page-flags.h 34.6 KB
Newer Older
1
/* SPDX-License-Identifier: GPL-2.0 */
L
Linus Torvalds 已提交
2 3 4 5 6 7 8
/*
 * Macros for manipulating and testing page->flags
 */

#ifndef PAGE_FLAGS_H
#define PAGE_FLAGS_H

A
Andrew Morton 已提交
9
#include <linux/types.h>
10
#include <linux/bug.h>
11
#include <linux/mmdebug.h>
12
#ifndef __GENERATING_BOUNDS_H
13
#include <linux/mm_types.h>
14
#include <generated/bounds.h>
15
#endif /* !__GENERATING_BOUNDS_H */
A
Andrew Morton 已提交
16

L
Linus Torvalds 已提交
17 18 19
/*
 * Various page->flags bits:
 *
20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
 * PG_reserved is set for special pages. The "struct page" of such a page
 * should in general not be touched (e.g. set dirty) except by its owner.
 * Pages marked as PG_reserved include:
 * - Pages part of the kernel image (including vDSO) and similar (e.g. BIOS,
 *   initrd, HW tables)
 * - Pages reserved or allocated early during boot (before the page allocator
 *   was initialized). This includes (depending on the architecture) the
 *   initial vmemmap, initial page tables, crashkernel, elfcorehdr, and much
 *   much more. Once (if ever) freed, PG_reserved is cleared and they will
 *   be given to the page allocator.
 * - Pages falling into physical memory gaps - not IORESOURCE_SYSRAM. Trying
 *   to read/write these pages might end badly. Don't touch!
 * - The zero page(s)
 * - Pages not added to the page allocator when onlining a section because
 *   they were excluded via the online_page_callback() or because they are
 *   PG_hwpoison.
 * - Pages allocated in the context of kexec/kdump (loaded kernel image,
 *   control pages, vmcoreinfo)
 * - MMIO/DMA pages. Some architectures don't allow to ioremap pages that are
 *   not marked PG_reserved (as they might be in use by somebody else who does
 *   not respect the caching strategy).
 * - Pages part of an offline section (struct pages of offline sections should
 *   not be trusted as they will be initialized when first onlined).
 * - MCA pages on ia64
 * - Pages holding CPU notes for POWER Firmware Assisted Dump
 * - Device memory (e.g. PMEM, DAX, HMM)
 * Some PG_reserved pages will be excluded from the hibernation image.
 * PG_reserved does in general not hinder anybody from dumping or swapping
 * and is no longer required for remap_pfn_range(). ioremap might require it.
 * Consequently, PG_reserved for a page mapped into user space can indicate
 * the zero page, the vDSO, MMIO pages or device memory.
L
Linus Torvalds 已提交
51
 *
N
Nick Piggin 已提交
52 53 54
 * The PG_private bitflag is set on pagecache pages if they contain filesystem
 * specific data (which is normally at page->private). It can be used by
 * private allocations for its own usage.
L
Linus Torvalds 已提交
55
 *
N
Nick Piggin 已提交
56 57 58 59 60 61 62 63 64
 * During initiation of disk I/O, PG_locked is set. This bit is set before I/O
 * and cleared when writeback _starts_ or when read _completes_. PG_writeback
 * is set before writeback starts and cleared when it finishes.
 *
 * PG_locked also pins a page in pagecache, and blocks truncation of the file
 * while it is held.
 *
 * page_waitqueue(page) is a wait queue of all tasks waiting for the page
 * to become unlocked.
L
Linus Torvalds 已提交
65
 *
H
Huang Ying 已提交
66 67 68 69 70
 * PG_swapbacked is set when a page uses swap as a backing storage.  This are
 * usually PageAnon or shmem pages but please note that even anonymous pages
 * might lose their PG_swapbacked flag when they simply can be dropped (e.g. as
 * a result of MADV_FREE).
 *
N
Nick Piggin 已提交
71 72
 * PG_referenced, PG_reclaim are used for page reclaim for anonymous and
 * file-backed pagecache (see mm/vmscan.c).
L
Linus Torvalds 已提交
73 74 75 76 77 78 79
 *
 * PG_error is set to indicate that an I/O error occurred on this page.
 *
 * PG_arch_1 is an architecture specific page state bit.  The generic code
 * guarantees that this bit is cleared for a page when it first is entered into
 * the page cache.
 *
80 81 82
 * PG_hwpoison indicates that a page got corrupted in hardware and contains
 * data with incorrect ECC bits that triggered a machine check. Accessing is
 * not safe since it may cause another machine check. Don't touch!
L
Linus Torvalds 已提交
83 84 85
 */

/*
86
 * Don't use the pageflags directly.  Use the PageFoo macros.
87 88 89 90 91 92
 *
 * The page flags field is split into two parts, the main flags area
 * which extends from the low bits upwards, and the fields area which
 * extends from the high bits downwards.
 *
 *  | FIELD | ... | FLAGS |
93 94
 *  N-1           ^       0
 *               (NR_PAGEFLAGS)
95
 *
96 97 98
 * The fields area is reserved for fields mapping zone, node (for NUMA) and
 * SPARSEMEM section (for variants of SPARSEMEM that require section ids like
 * SPARSEMEM_EXTREME with !SPARSEMEM_VMEMMAP).
L
Linus Torvalds 已提交
99
 */
100 101 102 103 104 105 106
enum pageflags {
	PG_locked,		/* Page is locked. Don't touch. */
	PG_referenced,
	PG_uptodate,
	PG_dirty,
	PG_lru,
	PG_active,
107
	PG_workingset,
108
	PG_waiters,		/* Page has waiters, check its waitqueue. Must be bit #7 and in the same byte as "PG_locked" */
109
	PG_error,
110 111 112 113 114
	PG_slab,
	PG_owner_priv_1,	/* Owner use. If pagecache, fs may use*/
	PG_arch_1,
	PG_reserved,
	PG_private,		/* If pagecache, has fs-private data */
115
	PG_private_2,		/* If pagecache, has fs aux data */
116
	PG_writeback,		/* Page is under writeback */
117
	PG_head,		/* A head page */
118 119
	PG_mappedtodisk,	/* Has blocks allocated on-disk */
	PG_reclaim,		/* To be reclaimed asap */
R
Rik van Riel 已提交
120
	PG_swapbacked,		/* Page is backed by RAM/swap */
L
Lee Schermerhorn 已提交
121
	PG_unevictable,		/* Page is "unevictable"  */
H
Hugh Dickins 已提交
122
#ifdef CONFIG_MMU
N
Nick Piggin 已提交
123
	PG_mlocked,		/* Page is vma mlocked */
L
Lee Schermerhorn 已提交
124
#endif
125
#ifdef CONFIG_ARCH_USES_PG_UNCACHED
126
	PG_uncached,		/* Page has been mapped as uncached */
127 128 129
#endif
#ifdef CONFIG_MEMORY_FAILURE
	PG_hwpoison,		/* hardware poisoned page. Don't touch */
A
Andrea Arcangeli 已提交
130
#endif
131
#if defined(CONFIG_PAGE_IDLE_FLAG) && defined(CONFIG_64BIT)
132 133
	PG_young,
	PG_idle,
S
Steven Price 已提交
134 135 136
#endif
#ifdef CONFIG_64BIT
	PG_arch_2,
137 138 139
#endif
#ifdef CONFIG_KASAN_HW_TAGS
	PG_skip_kasan_poison,
A
Andrew Morton 已提交
140
#endif
141 142
	__NR_PAGEFLAGS,

143 144
	PG_readahead = PG_reclaim,

145 146 147
	/* Filesystems */
	PG_checked = PG_owner_priv_1,

148 149 150
	/* SwapBacked */
	PG_swapcache = PG_owner_priv_1,	/* Swap page: swp_entry_t in private */

151 152 153 154 155 156
	/* Two page bits are conscripted by FS-Cache to maintain local caching
	 * state.  These bits are set on pages belonging to the netfs's inodes
	 * when those inodes are being locally cached.
	 */
	PG_fscache = PG_private_2,	/* page backed by cache */

157
	/* XEN */
158
	/* Pinned in Xen as a read-only pagetable page. */
159
	PG_pinned = PG_owner_priv_1,
160
	/* Pinned as part of domain save (see xen_mm_pin_all()). */
161
	PG_savepinned = PG_dirty,
162 163
	/* Has a grant mapping of another (foreign) domain's page. */
	PG_foreign = PG_owner_priv_1,
164 165
	/* Remapped by swiotlb-xen. */
	PG_xen_remapped = PG_owner_priv_1,
166

167 168
	/* SLOB */
	PG_slob_free = PG_private,
169 170

	/* Compound pages. Stored in first tail page's flags */
171
	PG_double_map = PG_workingset,
172

173 174 175 176 177 178 179 180 181
#ifdef CONFIG_MEMORY_FAILURE
	/*
	 * Compound pages. Stored in first tail page's flags.
	 * Indicates that at least one subpage is hwpoisoned in the
	 * THP.
	 */
	PG_has_hwpoisoned = PG_mappedtodisk,
#endif

182 183
	/* non-lru isolated movable page */
	PG_isolated = PG_reclaim,
A
Alexander Duyck 已提交
184 185 186

	/* Only valid for buddy pages. Used to track pages that are reported */
	PG_reported = PG_uptodate,
187
};
L
Linus Torvalds 已提交
188

189 190
#define PAGEFLAGS_MASK		((1UL << NR_PAGEFLAGS) - 1)

191 192
#ifndef __GENERATING_BOUNDS_H

193
#ifdef CONFIG_HUGETLB_PAGE_FREE_VMEMMAP
194 195 196 197 198 199 200 201
DECLARE_STATIC_KEY_MAYBE(CONFIG_HUGETLB_PAGE_FREE_VMEMMAP_DEFAULT_ON,
			 hugetlb_free_vmemmap_enabled_key);

static __always_inline bool hugetlb_free_vmemmap_enabled(void)
{
	return static_branch_maybe(CONFIG_HUGETLB_PAGE_FREE_VMEMMAP_DEFAULT_ON,
				   &hugetlb_free_vmemmap_enabled_key);
}
202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220

/*
 * If the feature of freeing some vmemmap pages associated with each HugeTLB
 * page is enabled, the head vmemmap page frame is reused and all of the tail
 * vmemmap addresses map to the head vmemmap page frame (furture details can
 * refer to the figure at the head of the mm/hugetlb_vmemmap.c).  In other
 * words, there are more than one page struct with PG_head associated with each
 * HugeTLB page.  We __know__ that there is only one head page struct, the tail
 * page structs with PG_head are fake head page structs.  We need an approach
 * to distinguish between those two different types of page structs so that
 * compound_head() can return the real head page struct when the parameter is
 * the tail page struct but with PG_head.
 *
 * The page_fixed_fake_head() returns the real head page struct if the @page is
 * fake page head, otherwise, returns @page which can either be a true page
 * head or tail.
 */
static __always_inline const struct page *page_fixed_fake_head(const struct page *page)
{
221
	if (!hugetlb_free_vmemmap_enabled())
222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248
		return page;

	/*
	 * Only addresses aligned with PAGE_SIZE of struct page may be fake head
	 * struct page. The alignment check aims to avoid access the fields (
	 * e.g. compound_head) of the @page[1]. It can avoid touch a (possibly)
	 * cold cacheline in some cases.
	 */
	if (IS_ALIGNED((unsigned long)page, PAGE_SIZE) &&
	    test_bit(PG_head, &page->flags)) {
		/*
		 * We can safely access the field of the @page[1] with PG_head
		 * because the @page is a compound page composed with at least
		 * two contiguous pages.
		 */
		unsigned long head = READ_ONCE(page[1].compound_head);

		if (likely(head & 1))
			return (const struct page *)(head - 1);
	}
	return page;
}
#else
static inline const struct page *page_fixed_fake_head(const struct page *page)
{
	return page;
}
249 250 251 252 253

static inline bool hugetlb_free_vmemmap_enabled(void)
{
	return false;
}
254 255 256 257 258 259 260
#endif

static __always_inline int page_is_fake_head(struct page *page)
{
	return page_fixed_fake_head(page) != page;
}

261
static inline unsigned long _compound_head(const struct page *page)
262 263 264 265
{
	unsigned long head = READ_ONCE(page->compound_head);

	if (unlikely(head & 1))
266
		return head - 1;
267
	return (unsigned long)page_fixed_fake_head(page);
268 269
}

270 271
#define compound_head(page)	((typeof(page))_compound_head(page))

272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299
/**
 * page_folio - Converts from page to folio.
 * @p: The page.
 *
 * Every page is part of a folio.  This function cannot be called on a
 * NULL pointer.
 *
 * Context: No reference, nor lock is required on @page.  If the caller
 * does not hold a reference, this call may race with a folio split, so
 * it should re-check the folio still contains this page after gaining
 * a reference on the folio.
 * Return: The folio which contains this page.
 */
#define page_folio(p)		(_Generic((p),				\
	const struct page *:	(const struct folio *)_compound_head(p), \
	struct page *:		(struct folio *)_compound_head(p)))

/**
 * folio_page - Return a page from a folio.
 * @folio: The folio.
 * @n: The page number to return.
 *
 * @n is relative to the start of the folio.  This function does not
 * check that the page number lies within @folio; the caller is presumed
 * to have a reference to the page.
 */
#define folio_page(folio, n)	nth_page(&(folio)->page, n)

300
static __always_inline int PageTail(struct page *page)
301
{
302
	return READ_ONCE(page->compound_head) & 1 || page_is_fake_head(page);
303 304
}

305
static __always_inline int PageCompound(struct page *page)
306
{
307 308
	return test_bit(PG_head, &page->flags) ||
	       READ_ONCE(page->compound_head) & 1;
309 310
}

311 312 313
#define	PAGE_POISON_PATTERN	-1l
static inline int PagePoisoned(const struct page *page)
{
M
Marco Elver 已提交
314
	return READ_ONCE(page->flags) == PAGE_POISON_PATTERN;
315 316
}

317 318 319 320 321 322 323 324
#ifdef CONFIG_DEBUG_VM
void page_init_poison(struct page *page, size_t size);
#else
static inline void page_init_poison(struct page *page, size_t size)
{
}
#endif

325 326 327 328 329 330 331 332 333
static unsigned long *folio_flags(struct folio *folio, unsigned n)
{
	struct page *page = &folio->page;

	VM_BUG_ON_PGFLAGS(PageTail(page), page);
	VM_BUG_ON_PGFLAGS(n > 0 && !test_bit(PG_head, &page->flags), page);
	return &page[n].flags;
}

334 335 336
/*
 * Page flags policies wrt compound pages
 *
337 338 339
 * PF_POISONED_CHECK
 *     check if this struct page poisoned/uninitialized
 *
340 341 342 343 344 345 346
 * PF_ANY:
 *     the page flag is relevant for small, head and tail pages.
 *
 * PF_HEAD:
 *     for compound page all operations related to the page flag applied to
 *     head page.
 *
347 348 349
 * PF_ONLY_HEAD:
 *     for compound page, callers only ever operate on the head page.
 *
350 351 352 353 354 355
 * PF_NO_TAIL:
 *     modifications of the page flag must be done on small or head pages,
 *     checks can be done on tail pages too.
 *
 * PF_NO_COMPOUND:
 *     the page flag is not relevant for compound pages.
356 357 358
 *
 * PF_SECOND:
 *     the page flag is stored in the first tail page.
359
 */
360 361 362 363 364
#define PF_POISONED_CHECK(page) ({					\
		VM_BUG_ON_PGFLAGS(PagePoisoned(page), page);		\
		page; })
#define PF_ANY(page, enforce)	PF_POISONED_CHECK(page)
#define PF_HEAD(page, enforce)	PF_POISONED_CHECK(compound_head(page))
365 366
#define PF_ONLY_HEAD(page, enforce) ({					\
		VM_BUG_ON_PGFLAGS(PageTail(page), page);		\
367
		PF_POISONED_CHECK(page); })
368 369
#define PF_NO_TAIL(page, enforce) ({					\
		VM_BUG_ON_PGFLAGS(enforce && PageTail(page), page);	\
370
		PF_POISONED_CHECK(compound_head(page)); })
371
#define PF_NO_COMPOUND(page, enforce) ({				\
372
		VM_BUG_ON_PGFLAGS(enforce && PageCompound(page), page);	\
373
		PF_POISONED_CHECK(page); })
374 375 376
#define PF_SECOND(page, enforce) ({					\
		VM_BUG_ON_PGFLAGS(!PageHead(page), page);		\
		PF_POISONED_CHECK(&page[1]); })
377

378 379 380 381 382 383 384 385
/* Which page is the flag stored in */
#define FOLIO_PF_ANY		0
#define FOLIO_PF_HEAD		0
#define FOLIO_PF_ONLY_HEAD	0
#define FOLIO_PF_NO_TAIL	0
#define FOLIO_PF_NO_COMPOUND	0
#define FOLIO_PF_SECOND		1

386 387 388
/*
 * Macros to create function definitions for page flags
 */
389
#define TESTPAGEFLAG(uname, lname, policy)				\
390 391
static __always_inline bool folio_test_##lname(struct folio *folio)	\
{ return test_bit(PG_##lname, folio_flags(folio, FOLIO_##policy)); }	\
392
static __always_inline int Page##uname(struct page *page)		\
393
{ return test_bit(PG_##lname, &policy(page, 0)->flags); }
394

395
#define SETPAGEFLAG(uname, lname, policy)				\
396 397 398
static __always_inline							\
void folio_set_##lname(struct folio *folio)				\
{ set_bit(PG_##lname, folio_flags(folio, FOLIO_##policy)); }		\
399
static __always_inline void SetPage##uname(struct page *page)		\
400
{ set_bit(PG_##lname, &policy(page, 1)->flags); }
401

402
#define CLEARPAGEFLAG(uname, lname, policy)				\
403 404 405
static __always_inline							\
void folio_clear_##lname(struct folio *folio)				\
{ clear_bit(PG_##lname, folio_flags(folio, FOLIO_##policy)); }		\
406
static __always_inline void ClearPage##uname(struct page *page)		\
407
{ clear_bit(PG_##lname, &policy(page, 1)->flags); }
408

409
#define __SETPAGEFLAG(uname, lname, policy)				\
410 411 412
static __always_inline							\
void __folio_set_##lname(struct folio *folio)				\
{ __set_bit(PG_##lname, folio_flags(folio, FOLIO_##policy)); }		\
413
static __always_inline void __SetPage##uname(struct page *page)		\
414
{ __set_bit(PG_##lname, &policy(page, 1)->flags); }
415

416
#define __CLEARPAGEFLAG(uname, lname, policy)				\
417 418 419
static __always_inline							\
void __folio_clear_##lname(struct folio *folio)				\
{ __clear_bit(PG_##lname, folio_flags(folio, FOLIO_##policy)); }	\
420
static __always_inline void __ClearPage##uname(struct page *page)	\
421
{ __clear_bit(PG_##lname, &policy(page, 1)->flags); }
422

423
#define TESTSETFLAG(uname, lname, policy)				\
424 425 426
static __always_inline							\
bool folio_test_set_##lname(struct folio *folio)			\
{ return test_and_set_bit(PG_##lname, folio_flags(folio, FOLIO_##policy)); } \
427
static __always_inline int TestSetPage##uname(struct page *page)	\
428
{ return test_and_set_bit(PG_##lname, &policy(page, 1)->flags); }
429

430
#define TESTCLEARFLAG(uname, lname, policy)				\
431 432 433
static __always_inline							\
bool folio_test_clear_##lname(struct folio *folio)			\
{ return test_and_clear_bit(PG_##lname, folio_flags(folio, FOLIO_##policy)); } \
434
static __always_inline int TestClearPage##uname(struct page *page)	\
435
{ return test_and_clear_bit(PG_##lname, &policy(page, 1)->flags); }
436

437 438 439 440
#define PAGEFLAG(uname, lname, policy)					\
	TESTPAGEFLAG(uname, lname, policy)				\
	SETPAGEFLAG(uname, lname, policy)				\
	CLEARPAGEFLAG(uname, lname, policy)
441

442 443 444 445
#define __PAGEFLAG(uname, lname, policy)				\
	TESTPAGEFLAG(uname, lname, policy)				\
	__SETPAGEFLAG(uname, lname, policy)				\
	__CLEARPAGEFLAG(uname, lname, policy)
446

447 448 449
#define TESTSCFLAG(uname, lname, policy)				\
	TESTSETFLAG(uname, lname, policy)				\
	TESTCLEARFLAG(uname, lname, policy)
450

451
#define TESTPAGEFLAG_FALSE(uname, lname)				\
452
static inline bool folio_test_##lname(const struct folio *folio) { return false; } \
453 454
static inline int Page##uname(const struct page *page) { return 0; }

455 456
#define SETPAGEFLAG_NOOP(uname, lname)					\
static inline void folio_set_##lname(struct folio *folio) { }		\
457 458
static inline void SetPage##uname(struct page *page) {  }

459 460
#define CLEARPAGEFLAG_NOOP(uname, lname)				\
static inline void folio_clear_##lname(struct folio *folio) { }		\
461 462
static inline void ClearPage##uname(struct page *page) {  }

463 464
#define __CLEARPAGEFLAG_NOOP(uname, lname)				\
static inline void __folio_clear_##lname(struct folio *folio) { }	\
465 466
static inline void __ClearPage##uname(struct page *page) {  }

467 468 469
#define TESTSETFLAG_FALSE(uname, lname)					\
static inline bool folio_test_set_##lname(struct folio *folio)		\
{ return 0; }								\
470 471
static inline int TestSetPage##uname(struct page *page) { return 0; }

472 473 474
#define TESTCLEARFLAG_FALSE(uname, lname)				\
static inline bool folio_test_clear_##lname(struct folio *folio)	\
{ return 0; }								\
475 476
static inline int TestClearPage##uname(struct page *page) { return 0; }

477 478
#define PAGEFLAG_FALSE(uname, lname) TESTPAGEFLAG_FALSE(uname, lname)	\
	SETPAGEFLAG_NOOP(uname, lname) CLEARPAGEFLAG_NOOP(uname, lname)
479

480 481
#define TESTSCFLAG_FALSE(uname, lname)					\
	TESTSETFLAG_FALSE(uname, lname) TESTCLEARFLAG_FALSE(uname, lname)
482

483
__PAGEFLAG(Locked, locked, PF_NO_TAIL)
484
PAGEFLAG(Waiters, waiters, PF_ONLY_HEAD) __CLEARPAGEFLAG(Waiters, waiters, PF_ONLY_HEAD)
485
PAGEFLAG(Error, error, PF_NO_TAIL) TESTCLEARFLAG(Error, error, PF_NO_TAIL)
486 487 488
PAGEFLAG(Referenced, referenced, PF_HEAD)
	TESTCLEARFLAG(Referenced, referenced, PF_HEAD)
	__SETPAGEFLAG(Referenced, referenced, PF_HEAD)
489 490
PAGEFLAG(Dirty, dirty, PF_HEAD) TESTSCFLAG(Dirty, dirty, PF_HEAD)
	__CLEARPAGEFLAG(Dirty, dirty, PF_HEAD)
491
PAGEFLAG(LRU, lru, PF_HEAD) __CLEARPAGEFLAG(LRU, lru, PF_HEAD)
A
Alex Shi 已提交
492
	TESTCLEARFLAG(LRU, lru, PF_HEAD)
493 494
PAGEFLAG(Active, active, PF_HEAD) __CLEARPAGEFLAG(Active, active, PF_HEAD)
	TESTCLEARFLAG(Active, active, PF_HEAD)
495 496
PAGEFLAG(Workingset, workingset, PF_HEAD)
	TESTCLEARFLAG(Workingset, workingset, PF_HEAD)
497 498
__PAGEFLAG(Slab, slab, PF_NO_TAIL)
__PAGEFLAG(SlobFree, slob_free, PF_NO_TAIL)
499
PAGEFLAG(Checked, checked, PF_NO_COMPOUND)	   /* Used by some filesystems */
500 501 502 503 504 505

/* Xen */
PAGEFLAG(Pinned, pinned, PF_NO_COMPOUND)
	TESTSCFLAG(Pinned, pinned, PF_NO_COMPOUND)
PAGEFLAG(SavePinned, savepinned, PF_NO_COMPOUND);
PAGEFLAG(Foreign, foreign, PF_NO_COMPOUND);
506 507
PAGEFLAG(XenRemapped, xen_remapped, PF_NO_COMPOUND)
	TESTCLEARFLAG(XenRemapped, xen_remapped, PF_NO_COMPOUND)
508

509 510
PAGEFLAG(Reserved, reserved, PF_NO_COMPOUND)
	__CLEARPAGEFLAG(Reserved, reserved, PF_NO_COMPOUND)
511
	__SETPAGEFLAG(Reserved, reserved, PF_NO_COMPOUND)
512 513 514
PAGEFLAG(SwapBacked, swapbacked, PF_NO_TAIL)
	__CLEARPAGEFLAG(SwapBacked, swapbacked, PF_NO_TAIL)
	__SETPAGEFLAG(SwapBacked, swapbacked, PF_NO_TAIL)
515

516 517 518 519 520
/*
 * Private page markings that may be used by the filesystem that owns the page
 * for its own purposes.
 * - PG_private and PG_private_2 cause releasepage() and co to be invoked
 */
521
PAGEFLAG(Private, private, PF_ANY)
522 523 524
PAGEFLAG(Private2, private_2, PF_ANY) TESTSCFLAG(Private2, private_2, PF_ANY)
PAGEFLAG(OwnerPriv1, owner_priv_1, PF_ANY)
	TESTCLEARFLAG(OwnerPriv1, owner_priv_1, PF_ANY)
525

526 527 528 529
/*
 * Only test-and-set exist for PG_writeback.  The unconditional operators are
 * risky: they bypass page accounting.
 */
530 531
TESTPAGEFLAG(Writeback, writeback, PF_NO_TAIL)
	TESTSCFLAG(Writeback, writeback, PF_NO_TAIL)
532
PAGEFLAG(MappedToDisk, mappedtodisk, PF_NO_TAIL)
533

534
/* PG_readahead is only used for reads; PG_reclaim is only for writes */
535 536
PAGEFLAG(Reclaim, reclaim, PF_NO_TAIL)
	TESTCLEARFLAG(Reclaim, reclaim, PF_NO_TAIL)
537 538
PAGEFLAG(Readahead, readahead, PF_NO_COMPOUND)
	TESTCLEARFLAG(Readahead, readahead, PF_NO_COMPOUND)
539 540

#ifdef CONFIG_HIGHMEM
L
Linus Torvalds 已提交
541
/*
542 543
 * Must use a macro here due to header dependency issues. page_zone() is not
 * available at this point.
L
Linus Torvalds 已提交
544
 */
V
Vineet Gupta 已提交
545
#define PageHighMem(__p) is_highmem_idx(page_zonenum(__p))
546
#else
547
PAGEFLAG_FALSE(HighMem, highmem)
548 549 550
#endif

#ifdef CONFIG_SWAP
551
static __always_inline bool folio_test_swapcache(struct folio *folio)
552
{
553 554 555
	return folio_test_swapbacked(folio) &&
			test_bit(PG_swapcache, folio_flags(folio, 0));
}
556

557 558 559
static __always_inline bool PageSwapCache(struct page *page)
{
	return folio_test_swapcache(page_folio(page));
560
}
561

562 563
SETPAGEFLAG(SwapCache, swapcache, PF_NO_TAIL)
CLEARPAGEFLAG(SwapCache, swapcache, PF_NO_TAIL)
564
#else
565
PAGEFLAG_FALSE(SwapCache, swapcache)
566 567
#endif

568 569 570
PAGEFLAG(Unevictable, unevictable, PF_HEAD)
	__CLEARPAGEFLAG(Unevictable, unevictable, PF_HEAD)
	TESTCLEARFLAG(Unevictable, unevictable, PF_HEAD)
N
Nick Piggin 已提交
571

H
Hugh Dickins 已提交
572
#ifdef CONFIG_MMU
573 574 575
PAGEFLAG(Mlocked, mlocked, PF_NO_TAIL)
	__CLEARPAGEFLAG(Mlocked, mlocked, PF_NO_TAIL)
	TESTSCFLAG(Mlocked, mlocked, PF_NO_TAIL)
L
Lee Schermerhorn 已提交
576
#else
577 578
PAGEFLAG_FALSE(Mlocked, mlocked) __CLEARPAGEFLAG_NOOP(Mlocked, mlocked)
	TESTSCFLAG_FALSE(Mlocked, mlocked)
L
Lee Schermerhorn 已提交
579 580
#endif

581
#ifdef CONFIG_ARCH_USES_PG_UNCACHED
582
PAGEFLAG(Uncached, uncached, PF_NO_COMPOUND)
583
#else
584
PAGEFLAG_FALSE(Uncached, uncached)
585
#endif
L
Linus Torvalds 已提交
586

587
#ifdef CONFIG_MEMORY_FAILURE
588 589
PAGEFLAG(HWPoison, hwpoison, PF_ANY)
TESTSCFLAG(HWPoison, hwpoison, PF_ANY)
590
#define __PG_HWPOISON (1UL << PG_hwpoison)
591 592 593
#define MAGIC_HWPOISON	0x48575053U	/* HWPS */
extern void SetPageHWPoisonTakenOff(struct page *page);
extern void ClearPageHWPoisonTakenOff(struct page *page);
594
extern bool take_page_off_buddy(struct page *page);
595
extern bool put_page_back_buddy(struct page *page);
596
#else
597
PAGEFLAG_FALSE(HWPoison, hwpoison)
598 599 600
#define __PG_HWPOISON 0
#endif

601
#if defined(CONFIG_PAGE_IDLE_FLAG) && defined(CONFIG_64BIT)
602 603 604 605
TESTPAGEFLAG(Young, young, PF_ANY)
SETPAGEFLAG(Young, young, PF_ANY)
TESTCLEARFLAG(Young, young, PF_ANY)
PAGEFLAG(Idle, idle, PF_ANY)
606 607
#endif

608 609 610
#ifdef CONFIG_KASAN_HW_TAGS
PAGEFLAG(SkipKASanPoison, skip_kasan_poison, PF_HEAD)
#else
611
PAGEFLAG_FALSE(SkipKASanPoison, skip_kasan_poison)
612 613
#endif

A
Alexander Duyck 已提交
614 615 616 617 618 619 620 621
/*
 * PageReported() is used to track reported free pages within the Buddy
 * allocator. We can use the non-atomic version of the test and set
 * operations as both should be shielded with the zone lock to prevent
 * any possible races on the setting or clearing of the bit.
 */
__PAGEFLAG(Reported, reported, PF_NO_COMPOUND)

622 623 624 625 626 627
/*
 * On an anonymous page mapped into a user virtual memory area,
 * page->mapping points to its anon_vma, not to a struct address_space;
 * with the PAGE_MAPPING_ANON bit set to distinguish it.  See rmap.h.
 *
 * On an anonymous page in a VM_MERGEABLE area, if CONFIG_KSM is enabled,
628 629
 * the PAGE_MAPPING_MOVABLE bit may be set along with the PAGE_MAPPING_ANON
 * bit; and then page->mapping points, not to an anon_vma, but to a private
630 631
 * structure which KSM associates with that merged page.  See ksm.h.
 *
632 633
 * PAGE_MAPPING_KSM without PAGE_MAPPING_ANON is used for non-lru movable
 * page and then page->mapping points a struct address_space.
634 635 636 637 638
 *
 * Please note that, confusingly, "page_mapping" refers to the inode
 * address_space which maps the page from disk; whereas "page_mapped"
 * refers to user virtual address space into which the page is mapped.
 */
639 640 641 642
#define PAGE_MAPPING_ANON	0x1
#define PAGE_MAPPING_MOVABLE	0x2
#define PAGE_MAPPING_KSM	(PAGE_MAPPING_ANON | PAGE_MAPPING_MOVABLE)
#define PAGE_MAPPING_FLAGS	(PAGE_MAPPING_ANON | PAGE_MAPPING_MOVABLE)
643

644
static __always_inline int PageMappingFlags(struct page *page)
645
{
646
	return ((unsigned long)page->mapping & PAGE_MAPPING_FLAGS) != 0;
647 648
}

649
static __always_inline bool folio_test_anon(struct folio *folio)
650
{
651 652 653 654
	return ((unsigned long)folio->mapping & PAGE_MAPPING_ANON) != 0;
}

static __always_inline bool PageAnon(struct page *page)
655
{
656
	return folio_test_anon(page_folio(page));
657 658 659 660 661 662
}

static __always_inline int __PageMovable(struct page *page)
{
	return ((unsigned long)page->mapping & PAGE_MAPPING_FLAGS) ==
				PAGE_MAPPING_MOVABLE;
663 664 665 666 667 668 669 670 671
}

#ifdef CONFIG_KSM
/*
 * A KSM page is one of those write-protected "shared pages" or "merged pages"
 * which KSM maps into multiple mms, wherever identical anonymous page content
 * is found in VM_MERGEABLE vmas.  It's a PageAnon page, pointing not to any
 * anon_vma, but to that page's node of the stable tree.
 */
672
static __always_inline bool folio_test_ksm(struct folio *folio)
673
{
674
	return ((unsigned long)folio->mapping & PAGE_MAPPING_FLAGS) ==
675
				PAGE_MAPPING_KSM;
676
}
677 678 679 680 681

static __always_inline bool PageKsm(struct page *page)
{
	return folio_test_ksm(page_folio(page));
}
682
#else
683
TESTPAGEFLAG_FALSE(Ksm, ksm)
684 685
#endif

W
Wu Fengguang 已提交
686 687
u64 stable_page_flags(struct page *page);

688 689 690 691 692 693 694 695 696 697
/**
 * folio_test_uptodate - Is this folio up to date?
 * @folio: The folio.
 *
 * The uptodate flag is set on a folio when every byte in the folio is
 * at least as new as the corresponding bytes on storage.  Anonymous
 * and CoW folios are always uptodate.  If the folio is not uptodate,
 * some of the bytes in it may be; see the is_partially_uptodate()
 * address_space operation.
 */
698
static inline bool folio_test_uptodate(struct folio *folio)
N
Nick Piggin 已提交
699
{
700
	bool ret = test_bit(PG_uptodate, folio_flags(folio, 0));
N
Nick Piggin 已提交
701
	/*
702 703 704
	 * Must ensure that the data we read out of the folio is loaded
	 * _after_ we've loaded folio->flags to check the uptodate bit.
	 * We can skip the barrier if the folio is not uptodate, because
N
Nick Piggin 已提交
705 706
	 * we wouldn't be reading anything from it.
	 *
707
	 * See folio_mark_uptodate() for the other side of the story.
N
Nick Piggin 已提交
708 709 710 711 712 713 714
	 */
	if (ret)
		smp_rmb();

	return ret;
}

715 716 717 718 719 720
static inline int PageUptodate(struct page *page)
{
	return folio_test_uptodate(page_folio(page));
}

static __always_inline void __folio_mark_uptodate(struct folio *folio)
N
Nick Piggin 已提交
721 722
{
	smp_wmb();
723
	__set_bit(PG_uptodate, folio_flags(folio, 0));
N
Nick Piggin 已提交
724 725
}

726
static __always_inline void folio_mark_uptodate(struct folio *folio)
727
{
N
Nick Piggin 已提交
728 729
	/*
	 * Memory barrier must be issued before setting the PG_uptodate bit,
730 731
	 * so that all previous stores issued in order to bring the folio
	 * uptodate are actually visible before folio_test_uptodate becomes true.
N
Nick Piggin 已提交
732 733
	 */
	smp_wmb();
734 735 736 737 738 739 740 741 742 743 744
	set_bit(PG_uptodate, folio_flags(folio, 0));
}

static __always_inline void __SetPageUptodate(struct page *page)
{
	__folio_mark_uptodate((struct folio *)page);
}

static __always_inline void SetPageUptodate(struct page *page)
{
	folio_mark_uptodate((struct folio *)page);
N
Nick Piggin 已提交
745 746
}

747
CLEARPAGEFLAG(Uptodate, uptodate, PF_NO_TAIL)
L
Linus Torvalds 已提交
748

749 750
bool __folio_start_writeback(struct folio *folio, bool keep_write);
bool set_page_writeback(struct page *page);
751

752 753 754 755
#define folio_start_writeback(folio)			\
	__folio_start_writeback(folio, false)
#define folio_start_writeback_keepwrite(folio)	\
	__folio_start_writeback(folio, true)
L
Linus Torvalds 已提交
756

757
static inline void set_page_writeback_keepwrite(struct page *page)
758
{
759
	folio_start_writeback_keepwrite(page_folio(page));
760
}
L
Linus Torvalds 已提交
761

762
static inline bool test_set_page_writeback(struct page *page)
763
{
764
	return set_page_writeback(page);
765 766
}

767 768 769 770 771 772 773 774 775 776 777 778 779 780
static __always_inline bool folio_test_head(struct folio *folio)
{
	return test_bit(PG_head, folio_flags(folio, FOLIO_PF_ANY));
}

static __always_inline int PageHead(struct page *page)
{
	PF_POISONED_CHECK(page);
	return test_bit(PG_head, &page->flags) && !page_is_fake_head(page);
}

__SETPAGEFLAG(Head, head, PF_ANY)
__CLEARPAGEFLAG(Head, head, PF_ANY)
CLEARPAGEFLAG(Head, head, PF_ANY)
781

782 783 784 785 786 787 788
/**
 * folio_test_large() - Does this folio contain more than one page?
 * @folio: The folio to test.
 *
 * Return: True if the folio is larger than one page.
 */
static inline bool folio_test_large(struct folio *folio)
789 790 791 792
{
	return folio_test_head(folio);
}

793
static __always_inline void set_compound_head(struct page *page, struct page *head)
794
{
795
	WRITE_ONCE(page->compound_head, (unsigned long)head + 1);
796 797
}

798
static __always_inline void clear_compound_head(struct page *page)
799
{
800
	WRITE_ONCE(page->compound_head, 0);
801
}
802

A
Andrea Arcangeli 已提交
803 804 805
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
static inline void ClearPageCompound(struct page *page)
{
806 807
	BUG_ON(!PageHead(page));
	ClearPageHead(page);
A
Andrea Arcangeli 已提交
808 809 810
}
#endif

811
#define PG_head_mask ((1UL << PG_head))
812

813 814 815
#ifdef CONFIG_HUGETLB_PAGE
int PageHuge(struct page *page);
int PageHeadHuge(struct page *page);
816 817 818 819
static inline bool folio_test_hugetlb(struct folio *folio)
{
	return PageHeadHuge(&folio->page);
}
820
#else
821 822
TESTPAGEFLAG_FALSE(Huge, hugetlb)
TESTPAGEFLAG_FALSE(HeadHuge, headhuge)
823 824
#endif

825
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
826 827 828 829 830 831 832 833 834 835
/*
 * PageHuge() only returns true for hugetlbfs pages, but not for
 * normal or transparent huge pages.
 *
 * PageTransHuge() returns true for both transparent huge and
 * hugetlbfs pages, but not normal pages. PageTransHuge() can only be
 * called only in the core VM paths where hugetlbfs pages can't exist.
 */
static inline int PageTransHuge(struct page *page)
{
836
	VM_BUG_ON_PAGE(PageTail(page), page);
837 838 839
	return PageHead(page);
}

840 841 842 843 844
static inline bool folio_test_transhuge(struct folio *folio)
{
	return folio_test_head(folio);
}

845 846 847 848 849
/*
 * PageTransCompound returns true for both transparent huge pages
 * and hugetlbfs pages, so it should only be called when it's known
 * that hugetlbfs pages aren't involved.
 */
850 851 852 853
static inline int PageTransCompound(struct page *page)
{
	return PageCompound(page);
}
854

855 856 857 858 859 860 861 862 863 864
/*
 * PageTransTail returns true for both transparent huge pages
 * and hugetlbfs pages, so it should only be called when it's known
 * that hugetlbfs pages aren't involved.
 */
static inline int PageTransTail(struct page *page)
{
	return PageTail(page);
}

865 866 867 868 869 870 871 872 873 874 875 876 877
/*
 * PageDoubleMap indicates that the compound page is mapped with PTEs as well
 * as PMDs.
 *
 * This is required for optimization of rmap operations for THP: we can postpone
 * per small page mapcount accounting (and its overhead from atomic operations)
 * until the first PMD split.
 *
 * For the page PageDoubleMap means ->_mapcount in all sub-pages is offset up
 * by one. This reference will go away with last compound_mapcount.
 *
 * See also __split_huge_pmd_locked() and page_remove_anon_compound_rmap().
 */
878 879
PAGEFLAG(DoubleMap, double_map, PF_SECOND)
	TESTSCFLAG(DoubleMap, double_map, PF_SECOND)
880
#else
881 882 883 884 885 886
TESTPAGEFLAG_FALSE(TransHuge, transhuge)
TESTPAGEFLAG_FALSE(TransCompound, transcompound)
TESTPAGEFLAG_FALSE(TransCompoundMap, transcompoundmap)
TESTPAGEFLAG_FALSE(TransTail, transtail)
PAGEFLAG_FALSE(DoubleMap, double_map)
	TESTSCFLAG_FALSE(DoubleMap, double_map)
887 888
#endif

889 890 891 892 893 894 895 896 897 898
#if defined(CONFIG_MEMORY_FAILURE) && defined(CONFIG_TRANSPARENT_HUGEPAGE)
/*
 * PageHasHWPoisoned indicates that at least one subpage is hwpoisoned in the
 * compound page.
 *
 * This flag is set by hwpoison handler.  Cleared by THP split or free page.
 */
PAGEFLAG(HasHWPoisoned, has_hwpoisoned, PF_SECOND)
	TESTSCFLAG(HasHWPoisoned, has_hwpoisoned, PF_SECOND)
#else
899 900
PAGEFLAG_FALSE(HasHWPoisoned, has_hwpoisoned)
	TESTSCFLAG_FALSE(HasHWPoisoned, has_hwpoisoned)
901 902
#endif

903 904 905 906 907 908 909 910 911 912 913 914
/*
 * Check if a page is currently marked HWPoisoned. Note that this check is
 * best effort only and inherently racy: there is no way to synchronize with
 * failing hardware.
 */
static inline bool is_page_hwpoison(struct page *page)
{
	if (PageHWPoison(page))
		return true;
	return PageHuge(page) && PageHWPoison(compound_head(page));
}

915
/*
916 917 918 919 920 921
 * For pages that are never mapped to userspace (and aren't PageSlab),
 * page_type may be used.  Because it is initialised to -1, we invert the
 * sense of the bit, so __SetPageFoo *clears* the bit used for PageFoo, and
 * __ClearPageFoo *sets* the bit used for PageFoo.  We reserve a few high and
 * low bits so that an underflow or overflow of page_mapcount() won't be
 * mistaken for a page type value.
922
 */
923 924 925

#define PAGE_TYPE_BASE	0xf0000000
/* Reserve		0x0000007f to catch underflows of page_mapcount */
926
#define PAGE_MAPCOUNT_RESERVE	-128
927
#define PG_buddy	0x00000080
928
#define PG_offline	0x00000100
929 930
#define PG_table	0x00000200
#define PG_guard	0x00000400
931 932 933 934

#define PageType(page, flag)						\
	((page->page_type & (PAGE_TYPE_BASE | flag)) == PAGE_TYPE_BASE)

935 936 937 938 939
static inline int page_has_type(struct page *page)
{
	return (int)page->page_type < PAGE_MAPCOUNT_RESERVE;
}

940
#define PAGE_TYPE_OPS(uname, lname)					\
941 942
static __always_inline int Page##uname(struct page *page)		\
{									\
943
	return PageType(page, PG_##lname);				\
944 945 946
}									\
static __always_inline void __SetPage##uname(struct page *page)		\
{									\
947 948
	VM_BUG_ON_PAGE(!PageType(page, 0), page);			\
	page->page_type &= ~PG_##lname;					\
949 950 951 952
}									\
static __always_inline void __ClearPage##uname(struct page *page)	\
{									\
	VM_BUG_ON_PAGE(!Page##uname(page), page);			\
953
	page->page_type |= PG_##lname;					\
954 955
}

956
/*
957
 * PageBuddy() indicates that the page is free and in the buddy system
958 959
 * (see mm/page_alloc.c).
 */
960
PAGE_TYPE_OPS(Buddy, buddy)
961

962
/*
963 964 965 966 967
 * PageOffline() indicates that the page is logically offline although the
 * containing section is online. (e.g. inflated in a balloon driver or
 * not onlined when onlining the section).
 * The content of these pages is effectively stale. Such pages should not
 * be touched (read/write/dump/save) except by their owner.
968 969 970 971 972 973 974 975 976 977
 *
 * If a driver wants to allow to offline unmovable PageOffline() pages without
 * putting them back to the buddy, it can do so via the memory notifier by
 * decrementing the reference count in MEM_GOING_OFFLINE and incrementing the
 * reference count in MEM_CANCEL_OFFLINE. When offlining, the PageOffline()
 * pages (now with a reference count of zero) are treated like free pages,
 * allowing the containing memory block to get offlined. A driver that
 * relies on this feature is aware that re-onlining the memory block will
 * require to re-set the pages PageOffline() and not giving them to the
 * buddy via online_page_callback_t.
978 979 980 981 982
 *
 * There are drivers that mark a page PageOffline() and expect there won't be
 * any further access to page content. PFN walkers that read content of random
 * pages should check PageOffline() and synchronize with such drivers using
 * page_offline_freeze()/page_offline_thaw().
983
 */
984
PAGE_TYPE_OPS(Offline, offline)
985

986 987 988 989 990
extern void page_offline_freeze(void);
extern void page_offline_thaw(void);
extern void page_offline_begin(void);
extern void page_offline_end(void);

991 992 993 994 995
/*
 * Marks pages in use as page tables.
 */
PAGE_TYPE_OPS(Table, table)

996 997 998 999 1000
/*
 * Marks guardpages used with debug_pagealloc.
 */
PAGE_TYPE_OPS(Guard, guard)

1001 1002
extern bool is_free_buddy_page(struct page *page);

1003
PAGEFLAG(Isolated, isolated, PF_ANY);
1004

H
Hugh Dickins 已提交
1005
#ifdef CONFIG_MMU
1006
#define __PG_MLOCKED		(1UL << PG_mlocked)
1007
#else
N
Nick Piggin 已提交
1008
#define __PG_MLOCKED		0
L
Lee Schermerhorn 已提交
1009 1010
#endif

1011 1012
/*
 * Flags checked when a page is freed.  Pages being freed should not have
G
Guo Ren 已提交
1013
 * these flags set.  If they are, there is a problem.
1014
 */
1015 1016 1017 1018 1019 1020
#define PAGE_FLAGS_CHECK_AT_FREE				\
	(1UL << PG_lru		| 1UL << PG_locked	|	\
	 1UL << PG_private	| 1UL << PG_private_2	|	\
	 1UL << PG_writeback	| 1UL << PG_reserved	|	\
	 1UL << PG_slab		| 1UL << PG_active 	|	\
	 1UL << PG_unevictable	| __PG_MLOCKED)
1021 1022 1023

/*
 * Flags checked when a page is prepped for return by the page allocator.
G
Guo Ren 已提交
1024
 * Pages being prepped should not have these flags set.  If they are set,
1025
 * there has been a kernel bug or struct page corruption.
1026 1027 1028
 *
 * __PG_HWPOISON is exceptional because it needs to be kept beyond page's
 * alloc-free cycle to prevent from reusing the page.
1029
 */
1030
#define PAGE_FLAGS_CHECK_AT_PREP	\
1031
	(PAGEFLAGS_MASK & ~__PG_HWPOISON)
1032

1033
#define PAGE_FLAGS_PRIVATE				\
1034
	(1UL << PG_private | 1UL << PG_private_2)
1035 1036 1037 1038 1039 1040 1041
/**
 * page_has_private - Determine if page has private stuff
 * @page: The page to be checked
 *
 * Determine if a page has private stuff, indicating that release routines
 * should be invoked upon it.
 */
1042 1043 1044 1045 1046
static inline int page_has_private(struct page *page)
{
	return !!(page->flags & PAGE_FLAGS_PRIVATE);
}

1047 1048 1049 1050 1051
static inline bool folio_has_private(struct folio *folio)
{
	return page_has_private(&folio->page);
}

1052 1053
#undef PF_ANY
#undef PF_HEAD
1054
#undef PF_ONLY_HEAD
1055 1056
#undef PF_NO_TAIL
#undef PF_NO_COMPOUND
1057
#undef PF_SECOND
1058
#endif /* !__GENERATING_BOUNDS_H */
1059

L
Linus Torvalds 已提交
1060
#endif	/* PAGE_FLAGS_H */