mmap.c 70.8 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5
/*
 * mm/mmap.c
 *
 * Written by obz.
 *
A
Alan Cox 已提交
6
 * Address space accounting code	<alan@lxorguk.ukuu.org.uk>
L
Linus Torvalds 已提交
7 8 9
 */

#include <linux/slab.h>
A
Alexey Dobriyan 已提交
10
#include <linux/backing-dev.h>
L
Linus Torvalds 已提交
11 12 13 14 15 16
#include <linux/mm.h>
#include <linux/shm.h>
#include <linux/mman.h>
#include <linux/pagemap.h>
#include <linux/swap.h>
#include <linux/syscalls.h>
17
#include <linux/capability.h>
L
Linus Torvalds 已提交
18 19 20 21 22 23 24
#include <linux/init.h>
#include <linux/file.h>
#include <linux/fs.h>
#include <linux/personality.h>
#include <linux/security.h>
#include <linux/hugetlb.h>
#include <linux/profile.h>
25
#include <linux/export.h>
L
Linus Torvalds 已提交
26 27 28
#include <linux/mount.h>
#include <linux/mempolicy.h>
#include <linux/rmap.h>
A
Andrea Arcangeli 已提交
29
#include <linux/mmu_notifier.h>
30
#include <linux/perf_event.h>
A
Al Viro 已提交
31
#include <linux/audit.h>
A
Andrea Arcangeli 已提交
32
#include <linux/khugepaged.h>
33
#include <linux/uprobes.h>
L
Linus Torvalds 已提交
34 35 36 37

#include <asm/uaccess.h>
#include <asm/cacheflush.h>
#include <asm/tlb.h>
38
#include <asm/mmu_context.h>
L
Linus Torvalds 已提交
39

40 41
#include "internal.h"

42 43 44 45
#ifndef arch_mmap_check
#define arch_mmap_check(addr, len, flags)	(0)
#endif

46 47 48 49
#ifndef arch_rebalance_pgtables
#define arch_rebalance_pgtables(addr, len)		(addr)
#endif

50 51 52 53
static void unmap_region(struct mm_struct *mm,
		struct vm_area_struct *vma, struct vm_area_struct *prev,
		unsigned long start, unsigned long end);

L
Linus Torvalds 已提交
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
/*
 * WARNING: the debugging will use recursive algorithms so never enable this
 * unless you know what you are doing.
 */
#undef DEBUG_MM_RB

/* description of effects of mapping type and prot in current implementation.
 * this is due to the limited x86 page protection hardware.  The expected
 * behavior is in parens:
 *
 * map_type	prot
 *		PROT_NONE	PROT_READ	PROT_WRITE	PROT_EXEC
 * MAP_SHARED	r: (no) no	r: (yes) yes	r: (no) yes	r: (no) yes
 *		w: (no) no	w: (no) no	w: (yes) yes	w: (no) no
 *		x: (no) no	x: (no) yes	x: (no) yes	x: (yes) yes
 *		
 * MAP_PRIVATE	r: (no) no	r: (yes) yes	r: (no) yes	r: (no) yes
 *		w: (no) no	w: (no) no	w: (copy) copy	w: (no) no
 *		x: (no) no	x: (no) yes	x: (no) yes	x: (yes) yes
 *
 */
pgprot_t protection_map[16] = {
	__P000, __P001, __P010, __P011, __P100, __P101, __P110, __P111,
	__S000, __S001, __S010, __S011, __S100, __S101, __S110, __S111
};

80 81
pgprot_t vm_get_page_prot(unsigned long vm_flags)
{
82 83 84
	return __pgprot(pgprot_val(protection_map[vm_flags &
				(VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]) |
			pgprot_val(arch_vm_get_page_prot(vm_flags)));
85 86 87
}
EXPORT_SYMBOL(vm_get_page_prot);

88 89
int sysctl_overcommit_memory __read_mostly = OVERCOMMIT_GUESS;  /* heuristic overcommit */
int sysctl_overcommit_ratio __read_mostly = 50;	/* default is 50% */
90
int sysctl_max_map_count __read_mostly = DEFAULT_MAX_MAP_COUNT;
91 92 93 94 95
/*
 * Make sure vm_committed_as in one cacheline and not cacheline shared with
 * other variables. It can be updated by several CPUs frequently.
 */
struct percpu_counter vm_committed_as ____cacheline_aligned_in_smp;
L
Linus Torvalds 已提交
96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112

/*
 * Check that a process has enough memory to allocate a new virtual
 * mapping. 0 means there is enough memory for the allocation to
 * succeed and -ENOMEM implies there is not.
 *
 * We currently support three overcommit policies, which are set via the
 * vm.overcommit_memory sysctl.  See Documentation/vm/overcommit-accounting
 *
 * Strict overcommit modes added 2002 Feb 26 by Alan Cox.
 * Additional code 2002 Jul 20 by Robert Love.
 *
 * cap_sys_admin is 1 if the process has admin privileges, 0 otherwise.
 *
 * Note this is a helper function intended to be used by LSMs which
 * wish to use this logic.
 */
113
int __vm_enough_memory(struct mm_struct *mm, long pages, int cap_sys_admin)
L
Linus Torvalds 已提交
114 115 116 117 118 119 120 121 122 123 124 125
{
	unsigned long free, allowed;

	vm_acct_memory(pages);

	/*
	 * Sometimes we want to use more memory than we have
	 */
	if (sysctl_overcommit_memory == OVERCOMMIT_ALWAYS)
		return 0;

	if (sysctl_overcommit_memory == OVERCOMMIT_GUESS) {
126 127 128 129 130 131 132 133 134 135
		free = global_page_state(NR_FREE_PAGES);
		free += global_page_state(NR_FILE_PAGES);

		/*
		 * shmem pages shouldn't be counted as free in this
		 * case, they can't be purged, only swapped out, and
		 * that won't affect the overall amount of available
		 * memory in the system.
		 */
		free -= global_page_state(NR_SHMEM);
L
Linus Torvalds 已提交
136 137 138 139 140 141 142 143 144

		free += nr_swap_pages;

		/*
		 * Any slabs which are created with the
		 * SLAB_RECLAIM_ACCOUNT flag claim to have contents
		 * which are reclaimable, under pressure.  The dentry
		 * cache and most inode caches should fall into this
		 */
145
		free += global_page_state(NR_SLAB_RECLAIMABLE);
L
Linus Torvalds 已提交
146

147 148 149
		/*
		 * Leave reserved pages. The pages are not for anonymous pages.
		 */
150
		if (free <= totalreserve_pages)
151 152
			goto error;
		else
153
			free -= totalreserve_pages;
154 155 156 157

		/*
		 * Leave the last 3% for root
		 */
L
Linus Torvalds 已提交
158
		if (!cap_sys_admin)
159
			free -= free / 32;
L
Linus Torvalds 已提交
160 161 162

		if (free > pages)
			return 0;
163 164

		goto error;
L
Linus Torvalds 已提交
165 166 167 168 169 170 171 172 173 174 175 176 177
	}

	allowed = (totalram_pages - hugetlb_total_pages())
	       	* sysctl_overcommit_ratio / 100;
	/*
	 * Leave the last 3% for root
	 */
	if (!cap_sys_admin)
		allowed -= allowed / 32;
	allowed += total_swap_pages;

	/* Don't let a single process grow too big:
	   leave 3% of the size of this process for other processes */
A
Alan Cox 已提交
178 179
	if (mm)
		allowed -= mm->total_vm / 32;
L
Linus Torvalds 已提交
180

181
	if (percpu_counter_read_positive(&vm_committed_as) < allowed)
L
Linus Torvalds 已提交
182
		return 0;
183
error:
L
Linus Torvalds 已提交
184 185 186 187 188 189
	vm_unacct_memory(pages);

	return -ENOMEM;
}

/*
190
 * Requires inode->i_mapping->i_mmap_mutex
L
Linus Torvalds 已提交
191 192 193 194 195
 */
static void __remove_shared_vm_struct(struct vm_area_struct *vma,
		struct file *file, struct address_space *mapping)
{
	if (vma->vm_flags & VM_DENYWRITE)
196
		atomic_inc(&file->f_path.dentry->d_inode->i_writecount);
L
Linus Torvalds 已提交
197 198 199 200 201 202 203 204 205 206 207 208
	if (vma->vm_flags & VM_SHARED)
		mapping->i_mmap_writable--;

	flush_dcache_mmap_lock(mapping);
	if (unlikely(vma->vm_flags & VM_NONLINEAR))
		list_del_init(&vma->shared.vm_set.list);
	else
		vma_prio_tree_remove(vma, &mapping->i_mmap);
	flush_dcache_mmap_unlock(mapping);
}

/*
209 210
 * Unlink a file-based vm structure from its prio_tree, to hide
 * vma from rmap and vmtruncate before freeing its page tables.
L
Linus Torvalds 已提交
211
 */
212
void unlink_file_vma(struct vm_area_struct *vma)
L
Linus Torvalds 已提交
213 214 215 216 217
{
	struct file *file = vma->vm_file;

	if (file) {
		struct address_space *mapping = file->f_mapping;
218
		mutex_lock(&mapping->i_mmap_mutex);
L
Linus Torvalds 已提交
219
		__remove_shared_vm_struct(vma, file, mapping);
220
		mutex_unlock(&mapping->i_mmap_mutex);
L
Linus Torvalds 已提交
221
	}
222 223 224 225 226 227 228 229 230 231
}

/*
 * Close a vm structure and free it, returning the next.
 */
static struct vm_area_struct *remove_vma(struct vm_area_struct *vma)
{
	struct vm_area_struct *next = vma->vm_next;

	might_sleep();
L
Linus Torvalds 已提交
232 233
	if (vma->vm_ops && vma->vm_ops->close)
		vma->vm_ops->close(vma);
M
Matt Helsley 已提交
234
	if (vma->vm_file) {
235
		fput(vma->vm_file);
M
Matt Helsley 已提交
236 237 238
		if (vma->vm_flags & VM_EXECUTABLE)
			removed_exe_file_vma(vma->vm_mm);
	}
239
	mpol_put(vma_policy(vma));
L
Linus Torvalds 已提交
240
	kmem_cache_free(vm_area_cachep, vma);
241
	return next;
L
Linus Torvalds 已提交
242 243
}

244 245
static unsigned long do_brk(unsigned long addr, unsigned long len);

246
SYSCALL_DEFINE1(brk, unsigned long, brk)
L
Linus Torvalds 已提交
247 248 249 250
{
	unsigned long rlim, retval;
	unsigned long newbrk, oldbrk;
	struct mm_struct *mm = current->mm;
251
	unsigned long min_brk;
L
Linus Torvalds 已提交
252 253 254

	down_write(&mm->mmap_sem);

255
#ifdef CONFIG_COMPAT_BRK
256 257 258 259 260
	/*
	 * CONFIG_COMPAT_BRK can still be overridden by setting
	 * randomize_va_space to 2, which will still cause mm->start_brk
	 * to be arbitrarily shifted
	 */
261
	if (current->brk_randomized)
262 263 264
		min_brk = mm->start_brk;
	else
		min_brk = mm->end_data;
265 266 267 268
#else
	min_brk = mm->start_brk;
#endif
	if (brk < min_brk)
L
Linus Torvalds 已提交
269
		goto out;
R
Ram Gupta 已提交
270 271 272 273 274 275 276

	/*
	 * Check against rlimit here. If this check is done later after the test
	 * of oldbrk with newbrk then it can escape the test and let the data
	 * segment grow beyond its set limit the in case where the limit is
	 * not page aligned -Ram Gupta
	 */
J
Jiri Slaby 已提交
277
	rlim = rlimit(RLIMIT_DATA);
J
Jiri Kosina 已提交
278 279
	if (rlim < RLIM_INFINITY && (brk - mm->start_brk) +
			(mm->end_data - mm->start_data) > rlim)
R
Ram Gupta 已提交
280 281
		goto out;

L
Linus Torvalds 已提交
282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326
	newbrk = PAGE_ALIGN(brk);
	oldbrk = PAGE_ALIGN(mm->brk);
	if (oldbrk == newbrk)
		goto set_brk;

	/* Always allow shrinking brk. */
	if (brk <= mm->brk) {
		if (!do_munmap(mm, newbrk, oldbrk-newbrk))
			goto set_brk;
		goto out;
	}

	/* Check against existing mmap mappings. */
	if (find_vma_intersection(mm, oldbrk, newbrk+PAGE_SIZE))
		goto out;

	/* Ok, looks good - let it rip. */
	if (do_brk(oldbrk, newbrk-oldbrk) != oldbrk)
		goto out;
set_brk:
	mm->brk = brk;
out:
	retval = mm->brk;
	up_write(&mm->mmap_sem);
	return retval;
}

#ifdef DEBUG_MM_RB
static int browse_rb(struct rb_root *root)
{
	int i = 0, j;
	struct rb_node *nd, *pn = NULL;
	unsigned long prev = 0, pend = 0;

	for (nd = rb_first(root); nd; nd = rb_next(nd)) {
		struct vm_area_struct *vma;
		vma = rb_entry(nd, struct vm_area_struct, vm_rb);
		if (vma->vm_start < prev)
			printk("vm_start %lx prev %lx\n", vma->vm_start, prev), i = -1;
		if (vma->vm_start < pend)
			printk("vm_start %lx pend %lx\n", vma->vm_start, pend);
		if (vma->vm_start > vma->vm_end)
			printk("vm_end %lx < vm_start %lx\n", vma->vm_end, vma->vm_start);
		i++;
		pn = nd;
D
David Miller 已提交
327 328
		prev = vma->vm_start;
		pend = vma->vm_end;
L
Linus Torvalds 已提交
329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352
	}
	j = 0;
	for (nd = pn; nd; nd = rb_prev(nd)) {
		j++;
	}
	if (i != j)
		printk("backwards %d, forwards %d\n", j, i), i = 0;
	return i;
}

void validate_mm(struct mm_struct *mm)
{
	int bug = 0;
	int i = 0;
	struct vm_area_struct *tmp = mm->mmap;
	while (tmp) {
		tmp = tmp->vm_next;
		i++;
	}
	if (i != mm->map_count)
		printk("map_count %d vm_next %d\n", mm->map_count, i), bug = 1;
	i = browse_rb(&mm->mm_rb);
	if (i != mm->map_count)
		printk("map_count %d rb %d\n", mm->map_count, i), bug = 1;
353
	BUG_ON(bug);
L
Linus Torvalds 已提交
354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379
}
#else
#define validate_mm(mm) do { } while (0)
#endif

static struct vm_area_struct *
find_vma_prepare(struct mm_struct *mm, unsigned long addr,
		struct vm_area_struct **pprev, struct rb_node ***rb_link,
		struct rb_node ** rb_parent)
{
	struct vm_area_struct * vma;
	struct rb_node ** __rb_link, * __rb_parent, * rb_prev;

	__rb_link = &mm->mm_rb.rb_node;
	rb_prev = __rb_parent = NULL;
	vma = NULL;

	while (*__rb_link) {
		struct vm_area_struct *vma_tmp;

		__rb_parent = *__rb_link;
		vma_tmp = rb_entry(__rb_parent, struct vm_area_struct, vm_rb);

		if (vma_tmp->vm_end > addr) {
			vma = vma_tmp;
			if (vma_tmp->vm_start <= addr)
380
				break;
L
Linus Torvalds 已提交
381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402
			__rb_link = &__rb_parent->rb_left;
		} else {
			rb_prev = __rb_parent;
			__rb_link = &__rb_parent->rb_right;
		}
	}

	*pprev = NULL;
	if (rb_prev)
		*pprev = rb_entry(rb_prev, struct vm_area_struct, vm_rb);
	*rb_link = __rb_link;
	*rb_parent = __rb_parent;
	return vma;
}

void __vma_link_rb(struct mm_struct *mm, struct vm_area_struct *vma,
		struct rb_node **rb_link, struct rb_node *rb_parent)
{
	rb_link_node(&vma->vm_rb, rb_parent, rb_link);
	rb_insert_color(&vma->vm_rb, &mm->mm_rb);
}

D
Denys Vlasenko 已提交
403
static void __vma_link_file(struct vm_area_struct *vma)
L
Linus Torvalds 已提交
404
{
Z
ZhenwenXu 已提交
405
	struct file *file;
L
Linus Torvalds 已提交
406 407 408 409 410 411

	file = vma->vm_file;
	if (file) {
		struct address_space *mapping = file->f_mapping;

		if (vma->vm_flags & VM_DENYWRITE)
412
			atomic_dec(&file->f_path.dentry->d_inode->i_writecount);
L
Linus Torvalds 已提交
413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442
		if (vma->vm_flags & VM_SHARED)
			mapping->i_mmap_writable++;

		flush_dcache_mmap_lock(mapping);
		if (unlikely(vma->vm_flags & VM_NONLINEAR))
			vma_nonlinear_insert(vma, &mapping->i_mmap_nonlinear);
		else
			vma_prio_tree_insert(vma, &mapping->i_mmap);
		flush_dcache_mmap_unlock(mapping);
	}
}

static void
__vma_link(struct mm_struct *mm, struct vm_area_struct *vma,
	struct vm_area_struct *prev, struct rb_node **rb_link,
	struct rb_node *rb_parent)
{
	__vma_link_list(mm, vma, prev, rb_parent);
	__vma_link_rb(mm, vma, rb_link, rb_parent);
}

static void vma_link(struct mm_struct *mm, struct vm_area_struct *vma,
			struct vm_area_struct *prev, struct rb_node **rb_link,
			struct rb_node *rb_parent)
{
	struct address_space *mapping = NULL;

	if (vma->vm_file)
		mapping = vma->vm_file->f_mapping;

443
	if (mapping)
444
		mutex_lock(&mapping->i_mmap_mutex);
L
Linus Torvalds 已提交
445 446 447 448 449

	__vma_link(mm, vma, prev, rb_link, rb_parent);
	__vma_link_file(vma);

	if (mapping)
450
		mutex_unlock(&mapping->i_mmap_mutex);
L
Linus Torvalds 已提交
451 452 453 454 455 456

	mm->map_count++;
	validate_mm(mm);
}

/*
457 458
 * Helper for vma_adjust() in the split_vma insert case: insert a vma into the
 * mm's list and rbtree.  It has already been inserted into the prio_tree.
L
Linus Torvalds 已提交
459
 */
Z
ZhenwenXu 已提交
460
static void __insert_vm_struct(struct mm_struct *mm, struct vm_area_struct *vma)
L
Linus Torvalds 已提交
461
{
Z
ZhenwenXu 已提交
462 463
	struct vm_area_struct *__vma, *prev;
	struct rb_node **rb_link, *rb_parent;
L
Linus Torvalds 已提交
464 465

	__vma = find_vma_prepare(mm, vma->vm_start,&prev, &rb_link, &rb_parent);
466
	BUG_ON(__vma && __vma->vm_start < vma->vm_end);
L
Linus Torvalds 已提交
467 468 469 470 471 472 473 474
	__vma_link(mm, vma, prev, rb_link, rb_parent);
	mm->map_count++;
}

static inline void
__vma_unlink(struct mm_struct *mm, struct vm_area_struct *vma,
		struct vm_area_struct *prev)
{
475 476 477 478 479
	struct vm_area_struct *next = vma->vm_next;

	prev->vm_next = next;
	if (next)
		next->vm_prev = prev;
L
Linus Torvalds 已提交
480 481 482 483 484 485 486 487 488 489 490 491
	rb_erase(&vma->vm_rb, &mm->mm_rb);
	if (mm->mmap_cache == vma)
		mm->mmap_cache = prev;
}

/*
 * We cannot adjust vm_start, vm_end, vm_pgoff fields of a vma that
 * is already present in an i_mmap tree without adjusting the tree.
 * The following helper function should be used when such adjustments
 * are necessary.  The "insert" vma (if any) is to be inserted
 * before we drop the necessary locks.
 */
492
int vma_adjust(struct vm_area_struct *vma, unsigned long start,
L
Linus Torvalds 已提交
493 494 495 496 497 498 499
	unsigned long end, pgoff_t pgoff, struct vm_area_struct *insert)
{
	struct mm_struct *mm = vma->vm_mm;
	struct vm_area_struct *next = vma->vm_next;
	struct vm_area_struct *importer = NULL;
	struct address_space *mapping = NULL;
	struct prio_tree_root *root = NULL;
500
	struct anon_vma *anon_vma = NULL;
L
Linus Torvalds 已提交
501 502 503 504 505
	struct file *file = vma->vm_file;
	long adjust_next = 0;
	int remove_next = 0;

	if (next && !insert) {
506 507
		struct vm_area_struct *exporter = NULL;

L
Linus Torvalds 已提交
508 509 510 511 512 513 514
		if (end >= next->vm_end) {
			/*
			 * vma expands, overlapping all the next, and
			 * perhaps the one after too (mprotect case 6).
			 */
again:			remove_next = 1 + (end > next->vm_end);
			end = next->vm_end;
515
			exporter = next;
L
Linus Torvalds 已提交
516 517 518 519 520 521 522
			importer = vma;
		} else if (end > next->vm_start) {
			/*
			 * vma expands, overlapping part of the next:
			 * mprotect case 5 shifting the boundary up.
			 */
			adjust_next = (end - next->vm_start) >> PAGE_SHIFT;
523
			exporter = next;
L
Linus Torvalds 已提交
524 525 526 527 528 529 530 531
			importer = vma;
		} else if (end < vma->vm_end) {
			/*
			 * vma shrinks, and !insert tells it's not
			 * split_vma inserting another: so it must be
			 * mprotect case 4 shifting the boundary down.
			 */
			adjust_next = - ((vma->vm_end - end) >> PAGE_SHIFT);
532
			exporter = vma;
L
Linus Torvalds 已提交
533 534 535
			importer = next;
		}

536 537 538 539 540
		/*
		 * Easily overlooked: when mprotect shifts the boundary,
		 * make sure the expanding vma has anon_vma set if the
		 * shrinking vma had, to cover any anon pages imported.
		 */
541 542
		if (exporter && exporter->anon_vma && !importer->anon_vma) {
			if (anon_vma_clone(importer, exporter))
543
				return -ENOMEM;
544
			importer->anon_vma = exporter->anon_vma;
545 546 547
		}
	}

L
Linus Torvalds 已提交
548 549
	if (file) {
		mapping = file->f_mapping;
550
		if (!(vma->vm_flags & VM_NONLINEAR)) {
L
Linus Torvalds 已提交
551
			root = &mapping->i_mmap;
552
			uprobe_munmap(vma, vma->vm_start, vma->vm_end);
553 554

			if (adjust_next)
555 556
				uprobe_munmap(next, next->vm_start,
							next->vm_end);
557 558
		}

559
		mutex_lock(&mapping->i_mmap_mutex);
L
Linus Torvalds 已提交
560 561 562 563 564 565 566 567 568 569 570
		if (insert) {
			/*
			 * Put into prio_tree now, so instantiated pages
			 * are visible to arm/parisc __flush_dcache_page
			 * throughout; but we cannot insert into address
			 * space until vma start or end is updated.
			 */
			__vma_link_file(insert);
		}
	}

571 572
	vma_adjust_trans_huge(vma, start, end, adjust_next);

573 574 575 576 577 578
	/*
	 * When changing only vma->vm_end, we don't really need anon_vma
	 * lock. This is a fairly rare case by itself, but the anon_vma
	 * lock may be shared between many sibling processes.  Skipping
	 * the lock for brk adjustments makes a difference sometimes.
	 */
579
	if (vma->anon_vma && (importer || start != vma->vm_start)) {
580 581 582 583
		anon_vma = vma->anon_vma;
		anon_vma_lock(anon_vma);
	}

L
Linus Torvalds 已提交
584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622
	if (root) {
		flush_dcache_mmap_lock(mapping);
		vma_prio_tree_remove(vma, root);
		if (adjust_next)
			vma_prio_tree_remove(next, root);
	}

	vma->vm_start = start;
	vma->vm_end = end;
	vma->vm_pgoff = pgoff;
	if (adjust_next) {
		next->vm_start += adjust_next << PAGE_SHIFT;
		next->vm_pgoff += adjust_next;
	}

	if (root) {
		if (adjust_next)
			vma_prio_tree_insert(next, root);
		vma_prio_tree_insert(vma, root);
		flush_dcache_mmap_unlock(mapping);
	}

	if (remove_next) {
		/*
		 * vma_merge has merged next into vma, and needs
		 * us to remove next before dropping the locks.
		 */
		__vma_unlink(mm, next, vma);
		if (file)
			__remove_shared_vm_struct(next, file, mapping);
	} else if (insert) {
		/*
		 * split_vma has split insert from vma, and needs
		 * us to insert it before dropping the locks
		 * (it may either follow vma or precede it).
		 */
		__insert_vm_struct(mm, insert);
	}

623 624
	if (anon_vma)
		anon_vma_unlock(anon_vma);
L
Linus Torvalds 已提交
625
	if (mapping)
626
		mutex_unlock(&mapping->i_mmap_mutex);
L
Linus Torvalds 已提交
627

628
	if (root) {
629
		uprobe_mmap(vma);
630 631

		if (adjust_next)
632
			uprobe_mmap(next);
633 634
	}

L
Linus Torvalds 已提交
635
	if (remove_next) {
M
Matt Helsley 已提交
636
		if (file) {
637
			uprobe_munmap(next, next->vm_start, next->vm_end);
L
Linus Torvalds 已提交
638
			fput(file);
M
Matt Helsley 已提交
639 640 641
			if (next->vm_flags & VM_EXECUTABLE)
				removed_exe_file_vma(mm);
		}
642 643
		if (next->anon_vma)
			anon_vma_merge(vma, next);
L
Linus Torvalds 已提交
644
		mm->map_count--;
645
		mpol_put(vma_policy(next));
L
Linus Torvalds 已提交
646 647 648 649 650 651 652 653 654 655 656
		kmem_cache_free(vm_area_cachep, next);
		/*
		 * In mprotect's case 6 (see comments on vma_merge),
		 * we must remove another next too. It would clutter
		 * up the code too much to do both in one go.
		 */
		if (remove_next == 2) {
			next = vma->vm_next;
			goto again;
		}
	}
657
	if (insert && file)
658
		uprobe_mmap(insert);
L
Linus Torvalds 已提交
659 660

	validate_mm(mm);
661 662

	return 0;
L
Linus Torvalds 已提交
663 664 665 666 667 668 669 670 671
}

/*
 * If the vma has a ->close operation then the driver probably needs to release
 * per-vma resources, so we don't attempt to merge those.
 */
static inline int is_mergeable_vma(struct vm_area_struct *vma,
			struct file *file, unsigned long vm_flags)
{
H
Hugh Dickins 已提交
672 673
	/* VM_CAN_NONLINEAR may get set later by f_op->mmap() */
	if ((vma->vm_flags ^ vm_flags) & ~VM_CAN_NONLINEAR)
L
Linus Torvalds 已提交
674 675 676 677 678 679 680 681 682
		return 0;
	if (vma->vm_file != file)
		return 0;
	if (vma->vm_ops && vma->vm_ops->close)
		return 0;
	return 1;
}

static inline int is_mergeable_anon_vma(struct anon_vma *anon_vma1,
S
Shaohua Li 已提交
683 684
					struct anon_vma *anon_vma2,
					struct vm_area_struct *vma)
L
Linus Torvalds 已提交
685
{
S
Shaohua Li 已提交
686 687 688 689 690 691 692 693
	/*
	 * The list_is_singular() test is to avoid merging VMA cloned from
	 * parents. This can improve scalability caused by anon_vma lock.
	 */
	if ((!anon_vma1 || !anon_vma2) && (!vma ||
		list_is_singular(&vma->anon_vma_chain)))
		return 1;
	return anon_vma1 == anon_vma2;
L
Linus Torvalds 已提交
694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711
}

/*
 * Return true if we can merge this (vm_flags,anon_vma,file,vm_pgoff)
 * in front of (at a lower virtual address and file offset than) the vma.
 *
 * We cannot merge two vmas if they have differently assigned (non-NULL)
 * anon_vmas, nor if same anon_vma is assigned but offsets incompatible.
 *
 * We don't check here for the merged mmap wrapping around the end of pagecache
 * indices (16TB on ia32) because do_mmap_pgoff() does not permit mmap's which
 * wrap, nor mmaps which cover the final page at index -1UL.
 */
static int
can_vma_merge_before(struct vm_area_struct *vma, unsigned long vm_flags,
	struct anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff)
{
	if (is_mergeable_vma(vma, file, vm_flags) &&
S
Shaohua Li 已提交
712
	    is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) {
L
Linus Torvalds 已提交
713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730
		if (vma->vm_pgoff == vm_pgoff)
			return 1;
	}
	return 0;
}

/*
 * Return true if we can merge this (vm_flags,anon_vma,file,vm_pgoff)
 * beyond (at a higher virtual address and file offset than) the vma.
 *
 * We cannot merge two vmas if they have differently assigned (non-NULL)
 * anon_vmas, nor if same anon_vma is assigned but offsets incompatible.
 */
static int
can_vma_merge_after(struct vm_area_struct *vma, unsigned long vm_flags,
	struct anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff)
{
	if (is_mergeable_vma(vma, file, vm_flags) &&
S
Shaohua Li 已提交
731
	    is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) {
L
Linus Torvalds 已提交
732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776
		pgoff_t vm_pglen;
		vm_pglen = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
		if (vma->vm_pgoff + vm_pglen == vm_pgoff)
			return 1;
	}
	return 0;
}

/*
 * Given a mapping request (addr,end,vm_flags,file,pgoff), figure out
 * whether that can be merged with its predecessor or its successor.
 * Or both (it neatly fills a hole).
 *
 * In most cases - when called for mmap, brk or mremap - [addr,end) is
 * certain not to be mapped by the time vma_merge is called; but when
 * called for mprotect, it is certain to be already mapped (either at
 * an offset within prev, or at the start of next), and the flags of
 * this area are about to be changed to vm_flags - and the no-change
 * case has already been eliminated.
 *
 * The following mprotect cases have to be considered, where AAAA is
 * the area passed down from mprotect_fixup, never extending beyond one
 * vma, PPPPPP is the prev vma specified, and NNNNNN the next vma after:
 *
 *     AAAA             AAAA                AAAA          AAAA
 *    PPPPPPNNNNNN    PPPPPPNNNNNN    PPPPPPNNNNNN    PPPPNNNNXXXX
 *    cannot merge    might become    might become    might become
 *                    PPNNNNNNNNNN    PPPPPPPPPPNN    PPPPPPPPPPPP 6 or
 *    mmap, brk or    case 4 below    case 5 below    PPPPPPPPXXXX 7 or
 *    mremap move:                                    PPPPNNNNNNNN 8
 *        AAAA
 *    PPPP    NNNN    PPPPPPPPPPPP    PPPPPPPPNNNN    PPPPNNNNNNNN
 *    might become    case 1 below    case 2 below    case 3 below
 *
 * Odd one out? Case 8, because it extends NNNN but needs flags of XXXX:
 * mprotect_fixup updates vm_flags & vm_page_prot on successful return.
 */
struct vm_area_struct *vma_merge(struct mm_struct *mm,
			struct vm_area_struct *prev, unsigned long addr,
			unsigned long end, unsigned long vm_flags,
		     	struct anon_vma *anon_vma, struct file *file,
			pgoff_t pgoff, struct mempolicy *policy)
{
	pgoff_t pglen = (end - addr) >> PAGE_SHIFT;
	struct vm_area_struct *area, *next;
777
	int err;
L
Linus Torvalds 已提交
778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808

	/*
	 * We later require that vma->vm_flags == vm_flags,
	 * so this tests vma->vm_flags & VM_SPECIAL, too.
	 */
	if (vm_flags & VM_SPECIAL)
		return NULL;

	if (prev)
		next = prev->vm_next;
	else
		next = mm->mmap;
	area = next;
	if (next && next->vm_end == end)		/* cases 6, 7, 8 */
		next = next->vm_next;

	/*
	 * Can it merge with the predecessor?
	 */
	if (prev && prev->vm_end == addr &&
  			mpol_equal(vma_policy(prev), policy) &&
			can_vma_merge_after(prev, vm_flags,
						anon_vma, file, pgoff)) {
		/*
		 * OK, it can.  Can we now merge in the successor as well?
		 */
		if (next && end == next->vm_start &&
				mpol_equal(policy, vma_policy(next)) &&
				can_vma_merge_before(next, vm_flags,
					anon_vma, file, pgoff+pglen) &&
				is_mergeable_anon_vma(prev->anon_vma,
S
Shaohua Li 已提交
809
						      next->anon_vma, NULL)) {
L
Linus Torvalds 已提交
810
							/* cases 1, 6 */
811
			err = vma_adjust(prev, prev->vm_start,
L
Linus Torvalds 已提交
812 813
				next->vm_end, prev->vm_pgoff, NULL);
		} else					/* cases 2, 5, 7 */
814
			err = vma_adjust(prev, prev->vm_start,
L
Linus Torvalds 已提交
815
				end, prev->vm_pgoff, NULL);
816 817
		if (err)
			return NULL;
A
Andrea Arcangeli 已提交
818
		khugepaged_enter_vma_merge(prev);
L
Linus Torvalds 已提交
819 820 821 822 823 824 825 826 827 828 829
		return prev;
	}

	/*
	 * Can this new request be merged in front of next?
	 */
	if (next && end == next->vm_start &&
 			mpol_equal(policy, vma_policy(next)) &&
			can_vma_merge_before(next, vm_flags,
					anon_vma, file, pgoff+pglen)) {
		if (prev && addr < prev->vm_end)	/* case 4 */
830
			err = vma_adjust(prev, prev->vm_start,
L
Linus Torvalds 已提交
831 832
				addr, prev->vm_pgoff, NULL);
		else					/* cases 3, 8 */
833
			err = vma_adjust(area, addr, next->vm_end,
L
Linus Torvalds 已提交
834
				next->vm_pgoff - pglen, NULL);
835 836
		if (err)
			return NULL;
A
Andrea Arcangeli 已提交
837
		khugepaged_enter_vma_merge(area);
L
Linus Torvalds 已提交
838 839 840 841 842 843
		return area;
	}

	return NULL;
}

844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898
/*
 * Rough compatbility check to quickly see if it's even worth looking
 * at sharing an anon_vma.
 *
 * They need to have the same vm_file, and the flags can only differ
 * in things that mprotect may change.
 *
 * NOTE! The fact that we share an anon_vma doesn't _have_ to mean that
 * we can merge the two vma's. For example, we refuse to merge a vma if
 * there is a vm_ops->close() function, because that indicates that the
 * driver is doing some kind of reference counting. But that doesn't
 * really matter for the anon_vma sharing case.
 */
static int anon_vma_compatible(struct vm_area_struct *a, struct vm_area_struct *b)
{
	return a->vm_end == b->vm_start &&
		mpol_equal(vma_policy(a), vma_policy(b)) &&
		a->vm_file == b->vm_file &&
		!((a->vm_flags ^ b->vm_flags) & ~(VM_READ|VM_WRITE|VM_EXEC)) &&
		b->vm_pgoff == a->vm_pgoff + ((b->vm_start - a->vm_start) >> PAGE_SHIFT);
}

/*
 * Do some basic sanity checking to see if we can re-use the anon_vma
 * from 'old'. The 'a'/'b' vma's are in VM order - one of them will be
 * the same as 'old', the other will be the new one that is trying
 * to share the anon_vma.
 *
 * NOTE! This runs with mm_sem held for reading, so it is possible that
 * the anon_vma of 'old' is concurrently in the process of being set up
 * by another page fault trying to merge _that_. But that's ok: if it
 * is being set up, that automatically means that it will be a singleton
 * acceptable for merging, so we can do all of this optimistically. But
 * we do that ACCESS_ONCE() to make sure that we never re-load the pointer.
 *
 * IOW: that the "list_is_singular()" test on the anon_vma_chain only
 * matters for the 'stable anon_vma' case (ie the thing we want to avoid
 * is to return an anon_vma that is "complex" due to having gone through
 * a fork).
 *
 * We also make sure that the two vma's are compatible (adjacent,
 * and with the same memory policies). That's all stable, even with just
 * a read lock on the mm_sem.
 */
static struct anon_vma *reusable_anon_vma(struct vm_area_struct *old, struct vm_area_struct *a, struct vm_area_struct *b)
{
	if (anon_vma_compatible(a, b)) {
		struct anon_vma *anon_vma = ACCESS_ONCE(old->anon_vma);

		if (anon_vma && list_is_singular(&old->anon_vma_chain))
			return anon_vma;
	}
	return NULL;
}

L
Linus Torvalds 已提交
899 900 901 902 903 904 905 906 907 908
/*
 * find_mergeable_anon_vma is used by anon_vma_prepare, to check
 * neighbouring vmas for a suitable anon_vma, before it goes off
 * to allocate a new anon_vma.  It checks because a repetitive
 * sequence of mprotects and faults may otherwise lead to distinct
 * anon_vmas being allocated, preventing vma merge in subsequent
 * mprotect.
 */
struct anon_vma *find_mergeable_anon_vma(struct vm_area_struct *vma)
{
909
	struct anon_vma *anon_vma;
L
Linus Torvalds 已提交
910 911 912 913 914 915
	struct vm_area_struct *near;

	near = vma->vm_next;
	if (!near)
		goto try_prev;

916 917 918
	anon_vma = reusable_anon_vma(near, vma, near);
	if (anon_vma)
		return anon_vma;
L
Linus Torvalds 已提交
919
try_prev:
920
	near = vma->vm_prev;
L
Linus Torvalds 已提交
921 922 923
	if (!near)
		goto none;

924 925 926
	anon_vma = reusable_anon_vma(near, near, vma);
	if (anon_vma)
		return anon_vma;
L
Linus Torvalds 已提交
927 928 929 930 931 932 933 934 935 936 937 938 939
none:
	/*
	 * There's no absolute need to look only at touching neighbours:
	 * we could search further afield for "compatible" anon_vmas.
	 * But it would probably just be a waste of time searching,
	 * or lead to too many vmas hanging off the same anon_vma.
	 * We're trying to allow mprotect remerging later on,
	 * not trying to minimize memory used for anon_vmas.
	 */
	return NULL;
}

#ifdef CONFIG_PROC_FS
940
void vm_stat_account(struct mm_struct *mm, unsigned long flags,
L
Linus Torvalds 已提交
941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956
						struct file *file, long pages)
{
	const unsigned long stack_flags
		= VM_STACK_FLAGS & (VM_GROWSUP|VM_GROWSDOWN);

	if (file) {
		mm->shared_vm += pages;
		if ((flags & (VM_EXEC|VM_WRITE)) == VM_EXEC)
			mm->exec_vm += pages;
	} else if (flags & stack_flags)
		mm->stack_vm += pages;
	if (flags & (VM_RESERVED|VM_IO))
		mm->reserved_vm += pages;
}
#endif /* CONFIG_PROC_FS */

A
Al Viro 已提交
957 958 959 960 961 962 963 964 965 966 967 968 969
/*
 * If a hint addr is less than mmap_min_addr change hint to be as
 * low as possible but still greater than mmap_min_addr
 */
static inline unsigned long round_hint_to_min(unsigned long hint)
{
	hint &= PAGE_MASK;
	if (((void *)hint != NULL) &&
	    (hint < mmap_min_addr))
		return PAGE_ALIGN(mmap_min_addr);
	return hint;
}

L
Linus Torvalds 已提交
970
/*
971
 * The caller must hold down_write(&current->mm->mmap_sem).
L
Linus Torvalds 已提交
972 973
 */

974
unsigned long do_mmap_pgoff(struct file *file, unsigned long addr,
L
Linus Torvalds 已提交
975 976 977 978 979
			unsigned long len, unsigned long prot,
			unsigned long flags, unsigned long pgoff)
{
	struct mm_struct * mm = current->mm;
	struct inode *inode;
980
	vm_flags_t vm_flags;
L
Linus Torvalds 已提交
981 982 983 984 985 986 987 988

	/*
	 * Does the application expect PROT_READ to imply PROT_EXEC?
	 *
	 * (the exception is when the underlying filesystem is noexec
	 *  mounted, in which case we dont add PROT_EXEC.)
	 */
	if ((prot & PROT_READ) && (current->personality & READ_IMPLIES_EXEC))
989
		if (!(file && (file->f_path.mnt->mnt_flags & MNT_NOEXEC)))
L
Linus Torvalds 已提交
990 991 992 993 994
			prot |= PROT_EXEC;

	if (!len)
		return -EINVAL;

995 996 997
	if (!(flags & MAP_FIXED))
		addr = round_hint_to_min(addr);

L
Linus Torvalds 已提交
998 999
	/* Careful about overflows.. */
	len = PAGE_ALIGN(len);
1000
	if (!len)
L
Linus Torvalds 已提交
1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024
		return -ENOMEM;

	/* offset overflow? */
	if ((pgoff + (len >> PAGE_SHIFT)) < pgoff)
               return -EOVERFLOW;

	/* Too many mappings? */
	if (mm->map_count > sysctl_max_map_count)
		return -ENOMEM;

	/* Obtain the address to map to. we verify (or select) it and ensure
	 * that it represents a valid section of the address space.
	 */
	addr = get_unmapped_area(file, addr, len, pgoff, flags);
	if (addr & ~PAGE_MASK)
		return addr;

	/* Do simple checking here so the lower-level routines won't have
	 * to. we assume access permissions have been handled by the open
	 * of the memory object, so we don't do any here.
	 */
	vm_flags = calc_vm_prot_bits(prot) | calc_vm_flag_bits(flags) |
			mm->def_flags | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC;

H
Huang Shijie 已提交
1025
	if (flags & MAP_LOCKED)
L
Linus Torvalds 已提交
1026 1027
		if (!can_do_mlock())
			return -EPERM;
1028

L
Linus Torvalds 已提交
1029 1030 1031
	/* mlock MCL_FUTURE? */
	if (vm_flags & VM_LOCKED) {
		unsigned long locked, lock_limit;
1032 1033
		locked = len >> PAGE_SHIFT;
		locked += mm->locked_vm;
J
Jiri Slaby 已提交
1034
		lock_limit = rlimit(RLIMIT_MEMLOCK);
1035
		lock_limit >>= PAGE_SHIFT;
L
Linus Torvalds 已提交
1036 1037 1038 1039
		if (locked > lock_limit && !capable(CAP_IPC_LOCK))
			return -EAGAIN;
	}

1040
	inode = file ? file->f_path.dentry->d_inode : NULL;
L
Linus Torvalds 已提交
1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068

	if (file) {
		switch (flags & MAP_TYPE) {
		case MAP_SHARED:
			if ((prot&PROT_WRITE) && !(file->f_mode&FMODE_WRITE))
				return -EACCES;

			/*
			 * Make sure we don't allow writing to an append-only
			 * file..
			 */
			if (IS_APPEND(inode) && (file->f_mode & FMODE_WRITE))
				return -EACCES;

			/*
			 * Make sure there are no mandatory locks on the file.
			 */
			if (locks_verify_locked(inode))
				return -EAGAIN;

			vm_flags |= VM_SHARED | VM_MAYSHARE;
			if (!(file->f_mode & FMODE_WRITE))
				vm_flags &= ~(VM_MAYWRITE | VM_SHARED);

			/* fall through */
		case MAP_PRIVATE:
			if (!(file->f_mode & FMODE_READ))
				return -EACCES;
1069
			if (file->f_path.mnt->mnt_flags & MNT_NOEXEC) {
L
Linus Torvalds 已提交
1070 1071 1072 1073 1074 1075 1076
				if (vm_flags & VM_EXEC)
					return -EPERM;
				vm_flags &= ~VM_MAYEXEC;
			}

			if (!file->f_op || !file->f_op->mmap)
				return -ENODEV;
L
Linus Torvalds 已提交
1077 1078 1079 1080 1081 1082 1083 1084
			break;

		default:
			return -EINVAL;
		}
	} else {
		switch (flags & MAP_TYPE) {
		case MAP_SHARED:
1085 1086 1087 1088
			/*
			 * Ignore pgoff.
			 */
			pgoff = 0;
L
Linus Torvalds 已提交
1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101
			vm_flags |= VM_SHARED | VM_MAYSHARE;
			break;
		case MAP_PRIVATE:
			/*
			 * Set pgoff according to addr for anon_vma.
			 */
			pgoff = addr >> PAGE_SHIFT;
			break;
		default:
			return -EINVAL;
		}
	}

1102
	return mmap_region(file, addr, len, flags, vm_flags, pgoff);
M
Miklos Szeredi 已提交
1103
}
1104

H
Hugh Dickins 已提交
1105 1106 1107 1108 1109 1110 1111 1112
SYSCALL_DEFINE6(mmap_pgoff, unsigned long, addr, unsigned long, len,
		unsigned long, prot, unsigned long, flags,
		unsigned long, fd, unsigned long, pgoff)
{
	struct file *file = NULL;
	unsigned long retval = -EBADF;

	if (!(flags & MAP_ANONYMOUS)) {
A
Al Viro 已提交
1113
		audit_mmap_fd(fd, flags);
H
Hugh Dickins 已提交
1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126
		if (unlikely(flags & MAP_HUGETLB))
			return -EINVAL;
		file = fget(fd);
		if (!file)
			goto out;
	} else if (flags & MAP_HUGETLB) {
		struct user_struct *user = NULL;
		/*
		 * VM_NORESERVE is used because the reservations will be
		 * taken when vm_ops->mmap() is called
		 * A dummy user value is used because we are not locking
		 * memory so no accounting is necessary
		 */
1127 1128 1129
		file = hugetlb_file_setup(HUGETLB_ANON_FILE, addr, len,
						VM_NORESERVE, &user,
						HUGETLB_ANONHUGE_INODE);
H
Hugh Dickins 已提交
1130 1131 1132 1133 1134 1135
		if (IS_ERR(file))
			return PTR_ERR(file);
	}

	flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE);

A
Al Viro 已提交
1136
	retval = vm_mmap_pgoff(file, addr, len, prot, flags, pgoff);
H
Hugh Dickins 已提交
1137 1138 1139 1140 1141 1142
	if (file)
		fput(file);
out:
	return retval;
}

C
Christoph Hellwig 已提交
1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166
#ifdef __ARCH_WANT_SYS_OLD_MMAP
struct mmap_arg_struct {
	unsigned long addr;
	unsigned long len;
	unsigned long prot;
	unsigned long flags;
	unsigned long fd;
	unsigned long offset;
};

SYSCALL_DEFINE1(old_mmap, struct mmap_arg_struct __user *, arg)
{
	struct mmap_arg_struct a;

	if (copy_from_user(&a, arg, sizeof(a)))
		return -EFAULT;
	if (a.offset & ~PAGE_MASK)
		return -EINVAL;

	return sys_mmap_pgoff(a.addr, a.len, a.prot, a.flags, a.fd,
			      a.offset >> PAGE_SHIFT);
}
#endif /* __ARCH_WANT_SYS_OLD_MMAP */

A
Alexey Dobriyan 已提交
1167 1168 1169 1170 1171 1172 1173 1174
/*
 * Some shared mappigns will want the pages marked read-only
 * to track write events. If so, we'll downgrade vm_page_prot
 * to the private version (using protection_map[] without the
 * VM_SHARED bit).
 */
int vma_wants_writenotify(struct vm_area_struct *vma)
{
1175
	vm_flags_t vm_flags = vma->vm_flags;
A
Alexey Dobriyan 已提交
1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186

	/* If it was private or non-writable, the write bit is already clear */
	if ((vm_flags & (VM_WRITE|VM_SHARED)) != ((VM_WRITE|VM_SHARED)))
		return 0;

	/* The backer wishes to know when pages are first written to? */
	if (vma->vm_ops && vma->vm_ops->page_mkwrite)
		return 1;

	/* The open routine did something to the protections already? */
	if (pgprot_val(vma->vm_page_prot) !=
1187
	    pgprot_val(vm_get_page_prot(vm_flags)))
A
Alexey Dobriyan 已提交
1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198
		return 0;

	/* Specialty mapping? */
	if (vm_flags & (VM_PFNMAP|VM_INSERTPAGE))
		return 0;

	/* Can the mapping track the dirty pages? */
	return vma->vm_file && vma->vm_file->f_mapping &&
		mapping_cap_account_dirty(vma->vm_file->f_mapping);
}

1199 1200
/*
 * We account for memory if it's a private writeable mapping,
1201
 * not hugepages and VM_NORESERVE wasn't set.
1202
 */
1203
static inline int accountable_mapping(struct file *file, vm_flags_t vm_flags)
1204
{
1205 1206 1207 1208 1209 1210 1211
	/*
	 * hugetlb has its own accounting separate from the core VM
	 * VM_HUGETLB may not be set yet so we cannot check for that flag.
	 */
	if (file && is_file_hugepages(file))
		return 0;

1212 1213 1214
	return (vm_flags & (VM_NORESERVE | VM_SHARED | VM_WRITE)) == VM_WRITE;
}

M
Miklos Szeredi 已提交
1215 1216
unsigned long mmap_region(struct file *file, unsigned long addr,
			  unsigned long len, unsigned long flags,
1217
			  vm_flags_t vm_flags, unsigned long pgoff)
M
Miklos Szeredi 已提交
1218 1219 1220 1221 1222 1223 1224 1225 1226
{
	struct mm_struct *mm = current->mm;
	struct vm_area_struct *vma, *prev;
	int correct_wcount = 0;
	int error;
	struct rb_node **rb_link, *rb_parent;
	unsigned long charged = 0;
	struct inode *inode =  file ? file->f_path.dentry->d_inode : NULL;

L
Linus Torvalds 已提交
1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237
	/* Clear old maps */
	error = -ENOMEM;
munmap_back:
	vma = find_vma_prepare(mm, addr, &prev, &rb_link, &rb_parent);
	if (vma && vma->vm_start < addr + len) {
		if (do_munmap(mm, addr, len))
			return -ENOMEM;
		goto munmap_back;
	}

	/* Check against address space limit. */
A
akpm@osdl.org 已提交
1238
	if (!may_expand_vm(mm, len >> PAGE_SHIFT))
L
Linus Torvalds 已提交
1239 1240
		return -ENOMEM;

1241 1242
	/*
	 * Set 'VM_NORESERVE' if we should not account for the
1243
	 * memory use of this mapping.
1244
	 */
1245 1246 1247 1248 1249 1250 1251 1252 1253
	if ((flags & MAP_NORESERVE)) {
		/* We honor MAP_NORESERVE if allowed to overcommit */
		if (sysctl_overcommit_memory != OVERCOMMIT_NEVER)
			vm_flags |= VM_NORESERVE;

		/* hugetlb applies strict overcommit unless MAP_NORESERVE */
		if (file && is_file_hugepages(file))
			vm_flags |= VM_NORESERVE;
	}
1254

1255 1256 1257
	/*
	 * Private writable mapping: check memory availability
	 */
1258
	if (accountable_mapping(file, vm_flags)) {
1259
		charged = len >> PAGE_SHIFT;
1260
		if (security_vm_enough_memory_mm(mm, charged))
1261 1262
			return -ENOMEM;
		vm_flags |= VM_ACCOUNT;
L
Linus Torvalds 已提交
1263 1264 1265
	}

	/*
1266
	 * Can we just expand an old mapping?
L
Linus Torvalds 已提交
1267
	 */
1268 1269 1270
	vma = vma_merge(mm, prev, addr, addr + len, vm_flags, NULL, file, pgoff, NULL);
	if (vma)
		goto out;
L
Linus Torvalds 已提交
1271 1272 1273 1274 1275 1276

	/*
	 * Determine the object being mapped and call the appropriate
	 * specific mapper. the address has already been validated, but
	 * not unmapped, but the maps are removed from the list.
	 */
P
Pekka Enberg 已提交
1277
	vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL);
L
Linus Torvalds 已提交
1278 1279 1280 1281 1282 1283 1284 1285 1286
	if (!vma) {
		error = -ENOMEM;
		goto unacct_error;
	}

	vma->vm_mm = mm;
	vma->vm_start = addr;
	vma->vm_end = addr + len;
	vma->vm_flags = vm_flags;
1287
	vma->vm_page_prot = vm_get_page_prot(vm_flags);
L
Linus Torvalds 已提交
1288
	vma->vm_pgoff = pgoff;
1289
	INIT_LIST_HEAD(&vma->anon_vma_chain);
L
Linus Torvalds 已提交
1290

1291 1292
	error = -EINVAL;	/* when rejecting VM_GROWSDOWN|VM_GROWSUP */

L
Linus Torvalds 已提交
1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306
	if (file) {
		if (vm_flags & (VM_GROWSDOWN|VM_GROWSUP))
			goto free_vma;
		if (vm_flags & VM_DENYWRITE) {
			error = deny_write_access(file);
			if (error)
				goto free_vma;
			correct_wcount = 1;
		}
		vma->vm_file = file;
		get_file(file);
		error = file->f_op->mmap(file, vma);
		if (error)
			goto unmap_and_free_vma;
M
Matt Helsley 已提交
1307 1308
		if (vm_flags & VM_EXECUTABLE)
			added_exe_file_vma(mm);
1309 1310 1311 1312 1313 1314 1315 1316 1317

		/* Can addr have changed??
		 *
		 * Answer: Yes, several device drivers can do it in their
		 *         f_op->mmap method. -DaveM
		 */
		addr = vma->vm_start;
		pgoff = vma->vm_pgoff;
		vm_flags = vma->vm_flags;
L
Linus Torvalds 已提交
1318
	} else if (vm_flags & VM_SHARED) {
1319 1320
		if (unlikely(vm_flags & (VM_GROWSDOWN|VM_GROWSUP)))
			goto free_vma;
L
Linus Torvalds 已提交
1321 1322 1323 1324 1325
		error = shmem_zero_setup(vma);
		if (error)
			goto free_vma;
	}

1326 1327 1328 1329 1330 1331 1332 1333 1334 1335
	if (vma_wants_writenotify(vma)) {
		pgprot_t pprot = vma->vm_page_prot;

		/* Can vma->vm_page_prot have changed??
		 *
		 * Answer: Yes, drivers may have changed it in their
		 *         f_op->mmap method.
		 *
		 * Ensures that vmas marked as uncached stay that way.
		 */
1336
		vma->vm_page_prot = vm_get_page_prot(vm_flags & ~VM_SHARED);
1337 1338 1339
		if (pgprot_val(pprot) == pgprot_val(pgprot_noncached(pprot)))
			vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
	}
1340

1341 1342
	vma_link(mm, vma, prev, rb_link, rb_parent);
	file = vma->vm_file;
1343 1344 1345 1346 1347

	/* Once vma denies write, undo our temporary denial count */
	if (correct_wcount)
		atomic_inc(&inode->i_writecount);
out:
1348
	perf_event_mmap(vma);
1349

L
Linus Torvalds 已提交
1350
	mm->total_vm += len >> PAGE_SHIFT;
1351
	vm_stat_account(mm, vm_flags, file, len >> PAGE_SHIFT);
L
Linus Torvalds 已提交
1352
	if (vm_flags & VM_LOCKED) {
1353 1354
		if (!mlock_vma_pages_range(vma, addr, addr + len))
			mm->locked_vm += (len >> PAGE_SHIFT);
1355
	} else if ((flags & MAP_POPULATE) && !(flags & MAP_NONBLOCK))
1356
		make_pages_present(addr, addr + len);
1357

1358
	if (file && uprobe_mmap(vma))
1359 1360 1361
		/* matching probes but cannot insert */
		goto unmap_and_free_vma;

L
Linus Torvalds 已提交
1362 1363 1364 1365 1366 1367 1368 1369 1370
	return addr;

unmap_and_free_vma:
	if (correct_wcount)
		atomic_inc(&inode->i_writecount);
	vma->vm_file = NULL;
	fput(file);

	/* Undo any partial mapping done by a device driver. */
1371 1372
	unmap_region(mm, vma, prev, vma->vm_start, vma->vm_end);
	charged = 0;
L
Linus Torvalds 已提交
1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403
free_vma:
	kmem_cache_free(vm_area_cachep, vma);
unacct_error:
	if (charged)
		vm_unacct_memory(charged);
	return error;
}

/* Get an address range which is currently unmapped.
 * For shmat() with addr=0.
 *
 * Ugly calling convention alert:
 * Return value with the low bits set means error value,
 * ie
 *	if (ret & ~PAGE_MASK)
 *		error = ret;
 *
 * This function "knows" that -ENOMEM has the bits set.
 */
#ifndef HAVE_ARCH_UNMAPPED_AREA
unsigned long
arch_get_unmapped_area(struct file *filp, unsigned long addr,
		unsigned long len, unsigned long pgoff, unsigned long flags)
{
	struct mm_struct *mm = current->mm;
	struct vm_area_struct *vma;
	unsigned long start_addr;

	if (len > TASK_SIZE)
		return -ENOMEM;

1404 1405 1406
	if (flags & MAP_FIXED)
		return addr;

L
Linus Torvalds 已提交
1407 1408 1409 1410 1411 1412 1413
	if (addr) {
		addr = PAGE_ALIGN(addr);
		vma = find_vma(mm, addr);
		if (TASK_SIZE - len >= addr &&
		    (!vma || addr + len <= vma->vm_start))
			return addr;
	}
1414 1415 1416 1417 1418 1419
	if (len > mm->cached_hole_size) {
	        start_addr = addr = mm->free_area_cache;
	} else {
	        start_addr = addr = TASK_UNMAPPED_BASE;
	        mm->cached_hole_size = 0;
	}
L
Linus Torvalds 已提交
1420 1421 1422 1423 1424 1425 1426 1427 1428 1429

full_search:
	for (vma = find_vma(mm, addr); ; vma = vma->vm_next) {
		/* At this point:  (!vma || addr < vma->vm_end). */
		if (TASK_SIZE - len < addr) {
			/*
			 * Start a new search - just in case we missed
			 * some holes.
			 */
			if (start_addr != TASK_UNMAPPED_BASE) {
1430 1431 1432
				addr = TASK_UNMAPPED_BASE;
			        start_addr = addr;
				mm->cached_hole_size = 0;
L
Linus Torvalds 已提交
1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443
				goto full_search;
			}
			return -ENOMEM;
		}
		if (!vma || addr + len <= vma->vm_start) {
			/*
			 * Remember the place where we stopped the search:
			 */
			mm->free_area_cache = addr + len;
			return addr;
		}
1444 1445
		if (addr + mm->cached_hole_size < vma->vm_start)
		        mm->cached_hole_size = vma->vm_start - addr;
L
Linus Torvalds 已提交
1446 1447 1448 1449 1450
		addr = vma->vm_end;
	}
}
#endif	

1451
void arch_unmap_area(struct mm_struct *mm, unsigned long addr)
L
Linus Torvalds 已提交
1452 1453 1454 1455
{
	/*
	 * Is this a new hole at the lowest possible address?
	 */
1456
	if (addr >= TASK_UNMAPPED_BASE && addr < mm->free_area_cache)
1457
		mm->free_area_cache = addr;
L
Linus Torvalds 已提交
1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471
}

/*
 * This mmap-allocator allocates new areas top-down from below the
 * stack's low limit (the base):
 */
#ifndef HAVE_ARCH_UNMAPPED_AREA_TOPDOWN
unsigned long
arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
			  const unsigned long len, const unsigned long pgoff,
			  const unsigned long flags)
{
	struct vm_area_struct *vma;
	struct mm_struct *mm = current->mm;
1472
	unsigned long addr = addr0, start_addr;
L
Linus Torvalds 已提交
1473 1474 1475 1476 1477

	/* requested length too big for entire address space */
	if (len > TASK_SIZE)
		return -ENOMEM;

1478 1479 1480
	if (flags & MAP_FIXED)
		return addr;

L
Linus Torvalds 已提交
1481 1482 1483 1484 1485 1486 1487 1488 1489
	/* requesting a specific address */
	if (addr) {
		addr = PAGE_ALIGN(addr);
		vma = find_vma(mm, addr);
		if (TASK_SIZE - len >= addr &&
				(!vma || addr + len <= vma->vm_start))
			return addr;
	}

1490 1491 1492 1493 1494 1495
	/* check if free_area_cache is useful for us */
	if (len <= mm->cached_hole_size) {
 	        mm->cached_hole_size = 0;
 		mm->free_area_cache = mm->mmap_base;
 	}

1496
try_again:
L
Linus Torvalds 已提交
1497
	/* either no address requested or can't fit in requested address hole */
1498
	start_addr = addr = mm->free_area_cache;
1499

1500 1501
	if (addr < len)
		goto fail;
L
Linus Torvalds 已提交
1502

1503
	addr -= len;
L
Linus Torvalds 已提交
1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514
	do {
		/*
		 * Lookup failure means no vma is above this address,
		 * else if new region fits below vma->vm_start,
		 * return with success:
		 */
		vma = find_vma(mm, addr);
		if (!vma || addr+len <= vma->vm_start)
			/* remember the address as a hint for next time */
			return (mm->free_area_cache = addr);

1515 1516 1517 1518
 		/* remember the largest hole we saw so far */
 		if (addr + mm->cached_hole_size < vma->vm_start)
 		        mm->cached_hole_size = vma->vm_start - addr;

L
Linus Torvalds 已提交
1519 1520
		/* try just below the current vma->vm_start */
		addr = vma->vm_start-len;
1521
	} while (len < vma->vm_start);
L
Linus Torvalds 已提交
1522

1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537
fail:
	/*
	 * if hint left us with no space for the requested
	 * mapping then try again:
	 *
	 * Note: this is different with the case of bottomup
	 * which does the fully line-search, but we use find_vma
	 * here that causes some holes skipped.
	 */
	if (start_addr != mm->mmap_base) {
		mm->free_area_cache = mm->mmap_base;
		mm->cached_hole_size = 0;
		goto try_again;
	}

L
Linus Torvalds 已提交
1538 1539 1540 1541 1542 1543
	/*
	 * A failed mmap() very likely causes application failure,
	 * so fall back to the bottom-up function here. This scenario
	 * can happen with large stack limits and large mmap()
	 * allocations.
	 */
1544 1545
	mm->cached_hole_size = ~0UL;
  	mm->free_area_cache = TASK_UNMAPPED_BASE;
L
Linus Torvalds 已提交
1546 1547 1548 1549 1550
	addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags);
	/*
	 * Restore the topdown base:
	 */
	mm->free_area_cache = mm->mmap_base;
1551
	mm->cached_hole_size = ~0UL;
L
Linus Torvalds 已提交
1552 1553 1554 1555 1556

	return addr;
}
#endif

1557
void arch_unmap_area_topdown(struct mm_struct *mm, unsigned long addr)
L
Linus Torvalds 已提交
1558 1559 1560 1561
{
	/*
	 * Is this a new hole at the highest possible address?
	 */
1562 1563
	if (addr > mm->free_area_cache)
		mm->free_area_cache = addr;
L
Linus Torvalds 已提交
1564 1565

	/* dont allow allocations above current base */
1566 1567
	if (mm->free_area_cache > mm->mmap_base)
		mm->free_area_cache = mm->mmap_base;
L
Linus Torvalds 已提交
1568 1569 1570 1571 1572 1573
}

unsigned long
get_unmapped_area(struct file *file, unsigned long addr, unsigned long len,
		unsigned long pgoff, unsigned long flags)
{
1574 1575 1576
	unsigned long (*get_area)(struct file *, unsigned long,
				  unsigned long, unsigned long, unsigned long);

1577 1578 1579 1580 1581 1582 1583 1584
	unsigned long error = arch_mmap_check(addr, len, flags);
	if (error)
		return error;

	/* Careful about overflows.. */
	if (len > TASK_SIZE)
		return -ENOMEM;

1585 1586 1587 1588 1589 1590
	get_area = current->mm->get_unmapped_area;
	if (file && file->f_op && file->f_op->get_unmapped_area)
		get_area = file->f_op->get_unmapped_area;
	addr = get_area(file, addr, len, pgoff, flags);
	if (IS_ERR_VALUE(addr))
		return addr;
L
Linus Torvalds 已提交
1591

1592 1593 1594 1595
	if (addr > TASK_SIZE - len)
		return -ENOMEM;
	if (addr & ~PAGE_MASK)
		return -EINVAL;
1596

1597 1598 1599
	addr = arch_rebalance_pgtables(addr, len);
	error = security_mmap_addr(addr);
	return error ? error : addr;
L
Linus Torvalds 已提交
1600 1601 1602 1603 1604
}

EXPORT_SYMBOL(get_unmapped_area);

/* Look up the first VMA which satisfies  addr < vm_end,  NULL if none. */
Z
ZhenwenXu 已提交
1605
struct vm_area_struct *find_vma(struct mm_struct *mm, unsigned long addr)
L
Linus Torvalds 已提交
1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641
{
	struct vm_area_struct *vma = NULL;

	if (mm) {
		/* Check the cache first. */
		/* (Cache hit rate is typically around 35%.) */
		vma = mm->mmap_cache;
		if (!(vma && vma->vm_end > addr && vma->vm_start <= addr)) {
			struct rb_node * rb_node;

			rb_node = mm->mm_rb.rb_node;
			vma = NULL;

			while (rb_node) {
				struct vm_area_struct * vma_tmp;

				vma_tmp = rb_entry(rb_node,
						struct vm_area_struct, vm_rb);

				if (vma_tmp->vm_end > addr) {
					vma = vma_tmp;
					if (vma_tmp->vm_start <= addr)
						break;
					rb_node = rb_node->rb_left;
				} else
					rb_node = rb_node->rb_right;
			}
			if (vma)
				mm->mmap_cache = vma;
		}
	}
	return vma;
}

EXPORT_SYMBOL(find_vma);

K
KOSAKI Motohiro 已提交
1642 1643 1644
/*
 * Same as find_vma, but also return a pointer to the previous VMA in *pprev.
 */
L
Linus Torvalds 已提交
1645 1646 1647 1648
struct vm_area_struct *
find_vma_prev(struct mm_struct *mm, unsigned long addr,
			struct vm_area_struct **pprev)
{
K
KOSAKI Motohiro 已提交
1649
	struct vm_area_struct *vma;
L
Linus Torvalds 已提交
1650

K
KOSAKI Motohiro 已提交
1651
	vma = find_vma(mm, addr);
M
Mikulas Patocka 已提交
1652 1653 1654 1655 1656 1657 1658 1659 1660 1661
	if (vma) {
		*pprev = vma->vm_prev;
	} else {
		struct rb_node *rb_node = mm->mm_rb.rb_node;
		*pprev = NULL;
		while (rb_node) {
			*pprev = rb_entry(rb_node, struct vm_area_struct, vm_rb);
			rb_node = rb_node->rb_right;
		}
	}
K
KOSAKI Motohiro 已提交
1662
	return vma;
L
Linus Torvalds 已提交
1663 1664 1665 1666 1667 1668 1669
}

/*
 * Verify that the stack growth is acceptable and
 * update accounting. This is shared with both the
 * grow-up and grow-down cases.
 */
Z
ZhenwenXu 已提交
1670
static int acct_stack_growth(struct vm_area_struct *vma, unsigned long size, unsigned long grow)
L
Linus Torvalds 已提交
1671 1672 1673
{
	struct mm_struct *mm = vma->vm_mm;
	struct rlimit *rlim = current->signal->rlim;
1674
	unsigned long new_start;
L
Linus Torvalds 已提交
1675 1676

	/* address space limit tests */
A
akpm@osdl.org 已提交
1677
	if (!may_expand_vm(mm, grow))
L
Linus Torvalds 已提交
1678 1679 1680
		return -ENOMEM;

	/* Stack limit test */
J
Jiri Slaby 已提交
1681
	if (size > ACCESS_ONCE(rlim[RLIMIT_STACK].rlim_cur))
L
Linus Torvalds 已提交
1682 1683 1684 1685 1686 1687 1688
		return -ENOMEM;

	/* mlock limit tests */
	if (vma->vm_flags & VM_LOCKED) {
		unsigned long locked;
		unsigned long limit;
		locked = mm->locked_vm + grow;
J
Jiri Slaby 已提交
1689 1690
		limit = ACCESS_ONCE(rlim[RLIMIT_MEMLOCK].rlim_cur);
		limit >>= PAGE_SHIFT;
L
Linus Torvalds 已提交
1691 1692 1693 1694
		if (locked > limit && !capable(CAP_IPC_LOCK))
			return -ENOMEM;
	}

1695 1696 1697 1698 1699 1700
	/* Check to ensure the stack will not grow into a hugetlb-only region */
	new_start = (vma->vm_flags & VM_GROWSUP) ? vma->vm_start :
			vma->vm_end - size;
	if (is_hugepage_only_range(vma->vm_mm, new_start, size))
		return -EFAULT;

L
Linus Torvalds 已提交
1701 1702 1703 1704
	/*
	 * Overcommit..  This must be the final test, as it will
	 * update security statistics.
	 */
1705
	if (security_vm_enough_memory_mm(mm, grow))
L
Linus Torvalds 已提交
1706 1707 1708 1709 1710 1711
		return -ENOMEM;

	/* Ok, everything looks good - let it rip */
	mm->total_vm += grow;
	if (vma->vm_flags & VM_LOCKED)
		mm->locked_vm += grow;
1712
	vm_stat_account(mm, vma->vm_flags, vma->vm_file, grow);
L
Linus Torvalds 已提交
1713 1714 1715
	return 0;
}

H
Hugh Dickins 已提交
1716
#if defined(CONFIG_STACK_GROWSUP) || defined(CONFIG_IA64)
L
Linus Torvalds 已提交
1717
/*
H
Hugh Dickins 已提交
1718 1719
 * PA-RISC uses this for its stack; IA64 for its Register Backing Store.
 * vma is the last one with address > vma->vm_end.  Have to extend vma.
L
Linus Torvalds 已提交
1720
 */
H
Hugh Dickins 已提交
1721
int expand_upwards(struct vm_area_struct *vma, unsigned long address)
L
Linus Torvalds 已提交
1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733
{
	int error;

	if (!(vma->vm_flags & VM_GROWSUP))
		return -EFAULT;

	/*
	 * We must make sure the anon_vma is allocated
	 * so that the anon_vma locking is not a noop.
	 */
	if (unlikely(anon_vma_prepare(vma)))
		return -ENOMEM;
1734
	vma_lock_anon_vma(vma);
L
Linus Torvalds 已提交
1735 1736 1737 1738 1739

	/*
	 * vma->vm_start/vm_end cannot change under us because the caller
	 * is required to hold the mmap_sem in read mode.  We need the
	 * anon_vma lock to serialize against concurrent expand_stacks.
1740
	 * Also guard against wrapping around to address 0.
L
Linus Torvalds 已提交
1741
	 */
1742 1743 1744
	if (address < PAGE_ALIGN(address+4))
		address = PAGE_ALIGN(address+4);
	else {
1745
		vma_unlock_anon_vma(vma);
1746 1747
		return -ENOMEM;
	}
L
Linus Torvalds 已提交
1748 1749 1750 1751 1752 1753 1754 1755 1756
	error = 0;

	/* Somebody else might have raced and expanded it already */
	if (address > vma->vm_end) {
		unsigned long size, grow;

		size = address - vma->vm_start;
		grow = (address - vma->vm_end) >> PAGE_SHIFT;

1757 1758 1759 1760 1761 1762 1763
		error = -ENOMEM;
		if (vma->vm_pgoff + (size >> PAGE_SHIFT) >= vma->vm_pgoff) {
			error = acct_stack_growth(vma, size, grow);
			if (!error) {
				vma->vm_end = address;
				perf_event_mmap(vma);
			}
1764
		}
L
Linus Torvalds 已提交
1765
	}
1766
	vma_unlock_anon_vma(vma);
A
Andrea Arcangeli 已提交
1767
	khugepaged_enter_vma_merge(vma);
L
Linus Torvalds 已提交
1768 1769
	return error;
}
H
Hugh Dickins 已提交
1770 1771
#endif /* CONFIG_STACK_GROWSUP || CONFIG_IA64 */

L
Linus Torvalds 已提交
1772 1773 1774
/*
 * vma is the first one with address < vma->vm_start.  Have to extend vma.
 */
1775
int expand_downwards(struct vm_area_struct *vma,
1776
				   unsigned long address)
L
Linus Torvalds 已提交
1777 1778 1779 1780 1781 1782 1783 1784 1785
{
	int error;

	/*
	 * We must make sure the anon_vma is allocated
	 * so that the anon_vma locking is not a noop.
	 */
	if (unlikely(anon_vma_prepare(vma)))
		return -ENOMEM;
1786 1787

	address &= PAGE_MASK;
1788
	error = security_mmap_addr(address);
1789 1790 1791
	if (error)
		return error;

1792
	vma_lock_anon_vma(vma);
L
Linus Torvalds 已提交
1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806

	/*
	 * vma->vm_start/vm_end cannot change under us because the caller
	 * is required to hold the mmap_sem in read mode.  We need the
	 * anon_vma lock to serialize against concurrent expand_stacks.
	 */

	/* Somebody else might have raced and expanded it already */
	if (address < vma->vm_start) {
		unsigned long size, grow;

		size = vma->vm_end - address;
		grow = (vma->vm_start - address) >> PAGE_SHIFT;

1807 1808 1809 1810 1811 1812 1813 1814
		error = -ENOMEM;
		if (grow <= vma->vm_pgoff) {
			error = acct_stack_growth(vma, size, grow);
			if (!error) {
				vma->vm_start = address;
				vma->vm_pgoff -= grow;
				perf_event_mmap(vma);
			}
L
Linus Torvalds 已提交
1815 1816
		}
	}
1817
	vma_unlock_anon_vma(vma);
A
Andrea Arcangeli 已提交
1818
	khugepaged_enter_vma_merge(vma);
L
Linus Torvalds 已提交
1819 1820 1821
	return error;
}

1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836
#ifdef CONFIG_STACK_GROWSUP
int expand_stack(struct vm_area_struct *vma, unsigned long address)
{
	return expand_upwards(vma, address);
}

struct vm_area_struct *
find_extend_vma(struct mm_struct *mm, unsigned long addr)
{
	struct vm_area_struct *vma, *prev;

	addr &= PAGE_MASK;
	vma = find_vma_prev(mm, addr, &prev);
	if (vma && (vma->vm_start <= addr))
		return vma;
1837
	if (!prev || expand_stack(prev, addr))
1838
		return NULL;
1839
	if (prev->vm_flags & VM_LOCKED) {
1840
		mlock_vma_pages_range(prev, addr, prev->vm_end);
1841
	}
1842 1843 1844 1845 1846 1847 1848 1849
	return prev;
}
#else
int expand_stack(struct vm_area_struct *vma, unsigned long address)
{
	return expand_downwards(vma, address);
}

L
Linus Torvalds 已提交
1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866
struct vm_area_struct *
find_extend_vma(struct mm_struct * mm, unsigned long addr)
{
	struct vm_area_struct * vma;
	unsigned long start;

	addr &= PAGE_MASK;
	vma = find_vma(mm,addr);
	if (!vma)
		return NULL;
	if (vma->vm_start <= addr)
		return vma;
	if (!(vma->vm_flags & VM_GROWSDOWN))
		return NULL;
	start = vma->vm_start;
	if (expand_stack(vma, addr))
		return NULL;
1867
	if (vma->vm_flags & VM_LOCKED) {
1868
		mlock_vma_pages_range(vma, addr, start);
1869
	}
L
Linus Torvalds 已提交
1870 1871 1872 1873 1874
	return vma;
}
#endif

/*
1875
 * Ok - we have the memory areas we should free on the vma list,
L
Linus Torvalds 已提交
1876
 * so release them, and do the vma updates.
1877 1878
 *
 * Called with the mm semaphore held.
L
Linus Torvalds 已提交
1879
 */
1880
static void remove_vma_list(struct mm_struct *mm, struct vm_area_struct *vma)
L
Linus Torvalds 已提交
1881
{
1882 1883
	unsigned long nr_accounted = 0;

1884 1885
	/* Update high watermark before we lower total_vm */
	update_hiwater_vm(mm);
L
Linus Torvalds 已提交
1886
	do {
1887 1888
		long nrpages = vma_pages(vma);

1889 1890
		if (vma->vm_flags & VM_ACCOUNT)
			nr_accounted += nrpages;
1891 1892
		mm->total_vm -= nrpages;
		vm_stat_account(mm, vma->vm_flags, vma->vm_file, -nrpages);
1893
		vma = remove_vma(vma);
1894
	} while (vma);
1895
	vm_unacct_memory(nr_accounted);
L
Linus Torvalds 已提交
1896 1897 1898 1899 1900 1901
	validate_mm(mm);
}

/*
 * Get rid of page table information in the indicated region.
 *
1902
 * Called with the mm semaphore held.
L
Linus Torvalds 已提交
1903 1904
 */
static void unmap_region(struct mm_struct *mm,
1905 1906
		struct vm_area_struct *vma, struct vm_area_struct *prev,
		unsigned long start, unsigned long end)
L
Linus Torvalds 已提交
1907
{
1908
	struct vm_area_struct *next = prev? prev->vm_next: mm->mmap;
P
Peter Zijlstra 已提交
1909
	struct mmu_gather tlb;
L
Linus Torvalds 已提交
1910 1911

	lru_add_drain();
P
Peter Zijlstra 已提交
1912
	tlb_gather_mmu(&tlb, mm, 0);
1913
	update_hiwater_rss(mm);
1914
	unmap_vmas(&tlb, vma, start, end);
P
Peter Zijlstra 已提交
1915 1916 1917
	free_pgtables(&tlb, vma, prev ? prev->vm_end : FIRST_USER_ADDRESS,
				 next ? next->vm_start : 0);
	tlb_finish_mmu(&tlb, start, end);
L
Linus Torvalds 已提交
1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929
}

/*
 * Create a list of vma's touched by the unmap, removing them from the mm's
 * vma list as we go..
 */
static void
detach_vmas_to_be_unmapped(struct mm_struct *mm, struct vm_area_struct *vma,
	struct vm_area_struct *prev, unsigned long end)
{
	struct vm_area_struct **insertion_point;
	struct vm_area_struct *tail_vma = NULL;
1930
	unsigned long addr;
L
Linus Torvalds 已提交
1931 1932

	insertion_point = (prev ? &prev->vm_next : &mm->mmap);
1933
	vma->vm_prev = NULL;
L
Linus Torvalds 已提交
1934 1935 1936 1937 1938 1939 1940
	do {
		rb_erase(&vma->vm_rb, &mm->mm_rb);
		mm->map_count--;
		tail_vma = vma;
		vma = vma->vm_next;
	} while (vma && vma->vm_start < end);
	*insertion_point = vma;
1941 1942
	if (vma)
		vma->vm_prev = prev;
L
Linus Torvalds 已提交
1943
	tail_vma->vm_next = NULL;
1944 1945 1946 1947 1948
	if (mm->unmap_area == arch_unmap_area)
		addr = prev ? prev->vm_end : mm->mmap_base;
	else
		addr = vma ?  vma->vm_start : mm->mmap_base;
	mm->unmap_area(mm, addr);
L
Linus Torvalds 已提交
1949 1950 1951 1952
	mm->mmap_cache = NULL;		/* Kill the cache. */
}

/*
1953 1954
 * __split_vma() bypasses sysctl_max_map_count checking.  We use this on the
 * munmap path where it doesn't make sense to fail.
L
Linus Torvalds 已提交
1955
 */
1956
static int __split_vma(struct mm_struct * mm, struct vm_area_struct * vma,
L
Linus Torvalds 已提交
1957 1958 1959 1960
	      unsigned long addr, int new_below)
{
	struct mempolicy *pol;
	struct vm_area_struct *new;
1961
	int err = -ENOMEM;
L
Linus Torvalds 已提交
1962

1963 1964
	if (is_vm_hugetlb_page(vma) && (addr &
					~(huge_page_mask(hstate_vma(vma)))))
L
Linus Torvalds 已提交
1965 1966
		return -EINVAL;

1967
	new = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL);
L
Linus Torvalds 已提交
1968
	if (!new)
1969
		goto out_err;
L
Linus Torvalds 已提交
1970 1971 1972 1973

	/* most fields are the same, copy all, and then fixup */
	*new = *vma;

1974 1975
	INIT_LIST_HEAD(&new->anon_vma_chain);

L
Linus Torvalds 已提交
1976 1977 1978 1979 1980 1981 1982
	if (new_below)
		new->vm_end = addr;
	else {
		new->vm_start = addr;
		new->vm_pgoff += ((addr - vma->vm_start) >> PAGE_SHIFT);
	}

1983
	pol = mpol_dup(vma_policy(vma));
L
Linus Torvalds 已提交
1984
	if (IS_ERR(pol)) {
1985 1986
		err = PTR_ERR(pol);
		goto out_free_vma;
L
Linus Torvalds 已提交
1987 1988 1989
	}
	vma_set_policy(new, pol);

1990 1991 1992
	if (anon_vma_clone(new, vma))
		goto out_free_mpol;

M
Matt Helsley 已提交
1993
	if (new->vm_file) {
L
Linus Torvalds 已提交
1994
		get_file(new->vm_file);
M
Matt Helsley 已提交
1995 1996 1997
		if (vma->vm_flags & VM_EXECUTABLE)
			added_exe_file_vma(mm);
	}
L
Linus Torvalds 已提交
1998 1999 2000 2001 2002

	if (new->vm_ops && new->vm_ops->open)
		new->vm_ops->open(new);

	if (new_below)
2003
		err = vma_adjust(vma, addr, vma->vm_end, vma->vm_pgoff +
L
Linus Torvalds 已提交
2004 2005
			((addr - new->vm_start) >> PAGE_SHIFT), new);
	else
2006
		err = vma_adjust(vma, vma->vm_start, addr, vma->vm_pgoff, new);
L
Linus Torvalds 已提交
2007

2008 2009 2010 2011 2012
	/* Success. */
	if (!err)
		return 0;

	/* Clean everything up if vma_adjust failed. */
2013 2014
	if (new->vm_ops && new->vm_ops->close)
		new->vm_ops->close(new);
2015 2016 2017 2018 2019
	if (new->vm_file) {
		if (vma->vm_flags & VM_EXECUTABLE)
			removed_exe_file_vma(mm);
		fput(new->vm_file);
	}
2020
	unlink_anon_vmas(new);
2021 2022 2023 2024 2025 2026
 out_free_mpol:
	mpol_put(pol);
 out_free_vma:
	kmem_cache_free(vm_area_cachep, new);
 out_err:
	return err;
L
Linus Torvalds 已提交
2027 2028
}

2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041
/*
 * Split a vma into two pieces at address 'addr', a new vma is allocated
 * either for the first part or the tail.
 */
int split_vma(struct mm_struct *mm, struct vm_area_struct *vma,
	      unsigned long addr, int new_below)
{
	if (mm->map_count >= sysctl_max_map_count)
		return -ENOMEM;

	return __split_vma(mm, vma, addr, new_below);
}

L
Linus Torvalds 已提交
2042 2043 2044 2045 2046 2047 2048 2049
/* Munmap is split into 2 main parts -- this part which finds
 * what needs doing, and the areas themselves, which do the
 * work.  This now handles partial unmappings.
 * Jeremy Fitzhardinge <jeremy@goop.org>
 */
int do_munmap(struct mm_struct *mm, unsigned long start, size_t len)
{
	unsigned long end;
2050
	struct vm_area_struct *vma, *prev, *last;
L
Linus Torvalds 已提交
2051 2052 2053 2054 2055 2056 2057 2058

	if ((start & ~PAGE_MASK) || start > TASK_SIZE || len > TASK_SIZE-start)
		return -EINVAL;

	if ((len = PAGE_ALIGN(len)) == 0)
		return -EINVAL;

	/* Find the first overlapping VMA */
2059
	vma = find_vma(mm, start);
2060
	if (!vma)
L
Linus Torvalds 已提交
2061
		return 0;
2062
	prev = vma->vm_prev;
2063
	/* we have  start < vma->vm_end  */
L
Linus Torvalds 已提交
2064 2065 2066

	/* if it doesn't overlap, we have nothing.. */
	end = start + len;
2067
	if (vma->vm_start >= end)
L
Linus Torvalds 已提交
2068 2069 2070 2071 2072 2073 2074 2075 2076
		return 0;

	/*
	 * If we need to split any vma, do it now to save pain later.
	 *
	 * Note: mremap's move_vma VM_ACCOUNT handling assumes a partially
	 * unmapped vm_area_struct will remain in use: so lower split_vma
	 * places tmp vma above, and higher split_vma places tmp vma below.
	 */
2077
	if (start > vma->vm_start) {
2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088
		int error;

		/*
		 * Make sure that map_count on return from munmap() will
		 * not exceed its limit; but let map_count go just above
		 * its limit temporarily, to help free resources as expected.
		 */
		if (end < vma->vm_end && mm->map_count >= sysctl_max_map_count)
			return -ENOMEM;

		error = __split_vma(mm, vma, start, 0);
L
Linus Torvalds 已提交
2089 2090
		if (error)
			return error;
2091
		prev = vma;
L
Linus Torvalds 已提交
2092 2093 2094 2095 2096
	}

	/* Does it split the last one? */
	last = find_vma(mm, end);
	if (last && end > last->vm_start) {
2097
		int error = __split_vma(mm, last, end, 1);
L
Linus Torvalds 已提交
2098 2099 2100
		if (error)
			return error;
	}
2101
	vma = prev? prev->vm_next: mm->mmap;
L
Linus Torvalds 已提交
2102

2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116
	/*
	 * unlock any mlock()ed ranges before detaching vmas
	 */
	if (mm->locked_vm) {
		struct vm_area_struct *tmp = vma;
		while (tmp && tmp->vm_start < end) {
			if (tmp->vm_flags & VM_LOCKED) {
				mm->locked_vm -= vma_pages(tmp);
				munlock_vma_pages_all(tmp);
			}
			tmp = tmp->vm_next;
		}
	}

L
Linus Torvalds 已提交
2117 2118 2119
	/*
	 * Remove the vma's, and unmap the actual pages
	 */
2120 2121
	detach_vmas_to_be_unmapped(mm, vma, prev, end);
	unmap_region(mm, vma, prev, start, end);
L
Linus Torvalds 已提交
2122 2123

	/* Fix up all other VM information */
2124
	remove_vma_list(mm, vma);
L
Linus Torvalds 已提交
2125 2126 2127 2128

	return 0;
}

A
Al Viro 已提交
2129
int vm_munmap(unsigned long start, size_t len)
L
Linus Torvalds 已提交
2130 2131
{
	int ret;
A
Al Viro 已提交
2132
	struct mm_struct *mm = current->mm;
L
Linus Torvalds 已提交
2133 2134

	down_write(&mm->mmap_sem);
2135
	ret = do_munmap(mm, start, len);
L
Linus Torvalds 已提交
2136 2137 2138
	up_write(&mm->mmap_sem);
	return ret;
}
2139 2140 2141 2142 2143
EXPORT_SYMBOL(vm_munmap);

SYSCALL_DEFINE2(munmap, unsigned long, addr, size_t, len)
{
	profile_munmap(addr);
A
Al Viro 已提交
2144
	return vm_munmap(addr, len);
2145
}
L
Linus Torvalds 已提交
2146 2147 2148

static inline void verify_mm_writelocked(struct mm_struct *mm)
{
2149
#ifdef CONFIG_DEBUG_VM
L
Linus Torvalds 已提交
2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161
	if (unlikely(down_read_trylock(&mm->mmap_sem))) {
		WARN_ON(1);
		up_read(&mm->mmap_sem);
	}
#endif
}

/*
 *  this is really a simplified "do_mmap".  it only handles
 *  anonymous maps.  eventually we may be able to do some
 *  brk-specific accounting here.
 */
2162
static unsigned long do_brk(unsigned long addr, unsigned long len)
L
Linus Torvalds 已提交
2163 2164 2165 2166 2167 2168
{
	struct mm_struct * mm = current->mm;
	struct vm_area_struct * vma, * prev;
	unsigned long flags;
	struct rb_node ** rb_link, * rb_parent;
	pgoff_t pgoff = addr >> PAGE_SHIFT;
2169
	int error;
L
Linus Torvalds 已提交
2170 2171 2172 2173 2174

	len = PAGE_ALIGN(len);
	if (!len)
		return addr;

2175 2176
	flags = VM_DATA_DEFAULT_FLAGS | VM_ACCOUNT | mm->def_flags;

A
Al Viro 已提交
2177 2178
	error = get_unmapped_area(NULL, addr, len, 0, MAP_FIXED);
	if (error & ~PAGE_MASK)
2179 2180
		return error;

L
Linus Torvalds 已提交
2181 2182 2183 2184 2185
	/*
	 * mlock MCL_FUTURE?
	 */
	if (mm->def_flags & VM_LOCKED) {
		unsigned long locked, lock_limit;
2186 2187
		locked = len >> PAGE_SHIFT;
		locked += mm->locked_vm;
J
Jiri Slaby 已提交
2188
		lock_limit = rlimit(RLIMIT_MEMLOCK);
2189
		lock_limit >>= PAGE_SHIFT;
L
Linus Torvalds 已提交
2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211
		if (locked > lock_limit && !capable(CAP_IPC_LOCK))
			return -EAGAIN;
	}

	/*
	 * mm->mmap_sem is required to protect against another thread
	 * changing the mappings in case we sleep.
	 */
	verify_mm_writelocked(mm);

	/*
	 * Clear old maps.  this also does some error checking for us
	 */
 munmap_back:
	vma = find_vma_prepare(mm, addr, &prev, &rb_link, &rb_parent);
	if (vma && vma->vm_start < addr + len) {
		if (do_munmap(mm, addr, len))
			return -ENOMEM;
		goto munmap_back;
	}

	/* Check against address space limits *after* clearing old maps... */
A
akpm@osdl.org 已提交
2212
	if (!may_expand_vm(mm, len >> PAGE_SHIFT))
L
Linus Torvalds 已提交
2213 2214 2215 2216 2217
		return -ENOMEM;

	if (mm->map_count > sysctl_max_map_count)
		return -ENOMEM;

2218
	if (security_vm_enough_memory_mm(mm, len >> PAGE_SHIFT))
L
Linus Torvalds 已提交
2219 2220 2221
		return -ENOMEM;

	/* Can we just expand an old private anonymous mapping? */
2222 2223 2224
	vma = vma_merge(mm, prev, addr, addr + len, flags,
					NULL, NULL, pgoff, NULL);
	if (vma)
L
Linus Torvalds 已提交
2225 2226 2227 2228 2229
		goto out;

	/*
	 * create a vma struct for an anonymous mapping
	 */
P
Pekka Enberg 已提交
2230
	vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL);
L
Linus Torvalds 已提交
2231 2232 2233 2234 2235
	if (!vma) {
		vm_unacct_memory(len >> PAGE_SHIFT);
		return -ENOMEM;
	}

2236
	INIT_LIST_HEAD(&vma->anon_vma_chain);
L
Linus Torvalds 已提交
2237 2238 2239 2240 2241
	vma->vm_mm = mm;
	vma->vm_start = addr;
	vma->vm_end = addr + len;
	vma->vm_pgoff = pgoff;
	vma->vm_flags = flags;
2242
	vma->vm_page_prot = vm_get_page_prot(flags);
L
Linus Torvalds 已提交
2243 2244
	vma_link(mm, vma, prev, rb_link, rb_parent);
out:
2245
	perf_event_mmap(vma);
L
Linus Torvalds 已提交
2246 2247
	mm->total_vm += len >> PAGE_SHIFT;
	if (flags & VM_LOCKED) {
2248 2249
		if (!mlock_vma_pages_range(vma, addr, addr + len))
			mm->locked_vm += (len >> PAGE_SHIFT);
L
Linus Torvalds 已提交
2250 2251 2252 2253
	}
	return addr;
}

2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264
unsigned long vm_brk(unsigned long addr, unsigned long len)
{
	struct mm_struct *mm = current->mm;
	unsigned long ret;

	down_write(&mm->mmap_sem);
	ret = do_brk(addr, len);
	up_write(&mm->mmap_sem);
	return ret;
}
EXPORT_SYMBOL(vm_brk);
L
Linus Torvalds 已提交
2265 2266 2267 2268

/* Release all mmaps. */
void exit_mmap(struct mm_struct *mm)
{
P
Peter Zijlstra 已提交
2269
	struct mmu_gather tlb;
2270
	struct vm_area_struct *vma;
L
Linus Torvalds 已提交
2271 2272
	unsigned long nr_accounted = 0;

2273
	/* mm's last user has gone, and its about to be pulled down */
A
Andrea Arcangeli 已提交
2274
	mmu_notifier_release(mm);
2275

2276 2277 2278 2279 2280 2281 2282 2283
	if (mm->locked_vm) {
		vma = mm->mmap;
		while (vma) {
			if (vma->vm_flags & VM_LOCKED)
				munlock_vma_pages_all(vma);
			vma = vma->vm_next;
		}
	}
2284 2285 2286

	arch_exit_mmap(mm);

2287
	vma = mm->mmap;
2288 2289 2290
	if (!vma)	/* Can happen if dup_mmap() received an OOM */
		return;

L
Linus Torvalds 已提交
2291 2292
	lru_add_drain();
	flush_cache_mm(mm);
P
Peter Zijlstra 已提交
2293
	tlb_gather_mmu(&tlb, mm, 1);
2294
	/* update_hiwater_rss(mm) here? but nobody should be looking */
2295
	/* Use -1 here to ensure all VMAs in the mm are unmapped */
2296
	unmap_vmas(&tlb, vma, 0, -1);
H
Hugh Dickins 已提交
2297

P
Peter Zijlstra 已提交
2298
	free_pgtables(&tlb, vma, FIRST_USER_ADDRESS, 0);
2299
	tlb_finish_mmu(&tlb, 0, -1);
L
Linus Torvalds 已提交
2300 2301

	/*
2302 2303
	 * Walk the list again, actually closing and freeing it,
	 * with preemption enabled, without holding any MM locks.
L
Linus Torvalds 已提交
2304
	 */
2305 2306 2307
	while (vma) {
		if (vma->vm_flags & VM_ACCOUNT)
			nr_accounted += vma_pages(vma);
2308
		vma = remove_vma(vma);
2309 2310
	}
	vm_unacct_memory(nr_accounted);
2311

2312
	BUG_ON(mm->nr_ptes > (FIRST_USER_ADDRESS+PMD_SIZE-1)>>PMD_SHIFT);
L
Linus Torvalds 已提交
2313 2314 2315 2316
}

/* Insert vm structure into process list sorted by address
 * and into the inode's i_mmap tree.  If vm_file is non-NULL
2317
 * then i_mmap_mutex is taken here.
L
Linus Torvalds 已提交
2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342
 */
int insert_vm_struct(struct mm_struct * mm, struct vm_area_struct * vma)
{
	struct vm_area_struct * __vma, * prev;
	struct rb_node ** rb_link, * rb_parent;

	/*
	 * The vm_pgoff of a purely anonymous vma should be irrelevant
	 * until its first write fault, when page's anon_vma and index
	 * are set.  But now set the vm_pgoff it will almost certainly
	 * end up with (unless mremap moves it elsewhere before that
	 * first wfault), so /proc/pid/maps tells a consistent story.
	 *
	 * By setting it to reflect the virtual start address of the
	 * vma, merges and splits can happen in a seamless way, just
	 * using the existing file pgoff checks and manipulations.
	 * Similarly in do_mmap_pgoff and in do_brk.
	 */
	if (!vma->vm_file) {
		BUG_ON(vma->anon_vma);
		vma->vm_pgoff = vma->vm_start >> PAGE_SHIFT;
	}
	__vma = find_vma_prepare(mm,vma->vm_start,&prev,&rb_link,&rb_parent);
	if (__vma && __vma->vm_start < vma->vm_end)
		return -ENOMEM;
2343
	if ((vma->vm_flags & VM_ACCOUNT) &&
2344
	     security_vm_enough_memory_mm(mm, vma_pages(vma)))
2345
		return -ENOMEM;
2346

2347
	if (vma->vm_file && uprobe_mmap(vma))
2348 2349
		return -EINVAL;

L
Linus Torvalds 已提交
2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366
	vma_link(mm, vma, prev, rb_link, rb_parent);
	return 0;
}

/*
 * Copy the vma structure to a new location in the same mm,
 * prior to moving page table entries, to effect an mremap move.
 */
struct vm_area_struct *copy_vma(struct vm_area_struct **vmap,
	unsigned long addr, unsigned long len, pgoff_t pgoff)
{
	struct vm_area_struct *vma = *vmap;
	unsigned long vma_start = vma->vm_start;
	struct mm_struct *mm = vma->vm_mm;
	struct vm_area_struct *new_vma, *prev;
	struct rb_node **rb_link, *rb_parent;
	struct mempolicy *pol;
2367
	bool faulted_in_anon_vma = true;
L
Linus Torvalds 已提交
2368 2369 2370 2371 2372

	/*
	 * If anonymous vma has not yet been faulted, update new pgoff
	 * to match new location, to increase its chance of merging.
	 */
2373
	if (unlikely(!vma->vm_file && !vma->anon_vma)) {
L
Linus Torvalds 已提交
2374
		pgoff = addr >> PAGE_SHIFT;
2375 2376
		faulted_in_anon_vma = false;
	}
L
Linus Torvalds 已提交
2377 2378 2379 2380 2381 2382 2383 2384

	find_vma_prepare(mm, addr, &prev, &rb_link, &rb_parent);
	new_vma = vma_merge(mm, prev, addr, addr + len, vma->vm_flags,
			vma->anon_vma, vma->vm_file, pgoff, vma_policy(vma));
	if (new_vma) {
		/*
		 * Source vma may have been merged into new_vma
		 */
2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399
		if (unlikely(vma_start >= new_vma->vm_start &&
			     vma_start < new_vma->vm_end)) {
			/*
			 * The only way we can get a vma_merge with
			 * self during an mremap is if the vma hasn't
			 * been faulted in yet and we were allowed to
			 * reset the dst vma->vm_pgoff to the
			 * destination address of the mremap to allow
			 * the merge to happen. mremap must change the
			 * vm_pgoff linearity between src and dst vmas
			 * (in turn preventing a vma_merge) to be
			 * safe. It is only safe to keep the vm_pgoff
			 * linear if there are no pages mapped yet.
			 */
			VM_BUG_ON(faulted_in_anon_vma);
L
Linus Torvalds 已提交
2400
			*vmap = new_vma;
2401 2402
		} else
			anon_vma_moveto_tail(new_vma);
L
Linus Torvalds 已提交
2403
	} else {
2404
		new_vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL);
L
Linus Torvalds 已提交
2405 2406
		if (new_vma) {
			*new_vma = *vma;
2407
			pol = mpol_dup(vma_policy(vma));
2408 2409 2410 2411 2412
			if (IS_ERR(pol))
				goto out_free_vma;
			INIT_LIST_HEAD(&new_vma->anon_vma_chain);
			if (anon_vma_clone(new_vma, vma))
				goto out_free_mempol;
L
Linus Torvalds 已提交
2413 2414 2415 2416
			vma_set_policy(new_vma, pol);
			new_vma->vm_start = addr;
			new_vma->vm_end = addr + len;
			new_vma->vm_pgoff = pgoff;
M
Matt Helsley 已提交
2417
			if (new_vma->vm_file) {
L
Linus Torvalds 已提交
2418
				get_file(new_vma->vm_file);
2419

2420
				if (uprobe_mmap(new_vma))
2421 2422
					goto out_free_mempol;

M
Matt Helsley 已提交
2423 2424 2425
				if (vma->vm_flags & VM_EXECUTABLE)
					added_exe_file_vma(mm);
			}
L
Linus Torvalds 已提交
2426 2427 2428 2429 2430 2431
			if (new_vma->vm_ops && new_vma->vm_ops->open)
				new_vma->vm_ops->open(new_vma);
			vma_link(mm, new_vma, prev, rb_link, rb_parent);
		}
	}
	return new_vma;
2432 2433 2434 2435 2436 2437

 out_free_mempol:
	mpol_put(pol);
 out_free_vma:
	kmem_cache_free(vm_area_cachep, new_vma);
	return NULL;
L
Linus Torvalds 已提交
2438
}
A
akpm@osdl.org 已提交
2439 2440 2441 2442 2443 2444 2445 2446 2447 2448

/*
 * Return true if the calling process may expand its vm space by the passed
 * number of pages
 */
int may_expand_vm(struct mm_struct *mm, unsigned long npages)
{
	unsigned long cur = mm->total_vm;	/* pages */
	unsigned long lim;

J
Jiri Slaby 已提交
2449
	lim = rlimit(RLIMIT_AS) >> PAGE_SHIFT;
A
akpm@osdl.org 已提交
2450 2451 2452 2453 2454

	if (cur + npages > lim)
		return 0;
	return 1;
}
2455 2456


N
Nick Piggin 已提交
2457 2458
static int special_mapping_fault(struct vm_area_struct *vma,
				struct vm_fault *vmf)
2459
{
N
Nick Piggin 已提交
2460
	pgoff_t pgoff;
2461 2462
	struct page **pages;

N
Nick Piggin 已提交
2463 2464 2465 2466 2467 2468 2469
	/*
	 * special mappings have no vm_file, and in that case, the mm
	 * uses vm_pgoff internally. So we have to subtract it from here.
	 * We are allowed to do this because we are the mm; do not copy
	 * this code into drivers!
	 */
	pgoff = vmf->pgoff - vma->vm_pgoff;
2470

N
Nick Piggin 已提交
2471 2472
	for (pages = vma->vm_private_data; pgoff && *pages; ++pages)
		pgoff--;
2473 2474 2475 2476

	if (*pages) {
		struct page *page = *pages;
		get_page(page);
N
Nick Piggin 已提交
2477 2478
		vmf->page = page;
		return 0;
2479 2480
	}

N
Nick Piggin 已提交
2481
	return VM_FAULT_SIGBUS;
2482 2483 2484 2485 2486 2487 2488 2489 2490
}

/*
 * Having a close hook prevents vma merging regardless of flags.
 */
static void special_mapping_close(struct vm_area_struct *vma)
{
}

2491
static const struct vm_operations_struct special_mapping_vmops = {
2492
	.close = special_mapping_close,
N
Nick Piggin 已提交
2493
	.fault = special_mapping_fault,
2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508
};

/*
 * Called with mm->mmap_sem held for writing.
 * Insert a new vma covering the given region, with the given flags.
 * Its pages are supplied by the given array of struct page *.
 * The array can be shorter than len >> PAGE_SHIFT if it's null-terminated.
 * The region past the last page supplied will always produce SIGBUS.
 * The array pointer and the pages it points to are assumed to stay alive
 * for as long as this mapping might exist.
 */
int install_special_mapping(struct mm_struct *mm,
			    unsigned long addr, unsigned long len,
			    unsigned long vm_flags, struct page **pages)
{
2509
	int ret;
2510 2511 2512 2513 2514 2515
	struct vm_area_struct *vma;

	vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL);
	if (unlikely(vma == NULL))
		return -ENOMEM;

2516
	INIT_LIST_HEAD(&vma->anon_vma_chain);
2517 2518 2519 2520
	vma->vm_mm = mm;
	vma->vm_start = addr;
	vma->vm_end = addr + len;

2521
	vma->vm_flags = vm_flags | mm->def_flags | VM_DONTEXPAND;
2522
	vma->vm_page_prot = vm_get_page_prot(vma->vm_flags);
2523 2524 2525 2526

	vma->vm_ops = &special_mapping_vmops;
	vma->vm_private_data = pages;

2527 2528 2529
	ret = insert_vm_struct(mm, vma);
	if (ret)
		goto out;
2530 2531 2532

	mm->total_vm += len >> PAGE_SHIFT;

2533
	perf_event_mmap(vma);
2534

2535
	return 0;
2536 2537 2538 2539

out:
	kmem_cache_free(vm_area_cachep, vma);
	return ret;
2540
}
2541 2542 2543

static DEFINE_MUTEX(mm_all_locks_mutex);

2544
static void vm_lock_anon_vma(struct mm_struct *mm, struct anon_vma *anon_vma)
2545
{
2546
	if (!test_bit(0, (unsigned long *) &anon_vma->root->head.next)) {
2547 2548 2549 2550
		/*
		 * The LSB of head.next can't change from under us
		 * because we hold the mm_all_locks_mutex.
		 */
2551
		mutex_lock_nest_lock(&anon_vma->root->mutex, &mm->mmap_sem);
2552 2553
		/*
		 * We can safely modify head.next after taking the
2554
		 * anon_vma->root->mutex. If some other vma in this mm shares
2555 2556 2557 2558
		 * the same anon_vma we won't take it again.
		 *
		 * No need of atomic instructions here, head.next
		 * can't change from under us thanks to the
2559
		 * anon_vma->root->mutex.
2560 2561
		 */
		if (__test_and_set_bit(0, (unsigned long *)
2562
				       &anon_vma->root->head.next))
2563 2564 2565 2566
			BUG();
	}
}

2567
static void vm_lock_mapping(struct mm_struct *mm, struct address_space *mapping)
2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580
{
	if (!test_bit(AS_MM_ALL_LOCKS, &mapping->flags)) {
		/*
		 * AS_MM_ALL_LOCKS can't change from under us because
		 * we hold the mm_all_locks_mutex.
		 *
		 * Operations on ->flags have to be atomic because
		 * even if AS_MM_ALL_LOCKS is stable thanks to the
		 * mm_all_locks_mutex, there may be other cpus
		 * changing other bitflags in parallel to us.
		 */
		if (test_and_set_bit(AS_MM_ALL_LOCKS, &mapping->flags))
			BUG();
2581
		mutex_lock_nest_lock(&mapping->i_mmap_mutex, &mm->mmap_sem);
2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607
	}
}

/*
 * This operation locks against the VM for all pte/vma/mm related
 * operations that could ever happen on a certain mm. This includes
 * vmtruncate, try_to_unmap, and all page faults.
 *
 * The caller must take the mmap_sem in write mode before calling
 * mm_take_all_locks(). The caller isn't allowed to release the
 * mmap_sem until mm_drop_all_locks() returns.
 *
 * mmap_sem in write mode is required in order to block all operations
 * that could modify pagetables and free pages without need of
 * altering the vma layout (for example populate_range() with
 * nonlinear vmas). It's also needed in write mode to avoid new
 * anon_vmas to be associated with existing vmas.
 *
 * A single task can't take more than one mm_take_all_locks() in a row
 * or it would deadlock.
 *
 * The LSB in anon_vma->head.next and the AS_MM_ALL_LOCKS bitflag in
 * mapping->flags avoid to take the same lock twice, if more than one
 * vma in this mm is backed by the same anon_vma or address_space.
 *
 * We can take all the locks in random order because the VM code
2608
 * taking i_mmap_mutex or anon_vma->mutex outside the mmap_sem never
2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619
 * takes more than one of them in a row. Secondly we're protected
 * against a concurrent mm_take_all_locks() by the mm_all_locks_mutex.
 *
 * mm_take_all_locks() and mm_drop_all_locks are expensive operations
 * that may have to take thousand of locks.
 *
 * mm_take_all_locks() can fail if it's interrupted by signals.
 */
int mm_take_all_locks(struct mm_struct *mm)
{
	struct vm_area_struct *vma;
2620
	struct anon_vma_chain *avc;
2621 2622 2623 2624 2625 2626 2627 2628 2629

	BUG_ON(down_read_trylock(&mm->mmap_sem));

	mutex_lock(&mm_all_locks_mutex);

	for (vma = mm->mmap; vma; vma = vma->vm_next) {
		if (signal_pending(current))
			goto out_unlock;
		if (vma->vm_file && vma->vm_file->f_mapping)
2630
			vm_lock_mapping(mm, vma->vm_file->f_mapping);
2631
	}
2632 2633 2634 2635 2636

	for (vma = mm->mmap; vma; vma = vma->vm_next) {
		if (signal_pending(current))
			goto out_unlock;
		if (vma->anon_vma)
2637 2638
			list_for_each_entry(avc, &vma->anon_vma_chain, same_vma)
				vm_lock_anon_vma(mm, avc->anon_vma);
2639
	}
2640

2641
	return 0;
2642 2643

out_unlock:
2644 2645
	mm_drop_all_locks(mm);
	return -EINTR;
2646 2647 2648 2649
}

static void vm_unlock_anon_vma(struct anon_vma *anon_vma)
{
2650
	if (test_bit(0, (unsigned long *) &anon_vma->root->head.next)) {
2651 2652 2653 2654 2655 2656 2657 2658 2659 2660
		/*
		 * The LSB of head.next can't change to 0 from under
		 * us because we hold the mm_all_locks_mutex.
		 *
		 * We must however clear the bitflag before unlocking
		 * the vma so the users using the anon_vma->head will
		 * never see our bitflag.
		 *
		 * No need of atomic instructions here, head.next
		 * can't change from under us until we release the
2661
		 * anon_vma->root->mutex.
2662 2663
		 */
		if (!__test_and_clear_bit(0, (unsigned long *)
2664
					  &anon_vma->root->head.next))
2665
			BUG();
2666
		anon_vma_unlock(anon_vma);
2667 2668 2669 2670 2671 2672 2673 2674 2675 2676
	}
}

static void vm_unlock_mapping(struct address_space *mapping)
{
	if (test_bit(AS_MM_ALL_LOCKS, &mapping->flags)) {
		/*
		 * AS_MM_ALL_LOCKS can't change to 0 from under us
		 * because we hold the mm_all_locks_mutex.
		 */
2677
		mutex_unlock(&mapping->i_mmap_mutex);
2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690
		if (!test_and_clear_bit(AS_MM_ALL_LOCKS,
					&mapping->flags))
			BUG();
	}
}

/*
 * The mmap_sem cannot be released by the caller until
 * mm_drop_all_locks() returns.
 */
void mm_drop_all_locks(struct mm_struct *mm)
{
	struct vm_area_struct *vma;
2691
	struct anon_vma_chain *avc;
2692 2693 2694 2695 2696 2697

	BUG_ON(down_read_trylock(&mm->mmap_sem));
	BUG_ON(!mutex_is_locked(&mm_all_locks_mutex));

	for (vma = mm->mmap; vma; vma = vma->vm_next) {
		if (vma->anon_vma)
2698 2699
			list_for_each_entry(avc, &vma->anon_vma_chain, same_vma)
				vm_unlock_anon_vma(avc->anon_vma);
2700 2701 2702 2703 2704 2705
		if (vma->vm_file && vma->vm_file->f_mapping)
			vm_unlock_mapping(vma->vm_file->f_mapping);
	}

	mutex_unlock(&mm_all_locks_mutex);
}
2706 2707 2708 2709 2710 2711

/*
 * initialise the VMA slab
 */
void __init mmap_init(void)
{
2712 2713 2714 2715
	int ret;

	ret = percpu_counter_init(&vm_committed_as, 0);
	VM_BUG_ON(ret);
2716
}