();
L
Linus Torvalds 已提交
1293
}
1294

1295 1296 1297

static void unmap_single_vma(struct mmu_gather *tlb,
		struct vm_area_struct *vma, unsigned long start_addr,
1298
		unsigned long end_addr,
1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309
		struct zap_details *details)
{
	unsigned long start = max(vma->vm_start, start_addr);
	unsigned long end;

	if (start >= vma->vm_end)
		return;
	end = min(vma->vm_end, end_addr);
	if (end <= vma->vm_start)
		return;

1310 1311 1312
	if (vma->vm_file)
		uprobe_munmap(vma, start, end);

1313
	if (unlikely(vma->vm_flags & VM_PFNMAP))
1314
		untrack_pfn(vma, 0, 0);
1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328

	if (start != end) {
		if (unlikely(is_vm_hugetlb_page(vma))) {
			/*
			 * It is undesirable to test vma->vm_file as it
			 * should be non-null for valid hugetlb area.
			 * However, vm_file will be NULL in the error
			 * cleanup path of do_mmap_pgoff. When
			 * hugetlbfs ->mmap method fails,
			 * do_mmap_pgoff() nullifies vma->vm_file
			 * before calling this function to clean up.
			 * Since no pte has actually been setup, it is
			 * safe to do nothing in this case.
			 */
1329 1330
			if (vma->vm_file) {
				mutex_lock(&vma->vm_file->f_mapping->i_mmap_mutex);
1331
				__unmap_hugepage_range_final(tlb, vma, start, end, NULL);
1332 1333
				mutex_unlock(&vma->vm_file->f_mapping->i_mmap_mutex);
			}
1334 1335 1336
		} else
			unmap_page_range(tlb, vma, start, end, details);
	}
L
Linus Torvalds 已提交
1337 1338 1339 1340
}

/**
 * unmap_vmas - unmap a range of memory covered by a list of vma's
1341
 * @tlb: address of the caller's struct mmu_gather
L
Linus Torvalds 已提交
1342 1343 1344 1345
 * @vma: the starting vma
 * @start_addr: virtual address at which to start unmapping
 * @end_addr: virtual address at which to end unmapping
 *
1346
 * Unmap all pages in the vma list.
L
Linus Torvalds 已提交
1347 1348 1349 1350 1351 1352 1353 1354 1355 1356
 *
 * Only addresses between `start' and `end' will be unmapped.
 *
 * The VMA list must be sorted in ascending virtual address order.
 *
 * unmap_vmas() assumes that the caller will flush the whole unmapped address
 * range after unmap_vmas() returns.  So the only responsibility here is to
 * ensure that any thus-far unmapped pages are flushed before unmap_vmas()
 * drops the lock and schedules.
 */
A
Al Viro 已提交
1357
void unmap_vmas(struct mmu_gather *tlb,
L
Linus Torvalds 已提交
1358
		struct vm_area_struct *vma, unsigned long start_addr,
1359
		unsigned long end_addr)
L
Linus Torvalds 已提交
1360
{
A
Andrea Arcangeli 已提交
1361
	struct mm_struct *mm = vma->vm_mm;
L
Linus Torvalds 已提交
1362

A
Andrea Arcangeli 已提交
1363
	mmu_notifier_invalidate_range_start(mm, start_addr, end_addr);
1364
	for ( ; vma && vma->vm_start < end_addr; vma = vma->vm_next)
1365
		unmap_single_vma(tlb, vma, start_addr, end_addr, NULL);
A
Andrea Arcangeli 已提交
1366
	mmu_notifier_invalidate_range_end(mm, start_addr, end_addr);
L
Linus Torvalds 已提交
1367 1368 1369 1370 1371
}

/**
 * zap_page_range - remove user pages in a given range
 * @vma: vm_area_struct holding the applicable pages
1372
 * @start: starting address of pages to zap
L
Linus Torvalds 已提交
1373 1374
 * @size: number of bytes to zap
 * @details: details of nonlinear truncation or shared cache invalidation
1375 1376
 *
 * Caller must protect the VMA list
L
Linus Torvalds 已提交
1377
 */
1378
void zap_page_range(struct vm_area_struct *vma, unsigned long start,
L
Linus Torvalds 已提交
1379 1380 1381
		unsigned long size, struct zap_details *details)
{
	struct mm_struct *mm = vma->vm_mm;
P
Peter Zijlstra 已提交
1382
	struct mmu_gather tlb;
1383
	unsigned long end = start + size;
L
Linus Torvalds 已提交
1384 1385

	lru_add_drain();
1386
	tlb_gather_mmu(&tlb, mm, start, end);
1387
	update_hiwater_rss(mm);
1388 1389
	mmu_notifier_invalidate_range_start(mm, start, end);
	for ( ; vma && vma->vm_start < end; vma = vma->vm_next)
1390
		unmap_single_vma(&tlb, vma, start, end, details);
1391 1392
	mmu_notifier_invalidate_range_end(mm, start, end);
	tlb_finish_mmu(&tlb, start, end);
L
Linus Torvalds 已提交
1393 1394
}

1395 1396 1397 1398 1399 1400 1401 1402
/**
 * zap_page_range_single - remove user pages in a given range
 * @vma: vm_area_struct holding the applicable pages
 * @address: starting address of pages to zap
 * @size: number of bytes to zap
 * @details: details of nonlinear truncation or shared cache invalidation
 *
 * The range must fit into one VMA.
L
Linus Torvalds 已提交
1403
 */
1404
static void zap_page_range_single(struct vm_area_struct *vma, unsigned long address,
L
Linus Torvalds 已提交
1405 1406 1407
		unsigned long size, struct zap_details *details)
{
	struct mm_struct *mm = vma->vm_mm;
P
Peter Zijlstra 已提交
1408
	struct mmu_gather tlb;
L
Linus Torvalds 已提交
1409 1410 1411
	unsigned long end = address + size;

	lru_add_drain();
1412
	tlb_gather_mmu(&tlb, mm, address, end);
1413
	update_hiwater_rss(mm);
1414
	mmu_notifier_invalidate_range_start(mm, address, end);
1415
	unmap_single_vma(&tlb, vma, address, end, details);
1416
	mmu_notifier_invalidate_range_end(mm, address, end);
P
Peter Zijlstra 已提交
1417
	tlb_finish_mmu(&tlb, address, end);
L
Linus Torvalds 已提交
1418 1419
}

1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437
/**
 * zap_vma_ptes - remove ptes mapping the vma
 * @vma: vm_area_struct holding ptes to be zapped
 * @address: starting address of pages to zap
 * @size: number of bytes to zap
 *
 * This function only unmaps ptes assigned to VM_PFNMAP vmas.
 *
 * The entire address range must be fully contained within the vma.
 *
 * Returns 0 if successful.
 */
int zap_vma_ptes(struct vm_area_struct *vma, unsigned long address,
		unsigned long size)
{
	if (address < vma->vm_start || address + size > vma->vm_end ||
	    		!(vma->vm_flags & VM_PFNMAP))
		return -1;
1438
	zap_page_range_single(vma, address, size, NULL);
1439 1440 1441 1442
	return 0;
}
EXPORT_SYMBOL_GPL(zap_vma_ptes);

J
Johannes Weiner 已提交
1443
/**
1444
 * follow_page_mask - look up a page descriptor from a user-virtual address
J
Johannes Weiner 已提交
1445 1446 1447
 * @vma: vm_area_struct mapping @address
 * @address: virtual address to look up
 * @flags: flags modifying lookup behaviour
1448
 * @page_mask: on output, *page_mask is set according to the size of the page
J
Johannes Weiner 已提交
1449 1450 1451 1452 1453 1454
 *
 * @flags can have FOLL_ flags set, defined in <linux/mm.h>
 *
 * Returns the mapped (struct page *), %NULL if no mapping exists, or
 * an error pointer if there is a mapping to something not represented
 * by a page descriptor (see also vm_normal_page()).
L
Linus Torvalds 已提交
1455
 */
1456 1457 1458
struct page *follow_page_mask(struct vm_area_struct *vma,
			      unsigned long address, unsigned int flags,
			      unsigned int *page_mask)
L
Linus Torvalds 已提交
1459 1460 1461 1462 1463
{
	pgd_t *pgd;
	pud_t *pud;
	pmd_t *pmd;
	pte_t *ptep, pte;
1464
	spinlock_t *ptl;
L
Linus Torvalds 已提交
1465
	struct page *page;
1466
	struct mm_struct *mm = vma->vm_mm;
L
Linus Torvalds 已提交
1467

1468 1469
	*page_mask = 0;

1470 1471 1472 1473 1474
	page = follow_huge_addr(mm, address, flags & FOLL_WRITE);
	if (!IS_ERR(page)) {
		BUG_ON(flags & FOLL_GET);
		goto out;
	}
L
Linus Torvalds 已提交
1475

1476
	page = NULL;
L
Linus Torvalds 已提交
1477 1478
	pgd = pgd_offset(mm, address);
	if (pgd_none(*pgd) || unlikely(pgd_bad(*pgd)))
1479
		goto no_page_table;
L
Linus Torvalds 已提交
1480 1481

	pud = pud_offset(pgd, address);
A
Andi Kleen 已提交
1482
	if (pud_none(*pud))
1483
		goto no_page_table;
1484
	if (pud_huge(*pud) && vma->vm_flags & VM_HUGETLB) {
1485 1486
		if (flags & FOLL_GET)
			goto out;
A
Andi Kleen 已提交
1487 1488 1489 1490 1491 1492
		page = follow_huge_pud(mm, address, pud, flags & FOLL_WRITE);
		goto out;
	}
	if (unlikely(pud_bad(*pud)))
		goto no_page_table;

L
Linus Torvalds 已提交
1493
	pmd = pmd_offset(pud, address);
1494
	if (pmd_none(*pmd))
1495
		goto no_page_table;
1496
	if (pmd_huge(*pmd) && vma->vm_flags & VM_HUGETLB) {
1497
		page = follow_huge_pmd(mm, address, pmd, flags & FOLL_WRITE);
1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510
		if (flags & FOLL_GET) {
			/*
			 * Refcount on tail pages are not well-defined and
			 * shouldn't be taken. The caller should handle a NULL
			 * return when trying to follow tail pages.
			 */
			if (PageHead(page))
				get_page(page);
			else {
				page = NULL;
				goto out;
			}
		}
L
Linus Torvalds 已提交
1511
		goto out;
1512
	}
1513 1514
	if ((flags & FOLL_NUMA) && pmd_numa(*pmd))
		goto no_page_table;
1515
	if (pmd_trans_huge(*pmd)) {
1516
		if (flags & FOLL_SPLIT) {
1517
			split_huge_page_pmd(vma, address, pmd);
1518 1519
			goto split_fallthrough;
		}
1520
		ptl = pmd_lock(mm, pmd);
1521 1522
		if (likely(pmd_trans_huge(*pmd))) {
			if (unlikely(pmd_trans_splitting(*pmd))) {
1523
				spin_unlock(ptl);
1524 1525
				wait_split_huge_page(vma->anon_vma, pmd);
			} else {
1526
				page = follow_trans_huge_pmd(vma, address,
1527
							     pmd, flags);
1528
				spin_unlock(ptl);
1529
				*page_mask = HPAGE_PMD_NR - 1;
1530 1531 1532
				goto out;
			}
		} else
1533
			spin_unlock(ptl);
1534 1535
		/* fall through */
	}
1536
split_fallthrough:
1537 1538 1539
	if (unlikely(pmd_bad(*pmd)))
		goto no_page_table;

1540
	ptep = pte_offset_map_lock(mm, pmd, address, &ptl);
L
Linus Torvalds 已提交
1541 1542

	pte = *ptep;
1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560
	if (!pte_present(pte)) {
		swp_entry_t entry;
		/*
		 * KSM's break_ksm() relies upon recognizing a ksm page
		 * even while it is being migrated, so for that case we
		 * need migration_entry_wait().
		 */
		if (likely(!(flags & FOLL_MIGRATION)))
			goto no_page;
		if (pte_none(pte) || pte_file(pte))
			goto no_page;
		entry = pte_to_swp_entry(pte);
		if (!is_migration_entry(entry))
			goto no_page;
		pte_unmap_unlock(ptep, ptl);
		migration_entry_wait(mm, pmd, address);
		goto split_fallthrough;
	}
1561 1562
	if ((flags & FOLL_NUMA) && pte_numa(pte))
		goto no_page;
1563 1564
	if ((flags & FOLL_WRITE) && !pte_write(pte))
		goto unlock;
H
Hugh Dickins 已提交
1565

1566
	page = vm_normal_page(vma, address, pte);
H
Hugh Dickins 已提交
1567 1568
	if (unlikely(!page)) {
		if ((flags & FOLL_DUMP) ||
H
Hugh Dickins 已提交
1569
		    !is_zero_pfn(pte_pfn(pte)))
H
Hugh Dickins 已提交
1570 1571 1572
			goto bad_page;
		page = pte_page(pte);
	}
L
Linus Torvalds 已提交
1573

1574
	if (flags & FOLL_GET)
1575
		get_page_foll(page);
1576 1577 1578 1579
	if (flags & FOLL_TOUCH) {
		if ((flags & FOLL_WRITE) &&
		    !pte_dirty(pte) && !PageDirty(page))
			set_page_dirty(page);
1580 1581 1582 1583 1584
		/*
		 * pte_mkyoung() would be more correct here, but atomic care
		 * is needed to avoid losing the dirty bit: it is easier to use
		 * mark_page_accessed().
		 */
1585 1586
		mark_page_accessed(page);
	}
1587
	if ((flags & FOLL_MLOCK) && (vma->vm_flags & VM_LOCKED)) {
1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599
		/*
		 * The preliminary mapping check is mainly to avoid the
		 * pointless overhead of lock_page on the ZERO_PAGE
		 * which might bounce very badly if there is contention.
		 *
		 * If the page is already locked, we don't need to
		 * handle it now - vmscan will handle it later if and
		 * when it attempts to reclaim the page.
		 */
		if (page->mapping && trylock_page(page)) {
			lru_add_drain();  /* push cached pages to LRU */
			/*
1600 1601 1602 1603
			 * Because we lock page here, and migration is
			 * blocked by the pte's page reference, and we
			 * know the page is still mapped, we don't even
			 * need to check for file-cache page truncation.
1604
			 */
1605
			mlock_vma_page(page);
1606 1607 1608
			unlock_page(page);
		}
	}
1609 1610
unlock:
	pte_unmap_unlock(ptep, ptl);
L
Linus Torvalds 已提交
1611
out:
1612
	return page;
L
Linus Torvalds 已提交
1613

1614 1615 1616 1617 1618 1619 1620 1621
bad_page:
	pte_unmap_unlock(ptep, ptl);
	return ERR_PTR(-EFAULT);

no_page:
	pte_unmap_unlock(ptep, ptl);
	if (!pte_none(pte))
		return page;
H
Hugh Dickins 已提交
1622

1623 1624 1625
no_page_table:
	/*
	 * When core dumping an enormous anonymous area that nobody
H
Hugh Dickins 已提交
1626 1627 1628 1629 1630
	 * has touched so far, we don't want to allocate unnecessary pages or
	 * page tables.  Return error instead of NULL to skip handle_mm_fault,
	 * then get_dump_page() will return NULL to leave a hole in the dump.
	 * But we can only make this optimization where a hole would surely
	 * be zero-filled if handle_mm_fault() actually did handle it.
1631
	 */
H
Hugh Dickins 已提交
1632 1633 1634
	if ((flags & FOLL_DUMP) &&
	    (!vma->vm_ops || !vma->vm_ops->fault))
		return ERR_PTR(-EFAULT);
1635
	return page;
L
Linus Torvalds 已提交
1636 1637
}

1638 1639
static inline int stack_guard_page(struct vm_area_struct *vma, unsigned long addr)
{
1640 1641
	return stack_guard_page_start(vma, addr) ||
	       stack_guard_page_end(vma, addr+PAGE_SIZE);
1642 1643
}

H
Huang Ying 已提交
1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692
/**
 * __get_user_pages() - pin user pages in memory
 * @tsk:	task_struct of target task
 * @mm:		mm_struct of target mm
 * @start:	starting user address
 * @nr_pages:	number of pages from start to pin
 * @gup_flags:	flags modifying pin behaviour
 * @pages:	array that receives pointers to the pages pinned.
 *		Should be at least nr_pages long. Or NULL, if caller
 *		only intends to ensure the pages are faulted in.
 * @vmas:	array of pointers to vmas corresponding to each page.
 *		Or NULL if the caller does not require them.
 * @nonblocking: whether waiting for disk IO or mmap_sem contention
 *
 * Returns number of pages pinned. This may be fewer than the number
 * requested. If nr_pages is 0 or negative, returns 0. If no pages
 * were pinned, returns -errno. Each page returned must be released
 * with a put_page() call when it is finished with. vmas will only
 * remain valid while mmap_sem is held.
 *
 * Must be called with mmap_sem held for read or write.
 *
 * __get_user_pages walks a process's page tables and takes a reference to
 * each struct page that each user address corresponds to at a given
 * instant. That is, it takes the page that would be accessed if a user
 * thread accesses the given user virtual address at that instant.
 *
 * This does not guarantee that the page exists in the user mappings when
 * __get_user_pages returns, and there may even be a completely different
 * page there in some cases (eg. if mmapped pagecache has been invalidated
 * and subsequently re faulted). However it does guarantee that the page
 * won't be freed completely. And mostly callers simply care that the page
 * contains data that was valid *at some point in time*. Typically, an IO
 * or similar operation cannot guarantee anything stronger anyway because
 * locks can't be held over the syscall boundary.
 *
 * If @gup_flags & FOLL_WRITE == 0, the page must not be written to. If
 * the page is written to, set_page_dirty (or set_page_dirty_lock, as
 * appropriate) must be called after the page is finished with, and
 * before put_page is called.
 *
 * If @nonblocking != NULL, __get_user_pages will not wait for disk IO
 * or mmap_sem contention, and if waiting is needed to pin all pages,
 * *@nonblocking will be set to 0.
 *
 * In most cases, get_user_pages or get_user_pages_fast should be used
 * instead of __get_user_pages. __get_user_pages should be used only if
 * you need some special @gup_flags.
 */
1693 1694 1695 1696
long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
		unsigned long start, unsigned long nr_pages,
		unsigned int gup_flags, struct page **pages,
		struct vm_area_struct **vmas, int *nonblocking)
L
Linus Torvalds 已提交
1697
{
1698
	long i;
H
Hugh Dickins 已提交
1699
	unsigned long vm_flags;
1700
	unsigned int page_mask;
L
Linus Torvalds 已提交
1701

1702
	if (!nr_pages)
1703
		return 0;
H
Hugh Dickins 已提交
1704 1705 1706

	VM_BUG_ON(!!pages != !!(gup_flags & FOLL_GET));

L
Linus Torvalds 已提交
1707 1708
	/* 
	 * Require read or write permissions.
H
Hugh Dickins 已提交
1709
	 * If FOLL_FORCE is set, we only require the "MAY" flags.
L
Linus Torvalds 已提交
1710
	 */
H
Hugh Dickins 已提交
1711 1712 1713 1714
	vm_flags  = (gup_flags & FOLL_WRITE) ?
			(VM_WRITE | VM_MAYWRITE) : (VM_READ | VM_MAYREAD);
	vm_flags &= (gup_flags & FOLL_FORCE) ?
			(VM_MAYREAD | VM_MAYWRITE) : (VM_READ | VM_WRITE);
1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727

	/*
	 * If FOLL_FORCE and FOLL_NUMA are both set, handle_mm_fault
	 * would be called on PROT_NONE ranges. We must never invoke
	 * handle_mm_fault on PROT_NONE ranges or the NUMA hinting
	 * page faults would unprotect the PROT_NONE ranges if
	 * _PAGE_NUMA and _PAGE_PROTNONE are sharing the same pte/pmd
	 * bitflag. So to avoid that, don't set FOLL_NUMA if
	 * FOLL_FORCE is set.
	 */
	if (!(gup_flags & FOLL_FORCE))
		gup_flags |= FOLL_NUMA;

L
Linus Torvalds 已提交
1728 1729 1730
	i = 0;

	do {
1731
		struct vm_area_struct *vma;
L
Linus Torvalds 已提交
1732 1733

		vma = find_extend_vma(mm, start);
1734
		if (!vma && in_gate_area(mm, start)) {
L
Linus Torvalds 已提交
1735 1736 1737 1738 1739
			unsigned long pg = start & PAGE_MASK;
			pgd_t *pgd;
			pud_t *pud;
			pmd_t *pmd;
			pte_t *pte;
N
Nick Piggin 已提交
1740 1741

			/* user gate pages are read-only */
H
Hugh Dickins 已提交
1742
			if (gup_flags & FOLL_WRITE)
L
Linus Torvalds 已提交
1743 1744 1745 1746 1747 1748 1749 1750 1751
				return i ? : -EFAULT;
			if (pg > TASK_SIZE)
				pgd = pgd_offset_k(pg);
			else
				pgd = pgd_offset_gate(mm, pg);
			BUG_ON(pgd_none(*pgd));
			pud = pud_offset(pgd, pg);
			BUG_ON(pud_none(*pud));
			pmd = pmd_offset(pud, pg);
1752 1753
			if (pmd_none(*pmd))
				return i ? : -EFAULT;
1754
			VM_BUG_ON(pmd_trans_huge(*pmd));
L
Linus Torvalds 已提交
1755
			pte = pte_offset_map(pmd, pg);
1756 1757 1758 1759
			if (pte_none(*pte)) {
				pte_unmap(pte);
				return i ? : -EFAULT;
			}
1760
			vma = get_gate_vma(mm);
L
Linus Torvalds 已提交
1761
			if (pages) {
1762 1763
				struct page *page;

1764
				page = vm_normal_page(vma, start, *pte);
1765 1766 1767 1768 1769 1770 1771 1772 1773
				if (!page) {
					if (!(gup_flags & FOLL_DUMP) &&
					     is_zero_pfn(pte_pfn(*pte)))
						page = pte_page(*pte);
					else {
						pte_unmap(pte);
						return i ? : -EFAULT;
					}
				}
1774
				pages[i] = page;
1775
				get_page(page);
L
Linus Torvalds 已提交
1776 1777
			}
			pte_unmap(pte);
1778
			page_mask = 0;
1779
			goto next_page;
L
Linus Torvalds 已提交
1780 1781
		}

N
Nick Piggin 已提交
1782 1783
		if (!vma ||
		    (vma->vm_flags & (VM_IO | VM_PFNMAP)) ||
H
Hugh Dickins 已提交
1784
		    !(vm_flags & vma->vm_flags))
L
Linus Torvalds 已提交
1785 1786
			return i ? : -EFAULT;

H
Hugh Dickins 已提交
1787 1788
		if (is_vm_hugetlb_page(vma)) {
			i = follow_hugetlb_page(mm, vma, pages, vmas,
H
Hugh Dickins 已提交
1789
					&start, &nr_pages, i, gup_flags);
H
Hugh Dickins 已提交
1790 1791
			continue;
		}
1792

L
Linus Torvalds 已提交
1793
		do {
1794
			struct page *page;
H
Hugh Dickins 已提交
1795
			unsigned int foll_flags = gup_flags;
1796
			unsigned int page_increm;
L
Linus Torvalds 已提交
1797

1798
			/*
1799
			 * If we have a pending SIGKILL, don't keep faulting
H
Hugh Dickins 已提交
1800
			 * pages and potentially allocating memory.
1801
			 */
H
Hugh Dickins 已提交
1802
			if (unlikely(fatal_signal_pending(current)))
1803
				return i ? i : -ERESTARTSYS;
1804

1805
			cond_resched();
1806 1807
			while (!(page = follow_page_mask(vma, start,
						foll_flags, &page_mask))) {
1808
				int ret;
1809 1810
				unsigned int fault_flags = 0;

1811 1812 1813 1814 1815
				/* For mlock, just skip the stack guard page. */
				if (foll_flags & FOLL_MLOCK) {
					if (stack_guard_page(vma, start))
						goto next_page;
				}
1816 1817 1818 1819
				if (foll_flags & FOLL_WRITE)
					fault_flags |= FAULT_FLAG_WRITE;
				if (nonblocking)
					fault_flags |= FAULT_FLAG_ALLOW_RETRY;
1820 1821
				if (foll_flags & FOLL_NOWAIT)
					fault_flags |= (FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_RETRY_NOWAIT);
1822

1823
				ret = handle_mm_fault(mm, vma, start,
1824
							fault_flags);
1825

N
Nick Piggin 已提交
1826 1827 1828
				if (ret & VM_FAULT_ERROR) {
					if (ret & VM_FAULT_OOM)
						return i ? i : -ENOMEM;
1829 1830 1831 1832 1833 1834 1835 1836 1837 1838
					if (ret & (VM_FAULT_HWPOISON |
						   VM_FAULT_HWPOISON_LARGE)) {
						if (i)
							return i;
						else if (gup_flags & FOLL_HWPOISON)
							return -EHWPOISON;
						else
							return -EFAULT;
					}
					if (ret & VM_FAULT_SIGBUS)
N
Nick Piggin 已提交
1839 1840 1841
						return i ? i : -EFAULT;
					BUG();
				}
1842 1843 1844 1845 1846 1847 1848

				if (tsk) {
					if (ret & VM_FAULT_MAJOR)
						tsk->maj_flt++;
					else
						tsk->min_flt++;
				}
N
Nick Piggin 已提交
1849

1850
				if (ret & VM_FAULT_RETRY) {
1851 1852
					if (nonblocking)
						*nonblocking = 0;
1853 1854 1855
					return i;
				}

1856
				/*
N
Nick Piggin 已提交
1857 1858 1859 1860
				 * The VM_FAULT_WRITE bit tells us that
				 * do_wp_page has broken COW when necessary,
				 * even if maybe_mkwrite decided not to set
				 * pte_write. We can thus safely do subsequent
1861 1862 1863 1864 1865 1866
				 * page lookups as if they were reads. But only
				 * do so when looping for pte_write is futile:
				 * in some cases userspace may also be wanting
				 * to write to the gotten user page, which a
				 * read fault here might prevent (a readonly
				 * page might get reCOWed by userspace write).
1867
				 */
1868 1869
				if ((ret & VM_FAULT_WRITE) &&
				    !(vma->vm_flags & VM_WRITE))
1870
					foll_flags &= ~FOLL_WRITE;
N
Nick Piggin 已提交
1871

1872
				cond_resched();
L
Linus Torvalds 已提交
1873
			}
1874 1875
			if (IS_ERR(page))
				return i ? i : PTR_ERR(page);
L
Linus Torvalds 已提交
1876
			if (pages) {
1877
				pages[i] = page;
1878

1879
				flush_anon_page(vma, page, start);
1880
				flush_dcache_page(page);
1881
				page_mask = 0;
L
Linus Torvalds 已提交
1882
			}
1883
next_page:
1884
			if (vmas) {
L
Linus Torvalds 已提交
1885
				vmas[i] = vma;
1886 1887 1888 1889 1890 1891 1892 1893
				page_mask = 0;
			}
			page_increm = 1 + (~(start >> PAGE_SHIFT) & page_mask);
			if (page_increm > nr_pages)
				page_increm = nr_pages;
			i += page_increm;
			start += page_increm * PAGE_SIZE;
			nr_pages -= page_increm;
1894 1895
		} while (nr_pages && start < vma->vm_end);
	} while (nr_pages);
L
Linus Torvalds 已提交
1896 1897
	return i;
}
H
Huang Ying 已提交
1898
EXPORT_SYMBOL(__get_user_pages);
N
Nick Piggin 已提交
1899

1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956
/*
 * fixup_user_fault() - manually resolve a user page fault
 * @tsk:	the task_struct to use for page fault accounting, or
 *		NULL if faults are not to be recorded.
 * @mm:		mm_struct of target mm
 * @address:	user address
 * @fault_flags:flags to pass down to handle_mm_fault()
 *
 * This is meant to be called in the specific scenario where for locking reasons
 * we try to access user memory in atomic context (within a pagefault_disable()
 * section), this returns -EFAULT, and we want to resolve the user fault before
 * trying again.
 *
 * Typically this is meant to be used by the futex code.
 *
 * The main difference with get_user_pages() is that this function will
 * unconditionally call handle_mm_fault() which will in turn perform all the
 * necessary SW fixup of the dirty and young bits in the PTE, while
 * handle_mm_fault() only guarantees to update these in the struct page.
 *
 * This is important for some architectures where those bits also gate the
 * access permission to the page because they are maintained in software.  On
 * such architectures, gup() will not be enough to make a subsequent access
 * succeed.
 *
 * This should be called with the mm_sem held for read.
 */
int fixup_user_fault(struct task_struct *tsk, struct mm_struct *mm,
		     unsigned long address, unsigned int fault_flags)
{
	struct vm_area_struct *vma;
	int ret;

	vma = find_extend_vma(mm, address);
	if (!vma || address < vma->vm_start)
		return -EFAULT;

	ret = handle_mm_fault(mm, vma, address, fault_flags);
	if (ret & VM_FAULT_ERROR) {
		if (ret & VM_FAULT_OOM)
			return -ENOMEM;
		if (ret & (VM_FAULT_HWPOISON | VM_FAULT_HWPOISON_LARGE))
			return -EHWPOISON;
		if (ret & VM_FAULT_SIGBUS)
			return -EFAULT;
		BUG();
	}
	if (tsk) {
		if (ret & VM_FAULT_MAJOR)
			tsk->maj_flt++;
		else
			tsk->min_flt++;
	}
	return 0;
}

/*
1957
 * get_user_pages() - pin user pages in memory
1958 1959
 * @tsk:	the task_struct to use for page fault accounting, or
 *		NULL if faults are not to be recorded.
1960 1961
 * @mm:		mm_struct of target mm
 * @start:	starting user address
1962
 * @nr_pages:	number of pages from start to pin
1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973
 * @write:	whether pages will be written to by the caller
 * @force:	whether to force write access even if user mapping is
 *		readonly. This will result in the page being COWed even
 *		in MAP_SHARED mappings. You do not want this.
 * @pages:	array that receives pointers to the pages pinned.
 *		Should be at least nr_pages long. Or NULL, if caller
 *		only intends to ensure the pages are faulted in.
 * @vmas:	array of pointers to vmas corresponding to each page.
 *		Or NULL if the caller does not require them.
 *
 * Returns number of pages pinned. This may be fewer than the number
1974
 * requested. If nr_pages is 0 or negative, returns 0. If no pages
1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
 * were pinned, returns -errno. Each page returned must be released
 * with a put_page() call when it is finished with. vmas will only
 * remain valid while mmap_sem is held.
 *
 * Must be called with mmap_sem held for read or write.
 *
 * get_user_pages walks a process's page tables and takes a reference to
 * each struct page that each user address corresponds to at a given
 * instant. That is, it takes the page that would be accessed if a user
 * thread accesses the given user virtual address at that instant.
 *
 * This does not guarantee that the page exists in the user mappings when
 * get_user_pages returns, and there may even be a completely different
 * page there in some cases (eg. if mmapped pagecache has been invalidated
 * and subsequently re faulted). However it does guarantee that the page
 * won't be freed completely. And mostly callers simply care that the page
 * contains data that was valid *at some point in time*. Typically, an IO
 * or similar operation cannot guarantee anything stronger anyway because
 * locks can't be held over the syscall boundary.
 *
 * If write=0, the page must not be written to. If the page is written to,
 * set_page_dirty (or set_page_dirty_lock, as appropriate) must be called
 * after the page is finished with, and before put_page is called.
 *
 * get_user_pages is typically used for fewer-copy IO operations, to get a
 * handle on the memory by some means other than accesses via the user virtual
 * addresses. The pages may be submitted for DMA to devices or accessed via
 * their kernel linear mapping (via the kmap APIs). Care should be taken to
 * use the correct cache flushing APIs.
 *
 * See also get_user_pages_fast, for performance critical applications.
 */
2007 2008 2009
long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
		unsigned long start, unsigned long nr_pages, int write,
		int force, struct page **pages, struct vm_area_struct **vmas)
N
Nick Piggin 已提交
2010
{
H
Hugh Dickins 已提交
2011
	int flags = FOLL_TOUCH;
N
Nick Piggin 已提交
2012

H
Hugh Dickins 已提交
2013 2014
	if (pages)
		flags |= FOLL_GET;
N
Nick Piggin 已提交
2015
	if (write)
H
Hugh Dickins 已提交
2016
		flags |= FOLL_WRITE;
N
Nick Piggin 已提交
2017
	if (force)
H
Hugh Dickins 已提交
2018
		flags |= FOLL_FORCE;
N
Nick Piggin 已提交
2019

2020 2021
	return __get_user_pages(tsk, mm, start, nr_pages, flags, pages, vmas,
				NULL);
N
Nick Piggin 已提交
2022
}
L
Linus Torvalds 已提交
2023 2024
EXPORT_SYMBOL(get_user_pages);

H
Hugh Dickins 已提交
2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045
/**
 * get_dump_page() - pin user page in memory while writing it to core dump
 * @addr: user address
 *
 * Returns struct page pointer of user page pinned for dump,
 * to be freed afterwards by page_cache_release() or put_page().
 *
 * Returns NULL on any kind of failure - a hole must then be inserted into
 * the corefile, to preserve alignment with its headers; and also returns
 * NULL wherever the ZERO_PAGE, or an anonymous pte_none, has been found -
 * allowing a hole to be left in the corefile to save diskspace.
 *
 * Called without mmap_sem, but after all other threads have been killed.
 */
#ifdef CONFIG_ELF_CORE
struct page *get_dump_page(unsigned long addr)
{
	struct vm_area_struct *vma;
	struct page *page;

	if (__get_user_pages(current, current->mm, addr, 1,
2046 2047
			     FOLL_FORCE | FOLL_DUMP | FOLL_GET, &page, &vma,
			     NULL) < 1)
H
Hugh Dickins 已提交
2048 2049 2050 2051 2052 2053
		return NULL;
	flush_cache_page(vma, addr, page_to_pfn(page));
	return page;
}
#endif /* CONFIG_ELF_CORE */

2054
pte_t *__get_locked_pte(struct mm_struct *mm, unsigned long addr,
H
Harvey Harrison 已提交
2055
			spinlock_t **ptl)
2056 2057 2058 2059
{
	pgd_t * pgd = pgd_offset(mm, addr);
	pud_t * pud = pud_alloc(mm, pgd, addr);
	if (pud) {
2060
		pmd_t * pmd = pmd_alloc(mm, pud, addr);
2061 2062
		if (pmd) {
			VM_BUG_ON(pmd_trans_huge(*pmd));
2063
			return pte_alloc_map_lock(mm, pmd, addr, ptl);
2064
		}
2065 2066 2067 2068
	}
	return NULL;
}

2069 2070 2071 2072 2073 2074 2075
/*
 * This is the old fallback for page remapping.
 *
 * For historical reasons, it only allows reserved pages. Only
 * old drivers should use this, and they needed to mark their
 * pages reserved for the old functions anyway.
 */
N
Nick Piggin 已提交
2076 2077
static int insert_page(struct vm_area_struct *vma, unsigned long addr,
			struct page *page, pgprot_t prot)
2078
{
N
Nick Piggin 已提交
2079
	struct mm_struct *mm = vma->vm_mm;
2080
	int retval;
2081
	pte_t *pte;
2082 2083
	spinlock_t *ptl;

2084
	retval = -EINVAL;
2085
	if (PageAnon(page))
2086
		goto out;
2087 2088
	retval = -ENOMEM;
	flush_dcache_page(page);
2089
	pte = get_locked_pte(mm, addr, &ptl);
2090
	if (!pte)
2091
		goto out;
2092 2093 2094 2095 2096 2097
	retval = -EBUSY;
	if (!pte_none(*pte))
		goto out_unlock;

	/* Ok, finally just insert the thing.. */
	get_page(page);
2098
	inc_mm_counter_fast(mm, MM_FILEPAGES);
2099 2100 2101 2102
	page_add_file_rmap(page);
	set_pte_at(mm, addr, pte, mk_pte(page, prot));

	retval = 0;
2103 2104
	pte_unmap_unlock(pte, ptl);
	return retval;
2105 2106 2107 2108 2109 2110
out_unlock:
	pte_unmap_unlock(pte, ptl);
out:
	return retval;
}

2111 2112 2113 2114 2115 2116
/**
 * vm_insert_page - insert single page into user vma
 * @vma: user vma to map to
 * @addr: target user address of this page
 * @page: source kernel page
 *
2117 2118 2119 2120 2121 2122
 * This allows drivers to insert individual pages they've allocated
 * into a user vma.
 *
 * The page has to be a nice clean _individual_ kernel allocation.
 * If you allocate a compound page, you need to have marked it as
 * such (__GFP_COMP), or manually just split the page up yourself
N
Nick Piggin 已提交
2123
 * (see split_page()).
2124 2125 2126 2127 2128 2129 2130 2131
 *
 * NOTE! Traditionally this was done with "remap_pfn_range()" which
 * took an arbitrary page protection parameter. This doesn't allow
 * that. Your vma protection will have to be set up correctly, which
 * means that if you want a shared writable mapping, you'd better
 * ask for a shared writable mapping!
 *
 * The page does not need to be reserved.
2132 2133 2134 2135 2136
 *
 * Usually this function is called from f_op->mmap() handler
 * under mm->mmap_sem write-lock, so it can change vma->vm_flags.
 * Caller must set VM_MIXEDMAP on vma if it wants to call this
 * function from other places, for example from page-fault handler.
2137
 */
N
Nick Piggin 已提交
2138 2139
int vm_insert_page(struct vm_area_struct *vma, unsigned long addr,
			struct page *page)
2140 2141 2142 2143 2144
{
	if (addr < vma->vm_start || addr >= vma->vm_end)
		return -EFAULT;
	if (!page_count(page))
		return -EINVAL;
2145 2146 2147 2148 2149
	if (!(vma->vm_flags & VM_MIXEDMAP)) {
		BUG_ON(down_read_trylock(&vma->vm_mm->mmap_sem));
		BUG_ON(vma->vm_flags & VM_PFNMAP);
		vma->vm_flags |= VM_MIXEDMAP;
	}
N
Nick Piggin 已提交
2150
	return insert_page(vma, addr, page, vma->vm_page_prot);
2151
}
2152
EXPORT_SYMBOL(vm_insert_page);
2153

N
Nick Piggin 已提交
2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172
static int insert_pfn(struct vm_area_struct *vma, unsigned long addr,
			unsigned long pfn, pgprot_t prot)
{
	struct mm_struct *mm = vma->vm_mm;
	int retval;
	pte_t *pte, entry;
	spinlock_t *ptl;

	retval = -ENOMEM;
	pte = get_locked_pte(mm, addr, &ptl);
	if (!pte)
		goto out;
	retval = -EBUSY;
	if (!pte_none(*pte))
		goto out_unlock;

	/* Ok, finally just insert the thing.. */
	entry = pte_mkspecial(pfn_pte(pfn, prot));
	set_pte_at(mm, addr, pte, entry);
2173
	update_mmu_cache(vma, addr, pte); /* XXX: why not for insert_page? */
N
Nick Piggin 已提交
2174 2175 2176 2177 2178 2179 2180 2181

	retval = 0;
out_unlock:
	pte_unmap_unlock(pte, ptl);
out:
	return retval;
}

N
Nick Piggin 已提交
2182 2183 2184 2185 2186 2187
/**
 * vm_insert_pfn - insert single pfn into user vma
 * @vma: user vma to map to
 * @addr: target user address of this page
 * @pfn: source kernel pfn
 *
2188
 * Similar to vm_insert_page, this allows drivers to insert individual pages
N
Nick Piggin 已提交
2189 2190 2191 2192
 * they've allocated into a user vma. Same comments apply.
 *
 * This function should only be called from a vm_ops->fault handler, and
 * in that case the handler should return NULL.
N
Nick Piggin 已提交
2193 2194 2195 2196 2197
 *
 * vma cannot be a COW mapping.
 *
 * As this is called only for pages that do not currently exist, we
 * do not need to flush old virtual caches or the TLB.
N
Nick Piggin 已提交
2198 2199
 */
int vm_insert_pfn(struct vm_area_struct *vma, unsigned long addr,
N
Nick Piggin 已提交
2200
			unsigned long pfn)
N
Nick Piggin 已提交
2201
{
2202
	int ret;
2203
	pgprot_t pgprot = vma->vm_page_prot;
N
Nick Piggin 已提交
2204 2205 2206 2207 2208 2209
	/*
	 * Technically, architectures with pte_special can avoid all these
	 * restrictions (same for remap_pfn_range).  However we would like
	 * consistency in testing and feature parity among all, so we should
	 * try to keep these invariants in place for everybody.
	 */
J
Jared Hulbert 已提交
2210 2211 2212 2213 2214
	BUG_ON(!(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP)));
	BUG_ON((vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP)) ==
						(VM_PFNMAP|VM_MIXEDMAP));
	BUG_ON((vma->vm_flags & VM_PFNMAP) && is_cow_mapping(vma->vm_flags));
	BUG_ON((vma->vm_flags & VM_MIXEDMAP) && pfn_valid(pfn));
N
Nick Piggin 已提交
2215

N
Nick Piggin 已提交
2216 2217
	if (addr < vma->vm_start || addr >= vma->vm_end)
		return -EFAULT;
2218
	if (track_pfn_insert(vma, &pgprot, pfn))
2219 2220
		return -EINVAL;

2221
	ret = insert_pfn(vma, addr, pfn, pgprot);
2222 2223

	return ret;
N
Nick Piggin 已提交
2224 2225
}
EXPORT_SYMBOL(vm_insert_pfn);
N
Nick Piggin 已提交
2226

N
Nick Piggin 已提交
2227 2228 2229 2230
int vm_insert_mixed(struct vm_area_struct *vma, unsigned long addr,
			unsigned long pfn)
{
	BUG_ON(!(vma->vm_flags & VM_MIXEDMAP));
N
Nick Piggin 已提交
2231

N
Nick Piggin 已提交
2232 2233
	if (addr < vma->vm_start || addr >= vma->vm_end)
		return -EFAULT;
N
Nick Piggin 已提交
2234

N
Nick Piggin 已提交
2235 2236 2237 2238
	/*
	 * If we don't have pte special, then we have to use the pfn_valid()
	 * based VM_MIXEDMAP scheme (see vm_normal_page), and thus we *must*
	 * refcount the page if pfn_valid is true (hence insert_page rather
H
Hugh Dickins 已提交
2239 2240
	 * than insert_pfn).  If a zero_pfn were inserted into a VM_MIXEDMAP
	 * without pte special, it would there be refcounted as a normal page.
N
Nick Piggin 已提交
2241 2242 2243 2244 2245 2246 2247 2248
	 */
	if (!HAVE_PTE_SPECIAL && pfn_valid(pfn)) {
		struct page *page;

		page = pfn_to_page(pfn);
		return insert_page(vma, addr, page, vma->vm_page_prot);
	}
	return insert_pfn(vma, addr, pfn, vma->vm_page_prot);
N
Nick Piggin 已提交
2249
}
N
Nick Piggin 已提交
2250
EXPORT_SYMBOL(vm_insert_mixed);
N
Nick Piggin 已提交
2251

L
Linus Torvalds 已提交
2252 2253 2254 2255 2256 2257 2258 2259 2260 2261
/*
 * maps a range of physical memory into the requested pages. the old
 * mappings are removed. any references to nonexistent pages results
 * in null mappings (currently treated as "copy-on-access")
 */
static int remap_pte_range(struct mm_struct *mm, pmd_t *pmd,
			unsigned long addr, unsigned long end,
			unsigned long pfn, pgprot_t prot)
{
	pte_t *pte;
H
Hugh Dickins 已提交
2262
	spinlock_t *ptl;
L
Linus Torvalds 已提交
2263

H
Hugh Dickins 已提交
2264
	pte = pte_alloc_map_lock(mm, pmd, addr, &ptl);
L
Linus Torvalds 已提交
2265 2266
	if (!pte)
		return -ENOMEM;
2267
	arch_enter_lazy_mmu_mode();
L
Linus Torvalds 已提交
2268 2269
	do {
		BUG_ON(!pte_none(*pte));
N
Nick Piggin 已提交
2270
		set_pte_at(mm, addr, pte, pte_mkspecial(pfn_pte(pfn, prot)));
L
Linus Torvalds 已提交
2271 2272
		pfn++;
	} while (pte++, addr += PAGE_SIZE, addr != end);
2273
	arch_leave_lazy_mmu_mode();
H
Hugh Dickins 已提交
2274
	pte_unmap_unlock(pte - 1, ptl);
L
Linus Torvalds 已提交
2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288
	return 0;
}

static inline int remap_pmd_range(struct mm_struct *mm, pud_t *pud,
			unsigned long addr, unsigned long end,
			unsigned long pfn, pgprot_t prot)
{
	pmd_t *pmd;
	unsigned long next;

	pfn -= addr >> PAGE_SHIFT;
	pmd = pmd_alloc(mm, pud, addr);
	if (!pmd)
		return -ENOMEM;
2289
	VM_BUG_ON(pmd_trans_huge(*pmd));
L
Linus Torvalds 已提交
2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318
	do {
		next = pmd_addr_end(addr, end);
		if (remap_pte_range(mm, pmd, addr, next,
				pfn + (addr >> PAGE_SHIFT), prot))
			return -ENOMEM;
	} while (pmd++, addr = next, addr != end);
	return 0;
}

static inline int remap_pud_range(struct mm_struct *mm, pgd_t *pgd,
			unsigned long addr, unsigned long end,
			unsigned long pfn, pgprot_t prot)
{
	pud_t *pud;
	unsigned long next;

	pfn -= addr >> PAGE_SHIFT;
	pud = pud_alloc(mm, pgd, addr);
	if (!pud)
		return -ENOMEM;
	do {
		next = pud_addr_end(addr, end);
		if (remap_pmd_range(mm, pud, addr, next,
				pfn + (addr >> PAGE_SHIFT), prot))
			return -ENOMEM;
	} while (pud++, addr = next, addr != end);
	return 0;
}

2319 2320 2321 2322 2323 2324 2325 2326 2327 2328
/**
 * remap_pfn_range - remap kernel memory to userspace
 * @vma: user vma to map to
 * @addr: target user address to start at
 * @pfn: physical address of kernel memory
 * @size: size of map area
 * @prot: page protection flags for this mapping
 *
 *  Note: this is only safe if the mm semaphore is held when called.
 */
L
Linus Torvalds 已提交
2329 2330 2331 2332 2333
int remap_pfn_range(struct vm_area_struct *vma, unsigned long addr,
		    unsigned long pfn, unsigned long size, pgprot_t prot)
{
	pgd_t *pgd;
	unsigned long next;
2334
	unsigned long end = addr + PAGE_ALIGN(size);
L
Linus Torvalds 已提交
2335 2336 2337 2338 2339 2340 2341 2342
	struct mm_struct *mm = vma->vm_mm;
	int err;

	/*
	 * Physically remapped pages are special. Tell the
	 * rest of the world about it:
	 *   VM_IO tells people not to look at these pages
	 *	(accesses can have side effects).
2343 2344 2345
	 *   VM_PFNMAP tells the core MM that the base pages are just
	 *	raw PFN mappings, and do not have a "struct page" associated
	 *	with them.
2346 2347 2348 2349
	 *   VM_DONTEXPAND
	 *      Disable vma merging and expanding with mremap().
	 *   VM_DONTDUMP
	 *      Omit vma from core dump, even when VM_IO turned off.
L
Linus Torvalds 已提交
2350 2351 2352 2353
	 *
	 * There's a horrible special case to handle copy-on-write
	 * behaviour that some programs depend on. We mark the "original"
	 * un-COW'ed pages by matching them up with "vma->vm_pgoff".
2354
	 * See vm_normal_page() for details.
L
Linus Torvalds 已提交
2355
	 */
2356 2357 2358
	if (is_cow_mapping(vma->vm_flags)) {
		if (addr != vma->vm_start || end != vma->vm_end)
			return -EINVAL;
L
Linus Torvalds 已提交
2359
		vma->vm_pgoff = pfn;
2360 2361 2362 2363
	}

	err = track_pfn_remap(vma, &prot, pfn, addr, PAGE_ALIGN(size));
	if (err)
2364
		return -EINVAL;
L
Linus Torvalds 已提交
2365

2366
	vma->vm_flags |= VM_IO | VM_PFNMAP | VM_DONTEXPAND | VM_DONTDUMP;
L
Linus Torvalds 已提交
2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378

	BUG_ON(addr >= end);
	pfn -= addr >> PAGE_SHIFT;
	pgd = pgd_offset(mm, addr);
	flush_cache_range(vma, addr, end);
	do {
		next = pgd_addr_end(addr, end);
		err = remap_pud_range(mm, pgd, addr, next,
				pfn + (addr >> PAGE_SHIFT), prot);
		if (err)
			break;
	} while (pgd++, addr = next, addr != end);
2379 2380

	if (err)
2381
		untrack_pfn(vma, pfn, PAGE_ALIGN(size));
2382

L
Linus Torvalds 已提交
2383 2384 2385 2386
	return err;
}
EXPORT_SYMBOL(remap_pfn_range);

2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433
/**
 * vm_iomap_memory - remap memory to userspace
 * @vma: user vma to map to
 * @start: start of area
 * @len: size of area
 *
 * This is a simplified io_remap_pfn_range() for common driver use. The
 * driver just needs to give us the physical memory range to be mapped,
 * we'll figure out the rest from the vma information.
 *
 * NOTE! Some drivers might want to tweak vma->vm_page_prot first to get
 * whatever write-combining details or similar.
 */
int vm_iomap_memory(struct vm_area_struct *vma, phys_addr_t start, unsigned long len)
{
	unsigned long vm_len, pfn, pages;

	/* Check that the physical memory area passed in looks valid */
	if (start + len < start)
		return -EINVAL;
	/*
	 * You *really* shouldn't map things that aren't page-aligned,
	 * but we've historically allowed it because IO memory might
	 * just have smaller alignment.
	 */
	len += start & ~PAGE_MASK;
	pfn = start >> PAGE_SHIFT;
	pages = (len + ~PAGE_MASK) >> PAGE_SHIFT;
	if (pfn + pages < pfn)
		return -EINVAL;

	/* We start the mapping 'vm_pgoff' pages into the area */
	if (vma->vm_pgoff > pages)
		return -EINVAL;
	pfn += vma->vm_pgoff;
	pages -= vma->vm_pgoff;

	/* Can we fit all of the mapping? */
	vm_len = vma->vm_end - vma->vm_start;
	if (vm_len >> PAGE_SHIFT > pages)
		return -EINVAL;

	/* Ok, let it rip */
	return io_remap_pfn_range(vma, vma->vm_start, pfn, vm_len, vma->vm_page_prot);
}
EXPORT_SYMBOL(vm_iomap_memory);

2434 2435 2436 2437 2438 2439
static int apply_to_pte_range(struct mm_struct *mm, pmd_t *pmd,
				     unsigned long addr, unsigned long end,
				     pte_fn_t fn, void *data)
{
	pte_t *pte;
	int err;
2440
	pgtable_t token;
2441
	spinlock_t *uninitialized_var(ptl);
2442 2443 2444 2445 2446 2447 2448 2449 2450

	pte = (mm == &init_mm) ?
		pte_alloc_kernel(pmd, addr) :
		pte_alloc_map_lock(mm, pmd, addr, &ptl);
	if (!pte)
		return -ENOMEM;

	BUG_ON(pmd_huge(*pmd));

2451 2452
	arch_enter_lazy_mmu_mode();

2453
	token = pmd_pgtable(*pmd);
2454 2455

	do {
2456
		err = fn(pte++, token, addr, data);
2457 2458
		if (err)
			break;
2459
	} while (addr += PAGE_SIZE, addr != end);
2460

2461 2462
	arch_leave_lazy_mmu_mode();

2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475
	if (mm != &init_mm)
		pte_unmap_unlock(pte-1, ptl);
	return err;
}

static int apply_to_pmd_range(struct mm_struct *mm, pud_t *pud,
				     unsigned long addr, unsigned long end,
				     pte_fn_t fn, void *data)
{
	pmd_t *pmd;
	unsigned long next;
	int err;

A
Andi Kleen 已提交
2476 2477
	BUG_ON(pud_huge(*pud));

2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518
	pmd = pmd_alloc(mm, pud, addr);
	if (!pmd)
		return -ENOMEM;
	do {
		next = pmd_addr_end(addr, end);
		err = apply_to_pte_range(mm, pmd, addr, next, fn, data);
		if (err)
			break;
	} while (pmd++, addr = next, addr != end);
	return err;
}

static int apply_to_pud_range(struct mm_struct *mm, pgd_t *pgd,
				     unsigned long addr, unsigned long end,
				     pte_fn_t fn, void *data)
{
	pud_t *pud;
	unsigned long next;
	int err;

	pud = pud_alloc(mm, pgd, addr);
	if (!pud)
		return -ENOMEM;
	do {
		next = pud_addr_end(addr, end);
		err = apply_to_pmd_range(mm, pud, addr, next, fn, data);
		if (err)
			break;
	} while (pud++, addr = next, addr != end);
	return err;
}

/*
 * Scan a region of virtual memory, filling in page tables as necessary
 * and calling a provided function on each leaf page table.
 */
int apply_to_page_range(struct mm_struct *mm, unsigned long addr,
			unsigned long size, pte_fn_t fn, void *data)
{
	pgd_t *pgd;
	unsigned long next;
2519
	unsigned long end = addr + size;
2520 2521 2522 2523 2524 2525 2526 2527 2528 2529
	int err;

	BUG_ON(addr >= end);
	pgd = pgd_offset(mm, addr);
	do {
		next = pgd_addr_end(addr, end);
		err = apply_to_pud_range(mm, pgd, addr, next, fn, data);
		if (err)
			break;
	} while (pgd++, addr = next, addr != end);
2530

2531 2532 2533 2534
	return err;
}
EXPORT_SYMBOL_GPL(apply_to_page_range);

2535 2536 2537 2538
/*
 * handle_pte_fault chooses page fault handler according to an entry
 * which was read non-atomically.  Before making any commitment, on
 * those architectures or configurations (e.g. i386 with PAE) which
2539
 * might give a mix of unmatched parts, do_swap_page and do_nonlinear_fault
2540 2541
 * must check under lock before unmapping the pte and proceeding
 * (but do_wp_page is only called after already making such a check;
2542
 * and do_anonymous_page can safely check later on).
2543
 */
H
Hugh Dickins 已提交
2544
static inline int pte_unmap_same(struct mm_struct *mm, pmd_t *pmd,
2545 2546 2547 2548 2549
				pte_t *page_table, pte_t orig_pte)
{
	int same = 1;
#if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT)
	if (sizeof(pte_t) > sizeof(unsigned long)) {
H
Hugh Dickins 已提交
2550 2551
		spinlock_t *ptl = pte_lockptr(mm, pmd);
		spin_lock(ptl);
2552
		same = pte_same(*page_table, orig_pte);
H
Hugh Dickins 已提交
2553
		spin_unlock(ptl);
2554 2555 2556 2557 2558 2559
	}
#endif
	pte_unmap(page_table);
	return same;
}

2560
static inline void cow_user_page(struct page *dst, struct page *src, unsigned long va, struct vm_area_struct *vma)
2561 2562 2563 2564 2565 2566 2567 2568
{
	/*
	 * If the source page was a PFN mapping, we don't have
	 * a "struct page" for it. We do a best-effort copy by
	 * just copying from the original user address. If that
	 * fails, we just zero-fill it. Live with it.
	 */
	if (unlikely(!src)) {
2569
		void *kaddr = kmap_atomic(dst);
L
Linus Torvalds 已提交
2570 2571 2572 2573 2574 2575 2576 2577 2578
		void __user *uaddr = (void __user *)(va & PAGE_MASK);

		/*
		 * This really shouldn't fail, because the page is there
		 * in the page tables. But it might just be unreadable,
		 * in which case we just give up and fill the result with
		 * zeroes.
		 */
		if (__copy_from_user_inatomic(kaddr, uaddr, PAGE_SIZE))
2579
			clear_page(kaddr);
2580
		kunmap_atomic(kaddr);
2581
		flush_dcache_page(dst);
N
Nick Piggin 已提交
2582 2583
	} else
		copy_user_highpage(dst, src, va, vma);
2584 2585
}

L
Linus Torvalds 已提交
2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599
/*
 * This routine handles present pages, when users try to write
 * to a shared page. It is done by copying the page to a new address
 * and decrementing the shared-page counter for the old page.
 *
 * Note that this routine assumes that the protection checks have been
 * done by the caller (the low-level page fault routine in most cases).
 * Thus we can safely just mark it writable once we've done any necessary
 * COW.
 *
 * We also mark the page dirty at this point even though the page will
 * change only once the write actually happens. This avoids a few races,
 * and potentially makes it more efficient.
 *
2600 2601 2602
 * We enter with non-exclusive mmap_sem (to exclude vma changes,
 * but allow concurrent faults), with pte both mapped and locked.
 * We return with mmap_sem still held, but pte unmapped and unlocked.
L
Linus Torvalds 已提交
2603
 */
2604 2605
static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
		unsigned long address, pte_t *page_table, pmd_t *pmd,
2606
		spinlock_t *ptl, pte_t orig_pte)
2607
	__releases(ptl)
L
Linus Torvalds 已提交
2608
{
2609
	struct page *old_page, *new_page = NULL;
L
Linus Torvalds 已提交
2610
	pte_t entry;
2611
	int ret = 0;
2612
	int page_mkwrite = 0;
2613
	struct page *dirty_page = NULL;
2614 2615
	unsigned long mmun_start = 0;	/* For mmu_notifiers */
	unsigned long mmun_end = 0;	/* For mmu_notifiers */
L
Linus Torvalds 已提交
2616

2617
	old_page = vm_normal_page(vma, address, orig_pte);
2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628
	if (!old_page) {
		/*
		 * VM_MIXEDMAP !pfn_valid() case
		 *
		 * We should not cow pages in a shared writeable mapping.
		 * Just mark the pages writable as we can't do any dirty
		 * accounting on raw pfn maps.
		 */
		if ((vma->vm_flags & (VM_WRITE|VM_SHARED)) ==
				     (VM_WRITE|VM_SHARED))
			goto reuse;
2629
		goto gotten;
2630
	}
L
Linus Torvalds 已提交
2631

2632
	/*
P
Peter Zijlstra 已提交
2633 2634
	 * Take out anonymous pages first, anonymous shared vmas are
	 * not dirty accountable.
2635
	 */
H
Hugh Dickins 已提交
2636
	if (PageAnon(old_page) && !PageKsm(old_page)) {
2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647
		if (!trylock_page(old_page)) {
			page_cache_get(old_page);
			pte_unmap_unlock(page_table, ptl);
			lock_page(old_page);
			page_table = pte_offset_map_lock(mm, pmd, address,
							 &ptl);
			if (!pte_same(*page_table, orig_pte)) {
				unlock_page(old_page);
				goto unlock;
			}
			page_cache_release(old_page);
P
Peter Zijlstra 已提交
2648
		}
2649
		if (reuse_swap_page(old_page)) {
2650 2651 2652 2653 2654 2655
			/*
			 * The page is all ours.  Move it to our anon_vma so
			 * the rmap code will not search our parent or siblings.
			 * Protected against the rmap code by the page lock.
			 */
			page_move_anon_rmap(old_page, vma, address);
2656 2657 2658
			unlock_page(old_page);
			goto reuse;
		}
2659
		unlock_page(old_page);
P
Peter Zijlstra 已提交
2660
	} else if (unlikely((vma->vm_flags & (VM_WRITE|VM_SHARED)) ==
2661
					(VM_WRITE|VM_SHARED))) {
P
Peter Zijlstra 已提交
2662 2663 2664 2665 2666
		/*
		 * Only catch write-faults on shared writable pages,
		 * read-only shared pages can get COWed by
		 * get_user_pages(.write=1, .force=1).
		 */
2667
		if (vma->vm_ops && vma->vm_ops->page_mkwrite) {
2668 2669 2670 2671 2672 2673 2674 2675 2676
			struct vm_fault vmf;
			int tmp;

			vmf.virtual_address = (void __user *)(address &
								PAGE_MASK);
			vmf.pgoff = old_page->index;
			vmf.flags = FAULT_FLAG_WRITE|FAULT_FLAG_MKWRITE;
			vmf.page = old_page;

2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687
			/*
			 * Notify the address space that the page is about to
			 * become writable so that it can prohibit this or wait
			 * for the page to get into an appropriate state.
			 *
			 * We do this without the lock held, so that it can
			 * sleep if it needs to.
			 */
			page_cache_get(old_page);
			pte_unmap_unlock(page_table, ptl);

2688 2689 2690 2691
			tmp = vma->vm_ops->page_mkwrite(vma, &vmf);
			if (unlikely(tmp &
					(VM_FAULT_ERROR | VM_FAULT_NOPAGE))) {
				ret = tmp;
2692
				goto unwritable_page;
2693
			}
N
Nick Piggin 已提交
2694 2695 2696 2697 2698 2699 2700 2701 2702
			if (unlikely(!(tmp & VM_FAULT_LOCKED))) {
				lock_page(old_page);
				if (!old_page->mapping) {
					ret = 0; /* retry the fault */
					unlock_page(old_page);
					goto unwritable_page;
				}
			} else
				VM_BUG_ON(!PageLocked(old_page));
2703 2704 2705 2706 2707 2708 2709 2710 2711

			/*
			 * Since we dropped the lock we need to revalidate
			 * the PTE as someone else may have changed it.  If
			 * they did, we just return, as we can count on the
			 * MMU to tell us if they didn't also make it writable.
			 */
			page_table = pte_offset_map_lock(mm, pmd, address,
							 &ptl);
N
Nick Piggin 已提交
2712 2713
			if (!pte_same(*page_table, orig_pte)) {
				unlock_page(old_page);
2714
				goto unlock;
N
Nick Piggin 已提交
2715
			}
2716 2717

			page_mkwrite = 1;
L
Linus Torvalds 已提交
2718
		}
2719 2720
		dirty_page = old_page;
		get_page(dirty_page);
2721

2722
reuse:
2723 2724 2725 2726 2727 2728 2729 2730
		/*
		 * Clear the pages cpupid information as the existing
		 * information potentially belongs to a now completely
		 * unrelated process.
		 */
		if (old_page)
			page_cpupid_xchg_last(old_page, (1 << LAST_CPUPID_SHIFT) - 1);

2731 2732 2733
		flush_cache_page(vma, address, pte_pfn(orig_pte));
		entry = pte_mkyoung(orig_pte);
		entry = maybe_mkwrite(pte_mkdirty(entry), vma);
2734
		if (ptep_set_access_flags(vma, address, page_table, entry,1))
2735
			update_mmu_cache(vma, address, page_table);
2736
		pte_unmap_unlock(page_table, ptl);
2737
		ret |= VM_FAULT_WRITE;
2738 2739 2740 2741 2742 2743 2744 2745 2746 2747

		if (!dirty_page)
			return ret;

		/*
		 * Yes, Virginia, this is actually required to prevent a race
		 * with clear_page_dirty_for_io() from clearing the page dirty
		 * bit after it clear all dirty ptes, but before a racing
		 * do_wp_page installs a dirty pte.
		 *
2748
		 * __do_fault is protected similarly.
2749 2750 2751 2752
		 */
		if (!page_mkwrite) {
			wait_on_page_locked(dirty_page);
			set_page_dirty_balance(dirty_page, page_mkwrite);
2753 2754 2755
			/* file_update_time outside page_lock */
			if (vma->vm_file)
				file_update_time(vma->vm_file);
2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773
		}
		put_page(dirty_page);
		if (page_mkwrite) {
			struct address_space *mapping = dirty_page->mapping;

			set_page_dirty(dirty_page);
			unlock_page(dirty_page);
			page_cache_release(dirty_page);
			if (mapping)	{
				/*
				 * Some device drivers do not set page.mapping
				 * but still dirty their pages
				 */
				balance_dirty_pages_ratelimited(mapping);
			}
		}

		return ret;
L
Linus Torvalds 已提交
2774 2775 2776 2777 2778
	}

	/*
	 * Ok, we need to copy. Oh, well..
	 */
N
Nick Piggin 已提交
2779
	page_cache_get(old_page);
H
Hugh Dickins 已提交
2780
gotten:
2781
	pte_unmap_unlock(page_table, ptl);
L
Linus Torvalds 已提交
2782 2783

	if (unlikely(anon_vma_prepare(vma)))
2784
		goto oom;
H
Hugh Dickins 已提交
2785

H
Hugh Dickins 已提交
2786
	if (is_zero_pfn(pte_pfn(orig_pte))) {
H
Hugh Dickins 已提交
2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797
		new_page = alloc_zeroed_user_highpage_movable(vma, address);
		if (!new_page)
			goto oom;
	} else {
		new_page = alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma, address);
		if (!new_page)
			goto oom;
		cow_user_page(new_page, old_page, address, vma);
	}
	__SetPageUptodate(new_page);

K
KAMEZAWA Hiroyuki 已提交
2798
	if (mem_cgroup_newpage_charge(new_page, mm, GFP_KERNEL))
2799 2800
		goto oom_free_new;

2801
	mmun_start  = address & PAGE_MASK;
2802
	mmun_end    = mmun_start + PAGE_SIZE;
2803 2804
	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);

L
Linus Torvalds 已提交
2805 2806 2807
	/*
	 * Re-check the pte - we dropped the lock
	 */
2808
	page_table = pte_offset_map_lock(mm, pmd, address, &ptl);
2809
	if (likely(pte_same(*page_table, orig_pte))) {
H
Hugh Dickins 已提交
2810 2811
		if (old_page) {
			if (!PageAnon(old_page)) {
2812 2813
				dec_mm_counter_fast(mm, MM_FILEPAGES);
				inc_mm_counter_fast(mm, MM_ANONPAGES);
H
Hugh Dickins 已提交
2814 2815
			}
		} else
2816
			inc_mm_counter_fast(mm, MM_ANONPAGES);
2817
		flush_cache_page(vma, address, pte_pfn(orig_pte));
2818 2819
		entry = mk_pte(new_page, vma->vm_page_prot);
		entry = maybe_mkwrite(pte_mkdirty(entry), vma);
2820 2821 2822 2823 2824 2825
		/*
		 * Clear the pte entry and flush it first, before updating the
		 * pte with the new entry. This will avoid a race condition
		 * seen in the presence of one thread doing SMC and another
		 * thread doing COW.
		 */
2826
		ptep_clear_flush(vma, address, page_table);
N
Nick Piggin 已提交
2827
		page_add_new_anon_rmap(new_page, vma, address);
2828 2829 2830 2831 2832 2833
		/*
		 * We call the notify macro here because, when using secondary
		 * mmu page tables (such as kvm shadow page tables), we want the
		 * new page to be mapped directly into the secondary page table.
		 */
		set_pte_at_notify(mm, address, page_table, entry);
2834
		update_mmu_cache(vma, address, page_table);
N
Nick Piggin 已提交
2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857
		if (old_page) {
			/*
			 * Only after switching the pte to the new page may
			 * we remove the mapcount here. Otherwise another
			 * process may come and find the rmap count decremented
			 * before the pte is switched to the new page, and
			 * "reuse" the old page writing into it while our pte
			 * here still points into it and can be read by other
			 * threads.
			 *
			 * The critical issue is to order this
			 * page_remove_rmap with the ptp_clear_flush above.
			 * Those stores are ordered by (if nothing else,)
			 * the barrier present in the atomic_add_negative
			 * in page_remove_rmap.
			 *
			 * Then the TLB flush in ptep_clear_flush ensures that
			 * no process can access the old page before the
			 * decremented mapcount is visible. And the old page
			 * cannot be reused until after the decremented
			 * mapcount is visible. So transitively, TLBs to
			 * old page will be flushed before it can be reused.
			 */
2858
			page_remove_rmap(old_page);
N
Nick Piggin 已提交
2859 2860
		}

L
Linus Torvalds 已提交
2861 2862
		/* Free the old page.. */
		new_page = old_page;
N
Nick Piggin 已提交
2863
		ret |= VM_FAULT_WRITE;
2864 2865 2866
	} else
		mem_cgroup_uncharge_page(new_page);

2867 2868
	if (new_page)
		page_cache_release(new_page);
2869
unlock:
2870
	pte_unmap_unlock(page_table, ptl);
2871
	if (mmun_end > mmun_start)
2872
		mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884
	if (old_page) {
		/*
		 * Don't let another task, with possibly unlocked vma,
		 * keep the mlocked page.
		 */
		if ((ret & VM_FAULT_WRITE) && (vma->vm_flags & VM_LOCKED)) {
			lock_page(old_page);	/* LRU manipulation */
			munlock_vma_page(old_page);
			unlock_page(old_page);
		}
		page_cache_release(old_page);
	}
N
Nick Piggin 已提交
2885
	return ret;
2886
oom_free_new:
2887
	page_cache_release(new_page);
2888
oom:
2889
	if (old_page)
H
Hugh Dickins 已提交
2890
		page_cache_release(old_page);
L
Linus Torvalds 已提交
2891
	return VM_FAULT_OOM;
2892 2893 2894

unwritable_page:
	page_cache_release(old_page);
2895
	return ret;
L
Linus Torvalds 已提交
2896 2897
}

2898
static void unmap_mapping_range_vma(struct vm_area_struct *vma,
L
Linus Torvalds 已提交
2899 2900 2901
		unsigned long start_addr, unsigned long end_addr,
		struct zap_details *details)
{
2902
	zap_page_range_single(vma, start_addr, end_addr - start_addr, details);
L
Linus Torvalds 已提交
2903 2904
}

2905
static inline void unmap_mapping_range_tree(struct rb_root *root,
L
Linus Torvalds 已提交
2906 2907 2908 2909 2910
					    struct zap_details *details)
{
	struct vm_area_struct *vma;
	pgoff_t vba, vea, zba, zea;

2911
	vma_interval_tree_foreach(vma, root,
L
Linus Torvalds 已提交
2912 2913 2914
			details->first_index, details->last_index) {

		vba = vma->vm_pgoff;
2915
		vea = vba + vma_pages(vma) - 1;
L
Linus Torvalds 已提交
2916 2917 2918 2919 2920 2921 2922 2923
		/* Assume for now that PAGE_CACHE_SHIFT == PAGE_SHIFT */
		zba = details->first_index;
		if (zba < vba)
			zba = vba;
		zea = details->last_index;
		if (zea > vea)
			zea = vea;

2924
		unmap_mapping_range_vma(vma,
L
Linus Torvalds 已提交
2925 2926
			((zba - vba) << PAGE_SHIFT) + vma->vm_start,
			((zea - vba + 1) << PAGE_SHIFT) + vma->vm_start,
2927
				details);
L
Linus Torvalds 已提交
2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941
	}
}

static inline void unmap_mapping_range_list(struct list_head *head,
					    struct zap_details *details)
{
	struct vm_area_struct *vma;

	/*
	 * In nonlinear VMAs there is no correspondence between virtual address
	 * offset and file offset.  So we must perform an exhaustive search
	 * across *all* the pages in each nonlinear VMA, not just the pages
	 * whose virtual address lies outside the file truncation point.
	 */
2942
	list_for_each_entry(vma, head, shared.nonlinear) {
L
Linus Torvalds 已提交
2943
		details->nonlinear_vma = vma;
2944
		unmap_mapping_range_vma(vma, vma->vm_start, vma->vm_end, details);
L
Linus Torvalds 已提交
2945 2946 2947 2948
	}
}

/**
2949
 * unmap_mapping_range - unmap the portion of all mmaps in the specified address_space corresponding to the specified page range in the underlying file.
M
Martin Waitz 已提交
2950
 * @mapping: the address space containing mmaps to be unmapped.
L
Linus Torvalds 已提交
2951 2952
 * @holebegin: byte in first page to unmap, relative to the start of
 * the underlying file.  This will be rounded down to a PAGE_SIZE
N
npiggin@suse.de 已提交
2953
 * boundary.  Note that this is different from truncate_pagecache(), which
L
Linus Torvalds 已提交
2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984
 * must keep the partial page.  In contrast, we must get rid of
 * partial pages.
 * @holelen: size of prospective hole in bytes.  This will be rounded
 * up to a PAGE_SIZE boundary.  A holelen of zero truncates to the
 * end of the file.
 * @even_cows: 1 when truncating a file, unmap even private COWed pages;
 * but 0 when invalidating pagecache, don't throw away private data.
 */
void unmap_mapping_range(struct address_space *mapping,
		loff_t const holebegin, loff_t const holelen, int even_cows)
{
	struct zap_details details;
	pgoff_t hba = holebegin >> PAGE_SHIFT;
	pgoff_t hlen = (holelen + PAGE_SIZE - 1) >> PAGE_SHIFT;

	/* Check for overflow. */
	if (sizeof(holelen) > sizeof(hlen)) {
		long long holeend =
			(holebegin + holelen + PAGE_SIZE - 1) >> PAGE_SHIFT;
		if (holeend & ~(long long)ULONG_MAX)
			hlen = ULONG_MAX - hba + 1;
	}

	details.check_mapping = even_cows? NULL: mapping;
	details.nonlinear_vma = NULL;
	details.first_index = hba;
	details.last_index = hba + hlen - 1;
	if (details.last_index < details.first_index)
		details.last_index = ULONG_MAX;


2985
	mutex_lock(&mapping->i_mmap_mutex);
2986
	if (unlikely(!RB_EMPTY_ROOT(&mapping->i_mmap)))
L
Linus Torvalds 已提交
2987 2988 2989
		unmap_mapping_range_tree(&mapping->i_mmap, &details);
	if (unlikely(!list_empty(&mapping->i_mmap_nonlinear)))
		unmap_mapping_range_list(&mapping->i_mmap_nonlinear, &details);
2990
	mutex_unlock(&mapping->i_mmap_mutex);
L
Linus Torvalds 已提交
2991 2992 2993 2994
}
EXPORT_SYMBOL(unmap_mapping_range);

/*
2995 2996 2997
 * We enter with non-exclusive mmap_sem (to exclude vma changes,
 * but allow concurrent faults), and pte mapped but not yet locked.
 * We return with mmap_sem still held, but pte unmapped and unlocked.
L
Linus Torvalds 已提交
2998
 */
2999 3000
static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
		unsigned long address, pte_t *page_table, pmd_t *pmd,
3001
		unsigned int flags, pte_t orig_pte)
L
Linus Torvalds 已提交
3002
{
3003
	spinlock_t *ptl;
3004
	struct page *page, *swapcache;
3005
	swp_entry_t entry;
L
Linus Torvalds 已提交
3006
	pte_t pte;
3007
	int locked;
3008
	struct mem_cgroup *ptr;
3009
	int exclusive = 0;
N
Nick Piggin 已提交
3010
	int ret = 0;
L
Linus Torvalds 已提交
3011

H
Hugh Dickins 已提交
3012
	if (!pte_unmap_same(mm, pmd, page_table, orig_pte))
3013
		goto out;
3014 3015

	entry = pte_to_swp_entry(orig_pte);
3016 3017 3018 3019 3020 3021 3022
	if (unlikely(non_swap_entry(entry))) {
		if (is_migration_entry(entry)) {
			migration_entry_wait(mm, pmd, address);
		} else if (is_hwpoison_entry(entry)) {
			ret = VM_FAULT_HWPOISON;
		} else {
			print_bad_pte(vma, address, orig_pte, NULL);
H
Hugh Dickins 已提交
3023
			ret = VM_FAULT_SIGBUS;
3024
		}
3025 3026
		goto out;
	}
3027
	delayacct_set_flag(DELAYACCT_PF_SWAPIN);
L
Linus Torvalds 已提交
3028 3029
	page = lookup_swap_cache(entry);
	if (!page) {
3030 3031
		page = swapin_readahead(entry,
					GFP_HIGHUSER_MOVABLE, vma, address);
L
Linus Torvalds 已提交
3032 3033
		if (!page) {
			/*
3034 3035
			 * Back out if somebody else faulted in this pte
			 * while we released the pte lock.
L
Linus Torvalds 已提交
3036
			 */
3037
			page_table = pte_offset_map_lock(mm, pmd, address, &ptl);
L
Linus Torvalds 已提交
3038 3039
			if (likely(pte_same(*page_table, orig_pte)))
				ret = VM_FAULT_OOM;
3040
			delayacct_clear_flag(DELAYACCT_PF_SWAPIN);
3041
			goto unlock;
L
Linus Torvalds 已提交
3042 3043 3044 3045
		}

		/* Had to read the page from swap area: Major fault */
		ret = VM_FAULT_MAJOR;
3046
		count_vm_event(PGMAJFAULT);
3047
		mem_cgroup_count_vm_event(mm, PGMAJFAULT);
3048
	} else if (PageHWPoison(page)) {
3049 3050 3051 3052
		/*
		 * hwpoisoned dirty swapcache pages are kept for killing
		 * owner processes (which may be unknown at hwpoison time)
		 */
3053 3054
		ret = VM_FAULT_HWPOISON;
		delayacct_clear_flag(DELAYACCT_PF_SWAPIN);
3055
		swapcache = page;
3056
		goto out_release;
L
Linus Torvalds 已提交
3057 3058
	}

3059
	swapcache = page;
3060
	locked = lock_page_or_retry(page, mm, flags);
R
Rik van Riel 已提交
3061

3062
	delayacct_clear_flag(DELAYACCT_PF_SWAPIN);
3063 3064 3065 3066
	if (!locked) {
		ret |= VM_FAULT_RETRY;
		goto out_release;
	}
3067

A
Andrea Arcangeli 已提交
3068
	/*
3069 3070 3071 3072
	 * Make sure try_to_free_swap or reuse_swap_page or swapoff did not
	 * release the swapcache from under us.  The page pin, and pte_same
	 * test below, are not enough to exclude that.  Even if it is still
	 * swapcache, we need to check that the page's swap has not changed.
A
Andrea Arcangeli 已提交
3073
	 */
3074
	if (unlikely(!PageSwapCache(page) || page_private(page) != entry.val))
A
Andrea Arcangeli 已提交
3075 3076
		goto out_page;

3077 3078 3079 3080 3081
	page = ksm_might_need_to_copy(page, vma, address);
	if (unlikely(!page)) {
		ret = VM_FAULT_OOM;
		page = swapcache;
		goto out_page;
H
Hugh Dickins 已提交
3082 3083
	}

K
KAMEZAWA Hiroyuki 已提交
3084
	if (mem_cgroup_try_charge_swapin(mm, page, GFP_KERNEL, &ptr)) {
3085
		ret = VM_FAULT_OOM;
3086
		goto out_page;
3087 3088
	}

L
Linus Torvalds 已提交
3089
	/*
3090
	 * Back out if somebody else already faulted in this pte.
L
Linus Torvalds 已提交
3091
	 */
3092
	page_table = pte_offset_map_lock(mm, pmd, address, &ptl);
H
Hugh Dickins 已提交
3093
	if (unlikely(!pte_same(*page_table, orig_pte)))
3094 3095 3096 3097 3098
		goto out_nomap;

	if (unlikely(!PageUptodate(page))) {
		ret = VM_FAULT_SIGBUS;
		goto out_nomap;
L
Linus Torvalds 已提交
3099 3100
	}

3101 3102 3103 3104 3105 3106 3107 3108
	/*
	 * The page isn't present yet, go ahead with the fault.
	 *
	 * Be careful about the sequence of operations here.
	 * To get its accounting right, reuse_swap_page() must be called
	 * while the page is counted on swap but not yet in mapcount i.e.
	 * before page_add_anon_rmap() and swap_free(); try_to_free_swap()
	 * must be called after the swap_free(), or it will never succeed.
3109 3110 3111 3112
	 * Because delete_from_swap_page() may be called by reuse_swap_page(),
	 * mem_cgroup_commit_charge_swapin() may not be able to find swp_entry
	 * in page->private. In this case, a record in swap_cgroup  is silently
	 * discarded at swap_free().
3113
	 */
L
Linus Torvalds 已提交
3114

3115
	inc_mm_counter_fast(mm, MM_ANONPAGES);
K
KAMEZAWA Hiroyuki 已提交
3116
	dec_mm_counter_fast(mm, MM_SWAPENTS);
L
Linus Torvalds 已提交
3117
	pte = mk_pte(page, vma->vm_page_prot);
3118
	if ((flags & FAULT_FLAG_WRITE) && reuse_swap_page(page)) {
L
Linus Torvalds 已提交
3119
		pte = maybe_mkwrite(pte_mkdirty(pte), vma);
3120
		flags &= ~FAULT_FLAG_WRITE;
3121
		ret |= VM_FAULT_WRITE;
3122
		exclusive = 1;
L
Linus Torvalds 已提交
3123 3124
	}
	flush_icache_page(vma, page);
3125 3126
	if (pte_swp_soft_dirty(orig_pte))
		pte = pte_mksoft_dirty(pte);
L
Linus Torvalds 已提交
3127
	set_pte_at(mm, address, page_table, pte);
3128
	if (page == swapcache)
3129
		do_page_add_anon_rmap(page, vma, address, exclusive);
3130 3131
	else /* ksm created a completely new copy */
		page_add_new_anon_rmap(page, vma, address);
3132 3133
	/* It's better to call commit-charge after rmap is established */
	mem_cgroup_commit_charge_swapin(page, ptr);
L
Linus Torvalds 已提交
3134

3135
	swap_free(entry);
N
Nick Piggin 已提交
3136
	if (vm_swap_full() || (vma->vm_flags & VM_LOCKED) || PageMlocked(page))
3137
		try_to_free_swap(page);
3138
	unlock_page(page);
3139
	if (page != swapcache) {
A
Andrea Arcangeli 已提交
3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150
		/*
		 * Hold the lock to avoid the swap entry to be reused
		 * until we take the PT lock for the pte_same() check
		 * (to avoid false positives from pte_same). For
		 * further safety release the lock after the swap_free
		 * so that the swap count won't change under a
		 * parallel locked swapcache.
		 */
		unlock_page(swapcache);
		page_cache_release(swapcache);
	}
3151

3152
	if (flags & FAULT_FLAG_WRITE) {
3153 3154 3155
		ret |= do_wp_page(mm, vma, address, page_table, pmd, ptl, pte);
		if (ret & VM_FAULT_ERROR)
			ret &= VM_FAULT_ERROR;
L
Linus Torvalds 已提交
3156 3157 3158 3159
		goto out;
	}

	/* No need to invalidate - it was non-present before */
3160
	update_mmu_cache(vma, address, page_table);
3161
unlock:
3162
	pte_unmap_unlock(page_table, ptl);
L
Linus Torvalds 已提交
3163 3164
out:
	return ret;
3165
out_nomap:
3166
	mem_cgroup_cancel_charge_swapin(ptr);
3167
	pte_unmap_unlock(page_table, ptl);
3168
out_page:
3169
	unlock_page(page);
3170
out_release:
3171
	page_cache_release(page);
3172
	if (page != swapcache) {
A
Andrea Arcangeli 已提交
3173 3174 3175
		unlock_page(swapcache);
		page_cache_release(swapcache);
	}
3176
	return ret;
L
Linus Torvalds 已提交
3177 3178
}

3179
/*
3180 3181
 * This is like a special single-page "expand_{down|up}wards()",
 * except we must first make sure that 'address{-|+}PAGE_SIZE'
3182 3183 3184 3185 3186 3187
 * doesn't hit another vma.
 */
static inline int check_stack_guard_page(struct vm_area_struct *vma, unsigned long address)
{
	address &= PAGE_MASK;
	if ((vma->vm_flags & VM_GROWSDOWN) && address == vma->vm_start) {
3188 3189 3190 3191 3192 3193 3194 3195 3196 3197
		struct vm_area_struct *prev = vma->vm_prev;

		/*
		 * Is there a mapping abutting this one below?
		 *
		 * That's only ok if it's the same stack mapping
		 * that has gotten split..
		 */
		if (prev && prev->vm_end == address)
			return prev->vm_flags & VM_GROWSDOWN ? 0 : -ENOMEM;
3198

3199
		expand_downwards(vma, address - PAGE_SIZE);
3200
	}
3201 3202 3203 3204 3205 3206 3207 3208 3209
	if ((vma->vm_flags & VM_GROWSUP) && address + PAGE_SIZE == vma->vm_end) {
		struct vm_area_struct *next = vma->vm_next;

		/* As VM_GROWSDOWN but s/below/above/ */
		if (next && next->vm_start == address + PAGE_SIZE)
			return next->vm_flags & VM_GROWSUP ? 0 : -ENOMEM;

		expand_upwards(vma, address + PAGE_SIZE);
	}
3210 3211 3212
	return 0;
}

L
Linus Torvalds 已提交
3213
/*
3214 3215 3216
 * We enter with non-exclusive mmap_sem (to exclude vma changes,
 * but allow concurrent faults), and pte mapped but not yet locked.
 * We return with mmap_sem still held, but pte unmapped and unlocked.
L
Linus Torvalds 已提交
3217
 */
3218 3219
static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
		unsigned long address, pte_t *page_table, pmd_t *pmd,
3220
		unsigned int flags)
L
Linus Torvalds 已提交
3221
{
3222 3223
	struct page *page;
	spinlock_t *ptl;
L
Linus Torvalds 已提交
3224 3225
	pte_t entry;

3226 3227 3228 3229
	pte_unmap(page_table);

	/* Check if we need to add a guard page to the stack */
	if (check_stack_guard_page(vma, address) < 0)
3230 3231
		return VM_FAULT_SIGBUS;

3232
	/* Use the zero-page for reads */
H
Hugh Dickins 已提交
3233 3234 3235
	if (!(flags & FAULT_FLAG_WRITE)) {
		entry = pte_mkspecial(pfn_pte(my_zero_pfn(address),
						vma->vm_page_prot));
3236
		page_table = pte_offset_map_lock(mm, pmd, address, &ptl);
H
Hugh Dickins 已提交
3237 3238 3239 3240 3241
		if (!pte_none(*page_table))
			goto unlock;
		goto setpte;
	}

N
Nick Piggin 已提交
3242 3243 3244 3245 3246 3247
	/* Allocate our own private page. */
	if (unlikely(anon_vma_prepare(vma)))
		goto oom;
	page = alloc_zeroed_user_highpage_movable(vma, address);
	if (!page)
		goto oom;
3248 3249 3250 3251 3252
	/*
	 * The memory barrier inside __SetPageUptodate makes sure that
	 * preceeding stores to the page contents become visible before
	 * the set_pte_at() write.
	 */
N
Nick Piggin 已提交
3253
	__SetPageUptodate(page);
3254

K
KAMEZAWA Hiroyuki 已提交
3255
	if (mem_cgroup_newpage_charge(page, mm, GFP_KERNEL))
3256 3257
		goto oom_free_page;

N
Nick Piggin 已提交
3258
	entry = mk_pte(page, vma->vm_page_prot);
H
Hugh Dickins 已提交
3259 3260
	if (vma->vm_flags & VM_WRITE)
		entry = pte_mkwrite(pte_mkdirty(entry));
L
Linus Torvalds 已提交
3261

N
Nick Piggin 已提交
3262
	page_table = pte_offset_map_lock(mm, pmd, address, &ptl);
3263
	if (!pte_none(*page_table))
N
Nick Piggin 已提交
3264
		goto release;
H
Hugh Dickins 已提交
3265

3266
	inc_mm_counter_fast(mm, MM_ANONPAGES);
N
Nick Piggin 已提交
3267
	page_add_new_anon_rmap(page, vma, address);
H
Hugh Dickins 已提交
3268
setpte:
3269
	set_pte_at(mm, address, page_table, entry);
L
Linus Torvalds 已提交
3270 3271

	/* No need to invalidate - it was non-present before */
3272
	update_mmu_cache(vma, address, page_table);
3273
unlock:
3274
	pte_unmap_unlock(page_table, ptl);
N
Nick Piggin 已提交
3275
	return 0;
3276
release:
3277
	mem_cgroup_uncharge_page(page);
3278 3279
	page_cache_release(page);
	goto unlock;
3280
oom_free_page:
3281
	page_cache_release(page);
3282
oom:
L
Linus Torvalds 已提交
3283 3284 3285 3286
	return VM_FAULT_OOM;
}

/*
3287
 * __do_fault() tries to create a new page mapping. It aggressively
L
Linus Torvalds 已提交
3288
 * tries to share with existing pages, but makes a separate copy if
3289 3290
 * the FAULT_FLAG_WRITE is set in the flags parameter in order to avoid
 * the next page fault.
L
Linus Torvalds 已提交
3291 3292 3293 3294
 *
 * As this is called only for pages that do not currently exist, we
 * do not need to flush old virtual caches or the TLB.
 *
3295
 * We enter with non-exclusive mmap_sem (to exclude vma changes,
3296
 * but allow concurrent faults), and pte neither mapped nor locked.
3297
 * We return with mmap_sem still held, but pte unmapped and unlocked.
L
Linus Torvalds 已提交
3298
 */
3299
static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma,
3300
		unsigned long address, pmd_t *pmd,
3301
		pgoff_t pgoff, unsigned int flags, pte_t orig_pte)
L
Linus Torvalds 已提交
3302
{
3303
	pte_t *page_table;
3304
	spinlock_t *ptl;
N
Nick Piggin 已提交
3305
	struct page *page;
3306
	struct page *cow_page;
L
Linus Torvalds 已提交
3307 3308
	pte_t entry;
	int anon = 0;
3309
	struct page *dirty_page = NULL;
N
Nick Piggin 已提交
3310 3311
	struct vm_fault vmf;
	int ret;
3312
	int page_mkwrite = 0;
3313

3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333
	/*
	 * If we do COW later, allocate page befor taking lock_page()
	 * on the file cache page. This will reduce lock holding time.
	 */
	if ((flags & FAULT_FLAG_WRITE) && !(vma->vm_flags & VM_SHARED)) {

		if (unlikely(anon_vma_prepare(vma)))
			return VM_FAULT_OOM;

		cow_page = alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma, address);
		if (!cow_page)
			return VM_FAULT_OOM;

		if (mem_cgroup_newpage_charge(cow_page, mm, GFP_KERNEL)) {
			page_cache_release(cow_page);
			return VM_FAULT_OOM;
		}
	} else
		cow_page = NULL;

N
Nick Piggin 已提交
3334 3335 3336 3337
	vmf.virtual_address = (void __user *)(address & PAGE_MASK);
	vmf.pgoff = pgoff;
	vmf.flags = flags;
	vmf.page = NULL;
L
Linus Torvalds 已提交
3338

N
Nick Piggin 已提交
3339
	ret = vma->vm_ops->fault(vma, &vmf);
3340 3341
	if (unlikely(ret & (VM_FAULT_ERROR | VM_FAULT_NOPAGE |
			    VM_FAULT_RETRY)))
3342
		goto uncharge_out;
L
Linus Torvalds 已提交
3343

3344 3345 3346
	if (unlikely(PageHWPoison(vmf.page))) {
		if (ret & VM_FAULT_LOCKED)
			unlock_page(vmf.page);
3347 3348
		ret = VM_FAULT_HWPOISON;
		goto uncharge_out;
3349 3350
	}

3351
	/*
N
Nick Piggin 已提交
3352
	 * For consistency in subsequent calls, make the faulted page always
3353 3354
	 * locked.
	 */
N
Nick Piggin 已提交
3355
	if (unlikely(!(ret & VM_FAULT_LOCKED)))
N
Nick Piggin 已提交
3356
		lock_page(vmf.page);
3357
	else
N
Nick Piggin 已提交
3358
		VM_BUG_ON(!PageLocked(vmf.page));
3359

L
Linus Torvalds 已提交
3360 3361 3362
	/*
	 * Should we do an early C-O-W break?
	 */
N
Nick Piggin 已提交
3363
	page = vmf.page;
3364
	if (flags & FAULT_FLAG_WRITE) {
3365
		if (!(vma->vm_flags & VM_SHARED)) {
3366
			page = cow_page;
3367
			anon = 1;
N
Nick Piggin 已提交
3368
			copy_user_highpage(page, vmf.page, address, vma);
N
Nick Piggin 已提交
3369
			__SetPageUptodate(page);
3370
		} else {
3371 3372
			/*
			 * If the page will be shareable, see if the backing
3373
			 * address space wants to know that the page is about
3374 3375
			 * to become writable
			 */
3376
			if (vma->vm_ops->page_mkwrite) {
3377 3378
				int tmp;

3379
				unlock_page(page);
N
Nick Piggin 已提交
3380
				vmf.flags = FAULT_FLAG_WRITE|FAULT_FLAG_MKWRITE;
3381 3382 3383 3384
				tmp = vma->vm_ops->page_mkwrite(vma, &vmf);
				if (unlikely(tmp &
					  (VM_FAULT_ERROR | VM_FAULT_NOPAGE))) {
					ret = tmp;
N
Nick Piggin 已提交
3385
					goto unwritable_page;
N
Nick Piggin 已提交
3386
				}
N
Nick Piggin 已提交
3387 3388 3389 3390 3391 3392 3393 3394 3395
				if (unlikely(!(tmp & VM_FAULT_LOCKED))) {
					lock_page(page);
					if (!page->mapping) {
						ret = 0; /* retry the fault */
						unlock_page(page);
						goto unwritable_page;
					}
				} else
					VM_BUG_ON(!PageLocked(page));
3396
				page_mkwrite = 1;
3397 3398
			}
		}
3399

L
Linus Torvalds 已提交
3400 3401
	}

3402
	page_table = pte_offset_map_lock(mm, pmd, address, &ptl);
L
Linus Torvalds 已提交
3403 3404 3405 3406 3407 3408

	/*
	 * This silly early PAGE_DIRTY setting removes a race
	 * due to the bad i386 page protection. But it's valid
	 * for other architectures too.
	 *
3409
	 * Note that if FAULT_FLAG_WRITE is set, we either now have
L
Linus Torvalds 已提交
3410 3411 3412 3413 3414
	 * an exclusive copy of the page, or this is a shared mapping,
	 * so we can make it writable and dirty to avoid having to
	 * handle that later.
	 */
	/* Only go through if we didn't race with anybody else... */
3415
	if (likely(pte_same(*page_table, orig_pte))) {
3416 3417
		flush_icache_page(vma, page);
		entry = mk_pte(page, vma->vm_page_prot);
3418
		if (flags & FAULT_FLAG_WRITE)
L
Linus Torvalds 已提交
3419
			entry = maybe_mkwrite(pte_mkdirty(entry), vma);
3420 3421
		else if (pte_file(orig_pte) && pte_file_soft_dirty(orig_pte))
			pte_mksoft_dirty(entry);
L
Linus Torvalds 已提交
3422
		if (anon) {
3423
			inc_mm_counter_fast(mm, MM_ANONPAGES);
3424
			page_add_new_anon_rmap(page, vma, address);
3425
		} else {
3426
			inc_mm_counter_fast(mm, MM_FILEPAGES);
3427
			page_add_file_rmap(page);
3428
			if (flags & FAULT_FLAG_WRITE) {
3429
				dirty_page = page;
3430 3431
				get_page(dirty_page);
			}
3432
		}
3433
		set_pte_at(mm, address, page_table, entry);
3434 3435

		/* no need to invalidate: a not-present page won't be cached */
3436
		update_mmu_cache(vma, address, page_table);
L
Linus Torvalds 已提交
3437
	} else {
3438 3439
		if (cow_page)
			mem_cgroup_uncharge_page(cow_page);
3440 3441 3442
		if (anon)
			page_cache_release(page);
		else
3443
			anon = 1; /* no anon but release faulted_page */
L
Linus Torvalds 已提交
3444 3445
	}

3446
	pte_unmap_unlock(page_table, ptl);
3447

N
Nick Piggin 已提交
3448 3449
	if (dirty_page) {
		struct address_space *mapping = page->mapping;
3450
		int dirtied = 0;
3451

N
Nick Piggin 已提交
3452
		if (set_page_dirty(dirty_page))
3453
			dirtied = 1;
N
Nick Piggin 已提交
3454
		unlock_page(dirty_page);
3455
		put_page(dirty_page);
3456
		if ((dirtied || page_mkwrite) && mapping) {
N
Nick Piggin 已提交
3457 3458 3459 3460 3461 3462 3463 3464
			/*
			 * Some device drivers do not set page.mapping but still
			 * dirty their pages
			 */
			balance_dirty_pages_ratelimited(mapping);
		}

		/* file_update_time outside page_lock */
3465
		if (vma->vm_file && !page_mkwrite)
N
Nick Piggin 已提交
3466 3467 3468 3469 3470
			file_update_time(vma->vm_file);
	} else {
		unlock_page(vmf.page);
		if (anon)
			page_cache_release(vmf.page);
3471
	}
3472

N
Nick Piggin 已提交
3473
	return ret;
N
Nick Piggin 已提交
3474 3475 3476 3477

unwritable_page:
	page_cache_release(page);
	return ret;
3478 3479 3480 3481 3482 3483 3484
uncharge_out:
	/* fs's fault handler get error */
	if (cow_page) {
		mem_cgroup_uncharge_page(cow_page);
		page_cache_release(cow_page);
	}
	return ret;
3485
}
3486

3487 3488
static int do_linear_fault(struct mm_struct *mm, struct vm_area_struct *vma,
		unsigned long address, pte_t *page_table, pmd_t *pmd,
3489
		unsigned int flags, pte_t orig_pte)
3490 3491
{
	pgoff_t pgoff = (((address & PAGE_MASK)
3492
			- vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff;
3493

3494 3495
	pte_unmap(page_table);
	return __do_fault(mm, vma, address, pmd, pgoff, flags, orig_pte);
3496 3497
}

L
Linus Torvalds 已提交
3498 3499 3500 3501
/*
 * Fault of a previously existing named mapping. Repopulate the pte
 * from the encoded file_pte if possible. This enables swappable
 * nonlinear vmas.
3502 3503 3504 3505
 *
 * We enter with non-exclusive mmap_sem (to exclude vma changes,
 * but allow concurrent faults), and pte mapped but not yet locked.
 * We return with mmap_sem still held, but pte unmapped and unlocked.
L
Linus Torvalds 已提交
3506
 */
N
Nick Piggin 已提交
3507
static int do_nonlinear_fault(struct mm_struct *mm, struct vm_area_struct *vma,
3508
		unsigned long address, pte_t *page_table, pmd_t *pmd,
3509
		unsigned int flags, pte_t orig_pte)
L
Linus Torvalds 已提交
3510
{
3511
	pgoff_t pgoff;
L
Linus Torvalds 已提交
3512

3513 3514
	flags |= FAULT_FLAG_NONLINEAR;

H
Hugh Dickins 已提交
3515
	if (!pte_unmap_same(mm, pmd, page_table, orig_pte))
N
Nick Piggin 已提交
3516
		return 0;
L
Linus Torvalds 已提交
3517

3518
	if (unlikely(!(vma->vm_flags & VM_NONLINEAR))) {
3519 3520 3521
		/*
		 * Page table corrupted: show pte and kill process.
		 */
3522
		print_bad_pte(vma, address, orig_pte, NULL);
H
Hugh Dickins 已提交
3523
		return VM_FAULT_SIGBUS;
3524 3525 3526
	}

	pgoff = pte_to_pgoff(orig_pte);
3527
	return __do_fault(mm, vma, address, pmd, pgoff, flags, orig_pte);
L
Linus Torvalds 已提交
3528 3529
}

3530
int numa_migrate_prep(struct page *page, struct vm_area_struct *vma,
3531 3532
				unsigned long addr, int page_nid,
				int *flags)
3533 3534 3535 3536
{
	get_page(page);

	count_vm_numa_event(NUMA_HINT_FAULTS);
3537
	if (page_nid == numa_node_id()) {
3538
		count_vm_numa_event(NUMA_HINT_FAULTS_LOCAL);
3539 3540
		*flags |= TNF_FAULT_LOCAL;
	}
3541 3542 3543 3544

	return mpol_misplaced(page, vma, addr);
}

3545 3546 3547
int do_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
		   unsigned long addr, pte_t pte, pte_t *ptep, pmd_t *pmd)
{
3548
	struct page *page = NULL;
3549
	spinlock_t *ptl;
3550
	int page_nid = -1;
3551
	int last_cpupid;
3552
	int target_nid;
3553
	bool migrated = false;
3554
	int flags = 0;
3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566

	/*
	* The "pte" at this point cannot be used safely without
	* validation through pte_unmap_same(). It's of NUMA type but
	* the pfn may be screwed if the read is non atomic.
	*
	* ptep_modify_prot_start is not called as this is clearing
	* the _PAGE_NUMA bit and it is not really expected that there
	* would be concurrent hardware modifications to the PTE.
	*/
	ptl = pte_lockptr(mm, pmd);
	spin_lock(ptl);
3567 3568 3569 3570 3571
	if (unlikely(!pte_same(*ptep, pte))) {
		pte_unmap_unlock(ptep, ptl);
		goto out;
	}

3572 3573 3574 3575 3576 3577 3578 3579 3580
	pte = pte_mknonnuma(pte);
	set_pte_at(mm, addr, ptep, pte);
	update_mmu_cache(vma, addr, ptep);

	page = vm_normal_page(vma, addr, pte);
	if (!page) {
		pte_unmap_unlock(ptep, ptl);
		return 0;
	}
3581
	BUG_ON(is_zero_pfn(page_to_pfn(page)));
3582

3583 3584 3585 3586 3587 3588 3589 3590
	/*
	 * Avoid grouping on DSO/COW pages in specific and RO pages
	 * in general, RO pages shouldn't hurt as much anyway since
	 * they can be in shared cache state.
	 */
	if (!pte_write(pte))
		flags |= TNF_NO_GROUP;

3591 3592 3593 3594 3595 3596 3597
	/*
	 * Flag if the page is shared between multiple address spaces. This
	 * is later used when determining whether to group tasks together
	 */
	if (page_mapcount(page) > 1 && (vma->vm_flags & VM_SHARED))
		flags |= TNF_SHARED;

3598
	last_cpupid = page_cpupid_last(page);
3599
	page_nid = page_to_nid(page);
3600
	target_nid = numa_migrate_prep(page, vma, addr, page_nid, &flags);
3601
	pte_unmap_unlock(ptep, ptl);
3602 3603 3604 3605 3606 3607
	if (target_nid == -1) {
		put_page(page);
		goto out;
	}

	/* Migrate to the requested node */
3608
	migrated = migrate_misplaced_page(page, vma, target_nid);
3609
	if (migrated) {
3610
		page_nid = target_nid;
3611 3612
		flags |= TNF_MIGRATED;
	}
3613 3614

out:
3615
	if (page_nid != -1)
3616
		task_numa_fault(last_cpupid, page_nid, 1, flags);
3617 3618 3619
	return 0;
}

L
Linus Torvalds 已提交
3620 3621 3622 3623 3624 3625 3626 3627 3628
/*
 * These routines also need to handle stuff like marking pages dirty
 * and/or accessed for architectures that don't do it in hardware (most
 * RISC architectures).  The early dirtying is also good on the i386.
 *
 * There is also a hook called "update_mmu_cache()" that architectures
 * with external mmu caches can use to update those (ie the Sparc or
 * PowerPC hashed page tables that act as extended TLBs).
 *
H
Hugh Dickins 已提交
3629 3630 3631
 * We enter with non-exclusive mmap_sem (to exclude vma changes,
 * but allow concurrent faults), and pte mapped but not yet locked.
 * We return with mmap_sem still held, but pte unmapped and unlocked.
L
Linus Torvalds 已提交
3632
 */
3633
static int handle_pte_fault(struct mm_struct *mm,
3634 3635
		     struct vm_area_struct *vma, unsigned long address,
		     pte_t *pte, pmd_t *pmd, unsigned int flags)
L
Linus Torvalds 已提交
3636 3637
{
	pte_t entry;
3638
	spinlock_t *ptl;
L
Linus Torvalds 已提交
3639

3640
	entry = *pte;
L
Linus Torvalds 已提交
3641
	if (!pte_present(entry)) {
3642
		if (pte_none(entry)) {
J
Jes Sorensen 已提交
3643
			if (vma->vm_ops) {
N
Nick Piggin 已提交
3644
				if (likely(vma->vm_ops->fault))
3645
					return do_linear_fault(mm, vma, address,
3646
						pte, pmd, flags, entry);
J
Jes Sorensen 已提交
3647 3648
			}
			return do_anonymous_page(mm, vma, address,
3649
						 pte, pmd, flags);
3650
		}
L
Linus Torvalds 已提交
3651
		if (pte_file(entry))
N
Nick Piggin 已提交
3652
			return do_nonlinear_fault(mm, vma, address,
3653
					pte, pmd, flags, entry);
3654
		return do_swap_page(mm, vma, address,
3655
					pte, pmd, flags, entry);
L
Linus Torvalds 已提交
3656 3657
	}

3658 3659 3660
	if (pte_numa(entry))
		return do_numa_page(mm, vma, address, entry, pte, pmd);

H
Hugh Dickins 已提交
3661
	ptl = pte_lockptr(mm, pmd);
3662 3663 3664
	spin_lock(ptl);
	if (unlikely(!pte_same(*pte, entry)))
		goto unlock;
3665
	if (flags & FAULT_FLAG_WRITE) {
L
Linus Torvalds 已提交
3666
		if (!pte_write(entry))
3667 3668
			return do_wp_page(mm, vma, address,
					pte, pmd, ptl, entry);
L
Linus Torvalds 已提交
3669 3670 3671
		entry = pte_mkdirty(entry);
	}
	entry = pte_mkyoung(entry);
3672
	if (ptep_set_access_flags(vma, address, pte, entry, flags & FAULT_FLAG_WRITE)) {
3673
		update_mmu_cache(vma, address, pte);
3674 3675 3676 3677 3678 3679 3680
	} else {
		/*
		 * This is needed only for protection faults but the arch code
		 * is not yet telling us if this is a protection fault or not.
		 * This still avoids useless tlb flushes for .text page faults
		 * with threads.
		 */
3681
		if (flags & FAULT_FLAG_WRITE)
3682
			flush_tlb_fix_spurious_fault(vma, address);
3683
	}
3684 3685
unlock:
	pte_unmap_unlock(pte, ptl);
N
Nick Piggin 已提交
3686
	return 0;
L
Linus Torvalds 已提交
3687 3688 3689 3690 3691
}

/*
 * By the time we get here, we already hold the mm semaphore
 */
3692 3693
static int __handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma,
			     unsigned long address, unsigned int flags)
L
Linus Torvalds 已提交
3694 3695 3696 3697 3698 3699
{
	pgd_t *pgd;
	pud_t *pud;
	pmd_t *pmd;
	pte_t *pte;

3700
	if (unlikely(is_vm_hugetlb_page(vma)))
3701
		return hugetlb_fault(mm, vma, address, flags);
L
Linus Torvalds 已提交
3702

3703
retry:
L
Linus Torvalds 已提交
3704 3705 3706
	pgd = pgd_offset(mm, address);
	pud = pud_alloc(mm, pgd, address);
	if (!pud)
H
Hugh Dickins 已提交
3707
		return VM_FAULT_OOM;
L
Linus Torvalds 已提交
3708 3709
	pmd = pmd_alloc(mm, pud, address);
	if (!pmd)
H
Hugh Dickins 已提交
3710
		return VM_FAULT_OOM;
3711
	if (pmd_none(*pmd) && transparent_hugepage_enabled(vma)) {
3712
		int ret = VM_FAULT_FALLBACK;
3713
		if (!vma->vm_ops)
3714 3715 3716 3717
			ret = do_huge_pmd_anonymous_page(mm, vma, address,
					pmd, flags);
		if (!(ret & VM_FAULT_FALLBACK))
			return ret;
3718 3719
	} else {
		pmd_t orig_pmd = *pmd;
3720 3721
		int ret;

3722 3723
		barrier();
		if (pmd_trans_huge(orig_pmd)) {
3724 3725
			unsigned int dirty = flags & FAULT_FLAG_WRITE;

3726 3727 3728 3729 3730 3731 3732 3733
			/*
			 * If the pmd is splitting, return and retry the
			 * the fault.  Alternative: wait until the split
			 * is done, and goto retry.
			 */
			if (pmd_trans_splitting(orig_pmd))
				return 0;

3734
			if (pmd_numa(orig_pmd))
3735
				return do_huge_pmd_numa_page(mm, vma, address,
3736 3737
							     orig_pmd, pmd);

3738
			if (dirty && !pmd_write(orig_pmd)) {
3739 3740 3741 3742 3743 3744 3745 3746 3747 3748
				ret = do_huge_pmd_wp_page(mm, vma, address, pmd,
							  orig_pmd);
				/*
				 * If COW results in an oom, the huge pmd will
				 * have been split, so retry the fault on the
				 * pte for a smaller charge.
				 */
				if (unlikely(ret & VM_FAULT_OOM))
					goto retry;
				return ret;
3749 3750 3751
			} else {
				huge_pmd_set_accessed(mm, vma, address, pmd,
						      orig_pmd, dirty);
3752
			}
3753

3754 3755 3756 3757
			return 0;
		}
	}

3758 3759
	/* THP should already have been handled */
	BUG_ON(pmd_numa(*pmd));
3760

3761 3762 3763 3764 3765
	/*
	 * Use __pte_alloc instead of pte_alloc_map, because we can't
	 * run pte_offset_map on the pmd, if an huge pmd could
	 * materialize from under us from a different thread.
	 */
3766 3767
	if (unlikely(pmd_none(*pmd)) &&
	    unlikely(__pte_alloc(mm, vma, pmd, address)))
H
Hugh Dickins 已提交
3768
		return VM_FAULT_OOM;
3769 3770 3771 3772 3773 3774 3775 3776 3777 3778
	/* if an huge pmd materialized from under us just retry later */
	if (unlikely(pmd_trans_huge(*pmd)))
		return 0;
	/*
	 * A regular pmd is established and it can't morph into a huge pmd
	 * from under us anymore at this point because we hold the mmap_sem
	 * read mode and khugepaged takes it in write mode. So now it's
	 * safe to run pte_offset_map().
	 */
	pte = pte_offset_map(pmd, address);
L
Linus Torvalds 已提交
3779

3780
	return handle_pte_fault(mm, vma, address, pte, pmd, flags);
L
Linus Torvalds 已提交
3781 3782
}

3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800
int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma,
		    unsigned long address, unsigned int flags)
{
	int ret;

	__set_current_state(TASK_RUNNING);

	count_vm_event(PGFAULT);
	mem_cgroup_count_vm_event(mm, PGFAULT);

	/* do counter updates before entering really critical section. */
	check_sync_rss_stat(current);

	/*
	 * Enable the memcg OOM handling for faults triggered in user
	 * space.  Kernel faults are handled more gracefully.
	 */
	if (flags & FAULT_FLAG_USER)
3801
		mem_cgroup_oom_enable();
3802 3803 3804

	ret = __handle_mm_fault(mm, vma, address, flags);

3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815
	if (flags & FAULT_FLAG_USER) {
		mem_cgroup_oom_disable();
                /*
                 * The task may have entered a memcg OOM situation but
                 * if the allocation error was handled gracefully (no
                 * VM_FAULT_OOM), there is no need to kill anything.
                 * Just clean up the OOM state peacefully.
                 */
                if (task_in_memcg_oom(current) && !(ret & VM_FAULT_OOM))
                        mem_cgroup_oom_synchronize(false);
	}
3816

3817 3818 3819
	return ret;
}

L
Linus Torvalds 已提交
3820 3821 3822
#ifndef __PAGETABLE_PUD_FOLDED
/*
 * Allocate page upper directory.
H
Hugh Dickins 已提交
3823
 * We've already handled the fast-path in-line.
L
Linus Torvalds 已提交
3824
 */
3825
int __pud_alloc(struct mm_struct *mm, pgd_t *pgd, unsigned long address)
L
Linus Torvalds 已提交
3826
{
H
Hugh Dickins 已提交
3827 3828
	pud_t *new = pud_alloc_one(mm, address);
	if (!new)
3829
		return -ENOMEM;
L
Linus Torvalds 已提交
3830

3831 3832
	smp_wmb(); /* See comment in __pte_alloc */

H
Hugh Dickins 已提交
3833
	spin_lock(&mm->page_table_lock);
3834
	if (pgd_present(*pgd))		/* Another has populated it */
3835
		pud_free(mm, new);
3836 3837
	else
		pgd_populate(mm, pgd, new);
H
Hugh Dickins 已提交
3838
	spin_unlock(&mm->page_table_lock);
3839
	return 0;
L
Linus Torvalds 已提交
3840 3841 3842 3843 3844 3845
}
#endif /* __PAGETABLE_PUD_FOLDED */

#ifndef __PAGETABLE_PMD_FOLDED
/*
 * Allocate page middle directory.
H
Hugh Dickins 已提交
3846
 * We've already handled the fast-path in-line.
L
Linus Torvalds 已提交
3847
 */
3848
int __pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address)
L
Linus Torvalds 已提交
3849
{
H
Hugh Dickins 已提交
3850 3851
	pmd_t *new = pmd_alloc_one(mm, address);
	if (!new)
3852
		return -ENOMEM;
L
Linus Torvalds 已提交
3853

3854 3855
	smp_wmb(); /* See comment in __pte_alloc */

H
Hugh Dickins 已提交
3856
	spin_lock(&mm->page_table_lock);
L
Linus Torvalds 已提交
3857
#ifndef __ARCH_HAS_4LEVEL_HACK
3858
	if (pud_present(*pud))		/* Another has populated it */
3859
		pmd_free(mm, new);
3860 3861
	else
		pud_populate(mm, pud, new);
L
Linus Torvalds 已提交
3862
#else
3863
	if (pgd_present(*pud))		/* Another has populated it */
3864
		pmd_free(mm, new);
3865 3866
	else
		pgd_populate(mm, pud, new);
L
Linus Torvalds 已提交
3867
#endif /* __ARCH_HAS_4LEVEL_HACK */
H
Hugh Dickins 已提交
3868
	spin_unlock(&mm->page_table_lock);
3869
	return 0;
3870
}
L
Linus Torvalds 已提交
3871 3872 3873 3874 3875
#endif /* __PAGETABLE_PMD_FOLDED */

#if !defined(__HAVE_ARCH_GATE_AREA)

#if defined(AT_SYSINFO_EHDR)
3876
static struct vm_area_struct gate_vma;
L
Linus Torvalds 已提交
3877 3878 3879 3880 3881 3882

static int __init gate_vma_init(void)
{
	gate_vma.vm_mm = NULL;
	gate_vma.vm_start = FIXADDR_USER_START;
	gate_vma.vm_end = FIXADDR_USER_END;
R
Roland McGrath 已提交
3883 3884
	gate_vma.vm_flags = VM_READ | VM_MAYREAD | VM_EXEC | VM_MAYEXEC;
	gate_vma.vm_page_prot = __P101;
3885

L
Linus Torvalds 已提交
3886 3887 3888 3889 3890
	return 0;
}
__initcall(gate_vma_init);
#endif

3891
struct vm_area_struct *get_gate_vma(struct mm_struct *mm)
L
Linus Torvalds 已提交
3892 3893 3894 3895 3896 3897 3898 3899
{
#ifdef AT_SYSINFO_EHDR
	return &gate_vma;
#else
	return NULL;
#endif
}

3900
int in_gate_area_no_mm(unsigned long addr)
L
Linus Torvalds 已提交
3901 3902 3903 3904 3905 3906 3907 3908 3909
{
#ifdef AT_SYSINFO_EHDR
	if ((addr >= FIXADDR_USER_START) && (addr < FIXADDR_USER_END))
		return 1;
#endif
	return 0;
}

#endif	/* __HAVE_ARCH_GATE_AREA */
3910

3911
static int __follow_pte(struct mm_struct *mm, unsigned long address,
J
Johannes Weiner 已提交
3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927
		pte_t **ptepp, spinlock_t **ptlp)
{
	pgd_t *pgd;
	pud_t *pud;
	pmd_t *pmd;
	pte_t *ptep;

	pgd = pgd_offset(mm, address);
	if (pgd_none(*pgd) || unlikely(pgd_bad(*pgd)))
		goto out;

	pud = pud_offset(pgd, address);
	if (pud_none(*pud) || unlikely(pud_bad(*pud)))
		goto out;

	pmd = pmd_offset(pud, address);
3928
	VM_BUG_ON(pmd_trans_huge(*pmd));
J
Johannes Weiner 已提交
3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948
	if (pmd_none(*pmd) || unlikely(pmd_bad(*pmd)))
		goto out;

	/* We cannot handle huge page PFN maps. Luckily they don't exist. */
	if (pmd_huge(*pmd))
		goto out;

	ptep = pte_offset_map_lock(mm, pmd, address, ptlp);
	if (!ptep)
		goto out;
	if (!pte_present(*ptep))
		goto unlock;
	*ptepp = ptep;
	return 0;
unlock:
	pte_unmap_unlock(ptep, *ptlp);
out:
	return -EINVAL;
}

3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959
static inline int follow_pte(struct mm_struct *mm, unsigned long address,
			     pte_t **ptepp, spinlock_t **ptlp)
{
	int res;

	/* (void) is needed to make gcc happy */
	(void) __cond_lock(*ptlp,
			   !(res = __follow_pte(mm, address, ptepp, ptlp)));
	return res;
}

J
Johannes Weiner 已提交
3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988
/**
 * follow_pfn - look up PFN at a user virtual address
 * @vma: memory mapping
 * @address: user virtual address
 * @pfn: location to store found PFN
 *
 * Only IO mappings and raw PFN mappings are allowed.
 *
 * Returns zero and the pfn at @pfn on success, -ve otherwise.
 */
int follow_pfn(struct vm_area_struct *vma, unsigned long address,
	unsigned long *pfn)
{
	int ret = -EINVAL;
	spinlock_t *ptl;
	pte_t *ptep;

	if (!(vma->vm_flags & (VM_IO | VM_PFNMAP)))
		return ret;

	ret = follow_pte(vma->vm_mm, address, &ptep, &ptl);
	if (ret)
		return ret;
	*pfn = pte_pfn(*ptep);
	pte_unmap_unlock(ptep, ptl);
	return 0;
}
EXPORT_SYMBOL(follow_pfn);

3989
#ifdef CONFIG_HAVE_IOREMAP_PROT
3990 3991 3992
int follow_phys(struct vm_area_struct *vma,
		unsigned long address, unsigned int flags,
		unsigned long *prot, resource_size_t *phys)
3993
{
3994
	int ret = -EINVAL;
3995 3996 3997
	pte_t *ptep, pte;
	spinlock_t *ptl;

3998 3999
	if (!(vma->vm_flags & (VM_IO | VM_PFNMAP)))
		goto out;
4000

4001
	if (follow_pte(vma->vm_mm, address, &ptep, &ptl))
4002
		goto out;
4003
	pte = *ptep;
4004

4005 4006 4007 4008
	if ((flags & FOLL_WRITE) && !pte_write(pte))
		goto unlock;

	*prot = pgprot_val(pte_pgprot(pte));
4009
	*phys = (resource_size_t)pte_pfn(pte) << PAGE_SHIFT;
4010

4011
	ret = 0;
4012 4013 4014
unlock:
	pte_unmap_unlock(ptep, ptl);
out:
4015
	return ret;
4016 4017 4018 4019 4020 4021 4022
}

int generic_access_phys(struct vm_area_struct *vma, unsigned long addr,
			void *buf, int len, int write)
{
	resource_size_t phys_addr;
	unsigned long prot = 0;
K
KOSAKI Motohiro 已提交
4023
	void __iomem *maddr;
4024 4025
	int offset = addr & (PAGE_SIZE-1);

4026
	if (follow_phys(vma, addr, write, &prot, &phys_addr))
4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037
		return -EINVAL;

	maddr = ioremap_prot(phys_addr, PAGE_SIZE, prot);
	if (write)
		memcpy_toio(maddr + offset, buf, len);
	else
		memcpy_fromio(buf, maddr + offset, len);
	iounmap(maddr);

	return len;
}
4038
EXPORT_SYMBOL_GPL(generic_access_phys);
4039 4040
#endif

4041
/*
4042 4043
 * Access another process' address space as given in mm.  If non-NULL, use the
 * given task for page fault accounting.
4044
 */
4045 4046
static int __access_remote_vm(struct task_struct *tsk, struct mm_struct *mm,
		unsigned long addr, void *buf, int len, int write)
4047 4048 4049 4050 4051
{
	struct vm_area_struct *vma;
	void *old_buf = buf;

	down_read(&mm->mmap_sem);
S
Simon Arlott 已提交
4052
	/* ignore errors, just check how much was successfully transferred */
4053 4054 4055
	while (len) {
		int bytes, ret, offset;
		void *maddr;
4056
		struct page *page = NULL;
4057 4058 4059

		ret = get_user_pages(tsk, mm, addr, 1,
				write, 1, &page, &vma);
4060 4061 4062 4063 4064 4065 4066
		if (ret <= 0) {
			/*
			 * Check if this is a VM_IO | VM_PFNMAP VMA, which
			 * we can access using slightly different code.
			 */
#ifdef CONFIG_HAVE_IOREMAP_PROT
			vma = find_vma(mm, addr);
4067
			if (!vma || vma->vm_start > addr)
4068 4069 4070 4071 4072 4073 4074 4075
				break;
			if (vma->vm_ops && vma->vm_ops->access)
				ret = vma->vm_ops->access(vma, addr, buf,
							  len, write);
			if (ret <= 0)
#endif
				break;
			bytes = ret;
4076
		} else {
4077 4078 4079 4080 4081 4082 4083 4084 4085 4086 4087 4088 4089 4090 4091 4092
			bytes = len;
			offset = addr & (PAGE_SIZE-1);
			if (bytes > PAGE_SIZE-offset)
				bytes = PAGE_SIZE-offset;

			maddr = kmap(page);
			if (write) {
				copy_to_user_page(vma, page, addr,
						  maddr + offset, buf, bytes);
				set_page_dirty_lock(page);
			} else {
				copy_from_user_page(vma, page, addr,
						    buf, maddr + offset, bytes);
			}
			kunmap(page);
			page_cache_release(page);
4093 4094 4095 4096 4097 4098 4099 4100 4101
		}
		len -= bytes;
		buf += bytes;
		addr += bytes;
	}
	up_read(&mm->mmap_sem);

	return buf - old_buf;
}
4102

S
Stephen Wilson 已提交
4103
/**
4104
 * access_remote_vm - access another process' address space
S
Stephen Wilson 已提交
4105 4106 4107 4108 4109 4110 4111 4112 4113 4114 4115 4116 4117 4118
 * @mm:		the mm_struct of the target address space
 * @addr:	start address to access
 * @buf:	source or destination buffer
 * @len:	number of bytes to transfer
 * @write:	whether the access is a write
 *
 * The caller must hold a reference on @mm.
 */
int access_remote_vm(struct mm_struct *mm, unsigned long addr,
		void *buf, int len, int write)
{
	return __access_remote_vm(NULL, mm, addr, buf, len, write);
}

4119 4120 4121 4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4132 4133 4134 4135 4136 4137 4138 4139
/*
 * Access another process' address space.
 * Source/target buffer must be kernel space,
 * Do not walk the page table directly, use get_user_pages
 */
int access_process_vm(struct task_struct *tsk, unsigned long addr,
		void *buf, int len, int write)
{
	struct mm_struct *mm;
	int ret;

	mm = get_task_mm(tsk);
	if (!mm)
		return 0;

	ret = __access_remote_vm(tsk, mm, addr, buf, len, write);
	mmput(mm);

	return ret;
}

4140 4141 4142 4143 4144 4145 4146 4147
/*
 * Print the name of a VMA.
 */
void print_vma_addr(char *prefix, unsigned long ip)
{
	struct mm_struct *mm = current->mm;
	struct vm_area_struct *vma;

4148 4149 4150 4151 4152 4153 4154
	/*
	 * Do not print if we are in atomic
	 * contexts (in exception stacks, etc.):
	 */
	if (preempt_count())
		return;

4155 4156 4157 4158 4159 4160
	down_read(&mm->mmap_sem);
	vma = find_vma(mm, ip);
	if (vma && vma->vm_file) {
		struct file *f = vma->vm_file;
		char *buf = (char *)__get_free_page(GFP_KERNEL);
		if (buf) {
A
Andy Shevchenko 已提交
4161
			char *p;
4162

4163
			p = d_path(&f->f_path, buf, PAGE_SIZE);
4164 4165
			if (IS_ERR(p))
				p = "?";
A
Andy Shevchenko 已提交
4166
			printk("%s%s[%lx+%lx]", prefix, kbasename(p),
4167 4168 4169 4170 4171
					vma->vm_start,
					vma->vm_end - vma->vm_start);
			free_page((unsigned long)buf);
		}
	}
4172
	up_read(&mm->mmap_sem);
4173
}
4174

4175
#if defined(CONFIG_PROVE_LOCKING) || defined(CONFIG_DEBUG_ATOMIC_SLEEP)
4176 4177
void might_fault(void)
{
4178 4179 4180 4181 4182 4183 4184 4185 4186
	/*
	 * Some code (nfs/sunrpc) uses socket ops on kernel memory while
	 * holding the mmap_sem, this is safe because kernel memory doesn't
	 * get paged out, therefore we'll never actually fault, and the
	 * below annotations will generate false positives.
	 */
	if (segment_eq(get_fs(), KERNEL_DS))
		return;

4187 4188 4189 4190 4191
	/*
	 * it would be nicer only to annotate paths which are not under
	 * pagefault_disable, however that requires a larger audit and
	 * providing helpers like get_user_atomic.
	 */
4192 4193 4194 4195 4196 4197
	if (in_atomic())
		return;

	__might_sleep(__FILE__, __LINE__, 0);

	if (current->mm)
4198 4199 4200 4201
		might_lock_read(&current->mm->mmap_sem);
}
EXPORT_SYMBOL(might_fault);
#endif
A
Andrea Arcangeli 已提交
4202 4203 4204 4205 4206 4207 4208 4209 4210 4211 4212 4213 4214 4215 4216 4217 4218 4219 4220 4221 4222 4223 4224 4225 4226 4227 4228 4229 4230 4231 4232 4233 4234 4235 4236 4237 4238 4239 4240 4241 4242 4243 4244 4245 4246 4247 4248 4249 4250 4251 4252 4253 4254 4255 4256 4257 4258 4259 4260 4261 4262 4263 4264 4265 4266 4267 4268 4269 4270 4271 4272

#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLBFS)
static void clear_gigantic_page(struct page *page,
				unsigned long addr,
				unsigned int pages_per_huge_page)
{
	int i;
	struct page *p = page;

	might_sleep();
	for (i = 0; i < pages_per_huge_page;
	     i++, p = mem_map_next(p, page, i)) {
		cond_resched();
		clear_user_highpage(p, addr + i * PAGE_SIZE);
	}
}
void clear_huge_page(struct page *page,
		     unsigned long addr, unsigned int pages_per_huge_page)
{
	int i;

	if (unlikely(pages_per_huge_page > MAX_ORDER_NR_PAGES)) {
		clear_gigantic_page(page, addr, pages_per_huge_page);
		return;
	}

	might_sleep();
	for (i = 0; i < pages_per_huge_page; i++) {
		cond_resched();
		clear_user_highpage(page + i, addr + i * PAGE_SIZE);
	}
}

static void copy_user_gigantic_page(struct page *dst, struct page *src,
				    unsigned long addr,
				    struct vm_area_struct *vma,
				    unsigned int pages_per_huge_page)
{
	int i;
	struct page *dst_base = dst;
	struct page *src_base = src;

	for (i = 0; i < pages_per_huge_page; ) {
		cond_resched();
		copy_user_highpage(dst, src, addr + i*PAGE_SIZE, vma);

		i++;
		dst = mem_map_next(dst, dst_base, i);
		src = mem_map_next(src, src_base, i);
	}
}

void copy_user_huge_page(struct page *dst, struct page *src,
			 unsigned long addr, struct vm_area_struct *vma,
			 unsigned int pages_per_huge_page)
{
	int i;

	if (unlikely(pages_per_huge_page > MAX_ORDER_NR_PAGES)) {
		copy_user_gigantic_page(dst, src, addr, vma,
					pages_per_huge_page);
		return;
	}

	might_sleep();
	for (i = 0; i < pages_per_huge_page; i++) {
		cond_resched();
		copy_user_highpage(dst + i, src + i, addr + i*PAGE_SIZE, vma);
	}
}
#endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLBFS */
4273 4274 4275 4276 4277 4278 4279 4280 4281 4282 4283 4284 4285 4286 4287 4288 4289 4290 4291

#if USE_SPLIT_PTE_PTLOCKS
bool __ptlock_alloc(struct page *page)
{
	spinlock_t *ptl;

	ptl = kmalloc(sizeof(spinlock_t), GFP_KERNEL);
	if (!ptl)
		return false;
	page->ptl = (unsigned long)ptl;
	return true;
}

void __ptlock_free(struct page *page)
{
	if (sizeof(spinlock_t) > sizeof(page->ptl))
		kfree((spinlock_t *)page->ptl);
}
#endif