Blame view
mm/mmap.c
88.5 KB
1da177e4c
|
1 2 3 4 5 |
/* * mm/mmap.c * * Written by obz. * |
046c68842
|
6 |
* Address space accounting code <alan@lxorguk.ukuu.org.uk> |
1da177e4c
|
7 |
*/ |
b1de0d139
|
8 |
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt |
e8420a8ec
|
9 |
#include <linux/kernel.h> |
1da177e4c
|
10 |
#include <linux/slab.h> |
4af3c9cc4
|
11 |
#include <linux/backing-dev.h> |
1da177e4c
|
12 |
#include <linux/mm.h> |
615d6e875
|
13 |
#include <linux/vmacache.h> |
1da177e4c
|
14 15 16 17 18 |
#include <linux/shm.h> #include <linux/mman.h> #include <linux/pagemap.h> #include <linux/swap.h> #include <linux/syscalls.h> |
c59ede7b7
|
19 |
#include <linux/capability.h> |
1da177e4c
|
20 21 22 23 24 25 26 |
#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> |
b95f1b31b
|
27 |
#include <linux/export.h> |
1da177e4c
|
28 29 30 |
#include <linux/mount.h> #include <linux/mempolicy.h> #include <linux/rmap.h> |
cddb8a5c1
|
31 |
#include <linux/mmu_notifier.h> |
82f71ae4a
|
32 |
#include <linux/mmdebug.h> |
cdd6c482c
|
33 |
#include <linux/perf_event.h> |
120a795da
|
34 |
#include <linux/audit.h> |
b15d00b6a
|
35 |
#include <linux/khugepaged.h> |
2b1444983
|
36 |
#include <linux/uprobes.h> |
d37371870
|
37 |
#include <linux/rbtree_augmented.h> |
1640879af
|
38 39 |
#include <linux/notifier.h> #include <linux/memory.h> |
b1de0d139
|
40 |
#include <linux/printk.h> |
19a809afe
|
41 |
#include <linux/userfaultfd_k.h> |
d977d56ce
|
42 |
#include <linux/moduleparam.h> |
62b5f7d01
|
43 |
#include <linux/pkeys.h> |
1da177e4c
|
44 45 46 47 |
#include <asm/uaccess.h> #include <asm/cacheflush.h> #include <asm/tlb.h> |
d6dd61c83
|
48 |
#include <asm/mmu_context.h> |
1da177e4c
|
49 |
|
42b777281
|
50 |
#include "internal.h" |
3a4597568
|
51 52 53 |
#ifndef arch_mmap_check #define arch_mmap_check(addr, len, flags) (0) #endif |
08e7d9b55
|
54 55 56 |
#ifndef arch_rebalance_pgtables #define arch_rebalance_pgtables(addr, len) (addr) #endif |
d07e22597
|
57 58 59 60 61 62 63 64 65 66 |
#ifdef CONFIG_HAVE_ARCH_MMAP_RND_BITS const int mmap_rnd_bits_min = CONFIG_ARCH_MMAP_RND_BITS_MIN; const int mmap_rnd_bits_max = CONFIG_ARCH_MMAP_RND_BITS_MAX; int mmap_rnd_bits __read_mostly = CONFIG_ARCH_MMAP_RND_BITS; #endif #ifdef CONFIG_HAVE_ARCH_MMAP_RND_COMPAT_BITS const int mmap_rnd_compat_bits_min = CONFIG_ARCH_MMAP_RND_COMPAT_BITS_MIN; const int mmap_rnd_compat_bits_max = CONFIG_ARCH_MMAP_RND_COMPAT_BITS_MAX; int mmap_rnd_compat_bits __read_mostly = CONFIG_ARCH_MMAP_RND_COMPAT_BITS; #endif |
d977d56ce
|
67 68 |
static bool ignore_rlimit_data = true; core_param(ignore_rlimit_data, ignore_rlimit_data, bool, 0644); |
d07e22597
|
69 |
|
e0da382c9
|
70 71 72 |
static void unmap_region(struct mm_struct *mm, struct vm_area_struct *vma, struct vm_area_struct *prev, unsigned long start, unsigned long end); |
1da177e4c
|
73 74 75 76 77 78 79 80 81 |
/* 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 |
cc71aba34
|
82 |
* |
1da177e4c
|
83 84 85 86 87 88 89 90 91 |
* 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 }; |
804af2cf6
|
92 93 |
pgprot_t vm_get_page_prot(unsigned long vm_flags) { |
b845f313d
|
94 95 96 |
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))); |
804af2cf6
|
97 98 |
} EXPORT_SYMBOL(vm_get_page_prot); |
64e455079
|
99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 |
static pgprot_t vm_pgprot_modify(pgprot_t oldprot, unsigned long vm_flags) { return pgprot_modify(oldprot, vm_get_page_prot(vm_flags)); } /* Update vma->vm_page_prot to reflect vma->vm_flags. */ void vma_set_page_prot(struct vm_area_struct *vma) { unsigned long vm_flags = vma->vm_flags; vma->vm_page_prot = vm_pgprot_modify(vma->vm_page_prot, vm_flags); if (vma_wants_writenotify(vma)) { vm_flags &= ~VM_SHARED; vma->vm_page_prot = vm_pgprot_modify(vma->vm_page_prot, vm_flags); } } |
1da177e4c
|
116 |
/* |
c8c06efa8
|
117 |
* Requires inode->i_mapping->i_mmap_rwsem |
1da177e4c
|
118 119 120 121 122 |
*/ static void __remove_shared_vm_struct(struct vm_area_struct *vma, struct file *file, struct address_space *mapping) { if (vma->vm_flags & VM_DENYWRITE) |
496ad9aa8
|
123 |
atomic_inc(&file_inode(file)->i_writecount); |
1da177e4c
|
124 |
if (vma->vm_flags & VM_SHARED) |
4bb5f5d93
|
125 |
mapping_unmap_writable(mapping); |
1da177e4c
|
126 127 |
flush_dcache_mmap_lock(mapping); |
27ba0644e
|
128 |
vma_interval_tree_remove(vma, &mapping->i_mmap); |
1da177e4c
|
129 130 131 132 |
flush_dcache_mmap_unlock(mapping); } /* |
6b2dbba8b
|
133 |
* Unlink a file-based vm structure from its interval tree, to hide |
a8fb5618d
|
134 |
* vma from rmap and vmtruncate before freeing its page tables. |
1da177e4c
|
135 |
*/ |
a8fb5618d
|
136 |
void unlink_file_vma(struct vm_area_struct *vma) |
1da177e4c
|
137 138 |
{ struct file *file = vma->vm_file; |
1da177e4c
|
139 140 |
if (file) { struct address_space *mapping = file->f_mapping; |
83cde9e8b
|
141 |
i_mmap_lock_write(mapping); |
1da177e4c
|
142 |
__remove_shared_vm_struct(vma, file, mapping); |
83cde9e8b
|
143 |
i_mmap_unlock_write(mapping); |
1da177e4c
|
144 |
} |
a8fb5618d
|
145 146 147 148 149 150 151 152 |
} /* * 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; |
a8fb5618d
|
153 |
might_sleep(); |
1da177e4c
|
154 155 |
if (vma->vm_ops && vma->vm_ops->close) vma->vm_ops->close(vma); |
e9714acf8
|
156 |
if (vma->vm_file) |
a8fb5618d
|
157 |
fput(vma->vm_file); |
f0be3d32b
|
158 |
mpol_put(vma_policy(vma)); |
1da177e4c
|
159 |
kmem_cache_free(vm_area_cachep, vma); |
a8fb5618d
|
160 |
return next; |
1da177e4c
|
161 |
} |
e4eb1ff61
|
162 |
static unsigned long do_brk(unsigned long addr, unsigned long len); |
6a6160a7b
|
163 |
SYSCALL_DEFINE1(brk, unsigned long, brk) |
1da177e4c
|
164 |
{ |
8764b338b
|
165 |
unsigned long retval; |
1da177e4c
|
166 167 |
unsigned long newbrk, oldbrk; struct mm_struct *mm = current->mm; |
a5b4592cf
|
168 |
unsigned long min_brk; |
128557ffe
|
169 |
bool populate; |
1da177e4c
|
170 171 |
down_write(&mm->mmap_sem); |
a5b4592cf
|
172 |
#ifdef CONFIG_COMPAT_BRK |
5520e8948
|
173 174 175 176 177 |
/* * 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 */ |
4471a675d
|
178 |
if (current->brk_randomized) |
5520e8948
|
179 180 181 |
min_brk = mm->start_brk; else min_brk = mm->end_data; |
a5b4592cf
|
182 183 184 185 |
#else min_brk = mm->start_brk; #endif if (brk < min_brk) |
1da177e4c
|
186 |
goto out; |
1e624196f
|
187 188 189 190 191 192 193 |
/* * 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 */ |
8764b338b
|
194 195 |
if (check_data_rlimit(rlimit(RLIMIT_DATA), brk, mm->start_brk, mm->end_data, mm->start_data)) |
1e624196f
|
196 |
goto out; |
1da177e4c
|
197 198 199 200 201 202 203 204 205 206 207 |
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; } |
1da177e4c
|
208 209 210 211 212 213 214 |
/* 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; |
128557ffe
|
215 |
|
1da177e4c
|
216 217 |
set_brk: mm->brk = brk; |
128557ffe
|
218 219 220 221 222 |
populate = newbrk > oldbrk && (mm->def_flags & VM_LOCKED) != 0; up_write(&mm->mmap_sem); if (populate) mm_populate(oldbrk, newbrk - oldbrk); return brk; |
1da177e4c
|
223 224 225 226 227 |
out: retval = mm->brk; up_write(&mm->mmap_sem); return retval; } |
d37371870
|
228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 |
static long vma_compute_subtree_gap(struct vm_area_struct *vma) { unsigned long max, subtree_gap; max = vma->vm_start; if (vma->vm_prev) max -= vma->vm_prev->vm_end; if (vma->vm_rb.rb_left) { subtree_gap = rb_entry(vma->vm_rb.rb_left, struct vm_area_struct, vm_rb)->rb_subtree_gap; if (subtree_gap > max) max = subtree_gap; } if (vma->vm_rb.rb_right) { subtree_gap = rb_entry(vma->vm_rb.rb_right, struct vm_area_struct, vm_rb)->rb_subtree_gap; if (subtree_gap > max) max = subtree_gap; } return max; } |
ed8ea8150
|
248 |
#ifdef CONFIG_DEBUG_VM_RB |
acf128d04
|
249 |
static int browse_rb(struct mm_struct *mm) |
1da177e4c
|
250 |
{ |
acf128d04
|
251 |
struct rb_root *root = &mm->mm_rb; |
5a0768f64
|
252 |
int i = 0, j, bug = 0; |
1da177e4c
|
253 254 255 256 257 258 |
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); |
5a0768f64
|
259 |
if (vma->vm_start < prev) { |
ff26f70f4
|
260 261 262 |
pr_emerg("vm_start %lx < prev %lx ", vma->vm_start, prev); |
5a0768f64
|
263 264 265 |
bug = 1; } if (vma->vm_start < pend) { |
ff26f70f4
|
266 267 268 |
pr_emerg("vm_start %lx < pend %lx ", vma->vm_start, pend); |
5a0768f64
|
269 270 271 |
bug = 1; } if (vma->vm_start > vma->vm_end) { |
ff26f70f4
|
272 273 274 |
pr_emerg("vm_start %lx > vm_end %lx ", vma->vm_start, vma->vm_end); |
5a0768f64
|
275 276 |
bug = 1; } |
acf128d04
|
277 |
spin_lock(&mm->page_table_lock); |
5a0768f64
|
278 |
if (vma->rb_subtree_gap != vma_compute_subtree_gap(vma)) { |
8542bdfc6
|
279 280 |
pr_emerg("free gap %lx, correct %lx ", |
5a0768f64
|
281 282 283 284 |
vma->rb_subtree_gap, vma_compute_subtree_gap(vma)); bug = 1; } |
acf128d04
|
285 |
spin_unlock(&mm->page_table_lock); |
1da177e4c
|
286 287 |
i++; pn = nd; |
d1af65d13
|
288 289 |
prev = vma->vm_start; pend = vma->vm_end; |
1da177e4c
|
290 291 |
} j = 0; |
5a0768f64
|
292 |
for (nd = pn; nd; nd = rb_prev(nd)) |
1da177e4c
|
293 |
j++; |
5a0768f64
|
294 |
if (i != j) { |
8542bdfc6
|
295 296 |
pr_emerg("backwards %d, forwards %d ", j, i); |
5a0768f64
|
297 |
bug = 1; |
1da177e4c
|
298 |
} |
5a0768f64
|
299 |
return bug ? -1 : i; |
1da177e4c
|
300 |
} |
d37371870
|
301 302 303 304 305 306 307 |
static void validate_mm_rb(struct rb_root *root, struct vm_area_struct *ignore) { struct rb_node *nd; 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); |
96dad67ff
|
308 309 310 |
VM_BUG_ON_VMA(vma != ignore && vma->rb_subtree_gap != vma_compute_subtree_gap(vma), vma); |
1da177e4c
|
311 |
} |
1da177e4c
|
312 |
} |
eafd4dc4d
|
313 |
static void validate_mm(struct mm_struct *mm) |
1da177e4c
|
314 315 316 |
{ int bug = 0; int i = 0; |
5a0768f64
|
317 |
unsigned long highest_address = 0; |
ed8ea8150
|
318 |
struct vm_area_struct *vma = mm->mmap; |
ff26f70f4
|
319 |
|
ed8ea8150
|
320 |
while (vma) { |
12352d3ca
|
321 |
struct anon_vma *anon_vma = vma->anon_vma; |
ed8ea8150
|
322 |
struct anon_vma_chain *avc; |
ff26f70f4
|
323 |
|
12352d3ca
|
324 325 326 327 328 329 |
if (anon_vma) { anon_vma_lock_read(anon_vma); list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) anon_vma_interval_tree_verify(avc); anon_vma_unlock_read(anon_vma); } |
5a0768f64
|
330 |
highest_address = vma->vm_end; |
ed8ea8150
|
331 |
vma = vma->vm_next; |
1da177e4c
|
332 333 |
i++; } |
5a0768f64
|
334 |
if (i != mm->map_count) { |
8542bdfc6
|
335 336 |
pr_emerg("map_count %d vm_next %d ", mm->map_count, i); |
5a0768f64
|
337 338 339 |
bug = 1; } if (highest_address != mm->highest_vm_end) { |
8542bdfc6
|
340 341 |
pr_emerg("mm->highest_vm_end %lx, found %lx ", |
ff26f70f4
|
342 |
mm->highest_vm_end, highest_address); |
5a0768f64
|
343 344 |
bug = 1; } |
acf128d04
|
345 |
i = browse_rb(mm); |
5a0768f64
|
346 |
if (i != mm->map_count) { |
ff26f70f4
|
347 348 349 |
if (i != -1) pr_emerg("map_count %d rb %d ", mm->map_count, i); |
5a0768f64
|
350 351 |
bug = 1; } |
96dad67ff
|
352 |
VM_BUG_ON_MM(bug, mm); |
1da177e4c
|
353 354 |
} #else |
d37371870
|
355 |
#define validate_mm_rb(root, ignore) do { } while (0) |
1da177e4c
|
356 357 |
#define validate_mm(mm) do { } while (0) #endif |
d37371870
|
358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 |
RB_DECLARE_CALLBACKS(static, vma_gap_callbacks, struct vm_area_struct, vm_rb, unsigned long, rb_subtree_gap, vma_compute_subtree_gap) /* * Update augmented rbtree rb_subtree_gap values after vma->vm_start or * vma->vm_prev->vm_end values changed, without modifying the vma's position * in the rbtree. */ static void vma_gap_update(struct vm_area_struct *vma) { /* * As it turns out, RB_DECLARE_CALLBACKS() already created a callback * function that does exacltly what we want. */ vma_gap_callbacks_propagate(&vma->vm_rb, NULL); } static inline void vma_rb_insert(struct vm_area_struct *vma, struct rb_root *root) { /* All rb_subtree_gap values must be consistent prior to insertion */ validate_mm_rb(root, NULL); rb_insert_augmented(&vma->vm_rb, root, &vma_gap_callbacks); } static void vma_rb_erase(struct vm_area_struct *vma, struct rb_root *root) { /* * All rb_subtree_gap values must be consistent prior to erase, * with the possible exception of the vma being erased. */ validate_mm_rb(root, vma); /* * Note rb_erase_augmented is a fairly large inline function, * so make sure we instantiate it only once with our desired * augmented rbtree callbacks. */ rb_erase_augmented(&vma->vm_rb, root, &vma_gap_callbacks); } |
bf181b9f9
|
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 |
/* * vma has some anon_vma assigned, and is already inserted on that * anon_vma's interval trees. * * Before updating the vma's vm_start / vm_end / vm_pgoff fields, the * vma must be removed from the anon_vma's interval trees using * anon_vma_interval_tree_pre_update_vma(). * * After the update, the vma will be reinserted using * anon_vma_interval_tree_post_update_vma(). * * The entire update must be protected by exclusive mmap_sem and by * the root anon_vma's mutex. */ static inline void anon_vma_interval_tree_pre_update_vma(struct vm_area_struct *vma) { struct anon_vma_chain *avc; list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) anon_vma_interval_tree_remove(avc, &avc->anon_vma->rb_root); } static inline void anon_vma_interval_tree_post_update_vma(struct vm_area_struct *vma) { struct anon_vma_chain *avc; list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) anon_vma_interval_tree_insert(avc, &avc->anon_vma->rb_root); } |
6597d7833
|
430 431 432 |
static int find_vma_links(struct mm_struct *mm, unsigned long addr, unsigned long end, struct vm_area_struct **pprev, struct rb_node ***rb_link, struct rb_node **rb_parent) |
1da177e4c
|
433 |
{ |
6597d7833
|
434 |
struct rb_node **__rb_link, *__rb_parent, *rb_prev; |
1da177e4c
|
435 436 437 |
__rb_link = &mm->mm_rb.rb_node; rb_prev = __rb_parent = NULL; |
1da177e4c
|
438 439 440 441 442 443 444 445 |
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) { |
6597d7833
|
446 447 448 |
/* Fail if an existing vma overlaps the area */ if (vma_tmp->vm_start < end) return -ENOMEM; |
1da177e4c
|
449 450 451 452 453 454 455 456 457 458 459 460 |
__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; |
6597d7833
|
461 |
return 0; |
1da177e4c
|
462 |
} |
e8420a8ec
|
463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 |
static unsigned long count_vma_pages_range(struct mm_struct *mm, unsigned long addr, unsigned long end) { unsigned long nr_pages = 0; struct vm_area_struct *vma; /* Find first overlaping mapping */ vma = find_vma_intersection(mm, addr, end); if (!vma) return 0; nr_pages = (min(end, vma->vm_end) - max(addr, vma->vm_start)) >> PAGE_SHIFT; /* Iterate over the rest of the overlaps */ for (vma = vma->vm_next; vma; vma = vma->vm_next) { unsigned long overlap_len; if (vma->vm_start > end) break; overlap_len = min(end, vma->vm_end) - vma->vm_start; nr_pages += overlap_len >> PAGE_SHIFT; } return nr_pages; } |
1da177e4c
|
490 491 492 |
void __vma_link_rb(struct mm_struct *mm, struct vm_area_struct *vma, struct rb_node **rb_link, struct rb_node *rb_parent) { |
d37371870
|
493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 |
/* Update tracking information for the gap following the new vma. */ if (vma->vm_next) vma_gap_update(vma->vm_next); else mm->highest_vm_end = vma->vm_end; /* * vma->vm_prev wasn't known when we followed the rbtree to find the * correct insertion point for that vma. As a result, we could not * update the vma vm_rb parents rb_subtree_gap values on the way down. * So, we first insert the vma with a zero rb_subtree_gap value * (to be consistent with what we did on the way down), and then * immediately update the gap to the correct value. Finally we * rebalance the rbtree after all augmented values have been set. */ |
1da177e4c
|
508 |
rb_link_node(&vma->vm_rb, rb_parent, rb_link); |
d37371870
|
509 510 511 |
vma->rb_subtree_gap = 0; vma_gap_update(vma); vma_rb_insert(vma, &mm->mm_rb); |
1da177e4c
|
512 |
} |
cb8f488c3
|
513 |
static void __vma_link_file(struct vm_area_struct *vma) |
1da177e4c
|
514 |
{ |
48aae4255
|
515 |
struct file *file; |
1da177e4c
|
516 517 518 519 520 521 |
file = vma->vm_file; if (file) { struct address_space *mapping = file->f_mapping; if (vma->vm_flags & VM_DENYWRITE) |
496ad9aa8
|
522 |
atomic_dec(&file_inode(file)->i_writecount); |
1da177e4c
|
523 |
if (vma->vm_flags & VM_SHARED) |
4bb5f5d93
|
524 |
atomic_inc(&mapping->i_mmap_writable); |
1da177e4c
|
525 526 |
flush_dcache_mmap_lock(mapping); |
27ba0644e
|
527 |
vma_interval_tree_insert(vma, &mapping->i_mmap); |
1da177e4c
|
528 529 530 531 532 533 534 535 536 537 538 |
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); |
1da177e4c
|
539 540 541 542 543 544 545 |
} 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; |
64ac4940d
|
546 |
if (vma->vm_file) { |
1da177e4c
|
547 |
mapping = vma->vm_file->f_mapping; |
83cde9e8b
|
548 |
i_mmap_lock_write(mapping); |
64ac4940d
|
549 |
} |
1da177e4c
|
550 551 552 |
__vma_link(mm, vma, prev, rb_link, rb_parent); __vma_link_file(vma); |
1da177e4c
|
553 |
if (mapping) |
83cde9e8b
|
554 |
i_mmap_unlock_write(mapping); |
1da177e4c
|
555 556 557 558 559 560 |
mm->map_count++; validate_mm(mm); } /* |
88f6b4c32
|
561 |
* Helper for vma_adjust() in the split_vma insert case: insert a vma into the |
6b2dbba8b
|
562 |
* mm's list and rbtree. It has already been inserted into the interval tree. |
1da177e4c
|
563 |
*/ |
48aae4255
|
564 |
static void __insert_vm_struct(struct mm_struct *mm, struct vm_area_struct *vma) |
1da177e4c
|
565 |
{ |
6597d7833
|
566 |
struct vm_area_struct *prev; |
48aae4255
|
567 |
struct rb_node **rb_link, *rb_parent; |
1da177e4c
|
568 |
|
6597d7833
|
569 570 571 |
if (find_vma_links(mm, vma->vm_start, vma->vm_end, &prev, &rb_link, &rb_parent)) BUG(); |
1da177e4c
|
572 573 574 575 576 577 578 579 |
__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) { |
d37371870
|
580 |
struct vm_area_struct *next; |
297c5eee3
|
581 |
|
d37371870
|
582 583 |
vma_rb_erase(vma, &mm->mm_rb); prev->vm_next = next = vma->vm_next; |
297c5eee3
|
584 585 |
if (next) next->vm_prev = prev; |
615d6e875
|
586 587 588 |
/* Kill the cache */ vmacache_invalidate(mm); |
1da177e4c
|
589 590 591 592 593 594 595 596 597 |
} /* * 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. */ |
5beb49305
|
598 |
int vma_adjust(struct vm_area_struct *vma, unsigned long start, |
1da177e4c
|
599 600 601 602 603 604 |
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; |
6b2dbba8b
|
605 |
struct rb_root *root = NULL; |
012f18004
|
606 |
struct anon_vma *anon_vma = NULL; |
1da177e4c
|
607 |
struct file *file = vma->vm_file; |
d37371870
|
608 |
bool start_changed = false, end_changed = false; |
1da177e4c
|
609 610 611 612 |
long adjust_next = 0; int remove_next = 0; if (next && !insert) { |
287d97ac0
|
613 |
struct vm_area_struct *exporter = NULL; |
1da177e4c
|
614 615 616 617 618 619 620 |
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; |
287d97ac0
|
621 |
exporter = next; |
1da177e4c
|
622 623 624 625 626 627 628 |
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; |
287d97ac0
|
629 |
exporter = next; |
1da177e4c
|
630 631 632 633 634 635 636 |
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. */ |
cc71aba34
|
637 |
adjust_next = -((vma->vm_end - end) >> PAGE_SHIFT); |
287d97ac0
|
638 |
exporter = vma; |
1da177e4c
|
639 640 |
importer = next; } |
1da177e4c
|
641 |
|
5beb49305
|
642 643 644 645 646 |
/* * 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. */ |
287d97ac0
|
647 |
if (exporter && exporter->anon_vma && !importer->anon_vma) { |
c4ea95d7c
|
648 |
int error; |
b800c91a0
|
649 |
importer->anon_vma = exporter->anon_vma; |
c4ea95d7c
|
650 |
error = anon_vma_clone(importer, exporter); |
3fe89b3e2
|
651 |
if (error) |
c4ea95d7c
|
652 |
return error; |
5beb49305
|
653 654 |
} } |
1da177e4c
|
655 656 |
if (file) { mapping = file->f_mapping; |
27ba0644e
|
657 658 |
root = &mapping->i_mmap; uprobe_munmap(vma, vma->vm_start, vma->vm_end); |
682968e0c
|
659 |
|
27ba0644e
|
660 661 |
if (adjust_next) uprobe_munmap(next, next->vm_start, next->vm_end); |
682968e0c
|
662 |
|
83cde9e8b
|
663 |
i_mmap_lock_write(mapping); |
1da177e4c
|
664 |
if (insert) { |
1da177e4c
|
665 |
/* |
6b2dbba8b
|
666 |
* Put into interval tree now, so instantiated pages |
1da177e4c
|
667 668 669 670 671 672 673 |
* 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); } } |
94fcc585f
|
674 |
vma_adjust_trans_huge(vma, start, end, adjust_next); |
bf181b9f9
|
675 676 677 678 |
anon_vma = vma->anon_vma; if (!anon_vma && adjust_next) anon_vma = next->anon_vma; if (anon_vma) { |
81d1b09c6
|
679 680 |
VM_BUG_ON_VMA(adjust_next && next->anon_vma && anon_vma != next->anon_vma, next); |
4fc3f1d66
|
681 |
anon_vma_lock_write(anon_vma); |
bf181b9f9
|
682 683 684 685 |
anon_vma_interval_tree_pre_update_vma(vma); if (adjust_next) anon_vma_interval_tree_pre_update_vma(next); } |
012f18004
|
686 |
|
1da177e4c
|
687 688 |
if (root) { flush_dcache_mmap_lock(mapping); |
6b2dbba8b
|
689 |
vma_interval_tree_remove(vma, root); |
1da177e4c
|
690 |
if (adjust_next) |
6b2dbba8b
|
691 |
vma_interval_tree_remove(next, root); |
1da177e4c
|
692 |
} |
d37371870
|
693 694 695 696 697 698 699 700 |
if (start != vma->vm_start) { vma->vm_start = start; start_changed = true; } if (end != vma->vm_end) { vma->vm_end = end; end_changed = true; } |
1da177e4c
|
701 702 703 704 705 706 707 708 |
vma->vm_pgoff = pgoff; if (adjust_next) { next->vm_start += adjust_next << PAGE_SHIFT; next->vm_pgoff += adjust_next; } if (root) { if (adjust_next) |
6b2dbba8b
|
709 710 |
vma_interval_tree_insert(next, root); vma_interval_tree_insert(vma, root); |
1da177e4c
|
711 712 713 714 715 716 717 718 719 720 721 |
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); |
1da177e4c
|
722 723 724 725 726 727 728 |
} 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); |
d37371870
|
729 730 731 732 733 734 735 736 737 |
} else { if (start_changed) vma_gap_update(vma); if (end_changed) { if (!next) mm->highest_vm_end = end; else if (!adjust_next) vma_gap_update(next); } |
1da177e4c
|
738 |
} |
bf181b9f9
|
739 740 741 742 |
if (anon_vma) { anon_vma_interval_tree_post_update_vma(vma); if (adjust_next) anon_vma_interval_tree_post_update_vma(next); |
08b52706d
|
743 |
anon_vma_unlock_write(anon_vma); |
bf181b9f9
|
744 |
} |
1da177e4c
|
745 |
if (mapping) |
83cde9e8b
|
746 |
i_mmap_unlock_write(mapping); |
1da177e4c
|
747 |
|
2b1444983
|
748 |
if (root) { |
7b2d81d48
|
749 |
uprobe_mmap(vma); |
2b1444983
|
750 751 |
if (adjust_next) |
7b2d81d48
|
752 |
uprobe_mmap(next); |
2b1444983
|
753 |
} |
1da177e4c
|
754 |
if (remove_next) { |
925d1c401
|
755 |
if (file) { |
cbc91f71b
|
756 |
uprobe_munmap(next, next->vm_start, next->vm_end); |
1da177e4c
|
757 |
fput(file); |
925d1c401
|
758 |
} |
5beb49305
|
759 760 |
if (next->anon_vma) anon_vma_merge(vma, next); |
1da177e4c
|
761 |
mm->map_count--; |
3964acd0d
|
762 |
mpol_put(vma_policy(next)); |
1da177e4c
|
763 764 765 766 767 768 |
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. */ |
d37371870
|
769 770 |
next = vma->vm_next; if (remove_next == 2) |
1da177e4c
|
771 |
goto again; |
d37371870
|
772 773 774 775 |
else if (next) vma_gap_update(next); else mm->highest_vm_end = end; |
1da177e4c
|
776 |
} |
2b1444983
|
777 |
if (insert && file) |
7b2d81d48
|
778 |
uprobe_mmap(insert); |
1da177e4c
|
779 780 |
validate_mm(mm); |
5beb49305
|
781 782 |
return 0; |
1da177e4c
|
783 784 785 786 787 788 |
} /* * 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. */ |
1da177e4c
|
789 |
static inline int is_mergeable_vma(struct vm_area_struct *vma, |
19a809afe
|
790 791 |
struct file *file, unsigned long vm_flags, struct vm_userfaultfd_ctx vm_userfaultfd_ctx) |
1da177e4c
|
792 |
{ |
34228d473
|
793 794 795 796 797 798 799 800 801 |
/* * VM_SOFTDIRTY should not prevent from VMA merging, if we * match the flags but dirty bit -- the caller should mark * merged VMA as dirty. If dirty bit won't be excluded from * comparison, we increase pressue on the memory system forcing * the kernel to generate new VMAs when old one could be * extended instead. */ if ((vma->vm_flags ^ vm_flags) & ~VM_SOFTDIRTY) |
1da177e4c
|
802 803 804 805 806 |
return 0; if (vma->vm_file != file) return 0; if (vma->vm_ops && vma->vm_ops->close) return 0; |
19a809afe
|
807 808 |
if (!is_mergeable_vm_userfaultfd_ctx(vma, vm_userfaultfd_ctx)) return 0; |
1da177e4c
|
809 810 811 812 |
return 1; } static inline int is_mergeable_anon_vma(struct anon_vma *anon_vma1, |
965f55dea
|
813 814 |
struct anon_vma *anon_vma2, struct vm_area_struct *vma) |
1da177e4c
|
815 |
{ |
965f55dea
|
816 817 818 819 820 821 822 823 |
/* * 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; |
1da177e4c
|
824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 |
} /* * 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, |
19a809afe
|
839 840 841 |
struct anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff, struct vm_userfaultfd_ctx vm_userfaultfd_ctx) |
1da177e4c
|
842 |
{ |
19a809afe
|
843 |
if (is_mergeable_vma(vma, file, vm_flags, vm_userfaultfd_ctx) && |
965f55dea
|
844 |
is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) { |
1da177e4c
|
845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 |
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, |
19a809afe
|
860 861 862 |
struct anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff, struct vm_userfaultfd_ctx vm_userfaultfd_ctx) |
1da177e4c
|
863 |
{ |
19a809afe
|
864 |
if (is_mergeable_vma(vma, file, vm_flags, vm_userfaultfd_ctx) && |
965f55dea
|
865 |
is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) { |
1da177e4c
|
866 |
pgoff_t vm_pglen; |
d6e932177
|
867 |
vm_pglen = vma_pages(vma); |
1da177e4c
|
868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 |
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, |
cc71aba34
|
906 |
struct anon_vma *anon_vma, struct file *file, |
19a809afe
|
907 908 |
pgoff_t pgoff, struct mempolicy *policy, struct vm_userfaultfd_ctx vm_userfaultfd_ctx) |
1da177e4c
|
909 910 911 |
{ pgoff_t pglen = (end - addr) >> PAGE_SHIFT; struct vm_area_struct *area, *next; |
5beb49305
|
912 |
int err; |
1da177e4c
|
913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 |
/* * 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 && |
cc71aba34
|
933 |
mpol_equal(vma_policy(prev), policy) && |
1da177e4c
|
934 |
can_vma_merge_after(prev, vm_flags, |
19a809afe
|
935 936 |
anon_vma, file, pgoff, vm_userfaultfd_ctx)) { |
1da177e4c
|
937 938 939 940 941 942 |
/* * 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, |
19a809afe
|
943 944 945 |
anon_vma, file, pgoff+pglen, vm_userfaultfd_ctx) && |
1da177e4c
|
946 |
is_mergeable_anon_vma(prev->anon_vma, |
965f55dea
|
947 |
next->anon_vma, NULL)) { |
1da177e4c
|
948 |
/* cases 1, 6 */ |
5beb49305
|
949 |
err = vma_adjust(prev, prev->vm_start, |
1da177e4c
|
950 951 |
next->vm_end, prev->vm_pgoff, NULL); } else /* cases 2, 5, 7 */ |
5beb49305
|
952 |
err = vma_adjust(prev, prev->vm_start, |
1da177e4c
|
953 |
end, prev->vm_pgoff, NULL); |
5beb49305
|
954 955 |
if (err) return NULL; |
6d50e60cd
|
956 |
khugepaged_enter_vma_merge(prev, vm_flags); |
1da177e4c
|
957 958 959 960 961 962 963 |
return prev; } /* * Can this new request be merged in front of next? */ if (next && end == next->vm_start && |
cc71aba34
|
964 |
mpol_equal(policy, vma_policy(next)) && |
1da177e4c
|
965 |
can_vma_merge_before(next, vm_flags, |
19a809afe
|
966 967 |
anon_vma, file, pgoff+pglen, vm_userfaultfd_ctx)) { |
1da177e4c
|
968 |
if (prev && addr < prev->vm_end) /* case 4 */ |
5beb49305
|
969 |
err = vma_adjust(prev, prev->vm_start, |
1da177e4c
|
970 971 |
addr, prev->vm_pgoff, NULL); else /* cases 3, 8 */ |
5beb49305
|
972 |
err = vma_adjust(area, addr, next->vm_end, |
1da177e4c
|
973 |
next->vm_pgoff - pglen, NULL); |
5beb49305
|
974 975 |
if (err) return NULL; |
6d50e60cd
|
976 |
khugepaged_enter_vma_merge(area, vm_flags); |
1da177e4c
|
977 978 979 980 981 982 983 |
return area; } return NULL; } /* |
d0e9fe175
|
984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 |
* 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 && |
34228d473
|
1001 |
!((a->vm_flags ^ b->vm_flags) & ~(VM_READ|VM_WRITE|VM_EXEC|VM_SOFTDIRTY)) && |
d0e9fe175
|
1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 |
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 |
4db0c3c29
|
1016 |
* we do that READ_ONCE() to make sure that we never re-load the pointer. |
d0e9fe175
|
1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 |
* * 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)) { |
4db0c3c29
|
1030 |
struct anon_vma *anon_vma = READ_ONCE(old->anon_vma); |
d0e9fe175
|
1031 1032 1033 1034 1035 1036 1037 1038 |
if (anon_vma && list_is_singular(&old->anon_vma_chain)) return anon_vma; } return NULL; } /* |
1da177e4c
|
1039 1040 1041 1042 1043 1044 1045 1046 1047 |
* 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) { |
d0e9fe175
|
1048 |
struct anon_vma *anon_vma; |
1da177e4c
|
1049 |
struct vm_area_struct *near; |
1da177e4c
|
1050 1051 1052 1053 |
near = vma->vm_next; if (!near) goto try_prev; |
d0e9fe175
|
1054 1055 1056 |
anon_vma = reusable_anon_vma(near, vma, near); if (anon_vma) return anon_vma; |
1da177e4c
|
1057 |
try_prev: |
9be34c9d5
|
1058 |
near = vma->vm_prev; |
1da177e4c
|
1059 1060 |
if (!near) goto none; |
d0e9fe175
|
1061 1062 1063 |
anon_vma = reusable_anon_vma(near, near, vma); if (anon_vma) return anon_vma; |
1da177e4c
|
1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 |
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; } |
1da177e4c
|
1075 |
/* |
404015308
|
1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 |
* 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; } |
363ee17f0
|
1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 |
static inline int mlock_future_check(struct mm_struct *mm, unsigned long flags, unsigned long len) { unsigned long locked, lock_limit; /* mlock MCL_FUTURE? */ if (flags & VM_LOCKED) { locked = len >> PAGE_SHIFT; locked += mm->locked_vm; lock_limit = rlimit(RLIMIT_MEMLOCK); lock_limit >>= PAGE_SHIFT; if (locked > lock_limit && !capable(CAP_IPC_LOCK)) return -EAGAIN; } return 0; } |
404015308
|
1104 |
/* |
27f5de796
|
1105 |
* The caller must hold down_write(¤t->mm->mmap_sem). |
1da177e4c
|
1106 |
*/ |
1fcfd8db7
|
1107 |
unsigned long do_mmap(struct file *file, unsigned long addr, |
1da177e4c
|
1108 |
unsigned long len, unsigned long prot, |
1fcfd8db7
|
1109 1110 |
unsigned long flags, vm_flags_t vm_flags, unsigned long pgoff, unsigned long *populate) |
1da177e4c
|
1111 |
{ |
cc71aba34
|
1112 |
struct mm_struct *mm = current->mm; |
62b5f7d01
|
1113 |
int pkey = 0; |
1da177e4c
|
1114 |
|
41badc15c
|
1115 |
*populate = 0; |
bebeb3d68
|
1116 |
|
e37609bb3
|
1117 1118 |
if (!len) return -EINVAL; |
1da177e4c
|
1119 1120 1121 1122 1123 1124 1125 |
/* * 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)) |
90f8572b0
|
1126 |
if (!(file && path_noexec(&file->f_path))) |
1da177e4c
|
1127 |
prot |= PROT_EXEC; |
7cd94146c
|
1128 1129 |
if (!(flags & MAP_FIXED)) addr = round_hint_to_min(addr); |
1da177e4c
|
1130 1131 |
/* Careful about overflows.. */ len = PAGE_ALIGN(len); |
9206de95b
|
1132 |
if (!len) |
1da177e4c
|
1133 1134 1135 1136 |
return -ENOMEM; /* offset overflow? */ if ((pgoff + (len >> PAGE_SHIFT)) < pgoff) |
cc71aba34
|
1137 |
return -EOVERFLOW; |
1da177e4c
|
1138 1139 1140 1141 1142 1143 1144 1145 1146 |
/* 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); |
de1741a13
|
1147 |
if (offset_in_page(addr)) |
1da177e4c
|
1148 |
return addr; |
62b5f7d01
|
1149 1150 1151 1152 1153 |
if (prot == PROT_EXEC) { pkey = execute_only_pkey(mm); if (pkey < 0) pkey = 0; } |
1da177e4c
|
1154 1155 1156 1157 |
/* 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. */ |
62b5f7d01
|
1158 |
vm_flags |= calc_vm_prot_bits(prot, pkey) | calc_vm_flag_bits(flags) | |
1da177e4c
|
1159 |
mm->def_flags | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC; |
cdf7b3418
|
1160 |
if (flags & MAP_LOCKED) |
1da177e4c
|
1161 1162 |
if (!can_do_mlock()) return -EPERM; |
ba470de43
|
1163 |
|
363ee17f0
|
1164 1165 |
if (mlock_future_check(mm, vm_flags, len)) return -EAGAIN; |
1da177e4c
|
1166 |
|
1da177e4c
|
1167 |
if (file) { |
077bf22b5
|
1168 |
struct inode *inode = file_inode(file); |
1da177e4c
|
1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 |
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. */ |
d7a06983a
|
1184 |
if (locks_verify_locked(file)) |
1da177e4c
|
1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 |
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; |
90f8572b0
|
1195 |
if (path_noexec(&file->f_path)) { |
80c5606c3
|
1196 1197 1198 1199 |
if (vm_flags & VM_EXEC) return -EPERM; vm_flags &= ~VM_MAYEXEC; } |
80c5606c3
|
1200 |
|
72c2d5319
|
1201 |
if (!file->f_op->mmap) |
80c5606c3
|
1202 |
return -ENODEV; |
b2c56e4f7
|
1203 1204 |
if (vm_flags & (VM_GROWSDOWN|VM_GROWSUP)) return -EINVAL; |
1da177e4c
|
1205 1206 1207 1208 1209 1210 1211 1212 |
break; default: return -EINVAL; } } else { switch (flags & MAP_TYPE) { case MAP_SHARED: |
b2c56e4f7
|
1213 1214 |
if (vm_flags & (VM_GROWSDOWN|VM_GROWSUP)) return -EINVAL; |
ce3639426
|
1215 1216 1217 1218 |
/* * Ignore pgoff. */ pgoff = 0; |
1da177e4c
|
1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 |
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; } } |
c22c0d634
|
1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 |
/* * Set 'VM_NORESERVE' if we should not account for the * memory use of this mapping. */ 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; } addr = mmap_region(file, addr, len, vm_flags, pgoff); |
09a9f1d27
|
1246 1247 1248 |
if (!IS_ERR_VALUE(addr) && ((vm_flags & VM_LOCKED) || (flags & (MAP_POPULATE | MAP_NONBLOCK)) == MAP_POPULATE)) |
41badc15c
|
1249 |
*populate = len; |
bebeb3d68
|
1250 |
return addr; |
0165ab443
|
1251 |
} |
6be5ceb02
|
1252 |
|
66f0dc481
|
1253 1254 1255 1256 1257 |
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; |
1e3ee14b9
|
1258 |
unsigned long retval; |
66f0dc481
|
1259 1260 |
if (!(flags & MAP_ANONYMOUS)) { |
120a795da
|
1261 |
audit_mmap_fd(fd, flags); |
66f0dc481
|
1262 1263 |
file = fget(fd); if (!file) |
1e3ee14b9
|
1264 |
return -EBADF; |
af73e4d95
|
1265 1266 |
if (is_file_hugepages(file)) len = ALIGN(len, huge_page_size(hstate_file(file))); |
493af5780
|
1267 1268 1269 |
retval = -EINVAL; if (unlikely(flags & MAP_HUGETLB && !is_file_hugepages(file))) goto out_fput; |
66f0dc481
|
1270 1271 |
} else if (flags & MAP_HUGETLB) { struct user_struct *user = NULL; |
c103a4dc4
|
1272 |
struct hstate *hs; |
af73e4d95
|
1273 |
|
c103a4dc4
|
1274 |
hs = hstate_sizelog((flags >> MAP_HUGE_SHIFT) & SHM_HUGE_MASK); |
091d0d55b
|
1275 1276 1277 1278 |
if (!hs) return -EINVAL; len = ALIGN(len, huge_page_size(hs)); |
66f0dc481
|
1279 1280 1281 1282 1283 1284 |
/* * 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 */ |
af73e4d95
|
1285 |
file = hugetlb_file_setup(HUGETLB_ANON_FILE, len, |
42d7395fe
|
1286 1287 1288 |
VM_NORESERVE, &user, HUGETLB_ANONHUGE_INODE, (flags >> MAP_HUGE_SHIFT) & MAP_HUGE_MASK); |
66f0dc481
|
1289 1290 1291 1292 1293 |
if (IS_ERR(file)) return PTR_ERR(file); } flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); |
eb36c5873
|
1294 |
retval = vm_mmap_pgoff(file, addr, len, prot, flags, pgoff); |
493af5780
|
1295 |
out_fput: |
66f0dc481
|
1296 1297 |
if (file) fput(file); |
66f0dc481
|
1298 1299 |
return retval; } |
a4679373c
|
1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 |
#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; |
de1741a13
|
1316 |
if (offset_in_page(a.offset)) |
a4679373c
|
1317 1318 1319 1320 1321 1322 |
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 */ |
4e950f6f0
|
1323 1324 1325 1326 1327 1328 1329 1330 |
/* * 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) { |
ca16d140a
|
1331 |
vm_flags_t vm_flags = vma->vm_flags; |
8a04446ab
|
1332 |
const struct vm_operations_struct *vm_ops = vma->vm_ops; |
4e950f6f0
|
1333 1334 1335 1336 1337 1338 |
/* 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? */ |
8a04446ab
|
1339 |
if (vm_ops && (vm_ops->page_mkwrite || vm_ops->pfn_mkwrite)) |
4e950f6f0
|
1340 |
return 1; |
64e455079
|
1341 1342 |
/* The open routine did something to the protections that pgprot_modify * won't preserve? */ |
4e950f6f0
|
1343 |
if (pgprot_val(vma->vm_page_prot) != |
64e455079
|
1344 |
pgprot_val(vm_pgprot_modify(vma->vm_page_prot, vm_flags))) |
4e950f6f0
|
1345 |
return 0; |
64e455079
|
1346 1347 1348 |
/* Do we need to track softdirty? */ if (IS_ENABLED(CONFIG_MEM_SOFT_DIRTY) && !(vm_flags & VM_SOFTDIRTY)) return 1; |
4e950f6f0
|
1349 |
/* Specialty mapping? */ |
4b6e1e370
|
1350 |
if (vm_flags & VM_PFNMAP) |
4e950f6f0
|
1351 1352 1353 1354 1355 1356 |
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); } |
fc8744adc
|
1357 1358 |
/* * We account for memory if it's a private writeable mapping, |
5a6fe1259
|
1359 |
* not hugepages and VM_NORESERVE wasn't set. |
fc8744adc
|
1360 |
*/ |
ca16d140a
|
1361 |
static inline int accountable_mapping(struct file *file, vm_flags_t vm_flags) |
fc8744adc
|
1362 |
{ |
5a6fe1259
|
1363 1364 1365 1366 1367 1368 |
/* * 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; |
fc8744adc
|
1369 1370 |
return (vm_flags & (VM_NORESERVE | VM_SHARED | VM_WRITE)) == VM_WRITE; } |
0165ab443
|
1371 |
unsigned long mmap_region(struct file *file, unsigned long addr, |
c22c0d634
|
1372 |
unsigned long len, vm_flags_t vm_flags, unsigned long pgoff) |
0165ab443
|
1373 1374 1375 |
{ struct mm_struct *mm = current->mm; struct vm_area_struct *vma, *prev; |
0165ab443
|
1376 1377 1378 |
int error; struct rb_node **rb_link, *rb_parent; unsigned long charged = 0; |
0165ab443
|
1379 |
|
e8420a8ec
|
1380 |
/* Check against address space limit. */ |
846383359
|
1381 |
if (!may_expand_vm(mm, vm_flags, len >> PAGE_SHIFT)) { |
e8420a8ec
|
1382 1383 1384 1385 1386 1387 |
unsigned long nr_pages; /* * MAP_FIXED may remove pages of mappings that intersects with * requested mapping. Account for the pages it would unmap. */ |
e8420a8ec
|
1388 |
nr_pages = count_vma_pages_range(mm, addr, addr + len); |
846383359
|
1389 1390 |
if (!may_expand_vm(mm, vm_flags, (len >> PAGE_SHIFT) - nr_pages)) |
e8420a8ec
|
1391 1392 |
return -ENOMEM; } |
1da177e4c
|
1393 |
/* Clear old maps */ |
9fcd14571
|
1394 1395 |
while (find_vma_links(mm, addr, addr + len, &prev, &rb_link, &rb_parent)) { |
1da177e4c
|
1396 1397 |
if (do_munmap(mm, addr, len)) return -ENOMEM; |
1da177e4c
|
1398 |
} |
fc8744adc
|
1399 |
/* |
fc8744adc
|
1400 1401 |
* Private writable mapping: check memory availability */ |
5a6fe1259
|
1402 |
if (accountable_mapping(file, vm_flags)) { |
fc8744adc
|
1403 |
charged = len >> PAGE_SHIFT; |
191c54244
|
1404 |
if (security_vm_enough_memory_mm(mm, charged)) |
fc8744adc
|
1405 1406 |
return -ENOMEM; vm_flags |= VM_ACCOUNT; |
1da177e4c
|
1407 1408 1409 |
} /* |
de33c8db5
|
1410 |
* Can we just expand an old mapping? |
1da177e4c
|
1411 |
*/ |
19a809afe
|
1412 1413 |
vma = vma_merge(mm, prev, addr, addr + len, vm_flags, NULL, file, pgoff, NULL, NULL_VM_UFFD_CTX); |
de33c8db5
|
1414 1415 |
if (vma) goto out; |
1da177e4c
|
1416 1417 1418 1419 1420 1421 |
/* * 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. */ |
c5e3b83e9
|
1422 |
vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); |
1da177e4c
|
1423 1424 1425 1426 |
if (!vma) { error = -ENOMEM; goto unacct_error; } |
1da177e4c
|
1427 1428 1429 1430 1431 |
vma->vm_mm = mm; vma->vm_start = addr; vma->vm_end = addr + len; vma->vm_flags = vm_flags; |
3ed75eb8f
|
1432 |
vma->vm_page_prot = vm_get_page_prot(vm_flags); |
1da177e4c
|
1433 |
vma->vm_pgoff = pgoff; |
5beb49305
|
1434 |
INIT_LIST_HEAD(&vma->anon_vma_chain); |
1da177e4c
|
1435 1436 |
if (file) { |
1da177e4c
|
1437 1438 1439 1440 |
if (vm_flags & VM_DENYWRITE) { error = deny_write_access(file); if (error) goto free_vma; |
1da177e4c
|
1441 |
} |
4bb5f5d93
|
1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 |
if (vm_flags & VM_SHARED) { error = mapping_map_writable(file->f_mapping); if (error) goto allow_write_and_free_vma; } /* ->mmap() can change vma->vm_file, but must guarantee that * vma_link() below can deny write-access if VM_DENYWRITE is set * and map writably if VM_SHARED is set. This usually means the * new file must not have been exposed to user-space, yet. */ |
cb0942b81
|
1453 |
vma->vm_file = get_file(file); |
1da177e4c
|
1454 1455 1456 |
error = file->f_op->mmap(file, vma); if (error) goto unmap_and_free_vma; |
f8dbf0a7a
|
1457 1458 1459 1460 1461 |
/* Can addr have changed?? * * Answer: Yes, several device drivers can do it in their * f_op->mmap method. -DaveM |
2897b4d29
|
1462 1463 |
* Bug: If addr is changed, prev, rb_link, rb_parent should * be updated for vma_link() |
f8dbf0a7a
|
1464 |
*/ |
2897b4d29
|
1465 |
WARN_ON_ONCE(addr != vma->vm_start); |
f8dbf0a7a
|
1466 |
addr = vma->vm_start; |
f8dbf0a7a
|
1467 |
vm_flags = vma->vm_flags; |
1da177e4c
|
1468 1469 1470 1471 1472 |
} else if (vm_flags & VM_SHARED) { error = shmem_zero_setup(vma); if (error) goto free_vma; } |
de33c8db5
|
1473 |
vma_link(mm, vma, prev, rb_link, rb_parent); |
4d3d5b41a
|
1474 |
/* Once vma denies write, undo our temporary denial count */ |
4bb5f5d93
|
1475 1476 1477 1478 1479 1480 |
if (file) { if (vm_flags & VM_SHARED) mapping_unmap_writable(file->f_mapping); if (vm_flags & VM_DENYWRITE) allow_write_access(file); } |
e86867720
|
1481 |
file = vma->vm_file; |
4d3d5b41a
|
1482 |
out: |
cdd6c482c
|
1483 |
perf_event_mmap(vma); |
0a4a93919
|
1484 |
|
846383359
|
1485 |
vm_stat_account(mm, vm_flags, len >> PAGE_SHIFT); |
1da177e4c
|
1486 |
if (vm_flags & VM_LOCKED) { |
bebeb3d68
|
1487 1488 |
if (!((vm_flags & VM_SPECIAL) || is_vm_hugetlb_page(vma) || vma == get_gate_vma(current->mm))) |
06f9d8c2b
|
1489 |
mm->locked_vm += (len >> PAGE_SHIFT); |
bebeb3d68
|
1490 |
else |
de60f5f10
|
1491 |
vma->vm_flags &= VM_LOCKED_CLEAR_MASK; |
bebeb3d68
|
1492 |
} |
2b1444983
|
1493 |
|
c7a3a88c9
|
1494 1495 |
if (file) uprobe_mmap(vma); |
2b1444983
|
1496 |
|
d9104d1ca
|
1497 1498 1499 1500 1501 1502 1503 1504 |
/* * New (or expanded) vma always get soft dirty status. * Otherwise user-space soft-dirty page tracker won't * be able to distinguish situation when vma area unmapped, * then new mapped in-place (which must be aimed as * a completely new data area). */ vma->vm_flags |= VM_SOFTDIRTY; |
64e455079
|
1505 |
vma_set_page_prot(vma); |
1da177e4c
|
1506 1507 1508 |
return addr; unmap_and_free_vma: |
1da177e4c
|
1509 1510 1511 1512 |
vma->vm_file = NULL; fput(file); /* Undo any partial mapping done by a device driver. */ |
e0da382c9
|
1513 1514 |
unmap_region(mm, vma, prev, vma->vm_start, vma->vm_end); charged = 0; |
4bb5f5d93
|
1515 1516 1517 1518 1519 |
if (vm_flags & VM_SHARED) mapping_unmap_writable(file->f_mapping); allow_write_and_free_vma: if (vm_flags & VM_DENYWRITE) allow_write_access(file); |
1da177e4c
|
1520 1521 1522 1523 1524 1525 1526 |
free_vma: kmem_cache_free(vm_area_cachep, vma); unacct_error: if (charged) vm_unacct_memory(charged); return error; } |
db4fbfb95
|
1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 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 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 |
unsigned long unmapped_area(struct vm_unmapped_area_info *info) { /* * We implement the search by looking for an rbtree node that * immediately follows a suitable gap. That is, * - gap_start = vma->vm_prev->vm_end <= info->high_limit - length; * - gap_end = vma->vm_start >= info->low_limit + length; * - gap_end - gap_start >= length */ struct mm_struct *mm = current->mm; struct vm_area_struct *vma; unsigned long length, low_limit, high_limit, gap_start, gap_end; /* Adjust search length to account for worst case alignment overhead */ length = info->length + info->align_mask; if (length < info->length) return -ENOMEM; /* Adjust search limits by the desired length */ if (info->high_limit < length) return -ENOMEM; high_limit = info->high_limit - length; if (info->low_limit > high_limit) return -ENOMEM; low_limit = info->low_limit + length; /* Check if rbtree root looks promising */ if (RB_EMPTY_ROOT(&mm->mm_rb)) goto check_highest; vma = rb_entry(mm->mm_rb.rb_node, struct vm_area_struct, vm_rb); if (vma->rb_subtree_gap < length) goto check_highest; while (true) { /* Visit left subtree if it looks promising */ gap_end = vma->vm_start; if (gap_end >= low_limit && vma->vm_rb.rb_left) { struct vm_area_struct *left = rb_entry(vma->vm_rb.rb_left, struct vm_area_struct, vm_rb); if (left->rb_subtree_gap >= length) { vma = left; continue; } } gap_start = vma->vm_prev ? vma->vm_prev->vm_end : 0; check_current: /* Check if current node has a suitable gap */ if (gap_start > high_limit) return -ENOMEM; if (gap_end >= low_limit && gap_end - gap_start >= length) goto found; /* Visit right subtree if it looks promising */ if (vma->vm_rb.rb_right) { struct vm_area_struct *right = rb_entry(vma->vm_rb.rb_right, struct vm_area_struct, vm_rb); if (right->rb_subtree_gap >= length) { vma = right; continue; } } /* Go back up the rbtree to find next candidate node */ while (true) { struct rb_node *prev = &vma->vm_rb; if (!rb_parent(prev)) goto check_highest; vma = rb_entry(rb_parent(prev), struct vm_area_struct, vm_rb); if (prev == vma->vm_rb.rb_left) { gap_start = vma->vm_prev->vm_end; gap_end = vma->vm_start; goto check_current; } } } check_highest: /* Check highest gap, which does not precede any rbtree node */ gap_start = mm->highest_vm_end; gap_end = ULONG_MAX; /* Only for VM_BUG_ON below */ if (gap_start > high_limit) return -ENOMEM; found: /* We found a suitable gap. Clip it with the original low_limit. */ if (gap_start < info->low_limit) gap_start = info->low_limit; /* Adjust gap address to the desired alignment */ gap_start += (info->align_offset - gap_start) & info->align_mask; VM_BUG_ON(gap_start + info->length > info->high_limit); VM_BUG_ON(gap_start + info->length > gap_end); return gap_start; } unsigned long unmapped_area_topdown(struct vm_unmapped_area_info *info) { struct mm_struct *mm = current->mm; struct vm_area_struct *vma; unsigned long length, low_limit, high_limit, gap_start, gap_end; /* Adjust search length to account for worst case alignment overhead */ length = info->length + info->align_mask; if (length < info->length) return -ENOMEM; /* * Adjust search limits by the desired length. * See implementation comment at top of unmapped_area(). */ gap_end = info->high_limit; if (gap_end < length) return -ENOMEM; high_limit = gap_end - length; if (info->low_limit > high_limit) return -ENOMEM; low_limit = info->low_limit + length; /* Check highest gap, which does not precede any rbtree node */ gap_start = mm->highest_vm_end; if (gap_start <= high_limit) goto found_highest; /* Check if rbtree root looks promising */ if (RB_EMPTY_ROOT(&mm->mm_rb)) return -ENOMEM; vma = rb_entry(mm->mm_rb.rb_node, struct vm_area_struct, vm_rb); if (vma->rb_subtree_gap < length) return -ENOMEM; while (true) { /* Visit right subtree if it looks promising */ gap_start = vma->vm_prev ? vma->vm_prev->vm_end : 0; if (gap_start <= high_limit && vma->vm_rb.rb_right) { struct vm_area_struct *right = rb_entry(vma->vm_rb.rb_right, struct vm_area_struct, vm_rb); if (right->rb_subtree_gap >= length) { vma = right; continue; } } check_current: /* Check if current node has a suitable gap */ gap_end = vma->vm_start; if (gap_end < low_limit) return -ENOMEM; if (gap_start <= high_limit && gap_end - gap_start >= length) goto found; /* Visit left subtree if it looks promising */ if (vma->vm_rb.rb_left) { struct vm_area_struct *left = rb_entry(vma->vm_rb.rb_left, struct vm_area_struct, vm_rb); if (left->rb_subtree_gap >= length) { vma = left; continue; } } /* Go back up the rbtree to find next candidate node */ while (true) { struct rb_node *prev = &vma->vm_rb; if (!rb_parent(prev)) return -ENOMEM; vma = rb_entry(rb_parent(prev), struct vm_area_struct, vm_rb); if (prev == vma->vm_rb.rb_right) { gap_start = vma->vm_prev ? vma->vm_prev->vm_end : 0; goto check_current; } } } found: /* We found a suitable gap. Clip it with the original high_limit. */ if (gap_end > info->high_limit) gap_end = info->high_limit; found_highest: /* Compute highest gap address at the desired alignment */ gap_end -= info->length; gap_end -= (gap_end - info->align_offset) & info->align_mask; VM_BUG_ON(gap_end < info->low_limit); VM_BUG_ON(gap_end < gap_start); return gap_end; } |
1da177e4c
|
1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 |
/* 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; |
db4fbfb95
|
1744 |
struct vm_unmapped_area_info info; |
1da177e4c
|
1745 |
|
2afc745f3
|
1746 |
if (len > TASK_SIZE - mmap_min_addr) |
1da177e4c
|
1747 |
return -ENOMEM; |
06abdfb47
|
1748 1749 |
if (flags & MAP_FIXED) return addr; |
1da177e4c
|
1750 1751 1752 |
if (addr) { addr = PAGE_ALIGN(addr); vma = find_vma(mm, addr); |
2afc745f3
|
1753 |
if (TASK_SIZE - len >= addr && addr >= mmap_min_addr && |
1da177e4c
|
1754 1755 1756 |
(!vma || addr + len <= vma->vm_start)) return addr; } |
1da177e4c
|
1757 |
|
db4fbfb95
|
1758 1759 |
info.flags = 0; info.length = len; |
4e99b0213
|
1760 |
info.low_limit = mm->mmap_base; |
db4fbfb95
|
1761 1762 1763 |
info.high_limit = TASK_SIZE; info.align_mask = 0; return vm_unmapped_area(&info); |
1da177e4c
|
1764 |
} |
cc71aba34
|
1765 |
#endif |
1da177e4c
|
1766 |
|
1da177e4c
|
1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 |
/* * 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; |
db4fbfb95
|
1779 1780 |
unsigned long addr = addr0; struct vm_unmapped_area_info info; |
1da177e4c
|
1781 1782 |
/* requested length too big for entire address space */ |
2afc745f3
|
1783 |
if (len > TASK_SIZE - mmap_min_addr) |
1da177e4c
|
1784 |
return -ENOMEM; |
06abdfb47
|
1785 1786 |
if (flags & MAP_FIXED) return addr; |
1da177e4c
|
1787 1788 1789 1790 |
/* requesting a specific address */ if (addr) { addr = PAGE_ALIGN(addr); vma = find_vma(mm, addr); |
2afc745f3
|
1791 |
if (TASK_SIZE - len >= addr && addr >= mmap_min_addr && |
1da177e4c
|
1792 1793 1794 |
(!vma || addr + len <= vma->vm_start)) return addr; } |
db4fbfb95
|
1795 1796 |
info.flags = VM_UNMAPPED_AREA_TOPDOWN; info.length = len; |
2afc745f3
|
1797 |
info.low_limit = max(PAGE_SIZE, mmap_min_addr); |
db4fbfb95
|
1798 1799 1800 |
info.high_limit = mm->mmap_base; info.align_mask = 0; addr = vm_unmapped_area(&info); |
b716ad953
|
1801 |
|
1da177e4c
|
1802 1803 1804 1805 1806 1807 |
/* * 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. */ |
de1741a13
|
1808 |
if (offset_in_page(addr)) { |
db4fbfb95
|
1809 1810 1811 1812 1813 1814 |
VM_BUG_ON(addr != -ENOMEM); info.flags = 0; info.low_limit = TASK_UNMAPPED_BASE; info.high_limit = TASK_SIZE; addr = vm_unmapped_area(&info); } |
1da177e4c
|
1815 1816 1817 1818 |
return addr; } #endif |
1da177e4c
|
1819 1820 1821 1822 |
unsigned long get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { |
06abdfb47
|
1823 1824 |
unsigned long (*get_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long); |
9206de95b
|
1825 1826 1827 1828 1829 1830 1831 |
unsigned long error = arch_mmap_check(addr, len, flags); if (error) return error; /* Careful about overflows.. */ if (len > TASK_SIZE) return -ENOMEM; |
06abdfb47
|
1832 |
get_area = current->mm->get_unmapped_area; |
72c2d5319
|
1833 |
if (file && file->f_op->get_unmapped_area) |
06abdfb47
|
1834 1835 1836 1837 |
get_area = file->f_op->get_unmapped_area; addr = get_area(file, addr, len, pgoff, flags); if (IS_ERR_VALUE(addr)) return addr; |
1da177e4c
|
1838 |
|
07ab67c8d
|
1839 1840 |
if (addr > TASK_SIZE - len) return -ENOMEM; |
de1741a13
|
1841 |
if (offset_in_page(addr)) |
07ab67c8d
|
1842 |
return -EINVAL; |
06abdfb47
|
1843 |
|
9ac4ed4bd
|
1844 1845 1846 |
addr = arch_rebalance_pgtables(addr, len); error = security_mmap_addr(addr); return error ? error : addr; |
1da177e4c
|
1847 1848 1849 1850 1851 |
} EXPORT_SYMBOL(get_unmapped_area); /* Look up the first VMA which satisfies addr < vm_end, NULL if none. */ |
48aae4255
|
1852 |
struct vm_area_struct *find_vma(struct mm_struct *mm, unsigned long addr) |
1da177e4c
|
1853 |
{ |
615d6e875
|
1854 1855 |
struct rb_node *rb_node; struct vm_area_struct *vma; |
1da177e4c
|
1856 |
|
841e31e5c
|
1857 |
/* Check the cache first. */ |
615d6e875
|
1858 1859 1860 |
vma = vmacache_find(mm, addr); if (likely(vma)) return vma; |
841e31e5c
|
1861 |
|
615d6e875
|
1862 |
rb_node = mm->mm_rb.rb_node; |
841e31e5c
|
1863 |
|
615d6e875
|
1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 |
while (rb_node) { struct vm_area_struct *tmp; tmp = rb_entry(rb_node, struct vm_area_struct, vm_rb); if (tmp->vm_end > addr) { vma = tmp; if (tmp->vm_start <= addr) break; rb_node = rb_node->rb_left; } else rb_node = rb_node->rb_right; |
1da177e4c
|
1876 |
} |
615d6e875
|
1877 1878 1879 |
if (vma) vmacache_update(addr, vma); |
1da177e4c
|
1880 1881 1882 1883 |
return vma; } EXPORT_SYMBOL(find_vma); |
6bd4837de
|
1884 1885 |
/* * Same as find_vma, but also return a pointer to the previous VMA in *pprev. |
6bd4837de
|
1886 |
*/ |
1da177e4c
|
1887 1888 1889 1890 |
struct vm_area_struct * find_vma_prev(struct mm_struct *mm, unsigned long addr, struct vm_area_struct **pprev) { |
6bd4837de
|
1891 |
struct vm_area_struct *vma; |
1da177e4c
|
1892 |
|
6bd4837de
|
1893 |
vma = find_vma(mm, addr); |
83cd904d2
|
1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 |
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; } } |
6bd4837de
|
1904 |
return vma; |
1da177e4c
|
1905 1906 1907 1908 1909 1910 1911 |
} /* * Verify that the stack growth is acceptable and * update accounting. This is shared with both the * grow-up and grow-down cases. */ |
48aae4255
|
1912 |
static int acct_stack_growth(struct vm_area_struct *vma, unsigned long size, unsigned long grow) |
1da177e4c
|
1913 1914 1915 |
{ struct mm_struct *mm = vma->vm_mm; struct rlimit *rlim = current->signal->rlim; |
690eac53d
|
1916 |
unsigned long new_start, actual_size; |
1da177e4c
|
1917 1918 |
/* address space limit tests */ |
846383359
|
1919 |
if (!may_expand_vm(mm, vma->vm_flags, grow)) |
1da177e4c
|
1920 1921 1922 |
return -ENOMEM; /* Stack limit test */ |
690eac53d
|
1923 1924 1925 |
actual_size = size; if (size && (vma->vm_flags & (VM_GROWSUP | VM_GROWSDOWN))) actual_size -= PAGE_SIZE; |
4db0c3c29
|
1926 |
if (actual_size > READ_ONCE(rlim[RLIMIT_STACK].rlim_cur)) |
1da177e4c
|
1927 1928 1929 1930 1931 1932 1933 |
return -ENOMEM; /* mlock limit tests */ if (vma->vm_flags & VM_LOCKED) { unsigned long locked; unsigned long limit; locked = mm->locked_vm + grow; |
4db0c3c29
|
1934 |
limit = READ_ONCE(rlim[RLIMIT_MEMLOCK].rlim_cur); |
59e99e5b9
|
1935 |
limit >>= PAGE_SHIFT; |
1da177e4c
|
1936 1937 1938 |
if (locked > limit && !capable(CAP_IPC_LOCK)) return -ENOMEM; } |
0d59a01bc
|
1939 1940 1941 1942 1943 |
/* 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; |
1da177e4c
|
1944 1945 1946 1947 |
/* * Overcommit.. This must be the final test, as it will * update security statistics. */ |
05fa199d4
|
1948 |
if (security_vm_enough_memory_mm(mm, grow)) |
1da177e4c
|
1949 |
return -ENOMEM; |
1da177e4c
|
1950 1951 |
return 0; } |
46dea3d09
|
1952 |
#if defined(CONFIG_STACK_GROWSUP) || defined(CONFIG_IA64) |
1da177e4c
|
1953 |
/* |
46dea3d09
|
1954 1955 |
* 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. |
1da177e4c
|
1956 |
*/ |
46dea3d09
|
1957 |
int expand_upwards(struct vm_area_struct *vma, unsigned long address) |
1da177e4c
|
1958 |
{ |
093578147
|
1959 |
struct mm_struct *mm = vma->vm_mm; |
12352d3ca
|
1960 |
int error = 0; |
1da177e4c
|
1961 1962 1963 |
if (!(vma->vm_flags & VM_GROWSUP)) return -EFAULT; |
12352d3ca
|
1964 1965 1966 1967 1968 1969 1970 |
/* Guard against wrapping around to address 0. */ if (address < PAGE_ALIGN(address+4)) address = PAGE_ALIGN(address+4); else return -ENOMEM; /* We must make sure the anon_vma is allocated. */ |
1da177e4c
|
1971 1972 |
if (unlikely(anon_vma_prepare(vma))) return -ENOMEM; |
1da177e4c
|
1973 1974 1975 1976 1977 1978 |
/* * 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. */ |
12352d3ca
|
1979 |
anon_vma_lock_write(vma->anon_vma); |
1da177e4c
|
1980 1981 1982 1983 1984 1985 1986 |
/* 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; |
42c36f63a
|
1987 1988 1989 1990 |
error = -ENOMEM; if (vma->vm_pgoff + (size >> PAGE_SHIFT) >= vma->vm_pgoff) { error = acct_stack_growth(vma, size, grow); if (!error) { |
4128997b5
|
1991 1992 1993 1994 1995 |
/* * vma_gap_update() doesn't support concurrent * updates, but we only hold a shared mmap_sem * lock here, so we need to protect against * concurrent vma expansions. |
12352d3ca
|
1996 |
* anon_vma_lock_write() doesn't help here, as |
4128997b5
|
1997 1998 1999 2000 2001 |
* we don't guarantee that all growable vmas * in a mm share the same root anon vma. * So, we reuse mm->page_table_lock to guard * against concurrent vma expansions. */ |
093578147
|
2002 |
spin_lock(&mm->page_table_lock); |
87e8827b3
|
2003 |
if (vma->vm_flags & VM_LOCKED) |
093578147
|
2004 |
mm->locked_vm += grow; |
846383359
|
2005 |
vm_stat_account(mm, vma->vm_flags, grow); |
bf181b9f9
|
2006 |
anon_vma_interval_tree_pre_update_vma(vma); |
42c36f63a
|
2007 |
vma->vm_end = address; |
bf181b9f9
|
2008 |
anon_vma_interval_tree_post_update_vma(vma); |
d37371870
|
2009 2010 2011 |
if (vma->vm_next) vma_gap_update(vma->vm_next); else |
093578147
|
2012 2013 |
mm->highest_vm_end = address; spin_unlock(&mm->page_table_lock); |
4128997b5
|
2014 |
|
42c36f63a
|
2015 2016 |
perf_event_mmap(vma); } |
3af9e8592
|
2017 |
} |
1da177e4c
|
2018 |
} |
12352d3ca
|
2019 |
anon_vma_unlock_write(vma->anon_vma); |
6d50e60cd
|
2020 |
khugepaged_enter_vma_merge(vma, vma->vm_flags); |
093578147
|
2021 |
validate_mm(mm); |
1da177e4c
|
2022 2023 |
return error; } |
46dea3d09
|
2024 |
#endif /* CONFIG_STACK_GROWSUP || CONFIG_IA64 */ |
1da177e4c
|
2025 2026 2027 |
/* * vma is the first one with address < vma->vm_start. Have to extend vma. */ |
d05f3169c
|
2028 |
int expand_downwards(struct vm_area_struct *vma, |
b6a2fea39
|
2029 |
unsigned long address) |
1da177e4c
|
2030 |
{ |
093578147
|
2031 |
struct mm_struct *mm = vma->vm_mm; |
1da177e4c
|
2032 |
int error; |
8869477a4
|
2033 |
address &= PAGE_MASK; |
e5467859f
|
2034 |
error = security_mmap_addr(address); |
8869477a4
|
2035 2036 |
if (error) return error; |
12352d3ca
|
2037 2038 2039 |
/* We must make sure the anon_vma is allocated. */ if (unlikely(anon_vma_prepare(vma))) return -ENOMEM; |
1da177e4c
|
2040 2041 2042 2043 2044 2045 |
/* * 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. */ |
12352d3ca
|
2046 |
anon_vma_lock_write(vma->anon_vma); |
1da177e4c
|
2047 2048 2049 2050 2051 2052 2053 |
/* 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; |
a626ca6a6
|
2054 2055 2056 2057 |
error = -ENOMEM; if (grow <= vma->vm_pgoff) { error = acct_stack_growth(vma, size, grow); if (!error) { |
4128997b5
|
2058 2059 2060 2061 2062 |
/* * vma_gap_update() doesn't support concurrent * updates, but we only hold a shared mmap_sem * lock here, so we need to protect against * concurrent vma expansions. |
12352d3ca
|
2063 |
* anon_vma_lock_write() doesn't help here, as |
4128997b5
|
2064 2065 2066 2067 2068 |
* we don't guarantee that all growable vmas * in a mm share the same root anon vma. * So, we reuse mm->page_table_lock to guard * against concurrent vma expansions. */ |
093578147
|
2069 |
spin_lock(&mm->page_table_lock); |
87e8827b3
|
2070 |
if (vma->vm_flags & VM_LOCKED) |
093578147
|
2071 |
mm->locked_vm += grow; |
846383359
|
2072 |
vm_stat_account(mm, vma->vm_flags, grow); |
bf181b9f9
|
2073 |
anon_vma_interval_tree_pre_update_vma(vma); |
a626ca6a6
|
2074 2075 |
vma->vm_start = address; vma->vm_pgoff -= grow; |
bf181b9f9
|
2076 |
anon_vma_interval_tree_post_update_vma(vma); |
d37371870
|
2077 |
vma_gap_update(vma); |
093578147
|
2078 |
spin_unlock(&mm->page_table_lock); |
4128997b5
|
2079 |
|
a626ca6a6
|
2080 2081 |
perf_event_mmap(vma); } |
1da177e4c
|
2082 2083 |
} } |
12352d3ca
|
2084 |
anon_vma_unlock_write(vma->anon_vma); |
6d50e60cd
|
2085 |
khugepaged_enter_vma_merge(vma, vma->vm_flags); |
093578147
|
2086 |
validate_mm(mm); |
1da177e4c
|
2087 2088 |
return error; } |
098849643
|
2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 |
/* * Note how expand_stack() refuses to expand the stack all the way to * abut the next virtual mapping, *unless* that mapping itself is also * a stack mapping. We want to leave room for a guard page, after all * (the guard page itself is not added here, that is done by the * actual page faulting logic) * * This matches the behavior of the guard page logic (see mm/memory.c: * check_stack_guard_page()), which only allows the guard page to be * removed under these circumstances. */ |
b6a2fea39
|
2100 2101 2102 |
#ifdef CONFIG_STACK_GROWSUP int expand_stack(struct vm_area_struct *vma, unsigned long address) { |
098849643
|
2103 2104 2105 2106 2107 2108 2109 2110 |
struct vm_area_struct *next; address &= PAGE_MASK; next = vma->vm_next; if (next && next->vm_start == address + PAGE_SIZE) { if (!(next->vm_flags & VM_GROWSUP)) return -ENOMEM; } |
b6a2fea39
|
2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 |
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; |
1c1271850
|
2123 |
if (!prev || expand_stack(prev, addr)) |
b6a2fea39
|
2124 |
return NULL; |
cea10a19b
|
2125 |
if (prev->vm_flags & VM_LOCKED) |
fc05f5662
|
2126 |
populate_vma_page_range(prev, addr, prev->vm_end, NULL); |
b6a2fea39
|
2127 2128 2129 2130 2131 |
return prev; } #else int expand_stack(struct vm_area_struct *vma, unsigned long address) { |
098849643
|
2132 2133 2134 2135 2136 2137 2138 2139 |
struct vm_area_struct *prev; address &= PAGE_MASK; prev = vma->vm_prev; if (prev && prev->vm_end == address) { if (!(prev->vm_flags & VM_GROWSDOWN)) return -ENOMEM; } |
b6a2fea39
|
2140 2141 |
return expand_downwards(vma, address); } |
1da177e4c
|
2142 |
struct vm_area_struct * |
cc71aba34
|
2143 |
find_extend_vma(struct mm_struct *mm, unsigned long addr) |
1da177e4c
|
2144 |
{ |
cc71aba34
|
2145 |
struct vm_area_struct *vma; |
1da177e4c
|
2146 2147 2148 |
unsigned long start; addr &= PAGE_MASK; |
cc71aba34
|
2149 |
vma = find_vma(mm, addr); |
1da177e4c
|
2150 2151 2152 2153 2154 2155 2156 2157 2158 |
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; |
cea10a19b
|
2159 |
if (vma->vm_flags & VM_LOCKED) |
fc05f5662
|
2160 |
populate_vma_page_range(vma, addr, start, NULL); |
1da177e4c
|
2161 2162 2163 |
return vma; } #endif |
e1d6d01ab
|
2164 |
EXPORT_SYMBOL_GPL(find_extend_vma); |
1da177e4c
|
2165 |
/* |
2c0b38146
|
2166 |
* Ok - we have the memory areas we should free on the vma list, |
1da177e4c
|
2167 |
* so release them, and do the vma updates. |
2c0b38146
|
2168 2169 |
* * Called with the mm semaphore held. |
1da177e4c
|
2170 |
*/ |
2c0b38146
|
2171 |
static void remove_vma_list(struct mm_struct *mm, struct vm_area_struct *vma) |
1da177e4c
|
2172 |
{ |
4f74d2c8e
|
2173 |
unsigned long nr_accounted = 0; |
365e9c87a
|
2174 2175 |
/* Update high watermark before we lower total_vm */ update_hiwater_vm(mm); |
1da177e4c
|
2176 |
do { |
2c0b38146
|
2177 |
long nrpages = vma_pages(vma); |
4f74d2c8e
|
2178 2179 |
if (vma->vm_flags & VM_ACCOUNT) nr_accounted += nrpages; |
846383359
|
2180 |
vm_stat_account(mm, vma->vm_flags, -nrpages); |
a8fb5618d
|
2181 |
vma = remove_vma(vma); |
146425a31
|
2182 |
} while (vma); |
4f74d2c8e
|
2183 |
vm_unacct_memory(nr_accounted); |
1da177e4c
|
2184 2185 2186 2187 2188 2189 |
validate_mm(mm); } /* * Get rid of page table information in the indicated region. * |
f10df6860
|
2190 |
* Called with the mm semaphore held. |
1da177e4c
|
2191 2192 |
*/ static void unmap_region(struct mm_struct *mm, |
e0da382c9
|
2193 2194 |
struct vm_area_struct *vma, struct vm_area_struct *prev, unsigned long start, unsigned long end) |
1da177e4c
|
2195 |
{ |
cc71aba34
|
2196 |
struct vm_area_struct *next = prev ? prev->vm_next : mm->mmap; |
d16dfc550
|
2197 |
struct mmu_gather tlb; |
1da177e4c
|
2198 2199 |
lru_add_drain(); |
2b047252d
|
2200 |
tlb_gather_mmu(&tlb, mm, start, end); |
365e9c87a
|
2201 |
update_hiwater_rss(mm); |
4f74d2c8e
|
2202 |
unmap_vmas(&tlb, vma, start, end); |
d16dfc550
|
2203 |
free_pgtables(&tlb, vma, prev ? prev->vm_end : FIRST_USER_ADDRESS, |
6ee8630e0
|
2204 |
next ? next->vm_start : USER_PGTABLES_CEILING); |
d16dfc550
|
2205 |
tlb_finish_mmu(&tlb, start, end); |
1da177e4c
|
2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 |
} /* * 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; insertion_point = (prev ? &prev->vm_next : &mm->mmap); |
297c5eee3
|
2220 |
vma->vm_prev = NULL; |
1da177e4c
|
2221 |
do { |
d37371870
|
2222 |
vma_rb_erase(vma, &mm->mm_rb); |
1da177e4c
|
2223 2224 2225 2226 2227 |
mm->map_count--; tail_vma = vma; vma = vma->vm_next; } while (vma && vma->vm_start < end); *insertion_point = vma; |
d37371870
|
2228 |
if (vma) { |
297c5eee3
|
2229 |
vma->vm_prev = prev; |
d37371870
|
2230 2231 2232 |
vma_gap_update(vma); } else mm->highest_vm_end = prev ? prev->vm_end : 0; |
1da177e4c
|
2233 |
tail_vma->vm_next = NULL; |
615d6e875
|
2234 2235 2236 |
/* Kill the cache */ vmacache_invalidate(mm); |
1da177e4c
|
2237 2238 2239 |
} /* |
659ace584
|
2240 2241 |
* __split_vma() bypasses sysctl_max_map_count checking. We use this on the * munmap path where it doesn't make sense to fail. |
1da177e4c
|
2242 |
*/ |
cc71aba34
|
2243 |
static int __split_vma(struct mm_struct *mm, struct vm_area_struct *vma, |
1da177e4c
|
2244 2245 |
unsigned long addr, int new_below) { |
1da177e4c
|
2246 |
struct vm_area_struct *new; |
e39758912
|
2247 |
int err; |
1da177e4c
|
2248 |
|
a55164389
|
2249 2250 |
if (is_vm_hugetlb_page(vma) && (addr & ~(huge_page_mask(hstate_vma(vma))))) |
1da177e4c
|
2251 |
return -EINVAL; |
e94b17660
|
2252 |
new = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); |
1da177e4c
|
2253 |
if (!new) |
e39758912
|
2254 |
return -ENOMEM; |
1da177e4c
|
2255 2256 2257 |
/* most fields are the same, copy all, and then fixup */ *new = *vma; |
5beb49305
|
2258 |
INIT_LIST_HEAD(&new->anon_vma_chain); |
1da177e4c
|
2259 2260 2261 2262 2263 2264 |
if (new_below) new->vm_end = addr; else { new->vm_start = addr; new->vm_pgoff += ((addr - vma->vm_start) >> PAGE_SHIFT); } |
ef0855d33
|
2265 2266 |
err = vma_dup_policy(vma, new); if (err) |
5beb49305
|
2267 |
goto out_free_vma; |
1da177e4c
|
2268 |
|
c4ea95d7c
|
2269 2270 |
err = anon_vma_clone(new, vma); if (err) |
5beb49305
|
2271 |
goto out_free_mpol; |
e9714acf8
|
2272 |
if (new->vm_file) |
1da177e4c
|
2273 2274 2275 2276 2277 2278 |
get_file(new->vm_file); if (new->vm_ops && new->vm_ops->open) new->vm_ops->open(new); if (new_below) |
5beb49305
|
2279 |
err = vma_adjust(vma, addr, vma->vm_end, vma->vm_pgoff + |
1da177e4c
|
2280 2281 |
((addr - new->vm_start) >> PAGE_SHIFT), new); else |
5beb49305
|
2282 |
err = vma_adjust(vma, vma->vm_start, addr, vma->vm_pgoff, new); |
1da177e4c
|
2283 |
|
5beb49305
|
2284 2285 2286 2287 2288 |
/* Success. */ if (!err) return 0; /* Clean everything up if vma_adjust failed. */ |
589275338
|
2289 2290 |
if (new->vm_ops && new->vm_ops->close) new->vm_ops->close(new); |
e9714acf8
|
2291 |
if (new->vm_file) |
5beb49305
|
2292 |
fput(new->vm_file); |
2aeadc30d
|
2293 |
unlink_anon_vmas(new); |
5beb49305
|
2294 |
out_free_mpol: |
ef0855d33
|
2295 |
mpol_put(vma_policy(new)); |
5beb49305
|
2296 2297 |
out_free_vma: kmem_cache_free(vm_area_cachep, new); |
5beb49305
|
2298 |
return err; |
1da177e4c
|
2299 |
} |
659ace584
|
2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 |
/* * 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); } |
1da177e4c
|
2312 2313 2314 2315 2316 2317 2318 2319 |
/* 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; |
146425a31
|
2320 |
struct vm_area_struct *vma, *prev, *last; |
1da177e4c
|
2321 |
|
de1741a13
|
2322 |
if ((offset_in_page(start)) || start > TASK_SIZE || len > TASK_SIZE-start) |
1da177e4c
|
2323 |
return -EINVAL; |
cc71aba34
|
2324 2325 |
len = PAGE_ALIGN(len); if (len == 0) |
1da177e4c
|
2326 2327 2328 |
return -EINVAL; /* Find the first overlapping VMA */ |
9be34c9d5
|
2329 |
vma = find_vma(mm, start); |
146425a31
|
2330 |
if (!vma) |
1da177e4c
|
2331 |
return 0; |
9be34c9d5
|
2332 |
prev = vma->vm_prev; |
146425a31
|
2333 |
/* we have start < vma->vm_end */ |
1da177e4c
|
2334 2335 2336 |
/* if it doesn't overlap, we have nothing.. */ end = start + len; |
146425a31
|
2337 |
if (vma->vm_start >= end) |
1da177e4c
|
2338 2339 2340 2341 2342 2343 2344 2345 2346 |
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. */ |
146425a31
|
2347 |
if (start > vma->vm_start) { |
659ace584
|
2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 |
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); |
1da177e4c
|
2359 2360 |
if (error) return error; |
146425a31
|
2361 |
prev = vma; |
1da177e4c
|
2362 2363 2364 2365 2366 |
} /* Does it split the last one? */ last = find_vma(mm, end); if (last && end > last->vm_start) { |
659ace584
|
2367 |
int error = __split_vma(mm, last, end, 1); |
1da177e4c
|
2368 2369 2370 |
if (error) return error; } |
cc71aba34
|
2371 |
vma = prev ? prev->vm_next : mm->mmap; |
1da177e4c
|
2372 2373 |
/* |
ba470de43
|
2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 |
* 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; } } /* |
1da177e4c
|
2388 2389 |
* Remove the vma's, and unmap the actual pages */ |
146425a31
|
2390 2391 |
detach_vmas_to_be_unmapped(mm, vma, prev, end); unmap_region(mm, vma, prev, start, end); |
1da177e4c
|
2392 |
|
1de4fa14e
|
2393 |
arch_unmap(mm, vma, start, end); |
1da177e4c
|
2394 |
/* Fix up all other VM information */ |
2c0b38146
|
2395 |
remove_vma_list(mm, vma); |
1da177e4c
|
2396 2397 2398 |
return 0; } |
1da177e4c
|
2399 |
|
bfce281c2
|
2400 |
int vm_munmap(unsigned long start, size_t len) |
1da177e4c
|
2401 2402 |
{ int ret; |
bfce281c2
|
2403 |
struct mm_struct *mm = current->mm; |
1da177e4c
|
2404 2405 |
down_write(&mm->mmap_sem); |
a46ef99d8
|
2406 |
ret = do_munmap(mm, start, len); |
1da177e4c
|
2407 2408 2409 |
up_write(&mm->mmap_sem); return ret; } |
a46ef99d8
|
2410 2411 2412 2413 2414 |
EXPORT_SYMBOL(vm_munmap); SYSCALL_DEFINE2(munmap, unsigned long, addr, size_t, len) { profile_munmap(addr); |
bfce281c2
|
2415 |
return vm_munmap(addr, len); |
a46ef99d8
|
2416 |
} |
1da177e4c
|
2417 |
|
c8d78c182
|
2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 |
/* * Emulation of deprecated remap_file_pages() syscall. */ SYSCALL_DEFINE5(remap_file_pages, unsigned long, start, unsigned long, size, unsigned long, prot, unsigned long, pgoff, unsigned long, flags) { struct mm_struct *mm = current->mm; struct vm_area_struct *vma; unsigned long populate = 0; unsigned long ret = -EINVAL; struct file *file; |
756a025f0
|
2431 2432 2433 |
pr_warn_once("%s (%d) uses deprecated remap_file_pages() syscall. See Documentation/vm/remap_file_pages.txt. ", current->comm, current->pid); |
c8d78c182
|
2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 |
if (prot) return ret; start = start & PAGE_MASK; size = size & PAGE_MASK; if (start + size <= start) return ret; /* Does pgoff wrap? */ if (pgoff + (size >> PAGE_SHIFT) < pgoff) return ret; down_write(&mm->mmap_sem); vma = find_vma(mm, start); if (!vma || !(vma->vm_flags & VM_SHARED)) goto out; |
48f7df329
|
2452 |
if (start < vma->vm_start) |
c8d78c182
|
2453 |
goto out; |
48f7df329
|
2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 |
if (start + size > vma->vm_end) { struct vm_area_struct *next; for (next = vma->vm_next; next; next = next->vm_next) { /* hole between vmas ? */ if (next->vm_start != next->vm_prev->vm_end) goto out; if (next->vm_file != vma->vm_file) goto out; if (next->vm_flags != vma->vm_flags) goto out; if (start + size <= next->vm_end) break; } if (!next) goto out; |
c8d78c182
|
2474 2475 2476 2477 2478 2479 2480 2481 2482 |
} prot |= vma->vm_flags & VM_READ ? PROT_READ : 0; prot |= vma->vm_flags & VM_WRITE ? PROT_WRITE : 0; prot |= vma->vm_flags & VM_EXEC ? PROT_EXEC : 0; flags &= MAP_NONBLOCK; flags |= MAP_SHARED | MAP_FIXED | MAP_POPULATE; if (vma->vm_flags & VM_LOCKED) { |
48f7df329
|
2483 |
struct vm_area_struct *tmp; |
c8d78c182
|
2484 |
flags |= MAP_LOCKED; |
48f7df329
|
2485 |
|
c8d78c182
|
2486 |
/* drop PG_Mlocked flag for over-mapped range */ |
48f7df329
|
2487 2488 2489 2490 2491 2492 |
for (tmp = vma; tmp->vm_start >= start + size; tmp = tmp->vm_next) { munlock_vma_pages_range(tmp, max(tmp->vm_start, start), min(tmp->vm_end, start + size)); } |
c8d78c182
|
2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 |
} file = get_file(vma->vm_file); ret = do_mmap_pgoff(vma->vm_file, start, size, prot, flags, pgoff, &populate); fput(file); out: up_write(&mm->mmap_sem); if (populate) mm_populate(ret, populate); if (!IS_ERR_VALUE(ret)) ret = 0; return ret; } |
1da177e4c
|
2507 2508 |
static inline void verify_mm_writelocked(struct mm_struct *mm) { |
a241ec65a
|
2509 |
#ifdef CONFIG_DEBUG_VM |
1da177e4c
|
2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 |
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. */ |
e4eb1ff61
|
2522 |
static unsigned long do_brk(unsigned long addr, unsigned long len) |
1da177e4c
|
2523 |
{ |
cc71aba34
|
2524 2525 |
struct mm_struct *mm = current->mm; struct vm_area_struct *vma, *prev; |
1da177e4c
|
2526 |
unsigned long flags; |
cc71aba34
|
2527 |
struct rb_node **rb_link, *rb_parent; |
1da177e4c
|
2528 |
pgoff_t pgoff = addr >> PAGE_SHIFT; |
3a4597568
|
2529 |
int error; |
1da177e4c
|
2530 2531 2532 2533 |
len = PAGE_ALIGN(len); if (!len) return addr; |
3a4597568
|
2534 |
flags = VM_DATA_DEFAULT_FLAGS | VM_ACCOUNT | mm->def_flags; |
2c6a10161
|
2535 |
error = get_unmapped_area(NULL, addr, len, 0, MAP_FIXED); |
de1741a13
|
2536 |
if (offset_in_page(error)) |
3a4597568
|
2537 |
return error; |
363ee17f0
|
2538 2539 2540 |
error = mlock_future_check(mm, mm->def_flags, len); if (error) return error; |
1da177e4c
|
2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 |
/* * 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 */ |
9fcd14571
|
2551 2552 |
while (find_vma_links(mm, addr, addr + len, &prev, &rb_link, &rb_parent)) { |
1da177e4c
|
2553 2554 |
if (do_munmap(mm, addr, len)) return -ENOMEM; |
1da177e4c
|
2555 2556 2557 |
} /* Check against address space limits *after* clearing old maps... */ |
846383359
|
2558 |
if (!may_expand_vm(mm, flags, len >> PAGE_SHIFT)) |
1da177e4c
|
2559 2560 2561 2562 |
return -ENOMEM; if (mm->map_count > sysctl_max_map_count) return -ENOMEM; |
191c54244
|
2563 |
if (security_vm_enough_memory_mm(mm, len >> PAGE_SHIFT)) |
1da177e4c
|
2564 |
return -ENOMEM; |
1da177e4c
|
2565 |
/* Can we just expand an old private anonymous mapping? */ |
ba470de43
|
2566 |
vma = vma_merge(mm, prev, addr, addr + len, flags, |
19a809afe
|
2567 |
NULL, NULL, pgoff, NULL, NULL_VM_UFFD_CTX); |
ba470de43
|
2568 |
if (vma) |
1da177e4c
|
2569 2570 2571 2572 2573 |
goto out; /* * create a vma struct for an anonymous mapping */ |
c5e3b83e9
|
2574 |
vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); |
1da177e4c
|
2575 2576 2577 2578 |
if (!vma) { vm_unacct_memory(len >> PAGE_SHIFT); return -ENOMEM; } |
1da177e4c
|
2579 |
|
5beb49305
|
2580 |
INIT_LIST_HEAD(&vma->anon_vma_chain); |
1da177e4c
|
2581 2582 2583 2584 2585 |
vma->vm_mm = mm; vma->vm_start = addr; vma->vm_end = addr + len; vma->vm_pgoff = pgoff; vma->vm_flags = flags; |
3ed75eb8f
|
2586 |
vma->vm_page_prot = vm_get_page_prot(flags); |
1da177e4c
|
2587 2588 |
vma_link(mm, vma, prev, rb_link, rb_parent); out: |
3af9e8592
|
2589 |
perf_event_mmap(vma); |
1da177e4c
|
2590 |
mm->total_vm += len >> PAGE_SHIFT; |
846383359
|
2591 |
mm->data_vm += len >> PAGE_SHIFT; |
128557ffe
|
2592 2593 |
if (flags & VM_LOCKED) mm->locked_vm += (len >> PAGE_SHIFT); |
d9104d1ca
|
2594 |
vma->vm_flags |= VM_SOFTDIRTY; |
1da177e4c
|
2595 2596 |
return addr; } |
e4eb1ff61
|
2597 2598 2599 2600 |
unsigned long vm_brk(unsigned long addr, unsigned long len) { struct mm_struct *mm = current->mm; unsigned long ret; |
128557ffe
|
2601 |
bool populate; |
e4eb1ff61
|
2602 2603 2604 |
down_write(&mm->mmap_sem); ret = do_brk(addr, len); |
128557ffe
|
2605 |
populate = ((mm->def_flags & VM_LOCKED) != 0); |
e4eb1ff61
|
2606 |
up_write(&mm->mmap_sem); |
128557ffe
|
2607 2608 |
if (populate) mm_populate(addr, len); |
e4eb1ff61
|
2609 2610 2611 |
return ret; } EXPORT_SYMBOL(vm_brk); |
1da177e4c
|
2612 2613 2614 2615 |
/* Release all mmaps. */ void exit_mmap(struct mm_struct *mm) { |
d16dfc550
|
2616 |
struct mmu_gather tlb; |
ba470de43
|
2617 |
struct vm_area_struct *vma; |
1da177e4c
|
2618 |
unsigned long nr_accounted = 0; |
d6dd61c83
|
2619 |
/* mm's last user has gone, and its about to be pulled down */ |
cddb8a5c1
|
2620 |
mmu_notifier_release(mm); |
d6dd61c83
|
2621 |
|
ba470de43
|
2622 2623 2624 2625 2626 2627 2628 2629 |
if (mm->locked_vm) { vma = mm->mmap; while (vma) { if (vma->vm_flags & VM_LOCKED) munlock_vma_pages_all(vma); vma = vma->vm_next; } } |
9480c53e9
|
2630 2631 |
arch_exit_mmap(mm); |
ba470de43
|
2632 |
vma = mm->mmap; |
9480c53e9
|
2633 2634 |
if (!vma) /* Can happen if dup_mmap() received an OOM */ return; |
1da177e4c
|
2635 |
lru_add_drain(); |
1da177e4c
|
2636 |
flush_cache_mm(mm); |
2b047252d
|
2637 |
tlb_gather_mmu(&tlb, mm, 0, -1); |
901608d90
|
2638 |
/* update_hiwater_rss(mm) here? but nobody should be looking */ |
e0da382c9
|
2639 |
/* Use -1 here to ensure all VMAs in the mm are unmapped */ |
4f74d2c8e
|
2640 |
unmap_vmas(&tlb, vma, 0, -1); |
9ba692948
|
2641 |
|
6ee8630e0
|
2642 |
free_pgtables(&tlb, vma, FIRST_USER_ADDRESS, USER_PGTABLES_CEILING); |
853f5e264
|
2643 |
tlb_finish_mmu(&tlb, 0, -1); |
1da177e4c
|
2644 |
|
1da177e4c
|
2645 |
/* |
8f4f8c164
|
2646 2647 |
* Walk the list again, actually closing and freeing it, * with preemption enabled, without holding any MM locks. |
1da177e4c
|
2648 |
*/ |
4f74d2c8e
|
2649 2650 2651 |
while (vma) { if (vma->vm_flags & VM_ACCOUNT) nr_accounted += vma_pages(vma); |
a8fb5618d
|
2652 |
vma = remove_vma(vma); |
4f74d2c8e
|
2653 2654 |
} vm_unacct_memory(nr_accounted); |
1da177e4c
|
2655 2656 2657 2658 |
} /* Insert vm structure into process list sorted by address * and into the inode's i_mmap tree. If vm_file is non-NULL |
c8c06efa8
|
2659 |
* then i_mmap_rwsem is taken here. |
1da177e4c
|
2660 |
*/ |
6597d7833
|
2661 |
int insert_vm_struct(struct mm_struct *mm, struct vm_area_struct *vma) |
1da177e4c
|
2662 |
{ |
6597d7833
|
2663 2664 |
struct vm_area_struct *prev; struct rb_node **rb_link, *rb_parent; |
1da177e4c
|
2665 |
|
c9d13f5fc
|
2666 2667 2668 2669 2670 2671 |
if (find_vma_links(mm, vma->vm_start, vma->vm_end, &prev, &rb_link, &rb_parent)) return -ENOMEM; if ((vma->vm_flags & VM_ACCOUNT) && security_vm_enough_memory_mm(mm, vma_pages(vma))) return -ENOMEM; |
1da177e4c
|
2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 |
/* * 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. */ |
8a9cc3b55
|
2684 |
if (vma_is_anonymous(vma)) { |
1da177e4c
|
2685 2686 2687 |
BUG_ON(vma->anon_vma); vma->vm_pgoff = vma->vm_start >> PAGE_SHIFT; } |
2b1444983
|
2688 |
|
1da177e4c
|
2689 2690 2691 2692 2693 2694 2695 2696 2697 |
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, |
38a76013a
|
2698 2699 |
unsigned long addr, unsigned long len, pgoff_t pgoff, bool *need_rmap_locks) |
1da177e4c
|
2700 2701 2702 2703 2704 2705 |
{ 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; |
948f017b0
|
2706 |
bool faulted_in_anon_vma = true; |
1da177e4c
|
2707 2708 2709 2710 2711 |
/* * If anonymous vma has not yet been faulted, update new pgoff * to match new location, to increase its chance of merging. */ |
ce75799b8
|
2712 |
if (unlikely(vma_is_anonymous(vma) && !vma->anon_vma)) { |
1da177e4c
|
2713 |
pgoff = addr >> PAGE_SHIFT; |
948f017b0
|
2714 2715 |
faulted_in_anon_vma = false; } |
1da177e4c
|
2716 |
|
6597d7833
|
2717 2718 |
if (find_vma_links(mm, addr, addr + len, &prev, &rb_link, &rb_parent)) return NULL; /* should never get here */ |
1da177e4c
|
2719 |
new_vma = vma_merge(mm, prev, addr, addr + len, vma->vm_flags, |
19a809afe
|
2720 2721 |
vma->anon_vma, vma->vm_file, pgoff, vma_policy(vma), vma->vm_userfaultfd_ctx); |
1da177e4c
|
2722 2723 2724 2725 |
if (new_vma) { /* * Source vma may have been merged into new_vma */ |
948f017b0
|
2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 |
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. */ |
81d1b09c6
|
2740 |
VM_BUG_ON_VMA(faulted_in_anon_vma, new_vma); |
38a76013a
|
2741 |
*vmap = vma = new_vma; |
108d6642a
|
2742 |
} |
38a76013a
|
2743 |
*need_rmap_locks = (new_vma->vm_pgoff <= vma->vm_pgoff); |
1da177e4c
|
2744 |
} else { |
e94b17660
|
2745 |
new_vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); |
e39758912
|
2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 |
if (!new_vma) goto out; *new_vma = *vma; new_vma->vm_start = addr; new_vma->vm_end = addr + len; new_vma->vm_pgoff = pgoff; if (vma_dup_policy(vma, new_vma)) goto out_free_vma; INIT_LIST_HEAD(&new_vma->anon_vma_chain); if (anon_vma_clone(new_vma, vma)) goto out_free_mempol; if (new_vma->vm_file) get_file(new_vma->vm_file); 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); *need_rmap_locks = false; |
1da177e4c
|
2763 2764 |
} return new_vma; |
5beb49305
|
2765 |
|
e39758912
|
2766 |
out_free_mempol: |
ef0855d33
|
2767 |
mpol_put(vma_policy(new_vma)); |
e39758912
|
2768 |
out_free_vma: |
5beb49305
|
2769 |
kmem_cache_free(vm_area_cachep, new_vma); |
e39758912
|
2770 |
out: |
5beb49305
|
2771 |
return NULL; |
1da177e4c
|
2772 |
} |
119f657c7
|
2773 2774 2775 2776 2777 |
/* * Return true if the calling process may expand its vm space by the passed * number of pages */ |
846383359
|
2778 |
bool may_expand_vm(struct mm_struct *mm, vm_flags_t flags, unsigned long npages) |
119f657c7
|
2779 |
{ |
846383359
|
2780 2781 |
if (mm->total_vm + npages > rlimit(RLIMIT_AS) >> PAGE_SHIFT) return false; |
119f657c7
|
2782 |
|
d977d56ce
|
2783 2784 2785 |
if (is_data_mapping(flags) && mm->data_vm + npages > rlimit(RLIMIT_DATA) >> PAGE_SHIFT) { if (ignore_rlimit_data) |
756a025f0
|
2786 2787 |
pr_warn_once("%s (%d): VmData %lu exceed data ulimit %lu. Will be forbidden soon. ", |
d977d56ce
|
2788 2789 2790 2791 2792 2793 |
current->comm, current->pid, (mm->data_vm + npages) << PAGE_SHIFT, rlimit(RLIMIT_DATA)); else return false; } |
119f657c7
|
2794 |
|
846383359
|
2795 2796 2797 2798 2799 2800 |
return true; } void vm_stat_account(struct mm_struct *mm, vm_flags_t flags, long npages) { mm->total_vm += npages; |
d977d56ce
|
2801 |
if (is_exec_mapping(flags)) |
846383359
|
2802 |
mm->exec_vm += npages; |
d977d56ce
|
2803 |
else if (is_stack_mapping(flags)) |
846383359
|
2804 |
mm->stack_vm += npages; |
d977d56ce
|
2805 |
else if (is_data_mapping(flags)) |
846383359
|
2806 |
mm->data_vm += npages; |
119f657c7
|
2807 |
} |
fa5dc22f8
|
2808 |
|
a62c34bd2
|
2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 |
static int special_mapping_fault(struct vm_area_struct *vma, struct vm_fault *vmf); /* * Having a close hook prevents vma merging regardless of flags. */ static void special_mapping_close(struct vm_area_struct *vma) { } static const char *special_mapping_name(struct vm_area_struct *vma) { return ((struct vm_special_mapping *)vma->vm_private_data)->name; } static const struct vm_operations_struct special_mapping_vmops = { .close = special_mapping_close, .fault = special_mapping_fault, .name = special_mapping_name, }; static const struct vm_operations_struct legacy_special_mapping_vmops = { .close = special_mapping_close, .fault = special_mapping_fault, }; |
fa5dc22f8
|
2834 |
|
b1d0e4f53
|
2835 2836 |
static int special_mapping_fault(struct vm_area_struct *vma, struct vm_fault *vmf) |
fa5dc22f8
|
2837 |
{ |
b1d0e4f53
|
2838 |
pgoff_t pgoff; |
fa5dc22f8
|
2839 |
struct page **pages; |
f872f5400
|
2840 |
if (vma->vm_ops == &legacy_special_mapping_vmops) { |
a62c34bd2
|
2841 |
pages = vma->vm_private_data; |
f872f5400
|
2842 2843 2844 2845 2846 2847 2848 2849 |
} else { struct vm_special_mapping *sm = vma->vm_private_data; if (sm->fault) return sm->fault(sm, vma, vmf); pages = sm->pages; } |
a62c34bd2
|
2850 |
|
8a9cc3b55
|
2851 |
for (pgoff = vmf->pgoff; pgoff && *pages; ++pages) |
b1d0e4f53
|
2852 |
pgoff--; |
fa5dc22f8
|
2853 2854 2855 2856 |
if (*pages) { struct page *page = *pages; get_page(page); |
b1d0e4f53
|
2857 2858 |
vmf->page = page; return 0; |
fa5dc22f8
|
2859 |
} |
b1d0e4f53
|
2860 |
return VM_FAULT_SIGBUS; |
fa5dc22f8
|
2861 |
} |
a62c34bd2
|
2862 2863 2864 |
static struct vm_area_struct *__install_special_mapping( struct mm_struct *mm, unsigned long addr, unsigned long len, |
27f28b972
|
2865 2866 |
unsigned long vm_flags, void *priv, const struct vm_operations_struct *ops) |
fa5dc22f8
|
2867 |
{ |
462e635e5
|
2868 |
int ret; |
fa5dc22f8
|
2869 2870 2871 2872 |
struct vm_area_struct *vma; vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); if (unlikely(vma == NULL)) |
3935ed6a3
|
2873 |
return ERR_PTR(-ENOMEM); |
fa5dc22f8
|
2874 |
|
5beb49305
|
2875 |
INIT_LIST_HEAD(&vma->anon_vma_chain); |
fa5dc22f8
|
2876 2877 2878 |
vma->vm_mm = mm; vma->vm_start = addr; vma->vm_end = addr + len; |
d9104d1ca
|
2879 |
vma->vm_flags = vm_flags | mm->def_flags | VM_DONTEXPAND | VM_SOFTDIRTY; |
3ed75eb8f
|
2880 |
vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); |
fa5dc22f8
|
2881 |
|
a62c34bd2
|
2882 2883 |
vma->vm_ops = ops; vma->vm_private_data = priv; |
fa5dc22f8
|
2884 |
|
462e635e5
|
2885 2886 2887 |
ret = insert_vm_struct(mm, vma); if (ret) goto out; |
fa5dc22f8
|
2888 |
|
846383359
|
2889 |
vm_stat_account(mm, vma->vm_flags, len >> PAGE_SHIFT); |
fa5dc22f8
|
2890 |
|
cdd6c482c
|
2891 |
perf_event_mmap(vma); |
089dd79db
|
2892 |
|
3935ed6a3
|
2893 |
return vma; |
462e635e5
|
2894 2895 2896 |
out: kmem_cache_free(vm_area_cachep, vma); |
3935ed6a3
|
2897 2898 |
return ERR_PTR(ret); } |
a62c34bd2
|
2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 |
/* * 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. */ struct vm_area_struct *_install_special_mapping( struct mm_struct *mm, unsigned long addr, unsigned long len, unsigned long vm_flags, const struct vm_special_mapping *spec) { |
27f28b972
|
2913 2914 |
return __install_special_mapping(mm, addr, len, vm_flags, (void *)spec, &special_mapping_vmops); |
a62c34bd2
|
2915 |
} |
3935ed6a3
|
2916 2917 2918 2919 |
int install_special_mapping(struct mm_struct *mm, unsigned long addr, unsigned long len, unsigned long vm_flags, struct page **pages) { |
a62c34bd2
|
2920 |
struct vm_area_struct *vma = __install_special_mapping( |
27f28b972
|
2921 2922 |
mm, addr, len, vm_flags, (void *)pages, &legacy_special_mapping_vmops); |
3935ed6a3
|
2923 |
|
14bd5b458
|
2924 |
return PTR_ERR_OR_ZERO(vma); |
fa5dc22f8
|
2925 |
} |
7906d00cd
|
2926 2927 |
static DEFINE_MUTEX(mm_all_locks_mutex); |
454ed842d
|
2928 |
static void vm_lock_anon_vma(struct mm_struct *mm, struct anon_vma *anon_vma) |
7906d00cd
|
2929 |
{ |
bf181b9f9
|
2930 |
if (!test_bit(0, (unsigned long *) &anon_vma->root->rb_root.rb_node)) { |
7906d00cd
|
2931 2932 2933 2934 |
/* * The LSB of head.next can't change from under us * because we hold the mm_all_locks_mutex. */ |
572043c90
|
2935 |
down_write_nest_lock(&anon_vma->root->rwsem, &mm->mmap_sem); |
7906d00cd
|
2936 2937 |
/* * We can safely modify head.next after taking the |
5a505085f
|
2938 |
* anon_vma->root->rwsem. If some other vma in this mm shares |
7906d00cd
|
2939 2940 2941 2942 |
* 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 |
5a505085f
|
2943 |
* anon_vma->root->rwsem. |
7906d00cd
|
2944 2945 |
*/ if (__test_and_set_bit(0, (unsigned long *) |
bf181b9f9
|
2946 |
&anon_vma->root->rb_root.rb_node)) |
7906d00cd
|
2947 2948 2949 |
BUG(); } } |
454ed842d
|
2950 |
static void vm_lock_mapping(struct mm_struct *mm, struct address_space *mapping) |
7906d00cd
|
2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 |
{ 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(); |
c8c06efa8
|
2964 |
down_write_nest_lock(&mapping->i_mmap_rwsem, &mm->mmap_sem); |
7906d00cd
|
2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 |
} } /* * 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 |
27ba0644e
|
2979 |
* altering the vma layout. It's also needed in write mode to avoid new |
7906d00cd
|
2980 2981 2982 2983 2984 |
* 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. * |
bf181b9f9
|
2985 |
* The LSB in anon_vma->rb_root.rb_node and the AS_MM_ALL_LOCKS bitflag in |
7906d00cd
|
2986 2987 2988 |
* 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. * |
88f306b68
|
2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 |
* We take locks in following order, accordingly to comment at beginning * of mm/rmap.c: * - all hugetlbfs_i_mmap_rwsem_key locks (aka mapping->i_mmap_rwsem for * hugetlb mapping); * - all i_mmap_rwsem locks; * - all anon_vma->rwseml * * We can take all locks within these types randomly because the VM code * doesn't nest them and we protected from parallel mm_take_all_locks() by * mm_all_locks_mutex. |
7906d00cd
|
2999 3000 3001 3002 3003 3004 3005 3006 3007 |
* * 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; |
5beb49305
|
3008 |
struct anon_vma_chain *avc; |
7906d00cd
|
3009 3010 3011 3012 3013 3014 3015 3016 |
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; |
88f306b68
|
3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 |
if (vma->vm_file && vma->vm_file->f_mapping && is_vm_hugetlb_page(vma)) vm_lock_mapping(mm, vma->vm_file->f_mapping); } 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 && !is_vm_hugetlb_page(vma)) |
454ed842d
|
3027 |
vm_lock_mapping(mm, vma->vm_file->f_mapping); |
7906d00cd
|
3028 |
} |
7cd5a02f5
|
3029 3030 3031 3032 3033 |
for (vma = mm->mmap; vma; vma = vma->vm_next) { if (signal_pending(current)) goto out_unlock; if (vma->anon_vma) |
5beb49305
|
3034 3035 |
list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) vm_lock_anon_vma(mm, avc->anon_vma); |
7906d00cd
|
3036 |
} |
7cd5a02f5
|
3037 |
|
584cff54e
|
3038 |
return 0; |
7906d00cd
|
3039 3040 |
out_unlock: |
584cff54e
|
3041 3042 |
mm_drop_all_locks(mm); return -EINTR; |
7906d00cd
|
3043 3044 3045 3046 |
} static void vm_unlock_anon_vma(struct anon_vma *anon_vma) { |
bf181b9f9
|
3047 |
if (test_bit(0, (unsigned long *) &anon_vma->root->rb_root.rb_node)) { |
7906d00cd
|
3048 3049 3050 3051 3052 |
/* * 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 |
bf181b9f9
|
3053 |
* the vma so the users using the anon_vma->rb_root will |
7906d00cd
|
3054 3055 3056 3057 |
* never see our bitflag. * * No need of atomic instructions here, head.next * can't change from under us until we release the |
5a505085f
|
3058 |
* anon_vma->root->rwsem. |
7906d00cd
|
3059 3060 |
*/ if (!__test_and_clear_bit(0, (unsigned long *) |
bf181b9f9
|
3061 |
&anon_vma->root->rb_root.rb_node)) |
7906d00cd
|
3062 |
BUG(); |
08b52706d
|
3063 |
anon_vma_unlock_write(anon_vma); |
7906d00cd
|
3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 |
} } 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. */ |
83cde9e8b
|
3074 |
i_mmap_unlock_write(mapping); |
7906d00cd
|
3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 |
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; |
5beb49305
|
3088 |
struct anon_vma_chain *avc; |
7906d00cd
|
3089 3090 3091 3092 3093 3094 |
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) |
5beb49305
|
3095 3096 |
list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) vm_unlock_anon_vma(avc->anon_vma); |
7906d00cd
|
3097 3098 3099 3100 3101 3102 |
if (vma->vm_file && vma->vm_file->f_mapping) vm_unlock_mapping(vma->vm_file->f_mapping); } mutex_unlock(&mm_all_locks_mutex); } |
8feae1311
|
3103 3104 3105 3106 3107 3108 |
/* * initialise the VMA slab */ void __init mmap_init(void) { |
00a62ce91
|
3109 |
int ret; |
908c7f194
|
3110 |
ret = percpu_counter_init(&vm_committed_as, 0, GFP_KERNEL); |
00a62ce91
|
3111 |
VM_BUG_ON(ret); |
8feae1311
|
3112 |
} |
c9b1d0981
|
3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 |
/* * Initialise sysctl_user_reserve_kbytes. * * This is intended to prevent a user from starting a single memory hogging * process, such that they cannot recover (kill the hog) in OVERCOMMIT_NEVER * mode. * * The default value is min(3% of free memory, 128MB) * 128MB is enough to recover with sshd/login, bash, and top/kill. */ |
1640879af
|
3124 |
static int init_user_reserve(void) |
c9b1d0981
|
3125 3126 3127 3128 3129 3130 3131 3132 |
{ unsigned long free_kbytes; free_kbytes = global_page_state(NR_FREE_PAGES) << (PAGE_SHIFT - 10); sysctl_user_reserve_kbytes = min(free_kbytes / 32, 1UL << 17); return 0; } |
a64fb3cd6
|
3133 |
subsys_initcall(init_user_reserve); |
4eeab4f55
|
3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 |
/* * Initialise sysctl_admin_reserve_kbytes. * * The purpose of sysctl_admin_reserve_kbytes is to allow the sys admin * to log in and kill a memory hogging process. * * Systems with more than 256MB will reserve 8MB, enough to recover * with sshd, bash, and top in OVERCOMMIT_GUESS. Smaller systems will * only reserve 3% of free pages by default. */ |
1640879af
|
3145 |
static int init_admin_reserve(void) |
4eeab4f55
|
3146 3147 3148 3149 3150 3151 3152 3153 |
{ unsigned long free_kbytes; free_kbytes = global_page_state(NR_FREE_PAGES) << (PAGE_SHIFT - 10); sysctl_admin_reserve_kbytes = min(free_kbytes / 32, 1UL << 13); return 0; } |
a64fb3cd6
|
3154 |
subsys_initcall(init_admin_reserve); |
1640879af
|
3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 |
/* * Reinititalise user and admin reserves if memory is added or removed. * * The default user reserve max is 128MB, and the default max for the * admin reserve is 8MB. These are usually, but not always, enough to * enable recovery from a memory hogging process using login/sshd, a shell, * and tools like top. It may make sense to increase or even disable the * reserve depending on the existence of swap or variations in the recovery * tools. So, the admin may have changed them. * * If memory is added and the reserves have been eliminated or increased above * the default max, then we'll trust the admin. * * If memory is removed and there isn't enough free memory, then we * need to reset the reserves. * * Otherwise keep the reserve set by the admin. */ static int reserve_mem_notifier(struct notifier_block *nb, unsigned long action, void *data) { unsigned long tmp, free_kbytes; switch (action) { case MEM_ONLINE: /* Default max is 128MB. Leave alone if modified by operator. */ tmp = sysctl_user_reserve_kbytes; if (0 < tmp && tmp < (1UL << 17)) init_user_reserve(); /* Default max is 8MB. Leave alone if modified by operator. */ tmp = sysctl_admin_reserve_kbytes; if (0 < tmp && tmp < (1UL << 13)) init_admin_reserve(); break; case MEM_OFFLINE: free_kbytes = global_page_state(NR_FREE_PAGES) << (PAGE_SHIFT - 10); if (sysctl_user_reserve_kbytes > free_kbytes) { init_user_reserve(); pr_info("vm.user_reserve_kbytes reset to %lu ", sysctl_user_reserve_kbytes); } if (sysctl_admin_reserve_kbytes > free_kbytes) { init_admin_reserve(); pr_info("vm.admin_reserve_kbytes reset to %lu ", sysctl_admin_reserve_kbytes); } break; default: break; } return NOTIFY_OK; } static struct notifier_block reserve_mem_nb = { .notifier_call = reserve_mem_notifier, }; static int __meminit init_reserve_notifier(void) { if (register_hotmemory_notifier(&reserve_mem_nb)) |
b1de0d139
|
3222 3223 |
pr_err("Failed registering memory add/remove notifier for admin reserve "); |
1640879af
|
3224 3225 3226 |
return 0; } |
a64fb3cd6
|
3227 |
subsys_initcall(init_reserve_notifier); |