mm.h 83.3 KB
Newer Older
1
/* SPDX-License-Identifier: GPL-2.0 */
L
Linus Torvalds 已提交
2 3 4 5 6 7 8
#ifndef _LINUX_MM_H
#define _LINUX_MM_H

#include <linux/errno.h>

#ifdef __KERNEL__

9
#include <linux/mmdebug.h>
L
Linus Torvalds 已提交
10
#include <linux/gfp.h>
11
#include <linux/bug.h>
L
Linus Torvalds 已提交
12 13 14
#include <linux/list.h>
#include <linux/mmzone.h>
#include <linux/rbtree.h>
15
#include <linux/atomic.h>
16
#include <linux/debug_locks.h>
17
#include <linux/mm_types.h>
18
#include <linux/range.h>
19
#include <linux/pfn.h>
20
#include <linux/percpu-refcount.h>
A
Andrea Arcangeli 已提交
21
#include <linux/bit_spinlock.h>
22
#include <linux/shrinker.h>
23
#include <linux/resource.h>
24
#include <linux/page_ext.h>
25
#include <linux/err.h>
26
#include <linux/page_ref.h>
27
#include <linux/memremap.h>
L
Linus Torvalds 已提交
28 29 30

struct mempolicy;
struct anon_vma;
31
struct anon_vma_chain;
A
Alexey Dobriyan 已提交
32
struct file_ra_state;
A
Alexey Dobriyan 已提交
33
struct user_struct;
A
Alexey Dobriyan 已提交
34
struct writeback_control;
35
struct bdi_writeback;
L
Linus Torvalds 已提交
36

37 38
void init_mm_internals(void);

39
#ifndef CONFIG_NEED_MULTIPLE_NODES	/* Don't use mapnrs, do it properly */
L
Linus Torvalds 已提交
40
extern unsigned long max_mapnr;
41 42 43 44 45 46 47

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
48 49
#endif

50
extern unsigned long totalram_pages;
L
Linus Torvalds 已提交
51 52 53 54 55 56 57 58 59
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

60 61 62 63 64 65 66 67 68 69 70
#ifdef CONFIG_HAVE_ARCH_MMAP_RND_BITS
extern const int mmap_rnd_bits_min;
extern const int mmap_rnd_bits_max;
extern int mmap_rnd_bits __read_mostly;
#endif
#ifdef CONFIG_HAVE_ARCH_MMAP_RND_COMPAT_BITS
extern const int mmap_rnd_compat_bits_min;
extern const int mmap_rnd_compat_bits_max;
extern int mmap_rnd_compat_bits __read_mostly;
#endif

L
Linus Torvalds 已提交
71 72 73 74
#include <asm/page.h>
#include <asm/pgtable.h>
#include <asm/processor.h>

75 76 77 78
#ifndef __pa_symbol
#define __pa_symbol(x)  __pa(RELOC_HIDE((unsigned long)(x), 0))
#endif

79 80 81 82
#ifndef page_to_virt
#define page_to_virt(x)	__va(PFN_PHYS(page_to_pfn(x)))
#endif

L
Laura Abbott 已提交
83 84 85 86
#ifndef lm_alias
#define lm_alias(x)	__va(__pa_symbol(x))
#endif

87 88 89 90 91 92 93 94 95 96 97
/*
 * To prevent common memory management code establishing
 * a zero page mapping on a read fault.
 * This macro should be defined within <asm/pgtable.h>.
 * s390 does this to prevent multiplexing of hardware bits
 * related to the physical page in case of virtualization.
 */
#ifndef mm_forbids_zeropage
#define mm_forbids_zeropage(X)	(0)
#endif

98 99 100 101 102 103 104 105 106
/*
 * On some architectures it is expensive to call memset() for small sizes.
 * Those architectures should provide their own implementation of "struct page"
 * zeroing by defining this macro in <asm/pgtable.h>.
 */
#ifndef mm_zero_struct_page
#define mm_zero_struct_page(pp)  ((void)memset((pp), 0, sizeof(struct page)))
#endif

107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127
/*
 * Default maximum number of active map areas, this limits the number of vmas
 * per mm struct. Users can overwrite this number by sysctl but there is a
 * problem.
 *
 * When a program's coredump is generated as ELF format, a section is created
 * per a vma. In ELF, the number of sections is represented in unsigned short.
 * This means the number of sections should be smaller than 65535 at coredump.
 * Because the kernel adds some informative sections to a image of program at
 * generating coredump, we need some margin. The number of extra sections is
 * 1-3 now and depends on arch. We use "5" as safe margin, here.
 *
 * ELF extended numbering allows more than 65535 sections, so 16-bit bound is
 * not a hard limit any more. Although some userspace tools can be surprised by
 * that.
 */
#define MAPCOUNT_ELF_CORE_MARGIN	(5)
#define DEFAULT_MAX_MAP_COUNT	(USHRT_MAX - MAPCOUNT_ELF_CORE_MARGIN)

extern int sysctl_max_map_count;

128
extern unsigned long sysctl_user_reserve_kbytes;
129
extern unsigned long sysctl_admin_reserve_kbytes;
130

131 132 133 134 135 136 137 138 139
extern int sysctl_overcommit_memory;
extern int sysctl_overcommit_ratio;
extern unsigned long sysctl_overcommit_kbytes;

extern int overcommit_ratio_handler(struct ctl_table *, int, void __user *,
				    size_t *, loff_t *);
extern int overcommit_kbytes_handler(struct ctl_table *, int, void __user *,
				    size_t *, loff_t *);

L
Linus Torvalds 已提交
140 141
#define nth_page(page,n) pfn_to_page(page_to_pfn((page)) + (n))

142 143 144
/* to align the pointer to the (next) page boundary */
#define PAGE_ALIGN(addr) ALIGN(addr, PAGE_SIZE)

145
/* test whether an address (unsigned long or pointer) is aligned to PAGE_SIZE */
146
#define PAGE_ALIGNED(addr)	IS_ALIGNED((unsigned long)(addr), PAGE_SIZE)
147

L
Linus Torvalds 已提交
148 149 150 151 152 153 154 155 156
/*
 * 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).
 */

157 158
extern struct kmem_cache *vm_area_cachep;

L
Linus Torvalds 已提交
159
#ifndef CONFIG_MMU
160 161
extern struct rb_root nommu_region_tree;
extern struct rw_semaphore nommu_region_sem;
L
Linus Torvalds 已提交
162 163 164 165 166

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

/*
H
Hugh Dickins 已提交
167
 * vm_flags in vm_area_struct, see mm_types.h.
168
 * When changing, update also include/trace/events/mmflags.h
L
Linus Torvalds 已提交
169
 */
170 171
#define VM_NONE		0x00000000

L
Linus Torvalds 已提交
172 173 174 175 176
#define VM_READ		0x00000001	/* currently active flags */
#define VM_WRITE	0x00000002
#define VM_EXEC		0x00000004
#define VM_SHARED	0x00000008

177
/* mprotect() hardcodes VM_MAYREAD >> 4 == VM_READ, and so for r/w/x bits. */
L
Linus Torvalds 已提交
178 179 180 181 182 183
#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 */
184
#define VM_UFFD_MISSING	0x00000200	/* missing pages tracking */
185
#define VM_PFNMAP	0x00000400	/* Page-ranges managed without "struct page", just pure PFN */
L
Linus Torvalds 已提交
186
#define VM_DENYWRITE	0x00000800	/* ETXTBSY on write attempts.. */
187
#define VM_UFFD_WP	0x00001000	/* wrprotect pages tracking */
L
Linus Torvalds 已提交
188 189 190 191 192 193 194 195 196 197

#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() */
E
Eric B Munson 已提交
198
#define VM_LOCKONFAULT	0x00080000	/* Lock the pages covered when they are faulted in */
L
Linus Torvalds 已提交
199
#define VM_ACCOUNT	0x00100000	/* Is a VM accounted object */
200
#define VM_NORESERVE	0x00200000	/* should the VM suppress accounting */
L
Linus Torvalds 已提交
201
#define VM_HUGETLB	0x00400000	/* Huge TLB Page VM */
J
Jan Kara 已提交
202
#define VM_SYNC		0x00800000	/* Synchronous page faults */
203
#define VM_ARCH_1	0x01000000	/* Architecture-specific flag */
204
#define VM_WIPEONFORK	0x02000000	/* Wipe VMA contents in child. */
205
#define VM_DONTDUMP	0x04000000	/* Do not include in the core dump */
206

207 208 209 210 211 212
#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
213
#define VM_MIXEDMAP	0x10000000	/* Can contain "struct page" and pure PFN pages */
214 215
#define VM_HUGEPAGE	0x20000000	/* MADV_HUGEPAGE marked this vma */
#define VM_NOHUGEPAGE	0x40000000	/* MADV_NOHUGEPAGE marked this vma */
H
Hugh Dickins 已提交
216
#define VM_MERGEABLE	0x80000000	/* KSM may merge identical pages */
L
Linus Torvalds 已提交
217

218 219 220 221 222
#ifdef CONFIG_ARCH_USES_HIGH_VMA_FLAGS
#define VM_HIGH_ARCH_BIT_0	32	/* bit only usable on 64-bit architectures */
#define VM_HIGH_ARCH_BIT_1	33	/* bit only usable on 64-bit architectures */
#define VM_HIGH_ARCH_BIT_2	34	/* bit only usable on 64-bit architectures */
#define VM_HIGH_ARCH_BIT_3	35	/* bit only usable on 64-bit architectures */
223
#define VM_HIGH_ARCH_BIT_4	36	/* bit only usable on 64-bit architectures */
224 225 226 227
#define VM_HIGH_ARCH_0	BIT(VM_HIGH_ARCH_BIT_0)
#define VM_HIGH_ARCH_1	BIT(VM_HIGH_ARCH_BIT_1)
#define VM_HIGH_ARCH_2	BIT(VM_HIGH_ARCH_BIT_2)
#define VM_HIGH_ARCH_3	BIT(VM_HIGH_ARCH_BIT_3)
228
#define VM_HIGH_ARCH_4	BIT(VM_HIGH_ARCH_BIT_4)
229 230
#endif /* CONFIG_ARCH_USES_HIGH_VMA_FLAGS */

231 232
#if defined(CONFIG_X86)
# define VM_PAT		VM_ARCH_1	/* PAT reserves whole VMA at once (x86) */
233 234 235 236 237 238 239
#if defined (CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS)
# define VM_PKEY_SHIFT	VM_HIGH_ARCH_BIT_0
# define VM_PKEY_BIT0	VM_HIGH_ARCH_0	/* A protection key is a 4-bit value */
# define VM_PKEY_BIT1	VM_HIGH_ARCH_1
# define VM_PKEY_BIT2	VM_HIGH_ARCH_2
# define VM_PKEY_BIT3	VM_HIGH_ARCH_3
#endif
240 241 242 243
#elif defined(CONFIG_PPC)
# define VM_SAO		VM_ARCH_1	/* Strong Access Ordering (powerpc) */
#elif defined(CONFIG_PARISC)
# define VM_GROWSUP	VM_ARCH_1
244 245
#elif defined(CONFIG_METAG)
# define VM_GROWSUP	VM_ARCH_1
246 247 248 249 250 251
#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

252
#if defined(CONFIG_X86_INTEL_MPX)
253
/* MPX specific bounds table or bounds directory */
254
# define VM_MPX		VM_HIGH_ARCH_4
255 256
#else
# define VM_MPX		VM_NONE
257 258
#endif

259 260 261 262
#ifndef VM_GROWSUP
# define VM_GROWSUP	VM_NONE
#endif

263 264 265
/* 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
266 267 268 269 270
#ifndef VM_STACK_DEFAULT_FLAGS		/* arch can override this */
#define VM_STACK_DEFAULT_FLAGS VM_DATA_DEFAULT_FLAGS
#endif

#ifdef CONFIG_STACK_GROWSUP
271
#define VM_STACK	VM_GROWSUP
L
Linus Torvalds 已提交
272
#else
273
#define VM_STACK	VM_GROWSDOWN
L
Linus Torvalds 已提交
274 275
#endif

276 277
#define VM_STACK_FLAGS	(VM_STACK | VM_STACK_DEFAULT_FLAGS | VM_ACCOUNT)

N
Nick Piggin 已提交
278
/*
279 280
 * 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
281
 */
282
#define VM_SPECIAL (VM_IO | VM_DONTEXPAND | VM_PFNMAP | VM_MIXEDMAP)
N
Nick Piggin 已提交
283

284 285 286
/* This mask defines which mm->def_flags a process can inherit its parent */
#define VM_INIT_DEF_MASK	VM_NOHUGEPAGE

E
Eric B Munson 已提交
287 288 289
/* This mask is used to clear all the VMA flags used by mlock */
#define VM_LOCKED_CLEAR_MASK	(~(VM_LOCKED | VM_LOCKONFAULT))

L
Linus Torvalds 已提交
290 291 292 293 294 295
/*
 * 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
296
#define FAULT_FLAG_WRITE	0x01	/* Fault was a write access */
297 298 299 300 301 302
#define FAULT_FLAG_MKWRITE	0x02	/* Fault was mkwrite of existing pte */
#define FAULT_FLAG_ALLOW_RETRY	0x04	/* Retry fault if blocking */
#define FAULT_FLAG_RETRY_NOWAIT	0x08	/* Don't drop mmap_sem and wait when retrying */
#define FAULT_FLAG_KILLABLE	0x10	/* The fault task is in SIGKILL killable region */
#define FAULT_FLAG_TRIED	0x20	/* Second try */
#define FAULT_FLAG_USER		0x40	/* The fault originated in userspace */
303
#define FAULT_FLAG_REMOTE	0x80	/* faulting for non current tsk/mm */
304
#define FAULT_FLAG_INSTRUCTION  0x100	/* The fault was during an instruction fetch */
N
Nick Piggin 已提交
305

306 307 308 309 310 311 312 313 314 315 316
#define FAULT_FLAG_TRACE \
	{ FAULT_FLAG_WRITE,		"WRITE" }, \
	{ FAULT_FLAG_MKWRITE,		"MKWRITE" }, \
	{ FAULT_FLAG_ALLOW_RETRY,	"ALLOW_RETRY" }, \
	{ FAULT_FLAG_RETRY_NOWAIT,	"RETRY_NOWAIT" }, \
	{ FAULT_FLAG_KILLABLE,		"KILLABLE" }, \
	{ FAULT_FLAG_TRIED,		"TRIED" }, \
	{ FAULT_FLAG_USER,		"USER" }, \
	{ FAULT_FLAG_REMOTE,		"REMOTE" }, \
	{ FAULT_FLAG_INSTRUCTION,	"INSTRUCTION" }

317
/*
N
Nick Piggin 已提交
318
 * vm_fault is filled by the the pagefault handler and passed to the vma's
N
Nick Piggin 已提交
319 320
 * ->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.
321
 *
322 323 324
 * MM layer fills up gfp_mask for page allocations but fault handler might
 * alter it if its implementation requires a different allocation context.
 *
325
 * pgoff should be used in favour of virtual_address, if possible.
326
 */
N
Nick Piggin 已提交
327
struct vm_fault {
J
Jan Kara 已提交
328
	struct vm_area_struct *vma;	/* Target VMA */
N
Nick Piggin 已提交
329
	unsigned int flags;		/* FAULT_FLAG_xxx flags */
330
	gfp_t gfp_mask;			/* gfp mask to be used for allocations */
N
Nick Piggin 已提交
331
	pgoff_t pgoff;			/* Logical page offset based on vma */
J
Jan Kara 已提交
332 333
	unsigned long address;		/* Faulting virtual address */
	pmd_t *pmd;			/* Pointer to pmd entry matching
J
Jan Kara 已提交
334
					 * the 'address' */
335 336 337
	pud_t *pud;			/* Pointer to pud entry matching
					 * the 'address'
					 */
J
Jan Kara 已提交
338
	pte_t orig_pte;			/* Value of PTE at the time of fault */
N
Nick Piggin 已提交
339

340 341
	struct page *cow_page;		/* Page handler may use for COW fault */
	struct mem_cgroup *memcg;	/* Cgroup cow_page belongs to */
N
Nick Piggin 已提交
342
	struct page *page;		/* ->fault handlers should return a
N
Nick Piggin 已提交
343
					 * page here, unless VM_FAULT_NOPAGE
N
Nick Piggin 已提交
344
					 * is set (which is also implied by
N
Nick Piggin 已提交
345
					 * VM_FAULT_ERROR).
N
Nick Piggin 已提交
346
					 */
J
Jan Kara 已提交
347
	/* These three entries are valid only while holding ptl lock */
K
Kirill A. Shutemov 已提交
348 349 350 351 352 353 354 355
	pte_t *pte;			/* Pointer to pte entry matching
					 * the 'address'. NULL if the page
					 * table hasn't been allocated.
					 */
	spinlock_t *ptl;		/* Page table lock.
					 * Protects pte page table if 'pte'
					 * is not NULL, otherwise pmd.
					 */
356 357 358 359 360 361 362
	pgtable_t prealloc_pte;		/* Pre-allocated pte page table.
					 * vm_ops->map_pages() calls
					 * alloc_set_pte() from atomic context.
					 * do_fault_around() pre-allocates
					 * page table to avoid allocation from
					 * atomic context.
					 */
363
};
L
Linus Torvalds 已提交
364

365 366 367 368 369 370 371
/* page entry size for vm->huge_fault() */
enum page_entry_size {
	PE_SIZE_PTE = 0,
	PE_SIZE_PMD,
	PE_SIZE_PUD,
};

L
Linus Torvalds 已提交
372 373 374 375 376 377 378 379
/*
 * 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);
380
	int (*split)(struct vm_area_struct * area, unsigned long addr);
381
	int (*mremap)(struct vm_area_struct * area);
382
	int (*fault)(struct vm_fault *vmf);
383
	int (*huge_fault)(struct vm_fault *vmf, enum page_entry_size pe_size);
J
Jan Kara 已提交
384
	void (*map_pages)(struct vm_fault *vmf,
K
Kirill A. Shutemov 已提交
385
			pgoff_t start_pgoff, pgoff_t end_pgoff);
386 387 388

	/* notification that a previously read-only page is about to become
	 * writable, if an error is returned it will cause a SIGBUS */
389
	int (*page_mkwrite)(struct vm_fault *vmf);
390

391
	/* same as page_mkwrite when using VM_PFNMAP|VM_MIXEDMAP */
392
	int (*pfn_mkwrite)(struct vm_fault *vmf);
393

394 395 396 397 398
	/* 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);
399 400 401 402 403 404

	/* Called by the /proc/PID/maps code to ask the vma whether it
	 * has a special name.  Returning non-NULL will also cause this
	 * vma to be dumped unconditionally. */
	const char *(*name)(struct vm_area_struct *vma);

L
Linus Torvalds 已提交
405
#ifdef CONFIG_NUMA
406 407 408 409 410 411 412
	/*
	 * 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
413
	int (*set_policy)(struct vm_area_struct *vma, struct mempolicy *new);
414 415 416 417 418 419 420 421 422 423 424

	/*
	 * 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
425 426 427
	struct mempolicy *(*get_policy)(struct vm_area_struct *vma,
					unsigned long addr);
#endif
428 429 430 431 432 433 434
	/*
	 * Called by vm_normal_page() for special PTEs to find the
	 * page for @addr.  This is useful if the default behavior
	 * (using pte_page()) would not find the correct page.
	 */
	struct page *(*find_special_page)(struct vm_area_struct *vma,
					  unsigned long addr);
L
Linus Torvalds 已提交
435 436 437 438 439
};

struct mmu_gather;
struct inode;

A
Andrew Morton 已提交
440 441
#define page_private(page)		((page)->private)
#define set_page_private(page, v)	((page)->private = (v))
H
Hugh Dickins 已提交
442

443 444 445 446 447
#if !defined(__HAVE_ARCH_PTE_DEVMAP) || !defined(CONFIG_TRANSPARENT_HUGEPAGE)
static inline int pmd_devmap(pmd_t pmd)
{
	return 0;
}
448 449 450 451
static inline int pud_devmap(pud_t pud)
{
	return 0;
}
452 453 454 455
static inline int pgd_devmap(pgd_t pgd)
{
	return 0;
}
456 457
#endif

L
Linus Torvalds 已提交
458 459 460 461 462
/*
 * FIXME: take this include out, include page-flags.h in
 * files which need it (119 of them)
 */
#include <linux/page-flags.h>
463
#include <linux/huge_mm.h>
L
Linus Torvalds 已提交
464 465 466 467 468 469 470 471 472 473 474 475 476 477 478

/*
 * 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
479
 * Drop a ref, return true if the refcount fell to zero (the page has no users)
L
Linus Torvalds 已提交
480
 */
481 482
static inline int put_page_testzero(struct page *page)
{
483 484
	VM_BUG_ON_PAGE(page_ref_count(page) == 0, page);
	return page_ref_dec_and_test(page);
485
}
L
Linus Torvalds 已提交
486 487

/*
488 489
 * Try to grab a ref unless the page has a refcount of zero, return false if
 * that is the case.
490 491
 * This can be called when MMU is off so it must not access
 * any of the virtual mappings.
L
Linus Torvalds 已提交
492
 */
493 494
static inline int get_page_unless_zero(struct page *page)
{
495
	return page_ref_add_unless(page, 1, 0);
496
}
L
Linus Torvalds 已提交
497

498
extern int page_is_ram(unsigned long pfn);
499 500 501 502 503 504 505

enum {
	REGION_INTERSECTS,
	REGION_DISJOINT,
	REGION_MIXED,
};

506 507
int region_intersects(resource_size_t offset, size_t size, unsigned long flags,
		      unsigned long desc);
508

509
/* Support for virtually mapped pages */
510 511
struct page *vmalloc_to_page(const void *addr);
unsigned long vmalloc_to_pfn(const void *addr);
512

513 514 515 516 517 518
/*
 * 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.
 */
519
static inline bool is_vmalloc_addr(const void *x)
520
{
521
#ifdef CONFIG_MMU
522 523 524
	unsigned long addr = (unsigned long)x;

	return addr >= VMALLOC_START && addr < VMALLOC_END;
525
#else
526
	return false;
527
#endif
528
}
529 530 531
#ifdef CONFIG_MMU
extern int is_vmalloc_or_module_addr(const void *x);
#else
532
static inline int is_vmalloc_or_module_addr(const void *x)
533 534 535 536
{
	return 0;
}
#endif
537

M
Michal Hocko 已提交
538 539 540 541 542 543 544 545 546 547 548 549 550 551
extern void *kvmalloc_node(size_t size, gfp_t flags, int node);
static inline void *kvmalloc(size_t size, gfp_t flags)
{
	return kvmalloc_node(size, flags, NUMA_NO_NODE);
}
static inline void *kvzalloc_node(size_t size, gfp_t flags, int node)
{
	return kvmalloc_node(size, flags | __GFP_ZERO, node);
}
static inline void *kvzalloc(size_t size, gfp_t flags)
{
	return kvmalloc(size, flags | __GFP_ZERO);
}

552 553 554 555 556 557 558 559
static inline void *kvmalloc_array(size_t n, size_t size, gfp_t flags)
{
	if (size != 0 && n > SIZE_MAX / size)
		return NULL;

	return kvmalloc(n * size, flags);
}

A
Al Viro 已提交
560 561
extern void kvfree(const void *addr);

562 563 564 565 566 567 568
static inline atomic_t *compound_mapcount_ptr(struct page *page)
{
	return &page[1].compound_mapcount;
}

static inline int compound_mapcount(struct page *page)
{
569
	VM_BUG_ON_PAGE(!PageCompound(page), page);
570 571 572 573
	page = compound_head(page);
	return atomic_read(compound_mapcount_ptr(page)) + 1;
}

574 575 576 577 578
/*
 * 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).
 */
579
static inline void page_mapcount_reset(struct page *page)
580 581 582 583
{
	atomic_set(&(page)->_mapcount, -1);
}

584 585
int __page_mapcount(struct page *page);

586 587
static inline int page_mapcount(struct page *page)
{
588
	VM_BUG_ON_PAGE(PageSlab(page), page);
589

590 591 592 593 594 595 596
	if (unlikely(PageCompound(page)))
		return __page_mapcount(page);
	return atomic_read(&page->_mapcount) + 1;
}

#ifdef CONFIG_TRANSPARENT_HUGEPAGE
int total_mapcount(struct page *page);
597
int page_trans_huge_mapcount(struct page *page, int *total_mapcount);
598 599 600 601
#else
static inline int total_mapcount(struct page *page)
{
	return page_mapcount(page);
602
}
603 604 605 606 607 608 609 610
static inline int page_trans_huge_mapcount(struct page *page,
					   int *total_mapcount)
{
	int mapcount = page_mapcount(page);
	if (total_mapcount)
		*total_mapcount = mapcount;
	return mapcount;
}
611
#endif
612

613 614 615
static inline struct page *virt_to_head_page(const void *x)
{
	struct page *page = virt_to_page(x);
616

617
	return compound_head(page);
618 619
}

620 621
void __put_page(struct page *page);

622
void put_pages_list(struct list_head *pages);
L
Linus Torvalds 已提交
623

N
Nick Piggin 已提交
624 625
void split_page(struct page *page, unsigned int order);

626 627 628
/*
 * Compound pages have a destructor function.  Provide a
 * prototype for that function and accessor functions.
629
 * These are _only_ valid on the head of a compound page.
630
 */
631 632 633 634 635 636 637 638
typedef void compound_page_dtor(struct page *);

/* Keep the enum in sync with compound_page_dtors array in mm/page_alloc.c */
enum compound_dtor_id {
	NULL_COMPOUND_DTOR,
	COMPOUND_PAGE_DTOR,
#ifdef CONFIG_HUGETLB_PAGE
	HUGETLB_PAGE_DTOR,
639 640 641
#endif
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
	TRANSHUGE_PAGE_DTOR,
642 643 644 645
#endif
	NR_COMPOUND_DTORS,
};
extern compound_page_dtor * const compound_page_dtors[];
646 647

static inline void set_compound_page_dtor(struct page *page,
648
		enum compound_dtor_id compound_dtor)
649
{
650 651
	VM_BUG_ON_PAGE(compound_dtor >= NR_COMPOUND_DTORS, page);
	page[1].compound_dtor = compound_dtor;
652 653 654 655
}

static inline compound_page_dtor *get_compound_page_dtor(struct page *page)
{
656 657
	VM_BUG_ON_PAGE(page[1].compound_dtor >= NR_COMPOUND_DTORS, page);
	return compound_page_dtors[page[1].compound_dtor];
658 659
}

660
static inline unsigned int compound_order(struct page *page)
661
{
662
	if (!PageHead(page))
663
		return 0;
664
	return page[1].compound_order;
665 666
}

667
static inline void set_compound_order(struct page *page, unsigned int order)
668
{
669
	page[1].compound_order = order;
670 671
}

672 673
void free_compound_page(struct page *page);

674
#ifdef CONFIG_MMU
A
Andrea Arcangeli 已提交
675 676 677 678 679 680 681 682 683 684 685 686
/*
 * 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;
}
687

J
Jan Kara 已提交
688
int alloc_set_pte(struct vm_fault *vmf, struct mem_cgroup *memcg,
689
		struct page *page);
690
int finish_fault(struct vm_fault *vmf);
691
int finish_mkwrite_fault(struct vm_fault *vmf);
692
#endif
A
Andrea Arcangeli 已提交
693

L
Linus Torvalds 已提交
694 695 696 697 698 699 700
/*
 * 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.
701 702
 *   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
703
 *   page_count() > 0  means the page has been allocated.
L
Linus Torvalds 已提交
704
 *
N
Nick Piggin 已提交
705 706 707 708 709
 * 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
710
 *
N
Nick Piggin 已提交
711 712 713 714 715 716 717 718 719
 * 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
720 721 722
 * managed by the Linux memory manager: I/O, buffers, swapping etc.
 * The following discussion applies only to them.
 *
N
Nick Piggin 已提交
723 724 725 726
 * 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
727
 *
N
Nick Piggin 已提交
728 729
 * 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,
730
 * in units of PAGE_SIZE.
L
Linus Torvalds 已提交
731
 *
N
Nick Piggin 已提交
732 733 734
 * 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
735
 *
N
Nick Piggin 已提交
736 737 738
 * 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
739
 *
N
Nick Piggin 已提交
740 741 742 743
 * 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
744
 *
N
Nick Piggin 已提交
745
 * All pagecache pages may be subject to I/O:
L
Linus Torvalds 已提交
746 747
 * - 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
748 749 750 751
 *   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
752 753 754 755 756 757
 */

/*
 * The zone field is never updated after free_area_init_core()
 * sets it, so none of the operations on it need to be atomic.
 */
758

759
/* Page flags: | [SECTION] | [NODE] | ZONE | [LAST_CPUPID] | ... | FLAGS | */
760
#define SECTIONS_PGOFF		((sizeof(unsigned long)*8) - SECTIONS_WIDTH)
A
Andy Whitcroft 已提交
761 762
#define NODES_PGOFF		(SECTIONS_PGOFF - NODES_WIDTH)
#define ZONES_PGOFF		(NODES_PGOFF - ZONES_WIDTH)
763
#define LAST_CPUPID_PGOFF	(ZONES_PGOFF - LAST_CPUPID_WIDTH)
A
Andy Whitcroft 已提交
764

765
/*
L
Lucas De Marchi 已提交
766
 * Define the bit shifts to access each section.  For non-existent
767 768 769
 * 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
770 771 772
#define SECTIONS_PGSHIFT	(SECTIONS_PGOFF * (SECTIONS_WIDTH != 0))
#define NODES_PGSHIFT		(NODES_PGOFF * (NODES_WIDTH != 0))
#define ZONES_PGSHIFT		(ZONES_PGOFF * (ZONES_WIDTH != 0))
773
#define LAST_CPUPID_PGSHIFT	(LAST_CPUPID_PGOFF * (LAST_CPUPID_WIDTH != 0))
774

775 776
/* NODE:ZONE or SECTION:ZONE is used to ID a zone for the buddy allocator */
#ifdef NODE_NOT_IN_PAGE_FLAGS
777
#define ZONEID_SHIFT		(SECTIONS_SHIFT + ZONES_SHIFT)
778 779
#define ZONEID_PGOFF		((SECTIONS_PGOFF < ZONES_PGOFF)? \
						SECTIONS_PGOFF : ZONES_PGOFF)
A
Andy Whitcroft 已提交
780
#else
781
#define ZONEID_SHIFT		(NODES_SHIFT + ZONES_SHIFT)
782 783
#define ZONEID_PGOFF		((NODES_PGOFF < ZONES_PGOFF)? \
						NODES_PGOFF : ZONES_PGOFF)
784 785
#endif

786
#define ZONEID_PGSHIFT		(ZONEID_PGOFF * (ZONEID_SHIFT != 0))
787

788 789
#if SECTIONS_WIDTH+NODES_WIDTH+ZONES_WIDTH > BITS_PER_LONG - NR_PAGEFLAGS
#error SECTIONS_WIDTH+NODES_WIDTH+ZONES_WIDTH > BITS_PER_LONG - NR_PAGEFLAGS
790 791
#endif

A
Andy Whitcroft 已提交
792 793 794
#define ZONES_MASK		((1UL << ZONES_WIDTH) - 1)
#define NODES_MASK		((1UL << NODES_WIDTH) - 1)
#define SECTIONS_MASK		((1UL << SECTIONS_WIDTH) - 1)
795
#define LAST_CPUPID_MASK	((1UL << LAST_CPUPID_SHIFT) - 1)
796
#define ZONEID_MASK		((1UL << ZONEID_SHIFT) - 1)
797

I
Ian Campbell 已提交
798
static inline enum zone_type page_zonenum(const struct page *page)
L
Linus Torvalds 已提交
799
{
800
	return (page->flags >> ZONES_PGSHIFT) & ZONES_MASK;
L
Linus Torvalds 已提交
801 802
}

803 804 805 806 807 808 809 810 811 812
#ifdef CONFIG_ZONE_DEVICE
static inline bool is_zone_device_page(const struct page *page)
{
	return page_zonenum(page) == ZONE_DEVICE;
}
#else
static inline bool is_zone_device_page(const struct page *page)
{
	return false;
}
813
#endif
814

815
#if defined(CONFIG_DEVICE_PRIVATE) || defined(CONFIG_DEVICE_PUBLIC)
816
void put_zone_device_private_or_public_page(struct page *page);
817 818 819 820 821
DECLARE_STATIC_KEY_FALSE(device_private_key);
#define IS_HMM_ENABLED static_branch_unlikely(&device_private_key)
static inline bool is_device_private_page(const struct page *page);
static inline bool is_device_public_page(const struct page *page);
#else /* CONFIG_DEVICE_PRIVATE || CONFIG_DEVICE_PUBLIC */
822
static inline void put_zone_device_private_or_public_page(struct page *page)
823 824
{
}
825 826 827 828 829 830 831 832 833
#define IS_HMM_ENABLED 0
static inline bool is_device_private_page(const struct page *page)
{
	return false;
}
static inline bool is_device_public_page(const struct page *page)
{
	return false;
}
834
#endif /* CONFIG_DEVICE_PRIVATE || CONFIG_DEVICE_PUBLIC */
835

836

837 838 839 840 841
static inline void get_page(struct page *page)
{
	page = compound_head(page);
	/*
	 * Getting a normal page or the head of a compound page
842
	 * requires to already have an elevated page->_refcount.
843
	 */
844 845
	VM_BUG_ON_PAGE(page_ref_count(page) <= 0, page);
	page_ref_inc(page);
846 847 848 849 850 851
}

static inline void put_page(struct page *page)
{
	page = compound_head(page);

852 853 854 855 856 857
	/*
	 * For private device pages we need to catch refcount transition from
	 * 2 to 1, when refcount reach one it means the private device page is
	 * free and we need to inform the device driver through callback. See
	 * include/linux/memremap.h and HMM for details.
	 */
858 859
	if (IS_HMM_ENABLED && unlikely(is_device_private_page(page) ||
	    unlikely(is_device_public_page(page)))) {
860
		put_zone_device_private_or_public_page(page);
861 862 863
		return;
	}

864 865 866 867
	if (put_page_testzero(page))
		__put_page(page);
}

C
Cody P Schafer 已提交
868 869 870 871
#if defined(CONFIG_SPARSEMEM) && !defined(CONFIG_SPARSEMEM_VMEMMAP)
#define SECTION_IN_PAGE_FLAGS
#endif

872
/*
873 874 875 876 877 878
 * 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.
879
 */
880 881
static inline int page_zone_id(struct page *page)
{
882
	return (page->flags >> ZONEID_PGSHIFT) & ZONEID_MASK;
883 884
}

885
static inline int zone_to_nid(struct zone *zone)
886
{
887 888 889 890 891
#ifdef CONFIG_NUMA
	return zone->node;
#else
	return 0;
#endif
892 893
}

894
#ifdef NODE_NOT_IN_PAGE_FLAGS
I
Ian Campbell 已提交
895
extern int page_to_nid(const struct page *page);
896
#else
I
Ian Campbell 已提交
897
static inline int page_to_nid(const struct page *page)
A
Andy Whitcroft 已提交
898
{
899
	return (page->flags >> NODES_PGSHIFT) & NODES_MASK;
A
Andy Whitcroft 已提交
900
}
901 902
#endif

903
#ifdef CONFIG_NUMA_BALANCING
904
static inline int cpu_pid_to_cpupid(int cpu, int pid)
905
{
906
	return ((cpu & LAST__CPU_MASK) << LAST__PID_SHIFT) | (pid & LAST__PID_MASK);
907 908
}

909
static inline int cpupid_to_pid(int cpupid)
910
{
911
	return cpupid & LAST__PID_MASK;
912
}
913

914
static inline int cpupid_to_cpu(int cpupid)
915
{
916
	return (cpupid >> LAST__PID_SHIFT) & LAST__CPU_MASK;
917 918
}

919
static inline int cpupid_to_nid(int cpupid)
920
{
921
	return cpu_to_node(cpupid_to_cpu(cpupid));
922 923
}

924
static inline bool cpupid_pid_unset(int cpupid)
925
{
926
	return cpupid_to_pid(cpupid) == (-1 & LAST__PID_MASK);
927 928
}

929
static inline bool cpupid_cpu_unset(int cpupid)
930
{
931
	return cpupid_to_cpu(cpupid) == (-1 & LAST__CPU_MASK);
932 933
}

934 935 936 937 938 939
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)
940 941
#ifdef LAST_CPUPID_NOT_IN_PAGE_FLAGS
static inline int page_cpupid_xchg_last(struct page *page, int cpupid)
942
{
943
	return xchg(&page->_last_cpupid, cpupid & LAST_CPUPID_MASK);
944
}
945 946 947 948 949 950

static inline int page_cpupid_last(struct page *page)
{
	return page->_last_cpupid;
}
static inline void page_cpupid_reset_last(struct page *page)
951
{
952
	page->_last_cpupid = -1 & LAST_CPUPID_MASK;
953 954
}
#else
955
static inline int page_cpupid_last(struct page *page)
956
{
957
	return (page->flags >> LAST_CPUPID_PGSHIFT) & LAST_CPUPID_MASK;
958 959
}

960
extern int page_cpupid_xchg_last(struct page *page, int cpupid);
961

962
static inline void page_cpupid_reset_last(struct page *page)
963
{
964
	page->flags |= LAST_CPUPID_MASK << LAST_CPUPID_PGSHIFT;
965
}
966 967 968
#endif /* LAST_CPUPID_NOT_IN_PAGE_FLAGS */
#else /* !CONFIG_NUMA_BALANCING */
static inline int page_cpupid_xchg_last(struct page *page, int cpupid)
969
{
970
	return page_to_nid(page); /* XXX */
971 972
}

973
static inline int page_cpupid_last(struct page *page)
974
{
975
	return page_to_nid(page); /* XXX */
976 977
}

978
static inline int cpupid_to_nid(int cpupid)
979 980 981 982
{
	return -1;
}

983
static inline int cpupid_to_pid(int cpupid)
984 985 986 987
{
	return -1;
}

988
static inline int cpupid_to_cpu(int cpupid)
989 990 991 992
{
	return -1;
}

993 994 995 996 997 998
static inline int cpu_pid_to_cpupid(int nid, int pid)
{
	return -1;
}

static inline bool cpupid_pid_unset(int cpupid)
999 1000 1001 1002
{
	return 1;
}

1003
static inline void page_cpupid_reset_last(struct page *page)
1004 1005
{
}
1006 1007 1008 1009 1010

static inline bool cpupid_match_pid(struct task_struct *task, int cpupid)
{
	return false;
}
1011
#endif /* CONFIG_NUMA_BALANCING */
1012

I
Ian Campbell 已提交
1013
static inline struct zone *page_zone(const struct page *page)
1014 1015 1016 1017
{
	return &NODE_DATA(page_to_nid(page))->node_zones[page_zonenum(page)];
}

1018 1019 1020 1021 1022
static inline pg_data_t *page_pgdat(const struct page *page)
{
	return NODE_DATA(page_to_nid(page));
}

C
Cody P Schafer 已提交
1023
#ifdef SECTION_IN_PAGE_FLAGS
1024 1025 1026 1027 1028 1029
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;
}

1030
static inline unsigned long page_to_section(const struct page *page)
A
Andy Whitcroft 已提交
1031 1032 1033
{
	return (page->flags >> SECTIONS_PGSHIFT) & SECTIONS_MASK;
}
1034
#endif
A
Andy Whitcroft 已提交
1035

1036
static inline void set_page_zone(struct page *page, enum zone_type zone)
1037 1038 1039 1040
{
	page->flags &= ~(ZONES_MASK << ZONES_PGSHIFT);
	page->flags |= (zone & ZONES_MASK) << ZONES_PGSHIFT;
}
1041

1042 1043 1044 1045
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
1046
}
1047

1048
static inline void set_page_links(struct page *page, enum zone_type zone,
A
Andy Whitcroft 已提交
1049
	unsigned long node, unsigned long pfn)
L
Linus Torvalds 已提交
1050
{
1051 1052
	set_page_zone(page, zone);
	set_page_node(page, node);
C
Cody P Schafer 已提交
1053
#ifdef SECTION_IN_PAGE_FLAGS
A
Andy Whitcroft 已提交
1054
	set_page_section(page, pfn_to_section_nr(pfn));
1055
#endif
L
Linus Torvalds 已提交
1056 1057
}

G
Greg Thelen 已提交
1058 1059 1060 1061 1062
#ifdef CONFIG_MEMCG
static inline struct mem_cgroup *page_memcg(struct page *page)
{
	return page->mem_cgroup;
}
1063 1064 1065 1066 1067
static inline struct mem_cgroup *page_memcg_rcu(struct page *page)
{
	WARN_ON_ONCE(!rcu_read_lock_held());
	return READ_ONCE(page->mem_cgroup);
}
G
Greg Thelen 已提交
1068 1069 1070 1071 1072
#else
static inline struct mem_cgroup *page_memcg(struct page *page)
{
	return NULL;
}
1073 1074 1075 1076 1077
static inline struct mem_cgroup *page_memcg_rcu(struct page *page)
{
	WARN_ON_ONCE(!rcu_read_lock_held());
	return NULL;
}
G
Greg Thelen 已提交
1078 1079
#endif

1080 1081 1082 1083 1084
/*
 * Some inline functions in vmstat.h depend on page_zone()
 */
#include <linux/vmstat.h>

I
Ian Campbell 已提交
1085
static __always_inline void *lowmem_page_address(const struct page *page)
L
Linus Torvalds 已提交
1086
{
1087
	return page_to_virt(page);
L
Linus Torvalds 已提交
1088 1089 1090 1091 1092 1093 1094
}

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

#if defined(WANT_PAGE_VIRTUAL)
1095 1096 1097 1098 1099 1100 1101 1102
static inline void *page_address(const struct page *page)
{
	return page->virtual;
}
static inline void set_page_address(struct page *page, void *address)
{
	page->virtual = address;
}
L
Linus Torvalds 已提交
1103 1104 1105 1106
#define page_address_init()  do { } while(0)
#endif

#if defined(HASHED_PAGE_VIRTUAL)
1107
void *page_address(const struct page *page);
L
Linus Torvalds 已提交
1108 1109 1110 1111 1112 1113 1114 1115 1116 1117
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

1118 1119
extern void *page_rmapping(struct page *page);
extern struct anon_vma *page_anon_vma(struct page *page);
S
Shaohua Li 已提交
1120
extern struct address_space *page_mapping(struct page *page);
L
Linus Torvalds 已提交
1121

1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132
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;
}

1133 1134
extern pgoff_t __page_file_index(struct page *page);

L
Linus Torvalds 已提交
1135 1136
/*
 * Return the pagecache index of the passed page.  Regular pagecache pages
1137
 * use ->index whereas swapcache pages use swp_offset(->private)
L
Linus Torvalds 已提交
1138 1139 1140 1141
 */
static inline pgoff_t page_index(struct page *page)
{
	if (unlikely(PageSwapCache(page)))
1142
		return __page_file_index(page);
L
Linus Torvalds 已提交
1143 1144 1145
	return page->index;
}

A
Andrew Morton 已提交
1146
bool page_mapped(struct page *page);
1147
struct address_space *page_mapping(struct page *page);
L
Linus Torvalds 已提交
1148

1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176
/*
 * Return true only if the page has been allocated with
 * ALLOC_NO_WATERMARKS and the low watermark was not
 * met implying that the system is under some pressure.
 */
static inline bool page_is_pfmemalloc(struct page *page)
{
	/*
	 * Page index cannot be this large so this must be
	 * a pfmemalloc page.
	 */
	return page->index == -1UL;
}

/*
 * Only to be called by the page allocator on a freshly allocated
 * page.
 */
static inline void set_page_pfmemalloc(struct page *page)
{
	page->index = -1UL;
}

static inline void clear_page_pfmemalloc(struct page *page)
{
	page->index = 0;
}

L
Linus Torvalds 已提交
1177 1178 1179 1180 1181
/*
 * 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
1182

N
Nick Piggin 已提交
1183 1184 1185 1186
#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 */
1187 1188
#define VM_FAULT_HWPOISON 0x0010	/* Hit poisoned small page */
#define VM_FAULT_HWPOISON_LARGE 0x0020  /* Hit poisoned large page. Index encoded in upper bits */
1189
#define VM_FAULT_SIGSEGV 0x0040
N
Nick Piggin 已提交
1190

N
Nick Piggin 已提交
1191 1192
#define VM_FAULT_NOPAGE	0x0100	/* ->fault installed the pte, not return page */
#define VM_FAULT_LOCKED	0x0200	/* ->fault locked the returned page */
1193
#define VM_FAULT_RETRY	0x0400	/* ->fault blocked, must retry */
1194
#define VM_FAULT_FALLBACK 0x0800	/* huge page fault failed, fall back to small */
1195
#define VM_FAULT_DONE_COW   0x1000	/* ->fault has fully handled COW */
1196 1197 1198
#define VM_FAULT_NEEDDSYNC  0x2000	/* ->fault did not modify page tables
					 * and needs fsync() to complete (for
					 * synchronous page faults in DAX) */
1199

1200 1201 1202
#define VM_FAULT_ERROR	(VM_FAULT_OOM | VM_FAULT_SIGBUS | VM_FAULT_SIGSEGV | \
			 VM_FAULT_HWPOISON | VM_FAULT_HWPOISON_LARGE | \
			 VM_FAULT_FALLBACK)
1203

1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215
#define VM_FAULT_RESULT_TRACE \
	{ VM_FAULT_OOM,			"OOM" }, \
	{ VM_FAULT_SIGBUS,		"SIGBUS" }, \
	{ VM_FAULT_MAJOR,		"MAJOR" }, \
	{ VM_FAULT_WRITE,		"WRITE" }, \
	{ VM_FAULT_HWPOISON,		"HWPOISON" }, \
	{ VM_FAULT_HWPOISON_LARGE,	"HWPOISON_LARGE" }, \
	{ VM_FAULT_SIGSEGV,		"SIGSEGV" }, \
	{ VM_FAULT_NOPAGE,		"NOPAGE" }, \
	{ VM_FAULT_LOCKED,		"LOCKED" }, \
	{ VM_FAULT_RETRY,		"RETRY" }, \
	{ VM_FAULT_FALLBACK,		"FALLBACK" }, \
1216 1217
	{ VM_FAULT_DONE_COW,		"DONE_COW" }, \
	{ VM_FAULT_NEEDDSYNC,		"NEEDDSYNC" }
1218

1219 1220 1221
/* 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
1222

1223 1224 1225 1226 1227
/*
 * 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
1228 1229
#define offset_in_page(p)	((unsigned long)(p) & ~PAGE_MASK)

1230
/*
1231
 * Flags passed to show_mem() and show_free_areas() to suppress output in
1232 1233
 * various contexts.
 */
1234
#define SHOW_MEM_FILTER_NODES		(0x0001u)	/* disallowed nodes */
1235

1236
extern void show_free_areas(unsigned int flags, nodemask_t *nodemask);
L
Linus Torvalds 已提交
1237

1238
extern bool can_do_mlock(void);
L
Linus Torvalds 已提交
1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250
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 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 */
};

1251 1252 1253 1254
struct page *_vm_normal_page(struct vm_area_struct *vma, unsigned long addr,
			     pte_t pte, bool with_public_device);
#define vm_normal_page(vma, addr, pte) _vm_normal_page(vma, addr, pte, false)

1255 1256
struct page *vm_normal_page_pmd(struct vm_area_struct *vma, unsigned long addr,
				pmd_t pmd);
N
Nick Piggin 已提交
1257

1258 1259
int zap_vma_ptes(struct vm_area_struct *vma, unsigned long address,
		unsigned long size);
A
Al Viro 已提交
1260
void zap_page_range(struct vm_area_struct *vma, unsigned long address,
1261
		unsigned long size);
1262 1263
void unmap_vmas(struct mmu_gather *tlb, struct vm_area_struct *start_vma,
		unsigned long start, unsigned long end);
1264 1265 1266

/**
 * mm_walk - callbacks for walk_page_range
1267 1268 1269 1270
 * @pud_entry: if set, called for each non-empty PUD (2nd-level) entry
 *	       this handler should only handle pud_trans_huge() puds.
 *	       the pmd_entry or pte_entry callbacks will be used for
 *	       regular PUDs.
1271
 * @pmd_entry: if set, called for each non-empty PMD (3rd-level) entry
1272 1273 1274
 *	       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.
1275 1276
 * @pte_entry: if set, called for each non-empty PTE (4th-level) entry
 * @pte_hole: if set, called for each hole at all levels
1277
 * @hugetlb_entry: if set, called for each hugetlb entry
N
Naoya Horiguchi 已提交
1278
 * @test_walk: caller specific callback function to determine whether
1279
 *             we walk over the current vma or not. Returning 0
N
Naoya Horiguchi 已提交
1280 1281
 *             value means "do page table walk over the current vma,"
 *             and a negative one means "abort current page table walk
1282
 *             right now." 1 means "skip the current vma."
N
Naoya Horiguchi 已提交
1283 1284 1285
 * @mm:        mm_struct representing the target process of page table walk
 * @vma:       vma currently walked (NULL if walking outside vmas)
 * @private:   private data for callbacks' usage
1286
 *
N
Naoya Horiguchi 已提交
1287
 * (see the comment on walk_page_range() for more details)
1288 1289
 */
struct mm_walk {
1290 1291
	int (*pud_entry)(pud_t *pud, unsigned long addr,
			 unsigned long next, struct mm_walk *walk);
1292 1293 1294 1295 1296 1297 1298 1299 1300
	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);
N
Naoya Horiguchi 已提交
1301 1302
	int (*test_walk)(unsigned long addr, unsigned long next,
			struct mm_walk *walk);
D
Dave Hansen 已提交
1303
	struct mm_struct *mm;
N
Naoya Horiguchi 已提交
1304
	struct vm_area_struct *vma;
D
Dave Hansen 已提交
1305
	void *private;
1306 1307
};

D
Dave Hansen 已提交
1308 1309
int walk_page_range(unsigned long addr, unsigned long end,
		struct mm_walk *walk);
N
Naoya Horiguchi 已提交
1310
int walk_page_vma(struct vm_area_struct *vma, struct mm_walk *walk);
1311
void free_pgd_range(struct mmu_gather *tlb, unsigned long addr,
1312
		unsigned long end, unsigned long floor, unsigned long ceiling);
L
Linus Torvalds 已提交
1313 1314 1315 1316
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);
R
Ross Zwisler 已提交
1317
int follow_pte_pmd(struct mm_struct *mm, unsigned long address,
1318
			     unsigned long *start, unsigned long *end,
R
Ross Zwisler 已提交
1319
			     pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp);
J
Johannes Weiner 已提交
1320 1321
int follow_pfn(struct vm_area_struct *vma, unsigned long address,
	unsigned long *pfn);
1322 1323
int follow_phys(struct vm_area_struct *vma, unsigned long address,
		unsigned int flags, unsigned long *prot, resource_size_t *phys);
1324 1325
int generic_access_phys(struct vm_area_struct *vma, unsigned long addr,
			void *buf, int len, int write);
L
Linus Torvalds 已提交
1326 1327 1328 1329 1330 1331 1332

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);
}

1333
extern void truncate_pagecache(struct inode *inode, loff_t new);
1334
extern void truncate_setsize(struct inode *inode, loff_t newsize);
1335
void pagecache_isize_extended(struct inode *inode, loff_t from, loff_t to);
1336
void truncate_pagecache_range(struct inode *inode, loff_t offset, loff_t end);
1337
int truncate_inode_page(struct address_space *mapping, struct page *page);
1338
int generic_error_remove_page(struct address_space *mapping, struct page *page);
1339 1340
int invalidate_inode_page(struct page *page);

1341
#ifdef CONFIG_MMU
1342 1343
extern int handle_mm_fault(struct vm_area_struct *vma, unsigned long address,
		unsigned int flags);
1344
extern int fixup_user_fault(struct task_struct *tsk, struct mm_struct *mm,
1345 1346
			    unsigned long address, unsigned int fault_flags,
			    bool *unlocked);
1347
#else
1348 1349
static inline int handle_mm_fault(struct vm_area_struct *vma,
		unsigned long address, unsigned int flags)
1350 1351 1352 1353 1354
{
	/* should never happen if there's no MMU */
	BUG();
	return VM_FAULT_SIGBUS;
}
1355 1356
static inline int fixup_user_fault(struct task_struct *tsk,
		struct mm_struct *mm, unsigned long address,
1357
		unsigned int fault_flags, bool *unlocked)
1358 1359 1360 1361 1362
{
	/* should never happen if there's no MMU */
	BUG();
	return -EFAULT;
}
1363
#endif
N
Nick Piggin 已提交
1364

1365 1366
extern int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len,
		unsigned int gup_flags);
S
Stephen Wilson 已提交
1367
extern int access_remote_vm(struct mm_struct *mm, unsigned long addr,
1368
		void *buf, int len, unsigned int gup_flags);
1369 1370
extern int __access_remote_vm(struct task_struct *tsk, struct mm_struct *mm,
		unsigned long addr, void *buf, int len, unsigned int gup_flags);
L
Linus Torvalds 已提交
1371

1372 1373
long get_user_pages_remote(struct task_struct *tsk, struct mm_struct *mm,
			    unsigned long start, unsigned long nr_pages,
1374
			    unsigned int gup_flags, struct page **pages,
1375
			    struct vm_area_struct **vmas, int *locked);
1376
long get_user_pages(unsigned long start, unsigned long nr_pages,
1377
			    unsigned int gup_flags, struct page **pages,
1378
			    struct vm_area_struct **vmas);
1379
long get_user_pages_locked(unsigned long start, unsigned long nr_pages,
1380
		    unsigned int gup_flags, struct page **pages, int *locked);
1381
long get_user_pages_unlocked(unsigned long start, unsigned long nr_pages,
1382
		    struct page **pages, unsigned int gup_flags);
1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395
#ifdef CONFIG_FS_DAX
long get_user_pages_longterm(unsigned long start, unsigned long nr_pages,
			    unsigned int gup_flags, struct page **pages,
			    struct vm_area_struct **vmas);
#else
static inline long get_user_pages_longterm(unsigned long start,
		unsigned long nr_pages, unsigned int gup_flags,
		struct page **pages, struct vm_area_struct **vmas)
{
	return get_user_pages(start, nr_pages, gup_flags, pages, vmas);
}
#endif /* CONFIG_FS_DAX */

1396 1397
int get_user_pages_fast(unsigned long start, int nr_pages, int write,
			struct page **pages);
1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412

/* Container for pinned pfns / pages */
struct frame_vector {
	unsigned int nr_allocated;	/* Number of frames we have space for */
	unsigned int nr_frames;	/* Number of frames stored in ptrs array */
	bool got_ref;		/* Did we pin pages by getting page ref? */
	bool is_pfns;		/* Does array contain pages or pfns? */
	void *ptrs[0];		/* Array of pinned pfns / pages. Use
				 * pfns_vector_pages() or pfns_vector_pfns()
				 * for access */
};

struct frame_vector *frame_vector_create(unsigned int nr_frames);
void frame_vector_destroy(struct frame_vector *vec);
int get_vaddr_frames(unsigned long start, unsigned int nr_pfns,
1413
		     unsigned int gup_flags, struct frame_vector *vec);
1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440
void put_vaddr_frames(struct frame_vector *vec);
int frame_vector_to_pages(struct frame_vector *vec);
void frame_vector_to_pfns(struct frame_vector *vec);

static inline unsigned int frame_vector_count(struct frame_vector *vec)
{
	return vec->nr_frames;
}

static inline struct page **frame_vector_pages(struct frame_vector *vec)
{
	if (vec->is_pfns) {
		int err = frame_vector_to_pages(vec);

		if (err)
			return ERR_PTR(err);
	}
	return (struct page **)(vec->ptrs);
}

static inline unsigned long *frame_vector_pfns(struct frame_vector *vec)
{
	if (!vec->is_pfns)
		frame_vector_to_pfns(vec);
	return (unsigned long *)(vec->ptrs);
}

1441 1442 1443 1444
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
1445
struct page *get_dump_page(unsigned long addr);
L
Linus Torvalds 已提交
1446

1447
extern int try_to_release_page(struct page * page, gfp_t gfp_mask);
1448 1449
extern void do_invalidatepage(struct page *page, unsigned int offset,
			      unsigned int length);
1450

L
Linus Torvalds 已提交
1451
int __set_page_dirty_nobuffers(struct page *page);
1452
int __set_page_dirty_no_writeback(struct page *page);
L
Linus Torvalds 已提交
1453 1454
int redirty_page_for_writepage(struct writeback_control *wbc,
				struct page *page);
J
Johannes Weiner 已提交
1455
void account_page_dirtied(struct page *page, struct address_space *mapping);
1456
void account_page_cleaned(struct page *page, struct address_space *mapping,
J
Johannes Weiner 已提交
1457
			  struct bdi_writeback *wb);
1458
int set_page_dirty(struct page *page);
L
Linus Torvalds 已提交
1459
int set_page_dirty_lock(struct page *page);
1460 1461 1462 1463 1464 1465 1466
void __cancel_dirty_page(struct page *page);
static inline void cancel_dirty_page(struct page *page)
{
	/* Avoid atomic ops, locking, etc. when not actually needed. */
	if (PageDirty(page))
		__cancel_dirty_page(page);
}
L
Linus Torvalds 已提交
1467
int clear_page_dirty_for_io(struct page *page);
1468

1469
int get_cmdline(struct task_struct *task, char *buffer, int buflen);
L
Linus Torvalds 已提交
1470

1471 1472 1473 1474 1475
static inline bool vma_is_anonymous(struct vm_area_struct *vma)
{
	return !vma->vm_ops;
}

1476 1477 1478 1479 1480 1481 1482 1483 1484 1485
#ifdef CONFIG_SHMEM
/*
 * The vma_is_shmem is not inline because it is used only by slow
 * paths in userfault.
 */
bool vma_is_shmem(struct vm_area_struct *vma);
#else
static inline bool vma_is_shmem(struct vm_area_struct *vma) { return false; }
#endif

1486
int vma_is_stack_for_current(struct vm_area_struct *vma);
1487

1488 1489
extern unsigned long move_page_tables(struct vm_area_struct *vma,
		unsigned long old_addr, struct vm_area_struct *new_vma,
1490 1491
		unsigned long new_addr, unsigned long len,
		bool need_rmap_locks);
1492 1493
extern unsigned long change_protection(struct vm_area_struct *vma, unsigned long start,
			      unsigned long end, pgprot_t newprot,
1494
			      int dirty_accountable, int prot_numa);
1495 1496 1497
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
1498

1499 1500 1501 1502 1503
/*
 * 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
1504 1505 1506 1507 1508
/*
 * per-process(per-mm_struct) statistics.
 */
static inline unsigned long get_mm_counter(struct mm_struct *mm, int member)
{
1509 1510 1511 1512 1513 1514 1515 1516 1517
	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;
1518
#endif
1519 1520
	return (unsigned long)val;
}
K
KAMEZAWA Hiroyuki 已提交
1521 1522 1523

static inline void add_mm_counter(struct mm_struct *mm, int member, long value)
{
1524
	atomic_long_add(value, &mm->rss_stat.count[member]);
K
KAMEZAWA Hiroyuki 已提交
1525 1526 1527 1528
}

static inline void inc_mm_counter(struct mm_struct *mm, int member)
{
1529
	atomic_long_inc(&mm->rss_stat.count[member]);
K
KAMEZAWA Hiroyuki 已提交
1530 1531 1532 1533
}

static inline void dec_mm_counter(struct mm_struct *mm, int member)
{
1534
	atomic_long_dec(&mm->rss_stat.count[member]);
K
KAMEZAWA Hiroyuki 已提交
1535 1536
}

1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551
/* Optimized variant when page is already known not to be PageAnon */
static inline int mm_counter_file(struct page *page)
{
	if (PageSwapBacked(page))
		return MM_SHMEMPAGES;
	return MM_FILEPAGES;
}

static inline int mm_counter(struct page *page)
{
	if (PageAnon(page))
		return MM_ANONPAGES;
	return mm_counter_file(page);
}

K
KAMEZAWA Hiroyuki 已提交
1552 1553 1554
static inline unsigned long get_mm_rss(struct mm_struct *mm)
{
	return get_mm_counter(mm, MM_FILEPAGES) +
1555 1556
		get_mm_counter(mm, MM_ANONPAGES) +
		get_mm_counter(mm, MM_SHMEMPAGES);
K
KAMEZAWA Hiroyuki 已提交
1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582
}

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;
}

1583 1584 1585 1586 1587
static inline void reset_mm_hiwater_rss(struct mm_struct *mm)
{
	mm->hiwater_rss = get_mm_rss(mm);
}

K
KAMEZAWA Hiroyuki 已提交
1588 1589 1590 1591 1592 1593 1594 1595 1596
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
1597
#if defined(SPLIT_RSS_COUNTING)
1598
void sync_mm_rss(struct mm_struct *mm);
K
KAMEZAWA Hiroyuki 已提交
1599
#else
1600
static inline void sync_mm_rss(struct mm_struct *mm)
K
KAMEZAWA Hiroyuki 已提交
1601 1602 1603
{
}
#endif
1604

1605 1606 1607 1608 1609 1610 1611
#ifndef __HAVE_ARCH_PTE_DEVMAP
static inline int pte_devmap(pte_t pte)
{
	return 0;
}
#endif

1612
int vma_wants_writenotify(struct vm_area_struct *vma, pgprot_t vm_page_prot);
1613

1614 1615 1616 1617 1618 1619 1620 1621 1622
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;
}
1623

1624 1625 1626 1627 1628 1629 1630 1631 1632 1633
#ifdef __PAGETABLE_P4D_FOLDED
static inline int __p4d_alloc(struct mm_struct *mm, pgd_t *pgd,
						unsigned long address)
{
	return 0;
}
#else
int __p4d_alloc(struct mm_struct *mm, pgd_t *pgd, unsigned long address);
#endif

K
Kirill A. Shutemov 已提交
1634
#if defined(__PAGETABLE_PUD_FOLDED) || !defined(CONFIG_MMU)
1635
static inline int __pud_alloc(struct mm_struct *mm, p4d_t *p4d,
N
Nick Piggin 已提交
1636 1637 1638 1639
						unsigned long address)
{
	return 0;
}
K
Kirill A. Shutemov 已提交
1640 1641 1642
static inline void mm_inc_nr_puds(struct mm_struct *mm) {}
static inline void mm_dec_nr_puds(struct mm_struct *mm) {}

N
Nick Piggin 已提交
1643
#else
1644
int __pud_alloc(struct mm_struct *mm, p4d_t *p4d, unsigned long address);
K
Kirill A. Shutemov 已提交
1645 1646 1647

static inline void mm_inc_nr_puds(struct mm_struct *mm)
{
1648
	atomic_long_add(PTRS_PER_PUD * sizeof(pud_t), &mm->pgtables_bytes);
K
Kirill A. Shutemov 已提交
1649 1650 1651 1652
}

static inline void mm_dec_nr_puds(struct mm_struct *mm)
{
1653
	atomic_long_sub(PTRS_PER_PUD * sizeof(pud_t), &mm->pgtables_bytes);
K
Kirill A. Shutemov 已提交
1654
}
N
Nick Piggin 已提交
1655 1656
#endif

1657
#if defined(__PAGETABLE_PMD_FOLDED) || !defined(CONFIG_MMU)
N
Nick Piggin 已提交
1658 1659 1660 1661 1662
static inline int __pmd_alloc(struct mm_struct *mm, pud_t *pud,
						unsigned long address)
{
	return 0;
}
1663 1664 1665 1666

static inline void mm_inc_nr_pmds(struct mm_struct *mm) {}
static inline void mm_dec_nr_pmds(struct mm_struct *mm) {}

N
Nick Piggin 已提交
1667
#else
1668
int __pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address);
1669 1670 1671

static inline void mm_inc_nr_pmds(struct mm_struct *mm)
{
1672
	atomic_long_add(PTRS_PER_PMD * sizeof(pmd_t), &mm->pgtables_bytes);
1673 1674 1675 1676
}

static inline void mm_dec_nr_pmds(struct mm_struct *mm)
{
1677
	atomic_long_sub(PTRS_PER_PMD * sizeof(pmd_t), &mm->pgtables_bytes);
1678
}
N
Nick Piggin 已提交
1679 1680
#endif

1681
#ifdef CONFIG_MMU
1682
static inline void mm_pgtables_bytes_init(struct mm_struct *mm)
1683
{
1684
	atomic_long_set(&mm->pgtables_bytes, 0);
1685 1686
}

1687
static inline unsigned long mm_pgtables_bytes(const struct mm_struct *mm)
1688
{
1689
	return atomic_long_read(&mm->pgtables_bytes);
1690 1691 1692 1693
}

static inline void mm_inc_nr_ptes(struct mm_struct *mm)
{
1694
	atomic_long_add(PTRS_PER_PTE * sizeof(pte_t), &mm->pgtables_bytes);
1695 1696 1697 1698
}

static inline void mm_dec_nr_ptes(struct mm_struct *mm)
{
1699
	atomic_long_sub(PTRS_PER_PTE * sizeof(pte_t), &mm->pgtables_bytes);
1700 1701 1702
}
#else

1703 1704
static inline void mm_pgtables_bytes_init(struct mm_struct *mm) {}
static inline unsigned long mm_pgtables_bytes(const struct mm_struct *mm)
1705 1706 1707 1708 1709 1710 1711 1712
{
	return 0;
}

static inline void mm_inc_nr_ptes(struct mm_struct *mm) {}
static inline void mm_dec_nr_ptes(struct mm_struct *mm) {}
#endif

1713
int __pte_alloc(struct mm_struct *mm, pmd_t *pmd, unsigned long address);
1714 1715
int __pte_alloc_kernel(pmd_t *pmd, unsigned long address);

L
Linus Torvalds 已提交
1716 1717 1718 1719
/*
 * The following ifdef needed to get the 4level-fixup.h header to work.
 * Remove it when 4level-fixup.h has been removed.
 */
1720
#if defined(CONFIG_MMU) && !defined(__ARCH_HAS_4LEVEL_HACK)
1721 1722

#ifndef __ARCH_HAS_5LEVEL_HACK
1723 1724 1725 1726 1727 1728 1729 1730 1731
static inline p4d_t *p4d_alloc(struct mm_struct *mm, pgd_t *pgd,
		unsigned long address)
{
	return (unlikely(pgd_none(*pgd)) && __p4d_alloc(mm, pgd, address)) ?
		NULL : p4d_offset(pgd, address);
}

static inline pud_t *pud_alloc(struct mm_struct *mm, p4d_t *p4d,
		unsigned long address)
L
Linus Torvalds 已提交
1732
{
1733 1734
	return (unlikely(p4d_none(*p4d)) && __pud_alloc(mm, p4d, address)) ?
		NULL : pud_offset(p4d, address);
L
Linus Torvalds 已提交
1735
}
1736
#endif /* !__ARCH_HAS_5LEVEL_HACK */
L
Linus Torvalds 已提交
1737 1738 1739

static inline pmd_t *pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address)
{
1740 1741
	return (unlikely(pud_none(*pud)) && __pmd_alloc(mm, pud, address))?
		NULL: pmd_offset(pud, address);
L
Linus Torvalds 已提交
1742
}
1743 1744
#endif /* CONFIG_MMU && !__ARCH_HAS_4LEVEL_HACK */

1745
#if USE_SPLIT_PTE_PTLOCKS
1746
#if ALLOC_SPLIT_PTLOCKS
1747
void __init ptlock_cache_init(void);
1748 1749 1750 1751 1752 1753 1754
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;
}
1755
#else /* ALLOC_SPLIT_PTLOCKS */
1756 1757 1758 1759
static inline void ptlock_cache_init(void)
{
}

1760 1761 1762 1763
static inline bool ptlock_alloc(struct page *page)
{
	return true;
}
1764

1765 1766 1767 1768 1769 1770
static inline void ptlock_free(struct page *page)
{
}

static inline spinlock_t *ptlock_ptr(struct page *page)
{
1771
	return &page->ptl;
1772
}
1773
#endif /* ALLOC_SPLIT_PTLOCKS */
1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786

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:
1787
	 * slab code uses page->slab_cache, which share storage with page->ptl.
1788
	 */
1789
	VM_BUG_ON_PAGE(*(unsigned long *)&page->ptl, page);
1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802
	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);
}

1803
#else	/* !USE_SPLIT_PTE_PTLOCKS */
H
Hugh Dickins 已提交
1804 1805 1806
/*
 * We use mm->page_table_lock to guard all pagetable pages of the mm.
 */
1807 1808 1809 1810
static inline spinlock_t *pte_lockptr(struct mm_struct *mm, pmd_t *pmd)
{
	return &mm->page_table_lock;
}
1811
static inline void ptlock_cache_init(void) {}
1812 1813
static inline bool ptlock_init(struct page *page) { return true; }
static inline void pte_lock_deinit(struct page *page) {}
1814
#endif /* USE_SPLIT_PTE_PTLOCKS */
H
Hugh Dickins 已提交
1815

1816 1817 1818 1819 1820 1821
static inline void pgtable_init(void)
{
	ptlock_cache_init();
	pgtable_cache_init();
}

1822
static inline bool pgtable_page_ctor(struct page *page)
1823
{
1824 1825
	if (!ptlock_init(page))
		return false;
1826
	inc_zone_page_state(page, NR_PAGETABLE);
1827
	return true;
1828 1829 1830 1831 1832 1833 1834 1835
}

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
1836 1837
#define pte_offset_map_lock(mm, pmd, address, ptlp)	\
({							\
H
Hugh Dickins 已提交
1838
	spinlock_t *__ptl = pte_lockptr(mm, pmd);	\
H
Hugh Dickins 已提交
1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849
	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)

1850 1851 1852 1853 1854
#define pte_alloc(mm, pmd, address)			\
	(unlikely(pmd_none(*(pmd))) && __pte_alloc(mm, pmd, address))

#define pte_alloc_map(mm, pmd, address)			\
	(pte_alloc(mm, pmd, address) ? NULL : pte_offset_map(pmd, address))
1855

H
Hugh Dickins 已提交
1856
#define pte_alloc_map_lock(mm, pmd, address, ptlp)	\
1857 1858
	(pte_alloc(mm, pmd, address) ?			\
		 NULL : pte_offset_map_lock(mm, pmd, address, ptlp))
H
Hugh Dickins 已提交
1859

1860
#define pte_alloc_kernel(pmd, address)			\
1861
	((unlikely(pmd_none(*(pmd))) && __pte_alloc_kernel(pmd, address))? \
1862
		NULL: pte_offset_kernel(pmd, address))
L
Linus Torvalds 已提交
1863

1864 1865
#if USE_SPLIT_PMD_PTLOCKS

1866 1867 1868 1869 1870 1871
static struct page *pmd_to_page(pmd_t *pmd)
{
	unsigned long mask = ~(PTRS_PER_PMD * sizeof(pmd_t) - 1);
	return virt_to_page((void *)((unsigned long) pmd & mask));
}

1872 1873
static inline spinlock_t *pmd_lockptr(struct mm_struct *mm, pmd_t *pmd)
{
1874
	return ptlock_ptr(pmd_to_page(pmd));
1875 1876 1877 1878 1879 1880 1881
}

static inline bool pgtable_pmd_page_ctor(struct page *page)
{
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
	page->pmd_huge_pte = NULL;
#endif
1882
	return ptlock_init(page);
1883 1884 1885 1886 1887
}

static inline void pgtable_pmd_page_dtor(struct page *page)
{
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
1888
	VM_BUG_ON_PAGE(page->pmd_huge_pte, page);
1889
#endif
1890
	ptlock_free(page);
1891 1892
}

1893
#define pmd_huge_pte(mm, pmd) (pmd_to_page(pmd)->pmd_huge_pte)
1894 1895 1896

#else

1897 1898 1899 1900 1901
static inline spinlock_t *pmd_lockptr(struct mm_struct *mm, pmd_t *pmd)
{
	return &mm->page_table_lock;
}

1902 1903 1904
static inline bool pgtable_pmd_page_ctor(struct page *page) { return true; }
static inline void pgtable_pmd_page_dtor(struct page *page) {}

1905
#define pmd_huge_pte(mm, pmd) ((mm)->pmd_huge_pte)
1906

1907 1908
#endif

1909 1910 1911 1912 1913 1914 1915
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;
}

1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933
/*
 * No scalability reason to split PUD locks yet, but follow the same pattern
 * as the PMD locks to make it easier if we decide to.  The VM should not be
 * considered ready to switch to split PUD locks yet; there may be places
 * which need to be converted from page_table_lock.
 */
static inline spinlock_t *pud_lockptr(struct mm_struct *mm, pud_t *pud)
{
	return &mm->page_table_lock;
}

static inline spinlock_t *pud_lock(struct mm_struct *mm, pud_t *pud)
{
	spinlock_t *ptl = pud_lockptr(mm, pud);

	spin_lock(ptl);
	return ptl;
}
1934

1935
extern void __init pagecache_init(void);
L
Linus Torvalds 已提交
1936
extern void free_area_init(unsigned long * zones_size);
1937 1938
extern void free_area_init_node(int nid, unsigned long * zones_size,
		unsigned long zone_start_pfn, unsigned long *zholes_size);
1939 1940
extern void free_initmem(void);

1941 1942 1943
/*
 * Free reserved pages within range [PAGE_ALIGN(start), end & PAGE_MASK)
 * into the buddy system. The freed pages will be poisoned with pattern
1944
 * "poison" if it's within range [0, UCHAR_MAX].
1945 1946
 * Return pages freed into the buddy system.
 */
1947
extern unsigned long free_reserved_area(void *start, void *end,
1948
					int poison, char *s);
1949

1950 1951 1952 1953 1954 1955 1956
#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
1957

1958
extern void adjust_managed_page_count(struct page *page, long count);
1959
extern void mem_init_print_info(const char *str);
1960

1961
extern void reserve_bootmem_region(phys_addr_t start, phys_addr_t end);
1962

1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984
/* 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.
1985 1986 1987
 * The freed pages will be poisoned with pattern "poison" if it's within
 * range [0, UCHAR_MAX].
 * Return pages freed into the buddy system.
1988 1989 1990 1991 1992
 */
static inline unsigned long free_initmem_default(int poison)
{
	extern char __init_begin[], __init_end[];

1993
	return free_reserved_area(&__init_begin, &__init_end,
1994 1995 1996
				  poison, "unused kernel");
}

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
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
2008
#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP
2009
/*
T
Tejun Heo 已提交
2010
 * With CONFIG_HAVE_MEMBLOCK_NODE_MAP set, an architecture may initialise its
2011 2012 2013 2014 2015 2016
 * 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
2017
 * physical memory with memblock_add[_node]() before calling
2018 2019 2020 2021 2022 2023
 * 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
2024
 * 	memblock_add_node(base, size, nid)
2025 2026
 * free_area_init_nodes(max_zone_pfns);
 *
T
Tejun Heo 已提交
2027 2028 2029 2030
 * 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.
2031 2032
 *
 * See mm/page_alloc.c for more information on each function exposed by
T
Tejun Heo 已提交
2033
 * CONFIG_HAVE_MEMBLOCK_NODE_MAP.
2034 2035
 */
extern void free_area_init_nodes(unsigned long *max_zone_pfn);
2036
unsigned long node_map_pfn_alignment(void);
2037 2038
unsigned long __absent_pages_in_range(int nid, unsigned long start_pfn,
						unsigned long end_pfn);
2039 2040 2041 2042 2043 2044 2045 2046
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);
2047

T
Tejun Heo 已提交
2048
#endif /* CONFIG_HAVE_MEMBLOCK_NODE_MAP */
2049

T
Tejun Heo 已提交
2050
#if !defined(CONFIG_HAVE_MEMBLOCK_NODE_MAP) && \
2051
    !defined(CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID)
2052 2053
static inline int __early_pfn_to_nid(unsigned long pfn,
					struct mminit_pfnnid_cache *state)
2054 2055 2056 2057 2058 2059 2060
{
	return 0;
}
#else
/* please see mm/page_alloc.c */
extern int __meminit early_pfn_to_nid(unsigned long pfn);
/* there is a per-arch backend function. */
2061 2062
extern int __meminit __early_pfn_to_nid(unsigned long pfn,
					struct mminit_pfnnid_cache *state);
2063 2064
#endif

2065 2066 2067 2068 2069 2070
#ifdef CONFIG_HAVE_MEMBLOCK
void zero_resv_unavail(void);
#else
static inline void zero_resv_unavail(void) {}
#endif

2071
extern void set_dma_reserve(unsigned long new_dma_reserve);
D
Dave Hansen 已提交
2072 2073
extern void memmap_init_zone(unsigned long, int, unsigned long,
				unsigned long, enum memmap_context);
2074
extern void setup_per_zone_wmarks(void);
2075
extern int __meminit init_per_zone_wmark_min(void);
L
Linus Torvalds 已提交
2076
extern void mem_init(void);
2077
extern void __init mmap_init(void);
2078
extern void show_mem(unsigned int flags, nodemask_t *nodemask);
2079
extern long si_mem_available(void);
L
Linus Torvalds 已提交
2080 2081
extern void si_meminfo(struct sysinfo * val);
extern void si_meminfo_node(struct sysinfo *val, int nid);
2082 2083 2084
#ifdef __HAVE_ARCH_RESERVED_KERNEL_PAGES
extern unsigned long arch_reserved_kernel_pages(void);
#endif
L
Linus Torvalds 已提交
2085

2086 2087
extern __printf(3, 4)
void warn_alloc(gfp_t gfp_mask, nodemask_t *nodemask, const char *fmt, ...);
2088

2089 2090
extern void setup_per_cpu_pageset(void);

2091
extern void zone_pcp_update(struct zone *zone);
2092
extern void zone_pcp_reset(struct zone *zone);
2093

2094 2095
/* page_alloc.c */
extern int min_free_kbytes;
2096
extern int watermark_scale_factor;
2097

2098
/* nommu.c */
2099
extern atomic_long_t mmap_pages_allocated;
2100
extern int nommu_shrink_inode_mappings(struct inode *, size_t, size_t);
2101

2102 2103
/* interval_tree.c */
void vma_interval_tree_insert(struct vm_area_struct *node,
2104
			      struct rb_root_cached *root);
M
Michel Lespinasse 已提交
2105 2106
void vma_interval_tree_insert_after(struct vm_area_struct *node,
				    struct vm_area_struct *prev,
2107
				    struct rb_root_cached *root);
2108
void vma_interval_tree_remove(struct vm_area_struct *node,
2109 2110
			      struct rb_root_cached *root);
struct vm_area_struct *vma_interval_tree_iter_first(struct rb_root_cached *root,
2111 2112 2113 2114 2115 2116 2117
				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
2118

2119
void anon_vma_interval_tree_insert(struct anon_vma_chain *node,
2120
				   struct rb_root_cached *root);
2121
void anon_vma_interval_tree_remove(struct anon_vma_chain *node,
2122 2123 2124 2125
				   struct rb_root_cached *root);
struct anon_vma_chain *
anon_vma_interval_tree_iter_first(struct rb_root_cached *root,
				  unsigned long start, unsigned long last);
2126 2127
struct anon_vma_chain *anon_vma_interval_tree_iter_next(
	struct anon_vma_chain *node, unsigned long start, unsigned long last);
2128 2129 2130
#ifdef CONFIG_DEBUG_VM_RB
void anon_vma_interval_tree_verify(struct anon_vma_chain *node);
#endif
2131 2132 2133 2134 2135

#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
2136
/* mmap.c */
2137
extern int __vm_enough_memory(struct mm_struct *mm, long pages, int cap_sys_admin);
2138 2139 2140 2141 2142 2143 2144 2145
extern int __vma_adjust(struct vm_area_struct *vma, unsigned long start,
	unsigned long end, pgoff_t pgoff, struct vm_area_struct *insert,
	struct vm_area_struct *expand);
static inline int vma_adjust(struct vm_area_struct *vma, unsigned long start,
	unsigned long end, pgoff_t pgoff, struct vm_area_struct *insert)
{
	return __vma_adjust(vma, start, end, pgoff, insert, NULL);
}
L
Linus Torvalds 已提交
2146 2147 2148
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,
2149
	struct mempolicy *, struct vm_userfaultfd_ctx);
L
Linus Torvalds 已提交
2150
extern struct anon_vma *find_mergeable_anon_vma(struct vm_area_struct *);
2151 2152 2153 2154
extern int __split_vma(struct mm_struct *, struct vm_area_struct *,
	unsigned long addr, int new_below);
extern int split_vma(struct mm_struct *, struct vm_area_struct *,
	unsigned long addr, int new_below);
L
Linus Torvalds 已提交
2155 2156 2157
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 *);
2158
extern void unlink_file_vma(struct vm_area_struct *);
L
Linus Torvalds 已提交
2159
extern struct vm_area_struct *copy_vma(struct vm_area_struct **,
2160 2161
	unsigned long addr, unsigned long len, pgoff_t pgoff,
	bool *need_rmap_locks);
L
Linus Torvalds 已提交
2162
extern void exit_mmap(struct mm_struct *);
M
Matt Helsley 已提交
2163

2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177
static inline int check_data_rlimit(unsigned long rlim,
				    unsigned long new,
				    unsigned long start,
				    unsigned long end_data,
				    unsigned long start_data)
{
	if (rlim < RLIM_INFINITY) {
		if (((new - start) + (end_data - start_data)) > rlim)
			return -ENOSPC;
	}

	return 0;
}

2178 2179 2180
extern int mm_take_all_locks(struct mm_struct *mm);
extern void mm_drop_all_locks(struct mm_struct *mm);

2181 2182
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
Mateusz Guzik 已提交
2183
extern struct file *get_task_exe_file(struct task_struct *task);
M
Matt Helsley 已提交
2184

2185 2186 2187
extern bool may_expand_vm(struct mm_struct *, vm_flags_t, unsigned long npages);
extern void vm_stat_account(struct mm_struct *, vm_flags_t, long npages);

2188 2189
extern bool vma_is_special_mapping(const struct vm_area_struct *vma,
				   const struct vm_special_mapping *sm);
2190 2191
extern struct vm_area_struct *_install_special_mapping(struct mm_struct *mm,
				   unsigned long addr, unsigned long len,
2192 2193 2194
				   unsigned long flags,
				   const struct vm_special_mapping *spec);
/* This is an obsolete alternative to _install_special_mapping. */
2195 2196 2197
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
2198 2199 2200

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

M
Miklos Szeredi 已提交
2201
extern unsigned long mmap_region(struct file *file, unsigned long addr,
2202 2203
	unsigned long len, vm_flags_t vm_flags, unsigned long pgoff,
	struct list_head *uf);
2204
extern unsigned long do_mmap(struct file *file, unsigned long addr,
2205
	unsigned long len, unsigned long prot, unsigned long flags,
2206 2207 2208 2209
	vm_flags_t vm_flags, unsigned long pgoff, unsigned long *populate,
	struct list_head *uf);
extern int do_munmap(struct mm_struct *, unsigned long, size_t,
		     struct list_head *uf);
L
Linus Torvalds 已提交
2210

2211 2212 2213
static inline unsigned long
do_mmap_pgoff(struct file *file, unsigned long addr,
	unsigned long len, unsigned long prot, unsigned long flags,
2214 2215
	unsigned long pgoff, unsigned long *populate,
	struct list_head *uf)
2216
{
2217
	return do_mmap(file, addr, len, prot, flags, 0, pgoff, populate, uf);
2218 2219
}

2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231
#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

2232
/* These take the mm semaphore themselves */
2233
extern int __must_check vm_brk(unsigned long, unsigned long);
2234
extern int __must_check vm_brk_flags(unsigned long, unsigned long, unsigned long);
A
Al Viro 已提交
2235
extern int vm_munmap(unsigned long, size_t);
M
Michal Hocko 已提交
2236
extern unsigned long __must_check vm_mmap(struct file *, unsigned long,
2237 2238
        unsigned long, unsigned long,
        unsigned long, unsigned long);
L
Linus Torvalds 已提交
2239

2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264
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)
{
2265
	if (info->flags & VM_UNMAPPED_AREA_TOPDOWN)
2266
		return unmapped_area_topdown(info);
2267 2268
	else
		return unmapped_area(info);
2269 2270
}

2271
/* truncate.c */
L
Linus Torvalds 已提交
2272
extern void truncate_inode_pages(struct address_space *, loff_t);
2273 2274
extern void truncate_inode_pages_range(struct address_space *,
				       loff_t lstart, loff_t lend);
2275
extern void truncate_inode_pages_final(struct address_space *);
L
Linus Torvalds 已提交
2276 2277

/* generic vm_area_ops exported for stackable file systems */
2278
extern int filemap_fault(struct vm_fault *vmf);
J
Jan Kara 已提交
2279
extern void filemap_map_pages(struct vm_fault *vmf,
K
Kirill A. Shutemov 已提交
2280
		pgoff_t start_pgoff, pgoff_t end_pgoff);
2281
extern int filemap_page_mkwrite(struct vm_fault *vmf);
L
Linus Torvalds 已提交
2282 2283

/* mm/page-writeback.c */
2284
int __must_check write_one_page(struct page *page);
N
Nick Piggin 已提交
2285
void task_dirty_inc(struct task_struct *tsk);
L
Linus Torvalds 已提交
2286 2287 2288 2289 2290 2291

/* 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
2292
			pgoff_t offset, unsigned long nr_to_read);
2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306

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);

2307
extern unsigned long stack_guard_gap;
2308
/* Generic expand stack which grows the stack according to GROWS{UP,DOWN} */
H
Hugh Dickins 已提交
2309
extern int expand_stack(struct vm_area_struct *vma, unsigned long address);
2310 2311 2312 2313

/* CONFIG_STACK_GROWSUP still needs to to grow downwards at some places */
extern int expand_downwards(struct vm_area_struct *vma,
		unsigned long address);
2314
#if VM_GROWSUP
H
Hugh Dickins 已提交
2315
extern int expand_upwards(struct vm_area_struct *vma, unsigned long address);
2316
#else
2317
  #define expand_upwards(vma, address) (0)
2318
#endif
L
Linus Torvalds 已提交
2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335

/* 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;
}

2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359
static inline unsigned long vm_start_gap(struct vm_area_struct *vma)
{
	unsigned long vm_start = vma->vm_start;

	if (vma->vm_flags & VM_GROWSDOWN) {
		vm_start -= stack_guard_gap;
		if (vm_start > vma->vm_start)
			vm_start = 0;
	}
	return vm_start;
}

static inline unsigned long vm_end_gap(struct vm_area_struct *vma)
{
	unsigned long vm_end = vma->vm_end;

	if (vma->vm_flags & VM_GROWSUP) {
		vm_end += stack_guard_gap;
		if (vm_end < vma->vm_end)
			vm_end = -PAGE_SIZE;
	}
	return vm_end;
}

L
Linus Torvalds 已提交
2360 2361 2362 2363 2364
static inline unsigned long vma_pages(struct vm_area_struct *vma)
{
	return (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
}

2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376
/* 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;
}

2377
#ifdef CONFIG_MMU
2378
pgprot_t vm_get_page_prot(unsigned long vm_flags);
2379
void vma_set_page_prot(struct vm_area_struct *vma);
2380 2381 2382 2383 2384
#else
static inline pgprot_t vm_get_page_prot(unsigned long vm_flags)
{
	return __pgprot(0);
}
2385 2386 2387 2388
static inline void vma_set_page_prot(struct vm_area_struct *vma)
{
	vma->vm_page_prot = vm_get_page_prot(vma->vm_flags);
}
2389 2390
#endif

2391
#ifdef CONFIG_NUMA_BALANCING
2392
unsigned long change_prot_numa(struct vm_area_struct *vma,
L
Lee Schermerhorn 已提交
2393 2394 2395
			unsigned long start, unsigned long end);
#endif

2396 2397 2398
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);
2399
int vm_insert_page(struct vm_area_struct *, unsigned long addr, struct page *);
N
Nick Piggin 已提交
2400 2401
int vm_insert_pfn(struct vm_area_struct *vma, unsigned long addr,
			unsigned long pfn);
A
Andy Lutomirski 已提交
2402 2403
int vm_insert_pfn_prot(struct vm_area_struct *vma, unsigned long addr,
			unsigned long pfn, pgprot_t pgprot);
N
Nick Piggin 已提交
2404
int vm_insert_mixed(struct vm_area_struct *vma, unsigned long addr,
2405
			pfn_t pfn);
R
Ross Zwisler 已提交
2406 2407
int vm_insert_mixed_mkwrite(struct vm_area_struct *vma, unsigned long addr,
			pfn_t pfn);
2408 2409
int vm_iomap_memory(struct vm_area_struct *vma, phys_addr_t start, unsigned long len);

2410

2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421
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);
}

2422 2423 2424
#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
2425
#define FOLL_DUMP	0x08	/* give error on hole if it would be zero */
H
Hugh Dickins 已提交
2426
#define FOLL_FORCE	0x10	/* get_user_pages read/write w/o permission */
2427 2428
#define FOLL_NOWAIT	0x20	/* if a disk transfer is needed, start the IO
				 * and return without waiting upon it */
2429
#define FOLL_POPULATE	0x40	/* fault in page */
2430
#define FOLL_SPLIT	0x80	/* don't return transhuge pages, split them */
2431
#define FOLL_HWPOISON	0x100	/* check page is hwpoisoned */
2432
#define FOLL_NUMA	0x200	/* force NUMA hinting page fault */
2433
#define FOLL_MIGRATION	0x400	/* wait for page to replace migration entry */
2434
#define FOLL_TRIED	0x800	/* a retry, previous pass started an IO */
E
Eric B Munson 已提交
2435
#define FOLL_MLOCK	0x1000	/* lock present pages */
2436
#define FOLL_REMOTE	0x2000	/* we are working on non-current tsk/mm */
2437
#define FOLL_COW	0x4000	/* internal GUP flag */
L
Linus Torvalds 已提交
2438

2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449
static inline int vm_fault_to_errno(int vm_fault, int foll_flags)
{
	if (vm_fault & VM_FAULT_OOM)
		return -ENOMEM;
	if (vm_fault & (VM_FAULT_HWPOISON | VM_FAULT_HWPOISON_LARGE))
		return (foll_flags & FOLL_HWPOISON) ? -EHWPOISON : -EFAULT;
	if (vm_fault & (VM_FAULT_SIGBUS | VM_FAULT_SIGSEGV))
		return -EFAULT;
	return 0;
}

2450
typedef int (*pte_fn_t)(pte_t *pte, pgtable_t token, unsigned long addr,
2451 2452 2453 2454
			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
2455

2456 2457 2458
#ifdef CONFIG_PAGE_POISONING
extern bool page_poisoning_enabled(void);
extern void kernel_poison_pages(struct page *page, int numpages, int enable);
2459
extern bool page_is_poisoned(struct page *page);
2460 2461 2462 2463
#else
static inline bool page_poisoning_enabled(void) { return false; }
static inline void kernel_poison_pages(struct page *page, int numpages,
					int enable) { }
2464
static inline bool page_is_poisoned(struct page *page) { return false; }
2465 2466
#endif

2467
#ifdef CONFIG_DEBUG_PAGEALLOC
2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483
extern bool _debug_pagealloc_enabled;
extern void __kernel_map_pages(struct page *page, int numpages, int enable);

static inline bool debug_pagealloc_enabled(void)
{
	return _debug_pagealloc_enabled;
}

static inline void
kernel_map_pages(struct page *page, int numpages, int enable)
{
	if (!debug_pagealloc_enabled())
		return;

	__kernel_map_pages(page, numpages, enable);
}
2484 2485
#ifdef CONFIG_HIBERNATION
extern bool kernel_page_present(struct page *page);
2486 2487
#endif	/* CONFIG_HIBERNATION */
#else	/* CONFIG_DEBUG_PAGEALLOC */
L
Linus Torvalds 已提交
2488
static inline void
N
Nick Piggin 已提交
2489
kernel_map_pages(struct page *page, int numpages, int enable) {}
2490 2491
#ifdef CONFIG_HIBERNATION
static inline bool kernel_page_present(struct page *page) { return true; }
2492 2493 2494 2495 2496 2497
#endif	/* CONFIG_HIBERNATION */
static inline bool debug_pagealloc_enabled(void)
{
	return false;
}
#endif	/* CONFIG_DEBUG_PAGEALLOC */
L
Linus Torvalds 已提交
2498

2499
#ifdef __HAVE_ARCH_GATE_AREA
2500
extern struct vm_area_struct *get_gate_vma(struct mm_struct *mm);
2501 2502
extern int in_gate_area_no_mm(unsigned long addr);
extern int in_gate_area(struct mm_struct *mm, unsigned long addr);
L
Linus Torvalds 已提交
2503
#else
2504 2505 2506 2507 2508 2509 2510 2511 2512
static inline struct vm_area_struct *get_gate_vma(struct mm_struct *mm)
{
	return NULL;
}
static inline int in_gate_area_no_mm(unsigned long addr) { return 0; }
static inline int in_gate_area(struct mm_struct *mm, unsigned long addr)
{
	return 0;
}
L
Linus Torvalds 已提交
2513 2514
#endif	/* __HAVE_ARCH_GATE_AREA */

2515 2516
extern bool process_shares_mm(struct task_struct *p, struct mm_struct *mm);

2517 2518
#ifdef CONFIG_SYSCTL
extern int sysctl_drop_caches;
2519
int drop_caches_sysctl_handler(struct ctl_table *, int,
A
Andrew Morton 已提交
2520
					void __user *, size_t *, loff_t *);
2521 2522
#endif

2523 2524
void drop_slab(void);
void drop_slab_node(int nid);
A
Andrew Morton 已提交
2525

2526 2527 2528
#ifndef CONFIG_MMU
#define randomize_va_space 0
#else
2529
extern int randomize_va_space;
2530
#endif
2531

2532
const char * arch_vma_name(struct vm_area_struct *vma);
2533
void print_vma_addr(char *prefix, unsigned long rip);
2534

2535 2536 2537 2538 2539 2540
void sparse_mem_maps_populate_node(struct page **map_map,
				   unsigned long pnum_begin,
				   unsigned long pnum_end,
				   unsigned long map_count,
				   int nodeid);

2541
struct page *sparse_mem_map_populate(unsigned long pnum, int nid);
2542
pgd_t *vmemmap_pgd_populate(unsigned long addr, int node);
2543 2544
p4d_t *vmemmap_p4d_populate(pgd_t *pgd, unsigned long addr, int node);
pud_t *vmemmap_pud_populate(p4d_t *p4d, unsigned long addr, int node);
2545 2546
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);
2547
void *vmemmap_alloc_block(unsigned long size, int node);
2548 2549 2550 2551 2552 2553 2554 2555
struct vmem_altmap;
void *__vmemmap_alloc_block_buf(unsigned long size, int node,
		struct vmem_altmap *altmap);
static inline void *vmemmap_alloc_block_buf(unsigned long size, int node)
{
	return __vmemmap_alloc_block_buf(size, node, NULL);
}

2556
void vmemmap_verify(pte_t *, int, unsigned long, unsigned long);
2557 2558 2559
int vmemmap_populate_basepages(unsigned long start, unsigned long end,
			       int node);
int vmemmap_populate(unsigned long start, unsigned long end, int node);
2560
void vmemmap_populate_print_last(void);
2561
#ifdef CONFIG_MEMORY_HOTPLUG
2562
void vmemmap_free(unsigned long start, unsigned long end);
2563
#endif
2564
void register_page_bootmem_memmap(unsigned long section_nr, struct page *map,
2565
				  unsigned long nr_pages);
2566

2567 2568
enum mf_flags {
	MF_COUNT_INCREASED = 1 << 0,
2569
	MF_ACTION_REQUIRED = 1 << 1,
2570
	MF_MUST_KILL = 1 << 2,
2571
	MF_SOFT_OFFLINE = 1 << 3,
2572
};
2573
extern int memory_failure(unsigned long pfn, int trapno, int flags);
2574
extern void memory_failure_queue(unsigned long pfn, int trapno, int flags);
W
Wu Fengguang 已提交
2575
extern int unpoison_memory(unsigned long pfn);
2576
extern int get_hwpoison_page(struct page *page);
2577
#define put_hwpoison_page(page)	put_page(page)
2578 2579
extern int sysctl_memory_failure_early_kill;
extern int sysctl_memory_failure_recovery;
2580
extern void shake_page(struct page *p, int access);
2581
extern atomic_long_t num_poisoned_pages;
2582
extern int soft_offline_page(struct page *page, int flags);
2583

2584 2585 2586 2587

/*
 * Error handlers for various types of pages.
 */
2588
enum mf_result {
2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617
	MF_IGNORED,	/* Error: cannot be handled */
	MF_FAILED,	/* Error: handling failed */
	MF_DELAYED,	/* Will be handled later */
	MF_RECOVERED,	/* Successfully recovered */
};

enum mf_action_page_type {
	MF_MSG_KERNEL,
	MF_MSG_KERNEL_HIGH_ORDER,
	MF_MSG_SLAB,
	MF_MSG_DIFFERENT_COMPOUND,
	MF_MSG_POISONED_HUGE,
	MF_MSG_HUGE,
	MF_MSG_FREE_HUGE,
	MF_MSG_UNMAP_FAILED,
	MF_MSG_DIRTY_SWAPCACHE,
	MF_MSG_CLEAN_SWAPCACHE,
	MF_MSG_DIRTY_MLOCKED_LRU,
	MF_MSG_CLEAN_MLOCKED_LRU,
	MF_MSG_DIRTY_UNEVICTABLE_LRU,
	MF_MSG_CLEAN_UNEVICTABLE_LRU,
	MF_MSG_DIRTY_LRU,
	MF_MSG_CLEAN_LRU,
	MF_MSG_TRUNCATED_LRU,
	MF_MSG_BUDDY,
	MF_MSG_BUDDY_2ND,
	MF_MSG_UNKNOWN,
};

A
Andrea Arcangeli 已提交
2618 2619
#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLBFS)
extern void clear_huge_page(struct page *page,
2620
			    unsigned long addr_hint,
A
Andrea Arcangeli 已提交
2621 2622 2623 2624
			    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);
2625 2626
extern long copy_huge_page_from_user(struct page *dst_page,
				const void __user *usr_src,
2627 2628
				unsigned int pages_per_huge_page,
				bool allow_pagefault);
A
Andrea Arcangeli 已提交
2629 2630
#endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLBFS */

2631 2632
extern struct page_ext_operations debug_guardpage_ops;

2633 2634
#ifdef CONFIG_DEBUG_PAGEALLOC
extern unsigned int _debug_guardpage_minorder;
2635
extern bool _debug_guardpage_enabled;
2636 2637 2638 2639 2640 2641

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

2642 2643 2644 2645 2646
static inline bool debug_guardpage_enabled(void)
{
	return _debug_guardpage_enabled;
}

2647 2648
static inline bool page_is_guard(struct page *page)
{
2649 2650 2651 2652 2653 2654
	struct page_ext *page_ext;

	if (!debug_guardpage_enabled())
		return false;

	page_ext = lookup_page_ext(page);
2655 2656 2657
	if (unlikely(!page_ext))
		return false;

2658
	return test_bit(PAGE_EXT_DEBUG_GUARD, &page_ext->flags);
2659 2660 2661
}
#else
static inline unsigned int debug_guardpage_minorder(void) { return 0; }
2662
static inline bool debug_guardpage_enabled(void) { return false; }
2663 2664 2665
static inline bool page_is_guard(struct page *page) { return false; }
#endif /* CONFIG_DEBUG_PAGEALLOC */

2666 2667 2668 2669 2670 2671
#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
2672 2673
#endif /* __KERNEL__ */
#endif /* _LINUX_MM_H */