gup.c 70.9 KB
Newer Older
1
// SPDX-License-Identifier: GPL-2.0-only
2 3 4 5 6 7
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/err.h>
#include <linux/spinlock.h>

#include <linux/mm.h>
8
#include <linux/memremap.h>
9 10 11 12 13
#include <linux/pagemap.h>
#include <linux/rmap.h>
#include <linux/swap.h>
#include <linux/swapops.h>

14
#include <linux/sched/signal.h>
15
#include <linux/rwsem.h>
16
#include <linux/hugetlb.h>
17 18 19
#include <linux/migrate.h>
#include <linux/mm_inline.h>
#include <linux/sched/mm.h>
20

21
#include <asm/mmu_context.h>
22
#include <asm/pgtable.h>
23
#include <asm/tlbflush.h>
24

25 26
#include "internal.h"

27 28 29 30 31
struct follow_page_context {
	struct dev_pagemap *pgmap;
	unsigned int page_mask;
};

32 33 34 35 36 37 38 39 40 41 42 43 44 45 46
/*
 * Return the compound head page with ref appropriately incremented,
 * or NULL if that failed.
 */
static inline struct page *try_get_compound_head(struct page *page, int refs)
{
	struct page *head = compound_head(page);

	if (WARN_ON_ONCE(page_ref_count(head) < 0))
		return NULL;
	if (unlikely(!page_cache_add_speculative(head, refs)))
		return NULL;
	return head;
}

47
/**
48 49
 * put_user_pages_dirty_lock() - release and optionally dirty gup-pinned pages
 * @pages:  array of pages to be maybe marked dirty, and definitely released.
50
 * @npages: number of pages in the @pages array.
51
 * @make_dirty: whether to mark the pages dirty
52 53 54 55 56
 *
 * "gup-pinned page" refers to a page that has had one of the get_user_pages()
 * variants called on that page.
 *
 * For each page in the @pages array, make that page (or its head page, if a
57 58 59
 * compound page) dirty, if @make_dirty is true, and if the page was previously
 * listed as clean. In any case, releases all pages using put_user_page(),
 * possibly via put_user_pages(), for the non-dirty case.
60 61 62
 *
 * Please see the put_user_page() documentation for details.
 *
63 64 65 66
 * set_page_dirty_lock() is used internally. If instead, set_page_dirty() is
 * required, then the caller should a) verify that this is really correct,
 * because _lock() is usually required, and b) hand code it:
 * set_page_dirty_lock(), put_user_page().
67 68
 *
 */
69 70
void put_user_pages_dirty_lock(struct page **pages, unsigned long npages,
			       bool make_dirty)
71
{
72
	unsigned long index;
73

74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110
	/*
	 * TODO: this can be optimized for huge pages: if a series of pages is
	 * physically contiguous and part of the same compound page, then a
	 * single operation to the head page should suffice.
	 */

	if (!make_dirty) {
		put_user_pages(pages, npages);
		return;
	}

	for (index = 0; index < npages; index++) {
		struct page *page = compound_head(pages[index]);
		/*
		 * Checking PageDirty at this point may race with
		 * clear_page_dirty_for_io(), but that's OK. Two key
		 * cases:
		 *
		 * 1) This code sees the page as already dirty, so it
		 * skips the call to set_page_dirty(). That could happen
		 * because clear_page_dirty_for_io() called
		 * page_mkclean(), followed by set_page_dirty().
		 * However, now the page is going to get written back,
		 * which meets the original intention of setting it
		 * dirty, so all is well: clear_page_dirty_for_io() goes
		 * on to call TestClearPageDirty(), and write the page
		 * back.
		 *
		 * 2) This code sees the page as clean, so it calls
		 * set_page_dirty(). The page stays dirty, despite being
		 * written back, so it gets written back again in the
		 * next writeback cycle. This is harmless.
		 */
		if (!PageDirty(page))
			set_page_dirty_lock(page);
		put_user_page(page);
	}
111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136
}
EXPORT_SYMBOL(put_user_pages_dirty_lock);

/**
 * put_user_pages() - release an array of gup-pinned pages.
 * @pages:  array of pages to be marked dirty and released.
 * @npages: number of pages in the @pages array.
 *
 * For each page in the @pages array, release the page using put_user_page().
 *
 * Please see the put_user_page() documentation for details.
 */
void put_user_pages(struct page **pages, unsigned long npages)
{
	unsigned long index;

	/*
	 * TODO: this can be optimized for huge pages: if a series of pages is
	 * physically contiguous and part of the same compound page, then a
	 * single operation to the head page should suffice.
	 */
	for (index = 0; index < npages; index++)
		put_user_page(pages[index]);
}
EXPORT_SYMBOL(put_user_pages);

137
#ifdef CONFIG_MMU
138 139
static struct page *no_page_table(struct vm_area_struct *vma,
		unsigned int flags)
140
{
141 142 143 144 145 146 147 148 149 150 151 152
	/*
	 * When core dumping an enormous anonymous area that nobody
	 * has touched so far, we don't want to allocate unnecessary pages or
	 * page tables.  Return error instead of NULL to skip handle_mm_fault,
	 * then get_dump_page() will return NULL to leave a hole in the dump.
	 * But we can only make this optimization where a hole would surely
	 * be zero-filled if handle_mm_fault() actually did handle it.
	 */
	if ((flags & FOLL_DUMP) && (!vma->vm_ops || !vma->vm_ops->fault))
		return ERR_PTR(-EFAULT);
	return NULL;
}
153

154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177
static int follow_pfn_pte(struct vm_area_struct *vma, unsigned long address,
		pte_t *pte, unsigned int flags)
{
	/* No page to get reference */
	if (flags & FOLL_GET)
		return -EFAULT;

	if (flags & FOLL_TOUCH) {
		pte_t entry = *pte;

		if (flags & FOLL_WRITE)
			entry = pte_mkdirty(entry);
		entry = pte_mkyoung(entry);

		if (!pte_same(*pte, entry)) {
			set_pte_at(vma->vm_mm, address, pte, entry);
			update_mmu_cache(vma, address, pte);
		}
	}

	/* Proper page table entry exists, but no corresponding struct page */
	return -EEXIST;
}

178 179 180 181 182 183
/*
 * FOLL_FORCE can write to even unwritable pte's, but only
 * after we've gone through a COW cycle and they are dirty.
 */
static inline bool can_follow_write_pte(pte_t pte, unsigned int flags)
{
184
	return pte_write(pte) ||
185 186 187
		((flags & FOLL_FORCE) && (flags & FOLL_COW) && pte_dirty(pte));
}

188
static struct page *follow_page_pte(struct vm_area_struct *vma,
189 190
		unsigned long address, pmd_t *pmd, unsigned int flags,
		struct dev_pagemap **pgmap)
191 192 193 194 195
{
	struct mm_struct *mm = vma->vm_mm;
	struct page *page;
	spinlock_t *ptl;
	pte_t *ptep, pte;
196

197 198 199 200
	/* FOLL_GET and FOLL_PIN are mutually exclusive. */
	if (WARN_ON_ONCE((flags & (FOLL_PIN | FOLL_GET)) ==
			 (FOLL_PIN | FOLL_GET)))
		return ERR_PTR(-EINVAL);
201
retry:
202
	if (unlikely(pmd_bad(*pmd)))
203
		return no_page_table(vma, flags);
204 205 206 207 208 209 210 211 212 213 214 215

	ptep = pte_offset_map_lock(mm, pmd, address, &ptl);
	pte = *ptep;
	if (!pte_present(pte)) {
		swp_entry_t entry;
		/*
		 * KSM's break_ksm() relies upon recognizing a ksm page
		 * even while it is being migrated, so for that case we
		 * need migration_entry_wait().
		 */
		if (likely(!(flags & FOLL_MIGRATION)))
			goto no_page;
216
		if (pte_none(pte))
217 218 219 220 221 222
			goto no_page;
		entry = pte_to_swp_entry(pte);
		if (!is_migration_entry(entry))
			goto no_page;
		pte_unmap_unlock(ptep, ptl);
		migration_entry_wait(mm, pmd, address);
223
		goto retry;
224
	}
225
	if ((flags & FOLL_NUMA) && pte_protnone(pte))
226
		goto no_page;
227
	if ((flags & FOLL_WRITE) && !can_follow_write_pte(pte, flags)) {
228 229 230
		pte_unmap_unlock(ptep, ptl);
		return NULL;
	}
231 232

	page = vm_normal_page(vma, address, pte);
233 234 235 236 237
	if (!page && pte_devmap(pte) && (flags & FOLL_GET)) {
		/*
		 * Only return device mapping pages in the FOLL_GET case since
		 * they are only valid while holding the pgmap reference.
		 */
238 239
		*pgmap = get_dev_pagemap(pte_pfn(pte), *pgmap);
		if (*pgmap)
240 241 242 243
			page = pte_page(pte);
		else
			goto no_page;
	} else if (unlikely(!page)) {
244 245 246 247 248 249 250 251 252 253 254 255 256 257 258
		if (flags & FOLL_DUMP) {
			/* Avoid special (like zero) pages in core dumps */
			page = ERR_PTR(-EFAULT);
			goto out;
		}

		if (is_zero_pfn(pte_pfn(pte))) {
			page = pte_page(pte);
		} else {
			int ret;

			ret = follow_pfn_pte(vma, address, ptep, flags);
			page = ERR_PTR(ret);
			goto out;
		}
259 260
	}

261 262 263 264 265 266 267 268 269 270 271 272 273
	if (flags & FOLL_SPLIT && PageTransCompound(page)) {
		int ret;
		get_page(page);
		pte_unmap_unlock(ptep, ptl);
		lock_page(page);
		ret = split_huge_page(page);
		unlock_page(page);
		put_page(page);
		if (ret)
			return ERR_PTR(ret);
		goto retry;
	}

274 275 276 277 278 279
	if (flags & FOLL_GET) {
		if (unlikely(!try_get_page(page))) {
			page = ERR_PTR(-ENOMEM);
			goto out;
		}
	}
280 281 282 283 284 285 286 287 288 289 290
	if (flags & FOLL_TOUCH) {
		if ((flags & FOLL_WRITE) &&
		    !pte_dirty(pte) && !PageDirty(page))
			set_page_dirty(page);
		/*
		 * pte_mkyoung() would be more correct here, but atomic care
		 * is needed to avoid losing the dirty bit: it is easier to use
		 * mark_page_accessed().
		 */
		mark_page_accessed(page);
	}
Eric B Munson's avatar
Eric B Munson committed
291
	if ((flags & FOLL_MLOCK) && (vma->vm_flags & VM_LOCKED)) {
292 293 294 295
		/* Do not mlock pte-mapped THP */
		if (PageTransCompound(page))
			goto out;

296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316
		/*
		 * The preliminary mapping check is mainly to avoid the
		 * pointless overhead of lock_page on the ZERO_PAGE
		 * which might bounce very badly if there is contention.
		 *
		 * If the page is already locked, we don't need to
		 * handle it now - vmscan will handle it later if and
		 * when it attempts to reclaim the page.
		 */
		if (page->mapping && trylock_page(page)) {
			lru_add_drain();  /* push cached pages to LRU */
			/*
			 * Because we lock page here, and migration is
			 * blocked by the pte's page reference, and we
			 * know the page is still mapped, we don't even
			 * need to check for file-cache page truncation.
			 */
			mlock_vma_page(page);
			unlock_page(page);
		}
	}
317
out:
318 319 320 321 322
	pte_unmap_unlock(ptep, ptl);
	return page;
no_page:
	pte_unmap_unlock(ptep, ptl);
	if (!pte_none(pte))
323 324 325 326
		return NULL;
	return no_page_table(vma, flags);
}

327 328
static struct page *follow_pmd_mask(struct vm_area_struct *vma,
				    unsigned long address, pud_t *pudp,
329 330
				    unsigned int flags,
				    struct follow_page_context *ctx)
331
{
332
	pmd_t *pmd, pmdval;
333 334 335 336
	spinlock_t *ptl;
	struct page *page;
	struct mm_struct *mm = vma->vm_mm;

337
	pmd = pmd_offset(pudp, address);
338 339 340 341 342 343
	/*
	 * The READ_ONCE() will stabilize the pmdval in a register or
	 * on the stack so that it will stop changing under the code.
	 */
	pmdval = READ_ONCE(*pmd);
	if (pmd_none(pmdval))
344
		return no_page_table(vma, flags);
345
	if (pmd_huge(pmdval) && is_vm_hugetlb_page(vma)) {
346 347 348 349
		page = follow_huge_pmd(mm, address, pmd, flags);
		if (page)
			return page;
		return no_page_table(vma, flags);
350
	}
351
	if (is_hugepd(__hugepd(pmd_val(pmdval)))) {
352
		page = follow_huge_pd(vma, address,
353
				      __hugepd(pmd_val(pmdval)), flags,
354 355 356 357 358
				      PMD_SHIFT);
		if (page)
			return page;
		return no_page_table(vma, flags);
	}
359
retry:
360
	if (!pmd_present(pmdval)) {
361 362 363
		if (likely(!(flags & FOLL_MIGRATION)))
			return no_page_table(vma, flags);
		VM_BUG_ON(thp_migration_supported() &&
364 365
				  !is_pmd_migration_entry(pmdval));
		if (is_pmd_migration_entry(pmdval))
366
			pmd_migration_entry_wait(mm, pmd);
367 368 369 370 371 372 373
		pmdval = READ_ONCE(*pmd);
		/*
		 * MADV_DONTNEED may convert the pmd to null because
		 * mmap_sem is held in read mode
		 */
		if (pmd_none(pmdval))
			return no_page_table(vma, flags);
374 375
		goto retry;
	}
376
	if (pmd_devmap(pmdval)) {
377
		ptl = pmd_lock(mm, pmd);
378
		page = follow_devmap_pmd(vma, address, pmd, flags, &ctx->pgmap);
379 380 381 382
		spin_unlock(ptl);
		if (page)
			return page;
	}
383
	if (likely(!pmd_trans_huge(pmdval)))
384
		return follow_page_pte(vma, address, pmd, flags, &ctx->pgmap);
385

386
	if ((flags & FOLL_NUMA) && pmd_protnone(pmdval))
387 388
		return no_page_table(vma, flags);

389
retry_locked:
390
	ptl = pmd_lock(mm, pmd);
391 392 393 394
	if (unlikely(pmd_none(*pmd))) {
		spin_unlock(ptl);
		return no_page_table(vma, flags);
	}
395 396 397 398 399 400 401
	if (unlikely(!pmd_present(*pmd))) {
		spin_unlock(ptl);
		if (likely(!(flags & FOLL_MIGRATION)))
			return no_page_table(vma, flags);
		pmd_migration_entry_wait(mm, pmd);
		goto retry_locked;
	}
402 403
	if (unlikely(!pmd_trans_huge(*pmd))) {
		spin_unlock(ptl);
404
		return follow_page_pte(vma, address, pmd, flags, &ctx->pgmap);
405
	}
Song Liu's avatar
Song Liu committed
406
	if (flags & (FOLL_SPLIT | FOLL_SPLIT_PMD)) {
407 408 409 410 411
		int ret;
		page = pmd_page(*pmd);
		if (is_huge_zero_page(page)) {
			spin_unlock(ptl);
			ret = 0;
412
			split_huge_pmd(vma, pmd, address);
413 414
			if (pmd_trans_unstable(pmd))
				ret = -EBUSY;
Song Liu's avatar
Song Liu committed
415
		} else if (flags & FOLL_SPLIT) {
416 417 418 419
			if (unlikely(!try_get_page(page))) {
				spin_unlock(ptl);
				return ERR_PTR(-ENOMEM);
			}
420
			spin_unlock(ptl);
421 422 423 424
			lock_page(page);
			ret = split_huge_page(page);
			unlock_page(page);
			put_page(page);
425 426
			if (pmd_none(*pmd))
				return no_page_table(vma, flags);
Song Liu's avatar
Song Liu committed
427 428 429 430
		} else {  /* flags & FOLL_SPLIT_PMD */
			spin_unlock(ptl);
			split_huge_pmd(vma, pmd, address);
			ret = pte_alloc(mm, pmd) ? -ENOMEM : 0;
431 432 433
		}

		return ret ? ERR_PTR(ret) :
434
			follow_page_pte(vma, address, pmd, flags, &ctx->pgmap);
435
	}
436 437
	page = follow_trans_huge_pmd(vma, address, pmd, flags);
	spin_unlock(ptl);
438
	ctx->page_mask = HPAGE_PMD_NR - 1;
439
	return page;
440 441
}

442 443
static struct page *follow_pud_mask(struct vm_area_struct *vma,
				    unsigned long address, p4d_t *p4dp,
444 445
				    unsigned int flags,
				    struct follow_page_context *ctx)
446 447 448 449 450 451 452 453 454
{
	pud_t *pud;
	spinlock_t *ptl;
	struct page *page;
	struct mm_struct *mm = vma->vm_mm;

	pud = pud_offset(p4dp, address);
	if (pud_none(*pud))
		return no_page_table(vma, flags);
455
	if (pud_huge(*pud) && is_vm_hugetlb_page(vma)) {
456 457 458 459 460
		page = follow_huge_pud(mm, address, pud, flags);
		if (page)
			return page;
		return no_page_table(vma, flags);
	}
461 462 463 464 465 466 467 468
	if (is_hugepd(__hugepd(pud_val(*pud)))) {
		page = follow_huge_pd(vma, address,
				      __hugepd(pud_val(*pud)), flags,
				      PUD_SHIFT);
		if (page)
			return page;
		return no_page_table(vma, flags);
	}
469 470
	if (pud_devmap(*pud)) {
		ptl = pud_lock(mm, pud);
471
		page = follow_devmap_pud(vma, address, pud, flags, &ctx->pgmap);
472 473 474 475 476 477 478
		spin_unlock(ptl);
		if (page)
			return page;
	}
	if (unlikely(pud_bad(*pud)))
		return no_page_table(vma, flags);

479
	return follow_pmd_mask(vma, address, pud, flags, ctx);
480 481 482 483
}

static struct page *follow_p4d_mask(struct vm_area_struct *vma,
				    unsigned long address, pgd_t *pgdp,
484 485
				    unsigned int flags,
				    struct follow_page_context *ctx)
486 487
{
	p4d_t *p4d;
488
	struct page *page;
489 490 491 492 493 494 495 496

	p4d = p4d_offset(pgdp, address);
	if (p4d_none(*p4d))
		return no_page_table(vma, flags);
	BUILD_BUG_ON(p4d_huge(*p4d));
	if (unlikely(p4d_bad(*p4d)))
		return no_page_table(vma, flags);

497 498 499 500 501 502 503 504
	if (is_hugepd(__hugepd(p4d_val(*p4d)))) {
		page = follow_huge_pd(vma, address,
				      __hugepd(p4d_val(*p4d)), flags,
				      P4D_SHIFT);
		if (page)
			return page;
		return no_page_table(vma, flags);
	}
505
	return follow_pud_mask(vma, address, p4d, flags, ctx);
506 507 508 509 510 511 512
}

/**
 * follow_page_mask - look up a page descriptor from a user-virtual address
 * @vma: vm_area_struct mapping @address
 * @address: virtual address to look up
 * @flags: flags modifying lookup behaviour
513 514
 * @ctx: contains dev_pagemap for %ZONE_DEVICE memory pinning and a
 *       pointer to output page_mask
515 516 517
 *
 * @flags can have FOLL_ flags set, defined in <linux/mm.h>
 *
518 519 520 521 522 523
 * When getting pages from ZONE_DEVICE memory, the @ctx->pgmap caches
 * the device's dev_pagemap metadata to avoid repeating expensive lookups.
 *
 * On output, the @ctx->page_mask is set according to the size of the page.
 *
 * Return: the mapped (struct page *), %NULL if no mapping exists, or
524 525 526
 * an error pointer if there is a mapping to something not represented
 * by a page descriptor (see also vm_normal_page()).
 */
527
static struct page *follow_page_mask(struct vm_area_struct *vma,
528
			      unsigned long address, unsigned int flags,
529
			      struct follow_page_context *ctx)
530 531 532 533 534
{
	pgd_t *pgd;
	struct page *page;
	struct mm_struct *mm = vma->vm_mm;

535
	ctx->page_mask = 0;
536 537 538 539 540 541 542 543 544 545 546 547 548

	/* make this handle hugepd */
	page = follow_huge_addr(mm, address, flags & FOLL_WRITE);
	if (!IS_ERR(page)) {
		BUG_ON(flags & FOLL_GET);
		return page;
	}

	pgd = pgd_offset(mm, address);

	if (pgd_none(*pgd) || unlikely(pgd_bad(*pgd)))
		return no_page_table(vma, flags);

549 550 551 552 553 554
	if (pgd_huge(*pgd)) {
		page = follow_huge_pgd(mm, address, pgd, flags);
		if (page)
			return page;
		return no_page_table(vma, flags);
	}
555 556 557 558 559 560 561 562
	if (is_hugepd(__hugepd(pgd_val(*pgd)))) {
		page = follow_huge_pd(vma, address,
				      __hugepd(pgd_val(*pgd)), flags,
				      PGDIR_SHIFT);
		if (page)
			return page;
		return no_page_table(vma, flags);
	}
563

564 565 566 567 568 569 570 571 572 573 574 575 576
	return follow_p4d_mask(vma, address, pgd, flags, ctx);
}

struct page *follow_page(struct vm_area_struct *vma, unsigned long address,
			 unsigned int foll_flags)
{
	struct follow_page_context ctx = { NULL };
	struct page *page;

	page = follow_page_mask(vma, address, foll_flags, &ctx);
	if (ctx.pgmap)
		put_dev_pagemap(ctx.pgmap);
	return page;
577 578
}

579 580 581 582 583
static int get_gate_page(struct mm_struct *mm, unsigned long address,
		unsigned int gup_flags, struct vm_area_struct **vma,
		struct page **page)
{
	pgd_t *pgd;
584
	p4d_t *p4d;
585 586 587 588 589 590 591 592 593 594 595 596
	pud_t *pud;
	pmd_t *pmd;
	pte_t *pte;
	int ret = -EFAULT;

	/* user gate pages are read-only */
	if (gup_flags & FOLL_WRITE)
		return -EFAULT;
	if (address > TASK_SIZE)
		pgd = pgd_offset_k(address);
	else
		pgd = pgd_offset_gate(mm, address);
597 598
	if (pgd_none(*pgd))
		return -EFAULT;
599
	p4d = p4d_offset(pgd, address);
600 601
	if (p4d_none(*p4d))
		return -EFAULT;
602
	pud = pud_offset(p4d, address);
603 604
	if (pud_none(*pud))
		return -EFAULT;
605
	pmd = pmd_offset(pud, address);
606
	if (!pmd_present(*pmd))
607 608 609 610 611 612 613 614 615 616 617 618 619 620
		return -EFAULT;
	VM_BUG_ON(pmd_trans_huge(*pmd));
	pte = pte_offset_map(pmd, address);
	if (pte_none(*pte))
		goto unmap;
	*vma = get_gate_vma(mm);
	if (!page)
		goto out;
	*page = vm_normal_page(*vma, address, *pte);
	if (!*page) {
		if ((gup_flags & FOLL_DUMP) || !is_zero_pfn(pte_pfn(*pte)))
			goto unmap;
		*page = pte_page(*pte);
	}
621 622 623 624
	if (unlikely(!try_get_page(*page))) {
		ret = -ENOMEM;
		goto unmap;
	}
625 626 627 628 629 630 631
out:
	ret = 0;
unmap:
	pte_unmap(pte);
	return ret;
}

632 633 634 635 636
/*
 * mmap_sem must be held on entry.  If @nonblocking != NULL and
 * *@flags does not include FOLL_NOWAIT, the mmap_sem may be released.
 * If it is, *@nonblocking will be set to 0 and -EBUSY returned.
 */
637 638 639 640
static int faultin_page(struct task_struct *tsk, struct vm_area_struct *vma,
		unsigned long address, unsigned int *flags, int *nonblocking)
{
	unsigned int fault_flags = 0;
641
	vm_fault_t ret;
642

Eric B Munson's avatar
Eric B Munson committed
643 644 645
	/* mlock all present pages, but do not fault in new pages */
	if ((*flags & (FOLL_POPULATE | FOLL_MLOCK)) == FOLL_MLOCK)
		return -ENOENT;
646 647
	if (*flags & FOLL_WRITE)
		fault_flags |= FAULT_FLAG_WRITE;
648 649
	if (*flags & FOLL_REMOTE)
		fault_flags |= FAULT_FLAG_REMOTE;
650 651 652 653
	if (nonblocking)
		fault_flags |= FAULT_FLAG_ALLOW_RETRY;
	if (*flags & FOLL_NOWAIT)
		fault_flags |= FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_RETRY_NOWAIT;
654 655 656 657
	if (*flags & FOLL_TRIED) {
		VM_WARN_ON_ONCE(fault_flags & FAULT_FLAG_ALLOW_RETRY);
		fault_flags |= FAULT_FLAG_TRIED;
	}
658

659
	ret = handle_mm_fault(vma, address, fault_flags);
660
	if (ret & VM_FAULT_ERROR) {
661 662 663 664
		int err = vm_fault_to_errno(ret, *flags);

		if (err)
			return err;
665 666 667 668 669 670 671 672 673 674 675
		BUG();
	}

	if (tsk) {
		if (ret & VM_FAULT_MAJOR)
			tsk->maj_flt++;
		else
			tsk->min_flt++;
	}

	if (ret & VM_FAULT_RETRY) {
676
		if (nonblocking && !(fault_flags & FAULT_FLAG_RETRY_NOWAIT))
677 678 679 680 681 682 683 684 685 686 687 688 689 690
			*nonblocking = 0;
		return -EBUSY;
	}

	/*
	 * The VM_FAULT_WRITE bit tells us that do_wp_page has broken COW when
	 * necessary, even if maybe_mkwrite decided not to set pte_write. We
	 * can thus safely do subsequent page lookups as if they were reads.
	 * But only do so when looping for pte_write is futile: in some cases
	 * userspace may also be wanting to write to the gotten user page,
	 * which a read fault here might prevent (a readonly page might get
	 * reCOWed by userspace write).
	 */
	if ((ret & VM_FAULT_WRITE) && !(vma->vm_flags & VM_WRITE))
691
		*flags |= FOLL_COW;
692 693 694
	return 0;
}

695 696 697
static int check_vma_flags(struct vm_area_struct *vma, unsigned long gup_flags)
{
	vm_flags_t vm_flags = vma->vm_flags;
698 699
	int write = (gup_flags & FOLL_WRITE);
	int foreign = (gup_flags & FOLL_REMOTE);
700 701 702 703

	if (vm_flags & (VM_IO | VM_PFNMAP))
		return -EFAULT;

704 705 706
	if (gup_flags & FOLL_ANON && !vma_is_anonymous(vma))
		return -EFAULT;

707
	if (write) {
708 709 710 711 712 713 714 715 716 717 718 719
		if (!(vm_flags & VM_WRITE)) {
			if (!(gup_flags & FOLL_FORCE))
				return -EFAULT;
			/*
			 * We used to let the write,force case do COW in a
			 * VM_MAYWRITE VM_SHARED !VM_WRITE vma, so ptrace could
			 * set a breakpoint in a read-only mapping of an
			 * executable, without corrupting the file (yet only
			 * when that file had been opened for writing!).
			 * Anon pages in shared mappings are surprising: now
			 * just reject it.
			 */
720
			if (!is_cow_mapping(vm_flags))
721 722 723 724 725 726 727 728 729 730 731 732
				return -EFAULT;
		}
	} else if (!(vm_flags & VM_READ)) {
		if (!(gup_flags & FOLL_FORCE))
			return -EFAULT;
		/*
		 * Is there actually any vma we can reach here which does not
		 * have VM_MAYREAD set?
		 */
		if (!(vm_flags & VM_MAYREAD))
			return -EFAULT;
	}
733 734 735 736 737
	/*
	 * gups are always data accesses, not instruction
	 * fetches, so execute=false here
	 */
	if (!arch_vma_access_permitted(vma, write, false, foreign))
738
		return -EFAULT;
739 740 741
	return 0;
}

742 743 744 745 746 747 748 749 750 751 752 753 754 755
/**
 * __get_user_pages() - pin user pages in memory
 * @tsk:	task_struct of target task
 * @mm:		mm_struct of target mm
 * @start:	starting user address
 * @nr_pages:	number of pages from start to pin
 * @gup_flags:	flags modifying pin behaviour
 * @pages:	array that receives pointers to the pages pinned.
 *		Should be at least nr_pages long. Or NULL, if caller
 *		only intends to ensure the pages are faulted in.
 * @vmas:	array of pointers to vmas corresponding to each page.
 *		Or NULL if the caller does not require them.
 * @nonblocking: whether waiting for disk IO or mmap_sem contention
 *
756 757 758 759 760 761 762 763 764 765 766
 * Returns either number of pages pinned (which may be less than the
 * number requested), or an error. Details about the return value:
 *
 * -- If nr_pages is 0, returns 0.
 * -- If nr_pages is >0, but no pages were pinned, returns -errno.
 * -- If nr_pages is >0, and some pages were pinned, returns the number of
 *    pages pinned. Again, this may be less than nr_pages.
 *
 * The caller is responsible for releasing returned @pages, via put_page().
 *
 * @vmas are valid only as long as mmap_sem is held.
767
 *
768
 * Must be called with mmap_sem held.  It may be released.  See below.
769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790
 *
 * __get_user_pages walks a process's page tables and takes a reference to
 * each struct page that each user address corresponds to at a given
 * instant. That is, it takes the page that would be accessed if a user
 * thread accesses the given user virtual address at that instant.
 *
 * This does not guarantee that the page exists in the user mappings when
 * __get_user_pages returns, and there may even be a completely different
 * page there in some cases (eg. if mmapped pagecache has been invalidated
 * and subsequently re faulted). However it does guarantee that the page
 * won't be freed completely. And mostly callers simply care that the page
 * contains data that was valid *at some point in time*. Typically, an IO
 * or similar operation cannot guarantee anything stronger anyway because
 * locks can't be held over the syscall boundary.
 *
 * If @gup_flags & FOLL_WRITE == 0, the page must not be written to. If
 * the page is written to, set_page_dirty (or set_page_dirty_lock, as
 * appropriate) must be called after the page is finished with, and
 * before put_page is called.
 *
 * If @nonblocking != NULL, __get_user_pages will not wait for disk IO
 * or mmap_sem contention, and if waiting is needed to pin all pages,
791 792 793 794 795 796 797 798
 * *@nonblocking will be set to 0.  Further, if @gup_flags does not
 * include FOLL_NOWAIT, the mmap_sem will be released via up_read() in
 * this case.
 *
 * A caller using such a combination of @nonblocking and @gup_flags
 * must therefore hold the mmap_sem for reading only, and recognize
 * when it's been released.  Otherwise, it must be held for either
 * reading or writing and will not be released.
799 800 801 802 803
 *
 * In most cases, get_user_pages or get_user_pages_fast should be used
 * instead of __get_user_pages. __get_user_pages should be used only if
 * you need some special @gup_flags.
 */
804
static long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
805 806 807 808
		unsigned long start, unsigned long nr_pages,
		unsigned int gup_flags, struct page **pages,
		struct vm_area_struct **vmas, int *nonblocking)
{
809
	long ret = 0, i = 0;
810
	struct vm_area_struct *vma = NULL;
811
	struct follow_page_context ctx = { NULL };
812 813 814 815

	if (!nr_pages)
		return 0;

816 817
	start = untagged_addr(start);

818
	VM_BUG_ON(!!pages != !!(gup_flags & (FOLL_GET | FOLL_PIN)));
819 820 821 822 823 824 825 826 827 828

	/*
	 * If FOLL_FORCE is set then do not force a full fault as the hinting
	 * fault information is unrelated to the reference behaviour of a task
	 * using the address space
	 */
	if (!(gup_flags & FOLL_FORCE))
		gup_flags |= FOLL_NUMA;

	do {
829 830 831 832 833 834 835 836 837 838 839 840
		struct page *page;
		unsigned int foll_flags = gup_flags;
		unsigned int page_increm;

		/* first iteration or cross vma bound */
		if (!vma || start >= vma->vm_end) {
			vma = find_extend_vma(mm, start);
			if (!vma && in_gate_area(mm, start)) {
				ret = get_gate_page(mm, start & PAGE_MASK,
						gup_flags, &vma,
						pages ? &pages[i] : NULL);
				if (ret)
841
					goto out;
842
				ctx.page_mask = 0;
843 844
				goto next_page;
			}
845

846 847 848 849
			if (!vma || check_vma_flags(vma, gup_flags)) {
				ret = -EFAULT;
				goto out;
			}
850 851 852
			if (is_vm_hugetlb_page(vma)) {
				i = follow_hugetlb_page(mm, vma, pages, vmas,
						&start, &nr_pages, i,
853
						gup_flags, nonblocking);
854
				continue;
855
			}
856 857 858 859 860 861
		}
retry:
		/*
		 * If we have a pending SIGKILL, don't keep faulting pages and
		 * potentially allocating memory.
		 */
862
		if (fatal_signal_pending(current)) {
863 864 865
			ret = -ERESTARTSYS;
			goto out;
		}
866
		cond_resched();
867 868

		page = follow_page_mask(vma, start, foll_flags, &ctx);
869 870 871 872 873 874
		if (!page) {
			ret = faultin_page(tsk, vma, start, &foll_flags,
					nonblocking);
			switch (ret) {
			case 0:
				goto retry;
875 876 877
			case -EBUSY:
				ret = 0;
				/* FALLTHRU */
878 879 880
			case -EFAULT:
			case -ENOMEM:
			case -EHWPOISON:
881
				goto out;
882 883
			case -ENOENT:
				goto next_page;
884
			}
885
			BUG();
886 887 888 889 890 891 892
		} else if (PTR_ERR(page) == -EEXIST) {
			/*
			 * Proper page table entry exists, but no corresponding
			 * struct page.
			 */
			goto next_page;
		} else if (IS_ERR(page)) {
893 894
			ret = PTR_ERR(page);
			goto out;
895
		}
896 897 898 899
		if (pages) {
			pages[i] = page;
			flush_anon_page(vma, page, start);
			flush_dcache_page(page);
900
			ctx.page_mask = 0;
901 902
		}
next_page:
903 904
		if (vmas) {
			vmas[i] = vma;
905
			ctx.page_mask = 0;
906
		}
907
		page_increm = 1 + (~(start >> PAGE_SHIFT) & ctx.page_mask);
908 909 910 911 912
		if (page_increm > nr_pages)
			page_increm = nr_pages;
		i += page_increm;
		start += page_increm * PAGE_SIZE;
		nr_pages -= page_increm;
913
	} while (nr_pages);
914 915 916 917
out:
	if (ctx.pgmap)
		put_dev_pagemap(ctx.pgmap);
	return i ? i : ret;
918 919
}

920 921
static bool vma_permits_fault(struct vm_area_struct *vma,
			      unsigned int fault_flags)
922
{
923 924
	bool write   = !!(fault_flags & FAULT_FLAG_WRITE);
	bool foreign = !!(fault_flags & FAULT_FLAG_REMOTE);
925
	vm_flags_t vm_flags = write ? VM_WRITE : VM_READ;
926 927 928 929

	if (!(vm_flags & vma->vm_flags))
		return false;

930 931
	/*
	 * The architecture might have a hardware protection
932
	 * mechanism other than read/write that can deny access.
933 934 935
	 *
	 * gup always represents data access, not instruction
	 * fetches, so execute=false here:
936
	 */
937
	if (!arch_vma_access_permitted(vma, write, false, foreign))
938 939
		return false;

940 941 942
	return true;
}

943 944 945 946 947 948 949
/*
 * fixup_user_fault() - manually resolve a user page fault
 * @tsk:	the task_struct to use for page fault accounting, or
 *		NULL if faults are not to be recorded.
 * @mm:		mm_struct of target mm
 * @address:	user address
 * @fault_flags:flags to pass down to handle_mm_fault()
950 951
 * @unlocked:	did we unlock the mmap_sem while retrying, maybe NULL if caller
 *		does not allow retry
952 953 954 955 956 957 958 959 960 961 962
 *
 * This is meant to be called in the specific scenario where for locking reasons
 * we try to access user memory in atomic context (within a pagefault_disable()
 * section), this returns -EFAULT, and we want to resolve the user fault before
 * trying again.
 *
 * Typically this is meant to be used by the futex code.
 *
 * The main difference with get_user_pages() is that this function will
 * unconditionally call handle_mm_fault() which will in turn perform all the
 * necessary SW fixup of the dirty and young bits in the PTE, while
963
 * get_user_pages() only guarantees to update these in the struct page.
964 965 966 967 968 969
 *
 * This is important for some architectures where those bits also gate the
 * access permission to the page because they are maintained in software.  On
 * such architectures, gup() will not be enough to make a subsequent access
 * succeed.
 *
970 971
 * This function will not return with an unlocked mmap_sem. So it has not the
 * same semantics wrt the @mm->mmap_sem as does filemap_fault().
972 973
 */
int fixup_user_fault(struct task_struct *tsk, struct mm_struct *mm,
974 975
		     unsigned long address, unsigned int fault_flags,
		     bool *unlocked)
976 977
{
	struct vm_area_struct *vma;
978
	vm_fault_t ret, major = 0;
979

980 981
	address = untagged_addr(address);

982 983
	if (unlocked)
		fault_flags |= FAULT_FLAG_ALLOW_RETRY;
984

985
retry:
986 987 988 989
	vma = find_extend_vma(mm, address);
	if (!vma || address < vma->vm_start)
		return -EFAULT;

990
	if (!vma_permits_fault(vma, fault_flags))
991 992
		return -EFAULT;

993
	ret = handle_mm_fault(vma, address, fault_flags);
994
	major |= ret & VM_FAULT_MAJOR;
995
	if (ret & VM_FAULT_ERROR) {
996 997 998 999
		int err = vm_fault_to_errno(ret, 0);

		if (err)
			return err;
1000 1001
		BUG();
	}
1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012

	if (ret & VM_FAULT_RETRY) {
		down_read(&mm->mmap_sem);
		if (!(fault_flags & FAULT_FLAG_TRIED)) {
			*unlocked = true;
			fault_flags &= ~FAULT_FLAG_ALLOW_RETRY;
			fault_flags |= FAULT_FLAG_TRIED;
			goto retry;
		}
	}

1013
	if (tsk) {
1014
		if (major)
1015 1016 1017 1018 1019 1020
			tsk->maj_flt++;
		else
			tsk->min_flt++;
	}
	return 0;
}
1021
EXPORT_SYMBOL_GPL(fixup_user_fault);
1022

1023 1024 1025 1026 1027 1028
static __always_inline long __get_user_pages_locked(struct task_struct *tsk,
						struct mm_struct *mm,
						unsigned long start,
						unsigned long nr_pages,
						struct page **pages,
						struct vm_area_struct **vmas,
1029
						int *locked,
1030
						unsigned int flags)
1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041
{
	long ret, pages_done;
	bool lock_dropped;

	if (locked) {
		/* if VM_FAULT_RETRY can be returned, vmas become invalid */
		BUG_ON(vmas);
		/* check caller initialized locked */
		BUG_ON(*locked != 1);
	}

1042 1043 1044 1045 1046 1047 1048 1049 1050 1051
	/*
	 * FOLL_PIN and FOLL_GET are mutually exclusive. Traditional behavior
	 * is to set FOLL_GET if the caller wants pages[] filled in (but has
	 * carelessly failed to specify FOLL_GET), so keep doing that, but only
	 * for FOLL_GET, not for the newer FOLL_PIN.
	 *
	 * FOLL_PIN always expects pages to be non-null, but no need to assert
	 * that here, as any failures will be obvious enough.
	 */
	if (pages && !(flags & FOLL_PIN))
1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075
		flags |= FOLL_GET;

	pages_done = 0;
	lock_dropped = false;
	for (;;) {
		ret = __get_user_pages(tsk, mm, start, nr_pages, flags, pages,
				       vmas, locked);
		if (!locked)
			/* VM_FAULT_RETRY couldn't trigger, bypass */
			return ret;

		/* VM_FAULT_RETRY cannot return errors */
		if (!*locked) {
			BUG_ON(ret < 0);
			BUG_ON(ret >= nr_pages);
		}

		if (ret > 0) {
			nr_pages -= ret;
			pages_done += ret;
			if (!nr_pages)
				break;
		}
		if (*locked) {
1076 1077 1078 1079
			/*
			 * VM_FAULT_RETRY didn't trigger or it was a
			 * FOLL_NOWAIT.
			 */
1080 1081 1082 1083
			if (!pages_done)
				pages_done = ret;
			break;
		}
1084 1085 1086 1087 1088 1089
		/*
		 * VM_FAULT_RETRY triggered, so seek to the faulting offset.
		 * For the prefault case (!pages) we only update counts.
		 */
		if (likely(pages))
			pages += ret;
1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111
		start += ret << PAGE_SHIFT;

		/*
		 * Repeat on the address that fired VM_FAULT_RETRY
		 * without FAULT_FLAG_ALLOW_RETRY but with
		 * FAULT_FLAG_TRIED.
		 */
		*locked = 1;
		lock_dropped = true;
		down_read(&mm->mmap_sem);
		ret = __get_user_pages(tsk, mm, start, 1, flags | FOLL_TRIED,
				       pages, NULL, NULL);
		if (ret != 1) {
			BUG_ON(ret > 1);
			if (!pages_done)
				pages_done = ret;
			break;
		}
		nr_pages--;
		pages_done++;
		if (!nr_pages)
			break;
1112 1113
		if (likely(pages))
			pages++;
1114 1115
		start += PAGE_SIZE;
	}
1116
	if (lock_dropped && *locked) {
1117 1118 1119 1120 1121 1122 1123 1124 1125 1126
		/*
		 * We must let the caller know we temporarily dropped the lock
		 * and so the critical section protected by it was lost.
		 */
		up_read(&mm->mmap_sem);
		*locked = 0;
	}
	return pages_done;
}

1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 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 1177 1178 1179