Blame view
mm/mmap.c
86.6 KB
1da177e4c Linux-2.6.12-rc2 |
1 2 3 4 5 |
/* * mm/mmap.c * * Written by obz. * |
046c68842 mm: update my add... |
6 |
* Address space accounting code <alan@lxorguk.ukuu.org.uk> |
1da177e4c Linux-2.6.12-rc2 |
7 |
*/ |
e8420a8ec mm/mmap: check fo... |
8 |
#include <linux/kernel.h> |
1da177e4c Linux-2.6.12-rc2 |
9 |
#include <linux/slab.h> |
4af3c9cc4 Drop some headers... |
10 |
#include <linux/backing-dev.h> |
1da177e4c Linux-2.6.12-rc2 |
11 12 13 14 15 16 |
#include <linux/mm.h> #include <linux/shm.h> #include <linux/mman.h> #include <linux/pagemap.h> #include <linux/swap.h> #include <linux/syscalls.h> |
c59ede7b7 [PATCH] move capa... |
17 |
#include <linux/capability.h> |
1da177e4c Linux-2.6.12-rc2 |
18 19 20 21 22 23 24 |
#include <linux/init.h> #include <linux/file.h> #include <linux/fs.h> #include <linux/personality.h> #include <linux/security.h> #include <linux/hugetlb.h> #include <linux/profile.h> |
b95f1b31b mm: Map most file... |
25 |
#include <linux/export.h> |
1da177e4c Linux-2.6.12-rc2 |
26 27 28 |
#include <linux/mount.h> #include <linux/mempolicy.h> #include <linux/rmap.h> |
cddb8a5c1 mmu-notifiers: core |
29 |
#include <linux/mmu_notifier.h> |
cdd6c482c perf: Do the big ... |
30 |
#include <linux/perf_event.h> |
120a795da audit mmap |
31 |
#include <linux/audit.h> |
b15d00b6a thp: khugepaged v... |
32 |
#include <linux/khugepaged.h> |
2b1444983 uprobes, mm, x86:... |
33 |
#include <linux/uprobes.h> |
d37371870 mm: augment vma r... |
34 |
#include <linux/rbtree_augmented.h> |
cf4aebc29 sched: Move sched... |
35 |
#include <linux/sched/sysctl.h> |
1640879af mm: reinititalise... |
36 37 |
#include <linux/notifier.h> #include <linux/memory.h> |
1da177e4c Linux-2.6.12-rc2 |
38 39 40 41 |
#include <asm/uaccess.h> #include <asm/cacheflush.h> #include <asm/tlb.h> |
d6dd61c83 [PATCH] x86: PARA... |
42 |
#include <asm/mmu_context.h> |
1da177e4c Linux-2.6.12-rc2 |
43 |
|
42b777281 mm: remove double... |
44 |
#include "internal.h" |
3a4597568 [PATCH] IA64,spar... |
45 46 47 |
#ifndef arch_mmap_check #define arch_mmap_check(addr, len, flags) (0) #endif |
08e7d9b55 arch_rebalance_pg... |
48 49 50 |
#ifndef arch_rebalance_pgtables #define arch_rebalance_pgtables(addr, len) (addr) #endif |
e0da382c9 [PATCH] freepgt: ... |
51 52 53 |
static void unmap_region(struct mm_struct *mm, struct vm_area_struct *vma, struct vm_area_struct *prev, unsigned long start, unsigned long end); |
1da177e4c Linux-2.6.12-rc2 |
54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 |
/* description of effects of mapping type and prot in current implementation. * this is due to the limited x86 page protection hardware. The expected * behavior is in parens: * * map_type prot * PROT_NONE PROT_READ PROT_WRITE PROT_EXEC * MAP_SHARED r: (no) no r: (yes) yes r: (no) yes r: (no) yes * w: (no) no w: (no) no w: (yes) yes w: (no) no * x: (no) no x: (no) yes x: (no) yes x: (yes) yes * * MAP_PRIVATE r: (no) no r: (yes) yes r: (no) yes r: (no) yes * w: (no) no w: (no) no w: (copy) copy w: (no) no * x: (no) no x: (no) yes x: (no) yes x: (yes) yes * */ pgprot_t protection_map[16] = { __P000, __P001, __P010, __P011, __P100, __P101, __P110, __P111, __S000, __S001, __S010, __S011, __S100, __S101, __S110, __S111 }; |
804af2cf6 [AGPGART] remove ... |
73 74 |
pgprot_t vm_get_page_prot(unsigned long vm_flags) { |
b845f313d mm: Allow archite... |
75 76 77 |
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 [AGPGART] remove ... |
78 79 |
} EXPORT_SYMBOL(vm_get_page_prot); |
34679d7ea mmap: add alignme... |
80 81 |
int sysctl_overcommit_memory __read_mostly = OVERCOMMIT_GUESS; /* heuristic overcommit */ int sysctl_overcommit_ratio __read_mostly = 50; /* default is 50% */ |
c3d8c1414 [PATCH] More __re... |
82 |
int sysctl_max_map_count __read_mostly = DEFAULT_MAX_MAP_COUNT; |
c9b1d0981 mm: limit growth ... |
83 |
unsigned long sysctl_user_reserve_kbytes __read_mostly = 1UL << 17; /* 128MB */ |
4eeab4f55 mm: replace hardc... |
84 |
unsigned long sysctl_admin_reserve_kbytes __read_mostly = 1UL << 13; /* 8MB */ |
34679d7ea mmap: add alignme... |
85 86 87 88 89 |
/* * Make sure vm_committed_as in one cacheline and not cacheline shared with * other variables. It can be updated by several CPUs frequently. */ struct percpu_counter vm_committed_as ____cacheline_aligned_in_smp; |
1da177e4c Linux-2.6.12-rc2 |
90 91 |
/* |
997071bcb mm: export a func... |
92 93 94 95 96 97 98 99 100 101 102 103 104 105 |
* The global memory commitment made in the system can be a metric * that can be used to drive ballooning decisions when Linux is hosted * as a guest. On Hyper-V, the host implements a policy engine for dynamically * balancing memory across competing virtual machines that are hosted. * Several metrics drive this policy engine including the guest reported * memory commitment. */ unsigned long vm_memory_committed(void) { return percpu_counter_read_positive(&vm_committed_as); } EXPORT_SYMBOL_GPL(vm_memory_committed); /* |
1da177e4c Linux-2.6.12-rc2 |
106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 |
* Check that a process has enough memory to allocate a new virtual * mapping. 0 means there is enough memory for the allocation to * succeed and -ENOMEM implies there is not. * * We currently support three overcommit policies, which are set via the * vm.overcommit_memory sysctl. See Documentation/vm/overcommit-accounting * * Strict overcommit modes added 2002 Feb 26 by Alan Cox. * Additional code 2002 Jul 20 by Robert Love. * * cap_sys_admin is 1 if the process has admin privileges, 0 otherwise. * * Note this is a helper function intended to be used by LSMs which * wish to use this logic. */ |
34b4e4aa3 fix NULL pointer ... |
121 |
int __vm_enough_memory(struct mm_struct *mm, long pages, int cap_sys_admin) |
1da177e4c Linux-2.6.12-rc2 |
122 |
{ |
c9b1d0981 mm: limit growth ... |
123 |
unsigned long free, allowed, reserve; |
1da177e4c Linux-2.6.12-rc2 |
124 125 126 127 128 129 130 131 132 133 |
vm_acct_memory(pages); /* * Sometimes we want to use more memory than we have */ if (sysctl_overcommit_memory == OVERCOMMIT_ALWAYS) return 0; if (sysctl_overcommit_memory == OVERCOMMIT_GUESS) { |
c15bef309 mmap: fix and tid... |
134 135 136 137 138 139 140 141 142 143 |
free = global_page_state(NR_FREE_PAGES); free += global_page_state(NR_FILE_PAGES); /* * shmem pages shouldn't be counted as free in this * case, they can't be purged, only swapped out, and * that won't affect the overall amount of available * memory in the system. */ free -= global_page_state(NR_SHMEM); |
1da177e4c Linux-2.6.12-rc2 |
144 |
|
ec8acf20a swap: add per-par... |
145 |
free += get_nr_swap_pages(); |
1da177e4c Linux-2.6.12-rc2 |
146 147 148 149 150 151 152 |
/* * Any slabs which are created with the * SLAB_RECLAIM_ACCOUNT flag claim to have contents * which are reclaimable, under pressure. The dentry * cache and most inode caches should fall into this */ |
972d1a7b1 [PATCH] ZVC: Supp... |
153 |
free += global_page_state(NR_SLAB_RECLAIMABLE); |
1da177e4c Linux-2.6.12-rc2 |
154 155 |
/* |
6d9f78396 [PATCH] overcommi... |
156 157 |
* Leave reserved pages. The pages are not for anonymous pages. */ |
c15bef309 mmap: fix and tid... |
158 |
if (free <= totalreserve_pages) |
6d9f78396 [PATCH] overcommi... |
159 160 |
goto error; else |
c15bef309 mmap: fix and tid... |
161 |
free -= totalreserve_pages; |
6d9f78396 [PATCH] overcommi... |
162 163 |
/* |
4eeab4f55 mm: replace hardc... |
164 |
* Reserve some for root |
6d9f78396 [PATCH] overcommi... |
165 |
*/ |
1da177e4c Linux-2.6.12-rc2 |
166 |
if (!cap_sys_admin) |
4eeab4f55 mm: replace hardc... |
167 |
free -= sysctl_admin_reserve_kbytes >> (PAGE_SHIFT - 10); |
1da177e4c Linux-2.6.12-rc2 |
168 169 170 |
if (free > pages) return 0; |
6d9f78396 [PATCH] overcommi... |
171 172 |
goto error; |
1da177e4c Linux-2.6.12-rc2 |
173 174 175 176 177 |
} allowed = (totalram_pages - hugetlb_total_pages()) * sysctl_overcommit_ratio / 100; /* |
4eeab4f55 mm: replace hardc... |
178 |
* Reserve some for root |
1da177e4c Linux-2.6.12-rc2 |
179 180 |
*/ if (!cap_sys_admin) |
4eeab4f55 mm: replace hardc... |
181 |
allowed -= sysctl_admin_reserve_kbytes >> (PAGE_SHIFT - 10); |
1da177e4c Linux-2.6.12-rc2 |
182 |
allowed += total_swap_pages; |
c9b1d0981 mm: limit growth ... |
183 184 185 186 187 188 189 |
/* * Don't let a single process grow so big a user can't recover */ if (mm) { reserve = sysctl_user_reserve_kbytes >> (PAGE_SHIFT - 10); allowed -= min(mm->total_vm / 32, reserve); } |
1da177e4c Linux-2.6.12-rc2 |
190 |
|
00a62ce91 mm: fix Committed... |
191 |
if (percpu_counter_read_positive(&vm_committed_as) < allowed) |
1da177e4c Linux-2.6.12-rc2 |
192 |
return 0; |
6d9f78396 [PATCH] overcommi... |
193 |
error: |
1da177e4c Linux-2.6.12-rc2 |
194 195 196 197 |
vm_unacct_memory(pages); return -ENOMEM; } |
1da177e4c Linux-2.6.12-rc2 |
198 |
/* |
3d48ae45e mm: Convert i_mma... |
199 |
* Requires inode->i_mapping->i_mmap_mutex |
1da177e4c Linux-2.6.12-rc2 |
200 201 202 203 204 |
*/ 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 new helper: file_... |
205 |
atomic_inc(&file_inode(file)->i_writecount); |
1da177e4c Linux-2.6.12-rc2 |
206 207 208 209 210 |
if (vma->vm_flags & VM_SHARED) mapping->i_mmap_writable--; flush_dcache_mmap_lock(mapping); if (unlikely(vma->vm_flags & VM_NONLINEAR)) |
6b2dbba8b mm: replace vma p... |
211 |
list_del_init(&vma->shared.nonlinear); |
1da177e4c Linux-2.6.12-rc2 |
212 |
else |
6b2dbba8b mm: replace vma p... |
213 |
vma_interval_tree_remove(vma, &mapping->i_mmap); |
1da177e4c Linux-2.6.12-rc2 |
214 215 216 217 |
flush_dcache_mmap_unlock(mapping); } /* |
6b2dbba8b mm: replace vma p... |
218 |
* Unlink a file-based vm structure from its interval tree, to hide |
a8fb5618d [PATCH] mm: unlin... |
219 |
* vma from rmap and vmtruncate before freeing its page tables. |
1da177e4c Linux-2.6.12-rc2 |
220 |
*/ |
a8fb5618d [PATCH] mm: unlin... |
221 |
void unlink_file_vma(struct vm_area_struct *vma) |
1da177e4c Linux-2.6.12-rc2 |
222 223 |
{ struct file *file = vma->vm_file; |
1da177e4c Linux-2.6.12-rc2 |
224 225 |
if (file) { struct address_space *mapping = file->f_mapping; |
3d48ae45e mm: Convert i_mma... |
226 |
mutex_lock(&mapping->i_mmap_mutex); |
1da177e4c Linux-2.6.12-rc2 |
227 |
__remove_shared_vm_struct(vma, file, mapping); |
3d48ae45e mm: Convert i_mma... |
228 |
mutex_unlock(&mapping->i_mmap_mutex); |
1da177e4c Linux-2.6.12-rc2 |
229 |
} |
a8fb5618d [PATCH] mm: unlin... |
230 231 232 233 234 235 236 237 |
} /* * 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 [PATCH] mm: unlin... |
238 |
might_sleep(); |
1da177e4c Linux-2.6.12-rc2 |
239 240 |
if (vma->vm_ops && vma->vm_ops->close) vma->vm_ops->close(vma); |
e9714acf8 mm: kill vma flag... |
241 |
if (vma->vm_file) |
a8fb5618d [PATCH] mm: unlin... |
242 |
fput(vma->vm_file); |
f0be3d32b mempolicy: rename... |
243 |
mpol_put(vma_policy(vma)); |
1da177e4c Linux-2.6.12-rc2 |
244 |
kmem_cache_free(vm_area_cachep, vma); |
a8fb5618d [PATCH] mm: unlin... |
245 |
return next; |
1da177e4c Linux-2.6.12-rc2 |
246 |
} |
e4eb1ff61 VM: add "vm_brk()... |
247 |
static unsigned long do_brk(unsigned long addr, unsigned long len); |
6a6160a7b [CVE-2009-0029] S... |
248 |
SYSCALL_DEFINE1(brk, unsigned long, brk) |
1da177e4c Linux-2.6.12-rc2 |
249 250 251 252 |
{ unsigned long rlim, retval; unsigned long newbrk, oldbrk; struct mm_struct *mm = current->mm; |
a5b4592cf brk: make sys_brk... |
253 |
unsigned long min_brk; |
128557ffe mm: use mm_popula... |
254 |
bool populate; |
1da177e4c Linux-2.6.12-rc2 |
255 256 |
down_write(&mm->mmap_sem); |
a5b4592cf brk: make sys_brk... |
257 |
#ifdef CONFIG_COMPAT_BRK |
5520e8948 brk: fix min_brk ... |
258 259 260 261 262 |
/* * 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 brk: COMPAT_BRK: ... |
263 |
if (current->brk_randomized) |
5520e8948 brk: fix min_brk ... |
264 265 266 |
min_brk = mm->start_brk; else min_brk = mm->end_data; |
a5b4592cf brk: make sys_brk... |
267 268 269 270 |
#else min_brk = mm->start_brk; #endif if (brk < min_brk) |
1da177e4c Linux-2.6.12-rc2 |
271 |
goto out; |
1e624196f [PATCH] mm: fix b... |
272 273 274 275 276 277 278 |
/* * 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 */ |
59e99e5b9 mm: use rlimit he... |
279 |
rlim = rlimit(RLIMIT_DATA); |
c1d171a00 x86: randomize brk |
280 281 |
if (rlim < RLIM_INFINITY && (brk - mm->start_brk) + (mm->end_data - mm->start_data) > rlim) |
1e624196f [PATCH] mm: fix b... |
282 |
goto out; |
1da177e4c Linux-2.6.12-rc2 |
283 284 285 286 287 288 289 290 291 292 293 |
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 Linux-2.6.12-rc2 |
294 295 296 297 298 299 300 |
/* 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 mm: use mm_popula... |
301 |
|
1da177e4c Linux-2.6.12-rc2 |
302 303 |
set_brk: mm->brk = brk; |
128557ffe mm: use mm_popula... |
304 305 306 307 308 |
populate = newbrk > oldbrk && (mm->def_flags & VM_LOCKED) != 0; up_write(&mm->mmap_sem); if (populate) mm_populate(oldbrk, newbrk - oldbrk); return brk; |
1da177e4c Linux-2.6.12-rc2 |
309 310 311 312 313 |
out: retval = mm->brk; up_write(&mm->mmap_sem); return retval; } |
d37371870 mm: augment vma r... |
314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 |
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 mm: add CONFIG_DE... |
334 |
#ifdef CONFIG_DEBUG_VM_RB |
1da177e4c Linux-2.6.12-rc2 |
335 336 |
static int browse_rb(struct rb_root *root) { |
5a0768f64 mm: check rb_subt... |
337 |
int i = 0, j, bug = 0; |
1da177e4c Linux-2.6.12-rc2 |
338 339 340 341 342 343 |
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 mm: check rb_subt... |
344 345 346 347 348 349 |
if (vma->vm_start < prev) { printk("vm_start %lx prev %lx ", vma->vm_start, prev); bug = 1; } if (vma->vm_start < pend) { |
1da177e4c Linux-2.6.12-rc2 |
350 351 |
printk("vm_start %lx pend %lx ", vma->vm_start, pend); |
5a0768f64 mm: check rb_subt... |
352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 |
bug = 1; } if (vma->vm_start > vma->vm_end) { printk("vm_end %lx < vm_start %lx ", vma->vm_end, vma->vm_start); bug = 1; } if (vma->rb_subtree_gap != vma_compute_subtree_gap(vma)) { printk("free gap %lx, correct %lx ", vma->rb_subtree_gap, vma_compute_subtree_gap(vma)); bug = 1; } |
1da177e4c Linux-2.6.12-rc2 |
367 368 |
i++; pn = nd; |
d1af65d13 [PATCH] Bug in MM... |
369 370 |
prev = vma->vm_start; pend = vma->vm_end; |
1da177e4c Linux-2.6.12-rc2 |
371 372 |
} j = 0; |
5a0768f64 mm: check rb_subt... |
373 |
for (nd = pn; nd; nd = rb_prev(nd)) |
1da177e4c Linux-2.6.12-rc2 |
374 |
j++; |
5a0768f64 mm: check rb_subt... |
375 376 377 378 |
if (i != j) { printk("backwards %d, forwards %d ", j, i); bug = 1; |
1da177e4c Linux-2.6.12-rc2 |
379 |
} |
5a0768f64 mm: check rb_subt... |
380 |
return bug ? -1 : i; |
1da177e4c Linux-2.6.12-rc2 |
381 |
} |
d37371870 mm: augment vma r... |
382 383 384 385 386 387 388 389 390 |
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); BUG_ON(vma != ignore && vma->rb_subtree_gap != vma_compute_subtree_gap(vma)); |
1da177e4c Linux-2.6.12-rc2 |
391 |
} |
1da177e4c Linux-2.6.12-rc2 |
392 393 394 395 396 397 |
} void validate_mm(struct mm_struct *mm) { int bug = 0; int i = 0; |
5a0768f64 mm: check rb_subt... |
398 |
unsigned long highest_address = 0; |
ed8ea8150 mm: add CONFIG_DE... |
399 400 401 |
struct vm_area_struct *vma = mm->mmap; while (vma) { struct anon_vma_chain *avc; |
63c3b902e mm: add anon_vma_... |
402 |
vma_lock_anon_vma(vma); |
ed8ea8150 mm: add CONFIG_DE... |
403 404 |
list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) anon_vma_interval_tree_verify(avc); |
63c3b902e mm: add anon_vma_... |
405 |
vma_unlock_anon_vma(vma); |
5a0768f64 mm: check rb_subt... |
406 |
highest_address = vma->vm_end; |
ed8ea8150 mm: add CONFIG_DE... |
407 |
vma = vma->vm_next; |
1da177e4c Linux-2.6.12-rc2 |
408 409 |
i++; } |
5a0768f64 mm: check rb_subt... |
410 411 412 413 414 415 416 417 418 419 420 |
if (i != mm->map_count) { printk("map_count %d vm_next %d ", mm->map_count, i); bug = 1; } if (highest_address != mm->highest_vm_end) { printk("mm->highest_vm_end %lx, found %lx ", mm->highest_vm_end, highest_address); bug = 1; } |
1da177e4c Linux-2.6.12-rc2 |
421 |
i = browse_rb(&mm->mm_rb); |
5a0768f64 mm: check rb_subt... |
422 423 424 425 426 |
if (i != mm->map_count) { printk("map_count %d rb %d ", mm->map_count, i); bug = 1; } |
46a350ef9 BUG_ON() Conversi... |
427 |
BUG_ON(bug); |
1da177e4c Linux-2.6.12-rc2 |
428 429 |
} #else |
d37371870 mm: augment vma r... |
430 |
#define validate_mm_rb(root, ignore) do { } while (0) |
1da177e4c Linux-2.6.12-rc2 |
431 432 |
#define validate_mm(mm) do { } while (0) #endif |
d37371870 mm: augment vma r... |
433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 |
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 mm anon rmap: rep... |
474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 |
/* * 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 mm/mmap.c: replac... |
505 506 507 |
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 Linux-2.6.12-rc2 |
508 |
{ |
6597d7833 mm/mmap.c: replac... |
509 |
struct rb_node **__rb_link, *__rb_parent, *rb_prev; |
1da177e4c Linux-2.6.12-rc2 |
510 511 512 |
__rb_link = &mm->mm_rb.rb_node; rb_prev = __rb_parent = NULL; |
1da177e4c Linux-2.6.12-rc2 |
513 514 515 516 517 518 519 520 |
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 mm/mmap.c: replac... |
521 522 523 |
/* Fail if an existing vma overlaps the area */ if (vma_tmp->vm_start < end) return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
524 525 526 527 528 529 530 531 532 533 534 535 |
__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 mm/mmap.c: replac... |
536 |
return 0; |
1da177e4c Linux-2.6.12-rc2 |
537 |
} |
e8420a8ec mm/mmap: check fo... |
538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 |
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 Linux-2.6.12-rc2 |
565 566 567 |
void __vma_link_rb(struct mm_struct *mm, struct vm_area_struct *vma, struct rb_node **rb_link, struct rb_node *rb_parent) { |
d37371870 mm: augment vma r... |
568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 |
/* 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 Linux-2.6.12-rc2 |
583 |
rb_link_node(&vma->vm_rb, rb_parent, rb_link); |
d37371870 mm: augment vma r... |
584 585 586 |
vma->rb_subtree_gap = 0; vma_gap_update(vma); vma_rb_insert(vma, &mm->mm_rb); |
1da177e4c Linux-2.6.12-rc2 |
587 |
} |
cb8f488c3 mmap.c: deinline ... |
588 |
static void __vma_link_file(struct vm_area_struct *vma) |
1da177e4c Linux-2.6.12-rc2 |
589 |
{ |
48aae4255 mm/mmap.c: fix co... |
590 |
struct file *file; |
1da177e4c Linux-2.6.12-rc2 |
591 592 593 594 595 596 |
file = vma->vm_file; if (file) { struct address_space *mapping = file->f_mapping; if (vma->vm_flags & VM_DENYWRITE) |
496ad9aa8 new helper: file_... |
597 |
atomic_dec(&file_inode(file)->i_writecount); |
1da177e4c Linux-2.6.12-rc2 |
598 599 600 601 602 603 604 |
if (vma->vm_flags & VM_SHARED) mapping->i_mmap_writable++; flush_dcache_mmap_lock(mapping); if (unlikely(vma->vm_flags & VM_NONLINEAR)) vma_nonlinear_insert(vma, &mapping->i_mmap_nonlinear); else |
6b2dbba8b mm: replace vma p... |
605 |
vma_interval_tree_insert(vma, &mapping->i_mmap); |
1da177e4c Linux-2.6.12-rc2 |
606 607 608 609 610 611 612 613 614 615 616 |
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 Linux-2.6.12-rc2 |
617 618 619 620 621 622 623 624 625 626 |
} static void vma_link(struct mm_struct *mm, struct vm_area_struct *vma, struct vm_area_struct *prev, struct rb_node **rb_link, struct rb_node *rb_parent) { struct address_space *mapping = NULL; if (vma->vm_file) mapping = vma->vm_file->f_mapping; |
97a894136 mm: Remove i_mmap... |
627 |
if (mapping) |
3d48ae45e mm: Convert i_mma... |
628 |
mutex_lock(&mapping->i_mmap_mutex); |
1da177e4c Linux-2.6.12-rc2 |
629 630 631 |
__vma_link(mm, vma, prev, rb_link, rb_parent); __vma_link_file(vma); |
1da177e4c Linux-2.6.12-rc2 |
632 |
if (mapping) |
3d48ae45e mm: Convert i_mma... |
633 |
mutex_unlock(&mapping->i_mmap_mutex); |
1da177e4c Linux-2.6.12-rc2 |
634 635 636 637 638 639 |
mm->map_count++; validate_mm(mm); } /* |
88f6b4c32 mmap.c: fix comme... |
640 |
* Helper for vma_adjust() in the split_vma insert case: insert a vma into the |
6b2dbba8b mm: replace vma p... |
641 |
* mm's list and rbtree. It has already been inserted into the interval tree. |
1da177e4c Linux-2.6.12-rc2 |
642 |
*/ |
48aae4255 mm/mmap.c: fix co... |
643 |
static void __insert_vm_struct(struct mm_struct *mm, struct vm_area_struct *vma) |
1da177e4c Linux-2.6.12-rc2 |
644 |
{ |
6597d7833 mm/mmap.c: replac... |
645 |
struct vm_area_struct *prev; |
48aae4255 mm/mmap.c: fix co... |
646 |
struct rb_node **rb_link, *rb_parent; |
1da177e4c Linux-2.6.12-rc2 |
647 |
|
6597d7833 mm/mmap.c: replac... |
648 649 650 |
if (find_vma_links(mm, vma->vm_start, vma->vm_end, &prev, &rb_link, &rb_parent)) BUG(); |
1da177e4c Linux-2.6.12-rc2 |
651 652 653 654 655 656 657 658 |
__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 mm: augment vma r... |
659 |
struct vm_area_struct *next; |
297c5eee3 mm: make the vma ... |
660 |
|
d37371870 mm: augment vma r... |
661 662 |
vma_rb_erase(vma, &mm->mm_rb); prev->vm_next = next = vma->vm_next; |
297c5eee3 mm: make the vma ... |
663 664 |
if (next) next->vm_prev = prev; |
1da177e4c Linux-2.6.12-rc2 |
665 666 667 668 669 670 671 672 673 674 675 |
if (mm->mmap_cache == vma) mm->mmap_cache = prev; } /* * We cannot adjust vm_start, vm_end, vm_pgoff fields of a vma that * is already present in an i_mmap tree without adjusting the tree. * The following helper function should be used when such adjustments * are necessary. The "insert" vma (if any) is to be inserted * before we drop the necessary locks. */ |
5beb49305 mm: change anon_v... |
676 |
int vma_adjust(struct vm_area_struct *vma, unsigned long start, |
1da177e4c Linux-2.6.12-rc2 |
677 678 679 680 681 682 |
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 mm: replace vma p... |
683 |
struct rb_root *root = NULL; |
012f18004 mm: always lock t... |
684 |
struct anon_vma *anon_vma = NULL; |
1da177e4c Linux-2.6.12-rc2 |
685 |
struct file *file = vma->vm_file; |
d37371870 mm: augment vma r... |
686 |
bool start_changed = false, end_changed = false; |
1da177e4c Linux-2.6.12-rc2 |
687 688 689 690 |
long adjust_next = 0; int remove_next = 0; if (next && !insert) { |
287d97ac0 vma_adjust: fix t... |
691 |
struct vm_area_struct *exporter = NULL; |
1da177e4c Linux-2.6.12-rc2 |
692 693 694 695 696 697 698 |
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 vma_adjust: fix t... |
699 |
exporter = next; |
1da177e4c Linux-2.6.12-rc2 |
700 701 702 703 704 705 706 |
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 vma_adjust: fix t... |
707 |
exporter = next; |
1da177e4c Linux-2.6.12-rc2 |
708 709 710 711 712 713 714 715 |
importer = vma; } else if (end < vma->vm_end) { /* * vma shrinks, and !insert tells it's not * split_vma inserting another: so it must be * mprotect case 4 shifting the boundary down. */ adjust_next = - ((vma->vm_end - end) >> PAGE_SHIFT); |
287d97ac0 vma_adjust: fix t... |
716 |
exporter = vma; |
1da177e4c Linux-2.6.12-rc2 |
717 718 |
importer = next; } |
1da177e4c Linux-2.6.12-rc2 |
719 |
|
5beb49305 mm: change anon_v... |
720 721 722 723 724 |
/* * 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 vma_adjust: fix t... |
725 726 |
if (exporter && exporter->anon_vma && !importer->anon_vma) { if (anon_vma_clone(importer, exporter)) |
5beb49305 mm: change anon_v... |
727 |
return -ENOMEM; |
287d97ac0 vma_adjust: fix t... |
728 |
importer->anon_vma = exporter->anon_vma; |
5beb49305 mm: change anon_v... |
729 730 |
} } |
1da177e4c Linux-2.6.12-rc2 |
731 732 |
if (file) { mapping = file->f_mapping; |
682968e0c uprobes/core: Opt... |
733 |
if (!(vma->vm_flags & VM_NONLINEAR)) { |
1da177e4c Linux-2.6.12-rc2 |
734 |
root = &mapping->i_mmap; |
cbc91f71b uprobes/core: Dec... |
735 |
uprobe_munmap(vma, vma->vm_start, vma->vm_end); |
682968e0c uprobes/core: Opt... |
736 737 |
if (adjust_next) |
cbc91f71b uprobes/core: Dec... |
738 739 |
uprobe_munmap(next, next->vm_start, next->vm_end); |
682968e0c uprobes/core: Opt... |
740 |
} |
3d48ae45e mm: Convert i_mma... |
741 |
mutex_lock(&mapping->i_mmap_mutex); |
1da177e4c Linux-2.6.12-rc2 |
742 |
if (insert) { |
1da177e4c Linux-2.6.12-rc2 |
743 |
/* |
6b2dbba8b mm: replace vma p... |
744 |
* Put into interval tree now, so instantiated pages |
1da177e4c Linux-2.6.12-rc2 |
745 746 747 748 749 750 751 |
* 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 thp: avoid breaki... |
752 |
vma_adjust_trans_huge(vma, start, end, adjust_next); |
bf181b9f9 mm anon rmap: rep... |
753 754 755 756 |
anon_vma = vma->anon_vma; if (!anon_vma && adjust_next) anon_vma = next->anon_vma; if (anon_vma) { |
ca42b26ab mm: fix potential... |
757 758 |
VM_BUG_ON(adjust_next && next->anon_vma && anon_vma != next->anon_vma); |
4fc3f1d66 mm/rmap, migratio... |
759 |
anon_vma_lock_write(anon_vma); |
bf181b9f9 mm anon rmap: rep... |
760 761 762 763 |
anon_vma_interval_tree_pre_update_vma(vma); if (adjust_next) anon_vma_interval_tree_pre_update_vma(next); } |
012f18004 mm: always lock t... |
764 |
|
1da177e4c Linux-2.6.12-rc2 |
765 766 |
if (root) { flush_dcache_mmap_lock(mapping); |
6b2dbba8b mm: replace vma p... |
767 |
vma_interval_tree_remove(vma, root); |
1da177e4c Linux-2.6.12-rc2 |
768 |
if (adjust_next) |
6b2dbba8b mm: replace vma p... |
769 |
vma_interval_tree_remove(next, root); |
1da177e4c Linux-2.6.12-rc2 |
770 |
} |
d37371870 mm: augment vma r... |
771 772 773 774 775 776 777 778 |
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 Linux-2.6.12-rc2 |
779 780 781 782 783 784 785 786 |
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 mm: replace vma p... |
787 788 |
vma_interval_tree_insert(next, root); vma_interval_tree_insert(vma, root); |
1da177e4c Linux-2.6.12-rc2 |
789 790 791 792 793 794 795 796 797 798 799 |
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 Linux-2.6.12-rc2 |
800 801 802 803 804 805 806 |
} 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 mm: augment vma r... |
807 808 809 810 811 812 813 814 815 |
} 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 Linux-2.6.12-rc2 |
816 |
} |
bf181b9f9 mm anon rmap: rep... |
817 818 819 820 |
if (anon_vma) { anon_vma_interval_tree_post_update_vma(vma); if (adjust_next) anon_vma_interval_tree_post_update_vma(next); |
08b52706d mm/rmap: rename a... |
821 |
anon_vma_unlock_write(anon_vma); |
bf181b9f9 mm anon rmap: rep... |
822 |
} |
1da177e4c Linux-2.6.12-rc2 |
823 |
if (mapping) |
3d48ae45e mm: Convert i_mma... |
824 |
mutex_unlock(&mapping->i_mmap_mutex); |
1da177e4c Linux-2.6.12-rc2 |
825 |
|
2b1444983 uprobes, mm, x86:... |
826 |
if (root) { |
7b2d81d48 uprobes/core: Cle... |
827 |
uprobe_mmap(vma); |
2b1444983 uprobes, mm, x86:... |
828 829 |
if (adjust_next) |
7b2d81d48 uprobes/core: Cle... |
830 |
uprobe_mmap(next); |
2b1444983 uprobes, mm, x86:... |
831 |
} |
1da177e4c Linux-2.6.12-rc2 |
832 |
if (remove_next) { |
925d1c401 procfs task exe s... |
833 |
if (file) { |
cbc91f71b uprobes/core: Dec... |
834 |
uprobe_munmap(next, next->vm_start, next->vm_end); |
1da177e4c Linux-2.6.12-rc2 |
835 |
fput(file); |
925d1c401 procfs task exe s... |
836 |
} |
5beb49305 mm: change anon_v... |
837 838 |
if (next->anon_vma) anon_vma_merge(vma, next); |
1da177e4c Linux-2.6.12-rc2 |
839 |
mm->map_count--; |
0bd6f78c3 mm: mempolicy: fi... |
840 |
mpol_put(vma_policy(next)); |
1da177e4c Linux-2.6.12-rc2 |
841 842 843 844 845 846 |
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 mm: augment vma r... |
847 848 |
next = vma->vm_next; if (remove_next == 2) |
1da177e4c Linux-2.6.12-rc2 |
849 |
goto again; |
d37371870 mm: augment vma r... |
850 851 852 853 |
else if (next) vma_gap_update(next); else mm->highest_vm_end = end; |
1da177e4c Linux-2.6.12-rc2 |
854 |
} |
2b1444983 uprobes, mm, x86:... |
855 |
if (insert && file) |
7b2d81d48 uprobes/core: Cle... |
856 |
uprobe_mmap(insert); |
1da177e4c Linux-2.6.12-rc2 |
857 858 |
validate_mm(mm); |
5beb49305 mm: change anon_v... |
859 860 |
return 0; |
1da177e4c Linux-2.6.12-rc2 |
861 862 863 864 865 866 |
} /* * 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 Linux-2.6.12-rc2 |
867 868 869 |
static inline int is_mergeable_vma(struct vm_area_struct *vma, struct file *file, unsigned long vm_flags) { |
0b173bc4d mm: kill vma flag... |
870 |
if (vma->vm_flags ^ vm_flags) |
1da177e4c Linux-2.6.12-rc2 |
871 872 873 874 875 876 877 878 879 |
return 0; if (vma->vm_file != file) return 0; if (vma->vm_ops && vma->vm_ops->close) return 0; return 1; } static inline int is_mergeable_anon_vma(struct anon_vma *anon_vma1, |
965f55dea mmap: avoid mergi... |
880 881 |
struct anon_vma *anon_vma2, struct vm_area_struct *vma) |
1da177e4c Linux-2.6.12-rc2 |
882 |
{ |
965f55dea mmap: avoid mergi... |
883 884 885 886 887 888 889 890 |
/* * 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 Linux-2.6.12-rc2 |
891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 |
} /* * Return true if we can merge this (vm_flags,anon_vma,file,vm_pgoff) * in front of (at a lower virtual address and file offset than) the vma. * * We cannot merge two vmas if they have differently assigned (non-NULL) * anon_vmas, nor if same anon_vma is assigned but offsets incompatible. * * We don't check here for the merged mmap wrapping around the end of pagecache * indices (16TB on ia32) because do_mmap_pgoff() does not permit mmap's which * wrap, nor mmaps which cover the final page at index -1UL. */ static int can_vma_merge_before(struct vm_area_struct *vma, unsigned long vm_flags, struct anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff) { if (is_mergeable_vma(vma, file, vm_flags) && |
965f55dea mmap: avoid mergi... |
909 |
is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) { |
1da177e4c Linux-2.6.12-rc2 |
910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 |
if (vma->vm_pgoff == vm_pgoff) return 1; } return 0; } /* * Return true if we can merge this (vm_flags,anon_vma,file,vm_pgoff) * beyond (at a higher virtual address and file offset than) the vma. * * We cannot merge two vmas if they have differently assigned (non-NULL) * anon_vmas, nor if same anon_vma is assigned but offsets incompatible. */ static int can_vma_merge_after(struct vm_area_struct *vma, unsigned long vm_flags, struct anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff) { if (is_mergeable_vma(vma, file, vm_flags) && |
965f55dea mmap: avoid mergi... |
928 |
is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) { |
1da177e4c Linux-2.6.12-rc2 |
929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 |
pgoff_t vm_pglen; vm_pglen = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT; if (vma->vm_pgoff + vm_pglen == vm_pgoff) return 1; } return 0; } /* * Given a mapping request (addr,end,vm_flags,file,pgoff), figure out * whether that can be merged with its predecessor or its successor. * Or both (it neatly fills a hole). * * In most cases - when called for mmap, brk or mremap - [addr,end) is * certain not to be mapped by the time vma_merge is called; but when * called for mprotect, it is certain to be already mapped (either at * an offset within prev, or at the start of next), and the flags of * this area are about to be changed to vm_flags - and the no-change * case has already been eliminated. * * The following mprotect cases have to be considered, where AAAA is * the area passed down from mprotect_fixup, never extending beyond one * vma, PPPPPP is the prev vma specified, and NNNNNN the next vma after: * * AAAA AAAA AAAA AAAA * PPPPPPNNNNNN PPPPPPNNNNNN PPPPPPNNNNNN PPPPNNNNXXXX * cannot merge might become might become might become * PPNNNNNNNNNN PPPPPPPPPPNN PPPPPPPPPPPP 6 or * mmap, brk or case 4 below case 5 below PPPPPPPPXXXX 7 or * mremap move: PPPPNNNNNNNN 8 * AAAA * PPPP NNNN PPPPPPPPPPPP PPPPPPPPNNNN PPPPNNNNNNNN * might become case 1 below case 2 below case 3 below * * Odd one out? Case 8, because it extends NNNN but needs flags of XXXX: * mprotect_fixup updates vm_flags & vm_page_prot on successful return. */ struct vm_area_struct *vma_merge(struct mm_struct *mm, struct vm_area_struct *prev, unsigned long addr, unsigned long end, unsigned long vm_flags, struct anon_vma *anon_vma, struct file *file, pgoff_t pgoff, struct mempolicy *policy) { pgoff_t pglen = (end - addr) >> PAGE_SHIFT; struct vm_area_struct *area, *next; |
5beb49305 mm: change anon_v... |
974 |
int err; |
1da177e4c Linux-2.6.12-rc2 |
975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 |
/* * We later require that vma->vm_flags == vm_flags, * so this tests vma->vm_flags & VM_SPECIAL, too. */ if (vm_flags & VM_SPECIAL) return NULL; if (prev) next = prev->vm_next; else next = mm->mmap; area = next; if (next && next->vm_end == end) /* cases 6, 7, 8 */ next = next->vm_next; /* * Can it merge with the predecessor? */ if (prev && prev->vm_end == addr && mpol_equal(vma_policy(prev), policy) && can_vma_merge_after(prev, vm_flags, anon_vma, file, pgoff)) { /* * OK, it can. Can we now merge in the successor as well? */ if (next && end == next->vm_start && mpol_equal(policy, vma_policy(next)) && can_vma_merge_before(next, vm_flags, anon_vma, file, pgoff+pglen) && is_mergeable_anon_vma(prev->anon_vma, |
965f55dea mmap: avoid mergi... |
1006 |
next->anon_vma, NULL)) { |
1da177e4c Linux-2.6.12-rc2 |
1007 |
/* cases 1, 6 */ |
5beb49305 mm: change anon_v... |
1008 |
err = vma_adjust(prev, prev->vm_start, |
1da177e4c Linux-2.6.12-rc2 |
1009 1010 |
next->vm_end, prev->vm_pgoff, NULL); } else /* cases 2, 5, 7 */ |
5beb49305 mm: change anon_v... |
1011 |
err = vma_adjust(prev, prev->vm_start, |
1da177e4c Linux-2.6.12-rc2 |
1012 |
end, prev->vm_pgoff, NULL); |
5beb49305 mm: change anon_v... |
1013 1014 |
if (err) return NULL; |
b15d00b6a thp: khugepaged v... |
1015 |
khugepaged_enter_vma_merge(prev); |
1da177e4c Linux-2.6.12-rc2 |
1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 |
return prev; } /* * Can this new request be merged in front of next? */ if (next && end == next->vm_start && mpol_equal(policy, vma_policy(next)) && can_vma_merge_before(next, vm_flags, anon_vma, file, pgoff+pglen)) { if (prev && addr < prev->vm_end) /* case 4 */ |
5beb49305 mm: change anon_v... |
1027 |
err = vma_adjust(prev, prev->vm_start, |
1da177e4c Linux-2.6.12-rc2 |
1028 1029 |
addr, prev->vm_pgoff, NULL); else /* cases 3, 8 */ |
5beb49305 mm: change anon_v... |
1030 |
err = vma_adjust(area, addr, next->vm_end, |
1da177e4c Linux-2.6.12-rc2 |
1031 |
next->vm_pgoff - pglen, NULL); |
5beb49305 mm: change anon_v... |
1032 1033 |
if (err) return NULL; |
b15d00b6a thp: khugepaged v... |
1034 |
khugepaged_enter_vma_merge(area); |
1da177e4c Linux-2.6.12-rc2 |
1035 1036 1037 1038 1039 1040 1041 |
return area; } return NULL; } /* |
d0e9fe175 Simplify and comm... |
1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 |
* Rough compatbility check to quickly see if it's even worth looking * at sharing an anon_vma. * * They need to have the same vm_file, and the flags can only differ * in things that mprotect may change. * * NOTE! The fact that we share an anon_vma doesn't _have_ to mean that * we can merge the two vma's. For example, we refuse to merge a vma if * there is a vm_ops->close() function, because that indicates that the * driver is doing some kind of reference counting. But that doesn't * really matter for the anon_vma sharing case. */ static int anon_vma_compatible(struct vm_area_struct *a, struct vm_area_struct *b) { return a->vm_end == b->vm_start && mpol_equal(vma_policy(a), vma_policy(b)) && a->vm_file == b->vm_file && !((a->vm_flags ^ b->vm_flags) & ~(VM_READ|VM_WRITE|VM_EXEC)) && b->vm_pgoff == a->vm_pgoff + ((b->vm_start - a->vm_start) >> PAGE_SHIFT); } /* * Do some basic sanity checking to see if we can re-use the anon_vma * from 'old'. The 'a'/'b' vma's are in VM order - one of them will be * the same as 'old', the other will be the new one that is trying * to share the anon_vma. * * NOTE! This runs with mm_sem held for reading, so it is possible that * the anon_vma of 'old' is concurrently in the process of being set up * by another page fault trying to merge _that_. But that's ok: if it * is being set up, that automatically means that it will be a singleton * acceptable for merging, so we can do all of this optimistically. But * we do that ACCESS_ONCE() to make sure that we never re-load the pointer. * * IOW: that the "list_is_singular()" test on the anon_vma_chain only * matters for the 'stable anon_vma' case (ie the thing we want to avoid * is to return an anon_vma that is "complex" due to having gone through * a fork). * * We also make sure that the two vma's are compatible (adjacent, * and with the same memory policies). That's all stable, even with just * a read lock on the mm_sem. */ static struct anon_vma *reusable_anon_vma(struct vm_area_struct *old, struct vm_area_struct *a, struct vm_area_struct *b) { if (anon_vma_compatible(a, b)) { struct anon_vma *anon_vma = ACCESS_ONCE(old->anon_vma); if (anon_vma && list_is_singular(&old->anon_vma_chain)) return anon_vma; } return NULL; } /* |
1da177e4c Linux-2.6.12-rc2 |
1097 1098 1099 1100 1101 1102 1103 1104 1105 |
* 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 Simplify and comm... |
1106 |
struct anon_vma *anon_vma; |
1da177e4c Linux-2.6.12-rc2 |
1107 |
struct vm_area_struct *near; |
1da177e4c Linux-2.6.12-rc2 |
1108 1109 1110 1111 |
near = vma->vm_next; if (!near) goto try_prev; |
d0e9fe175 Simplify and comm... |
1112 1113 1114 |
anon_vma = reusable_anon_vma(near, vma, near); if (anon_vma) return anon_vma; |
1da177e4c Linux-2.6.12-rc2 |
1115 |
try_prev: |
9be34c9d5 mm: get rid of th... |
1116 |
near = vma->vm_prev; |
1da177e4c Linux-2.6.12-rc2 |
1117 1118 |
if (!near) goto none; |
d0e9fe175 Simplify and comm... |
1119 1120 1121 |
anon_vma = reusable_anon_vma(near, near, vma); if (anon_vma) return anon_vma; |
1da177e4c Linux-2.6.12-rc2 |
1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 |
none: /* * There's no absolute need to look only at touching neighbours: * we could search further afield for "compatible" anon_vmas. * But it would probably just be a waste of time searching, * or lead to too many vmas hanging off the same anon_vma. * We're trying to allow mprotect remerging later on, * not trying to minimize memory used for anon_vmas. */ return NULL; } #ifdef CONFIG_PROC_FS |
ab50b8ed8 [PATCH] mm: vm_st... |
1135 |
void vm_stat_account(struct mm_struct *mm, unsigned long flags, |
1da177e4c Linux-2.6.12-rc2 |
1136 1137 1138 1139 |
struct file *file, long pages) { const unsigned long stack_flags = VM_STACK_FLAGS & (VM_GROWSUP|VM_GROWSDOWN); |
44de9d0ca mm: account the t... |
1140 |
mm->total_vm += pages; |
1da177e4c Linux-2.6.12-rc2 |
1141 1142 1143 1144 1145 1146 |
if (file) { mm->shared_vm += pages; if ((flags & (VM_EXEC|VM_WRITE)) == VM_EXEC) mm->exec_vm += pages; } else if (flags & stack_flags) mm->stack_vm += pages; |
1da177e4c Linux-2.6.12-rc2 |
1147 1148 1149 1150 |
} #endif /* CONFIG_PROC_FS */ /* |
404015308 security: trim se... |
1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 |
* 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; } /* |
27f5de796 mm: Fix problem o... |
1164 |
* The caller must hold down_write(¤t->mm->mmap_sem). |
1da177e4c Linux-2.6.12-rc2 |
1165 |
*/ |
e3fc629d7 switch aio and sh... |
1166 |
unsigned long do_mmap_pgoff(struct file *file, unsigned long addr, |
1da177e4c Linux-2.6.12-rc2 |
1167 |
unsigned long len, unsigned long prot, |
bebeb3d68 mm: introduce mm_... |
1168 |
unsigned long flags, unsigned long pgoff, |
41badc15c mm: make do_mmap_... |
1169 |
unsigned long *populate) |
1da177e4c Linux-2.6.12-rc2 |
1170 1171 |
{ struct mm_struct * mm = current->mm; |
1da177e4c Linux-2.6.12-rc2 |
1172 |
struct inode *inode; |
ca16d140a mm: don't access ... |
1173 |
vm_flags_t vm_flags; |
1da177e4c Linux-2.6.12-rc2 |
1174 |
|
41badc15c mm: make do_mmap_... |
1175 |
*populate = 0; |
bebeb3d68 mm: introduce mm_... |
1176 |
|
1da177e4c Linux-2.6.12-rc2 |
1177 1178 1179 1180 1181 1182 1183 |
/* * 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)) |
d3ac7f892 [PATCH] mm: chang... |
1184 |
if (!(file && (file->f_path.mnt->mnt_flags & MNT_NOEXEC))) |
1da177e4c Linux-2.6.12-rc2 |
1185 1186 1187 1188 |
prot |= PROT_EXEC; if (!len) return -EINVAL; |
7cd94146c Security: round m... |
1189 1190 |
if (!(flags & MAP_FIXED)) addr = round_hint_to_min(addr); |
1da177e4c Linux-2.6.12-rc2 |
1191 1192 |
/* Careful about overflows.. */ len = PAGE_ALIGN(len); |
9206de95b Take arch_mmap_ch... |
1193 |
if (!len) |
1da177e4c Linux-2.6.12-rc2 |
1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 |
return -ENOMEM; /* offset overflow? */ if ((pgoff + (len >> PAGE_SHIFT)) < pgoff) return -EOVERFLOW; /* Too many mappings? */ if (mm->map_count > sysctl_max_map_count) return -ENOMEM; /* Obtain the address to map to. we verify (or select) it and ensure * that it represents a valid section of the address space. */ addr = get_unmapped_area(file, addr, len, pgoff, flags); if (addr & ~PAGE_MASK) return addr; /* Do simple checking here so the lower-level routines won't have * to. we assume access permissions have been handled by the open * of the memory object, so we don't do any here. */ vm_flags = calc_vm_prot_bits(prot) | calc_vm_flag_bits(flags) | mm->def_flags | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC; |
cdf7b3418 mmap: remove unne... |
1217 |
if (flags & MAP_LOCKED) |
1da177e4c Linux-2.6.12-rc2 |
1218 1219 |
if (!can_do_mlock()) return -EPERM; |
ba470de43 mmap: handle mloc... |
1220 |
|
1da177e4c Linux-2.6.12-rc2 |
1221 1222 1223 |
/* mlock MCL_FUTURE? */ if (vm_flags & VM_LOCKED) { unsigned long locked, lock_limit; |
93ea1d0a1 [PATCH] RLIMIT_ME... |
1224 1225 |
locked = len >> PAGE_SHIFT; locked += mm->locked_vm; |
59e99e5b9 mm: use rlimit he... |
1226 |
lock_limit = rlimit(RLIMIT_MEMLOCK); |
93ea1d0a1 [PATCH] RLIMIT_ME... |
1227 |
lock_limit >>= PAGE_SHIFT; |
1da177e4c Linux-2.6.12-rc2 |
1228 1229 1230 |
if (locked > lock_limit && !capable(CAP_IPC_LOCK)) return -EAGAIN; } |
496ad9aa8 new helper: file_... |
1231 |
inode = file ? file_inode(file) : NULL; |
1da177e4c Linux-2.6.12-rc2 |
1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 |
if (file) { switch (flags & MAP_TYPE) { case MAP_SHARED: if ((prot&PROT_WRITE) && !(file->f_mode&FMODE_WRITE)) return -EACCES; /* * Make sure we don't allow writing to an append-only * file.. */ if (IS_APPEND(inode) && (file->f_mode & FMODE_WRITE)) return -EACCES; /* * Make sure there are no mandatory locks on the file. */ if (locks_verify_locked(inode)) return -EAGAIN; vm_flags |= VM_SHARED | VM_MAYSHARE; if (!(file->f_mode & FMODE_WRITE)) vm_flags &= ~(VM_MAYWRITE | VM_SHARED); /* fall through */ case MAP_PRIVATE: if (!(file->f_mode & FMODE_READ)) return -EACCES; |
d3ac7f892 [PATCH] mm: chang... |
1260 |
if (file->f_path.mnt->mnt_flags & MNT_NOEXEC) { |
80c5606c3 Fix VM_MAYEXEC ca... |
1261 1262 1263 1264 |
if (vm_flags & VM_EXEC) return -EPERM; vm_flags &= ~VM_MAYEXEC; } |
80c5606c3 Fix VM_MAYEXEC ca... |
1265 1266 1267 |
if (!file->f_op || !file->f_op->mmap) return -ENODEV; |
1da177e4c Linux-2.6.12-rc2 |
1268 1269 1270 1271 1272 1273 1274 1275 |
break; default: return -EINVAL; } } else { switch (flags & MAP_TYPE) { case MAP_SHARED: |
ce3639426 mmap: fix petty b... |
1276 1277 1278 1279 |
/* * Ignore pgoff. */ pgoff = 0; |
1da177e4c Linux-2.6.12-rc2 |
1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 |
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 mm: remove flags ... |
1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 |
/* * 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 Revert "mm: intro... |
1307 1308 1309 |
if (!IS_ERR_VALUE(addr) && ((vm_flags & VM_LOCKED) || (flags & (MAP_POPULATE | MAP_NONBLOCK)) == MAP_POPULATE)) |
41badc15c mm: make do_mmap_... |
1310 |
*populate = len; |
bebeb3d68 mm: introduce mm_... |
1311 |
return addr; |
0165ab443 split mmap |
1312 |
} |
6be5ceb02 VM: add "vm_mmap(... |
1313 |
|
66f0dc481 mm: move sys_mmap... |
1314 1315 1316 1317 1318 1319 1320 1321 |
SYSCALL_DEFINE6(mmap_pgoff, unsigned long, addr, unsigned long, len, unsigned long, prot, unsigned long, flags, unsigned long, fd, unsigned long, pgoff) { struct file *file = NULL; unsigned long retval = -EBADF; if (!(flags & MAP_ANONYMOUS)) { |
120a795da audit mmap |
1322 |
audit_mmap_fd(fd, flags); |
66f0dc481 mm: move sys_mmap... |
1323 1324 1325 1326 1327 |
if (unlikely(flags & MAP_HUGETLB)) return -EINVAL; file = fget(fd); if (!file) goto out; |
af73e4d95 hugetlbfs: fix mm... |
1328 1329 |
if (is_file_hugepages(file)) len = ALIGN(len, huge_page_size(hstate_file(file))); |
66f0dc481 mm: move sys_mmap... |
1330 1331 |
} else if (flags & MAP_HUGETLB) { struct user_struct *user = NULL; |
091d0d55b shm: fix null poi... |
1332 1333 |
struct hstate *hs = hstate_sizelog((flags >> MAP_HUGE_SHIFT) & SHM_HUGE_MASK); |
af73e4d95 hugetlbfs: fix mm... |
1334 |
|
091d0d55b shm: fix null poi... |
1335 1336 1337 1338 |
if (!hs) return -EINVAL; len = ALIGN(len, huge_page_size(hs)); |
66f0dc481 mm: move sys_mmap... |
1339 1340 1341 1342 1343 1344 |
/* * 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 hugetlbfs: fix mm... |
1345 |
file = hugetlb_file_setup(HUGETLB_ANON_FILE, len, |
42d7395fe mm: support more ... |
1346 1347 1348 |
VM_NORESERVE, &user, HUGETLB_ANONHUGE_INODE, (flags >> MAP_HUGE_SHIFT) & MAP_HUGE_MASK); |
66f0dc481 mm: move sys_mmap... |
1349 1350 1351 1352 1353 |
if (IS_ERR(file)) return PTR_ERR(file); } flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); |
eb36c5873 new helper: vm_mm... |
1354 |
retval = vm_mmap_pgoff(file, addr, len, prot, flags, pgoff); |
66f0dc481 mm: move sys_mmap... |
1355 1356 1357 1358 1359 |
if (file) fput(file); out: return retval; } |
a4679373c Add generic sys_o... |
1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 |
#ifdef __ARCH_WANT_SYS_OLD_MMAP struct mmap_arg_struct { unsigned long addr; unsigned long len; unsigned long prot; unsigned long flags; unsigned long fd; unsigned long offset; }; SYSCALL_DEFINE1(old_mmap, struct mmap_arg_struct __user *, arg) { struct mmap_arg_struct a; if (copy_from_user(&a, arg, sizeof(a))) return -EFAULT; if (a.offset & ~PAGE_MASK) return -EINVAL; return sys_mmap_pgoff(a.addr, a.len, a.prot, a.flags, a.fd, a.offset >> PAGE_SHIFT); } #endif /* __ARCH_WANT_SYS_OLD_MMAP */ |
4e950f6f0 Remove fs.h from ... |
1383 1384 1385 1386 1387 1388 1389 1390 |
/* * 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 mm: don't access ... |
1391 |
vm_flags_t vm_flags = vma->vm_flags; |
4e950f6f0 Remove fs.h from ... |
1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 |
/* If it was private or non-writable, the write bit is already clear */ if ((vm_flags & (VM_WRITE|VM_SHARED)) != ((VM_WRITE|VM_SHARED))) return 0; /* The backer wishes to know when pages are first written to? */ if (vma->vm_ops && vma->vm_ops->page_mkwrite) return 1; /* The open routine did something to the protections already? */ if (pgprot_val(vma->vm_page_prot) != |
3ed75eb8f setup vma->vm_pag... |
1403 |
pgprot_val(vm_get_page_prot(vm_flags))) |
4e950f6f0 Remove fs.h from ... |
1404 1405 1406 |
return 0; /* Specialty mapping? */ |
4b6e1e370 mm: kill vma flag... |
1407 |
if (vm_flags & VM_PFNMAP) |
4e950f6f0 Remove fs.h from ... |
1408 1409 1410 1411 1412 1413 |
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 Stop playing sill... |
1414 1415 |
/* * We account for memory if it's a private writeable mapping, |
5a6fe1259 Do not account fo... |
1416 |
* not hugepages and VM_NORESERVE wasn't set. |
fc8744adc Stop playing sill... |
1417 |
*/ |
ca16d140a mm: don't access ... |
1418 |
static inline int accountable_mapping(struct file *file, vm_flags_t vm_flags) |
fc8744adc Stop playing sill... |
1419 |
{ |
5a6fe1259 Do not account fo... |
1420 1421 1422 1423 1424 1425 |
/* * 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 Stop playing sill... |
1426 1427 |
return (vm_flags & (VM_NORESERVE | VM_SHARED | VM_WRITE)) == VM_WRITE; } |
0165ab443 split mmap |
1428 |
unsigned long mmap_region(struct file *file, unsigned long addr, |
c22c0d634 mm: remove flags ... |
1429 |
unsigned long len, vm_flags_t vm_flags, unsigned long pgoff) |
0165ab443 split mmap |
1430 1431 1432 1433 1434 1435 1436 |
{ struct mm_struct *mm = current->mm; struct vm_area_struct *vma, *prev; int correct_wcount = 0; int error; struct rb_node **rb_link, *rb_parent; unsigned long charged = 0; |
496ad9aa8 new helper: file_... |
1437 |
struct inode *inode = file ? file_inode(file) : NULL; |
0165ab443 split mmap |
1438 |
|
e8420a8ec mm/mmap: check fo... |
1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 |
/* Check against address space limit. */ if (!may_expand_vm(mm, len >> PAGE_SHIFT)) { unsigned long nr_pages; /* * MAP_FIXED may remove pages of mappings that intersects with * requested mapping. Account for the pages it would unmap. */ if (!(vm_flags & MAP_FIXED)) return -ENOMEM; nr_pages = count_vma_pages_range(mm, addr, addr + len); if (!may_expand_vm(mm, (len >> PAGE_SHIFT) - nr_pages)) return -ENOMEM; } |
1da177e4c Linux-2.6.12-rc2 |
1455 1456 1457 |
/* Clear old maps */ error = -ENOMEM; munmap_back: |
6597d7833 mm/mmap.c: replac... |
1458 |
if (find_vma_links(mm, addr, addr + len, &prev, &rb_link, &rb_parent)) { |
1da177e4c Linux-2.6.12-rc2 |
1459 1460 1461 1462 |
if (do_munmap(mm, addr, len)) return -ENOMEM; goto munmap_back; } |
fc8744adc Stop playing sill... |
1463 |
/* |
fc8744adc Stop playing sill... |
1464 1465 |
* Private writable mapping: check memory availability */ |
5a6fe1259 Do not account fo... |
1466 |
if (accountable_mapping(file, vm_flags)) { |
fc8744adc Stop playing sill... |
1467 |
charged = len >> PAGE_SHIFT; |
191c54244 mm: collapse secu... |
1468 |
if (security_vm_enough_memory_mm(mm, charged)) |
fc8744adc Stop playing sill... |
1469 1470 |
return -ENOMEM; vm_flags |= VM_ACCOUNT; |
1da177e4c Linux-2.6.12-rc2 |
1471 1472 1473 |
} /* |
de33c8db5 Fix OOPS in mmap_... |
1474 |
* Can we just expand an old mapping? |
1da177e4c Linux-2.6.12-rc2 |
1475 |
*/ |
de33c8db5 Fix OOPS in mmap_... |
1476 1477 1478 |
vma = vma_merge(mm, prev, addr, addr + len, vm_flags, NULL, file, pgoff, NULL); if (vma) goto out; |
1da177e4c Linux-2.6.12-rc2 |
1479 1480 1481 1482 1483 1484 |
/* * 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 [PATCH] mm: use k... |
1485 |
vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); |
1da177e4c Linux-2.6.12-rc2 |
1486 1487 1488 1489 |
if (!vma) { error = -ENOMEM; goto unacct_error; } |
1da177e4c Linux-2.6.12-rc2 |
1490 1491 1492 1493 1494 |
vma->vm_mm = mm; vma->vm_start = addr; vma->vm_end = addr + len; vma->vm_flags = vm_flags; |
3ed75eb8f setup vma->vm_pag... |
1495 |
vma->vm_page_prot = vm_get_page_prot(vm_flags); |
1da177e4c Linux-2.6.12-rc2 |
1496 |
vma->vm_pgoff = pgoff; |
5beb49305 mm: change anon_v... |
1497 |
INIT_LIST_HEAD(&vma->anon_vma_chain); |
1da177e4c Linux-2.6.12-rc2 |
1498 |
|
ce8fea7aa mmap: EINVAL not ... |
1499 |
error = -EINVAL; /* when rejecting VM_GROWSDOWN|VM_GROWSUP */ |
1da177e4c Linux-2.6.12-rc2 |
1500 |
if (file) { |
1da177e4c Linux-2.6.12-rc2 |
1501 1502 1503 1504 1505 1506 1507 1508 |
if (vm_flags & (VM_GROWSDOWN|VM_GROWSUP)) goto free_vma; if (vm_flags & VM_DENYWRITE) { error = deny_write_access(file); if (error) goto free_vma; correct_wcount = 1; } |
cb0942b81 make get_file() r... |
1509 |
vma->vm_file = get_file(file); |
1da177e4c Linux-2.6.12-rc2 |
1510 1511 1512 |
error = file->f_op->mmap(file, vma); if (error) goto unmap_and_free_vma; |
f8dbf0a7a mmap: save some c... |
1513 1514 1515 1516 1517 |
/* Can addr have changed?? * * Answer: Yes, several device drivers can do it in their * f_op->mmap method. -DaveM |
2897b4d29 mm: WARN_ON_ONCE ... |
1518 1519 |
* Bug: If addr is changed, prev, rb_link, rb_parent should * be updated for vma_link() |
f8dbf0a7a mmap: save some c... |
1520 |
*/ |
2897b4d29 mm: WARN_ON_ONCE ... |
1521 |
WARN_ON_ONCE(addr != vma->vm_start); |
f8dbf0a7a mmap: save some c... |
1522 1523 1524 |
addr = vma->vm_start; pgoff = vma->vm_pgoff; vm_flags = vma->vm_flags; |
1da177e4c Linux-2.6.12-rc2 |
1525 |
} else if (vm_flags & VM_SHARED) { |
835ee7978 VM_GROWS{UP,DOWN}... |
1526 1527 |
if (unlikely(vm_flags & (VM_GROWSDOWN|VM_GROWSUP))) goto free_vma; |
1da177e4c Linux-2.6.12-rc2 |
1528 1529 1530 1531 |
error = shmem_zero_setup(vma); if (error) goto free_vma; } |
c9d0bf241 mm: uncached vma ... |
1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 |
if (vma_wants_writenotify(vma)) { pgprot_t pprot = vma->vm_page_prot; /* Can vma->vm_page_prot have changed?? * * Answer: Yes, drivers may have changed it in their * f_op->mmap method. * * Ensures that vmas marked as uncached stay that way. */ |
1ddd439ef fix mprotect vma_... |
1542 |
vma->vm_page_prot = vm_get_page_prot(vm_flags & ~VM_SHARED); |
c9d0bf241 mm: uncached vma ... |
1543 1544 1545 |
if (pgprot_val(pprot) == pgprot_val(pgprot_noncached(pprot))) vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); } |
d08b3851d [PATCH] mm: track... |
1546 |
|
de33c8db5 Fix OOPS in mmap_... |
1547 1548 |
vma_link(mm, vma, prev, rb_link, rb_parent); file = vma->vm_file; |
4d3d5b41a mmap_region: clea... |
1549 1550 1551 1552 1553 |
/* Once vma denies write, undo our temporary denial count */ if (correct_wcount) atomic_inc(&inode->i_writecount); out: |
cdd6c482c perf: Do the big ... |
1554 |
perf_event_mmap(vma); |
0a4a93919 perf_counter: exe... |
1555 |
|
ab50b8ed8 [PATCH] mm: vm_st... |
1556 |
vm_stat_account(mm, vm_flags, file, len >> PAGE_SHIFT); |
1da177e4c Linux-2.6.12-rc2 |
1557 |
if (vm_flags & VM_LOCKED) { |
bebeb3d68 mm: introduce mm_... |
1558 1559 |
if (!((vm_flags & VM_SPECIAL) || is_vm_hugetlb_page(vma) || vma == get_gate_vma(current->mm))) |
06f9d8c2b mm: mlock_vma_pag... |
1560 |
mm->locked_vm += (len >> PAGE_SHIFT); |
bebeb3d68 mm: introduce mm_... |
1561 1562 1563 |
else vma->vm_flags &= ~VM_LOCKED; } |
2b1444983 uprobes, mm, x86:... |
1564 |
|
c7a3a88c9 uprobes: Fix mmap... |
1565 1566 |
if (file) uprobe_mmap(vma); |
2b1444983 uprobes, mm, x86:... |
1567 |
|
1da177e4c Linux-2.6.12-rc2 |
1568 1569 1570 1571 1572 1573 1574 1575 1576 |
return addr; unmap_and_free_vma: if (correct_wcount) atomic_inc(&inode->i_writecount); vma->vm_file = NULL; fput(file); /* Undo any partial mapping done by a device driver. */ |
e0da382c9 [PATCH] freepgt: ... |
1577 1578 |
unmap_region(mm, vma, prev, vma->vm_start, vma->vm_end); charged = 0; |
1da177e4c Linux-2.6.12-rc2 |
1579 1580 1581 1582 1583 1584 1585 |
free_vma: kmem_cache_free(vm_area_cachep, vma); unacct_error: if (charged) vm_unacct_memory(charged); return error; } |
db4fbfb95 mm: vm_unmapped_a... |
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 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 |
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 Linux-2.6.12-rc2 |
1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 |
/* 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 mm: vm_unmapped_a... |
1803 |
struct vm_unmapped_area_info info; |
1da177e4c Linux-2.6.12-rc2 |
1804 1805 1806 |
if (len > TASK_SIZE) return -ENOMEM; |
06abdfb47 get_unmapped_area... |
1807 1808 |
if (flags & MAP_FIXED) return addr; |
1da177e4c Linux-2.6.12-rc2 |
1809 1810 1811 1812 1813 1814 1815 |
if (addr) { addr = PAGE_ALIGN(addr); vma = find_vma(mm, addr); if (TASK_SIZE - len >= addr && (!vma || addr + len <= vma->vm_start)) return addr; } |
1da177e4c Linux-2.6.12-rc2 |
1816 |
|
db4fbfb95 mm: vm_unmapped_a... |
1817 1818 1819 1820 1821 1822 |
info.flags = 0; info.length = len; info.low_limit = TASK_UNMAPPED_BASE; info.high_limit = TASK_SIZE; info.align_mask = 0; return vm_unmapped_area(&info); |
1da177e4c Linux-2.6.12-rc2 |
1823 1824 |
} #endif |
1363c3cd8 [PATCH] Avoiding ... |
1825 |
void arch_unmap_area(struct mm_struct *mm, unsigned long addr) |
1da177e4c Linux-2.6.12-rc2 |
1826 1827 1828 1829 |
{ /* * Is this a new hole at the lowest possible address? */ |
f44d21985 mm: do not reset ... |
1830 |
if (addr >= TASK_UNMAPPED_BASE && addr < mm->free_area_cache) |
1363c3cd8 [PATCH] Avoiding ... |
1831 |
mm->free_area_cache = addr; |
1da177e4c Linux-2.6.12-rc2 |
1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 |
} /* * 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 mm: vm_unmapped_a... |
1846 1847 |
unsigned long addr = addr0; struct vm_unmapped_area_info info; |
1da177e4c Linux-2.6.12-rc2 |
1848 1849 1850 1851 |
/* requested length too big for entire address space */ if (len > TASK_SIZE) return -ENOMEM; |
06abdfb47 get_unmapped_area... |
1852 1853 |
if (flags & MAP_FIXED) return addr; |
1da177e4c Linux-2.6.12-rc2 |
1854 1855 1856 1857 1858 1859 1860 1861 |
/* requesting a specific address */ if (addr) { addr = PAGE_ALIGN(addr); vma = find_vma(mm, addr); if (TASK_SIZE - len >= addr && (!vma || addr + len <= vma->vm_start)) return addr; } |
db4fbfb95 mm: vm_unmapped_a... |
1862 1863 1864 1865 1866 1867 |
info.flags = VM_UNMAPPED_AREA_TOPDOWN; info.length = len; info.low_limit = PAGE_SIZE; info.high_limit = mm->mmap_base; info.align_mask = 0; addr = vm_unmapped_area(&info); |
b716ad953 mm: search from f... |
1868 |
|
1da177e4c Linux-2.6.12-rc2 |
1869 1870 1871 1872 1873 1874 |
/* * 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. */ |
db4fbfb95 mm: vm_unmapped_a... |
1875 1876 1877 1878 1879 1880 1881 |
if (addr & ~PAGE_MASK) { 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 Linux-2.6.12-rc2 |
1882 1883 1884 1885 |
return addr; } #endif |
1363c3cd8 [PATCH] Avoiding ... |
1886 |
void arch_unmap_area_topdown(struct mm_struct *mm, unsigned long addr) |
1da177e4c Linux-2.6.12-rc2 |
1887 1888 1889 1890 |
{ /* * Is this a new hole at the highest possible address? */ |
1363c3cd8 [PATCH] Avoiding ... |
1891 1892 |
if (addr > mm->free_area_cache) mm->free_area_cache = addr; |
1da177e4c Linux-2.6.12-rc2 |
1893 1894 |
/* dont allow allocations above current base */ |
1363c3cd8 [PATCH] Avoiding ... |
1895 1896 |
if (mm->free_area_cache > mm->mmap_base) mm->free_area_cache = mm->mmap_base; |
1da177e4c Linux-2.6.12-rc2 |
1897 1898 1899 1900 1901 1902 |
} unsigned long get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { |
06abdfb47 get_unmapped_area... |
1903 1904 |
unsigned long (*get_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long); |
9206de95b Take arch_mmap_ch... |
1905 1906 1907 1908 1909 1910 1911 |
unsigned long error = arch_mmap_check(addr, len, flags); if (error) return error; /* Careful about overflows.. */ if (len > TASK_SIZE) return -ENOMEM; |
06abdfb47 get_unmapped_area... |
1912 1913 1914 1915 1916 1917 |
get_area = current->mm->get_unmapped_area; if (file && file->f_op && file->f_op->get_unmapped_area) get_area = file->f_op->get_unmapped_area; addr = get_area(file, addr, len, pgoff, flags); if (IS_ERR_VALUE(addr)) return addr; |
1da177e4c Linux-2.6.12-rc2 |
1918 |
|
07ab67c8d Fix get_unmapped_... |
1919 1920 1921 1922 |
if (addr > TASK_SIZE - len) return -ENOMEM; if (addr & ~PAGE_MASK) return -EINVAL; |
06abdfb47 get_unmapped_area... |
1923 |
|
9ac4ed4bd move security_mma... |
1924 1925 1926 |
addr = arch_rebalance_pgtables(addr, len); error = security_mmap_addr(addr); return error ? error : addr; |
1da177e4c Linux-2.6.12-rc2 |
1927 1928 1929 1930 1931 |
} EXPORT_SYMBOL(get_unmapped_area); /* Look up the first VMA which satisfies addr < vm_end, NULL if none. */ |
48aae4255 mm/mmap.c: fix co... |
1932 |
struct vm_area_struct *find_vma(struct mm_struct *mm, unsigned long addr) |
1da177e4c Linux-2.6.12-rc2 |
1933 1934 |
{ struct vm_area_struct *vma = NULL; |
841e31e5c mm/mmap.c: find_v... |
1935 1936 |
/* Check the cache first. */ /* (Cache hit rate is typically around 35%.) */ |
b6a9b7f6b mm: prevent mmap_... |
1937 |
vma = ACCESS_ONCE(mm->mmap_cache); |
841e31e5c mm/mmap.c: find_v... |
1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 |
if (!(vma && vma->vm_end > addr && vma->vm_start <= addr)) { struct rb_node *rb_node; rb_node = mm->mm_rb.rb_node; vma = NULL; while (rb_node) { struct vm_area_struct *vma_tmp; vma_tmp = rb_entry(rb_node, struct vm_area_struct, vm_rb); if (vma_tmp->vm_end > addr) { vma = vma_tmp; if (vma_tmp->vm_start <= addr) break; rb_node = rb_node->rb_left; } else rb_node = rb_node->rb_right; |
1da177e4c Linux-2.6.12-rc2 |
1957 |
} |
841e31e5c mm/mmap.c: find_v... |
1958 1959 |
if (vma) mm->mmap_cache = vma; |
1da177e4c Linux-2.6.12-rc2 |
1960 1961 1962 1963 1964 |
} return vma; } EXPORT_SYMBOL(find_vma); |
6bd4837de mm: simplify find... |
1965 1966 |
/* * Same as find_vma, but also return a pointer to the previous VMA in *pprev. |
6bd4837de mm: simplify find... |
1967 |
*/ |
1da177e4c Linux-2.6.12-rc2 |
1968 1969 1970 1971 |
struct vm_area_struct * find_vma_prev(struct mm_struct *mm, unsigned long addr, struct vm_area_struct **pprev) { |
6bd4837de mm: simplify find... |
1972 |
struct vm_area_struct *vma; |
1da177e4c Linux-2.6.12-rc2 |
1973 |
|
6bd4837de mm: simplify find... |
1974 |
vma = find_vma(mm, addr); |
83cd904d2 mm: fix find_vma_... |
1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 |
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 mm: simplify find... |
1985 |
return vma; |
1da177e4c Linux-2.6.12-rc2 |
1986 1987 1988 1989 1990 1991 1992 |
} /* * Verify that the stack growth is acceptable and * update accounting. This is shared with both the * grow-up and grow-down cases. */ |
48aae4255 mm/mmap.c: fix co... |
1993 |
static int acct_stack_growth(struct vm_area_struct *vma, unsigned long size, unsigned long grow) |
1da177e4c Linux-2.6.12-rc2 |
1994 1995 1996 |
{ struct mm_struct *mm = vma->vm_mm; struct rlimit *rlim = current->signal->rlim; |
0d59a01bc [PATCH] Don't all... |
1997 |
unsigned long new_start; |
1da177e4c Linux-2.6.12-rc2 |
1998 1999 |
/* address space limit tests */ |
119f657c7 [PATCH] RLIMIT_AS... |
2000 |
if (!may_expand_vm(mm, grow)) |
1da177e4c Linux-2.6.12-rc2 |
2001 2002 2003 |
return -ENOMEM; /* Stack limit test */ |
59e99e5b9 mm: use rlimit he... |
2004 |
if (size > ACCESS_ONCE(rlim[RLIMIT_STACK].rlim_cur)) |
1da177e4c Linux-2.6.12-rc2 |
2005 2006 2007 2008 2009 2010 2011 |
return -ENOMEM; /* mlock limit tests */ if (vma->vm_flags & VM_LOCKED) { unsigned long locked; unsigned long limit; locked = mm->locked_vm + grow; |
59e99e5b9 mm: use rlimit he... |
2012 2013 |
limit = ACCESS_ONCE(rlim[RLIMIT_MEMLOCK].rlim_cur); limit >>= PAGE_SHIFT; |
1da177e4c Linux-2.6.12-rc2 |
2014 2015 2016 |
if (locked > limit && !capable(CAP_IPC_LOCK)) return -ENOMEM; } |
0d59a01bc [PATCH] Don't all... |
2017 2018 2019 2020 2021 |
/* 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 Linux-2.6.12-rc2 |
2022 2023 2024 2025 |
/* * Overcommit.. This must be the final test, as it will * update security statistics. */ |
05fa199d4 mm: pass correct ... |
2026 |
if (security_vm_enough_memory_mm(mm, grow)) |
1da177e4c Linux-2.6.12-rc2 |
2027 2028 2029 |
return -ENOMEM; /* Ok, everything looks good - let it rip */ |
1da177e4c Linux-2.6.12-rc2 |
2030 2031 |
if (vma->vm_flags & VM_LOCKED) mm->locked_vm += grow; |
ab50b8ed8 [PATCH] mm: vm_st... |
2032 |
vm_stat_account(mm, vma->vm_flags, vma->vm_file, grow); |
1da177e4c Linux-2.6.12-rc2 |
2033 2034 |
return 0; } |
46dea3d09 [PATCH] mm: ia64 ... |
2035 |
#if defined(CONFIG_STACK_GROWSUP) || defined(CONFIG_IA64) |
1da177e4c Linux-2.6.12-rc2 |
2036 |
/* |
46dea3d09 [PATCH] mm: ia64 ... |
2037 2038 |
* 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 Linux-2.6.12-rc2 |
2039 |
*/ |
46dea3d09 [PATCH] mm: ia64 ... |
2040 |
int expand_upwards(struct vm_area_struct *vma, unsigned long address) |
1da177e4c Linux-2.6.12-rc2 |
2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 |
{ int error; if (!(vma->vm_flags & VM_GROWSUP)) return -EFAULT; /* * We must make sure the anon_vma is allocated * so that the anon_vma locking is not a noop. */ if (unlikely(anon_vma_prepare(vma))) return -ENOMEM; |
bb4a340e0 mm: rename anon_v... |
2053 |
vma_lock_anon_vma(vma); |
1da177e4c Linux-2.6.12-rc2 |
2054 2055 2056 2057 2058 |
/* * 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. |
06b32f3ab [PARISC] Handle w... |
2059 |
* Also guard against wrapping around to address 0. |
1da177e4c Linux-2.6.12-rc2 |
2060 |
*/ |
06b32f3ab [PARISC] Handle w... |
2061 2062 2063 |
if (address < PAGE_ALIGN(address+4)) address = PAGE_ALIGN(address+4); else { |
bb4a340e0 mm: rename anon_v... |
2064 |
vma_unlock_anon_vma(vma); |
06b32f3ab [PARISC] Handle w... |
2065 2066 |
return -ENOMEM; } |
1da177e4c Linux-2.6.12-rc2 |
2067 2068 2069 2070 2071 2072 2073 2074 |
error = 0; /* Somebody else might have raced and expanded it already */ if (address > vma->vm_end) { unsigned long size, grow; size = address - vma->vm_start; grow = (address - vma->vm_end) >> PAGE_SHIFT; |
42c36f63a vm: fix vm_pgoff ... |
2075 2076 2077 2078 |
error = -ENOMEM; if (vma->vm_pgoff + (size >> PAGE_SHIFT) >= vma->vm_pgoff) { error = acct_stack_growth(vma, size, grow); if (!error) { |
4128997b5 mm: protect again... |
2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 |
/* * 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. * vma_lock_anon_vma() doesn't help here, as * 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. */ spin_lock(&vma->vm_mm->page_table_lock); |
bf181b9f9 mm anon rmap: rep... |
2091 |
anon_vma_interval_tree_pre_update_vma(vma); |
42c36f63a vm: fix vm_pgoff ... |
2092 |
vma->vm_end = address; |
bf181b9f9 mm anon rmap: rep... |
2093 |
anon_vma_interval_tree_post_update_vma(vma); |
d37371870 mm: augment vma r... |
2094 2095 2096 2097 |
if (vma->vm_next) vma_gap_update(vma->vm_next); else vma->vm_mm->highest_vm_end = address; |
4128997b5 mm: protect again... |
2098 |
spin_unlock(&vma->vm_mm->page_table_lock); |
42c36f63a vm: fix vm_pgoff ... |
2099 2100 |
perf_event_mmap(vma); } |
3af9e8592 perf: Add non-exe... |
2101 |
} |
1da177e4c Linux-2.6.12-rc2 |
2102 |
} |
bb4a340e0 mm: rename anon_v... |
2103 |
vma_unlock_anon_vma(vma); |
b15d00b6a thp: khugepaged v... |
2104 |
khugepaged_enter_vma_merge(vma); |
ed8ea8150 mm: add CONFIG_DE... |
2105 |
validate_mm(vma->vm_mm); |
1da177e4c Linux-2.6.12-rc2 |
2106 2107 |
return error; } |
46dea3d09 [PATCH] mm: ia64 ... |
2108 |
#endif /* CONFIG_STACK_GROWSUP || CONFIG_IA64 */ |
1da177e4c Linux-2.6.12-rc2 |
2109 2110 2111 |
/* * vma is the first one with address < vma->vm_start. Have to extend vma. */ |
d05f3169c mm: make expand_d... |
2112 |
int expand_downwards(struct vm_area_struct *vma, |
b6a2fea39 mm: variable leng... |
2113 |
unsigned long address) |
1da177e4c Linux-2.6.12-rc2 |
2114 2115 2116 2117 2118 2119 2120 2121 2122 |
{ int error; /* * We must make sure the anon_vma is allocated * so that the anon_vma locking is not a noop. */ if (unlikely(anon_vma_prepare(vma))) return -ENOMEM; |
8869477a4 security: protect... |
2123 2124 |
address &= PAGE_MASK; |
e5467859f split ->file_mmap... |
2125 |
error = security_mmap_addr(address); |
8869477a4 security: protect... |
2126 2127 |
if (error) return error; |
bb4a340e0 mm: rename anon_v... |
2128 |
vma_lock_anon_vma(vma); |
1da177e4c Linux-2.6.12-rc2 |
2129 2130 2131 2132 2133 2134 |
/* * 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. */ |
1da177e4c Linux-2.6.12-rc2 |
2135 2136 2137 2138 2139 2140 2141 |
/* 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 vm: fix vm_pgoff ... |
2142 2143 2144 2145 |
error = -ENOMEM; if (grow <= vma->vm_pgoff) { error = acct_stack_growth(vma, size, grow); if (!error) { |
4128997b5 mm: protect again... |
2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 |
/* * 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. * vma_lock_anon_vma() doesn't help here, as * 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. */ spin_lock(&vma->vm_mm->page_table_lock); |
bf181b9f9 mm anon rmap: rep... |
2158 |
anon_vma_interval_tree_pre_update_vma(vma); |
a626ca6a6 vm: fix vm_pgoff ... |
2159 2160 |
vma->vm_start = address; vma->vm_pgoff -= grow; |
bf181b9f9 mm anon rmap: rep... |
2161 |
anon_vma_interval_tree_post_update_vma(vma); |
d37371870 mm: augment vma r... |
2162 |
vma_gap_update(vma); |
4128997b5 mm: protect again... |
2163 |
spin_unlock(&vma->vm_mm->page_table_lock); |
a626ca6a6 vm: fix vm_pgoff ... |
2164 2165 |
perf_event_mmap(vma); } |
1da177e4c Linux-2.6.12-rc2 |
2166 2167 |
} } |
bb4a340e0 mm: rename anon_v... |
2168 |
vma_unlock_anon_vma(vma); |
b15d00b6a thp: khugepaged v... |
2169 |
khugepaged_enter_vma_merge(vma); |
ed8ea8150 mm: add CONFIG_DE... |
2170 |
validate_mm(vma->vm_mm); |
1da177e4c Linux-2.6.12-rc2 |
2171 2172 |
return error; } |
098849643 mm: do not grow t... |
2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 |
/* * 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 mm: variable leng... |
2184 2185 2186 |
#ifdef CONFIG_STACK_GROWSUP int expand_stack(struct vm_area_struct *vma, unsigned long address) { |
098849643 mm: do not grow t... |
2187 2188 2189 2190 2191 2192 2193 2194 |
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 mm: variable leng... |
2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 |
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 parisc: fix find_... |
2207 |
if (!prev || expand_stack(prev, addr)) |
b6a2fea39 mm: variable leng... |
2208 |
return NULL; |
cea10a19b mm: directly use ... |
2209 2210 |
if (prev->vm_flags & VM_LOCKED) __mlock_vma_pages_range(prev, addr, prev->vm_end, NULL); |
b6a2fea39 mm: variable leng... |
2211 2212 2213 2214 2215 |
return prev; } #else int expand_stack(struct vm_area_struct *vma, unsigned long address) { |
098849643 mm: do not grow t... |
2216 2217 2218 2219 2220 2221 2222 2223 |
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 mm: variable leng... |
2224 2225 |
return expand_downwards(vma, address); } |
1da177e4c Linux-2.6.12-rc2 |
2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 |
struct vm_area_struct * find_extend_vma(struct mm_struct * mm, unsigned long addr) { struct vm_area_struct * vma; unsigned long start; addr &= PAGE_MASK; vma = find_vma(mm,addr); if (!vma) return NULL; if (vma->vm_start <= addr) return vma; if (!(vma->vm_flags & VM_GROWSDOWN)) return NULL; start = vma->vm_start; if (expand_stack(vma, addr)) return NULL; |
cea10a19b mm: directly use ... |
2243 2244 |
if (vma->vm_flags & VM_LOCKED) __mlock_vma_pages_range(vma, addr, start, NULL); |
1da177e4c Linux-2.6.12-rc2 |
2245 2246 2247 |
return vma; } #endif |
1da177e4c Linux-2.6.12-rc2 |
2248 |
/* |
2c0b38146 [PATCH] mm: remov... |
2249 |
* Ok - we have the memory areas we should free on the vma list, |
1da177e4c Linux-2.6.12-rc2 |
2250 |
* so release them, and do the vma updates. |
2c0b38146 [PATCH] mm: remov... |
2251 2252 |
* * Called with the mm semaphore held. |
1da177e4c Linux-2.6.12-rc2 |
2253 |
*/ |
2c0b38146 [PATCH] mm: remov... |
2254 |
static void remove_vma_list(struct mm_struct *mm, struct vm_area_struct *vma) |
1da177e4c Linux-2.6.12-rc2 |
2255 |
{ |
4f74d2c8e vm: remove 'nr_ac... |
2256 |
unsigned long nr_accounted = 0; |
365e9c87a [PATCH] mm: updat... |
2257 2258 |
/* Update high watermark before we lower total_vm */ update_hiwater_vm(mm); |
1da177e4c Linux-2.6.12-rc2 |
2259 |
do { |
2c0b38146 [PATCH] mm: remov... |
2260 |
long nrpages = vma_pages(vma); |
4f74d2c8e vm: remove 'nr_ac... |
2261 2262 |
if (vma->vm_flags & VM_ACCOUNT) nr_accounted += nrpages; |
2c0b38146 [PATCH] mm: remov... |
2263 |
vm_stat_account(mm, vma->vm_flags, vma->vm_file, -nrpages); |
a8fb5618d [PATCH] mm: unlin... |
2264 |
vma = remove_vma(vma); |
146425a31 [PATCH] freepgt: ... |
2265 |
} while (vma); |
4f74d2c8e vm: remove 'nr_ac... |
2266 |
vm_unacct_memory(nr_accounted); |
1da177e4c Linux-2.6.12-rc2 |
2267 2268 2269 2270 2271 2272 |
validate_mm(mm); } /* * Get rid of page table information in the indicated region. * |
f10df6860 [PATCH] fix locki... |
2273 |
* Called with the mm semaphore held. |
1da177e4c Linux-2.6.12-rc2 |
2274 2275 |
*/ static void unmap_region(struct mm_struct *mm, |
e0da382c9 [PATCH] freepgt: ... |
2276 2277 |
struct vm_area_struct *vma, struct vm_area_struct *prev, unsigned long start, unsigned long end) |
1da177e4c Linux-2.6.12-rc2 |
2278 |
{ |
e0da382c9 [PATCH] freepgt: ... |
2279 |
struct vm_area_struct *next = prev? prev->vm_next: mm->mmap; |
d16dfc550 mm: mmu_gather re... |
2280 |
struct mmu_gather tlb; |
1da177e4c Linux-2.6.12-rc2 |
2281 2282 |
lru_add_drain(); |
8e220cfd1 Fix TLB gather vi... |
2283 |
tlb_gather_mmu(&tlb, mm, start, end); |
365e9c87a [PATCH] mm: updat... |
2284 |
update_hiwater_rss(mm); |
4f74d2c8e vm: remove 'nr_ac... |
2285 |
unmap_vmas(&tlb, vma, start, end); |
d16dfc550 mm: mmu_gather re... |
2286 |
free_pgtables(&tlb, vma, prev ? prev->vm_end : FIRST_USER_ADDRESS, |
6ee8630e0 mm: allow arch co... |
2287 |
next ? next->vm_start : USER_PGTABLES_CEILING); |
d16dfc550 mm: mmu_gather re... |
2288 |
tlb_finish_mmu(&tlb, start, end); |
1da177e4c Linux-2.6.12-rc2 |
2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 |
} /* * 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; |
1363c3cd8 [PATCH] Avoiding ... |
2301 |
unsigned long addr; |
1da177e4c Linux-2.6.12-rc2 |
2302 2303 |
insertion_point = (prev ? &prev->vm_next : &mm->mmap); |
297c5eee3 mm: make the vma ... |
2304 |
vma->vm_prev = NULL; |
1da177e4c Linux-2.6.12-rc2 |
2305 |
do { |
d37371870 mm: augment vma r... |
2306 |
vma_rb_erase(vma, &mm->mm_rb); |
1da177e4c Linux-2.6.12-rc2 |
2307 2308 2309 2310 2311 |
mm->map_count--; tail_vma = vma; vma = vma->vm_next; } while (vma && vma->vm_start < end); *insertion_point = vma; |
d37371870 mm: augment vma r... |
2312 |
if (vma) { |
297c5eee3 mm: make the vma ... |
2313 |
vma->vm_prev = prev; |
d37371870 mm: augment vma r... |
2314 2315 2316 |
vma_gap_update(vma); } else mm->highest_vm_end = prev ? prev->vm_end : 0; |
1da177e4c Linux-2.6.12-rc2 |
2317 |
tail_vma->vm_next = NULL; |
1363c3cd8 [PATCH] Avoiding ... |
2318 2319 2320 2321 2322 |
if (mm->unmap_area == arch_unmap_area) addr = prev ? prev->vm_end : mm->mmap_base; else addr = vma ? vma->vm_start : mm->mmap_base; mm->unmap_area(mm, addr); |
1da177e4c Linux-2.6.12-rc2 |
2323 2324 2325 2326 |
mm->mmap_cache = NULL; /* Kill the cache. */ } /* |
659ace584 mmap: don't retur... |
2327 2328 |
* __split_vma() bypasses sysctl_max_map_count checking. We use this on the * munmap path where it doesn't make sense to fail. |
1da177e4c Linux-2.6.12-rc2 |
2329 |
*/ |
659ace584 mmap: don't retur... |
2330 |
static int __split_vma(struct mm_struct * mm, struct vm_area_struct * vma, |
1da177e4c Linux-2.6.12-rc2 |
2331 2332 2333 2334 |
unsigned long addr, int new_below) { struct mempolicy *pol; struct vm_area_struct *new; |
5beb49305 mm: change anon_v... |
2335 |
int err = -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
2336 |
|
a55164389 hugetlb: modular ... |
2337 2338 |
if (is_vm_hugetlb_page(vma) && (addr & ~(huge_page_mask(hstate_vma(vma))))) |
1da177e4c Linux-2.6.12-rc2 |
2339 |
return -EINVAL; |
e94b17660 [PATCH] slab: rem... |
2340 |
new = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); |
1da177e4c Linux-2.6.12-rc2 |
2341 |
if (!new) |
5beb49305 mm: change anon_v... |
2342 |
goto out_err; |
1da177e4c Linux-2.6.12-rc2 |
2343 2344 2345 |
/* most fields are the same, copy all, and then fixup */ *new = *vma; |
5beb49305 mm: change anon_v... |
2346 |
INIT_LIST_HEAD(&new->anon_vma_chain); |
1da177e4c Linux-2.6.12-rc2 |
2347 2348 2349 2350 2351 2352 |
if (new_below) new->vm_end = addr; else { new->vm_start = addr; new->vm_pgoff += ((addr - vma->vm_start) >> PAGE_SHIFT); } |
846a16bf0 mempolicy: rename... |
2353 |
pol = mpol_dup(vma_policy(vma)); |
1da177e4c Linux-2.6.12-rc2 |
2354 |
if (IS_ERR(pol)) { |
5beb49305 mm: change anon_v... |
2355 2356 |
err = PTR_ERR(pol); goto out_free_vma; |
1da177e4c Linux-2.6.12-rc2 |
2357 2358 |
} vma_set_policy(new, pol); |
5beb49305 mm: change anon_v... |
2359 2360 |
if (anon_vma_clone(new, vma)) goto out_free_mpol; |
e9714acf8 mm: kill vma flag... |
2361 |
if (new->vm_file) |
1da177e4c Linux-2.6.12-rc2 |
2362 2363 2364 2365 2366 2367 |
get_file(new->vm_file); if (new->vm_ops && new->vm_ops->open) new->vm_ops->open(new); if (new_below) |
5beb49305 mm: change anon_v... |
2368 |
err = vma_adjust(vma, addr, vma->vm_end, vma->vm_pgoff + |
1da177e4c Linux-2.6.12-rc2 |
2369 2370 |
((addr - new->vm_start) >> PAGE_SHIFT), new); else |
5beb49305 mm: change anon_v... |
2371 |
err = vma_adjust(vma, vma->vm_start, addr, vma->vm_pgoff, new); |
1da177e4c Linux-2.6.12-rc2 |
2372 |
|
5beb49305 mm: change anon_v... |
2373 2374 2375 2376 2377 |
/* Success. */ if (!err) return 0; /* Clean everything up if vma_adjust failed. */ |
589275338 mmap: check ->vm_... |
2378 2379 |
if (new->vm_ops && new->vm_ops->close) new->vm_ops->close(new); |
e9714acf8 mm: kill vma flag... |
2380 |
if (new->vm_file) |
5beb49305 mm: change anon_v... |
2381 |
fput(new->vm_file); |
2aeadc30d mmap: call unlink... |
2382 |
unlink_anon_vmas(new); |
5beb49305 mm: change anon_v... |
2383 2384 2385 2386 2387 2388 |
out_free_mpol: mpol_put(pol); out_free_vma: kmem_cache_free(vm_area_cachep, new); out_err: return err; |
1da177e4c Linux-2.6.12-rc2 |
2389 |
} |
659ace584 mmap: don't retur... |
2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 |
/* * 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 Linux-2.6.12-rc2 |
2402 2403 2404 2405 2406 2407 2408 2409 |
/* 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 [PATCH] freepgt: ... |
2410 |
struct vm_area_struct *vma, *prev, *last; |
1da177e4c Linux-2.6.12-rc2 |
2411 2412 2413 2414 2415 2416 2417 2418 |
if ((start & ~PAGE_MASK) || start > TASK_SIZE || len > TASK_SIZE-start) return -EINVAL; if ((len = PAGE_ALIGN(len)) == 0) return -EINVAL; /* Find the first overlapping VMA */ |
9be34c9d5 mm: get rid of th... |
2419 |
vma = find_vma(mm, start); |
146425a31 [PATCH] freepgt: ... |
2420 |
if (!vma) |
1da177e4c Linux-2.6.12-rc2 |
2421 |
return 0; |
9be34c9d5 mm: get rid of th... |
2422 |
prev = vma->vm_prev; |
146425a31 [PATCH] freepgt: ... |
2423 |
/* we have start < vma->vm_end */ |
1da177e4c Linux-2.6.12-rc2 |
2424 2425 2426 |
/* if it doesn't overlap, we have nothing.. */ end = start + len; |
146425a31 [PATCH] freepgt: ... |
2427 |
if (vma->vm_start >= end) |
1da177e4c Linux-2.6.12-rc2 |
2428 2429 2430 2431 2432 2433 2434 2435 2436 |
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 [PATCH] freepgt: ... |
2437 |
if (start > vma->vm_start) { |
659ace584 mmap: don't retur... |
2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 |
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 Linux-2.6.12-rc2 |
2449 2450 |
if (error) return error; |
146425a31 [PATCH] freepgt: ... |
2451 |
prev = vma; |
1da177e4c Linux-2.6.12-rc2 |
2452 2453 2454 2455 2456 |
} /* Does it split the last one? */ last = find_vma(mm, end); if (last && end > last->vm_start) { |
659ace584 mmap: don't retur... |
2457 |
int error = __split_vma(mm, last, end, 1); |
1da177e4c Linux-2.6.12-rc2 |
2458 2459 2460 |
if (error) return error; } |
146425a31 [PATCH] freepgt: ... |
2461 |
vma = prev? prev->vm_next: mm->mmap; |
1da177e4c Linux-2.6.12-rc2 |
2462 2463 |
/* |
ba470de43 mmap: handle mloc... |
2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 |
* 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 Linux-2.6.12-rc2 |
2478 2479 |
* Remove the vma's, and unmap the actual pages */ |
146425a31 [PATCH] freepgt: ... |
2480 2481 |
detach_vmas_to_be_unmapped(mm, vma, prev, end); unmap_region(mm, vma, prev, start, end); |
1da177e4c Linux-2.6.12-rc2 |
2482 2483 |
/* Fix up all other VM information */ |
2c0b38146 [PATCH] mm: remov... |
2484 |
remove_vma_list(mm, vma); |
1da177e4c Linux-2.6.12-rc2 |
2485 2486 2487 |
return 0; } |
1da177e4c Linux-2.6.12-rc2 |
2488 |
|
bfce281c2 kill mm argument ... |
2489 |
int vm_munmap(unsigned long start, size_t len) |
1da177e4c Linux-2.6.12-rc2 |
2490 2491 |
{ int ret; |
bfce281c2 kill mm argument ... |
2492 |
struct mm_struct *mm = current->mm; |
1da177e4c Linux-2.6.12-rc2 |
2493 2494 |
down_write(&mm->mmap_sem); |
a46ef99d8 VM: add "vm_munma... |
2495 |
ret = do_munmap(mm, start, len); |
1da177e4c Linux-2.6.12-rc2 |
2496 2497 2498 |
up_write(&mm->mmap_sem); return ret; } |
a46ef99d8 VM: add "vm_munma... |
2499 2500 2501 2502 2503 |
EXPORT_SYMBOL(vm_munmap); SYSCALL_DEFINE2(munmap, unsigned long, addr, size_t, len) { profile_munmap(addr); |
bfce281c2 kill mm argument ... |
2504 |
return vm_munmap(addr, len); |
a46ef99d8 VM: add "vm_munma... |
2505 |
} |
1da177e4c Linux-2.6.12-rc2 |
2506 2507 2508 |
static inline void verify_mm_writelocked(struct mm_struct *mm) { |
a241ec65a [PATCH] RCU tortu... |
2509 |
#ifdef CONFIG_DEBUG_VM |
1da177e4c Linux-2.6.12-rc2 |
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 VM: add "vm_brk()... |
2522 |
static unsigned long do_brk(unsigned long addr, unsigned long len) |
1da177e4c Linux-2.6.12-rc2 |
2523 2524 2525 2526 2527 2528 |
{ struct mm_struct * mm = current->mm; struct vm_area_struct * vma, * prev; unsigned long flags; struct rb_node ** rb_link, * rb_parent; pgoff_t pgoff = addr >> PAGE_SHIFT; |
3a4597568 [PATCH] IA64,spar... |
2529 |
int error; |
1da177e4c Linux-2.6.12-rc2 |
2530 2531 2532 2533 |
len = PAGE_ALIGN(len); if (!len) return addr; |
3a4597568 [PATCH] IA64,spar... |
2534 |
flags = VM_DATA_DEFAULT_FLAGS | VM_ACCOUNT | mm->def_flags; |
2c6a10161 switch do_brk() t... |
2535 2536 |
error = get_unmapped_area(NULL, addr, len, 0, MAP_FIXED); if (error & ~PAGE_MASK) |
3a4597568 [PATCH] IA64,spar... |
2537 |
return error; |
1da177e4c Linux-2.6.12-rc2 |
2538 2539 2540 2541 2542 |
/* * mlock MCL_FUTURE? */ if (mm->def_flags & VM_LOCKED) { unsigned long locked, lock_limit; |
93ea1d0a1 [PATCH] RLIMIT_ME... |
2543 2544 |
locked = len >> PAGE_SHIFT; locked += mm->locked_vm; |
59e99e5b9 mm: use rlimit he... |
2545 |
lock_limit = rlimit(RLIMIT_MEMLOCK); |
93ea1d0a1 [PATCH] RLIMIT_ME... |
2546 |
lock_limit >>= PAGE_SHIFT; |
1da177e4c Linux-2.6.12-rc2 |
2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 |
if (locked > lock_limit && !capable(CAP_IPC_LOCK)) return -EAGAIN; } /* * mm->mmap_sem is required to protect against another thread * changing the mappings in case we sleep. */ verify_mm_writelocked(mm); /* * Clear old maps. this also does some error checking for us */ munmap_back: |
6597d7833 mm/mmap.c: replac... |
2561 |
if (find_vma_links(mm, addr, addr + len, &prev, &rb_link, &rb_parent)) { |
1da177e4c Linux-2.6.12-rc2 |
2562 2563 2564 2565 2566 2567 |
if (do_munmap(mm, addr, len)) return -ENOMEM; goto munmap_back; } /* Check against address space limits *after* clearing old maps... */ |
119f657c7 [PATCH] RLIMIT_AS... |
2568 |
if (!may_expand_vm(mm, len >> PAGE_SHIFT)) |
1da177e4c Linux-2.6.12-rc2 |
2569 2570 2571 2572 |
return -ENOMEM; if (mm->map_count > sysctl_max_map_count) return -ENOMEM; |
191c54244 mm: collapse secu... |
2573 |
if (security_vm_enough_memory_mm(mm, len >> PAGE_SHIFT)) |
1da177e4c Linux-2.6.12-rc2 |
2574 |
return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
2575 |
/* Can we just expand an old private anonymous mapping? */ |
ba470de43 mmap: handle mloc... |
2576 2577 2578 |
vma = vma_merge(mm, prev, addr, addr + len, flags, NULL, NULL, pgoff, NULL); if (vma) |
1da177e4c Linux-2.6.12-rc2 |
2579 2580 2581 2582 2583 |
goto out; /* * create a vma struct for an anonymous mapping */ |
c5e3b83e9 [PATCH] mm: use k... |
2584 |
vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); |
1da177e4c Linux-2.6.12-rc2 |
2585 2586 2587 2588 |
if (!vma) { vm_unacct_memory(len >> PAGE_SHIFT); return -ENOMEM; } |
1da177e4c Linux-2.6.12-rc2 |
2589 |
|
5beb49305 mm: change anon_v... |
2590 |
INIT_LIST_HEAD(&vma->anon_vma_chain); |
1da177e4c Linux-2.6.12-rc2 |
2591 2592 2593 2594 2595 |
vma->vm_mm = mm; vma->vm_start = addr; vma->vm_end = addr + len; vma->vm_pgoff = pgoff; vma->vm_flags = flags; |
3ed75eb8f setup vma->vm_pag... |
2596 |
vma->vm_page_prot = vm_get_page_prot(flags); |
1da177e4c Linux-2.6.12-rc2 |
2597 2598 |
vma_link(mm, vma, prev, rb_link, rb_parent); out: |
3af9e8592 perf: Add non-exe... |
2599 |
perf_event_mmap(vma); |
1da177e4c Linux-2.6.12-rc2 |
2600 |
mm->total_vm += len >> PAGE_SHIFT; |
128557ffe mm: use mm_popula... |
2601 2602 |
if (flags & VM_LOCKED) mm->locked_vm += (len >> PAGE_SHIFT); |
1da177e4c Linux-2.6.12-rc2 |
2603 2604 |
return addr; } |
e4eb1ff61 VM: add "vm_brk()... |
2605 2606 2607 2608 |
unsigned long vm_brk(unsigned long addr, unsigned long len) { struct mm_struct *mm = current->mm; unsigned long ret; |
128557ffe mm: use mm_popula... |
2609 |
bool populate; |
e4eb1ff61 VM: add "vm_brk()... |
2610 2611 2612 |
down_write(&mm->mmap_sem); ret = do_brk(addr, len); |
128557ffe mm: use mm_popula... |
2613 |
populate = ((mm->def_flags & VM_LOCKED) != 0); |
e4eb1ff61 VM: add "vm_brk()... |
2614 |
up_write(&mm->mmap_sem); |
128557ffe mm: use mm_popula... |
2615 2616 |
if (populate) mm_populate(addr, len); |
e4eb1ff61 VM: add "vm_brk()... |
2617 2618 2619 |
return ret; } EXPORT_SYMBOL(vm_brk); |
1da177e4c Linux-2.6.12-rc2 |
2620 2621 2622 2623 |
/* Release all mmaps. */ void exit_mmap(struct mm_struct *mm) { |
d16dfc550 mm: mmu_gather re... |
2624 |
struct mmu_gather tlb; |
ba470de43 mmap: handle mloc... |
2625 |
struct vm_area_struct *vma; |
1da177e4c Linux-2.6.12-rc2 |
2626 |
unsigned long nr_accounted = 0; |
d6dd61c83 [PATCH] x86: PARA... |
2627 |
/* mm's last user has gone, and its about to be pulled down */ |
cddb8a5c1 mmu-notifiers: core |
2628 |
mmu_notifier_release(mm); |
d6dd61c83 [PATCH] x86: PARA... |
2629 |
|
ba470de43 mmap: handle mloc... |
2630 2631 2632 2633 2634 2635 2636 2637 |
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 mm: rearrange exi... |
2638 2639 |
arch_exit_mmap(mm); |
ba470de43 mmap: handle mloc... |
2640 |
vma = mm->mmap; |
9480c53e9 mm: rearrange exi... |
2641 2642 |
if (!vma) /* Can happen if dup_mmap() received an OOM */ return; |
1da177e4c Linux-2.6.12-rc2 |
2643 |
lru_add_drain(); |
1da177e4c Linux-2.6.12-rc2 |
2644 |
flush_cache_mm(mm); |
8e220cfd1 Fix TLB gather vi... |
2645 |
tlb_gather_mmu(&tlb, mm, 0, -1); |
901608d90 mm: introduce get... |
2646 |
/* update_hiwater_rss(mm) here? but nobody should be looking */ |
e0da382c9 [PATCH] freepgt: ... |
2647 |
/* Use -1 here to ensure all VMAs in the mm are unmapped */ |
4f74d2c8e vm: remove 'nr_ac... |
2648 |
unmap_vmas(&tlb, vma, 0, -1); |
9ba692948 ksm: fix oom dead... |
2649 |
|
6ee8630e0 mm: allow arch co... |
2650 |
free_pgtables(&tlb, vma, FIRST_USER_ADDRESS, USER_PGTABLES_CEILING); |
853f5e264 VM: don't bother ... |
2651 |
tlb_finish_mmu(&tlb, 0, -1); |
1da177e4c Linux-2.6.12-rc2 |
2652 |
|
1da177e4c Linux-2.6.12-rc2 |
2653 |
/* |
8f4f8c164 [PATCH] mm: unlin... |
2654 2655 |
* Walk the list again, actually closing and freeing it, * with preemption enabled, without holding any MM locks. |
1da177e4c Linux-2.6.12-rc2 |
2656 |
*/ |
4f74d2c8e vm: remove 'nr_ac... |
2657 2658 2659 |
while (vma) { if (vma->vm_flags & VM_ACCOUNT) nr_accounted += vma_pages(vma); |
a8fb5618d [PATCH] mm: unlin... |
2660 |
vma = remove_vma(vma); |
4f74d2c8e vm: remove 'nr_ac... |
2661 2662 |
} vm_unacct_memory(nr_accounted); |
e0da382c9 [PATCH] freepgt: ... |
2663 |
|
f9aed62a2 mm: change nr_pte... |
2664 |
WARN_ON(mm->nr_ptes > (FIRST_USER_ADDRESS+PMD_SIZE-1)>>PMD_SHIFT); |
1da177e4c Linux-2.6.12-rc2 |
2665 2666 2667 2668 |
} /* Insert vm structure into process list sorted by address * and into the inode's i_mmap tree. If vm_file is non-NULL |
3d48ae45e mm: Convert i_mma... |
2669 |
* then i_mmap_mutex is taken here. |
1da177e4c Linux-2.6.12-rc2 |
2670 |
*/ |
6597d7833 mm/mmap.c: replac... |
2671 |
int insert_vm_struct(struct mm_struct *mm, struct vm_area_struct *vma) |
1da177e4c Linux-2.6.12-rc2 |
2672 |
{ |
6597d7833 mm/mmap.c: replac... |
2673 2674 |
struct vm_area_struct *prev; struct rb_node **rb_link, *rb_parent; |
1da177e4c Linux-2.6.12-rc2 |
2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 |
/* * The vm_pgoff of a purely anonymous vma should be irrelevant * until its first write fault, when page's anon_vma and index * are set. But now set the vm_pgoff it will almost certainly * end up with (unless mremap moves it elsewhere before that * first wfault), so /proc/pid/maps tells a consistent story. * * By setting it to reflect the virtual start address of the * vma, merges and splits can happen in a seamless way, just * using the existing file pgoff checks and manipulations. * Similarly in do_mmap_pgoff and in do_brk. */ if (!vma->vm_file) { BUG_ON(vma->anon_vma); vma->vm_pgoff = vma->vm_start >> PAGE_SHIFT; } |
6597d7833 mm/mmap.c: replac... |
2692 2693 |
if (find_vma_links(mm, vma->vm_start, vma->vm_end, &prev, &rb_link, &rb_parent)) |
1da177e4c Linux-2.6.12-rc2 |
2694 |
return -ENOMEM; |
2fd4ef85e [PATCH] error pat... |
2695 |
if ((vma->vm_flags & VM_ACCOUNT) && |
34b4e4aa3 fix NULL pointer ... |
2696 |
security_vm_enough_memory_mm(mm, vma_pages(vma))) |
2fd4ef85e [PATCH] error pat... |
2697 |
return -ENOMEM; |
2b1444983 uprobes, mm, x86:... |
2698 |
|
1da177e4c Linux-2.6.12-rc2 |
2699 2700 2701 2702 2703 2704 2705 2706 2707 |
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 mm: avoid taking ... |
2708 2709 |
unsigned long addr, unsigned long len, pgoff_t pgoff, bool *need_rmap_locks) |
1da177e4c Linux-2.6.12-rc2 |
2710 2711 2712 2713 2714 2715 2716 |
{ struct vm_area_struct *vma = *vmap; unsigned long vma_start = vma->vm_start; struct mm_struct *mm = vma->vm_mm; struct vm_area_struct *new_vma, *prev; struct rb_node **rb_link, *rb_parent; struct mempolicy *pol; |
948f017b0 mremap: enforce r... |
2717 |
bool faulted_in_anon_vma = true; |
1da177e4c Linux-2.6.12-rc2 |
2718 2719 2720 2721 2722 |
/* * If anonymous vma has not yet been faulted, update new pgoff * to match new location, to increase its chance of merging. */ |
948f017b0 mremap: enforce r... |
2723 |
if (unlikely(!vma->vm_file && !vma->anon_vma)) { |
1da177e4c Linux-2.6.12-rc2 |
2724 |
pgoff = addr >> PAGE_SHIFT; |
948f017b0 mremap: enforce r... |
2725 2726 |
faulted_in_anon_vma = false; } |
1da177e4c Linux-2.6.12-rc2 |
2727 |
|
6597d7833 mm/mmap.c: replac... |
2728 2729 |
if (find_vma_links(mm, addr, addr + len, &prev, &rb_link, &rb_parent)) return NULL; /* should never get here */ |
1da177e4c Linux-2.6.12-rc2 |
2730 2731 2732 2733 2734 2735 |
new_vma = vma_merge(mm, prev, addr, addr + len, vma->vm_flags, vma->anon_vma, vma->vm_file, pgoff, vma_policy(vma)); if (new_vma) { /* * Source vma may have been merged into new_vma */ |
948f017b0 mremap: enforce r... |
2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 |
if (unlikely(vma_start >= new_vma->vm_start && vma_start < new_vma->vm_end)) { /* * The only way we can get a vma_merge with * self during an mremap is if the vma hasn't * been faulted in yet and we were allowed to * reset the dst vma->vm_pgoff to the * destination address of the mremap to allow * the merge to happen. mremap must change the * vm_pgoff linearity between src and dst vmas * (in turn preventing a vma_merge) to be * safe. It is only safe to keep the vm_pgoff * linear if there are no pages mapped yet. */ VM_BUG_ON(faulted_in_anon_vma); |
38a76013a mm: avoid taking ... |
2751 |
*vmap = vma = new_vma; |
108d6642a mm anon rmap: rem... |
2752 |
} |
38a76013a mm: avoid taking ... |
2753 |
*need_rmap_locks = (new_vma->vm_pgoff <= vma->vm_pgoff); |
1da177e4c Linux-2.6.12-rc2 |
2754 |
} else { |
e94b17660 [PATCH] slab: rem... |
2755 |
new_vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); |
1da177e4c Linux-2.6.12-rc2 |
2756 2757 |
if (new_vma) { *new_vma = *vma; |
523d4e200 mm anon rmap: in ... |
2758 2759 2760 |
new_vma->vm_start = addr; new_vma->vm_end = addr + len; new_vma->vm_pgoff = pgoff; |
846a16bf0 mempolicy: rename... |
2761 |
pol = mpol_dup(vma_policy(vma)); |
5beb49305 mm: change anon_v... |
2762 2763 |
if (IS_ERR(pol)) goto out_free_vma; |
523d4e200 mm anon rmap: in ... |
2764 |
vma_set_policy(new_vma, pol); |
5beb49305 mm: change anon_v... |
2765 2766 2767 |
INIT_LIST_HEAD(&new_vma->anon_vma_chain); if (anon_vma_clone(new_vma, vma)) goto out_free_mempol; |
e9714acf8 mm: kill vma flag... |
2768 |
if (new_vma->vm_file) |
1da177e4c Linux-2.6.12-rc2 |
2769 2770 2771 2772 |
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); |
38a76013a mm: avoid taking ... |
2773 |
*need_rmap_locks = false; |
1da177e4c Linux-2.6.12-rc2 |
2774 2775 2776 |
} } return new_vma; |
5beb49305 mm: change anon_v... |
2777 2778 2779 2780 2781 2782 |
out_free_mempol: mpol_put(pol); out_free_vma: kmem_cache_free(vm_area_cachep, new_vma); return NULL; |
1da177e4c Linux-2.6.12-rc2 |
2783 |
} |
119f657c7 [PATCH] RLIMIT_AS... |
2784 2785 2786 2787 2788 2789 2790 2791 2792 |
/* * Return true if the calling process may expand its vm space by the passed * number of pages */ int may_expand_vm(struct mm_struct *mm, unsigned long npages) { unsigned long cur = mm->total_vm; /* pages */ unsigned long lim; |
59e99e5b9 mm: use rlimit he... |
2793 |
lim = rlimit(RLIMIT_AS) >> PAGE_SHIFT; |
119f657c7 [PATCH] RLIMIT_AS... |
2794 2795 2796 2797 2798 |
if (cur + npages > lim) return 0; return 1; } |
fa5dc22f8 [PATCH] Add insta... |
2799 |
|
b1d0e4f53 mm: special mappi... |
2800 2801 |
static int special_mapping_fault(struct vm_area_struct *vma, struct vm_fault *vmf) |
fa5dc22f8 [PATCH] Add insta... |
2802 |
{ |
b1d0e4f53 mm: special mappi... |
2803 |
pgoff_t pgoff; |
fa5dc22f8 [PATCH] Add insta... |
2804 |
struct page **pages; |
b1d0e4f53 mm: special mappi... |
2805 2806 2807 2808 2809 2810 2811 |
/* * special mappings have no vm_file, and in that case, the mm * uses vm_pgoff internally. So we have to subtract it from here. * We are allowed to do this because we are the mm; do not copy * this code into drivers! */ pgoff = vmf->pgoff - vma->vm_pgoff; |
fa5dc22f8 [PATCH] Add insta... |
2812 |
|
b1d0e4f53 mm: special mappi... |
2813 2814 |
for (pages = vma->vm_private_data; pgoff && *pages; ++pages) pgoff--; |
fa5dc22f8 [PATCH] Add insta... |
2815 2816 2817 2818 |
if (*pages) { struct page *page = *pages; get_page(page); |
b1d0e4f53 mm: special mappi... |
2819 2820 |
vmf->page = page; return 0; |
fa5dc22f8 [PATCH] Add insta... |
2821 |
} |
b1d0e4f53 mm: special mappi... |
2822 |
return VM_FAULT_SIGBUS; |
fa5dc22f8 [PATCH] Add insta... |
2823 2824 2825 2826 2827 2828 2829 2830 |
} /* * Having a close hook prevents vma merging regardless of flags. */ static void special_mapping_close(struct vm_area_struct *vma) { } |
f0f37e2f7 const: mark struc... |
2831 |
static const struct vm_operations_struct special_mapping_vmops = { |
fa5dc22f8 [PATCH] Add insta... |
2832 |
.close = special_mapping_close, |
b1d0e4f53 mm: special mappi... |
2833 |
.fault = special_mapping_fault, |
fa5dc22f8 [PATCH] Add insta... |
2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 |
}; /* * Called with mm->mmap_sem held for writing. * Insert a new vma covering the given region, with the given flags. * Its pages are supplied by the given array of struct page *. * The array can be shorter than len >> PAGE_SHIFT if it's null-terminated. * The region past the last page supplied will always produce SIGBUS. * The array pointer and the pages it points to are assumed to stay alive * for as long as this mapping might exist. */ int install_special_mapping(struct mm_struct *mm, unsigned long addr, unsigned long len, unsigned long vm_flags, struct page **pages) { |
462e635e5 install_special_m... |
2849 |
int ret; |
fa5dc22f8 [PATCH] Add insta... |
2850 2851 2852 2853 2854 |
struct vm_area_struct *vma; vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); if (unlikely(vma == NULL)) return -ENOMEM; |
5beb49305 mm: change anon_v... |
2855 |
INIT_LIST_HEAD(&vma->anon_vma_chain); |
fa5dc22f8 [PATCH] Add insta... |
2856 2857 2858 |
vma->vm_mm = mm; vma->vm_start = addr; vma->vm_end = addr + len; |
2f98735c9 vm audit: add VM_... |
2859 |
vma->vm_flags = vm_flags | mm->def_flags | VM_DONTEXPAND; |
3ed75eb8f setup vma->vm_pag... |
2860 |
vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); |
fa5dc22f8 [PATCH] Add insta... |
2861 2862 2863 |
vma->vm_ops = &special_mapping_vmops; vma->vm_private_data = pages; |
462e635e5 install_special_m... |
2864 2865 2866 |
ret = insert_vm_struct(mm, vma); if (ret) goto out; |
fa5dc22f8 [PATCH] Add insta... |
2867 2868 |
mm->total_vm += len >> PAGE_SHIFT; |
cdd6c482c perf: Do the big ... |
2869 |
perf_event_mmap(vma); |
089dd79db perf_counter: Gen... |
2870 |
|
fa5dc22f8 [PATCH] Add insta... |
2871 |
return 0; |
462e635e5 install_special_m... |
2872 2873 2874 2875 |
out: kmem_cache_free(vm_area_cachep, vma); return ret; |
fa5dc22f8 [PATCH] Add insta... |
2876 |
} |
7906d00cd mmu-notifiers: ad... |
2877 2878 |
static DEFINE_MUTEX(mm_all_locks_mutex); |
454ed842d lockdep: annotate... |
2879 |
static void vm_lock_anon_vma(struct mm_struct *mm, struct anon_vma *anon_vma) |
7906d00cd mmu-notifiers: ad... |
2880 |
{ |
bf181b9f9 mm anon rmap: rep... |
2881 |
if (!test_bit(0, (unsigned long *) &anon_vma->root->rb_root.rb_node)) { |
7906d00cd mmu-notifiers: ad... |
2882 2883 2884 2885 |
/* * The LSB of head.next can't change from under us * because we hold the mm_all_locks_mutex. */ |
572043c90 mm: mmap: annotat... |
2886 |
down_write_nest_lock(&anon_vma->root->rwsem, &mm->mmap_sem); |
7906d00cd mmu-notifiers: ad... |
2887 2888 |
/* * We can safely modify head.next after taking the |
5a505085f mm/rmap: Convert ... |
2889 |
* anon_vma->root->rwsem. If some other vma in this mm shares |
7906d00cd mmu-notifiers: ad... |
2890 2891 2892 2893 |
* 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 mm/rmap: Convert ... |
2894 |
* anon_vma->root->rwsem. |
7906d00cd mmu-notifiers: ad... |
2895 2896 |
*/ if (__test_and_set_bit(0, (unsigned long *) |
bf181b9f9 mm anon rmap: rep... |
2897 |
&anon_vma->root->rb_root.rb_node)) |
7906d00cd mmu-notifiers: ad... |
2898 2899 2900 |
BUG(); } } |
454ed842d lockdep: annotate... |
2901 |
static void vm_lock_mapping(struct mm_struct *mm, struct address_space *mapping) |
7906d00cd mmu-notifiers: ad... |
2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 |
{ 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(); |
3d48ae45e mm: Convert i_mma... |
2915 |
mutex_lock_nest_lock(&mapping->i_mmap_mutex, &mm->mmap_sem); |
7906d00cd mmu-notifiers: ad... |
2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 |
} } /* * This operation locks against the VM for all pte/vma/mm related * operations that could ever happen on a certain mm. This includes * vmtruncate, try_to_unmap, and all page faults. * * The caller must take the mmap_sem in write mode before calling * mm_take_all_locks(). The caller isn't allowed to release the * mmap_sem until mm_drop_all_locks() returns. * * mmap_sem in write mode is required in order to block all operations * that could modify pagetables and free pages without need of * altering the vma layout (for example populate_range() with * nonlinear vmas). It's also needed in write mode to avoid new * anon_vmas to be associated with existing vmas. * * A single task can't take more than one mm_take_all_locks() in a row * or it would deadlock. * |
bf181b9f9 mm anon rmap: rep... |
2937 |
* The LSB in anon_vma->rb_root.rb_node and the AS_MM_ALL_LOCKS bitflag in |
7906d00cd mmu-notifiers: ad... |
2938 2939 2940 2941 |
* mapping->flags avoid to take the same lock twice, if more than one * vma in this mm is backed by the same anon_vma or address_space. * * We can take all the locks in random order because the VM code |
631b0cfdb mm: fix wrong com... |
2942 |
* taking i_mmap_mutex or anon_vma->rwsem outside the mmap_sem never |
7906d00cd mmu-notifiers: ad... |
2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 |
* takes more than one of them in a row. Secondly we're protected * against a concurrent mm_take_all_locks() by the mm_all_locks_mutex. * * mm_take_all_locks() and mm_drop_all_locks are expensive operations * that may have to take thousand of locks. * * mm_take_all_locks() can fail if it's interrupted by signals. */ int mm_take_all_locks(struct mm_struct *mm) { struct vm_area_struct *vma; |
5beb49305 mm: change anon_v... |
2954 |
struct anon_vma_chain *avc; |
7906d00cd mmu-notifiers: ad... |
2955 2956 2957 2958 2959 2960 2961 2962 |
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; |
7906d00cd mmu-notifiers: ad... |
2963 |
if (vma->vm_file && vma->vm_file->f_mapping) |
454ed842d lockdep: annotate... |
2964 |
vm_lock_mapping(mm, vma->vm_file->f_mapping); |
7906d00cd mmu-notifiers: ad... |
2965 |
} |
7cd5a02f5 mm: fix mm_take_a... |
2966 2967 2968 2969 2970 |
for (vma = mm->mmap; vma; vma = vma->vm_next) { if (signal_pending(current)) goto out_unlock; if (vma->anon_vma) |
5beb49305 mm: change anon_v... |
2971 2972 |
list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) vm_lock_anon_vma(mm, avc->anon_vma); |
7906d00cd mmu-notifiers: ad... |
2973 |
} |
7cd5a02f5 mm: fix mm_take_a... |
2974 |
|
584cff54e mm/mmap.c: elimin... |
2975 |
return 0; |
7906d00cd mmu-notifiers: ad... |
2976 2977 |
out_unlock: |
584cff54e mm/mmap.c: elimin... |
2978 2979 |
mm_drop_all_locks(mm); return -EINTR; |
7906d00cd mmu-notifiers: ad... |
2980 2981 2982 2983 |
} static void vm_unlock_anon_vma(struct anon_vma *anon_vma) { |
bf181b9f9 mm anon rmap: rep... |
2984 |
if (test_bit(0, (unsigned long *) &anon_vma->root->rb_root.rb_node)) { |
7906d00cd mmu-notifiers: ad... |
2985 2986 2987 2988 2989 |
/* * 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 mm anon rmap: rep... |
2990 |
* the vma so the users using the anon_vma->rb_root will |
7906d00cd mmu-notifiers: ad... |
2991 2992 2993 2994 |
* never see our bitflag. * * No need of atomic instructions here, head.next * can't change from under us until we release the |
5a505085f mm/rmap: Convert ... |
2995 |
* anon_vma->root->rwsem. |
7906d00cd mmu-notifiers: ad... |
2996 2997 |
*/ if (!__test_and_clear_bit(0, (unsigned long *) |
bf181b9f9 mm anon rmap: rep... |
2998 |
&anon_vma->root->rb_root.rb_node)) |
7906d00cd mmu-notifiers: ad... |
2999 |
BUG(); |
08b52706d mm/rmap: rename a... |
3000 |
anon_vma_unlock_write(anon_vma); |
7906d00cd mmu-notifiers: ad... |
3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 |
} } 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. */ |
3d48ae45e mm: Convert i_mma... |
3011 |
mutex_unlock(&mapping->i_mmap_mutex); |
7906d00cd mmu-notifiers: ad... |
3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 |
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 mm: change anon_v... |
3025 |
struct anon_vma_chain *avc; |
7906d00cd mmu-notifiers: ad... |
3026 3027 3028 3029 3030 3031 |
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 mm: change anon_v... |
3032 3033 |
list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) vm_unlock_anon_vma(avc->anon_vma); |
7906d00cd mmu-notifiers: ad... |
3034 3035 3036 3037 3038 3039 |
if (vma->vm_file && vma->vm_file->f_mapping) vm_unlock_mapping(vma->vm_file->f_mapping); } mutex_unlock(&mm_all_locks_mutex); } |
8feae1311 NOMMU: Make VMAs ... |
3040 3041 3042 3043 3044 3045 |
/* * initialise the VMA slab */ void __init mmap_init(void) { |
00a62ce91 mm: fix Committed... |
3046 3047 3048 3049 |
int ret; ret = percpu_counter_init(&vm_committed_as, 0); VM_BUG_ON(ret); |
8feae1311 NOMMU: Make VMAs ... |
3050 |
} |
c9b1d0981 mm: limit growth ... |
3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 |
/* * 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 mm: reinititalise... |
3062 |
static int init_user_reserve(void) |
c9b1d0981 mm: limit growth ... |
3063 3064 3065 3066 3067 3068 3069 3070 3071 |
{ 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; } module_init(init_user_reserve) |
4eeab4f55 mm: replace hardc... |
3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 |
/* * 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 mm: reinititalise... |
3083 |
static int init_admin_reserve(void) |
4eeab4f55 mm: replace hardc... |
3084 3085 3086 3087 3088 3089 3090 3091 3092 |
{ 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; } module_init(init_admin_reserve) |
1640879af mm: reinititalise... |
3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 |
/* * 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)) printk("Failed registering memory add/remove notifier for admin reserve"); return 0; } module_init(init_reserve_notifier) |