gup.c 66.3 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
 * put_user_pages_dirty_lock() - release and optionally dirty gup-pinned pages
 * @pages:  array of pages to be maybe marked dirty, and definitely released.
35
 * @npages: number of pages in the @pages array.
36
 * @make_dirty: whether to mark the pages dirty
37 38 39 40 41
 *
 * "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
42 43 44
 * 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.
45 46 47
 *
 * Please see the put_user_page() documentation for details.
 *
48 49 50 51
 * 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().
52 53
 *
 */
54 55
void put_user_pages_dirty_lock(struct page **pages, unsigned long npages,
			       bool make_dirty)
56
{
57
	unsigned long index;
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
	/*
	 * 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);
	}
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
}
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);

122
#ifdef CONFIG_MMU
123 124
static struct page *no_page_table(struct vm_area_struct *vma,
		unsigned int flags)
125
{
126 127 128 129 130 131 132 133 134 135 136 137
	/*
	 * 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;
}
138

139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162
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;
}

163 164 165 166 167 168
/*
 * 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)
{
169
	return pte_write(pte) ||
170 171 172
		((flags & FOLL_FORCE) && (flags & FOLL_COW) && pte_dirty(pte));
}

173
static struct page *follow_page_pte(struct vm_area_struct *vma,
174 175
		unsigned long address, pmd_t *pmd, unsigned int flags,
		struct dev_pagemap **pgmap)
176 177 178 179 180
{
	struct mm_struct *mm = vma->vm_mm;
	struct page *page;
	spinlock_t *ptl;
	pte_t *ptep, pte;
181

182
retry:
183
	if (unlikely(pmd_bad(*pmd)))
184
		return no_page_table(vma, flags);
185 186 187 188 189 190 191 192 193 194 195 196

	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;
197
		if (pte_none(pte))
198 199 200 201 202 203
			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);
204
		goto retry;
205
	}
206
	if ((flags & FOLL_NUMA) && pte_protnone(pte))
207
		goto no_page;
208
	if ((flags & FOLL_WRITE) && !can_follow_write_pte(pte, flags)) {
209 210 211
		pte_unmap_unlock(ptep, ptl);
		return NULL;
	}
212 213

	page = vm_normal_page(vma, address, pte);
214 215 216 217 218
	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.
		 */
219 220
		*pgmap = get_dev_pagemap(pte_pfn(pte), *pgmap);
		if (*pgmap)
221 222 223 224
			page = pte_page(pte);
		else
			goto no_page;
	} else if (unlikely(!page)) {
225 226 227 228 229 230 231 232 233 234 235 236 237 238 239
		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;
		}
240 241
	}

242 243 244 245 246 247 248 249 250 251 252 253 254
	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;
	}

255 256 257 258 259 260
	if (flags & FOLL_GET) {
		if (unlikely(!try_get_page(page))) {
			page = ERR_PTR(-ENOMEM);
			goto out;
		}
	}
261 262 263 264 265 266 267 268 269 270 271
	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
272
	if ((flags & FOLL_MLOCK) && (vma->vm_flags & VM_LOCKED)) {
273 274 275 276
		/* Do not mlock pte-mapped THP */
		if (PageTransCompound(page))
			goto out;

277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297
		/*
		 * 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);
		}
	}
298
out:
299 300 301 302 303
	pte_unmap_unlock(ptep, ptl);
	return page;
no_page:
	pte_unmap_unlock(ptep, ptl);
	if (!pte_none(pte))
304 305 306 307
		return NULL;
	return no_page_table(vma, flags);
}

308 309
static struct page *follow_pmd_mask(struct vm_area_struct *vma,
				    unsigned long address, pud_t *pudp,
310 311
				    unsigned int flags,
				    struct follow_page_context *ctx)
312
{
313
	pmd_t *pmd, pmdval;
314 315 316 317
	spinlock_t *ptl;
	struct page *page;
	struct mm_struct *mm = vma->vm_mm;

318
	pmd = pmd_offset(pudp, address);
319 320 321 322 323 324
	/*
	 * 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))
325
		return no_page_table(vma, flags);
326
	if (pmd_huge(pmdval) && is_vm_hugetlb_page(vma)) {
327 328 329 330
		page = follow_huge_pmd(mm, address, pmd, flags);
		if (page)
			return page;
		return no_page_table(vma, flags);
331
	}
332
	if (is_hugepd(__hugepd(pmd_val(pmdval)))) {
333
		page = follow_huge_pd(vma, address,
334
				      __hugepd(pmd_val(pmdval)), flags,
335 336 337 338 339
				      PMD_SHIFT);
		if (page)
			return page;
		return no_page_table(vma, flags);
	}
340
retry:
341
	if (!pmd_present(pmdval)) {
342 343 344
		if (likely(!(flags & FOLL_MIGRATION)))
			return no_page_table(vma, flags);
		VM_BUG_ON(thp_migration_supported() &&
345 346
				  !is_pmd_migration_entry(pmdval));
		if (is_pmd_migration_entry(pmdval))
347
			pmd_migration_entry_wait(mm, pmd);
348 349 350 351 352 353 354
		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);
355 356
		goto retry;
	}
357
	if (pmd_devmap(pmdval)) {
358
		ptl = pmd_lock(mm, pmd);
359
		page = follow_devmap_pmd(vma, address, pmd, flags, &ctx->pgmap);
360 361 362 363
		spin_unlock(ptl);
		if (page)
			return page;
	}
364
	if (likely(!pmd_trans_huge(pmdval)))
365
		return follow_page_pte(vma, address, pmd, flags, &ctx->pgmap);
366

367
	if ((flags & FOLL_NUMA) && pmd_protnone(pmdval))
368 369
		return no_page_table(vma, flags);

370
retry_locked:
371
	ptl = pmd_lock(mm, pmd);
372 373 374 375
	if (unlikely(pmd_none(*pmd))) {
		spin_unlock(ptl);
		return no_page_table(vma, flags);
	}
376 377 378 379 380 381 382
	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;
	}
383 384
	if (unlikely(!pmd_trans_huge(*pmd))) {
		spin_unlock(ptl);
385
		return follow_page_pte(vma, address, pmd, flags, &ctx->pgmap);
386
	}
Song Liu's avatar
Song Liu committed
387
	if (flags & (FOLL_SPLIT | FOLL_SPLIT_PMD)) {
388 389 390 391 392
		int ret;
		page = pmd_page(*pmd);
		if (is_huge_zero_page(page)) {
			spin_unlock(ptl);
			ret = 0;
393
			split_huge_pmd(vma, pmd, address);
394 395
			if (pmd_trans_unstable(pmd))
				ret = -EBUSY;
Song Liu's avatar
Song Liu committed
396
		} else if (flags & FOLL_SPLIT) {
397 398 399 400
			if (unlikely(!try_get_page(page))) {
				spin_unlock(ptl);
				return ERR_PTR(-ENOMEM);
			}
401
			spin_unlock(ptl);
402 403 404 405
			lock_page(page);
			ret = split_huge_page(page);
			unlock_page(page);
			put_page(page);
406 407
			if (pmd_none(*pmd))
				return no_page_table(vma, flags);
Song Liu's avatar
Song Liu committed
408 409 410 411
		} else {  /* flags & FOLL_SPLIT_PMD */
			spin_unlock(ptl);
			split_huge_pmd(vma, pmd, address);
			ret = pte_alloc(mm, pmd) ? -ENOMEM : 0;
412 413 414
		}

		return ret ? ERR_PTR(ret) :
415
			follow_page_pte(vma, address, pmd, flags, &ctx->pgmap);
416
	}
417 418
	page = follow_trans_huge_pmd(vma, address, pmd, flags);
	spin_unlock(ptl);
419
	ctx->page_mask = HPAGE_PMD_NR - 1;
420
	return page;
421 422
}

423 424
static struct page *follow_pud_mask(struct vm_area_struct *vma,
				    unsigned long address, p4d_t *p4dp,
425 426
				    unsigned int flags,
				    struct follow_page_context *ctx)
427 428 429 430 431 432 433 434 435
{
	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);
436
	if (pud_huge(*pud) && is_vm_hugetlb_page(vma)) {
437 438 439 440 441
		page = follow_huge_pud(mm, address, pud, flags);
		if (page)
			return page;
		return no_page_table(vma, flags);
	}
442 443 444 445 446 447 448 449
	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);
	}
450 451
	if (pud_devmap(*pud)) {
		ptl = pud_lock(mm, pud);
452
		page = follow_devmap_pud(vma, address, pud, flags, &ctx->pgmap);
453 454 455 456 457 458 459
		spin_unlock(ptl);
		if (page)
			return page;
	}
	if (unlikely(pud_bad(*pud)))
		return no_page_table(vma, flags);

460
	return follow_pmd_mask(vma, address, pud, flags, ctx);
461 462 463 464
}

static struct page *follow_p4d_mask(struct vm_area_struct *vma,
				    unsigned long address, pgd_t *pgdp,
465 466
				    unsigned int flags,
				    struct follow_page_context *ctx)
467 468
{
	p4d_t *p4d;
469
	struct page *page;
470 471 472 473 474 475 476 477

	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);

478 479 480 481 482 483 484 485
	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);
	}
486
	return follow_pud_mask(vma, address, p4d, flags, ctx);
487 488 489 490 491 492 493
}

/**
 * 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
494 495
 * @ctx: contains dev_pagemap for %ZONE_DEVICE memory pinning and a
 *       pointer to output page_mask
496 497 498
 *
 * @flags can have FOLL_ flags set, defined in <linux/mm.h>
 *
499 500 501 502 503 504
 * 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
505 506 507
 * an error pointer if there is a mapping to something not represented
 * by a page descriptor (see also vm_normal_page()).
 */
508
static struct page *follow_page_mask(struct vm_area_struct *vma,
509
			      unsigned long address, unsigned int flags,
510
			      struct follow_page_context *ctx)
511 512 513 514 515
{
	pgd_t *pgd;
	struct page *page;
	struct mm_struct *mm = vma->vm_mm;

516
	ctx->page_mask = 0;
517 518 519 520 521 522 523 524 525 526 527 528 529

	/* 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);

530 531 532 533 534 535
	if (pgd_huge(*pgd)) {
		page = follow_huge_pgd(mm, address, pgd, flags);
		if (page)
			return page;
		return no_page_table(vma, flags);
	}
536 537 538 539 540 541 542 543
	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);
	}
544

545 546 547 548 549 550 551 552 553 554 555 556 557
	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;
558 559
}

560 561 562 563 564
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;
565
	p4d_t *p4d;
566 567 568 569 570 571 572 573 574 575 576 577
	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);
578 579
	if (pgd_none(*pgd))
		return -EFAULT;
580
	p4d = p4d_offset(pgd, address);
581 582
	if (p4d_none(*p4d))
		return -EFAULT;
583
	pud = pud_offset(p4d, address);
584 585
	if (pud_none(*pud))
		return -EFAULT;
586
	pmd = pmd_offset(pud, address);
587
	if (!pmd_present(*pmd))
588 589 590 591 592 593 594 595 596 597 598 599 600 601
		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);
	}
602 603 604 605
	if (unlikely(!try_get_page(*page))) {
		ret = -ENOMEM;
		goto unmap;
	}
606 607 608 609 610 611 612
out:
	ret = 0;
unmap:
	pte_unmap(pte);
	return ret;
}

613 614 615 616 617
/*
 * 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.
 */
618 619 620 621
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;
622
	vm_fault_t ret;
623

Eric B Munson's avatar
Eric B Munson committed
624 625 626
	/* mlock all present pages, but do not fault in new pages */
	if ((*flags & (FOLL_POPULATE | FOLL_MLOCK)) == FOLL_MLOCK)
		return -ENOENT;
627 628
	if (*flags & FOLL_WRITE)
		fault_flags |= FAULT_FLAG_WRITE;
629 630
	if (*flags & FOLL_REMOTE)
		fault_flags |= FAULT_FLAG_REMOTE;
631 632 633 634
	if (nonblocking)
		fault_flags |= FAULT_FLAG_ALLOW_RETRY;
	if (*flags & FOLL_NOWAIT)
		fault_flags |= FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_RETRY_NOWAIT;
635 636 637 638
	if (*flags & FOLL_TRIED) {
		VM_WARN_ON_ONCE(fault_flags & FAULT_FLAG_ALLOW_RETRY);
		fault_flags |= FAULT_FLAG_TRIED;
	}
639

640
	ret = handle_mm_fault(vma, address, fault_flags);
641
	if (ret & VM_FAULT_ERROR) {
642 643 644 645
		int err = vm_fault_to_errno(ret, *flags);

		if (err)
			return err;
646 647 648 649 650 651 652 653 654 655 656
		BUG();
	}

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

	if (ret & VM_FAULT_RETRY) {
657
		if (nonblocking && !(fault_flags & FAULT_FLAG_RETRY_NOWAIT))
658 659 660 661 662 663 664 665 666 667 668 669 670 671
			*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))
672
		*flags |= FOLL_COW;
673 674 675
	return 0;
}

676 677 678
static int check_vma_flags(struct vm_area_struct *vma, unsigned long gup_flags)
{
	vm_flags_t vm_flags = vma->vm_flags;
679 680
	int write = (gup_flags & FOLL_WRITE);
	int foreign = (gup_flags & FOLL_REMOTE);
681 682 683 684

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

685 686 687
	if (gup_flags & FOLL_ANON && !vma_is_anonymous(vma))
		return -EFAULT;

688
	if (write) {
689 690 691 692 693 694 695 696 697 698 699 700
		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.
			 */
701
			if (!is_cow_mapping(vm_flags))
702 703 704 705 706 707 708 709 710 711 712 713
				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;
	}
714 715 716 717 718
	/*
	 * gups are always data accesses, not instruction
	 * fetches, so execute=false here
	 */
	if (!arch_vma_access_permitted(vma, write, false, foreign))
719
		return -EFAULT;
720 721 722
	return 0;
}

723 724 725 726 727 728 729 730 731 732 733 734 735 736
/**
 * __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
 *
737 738 739 740 741 742 743 744 745 746 747
 * 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.
748
 *
749
 * Must be called with mmap_sem held.  It may be released.  See below.
750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771
 *
 * __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,
772 773 774 775 776 777 778 779
 * *@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.
780 781 782 783 784
 *
 * 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.
 */
785
static long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
786 787 788 789
		unsigned long start, unsigned long nr_pages,
		unsigned int gup_flags, struct page **pages,
		struct vm_area_struct **vmas, int *nonblocking)
{
790
	long ret = 0, i = 0;
791
	struct vm_area_struct *vma = NULL;
792
	struct follow_page_context ctx = { NULL };
793 794 795 796

	if (!nr_pages)
		return 0;

797 798
	start = untagged_addr(start);

799 800 801 802 803 804 805 806 807 808 809
	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 {
810 811 812 813 814 815 816 817 818 819 820 821
		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)
822
					goto out;
823
				ctx.page_mask = 0;
824 825
				goto next_page;
			}
826

827 828 829 830
			if (!vma || check_vma_flags(vma, gup_flags)) {
				ret = -EFAULT;
				goto out;
			}
831 832 833
			if (is_vm_hugetlb_page(vma)) {
				i = follow_hugetlb_page(mm, vma, pages, vmas,
						&start, &nr_pages, i,
834
						gup_flags, nonblocking);
835
				continue;
836
			}
837 838 839 840 841 842
		}
retry:
		/*
		 * If we have a pending SIGKILL, don't keep faulting pages and
		 * potentially allocating memory.
		 */
843
		if (fatal_signal_pending(current)) {
844 845 846
			ret = -ERESTARTSYS;
			goto out;
		}
847
		cond_resched();
848 849

		page = follow_page_mask(vma, start, foll_flags, &ctx);
850 851 852 853 854 855
		if (!page) {
			ret = faultin_page(tsk, vma, start, &foll_flags,
					nonblocking);
			switch (ret) {
			case 0:
				goto retry;
856 857 858
			case -EBUSY:
				ret = 0;
				/* FALLTHRU */
859 860 861
			case -EFAULT:
			case -ENOMEM:
			case -EHWPOISON:
862
				goto out;
863 864
			case -ENOENT:
				goto next_page;
865
			}
866
			BUG();
867 868 869 870 871 872 873
		} else if (PTR_ERR(page) == -EEXIST) {
			/*
			 * Proper page table entry exists, but no corresponding
			 * struct page.
			 */
			goto next_page;
		} else if (IS_ERR(page)) {
874 875
			ret = PTR_ERR(page);
			goto out;
876
		}
877 878 879 880
		if (pages) {
			pages[i] = page;
			flush_anon_page(vma, page, start);
			flush_dcache_page(page);
881
			ctx.page_mask = 0;
882 883
		}
next_page:
884 885
		if (vmas) {
			vmas[i] = vma;
886
			ctx.page_mask = 0;
887
		}
888
		page_increm = 1 + (~(start >> PAGE_SHIFT) & ctx.page_mask);
889 890 891 892 893
		if (page_increm > nr_pages)
			page_increm = nr_pages;
		i += page_increm;
		start += page_increm * PAGE_SIZE;
		nr_pages -= page_increm;
894
	} while (nr_pages);
895 896 897 898
out:
	if (ctx.pgmap)
		put_dev_pagemap(ctx.pgmap);
	return i ? i : ret;
899 900
}

901 902
static bool vma_permits_fault(struct vm_area_struct *vma,
			      unsigned int fault_flags)
903
{
904 905
	bool write   = !!(fault_flags & FAULT_FLAG_WRITE);
	bool foreign = !!(fault_flags & FAULT_FLAG_REMOTE);
906
	vm_flags_t vm_flags = write ? VM_WRITE : VM_READ;
907 908 909 910

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

911 912
	/*
	 * The architecture might have a hardware protection
913
	 * mechanism other than read/write that can deny access.
914 915 916
	 *
	 * gup always represents data access, not instruction
	 * fetches, so execute=false here:
917
	 */
918
	if (!arch_vma_access_permitted(vma, write, false, foreign))
919 920
		return false;

921 922 923
	return true;
}

924 925 926 927 928 929 930
/*
 * 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()
931 932
 * @unlocked:	did we unlock the mmap_sem while retrying, maybe NULL if caller
 *		does not allow retry
933 934 935 936 937 938 939 940 941 942 943
 *
 * 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
944
 * get_user_pages() only guarantees to update these in the struct page.
945 946 947 948 949 950
 *
 * 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.
 *
951 952
 * 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().
953 954
 */
int fixup_user_fault(struct task_struct *tsk, struct mm_struct *mm,
955 956
		     unsigned long address, unsigned int fault_flags,
		     bool *unlocked)
957 958
{
	struct vm_area_struct *vma;
959
	vm_fault_t ret, major = 0;
960

961 962
	address = untagged_addr(address);

963 964
	if (unlocked)
		fault_flags |= FAULT_FLAG_ALLOW_RETRY;
965

966
retry:
967 968 969 970
	vma = find_extend_vma(mm, address);
	if (!vma || address < vma->vm_start)
		return -EFAULT;

971
	if (!vma_permits_fault(vma, fault_flags))
972 973
		return -EFAULT;

974
	ret = handle_mm_fault(vma, address, fault_flags);
975
	major |= ret & VM_FAULT_MAJOR;
976
	if (ret & VM_FAULT_ERROR) {
977 978 979 980
		int err = vm_fault_to_errno(ret, 0);

		if (err)
			return err;
981 982
		BUG();
	}
983 984 985 986 987 988 989 990 991 992 993

	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;
		}
	}

994
	if (tsk) {
995
		if (major)
996 997 998 999 1000 1001
			tsk->maj_flt++;
		else
			tsk->min_flt++;
	}
	return 0;
}
1002
EXPORT_SYMBOL_GPL(fixup_user_fault);
1003

1004 1005 1006 1007 1008 1009
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,
1010
						int *locked,
1011
						unsigned int flags)
1012 1013 1014 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
{
	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 (ret > 0) {
			nr_pages -= ret;
			pages_done += ret;
			if (!nr_pages)
				break;
		}
		if (*locked) {
1048 1049 1050 1051
			/*
			 * VM_FAULT_RETRY didn't trigger or it was a
			 * FOLL_NOWAIT.
			 */
1052 1053 1054 1055
			if (!pages_done)
				pages_done = ret;
			break;
		}
1056 1057 1058 1059 1060 1061
		/*
		 * VM_FAULT_RETRY triggered, so seek to the faulting offset.
		 * For the prefault case (!pages) we only update counts.
		 */
		if (likely(pages))
			pages += ret;
1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083
		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;
1084 1085
		if (likely(pages))
			pages++;
1086 1087
		start += PAGE_SIZE;
	}
1088
	if (lock_dropped && *locked) {
1089 1090 1091 1092 1093 1094 1095 1096 1097 1098
		/*
		 * 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;
}

1099
/*
1100
 * get_user_pages_remote() - pin user pages in memory
1101 1102 1103 1104 1105
 * @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
1106
 * @gup_flags:	flags modifying lookup behaviour
1107 1108 1109 1110 1111
 * @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.
1112 1113 1114
 * @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.
1115
 *
1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126
 * 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.
1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143
 *
 * Must be called with mmap_sem held for read or write.
 *
 * 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.
 *
1144 1145 1146
 * 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.
1147 1148