pgtable.h 21.7 KB
Newer Older
H
H. Peter Anvin 已提交
1 2
#ifndef _ASM_X86_PGTABLE_H
#define _ASM_X86_PGTABLE_H
3

4
#include <asm/page.h>
5
#include <asm/e820.h>
6

7
#include <asm/pgtable_types.h>
8

9 10 11
/*
 * Macro to mark a page protection value as UC-
 */
12 13 14 15
#define pgprot_noncached(prot)						\
	((boot_cpu_data.x86 > 3)					\
	 ? (__pgprot(pgprot_val(prot) |					\
		     cachemode2protval(_PAGE_CACHE_MODE_UC_MINUS)))	\
16 17
	 : (prot))

18
#ifndef __ASSEMBLY__
19 20
#include <asm/x86_init.h>

21
void ptdump_walk_pgd_level(struct seq_file *m, pgd_t *pgd);
S
Stephen Smalley 已提交
22 23 24 25 26 27 28
void ptdump_walk_pgd_level_checkwx(void);

#ifdef CONFIG_DEBUG_WX
#define debug_checkwx() ptdump_walk_pgd_level_checkwx()
#else
#define debug_checkwx() do { } while (0)
#endif
29

30 31 32 33
/*
 * ZERO_PAGE is a global shared page that is always zero: used
 * for zero-mapped memory areas etc..
 */
34 35
extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)]
	__visible;
36 37
#define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))

38 39
extern spinlock_t pgd_lock;
extern struct list_head pgd_list;
40

41 42
extern struct mm_struct *pgd_page_get_mm(struct page *page);

43 44 45 46 47
#ifdef CONFIG_PARAVIRT
#include <asm/paravirt.h>
#else  /* !CONFIG_PARAVIRT */
#define set_pte(ptep, pte)		native_set_pte(ptep, pte)
#define set_pte_at(mm, addr, ptep, pte)	native_set_pte_at(mm, addr, ptep, pte)
48
#define set_pmd_at(mm, addr, pmdp, pmd)	native_set_pmd_at(mm, addr, pmdp, pmd)
49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72

#define set_pte_atomic(ptep, pte)					\
	native_set_pte_atomic(ptep, pte)

#define set_pmd(pmdp, pmd)		native_set_pmd(pmdp, pmd)

#ifndef __PAGETABLE_PUD_FOLDED
#define set_pgd(pgdp, pgd)		native_set_pgd(pgdp, pgd)
#define pgd_clear(pgd)			native_pgd_clear(pgd)
#endif

#ifndef set_pud
# define set_pud(pudp, pud)		native_set_pud(pudp, pud)
#endif

#ifndef __PAGETABLE_PMD_FOLDED
#define pud_clear(pud)			native_pud_clear(pud)
#endif

#define pte_clear(mm, addr, ptep)	native_pte_clear(mm, addr, ptep)
#define pmd_clear(pmd)			native_pmd_clear(pmd)

#define pte_update(mm, addr, ptep)              do { } while (0)
#define pte_update_defer(mm, addr, ptep)        do { } while (0)
73 74
#define pmd_update(mm, addr, ptep)              do { } while (0)
#define pmd_update_defer(mm, addr, ptep)        do { } while (0)
75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91

#define pgd_val(x)	native_pgd_val(x)
#define __pgd(x)	native_make_pgd(x)

#ifndef __PAGETABLE_PUD_FOLDED
#define pud_val(x)	native_pud_val(x)
#define __pud(x)	native_make_pud(x)
#endif

#ifndef __PAGETABLE_PMD_FOLDED
#define pmd_val(x)	native_pmd_val(x)
#define __pmd(x)	native_make_pmd(x)
#endif

#define pte_val(x)	native_pte_val(x)
#define __pte(x)	native_make_pte(x)

92 93
#define arch_end_context_switch(prev)	do {} while(0)

94 95
#endif	/* CONFIG_PARAVIRT */

96 97 98 99
/*
 * The following only work if pte_present() is true.
 * Undefined behaviour if not..
 */
100 101
static inline int pte_dirty(pte_t pte)
{
102
	return pte_flags(pte) & _PAGE_DIRTY;
103 104 105 106
}

static inline int pte_young(pte_t pte)
{
107
	return pte_flags(pte) & _PAGE_ACCESSED;
108 109
}

110 111 112 113
static inline int pmd_dirty(pmd_t pmd)
{
	return pmd_flags(pmd) & _PAGE_DIRTY;
}
114

J
Johannes Weiner 已提交
115 116 117 118 119
static inline int pmd_young(pmd_t pmd)
{
	return pmd_flags(pmd) & _PAGE_ACCESSED;
}

120 121
static inline int pte_write(pte_t pte)
{
122
	return pte_flags(pte) & _PAGE_RW;
123 124 125 126
}

static inline int pte_huge(pte_t pte)
{
127
	return pte_flags(pte) & _PAGE_PSE;
128 129
}

130 131
static inline int pte_global(pte_t pte)
{
132
	return pte_flags(pte) & _PAGE_GLOBAL;
133 134 135 136
}

static inline int pte_exec(pte_t pte)
{
137
	return !(pte_flags(pte) & _PAGE_NX);
138 139
}

N
Nick Piggin 已提交
140 141
static inline int pte_special(pte_t pte)
{
142
	return pte_flags(pte) & _PAGE_SPECIAL;
N
Nick Piggin 已提交
143 144
}

H
Hugh Dickins 已提交
145 146 147 148 149
static inline unsigned long pte_pfn(pte_t pte)
{
	return (pte_val(pte) & PTE_PFN_MASK) >> PAGE_SHIFT;
}

A
Akinobu Mita 已提交
150 151
static inline unsigned long pmd_pfn(pmd_t pmd)
{
152
	return (pmd_val(pmd) & pmd_pfn_mask(pmd)) >> PAGE_SHIFT;
A
Akinobu Mita 已提交
153 154
}

155 156
static inline unsigned long pud_pfn(pud_t pud)
{
157
	return (pud_val(pud) & pud_pfn_mask(pud)) >> PAGE_SHIFT;
158 159
}

H
Hugh Dickins 已提交
160 161
#define pte_page(pte)	pfn_to_page(pte_pfn(pte))

162 163
static inline int pmd_large(pmd_t pte)
{
164
	return pmd_flags(pte) & _PAGE_PSE;
165 166
}

J
Johannes Weiner 已提交
167 168 169 170 171 172 173 174 175 176
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
static inline int pmd_trans_splitting(pmd_t pmd)
{
	return pmd_val(pmd) & _PAGE_SPLITTING;
}

static inline int pmd_trans_huge(pmd_t pmd)
{
	return pmd_val(pmd) & _PAGE_PSE;
}
177 178 179 180 181

static inline int has_transparent_hugepage(void)
{
	return cpu_has_pse;
}
J
Johannes Weiner 已提交
182 183
#endif /* CONFIG_TRANSPARENT_HUGEPAGE */

184 185 186 187 188 189 190 191 192 193 194 195 196 197
static inline pte_t pte_set_flags(pte_t pte, pteval_t set)
{
	pteval_t v = native_pte_val(pte);

	return native_make_pte(v | set);
}

static inline pte_t pte_clear_flags(pte_t pte, pteval_t clear)
{
	pteval_t v = native_pte_val(pte);

	return native_make_pte(v & ~clear);
}

198 199
static inline pte_t pte_mkclean(pte_t pte)
{
200
	return pte_clear_flags(pte, _PAGE_DIRTY);
201 202 203 204
}

static inline pte_t pte_mkold(pte_t pte)
{
205
	return pte_clear_flags(pte, _PAGE_ACCESSED);
206 207 208 209
}

static inline pte_t pte_wrprotect(pte_t pte)
{
210
	return pte_clear_flags(pte, _PAGE_RW);
211 212 213 214
}

static inline pte_t pte_mkexec(pte_t pte)
{
215
	return pte_clear_flags(pte, _PAGE_NX);
216 217 218 219
}

static inline pte_t pte_mkdirty(pte_t pte)
{
220
	return pte_set_flags(pte, _PAGE_DIRTY | _PAGE_SOFT_DIRTY);
221 222 223 224
}

static inline pte_t pte_mkyoung(pte_t pte)
{
225
	return pte_set_flags(pte, _PAGE_ACCESSED);
226 227 228 229
}

static inline pte_t pte_mkwrite(pte_t pte)
{
230
	return pte_set_flags(pte, _PAGE_RW);
231 232 233 234
}

static inline pte_t pte_mkhuge(pte_t pte)
{
235
	return pte_set_flags(pte, _PAGE_PSE);
236 237 238 239
}

static inline pte_t pte_clrhuge(pte_t pte)
{
240
	return pte_clear_flags(pte, _PAGE_PSE);
241 242 243 244
}

static inline pte_t pte_mkglobal(pte_t pte)
{
245
	return pte_set_flags(pte, _PAGE_GLOBAL);
246 247 248 249
}

static inline pte_t pte_clrglobal(pte_t pte)
{
250
	return pte_clear_flags(pte, _PAGE_GLOBAL);
251
}
252

N
Nick Piggin 已提交
253 254
static inline pte_t pte_mkspecial(pte_t pte)
{
255
	return pte_set_flags(pte, _PAGE_SPECIAL);
N
Nick Piggin 已提交
256 257
}

J
Johannes Weiner 已提交
258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283
static inline pmd_t pmd_set_flags(pmd_t pmd, pmdval_t set)
{
	pmdval_t v = native_pmd_val(pmd);

	return __pmd(v | set);
}

static inline pmd_t pmd_clear_flags(pmd_t pmd, pmdval_t clear)
{
	pmdval_t v = native_pmd_val(pmd);

	return __pmd(v & ~clear);
}

static inline pmd_t pmd_mkold(pmd_t pmd)
{
	return pmd_clear_flags(pmd, _PAGE_ACCESSED);
}

static inline pmd_t pmd_wrprotect(pmd_t pmd)
{
	return pmd_clear_flags(pmd, _PAGE_RW);
}

static inline pmd_t pmd_mkdirty(pmd_t pmd)
{
284
	return pmd_set_flags(pmd, _PAGE_DIRTY | _PAGE_SOFT_DIRTY);
J
Johannes Weiner 已提交
285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303
}

static inline pmd_t pmd_mkhuge(pmd_t pmd)
{
	return pmd_set_flags(pmd, _PAGE_PSE);
}

static inline pmd_t pmd_mkyoung(pmd_t pmd)
{
	return pmd_set_flags(pmd, _PAGE_ACCESSED);
}

static inline pmd_t pmd_mkwrite(pmd_t pmd)
{
	return pmd_set_flags(pmd, _PAGE_RW);
}

static inline pmd_t pmd_mknotpresent(pmd_t pmd)
{
304
	return pmd_clear_flags(pmd, _PAGE_PRESENT | _PAGE_PROTNONE);
J
Johannes Weiner 已提交
305 306
}

307
#ifdef CONFIG_HAVE_ARCH_SOFT_DIRTY
308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327
static inline int pte_soft_dirty(pte_t pte)
{
	return pte_flags(pte) & _PAGE_SOFT_DIRTY;
}

static inline int pmd_soft_dirty(pmd_t pmd)
{
	return pmd_flags(pmd) & _PAGE_SOFT_DIRTY;
}

static inline pte_t pte_mksoft_dirty(pte_t pte)
{
	return pte_set_flags(pte, _PAGE_SOFT_DIRTY);
}

static inline pmd_t pmd_mksoft_dirty(pmd_t pmd)
{
	return pmd_set_flags(pmd, _PAGE_SOFT_DIRTY);
}

328 329
#endif /* CONFIG_HAVE_ARCH_SOFT_DIRTY */

330 331 332 333 334 335 336 337 338 339 340 341 342 343
/*
 * Mask out unsupported bits in a present pgprot.  Non-present pgprots
 * can use those bits for other purposes, so leave them be.
 */
static inline pgprotval_t massage_pgprot(pgprot_t pgprot)
{
	pgprotval_t protval = pgprot_val(pgprot);

	if (protval & _PAGE_PRESENT)
		protval &= __supported_pte_mask;

	return protval;
}

344 345
static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot)
{
346 347
	return __pte(((phys_addr_t)page_nr << PAGE_SHIFT) |
		     massage_pgprot(pgprot));
348 349 350 351
}

static inline pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot)
{
352 353
	return __pmd(((phys_addr_t)page_nr << PAGE_SHIFT) |
		     massage_pgprot(pgprot));
354 355
}

356 357 358 359 360 361 362 363
static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
{
	pteval_t val = pte_val(pte);

	/*
	 * Chop off the NX bit (if present), and add the NX portion of
	 * the newprot (if present):
	 */
364
	val &= _PAGE_CHG_MASK;
365
	val |= massage_pgprot(newprot) & ~_PAGE_CHG_MASK;
366 367 368 369

	return __pte(val);
}

J
Johannes Weiner 已提交
370 371 372 373 374 375 376 377 378 379
static inline pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot)
{
	pmdval_t val = pmd_val(pmd);

	val &= _HPAGE_CHG_MASK;
	val |= massage_pgprot(newprot) & ~_HPAGE_CHG_MASK;

	return __pmd(val);
}

380 381 382 383 384 385 386 387 388
/* mprotect needs to preserve PAT bits when updating vm_page_prot */
#define pgprot_modify pgprot_modify
static inline pgprot_t pgprot_modify(pgprot_t oldprot, pgprot_t newprot)
{
	pgprotval_t preservebits = pgprot_val(oldprot) & _PAGE_CHG_MASK;
	pgprotval_t addbits = pgprot_val(newprot);
	return __pgprot(preservebits | addbits);
}

389 390 391
#define pte_pgprot(x) __pgprot(pte_flags(x))
#define pmd_pgprot(x) __pgprot(pmd_flags(x))
#define pud_pgprot(x) __pgprot(pud_flags(x))
A
Andi Kleen 已提交
392

393
#define canon_pgprot(p) __pgprot(massage_pgprot(p))
A
Andi Kleen 已提交
394

395
static inline int is_new_memtype_allowed(u64 paddr, unsigned long size,
396 397
					 enum page_cache_mode pcm,
					 enum page_cache_mode new_pcm)
398
{
399
	/*
400
	 * PAT type is always WB for untracked ranges, so no need to check.
401
	 */
402
	if (x86_platform.is_untracked_pat_range(paddr, paddr + size))
403 404
		return 1;

405 406 407 408 409
	/*
	 * Certain new memtypes are not allowed with certain
	 * requested memtype:
	 * - request is uncached, return cannot be write-back
	 * - request is write-combine, return cannot be write-back
410 411
	 * - request is write-through, return cannot be write-back
	 * - request is write-through, return cannot be write-combine
412
	 */
413 414 415
	if ((pcm == _PAGE_CACHE_MODE_UC_MINUS &&
	     new_pcm == _PAGE_CACHE_MODE_WB) ||
	    (pcm == _PAGE_CACHE_MODE_WC &&
416 417 418 419 420
	     new_pcm == _PAGE_CACHE_MODE_WB) ||
	    (pcm == _PAGE_CACHE_MODE_WT &&
	     new_pcm == _PAGE_CACHE_MODE_WB) ||
	    (pcm == _PAGE_CACHE_MODE_WT &&
	     new_pcm == _PAGE_CACHE_MODE_WC)) {
421 422 423 424 425 426
		return 0;
	}

	return 1;
}

427 428
pmd_t *populate_extra_pmd(unsigned long vaddr);
pte_t *populate_extra_pte(unsigned long vaddr);
429 430
#endif	/* __ASSEMBLY__ */

431
#ifdef CONFIG_X86_32
432
# include <asm/pgtable_32.h>
433
#else
434
# include <asm/pgtable_64.h>
435
#endif
436

437
#ifndef __ASSEMBLY__
J
Jeremy Fitzhardinge 已提交
438
#include <linux/mm_types.h>
439
#include <linux/mmdebug.h>
440
#include <linux/log2.h>
441

J
Jeremy Fitzhardinge 已提交
442 443 444 445 446
static inline int pte_none(pte_t pte)
{
	return !pte.pte;
}

J
Jeremy Fitzhardinge 已提交
447 448 449 450 451 452
#define __HAVE_ARCH_PTE_SAME
static inline int pte_same(pte_t a, pte_t b)
{
	return a.pte == b.pte;
}

J
Jeremy Fitzhardinge 已提交
453
static inline int pte_present(pte_t a)
454 455 456 457
{
	return pte_flags(a) & (_PAGE_PRESENT | _PAGE_PROTNONE);
}

R
Rik van Riel 已提交
458
#define pte_accessible pte_accessible
459
static inline bool pte_accessible(struct mm_struct *mm, pte_t a)
R
Rik van Riel 已提交
460
{
461 462 463
	if (pte_flags(a) & _PAGE_PRESENT)
		return true;

464
	if ((pte_flags(a) & _PAGE_PROTNONE) &&
465 466 467 468
			mm_tlb_flush_pending(mm))
		return true;

	return false;
R
Rik van Riel 已提交
469 470
}

J
Jeremy Fitzhardinge 已提交
471
static inline int pte_hidden(pte_t pte)
472
{
J
Jeremy Fitzhardinge 已提交
473
	return pte_flags(pte) & _PAGE_HIDDEN;
474 475
}

J
Jeremy Fitzhardinge 已提交
476 477
static inline int pmd_present(pmd_t pmd)
{
478 479 480 481 482 483
	/*
	 * Checking for _PAGE_PSE is needed too because
	 * split_huge_page will temporarily clear the present bit (but
	 * the _PAGE_PSE flag will remain set at all times while the
	 * _PAGE_PRESENT bit is clear).
	 */
484
	return pmd_flags(pmd) & (_PAGE_PRESENT | _PAGE_PROTNONE | _PAGE_PSE);
J
Jeremy Fitzhardinge 已提交
485 486
}

487 488 489 490 491 492 493
#ifdef CONFIG_NUMA_BALANCING
/*
 * These work without NUMA balancing but the kernel does not care. See the
 * comment in include/asm-generic/pgtable.h
 */
static inline int pte_protnone(pte_t pte)
{
494 495
	return (pte_flags(pte) & (_PAGE_PROTNONE | _PAGE_PRESENT))
		== _PAGE_PROTNONE;
496 497 498 499
}

static inline int pmd_protnone(pmd_t pmd)
{
500 501
	return (pmd_flags(pmd) & (_PAGE_PROTNONE | _PAGE_PRESENT))
		== _PAGE_PROTNONE;
502 503 504
}
#endif /* CONFIG_NUMA_BALANCING */

J
Jeremy Fitzhardinge 已提交
505 506 507 508
static inline int pmd_none(pmd_t pmd)
{
	/* Only check low word on 32-bit platforms, since it might be
	   out of sync with upper half. */
509
	return (unsigned long)native_pmd_val(pmd) == 0;
J
Jeremy Fitzhardinge 已提交
510 511
}

J
Jeremy Fitzhardinge 已提交
512 513
static inline unsigned long pmd_page_vaddr(pmd_t pmd)
{
514
	return (unsigned long)__va(pmd_val(pmd) & pmd_pfn_mask(pmd));
J
Jeremy Fitzhardinge 已提交
515 516
}

517 518 519 520
/*
 * Currently stuck as a macro due to indirect forward reference to
 * linux/mmzone.h's __section_mem_map_addr() definition:
 */
521 522
#define pmd_page(pmd)		\
	pfn_to_page((pmd_val(pmd) & pmd_pfn_mask(pmd)) >> PAGE_SHIFT)
J
Jeremy Fitzhardinge 已提交
523

J
Jeremy Fitzhardinge 已提交
524 525 526 527 528 529
/*
 * the pmd page can be thought of an array like this: pmd_t[PTRS_PER_PMD]
 *
 * this macro returns the index of the entry in the pmd page which would
 * control the given virtual address
 */
F
Figo.zhang 已提交
530
static inline unsigned long pmd_index(unsigned long address)
J
Jeremy Fitzhardinge 已提交
531 532 533 534
{
	return (address >> PMD_SHIFT) & (PTRS_PER_PMD - 1);
}

J
Jeremy Fitzhardinge 已提交
535 536 537 538 539 540 541 542 543
/*
 * Conversion functions: convert a page and protection to a page entry,
 * and a page entry and page directory to the page they refer to.
 *
 * (Currently stuck as a macro because of indirect forward reference
 * to linux/mm.h:page_to_nid())
 */
#define mk_pte(page, pgprot)   pfn_pte(page_to_pfn(page), (pgprot))

J
Jeremy Fitzhardinge 已提交
544 545 546 547 548 549
/*
 * the pte page can be thought of an array like this: pte_t[PTRS_PER_PTE]
 *
 * this function returns the index of the entry in the pte page which would
 * control the given virtual address
 */
F
Figo.zhang 已提交
550
static inline unsigned long pte_index(unsigned long address)
J
Jeremy Fitzhardinge 已提交
551 552 553 554
{
	return (address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1);
}

555 556 557 558 559
static inline pte_t *pte_offset_kernel(pmd_t *pmd, unsigned long address)
{
	return (pte_t *)pmd_page_vaddr(*pmd) + pte_index(address);
}

J
Jeremy Fitzhardinge 已提交
560 561
static inline int pmd_bad(pmd_t pmd)
{
562
	return (pmd_flags(pmd) & ~_PAGE_USER) != _KERNPG_TABLE;
J
Jeremy Fitzhardinge 已提交
563 564
}

J
Jeremy Fitzhardinge 已提交
565 566 567 568 569
static inline unsigned long pages_to_mb(unsigned long npg)
{
	return npg >> (20 - PAGE_SHIFT);
}

570
#if CONFIG_PGTABLE_LEVELS > 2
J
Jeremy Fitzhardinge 已提交
571 572
static inline int pud_none(pud_t pud)
{
573
	return native_pud_val(pud) == 0;
J
Jeremy Fitzhardinge 已提交
574 575
}

J
Jeremy Fitzhardinge 已提交
576 577
static inline int pud_present(pud_t pud)
{
578
	return pud_flags(pud) & _PAGE_PRESENT;
J
Jeremy Fitzhardinge 已提交
579
}
J
Jeremy Fitzhardinge 已提交
580 581 582

static inline unsigned long pud_page_vaddr(pud_t pud)
{
583
	return (unsigned long)__va(pud_val(pud) & pud_pfn_mask(pud));
J
Jeremy Fitzhardinge 已提交
584
}
J
Jeremy Fitzhardinge 已提交
585

586 587 588 589
/*
 * Currently stuck as a macro due to indirect forward reference to
 * linux/mmzone.h's __section_mem_map_addr() definition:
 */
590 591
#define pud_page(pud)		\
	pfn_to_page((pud_val(pud) & pud_pfn_mask(pud)) >> PAGE_SHIFT)
J
Jeremy Fitzhardinge 已提交
592 593 594 595 596 597

/* Find an entry in the second-level page table.. */
static inline pmd_t *pmd_offset(pud_t *pud, unsigned long address)
{
	return (pmd_t *)pud_page_vaddr(*pud) + pmd_index(address);
}
J
Jeremy Fitzhardinge 已提交
598

J
Jeremy Fitzhardinge 已提交
599 600
static inline int pud_large(pud_t pud)
{
601
	return (pud_val(pud) & (_PAGE_PSE | _PAGE_PRESENT)) ==
J
Jeremy Fitzhardinge 已提交
602 603
		(_PAGE_PSE | _PAGE_PRESENT);
}
J
Jeremy Fitzhardinge 已提交
604 605 606

static inline int pud_bad(pud_t pud)
{
607
	return (pud_flags(pud) & ~(_KERNPG_TABLE | _PAGE_USER)) != 0;
J
Jeremy Fitzhardinge 已提交
608
}
609 610 611 612 613
#else
static inline int pud_large(pud_t pud)
{
	return 0;
}
614
#endif	/* CONFIG_PGTABLE_LEVELS > 2 */
J
Jeremy Fitzhardinge 已提交
615

616
#if CONFIG_PGTABLE_LEVELS > 3
J
Jeremy Fitzhardinge 已提交
617 618
static inline int pgd_present(pgd_t pgd)
{
619
	return pgd_flags(pgd) & _PAGE_PRESENT;
J
Jeremy Fitzhardinge 已提交
620
}
J
Jeremy Fitzhardinge 已提交
621 622 623 624 625

static inline unsigned long pgd_page_vaddr(pgd_t pgd)
{
	return (unsigned long)__va((unsigned long)pgd_val(pgd) & PTE_PFN_MASK);
}
J
Jeremy Fitzhardinge 已提交
626

627 628 629 630 631
/*
 * Currently stuck as a macro due to indirect forward reference to
 * linux/mmzone.h's __section_mem_map_addr() definition:
 */
#define pgd_page(pgd)		pfn_to_page(pgd_val(pgd) >> PAGE_SHIFT)
J
Jeremy Fitzhardinge 已提交
632 633

/* to find an entry in a page-table-directory. */
F
Figo.zhang 已提交
634
static inline unsigned long pud_index(unsigned long address)
J
Jeremy Fitzhardinge 已提交
635 636 637
{
	return (address >> PUD_SHIFT) & (PTRS_PER_PUD - 1);
}
J
Jeremy Fitzhardinge 已提交
638 639 640 641 642

static inline pud_t *pud_offset(pgd_t *pgd, unsigned long address)
{
	return (pud_t *)pgd_page_vaddr(*pgd) + pud_index(address);
}
J
Jeremy Fitzhardinge 已提交
643 644 645

static inline int pgd_bad(pgd_t pgd)
{
646
	return (pgd_flags(pgd) & ~_PAGE_USER) != _KERNPG_TABLE;
J
Jeremy Fitzhardinge 已提交
647
}
J
Jeremy Fitzhardinge 已提交
648 649 650

static inline int pgd_none(pgd_t pgd)
{
651
	return !native_pgd_val(pgd);
J
Jeremy Fitzhardinge 已提交
652
}
653
#endif	/* CONFIG_PGTABLE_LEVELS > 3 */
J
Jeremy Fitzhardinge 已提交
654

655 656
#endif	/* __ASSEMBLY__ */

J
Jeremy Fitzhardinge 已提交
657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676
/*
 * the pgd page can be thought of an array like this: pgd_t[PTRS_PER_PGD]
 *
 * this macro returns the index of the entry in the pgd page which would
 * control the given virtual address
 */
#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD - 1))

/*
 * pgd_offset() returns a (pgd_t *)
 * pgd_index() is used get the offset into the pgd page's array of pgd_t's;
 */
#define pgd_offset(mm, address) ((mm)->pgd + pgd_index((address)))
/*
 * a shortcut which implies the use of the kernel's pgd, instead
 * of a process's
 */
#define pgd_offset_k(address) pgd_offset(&init_mm, (address))


J
Jeremy Fitzhardinge 已提交
677 678 679
#define KERNEL_PGD_BOUNDARY	pgd_index(PAGE_OFFSET)
#define KERNEL_PGD_PTRS		(PTRS_PER_PGD - KERNEL_PGD_BOUNDARY)

680 681
#ifndef __ASSEMBLY__

682
extern int direct_gbpages;
683
void init_mem_mapping(void);
Y
Yinghai Lu 已提交
684
void early_alloc_pgt_buf(void);
685

686 687 688 689 690 691 692 693 694 695
/* local pte updates need not use xchg for locking */
static inline pte_t native_local_ptep_get_and_clear(pte_t *ptep)
{
	pte_t res = *ptep;

	/* Pure native function needs no input for mm, addr */
	native_pte_clear(NULL, 0, ptep);
	return res;
}

J
Johannes Weiner 已提交
696 697 698 699 700 701 702 703
static inline pmd_t native_local_pmdp_get_and_clear(pmd_t *pmdp)
{
	pmd_t res = *pmdp;

	native_pmd_clear(pmdp);
	return res;
}

704 705 706 707 708 709
static inline void native_set_pte_at(struct mm_struct *mm, unsigned long addr,
				     pte_t *ptep , pte_t pte)
{
	native_set_pte(ptep, pte);
}

A
Andrea Arcangeli 已提交
710 711 712 713 714 715
static inline void native_set_pmd_at(struct mm_struct *mm, unsigned long addr,
				     pmd_t *pmdp , pmd_t pmd)
{
	native_set_pmd(pmdp, pmd);
}

716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740
#ifndef CONFIG_PARAVIRT
/*
 * Rules for using pte_update - it must be called after any PTE update which
 * has not been done using the set_pte / clear_pte interfaces.  It is used by
 * shadow mode hypervisors to resynchronize the shadow page tables.  Kernel PTE
 * updates should either be sets, clears, or set_pte_atomic for P->P
 * transitions, which means this hook should only be called for user PTEs.
 * This hook implies a P->P protection or access change has taken place, which
 * requires a subsequent TLB flush.  The notification can optionally be delayed
 * until the TLB flush event by using the pte_update_defer form of the
 * interface, but care must be taken to assure that the flush happens while
 * still holding the same page table lock so that the shadow and primary pages
 * do not become out of sync on SMP.
 */
#define pte_update(mm, addr, ptep)		do { } while (0)
#define pte_update_defer(mm, addr, ptep)	do { } while (0)
#endif

/*
 * We only update the dirty/accessed state if we set
 * the dirty bit by hand in the kernel, since the hardware
 * will do the accessed bit for us, and we don't want to
 * race with other CPU's that might be updating the dirty
 * bit at the same time.
 */
741 742
struct vm_area_struct;

743
#define  __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
744 745 746
extern int ptep_set_access_flags(struct vm_area_struct *vma,
				 unsigned long address, pte_t *ptep,
				 pte_t entry, int dirty);
747 748

#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
749 750
extern int ptep_test_and_clear_young(struct vm_area_struct *vma,
				     unsigned long addr, pte_t *ptep);
751 752

#define __HAVE_ARCH_PTEP_CLEAR_YOUNG_FLUSH
753 754
extern int ptep_clear_flush_young(struct vm_area_struct *vma,
				  unsigned long address, pte_t *ptep);
755 756

#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
757 758
static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
				       pte_t *ptep)
759 760 761 762 763 764 765
{
	pte_t pte = native_ptep_get_and_clear(ptep);
	pte_update(mm, addr, ptep);
	return pte;
}

#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
766 767 768
static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm,
					    unsigned long addr, pte_t *ptep,
					    int full)
769 770 771 772 773 774 775 776 777 778 779 780 781 782 783
{
	pte_t pte;
	if (full) {
		/*
		 * Full address destruction in progress; paravirt does not
		 * care about updates and native needs no locking
		 */
		pte = native_local_ptep_get_and_clear(ptep);
	} else {
		pte = ptep_get_and_clear(mm, addr, ptep);
	}
	return pte;
}

#define __HAVE_ARCH_PTEP_SET_WRPROTECT
784 785
static inline void ptep_set_wrprotect(struct mm_struct *mm,
				      unsigned long addr, pte_t *ptep)
786
{
J
Jeremy Fitzhardinge 已提交
787
	clear_bit(_PAGE_BIT_RW, (unsigned long *)&ptep->pte);
788 789 790
	pte_update(mm, addr, ptep);
}

791
#define flush_tlb_fix_spurious_fault(vma, address) do { } while (0)
792

J
Johannes Weiner 已提交
793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818
#define mk_pmd(page, pgprot)   pfn_pmd(page_to_pfn(page), (pgprot))

#define  __HAVE_ARCH_PMDP_SET_ACCESS_FLAGS
extern int pmdp_set_access_flags(struct vm_area_struct *vma,
				 unsigned long address, pmd_t *pmdp,
				 pmd_t entry, int dirty);

#define __HAVE_ARCH_PMDP_TEST_AND_CLEAR_YOUNG
extern int pmdp_test_and_clear_young(struct vm_area_struct *vma,
				     unsigned long addr, pmd_t *pmdp);

#define __HAVE_ARCH_PMDP_CLEAR_YOUNG_FLUSH
extern int pmdp_clear_flush_young(struct vm_area_struct *vma,
				  unsigned long address, pmd_t *pmdp);


#define __HAVE_ARCH_PMDP_SPLITTING_FLUSH
extern void pmdp_splitting_flush(struct vm_area_struct *vma,
				 unsigned long addr, pmd_t *pmdp);

#define __HAVE_ARCH_PMD_WRITE
static inline int pmd_write(pmd_t pmd)
{
	return pmd_flags(pmd) & _PAGE_RW;
}

819 820
#define __HAVE_ARCH_PMDP_HUGE_GET_AND_CLEAR
static inline pmd_t pmdp_huge_get_and_clear(struct mm_struct *mm, unsigned long addr,
J
Johannes Weiner 已提交
821 822 823 824 825 826 827 828 829 830 831 832 833 834 835
				       pmd_t *pmdp)
{
	pmd_t pmd = native_pmdp_get_and_clear(pmdp);
	pmd_update(mm, addr, pmdp);
	return pmd;
}

#define __HAVE_ARCH_PMDP_SET_WRPROTECT
static inline void pmdp_set_wrprotect(struct mm_struct *mm,
				      unsigned long addr, pmd_t *pmdp)
{
	clear_bit(_PAGE_BIT_RW, (unsigned long *)pmdp);
	pmd_update(mm, addr, pmdp);
}

J
Jeremy Fitzhardinge 已提交
836 837 838 839 840 841 842 843 844 845 846 847 848 849 850
/*
 * clone_pgd_range(pgd_t *dst, pgd_t *src, int count);
 *
 *  dst - pointer to pgd range anwhere on a pgd page
 *  src - ""
 *  count - the number of pgds to copy.
 *
 * dst and src can be on the same page, but the range must not overlap,
 * and must not cross a page boundary.
 */
static inline void clone_pgd_range(pgd_t *dst, pgd_t *src, int count)
{
       memcpy(dst, src, count * sizeof(pgd_t));
}

851 852 853 854 855 856 857 858 859 860 861 862 863
#define PTE_SHIFT ilog2(PTRS_PER_PTE)
static inline int page_level_shift(enum pg_level level)
{
	return (PAGE_SHIFT - PTE_SHIFT) + level * PTE_SHIFT;
}
static inline unsigned long page_level_size(enum pg_level level)
{
	return 1UL << page_level_shift(level);
}
static inline unsigned long page_level_mask(enum pg_level level)
{
	return ~(page_level_size(level) - 1);
}
J
Jeremy Fitzhardinge 已提交
864

865 866 867 868 869 870 871 872 873 874 875 876
/*
 * The x86 doesn't have any external MMU info: the kernel page
 * tables contain all the necessary information.
 */
static inline void update_mmu_cache(struct vm_area_struct *vma,
		unsigned long addr, pte_t *ptep)
{
}
static inline void update_mmu_cache_pmd(struct vm_area_struct *vma,
		unsigned long addr, pmd_t *pmd)
{
}
J
Jeremy Fitzhardinge 已提交
877

878
#ifdef CONFIG_HAVE_ARCH_SOFT_DIRTY
879 880 881 882 883 884 885 886 887 888 889 890 891 892
static inline pte_t pte_swp_mksoft_dirty(pte_t pte)
{
	return pte_set_flags(pte, _PAGE_SWP_SOFT_DIRTY);
}

static inline int pte_swp_soft_dirty(pte_t pte)
{
	return pte_flags(pte) & _PAGE_SWP_SOFT_DIRTY;
}

static inline pte_t pte_swp_clear_soft_dirty(pte_t pte)
{
	return pte_clear_flags(pte, _PAGE_SWP_SOFT_DIRTY);
}
893
#endif
894

895 896 897
#include <asm-generic/pgtable.h>
#endif	/* __ASSEMBLY__ */

H
H. Peter Anvin 已提交
898
#endif /* _ASM_X86_PGTABLE_H */