gup.c 60.8 KB
Newer Older
1 2 3 4 5 6
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/err.h>
#include <linux/spinlock.h>

#include <linux/mm.h>
7
#include <linux/memremap.h>
8 9 10 11 12
#include <linux/pagemap.h>
#include <linux/rmap.h>
#include <linux/swap.h>
#include <linux/swapops.h>

13
#include <linux/sched/signal.h>
14
#include <linux/rwsem.h>
15
#include <linux/hugetlb.h>
16 17 18
#include <linux/migrate.h>
#include <linux/mm_inline.h>
#include <linux/sched/mm.h>
19

20
#include <asm/mmu_context.h>
21
#include <asm/pgtable.h>
22
#include <asm/tlbflush.h>
23

24 25
#include "internal.h"

26 27 28 29 30
struct follow_page_context {
	struct dev_pagemap *pgmap;
	unsigned int page_mask;
};

31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135
typedef int (*set_dirty_func_t)(struct page *page);

static void __put_user_pages_dirty(struct page **pages,
				   unsigned long npages,
				   set_dirty_func_t sdf)
{
	unsigned long index;

	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 sdf(). 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 sdf().
		 * 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))
			sdf(page);

		put_user_page(page);
	}
}

/**
 * put_user_pages_dirty() - release and dirty an array of gup-pinned pages
 * @pages:  array of pages to be marked dirty and released.
 * @npages: number of pages in the @pages array.
 *
 * "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
 * compound page) dirty, if it was previously listed as clean. Then, release
 * the page using put_user_page().
 *
 * Please see the put_user_page() documentation for details.
 *
 * set_page_dirty(), which does not lock the page, is used here.
 * Therefore, it is the caller's responsibility to ensure that this is
 * safe. If not, then put_user_pages_dirty_lock() should be called instead.
 *
 */
void put_user_pages_dirty(struct page **pages, unsigned long npages)
{
	__put_user_pages_dirty(pages, npages, set_page_dirty);
}
EXPORT_SYMBOL(put_user_pages_dirty);

/**
 * put_user_pages_dirty_lock() - release and dirty 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, make that page (or its head page, if a
 * compound page) dirty, if it was previously listed as clean. Then, release
 * the page using put_user_page().
 *
 * Please see the put_user_page() documentation for details.
 *
 * This is just like put_user_pages_dirty(), except that it invokes
 * set_page_dirty_lock(), instead of set_page_dirty().
 *
 */
void put_user_pages_dirty_lock(struct page **pages, unsigned long npages)
{
	__put_user_pages_dirty(pages, npages, set_page_dirty_lock);
}
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);

136 137
static struct page *no_page_table(struct vm_area_struct *vma,
		unsigned int flags)
138
{
139 140 141 142 143 144 145 146 147 148 149 150
	/*
	 * 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;
}
151

152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175
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;
}

176 177 178 179 180 181
/*
 * 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)
{
182
	return pte_write(pte) ||
183 184 185
		((flags & FOLL_FORCE) && (flags & FOLL_COW) && pte_dirty(pte));
}

186
static struct page *follow_page_pte(struct vm_area_struct *vma,
187 188
		unsigned long address, pmd_t *pmd, unsigned int flags,
		struct dev_pagemap **pgmap)
189 190 191 192 193
{
	struct mm_struct *mm = vma->vm_mm;
	struct page *page;
	spinlock_t *ptl;
	pte_t *ptep, pte;
194

195
retry:
196
	if (unlikely(pmd_bad(*pmd)))
197
		return no_page_table(vma, flags);
198 199 200 201 202 203 204 205 206 207 208 209

	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;
210
		if (pte_none(pte))
211 212 213 214 215 216
			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);
217
		goto retry;
218
	}
219
	if ((flags & FOLL_NUMA) && pte_protnone(pte))
220
		goto no_page;
221
	if ((flags & FOLL_WRITE) && !can_follow_write_pte(pte, flags)) {
222 223 224
		pte_unmap_unlock(ptep, ptl);
		return NULL;
	}
225 226

	page = vm_normal_page(vma, address, pte);
227 228 229 230 231
	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.
		 */
232 233
		*pgmap = get_dev_pagemap(pte_pfn(pte), *pgmap);
		if (*pgmap)
234 235 236 237
			page = pte_page(pte);
		else
			goto no_page;
	} else if (unlikely(!page)) {
238 239 240 241 242 243 244 245 246 247 248 249 250 251 252
		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;
		}
253 254
	}

255 256 257 258 259 260 261 262 263 264 265 266 267
	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;
	}

268 269 270 271 272 273
	if (flags & FOLL_GET) {
		if (unlikely(!try_get_page(page))) {
			page = ERR_PTR(-ENOMEM);
			goto out;
		}
	}
274 275 276 277 278 279 280 281 282 283 284
	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
285
	if ((flags & FOLL_MLOCK) && (vma->vm_flags & VM_LOCKED)) {
286 287 288 289
		/* Do not mlock pte-mapped THP */
		if (PageTransCompound(page))
			goto out;

290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310
		/*
		 * 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);
		}
	}
311
out:
312 313 314 315 316
	pte_unmap_unlock(ptep, ptl);
	return page;
no_page:
	pte_unmap_unlock(ptep, ptl);
	if (!pte_none(pte))
317 318 319 320
		return NULL;
	return no_page_table(vma, flags);
}

321 322
static struct page *follow_pmd_mask(struct vm_area_struct *vma,
				    unsigned long address, pud_t *pudp,
323 324
				    unsigned int flags,
				    struct follow_page_context *ctx)
325
{
326
	pmd_t *pmd, pmdval;
327 328 329 330
	spinlock_t *ptl;
	struct page *page;
	struct mm_struct *mm = vma->vm_mm;

331
	pmd = pmd_offset(pudp, address);
332 333 334 335 336 337
	/*
	 * 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))
338
		return no_page_table(vma, flags);
339
	if (pmd_huge(pmdval) && vma->vm_flags & VM_HUGETLB) {
340 341 342 343
		page = follow_huge_pmd(mm, address, pmd, flags);
		if (page)
			return page;
		return no_page_table(vma, flags);
344
	}
345
	if (is_hugepd(__hugepd(pmd_val(pmdval)))) {
346
		page = follow_huge_pd(vma, address,
347
				      __hugepd(pmd_val(pmdval)), flags,
348 349 350 351 352
				      PMD_SHIFT);
		if (page)
			return page;
		return no_page_table(vma, flags);
	}
353
retry:
354
	if (!pmd_present(pmdval)) {
355 356 357
		if (likely(!(flags & FOLL_MIGRATION)))
			return no_page_table(vma, flags);
		VM_BUG_ON(thp_migration_supported() &&
358 359
				  !is_pmd_migration_entry(pmdval));
		if (is_pmd_migration_entry(pmdval))
360
			pmd_migration_entry_wait(mm, pmd);
361 362 363 364 365 366 367
		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);
368 369
		goto retry;
	}
370
	if (pmd_devmap(pmdval)) {
371
		ptl = pmd_lock(mm, pmd);
372
		page = follow_devmap_pmd(vma, address, pmd, flags, &ctx->pgmap);
373 374 375 376
		spin_unlock(ptl);
		if (page)
			return page;
	}
377
	if (likely(!pmd_trans_huge(pmdval)))
378
		return follow_page_pte(vma, address, pmd, flags, &ctx->pgmap);
379

380
	if ((flags & FOLL_NUMA) && pmd_protnone(pmdval))
381 382
		return no_page_table(vma, flags);

383
retry_locked:
384
	ptl = pmd_lock(mm, pmd);
385 386 387 388
	if (unlikely(pmd_none(*pmd))) {
		spin_unlock(ptl);
		return no_page_table(vma, flags);
	}
389 390 391 392 393 394 395
	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;
	}
396 397
	if (unlikely(!pmd_trans_huge(*pmd))) {
		spin_unlock(ptl);
398
		return follow_page_pte(vma, address, pmd, flags, &ctx->pgmap);
399 400 401 402 403 404 405
	}
	if (flags & FOLL_SPLIT) {
		int ret;
		page = pmd_page(*pmd);
		if (is_huge_zero_page(page)) {
			spin_unlock(ptl);
			ret = 0;
406
			split_huge_pmd(vma, pmd, address);
407 408
			if (pmd_trans_unstable(pmd))
				ret = -EBUSY;
409
		} else {
410 411 412 413
			if (unlikely(!try_get_page(page))) {
				spin_unlock(ptl);
				return ERR_PTR(-ENOMEM);
			}
414
			spin_unlock(ptl);
415 416 417 418
			lock_page(page);
			ret = split_huge_page(page);
			unlock_page(page);
			put_page(page);
419 420
			if (pmd_none(*pmd))
				return no_page_table(vma, flags);
421 422 423
		}

		return ret ? ERR_PTR(ret) :
424
			follow_page_pte(vma, address, pmd, flags, &ctx->pgmap);
425
	}
426 427
	page = follow_trans_huge_pmd(vma, address, pmd, flags);
	spin_unlock(ptl);
428
	ctx->page_mask = HPAGE_PMD_NR - 1;
429
	return page;
430 431
}

432 433
static struct page *follow_pud_mask(struct vm_area_struct *vma,
				    unsigned long address, p4d_t *p4dp,
434 435
				    unsigned int flags,
				    struct follow_page_context *ctx)
436 437 438 439 440 441 442 443 444 445 446 447 448 449 450
{
	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);
	if (pud_huge(*pud) && vma->vm_flags & VM_HUGETLB) {
		page = follow_huge_pud(mm, address, pud, flags);
		if (page)
			return page;
		return no_page_table(vma, flags);
	}
451 452 453 454 455 456 457 458
	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);
	}
459 460
	if (pud_devmap(*pud)) {
		ptl = pud_lock(mm, pud);
461
		page = follow_devmap_pud(vma, address, pud, flags, &ctx->pgmap);
462 463 464 465 466 467 468
		spin_unlock(ptl);
		if (page)
			return page;
	}
	if (unlikely(pud_bad(*pud)))
		return no_page_table(vma, flags);

469
	return follow_pmd_mask(vma, address, pud, flags, ctx);
470 471 472 473
}

static struct page *follow_p4d_mask(struct vm_area_struct *vma,
				    unsigned long address, pgd_t *pgdp,
474 475
				    unsigned int flags,
				    struct follow_page_context *ctx)
476 477
{
	p4d_t *p4d;
478
	struct page *page;
479 480 481 482 483 484 485 486

	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);

487 488 489 490 491 492 493 494
	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);
	}
495
	return follow_pud_mask(vma, address, p4d, flags, ctx);
496 497 498 499 500 501 502
}

/**
 * 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
503 504
 * @ctx: contains dev_pagemap for %ZONE_DEVICE memory pinning and a
 *       pointer to output page_mask
505 506 507
 *
 * @flags can have FOLL_ flags set, defined in <linux/mm.h>
 *
508 509 510 511 512 513
 * 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
514 515 516 517 518
 * an error pointer if there is a mapping to something not represented
 * by a page descriptor (see also vm_normal_page()).
 */
struct page *follow_page_mask(struct vm_area_struct *vma,
			      unsigned long address, unsigned int flags,
519
			      struct follow_page_context *ctx)
520 521 522 523 524
{
	pgd_t *pgd;
	struct page *page;
	struct mm_struct *mm = vma->vm_mm;

525
	ctx->page_mask = 0;
526 527 528 529 530 531 532 533 534 535 536 537 538

	/* 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);

539 540 541 542 543 544
	if (pgd_huge(*pgd)) {
		page = follow_huge_pgd(mm, address, pgd, flags);
		if (page)
			return page;
		return no_page_table(vma, flags);
	}
545 546 547 548 549 550 551 552
	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);
	}
553

554 555 556 557 558 559 560 561 562 563 564 565 566
	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;
567 568
}

569 570 571 572 573
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;
574
	p4d_t *p4d;
575 576 577 578 579 580 581 582 583 584 585 586 587
	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);
	BUG_ON(pgd_none(*pgd));
588 589 590
	p4d = p4d_offset(pgd, address);
	BUG_ON(p4d_none(*p4d));
	pud = pud_offset(p4d, address);
591 592
	BUG_ON(pud_none(*pud));
	pmd = pmd_offset(pud, address);
593
	if (!pmd_present(*pmd))
594 595 596 597 598 599 600 601 602 603 604 605 606
		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);
607 608 609 610 611 612 613

		/*
		 * This should never happen (a device public page in the gate
		 * area).
		 */
		if (is_device_public_page(*page))
			goto unmap;
614
	}
615 616 617 618
	if (unlikely(!try_get_page(*page))) {
		ret = -ENOMEM;
		goto unmap;
	}
619 620 621 622 623 624 625
out:
	ret = 0;
unmap:
	pte_unmap(pte);
	return ret;
}

626 627 628 629 630
/*
 * 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.
 */
631 632 633 634
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;
635
	vm_fault_t ret;
636

Eric B Munson's avatar
Eric B Munson committed
637 638 639
	/* mlock all present pages, but do not fault in new pages */
	if ((*flags & (FOLL_POPULATE | FOLL_MLOCK)) == FOLL_MLOCK)
		return -ENOENT;
640 641
	if (*flags & FOLL_WRITE)
		fault_flags |= FAULT_FLAG_WRITE;
642 643
	if (*flags & FOLL_REMOTE)
		fault_flags |= FAULT_FLAG_REMOTE;
644 645 646 647
	if (nonblocking)
		fault_flags |= FAULT_FLAG_ALLOW_RETRY;
	if (*flags & FOLL_NOWAIT)
		fault_flags |= FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_RETRY_NOWAIT;
648 649 650 651
	if (*flags & FOLL_TRIED) {
		VM_WARN_ON_ONCE(fault_flags & FAULT_FLAG_ALLOW_RETRY);
		fault_flags |= FAULT_FLAG_TRIED;
	}
652

653
	ret = handle_mm_fault(vma, address, fault_flags);
654
	if (ret & VM_FAULT_ERROR) {
655 656 657 658
		int err = vm_fault_to_errno(ret, *flags);

		if (err)
			return err;
659 660 661 662 663 664 665 666 667 668 669
		BUG();
	}

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

	if (ret & VM_FAULT_RETRY) {
670
		if (nonblocking && !(fault_flags & FAULT_FLAG_RETRY_NOWAIT))
671 672 673 674 675 676 677 678 679 680 681 682 683 684
			*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))
685
		*flags |= FOLL_COW;
686 687 688
	return 0;
}

689 690 691
static int check_vma_flags(struct vm_area_struct *vma, unsigned long gup_flags)
{
	vm_flags_t vm_flags = vma->vm_flags;
692 693
	int write = (gup_flags & FOLL_WRITE);
	int foreign = (gup_flags & FOLL_REMOTE);
694 695 696 697

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

698 699 700
	if (gup_flags & FOLL_ANON && !vma_is_anonymous(vma))
		return -EFAULT;

701
	if (write) {
702 703 704 705 706 707 708 709 710 711 712 713
		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.
			 */
714
			if (!is_cow_mapping(vm_flags))
715 716 717 718 719 720 721 722 723 724 725 726
				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;
	}
727 728 729 730 731
	/*
	 * gups are always data accesses, not instruction
	 * fetches, so execute=false here
	 */
	if (!arch_vma_access_permitted(vma, write, false, foreign))
732
		return -EFAULT;
733 734 735
	return 0;
}

736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755
/**
 * __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
 *
 * Returns number of pages pinned. This may be fewer than the number
 * requested. If nr_pages is 0 or negative, returns 0. If no pages
 * were pinned, returns -errno. Each page returned must be released
 * with a put_page() call when it is finished with. vmas will only
 * remain valid while mmap_sem is held.
 *
756
 * Must be called with mmap_sem held.  It may be released.  See below.
757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778
 *
 * __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,
779 780 781 782 783 784 785 786
 * *@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.
787 788 789 790 791
 *
 * 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.
 */
792
static long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
793 794 795 796
		unsigned long start, unsigned long nr_pages,
		unsigned int gup_flags, struct page **pages,
		struct vm_area_struct **vmas, int *nonblocking)
{
797
	long ret = 0, i = 0;
798
	struct vm_area_struct *vma = NULL;
799
	struct follow_page_context ctx = { NULL };
800 801 802 803 804 805 806 807 808 809 810 811 812 813 814

	if (!nr_pages)
		return 0;

	VM_BUG_ON(!!pages != !!(gup_flags & FOLL_GET));

	/*
	 * 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 {
815 816 817 818 819 820 821 822 823 824 825 826
		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)
827
					goto out;
828
				ctx.page_mask = 0;
829 830
				goto next_page;
			}
831

832 833 834 835
			if (!vma || check_vma_flags(vma, gup_flags)) {
				ret = -EFAULT;
				goto out;
			}
836 837 838
			if (is_vm_hugetlb_page(vma)) {
				i = follow_hugetlb_page(mm, vma, pages, vmas,
						&start, &nr_pages, i,
839
						gup_flags, nonblocking);
840
				continue;
841
			}
842 843 844 845 846 847
		}
retry:
		/*
		 * If we have a pending SIGKILL, don't keep faulting pages and
		 * potentially allocating memory.
		 */
848
		if (fatal_signal_pending(current)) {
849 850 851
			ret = -ERESTARTSYS;
			goto out;
		}
852
		cond_resched();
853 854

		page = follow_page_mask(vma, start, foll_flags, &ctx);
855 856 857 858 859 860
		if (!page) {
			ret = faultin_page(tsk, vma, start, &foll_flags,
					nonblocking);
			switch (ret) {
			case 0:
				goto retry;
861 862 863
			case -EBUSY:
				ret = 0;
				/* FALLTHRU */
864 865 866
			case -EFAULT:
			case -ENOMEM:
			case -EHWPOISON:
867
				goto out;
868 869
			case -ENOENT:
				goto next_page;
870
			}
871
			BUG();
872 873 874 875 876 877 878
		} else if (PTR_ERR(page) == -EEXIST) {
			/*
			 * Proper page table entry exists, but no corresponding
			 * struct page.
			 */
			goto next_page;
		} else if (IS_ERR(page)) {
879 880
			ret = PTR_ERR(page);
			goto out;
881
		}
882 883 884 885
		if (pages) {
			pages[i] = page;
			flush_anon_page(vma, page, start);
			flush_dcache_page(page);
886
			ctx.page_mask = 0;
887 888
		}
next_page:
889 890
		if (vmas) {
			vmas[i] = vma;
891
			ctx.page_mask = 0;
892
		}
893
		page_increm = 1 + (~(start >> PAGE_SHIFT) & ctx.page_mask);
894 895 896 897 898
		if (page_increm > nr_pages)
			page_increm = nr_pages;
		i += page_increm;
		start += page_increm * PAGE_SIZE;
		nr_pages -= page_increm;
899
	} while (nr_pages);
900 901 902 903
out:
	if (ctx.pgmap)
		put_dev_pagemap(ctx.pgmap);
	return i ? i : ret;
904 905
}

906 907
static bool vma_permits_fault(struct vm_area_struct *vma,
			      unsigned int fault_flags)
908
{
909 910
	bool write   = !!(fault_flags & FAULT_FLAG_WRITE);
	bool foreign = !!(fault_flags & FAULT_FLAG_REMOTE);
911
	vm_flags_t vm_flags = write ? VM_WRITE : VM_READ;
912 913 914 915

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

916 917
	/*
	 * The architecture might have a hardware protection
918
	 * mechanism other than read/write that can deny access.
919 920 921
	 *
	 * gup always represents data access, not instruction
	 * fetches, so execute=false here:
922
	 */
923
	if (!arch_vma_access_permitted(vma, write, false, foreign))
924 925
		return false;

926 927 928
	return true;
}

929 930 931 932 933 934 935
/*
 * 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()
936 937
 * @unlocked:	did we unlock the mmap_sem while retrying, maybe NULL if caller
 *		does not allow retry
938 939 940 941 942 943 944 945 946 947 948
 *
 * 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
949
 * get_user_pages() only guarantees to update these in the struct page.
950 951 952 953 954 955
 *
 * 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.
 *
956 957
 * 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().
958 959
 */
int fixup_user_fault(struct task_struct *tsk, struct mm_struct *mm,
960 961
		     unsigned long address, unsigned int fault_flags,
		     bool *unlocked)
962 963
{
	struct vm_area_struct *vma;
964
	vm_fault_t ret, major = 0;
965 966 967

	if (unlocked)
		fault_flags |= FAULT_FLAG_ALLOW_RETRY;
968

969
retry:
970 971 972 973
	vma = find_extend_vma(mm, address);
	if (!vma || address < vma->vm_start)
		return -EFAULT;

974
	if (!vma_permits_fault(vma, fault_flags))
975 976
		return -EFAULT;

977
	ret = handle_mm_fault(vma, address, fault_flags);
978
	major |= ret & VM_FAULT_MAJOR;
979
	if (ret & VM_FAULT_ERROR) {
980 981 982 983
		int err = vm_fault_to_errno(ret, 0);

		if (err)
			return err;
984 985
		BUG();
	}
986 987 988 989 990 991 992 993 994 995 996

	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;
		}
	}

997
	if (tsk) {
998
		if (major)
999 1000 1001 1002 1003 1004
			tsk->maj_flt++;
		else
			tsk->min_flt++;
	}
	return 0;
}
1005
EXPORT_SYMBOL_GPL(fixup_user_fault);
1006

1007 1008 1009 1010 1011 1012
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,
1013
						int *locked,
1014
						unsigned int flags)
1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054
{
	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);
	}

	if (pages)
		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 (!pages)
			/* If it's a prefault don't insist harder */
			return ret;

		if (ret > 0) {
			nr_pages -= ret;
			pages_done += ret;
			if (!nr_pages)
				break;
		}
		if (*locked) {
1055 1056 1057 1058
			/*
			 * VM_FAULT_RETRY didn't trigger or it was a
			 * FOLL_NOWAIT.
			 */
1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089
			if (!pages_done)
				pages_done = ret;
			break;
		}
		/* VM_FAULT_RETRY triggered, so seek to the faulting offset */
		pages += ret;
		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;
		pages++;
		start += PAGE_SIZE;
	}
1090
	if (lock_dropped && *locked) {
1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121
		/*
		 * 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;
}

/*
 * We can leverage the VM_FAULT_RETRY functionality in the page fault
 * paths better by using either get_user_pages_locked() or
 * get_user_pages_unlocked().
 *
 * get_user_pages_locked() is suitable to replace the form:
 *
 *      down_read(&mm->mmap_sem);
 *      do_something()
 *      get_user_pages(tsk, mm, ..., pages, NULL);
 *      up_read(&mm->mmap_sem);
 *
 *  to:
 *
 *      int locked = 1;
 *      down_read(&mm->mmap_sem);
 *      do_something()
 *      get_user_pages_locked(tsk, mm, ..., pages, &locked);
 *      if (locked)
 *          up_read(&mm->mmap_sem);
 */
1122
long get_user_pages_locked(unsigned long start, unsigned long nr_pages,
1123
			   unsigned int gup_flags, struct page **pages,
1124 1125
			   int *locked)
{
1126 1127 1128 1129 1130 1131 1132 1133 1134
	/*
	 * FIXME: Current FOLL_LONGTERM behavior is incompatible with
	 * FAULT_FLAG_ALLOW_RETRY because of the FS DAX check requirement on
	 * vmas.  As there are no users of this flag in this call we simply
	 * disallow this option for now.
	 */
	if (WARN_ON_ONCE(gup_flags & FOLL_LONGTERM))
		return -EINVAL;

1135
	return __get_user_pages_locked(current, current->mm, start, nr_pages,
1136
				       pages, NULL, locked,
1137
				       gup_flags | FOLL_TOUCH);
1138
}
1139
EXPORT_SYMBOL(get_user_pages_locked);
1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152

/*
 * get_user_pages_unlocked() is suitable to replace the form:
 *
 *      down_read(&mm->mmap_sem);
 *      get_user_pages(tsk, mm, ..., pages, NULL);
 *      up_read(&mm->mmap_sem);
 *
 *  with:
 *
 *      get_user_pages_unlocked(tsk, mm, ..., pages);
 *
 * It is functionally equivalent to get_user_pages_fast so
1153 1154
 * get_user_pages_fast should be used instead if specific gup_flags
 * (e.g. FOLL_FORCE) are not required.
1155
 */
1156
long get_user_pages_unlocked(unsigned long start, unsigned long nr_pages,
1157
			     struct page **pages, unsigned int gup_flags)
1158
{
1159 1160 1161 1162
	struct mm_struct *mm = current->mm;
	int locked = 1;
	long ret;

1163 1164 1165 1166 1167 1168 1169 1170 1171
	/*
	 * FIXME: Current FOLL_LONGTERM behavior is incompatible with
	 * FAULT_FLAG_ALLOW_RETRY because of the FS DAX check requirement on
	 * vmas.  As there are no users of this flag in this call we simply
	 * disallow this option for now.
	 */
	if (WARN_ON_ONCE(gup_flags & FOLL_LONGTERM))
		return -EINVAL;

1172 1173
	down_read(&mm->mmap_sem);
	ret = __get_user_pages_locked(current, mm, start, nr_pages, pages, NULL,
1174
				      &locked, gup_flags | FOLL_TOUCH);
1175 1176 1177
	if (locked)
		up_read(&mm->mmap_sem);
	return ret;
1178
}
1179
EXPORT_SYMBOL(get_user_pages_unlocked);
1180

1181
/*
1182
 * get_user_pages_remote() - pin user pages in memory
1183 1184 1185 1186 1187
 * @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
 * @start:	starting user address
 * @nr_pages:	number of pages from start to pin
1188
 * @gup_flags:	flags modifying lookup behaviour
1189 1190 1191 1192 1193
 * @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.
1194 1195 1196
 * @locked:	pointer to lock flag indicating whether lock is held and
 *		subsequently whether VM_FAULT_RETRY functionality can be
 *		utilised. Lock must initially be held.
1197 1198 1199 1200 1201 1202 1203 1204 1205