mm.h 63.7 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8
#ifndef _LINUX_MM_H
#define _LINUX_MM_H

#include <linux/errno.h>

#ifdef __KERNEL__

#include <linux/gfp.h>
9
#include <linux/bug.h>
L
Linus Torvalds 已提交
10 11 12
#include <linux/list.h>
#include <linux/mmzone.h>
#include <linux/rbtree.h>
13
#include <linux/atomic.h>
14
#include <linux/debug_locks.h>
15
#include <linux/mm_types.h>
16
#include <linux/range.h>
17
#include <linux/pfn.h>
A
Andrea Arcangeli 已提交
18
#include <linux/bit_spinlock.h>
19
#include <linux/shrinker.h>
L
Linus Torvalds 已提交
20 21 22

struct mempolicy;
struct anon_vma;
23
struct anon_vma_chain;
A
Alexey Dobriyan 已提交
24
struct file_ra_state;
A
Alexey Dobriyan 已提交
25
struct user_struct;
A
Alexey Dobriyan 已提交
26
struct writeback_control;
L
Linus Torvalds 已提交
27

28
#ifndef CONFIG_NEED_MULTIPLE_NODES	/* Don't use mapnrs, do it properly */
L
Linus Torvalds 已提交
29
extern unsigned long max_mapnr;
30 31 32 33 34 35 36

static inline void set_max_mapnr(unsigned long limit)
{
	max_mapnr = limit;
}
#else
static inline void set_max_mapnr(unsigned long limit) { }
L
Linus Torvalds 已提交
37 38
#endif

39
extern unsigned long totalram_pages;
L
Linus Torvalds 已提交
40 41 42 43 44 45 46 47 48 49 50 51 52
extern void * high_memory;
extern int page_cluster;

#ifdef CONFIG_SYSCTL
extern int sysctl_legacy_va_layout;
#else
#define sysctl_legacy_va_layout 0
#endif

#include <asm/page.h>
#include <asm/pgtable.h>
#include <asm/processor.h>

53 54 55 56
#ifndef __pa_symbol
#define __pa_symbol(x)  __pa(RELOC_HIDE((unsigned long)(x), 0))
#endif

57
extern unsigned long sysctl_user_reserve_kbytes;
58
extern unsigned long sysctl_admin_reserve_kbytes;
59

L
Linus Torvalds 已提交
60 61
#define nth_page(page,n) pfn_to_page(page_to_pfn((page)) + (n))

62 63 64
/* to align the pointer to the (next) page boundary */
#define PAGE_ALIGN(addr) ALIGN(addr, PAGE_SIZE)

65 66 67
/* test whether an address (unsigned long or pointer) is aligned to PAGE_SIZE */
#define PAGE_ALIGNED(addr)	IS_ALIGNED((unsigned long)addr, PAGE_SIZE)

L
Linus Torvalds 已提交
68 69 70 71 72 73 74 75 76
/*
 * Linux kernel virtual memory manager primitives.
 * The idea being to have a "virtual" mm in the same way
 * we have a virtual fs - giving a cleaner interface to the
 * mm details, and allowing different kinds of memory mappings
 * (from shared memory to executable loading to arbitrary
 * mmap() functions).
 */

77 78
extern struct kmem_cache *vm_area_cachep;

L
Linus Torvalds 已提交
79
#ifndef CONFIG_MMU
80 81
extern struct rb_root nommu_region_tree;
extern struct rw_semaphore nommu_region_sem;
L
Linus Torvalds 已提交
82 83 84 85 86

extern unsigned int kobjsize(const void *objp);
#endif

/*
H
Hugh Dickins 已提交
87
 * vm_flags in vm_area_struct, see mm_types.h.
L
Linus Torvalds 已提交
88
 */
89 90
#define VM_NONE		0x00000000

L
Linus Torvalds 已提交
91 92 93 94 95
#define VM_READ		0x00000001	/* currently active flags */
#define VM_WRITE	0x00000002
#define VM_EXEC		0x00000004
#define VM_SHARED	0x00000008

96
/* mprotect() hardcodes VM_MAYREAD >> 4 == VM_READ, and so for r/w/x bits. */
L
Linus Torvalds 已提交
97 98 99 100 101 102
#define VM_MAYREAD	0x00000010	/* limits for mprotect() etc */
#define VM_MAYWRITE	0x00000020
#define VM_MAYEXEC	0x00000040
#define VM_MAYSHARE	0x00000080

#define VM_GROWSDOWN	0x00000100	/* general info on the segment */
103
#define VM_PFNMAP	0x00000400	/* Page-ranges managed without "struct page", just pure PFN */
L
Linus Torvalds 已提交
104 105 106 107 108 109 110 111 112 113 114 115
#define VM_DENYWRITE	0x00000800	/* ETXTBSY on write attempts.. */

#define VM_LOCKED	0x00002000
#define VM_IO           0x00004000	/* Memory mapped I/O or similar */

					/* Used by sys_madvise() */
#define VM_SEQ_READ	0x00008000	/* App will access data sequentially */
#define VM_RAND_READ	0x00010000	/* App will not benefit from clustered reads */

#define VM_DONTCOPY	0x00020000      /* Do not copy this vma on fork */
#define VM_DONTEXPAND	0x00040000	/* Cannot expand with mremap() */
#define VM_ACCOUNT	0x00100000	/* Is a VM accounted object */
116
#define VM_NORESERVE	0x00200000	/* should the VM suppress accounting */
L
Linus Torvalds 已提交
117 118
#define VM_HUGETLB	0x00400000	/* Huge TLB Page VM */
#define VM_NONLINEAR	0x00800000	/* Is non-linear (remap_file_pages) */
119
#define VM_ARCH_1	0x01000000	/* Architecture-specific flag */
120
#define VM_DONTDUMP	0x04000000	/* Do not include in the core dump */
121

122 123 124 125 126 127
#ifdef CONFIG_MEM_SOFT_DIRTY
# define VM_SOFTDIRTY	0x08000000	/* Not soft dirty clean area */
#else
# define VM_SOFTDIRTY	0
#endif

J
Jared Hulbert 已提交
128
#define VM_MIXEDMAP	0x10000000	/* Can contain "struct page" and pure PFN pages */
129 130
#define VM_HUGEPAGE	0x20000000	/* MADV_HUGEPAGE marked this vma */
#define VM_NOHUGEPAGE	0x40000000	/* MADV_NOHUGEPAGE marked this vma */
H
Hugh Dickins 已提交
131
#define VM_MERGEABLE	0x80000000	/* KSM may merge identical pages */
L
Linus Torvalds 已提交
132

133 134 135 136 137 138
#if defined(CONFIG_X86)
# define VM_PAT		VM_ARCH_1	/* PAT reserves whole VMA at once (x86) */
#elif defined(CONFIG_PPC)
# define VM_SAO		VM_ARCH_1	/* Strong Access Ordering (powerpc) */
#elif defined(CONFIG_PARISC)
# define VM_GROWSUP	VM_ARCH_1
139 140
#elif defined(CONFIG_METAG)
# define VM_GROWSUP	VM_ARCH_1
141 142 143 144 145 146 147 148 149 150
#elif defined(CONFIG_IA64)
# define VM_GROWSUP	VM_ARCH_1
#elif !defined(CONFIG_MMU)
# define VM_MAPPED_COPY	VM_ARCH_1	/* T if mapped copy of data (nommu mmap) */
#endif

#ifndef VM_GROWSUP
# define VM_GROWSUP	VM_NONE
#endif

151 152 153
/* Bits set in the VMA until the stack is in its final location */
#define VM_STACK_INCOMPLETE_SETUP	(VM_RAND_READ | VM_SEQ_READ)

L
Linus Torvalds 已提交
154 155 156 157 158 159 160 161 162 163
#ifndef VM_STACK_DEFAULT_FLAGS		/* arch can override this */
#define VM_STACK_DEFAULT_FLAGS VM_DATA_DEFAULT_FLAGS
#endif

#ifdef CONFIG_STACK_GROWSUP
#define VM_STACK_FLAGS	(VM_GROWSUP | VM_STACK_DEFAULT_FLAGS | VM_ACCOUNT)
#else
#define VM_STACK_FLAGS	(VM_GROWSDOWN | VM_STACK_DEFAULT_FLAGS | VM_ACCOUNT)
#endif

N
Nick Piggin 已提交
164
/*
165 166
 * Special vmas that are non-mergable, non-mlock()able.
 * Note: mm/huge_memory.c VM_NO_THP depends on this definition.
N
Nick Piggin 已提交
167
 */
168
#define VM_SPECIAL (VM_IO | VM_DONTEXPAND | VM_PFNMAP)
N
Nick Piggin 已提交
169

L
Linus Torvalds 已提交
170 171 172 173 174 175
/*
 * mapping from the currently active vm_flags protection bits (the
 * low four bits) to a page protection mask..
 */
extern pgprot_t protection_map[16];

N
Nick Piggin 已提交
176 177
#define FAULT_FLAG_WRITE	0x01	/* Fault was a write access */
#define FAULT_FLAG_NONLINEAR	0x02	/* Fault was via a nonlinear mapping */
178
#define FAULT_FLAG_MKWRITE	0x04	/* Fault was mkwrite of existing pte */
179
#define FAULT_FLAG_ALLOW_RETRY	0x08	/* Retry fault if blocking */
180
#define FAULT_FLAG_RETRY_NOWAIT	0x10	/* Don't drop mmap_sem and wait when retrying */
181
#define FAULT_FLAG_KILLABLE	0x20	/* The fault task is in SIGKILL killable region */
182
#define FAULT_FLAG_TRIED	0x40	/* second try */
183
#define FAULT_FLAG_USER		0x80	/* The fault originated in userspace */
N
Nick Piggin 已提交
184

185
/*
N
Nick Piggin 已提交
186
 * vm_fault is filled by the the pagefault handler and passed to the vma's
N
Nick Piggin 已提交
187 188
 * ->fault function. The vma's ->fault is responsible for returning a bitmask
 * of VM_FAULT_xxx flags that give details about how the fault was handled.
189
 *
N
Nick Piggin 已提交
190
 * pgoff should be used in favour of virtual_address, if possible. If pgoff
191
 * is used, one may implement ->remap_pages to get nonlinear mapping support.
192
 */
N
Nick Piggin 已提交
193 194 195 196 197 198
struct vm_fault {
	unsigned int flags;		/* FAULT_FLAG_xxx flags */
	pgoff_t pgoff;			/* Logical page offset based on vma */
	void __user *virtual_address;	/* Faulting virtual address */

	struct page *page;		/* ->fault handlers should return a
N
Nick Piggin 已提交
199
					 * page here, unless VM_FAULT_NOPAGE
N
Nick Piggin 已提交
200
					 * is set (which is also implied by
N
Nick Piggin 已提交
201
					 * VM_FAULT_ERROR).
N
Nick Piggin 已提交
202
					 */
203
};
L
Linus Torvalds 已提交
204 205 206 207 208 209 210 211 212

/*
 * These are the virtual MM functions - opening of an area, closing and
 * unmapping it (needed to keep files on disk up-to-date etc), pointer
 * to the functions called when a no-page or a wp-page exception occurs. 
 */
struct vm_operations_struct {
	void (*open)(struct vm_area_struct * area);
	void (*close)(struct vm_area_struct * area);
N
Nick Piggin 已提交
213
	int (*fault)(struct vm_area_struct *vma, struct vm_fault *vmf);
214 215 216

	/* notification that a previously read-only page is about to become
	 * writable, if an error is returned it will cause a SIGBUS */
217
	int (*page_mkwrite)(struct vm_area_struct *vma, struct vm_fault *vmf);
218 219 220 221 222 223

	/* called by access_process_vm when get_user_pages() fails, typically
	 * for use by special VMAs that can switch between memory and hardware
	 */
	int (*access)(struct vm_area_struct *vma, unsigned long addr,
		      void *buf, int len, int write);
L
Linus Torvalds 已提交
224
#ifdef CONFIG_NUMA
225 226 227 228 229 230 231
	/*
	 * set_policy() op must add a reference to any non-NULL @new mempolicy
	 * to hold the policy upon return.  Caller should pass NULL @new to
	 * remove a policy and fall back to surrounding context--i.e. do not
	 * install a MPOL_DEFAULT policy, nor the task or system default
	 * mempolicy.
	 */
L
Linus Torvalds 已提交
232
	int (*set_policy)(struct vm_area_struct *vma, struct mempolicy *new);
233 234 235 236 237 238 239 240 241 242 243

	/*
	 * get_policy() op must add reference [mpol_get()] to any policy at
	 * (vma,addr) marked as MPOL_SHARED.  The shared policy infrastructure
	 * in mm/mempolicy.c will do this automatically.
	 * get_policy() must NOT add a ref if the policy at (vma,addr) is not
	 * marked as MPOL_SHARED. vma policies are protected by the mmap_sem.
	 * If no [shared/vma] mempolicy exists at the addr, get_policy() op
	 * must return NULL--i.e., do not "fallback" to task or system default
	 * policy.
	 */
L
Linus Torvalds 已提交
244 245
	struct mempolicy *(*get_policy)(struct vm_area_struct *vma,
					unsigned long addr);
246 247
	int (*migrate)(struct vm_area_struct *vma, const nodemask_t *from,
		const nodemask_t *to, unsigned long flags);
L
Linus Torvalds 已提交
248
#endif
249 250 251
	/* called by sys_remap_file_pages() to populate non-linear mapping */
	int (*remap_pages)(struct vm_area_struct *vma, unsigned long addr,
			   unsigned long size, pgoff_t pgoff);
L
Linus Torvalds 已提交
252 253 254 255 256
};

struct mmu_gather;
struct inode;

A
Andrew Morton 已提交
257 258
#define page_private(page)		((page)->private)
#define set_page_private(page, v)	((page)->private = (v))
H
Hugh Dickins 已提交
259

260 261 262
/* It's valid only if the page is free path or free_list */
static inline void set_freepage_migratetype(struct page *page, int migratetype)
{
M
Minchan Kim 已提交
263
	page->index = migratetype;
264 265 266 267 268
}

/* It's valid only if the page is free path or free_list */
static inline int get_freepage_migratetype(struct page *page)
{
M
Minchan Kim 已提交
269
	return page->index;
270 271
}

L
Linus Torvalds 已提交
272 273 274 275 276
/*
 * FIXME: take this include out, include page-flags.h in
 * files which need it (119 of them)
 */
#include <linux/page-flags.h>
277
#include <linux/huge_mm.h>
L
Linus Torvalds 已提交
278 279 280 281 282 283 284 285 286 287 288 289 290 291 292

/*
 * Methods to modify the page usage count.
 *
 * What counts for a page usage:
 * - cache mapping   (page->mapping)
 * - private data    (page->private)
 * - page mapped in a task's page tables, each mapping
 *   is counted separately
 *
 * Also, many kernel routines increase the page count before a critical
 * routine so they can be sure the page doesn't go away from under them.
 */

/*
N
Nick Piggin 已提交
293
 * Drop a ref, return true if the refcount fell to zero (the page has no users)
L
Linus Torvalds 已提交
294
 */
295 296
static inline int put_page_testzero(struct page *page)
{
N
Nick Piggin 已提交
297
	VM_BUG_ON(atomic_read(&page->_count) == 0);
298
	return atomic_dec_and_test(&page->_count);
299
}
L
Linus Torvalds 已提交
300 301

/*
302 303
 * Try to grab a ref unless the page has a refcount of zero, return false if
 * that is the case.
304 305
 * This can be called when MMU is off so it must not access
 * any of the virtual mappings.
L
Linus Torvalds 已提交
306
 */
307 308
static inline int get_page_unless_zero(struct page *page)
{
309
	return atomic_inc_not_zero(&page->_count);
310
}
L
Linus Torvalds 已提交
311

312 313 314 315 316 317 318 319 320 321 322 323
/*
 * Try to drop a ref unless the page has a refcount of one, return false if
 * that is the case.
 * This is to make sure that the refcount won't become zero after this drop.
 * This can be called when MMU is off so it must not access
 * any of the virtual mappings.
 */
static inline int put_page_unless_one(struct page *page)
{
	return atomic_add_unless(&page->_count, -1, 1);
}

324 325
extern int page_is_ram(unsigned long pfn);

326
/* Support for virtually mapped pages */
327 328
struct page *vmalloc_to_page(const void *addr);
unsigned long vmalloc_to_pfn(const void *addr);
329

330 331 332 333 334 335
/*
 * Determine if an address is within the vmalloc range
 *
 * On nommu, vmalloc/vfree wrap through kmalloc/kfree directly, so there
 * is no special casing required.
 */
336 337
static inline int is_vmalloc_addr(const void *x)
{
338
#ifdef CONFIG_MMU
339 340 341
	unsigned long addr = (unsigned long)x;

	return addr >= VMALLOC_START && addr < VMALLOC_END;
342 343
#else
	return 0;
344
#endif
345
}
346 347 348
#ifdef CONFIG_MMU
extern int is_vmalloc_or_module_addr(const void *x);
#else
349
static inline int is_vmalloc_or_module_addr(const void *x)
350 351 352 353
{
	return 0;
}
#endif
354

A
Andrea Arcangeli 已提交
355 356 357
static inline void compound_lock(struct page *page)
{
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
358
	VM_BUG_ON(PageSlab(page));
A
Andrea Arcangeli 已提交
359 360 361 362 363 364 365
	bit_spin_lock(PG_compound_lock, &page->flags);
#endif
}

static inline void compound_unlock(struct page *page)
{
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
366
	VM_BUG_ON(PageSlab(page));
A
Andrea Arcangeli 已提交
367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389
	bit_spin_unlock(PG_compound_lock, &page->flags);
#endif
}

static inline unsigned long compound_lock_irqsave(struct page *page)
{
	unsigned long uninitialized_var(flags);
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
	local_irq_save(flags);
	compound_lock(page);
#endif
	return flags;
}

static inline void compound_unlock_irqrestore(struct page *page,
					      unsigned long flags)
{
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
	compound_unlock(page);
	local_irq_restore(flags);
#endif
}

390 391
static inline struct page *compound_head(struct page *page)
{
392
	if (unlikely(PageTail(page)))
393 394 395 396
		return page->first_page;
	return page;
}

397 398 399 400 401
/*
 * The atomic page->_mapcount, starts from -1: so that transitions
 * both from it and to it can be tracked, using atomic_inc_and_test
 * and atomic_add_negative(-1).
 */
402
static inline void page_mapcount_reset(struct page *page)
403 404 405 406 407 408 409 410 411
{
	atomic_set(&(page)->_mapcount, -1);
}

static inline int page_mapcount(struct page *page)
{
	return atomic_read(&(page)->_mapcount) + 1;
}

H
Hugh Dickins 已提交
412
static inline int page_count(struct page *page)
L
Linus Torvalds 已提交
413
{
414
	return atomic_read(&compound_head(page)->_count);
L
Linus Torvalds 已提交
415 416
}

A
Andrea Arcangeli 已提交
417 418 419 420 421 422 423 424 425 426 427
static inline void get_huge_page_tail(struct page *page)
{
	/*
	 * __split_huge_page_refcount() cannot run
	 * from under us.
	 */
	VM_BUG_ON(page_mapcount(page) < 0);
	VM_BUG_ON(atomic_read(&page->_count) != 0);
	atomic_inc(&page->_mapcount);
}

428 429
extern bool __get_page_tail(struct page *page);

L
Linus Torvalds 已提交
430 431
static inline void get_page(struct page *page)
{
432 433 434
	if (unlikely(PageTail(page)))
		if (likely(__get_page_tail(page)))
			return;
435 436
	/*
	 * Getting a normal page or the head of a compound page
437
	 * requires to already have an elevated page->_count.
438
	 */
439
	VM_BUG_ON(atomic_read(&page->_count) <= 0);
L
Linus Torvalds 已提交
440 441 442
	atomic_inc(&page->_count);
}

443 444 445 446 447 448
static inline struct page *virt_to_head_page(const void *x)
{
	struct page *page = virt_to_page(x);
	return compound_head(page);
}

449 450 451 452 453 454 455 456 457
/*
 * Setup the page count before being freed into the page allocator for
 * the first time (boot or memory hotplug)
 */
static inline void init_page_count(struct page *page)
{
	atomic_set(&page->_count, 1);
}

A
Andrea Arcangeli 已提交
458 459 460
/*
 * PageBuddy() indicate that the page is free and in the buddy system
 * (see mm/page_alloc.c).
461 462 463 464 465
 *
 * PAGE_BUDDY_MAPCOUNT_VALUE must be <= -2 but better not too close to
 * -2 so that an underflow of the page_mapcount() won't be mistaken
 * for a genuine PAGE_BUDDY_MAPCOUNT_VALUE. -128 can be created very
 * efficiently by most CPU architectures.
A
Andrea Arcangeli 已提交
466
 */
467 468
#define PAGE_BUDDY_MAPCOUNT_VALUE (-128)

A
Andrea Arcangeli 已提交
469 470
static inline int PageBuddy(struct page *page)
{
471
	return atomic_read(&page->_mapcount) == PAGE_BUDDY_MAPCOUNT_VALUE;
A
Andrea Arcangeli 已提交
472 473 474 475 476
}

static inline void __SetPageBuddy(struct page *page)
{
	VM_BUG_ON(atomic_read(&page->_mapcount) != -1);
477
	atomic_set(&page->_mapcount, PAGE_BUDDY_MAPCOUNT_VALUE);
A
Andrea Arcangeli 已提交
478 479 480 481 482 483 484 485
}

static inline void __ClearPageBuddy(struct page *page)
{
	VM_BUG_ON(!PageBuddy(page));
	atomic_set(&page->_mapcount, -1);
}

L
Linus Torvalds 已提交
486
void put_page(struct page *page);
487
void put_pages_list(struct list_head *pages);
L
Linus Torvalds 已提交
488

N
Nick Piggin 已提交
489
void split_page(struct page *page, unsigned int order);
490
int split_free_page(struct page *page);
N
Nick Piggin 已提交
491

492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509
/*
 * Compound pages have a destructor function.  Provide a
 * prototype for that function and accessor functions.
 * These are _only_ valid on the head of a PG_compound page.
 */
typedef void compound_page_dtor(struct page *);

static inline void set_compound_page_dtor(struct page *page,
						compound_page_dtor *dtor)
{
	page[1].lru.next = (void *)dtor;
}

static inline compound_page_dtor *get_compound_page_dtor(struct page *page)
{
	return (compound_page_dtor *)page[1].lru.next;
}

510 511
static inline int compound_order(struct page *page)
{
512
	if (!PageHead(page))
513 514 515 516 517 518 519 520 521
		return 0;
	return (unsigned long)page[1].lru.prev;
}

static inline void set_compound_order(struct page *page, unsigned long order)
{
	page[1].lru.prev = (void *)order;
}

522
#ifdef CONFIG_MMU
A
Andrea Arcangeli 已提交
523 524 525 526 527 528 529 530 531 532 533 534
/*
 * Do pte_mkwrite, but only if the vma says VM_WRITE.  We do this when
 * servicing faults for write access.  In the normal case, do always want
 * pte_mkwrite.  But get_user_pages can cause write faults for mappings
 * that do not have writing enabled, when used by access_process_vm.
 */
static inline pte_t maybe_mkwrite(pte_t pte, struct vm_area_struct *vma)
{
	if (likely(vma->vm_flags & VM_WRITE))
		pte = pte_mkwrite(pte);
	return pte;
}
535
#endif
A
Andrea Arcangeli 已提交
536

L
Linus Torvalds 已提交
537 538 539 540 541 542 543
/*
 * Multiple processes may "see" the same page. E.g. for untouched
 * mappings of /dev/null, all processes see the same page full of
 * zeroes, and text pages of executables and shared libraries have
 * only one copy in memory, at most, normally.
 *
 * For the non-reserved pages, page_count(page) denotes a reference count.
544 545
 *   page_count() == 0 means the page is free. page->lru is then used for
 *   freelist management in the buddy allocator.
N
Nick Piggin 已提交
546
 *   page_count() > 0  means the page has been allocated.
L
Linus Torvalds 已提交
547
 *
N
Nick Piggin 已提交
548 549 550 551 552
 * Pages are allocated by the slab allocator in order to provide memory
 * to kmalloc and kmem_cache_alloc. In this case, the management of the
 * page, and the fields in 'struct page' are the responsibility of mm/slab.c
 * unless a particular usage is carefully commented. (the responsibility of
 * freeing the kmalloc memory is the caller's, of course).
L
Linus Torvalds 已提交
553
 *
N
Nick Piggin 已提交
554 555 556 557 558 559 560 561 562
 * A page may be used by anyone else who does a __get_free_page().
 * In this case, page_count still tracks the references, and should only
 * be used through the normal accessor functions. The top bits of page->flags
 * and page->virtual store page management information, but all other fields
 * are unused and could be used privately, carefully. The management of this
 * page is the responsibility of the one who allocated it, and those who have
 * subsequently been given references to it.
 *
 * The other pages (we may call them "pagecache pages") are completely
L
Linus Torvalds 已提交
563 564 565
 * managed by the Linux memory manager: I/O, buffers, swapping etc.
 * The following discussion applies only to them.
 *
N
Nick Piggin 已提交
566 567 568 569
 * A pagecache page contains an opaque `private' member, which belongs to the
 * page's address_space. Usually, this is the address of a circular list of
 * the page's disk buffers. PG_private must be set to tell the VM to call
 * into the filesystem to release these pages.
L
Linus Torvalds 已提交
570
 *
N
Nick Piggin 已提交
571 572 573
 * A page may belong to an inode's memory mapping. In this case, page->mapping
 * is the pointer to the inode, and page->index is the file offset of the page,
 * in units of PAGE_CACHE_SIZE.
L
Linus Torvalds 已提交
574
 *
N
Nick Piggin 已提交
575 576 577
 * If pagecache pages are not associated with an inode, they are said to be
 * anonymous pages. These may become associated with the swapcache, and in that
 * case PG_swapcache is set, and page->private is an offset into the swapcache.
L
Linus Torvalds 已提交
578
 *
N
Nick Piggin 已提交
579 580 581
 * In either case (swapcache or inode backed), the pagecache itself holds one
 * reference to the page. Setting PG_private should also increment the
 * refcount. The each user mapping also has a reference to the page.
L
Linus Torvalds 已提交
582
 *
N
Nick Piggin 已提交
583 584 585 586
 * The pagecache pages are stored in a per-mapping radix tree, which is
 * rooted at mapping->page_tree, and indexed by offset.
 * Where 2.4 and early 2.6 kernels kept dirty/clean pages in per-address_space
 * lists, we instead now tag pages as dirty/writeback in the radix tree.
L
Linus Torvalds 已提交
587
 *
N
Nick Piggin 已提交
588
 * All pagecache pages may be subject to I/O:
L
Linus Torvalds 已提交
589 590
 * - inode pages may need to be read from disk,
 * - inode pages which have been modified and are MAP_SHARED may need
N
Nick Piggin 已提交
591 592 593 594
 *   to be written back to the inode on disk,
 * - anonymous pages (including MAP_PRIVATE file mappings) which have been
 *   modified may need to be swapped out to swap space and (later) to be read
 *   back into memory.
L
Linus Torvalds 已提交
595 596 597 598 599 600
 */

/*
 * The zone field is never updated after free_area_init_core()
 * sets it, so none of the operations on it need to be atomic.
 */
601

602
/* Page flags: | [SECTION] | [NODE] | ZONE | [LAST_CPUPID] | ... | FLAGS | */
603
#define SECTIONS_PGOFF		((sizeof(unsigned long)*8) - SECTIONS_WIDTH)
A
Andy Whitcroft 已提交
604 605
#define NODES_PGOFF		(SECTIONS_PGOFF - NODES_WIDTH)
#define ZONES_PGOFF		(NODES_PGOFF - ZONES_WIDTH)
606
#define LAST_CPUPID_PGOFF	(ZONES_PGOFF - LAST_CPUPID_WIDTH)
A
Andy Whitcroft 已提交
607

608
/*
L
Lucas De Marchi 已提交
609
 * Define the bit shifts to access each section.  For non-existent
610 611 612
 * sections we define the shift as 0; that plus a 0 mask ensures
 * the compiler will optimise away reference to them.
 */
A
Andy Whitcroft 已提交
613 614 615
#define SECTIONS_PGSHIFT	(SECTIONS_PGOFF * (SECTIONS_WIDTH != 0))
#define NODES_PGSHIFT		(NODES_PGOFF * (NODES_WIDTH != 0))
#define ZONES_PGSHIFT		(ZONES_PGOFF * (ZONES_WIDTH != 0))
616
#define LAST_CPUPID_PGSHIFT	(LAST_CPUPID_PGOFF * (LAST_CPUPID_WIDTH != 0))
617

618 619
/* NODE:ZONE or SECTION:ZONE is used to ID a zone for the buddy allocator */
#ifdef NODE_NOT_IN_PAGE_FLAGS
620
#define ZONEID_SHIFT		(SECTIONS_SHIFT + ZONES_SHIFT)
621 622
#define ZONEID_PGOFF		((SECTIONS_PGOFF < ZONES_PGOFF)? \
						SECTIONS_PGOFF : ZONES_PGOFF)
A
Andy Whitcroft 已提交
623
#else
624
#define ZONEID_SHIFT		(NODES_SHIFT + ZONES_SHIFT)
625 626
#define ZONEID_PGOFF		((NODES_PGOFF < ZONES_PGOFF)? \
						NODES_PGOFF : ZONES_PGOFF)
627 628
#endif

629
#define ZONEID_PGSHIFT		(ZONEID_PGOFF * (ZONEID_SHIFT != 0))
630

631 632
#if SECTIONS_WIDTH+NODES_WIDTH+ZONES_WIDTH > BITS_PER_LONG - NR_PAGEFLAGS
#error SECTIONS_WIDTH+NODES_WIDTH+ZONES_WIDTH > BITS_PER_LONG - NR_PAGEFLAGS
633 634
#endif

A
Andy Whitcroft 已提交
635 636 637
#define ZONES_MASK		((1UL << ZONES_WIDTH) - 1)
#define NODES_MASK		((1UL << NODES_WIDTH) - 1)
#define SECTIONS_MASK		((1UL << SECTIONS_WIDTH) - 1)
638
#define LAST_CPUPID_MASK	((1UL << LAST_CPUPID_WIDTH) - 1)
639
#define ZONEID_MASK		((1UL << ZONEID_SHIFT) - 1)
640

I
Ian Campbell 已提交
641
static inline enum zone_type page_zonenum(const struct page *page)
L
Linus Torvalds 已提交
642
{
643
	return (page->flags >> ZONES_PGSHIFT) & ZONES_MASK;
L
Linus Torvalds 已提交
644 645
}

C
Cody P Schafer 已提交
646 647 648 649
#if defined(CONFIG_SPARSEMEM) && !defined(CONFIG_SPARSEMEM_VMEMMAP)
#define SECTION_IN_PAGE_FLAGS
#endif

650
/*
651 652 653 654 655 656
 * The identification function is mainly used by the buddy allocator for
 * determining if two pages could be buddies. We are not really identifying
 * the zone since we could be using the section number id if we do not have
 * node id available in page flags.
 * We only guarantee that it will return the same value for two combinable
 * pages in a zone.
657
 */
658 659
static inline int page_zone_id(struct page *page)
{
660
	return (page->flags >> ZONEID_PGSHIFT) & ZONEID_MASK;
661 662
}

663
static inline int zone_to_nid(struct zone *zone)
664
{
665 666 667 668 669
#ifdef CONFIG_NUMA
	return zone->node;
#else
	return 0;
#endif
670 671
}

672
#ifdef NODE_NOT_IN_PAGE_FLAGS
I
Ian Campbell 已提交
673
extern int page_to_nid(const struct page *page);
674
#else
I
Ian Campbell 已提交
675
static inline int page_to_nid(const struct page *page)
A
Andy Whitcroft 已提交
676
{
677
	return (page->flags >> NODES_PGSHIFT) & NODES_MASK;
A
Andy Whitcroft 已提交
678
}
679 680
#endif

681
#ifdef CONFIG_NUMA_BALANCING
682
static inline int cpu_pid_to_cpupid(int cpu, int pid)
683
{
684
	return ((cpu & LAST__CPU_MASK) << LAST__PID_SHIFT) | (pid & LAST__PID_MASK);
685 686
}

687
static inline int cpupid_to_pid(int cpupid)
688
{
689
	return cpupid & LAST__PID_MASK;
690
}
691

692
static inline int cpupid_to_cpu(int cpupid)
693
{
694
	return (cpupid >> LAST__PID_SHIFT) & LAST__CPU_MASK;
695 696
}

697
static inline int cpupid_to_nid(int cpupid)
698
{
699
	return cpu_to_node(cpupid_to_cpu(cpupid));
700 701
}

702
static inline bool cpupid_pid_unset(int cpupid)
703
{
704
	return cpupid_to_pid(cpupid) == (-1 & LAST__PID_MASK);
705 706
}

707
static inline bool cpupid_cpu_unset(int cpupid)
708
{
709
	return cpupid_to_cpu(cpupid) == (-1 & LAST__CPU_MASK);
710 711
}

712 713 714 715 716 717
static inline bool __cpupid_match_pid(pid_t task_pid, int cpupid)
{
	return (task_pid & LAST__PID_MASK) == cpupid_to_pid(cpupid);
}

#define cpupid_match_pid(task, cpupid) __cpupid_match_pid(task->pid, cpupid)
718 719
#ifdef LAST_CPUPID_NOT_IN_PAGE_FLAGS
static inline int page_cpupid_xchg_last(struct page *page, int cpupid)
720
{
721
	return xchg(&page->_last_cpupid, cpupid);
722
}
723 724 725 726 727 728

static inline int page_cpupid_last(struct page *page)
{
	return page->_last_cpupid;
}
static inline void page_cpupid_reset_last(struct page *page)
729
{
730
	page->_last_cpupid = -1;
731 732
}
#else
733
static inline int page_cpupid_last(struct page *page)
734
{
735
	return (page->flags >> LAST_CPUPID_PGSHIFT) & LAST_CPUPID_MASK;
736 737
}

738
extern int page_cpupid_xchg_last(struct page *page, int cpupid);
739

740
static inline void page_cpupid_reset_last(struct page *page)
741
{
742
	int cpupid = (1 << LAST_CPUPID_SHIFT) - 1;
M
Mel Gorman 已提交
743

744 745
	page->flags &= ~(LAST_CPUPID_MASK << LAST_CPUPID_PGSHIFT);
	page->flags |= (cpupid & LAST_CPUPID_MASK) << LAST_CPUPID_PGSHIFT;
746
}
747 748 749
#endif /* LAST_CPUPID_NOT_IN_PAGE_FLAGS */
#else /* !CONFIG_NUMA_BALANCING */
static inline int page_cpupid_xchg_last(struct page *page, int cpupid)
750
{
751
	return page_to_nid(page); /* XXX */
752 753
}

754
static inline int page_cpupid_last(struct page *page)
755
{
756
	return page_to_nid(page); /* XXX */
757 758
}

759
static inline int cpupid_to_nid(int cpupid)
760 761 762 763
{
	return -1;
}

764
static inline int cpupid_to_pid(int cpupid)
765 766 767 768
{
	return -1;
}

769
static inline int cpupid_to_cpu(int cpupid)
770 771 772 773
{
	return -1;
}

774 775 776 777 778 779
static inline int cpu_pid_to_cpupid(int nid, int pid)
{
	return -1;
}

static inline bool cpupid_pid_unset(int cpupid)
780 781 782 783
{
	return 1;
}

784
static inline void page_cpupid_reset_last(struct page *page)
785 786
{
}
787 788 789 790 791

static inline bool cpupid_match_pid(struct task_struct *task, int cpupid)
{
	return false;
}
792
#endif /* CONFIG_NUMA_BALANCING */
793

I
Ian Campbell 已提交
794
static inline struct zone *page_zone(const struct page *page)
795 796 797 798
{
	return &NODE_DATA(page_to_nid(page))->node_zones[page_zonenum(page)];
}

C
Cody P Schafer 已提交
799
#ifdef SECTION_IN_PAGE_FLAGS
800 801 802 803 804 805
static inline void set_page_section(struct page *page, unsigned long section)
{
	page->flags &= ~(SECTIONS_MASK << SECTIONS_PGSHIFT);
	page->flags |= (section & SECTIONS_MASK) << SECTIONS_PGSHIFT;
}

806
static inline unsigned long page_to_section(const struct page *page)
A
Andy Whitcroft 已提交
807 808 809
{
	return (page->flags >> SECTIONS_PGSHIFT) & SECTIONS_MASK;
}
810
#endif
A
Andy Whitcroft 已提交
811

812
static inline void set_page_zone(struct page *page, enum zone_type zone)
813 814 815 816
{
	page->flags &= ~(ZONES_MASK << ZONES_PGSHIFT);
	page->flags |= (zone & ZONES_MASK) << ZONES_PGSHIFT;
}
817

818 819 820 821
static inline void set_page_node(struct page *page, unsigned long node)
{
	page->flags &= ~(NODES_MASK << NODES_PGSHIFT);
	page->flags |= (node & NODES_MASK) << NODES_PGSHIFT;
L
Linus Torvalds 已提交
822
}
823

824
static inline void set_page_links(struct page *page, enum zone_type zone,
A
Andy Whitcroft 已提交
825
	unsigned long node, unsigned long pfn)
L
Linus Torvalds 已提交
826
{
827 828
	set_page_zone(page, zone);
	set_page_node(page, node);
C
Cody P Schafer 已提交
829
#ifdef SECTION_IN_PAGE_FLAGS
A
Andy Whitcroft 已提交
830
	set_page_section(page, pfn_to_section_nr(pfn));
831
#endif
L
Linus Torvalds 已提交
832 833
}

834 835 836 837 838
/*
 * Some inline functions in vmstat.h depend on page_zone()
 */
#include <linux/vmstat.h>

I
Ian Campbell 已提交
839
static __always_inline void *lowmem_page_address(const struct page *page)
L
Linus Torvalds 已提交
840
{
841
	return __va(PFN_PHYS(page_to_pfn(page)));
L
Linus Torvalds 已提交
842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857
}

#if defined(CONFIG_HIGHMEM) && !defined(WANT_PAGE_VIRTUAL)
#define HASHED_PAGE_VIRTUAL
#endif

#if defined(WANT_PAGE_VIRTUAL)
#define page_address(page) ((page)->virtual)
#define set_page_address(page, address)			\
	do {						\
		(page)->virtual = (address);		\
	} while(0)
#define page_address_init()  do { } while(0)
#endif

#if defined(HASHED_PAGE_VIRTUAL)
858
void *page_address(const struct page *page);
L
Linus Torvalds 已提交
859 860 861 862 863 864 865 866 867 868 869 870 871
void set_page_address(struct page *page, void *virtual);
void page_address_init(void);
#endif

#if !defined(HASHED_PAGE_VIRTUAL) && !defined(WANT_PAGE_VIRTUAL)
#define page_address(page) lowmem_page_address(page)
#define set_page_address(page, address)  do { } while(0)
#define page_address_init()  do { } while(0)
#endif

/*
 * On an anonymous page mapped into a user virtual memory area,
 * page->mapping points to its anon_vma, not to a struct address_space;
H
Hugh Dickins 已提交
872 873 874 875 876 877 878 879
 * with the PAGE_MAPPING_ANON bit set to distinguish it.  See rmap.h.
 *
 * On an anonymous page in a VM_MERGEABLE area, if CONFIG_KSM is enabled,
 * the PAGE_MAPPING_KSM bit may be set along with the PAGE_MAPPING_ANON bit;
 * and then page->mapping points, not to an anon_vma, but to a private
 * structure which KSM associates with that merged page.  See ksm.h.
 *
 * PAGE_MAPPING_KSM without PAGE_MAPPING_ANON is currently never used.
L
Linus Torvalds 已提交
880 881 882 883 884 885
 *
 * Please note that, confusingly, "page_mapping" refers to the inode
 * address_space which maps the page from disk; whereas "page_mapped"
 * refers to user virtual address space into which the page is mapped.
 */
#define PAGE_MAPPING_ANON	1
H
Hugh Dickins 已提交
886 887
#define PAGE_MAPPING_KSM	2
#define PAGE_MAPPING_FLAGS	(PAGE_MAPPING_ANON | PAGE_MAPPING_KSM)
L
Linus Torvalds 已提交
888

S
Shaohua Li 已提交
889
extern struct address_space *page_mapping(struct page *page);
L
Linus Torvalds 已提交
890

H
Hugh Dickins 已提交
891 892 893 894 895 896
/* Neutral page->mapping pointer to address_space or anon_vma or other */
static inline void *page_rmapping(struct page *page)
{
	return (void *)((unsigned long)page->mapping & ~PAGE_MAPPING_FLAGS);
}

897 898 899 900 901 902 903 904 905 906 907
extern struct address_space *__page_file_mapping(struct page *);

static inline
struct address_space *page_file_mapping(struct page *page)
{
	if (unlikely(PageSwapCache(page)))
		return __page_file_mapping(page);

	return page->mapping;
}

L
Linus Torvalds 已提交
908 909 910 911 912 913 914 915 916 917 918 919
static inline int PageAnon(struct page *page)
{
	return ((unsigned long)page->mapping & PAGE_MAPPING_ANON) != 0;
}

/*
 * Return the pagecache index of the passed page.  Regular pagecache pages
 * use ->index whereas swapcache pages use ->private
 */
static inline pgoff_t page_index(struct page *page)
{
	if (unlikely(PageSwapCache(page)))
H
Hugh Dickins 已提交
920
		return page_private(page);
L
Linus Torvalds 已提交
921 922 923
	return page->index;
}

924 925 926 927 928 929 930 931 932 933 934 935 936 937
extern pgoff_t __page_file_index(struct page *page);

/*
 * Return the file index of the page. Regular pagecache pages use ->index
 * whereas swapcache pages use swp_offset(->private)
 */
static inline pgoff_t page_file_index(struct page *page)
{
	if (unlikely(PageSwapCache(page)))
		return __page_file_index(page);

	return page->index;
}

L
Linus Torvalds 已提交
938 939 940 941 942 943 944 945 946 947 948 949 950
/*
 * Return true if this page is mapped into pagetables.
 */
static inline int page_mapped(struct page *page)
{
	return atomic_read(&(page)->_mapcount) >= 0;
}

/*
 * Different kinds of faults, as returned by handle_mm_fault().
 * Used to decide whether a process gets delivered SIGBUS or
 * just gets major/minor fault counters bumped up.
 */
N
Nick Piggin 已提交
951

N
Nick Piggin 已提交
952
#define VM_FAULT_MINOR	0 /* For backwards compat. Remove me quickly. */
N
Nick Piggin 已提交
953

N
Nick Piggin 已提交
954 955 956 957
#define VM_FAULT_OOM	0x0001
#define VM_FAULT_SIGBUS	0x0002
#define VM_FAULT_MAJOR	0x0004
#define VM_FAULT_WRITE	0x0008	/* Special case for get_user_pages */
958 959
#define VM_FAULT_HWPOISON 0x0010	/* Hit poisoned small page */
#define VM_FAULT_HWPOISON_LARGE 0x0020  /* Hit poisoned large page. Index encoded in upper bits */
N
Nick Piggin 已提交
960

N
Nick Piggin 已提交
961 962
#define VM_FAULT_NOPAGE	0x0100	/* ->fault installed the pte, not return page */
#define VM_FAULT_LOCKED	0x0200	/* ->fault locked the returned page */
963
#define VM_FAULT_RETRY	0x0400	/* ->fault blocked, must retry */
964
#define VM_FAULT_FALLBACK 0x0800	/* huge page fault failed, fall back to small */
L
Linus Torvalds 已提交
965

966 967 968
#define VM_FAULT_HWPOISON_LARGE_MASK 0xf000 /* encodes hpage index for large hwpoison */

#define VM_FAULT_ERROR	(VM_FAULT_OOM | VM_FAULT_SIGBUS | VM_FAULT_HWPOISON | \
969
			 VM_FAULT_FALLBACK | VM_FAULT_HWPOISON_LARGE)
970 971 972 973

/* Encode hstate index for a hwpoisoned large page */
#define VM_FAULT_SET_HINDEX(x) ((x) << 12)
#define VM_FAULT_GET_HINDEX(x) (((x) >> 12) & 0xf)
N
Nick Piggin 已提交
974

975 976 977 978 979
/*
 * Can be called by the pagefault handler when it gets a VM_FAULT_OOM.
 */
extern void pagefault_out_of_memory(void);

L
Linus Torvalds 已提交
980 981
#define offset_in_page(p)	((unsigned long)(p) & ~PAGE_MASK)

982
/*
983
 * Flags passed to show_mem() and show_free_areas() to suppress output in
984 985
 * various contexts.
 */
986 987
#define SHOW_MEM_FILTER_NODES		(0x0001u)	/* disallowed nodes */
#define SHOW_MEM_FILTER_PAGE_COUNT	(0x0002u)	/* page type count */
988

989 990
extern void show_free_areas(unsigned int flags);
extern bool skip_free_areas_node(unsigned int flags, int nid);
L
Linus Torvalds 已提交
991 992 993

int shmem_zero_setup(struct vm_area_struct *);

A
Alexey Dobriyan 已提交
994
extern int can_do_mlock(void);
L
Linus Torvalds 已提交
995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007
extern int user_shm_lock(size_t, struct user_struct *);
extern void user_shm_unlock(size_t, struct user_struct *);

/*
 * Parameter block passed down to zap_pte_range in exceptional cases.
 */
struct zap_details {
	struct vm_area_struct *nonlinear_vma;	/* Check page->index if set */
	struct address_space *check_mapping;	/* Check page->mapping if set */
	pgoff_t	first_index;			/* Lowest page->index to unmap */
	pgoff_t last_index;			/* Highest page->index to unmap */
};

N
Nick Piggin 已提交
1008 1009 1010
struct page *vm_normal_page(struct vm_area_struct *vma, unsigned long addr,
		pte_t pte);

1011 1012
int zap_vma_ptes(struct vm_area_struct *vma, unsigned long address,
		unsigned long size);
A
Al Viro 已提交
1013
void zap_page_range(struct vm_area_struct *vma, unsigned long address,
L
Linus Torvalds 已提交
1014
		unsigned long size, struct zap_details *);
1015 1016
void unmap_vmas(struct mmu_gather *tlb, struct vm_area_struct *start_vma,
		unsigned long start, unsigned long end);
1017 1018 1019 1020 1021 1022

/**
 * mm_walk - callbacks for walk_page_range
 * @pgd_entry: if set, called for each non-empty PGD (top-level) entry
 * @pud_entry: if set, called for each non-empty PUD (2nd-level) entry
 * @pmd_entry: if set, called for each non-empty PMD (3rd-level) entry
1023 1024 1025
 *	       this handler is required to be able to handle
 *	       pmd_trans_huge() pmds.  They may simply choose to
 *	       split_huge_page() instead of handling it explicitly.
1026 1027
 * @pte_entry: if set, called for each non-empty PTE (4th-level) entry
 * @pte_hole: if set, called for each hole at all levels
1028
 * @hugetlb_entry: if set, called for each hugetlb entry
1029 1030
 *		   *Caution*: The caller must hold mmap_sem() if @hugetlb_entry
 * 			      is used.
1031 1032 1033 1034
 *
 * (see walk_page_range for more details)
 */
struct mm_walk {
1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047
	int (*pgd_entry)(pgd_t *pgd, unsigned long addr,
			 unsigned long next, struct mm_walk *walk);
	int (*pud_entry)(pud_t *pud, unsigned long addr,
	                 unsigned long next, struct mm_walk *walk);
	int (*pmd_entry)(pmd_t *pmd, unsigned long addr,
			 unsigned long next, struct mm_walk *walk);
	int (*pte_entry)(pte_t *pte, unsigned long addr,
			 unsigned long next, struct mm_walk *walk);
	int (*pte_hole)(unsigned long addr, unsigned long next,
			struct mm_walk *walk);
	int (*hugetlb_entry)(pte_t *pte, unsigned long hmask,
			     unsigned long addr, unsigned long next,
			     struct mm_walk *walk);
D
Dave Hansen 已提交
1048 1049
	struct mm_struct *mm;
	void *private;
1050 1051
};

D
Dave Hansen 已提交
1052 1053
int walk_page_range(unsigned long addr, unsigned long end,
		struct mm_walk *walk);
1054
void free_pgd_range(struct mmu_gather *tlb, unsigned long addr,
1055
		unsigned long end, unsigned long floor, unsigned long ceiling);
L
Linus Torvalds 已提交
1056 1057 1058 1059
int copy_page_range(struct mm_struct *dst, struct mm_struct *src,
			struct vm_area_struct *vma);
void unmap_mapping_range(struct address_space *mapping,
		loff_t const holebegin, loff_t const holelen, int even_cows);
J
Johannes Weiner 已提交
1060 1061
int follow_pfn(struct vm_area_struct *vma, unsigned long address,
	unsigned long *pfn);
1062 1063
int follow_phys(struct vm_area_struct *vma, unsigned long address,
		unsigned int flags, unsigned long *prot, resource_size_t *phys);
1064 1065
int generic_access_phys(struct vm_area_struct *vma, unsigned long addr,
			void *buf, int len, int write);
L
Linus Torvalds 已提交
1066 1067 1068 1069 1070 1071 1072

static inline void unmap_shared_mapping_range(struct address_space *mapping,
		loff_t const holebegin, loff_t const holelen)
{
	unmap_mapping_range(mapping, holebegin, holelen, 0);
}

1073
extern void truncate_pagecache(struct inode *inode, loff_t new);
1074
extern void truncate_setsize(struct inode *inode, loff_t newsize);
1075
void truncate_pagecache_range(struct inode *inode, loff_t offset, loff_t end);
1076
int truncate_inode_page(struct address_space *mapping, struct page *page);
1077
int generic_error_remove_page(struct address_space *mapping, struct page *page);
1078 1079
int invalidate_inode_page(struct page *page);

1080
#ifdef CONFIG_MMU
N
Nick Piggin 已提交
1081
extern int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma,
1082
			unsigned long address, unsigned int flags);
1083 1084
extern int fixup_user_fault(struct task_struct *tsk, struct mm_struct *mm,
			    unsigned long address, unsigned int fault_flags);
1085 1086 1087
#else
static inline int handle_mm_fault(struct mm_struct *mm,
			struct vm_area_struct *vma, unsigned long address,
1088
			unsigned int flags)
1089 1090 1091 1092 1093
{
	/* should never happen if there's no MMU */
	BUG();
	return VM_FAULT_SIGBUS;
}
1094 1095 1096 1097 1098 1099 1100 1101
static inline int fixup_user_fault(struct task_struct *tsk,
		struct mm_struct *mm, unsigned long address,
		unsigned int fault_flags)
{
	/* should never happen if there's no MMU */
	BUG();
	return -EFAULT;
}
1102
#endif
N
Nick Piggin 已提交
1103

L
Linus Torvalds 已提交
1104
extern int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write);
S
Stephen Wilson 已提交
1105 1106
extern int access_remote_vm(struct mm_struct *mm, unsigned long addr,
		void *buf, int len, int write);
L
Linus Torvalds 已提交
1107

1108 1109 1110 1111 1112 1113 1114 1115
long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
		      unsigned long start, unsigned long nr_pages,
		      unsigned int foll_flags, struct page **pages,
		      struct vm_area_struct **vmas, int *nonblocking);
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);
1116 1117
int get_user_pages_fast(unsigned long start, int nr_pages, int write,
			struct page **pages);
1118 1119 1120 1121
struct kvec;
int get_kernel_pages(const struct kvec *iov, int nr_pages, int write,
			struct page **pages);
int get_kernel_page(unsigned long start, int write, struct page **pages);
H
Hugh Dickins 已提交
1122
struct page *get_dump_page(unsigned long addr);
L
Linus Torvalds 已提交
1123

1124
extern int try_to_release_page(struct page * page, gfp_t gfp_mask);
1125 1126
extern void do_invalidatepage(struct page *page, unsigned int offset,
			      unsigned int length);
1127

L
Linus Torvalds 已提交
1128
int __set_page_dirty_nobuffers(struct page *page);
1129
int __set_page_dirty_no_writeback(struct page *page);
L
Linus Torvalds 已提交
1130 1131
int redirty_page_for_writepage(struct writeback_control *wbc,
				struct page *page);
1132
void account_page_dirtied(struct page *page, struct address_space *mapping);
M
Michael Rubin 已提交
1133
void account_page_writeback(struct page *page);
1134
int set_page_dirty(struct page *page);
L
Linus Torvalds 已提交
1135 1136 1137
int set_page_dirty_lock(struct page *page);
int clear_page_dirty_for_io(struct page *page);

1138
/* Is the vma a continuation of the stack vma above it? */
1139
static inline int vma_growsdown(struct vm_area_struct *vma, unsigned long addr)
1140 1141 1142 1143
{
	return vma && (vma->vm_end == addr) && (vma->vm_flags & VM_GROWSDOWN);
}

1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165
static inline int stack_guard_page_start(struct vm_area_struct *vma,
					     unsigned long addr)
{
	return (vma->vm_flags & VM_GROWSDOWN) &&
		(vma->vm_start == addr) &&
		!vma_growsdown(vma->vm_prev, addr);
}

/* Is the vma a continuation of the stack vma below it? */
static inline int vma_growsup(struct vm_area_struct *vma, unsigned long addr)
{
	return vma && (vma->vm_start == addr) && (vma->vm_flags & VM_GROWSUP);
}

static inline int stack_guard_page_end(struct vm_area_struct *vma,
					   unsigned long addr)
{
	return (vma->vm_flags & VM_GROWSUP) &&
		(vma->vm_end == addr) &&
		!vma_growsup(vma->vm_next, addr);
}

1166 1167 1168
extern pid_t
vm_is_stack(struct task_struct *task, struct vm_area_struct *vma, int in_group);

1169 1170
extern unsigned long move_page_tables(struct vm_area_struct *vma,
		unsigned long old_addr, struct vm_area_struct *new_vma,
1171 1172
		unsigned long new_addr, unsigned long len,
		bool need_rmap_locks);
1173 1174
extern unsigned long change_protection(struct vm_area_struct *vma, unsigned long start,
			      unsigned long end, pgprot_t newprot,
1175
			      int dirty_accountable, int prot_numa);
1176 1177 1178
extern int mprotect_fixup(struct vm_area_struct *vma,
			  struct vm_area_struct **pprev, unsigned long start,
			  unsigned long end, unsigned long newflags);
L
Linus Torvalds 已提交
1179

1180 1181 1182 1183 1184
/*
 * doesn't attempt to fault and will return short.
 */
int __get_user_pages_fast(unsigned long start, int nr_pages, int write,
			  struct page **pages);
K
KAMEZAWA Hiroyuki 已提交
1185 1186 1187 1188 1189
/*
 * per-process(per-mm_struct) statistics.
 */
static inline unsigned long get_mm_counter(struct mm_struct *mm, int member)
{
1190 1191 1192 1193 1194 1195 1196 1197 1198
	long val = atomic_long_read(&mm->rss_stat.count[member]);

#ifdef SPLIT_RSS_COUNTING
	/*
	 * counter is updated in asynchronous manner and may go to minus.
	 * But it's never be expected number for users.
	 */
	if (val < 0)
		val = 0;
1199
#endif
1200 1201
	return (unsigned long)val;
}
K
KAMEZAWA Hiroyuki 已提交
1202 1203 1204

static inline void add_mm_counter(struct mm_struct *mm, int member, long value)
{
1205
	atomic_long_add(value, &mm->rss_stat.count[member]);
K
KAMEZAWA Hiroyuki 已提交
1206 1207 1208 1209
}

static inline void inc_mm_counter(struct mm_struct *mm, int member)
{
1210
	atomic_long_inc(&mm->rss_stat.count[member]);
K
KAMEZAWA Hiroyuki 已提交
1211 1212 1213 1214
}

static inline void dec_mm_counter(struct mm_struct *mm, int member)
{
1215
	atomic_long_dec(&mm->rss_stat.count[member]);
K
KAMEZAWA Hiroyuki 已提交
1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256
}

static inline unsigned long get_mm_rss(struct mm_struct *mm)
{
	return get_mm_counter(mm, MM_FILEPAGES) +
		get_mm_counter(mm, MM_ANONPAGES);
}

static inline unsigned long get_mm_hiwater_rss(struct mm_struct *mm)
{
	return max(mm->hiwater_rss, get_mm_rss(mm));
}

static inline unsigned long get_mm_hiwater_vm(struct mm_struct *mm)
{
	return max(mm->hiwater_vm, mm->total_vm);
}

static inline void update_hiwater_rss(struct mm_struct *mm)
{
	unsigned long _rss = get_mm_rss(mm);

	if ((mm)->hiwater_rss < _rss)
		(mm)->hiwater_rss = _rss;
}

static inline void update_hiwater_vm(struct mm_struct *mm)
{
	if (mm->hiwater_vm < mm->total_vm)
		mm->hiwater_vm = mm->total_vm;
}

static inline void setmax_mm_hiwater_rss(unsigned long *maxrss,
					 struct mm_struct *mm)
{
	unsigned long hiwater_rss = get_mm_hiwater_rss(mm);

	if (*maxrss < hiwater_rss)
		*maxrss = hiwater_rss;
}

K
KAMEZAWA Hiroyuki 已提交
1257
#if defined(SPLIT_RSS_COUNTING)
1258
void sync_mm_rss(struct mm_struct *mm);
K
KAMEZAWA Hiroyuki 已提交
1259
#else
1260
static inline void sync_mm_rss(struct mm_struct *mm)
K
KAMEZAWA Hiroyuki 已提交
1261 1262 1263
{
}
#endif
1264

A
Alexey Dobriyan 已提交
1265
int vma_wants_writenotify(struct vm_area_struct *vma);
1266

1267 1268 1269 1270 1271 1272 1273 1274 1275
extern pte_t *__get_locked_pte(struct mm_struct *mm, unsigned long addr,
			       spinlock_t **ptl);
static inline pte_t *get_locked_pte(struct mm_struct *mm, unsigned long addr,
				    spinlock_t **ptl)
{
	pte_t *ptep;
	__cond_lock(*ptl, ptep = __get_locked_pte(mm, addr, ptl));
	return ptep;
}
1276

N
Nick Piggin 已提交
1277 1278 1279 1280 1281 1282 1283
#ifdef __PAGETABLE_PUD_FOLDED
static inline int __pud_alloc(struct mm_struct *mm, pgd_t *pgd,
						unsigned long address)
{
	return 0;
}
#else
1284
int __pud_alloc(struct mm_struct *mm, pgd_t *pgd, unsigned long address);
N
Nick Piggin 已提交
1285 1286 1287 1288 1289 1290 1291 1292 1293
#endif

#ifdef __PAGETABLE_PMD_FOLDED
static inline int __pmd_alloc(struct mm_struct *mm, pud_t *pud,
						unsigned long address)
{
	return 0;
}
#else
1294
int __pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address);
N
Nick Piggin 已提交
1295 1296
#endif

1297 1298
int __pte_alloc(struct mm_struct *mm, struct vm_area_struct *vma,
		pmd_t *pmd, unsigned long address);
1299 1300
int __pte_alloc_kernel(pmd_t *pmd, unsigned long address);

L
Linus Torvalds 已提交
1301 1302 1303 1304
/*
 * The following ifdef needed to get the 4level-fixup.h header to work.
 * Remove it when 4level-fixup.h has been removed.
 */
1305
#if defined(CONFIG_MMU) && !defined(__ARCH_HAS_4LEVEL_HACK)
L
Linus Torvalds 已提交
1306 1307
static inline pud_t *pud_alloc(struct mm_struct *mm, pgd_t *pgd, unsigned long address)
{
1308 1309
	return (unlikely(pgd_none(*pgd)) && __pud_alloc(mm, pgd, address))?
		NULL: pud_offset(pgd, address);
L
Linus Torvalds 已提交
1310 1311 1312 1313
}

static inline pmd_t *pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address)
{
1314 1315
	return (unlikely(pud_none(*pud)) && __pmd_alloc(mm, pud, address))?
		NULL: pmd_offset(pud, address);
L
Linus Torvalds 已提交
1316
}
1317 1318
#endif /* CONFIG_MMU && !__ARCH_HAS_4LEVEL_HACK */

1319
#if USE_SPLIT_PTE_PTLOCKS
1320
#if ALLOC_SPLIT_PTLOCKS
1321 1322 1323 1324 1325 1326 1327
extern bool ptlock_alloc(struct page *page);
extern void ptlock_free(struct page *page);

static inline spinlock_t *ptlock_ptr(struct page *page)
{
	return page->ptl;
}
1328
#else /* ALLOC_SPLIT_PTLOCKS */
1329 1330 1331 1332
static inline bool ptlock_alloc(struct page *page)
{
	return true;
}
1333

1334 1335 1336 1337 1338 1339
static inline void ptlock_free(struct page *page)
{
}

static inline spinlock_t *ptlock_ptr(struct page *page)
{
1340
	return &page->ptl;
1341
}
1342
#endif /* ALLOC_SPLIT_PTLOCKS */
1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358

static inline spinlock_t *pte_lockptr(struct mm_struct *mm, pmd_t *pmd)
{
	return ptlock_ptr(pmd_page(*pmd));
}

static inline bool ptlock_init(struct page *page)
{
	/*
	 * prep_new_page() initialize page->private (and therefore page->ptl)
	 * with 0. Make sure nobody took it in use in between.
	 *
	 * It can happen if arch try to use slab for page table allocation:
	 * slab code uses page->slab_cache and page->first_page (for tail
	 * pages), which share storage with page->ptl.
	 */
1359
	VM_BUG_ON(*(unsigned long *)&page->ptl);
1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372
	if (!ptlock_alloc(page))
		return false;
	spin_lock_init(ptlock_ptr(page));
	return true;
}

/* Reset page->mapping so free_pages_check won't complain. */
static inline void pte_lock_deinit(struct page *page)
{
	page->mapping = NULL;
	ptlock_free(page);
}

1373
#else	/* !USE_SPLIT_PTE_PTLOCKS */
H
Hugh Dickins 已提交
1374 1375 1376
/*
 * We use mm->page_table_lock to guard all pagetable pages of the mm.
 */
1377 1378 1379 1380 1381 1382
static inline spinlock_t *pte_lockptr(struct mm_struct *mm, pmd_t *pmd)
{
	return &mm->page_table_lock;
}
static inline bool ptlock_init(struct page *page) { return true; }
static inline void pte_lock_deinit(struct page *page) {}
1383
#endif /* USE_SPLIT_PTE_PTLOCKS */
H
Hugh Dickins 已提交
1384

1385
static inline bool pgtable_page_ctor(struct page *page)
1386 1387
{
	inc_zone_page_state(page, NR_PAGETABLE);
1388
	return ptlock_init(page);
1389 1390 1391 1392 1393 1394 1395 1396
}

static inline void pgtable_page_dtor(struct page *page)
{
	pte_lock_deinit(page);
	dec_zone_page_state(page, NR_PAGETABLE);
}

H
Hugh Dickins 已提交
1397 1398
#define pte_offset_map_lock(mm, pmd, address, ptlp)	\
({							\
H
Hugh Dickins 已提交
1399
	spinlock_t *__ptl = pte_lockptr(mm, pmd);	\
H
Hugh Dickins 已提交
1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410
	pte_t *__pte = pte_offset_map(pmd, address);	\
	*(ptlp) = __ptl;				\
	spin_lock(__ptl);				\
	__pte;						\
})

#define pte_unmap_unlock(pte, ptl)	do {		\
	spin_unlock(ptl);				\
	pte_unmap(pte);					\
} while (0)

1411 1412 1413 1414
#define pte_alloc_map(mm, vma, pmd, address)				\
	((unlikely(pmd_none(*(pmd))) && __pte_alloc(mm, vma,	\
							pmd, address))?	\
	 NULL: pte_offset_map(pmd, address))
1415

H
Hugh Dickins 已提交
1416
#define pte_alloc_map_lock(mm, pmd, address, ptlp)	\
1417 1418
	((unlikely(pmd_none(*(pmd))) && __pte_alloc(mm, NULL,	\
							pmd, address))?	\
H
Hugh Dickins 已提交
1419 1420
		NULL: pte_offset_map_lock(mm, pmd, address, ptlp))

1421
#define pte_alloc_kernel(pmd, address)			\
1422
	((unlikely(pmd_none(*(pmd))) && __pte_alloc_kernel(pmd, address))? \
1423
		NULL: pte_offset_kernel(pmd, address))
L
Linus Torvalds 已提交
1424

1425 1426 1427 1428
#if USE_SPLIT_PMD_PTLOCKS

static inline spinlock_t *pmd_lockptr(struct mm_struct *mm, pmd_t *pmd)
{
1429
	return ptlock_ptr(virt_to_page(pmd));
1430 1431 1432 1433 1434 1435 1436
}

static inline bool pgtable_pmd_page_ctor(struct page *page)
{
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
	page->pmd_huge_pte = NULL;
#endif
1437
	return ptlock_init(page);
1438 1439 1440 1441 1442 1443 1444
}

static inline void pgtable_pmd_page_dtor(struct page *page)
{
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
	VM_BUG_ON(page->pmd_huge_pte);
#endif
1445
	ptlock_free(page);
1446 1447 1448 1449 1450 1451
}

#define pmd_huge_pte(mm, pmd) (virt_to_page(pmd)->pmd_huge_pte)

#else

1452 1453 1454 1455 1456
static inline spinlock_t *pmd_lockptr(struct mm_struct *mm, pmd_t *pmd)
{
	return &mm->page_table_lock;
}

1457 1458 1459
static inline bool pgtable_pmd_page_ctor(struct page *page) { return true; }
static inline void pgtable_pmd_page_dtor(struct page *page) {}

1460
#define pmd_huge_pte(mm, pmd) ((mm)->pmd_huge_pte)
1461

1462 1463
#endif

1464 1465 1466 1467 1468 1469 1470
static inline spinlock_t *pmd_lock(struct mm_struct *mm, pmd_t *pmd)
{
	spinlock_t *ptl = pmd_lockptr(mm, pmd);
	spin_lock(ptl);
	return ptl;
}

L
Linus Torvalds 已提交
1471
extern void free_area_init(unsigned long * zones_size);
1472 1473
extern void free_area_init_node(int nid, unsigned long * zones_size,
		unsigned long zone_start_pfn, unsigned long *zholes_size);
1474 1475
extern void free_initmem(void);

1476 1477 1478
/*
 * Free reserved pages within range [PAGE_ALIGN(start), end & PAGE_MASK)
 * into the buddy system. The freed pages will be poisoned with pattern
1479
 * "poison" if it's within range [0, UCHAR_MAX].
1480 1481
 * Return pages freed into the buddy system.
 */
1482
extern unsigned long free_reserved_area(void *start, void *end,
1483
					int poison, char *s);
1484

1485 1486 1487 1488 1489 1490 1491
#ifdef	CONFIG_HIGHMEM
/*
 * Free a highmem page into the buddy system, adjusting totalhigh_pages
 * and totalram_pages.
 */
extern void free_highmem_page(struct page *page);
#endif
1492

1493
extern void adjust_managed_page_count(struct page *page, long count);
1494
extern void mem_init_print_info(const char *str);
1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517

/* Free the reserved page into the buddy system, so it gets managed. */
static inline void __free_reserved_page(struct page *page)
{
	ClearPageReserved(page);
	init_page_count(page);
	__free_page(page);
}

static inline void free_reserved_page(struct page *page)
{
	__free_reserved_page(page);
	adjust_managed_page_count(page, 1);
}

static inline void mark_page_reserved(struct page *page)
{
	SetPageReserved(page);
	adjust_managed_page_count(page, -1);
}

/*
 * Default method to free all the __init memory into the buddy system.
1518 1519 1520
 * The freed pages will be poisoned with pattern "poison" if it's within
 * range [0, UCHAR_MAX].
 * Return pages freed into the buddy system.
1521 1522 1523 1524 1525
 */
static inline unsigned long free_initmem_default(int poison)
{
	extern char __init_begin[], __init_end[];

1526
	return free_reserved_area(&__init_begin, &__init_end,
1527 1528 1529
				  poison, "unused kernel");
}

1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540
static inline unsigned long get_num_physpages(void)
{
	int nid;
	unsigned long phys_pages = 0;

	for_each_online_node(nid)
		phys_pages += node_present_pages(nid);

	return phys_pages;
}

T
Tejun Heo 已提交
1541
#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP
1542
/*
T
Tejun Heo 已提交
1543
 * With CONFIG_HAVE_MEMBLOCK_NODE_MAP set, an architecture may initialise its
1544 1545 1546 1547 1548 1549
 * zones, allocate the backing mem_map and account for memory holes in a more
 * architecture independent manner. This is a substitute for creating the
 * zone_sizes[] and zholes_size[] arrays and passing them to
 * free_area_init_node()
 *
 * An architecture is expected to register range of page frames backed by
T
Tejun Heo 已提交
1550
 * physical memory with memblock_add[_node]() before calling
1551 1552 1553 1554 1555 1556
 * free_area_init_nodes() passing in the PFN each zone ends at. At a basic
 * usage, an architecture is expected to do something like
 *
 * unsigned long max_zone_pfns[MAX_NR_ZONES] = {max_dma, max_normal_pfn,
 * 							 max_highmem_pfn};
 * for_each_valid_physical_page_range()
T
Tejun Heo 已提交
1557
 * 	memblock_add_node(base, size, nid)
1558 1559
 * free_area_init_nodes(max_zone_pfns);
 *
T
Tejun Heo 已提交
1560 1561 1562 1563
 * free_bootmem_with_active_regions() calls free_bootmem_node() for each
 * registered physical page range.  Similarly
 * sparse_memory_present_with_active_regions() calls memory_present() for
 * each range when SPARSEMEM is enabled.
1564 1565
 *
 * See mm/page_alloc.c for more information on each function exposed by
T
Tejun Heo 已提交
1566
 * CONFIG_HAVE_MEMBLOCK_NODE_MAP.
1567 1568
 */
extern void free_area_init_nodes(unsigned long *max_zone_pfn);
1569
unsigned long node_map_pfn_alignment(void);
1570 1571
unsigned long __absent_pages_in_range(int nid, unsigned long start_pfn,
						unsigned long end_pfn);
1572 1573 1574 1575 1576 1577 1578 1579
extern unsigned long absent_pages_in_range(unsigned long start_pfn,
						unsigned long end_pfn);
extern void get_pfn_range_for_nid(unsigned int nid,
			unsigned long *start_pfn, unsigned long *end_pfn);
extern unsigned long find_min_pfn_with_active_regions(void);
extern void free_bootmem_with_active_regions(int nid,
						unsigned long max_low_pfn);
extern void sparse_memory_present_with_active_regions(int nid);
1580

T
Tejun Heo 已提交
1581
#endif /* CONFIG_HAVE_MEMBLOCK_NODE_MAP */
1582

T
Tejun Heo 已提交
1583
#if !defined(CONFIG_HAVE_MEMBLOCK_NODE_MAP) && \
1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597
    !defined(CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID)
static inline int __early_pfn_to_nid(unsigned long pfn)
{
	return 0;
}
#else
/* please see mm/page_alloc.c */
extern int __meminit early_pfn_to_nid(unsigned long pfn);
#ifdef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID
/* there is a per-arch backend function. */
extern int __meminit __early_pfn_to_nid(unsigned long pfn);
#endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */
#endif

1598
extern void set_dma_reserve(unsigned long new_dma_reserve);
D
Dave Hansen 已提交
1599 1600
extern void memmap_init_zone(unsigned long, int, unsigned long,
				unsigned long, enum memmap_context);
1601
extern void setup_per_zone_wmarks(void);
1602
extern int __meminit init_per_zone_wmark_min(void);
L
Linus Torvalds 已提交
1603
extern void mem_init(void);
1604
extern void __init mmap_init(void);
1605
extern void show_mem(unsigned int flags);
L
Linus Torvalds 已提交
1606 1607 1608
extern void si_meminfo(struct sysinfo * val);
extern void si_meminfo_node(struct sysinfo *val, int nid);

J
Joe Perches 已提交
1609 1610
extern __printf(3, 4)
void warn_alloc_failed(gfp_t gfp_mask, int order, const char *fmt, ...);
1611

1612 1613
extern void setup_per_cpu_pageset(void);

1614
extern void zone_pcp_update(struct zone *zone);
1615
extern void zone_pcp_reset(struct zone *zone);
1616

1617 1618 1619
/* page_alloc.c */
extern int min_free_kbytes;

1620
/* nommu.c */
1621
extern atomic_long_t mmap_pages_allocated;
1622
extern int nommu_shrink_inode_mappings(struct inode *, size_t, size_t);
1623

1624 1625 1626
/* interval_tree.c */
void vma_interval_tree_insert(struct vm_area_struct *node,
			      struct rb_root *root);
M
Michel Lespinasse 已提交
1627 1628 1629
void vma_interval_tree_insert_after(struct vm_area_struct *node,
				    struct vm_area_struct *prev,
				    struct rb_root *root);
1630 1631 1632 1633 1634 1635 1636 1637 1638 1639
void vma_interval_tree_remove(struct vm_area_struct *node,
			      struct rb_root *root);
struct vm_area_struct *vma_interval_tree_iter_first(struct rb_root *root,
				unsigned long start, unsigned long last);
struct vm_area_struct *vma_interval_tree_iter_next(struct vm_area_struct *node,
				unsigned long start, unsigned long last);

#define vma_interval_tree_foreach(vma, root, start, last)		\
	for (vma = vma_interval_tree_iter_first(root, start, last);	\
	     vma; vma = vma_interval_tree_iter_next(vma, start, last))
L
Linus Torvalds 已提交
1640 1641 1642 1643

static inline void vma_nonlinear_insert(struct vm_area_struct *vma,
					struct list_head *list)
{
1644
	list_add_tail(&vma->shared.nonlinear, list);
L
Linus Torvalds 已提交
1645 1646
}

1647 1648 1649 1650 1651 1652 1653 1654
void anon_vma_interval_tree_insert(struct anon_vma_chain *node,
				   struct rb_root *root);
void anon_vma_interval_tree_remove(struct anon_vma_chain *node,
				   struct rb_root *root);
struct anon_vma_chain *anon_vma_interval_tree_iter_first(
	struct rb_root *root, unsigned long start, unsigned long last);
struct anon_vma_chain *anon_vma_interval_tree_iter_next(
	struct anon_vma_chain *node, unsigned long start, unsigned long last);
1655 1656 1657
#ifdef CONFIG_DEBUG_VM_RB
void anon_vma_interval_tree_verify(struct anon_vma_chain *node);
#endif
1658 1659 1660 1661 1662

#define anon_vma_interval_tree_foreach(avc, root, start, last)		 \
	for (avc = anon_vma_interval_tree_iter_first(root, start, last); \
	     avc; avc = anon_vma_interval_tree_iter_next(avc, start, last))

L
Linus Torvalds 已提交
1663
/* mmap.c */
1664
extern int __vm_enough_memory(struct mm_struct *mm, long pages, int cap_sys_admin);
1665
extern int vma_adjust(struct vm_area_struct *vma, unsigned long start,
L
Linus Torvalds 已提交
1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676
	unsigned long end, pgoff_t pgoff, struct vm_area_struct *insert);
extern struct vm_area_struct *vma_merge(struct mm_struct *,
	struct vm_area_struct *prev, unsigned long addr, unsigned long end,
	unsigned long vm_flags, struct anon_vma *, struct file *, pgoff_t,
	struct mempolicy *);
extern struct anon_vma *find_mergeable_anon_vma(struct vm_area_struct *);
extern int split_vma(struct mm_struct *,
	struct vm_area_struct *, unsigned long addr, int new_below);
extern int insert_vm_struct(struct mm_struct *, struct vm_area_struct *);
extern void __vma_link_rb(struct mm_struct *, struct vm_area_struct *,
	struct rb_node **, struct rb_node *);
1677
extern void unlink_file_vma(struct vm_area_struct *);
L
Linus Torvalds 已提交
1678
extern struct vm_area_struct *copy_vma(struct vm_area_struct **,
1679 1680
	unsigned long addr, unsigned long len, pgoff_t pgoff,
	bool *need_rmap_locks);
L
Linus Torvalds 已提交
1681
extern void exit_mmap(struct mm_struct *);
M
Matt Helsley 已提交
1682

1683 1684 1685
extern int mm_take_all_locks(struct mm_struct *mm);
extern void mm_drop_all_locks(struct mm_struct *mm);

1686 1687
extern void set_mm_exe_file(struct mm_struct *mm, struct file *new_exe_file);
extern struct file *get_mm_exe_file(struct mm_struct *mm);
M
Matt Helsley 已提交
1688

A
akpm@osdl.org 已提交
1689
extern int may_expand_vm(struct mm_struct *mm, unsigned long npages);
1690 1691 1692
extern int install_special_mapping(struct mm_struct *mm,
				   unsigned long addr, unsigned long len,
				   unsigned long flags, struct page **pages);
L
Linus Torvalds 已提交
1693 1694 1695

extern unsigned long get_unmapped_area(struct file *, unsigned long, unsigned long, unsigned long, unsigned long);

M
Miklos Szeredi 已提交
1696
extern unsigned long mmap_region(struct file *file, unsigned long addr,
1697
	unsigned long len, vm_flags_t vm_flags, unsigned long pgoff);
1698 1699
extern unsigned long do_mmap_pgoff(struct file *file, unsigned long addr,
	unsigned long len, unsigned long prot, unsigned long flags,
1700
	unsigned long pgoff, unsigned long *populate);
L
Linus Torvalds 已提交
1701 1702
extern int do_munmap(struct mm_struct *, unsigned long, size_t);

1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714
#ifdef CONFIG_MMU
extern int __mm_populate(unsigned long addr, unsigned long len,
			 int ignore_errors);
static inline void mm_populate(unsigned long addr, unsigned long len)
{
	/* Ignore errors */
	(void) __mm_populate(addr, len, 1);
}
#else
static inline void mm_populate(unsigned long addr, unsigned long len) {}
#endif

1715 1716
/* These take the mm semaphore themselves */
extern unsigned long vm_brk(unsigned long, unsigned long);
A
Al Viro 已提交
1717
extern int vm_munmap(unsigned long, size_t);
1718 1719 1720
extern unsigned long vm_mmap(struct file *, unsigned long,
        unsigned long, unsigned long,
        unsigned long, unsigned long);
L
Linus Torvalds 已提交
1721

1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752
struct vm_unmapped_area_info {
#define VM_UNMAPPED_AREA_TOPDOWN 1
	unsigned long flags;
	unsigned long length;
	unsigned long low_limit;
	unsigned long high_limit;
	unsigned long align_mask;
	unsigned long align_offset;
};

extern unsigned long unmapped_area(struct vm_unmapped_area_info *info);
extern unsigned long unmapped_area_topdown(struct vm_unmapped_area_info *info);

/*
 * Search for an unmapped address range.
 *
 * We are looking for a range that:
 * - does not intersect with any VMA;
 * - is contained within the [low_limit, high_limit) interval;
 * - is at least the desired size.
 * - satisfies (begin_addr & align_mask) == (align_offset & align_mask)
 */
static inline unsigned long
vm_unmapped_area(struct vm_unmapped_area_info *info)
{
	if (!(info->flags & VM_UNMAPPED_AREA_TOPDOWN))
		return unmapped_area(info);
	else
		return unmapped_area_topdown(info);
}

1753
/* truncate.c */
L
Linus Torvalds 已提交
1754
extern void truncate_inode_pages(struct address_space *, loff_t);
1755 1756
extern void truncate_inode_pages_range(struct address_space *,
				       loff_t lstart, loff_t lend);
L
Linus Torvalds 已提交
1757 1758

/* generic vm_area_ops exported for stackable file systems */
N
Nick Piggin 已提交
1759
extern int filemap_fault(struct vm_area_struct *, struct vm_fault *);
1760
extern int filemap_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf);
L
Linus Torvalds 已提交
1761 1762 1763

/* mm/page-writeback.c */
int write_one_page(struct page *page, int wait);
N
Nick Piggin 已提交
1764
void task_dirty_inc(struct task_struct *tsk);
L
Linus Torvalds 已提交
1765 1766 1767 1768 1769 1770

/* readahead.c */
#define VM_MAX_READAHEAD	128	/* kbytes */
#define VM_MIN_READAHEAD	16	/* kbytes (includes current page) */

int force_page_cache_readahead(struct address_space *mapping, struct file *filp,
A
Andrew Morton 已提交
1771
			pgoff_t offset, unsigned long nr_to_read);
1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785

void page_cache_sync_readahead(struct address_space *mapping,
			       struct file_ra_state *ra,
			       struct file *filp,
			       pgoff_t offset,
			       unsigned long size);

void page_cache_async_readahead(struct address_space *mapping,
				struct file_ra_state *ra,
				struct file *filp,
				struct page *pg,
				pgoff_t offset,
				unsigned long size);

L
Linus Torvalds 已提交
1786
unsigned long max_sane_readahead(unsigned long nr);
1787 1788 1789
unsigned long ra_submit(struct file_ra_state *ra,
			struct address_space *mapping,
			struct file *filp);
L
Linus Torvalds 已提交
1790

1791
/* Generic expand stack which grows the stack according to GROWS{UP,DOWN} */
H
Hugh Dickins 已提交
1792
extern int expand_stack(struct vm_area_struct *vma, unsigned long address);
1793 1794 1795 1796

/* CONFIG_STACK_GROWSUP still needs to to grow downwards at some places */
extern int expand_downwards(struct vm_area_struct *vma,
		unsigned long address);
1797
#if VM_GROWSUP
H
Hugh Dickins 已提交
1798
extern int expand_upwards(struct vm_area_struct *vma, unsigned long address);
1799 1800
#else
  #define expand_upwards(vma, address) do { } while (0)
1801
#endif
L
Linus Torvalds 已提交
1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823

/* Look up the first VMA which satisfies  addr < vm_end,  NULL if none. */
extern struct vm_area_struct * find_vma(struct mm_struct * mm, unsigned long addr);
extern struct vm_area_struct * find_vma_prev(struct mm_struct * mm, unsigned long addr,
					     struct vm_area_struct **pprev);

/* Look up the first VMA which intersects the interval start_addr..end_addr-1,
   NULL if none.  Assume start_addr < end_addr. */
static inline struct vm_area_struct * find_vma_intersection(struct mm_struct * mm, unsigned long start_addr, unsigned long end_addr)
{
	struct vm_area_struct * vma = find_vma(mm,start_addr);

	if (vma && end_addr <= vma->vm_start)
		vma = NULL;
	return vma;
}

static inline unsigned long vma_pages(struct vm_area_struct *vma)
{
	return (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
}

1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835
/* Look up the first VMA which exactly match the interval vm_start ... vm_end */
static inline struct vm_area_struct *find_exact_vma(struct mm_struct *mm,
				unsigned long vm_start, unsigned long vm_end)
{
	struct vm_area_struct *vma = find_vma(mm, vm_start);

	if (vma && (vma->vm_start != vm_start || vma->vm_end != vm_end))
		vma = NULL;

	return vma;
}

1836
#ifdef CONFIG_MMU
1837
pgprot_t vm_get_page_prot(unsigned long vm_flags);
1838 1839 1840 1841 1842 1843 1844
#else
static inline pgprot_t vm_get_page_prot(unsigned long vm_flags)
{
	return __pgprot(0);
}
#endif

L
Lee Schermerhorn 已提交
1845
#ifdef CONFIG_ARCH_USES_NUMA_PROT_NONE
1846
unsigned long change_prot_numa(struct vm_area_struct *vma,
L
Lee Schermerhorn 已提交
1847 1848 1849
			unsigned long start, unsigned long end);
#endif

1850 1851 1852
struct vm_area_struct *find_extend_vma(struct mm_struct *, unsigned long addr);
int remap_pfn_range(struct vm_area_struct *, unsigned long addr,
			unsigned long pfn, unsigned long size, pgprot_t);
1853
int vm_insert_page(struct vm_area_struct *, unsigned long addr, struct page *);
N
Nick Piggin 已提交
1854 1855
int vm_insert_pfn(struct vm_area_struct *vma, unsigned long addr,
			unsigned long pfn);
N
Nick Piggin 已提交
1856 1857
int vm_insert_mixed(struct vm_area_struct *vma, unsigned long addr,
			unsigned long pfn);
1858 1859
int vm_iomap_memory(struct vm_area_struct *vma, phys_addr_t start, unsigned long len);

1860

1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871
struct page *follow_page_mask(struct vm_area_struct *vma,
			      unsigned long address, unsigned int foll_flags,
			      unsigned int *page_mask);

static inline struct page *follow_page(struct vm_area_struct *vma,
		unsigned long address, unsigned int foll_flags)
{
	unsigned int unused_page_mask;
	return follow_page_mask(vma, address, foll_flags, &unused_page_mask);
}

1872 1873 1874
#define FOLL_WRITE	0x01	/* check pte is writable */
#define FOLL_TOUCH	0x02	/* mark page accessed */
#define FOLL_GET	0x04	/* do get_page on page */
H
Hugh Dickins 已提交
1875
#define FOLL_DUMP	0x08	/* give error on hole if it would be zero */
H
Hugh Dickins 已提交
1876
#define FOLL_FORCE	0x10	/* get_user_pages read/write w/o permission */
1877 1878
#define FOLL_NOWAIT	0x20	/* if a disk transfer is needed, start the IO
				 * and return without waiting upon it */
1879
#define FOLL_MLOCK	0x40	/* mark page as mlocked */
1880
#define FOLL_SPLIT	0x80	/* don't return transhuge pages, split them */
1881
#define FOLL_HWPOISON	0x100	/* check page is hwpoisoned */
1882
#define FOLL_NUMA	0x200	/* force NUMA hinting page fault */
1883
#define FOLL_MIGRATION	0x400	/* wait for page to replace migration entry */
L
Linus Torvalds 已提交
1884

1885
typedef int (*pte_fn_t)(pte_t *pte, pgtable_t token, unsigned long addr,
1886 1887 1888 1889
			void *data);
extern int apply_to_page_range(struct mm_struct *mm, unsigned long address,
			       unsigned long size, pte_fn_t fn, void *data);

L
Linus Torvalds 已提交
1890
#ifdef CONFIG_PROC_FS
1891
void vm_stat_account(struct mm_struct *, unsigned long, struct file *, long);
L
Linus Torvalds 已提交
1892
#else
1893
static inline void vm_stat_account(struct mm_struct *mm,
L
Linus Torvalds 已提交
1894 1895
			unsigned long flags, struct file *file, long pages)
{
1896
	mm->total_vm += pages;
L
Linus Torvalds 已提交
1897 1898 1899
}
#endif /* CONFIG_PROC_FS */

1900 1901
#ifdef CONFIG_DEBUG_PAGEALLOC
extern void kernel_map_pages(struct page *page, int numpages, int enable);
1902 1903 1904
#ifdef CONFIG_HIBERNATION
extern bool kernel_page_present(struct page *page);
#endif /* CONFIG_HIBERNATION */
1905
#else
L
Linus Torvalds 已提交
1906
static inline void
N
Nick Piggin 已提交
1907
kernel_map_pages(struct page *page, int numpages, int enable) {}
1908 1909 1910
#ifdef CONFIG_HIBERNATION
static inline bool kernel_page_present(struct page *page) { return true; }
#endif /* CONFIG_HIBERNATION */
L
Linus Torvalds 已提交
1911 1912
#endif

1913
extern struct vm_area_struct *get_gate_vma(struct mm_struct *mm);
L
Linus Torvalds 已提交
1914
#ifdef	__HAVE_ARCH_GATE_AREA
1915
int in_gate_area_no_mm(unsigned long addr);
1916
int in_gate_area(struct mm_struct *mm, unsigned long addr);
L
Linus Torvalds 已提交
1917
#else
1918 1919
int in_gate_area_no_mm(unsigned long addr);
#define in_gate_area(mm, addr) ({(void)mm; in_gate_area_no_mm(addr);})
L
Linus Torvalds 已提交
1920 1921
#endif	/* __HAVE_ARCH_GATE_AREA */

1922 1923
#ifdef CONFIG_SYSCTL
extern int sysctl_drop_caches;
1924
int drop_caches_sysctl_handler(struct ctl_table *, int,
A
Andrew Morton 已提交
1925
					void __user *, size_t *, loff_t *);
1926 1927
#endif

1928
unsigned long shrink_slab(struct shrink_control *shrink,
1929 1930
			  unsigned long nr_pages_scanned,
			  unsigned long lru_pages);
A
Andrew Morton 已提交
1931

1932 1933 1934
#ifndef CONFIG_MMU
#define randomize_va_space 0
#else
1935
extern int randomize_va_space;
1936
#endif
1937

1938
const char * arch_vma_name(struct vm_area_struct *vma);
1939
void print_vma_addr(char *prefix, unsigned long rip);
1940

1941 1942 1943 1944 1945 1946
void sparse_mem_maps_populate_node(struct page **map_map,
				   unsigned long pnum_begin,
				   unsigned long pnum_end,
				   unsigned long map_count,
				   int nodeid);

1947
struct page *sparse_mem_map_populate(unsigned long pnum, int nid);
1948 1949 1950 1951
pgd_t *vmemmap_pgd_populate(unsigned long addr, int node);
pud_t *vmemmap_pud_populate(pgd_t *pgd, unsigned long addr, int node);
pmd_t *vmemmap_pmd_populate(pud_t *pud, unsigned long addr, int node);
pte_t *vmemmap_pte_populate(pmd_t *pmd, unsigned long addr, int node);
1952
void *vmemmap_alloc_block(unsigned long size, int node);
1953
void *vmemmap_alloc_block_buf(unsigned long size, int node);
1954
void vmemmap_verify(pte_t *, int, unsigned long, unsigned long);
1955 1956 1957
int vmemmap_populate_basepages(unsigned long start, unsigned long end,
			       int node);
int vmemmap_populate(unsigned long start, unsigned long end, int node);
1958
void vmemmap_populate_print_last(void);
1959
#ifdef CONFIG_MEMORY_HOTPLUG
1960
void vmemmap_free(unsigned long start, unsigned long end);
1961
#endif
1962 1963
void register_page_bootmem_memmap(unsigned long section_nr, struct page *map,
				  unsigned long size);
1964

1965 1966
enum mf_flags {
	MF_COUNT_INCREASED = 1 << 0,
1967
	MF_ACTION_REQUIRED = 1 << 1,
1968
	MF_MUST_KILL = 1 << 2,
1969
	MF_SOFT_OFFLINE = 1 << 3,
1970
};
1971
extern int memory_failure(unsigned long pfn, int trapno, int flags);
1972
extern void memory_failure_queue(unsigned long pfn, int trapno, int flags);
W
Wu Fengguang 已提交
1973
extern int unpoison_memory(unsigned long pfn);
1974 1975
extern int sysctl_memory_failure_early_kill;
extern int sysctl_memory_failure_recovery;
1976
extern void shake_page(struct page *p, int access);
1977
extern atomic_long_t num_poisoned_pages;
1978
extern int soft_offline_page(struct page *page, int flags);
1979

1980 1981
extern void dump_page(struct page *page);

A
Andrea Arcangeli 已提交
1982 1983 1984 1985 1986 1987 1988 1989 1990
#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLBFS)
extern void clear_huge_page(struct page *page,
			    unsigned long addr,
			    unsigned int pages_per_huge_page);
extern 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);
#endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLBFS */

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
#ifdef CONFIG_DEBUG_PAGEALLOC
extern unsigned int _debug_guardpage_minorder;

static inline unsigned int debug_guardpage_minorder(void)
{
	return _debug_guardpage_minorder;
}

static inline bool page_is_guard(struct page *page)
{
	return test_bit(PAGE_DEBUG_FLAG_GUARD, &page->debug_flags);
}
#else
static inline unsigned int debug_guardpage_minorder(void) { return 0; }
static inline bool page_is_guard(struct page *page) { return false; }
#endif /* CONFIG_DEBUG_PAGEALLOC */

2008 2009 2010 2011 2012 2013
#if MAX_NUMNODES > 1
void __init setup_nr_node_ids(void);
#else
static inline void setup_nr_node_ids(void) {}
#endif

L
Linus Torvalds 已提交
2014 2015
#endif /* __KERNEL__ */
#endif /* _LINUX_MM_H */