gup.c 80.9 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 47
static void hpage_pincount_add(struct page *page, int refs)
{
	VM_BUG_ON_PAGE(!hpage_pincount_available(page), page);
	VM_BUG_ON_PAGE(page != compound_head(page), page);

	atomic_add(refs, compound_pincount_ptr(page));
}

static void hpage_pincount_sub(struct page *page, int refs)
{
	VM_BUG_ON_PAGE(!hpage_pincount_available(page), page);
	VM_BUG_ON_PAGE(page != compound_head(page), page);

	atomic_sub(refs, compound_pincount_ptr(page));
}

48 49 50 51 52 53 54 55 56 57 58 59 60 61 62
/*
 * 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;
}

John Hubbard's avatar
John Hubbard committed
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
/*
 * try_grab_compound_head() - attempt to elevate a page's refcount, by a
 * flags-dependent amount.
 *
 * "grab" names in this file mean, "look at flags to decide whether to use
 * FOLL_PIN or FOLL_GET behavior, when incrementing the page's refcount.
 *
 * Either FOLL_PIN or FOLL_GET (or neither) must be set, but not both at the
 * same time. (That's true throughout the get_user_pages*() and
 * pin_user_pages*() APIs.) Cases:
 *
 *    FOLL_GET: page's refcount will be incremented by 1.
 *    FOLL_PIN: page's refcount will be incremented by GUP_PIN_COUNTING_BIAS.
 *
 * Return: head page (with refcount appropriately incremented) for success, or
 * NULL upon failure. If neither FOLL_GET nor FOLL_PIN was set, that's
 * considered failure, and furthermore, a likely bug in the caller, so a warning
 * is also emitted.
 */
static __maybe_unused struct page *try_grab_compound_head(struct page *page,
							  int refs,
							  unsigned int flags)
{
	if (flags & FOLL_GET)
		return try_get_compound_head(page, refs);
	else if (flags & FOLL_PIN) {
89 90
		int orig_refs = refs;

91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108
		/*
		 * When pinning a compound page of order > 1 (which is what
		 * hpage_pincount_available() checks for), use an exact count to
		 * track it, via hpage_pincount_add/_sub().
		 *
		 * However, be sure to *also* increment the normal page refcount
		 * field at least once, so that the page really is pinned.
		 */
		if (!hpage_pincount_available(page))
			refs *= GUP_PIN_COUNTING_BIAS;

		page = try_get_compound_head(page, refs);
		if (!page)
			return NULL;

		if (hpage_pincount_available(page))
			hpage_pincount_add(page, refs);

109 110 111
		mod_node_page_state(page_pgdat(page), NR_FOLL_PIN_ACQUIRED,
				    orig_refs);

112
		return page;
John Hubbard's avatar
John Hubbard committed
113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146
	}

	WARN_ON_ONCE(1);
	return NULL;
}

/**
 * try_grab_page() - elevate a page's refcount by a flag-dependent amount
 *
 * This might not do anything at all, depending on the flags argument.
 *
 * "grab" names in this file mean, "look at flags to decide whether to use
 * FOLL_PIN or FOLL_GET behavior, when incrementing the page's refcount.
 *
 * @page:    pointer to page to be grabbed
 * @flags:   gup flags: these are the FOLL_* flag values.
 *
 * Either FOLL_PIN or FOLL_GET (or neither) may be set, but not both at the same
 * time. Cases:
 *
 *    FOLL_GET: page's refcount will be incremented by 1.
 *    FOLL_PIN: page's refcount will be incremented by GUP_PIN_COUNTING_BIAS.
 *
 * Return: true for success, or if no action was required (if neither FOLL_PIN
 * nor FOLL_GET was set, nothing is done). False for failure: FOLL_GET or
 * FOLL_PIN was set, but the page could not be grabbed.
 */
bool __must_check try_grab_page(struct page *page, unsigned int flags)
{
	WARN_ON_ONCE((flags & (FOLL_GET | FOLL_PIN)) == (FOLL_GET | FOLL_PIN));

	if (flags & FOLL_GET)
		return try_get_page(page);
	else if (flags & FOLL_PIN) {
147 148
		int refs = 1;

John Hubbard's avatar
John Hubbard committed
149 150 151 152 153
		page = compound_head(page);

		if (WARN_ON_ONCE(page_ref_count(page) <= 0))
			return false;

154 155 156 157 158 159 160 161 162 163 164 165
		if (hpage_pincount_available(page))
			hpage_pincount_add(page, 1);
		else
			refs = GUP_PIN_COUNTING_BIAS;

		/*
		 * Similar to try_grab_compound_head(): even if using the
		 * hpage_pincount_add/_sub() routines, be sure to
		 * *also* increment the normal page refcount field at least
		 * once, so that the page really is pinned.
		 */
		page_ref_add(page, refs);
166 167

		mod_node_page_state(page_pgdat(page), NR_FOLL_PIN_ACQUIRED, 1);
John Hubbard's avatar
John Hubbard committed
168 169 170 171 172 173 174 175
	}

	return true;
}

#ifdef CONFIG_DEV_PAGEMAP_OPS
static bool __unpin_devmap_managed_user_page(struct page *page)
{
176
	int count, refs = 1;
John Hubbard's avatar
John Hubbard committed
177 178 179 180

	if (!page_is_devmap_managed(page))
		return false;

181 182 183 184 185 186
	if (hpage_pincount_available(page))
		hpage_pincount_sub(page, 1);
	else
		refs = GUP_PIN_COUNTING_BIAS;

	count = page_ref_sub_return(page, refs);
John Hubbard's avatar
John Hubbard committed
187

188
	mod_node_page_state(page_pgdat(page), NR_FOLL_PIN_RELEASED, 1);
John Hubbard's avatar
John Hubbard committed
189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218
	/*
	 * devmap page refcounts are 1-based, rather than 0-based: if
	 * refcount is 1, then the page is free and the refcount is
	 * stable because nobody holds a reference on the page.
	 */
	if (count == 1)
		free_devmap_managed_page(page);
	else if (!count)
		__put_page(page);

	return true;
}
#else
static bool __unpin_devmap_managed_user_page(struct page *page)
{
	return false;
}
#endif /* CONFIG_DEV_PAGEMAP_OPS */

/**
 * unpin_user_page() - release a dma-pinned page
 * @page:            pointer to page to be released
 *
 * Pages that were pinned via pin_user_pages*() must be released via either
 * unpin_user_page(), or one of the unpin_user_pages*() routines. This is so
 * that such pages can be separately tracked and uniquely handled. In
 * particular, interactions with RDMA and filesystems need special handling.
 */
void unpin_user_page(struct page *page)
{
219 220
	int refs = 1;

John Hubbard's avatar
John Hubbard committed
221 222 223 224 225 226 227 228 229 230 231
	page = compound_head(page);

	/*
	 * For devmap managed pages we need to catch refcount transition from
	 * GUP_PIN_COUNTING_BIAS to 1, when refcount reach one it means the
	 * page is free and we need to inform the device driver through
	 * callback. See include/linux/memremap.h and HMM for details.
	 */
	if (__unpin_devmap_managed_user_page(page))
		return;

232 233 234 235 236 237
	if (hpage_pincount_available(page))
		hpage_pincount_sub(page, 1);
	else
		refs = GUP_PIN_COUNTING_BIAS;

	if (page_ref_sub_and_test(page, refs))
John Hubbard's avatar
John Hubbard committed
238
		__put_page(page);
239 240

	mod_node_page_state(page_pgdat(page), NR_FOLL_PIN_RELEASED, 1);
John Hubbard's avatar
John Hubbard committed
241 242 243
}
EXPORT_SYMBOL(unpin_user_page);

244
/**
245
 * unpin_user_pages_dirty_lock() - release and optionally dirty gup-pinned pages
246
 * @pages:  array of pages to be maybe marked dirty, and definitely released.
247
 * @npages: number of pages in the @pages array.
248
 * @make_dirty: whether to mark the pages dirty
249 250 251 252 253
 *
 * "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
254
 * compound page) dirty, if @make_dirty is true, and if the page was previously
255 256
 * listed as clean. In any case, releases all pages using unpin_user_page(),
 * possibly via unpin_user_pages(), for the non-dirty case.
257
 *
258
 * Please see the unpin_user_page() documentation for details.
259
 *
260 261 262
 * 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:
263
 * set_page_dirty_lock(), unpin_user_page().
264 265
 *
 */
266 267
void unpin_user_pages_dirty_lock(struct page **pages, unsigned long npages,
				 bool make_dirty)
268
{
269
	unsigned long index;
270

271 272 273 274 275 276 277
	/*
	 * 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) {
278
		unpin_user_pages(pages, npages);
279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305
		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);
306
		unpin_user_page(page);
307
	}
308
}
309
EXPORT_SYMBOL(unpin_user_pages_dirty_lock);
310 311

/**
312
 * unpin_user_pages() - release an array of gup-pinned pages.
313 314 315
 * @pages:  array of pages to be marked dirty and released.
 * @npages: number of pages in the @pages array.
 *
316
 * For each page in the @pages array, release the page using unpin_user_page().
317
 *
318
 * Please see the unpin_user_page() documentation for details.
319
 */
320
void unpin_user_pages(struct page **pages, unsigned long npages)
321 322 323 324 325 326 327 328 329
{
	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++)
330
		unpin_user_page(pages[index]);
331
}
332
EXPORT_SYMBOL(unpin_user_pages);
333

334
#ifdef CONFIG_MMU
335 336
static struct page *no_page_table(struct vm_area_struct *vma,
		unsigned int flags)
337
{
338 339 340 341 342 343 344 345 346 347 348 349
	/*
	 * 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;
}
350

351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374
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;
}

375 376 377 378 379 380
/*
 * 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)
{
381
	return pte_write(pte) ||
382 383 384
		((flags & FOLL_FORCE) && (flags & FOLL_COW) && pte_dirty(pte));
}

385
static struct page *follow_page_pte(struct vm_area_struct *vma,
386 387
		unsigned long address, pmd_t *pmd, unsigned int flags,
		struct dev_pagemap **pgmap)
388 389 390 391 392
{
	struct mm_struct *mm = vma->vm_mm;
	struct page *page;
	spinlock_t *ptl;
	pte_t *ptep, pte;
393
	int ret;
394

395 396 397 398
	/* 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);
399
retry:
400
	if (unlikely(pmd_bad(*pmd)))
401
		return no_page_table(vma, flags);
402 403 404 405 406 407 408 409 410 411 412 413

	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;
414
		if (pte_none(pte))
415 416 417 418 419 420
			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);
421
		goto retry;
422
	}
423
	if ((flags & FOLL_NUMA) && pte_protnone(pte))
424
		goto no_page;
425
	if ((flags & FOLL_WRITE) && !can_follow_write_pte(pte, flags)) {
426 427 428
		pte_unmap_unlock(ptep, ptl);
		return NULL;
	}
429 430

	page = vm_normal_page(vma, address, pte);
John Hubbard's avatar
John Hubbard committed
431
	if (!page && pte_devmap(pte) && (flags & (FOLL_GET | FOLL_PIN))) {
432
		/*
John Hubbard's avatar
John Hubbard committed
433 434 435
		 * Only return device mapping pages in the FOLL_GET or FOLL_PIN
		 * case since they are only valid while holding the pgmap
		 * reference.
436
		 */
437 438
		*pgmap = get_dev_pagemap(pte_pfn(pte), *pgmap);
		if (*pgmap)
439 440 441 442
			page = pte_page(pte);
		else
			goto no_page;
	} else if (unlikely(!page)) {
443 444 445 446 447 448 449 450 451 452 453 454 455
		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 {
			ret = follow_pfn_pte(vma, address, ptep, flags);
			page = ERR_PTR(ret);
			goto out;
		}
456 457
	}

458 459 460 461 462 463 464 465 466 467 468 469
	if (flags & FOLL_SPLIT && PageTransCompound(page)) {
		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;
	}

John Hubbard's avatar
John Hubbard committed
470 471 472 473
	/* try_grab_page() does nothing unless FOLL_GET or FOLL_PIN is set. */
	if (unlikely(!try_grab_page(page, flags))) {
		page = ERR_PTR(-ENOMEM);
		goto out;
474
	}
475 476 477 478 479 480 481 482 483 484 485 486 487
	/*
	 * We need to make the page accessible if and only if we are going
	 * to access its content (the FOLL_PIN case).  Please see
	 * Documentation/core-api/pin_user_pages.rst for details.
	 */
	if (flags & FOLL_PIN) {
		ret = arch_make_page_accessible(page);
		if (ret) {
			unpin_user_page(page);
			page = ERR_PTR(ret);
			goto out;
		}
	}
488 489 490 491 492 493 494 495 496 497 498
	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
499
	if ((flags & FOLL_MLOCK) && (vma->vm_flags & VM_LOCKED)) {
500 501 502 503
		/* Do not mlock pte-mapped THP */
		if (PageTransCompound(page))
			goto out;

504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524
		/*
		 * 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);
		}
	}
525
out:
526 527 528 529 530
	pte_unmap_unlock(ptep, ptl);
	return page;
no_page:
	pte_unmap_unlock(ptep, ptl);
	if (!pte_none(pte))
531 532 533 534
		return NULL;
	return no_page_table(vma, flags);
}

535 536
static struct page *follow_pmd_mask(struct vm_area_struct *vma,
				    unsigned long address, pud_t *pudp,
537 538
				    unsigned int flags,
				    struct follow_page_context *ctx)
539
{
540
	pmd_t *pmd, pmdval;
541 542 543 544
	spinlock_t *ptl;
	struct page *page;
	struct mm_struct *mm = vma->vm_mm;

545
	pmd = pmd_offset(pudp, address);
546 547 548 549 550 551
	/*
	 * 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))
552
		return no_page_table(vma, flags);
553
	if (pmd_huge(pmdval) && is_vm_hugetlb_page(vma)) {
554 555 556 557
		page = follow_huge_pmd(mm, address, pmd, flags);
		if (page)
			return page;
		return no_page_table(vma, flags);
558
	}
559
	if (is_hugepd(__hugepd(pmd_val(pmdval)))) {
560
		page = follow_huge_pd(vma, address,
561
				      __hugepd(pmd_val(pmdval)), flags,
562 563 564 565 566
				      PMD_SHIFT);
		if (page)
			return page;
		return no_page_table(vma, flags);
	}
567
retry:
568
	if (!pmd_present(pmdval)) {
569 570 571
		if (likely(!(flags & FOLL_MIGRATION)))
			return no_page_table(vma, flags);
		VM_BUG_ON(thp_migration_supported() &&
572 573
				  !is_pmd_migration_entry(pmdval));
		if (is_pmd_migration_entry(pmdval))
574
			pmd_migration_entry_wait(mm, pmd);
575 576 577 578 579 580 581
		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);
582 583
		goto retry;
	}
584
	if (pmd_devmap(pmdval)) {
585
		ptl = pmd_lock(mm, pmd);
586
		page = follow_devmap_pmd(vma, address, pmd, flags, &ctx->pgmap);
587 588 589 590
		spin_unlock(ptl);
		if (page)
			return page;
	}
591
	if (likely(!pmd_trans_huge(pmdval)))
592
		return follow_page_pte(vma, address, pmd, flags, &ctx->pgmap);
593

594
	if ((flags & FOLL_NUMA) && pmd_protnone(pmdval))
595 596
		return no_page_table(vma, flags);

597
retry_locked:
598
	ptl = pmd_lock(mm, pmd);
599 600 601 602
	if (unlikely(pmd_none(*pmd))) {
		spin_unlock(ptl);
		return no_page_table(vma, flags);
	}
603 604 605 606 607 608 609
	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;
	}
610 611
	if (unlikely(!pmd_trans_huge(*pmd))) {
		spin_unlock(ptl);
612
		return follow_page_pte(vma, address, pmd, flags, &ctx->pgmap);
613
	}
Song Liu's avatar
Song Liu committed
614
	if (flags & (FOLL_SPLIT | FOLL_SPLIT_PMD)) {
615 616 617 618 619
		int ret;
		page = pmd_page(*pmd);
		if (is_huge_zero_page(page)) {
			spin_unlock(ptl);
			ret = 0;
620
			split_huge_pmd(vma, pmd, address);
621 622
			if (pmd_trans_unstable(pmd))
				ret = -EBUSY;
Song Liu's avatar
Song Liu committed
623
		} else if (flags & FOLL_SPLIT) {
624 625 626 627
			if (unlikely(!try_get_page(page))) {
				spin_unlock(ptl);
				return ERR_PTR(-ENOMEM);
			}
628
			spin_unlock(ptl);
629 630 631 632
			lock_page(page);
			ret = split_huge_page(page);
			unlock_page(page);
			put_page(page);
633 634
			if (pmd_none(*pmd))
				return no_page_table(vma, flags);
Song Liu's avatar
Song Liu committed
635 636 637 638
		} else {  /* flags & FOLL_SPLIT_PMD */
			spin_unlock(ptl);
			split_huge_pmd(vma, pmd, address);
			ret = pte_alloc(mm, pmd) ? -ENOMEM : 0;
639 640 641
		}

		return ret ? ERR_PTR(ret) :
642
			follow_page_pte(vma, address, pmd, flags, &ctx->pgmap);
643
	}
644 645
	page = follow_trans_huge_pmd(vma, address, pmd, flags);
	spin_unlock(ptl);
646
	ctx->page_mask = HPAGE_PMD_NR - 1;
647
	return page;
648 649
}

650 651
static struct page *follow_pud_mask(struct vm_area_struct *vma,
				    unsigned long address, p4d_t *p4dp,
652 653
				    unsigned int flags,
				    struct follow_page_context *ctx)
654 655 656 657 658 659 660 661 662
{
	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);
663
	if (pud_huge(*pud) && is_vm_hugetlb_page(vma)) {
664 665 666 667 668
		page = follow_huge_pud(mm, address, pud, flags);
		if (page)
			return page;
		return no_page_table(vma, flags);
	}
669 670 671 672 673 674 675 676
	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);
	}
677 678
	if (pud_devmap(*pud)) {
		ptl = pud_lock(mm, pud);
679
		page = follow_devmap_pud(vma, address, pud, flags, &ctx->pgmap);
680 681 682 683 684 685 686
		spin_unlock(ptl);
		if (page)
			return page;
	}
	if (unlikely(pud_bad(*pud)))
		return no_page_table(vma, flags);

687
	return follow_pmd_mask(vma, address, pud, flags, ctx);
688 689 690 691
}

static struct page *follow_p4d_mask(struct vm_area_struct *vma,
				    unsigned long address, pgd_t *pgdp,
692 693
				    unsigned int flags,
				    struct follow_page_context *ctx)
694 695
{
	p4d_t *p4d;
696
	struct page *page;
697 698 699 700 701 702 703 704

	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);

705 706 707 708 709 710 711 712
	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);
	}
713
	return follow_pud_mask(vma, address, p4d, flags, ctx);
714 715 716 717 718 719 720
}

/**
 * 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
721 722
 * @ctx: contains dev_pagemap for %ZONE_DEVICE memory pinning and a
 *       pointer to output page_mask
723 724 725
 *
 * @flags can have FOLL_ flags set, defined in <linux/mm.h>
 *
726 727 728 729 730 731
 * 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
732 733 734
 * an error pointer if there is a mapping to something not represented
 * by a page descriptor (see also vm_normal_page()).
 */
735
static struct page *follow_page_mask(struct vm_area_struct *vma,
736
			      unsigned long address, unsigned int flags,
737
			      struct follow_page_context *ctx)
738 739 740 741 742
{
	pgd_t *pgd;
	struct page *page;
	struct mm_struct *mm = vma->vm_mm;

743
	ctx->page_mask = 0;
744 745 746 747

	/* make this handle hugepd */
	page = follow_huge_addr(mm, address, flags & FOLL_WRITE);
	if (!IS_ERR(page)) {
John Hubbard's avatar
John Hubbard committed
748
		WARN_ON_ONCE(flags & (FOLL_GET | FOLL_PIN));
749 750 751 752 753 754 755 756
		return page;
	}

	pgd = pgd_offset(mm, address);

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

757 758 759 760 761 762
	if (pgd_huge(*pgd)) {
		page = follow_huge_pgd(mm, address, pgd, flags);
		if (page)
			return page;
		return no_page_table(vma, flags);
	}
763 764 765 766 767 768 769 770
	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);
	}
771

772 773 774 775 776 777 778 779 780 781 782 783 784
	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;
785 786
}

787 788 789 790 791
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;
792
	p4d_t *p4d;
793 794 795 796 797 798 799 800 801 802 803 804
	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);
805 806
	if (pgd_none(*pgd))
		return -EFAULT;
807
	p4d = p4d_offset(pgd, address);
808 809
	if (p4d_none(*p4d))
		return -EFAULT;
810
	pud = pud_offset(p4d, address);
811 812
	if (pud_none(*pud))
		return -EFAULT;
813
	pmd = pmd_offset(pud, address);
814
	if (!pmd_present(*pmd))
815 816 817 818 819 820 821 822 823 824 825 826 827 828
		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);
	}
829 830 831 832
	if (unlikely(!try_get_page(*page))) {
		ret = -ENOMEM;
		goto unmap;
	}
833 834 835 836 837 838 839
out:
	ret = 0;
unmap:
	pte_unmap(pte);
	return ret;
}

840 841 842 843 844
/*
 * 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.
 */
845 846 847 848
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;
849
	vm_fault_t ret;
850

Eric B Munson's avatar
Eric B Munson committed
851 852 853
	/* mlock all present pages, but do not fault in new pages */
	if ((*flags & (FOLL_POPULATE | FOLL_MLOCK)) == FOLL_MLOCK)
		return -ENOENT;
854 855
	if (*flags & FOLL_WRITE)
		fault_flags |= FAULT_FLAG_WRITE;
856 857
	if (*flags & FOLL_REMOTE)
		fault_flags |= FAULT_FLAG_REMOTE;
858 859 860 861
	if (nonblocking)
		fault_flags |= FAULT_FLAG_ALLOW_RETRY;
	if (*flags & FOLL_NOWAIT)
		fault_flags |= FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_RETRY_NOWAIT;
862 863 864 865
	if (*flags & FOLL_TRIED) {
		VM_WARN_ON_ONCE(fault_flags & FAULT_FLAG_ALLOW_RETRY);
		fault_flags |= FAULT_FLAG_TRIED;
	}
866

867
	ret = handle_mm_fault(vma, address, fault_flags);
868
	if (ret & VM_FAULT_ERROR) {
869 870 871 872
		int err = vm_fault_to_errno(ret, *flags);

		if (err)
			return err;
873 874 875 876 877 878 879 880 881 882 883
		BUG();
	}

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

	if (ret & VM_FAULT_RETRY) {
884
		if (nonblocking && !(fault_flags & FAULT_FLAG_RETRY_NOWAIT))
885 886 887 888 889 890 891 892 893 894 895 896 897 898
			*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))
899
		*flags |= FOLL_COW;
900 901 902
	return 0;
}

903 904 905
static int check_vma_flags(struct vm_area_struct *vma, unsigned long gup_flags)
{
	vm_flags_t vm_flags = vma->vm_flags;
906 907
	int write = (gup_flags & FOLL_WRITE);
	int foreign = (gup_flags & FOLL_REMOTE);
908 909 910 911

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

912 913 914
	if (gup_flags & FOLL_ANON && !vma_is_anonymous(vma))
		return -EFAULT;

915
	if (write) {
916 917 918 919 920 921 922 923 924 925 926 927
		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.
			 */
928
			if (!is_cow_mapping(vm_flags))
929 930 931 932 933 934 935 936 937 938 939 940
				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;
	}
941 942 943 944 945
	/*
	 * gups are always data accesses, not instruction
	 * fetches, so execute=false here
	 */
	if (!arch_vma_access_permitted(vma, write, false, foreign))
946
		return -EFAULT;
947 948 949
	return 0;
}

950 951 952 953 954 955 956 957 958 959 960 961 962 963
/**
 * __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
 *
964 965 966 967 968 969 970 971 972 973 974
 * 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.
975
 *
976
 * Must be called with mmap_sem held.  It may be released.  See below.
977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998
 *
 * __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,
999 1000 1001 1002 1003 1004 1005 1006
 * *@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.
1007 1008 1009 1010 1011
 *
 * 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.
 */
1012
static long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
1013 1014 1015 1016
		unsigned long start, unsigned long nr_pages,
		unsigned int gup_flags, struct page **pages,
		struct vm_area_struct **vmas, int *nonblocking)
{
1017
	long ret = 0, i = 0;
1018
	struct vm_area_struct *vma = NULL;
1019
	struct follow_page_context ctx = { NULL };
1020 1021 1022 1023

	if (!nr_pages)
		return 0;

1024 1025
	start = untagged_addr(start);

1026
	VM_BUG_ON(!!pages != !!(gup_flags & (FOLL_GET | FOLL_PIN)));
1027 1028 1029 1030 1031 1032 1033 1034 1035 1036

	/*
	 * 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 {
1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048
		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)
1049
					goto out;
1050
				ctx.page_mask = 0;
1051 1052
				goto next_page;
			}
1053

1054 1055 1056 1057
			if (!vma || check_vma_flags(vma, gup_flags)) {
				ret = -EFAULT;
				goto out;
			}
1058 1059 1060
			if (is_vm_hugetlb_page(vma)) {
				i = follow_hugetlb_page(mm, vma, pages, vmas,
						&start, &nr_pages, i,
1061
						gup_flags, nonblocking);
1062
				continue;
1063
			}
1064 1065 1066 1067 1068 1069
		}
retry:
		/*
		 * If we have a pending SIGKILL, don't keep faulting pages and
		 * potentially allocating memory.
		 */
1070
		if (fatal_signal_pending(current)) {
1071 1072 1073
			ret = -ERESTARTSYS;
			goto out;
		}
1074
		cond_resched();
1075 1076

		page = follow_page_mask(vma, start, foll_flags, &ctx);
1077 1078 1079 1080 1081 1082
		if (!page) {
			ret = faultin_page(tsk, vma, start, &foll_flags,
					nonblocking);
			switch (ret) {
			case 0:
				goto retry;
1083 1084 1085
			case -EBUSY:
				ret = 0;
				/* FALLTHRU */
1086 1087 1088
			case -EFAULT:
			case -ENOMEM:
			case -EHWPOISON:
1089
				goto out;
1090 1091
			case -ENOENT:
				goto next_page;
1092
			}
1093
			BUG();
1094 1095 1096 1097 1098 1099 1100
		} else if (PTR_ERR(page) == -EEXIST) {
			/*
			 * Proper page table entry exists, but no corresponding
			 * struct page.
			 */
			goto next_page;
		} else if (IS_ERR(page)) {
1101 1102
			ret = PTR_ERR(page);
			goto out;
1103
		}
1104 1105 1106 1107
		if (pages) {
			pages[i] = page;
			flush_anon_page(vma, page, start);
			flush_dcache_page(page);
1108
			ctx.page_mask = 0;
1109 1110
		}
next_page:
1111 1112
		if (vmas) {
			vmas[i] = vma;
1113
			ctx.page_mask = 0;
1114
		}
1115
		page_increm = 1 + (~(start >> PAGE_SHIFT) & ctx.page_mask);
1116 1117 1118 1119 1120
		if (page_increm > nr_pages)
			page_increm = nr_pages;
		i += page_increm;
		start += page_increm * PAGE_SIZE;
		nr_pages -= page_increm;
1121
	} while (nr_pages);
1122 1123 1124 1125
out:
	if (ctx.pgmap)
		put_dev_pagemap(ctx.pgmap);
	return i ? i : ret;
1126 1127
}

1128 1129
static bool vma_permits_fault(struct vm_area_struct *vma,
			      unsigned int fault_flags)
1130
{
1131 1132
	bool write   = !!(fault_flags & FAULT_FLAG_WRITE);
	bool foreign = !!(fault_flags & FAULT_FLAG_REMOTE);
1133
	vm_flags_t vm_flags = write ? VM_WRITE : VM_READ;
1134 1135 1136 1137

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

1138 1139
	/*
	 * The architecture might have a hardware protection
1140
	 * mechanism other than read/write that can deny access.
1141 1142 1143