Blame view

mm/util.c 22.6 KB
457c89965   Thomas Gleixner   treewide: Add SPD...
1
  // SPDX-License-Identifier: GPL-2.0-only
16d69265b   Andrew Morton   uninline arch_pic...
2
  #include <linux/mm.h>
30992c97a   Matt Mackall   [PATCH] slob: int...
3
4
  #include <linux/slab.h>
  #include <linux/string.h>
3b32123d7   Gideon Israel Dsouza   mm: use macros fr...
5
  #include <linux/compiler.h>
b95f1b31b   Paul Gortmaker   mm: Map most file...
6
  #include <linux/export.h>
96840aa00   Davi Arnaut   [PATCH] strndup_u...
7
  #include <linux/err.h>
3b8f14b41   Adrian Bunk   mm/util.c must #i...
8
  #include <linux/sched.h>
6e84f3152   Ingo Molnar   sched/headers: Pr...
9
  #include <linux/sched/mm.h>
79eb597cb   Daniel Jordan   mm: add account_l...
10
  #include <linux/sched/signal.h>
68db0cf10   Ingo Molnar   sched/headers: Pr...
11
  #include <linux/sched/task_stack.h>
eb36c5873   Al Viro   new helper: vm_mm...
12
  #include <linux/security.h>
9800339b5   Shaohua Li   mm: don't inline ...
13
  #include <linux/swap.h>
33806f06d   Shaohua Li   swap: make each s...
14
  #include <linux/swapops.h>
00619bcc4   Jerome Marchand   mm: factor commit...
15
16
  #include <linux/mman.h>
  #include <linux/hugetlb.h>
39f1f78d5   Al Viro   nick kvfree() fro...
17
  #include <linux/vmalloc.h>
897ab3e0c   Mike Rapoport   userfaultfd: non-...
18
  #include <linux/userfaultfd_k.h>
649775be6   Alexandre Ghiti   mm, fs: move rand...
19
  #include <linux/elf.h>
67f3977f8   Alexandre Ghiti   arm64, mm: move g...
20
21
  #include <linux/elf-randomize.h>
  #include <linux/personality.h>
649775be6   Alexandre Ghiti   mm, fs: move rand...
22
  #include <linux/random.h>
67f3977f8   Alexandre Ghiti   arm64, mm: move g...
23
24
25
  #include <linux/processor.h>
  #include <linux/sizes.h>
  #include <linux/compat.h>
00619bcc4   Jerome Marchand   mm: factor commit...
26

7c0f6ba68   Linus Torvalds   Replace <asm/uacc...
27
  #include <linux/uaccess.h>
30992c97a   Matt Mackall   [PATCH] slob: int...
28

6038def0d   Namhyung Kim   mm: nommu: sort m...
29
  #include "internal.h"
a4bb1e43e   Andrzej Hajda   mm/util: add kstr...
30
31
32
33
34
35
36
37
38
39
40
41
  /**
   * kfree_const - conditionally free memory
   * @x: pointer to the memory
   *
   * Function calls kfree only if @x is not in .rodata section.
   */
  void kfree_const(const void *x)
  {
  	if (!is_kernel_rodata((unsigned long)x))
  		kfree(x);
  }
  EXPORT_SYMBOL(kfree_const);
30992c97a   Matt Mackall   [PATCH] slob: int...
42
  /**
30992c97a   Matt Mackall   [PATCH] slob: int...
43
   * kstrdup - allocate space for and copy an existing string
30992c97a   Matt Mackall   [PATCH] slob: int...
44
45
   * @s: the string to duplicate
   * @gfp: the GFP mask used in the kmalloc() call when allocating memory
a862f68a8   Mike Rapoport   docs/core-api/mm:...
46
47
   *
   * Return: newly allocated copy of @s or %NULL in case of error
30992c97a   Matt Mackall   [PATCH] slob: int...
48
49
50
51
52
53
54
55
56
57
   */
  char *kstrdup(const char *s, gfp_t gfp)
  {
  	size_t len;
  	char *buf;
  
  	if (!s)
  		return NULL;
  
  	len = strlen(s) + 1;
1d2c8eea6   Christoph Hellwig   [PATCH] slab: cle...
58
  	buf = kmalloc_track_caller(len, gfp);
30992c97a   Matt Mackall   [PATCH] slob: int...
59
60
61
62
63
  	if (buf)
  		memcpy(buf, s, len);
  	return buf;
  }
  EXPORT_SYMBOL(kstrdup);
96840aa00   Davi Arnaut   [PATCH] strndup_u...
64

1a2f67b45   Alexey Dobriyan   [PATCH] kmemdup: ...
65
  /**
a4bb1e43e   Andrzej Hajda   mm/util: add kstr...
66
67
68
69
   * kstrdup_const - conditionally duplicate an existing const string
   * @s: the string to duplicate
   * @gfp: the GFP mask used in the kmalloc() call when allocating memory
   *
a862f68a8   Mike Rapoport   docs/core-api/mm:...
70
71
72
73
   * Note: Strings allocated by kstrdup_const should be freed by kfree_const.
   *
   * Return: source string if it is in .rodata section otherwise
   * fallback to kstrdup.
a4bb1e43e   Andrzej Hajda   mm/util: add kstr...
74
75
76
77
78
79
80
81
82
83
84
   */
  const char *kstrdup_const(const char *s, gfp_t gfp)
  {
  	if (is_kernel_rodata((unsigned long)s))
  		return s;
  
  	return kstrdup(s, gfp);
  }
  EXPORT_SYMBOL(kstrdup_const);
  
  /**
1e66df3ee   Jeremy Fitzhardinge   add kstrndup
85
86
87
88
   * kstrndup - allocate space for and copy an existing string
   * @s: the string to duplicate
   * @max: read at most @max chars from @s
   * @gfp: the GFP mask used in the kmalloc() call when allocating memory
f35157417   David Howells   Provide a functio...
89
90
   *
   * Note: Use kmemdup_nul() instead if the size is known exactly.
a862f68a8   Mike Rapoport   docs/core-api/mm:...
91
92
   *
   * Return: newly allocated copy of @s or %NULL in case of error
1e66df3ee   Jeremy Fitzhardinge   add kstrndup
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
   */
  char *kstrndup(const char *s, size_t max, gfp_t gfp)
  {
  	size_t len;
  	char *buf;
  
  	if (!s)
  		return NULL;
  
  	len = strnlen(s, max);
  	buf = kmalloc_track_caller(len+1, gfp);
  	if (buf) {
  		memcpy(buf, s, len);
  		buf[len] = '\0';
  	}
  	return buf;
  }
  EXPORT_SYMBOL(kstrndup);
  
  /**
1a2f67b45   Alexey Dobriyan   [PATCH] kmemdup: ...
113
114
115
116
117
   * kmemdup - duplicate region of memory
   *
   * @src: memory region to duplicate
   * @len: memory region length
   * @gfp: GFP mask to use
a862f68a8   Mike Rapoport   docs/core-api/mm:...
118
119
   *
   * Return: newly allocated copy of @src or %NULL in case of error
1a2f67b45   Alexey Dobriyan   [PATCH] kmemdup: ...
120
121
122
123
   */
  void *kmemdup(const void *src, size_t len, gfp_t gfp)
  {
  	void *p;
1d2c8eea6   Christoph Hellwig   [PATCH] slab: cle...
124
  	p = kmalloc_track_caller(len, gfp);
1a2f67b45   Alexey Dobriyan   [PATCH] kmemdup: ...
125
126
127
128
129
  	if (p)
  		memcpy(p, src, len);
  	return p;
  }
  EXPORT_SYMBOL(kmemdup);
ef2ad80c7   Christoph Lameter   Slab allocators: ...
130
  /**
f35157417   David Howells   Provide a functio...
131
132
133
134
   * kmemdup_nul - Create a NUL-terminated string from unterminated data
   * @s: The data to stringify
   * @len: The size of the data
   * @gfp: the GFP mask used in the kmalloc() call when allocating memory
a862f68a8   Mike Rapoport   docs/core-api/mm:...
135
136
137
   *
   * Return: newly allocated copy of @s with NUL-termination or %NULL in
   * case of error
f35157417   David Howells   Provide a functio...
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
   */
  char *kmemdup_nul(const char *s, size_t len, gfp_t gfp)
  {
  	char *buf;
  
  	if (!s)
  		return NULL;
  
  	buf = kmalloc_track_caller(len + 1, gfp);
  	if (buf) {
  		memcpy(buf, s, len);
  		buf[len] = '\0';
  	}
  	return buf;
  }
  EXPORT_SYMBOL(kmemdup_nul);
  
  /**
610a77e04   Li Zefan   memdup_user(): in...
156
157
158
159
160
   * memdup_user - duplicate memory region from user space
   *
   * @src: source address in user space
   * @len: number of bytes to copy
   *
a862f68a8   Mike Rapoport   docs/core-api/mm:...
161
   * Return: an ERR_PTR() on failure.  Result is physically
50fd2f298   Al Viro   new primitive: vm...
162
   * contiguous, to be freed by kfree().
610a77e04   Li Zefan   memdup_user(): in...
163
164
165
166
   */
  void *memdup_user(const void __user *src, size_t len)
  {
  	void *p;
6c8fcc096   Daniel Vetter   mm: don't let use...
167
  	p = kmalloc_track_caller(len, GFP_USER | __GFP_NOWARN);
610a77e04   Li Zefan   memdup_user(): in...
168
169
170
171
172
173
174
175
176
177
178
  	if (!p)
  		return ERR_PTR(-ENOMEM);
  
  	if (copy_from_user(p, src, len)) {
  		kfree(p);
  		return ERR_PTR(-EFAULT);
  	}
  
  	return p;
  }
  EXPORT_SYMBOL(memdup_user);
50fd2f298   Al Viro   new primitive: vm...
179
180
181
182
183
184
  /**
   * vmemdup_user - duplicate memory region from user space
   *
   * @src: source address in user space
   * @len: number of bytes to copy
   *
a862f68a8   Mike Rapoport   docs/core-api/mm:...
185
   * Return: an ERR_PTR() on failure.  Result may be not
50fd2f298   Al Viro   new primitive: vm...
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
   * physically contiguous.  Use kvfree() to free.
   */
  void *vmemdup_user(const void __user *src, size_t len)
  {
  	void *p;
  
  	p = kvmalloc(len, GFP_USER);
  	if (!p)
  		return ERR_PTR(-ENOMEM);
  
  	if (copy_from_user(p, src, len)) {
  		kvfree(p);
  		return ERR_PTR(-EFAULT);
  	}
  
  	return p;
  }
  EXPORT_SYMBOL(vmemdup_user);
b86181f1a   Mike Rapoport   mm/util: make str...
204
  /**
96840aa00   Davi Arnaut   [PATCH] strndup_u...
205
   * strndup_user - duplicate an existing string from user space
96840aa00   Davi Arnaut   [PATCH] strndup_u...
206
207
   * @s: The string to duplicate
   * @n: Maximum number of bytes to copy, including the trailing NUL.
a862f68a8   Mike Rapoport   docs/core-api/mm:...
208
   *
e91455217   Andrew Morton   mm/util.c: fix st...
209
   * Return: newly allocated copy of @s or an ERR_PTR() in case of error
96840aa00   Davi Arnaut   [PATCH] strndup_u...
210
211
212
213
214
215
216
217
218
219
220
221
222
   */
  char *strndup_user(const char __user *s, long n)
  {
  	char *p;
  	long length;
  
  	length = strnlen_user(s, n);
  
  	if (!length)
  		return ERR_PTR(-EFAULT);
  
  	if (length > n)
  		return ERR_PTR(-EINVAL);
90d740455   Julia Lawall   mm: use memdup_user
223
  	p = memdup_user(s, length);
96840aa00   Davi Arnaut   [PATCH] strndup_u...
224

90d740455   Julia Lawall   mm: use memdup_user
225
226
  	if (IS_ERR(p))
  		return p;
96840aa00   Davi Arnaut   [PATCH] strndup_u...
227
228
229
230
231
232
  
  	p[length - 1] = '\0';
  
  	return p;
  }
  EXPORT_SYMBOL(strndup_user);
16d69265b   Andrew Morton   uninline arch_pic...
233

e9d408e10   Al Viro   new helper: memdu...
234
235
236
237
238
239
  /**
   * memdup_user_nul - duplicate memory region from user space and NUL-terminate
   *
   * @src: source address in user space
   * @len: number of bytes to copy
   *
a862f68a8   Mike Rapoport   docs/core-api/mm:...
240
   * Return: an ERR_PTR() on failure.
e9d408e10   Al Viro   new helper: memdu...
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
   */
  void *memdup_user_nul(const void __user *src, size_t len)
  {
  	char *p;
  
  	/*
  	 * Always use GFP_KERNEL, since copy_from_user() can sleep and
  	 * cause pagefault, which makes it pointless to use GFP_NOFS
  	 * or GFP_ATOMIC.
  	 */
  	p = kmalloc_track_caller(len + 1, GFP_KERNEL);
  	if (!p)
  		return ERR_PTR(-ENOMEM);
  
  	if (copy_from_user(p, src, len)) {
  		kfree(p);
  		return ERR_PTR(-EFAULT);
  	}
  	p[len] = '\0';
  
  	return p;
  }
  EXPORT_SYMBOL(memdup_user_nul);
6038def0d   Namhyung Kim   mm: nommu: sort m...
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
  void __vma_link_list(struct mm_struct *mm, struct vm_area_struct *vma,
  		struct vm_area_struct *prev, struct rb_node *rb_parent)
  {
  	struct vm_area_struct *next;
  
  	vma->vm_prev = prev;
  	if (prev) {
  		next = prev->vm_next;
  		prev->vm_next = vma;
  	} else {
  		mm->mmap = vma;
  		if (rb_parent)
  			next = rb_entry(rb_parent,
  					struct vm_area_struct, vm_rb);
  		else
  			next = NULL;
  	}
  	vma->vm_next = next;
  	if (next)
  		next->vm_prev = vma;
  }
b76437579   Siddhesh Poyarekar   procfs: mark thre...
285
  /* Check if the vma is being used as a stack by this task */
d17af5056   Andy Lutomirski   mm: Change vm_is_...
286
  int vma_is_stack_for_current(struct vm_area_struct *vma)
b76437579   Siddhesh Poyarekar   procfs: mark thre...
287
  {
d17af5056   Andy Lutomirski   mm: Change vm_is_...
288
  	struct task_struct * __maybe_unused t = current;
b76437579   Siddhesh Poyarekar   procfs: mark thre...
289
290
  	return (vma->vm_start <= KSTK_ESP(t) && vma->vm_end >= KSTK_ESP(t));
  }
649775be6   Alexandre Ghiti   mm, fs: move rand...
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
  #ifndef STACK_RND_MASK
  #define STACK_RND_MASK (0x7ff >> (PAGE_SHIFT - 12))     /* 8MB of VA */
  #endif
  
  unsigned long randomize_stack_top(unsigned long stack_top)
  {
  	unsigned long random_variable = 0;
  
  	if (current->flags & PF_RANDOMIZE) {
  		random_variable = get_random_long();
  		random_variable &= STACK_RND_MASK;
  		random_variable <<= PAGE_SHIFT;
  	}
  #ifdef CONFIG_STACK_GROWSUP
  	return PAGE_ALIGN(stack_top) + random_variable;
  #else
  	return PAGE_ALIGN(stack_top) - random_variable;
  #endif
  }
67f3977f8   Alexandre Ghiti   arm64, mm: move g...
310
  #ifdef CONFIG_ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT
e7142bf5d   Alexandre Ghiti   arm64, mm: make r...
311
312
313
314
315
316
317
318
  unsigned long arch_randomize_brk(struct mm_struct *mm)
  {
  	/* Is the current task 32bit ? */
  	if (!IS_ENABLED(CONFIG_64BIT) || is_compat_task())
  		return randomize_page(mm->brk, SZ_32M);
  
  	return randomize_page(mm->brk, SZ_1G);
  }
67f3977f8   Alexandre Ghiti   arm64, mm: move g...
319
320
321
322
323
324
325
326
327
328
329
330
331
  unsigned long arch_mmap_rnd(void)
  {
  	unsigned long rnd;
  
  #ifdef CONFIG_HAVE_ARCH_MMAP_RND_COMPAT_BITS
  	if (is_compat_task())
  		rnd = get_random_long() & ((1UL << mmap_rnd_compat_bits) - 1);
  	else
  #endif /* CONFIG_HAVE_ARCH_MMAP_RND_COMPAT_BITS */
  		rnd = get_random_long() & ((1UL << mmap_rnd_bits) - 1);
  
  	return rnd << PAGE_SHIFT;
  }
67f3977f8   Alexandre Ghiti   arm64, mm: move g...
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
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
375
376
377
378
379
380
381
382
383
384
385
386
387
  
  static int mmap_is_legacy(struct rlimit *rlim_stack)
  {
  	if (current->personality & ADDR_COMPAT_LAYOUT)
  		return 1;
  
  	if (rlim_stack->rlim_cur == RLIM_INFINITY)
  		return 1;
  
  	return sysctl_legacy_va_layout;
  }
  
  /*
   * Leave enough space between the mmap area and the stack to honour ulimit in
   * the face of randomisation.
   */
  #define MIN_GAP		(SZ_128M)
  #define MAX_GAP		(STACK_TOP / 6 * 5)
  
  static unsigned long mmap_base(unsigned long rnd, struct rlimit *rlim_stack)
  {
  	unsigned long gap = rlim_stack->rlim_cur;
  	unsigned long pad = stack_guard_gap;
  
  	/* Account for stack randomization if necessary */
  	if (current->flags & PF_RANDOMIZE)
  		pad += (STACK_RND_MASK << PAGE_SHIFT);
  
  	/* Values close to RLIM_INFINITY can overflow. */
  	if (gap + pad > gap)
  		gap += pad;
  
  	if (gap < MIN_GAP)
  		gap = MIN_GAP;
  	else if (gap > MAX_GAP)
  		gap = MAX_GAP;
  
  	return PAGE_ALIGN(STACK_TOP - gap - rnd);
  }
  
  void arch_pick_mmap_layout(struct mm_struct *mm, struct rlimit *rlim_stack)
  {
  	unsigned long random_factor = 0UL;
  
  	if (current->flags & PF_RANDOMIZE)
  		random_factor = arch_mmap_rnd();
  
  	if (mmap_is_legacy(rlim_stack)) {
  		mm->mmap_base = TASK_UNMAPPED_BASE + random_factor;
  		mm->get_unmapped_area = arch_get_unmapped_area;
  	} else {
  		mm->mmap_base = mmap_base(random_factor, rlim_stack);
  		mm->get_unmapped_area = arch_get_unmapped_area_topdown;
  	}
  }
  #elif defined(CONFIG_MMU) && !defined(HAVE_ARCH_PICK_MMAP_LAYOUT)
8f2af155b   Kees Cook   exec: pass stack ...
388
  void arch_pick_mmap_layout(struct mm_struct *mm, struct rlimit *rlim_stack)
16d69265b   Andrew Morton   uninline arch_pic...
389
390
391
  {
  	mm->mmap_base = TASK_UNMAPPED_BASE;
  	mm->get_unmapped_area = arch_get_unmapped_area;
16d69265b   Andrew Morton   uninline arch_pic...
392
393
  }
  #endif
912985dce   Rusty Russell   mm: Make generic ...
394

79eb597cb   Daniel Jordan   mm: add account_l...
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
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
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
  /**
   * __account_locked_vm - account locked pages to an mm's locked_vm
   * @mm:          mm to account against
   * @pages:       number of pages to account
   * @inc:         %true if @pages should be considered positive, %false if not
   * @task:        task used to check RLIMIT_MEMLOCK
   * @bypass_rlim: %true if checking RLIMIT_MEMLOCK should be skipped
   *
   * Assumes @task and @mm are valid (i.e. at least one reference on each), and
   * that mmap_sem is held as writer.
   *
   * Return:
   * * 0       on success
   * * -ENOMEM if RLIMIT_MEMLOCK would be exceeded.
   */
  int __account_locked_vm(struct mm_struct *mm, unsigned long pages, bool inc,
  			struct task_struct *task, bool bypass_rlim)
  {
  	unsigned long locked_vm, limit;
  	int ret = 0;
  
  	lockdep_assert_held_write(&mm->mmap_sem);
  
  	locked_vm = mm->locked_vm;
  	if (inc) {
  		if (!bypass_rlim) {
  			limit = task_rlimit(task, RLIMIT_MEMLOCK) >> PAGE_SHIFT;
  			if (locked_vm + pages > limit)
  				ret = -ENOMEM;
  		}
  		if (!ret)
  			mm->locked_vm = locked_vm + pages;
  	} else {
  		WARN_ON_ONCE(pages > locked_vm);
  		mm->locked_vm = locked_vm - pages;
  	}
  
  	pr_debug("%s: [%d] caller %ps %c%lu %lu/%lu%s
  ", __func__, task->pid,
  		 (void *)_RET_IP_, (inc) ? '+' : '-', pages << PAGE_SHIFT,
  		 locked_vm << PAGE_SHIFT, task_rlimit(task, RLIMIT_MEMLOCK),
  		 ret ? " - exceeded" : "");
  
  	return ret;
  }
  EXPORT_SYMBOL_GPL(__account_locked_vm);
  
  /**
   * account_locked_vm - account locked pages to an mm's locked_vm
   * @mm:          mm to account against, may be NULL
   * @pages:       number of pages to account
   * @inc:         %true if @pages should be considered positive, %false if not
   *
   * Assumes a non-NULL @mm is valid (i.e. at least one reference on it).
   *
   * Return:
   * * 0       on success, or if mm is NULL
   * * -ENOMEM if RLIMIT_MEMLOCK would be exceeded.
   */
  int account_locked_vm(struct mm_struct *mm, unsigned long pages, bool inc)
  {
  	int ret;
  
  	if (pages == 0 || !mm)
  		return 0;
  
  	down_write(&mm->mmap_sem);
  	ret = __account_locked_vm(mm, pages, inc, current,
  				  capable(CAP_IPC_LOCK));
  	up_write(&mm->mmap_sem);
  
  	return ret;
  }
  EXPORT_SYMBOL_GPL(account_locked_vm);
eb36c5873   Al Viro   new helper: vm_mm...
469
470
  unsigned long vm_mmap_pgoff(struct file *file, unsigned long addr,
  	unsigned long len, unsigned long prot,
9fbeb5ab5   Michal Hocko   mm: make vm_mmap ...
471
  	unsigned long flag, unsigned long pgoff)
eb36c5873   Al Viro   new helper: vm_mm...
472
473
474
  {
  	unsigned long ret;
  	struct mm_struct *mm = current->mm;
41badc15c   Michel Lespinasse   mm: make do_mmap_...
475
  	unsigned long populate;
897ab3e0c   Mike Rapoport   userfaultfd: non-...
476
  	LIST_HEAD(uf);
eb36c5873   Al Viro   new helper: vm_mm...
477
478
479
  
  	ret = security_mmap_file(file, prot, flag);
  	if (!ret) {
9fbeb5ab5   Michal Hocko   mm: make vm_mmap ...
480
481
  		if (down_write_killable(&mm->mmap_sem))
  			return -EINTR;
bebeb3d68   Michel Lespinasse   mm: introduce mm_...
482
  		ret = do_mmap_pgoff(file, addr, len, prot, flag, pgoff,
897ab3e0c   Mike Rapoport   userfaultfd: non-...
483
  				    &populate, &uf);
eb36c5873   Al Viro   new helper: vm_mm...
484
  		up_write(&mm->mmap_sem);
897ab3e0c   Mike Rapoport   userfaultfd: non-...
485
  		userfaultfd_unmap_complete(mm, &uf);
41badc15c   Michel Lespinasse   mm: make do_mmap_...
486
487
  		if (populate)
  			mm_populate(ret, populate);
eb36c5873   Al Viro   new helper: vm_mm...
488
489
490
491
492
493
494
495
496
497
  	}
  	return ret;
  }
  
  unsigned long vm_mmap(struct file *file, unsigned long addr,
  	unsigned long len, unsigned long prot,
  	unsigned long flag, unsigned long offset)
  {
  	if (unlikely(offset + PAGE_ALIGN(len) < offset))
  		return -EINVAL;
ea53cde08   Alexander Kuleshov   mm/util: use offs...
498
  	if (unlikely(offset_in_page(offset)))
eb36c5873   Al Viro   new helper: vm_mm...
499
  		return -EINVAL;
9fbeb5ab5   Michal Hocko   mm: make vm_mmap ...
500
  	return vm_mmap_pgoff(file, addr, len, prot, flag, offset >> PAGE_SHIFT);
eb36c5873   Al Viro   new helper: vm_mm...
501
502
  }
  EXPORT_SYMBOL(vm_mmap);
a7c3e901a   Michal Hocko   mm: introduce kv[...
503
504
505
506
507
508
509
510
511
512
  /**
   * kvmalloc_node - attempt to allocate physically contiguous memory, but upon
   * failure, fall back to non-contiguous (vmalloc) allocation.
   * @size: size of the request.
   * @flags: gfp mask for the allocation - must be compatible (superset) with GFP_KERNEL.
   * @node: numa node to allocate from
   *
   * Uses kmalloc to get the memory but if the allocation fails then falls back
   * to the vmalloc allocator. Use kvfree for freeing the memory.
   *
cc965a29d   Michal Hocko   mm: kvmalloc supp...
513
514
515
   * Reclaim modifiers - __GFP_NORETRY and __GFP_NOFAIL are not supported.
   * __GFP_RETRY_MAYFAIL is supported, and it should be used only if kmalloc is
   * preferable to the vmalloc fallback, due to visible performance drawbacks.
a7c3e901a   Michal Hocko   mm: introduce kv[...
516
   *
ce91f6ee5   Michal Hocko   mm: kvmalloc does...
517
518
   * Please note that any use of gfp flags outside of GFP_KERNEL is careful to not
   * fall back to vmalloc.
a862f68a8   Mike Rapoport   docs/core-api/mm:...
519
520
   *
   * Return: pointer to the allocated memory of %NULL in case of failure
a7c3e901a   Michal Hocko   mm: introduce kv[...
521
522
523
524
525
526
527
528
529
530
   */
  void *kvmalloc_node(size_t size, gfp_t flags, int node)
  {
  	gfp_t kmalloc_flags = flags;
  	void *ret;
  
  	/*
  	 * vmalloc uses GFP_KERNEL for some internal allocations (e.g page tables)
  	 * so the given set of flags has to be compatible.
  	 */
ce91f6ee5   Michal Hocko   mm: kvmalloc does...
531
532
  	if ((flags & GFP_KERNEL) != GFP_KERNEL)
  		return kmalloc_node(size, flags, node);
a7c3e901a   Michal Hocko   mm: introduce kv[...
533
534
  
  	/*
4f4f2ba9c   Michal Hocko   mm: clarify why w...
535
536
537
538
539
  	 * We want to attempt a large physically contiguous block first because
  	 * it is less likely to fragment multiple larger blocks and therefore
  	 * contribute to a long term fragmentation less than vmalloc fallback.
  	 * However make sure that larger requests are not too disruptive - no
  	 * OOM killer and no allocation failure warnings as we have a fallback.
a7c3e901a   Michal Hocko   mm: introduce kv[...
540
  	 */
6c5ab6511   Michal Hocko   mm: support __GFP...
541
542
  	if (size > PAGE_SIZE) {
  		kmalloc_flags |= __GFP_NOWARN;
cc965a29d   Michal Hocko   mm: kvmalloc supp...
543
  		if (!(kmalloc_flags & __GFP_RETRY_MAYFAIL))
6c5ab6511   Michal Hocko   mm: support __GFP...
544
545
  			kmalloc_flags |= __GFP_NORETRY;
  	}
a7c3e901a   Michal Hocko   mm: introduce kv[...
546
547
548
549
550
551
552
553
554
  
  	ret = kmalloc_node(size, kmalloc_flags, node);
  
  	/*
  	 * It doesn't really make sense to fallback to vmalloc for sub page
  	 * requests
  	 */
  	if (ret || size <= PAGE_SIZE)
  		return ret;
8594a21cf   Michal Hocko   mm, vmalloc: fix ...
555
556
  	return __vmalloc_node_flags_caller(size, node, flags,
  			__builtin_return_address(0));
a7c3e901a   Michal Hocko   mm: introduce kv[...
557
558
  }
  EXPORT_SYMBOL(kvmalloc_node);
ff4dc7729   Mike Rapoport   mm/util: add kern...
559
  /**
04b8e9460   Andrew Morton   mm/util.c: improv...
560
561
   * kvfree() - Free memory.
   * @addr: Pointer to allocated memory.
ff4dc7729   Mike Rapoport   mm/util: add kern...
562
   *
04b8e9460   Andrew Morton   mm/util.c: improv...
563
564
565
566
   * kvfree frees memory allocated by any of vmalloc(), kmalloc() or kvmalloc().
   * It is slightly more efficient to use kfree() or vfree() if you are certain
   * that you know which one to use.
   *
52414d330   Andrey Ryabinin   kvfree(): fix mis...
567
   * Context: Either preemptible task context or not-NMI interrupt.
ff4dc7729   Mike Rapoport   mm/util: add kern...
568
   */
39f1f78d5   Al Viro   nick kvfree() fro...
569
570
571
572
573
574
575
576
  void kvfree(const void *addr)
  {
  	if (is_vmalloc_addr(addr))
  		vfree(addr);
  	else
  		kfree(addr);
  }
  EXPORT_SYMBOL(kvfree);
0b11ec4ae   Waiman Long   mm: add kvfree_se...
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
  /**
   * kvfree_sensitive - Free a data object containing sensitive information.
   * @addr: address of the data object to be freed.
   * @len: length of the data object.
   *
   * Use the special memzero_explicit() function to clear the content of a
   * kvmalloc'ed object containing sensitive data to make sure that the
   * compiler won't optimize out the data clearing.
   */
  void kvfree_sensitive(const void *addr, size_t len)
  {
  	if (likely(!ZERO_OR_NULL_PTR(addr))) {
  		memzero_explicit((void *)addr, len);
  		kvfree(addr);
  	}
  }
  EXPORT_SYMBOL(kvfree_sensitive);
e39155ea1   Kirill A. Shutemov   mm: uninline and ...
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
  static inline void *__page_rmapping(struct page *page)
  {
  	unsigned long mapping;
  
  	mapping = (unsigned long)page->mapping;
  	mapping &= ~PAGE_MAPPING_FLAGS;
  
  	return (void *)mapping;
  }
  
  /* Neutral page->mapping pointer to address_space or anon_vma or other */
  void *page_rmapping(struct page *page)
  {
  	page = compound_head(page);
  	return __page_rmapping(page);
  }
1aa8aea53   Andrew Morton   mm: uninline page...
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
  /*
   * Return true if this page is mapped into pagetables.
   * For compound page it returns true if any subpage of compound page is mapped.
   */
  bool page_mapped(struct page *page)
  {
  	int i;
  
  	if (likely(!PageCompound(page)))
  		return atomic_read(&page->_mapcount) >= 0;
  	page = compound_head(page);
  	if (atomic_read(compound_mapcount_ptr(page)) >= 0)
  		return true;
  	if (PageHuge(page))
  		return false;
d8c6546b1   Matthew Wilcox (Oracle)   mm: introduce com...
625
  	for (i = 0; i < compound_nr(page); i++) {
1aa8aea53   Andrew Morton   mm: uninline page...
626
627
628
629
630
631
  		if (atomic_read(&page[i]._mapcount) >= 0)
  			return true;
  	}
  	return false;
  }
  EXPORT_SYMBOL(page_mapped);
e39155ea1   Kirill A. Shutemov   mm: uninline and ...
632
633
634
635
636
637
638
639
640
641
  struct anon_vma *page_anon_vma(struct page *page)
  {
  	unsigned long mapping;
  
  	page = compound_head(page);
  	mapping = (unsigned long)page->mapping;
  	if ((mapping & PAGE_MAPPING_FLAGS) != PAGE_MAPPING_ANON)
  		return NULL;
  	return __page_rmapping(page);
  }
9800339b5   Shaohua Li   mm: don't inline ...
642
643
  struct address_space *page_mapping(struct page *page)
  {
1c290f642   Kirill A. Shutemov   mm: sanitize page...
644
645
646
  	struct address_space *mapping;
  
  	page = compound_head(page);
9800339b5   Shaohua Li   mm: don't inline ...
647

03e5ac2fc   Mikulas Patocka   mm: fix crash whe...
648
649
650
  	/* This happens if someone calls flush_dcache_page on slab page */
  	if (unlikely(PageSlab(page)))
  		return NULL;
33806f06d   Shaohua Li   swap: make each s...
651
652
653
654
  	if (unlikely(PageSwapCache(page))) {
  		swp_entry_t entry;
  
  		entry.val = page_private(page);
e39155ea1   Kirill A. Shutemov   mm: uninline and ...
655
656
  		return swap_address_space(entry);
  	}
1c290f642   Kirill A. Shutemov   mm: sanitize page...
657
  	mapping = page->mapping;
bda807d44   Minchan Kim   mm: migrate: supp...
658
  	if ((unsigned long)mapping & PAGE_MAPPING_ANON)
e39155ea1   Kirill A. Shutemov   mm: uninline and ...
659
  		return NULL;
bda807d44   Minchan Kim   mm: migrate: supp...
660
661
  
  	return (void *)((unsigned long)mapping & ~PAGE_MAPPING_FLAGS);
9800339b5   Shaohua Li   mm: don't inline ...
662
  }
bda807d44   Minchan Kim   mm: migrate: supp...
663
  EXPORT_SYMBOL(page_mapping);
9800339b5   Shaohua Li   mm: don't inline ...
664

cb9f753a3   Huang Ying   mm: fix races bet...
665
666
667
668
669
670
671
672
673
  /*
   * For file cache pages, return the address_space, otherwise return NULL
   */
  struct address_space *page_mapping_file(struct page *page)
  {
  	if (unlikely(PageSwapCache(page)))
  		return NULL;
  	return page_mapping(page);
  }
b20ce5e03   Kirill A. Shutemov   mm: prepare page_...
674
675
676
677
678
679
  /* Slow path of page_mapcount() for compound pages */
  int __page_mapcount(struct page *page)
  {
  	int ret;
  
  	ret = atomic_read(&page->_mapcount) + 1;
dd78fedde   Kirill A. Shutemov   rmap: support fil...
680
681
682
683
684
685
  	/*
  	 * For file THP page->_mapcount contains total number of mapping
  	 * of the page: no need to look into compound_mapcount.
  	 */
  	if (!PageAnon(page) && !PageHuge(page))
  		return ret;
b20ce5e03   Kirill A. Shutemov   mm: prepare page_...
686
687
688
689
690
691
692
  	page = compound_head(page);
  	ret += atomic_read(compound_mapcount_ptr(page)) + 1;
  	if (PageDoubleMap(page))
  		ret--;
  	return ret;
  }
  EXPORT_SYMBOL_GPL(__page_mapcount);
39a1aa8e1   Andrey Ryabinin   mm: deduplicate m...
693
694
695
696
697
698
  int sysctl_overcommit_memory __read_mostly = OVERCOMMIT_GUESS;
  int sysctl_overcommit_ratio __read_mostly = 50;
  unsigned long sysctl_overcommit_kbytes __read_mostly;
  int sysctl_max_map_count __read_mostly = DEFAULT_MAX_MAP_COUNT;
  unsigned long sysctl_user_reserve_kbytes __read_mostly = 1UL << 17; /* 128MB */
  unsigned long sysctl_admin_reserve_kbytes __read_mostly = 1UL << 13; /* 8MB */
49f0ce5f9   Jerome Marchand   mm: add overcommi...
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
  int overcommit_ratio_handler(struct ctl_table *table, int write,
  			     void __user *buffer, size_t *lenp,
  			     loff_t *ppos)
  {
  	int ret;
  
  	ret = proc_dointvec(table, write, buffer, lenp, ppos);
  	if (ret == 0 && write)
  		sysctl_overcommit_kbytes = 0;
  	return ret;
  }
  
  int overcommit_kbytes_handler(struct ctl_table *table, int write,
  			     void __user *buffer, size_t *lenp,
  			     loff_t *ppos)
  {
  	int ret;
  
  	ret = proc_doulongvec_minmax(table, write, buffer, lenp, ppos);
  	if (ret == 0 && write)
  		sysctl_overcommit_ratio = 0;
  	return ret;
  }
00619bcc4   Jerome Marchand   mm: factor commit...
722
723
724
725
726
  /*
   * Committed memory limit enforced when OVERCOMMIT_NEVER policy is used
   */
  unsigned long vm_commit_limit(void)
  {
49f0ce5f9   Jerome Marchand   mm: add overcommi...
727
728
729
730
731
  	unsigned long allowed;
  
  	if (sysctl_overcommit_kbytes)
  		allowed = sysctl_overcommit_kbytes >> (PAGE_SHIFT - 10);
  	else
ca79b0c21   Arun KS   mm: convert total...
732
  		allowed = ((totalram_pages() - hugetlb_total_pages())
49f0ce5f9   Jerome Marchand   mm: add overcommi...
733
734
735
736
  			   * sysctl_overcommit_ratio / 100);
  	allowed += total_swap_pages;
  
  	return allowed;
00619bcc4   Jerome Marchand   mm: factor commit...
737
  }
39a1aa8e1   Andrey Ryabinin   mm: deduplicate m...
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
  /*
   * 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;
  
  /*
   * The global memory commitment made in the system can be a metric
   * that can be used to drive ballooning decisions when Linux is hosted
   * as a guest. On Hyper-V, the host implements a policy engine for dynamically
   * balancing memory across competing virtual machines that are hosted.
   * Several metrics drive this policy engine including the guest reported
   * memory commitment.
   */
  unsigned long vm_memory_committed(void)
  {
  	return percpu_counter_read_positive(&vm_committed_as);
  }
  EXPORT_SYMBOL_GPL(vm_memory_committed);
  
  /*
   * 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
ad56b738c   Mike Rapoport   docs/vm: rename d...
764
   * vm.overcommit_memory sysctl.  See Documentation/vm/overcommit-accounting.rst
39a1aa8e1   Andrey Ryabinin   mm: deduplicate m...
765
766
767
768
769
770
771
772
773
774
775
   *
   * 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.
   */
  int __vm_enough_memory(struct mm_struct *mm, long pages, int cap_sys_admin)
  {
8c7829b04   Johannes Weiner   mm: fix false-pos...
776
  	long allowed;
39a1aa8e1   Andrey Ryabinin   mm: deduplicate m...
777
778
779
780
781
782
783
784
785
786
787
788
789
790
  
  	VM_WARN_ONCE(percpu_counter_read(&vm_committed_as) <
  			-(s64)vm_committed_as_batch * num_online_cpus(),
  			"memory commitment underflow");
  
  	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) {
8c7829b04   Johannes Weiner   mm: fix false-pos...
791
  		if (pages > totalram_pages() + total_swap_pages)
39a1aa8e1   Andrey Ryabinin   mm: deduplicate m...
792
  			goto error;
8c7829b04   Johannes Weiner   mm: fix false-pos...
793
  		return 0;
39a1aa8e1   Andrey Ryabinin   mm: deduplicate m...
794
795
796
797
798
799
800
801
802
803
804
805
806
  	}
  
  	allowed = vm_commit_limit();
  	/*
  	 * Reserve some for root
  	 */
  	if (!cap_sys_admin)
  		allowed -= sysctl_admin_reserve_kbytes >> (PAGE_SHIFT - 10);
  
  	/*
  	 * Don't let a single process grow so big a user can't recover
  	 */
  	if (mm) {
8c7829b04   Johannes Weiner   mm: fix false-pos...
807
  		long reserve = sysctl_user_reserve_kbytes >> (PAGE_SHIFT - 10);
39a1aa8e1   Andrey Ryabinin   mm: deduplicate m...
808
809
810
811
812
813
814
815
816
817
  		allowed -= min_t(long, mm->total_vm / 32, reserve);
  	}
  
  	if (percpu_counter_read_positive(&vm_committed_as) < allowed)
  		return 0;
  error:
  	vm_unacct_memory(pages);
  
  	return -ENOMEM;
  }
a90902531   William Roberts   mm: Create utilit...
818
819
820
821
822
823
  /**
   * get_cmdline() - copy the cmdline value to a buffer.
   * @task:     the task whose cmdline value to copy.
   * @buffer:   the buffer to copy to.
   * @buflen:   the length of the buffer. Larger cmdline values are truncated
   *            to this length.
a862f68a8   Mike Rapoport   docs/core-api/mm:...
824
825
   *
   * Return: the size of the cmdline field copied. Note that the copy does
a90902531   William Roberts   mm: Create utilit...
826
827
828
829
830
831
832
   * not guarantee an ending NULL byte.
   */
  int get_cmdline(struct task_struct *task, char *buffer, int buflen)
  {
  	int res = 0;
  	unsigned int len;
  	struct mm_struct *mm = get_task_mm(task);
a3b609ef9   Mateusz Guzik   proc read mm's {a...
833
  	unsigned long arg_start, arg_end, env_start, env_end;
a90902531   William Roberts   mm: Create utilit...
834
835
836
837
  	if (!mm)
  		goto out;
  	if (!mm->arg_end)
  		goto out_mm;	/* Shh! No looking before we're done */
bc81426f5   Michal Koutný   prctl_set_mm: dow...
838
  	spin_lock(&mm->arg_lock);
a3b609ef9   Mateusz Guzik   proc read mm's {a...
839
840
841
842
  	arg_start = mm->arg_start;
  	arg_end = mm->arg_end;
  	env_start = mm->env_start;
  	env_end = mm->env_end;
bc81426f5   Michal Koutný   prctl_set_mm: dow...
843
  	spin_unlock(&mm->arg_lock);
a3b609ef9   Mateusz Guzik   proc read mm's {a...
844
845
  
  	len = arg_end - arg_start;
a90902531   William Roberts   mm: Create utilit...
846
847
848
  
  	if (len > buflen)
  		len = buflen;
f307ab6dc   Lorenzo Stoakes   mm: replace acces...
849
  	res = access_process_vm(task, arg_start, buffer, len, FOLL_FORCE);
a90902531   William Roberts   mm: Create utilit...
850
851
852
853
854
855
856
857
858
859
  
  	/*
  	 * If the nul at the end of args has been overwritten, then
  	 * assume application is using setproctitle(3).
  	 */
  	if (res > 0 && buffer[res-1] != '\0' && len < buflen) {
  		len = strnlen(buffer, res);
  		if (len < res) {
  			res = len;
  		} else {
a3b609ef9   Mateusz Guzik   proc read mm's {a...
860
  			len = env_end - env_start;
a90902531   William Roberts   mm: Create utilit...
861
862
  			if (len > buflen - res)
  				len = buflen - res;
a3b609ef9   Mateusz Guzik   proc read mm's {a...
863
  			res += access_process_vm(task, env_start,
f307ab6dc   Lorenzo Stoakes   mm: replace acces...
864
865
  						 buffer+res, len,
  						 FOLL_FORCE);
a90902531   William Roberts   mm: Create utilit...
866
867
868
869
870
871
872
873
  			res = strnlen(buffer, res);
  		}
  	}
  out_mm:
  	mmput(mm);
  out:
  	return res;
  }
010c164a5   Song Liu   mm: move memcmp_p...
874
875
876
877
878
879
880
881
882
883
884
885
886
  
  int memcmp_pages(struct page *page1, struct page *page2)
  {
  	char *addr1, *addr2;
  	int ret;
  
  	addr1 = kmap_atomic(page1);
  	addr2 = kmap_atomic(page2);
  	ret = memcmp(addr1, addr2, PAGE_SIZE);
  	kunmap_atomic(addr2);
  	kunmap_atomic(addr1);
  	return ret;
  }