Blame view
mm/mmap.c
94.7 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 |
*/ |
b1de0d139 mm: convert some ... |
8 |
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt |
e8420a8ec mm/mmap: check fo... |
9 |
#include <linux/kernel.h> |
1da177e4c Linux-2.6.12-rc2 |
10 |
#include <linux/slab.h> |
4af3c9cc4 Drop some headers... |
11 |
#include <linux/backing-dev.h> |
1da177e4c Linux-2.6.12-rc2 |
12 |
#include <linux/mm.h> |
615d6e875 mm: per-thread vm... |
13 |
#include <linux/vmacache.h> |
1da177e4c Linux-2.6.12-rc2 |
14 15 16 17 18 |
#include <linux/shm.h> #include <linux/mman.h> #include <linux/pagemap.h> #include <linux/swap.h> #include <linux/syscalls.h> |
c59ede7b7 [PATCH] move capa... |
19 |
#include <linux/capability.h> |
1da177e4c Linux-2.6.12-rc2 |
20 21 22 23 24 25 |
#include <linux/init.h> #include <linux/file.h> #include <linux/fs.h> #include <linux/personality.h> #include <linux/security.h> #include <linux/hugetlb.h> |
c01d5b300 shmem: get_unmapp... |
26 |
#include <linux/shmem_fs.h> |
1da177e4c Linux-2.6.12-rc2 |
27 |
#include <linux/profile.h> |
b95f1b31b mm: Map most file... |
28 |
#include <linux/export.h> |
1da177e4c Linux-2.6.12-rc2 |
29 30 31 |
#include <linux/mount.h> #include <linux/mempolicy.h> #include <linux/rmap.h> |
cddb8a5c1 mmu-notifiers: core |
32 |
#include <linux/mmu_notifier.h> |
82f71ae4a mm: catch memory ... |
33 |
#include <linux/mmdebug.h> |
cdd6c482c perf: Do the big ... |
34 |
#include <linux/perf_event.h> |
120a795da audit mmap |
35 |
#include <linux/audit.h> |
b15d00b6a thp: khugepaged v... |
36 |
#include <linux/khugepaged.h> |
2b1444983 uprobes, mm, x86:... |
37 |
#include <linux/uprobes.h> |
d37371870 mm: augment vma r... |
38 |
#include <linux/rbtree_augmented.h> |
1640879af mm: reinititalise... |
39 40 |
#include <linux/notifier.h> #include <linux/memory.h> |
b1de0d139 mm: convert some ... |
41 |
#include <linux/printk.h> |
19a809afe userfaultfd: teac... |
42 |
#include <linux/userfaultfd_k.h> |
d977d56ce mm: warn about Vm... |
43 |
#include <linux/moduleparam.h> |
62b5f7d01 mm/core, x86/mm/p... |
44 |
#include <linux/pkeys.h> |
1da177e4c Linux-2.6.12-rc2 |
45 46 47 48 |
#include <asm/uaccess.h> #include <asm/cacheflush.h> #include <asm/tlb.h> |
d6dd61c83 [PATCH] x86: PARA... |
49 |
#include <asm/mmu_context.h> |
1da177e4c Linux-2.6.12-rc2 |
50 |
|
42b777281 mm: remove double... |
51 |
#include "internal.h" |
3a4597568 [PATCH] IA64,spar... |
52 53 54 |
#ifndef arch_mmap_check #define arch_mmap_check(addr, len, flags) (0) #endif |
d07e22597 mm: mmap: add new... |
55 56 57 58 59 60 61 62 63 64 |
#ifdef CONFIG_HAVE_ARCH_MMAP_RND_BITS const int mmap_rnd_bits_min = CONFIG_ARCH_MMAP_RND_BITS_MIN; const int mmap_rnd_bits_max = CONFIG_ARCH_MMAP_RND_BITS_MAX; int mmap_rnd_bits __read_mostly = CONFIG_ARCH_MMAP_RND_BITS; #endif #ifdef CONFIG_HAVE_ARCH_MMAP_RND_COMPAT_BITS const int mmap_rnd_compat_bits_min = CONFIG_ARCH_MMAP_RND_COMPAT_BITS_MIN; const int mmap_rnd_compat_bits_max = CONFIG_ARCH_MMAP_RND_COMPAT_BITS_MAX; int mmap_rnd_compat_bits __read_mostly = CONFIG_ARCH_MMAP_RND_COMPAT_BITS; #endif |
f4fcd5584 mm: enable RLIMIT... |
65 |
static bool ignore_rlimit_data; |
d977d56ce mm: warn about Vm... |
66 |
core_param(ignore_rlimit_data, ignore_rlimit_data, bool, 0644); |
d07e22597 mm: mmap: add new... |
67 |
|
e0da382c9 [PATCH] freepgt: ... |
68 69 70 |
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 |
71 72 73 74 75 76 77 78 79 |
/* description of effects of mapping type and prot in current implementation. * this is due to the limited x86 page protection hardware. The expected * behavior is in parens: * * map_type prot * PROT_NONE PROT_READ PROT_WRITE PROT_EXEC * MAP_SHARED r: (no) no r: (yes) yes r: (no) yes r: (no) yes * w: (no) no w: (no) no w: (yes) yes w: (no) no * x: (no) no x: (no) yes x: (no) yes x: (yes) yes |
cc71aba34 mm/mmap.c: whites... |
80 |
* |
1da177e4c Linux-2.6.12-rc2 |
81 82 83 84 |
* 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 * |
cab15ce60 arm64: Introduce ... |
85 86 87 88 89 |
* On arm64, PROT_EXEC has the following behaviour for both MAP_SHARED and * MAP_PRIVATE: * r: (no) no * w: (no) no * x: (yes) yes |
1da177e4c Linux-2.6.12-rc2 |
90 91 92 93 94 |
*/ pgprot_t protection_map[16] = { __P000, __P001, __P010, __P011, __P100, __P101, __P110, __P111, __S000, __S001, __S010, __S011, __S100, __S101, __S110, __S111 }; |
804af2cf6 [AGPGART] remove ... |
95 96 |
pgprot_t vm_get_page_prot(unsigned long vm_flags) { |
b845f313d mm: Allow archite... |
97 98 99 |
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 ... |
100 101 |
} EXPORT_SYMBOL(vm_get_page_prot); |
64e455079 mm: softdirty: en... |
102 103 104 105 106 107 108 109 110 |
static pgprot_t vm_pgprot_modify(pgprot_t oldprot, unsigned long vm_flags) { return pgprot_modify(oldprot, vm_get_page_prot(vm_flags)); } /* Update vma->vm_page_prot to reflect vma->vm_flags. */ void vma_set_page_prot(struct vm_area_struct *vma) { unsigned long vm_flags = vma->vm_flags; |
6d2329f88 mm: vm_page_prot:... |
111 |
pgprot_t vm_page_prot; |
64e455079 mm: softdirty: en... |
112 |
|
6d2329f88 mm: vm_page_prot:... |
113 114 |
vm_page_prot = vm_pgprot_modify(vma->vm_page_prot, vm_flags); if (vma_wants_writenotify(vma, vm_page_prot)) { |
64e455079 mm: softdirty: en... |
115 |
vm_flags &= ~VM_SHARED; |
6d2329f88 mm: vm_page_prot:... |
116 |
vm_page_prot = vm_pgprot_modify(vm_page_prot, vm_flags); |
64e455079 mm: softdirty: en... |
117 |
} |
6d2329f88 mm: vm_page_prot:... |
118 119 |
/* remove_protection_ptes reads vma->vm_page_prot without mmap_sem */ WRITE_ONCE(vma->vm_page_prot, vm_page_prot); |
64e455079 mm: softdirty: en... |
120 |
} |
1da177e4c Linux-2.6.12-rc2 |
121 |
/* |
c8c06efa8 mm: convert i_mma... |
122 |
* Requires inode->i_mapping->i_mmap_rwsem |
1da177e4c Linux-2.6.12-rc2 |
123 124 125 126 127 |
*/ 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_... |
128 |
atomic_inc(&file_inode(file)->i_writecount); |
1da177e4c Linux-2.6.12-rc2 |
129 |
if (vma->vm_flags & VM_SHARED) |
4bb5f5d93 mm: allow drivers... |
130 |
mapping_unmap_writable(mapping); |
1da177e4c Linux-2.6.12-rc2 |
131 132 |
flush_dcache_mmap_lock(mapping); |
27ba0644e rmap: drop suppor... |
133 |
vma_interval_tree_remove(vma, &mapping->i_mmap); |
1da177e4c Linux-2.6.12-rc2 |
134 135 136 137 |
flush_dcache_mmap_unlock(mapping); } /* |
6b2dbba8b mm: replace vma p... |
138 |
* Unlink a file-based vm structure from its interval tree, to hide |
a8fb5618d [PATCH] mm: unlin... |
139 |
* vma from rmap and vmtruncate before freeing its page tables. |
1da177e4c Linux-2.6.12-rc2 |
140 |
*/ |
a8fb5618d [PATCH] mm: unlin... |
141 |
void unlink_file_vma(struct vm_area_struct *vma) |
1da177e4c Linux-2.6.12-rc2 |
142 143 |
{ struct file *file = vma->vm_file; |
1da177e4c Linux-2.6.12-rc2 |
144 145 |
if (file) { struct address_space *mapping = file->f_mapping; |
83cde9e8b mm: use new helpe... |
146 |
i_mmap_lock_write(mapping); |
1da177e4c Linux-2.6.12-rc2 |
147 |
__remove_shared_vm_struct(vma, file, mapping); |
83cde9e8b mm: use new helpe... |
148 |
i_mmap_unlock_write(mapping); |
1da177e4c Linux-2.6.12-rc2 |
149 |
} |
a8fb5618d [PATCH] mm: unlin... |
150 151 152 153 154 155 156 157 |
} /* * 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... |
158 |
might_sleep(); |
1da177e4c Linux-2.6.12-rc2 |
159 160 |
if (vma->vm_ops && vma->vm_ops->close) vma->vm_ops->close(vma); |
e9714acf8 mm: kill vma flag... |
161 |
if (vma->vm_file) |
a8fb5618d [PATCH] mm: unlin... |
162 |
fput(vma->vm_file); |
f0be3d32b mempolicy: rename... |
163 |
mpol_put(vma_policy(vma)); |
1da177e4c Linux-2.6.12-rc2 |
164 |
kmem_cache_free(vm_area_cachep, vma); |
a8fb5618d [PATCH] mm: unlin... |
165 |
return next; |
1da177e4c Linux-2.6.12-rc2 |
166 |
} |
5d22fc25d mm: remove more I... |
167 |
static int do_brk(unsigned long addr, unsigned long len); |
e4eb1ff61 VM: add "vm_brk()... |
168 |
|
6a6160a7b [CVE-2009-0029] S... |
169 |
SYSCALL_DEFINE1(brk, unsigned long, brk) |
1da177e4c Linux-2.6.12-rc2 |
170 |
{ |
8764b338b mm: use may_adjus... |
171 |
unsigned long retval; |
1da177e4c Linux-2.6.12-rc2 |
172 173 |
unsigned long newbrk, oldbrk; struct mm_struct *mm = current->mm; |
a5b4592cf brk: make sys_brk... |
174 |
unsigned long min_brk; |
128557ffe mm: use mm_popula... |
175 |
bool populate; |
1da177e4c Linux-2.6.12-rc2 |
176 |
|
dc0ef0df7 mm: make mmap_sem... |
177 178 |
if (down_write_killable(&mm->mmap_sem)) return -EINTR; |
1da177e4c Linux-2.6.12-rc2 |
179 |
|
a5b4592cf brk: make sys_brk... |
180 |
#ifdef CONFIG_COMPAT_BRK |
5520e8948 brk: fix min_brk ... |
181 182 183 184 185 |
/* * 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: ... |
186 |
if (current->brk_randomized) |
5520e8948 brk: fix min_brk ... |
187 188 189 |
min_brk = mm->start_brk; else min_brk = mm->end_data; |
a5b4592cf brk: make sys_brk... |
190 191 192 193 |
#else min_brk = mm->start_brk; #endif if (brk < min_brk) |
1da177e4c Linux-2.6.12-rc2 |
194 |
goto out; |
1e624196f [PATCH] mm: fix b... |
195 196 197 198 199 200 201 |
/* * Check against rlimit here. If this check is done later after the test * of oldbrk with newbrk then it can escape the test and let the data * segment grow beyond its set limit the in case where the limit is * not page aligned -Ram Gupta */ |
8764b338b mm: use may_adjus... |
202 203 |
if (check_data_rlimit(rlimit(RLIMIT_DATA), brk, mm->start_brk, mm->end_data, mm->start_data)) |
1e624196f [PATCH] mm: fix b... |
204 |
goto out; |
1da177e4c Linux-2.6.12-rc2 |
205 206 207 208 209 210 211 212 213 214 215 |
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 |
216 217 218 219 220 |
/* Check against existing mmap mappings. */ if (find_vma_intersection(mm, oldbrk, newbrk+PAGE_SIZE)) goto out; /* Ok, looks good - let it rip. */ |
5d22fc25d mm: remove more I... |
221 |
if (do_brk(oldbrk, newbrk-oldbrk) < 0) |
1da177e4c Linux-2.6.12-rc2 |
222 |
goto out; |
128557ffe mm: use mm_popula... |
223 |
|
1da177e4c Linux-2.6.12-rc2 |
224 225 |
set_brk: mm->brk = brk; |
128557ffe mm: use mm_popula... |
226 227 228 229 230 |
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 |
231 232 233 234 235 |
out: retval = mm->brk; up_write(&mm->mmap_sem); return retval; } |
d37371870 mm: augment vma r... |
236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 |
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... |
256 |
#ifdef CONFIG_DEBUG_VM_RB |
acf128d04 mm: validate_mm b... |
257 |
static int browse_rb(struct mm_struct *mm) |
1da177e4c Linux-2.6.12-rc2 |
258 |
{ |
acf128d04 mm: validate_mm b... |
259 |
struct rb_root *root = &mm->mm_rb; |
5a0768f64 mm: check rb_subt... |
260 |
int i = 0, j, bug = 0; |
1da177e4c Linux-2.6.12-rc2 |
261 262 263 264 265 266 |
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... |
267 |
if (vma->vm_start < prev) { |
ff26f70f4 mm/mmap.c: clean ... |
268 269 270 |
pr_emerg("vm_start %lx < prev %lx ", vma->vm_start, prev); |
5a0768f64 mm: check rb_subt... |
271 272 273 |
bug = 1; } if (vma->vm_start < pend) { |
ff26f70f4 mm/mmap.c: clean ... |
274 275 276 |
pr_emerg("vm_start %lx < pend %lx ", vma->vm_start, pend); |
5a0768f64 mm: check rb_subt... |
277 278 279 |
bug = 1; } if (vma->vm_start > vma->vm_end) { |
ff26f70f4 mm/mmap.c: clean ... |
280 281 282 |
pr_emerg("vm_start %lx > vm_end %lx ", vma->vm_start, vma->vm_end); |
5a0768f64 mm: check rb_subt... |
283 284 |
bug = 1; } |
acf128d04 mm: validate_mm b... |
285 |
spin_lock(&mm->page_table_lock); |
5a0768f64 mm: check rb_subt... |
286 |
if (vma->rb_subtree_gap != vma_compute_subtree_gap(vma)) { |
8542bdfc6 mm/mmap.c: use pr... |
287 288 |
pr_emerg("free gap %lx, correct %lx ", |
5a0768f64 mm: check rb_subt... |
289 290 291 292 |
vma->rb_subtree_gap, vma_compute_subtree_gap(vma)); bug = 1; } |
acf128d04 mm: validate_mm b... |
293 |
spin_unlock(&mm->page_table_lock); |
1da177e4c Linux-2.6.12-rc2 |
294 295 |
i++; pn = nd; |
d1af65d13 [PATCH] Bug in MM... |
296 297 |
prev = vma->vm_start; pend = vma->vm_end; |
1da177e4c Linux-2.6.12-rc2 |
298 299 |
} j = 0; |
5a0768f64 mm: check rb_subt... |
300 |
for (nd = pn; nd; nd = rb_prev(nd)) |
1da177e4c Linux-2.6.12-rc2 |
301 |
j++; |
5a0768f64 mm: check rb_subt... |
302 |
if (i != j) { |
8542bdfc6 mm/mmap.c: use pr... |
303 304 |
pr_emerg("backwards %d, forwards %d ", j, i); |
5a0768f64 mm: check rb_subt... |
305 |
bug = 1; |
1da177e4c Linux-2.6.12-rc2 |
306 |
} |
5a0768f64 mm: check rb_subt... |
307 |
return bug ? -1 : i; |
1da177e4c Linux-2.6.12-rc2 |
308 |
} |
d37371870 mm: augment vma r... |
309 310 311 312 313 314 315 |
static void validate_mm_rb(struct rb_root *root, struct vm_area_struct *ignore) { struct rb_node *nd; for (nd = rb_first(root); nd; nd = rb_next(nd)) { struct vm_area_struct *vma; vma = rb_entry(nd, struct vm_area_struct, vm_rb); |
96dad67ff mm: use VM_BUG_ON... |
316 317 318 |
VM_BUG_ON_VMA(vma != ignore && vma->rb_subtree_gap != vma_compute_subtree_gap(vma), vma); |
1da177e4c Linux-2.6.12-rc2 |
319 |
} |
1da177e4c Linux-2.6.12-rc2 |
320 |
} |
eafd4dc4d mm/mmap.c: mark f... |
321 |
static void validate_mm(struct mm_struct *mm) |
1da177e4c Linux-2.6.12-rc2 |
322 323 324 |
{ int bug = 0; int i = 0; |
5a0768f64 mm: check rb_subt... |
325 |
unsigned long highest_address = 0; |
ed8ea8150 mm: add CONFIG_DE... |
326 |
struct vm_area_struct *vma = mm->mmap; |
ff26f70f4 mm/mmap.c: clean ... |
327 |
|
ed8ea8150 mm: add CONFIG_DE... |
328 |
while (vma) { |
12352d3ca mm: replace vma_l... |
329 |
struct anon_vma *anon_vma = vma->anon_vma; |
ed8ea8150 mm: add CONFIG_DE... |
330 |
struct anon_vma_chain *avc; |
ff26f70f4 mm/mmap.c: clean ... |
331 |
|
12352d3ca mm: replace vma_l... |
332 333 334 335 336 337 |
if (anon_vma) { anon_vma_lock_read(anon_vma); list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) anon_vma_interval_tree_verify(avc); anon_vma_unlock_read(anon_vma); } |
5a0768f64 mm: check rb_subt... |
338 |
highest_address = vma->vm_end; |
ed8ea8150 mm: add CONFIG_DE... |
339 |
vma = vma->vm_next; |
1da177e4c Linux-2.6.12-rc2 |
340 341 |
i++; } |
5a0768f64 mm: check rb_subt... |
342 |
if (i != mm->map_count) { |
8542bdfc6 mm/mmap.c: use pr... |
343 344 |
pr_emerg("map_count %d vm_next %d ", mm->map_count, i); |
5a0768f64 mm: check rb_subt... |
345 346 347 |
bug = 1; } if (highest_address != mm->highest_vm_end) { |
8542bdfc6 mm/mmap.c: use pr... |
348 349 |
pr_emerg("mm->highest_vm_end %lx, found %lx ", |
ff26f70f4 mm/mmap.c: clean ... |
350 |
mm->highest_vm_end, highest_address); |
5a0768f64 mm: check rb_subt... |
351 352 |
bug = 1; } |
acf128d04 mm: validate_mm b... |
353 |
i = browse_rb(mm); |
5a0768f64 mm: check rb_subt... |
354 |
if (i != mm->map_count) { |
ff26f70f4 mm/mmap.c: clean ... |
355 356 357 |
if (i != -1) pr_emerg("map_count %d rb %d ", mm->map_count, i); |
5a0768f64 mm: check rb_subt... |
358 359 |
bug = 1; } |
96dad67ff mm: use VM_BUG_ON... |
360 |
VM_BUG_ON_MM(bug, mm); |
1da177e4c Linux-2.6.12-rc2 |
361 362 |
} #else |
d37371870 mm: augment vma r... |
363 |
#define validate_mm_rb(root, ignore) do { } while (0) |
1da177e4c Linux-2.6.12-rc2 |
364 365 |
#define validate_mm(mm) do { } while (0) #endif |
d37371870 mm: augment vma r... |
366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 |
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); } |
8f26e0b17 mm: vma_merge: co... |
391 |
static void __vma_rb_erase(struct vm_area_struct *vma, struct rb_root *root) |
d37371870 mm: augment vma r... |
392 393 |
{ /* |
d37371870 mm: augment vma r... |
394 395 396 397 398 399 |
* 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); } |
8f26e0b17 mm: vma_merge: co... |
400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 |
static __always_inline void vma_rb_erase_ignore(struct vm_area_struct *vma, struct rb_root *root, struct vm_area_struct *ignore) { /* * All rb_subtree_gap values must be consistent prior to erase, * with the possible exception of the "next" vma being erased if * next->vm_start was reduced. */ validate_mm_rb(root, ignore); __vma_rb_erase(vma, root); } static __always_inline 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); __vma_rb_erase(vma, root); } |
bf181b9f9 mm anon rmap: rep... |
425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 |
/* * 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... |
456 457 458 |
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 |
459 |
{ |
6597d7833 mm/mmap.c: replac... |
460 |
struct rb_node **__rb_link, *__rb_parent, *rb_prev; |
1da177e4c Linux-2.6.12-rc2 |
461 462 463 |
__rb_link = &mm->mm_rb.rb_node; rb_prev = __rb_parent = NULL; |
1da177e4c Linux-2.6.12-rc2 |
464 465 466 467 468 469 470 471 |
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... |
472 473 474 |
/* Fail if an existing vma overlaps the area */ if (vma_tmp->vm_start < end) return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
475 476 477 478 479 480 481 482 483 484 485 486 |
__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... |
487 |
return 0; |
1da177e4c Linux-2.6.12-rc2 |
488 |
} |
e8420a8ec mm/mmap: check fo... |
489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 |
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 |
516 517 518 |
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... |
519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 |
/* 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 |
534 |
rb_link_node(&vma->vm_rb, rb_parent, rb_link); |
d37371870 mm: augment vma r... |
535 536 537 |
vma->rb_subtree_gap = 0; vma_gap_update(vma); vma_rb_insert(vma, &mm->mm_rb); |
1da177e4c Linux-2.6.12-rc2 |
538 |
} |
cb8f488c3 mmap.c: deinline ... |
539 |
static void __vma_link_file(struct vm_area_struct *vma) |
1da177e4c Linux-2.6.12-rc2 |
540 |
{ |
48aae4255 mm/mmap.c: fix co... |
541 |
struct file *file; |
1da177e4c Linux-2.6.12-rc2 |
542 543 544 545 546 547 |
file = vma->vm_file; if (file) { struct address_space *mapping = file->f_mapping; if (vma->vm_flags & VM_DENYWRITE) |
496ad9aa8 new helper: file_... |
548 |
atomic_dec(&file_inode(file)->i_writecount); |
1da177e4c Linux-2.6.12-rc2 |
549 |
if (vma->vm_flags & VM_SHARED) |
4bb5f5d93 mm: allow drivers... |
550 |
atomic_inc(&mapping->i_mmap_writable); |
1da177e4c Linux-2.6.12-rc2 |
551 552 |
flush_dcache_mmap_lock(mapping); |
27ba0644e rmap: drop suppor... |
553 |
vma_interval_tree_insert(vma, &mapping->i_mmap); |
1da177e4c Linux-2.6.12-rc2 |
554 555 556 557 558 559 560 561 562 563 564 |
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 |
565 566 567 568 569 570 571 |
} static void vma_link(struct mm_struct *mm, struct vm_area_struct *vma, struct vm_area_struct *prev, struct rb_node **rb_link, struct rb_node *rb_parent) { struct address_space *mapping = NULL; |
64ac4940d mm/mmap.c: remove... |
572 |
if (vma->vm_file) { |
1da177e4c Linux-2.6.12-rc2 |
573 |
mapping = vma->vm_file->f_mapping; |
83cde9e8b mm: use new helpe... |
574 |
i_mmap_lock_write(mapping); |
64ac4940d mm/mmap.c: remove... |
575 |
} |
1da177e4c Linux-2.6.12-rc2 |
576 577 578 |
__vma_link(mm, vma, prev, rb_link, rb_parent); __vma_link_file(vma); |
1da177e4c Linux-2.6.12-rc2 |
579 |
if (mapping) |
83cde9e8b mm: use new helpe... |
580 |
i_mmap_unlock_write(mapping); |
1da177e4c Linux-2.6.12-rc2 |
581 582 583 584 585 586 |
mm->map_count++; validate_mm(mm); } /* |
88f6b4c32 mmap.c: fix comme... |
587 |
* Helper for vma_adjust() in the split_vma insert case: insert a vma into the |
6b2dbba8b mm: replace vma p... |
588 |
* mm's list and rbtree. It has already been inserted into the interval tree. |
1da177e4c Linux-2.6.12-rc2 |
589 |
*/ |
48aae4255 mm/mmap.c: fix co... |
590 |
static void __insert_vm_struct(struct mm_struct *mm, struct vm_area_struct *vma) |
1da177e4c Linux-2.6.12-rc2 |
591 |
{ |
6597d7833 mm/mmap.c: replac... |
592 |
struct vm_area_struct *prev; |
48aae4255 mm/mmap.c: fix co... |
593 |
struct rb_node **rb_link, *rb_parent; |
1da177e4c Linux-2.6.12-rc2 |
594 |
|
6597d7833 mm/mmap.c: replac... |
595 596 597 |
if (find_vma_links(mm, vma->vm_start, vma->vm_end, &prev, &rb_link, &rb_parent)) BUG(); |
1da177e4c Linux-2.6.12-rc2 |
598 599 600 |
__vma_link(mm, vma, prev, rb_link, rb_parent); mm->map_count++; } |
e86f15ee6 mm: vma_merge: fi... |
601 602 603 |
static __always_inline void __vma_unlink_common(struct mm_struct *mm, struct vm_area_struct *vma, struct vm_area_struct *prev, |
8f26e0b17 mm: vma_merge: co... |
604 605 |
bool has_prev, struct vm_area_struct *ignore) |
1da177e4c Linux-2.6.12-rc2 |
606 |
{ |
d37371870 mm: augment vma r... |
607 |
struct vm_area_struct *next; |
297c5eee3 mm: make the vma ... |
608 |
|
8f26e0b17 mm: vma_merge: co... |
609 |
vma_rb_erase_ignore(vma, &mm->mm_rb, ignore); |
e86f15ee6 mm: vma_merge: fi... |
610 611 612 613 614 615 616 617 618 619 |
next = vma->vm_next; if (has_prev) prev->vm_next = next; else { prev = vma->vm_prev; if (prev) prev->vm_next = next; else mm->mmap = next; } |
297c5eee3 mm: make the vma ... |
620 621 |
if (next) next->vm_prev = prev; |
615d6e875 mm: per-thread vm... |
622 623 624 |
/* Kill the cache */ vmacache_invalidate(mm); |
1da177e4c Linux-2.6.12-rc2 |
625 |
} |
e86f15ee6 mm: vma_merge: fi... |
626 627 628 629 |
static inline void __vma_unlink_prev(struct mm_struct *mm, struct vm_area_struct *vma, struct vm_area_struct *prev) { |
8f26e0b17 mm: vma_merge: co... |
630 |
__vma_unlink_common(mm, vma, prev, true, vma); |
e86f15ee6 mm: vma_merge: fi... |
631 |
} |
1da177e4c Linux-2.6.12-rc2 |
632 633 634 635 636 637 638 |
/* * 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. */ |
e86f15ee6 mm: vma_merge: fi... |
639 640 641 |
int __vma_adjust(struct vm_area_struct *vma, unsigned long start, unsigned long end, pgoff_t pgoff, struct vm_area_struct *insert, struct vm_area_struct *expand) |
1da177e4c Linux-2.6.12-rc2 |
642 643 |
{ struct mm_struct *mm = vma->vm_mm; |
e86f15ee6 mm: vma_merge: fi... |
644 |
struct vm_area_struct *next = vma->vm_next, *orig_vma = vma; |
1da177e4c Linux-2.6.12-rc2 |
645 |
struct address_space *mapping = NULL; |
6b2dbba8b mm: replace vma p... |
646 |
struct rb_root *root = NULL; |
012f18004 mm: always lock t... |
647 |
struct anon_vma *anon_vma = NULL; |
1da177e4c Linux-2.6.12-rc2 |
648 |
struct file *file = vma->vm_file; |
d37371870 mm: augment vma r... |
649 |
bool start_changed = false, end_changed = false; |
1da177e4c Linux-2.6.12-rc2 |
650 651 652 653 |
long adjust_next = 0; int remove_next = 0; if (next && !insert) { |
734537c9c mm: fix use-after... |
654 |
struct vm_area_struct *exporter = NULL, *importer = NULL; |
287d97ac0 vma_adjust: fix t... |
655 |
|
1da177e4c Linux-2.6.12-rc2 |
656 657 658 659 |
if (end >= next->vm_end) { /* * vma expands, overlapping all the next, and * perhaps the one after too (mprotect case 6). |
86d12e471 mm: vma_adjust: m... |
660 |
* The only other cases that gets here are |
e86f15ee6 mm: vma_merge: fi... |
661 |
* case 1, case 7 and case 8. |
1da177e4c Linux-2.6.12-rc2 |
662 |
*/ |
e86f15ee6 mm: vma_merge: fi... |
663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 |
if (next == expand) { /* * The only case where we don't expand "vma" * and we expand "next" instead is case 8. */ VM_WARN_ON(end != next->vm_end); /* * remove_next == 3 means we're * removing "vma" and that to do so we * swapped "vma" and "next". */ remove_next = 3; VM_WARN_ON(file != next->vm_file); swap(vma, next); } else { VM_WARN_ON(expand != vma); /* * case 1, 6, 7, remove_next == 2 is case 6, * remove_next == 1 is case 1 or 7. */ remove_next = 1 + (end > next->vm_end); VM_WARN_ON(remove_next == 2 && end != next->vm_next->vm_end); VM_WARN_ON(remove_next == 1 && end != next->vm_end); /* trim end to next, for case 6 first pass */ end = next->vm_end; } |
287d97ac0 vma_adjust: fix t... |
691 |
exporter = next; |
1da177e4c Linux-2.6.12-rc2 |
692 |
importer = vma; |
734537c9c mm: fix use-after... |
693 694 695 696 697 |
/* * If next doesn't have anon_vma, import from vma after * next, if the vma overlaps with it. */ |
97a42cd43 mm: vma_adjust: r... |
698 |
if (remove_next == 2 && !next->anon_vma) |
734537c9c mm: fix use-after... |
699 |
exporter = next->vm_next; |
1da177e4c Linux-2.6.12-rc2 |
700 701 702 703 704 705 |
} 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... |
706 |
exporter = next; |
1da177e4c Linux-2.6.12-rc2 |
707 |
importer = vma; |
e86f15ee6 mm: vma_merge: fi... |
708 |
VM_WARN_ON(expand != importer); |
1da177e4c Linux-2.6.12-rc2 |
709 710 711 712 713 714 |
} else if (end < vma->vm_end) { /* * vma shrinks, and !insert tells it's not * split_vma inserting another: so it must be * mprotect case 4 shifting the boundary down. */ |
cc71aba34 mm/mmap.c: whites... |
715 |
adjust_next = -((vma->vm_end - end) >> PAGE_SHIFT); |
287d97ac0 vma_adjust: fix t... |
716 |
exporter = vma; |
1da177e4c Linux-2.6.12-rc2 |
717 |
importer = next; |
e86f15ee6 mm: vma_merge: fi... |
718 |
VM_WARN_ON(expand != importer); |
1da177e4c Linux-2.6.12-rc2 |
719 |
} |
1da177e4c Linux-2.6.12-rc2 |
720 |
|
5beb49305 mm: change anon_v... |
721 722 723 724 725 |
/* * 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... |
726 |
if (exporter && exporter->anon_vma && !importer->anon_vma) { |
c4ea95d7c mm: fix anon_vma_... |
727 |
int error; |
b800c91a0 mm: fix corner ca... |
728 |
importer->anon_vma = exporter->anon_vma; |
c4ea95d7c mm: fix anon_vma_... |
729 |
error = anon_vma_clone(importer, exporter); |
3fe89b3e2 mm: fix anon_vma-... |
730 |
if (error) |
c4ea95d7c mm: fix anon_vma_... |
731 |
return error; |
5beb49305 mm: change anon_v... |
732 733 |
} } |
734537c9c mm: fix use-after... |
734 |
again: |
e86f15ee6 mm: vma_merge: fi... |
735 |
vma_adjust_trans_huge(orig_vma, start, end, adjust_next); |
37f9f5595 thp: run vma_adju... |
736 |
|
1da177e4c Linux-2.6.12-rc2 |
737 738 |
if (file) { mapping = file->f_mapping; |
27ba0644e rmap: drop suppor... |
739 740 |
root = &mapping->i_mmap; uprobe_munmap(vma, vma->vm_start, vma->vm_end); |
682968e0c uprobes/core: Opt... |
741 |
|
27ba0644e rmap: drop suppor... |
742 743 |
if (adjust_next) uprobe_munmap(next, next->vm_start, next->vm_end); |
682968e0c uprobes/core: Opt... |
744 |
|
83cde9e8b mm: use new helpe... |
745 |
i_mmap_lock_write(mapping); |
1da177e4c Linux-2.6.12-rc2 |
746 |
if (insert) { |
1da177e4c Linux-2.6.12-rc2 |
747 |
/* |
6b2dbba8b mm: replace vma p... |
748 |
* Put into interval tree now, so instantiated pages |
1da177e4c Linux-2.6.12-rc2 |
749 750 751 752 753 754 755 |
* 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); } } |
bf181b9f9 mm anon rmap: rep... |
756 757 758 759 |
anon_vma = vma->anon_vma; if (!anon_vma && adjust_next) anon_vma = next->anon_vma; if (anon_vma) { |
e86f15ee6 mm: vma_merge: fi... |
760 761 |
VM_WARN_ON(adjust_next && next->anon_vma && anon_vma != next->anon_vma); |
4fc3f1d66 mm/rmap, migratio... |
762 |
anon_vma_lock_write(anon_vma); |
bf181b9f9 mm anon rmap: rep... |
763 764 765 766 |
anon_vma_interval_tree_pre_update_vma(vma); if (adjust_next) anon_vma_interval_tree_pre_update_vma(next); } |
012f18004 mm: always lock t... |
767 |
|
1da177e4c Linux-2.6.12-rc2 |
768 769 |
if (root) { flush_dcache_mmap_lock(mapping); |
6b2dbba8b mm: replace vma p... |
770 |
vma_interval_tree_remove(vma, root); |
1da177e4c Linux-2.6.12-rc2 |
771 |
if (adjust_next) |
6b2dbba8b mm: replace vma p... |
772 |
vma_interval_tree_remove(next, root); |
1da177e4c Linux-2.6.12-rc2 |
773 |
} |
d37371870 mm: augment vma r... |
774 775 776 777 778 779 780 781 |
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 |
782 783 784 785 786 787 788 789 |
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... |
790 791 |
vma_interval_tree_insert(next, root); vma_interval_tree_insert(vma, root); |
1da177e4c Linux-2.6.12-rc2 |
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. */ |
e86f15ee6 mm: vma_merge: fi... |
800 801 802 |
if (remove_next != 3) __vma_unlink_prev(mm, next, vma); else |
8f26e0b17 mm: vma_merge: co... |
803 804 805 806 807 808 809 810 811 812 |
/* * vma is not before next if they've been * swapped. * * pre-swap() next->vm_start was reduced so * tell validate_mm_rb to ignore pre-swap() * "next" (which is stored in post-swap() * "vma"). */ __vma_unlink_common(mm, next, NULL, false, vma); |
1da177e4c Linux-2.6.12-rc2 |
813 814 |
if (file) __remove_shared_vm_struct(next, file, mapping); |
1da177e4c Linux-2.6.12-rc2 |
815 816 817 818 819 820 821 |
} 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... |
822 823 824 825 826 827 828 829 830 |
} 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 |
831 |
} |
bf181b9f9 mm anon rmap: rep... |
832 833 834 835 |
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... |
836 |
anon_vma_unlock_write(anon_vma); |
bf181b9f9 mm anon rmap: rep... |
837 |
} |
1da177e4c Linux-2.6.12-rc2 |
838 |
if (mapping) |
83cde9e8b mm: use new helpe... |
839 |
i_mmap_unlock_write(mapping); |
1da177e4c Linux-2.6.12-rc2 |
840 |
|
2b1444983 uprobes, mm, x86:... |
841 |
if (root) { |
7b2d81d48 uprobes/core: Cle... |
842 |
uprobe_mmap(vma); |
2b1444983 uprobes, mm, x86:... |
843 844 |
if (adjust_next) |
7b2d81d48 uprobes/core: Cle... |
845 |
uprobe_mmap(next); |
2b1444983 uprobes, mm, x86:... |
846 |
} |
1da177e4c Linux-2.6.12-rc2 |
847 |
if (remove_next) { |
925d1c401 procfs task exe s... |
848 |
if (file) { |
cbc91f71b uprobes/core: Dec... |
849 |
uprobe_munmap(next, next->vm_start, next->vm_end); |
1da177e4c Linux-2.6.12-rc2 |
850 |
fput(file); |
925d1c401 procfs task exe s... |
851 |
} |
5beb49305 mm: change anon_v... |
852 853 |
if (next->anon_vma) anon_vma_merge(vma, next); |
1da177e4c Linux-2.6.12-rc2 |
854 |
mm->map_count--; |
3964acd0d mm: mempolicy: fi... |
855 |
mpol_put(vma_policy(next)); |
1da177e4c Linux-2.6.12-rc2 |
856 857 858 859 860 861 |
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. */ |
e86f15ee6 mm: vma_merge: fi... |
862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 |
if (remove_next != 3) { /* * If "next" was removed and vma->vm_end was * expanded (up) over it, in turn * "next->vm_prev->vm_end" changed and the * "vma->vm_next" gap must be updated. */ next = vma->vm_next; } else { /* * For the scope of the comment "next" and * "vma" considered pre-swap(): if "vma" was * removed, next->vm_start was expanded (down) * over it and the "next" gap must be updated. * Because of the swap() the post-swap() "vma" * actually points to pre-swap() "next" * (post-swap() "next" as opposed is now a * dangling pointer). */ next = vma; } |
734537c9c mm: fix use-after... |
883 884 885 |
if (remove_next == 2) { remove_next = 1; end = next->vm_end; |
1da177e4c Linux-2.6.12-rc2 |
886 |
goto again; |
734537c9c mm: fix use-after... |
887 |
} |
d37371870 mm: augment vma r... |
888 889 |
else if (next) vma_gap_update(next); |
fb8c41e9a mm: vma_adjust: r... |
890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 |
else { /* * If remove_next == 2 we obviously can't * reach this path. * * If remove_next == 3 we can't reach this * path because pre-swap() next is always not * NULL. pre-swap() "next" is not being * removed and its next->vm_end is not altered * (and furthermore "end" already matches * next->vm_end in remove_next == 3). * * We reach this only in the remove_next == 1 * case if the "next" vma that was removed was * the highest vma of the mm. However in such * case next->vm_end == "end" and the extended * "vma" has vma->vm_end == next->vm_end so * mm->highest_vm_end doesn't need any update * in remove_next == 1 case. */ VM_WARN_ON(mm->highest_vm_end != end); } |
1da177e4c Linux-2.6.12-rc2 |
912 |
} |
2b1444983 uprobes, mm, x86:... |
913 |
if (insert && file) |
7b2d81d48 uprobes/core: Cle... |
914 |
uprobe_mmap(insert); |
1da177e4c Linux-2.6.12-rc2 |
915 916 |
validate_mm(mm); |
5beb49305 mm: change anon_v... |
917 918 |
return 0; |
1da177e4c Linux-2.6.12-rc2 |
919 920 921 922 923 924 |
} /* * 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 |
925 |
static inline int is_mergeable_vma(struct vm_area_struct *vma, |
19a809afe userfaultfd: teac... |
926 927 |
struct file *file, unsigned long vm_flags, struct vm_userfaultfd_ctx vm_userfaultfd_ctx) |
1da177e4c Linux-2.6.12-rc2 |
928 |
{ |
34228d473 mm: ignore VM_SOF... |
929 930 931 932 933 934 935 936 937 |
/* * VM_SOFTDIRTY should not prevent from VMA merging, if we * match the flags but dirty bit -- the caller should mark * merged VMA as dirty. If dirty bit won't be excluded from * comparison, we increase pressue on the memory system forcing * the kernel to generate new VMAs when old one could be * extended instead. */ if ((vma->vm_flags ^ vm_flags) & ~VM_SOFTDIRTY) |
1da177e4c Linux-2.6.12-rc2 |
938 939 940 941 942 |
return 0; if (vma->vm_file != file) return 0; if (vma->vm_ops && vma->vm_ops->close) return 0; |
19a809afe userfaultfd: teac... |
943 944 |
if (!is_mergeable_vm_userfaultfd_ctx(vma, vm_userfaultfd_ctx)) return 0; |
1da177e4c Linux-2.6.12-rc2 |
945 946 947 948 |
return 1; } static inline int is_mergeable_anon_vma(struct anon_vma *anon_vma1, |
965f55dea mmap: avoid mergi... |
949 950 |
struct anon_vma *anon_vma2, struct vm_area_struct *vma) |
1da177e4c Linux-2.6.12-rc2 |
951 |
{ |
965f55dea mmap: avoid mergi... |
952 953 954 955 956 957 958 959 |
/* * 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 |
960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 |
} /* * Return true if we can merge this (vm_flags,anon_vma,file,vm_pgoff) * in front of (at a lower virtual address and file offset than) the vma. * * We cannot merge two vmas if they have differently assigned (non-NULL) * anon_vmas, nor if same anon_vma is assigned but offsets incompatible. * * We don't check here for the merged mmap wrapping around the end of pagecache * indices (16TB on ia32) because do_mmap_pgoff() does not permit mmap's which * wrap, nor mmaps which cover the final page at index -1UL. */ static int can_vma_merge_before(struct vm_area_struct *vma, unsigned long vm_flags, |
19a809afe userfaultfd: teac... |
975 976 977 |
struct anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff, struct vm_userfaultfd_ctx vm_userfaultfd_ctx) |
1da177e4c Linux-2.6.12-rc2 |
978 |
{ |
19a809afe userfaultfd: teac... |
979 |
if (is_mergeable_vma(vma, file, vm_flags, vm_userfaultfd_ctx) && |
965f55dea mmap: avoid mergi... |
980 |
is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) { |
1da177e4c Linux-2.6.12-rc2 |
981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 |
if (vma->vm_pgoff == vm_pgoff) return 1; } return 0; } /* * Return true if we can merge this (vm_flags,anon_vma,file,vm_pgoff) * beyond (at a higher virtual address and file offset than) the vma. * * We cannot merge two vmas if they have differently assigned (non-NULL) * anon_vmas, nor if same anon_vma is assigned but offsets incompatible. */ static int can_vma_merge_after(struct vm_area_struct *vma, unsigned long vm_flags, |
19a809afe userfaultfd: teac... |
996 997 998 |
struct anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff, struct vm_userfaultfd_ctx vm_userfaultfd_ctx) |
1da177e4c Linux-2.6.12-rc2 |
999 |
{ |
19a809afe userfaultfd: teac... |
1000 |
if (is_mergeable_vma(vma, file, vm_flags, vm_userfaultfd_ctx) && |
965f55dea mmap: avoid mergi... |
1001 |
is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) { |
1da177e4c Linux-2.6.12-rc2 |
1002 |
pgoff_t vm_pglen; |
d6e932177 mm: use vma_pages... |
1003 |
vm_pglen = vma_pages(vma); |
1da177e4c Linux-2.6.12-rc2 |
1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 |
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 |
e86f15ee6 mm: vma_merge: fi... |
1031 |
* mremap move: PPPPXXXXXXXX 8 |
1da177e4c Linux-2.6.12-rc2 |
1032 1033 1034 1035 |
* AAAA * PPPP NNNN PPPPPPPPPPPP PPPPPPPPNNNN PPPPNNNNNNNN * might become case 1 below case 2 below case 3 below * |
e86f15ee6 mm: vma_merge: fi... |
1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 |
* It is important for case 8 that the the vma NNNN overlapping the * region AAAA is never going to extended over XXXX. Instead XXXX must * be extended in region AAAA and NNNN must be removed. This way in * all cases where vma_merge succeeds, the moment vma_adjust drops the * rmap_locks, the properties of the merged vma will be already * correct for the whole merged range. Some of those properties like * vm_page_prot/vm_flags may be accessed by rmap_walks and they must * be correct for the whole merged range immediately after the * rmap_locks are released. Otherwise if XXXX would be removed and * NNNN would be extended over the XXXX range, remove_migration_ptes * or other rmap walkers (if working on addresses beyond the "end" * parameter) may establish ptes with the wrong permissions of NNNN * instead of the right permissions of XXXX. |
1da177e4c Linux-2.6.12-rc2 |
1049 1050 1051 1052 |
*/ struct vm_area_struct *vma_merge(struct mm_struct *mm, struct vm_area_struct *prev, unsigned long addr, unsigned long end, unsigned long vm_flags, |
cc71aba34 mm/mmap.c: whites... |
1053 |
struct anon_vma *anon_vma, struct file *file, |
19a809afe userfaultfd: teac... |
1054 1055 |
pgoff_t pgoff, struct mempolicy *policy, struct vm_userfaultfd_ctx vm_userfaultfd_ctx) |
1da177e4c Linux-2.6.12-rc2 |
1056 1057 1058 |
{ pgoff_t pglen = (end - addr) >> PAGE_SHIFT; struct vm_area_struct *area, *next; |
5beb49305 mm: change anon_v... |
1059 |
int err; |
1da177e4c Linux-2.6.12-rc2 |
1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 |
/* * 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; |
e86f15ee6 mm: vma_merge: fi... |
1073 |
if (area && area->vm_end == end) /* cases 6, 7, 8 */ |
1da177e4c Linux-2.6.12-rc2 |
1074 |
next = next->vm_next; |
e86f15ee6 mm: vma_merge: fi... |
1075 1076 1077 1078 |
/* verify some invariant that must be enforced by the caller */ VM_WARN_ON(prev && addr <= prev->vm_start); VM_WARN_ON(area && end > area->vm_end); VM_WARN_ON(addr >= end); |
1da177e4c Linux-2.6.12-rc2 |
1079 1080 1081 1082 |
/* * Can it merge with the predecessor? */ if (prev && prev->vm_end == addr && |
cc71aba34 mm/mmap.c: whites... |
1083 |
mpol_equal(vma_policy(prev), policy) && |
1da177e4c Linux-2.6.12-rc2 |
1084 |
can_vma_merge_after(prev, vm_flags, |
19a809afe userfaultfd: teac... |
1085 1086 |
anon_vma, file, pgoff, vm_userfaultfd_ctx)) { |
1da177e4c Linux-2.6.12-rc2 |
1087 1088 1089 1090 1091 1092 |
/* * OK, it can. Can we now merge in the successor as well? */ if (next && end == next->vm_start && mpol_equal(policy, vma_policy(next)) && can_vma_merge_before(next, vm_flags, |
19a809afe userfaultfd: teac... |
1093 1094 1095 |
anon_vma, file, pgoff+pglen, vm_userfaultfd_ctx) && |
1da177e4c Linux-2.6.12-rc2 |
1096 |
is_mergeable_anon_vma(prev->anon_vma, |
965f55dea mmap: avoid mergi... |
1097 |
next->anon_vma, NULL)) { |
1da177e4c Linux-2.6.12-rc2 |
1098 |
/* cases 1, 6 */ |
e86f15ee6 mm: vma_merge: fi... |
1099 1100 1101 |
err = __vma_adjust(prev, prev->vm_start, next->vm_end, prev->vm_pgoff, NULL, prev); |
1da177e4c Linux-2.6.12-rc2 |
1102 |
} else /* cases 2, 5, 7 */ |
e86f15ee6 mm: vma_merge: fi... |
1103 1104 |
err = __vma_adjust(prev, prev->vm_start, end, prev->vm_pgoff, NULL, prev); |
5beb49305 mm: change anon_v... |
1105 1106 |
if (err) return NULL; |
6d50e60cd mm, thp: fix coll... |
1107 |
khugepaged_enter_vma_merge(prev, vm_flags); |
1da177e4c Linux-2.6.12-rc2 |
1108 1109 1110 1111 1112 1113 1114 |
return prev; } /* * Can this new request be merged in front of next? */ if (next && end == next->vm_start && |
cc71aba34 mm/mmap.c: whites... |
1115 |
mpol_equal(policy, vma_policy(next)) && |
1da177e4c Linux-2.6.12-rc2 |
1116 |
can_vma_merge_before(next, vm_flags, |
19a809afe userfaultfd: teac... |
1117 1118 |
anon_vma, file, pgoff+pglen, vm_userfaultfd_ctx)) { |
1da177e4c Linux-2.6.12-rc2 |
1119 |
if (prev && addr < prev->vm_end) /* case 4 */ |
e86f15ee6 mm: vma_merge: fi... |
1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 |
err = __vma_adjust(prev, prev->vm_start, addr, prev->vm_pgoff, NULL, next); else { /* cases 3, 8 */ err = __vma_adjust(area, addr, next->vm_end, next->vm_pgoff - pglen, NULL, next); /* * In case 3 area is already equal to next and * this is a noop, but in case 8 "area" has * been removed and next was expanded over it. */ area = next; } |
5beb49305 mm: change anon_v... |
1132 1133 |
if (err) return NULL; |
6d50e60cd mm, thp: fix coll... |
1134 |
khugepaged_enter_vma_merge(area, vm_flags); |
1da177e4c Linux-2.6.12-rc2 |
1135 1136 1137 1138 1139 1140 1141 |
return area; } return NULL; } /* |
d0e9fe175 Simplify and comm... |
1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 |
* Rough compatbility check to quickly see if it's even worth looking * at sharing an anon_vma. * * They need to have the same vm_file, and the flags can only differ * in things that mprotect may change. * * NOTE! The fact that we share an anon_vma doesn't _have_ to mean that * we can merge the two vma's. For example, we refuse to merge a vma if * there is a vm_ops->close() function, because that indicates that the * driver is doing some kind of reference counting. But that doesn't * really matter for the anon_vma sharing case. */ static int anon_vma_compatible(struct vm_area_struct *a, struct vm_area_struct *b) { return a->vm_end == b->vm_start && mpol_equal(vma_policy(a), vma_policy(b)) && a->vm_file == b->vm_file && |
34228d473 mm: ignore VM_SOF... |
1159 |
!((a->vm_flags ^ b->vm_flags) & ~(VM_READ|VM_WRITE|VM_EXEC|VM_SOFTDIRTY)) && |
d0e9fe175 Simplify and comm... |
1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 |
b->vm_pgoff == a->vm_pgoff + ((b->vm_start - a->vm_start) >> PAGE_SHIFT); } /* * Do some basic sanity checking to see if we can re-use the anon_vma * from 'old'. The 'a'/'b' vma's are in VM order - one of them will be * the same as 'old', the other will be the new one that is trying * to share the anon_vma. * * NOTE! This runs with mm_sem held for reading, so it is possible that * the anon_vma of 'old' is concurrently in the process of being set up * by another page fault trying to merge _that_. But that's ok: if it * is being set up, that automatically means that it will be a singleton * acceptable for merging, so we can do all of this optimistically. But |
4db0c3c29 mm: remove rest o... |
1174 |
* we do that READ_ONCE() to make sure that we never re-load the pointer. |
d0e9fe175 Simplify and comm... |
1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 |
* * IOW: that the "list_is_singular()" test on the anon_vma_chain only * matters for the 'stable anon_vma' case (ie the thing we want to avoid * is to return an anon_vma that is "complex" due to having gone through * a fork). * * We also make sure that the two vma's are compatible (adjacent, * and with the same memory policies). That's all stable, even with just * a read lock on the mm_sem. */ static struct anon_vma *reusable_anon_vma(struct vm_area_struct *old, struct vm_area_struct *a, struct vm_area_struct *b) { if (anon_vma_compatible(a, b)) { |
4db0c3c29 mm: remove rest o... |
1188 |
struct anon_vma *anon_vma = READ_ONCE(old->anon_vma); |
d0e9fe175 Simplify and comm... |
1189 1190 1191 1192 1193 1194 1195 1196 |
if (anon_vma && list_is_singular(&old->anon_vma_chain)) return anon_vma; } return NULL; } /* |
1da177e4c Linux-2.6.12-rc2 |
1197 1198 1199 1200 1201 1202 1203 1204 1205 |
* 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... |
1206 |
struct anon_vma *anon_vma; |
1da177e4c Linux-2.6.12-rc2 |
1207 |
struct vm_area_struct *near; |
1da177e4c Linux-2.6.12-rc2 |
1208 1209 1210 1211 |
near = vma->vm_next; if (!near) goto try_prev; |
d0e9fe175 Simplify and comm... |
1212 1213 1214 |
anon_vma = reusable_anon_vma(near, vma, near); if (anon_vma) return anon_vma; |
1da177e4c Linux-2.6.12-rc2 |
1215 |
try_prev: |
9be34c9d5 mm: get rid of th... |
1216 |
near = vma->vm_prev; |
1da177e4c Linux-2.6.12-rc2 |
1217 1218 |
if (!near) goto none; |
d0e9fe175 Simplify and comm... |
1219 1220 1221 |
anon_vma = reusable_anon_vma(near, near, vma); if (anon_vma) return anon_vma; |
1da177e4c Linux-2.6.12-rc2 |
1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 |
none: /* * There's no absolute need to look only at touching neighbours: * we could search further afield for "compatible" anon_vmas. * But it would probably just be a waste of time searching, * or lead to too many vmas hanging off the same anon_vma. * We're trying to allow mprotect remerging later on, * not trying to minimize memory used for anon_vmas. */ return NULL; } |
1da177e4c Linux-2.6.12-rc2 |
1233 |
/* |
404015308 security: trim se... |
1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 |
* If a hint addr is less than mmap_min_addr change hint to be as * low as possible but still greater than mmap_min_addr */ static inline unsigned long round_hint_to_min(unsigned long hint) { hint &= PAGE_MASK; if (((void *)hint != NULL) && (hint < mmap_min_addr)) return PAGE_ALIGN(mmap_min_addr); return hint; } |
363ee17f0 mm/mmap.c: add ml... |
1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 |
static inline int mlock_future_check(struct mm_struct *mm, unsigned long flags, unsigned long len) { unsigned long locked, lock_limit; /* mlock MCL_FUTURE? */ if (flags & VM_LOCKED) { locked = len >> PAGE_SHIFT; locked += mm->locked_vm; lock_limit = rlimit(RLIMIT_MEMLOCK); lock_limit >>= PAGE_SHIFT; if (locked > lock_limit && !capable(CAP_IPC_LOCK)) return -EAGAIN; } return 0; } |
404015308 security: trim se... |
1262 |
/* |
27f5de796 mm: Fix problem o... |
1263 |
* The caller must hold down_write(¤t->mm->mmap_sem). |
1da177e4c Linux-2.6.12-rc2 |
1264 |
*/ |
1fcfd8db7 mm, mpx: add "vm_... |
1265 |
unsigned long do_mmap(struct file *file, unsigned long addr, |
1da177e4c Linux-2.6.12-rc2 |
1266 |
unsigned long len, unsigned long prot, |
1fcfd8db7 mm, mpx: add "vm_... |
1267 1268 |
unsigned long flags, vm_flags_t vm_flags, unsigned long pgoff, unsigned long *populate) |
1da177e4c Linux-2.6.12-rc2 |
1269 |
{ |
cc71aba34 mm/mmap.c: whites... |
1270 |
struct mm_struct *mm = current->mm; |
62b5f7d01 mm/core, x86/mm/p... |
1271 |
int pkey = 0; |
1da177e4c Linux-2.6.12-rc2 |
1272 |
|
41badc15c mm: make do_mmap_... |
1273 |
*populate = 0; |
bebeb3d68 mm: introduce mm_... |
1274 |
|
e37609bb3 mm/mmap.c: optimi... |
1275 1276 |
if (!len) return -EINVAL; |
1da177e4c Linux-2.6.12-rc2 |
1277 1278 1279 1280 1281 1282 1283 |
/* * Does the application expect PROT_READ to imply PROT_EXEC? * * (the exception is when the underlying filesystem is noexec * mounted, in which case we dont add PROT_EXEC.) */ if ((prot & PROT_READ) && (current->personality & READ_IMPLIES_EXEC)) |
90f8572b0 vfs: Commit to ne... |
1284 |
if (!(file && path_noexec(&file->f_path))) |
1da177e4c Linux-2.6.12-rc2 |
1285 |
prot |= PROT_EXEC; |
7cd94146c Security: round m... |
1286 1287 |
if (!(flags & MAP_FIXED)) addr = round_hint_to_min(addr); |
1da177e4c Linux-2.6.12-rc2 |
1288 1289 |
/* Careful about overflows.. */ len = PAGE_ALIGN(len); |
9206de95b Take arch_mmap_ch... |
1290 |
if (!len) |
1da177e4c Linux-2.6.12-rc2 |
1291 1292 1293 1294 |
return -ENOMEM; /* offset overflow? */ if ((pgoff + (len >> PAGE_SHIFT)) < pgoff) |
cc71aba34 mm/mmap.c: whites... |
1295 |
return -EOVERFLOW; |
1da177e4c Linux-2.6.12-rc2 |
1296 1297 1298 1299 1300 1301 1302 1303 1304 |
/* Too many mappings? */ if (mm->map_count > sysctl_max_map_count) return -ENOMEM; /* Obtain the address to map to. we verify (or select) it and ensure * that it represents a valid section of the address space. */ addr = get_unmapped_area(file, addr, len, pgoff, flags); |
de1741a13 mm/mmap: use offs... |
1305 |
if (offset_in_page(addr)) |
1da177e4c Linux-2.6.12-rc2 |
1306 |
return addr; |
62b5f7d01 mm/core, x86/mm/p... |
1307 1308 1309 1310 1311 |
if (prot == PROT_EXEC) { pkey = execute_only_pkey(mm); if (pkey < 0) pkey = 0; } |
1da177e4c Linux-2.6.12-rc2 |
1312 1313 1314 1315 |
/* Do simple checking here so the lower-level routines won't have * to. we assume access permissions have been handled by the open * of the memory object, so we don't do any here. */ |
62b5f7d01 mm/core, x86/mm/p... |
1316 |
vm_flags |= calc_vm_prot_bits(prot, pkey) | calc_vm_flag_bits(flags) | |
1da177e4c Linux-2.6.12-rc2 |
1317 |
mm->def_flags | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC; |
cdf7b3418 mmap: remove unne... |
1318 |
if (flags & MAP_LOCKED) |
1da177e4c Linux-2.6.12-rc2 |
1319 1320 |
if (!can_do_mlock()) return -EPERM; |
ba470de43 mmap: handle mloc... |
1321 |
|
363ee17f0 mm/mmap.c: add ml... |
1322 1323 |
if (mlock_future_check(mm, vm_flags, len)) return -EAGAIN; |
1da177e4c Linux-2.6.12-rc2 |
1324 |
|
1da177e4c Linux-2.6.12-rc2 |
1325 |
if (file) { |
077bf22b5 mm: do_mmap_pgoff... |
1326 |
struct inode *inode = file_inode(file); |
1da177e4c Linux-2.6.12-rc2 |
1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 |
switch (flags & MAP_TYPE) { case MAP_SHARED: if ((prot&PROT_WRITE) && !(file->f_mode&FMODE_WRITE)) return -EACCES; /* * Make sure we don't allow writing to an append-only * file.. */ if (IS_APPEND(inode) && (file->f_mode & FMODE_WRITE)) return -EACCES; /* * Make sure there are no mandatory locks on the file. */ |
d7a06983a locks: fix locks_... |
1342 |
if (locks_verify_locked(file)) |
1da177e4c Linux-2.6.12-rc2 |
1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 |
return -EAGAIN; vm_flags |= VM_SHARED | VM_MAYSHARE; if (!(file->f_mode & FMODE_WRITE)) vm_flags &= ~(VM_MAYWRITE | VM_SHARED); /* fall through */ case MAP_PRIVATE: if (!(file->f_mode & FMODE_READ)) return -EACCES; |
90f8572b0 vfs: Commit to ne... |
1353 |
if (path_noexec(&file->f_path)) { |
80c5606c3 Fix VM_MAYEXEC ca... |
1354 1355 1356 1357 |
if (vm_flags & VM_EXEC) return -EPERM; vm_flags &= ~VM_MAYEXEC; } |
80c5606c3 Fix VM_MAYEXEC ca... |
1358 |
|
72c2d5319 file->f_op is nev... |
1359 |
if (!file->f_op->mmap) |
80c5606c3 Fix VM_MAYEXEC ca... |
1360 |
return -ENODEV; |
b2c56e4f7 mm: shift VM_GROW... |
1361 1362 |
if (vm_flags & (VM_GROWSDOWN|VM_GROWSUP)) return -EINVAL; |
1da177e4c Linux-2.6.12-rc2 |
1363 1364 1365 1366 1367 1368 1369 1370 |
break; default: return -EINVAL; } } else { switch (flags & MAP_TYPE) { case MAP_SHARED: |
b2c56e4f7 mm: shift VM_GROW... |
1371 1372 |
if (vm_flags & (VM_GROWSDOWN|VM_GROWSUP)) return -EINVAL; |
ce3639426 mmap: fix petty b... |
1373 1374 1375 1376 |
/* * Ignore pgoff. */ pgoff = 0; |
1da177e4c Linux-2.6.12-rc2 |
1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 |
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 ... |
1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 |
/* * 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... |
1404 1405 1406 |
if (!IS_ERR_VALUE(addr) && ((vm_flags & VM_LOCKED) || (flags & (MAP_POPULATE | MAP_NONBLOCK)) == MAP_POPULATE)) |
41badc15c mm: make do_mmap_... |
1407 |
*populate = len; |
bebeb3d68 mm: introduce mm_... |
1408 |
return addr; |
0165ab443 split mmap |
1409 |
} |
6be5ceb02 VM: add "vm_mmap(... |
1410 |
|
66f0dc481 mm: move sys_mmap... |
1411 1412 1413 1414 1415 |
SYSCALL_DEFINE6(mmap_pgoff, unsigned long, addr, unsigned long, len, unsigned long, prot, unsigned long, flags, unsigned long, fd, unsigned long, pgoff) { struct file *file = NULL; |
1e3ee14b9 mm/mmap.c: do not... |
1416 |
unsigned long retval; |
66f0dc481 mm: move sys_mmap... |
1417 1418 |
if (!(flags & MAP_ANONYMOUS)) { |
120a795da audit mmap |
1419 |
audit_mmap_fd(fd, flags); |
66f0dc481 mm: move sys_mmap... |
1420 1421 |
file = fget(fd); if (!file) |
1e3ee14b9 mm/mmap.c: do not... |
1422 |
return -EBADF; |
af73e4d95 hugetlbfs: fix mm... |
1423 1424 |
if (is_file_hugepages(file)) len = ALIGN(len, huge_page_size(hstate_file(file))); |
493af5780 mmap: allow MAP_H... |
1425 1426 1427 |
retval = -EINVAL; if (unlikely(flags & MAP_HUGETLB && !is_file_hugepages(file))) goto out_fput; |
66f0dc481 mm: move sys_mmap... |
1428 1429 |
} else if (flags & MAP_HUGETLB) { struct user_struct *user = NULL; |
c103a4dc4 ipc/shmc.c: elimi... |
1430 |
struct hstate *hs; |
af73e4d95 hugetlbfs: fix mm... |
1431 |
|
c103a4dc4 ipc/shmc.c: elimi... |
1432 |
hs = hstate_sizelog((flags >> MAP_HUGE_SHIFT) & SHM_HUGE_MASK); |
091d0d55b shm: fix null poi... |
1433 1434 1435 1436 |
if (!hs) return -EINVAL; len = ALIGN(len, huge_page_size(hs)); |
66f0dc481 mm: move sys_mmap... |
1437 1438 1439 1440 1441 1442 |
/* * 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... |
1443 |
file = hugetlb_file_setup(HUGETLB_ANON_FILE, len, |
42d7395fe mm: support more ... |
1444 1445 1446 |
VM_NORESERVE, &user, HUGETLB_ANONHUGE_INODE, (flags >> MAP_HUGE_SHIFT) & MAP_HUGE_MASK); |
66f0dc481 mm: move sys_mmap... |
1447 1448 1449 1450 1451 |
if (IS_ERR(file)) return PTR_ERR(file); } flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); |
9fbeb5ab5 mm: make vm_mmap ... |
1452 |
retval = vm_mmap_pgoff(file, addr, len, prot, flags, pgoff); |
493af5780 mmap: allow MAP_H... |
1453 |
out_fput: |
66f0dc481 mm: move sys_mmap... |
1454 1455 |
if (file) fput(file); |
66f0dc481 mm: move sys_mmap... |
1456 1457 |
return retval; } |
a4679373c Add generic sys_o... |
1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 |
#ifdef __ARCH_WANT_SYS_OLD_MMAP struct mmap_arg_struct { unsigned long addr; unsigned long len; unsigned long prot; unsigned long flags; unsigned long fd; unsigned long offset; }; SYSCALL_DEFINE1(old_mmap, struct mmap_arg_struct __user *, arg) { struct mmap_arg_struct a; if (copy_from_user(&a, arg, sizeof(a))) return -EFAULT; |
de1741a13 mm/mmap: use offs... |
1474 |
if (offset_in_page(a.offset)) |
a4679373c Add generic sys_o... |
1475 1476 1477 1478 1479 1480 |
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 ... |
1481 1482 1483 1484 1485 1486 |
/* * 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). */ |
6d2329f88 mm: vm_page_prot:... |
1487 |
int vma_wants_writenotify(struct vm_area_struct *vma, pgprot_t vm_page_prot) |
4e950f6f0 Remove fs.h from ... |
1488 |
{ |
ca16d140a mm: don't access ... |
1489 |
vm_flags_t vm_flags = vma->vm_flags; |
8a04446ab mm, dax: VMA with... |
1490 |
const struct vm_operations_struct *vm_ops = vma->vm_ops; |
4e950f6f0 Remove fs.h from ... |
1491 1492 1493 1494 1495 1496 |
/* If it was private or non-writable, the write bit is already clear */ if ((vm_flags & (VM_WRITE|VM_SHARED)) != ((VM_WRITE|VM_SHARED))) return 0; /* The backer wishes to know when pages are first written to? */ |
8a04446ab mm, dax: VMA with... |
1497 |
if (vm_ops && (vm_ops->page_mkwrite || vm_ops->pfn_mkwrite)) |
4e950f6f0 Remove fs.h from ... |
1498 |
return 1; |
64e455079 mm: softdirty: en... |
1499 1500 |
/* The open routine did something to the protections that pgprot_modify * won't preserve? */ |
6d2329f88 mm: vm_page_prot:... |
1501 1502 |
if (pgprot_val(vm_page_prot) != pgprot_val(vm_pgprot_modify(vm_page_prot, vm_flags))) |
4e950f6f0 Remove fs.h from ... |
1503 |
return 0; |
64e455079 mm: softdirty: en... |
1504 1505 1506 |
/* Do we need to track softdirty? */ if (IS_ENABLED(CONFIG_MEM_SOFT_DIRTY) && !(vm_flags & VM_SOFTDIRTY)) return 1; |
4e950f6f0 Remove fs.h from ... |
1507 |
/* Specialty mapping? */ |
4b6e1e370 mm: kill vma flag... |
1508 |
if (vm_flags & VM_PFNMAP) |
4e950f6f0 Remove fs.h from ... |
1509 1510 1511 1512 1513 1514 |
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... |
1515 1516 |
/* * We account for memory if it's a private writeable mapping, |
5a6fe1259 Do not account fo... |
1517 |
* not hugepages and VM_NORESERVE wasn't set. |
fc8744adc Stop playing sill... |
1518 |
*/ |
ca16d140a mm: don't access ... |
1519 |
static inline int accountable_mapping(struct file *file, vm_flags_t vm_flags) |
fc8744adc Stop playing sill... |
1520 |
{ |
5a6fe1259 Do not account fo... |
1521 1522 1523 1524 1525 1526 |
/* * 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... |
1527 1528 |
return (vm_flags & (VM_NORESERVE | VM_SHARED | VM_WRITE)) == VM_WRITE; } |
0165ab443 split mmap |
1529 |
unsigned long mmap_region(struct file *file, unsigned long addr, |
c22c0d634 mm: remove flags ... |
1530 |
unsigned long len, vm_flags_t vm_flags, unsigned long pgoff) |
0165ab443 split mmap |
1531 1532 1533 |
{ struct mm_struct *mm = current->mm; struct vm_area_struct *vma, *prev; |
0165ab443 split mmap |
1534 1535 1536 |
int error; struct rb_node **rb_link, *rb_parent; unsigned long charged = 0; |
0165ab443 split mmap |
1537 |
|
e8420a8ec mm/mmap: check fo... |
1538 |
/* Check against address space limit. */ |
846383359 mm: rework virtua... |
1539 |
if (!may_expand_vm(mm, vm_flags, len >> PAGE_SHIFT)) { |
e8420a8ec mm/mmap: check fo... |
1540 1541 1542 1543 1544 1545 |
unsigned long nr_pages; /* * MAP_FIXED may remove pages of mappings that intersects with * requested mapping. Account for the pages it would unmap. */ |
e8420a8ec mm/mmap: check fo... |
1546 |
nr_pages = count_vma_pages_range(mm, addr, addr + len); |
846383359 mm: rework virtua... |
1547 1548 |
if (!may_expand_vm(mm, vm_flags, (len >> PAGE_SHIFT) - nr_pages)) |
e8420a8ec mm/mmap: check fo... |
1549 1550 |
return -ENOMEM; } |
1da177e4c Linux-2.6.12-rc2 |
1551 |
/* Clear old maps */ |
9fcd14571 mm/mmap.c: use wh... |
1552 1553 |
while (find_vma_links(mm, addr, addr + len, &prev, &rb_link, &rb_parent)) { |
1da177e4c Linux-2.6.12-rc2 |
1554 1555 |
if (do_munmap(mm, addr, len)) return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
1556 |
} |
fc8744adc Stop playing sill... |
1557 |
/* |
fc8744adc Stop playing sill... |
1558 1559 |
* Private writable mapping: check memory availability */ |
5a6fe1259 Do not account fo... |
1560 |
if (accountable_mapping(file, vm_flags)) { |
fc8744adc Stop playing sill... |
1561 |
charged = len >> PAGE_SHIFT; |
191c54244 mm: collapse secu... |
1562 |
if (security_vm_enough_memory_mm(mm, charged)) |
fc8744adc Stop playing sill... |
1563 1564 |
return -ENOMEM; vm_flags |= VM_ACCOUNT; |
1da177e4c Linux-2.6.12-rc2 |
1565 1566 1567 |
} /* |
de33c8db5 Fix OOPS in mmap_... |
1568 |
* Can we just expand an old mapping? |
1da177e4c Linux-2.6.12-rc2 |
1569 |
*/ |
19a809afe userfaultfd: teac... |
1570 1571 |
vma = vma_merge(mm, prev, addr, addr + len, vm_flags, NULL, file, pgoff, NULL, NULL_VM_UFFD_CTX); |
de33c8db5 Fix OOPS in mmap_... |
1572 1573 |
if (vma) goto out; |
1da177e4c Linux-2.6.12-rc2 |
1574 1575 1576 1577 1578 1579 |
/* * 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... |
1580 |
vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); |
1da177e4c Linux-2.6.12-rc2 |
1581 1582 1583 1584 |
if (!vma) { error = -ENOMEM; goto unacct_error; } |
1da177e4c Linux-2.6.12-rc2 |
1585 1586 1587 1588 1589 |
vma->vm_mm = mm; vma->vm_start = addr; vma->vm_end = addr + len; vma->vm_flags = vm_flags; |
3ed75eb8f setup vma->vm_pag... |
1590 |
vma->vm_page_prot = vm_get_page_prot(vm_flags); |
1da177e4c Linux-2.6.12-rc2 |
1591 |
vma->vm_pgoff = pgoff; |
5beb49305 mm: change anon_v... |
1592 |
INIT_LIST_HEAD(&vma->anon_vma_chain); |
1da177e4c Linux-2.6.12-rc2 |
1593 1594 |
if (file) { |
1da177e4c Linux-2.6.12-rc2 |
1595 1596 1597 1598 |
if (vm_flags & VM_DENYWRITE) { error = deny_write_access(file); if (error) goto free_vma; |
1da177e4c Linux-2.6.12-rc2 |
1599 |
} |
4bb5f5d93 mm: allow drivers... |
1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 |
if (vm_flags & VM_SHARED) { error = mapping_map_writable(file->f_mapping); if (error) goto allow_write_and_free_vma; } /* ->mmap() can change vma->vm_file, but must guarantee that * vma_link() below can deny write-access if VM_DENYWRITE is set * and map writably if VM_SHARED is set. This usually means the * new file must not have been exposed to user-space, yet. */ |
cb0942b81 make get_file() r... |
1611 |
vma->vm_file = get_file(file); |
1da177e4c Linux-2.6.12-rc2 |
1612 1613 1614 |
error = file->f_op->mmap(file, vma); if (error) goto unmap_and_free_vma; |
f8dbf0a7a mmap: save some c... |
1615 1616 1617 1618 1619 |
/* Can addr have changed?? * * Answer: Yes, several device drivers can do it in their * f_op->mmap method. -DaveM |
2897b4d29 mm: WARN_ON_ONCE ... |
1620 1621 |
* Bug: If addr is changed, prev, rb_link, rb_parent should * be updated for vma_link() |
f8dbf0a7a mmap: save some c... |
1622 |
*/ |
2897b4d29 mm: WARN_ON_ONCE ... |
1623 |
WARN_ON_ONCE(addr != vma->vm_start); |
f8dbf0a7a mmap: save some c... |
1624 |
addr = vma->vm_start; |
f8dbf0a7a mmap: save some c... |
1625 |
vm_flags = vma->vm_flags; |
1da177e4c Linux-2.6.12-rc2 |
1626 1627 1628 1629 1630 |
} else if (vm_flags & VM_SHARED) { error = shmem_zero_setup(vma); if (error) goto free_vma; } |
de33c8db5 Fix OOPS in mmap_... |
1631 |
vma_link(mm, vma, prev, rb_link, rb_parent); |
4d3d5b41a mmap_region: clea... |
1632 |
/* Once vma denies write, undo our temporary denial count */ |
4bb5f5d93 mm: allow drivers... |
1633 1634 1635 1636 1637 1638 |
if (file) { if (vm_flags & VM_SHARED) mapping_unmap_writable(file->f_mapping); if (vm_flags & VM_DENYWRITE) allow_write_access(file); } |
e86867720 mm: mmap_region: ... |
1639 |
file = vma->vm_file; |
4d3d5b41a mmap_region: clea... |
1640 |
out: |
cdd6c482c perf: Do the big ... |
1641 |
perf_event_mmap(vma); |
0a4a93919 perf_counter: exe... |
1642 |
|
846383359 mm: rework virtua... |
1643 |
vm_stat_account(mm, vm_flags, len >> PAGE_SHIFT); |
1da177e4c Linux-2.6.12-rc2 |
1644 |
if (vm_flags & VM_LOCKED) { |
bebeb3d68 mm: introduce mm_... |
1645 1646 |
if (!((vm_flags & VM_SPECIAL) || is_vm_hugetlb_page(vma) || vma == get_gate_vma(current->mm))) |
06f9d8c2b mm: mlock_vma_pag... |
1647 |
mm->locked_vm += (len >> PAGE_SHIFT); |
bebeb3d68 mm: introduce mm_... |
1648 |
else |
de60f5f10 mm: introduce VM_... |
1649 |
vma->vm_flags &= VM_LOCKED_CLEAR_MASK; |
bebeb3d68 mm: introduce mm_... |
1650 |
} |
2b1444983 uprobes, mm, x86:... |
1651 |
|
c7a3a88c9 uprobes: Fix mmap... |
1652 1653 |
if (file) uprobe_mmap(vma); |
2b1444983 uprobes, mm, x86:... |
1654 |
|
d9104d1ca mm: track vma cha... |
1655 1656 1657 1658 1659 1660 1661 1662 |
/* * New (or expanded) vma always get soft dirty status. * Otherwise user-space soft-dirty page tracker won't * be able to distinguish situation when vma area unmapped, * then new mapped in-place (which must be aimed as * a completely new data area). */ vma->vm_flags |= VM_SOFTDIRTY; |
64e455079 mm: softdirty: en... |
1663 |
vma_set_page_prot(vma); |
1da177e4c Linux-2.6.12-rc2 |
1664 1665 1666 |
return addr; unmap_and_free_vma: |
1da177e4c Linux-2.6.12-rc2 |
1667 1668 1669 1670 |
vma->vm_file = NULL; fput(file); /* Undo any partial mapping done by a device driver. */ |
e0da382c9 [PATCH] freepgt: ... |
1671 1672 |
unmap_region(mm, vma, prev, vma->vm_start, vma->vm_end); charged = 0; |
4bb5f5d93 mm: allow drivers... |
1673 1674 1675 1676 1677 |
if (vm_flags & VM_SHARED) mapping_unmap_writable(file->f_mapping); allow_write_and_free_vma: if (vm_flags & VM_DENYWRITE) allow_write_access(file); |
1da177e4c Linux-2.6.12-rc2 |
1678 1679 1680 1681 1682 1683 1684 |
free_vma: kmem_cache_free(vm_area_cachep, vma); unacct_error: if (charged) vm_unacct_memory(charged); return error; } |
db4fbfb95 mm: vm_unmapped_a... |
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 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 |
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 |
1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 |
/* 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... |
1902 |
struct vm_unmapped_area_info info; |
1da177e4c Linux-2.6.12-rc2 |
1903 |
|
2afc745f3 mm: ensure get_un... |
1904 |
if (len > TASK_SIZE - mmap_min_addr) |
1da177e4c Linux-2.6.12-rc2 |
1905 |
return -ENOMEM; |
06abdfb47 get_unmapped_area... |
1906 1907 |
if (flags & MAP_FIXED) return addr; |
1da177e4c Linux-2.6.12-rc2 |
1908 1909 1910 |
if (addr) { addr = PAGE_ALIGN(addr); vma = find_vma(mm, addr); |
2afc745f3 mm: ensure get_un... |
1911 |
if (TASK_SIZE - len >= addr && addr >= mmap_min_addr && |
1da177e4c Linux-2.6.12-rc2 |
1912 1913 1914 |
(!vma || addr + len <= vma->vm_start)) return addr; } |
1da177e4c Linux-2.6.12-rc2 |
1915 |
|
db4fbfb95 mm: vm_unmapped_a... |
1916 1917 |
info.flags = 0; info.length = len; |
4e99b0213 mmap: arch_get_un... |
1918 |
info.low_limit = mm->mmap_base; |
db4fbfb95 mm: vm_unmapped_a... |
1919 1920 1921 |
info.high_limit = TASK_SIZE; info.align_mask = 0; return vm_unmapped_area(&info); |
1da177e4c Linux-2.6.12-rc2 |
1922 |
} |
cc71aba34 mm/mmap.c: whites... |
1923 |
#endif |
1da177e4c Linux-2.6.12-rc2 |
1924 |
|
1da177e4c Linux-2.6.12-rc2 |
1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 |
/* * 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... |
1937 1938 |
unsigned long addr = addr0; struct vm_unmapped_area_info info; |
1da177e4c Linux-2.6.12-rc2 |
1939 1940 |
/* requested length too big for entire address space */ |
2afc745f3 mm: ensure get_un... |
1941 |
if (len > TASK_SIZE - mmap_min_addr) |
1da177e4c Linux-2.6.12-rc2 |
1942 |
return -ENOMEM; |
06abdfb47 get_unmapped_area... |
1943 1944 |
if (flags & MAP_FIXED) return addr; |
1da177e4c Linux-2.6.12-rc2 |
1945 1946 1947 1948 |
/* requesting a specific address */ if (addr) { addr = PAGE_ALIGN(addr); vma = find_vma(mm, addr); |
2afc745f3 mm: ensure get_un... |
1949 |
if (TASK_SIZE - len >= addr && addr >= mmap_min_addr && |
1da177e4c Linux-2.6.12-rc2 |
1950 1951 1952 |
(!vma || addr + len <= vma->vm_start)) return addr; } |
db4fbfb95 mm: vm_unmapped_a... |
1953 1954 |
info.flags = VM_UNMAPPED_AREA_TOPDOWN; info.length = len; |
2afc745f3 mm: ensure get_un... |
1955 |
info.low_limit = max(PAGE_SIZE, mmap_min_addr); |
db4fbfb95 mm: vm_unmapped_a... |
1956 1957 1958 |
info.high_limit = mm->mmap_base; info.align_mask = 0; addr = vm_unmapped_area(&info); |
b716ad953 mm: search from f... |
1959 |
|
1da177e4c Linux-2.6.12-rc2 |
1960 1961 1962 1963 1964 1965 |
/* * A failed mmap() very likely causes application failure, * so fall back to the bottom-up function here. This scenario * can happen with large stack limits and large mmap() * allocations. */ |
de1741a13 mm/mmap: use offs... |
1966 |
if (offset_in_page(addr)) { |
db4fbfb95 mm: vm_unmapped_a... |
1967 1968 1969 1970 1971 1972 |
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 |
1973 1974 1975 1976 |
return addr; } #endif |
1da177e4c Linux-2.6.12-rc2 |
1977 1978 1979 1980 |
unsigned long get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { |
06abdfb47 get_unmapped_area... |
1981 1982 |
unsigned long (*get_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long); |
9206de95b Take arch_mmap_ch... |
1983 1984 1985 1986 1987 1988 1989 |
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... |
1990 |
get_area = current->mm->get_unmapped_area; |
c01d5b300 shmem: get_unmapp... |
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 |
if (file) { if (file->f_op->get_unmapped_area) get_area = file->f_op->get_unmapped_area; } else if (flags & MAP_SHARED) { /* * mmap_region() will call shmem_zero_setup() to create a file, * so use shmem's get_unmapped_area in case it can be huge. * do_mmap_pgoff() will clear pgoff, so match alignment. */ pgoff = 0; get_area = shmem_get_unmapped_area; } |
06abdfb47 get_unmapped_area... |
2003 2004 2005 |
addr = get_area(file, addr, len, pgoff, flags); if (IS_ERR_VALUE(addr)) return addr; |
1da177e4c Linux-2.6.12-rc2 |
2006 |
|
07ab67c8d Fix get_unmapped_... |
2007 2008 |
if (addr > TASK_SIZE - len) return -ENOMEM; |
de1741a13 mm/mmap: use offs... |
2009 |
if (offset_in_page(addr)) |
07ab67c8d Fix get_unmapped_... |
2010 |
return -EINVAL; |
06abdfb47 get_unmapped_area... |
2011 |
|
9ac4ed4bd move security_mma... |
2012 2013 |
error = security_mmap_addr(addr); return error ? error : addr; |
1da177e4c Linux-2.6.12-rc2 |
2014 2015 2016 2017 2018 |
} EXPORT_SYMBOL(get_unmapped_area); /* Look up the first VMA which satisfies addr < vm_end, NULL if none. */ |
48aae4255 mm/mmap.c: fix co... |
2019 |
struct vm_area_struct *find_vma(struct mm_struct *mm, unsigned long addr) |
1da177e4c Linux-2.6.12-rc2 |
2020 |
{ |
615d6e875 mm: per-thread vm... |
2021 2022 |
struct rb_node *rb_node; struct vm_area_struct *vma; |
1da177e4c Linux-2.6.12-rc2 |
2023 |
|
841e31e5c mm/mmap.c: find_v... |
2024 |
/* Check the cache first. */ |
615d6e875 mm: per-thread vm... |
2025 2026 2027 |
vma = vmacache_find(mm, addr); if (likely(vma)) return vma; |
841e31e5c mm/mmap.c: find_v... |
2028 |
|
615d6e875 mm: per-thread vm... |
2029 |
rb_node = mm->mm_rb.rb_node; |
841e31e5c mm/mmap.c: find_v... |
2030 |
|
615d6e875 mm: per-thread vm... |
2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 |
while (rb_node) { struct vm_area_struct *tmp; tmp = rb_entry(rb_node, struct vm_area_struct, vm_rb); if (tmp->vm_end > addr) { vma = tmp; if (tmp->vm_start <= addr) break; rb_node = rb_node->rb_left; } else rb_node = rb_node->rb_right; |
1da177e4c Linux-2.6.12-rc2 |
2043 |
} |
615d6e875 mm: per-thread vm... |
2044 2045 2046 |
if (vma) vmacache_update(addr, vma); |
1da177e4c Linux-2.6.12-rc2 |
2047 2048 2049 2050 |
return vma; } EXPORT_SYMBOL(find_vma); |
6bd4837de mm: simplify find... |
2051 2052 |
/* * Same as find_vma, but also return a pointer to the previous VMA in *pprev. |
6bd4837de mm: simplify find... |
2053 |
*/ |
1da177e4c Linux-2.6.12-rc2 |
2054 2055 2056 2057 |
struct vm_area_struct * find_vma_prev(struct mm_struct *mm, unsigned long addr, struct vm_area_struct **pprev) { |
6bd4837de mm: simplify find... |
2058 |
struct vm_area_struct *vma; |
1da177e4c Linux-2.6.12-rc2 |
2059 |
|
6bd4837de mm: simplify find... |
2060 |
vma = find_vma(mm, addr); |
83cd904d2 mm: fix find_vma_... |
2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 |
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... |
2071 |
return vma; |
1da177e4c Linux-2.6.12-rc2 |
2072 2073 2074 2075 2076 2077 2078 |
} /* * 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... |
2079 |
static int acct_stack_growth(struct vm_area_struct *vma, unsigned long size, unsigned long grow) |
1da177e4c Linux-2.6.12-rc2 |
2080 2081 2082 |
{ struct mm_struct *mm = vma->vm_mm; struct rlimit *rlim = current->signal->rlim; |
690eac53d mm: Don't count t... |
2083 |
unsigned long new_start, actual_size; |
1da177e4c Linux-2.6.12-rc2 |
2084 2085 |
/* address space limit tests */ |
846383359 mm: rework virtua... |
2086 |
if (!may_expand_vm(mm, vma->vm_flags, grow)) |
1da177e4c Linux-2.6.12-rc2 |
2087 2088 2089 |
return -ENOMEM; /* Stack limit test */ |
690eac53d mm: Don't count t... |
2090 2091 2092 |
actual_size = size; if (size && (vma->vm_flags & (VM_GROWSUP | VM_GROWSDOWN))) actual_size -= PAGE_SIZE; |
4db0c3c29 mm: remove rest o... |
2093 |
if (actual_size > READ_ONCE(rlim[RLIMIT_STACK].rlim_cur)) |
1da177e4c Linux-2.6.12-rc2 |
2094 2095 2096 2097 2098 2099 2100 |
return -ENOMEM; /* mlock limit tests */ if (vma->vm_flags & VM_LOCKED) { unsigned long locked; unsigned long limit; locked = mm->locked_vm + grow; |
4db0c3c29 mm: remove rest o... |
2101 |
limit = READ_ONCE(rlim[RLIMIT_MEMLOCK].rlim_cur); |
59e99e5b9 mm: use rlimit he... |
2102 |
limit >>= PAGE_SHIFT; |
1da177e4c Linux-2.6.12-rc2 |
2103 2104 2105 |
if (locked > limit && !capable(CAP_IPC_LOCK)) return -ENOMEM; } |
0d59a01bc [PATCH] Don't all... |
2106 2107 2108 2109 2110 |
/* 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 |
2111 2112 2113 2114 |
/* * Overcommit.. This must be the final test, as it will * update security statistics. */ |
05fa199d4 mm: pass correct ... |
2115 |
if (security_vm_enough_memory_mm(mm, grow)) |
1da177e4c Linux-2.6.12-rc2 |
2116 |
return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
2117 2118 |
return 0; } |
46dea3d09 [PATCH] mm: ia64 ... |
2119 |
#if defined(CONFIG_STACK_GROWSUP) || defined(CONFIG_IA64) |
1da177e4c Linux-2.6.12-rc2 |
2120 |
/* |
46dea3d09 [PATCH] mm: ia64 ... |
2121 2122 |
* 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 |
2123 |
*/ |
46dea3d09 [PATCH] mm: ia64 ... |
2124 |
int expand_upwards(struct vm_area_struct *vma, unsigned long address) |
1da177e4c Linux-2.6.12-rc2 |
2125 |
{ |
093578147 mm: add the "stru... |
2126 |
struct mm_struct *mm = vma->vm_mm; |
12352d3ca mm: replace vma_l... |
2127 |
int error = 0; |
1da177e4c Linux-2.6.12-rc2 |
2128 2129 2130 |
if (!(vma->vm_flags & VM_GROWSUP)) return -EFAULT; |
12352d3ca mm: replace vma_l... |
2131 2132 2133 2134 2135 2136 2137 |
/* Guard against wrapping around to address 0. */ if (address < PAGE_ALIGN(address+4)) address = PAGE_ALIGN(address+4); else return -ENOMEM; /* We must make sure the anon_vma is allocated. */ |
1da177e4c Linux-2.6.12-rc2 |
2138 2139 |
if (unlikely(anon_vma_prepare(vma))) return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
2140 2141 2142 2143 2144 2145 |
/* * vma->vm_start/vm_end cannot change under us because the caller * is required to hold the mmap_sem in read mode. We need the * anon_vma lock to serialize against concurrent expand_stacks. */ |
12352d3ca mm: replace vma_l... |
2146 |
anon_vma_lock_write(vma->anon_vma); |
1da177e4c Linux-2.6.12-rc2 |
2147 2148 2149 2150 2151 2152 2153 |
/* 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 ... |
2154 2155 2156 2157 |
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... |
2158 2159 2160 2161 2162 |
/* * vma_gap_update() doesn't support concurrent * updates, but we only hold a shared mmap_sem * lock here, so we need to protect against * concurrent vma expansions. |
12352d3ca mm: replace vma_l... |
2163 |
* anon_vma_lock_write() doesn't help here, as |
4128997b5 mm: protect again... |
2164 2165 2166 2167 2168 |
* we don't guarantee that all growable vmas * in a mm share the same root anon vma. * So, we reuse mm->page_table_lock to guard * against concurrent vma expansions. */ |
093578147 mm: add the "stru... |
2169 |
spin_lock(&mm->page_table_lock); |
87e8827b3 mm: fix the racy ... |
2170 |
if (vma->vm_flags & VM_LOCKED) |
093578147 mm: add the "stru... |
2171 |
mm->locked_vm += grow; |
846383359 mm: rework virtua... |
2172 |
vm_stat_account(mm, vma->vm_flags, grow); |
bf181b9f9 mm anon rmap: rep... |
2173 |
anon_vma_interval_tree_pre_update_vma(vma); |
42c36f63a vm: fix vm_pgoff ... |
2174 |
vma->vm_end = address; |
bf181b9f9 mm anon rmap: rep... |
2175 |
anon_vma_interval_tree_post_update_vma(vma); |
d37371870 mm: augment vma r... |
2176 2177 2178 |
if (vma->vm_next) vma_gap_update(vma->vm_next); else |
093578147 mm: add the "stru... |
2179 2180 |
mm->highest_vm_end = address; spin_unlock(&mm->page_table_lock); |
4128997b5 mm: protect again... |
2181 |
|
42c36f63a vm: fix vm_pgoff ... |
2182 2183 |
perf_event_mmap(vma); } |
3af9e8592 perf: Add non-exe... |
2184 |
} |
1da177e4c Linux-2.6.12-rc2 |
2185 |
} |
12352d3ca mm: replace vma_l... |
2186 |
anon_vma_unlock_write(vma->anon_vma); |
6d50e60cd mm, thp: fix coll... |
2187 |
khugepaged_enter_vma_merge(vma, vma->vm_flags); |
093578147 mm: add the "stru... |
2188 |
validate_mm(mm); |
1da177e4c Linux-2.6.12-rc2 |
2189 2190 |
return error; } |
46dea3d09 [PATCH] mm: ia64 ... |
2191 |
#endif /* CONFIG_STACK_GROWSUP || CONFIG_IA64 */ |
1da177e4c Linux-2.6.12-rc2 |
2192 2193 2194 |
/* * vma is the first one with address < vma->vm_start. Have to extend vma. */ |
d05f3169c mm: make expand_d... |
2195 |
int expand_downwards(struct vm_area_struct *vma, |
b6a2fea39 mm: variable leng... |
2196 |
unsigned long address) |
1da177e4c Linux-2.6.12-rc2 |
2197 |
{ |
093578147 mm: add the "stru... |
2198 |
struct mm_struct *mm = vma->vm_mm; |
1da177e4c Linux-2.6.12-rc2 |
2199 |
int error; |
8869477a4 security: protect... |
2200 |
address &= PAGE_MASK; |
e5467859f split ->file_mmap... |
2201 |
error = security_mmap_addr(address); |
8869477a4 security: protect... |
2202 2203 |
if (error) return error; |
12352d3ca mm: replace vma_l... |
2204 2205 2206 |
/* We must make sure the anon_vma is allocated. */ if (unlikely(anon_vma_prepare(vma))) return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
2207 2208 2209 2210 2211 2212 |
/* * vma->vm_start/vm_end cannot change under us because the caller * is required to hold the mmap_sem in read mode. We need the * anon_vma lock to serialize against concurrent expand_stacks. */ |
12352d3ca mm: replace vma_l... |
2213 |
anon_vma_lock_write(vma->anon_vma); |
1da177e4c Linux-2.6.12-rc2 |
2214 2215 2216 2217 2218 2219 2220 |
/* 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 ... |
2221 2222 2223 2224 |
error = -ENOMEM; if (grow <= vma->vm_pgoff) { error = acct_stack_growth(vma, size, grow); if (!error) { |
4128997b5 mm: protect again... |
2225 2226 2227 2228 2229 |
/* * vma_gap_update() doesn't support concurrent * updates, but we only hold a shared mmap_sem * lock here, so we need to protect against * concurrent vma expansions. |
12352d3ca mm: replace vma_l... |
2230 |
* anon_vma_lock_write() doesn't help here, as |
4128997b5 mm: protect again... |
2231 2232 2233 2234 2235 |
* we don't guarantee that all growable vmas * in a mm share the same root anon vma. * So, we reuse mm->page_table_lock to guard * against concurrent vma expansions. */ |
093578147 mm: add the "stru... |
2236 |
spin_lock(&mm->page_table_lock); |
87e8827b3 mm: fix the racy ... |
2237 |
if (vma->vm_flags & VM_LOCKED) |
093578147 mm: add the "stru... |
2238 |
mm->locked_vm += grow; |
846383359 mm: rework virtua... |
2239 |
vm_stat_account(mm, vma->vm_flags, grow); |
bf181b9f9 mm anon rmap: rep... |
2240 |
anon_vma_interval_tree_pre_update_vma(vma); |
a626ca6a6 vm: fix vm_pgoff ... |
2241 2242 |
vma->vm_start = address; vma->vm_pgoff -= grow; |
bf181b9f9 mm anon rmap: rep... |
2243 |
anon_vma_interval_tree_post_update_vma(vma); |
d37371870 mm: augment vma r... |
2244 |
vma_gap_update(vma); |
093578147 mm: add the "stru... |
2245 |
spin_unlock(&mm->page_table_lock); |
4128997b5 mm: protect again... |
2246 |
|
a626ca6a6 vm: fix vm_pgoff ... |
2247 2248 |
perf_event_mmap(vma); } |
1da177e4c Linux-2.6.12-rc2 |
2249 2250 |
} } |
12352d3ca mm: replace vma_l... |
2251 |
anon_vma_unlock_write(vma->anon_vma); |
6d50e60cd mm, thp: fix coll... |
2252 |
khugepaged_enter_vma_merge(vma, vma->vm_flags); |
093578147 mm: add the "stru... |
2253 |
validate_mm(mm); |
1da177e4c Linux-2.6.12-rc2 |
2254 2255 |
return error; } |
098849643 mm: do not grow t... |
2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 |
/* * 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... |
2267 2268 2269 |
#ifdef CONFIG_STACK_GROWSUP int expand_stack(struct vm_area_struct *vma, unsigned long address) { |
098849643 mm: do not grow t... |
2270 2271 2272 2273 2274 2275 2276 2277 |
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... |
2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 |
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_... |
2290 |
if (!prev || expand_stack(prev, addr)) |
b6a2fea39 mm: variable leng... |
2291 |
return NULL; |
cea10a19b mm: directly use ... |
2292 |
if (prev->vm_flags & VM_LOCKED) |
fc05f5662 mm: rename __mloc... |
2293 |
populate_vma_page_range(prev, addr, prev->vm_end, NULL); |
b6a2fea39 mm: variable leng... |
2294 2295 2296 2297 2298 |
return prev; } #else int expand_stack(struct vm_area_struct *vma, unsigned long address) { |
098849643 mm: do not grow t... |
2299 2300 2301 2302 2303 2304 2305 2306 |
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... |
2307 2308 |
return expand_downwards(vma, address); } |
1da177e4c Linux-2.6.12-rc2 |
2309 |
struct vm_area_struct * |
cc71aba34 mm/mmap.c: whites... |
2310 |
find_extend_vma(struct mm_struct *mm, unsigned long addr) |
1da177e4c Linux-2.6.12-rc2 |
2311 |
{ |
cc71aba34 mm/mmap.c: whites... |
2312 |
struct vm_area_struct *vma; |
1da177e4c Linux-2.6.12-rc2 |
2313 2314 2315 |
unsigned long start; addr &= PAGE_MASK; |
cc71aba34 mm/mmap.c: whites... |
2316 |
vma = find_vma(mm, addr); |
1da177e4c Linux-2.6.12-rc2 |
2317 2318 2319 2320 2321 2322 2323 2324 2325 |
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 ... |
2326 |
if (vma->vm_flags & VM_LOCKED) |
fc05f5662 mm: rename __mloc... |
2327 |
populate_vma_page_range(vma, addr, start, NULL); |
1da177e4c Linux-2.6.12-rc2 |
2328 2329 2330 |
return vma; } #endif |
e1d6d01ab mm: export find_e... |
2331 |
EXPORT_SYMBOL_GPL(find_extend_vma); |
1da177e4c Linux-2.6.12-rc2 |
2332 |
/* |
2c0b38146 [PATCH] mm: remov... |
2333 |
* Ok - we have the memory areas we should free on the vma list, |
1da177e4c Linux-2.6.12-rc2 |
2334 |
* so release them, and do the vma updates. |
2c0b38146 [PATCH] mm: remov... |
2335 2336 |
* * Called with the mm semaphore held. |
1da177e4c Linux-2.6.12-rc2 |
2337 |
*/ |
2c0b38146 [PATCH] mm: remov... |
2338 |
static void remove_vma_list(struct mm_struct *mm, struct vm_area_struct *vma) |
1da177e4c Linux-2.6.12-rc2 |
2339 |
{ |
4f74d2c8e vm: remove 'nr_ac... |
2340 |
unsigned long nr_accounted = 0; |
365e9c87a [PATCH] mm: updat... |
2341 2342 |
/* Update high watermark before we lower total_vm */ update_hiwater_vm(mm); |
1da177e4c Linux-2.6.12-rc2 |
2343 |
do { |
2c0b38146 [PATCH] mm: remov... |
2344 |
long nrpages = vma_pages(vma); |
4f74d2c8e vm: remove 'nr_ac... |
2345 2346 |
if (vma->vm_flags & VM_ACCOUNT) nr_accounted += nrpages; |
846383359 mm: rework virtua... |
2347 |
vm_stat_account(mm, vma->vm_flags, -nrpages); |
a8fb5618d [PATCH] mm: unlin... |
2348 |
vma = remove_vma(vma); |
146425a31 [PATCH] freepgt: ... |
2349 |
} while (vma); |
4f74d2c8e vm: remove 'nr_ac... |
2350 |
vm_unacct_memory(nr_accounted); |
1da177e4c Linux-2.6.12-rc2 |
2351 2352 2353 2354 2355 2356 |
validate_mm(mm); } /* * Get rid of page table information in the indicated region. * |
f10df6860 [PATCH] fix locki... |
2357 |
* Called with the mm semaphore held. |
1da177e4c Linux-2.6.12-rc2 |
2358 2359 |
*/ static void unmap_region(struct mm_struct *mm, |
e0da382c9 [PATCH] freepgt: ... |
2360 2361 |
struct vm_area_struct *vma, struct vm_area_struct *prev, unsigned long start, unsigned long end) |
1da177e4c Linux-2.6.12-rc2 |
2362 |
{ |
cc71aba34 mm/mmap.c: whites... |
2363 |
struct vm_area_struct *next = prev ? prev->vm_next : mm->mmap; |
d16dfc550 mm: mmu_gather re... |
2364 |
struct mmu_gather tlb; |
1da177e4c Linux-2.6.12-rc2 |
2365 2366 |
lru_add_drain(); |
2b047252d Fix TLB gather vi... |
2367 |
tlb_gather_mmu(&tlb, mm, start, end); |
365e9c87a [PATCH] mm: updat... |
2368 |
update_hiwater_rss(mm); |
4f74d2c8e vm: remove 'nr_ac... |
2369 |
unmap_vmas(&tlb, vma, start, end); |
d16dfc550 mm: mmu_gather re... |
2370 |
free_pgtables(&tlb, vma, prev ? prev->vm_end : FIRST_USER_ADDRESS, |
6ee8630e0 mm: allow arch co... |
2371 |
next ? next->vm_start : USER_PGTABLES_CEILING); |
d16dfc550 mm: mmu_gather re... |
2372 |
tlb_finish_mmu(&tlb, start, end); |
1da177e4c Linux-2.6.12-rc2 |
2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 |
} /* * Create a list of vma's touched by the unmap, removing them from the mm's * vma list as we go.. */ static void detach_vmas_to_be_unmapped(struct mm_struct *mm, struct vm_area_struct *vma, struct vm_area_struct *prev, unsigned long end) { struct vm_area_struct **insertion_point; struct vm_area_struct *tail_vma = NULL; insertion_point = (prev ? &prev->vm_next : &mm->mmap); |
297c5eee3 mm: make the vma ... |
2387 |
vma->vm_prev = NULL; |
1da177e4c Linux-2.6.12-rc2 |
2388 |
do { |
d37371870 mm: augment vma r... |
2389 |
vma_rb_erase(vma, &mm->mm_rb); |
1da177e4c Linux-2.6.12-rc2 |
2390 2391 2392 2393 2394 |
mm->map_count--; tail_vma = vma; vma = vma->vm_next; } while (vma && vma->vm_start < end); *insertion_point = vma; |
d37371870 mm: augment vma r... |
2395 |
if (vma) { |
297c5eee3 mm: make the vma ... |
2396 |
vma->vm_prev = prev; |
d37371870 mm: augment vma r... |
2397 2398 2399 |
vma_gap_update(vma); } else mm->highest_vm_end = prev ? prev->vm_end : 0; |
1da177e4c Linux-2.6.12-rc2 |
2400 |
tail_vma->vm_next = NULL; |
615d6e875 mm: per-thread vm... |
2401 2402 2403 |
/* Kill the cache */ vmacache_invalidate(mm); |
1da177e4c Linux-2.6.12-rc2 |
2404 2405 2406 |
} /* |
659ace584 mmap: don't retur... |
2407 2408 |
* __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 |
2409 |
*/ |
cc71aba34 mm/mmap.c: whites... |
2410 |
static int __split_vma(struct mm_struct *mm, struct vm_area_struct *vma, |
1da177e4c Linux-2.6.12-rc2 |
2411 2412 |
unsigned long addr, int new_below) { |
1da177e4c Linux-2.6.12-rc2 |
2413 |
struct vm_area_struct *new; |
e39758912 mm/mmap.c: simpli... |
2414 |
int err; |
1da177e4c Linux-2.6.12-rc2 |
2415 |
|
a55164389 hugetlb: modular ... |
2416 2417 |
if (is_vm_hugetlb_page(vma) && (addr & ~(huge_page_mask(hstate_vma(vma))))) |
1da177e4c Linux-2.6.12-rc2 |
2418 |
return -EINVAL; |
e94b17660 [PATCH] slab: rem... |
2419 |
new = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); |
1da177e4c Linux-2.6.12-rc2 |
2420 |
if (!new) |
e39758912 mm/mmap.c: simpli... |
2421 |
return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
2422 2423 2424 |
/* most fields are the same, copy all, and then fixup */ *new = *vma; |
5beb49305 mm: change anon_v... |
2425 |
INIT_LIST_HEAD(&new->anon_vma_chain); |
1da177e4c Linux-2.6.12-rc2 |
2426 2427 2428 2429 2430 2431 |
if (new_below) new->vm_end = addr; else { new->vm_start = addr; new->vm_pgoff += ((addr - vma->vm_start) >> PAGE_SHIFT); } |
ef0855d33 mm: mempolicy: tu... |
2432 2433 |
err = vma_dup_policy(vma, new); if (err) |
5beb49305 mm: change anon_v... |
2434 |
goto out_free_vma; |
1da177e4c Linux-2.6.12-rc2 |
2435 |
|
c4ea95d7c mm: fix anon_vma_... |
2436 2437 |
err = anon_vma_clone(new, vma); if (err) |
5beb49305 mm: change anon_v... |
2438 |
goto out_free_mpol; |
e9714acf8 mm: kill vma flag... |
2439 |
if (new->vm_file) |
1da177e4c Linux-2.6.12-rc2 |
2440 2441 2442 2443 2444 2445 |
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... |
2446 |
err = vma_adjust(vma, addr, vma->vm_end, vma->vm_pgoff + |
1da177e4c Linux-2.6.12-rc2 |
2447 2448 |
((addr - new->vm_start) >> PAGE_SHIFT), new); else |
5beb49305 mm: change anon_v... |
2449 |
err = vma_adjust(vma, vma->vm_start, addr, vma->vm_pgoff, new); |
1da177e4c Linux-2.6.12-rc2 |
2450 |
|
5beb49305 mm: change anon_v... |
2451 2452 2453 2454 2455 |
/* Success. */ if (!err) return 0; /* Clean everything up if vma_adjust failed. */ |
589275338 mmap: check ->vm_... |
2456 2457 |
if (new->vm_ops && new->vm_ops->close) new->vm_ops->close(new); |
e9714acf8 mm: kill vma flag... |
2458 |
if (new->vm_file) |
5beb49305 mm: change anon_v... |
2459 |
fput(new->vm_file); |
2aeadc30d mmap: call unlink... |
2460 |
unlink_anon_vmas(new); |
5beb49305 mm: change anon_v... |
2461 |
out_free_mpol: |
ef0855d33 mm: mempolicy: tu... |
2462 |
mpol_put(vma_policy(new)); |
5beb49305 mm: change anon_v... |
2463 2464 |
out_free_vma: kmem_cache_free(vm_area_cachep, new); |
5beb49305 mm: change anon_v... |
2465 |
return err; |
1da177e4c Linux-2.6.12-rc2 |
2466 |
} |
659ace584 mmap: don't retur... |
2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 |
/* * 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 |
2479 2480 2481 2482 2483 2484 2485 2486 |
/* 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: ... |
2487 |
struct vm_area_struct *vma, *prev, *last; |
1da177e4c Linux-2.6.12-rc2 |
2488 |
|
de1741a13 mm/mmap: use offs... |
2489 |
if ((offset_in_page(start)) || start > TASK_SIZE || len > TASK_SIZE-start) |
1da177e4c Linux-2.6.12-rc2 |
2490 |
return -EINVAL; |
cc71aba34 mm/mmap.c: whites... |
2491 2492 |
len = PAGE_ALIGN(len); if (len == 0) |
1da177e4c Linux-2.6.12-rc2 |
2493 2494 2495 |
return -EINVAL; /* Find the first overlapping VMA */ |
9be34c9d5 mm: get rid of th... |
2496 |
vma = find_vma(mm, start); |
146425a31 [PATCH] freepgt: ... |
2497 |
if (!vma) |
1da177e4c Linux-2.6.12-rc2 |
2498 |
return 0; |
9be34c9d5 mm: get rid of th... |
2499 |
prev = vma->vm_prev; |
146425a31 [PATCH] freepgt: ... |
2500 |
/* we have start < vma->vm_end */ |
1da177e4c Linux-2.6.12-rc2 |
2501 2502 2503 |
/* if it doesn't overlap, we have nothing.. */ end = start + len; |
146425a31 [PATCH] freepgt: ... |
2504 |
if (vma->vm_start >= end) |
1da177e4c Linux-2.6.12-rc2 |
2505 2506 2507 2508 2509 2510 2511 2512 2513 |
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: ... |
2514 |
if (start > vma->vm_start) { |
659ace584 mmap: don't retur... |
2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 |
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 |
2526 2527 |
if (error) return error; |
146425a31 [PATCH] freepgt: ... |
2528 |
prev = vma; |
1da177e4c Linux-2.6.12-rc2 |
2529 2530 2531 2532 2533 |
} /* Does it split the last one? */ last = find_vma(mm, end); if (last && end > last->vm_start) { |
659ace584 mmap: don't retur... |
2534 |
int error = __split_vma(mm, last, end, 1); |
1da177e4c Linux-2.6.12-rc2 |
2535 2536 2537 |
if (error) return error; } |
cc71aba34 mm/mmap.c: whites... |
2538 |
vma = prev ? prev->vm_next : mm->mmap; |
1da177e4c Linux-2.6.12-rc2 |
2539 2540 |
/* |
ba470de43 mmap: handle mloc... |
2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 |
* 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 |
2555 2556 |
* Remove the vma's, and unmap the actual pages */ |
146425a31 [PATCH] freepgt: ... |
2557 2558 |
detach_vmas_to_be_unmapped(mm, vma, prev, end); unmap_region(mm, vma, prev, start, end); |
1da177e4c Linux-2.6.12-rc2 |
2559 |
|
1de4fa14e x86, mpx: Cleanup... |
2560 |
arch_unmap(mm, vma, start, end); |
1da177e4c Linux-2.6.12-rc2 |
2561 |
/* Fix up all other VM information */ |
2c0b38146 [PATCH] mm: remov... |
2562 |
remove_vma_list(mm, vma); |
1da177e4c Linux-2.6.12-rc2 |
2563 2564 2565 |
return 0; } |
1da177e4c Linux-2.6.12-rc2 |
2566 |
|
bfce281c2 kill mm argument ... |
2567 |
int vm_munmap(unsigned long start, size_t len) |
1da177e4c Linux-2.6.12-rc2 |
2568 2569 |
{ int ret; |
bfce281c2 kill mm argument ... |
2570 |
struct mm_struct *mm = current->mm; |
1da177e4c Linux-2.6.12-rc2 |
2571 |
|
ae7987835 mm: make vm_munma... |
2572 2573 |
if (down_write_killable(&mm->mmap_sem)) return -EINTR; |
a46ef99d8 VM: add "vm_munma... |
2574 |
ret = do_munmap(mm, start, len); |
1da177e4c Linux-2.6.12-rc2 |
2575 2576 2577 |
up_write(&mm->mmap_sem); return ret; } |
a46ef99d8 VM: add "vm_munma... |
2578 2579 2580 2581 |
EXPORT_SYMBOL(vm_munmap); SYSCALL_DEFINE2(munmap, unsigned long, addr, size_t, len) { |
dc0ef0df7 mm: make mmap_sem... |
2582 2583 |
int ret; struct mm_struct *mm = current->mm; |
a46ef99d8 VM: add "vm_munma... |
2584 |
profile_munmap(addr); |
dc0ef0df7 mm: make mmap_sem... |
2585 2586 2587 2588 2589 |
if (down_write_killable(&mm->mmap_sem)) return -EINTR; ret = do_munmap(mm, addr, len); up_write(&mm->mmap_sem); return ret; |
a46ef99d8 VM: add "vm_munma... |
2590 |
} |
1da177e4c Linux-2.6.12-rc2 |
2591 |
|
c8d78c182 mm: replace remap... |
2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 |
/* * Emulation of deprecated remap_file_pages() syscall. */ SYSCALL_DEFINE5(remap_file_pages, unsigned long, start, unsigned long, size, unsigned long, prot, unsigned long, pgoff, unsigned long, flags) { struct mm_struct *mm = current->mm; struct vm_area_struct *vma; unsigned long populate = 0; unsigned long ret = -EINVAL; struct file *file; |
756a025f0 mm: coalesce spli... |
2605 2606 2607 |
pr_warn_once("%s (%d) uses deprecated remap_file_pages() syscall. See Documentation/vm/remap_file_pages.txt. ", current->comm, current->pid); |
c8d78c182 mm: replace remap... |
2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 |
if (prot) return ret; start = start & PAGE_MASK; size = size & PAGE_MASK; if (start + size <= start) return ret; /* Does pgoff wrap? */ if (pgoff + (size >> PAGE_SHIFT) < pgoff) return ret; |
dc0ef0df7 mm: make mmap_sem... |
2620 2621 |
if (down_write_killable(&mm->mmap_sem)) return -EINTR; |
c8d78c182 mm: replace remap... |
2622 2623 2624 2625 |
vma = find_vma(mm, start); if (!vma || !(vma->vm_flags & VM_SHARED)) goto out; |
48f7df329 mm: fix regressio... |
2626 |
if (start < vma->vm_start) |
c8d78c182 mm: replace remap... |
2627 |
goto out; |
48f7df329 mm: fix regressio... |
2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 |
if (start + size > vma->vm_end) { struct vm_area_struct *next; for (next = vma->vm_next; next; next = next->vm_next) { /* hole between vmas ? */ if (next->vm_start != next->vm_prev->vm_end) goto out; if (next->vm_file != vma->vm_file) goto out; if (next->vm_flags != vma->vm_flags) goto out; if (start + size <= next->vm_end) break; } if (!next) goto out; |
c8d78c182 mm: replace remap... |
2648 2649 2650 2651 2652 2653 2654 2655 2656 |
} prot |= vma->vm_flags & VM_READ ? PROT_READ : 0; prot |= vma->vm_flags & VM_WRITE ? PROT_WRITE : 0; prot |= vma->vm_flags & VM_EXEC ? PROT_EXEC : 0; flags &= MAP_NONBLOCK; flags |= MAP_SHARED | MAP_FIXED | MAP_POPULATE; if (vma->vm_flags & VM_LOCKED) { |
48f7df329 mm: fix regressio... |
2657 |
struct vm_area_struct *tmp; |
c8d78c182 mm: replace remap... |
2658 |
flags |= MAP_LOCKED; |
48f7df329 mm: fix regressio... |
2659 |
|
c8d78c182 mm: replace remap... |
2660 |
/* drop PG_Mlocked flag for over-mapped range */ |
48f7df329 mm: fix regressio... |
2661 2662 |
for (tmp = vma; tmp->vm_start >= start + size; tmp = tmp->vm_next) { |
9a73f61bd thp, mlock: do no... |
2663 2664 2665 2666 2667 |
/* * Split pmd and munlock page on the border * of the range. */ vma_adjust_trans_huge(tmp, start, start + size, 0); |
48f7df329 mm: fix regressio... |
2668 2669 2670 2671 |
munlock_vma_pages_range(tmp, max(tmp->vm_start, start), min(tmp->vm_end, start + size)); } |
c8d78c182 mm: replace remap... |
2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 |
} file = get_file(vma->vm_file); ret = do_mmap_pgoff(vma->vm_file, start, size, prot, flags, pgoff, &populate); fput(file); out: up_write(&mm->mmap_sem); if (populate) mm_populate(ret, populate); if (!IS_ERR_VALUE(ret)) ret = 0; return ret; } |
1da177e4c Linux-2.6.12-rc2 |
2686 2687 |
static inline void verify_mm_writelocked(struct mm_struct *mm) { |
a241ec65a [PATCH] RCU tortu... |
2688 |
#ifdef CONFIG_DEBUG_VM |
1da177e4c Linux-2.6.12-rc2 |
2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 |
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. */ |
ba093a6d9 mm: refuse wrappe... |
2701 |
static int do_brk(unsigned long addr, unsigned long request) |
1da177e4c Linux-2.6.12-rc2 |
2702 |
{ |
cc71aba34 mm/mmap.c: whites... |
2703 2704 |
struct mm_struct *mm = current->mm; struct vm_area_struct *vma, *prev; |
ba093a6d9 mm: refuse wrappe... |
2705 |
unsigned long flags, len; |
cc71aba34 mm/mmap.c: whites... |
2706 |
struct rb_node **rb_link, *rb_parent; |
1da177e4c Linux-2.6.12-rc2 |
2707 |
pgoff_t pgoff = addr >> PAGE_SHIFT; |
3a4597568 [PATCH] IA64,spar... |
2708 |
int error; |
1da177e4c Linux-2.6.12-rc2 |
2709 |
|
ba093a6d9 mm: refuse wrappe... |
2710 2711 2712 |
len = PAGE_ALIGN(request); if (len < request) return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
2713 |
if (!len) |
5d22fc25d mm: remove more I... |
2714 |
return 0; |
1da177e4c Linux-2.6.12-rc2 |
2715 |
|
3a4597568 [PATCH] IA64,spar... |
2716 |
flags = VM_DATA_DEFAULT_FLAGS | VM_ACCOUNT | mm->def_flags; |
2c6a10161 switch do_brk() t... |
2717 |
error = get_unmapped_area(NULL, addr, len, 0, MAP_FIXED); |
de1741a13 mm/mmap: use offs... |
2718 |
if (offset_in_page(error)) |
3a4597568 [PATCH] IA64,spar... |
2719 |
return error; |
363ee17f0 mm/mmap.c: add ml... |
2720 2721 2722 |
error = mlock_future_check(mm, mm->def_flags, len); if (error) return error; |
1da177e4c Linux-2.6.12-rc2 |
2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 |
/* * mm->mmap_sem is required to protect against another thread * changing the mappings in case we sleep. */ verify_mm_writelocked(mm); /* * Clear old maps. this also does some error checking for us */ |
9fcd14571 mm/mmap.c: use wh... |
2733 2734 |
while (find_vma_links(mm, addr, addr + len, &prev, &rb_link, &rb_parent)) { |
1da177e4c Linux-2.6.12-rc2 |
2735 2736 |
if (do_munmap(mm, addr, len)) return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
2737 2738 2739 |
} /* Check against address space limits *after* clearing old maps... */ |
846383359 mm: rework virtua... |
2740 |
if (!may_expand_vm(mm, flags, len >> PAGE_SHIFT)) |
1da177e4c Linux-2.6.12-rc2 |
2741 2742 2743 2744 |
return -ENOMEM; if (mm->map_count > sysctl_max_map_count) return -ENOMEM; |
191c54244 mm: collapse secu... |
2745 |
if (security_vm_enough_memory_mm(mm, len >> PAGE_SHIFT)) |
1da177e4c Linux-2.6.12-rc2 |
2746 |
return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
2747 |
/* Can we just expand an old private anonymous mapping? */ |
ba470de43 mmap: handle mloc... |
2748 |
vma = vma_merge(mm, prev, addr, addr + len, flags, |
19a809afe userfaultfd: teac... |
2749 |
NULL, NULL, pgoff, NULL, NULL_VM_UFFD_CTX); |
ba470de43 mmap: handle mloc... |
2750 |
if (vma) |
1da177e4c Linux-2.6.12-rc2 |
2751 2752 2753 2754 2755 |
goto out; /* * create a vma struct for an anonymous mapping */ |
c5e3b83e9 [PATCH] mm: use k... |
2756 |
vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); |
1da177e4c Linux-2.6.12-rc2 |
2757 2758 2759 2760 |
if (!vma) { vm_unacct_memory(len >> PAGE_SHIFT); return -ENOMEM; } |
1da177e4c Linux-2.6.12-rc2 |
2761 |
|
5beb49305 mm: change anon_v... |
2762 |
INIT_LIST_HEAD(&vma->anon_vma_chain); |
1da177e4c Linux-2.6.12-rc2 |
2763 2764 2765 2766 2767 |
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... |
2768 |
vma->vm_page_prot = vm_get_page_prot(flags); |
1da177e4c Linux-2.6.12-rc2 |
2769 2770 |
vma_link(mm, vma, prev, rb_link, rb_parent); out: |
3af9e8592 perf: Add non-exe... |
2771 |
perf_event_mmap(vma); |
1da177e4c Linux-2.6.12-rc2 |
2772 |
mm->total_vm += len >> PAGE_SHIFT; |
846383359 mm: rework virtua... |
2773 |
mm->data_vm += len >> PAGE_SHIFT; |
128557ffe mm: use mm_popula... |
2774 2775 |
if (flags & VM_LOCKED) mm->locked_vm += (len >> PAGE_SHIFT); |
d9104d1ca mm: track vma cha... |
2776 |
vma->vm_flags |= VM_SOFTDIRTY; |
5d22fc25d mm: remove more I... |
2777 |
return 0; |
1da177e4c Linux-2.6.12-rc2 |
2778 |
} |
5d22fc25d mm: remove more I... |
2779 |
int vm_brk(unsigned long addr, unsigned long len) |
e4eb1ff61 VM: add "vm_brk()... |
2780 2781 |
{ struct mm_struct *mm = current->mm; |
5d22fc25d mm: remove more I... |
2782 |
int ret; |
128557ffe mm: use mm_popula... |
2783 |
bool populate; |
e4eb1ff61 VM: add "vm_brk()... |
2784 |
|
2d6c92824 mm: make vm_brk k... |
2785 2786 |
if (down_write_killable(&mm->mmap_sem)) return -EINTR; |
e4eb1ff61 VM: add "vm_brk()... |
2787 |
ret = do_brk(addr, len); |
128557ffe mm: use mm_popula... |
2788 |
populate = ((mm->def_flags & VM_LOCKED) != 0); |
e4eb1ff61 VM: add "vm_brk()... |
2789 |
up_write(&mm->mmap_sem); |
5d22fc25d mm: remove more I... |
2790 |
if (populate && !ret) |
128557ffe mm: use mm_popula... |
2791 |
mm_populate(addr, len); |
e4eb1ff61 VM: add "vm_brk()... |
2792 2793 2794 |
return ret; } EXPORT_SYMBOL(vm_brk); |
1da177e4c Linux-2.6.12-rc2 |
2795 2796 2797 2798 |
/* Release all mmaps. */ void exit_mmap(struct mm_struct *mm) { |
d16dfc550 mm: mmu_gather re... |
2799 |
struct mmu_gather tlb; |
ba470de43 mmap: handle mloc... |
2800 |
struct vm_area_struct *vma; |
1da177e4c Linux-2.6.12-rc2 |
2801 |
unsigned long nr_accounted = 0; |
d6dd61c83 [PATCH] x86: PARA... |
2802 |
/* mm's last user has gone, and its about to be pulled down */ |
cddb8a5c1 mmu-notifiers: core |
2803 |
mmu_notifier_release(mm); |
d6dd61c83 [PATCH] x86: PARA... |
2804 |
|
ba470de43 mmap: handle mloc... |
2805 2806 2807 2808 2809 2810 2811 2812 |
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... |
2813 2814 |
arch_exit_mmap(mm); |
ba470de43 mmap: handle mloc... |
2815 |
vma = mm->mmap; |
9480c53e9 mm: rearrange exi... |
2816 2817 |
if (!vma) /* Can happen if dup_mmap() received an OOM */ return; |
1da177e4c Linux-2.6.12-rc2 |
2818 |
lru_add_drain(); |
1da177e4c Linux-2.6.12-rc2 |
2819 |
flush_cache_mm(mm); |
2b047252d Fix TLB gather vi... |
2820 |
tlb_gather_mmu(&tlb, mm, 0, -1); |
901608d90 mm: introduce get... |
2821 |
/* update_hiwater_rss(mm) here? but nobody should be looking */ |
e0da382c9 [PATCH] freepgt: ... |
2822 |
/* Use -1 here to ensure all VMAs in the mm are unmapped */ |
4f74d2c8e vm: remove 'nr_ac... |
2823 |
unmap_vmas(&tlb, vma, 0, -1); |
9ba692948 ksm: fix oom dead... |
2824 |
|
6ee8630e0 mm: allow arch co... |
2825 |
free_pgtables(&tlb, vma, FIRST_USER_ADDRESS, USER_PGTABLES_CEILING); |
853f5e264 VM: don't bother ... |
2826 |
tlb_finish_mmu(&tlb, 0, -1); |
1da177e4c Linux-2.6.12-rc2 |
2827 |
|
1da177e4c Linux-2.6.12-rc2 |
2828 |
/* |
8f4f8c164 [PATCH] mm: unlin... |
2829 2830 |
* Walk the list again, actually closing and freeing it, * with preemption enabled, without holding any MM locks. |
1da177e4c Linux-2.6.12-rc2 |
2831 |
*/ |
4f74d2c8e vm: remove 'nr_ac... |
2832 2833 2834 |
while (vma) { if (vma->vm_flags & VM_ACCOUNT) nr_accounted += vma_pages(vma); |
a8fb5618d [PATCH] mm: unlin... |
2835 |
vma = remove_vma(vma); |
4f74d2c8e vm: remove 'nr_ac... |
2836 2837 |
} vm_unacct_memory(nr_accounted); |
1da177e4c Linux-2.6.12-rc2 |
2838 2839 2840 2841 |
} /* Insert vm structure into process list sorted by address * and into the inode's i_mmap tree. If vm_file is non-NULL |
c8c06efa8 mm: convert i_mma... |
2842 |
* then i_mmap_rwsem is taken here. |
1da177e4c Linux-2.6.12-rc2 |
2843 |
*/ |
6597d7833 mm/mmap.c: replac... |
2844 |
int insert_vm_struct(struct mm_struct *mm, struct vm_area_struct *vma) |
1da177e4c Linux-2.6.12-rc2 |
2845 |
{ |
6597d7833 mm/mmap.c: replac... |
2846 2847 |
struct vm_area_struct *prev; struct rb_node **rb_link, *rb_parent; |
1da177e4c Linux-2.6.12-rc2 |
2848 |
|
c9d13f5fc mm/mmap.c:insert_... |
2849 2850 2851 2852 2853 2854 |
if (find_vma_links(mm, vma->vm_start, vma->vm_end, &prev, &rb_link, &rb_parent)) return -ENOMEM; if ((vma->vm_flags & VM_ACCOUNT) && security_vm_enough_memory_mm(mm, vma_pages(vma))) return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 |
/* * The vm_pgoff of a purely anonymous vma should be irrelevant * until its first write fault, when page's anon_vma and index * are set. But now set the vm_pgoff it will almost certainly * end up with (unless mremap moves it elsewhere before that * first wfault), so /proc/pid/maps tells a consistent story. * * By setting it to reflect the virtual start address of the * vma, merges and splits can happen in a seamless way, just * using the existing file pgoff checks and manipulations. * Similarly in do_mmap_pgoff and in do_brk. */ |
8a9cc3b55 mmap: fix the usa... |
2867 |
if (vma_is_anonymous(vma)) { |
1da177e4c Linux-2.6.12-rc2 |
2868 2869 2870 |
BUG_ON(vma->anon_vma); vma->vm_pgoff = vma->vm_start >> PAGE_SHIFT; } |
2b1444983 uprobes, mm, x86:... |
2871 |
|
1da177e4c Linux-2.6.12-rc2 |
2872 2873 2874 2875 2876 2877 2878 2879 2880 |
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 ... |
2881 2882 |
unsigned long addr, unsigned long len, pgoff_t pgoff, bool *need_rmap_locks) |
1da177e4c Linux-2.6.12-rc2 |
2883 2884 2885 2886 2887 2888 |
{ struct vm_area_struct *vma = *vmap; unsigned long vma_start = vma->vm_start; struct mm_struct *mm = vma->vm_mm; struct vm_area_struct *new_vma, *prev; struct rb_node **rb_link, *rb_parent; |
948f017b0 mremap: enforce r... |
2889 |
bool faulted_in_anon_vma = true; |
1da177e4c Linux-2.6.12-rc2 |
2890 2891 2892 2893 2894 |
/* * If anonymous vma has not yet been faulted, update new pgoff * to match new location, to increase its chance of merging. */ |
ce75799b8 mremap: fix the w... |
2895 |
if (unlikely(vma_is_anonymous(vma) && !vma->anon_vma)) { |
1da177e4c Linux-2.6.12-rc2 |
2896 |
pgoff = addr >> PAGE_SHIFT; |
948f017b0 mremap: enforce r... |
2897 2898 |
faulted_in_anon_vma = false; } |
1da177e4c Linux-2.6.12-rc2 |
2899 |
|
6597d7833 mm/mmap.c: replac... |
2900 2901 |
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 |
2902 |
new_vma = vma_merge(mm, prev, addr, addr + len, vma->vm_flags, |
19a809afe userfaultfd: teac... |
2903 2904 |
vma->anon_vma, vma->vm_file, pgoff, vma_policy(vma), vma->vm_userfaultfd_ctx); |
1da177e4c Linux-2.6.12-rc2 |
2905 2906 2907 2908 |
if (new_vma) { /* * Source vma may have been merged into new_vma */ |
948f017b0 mremap: enforce r... |
2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 |
if (unlikely(vma_start >= new_vma->vm_start && vma_start < new_vma->vm_end)) { /* * The only way we can get a vma_merge with * self during an mremap is if the vma hasn't * been faulted in yet and we were allowed to * reset the dst vma->vm_pgoff to the * destination address of the mremap to allow * the merge to happen. mremap must change the * vm_pgoff linearity between src and dst vmas * (in turn preventing a vma_merge) to be * safe. It is only safe to keep the vm_pgoff * linear if there are no pages mapped yet. */ |
81d1b09c6 mm: convert a few... |
2923 |
VM_BUG_ON_VMA(faulted_in_anon_vma, new_vma); |
38a76013a mm: avoid taking ... |
2924 |
*vmap = vma = new_vma; |
108d6642a mm anon rmap: rem... |
2925 |
} |
38a76013a mm: avoid taking ... |
2926 |
*need_rmap_locks = (new_vma->vm_pgoff <= vma->vm_pgoff); |
1da177e4c Linux-2.6.12-rc2 |
2927 |
} else { |
e94b17660 [PATCH] slab: rem... |
2928 |
new_vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); |
e39758912 mm/mmap.c: simpli... |
2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 |
if (!new_vma) goto out; *new_vma = *vma; new_vma->vm_start = addr; new_vma->vm_end = addr + len; new_vma->vm_pgoff = pgoff; if (vma_dup_policy(vma, new_vma)) goto out_free_vma; INIT_LIST_HEAD(&new_vma->anon_vma_chain); if (anon_vma_clone(new_vma, vma)) goto out_free_mempol; if (new_vma->vm_file) get_file(new_vma->vm_file); if (new_vma->vm_ops && new_vma->vm_ops->open) new_vma->vm_ops->open(new_vma); vma_link(mm, new_vma, prev, rb_link, rb_parent); *need_rmap_locks = false; |
1da177e4c Linux-2.6.12-rc2 |
2946 2947 |
} return new_vma; |
5beb49305 mm: change anon_v... |
2948 |
|
e39758912 mm/mmap.c: simpli... |
2949 |
out_free_mempol: |
ef0855d33 mm: mempolicy: tu... |
2950 |
mpol_put(vma_policy(new_vma)); |
e39758912 mm/mmap.c: simpli... |
2951 |
out_free_vma: |
5beb49305 mm: change anon_v... |
2952 |
kmem_cache_free(vm_area_cachep, new_vma); |
e39758912 mm/mmap.c: simpli... |
2953 |
out: |
5beb49305 mm: change anon_v... |
2954 |
return NULL; |
1da177e4c Linux-2.6.12-rc2 |
2955 |
} |
119f657c7 [PATCH] RLIMIT_AS... |
2956 2957 2958 2959 2960 |
/* * Return true if the calling process may expand its vm space by the passed * number of pages */ |
846383359 mm: rework virtua... |
2961 |
bool may_expand_vm(struct mm_struct *mm, vm_flags_t flags, unsigned long npages) |
119f657c7 [PATCH] RLIMIT_AS... |
2962 |
{ |
846383359 mm: rework virtua... |
2963 2964 |
if (mm->total_vm + npages > rlimit(RLIMIT_AS) >> PAGE_SHIFT) return false; |
119f657c7 [PATCH] RLIMIT_AS... |
2965 |
|
d977d56ce mm: warn about Vm... |
2966 2967 |
if (is_data_mapping(flags) && mm->data_vm + npages > rlimit(RLIMIT_DATA) >> PAGE_SHIFT) { |
f4fcd5584 mm: enable RLIMIT... |
2968 2969 2970 2971 2972 2973 2974 |
/* Workaround for Valgrind */ if (rlimit(RLIMIT_DATA) == 0 && mm->data_vm + npages <= rlimit_max(RLIMIT_DATA) >> PAGE_SHIFT) return true; if (!ignore_rlimit_data) { pr_warn_once("%s (%d): VmData %lu exceed data ulimit %lu. Update limits or use boot option ignore_rlimit_data. ", |
d977d56ce mm: warn about Vm... |
2975 2976 2977 |
current->comm, current->pid, (mm->data_vm + npages) << PAGE_SHIFT, rlimit(RLIMIT_DATA)); |
d977d56ce mm: warn about Vm... |
2978 |
return false; |
f4fcd5584 mm: enable RLIMIT... |
2979 |
} |
d977d56ce mm: warn about Vm... |
2980 |
} |
119f657c7 [PATCH] RLIMIT_AS... |
2981 |
|
846383359 mm: rework virtua... |
2982 2983 2984 2985 2986 2987 |
return true; } void vm_stat_account(struct mm_struct *mm, vm_flags_t flags, long npages) { mm->total_vm += npages; |
d977d56ce mm: warn about Vm... |
2988 |
if (is_exec_mapping(flags)) |
846383359 mm: rework virtua... |
2989 |
mm->exec_vm += npages; |
d977d56ce mm: warn about Vm... |
2990 |
else if (is_stack_mapping(flags)) |
846383359 mm: rework virtua... |
2991 |
mm->stack_vm += npages; |
d977d56ce mm: warn about Vm... |
2992 |
else if (is_data_mapping(flags)) |
846383359 mm: rework virtua... |
2993 |
mm->data_vm += npages; |
119f657c7 [PATCH] RLIMIT_AS... |
2994 |
} |
fa5dc22f8 [PATCH] Add insta... |
2995 |
|
a62c34bd2 x86, mm: Improve ... |
2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 |
static int special_mapping_fault(struct vm_area_struct *vma, struct vm_fault *vmf); /* * Having a close hook prevents vma merging regardless of flags. */ static void special_mapping_close(struct vm_area_struct *vma) { } static const char *special_mapping_name(struct vm_area_struct *vma) { return ((struct vm_special_mapping *)vma->vm_private_data)->name; } |
b059a453b x86/vdso: Add mre... |
3010 3011 3012 3013 3014 3015 3016 3017 |
static int special_mapping_mremap(struct vm_area_struct *new_vma) { struct vm_special_mapping *sm = new_vma->vm_private_data; if (sm->mremap) return sm->mremap(sm, new_vma); return 0; } |
a62c34bd2 x86, mm: Improve ... |
3018 3019 3020 |
static const struct vm_operations_struct special_mapping_vmops = { .close = special_mapping_close, .fault = special_mapping_fault, |
b059a453b x86/vdso: Add mre... |
3021 |
.mremap = special_mapping_mremap, |
a62c34bd2 x86, mm: Improve ... |
3022 3023 3024 3025 3026 3027 3028 |
.name = special_mapping_name, }; static const struct vm_operations_struct legacy_special_mapping_vmops = { .close = special_mapping_close, .fault = special_mapping_fault, }; |
fa5dc22f8 [PATCH] Add insta... |
3029 |
|
b1d0e4f53 mm: special mappi... |
3030 3031 |
static int special_mapping_fault(struct vm_area_struct *vma, struct vm_fault *vmf) |
fa5dc22f8 [PATCH] Add insta... |
3032 |
{ |
b1d0e4f53 mm: special mappi... |
3033 |
pgoff_t pgoff; |
fa5dc22f8 [PATCH] Add insta... |
3034 |
struct page **pages; |
f872f5400 mm: Add a vm_spec... |
3035 |
if (vma->vm_ops == &legacy_special_mapping_vmops) { |
a62c34bd2 x86, mm: Improve ... |
3036 |
pages = vma->vm_private_data; |
f872f5400 mm: Add a vm_spec... |
3037 3038 3039 3040 3041 3042 3043 3044 |
} else { struct vm_special_mapping *sm = vma->vm_private_data; if (sm->fault) return sm->fault(sm, vma, vmf); pages = sm->pages; } |
a62c34bd2 x86, mm: Improve ... |
3045 |
|
8a9cc3b55 mmap: fix the usa... |
3046 |
for (pgoff = vmf->pgoff; pgoff && *pages; ++pages) |
b1d0e4f53 mm: special mappi... |
3047 |
pgoff--; |
fa5dc22f8 [PATCH] Add insta... |
3048 3049 3050 3051 |
if (*pages) { struct page *page = *pages; get_page(page); |
b1d0e4f53 mm: special mappi... |
3052 3053 |
vmf->page = page; return 0; |
fa5dc22f8 [PATCH] Add insta... |
3054 |
} |
b1d0e4f53 mm: special mappi... |
3055 |
return VM_FAULT_SIGBUS; |
fa5dc22f8 [PATCH] Add insta... |
3056 |
} |
a62c34bd2 x86, mm: Improve ... |
3057 3058 3059 |
static struct vm_area_struct *__install_special_mapping( struct mm_struct *mm, unsigned long addr, unsigned long len, |
27f28b972 mm/mmap.c: change... |
3060 3061 |
unsigned long vm_flags, void *priv, const struct vm_operations_struct *ops) |
fa5dc22f8 [PATCH] Add insta... |
3062 |
{ |
462e635e5 install_special_m... |
3063 |
int ret; |
fa5dc22f8 [PATCH] Add insta... |
3064 3065 3066 3067 |
struct vm_area_struct *vma; vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); if (unlikely(vma == NULL)) |
3935ed6a3 mm: Add new func ... |
3068 |
return ERR_PTR(-ENOMEM); |
fa5dc22f8 [PATCH] Add insta... |
3069 |
|
5beb49305 mm: change anon_v... |
3070 |
INIT_LIST_HEAD(&vma->anon_vma_chain); |
fa5dc22f8 [PATCH] Add insta... |
3071 3072 3073 |
vma->vm_mm = mm; vma->vm_start = addr; vma->vm_end = addr + len; |
d9104d1ca mm: track vma cha... |
3074 |
vma->vm_flags = vm_flags | mm->def_flags | VM_DONTEXPAND | VM_SOFTDIRTY; |
3ed75eb8f setup vma->vm_pag... |
3075 |
vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); |
fa5dc22f8 [PATCH] Add insta... |
3076 |
|
a62c34bd2 x86, mm: Improve ... |
3077 3078 |
vma->vm_ops = ops; vma->vm_private_data = priv; |
fa5dc22f8 [PATCH] Add insta... |
3079 |
|
462e635e5 install_special_m... |
3080 3081 3082 |
ret = insert_vm_struct(mm, vma); if (ret) goto out; |
fa5dc22f8 [PATCH] Add insta... |
3083 |
|
846383359 mm: rework virtua... |
3084 |
vm_stat_account(mm, vma->vm_flags, len >> PAGE_SHIFT); |
fa5dc22f8 [PATCH] Add insta... |
3085 |
|
cdd6c482c perf: Do the big ... |
3086 |
perf_event_mmap(vma); |
089dd79db perf_counter: Gen... |
3087 |
|
3935ed6a3 mm: Add new func ... |
3088 |
return vma; |
462e635e5 install_special_m... |
3089 3090 3091 |
out: kmem_cache_free(vm_area_cachep, vma); |
3935ed6a3 mm: Add new func ... |
3092 3093 |
return ERR_PTR(ret); } |
2eefd8789 x86/arch_prctl/vd... |
3094 3095 3096 3097 3098 3099 3100 |
bool vma_is_special_mapping(const struct vm_area_struct *vma, const struct vm_special_mapping *sm) { return vma->vm_private_data == sm && (vma->vm_ops == &special_mapping_vmops || vma->vm_ops == &legacy_special_mapping_vmops); } |
a62c34bd2 x86, mm: Improve ... |
3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 |
/* * Called with mm->mmap_sem held for writing. * Insert a new vma covering the given region, with the given flags. * Its pages are supplied by the given array of struct page *. * The array can be shorter than len >> PAGE_SHIFT if it's null-terminated. * The region past the last page supplied will always produce SIGBUS. * The array pointer and the pages it points to are assumed to stay alive * for as long as this mapping might exist. */ struct vm_area_struct *_install_special_mapping( struct mm_struct *mm, unsigned long addr, unsigned long len, unsigned long vm_flags, const struct vm_special_mapping *spec) { |
27f28b972 mm/mmap.c: change... |
3115 3116 |
return __install_special_mapping(mm, addr, len, vm_flags, (void *)spec, &special_mapping_vmops); |
a62c34bd2 x86, mm: Improve ... |
3117 |
} |
3935ed6a3 mm: Add new func ... |
3118 3119 3120 3121 |
int install_special_mapping(struct mm_struct *mm, unsigned long addr, unsigned long len, unsigned long vm_flags, struct page **pages) { |
a62c34bd2 x86, mm: Improve ... |
3122 |
struct vm_area_struct *vma = __install_special_mapping( |
27f28b972 mm/mmap.c: change... |
3123 3124 |
mm, addr, len, vm_flags, (void *)pages, &legacy_special_mapping_vmops); |
3935ed6a3 mm: Add new func ... |
3125 |
|
14bd5b458 mm/mmap.c: replac... |
3126 |
return PTR_ERR_OR_ZERO(vma); |
fa5dc22f8 [PATCH] Add insta... |
3127 |
} |
7906d00cd mmu-notifiers: ad... |
3128 3129 |
static DEFINE_MUTEX(mm_all_locks_mutex); |
454ed842d lockdep: annotate... |
3130 |
static void vm_lock_anon_vma(struct mm_struct *mm, struct anon_vma *anon_vma) |
7906d00cd mmu-notifiers: ad... |
3131 |
{ |
bf181b9f9 mm anon rmap: rep... |
3132 |
if (!test_bit(0, (unsigned long *) &anon_vma->root->rb_root.rb_node)) { |
7906d00cd mmu-notifiers: ad... |
3133 3134 3135 3136 |
/* * The LSB of head.next can't change from under us * because we hold the mm_all_locks_mutex. */ |
572043c90 mm: mmap: annotat... |
3137 |
down_write_nest_lock(&anon_vma->root->rwsem, &mm->mmap_sem); |
7906d00cd mmu-notifiers: ad... |
3138 3139 |
/* * We can safely modify head.next after taking the |
5a505085f mm/rmap: Convert ... |
3140 |
* anon_vma->root->rwsem. If some other vma in this mm shares |
7906d00cd mmu-notifiers: ad... |
3141 3142 3143 3144 |
* 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 ... |
3145 |
* anon_vma->root->rwsem. |
7906d00cd mmu-notifiers: ad... |
3146 3147 |
*/ if (__test_and_set_bit(0, (unsigned long *) |
bf181b9f9 mm anon rmap: rep... |
3148 |
&anon_vma->root->rb_root.rb_node)) |
7906d00cd mmu-notifiers: ad... |
3149 3150 3151 |
BUG(); } } |
454ed842d lockdep: annotate... |
3152 |
static void vm_lock_mapping(struct mm_struct *mm, struct address_space *mapping) |
7906d00cd mmu-notifiers: ad... |
3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 |
{ if (!test_bit(AS_MM_ALL_LOCKS, &mapping->flags)) { /* * AS_MM_ALL_LOCKS can't change from under us because * we hold the mm_all_locks_mutex. * * Operations on ->flags have to be atomic because * even if AS_MM_ALL_LOCKS is stable thanks to the * mm_all_locks_mutex, there may be other cpus * changing other bitflags in parallel to us. */ if (test_and_set_bit(AS_MM_ALL_LOCKS, &mapping->flags)) BUG(); |
c8c06efa8 mm: convert i_mma... |
3166 |
down_write_nest_lock(&mapping->i_mmap_rwsem, &mm->mmap_sem); |
7906d00cd mmu-notifiers: ad... |
3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 |
} } /* * This operation locks against the VM for all pte/vma/mm related * operations that could ever happen on a certain mm. This includes * vmtruncate, try_to_unmap, and all page faults. * * The caller must take the mmap_sem in write mode before calling * mm_take_all_locks(). The caller isn't allowed to release the * mmap_sem until mm_drop_all_locks() returns. * * mmap_sem in write mode is required in order to block all operations * that could modify pagetables and free pages without need of |
27ba0644e rmap: drop suppor... |
3181 |
* altering the vma layout. It's also needed in write mode to avoid new |
7906d00cd mmu-notifiers: ad... |
3182 3183 3184 3185 3186 |
* 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... |
3187 |
* The LSB in anon_vma->rb_root.rb_node and the AS_MM_ALL_LOCKS bitflag in |
7906d00cd mmu-notifiers: ad... |
3188 3189 3190 |
* mapping->flags avoid to take the same lock twice, if more than one * vma in this mm is backed by the same anon_vma or address_space. * |
88f306b68 mm: fix locking o... |
3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 |
* We take locks in following order, accordingly to comment at beginning * of mm/rmap.c: * - all hugetlbfs_i_mmap_rwsem_key locks (aka mapping->i_mmap_rwsem for * hugetlb mapping); * - all i_mmap_rwsem locks; * - all anon_vma->rwseml * * We can take all locks within these types randomly because the VM code * doesn't nest them and we protected from parallel mm_take_all_locks() by * mm_all_locks_mutex. |
7906d00cd mmu-notifiers: ad... |
3201 3202 3203 3204 3205 3206 3207 3208 3209 |
* * 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... |
3210 |
struct anon_vma_chain *avc; |
7906d00cd mmu-notifiers: ad... |
3211 3212 3213 3214 3215 3216 3217 3218 |
BUG_ON(down_read_trylock(&mm->mmap_sem)); mutex_lock(&mm_all_locks_mutex); for (vma = mm->mmap; vma; vma = vma->vm_next) { if (signal_pending(current)) goto out_unlock; |
88f306b68 mm: fix locking o... |
3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 |
if (vma->vm_file && vma->vm_file->f_mapping && is_vm_hugetlb_page(vma)) vm_lock_mapping(mm, vma->vm_file->f_mapping); } for (vma = mm->mmap; vma; vma = vma->vm_next) { if (signal_pending(current)) goto out_unlock; if (vma->vm_file && vma->vm_file->f_mapping && !is_vm_hugetlb_page(vma)) |
454ed842d lockdep: annotate... |
3229 |
vm_lock_mapping(mm, vma->vm_file->f_mapping); |
7906d00cd mmu-notifiers: ad... |
3230 |
} |
7cd5a02f5 mm: fix mm_take_a... |
3231 3232 3233 3234 3235 |
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... |
3236 3237 |
list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) vm_lock_anon_vma(mm, avc->anon_vma); |
7906d00cd mmu-notifiers: ad... |
3238 |
} |
7cd5a02f5 mm: fix mm_take_a... |
3239 |
|
584cff54e mm/mmap.c: elimin... |
3240 |
return 0; |
7906d00cd mmu-notifiers: ad... |
3241 3242 |
out_unlock: |
584cff54e mm/mmap.c: elimin... |
3243 3244 |
mm_drop_all_locks(mm); return -EINTR; |
7906d00cd mmu-notifiers: ad... |
3245 3246 3247 3248 |
} static void vm_unlock_anon_vma(struct anon_vma *anon_vma) { |
bf181b9f9 mm anon rmap: rep... |
3249 |
if (test_bit(0, (unsigned long *) &anon_vma->root->rb_root.rb_node)) { |
7906d00cd mmu-notifiers: ad... |
3250 3251 3252 3253 3254 |
/* * 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... |
3255 |
* the vma so the users using the anon_vma->rb_root will |
7906d00cd mmu-notifiers: ad... |
3256 3257 3258 3259 |
* 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 ... |
3260 |
* anon_vma->root->rwsem. |
7906d00cd mmu-notifiers: ad... |
3261 3262 |
*/ if (!__test_and_clear_bit(0, (unsigned long *) |
bf181b9f9 mm anon rmap: rep... |
3263 |
&anon_vma->root->rb_root.rb_node)) |
7906d00cd mmu-notifiers: ad... |
3264 |
BUG(); |
08b52706d mm/rmap: rename a... |
3265 |
anon_vma_unlock_write(anon_vma); |
7906d00cd mmu-notifiers: ad... |
3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 |
} } static void vm_unlock_mapping(struct address_space *mapping) { if (test_bit(AS_MM_ALL_LOCKS, &mapping->flags)) { /* * AS_MM_ALL_LOCKS can't change to 0 from under us * because we hold the mm_all_locks_mutex. */ |
83cde9e8b mm: use new helpe... |
3276 |
i_mmap_unlock_write(mapping); |
7906d00cd mmu-notifiers: ad... |
3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 |
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... |
3290 |
struct anon_vma_chain *avc; |
7906d00cd mmu-notifiers: ad... |
3291 3292 3293 3294 3295 3296 |
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... |
3297 3298 |
list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) vm_unlock_anon_vma(avc->anon_vma); |
7906d00cd mmu-notifiers: ad... |
3299 3300 3301 3302 3303 3304 |
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 ... |
3305 3306 3307 3308 3309 3310 |
/* * initialise the VMA slab */ void __init mmap_init(void) { |
00a62ce91 mm: fix Committed... |
3311 |
int ret; |
908c7f194 percpu_counter: a... |
3312 |
ret = percpu_counter_init(&vm_committed_as, 0, GFP_KERNEL); |
00a62ce91 mm: fix Committed... |
3313 |
VM_BUG_ON(ret); |
8feae1311 NOMMU: Make VMAs ... |
3314 |
} |
c9b1d0981 mm: limit growth ... |
3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 |
/* * 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... |
3326 |
static int init_user_reserve(void) |
c9b1d0981 mm: limit growth ... |
3327 3328 3329 3330 3331 3332 3333 3334 |
{ unsigned long free_kbytes; free_kbytes = global_page_state(NR_FREE_PAGES) << (PAGE_SHIFT - 10); sysctl_user_reserve_kbytes = min(free_kbytes / 32, 1UL << 17); return 0; } |
a64fb3cd6 mm: audit/fix non... |
3335 |
subsys_initcall(init_user_reserve); |
4eeab4f55 mm: replace hardc... |
3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 |
/* * 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... |
3347 |
static int init_admin_reserve(void) |
4eeab4f55 mm: replace hardc... |
3348 3349 3350 3351 3352 3353 3354 3355 |
{ unsigned long free_kbytes; free_kbytes = global_page_state(NR_FREE_PAGES) << (PAGE_SHIFT - 10); sysctl_admin_reserve_kbytes = min(free_kbytes / 32, 1UL << 13); return 0; } |
a64fb3cd6 mm: audit/fix non... |
3356 |
subsys_initcall(init_admin_reserve); |
1640879af mm: reinititalise... |
3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 |
/* * Reinititalise user and admin reserves if memory is added or removed. * * The default user reserve max is 128MB, and the default max for the * admin reserve is 8MB. These are usually, but not always, enough to * enable recovery from a memory hogging process using login/sshd, a shell, * and tools like top. It may make sense to increase or even disable the * reserve depending on the existence of swap or variations in the recovery * tools. So, the admin may have changed them. * * If memory is added and the reserves have been eliminated or increased above * the default max, then we'll trust the admin. * * If memory is removed and there isn't enough free memory, then we * need to reset the reserves. * * Otherwise keep the reserve set by the admin. */ static int reserve_mem_notifier(struct notifier_block *nb, unsigned long action, void *data) { unsigned long tmp, free_kbytes; switch (action) { case MEM_ONLINE: /* Default max is 128MB. Leave alone if modified by operator. */ tmp = sysctl_user_reserve_kbytes; if (0 < tmp && tmp < (1UL << 17)) init_user_reserve(); /* Default max is 8MB. Leave alone if modified by operator. */ tmp = sysctl_admin_reserve_kbytes; if (0 < tmp && tmp < (1UL << 13)) init_admin_reserve(); break; case MEM_OFFLINE: free_kbytes = global_page_state(NR_FREE_PAGES) << (PAGE_SHIFT - 10); if (sysctl_user_reserve_kbytes > free_kbytes) { init_user_reserve(); pr_info("vm.user_reserve_kbytes reset to %lu ", sysctl_user_reserve_kbytes); } if (sysctl_admin_reserve_kbytes > free_kbytes) { init_admin_reserve(); pr_info("vm.admin_reserve_kbytes reset to %lu ", sysctl_admin_reserve_kbytes); } break; default: break; } return NOTIFY_OK; } static struct notifier_block reserve_mem_nb = { .notifier_call = reserve_mem_notifier, }; static int __meminit init_reserve_notifier(void) { if (register_hotmemory_notifier(&reserve_mem_nb)) |
b1de0d139 mm: convert some ... |
3424 3425 |
pr_err("Failed registering memory add/remove notifier for admin reserve "); |
1640879af mm: reinititalise... |
3426 3427 3428 |
return 0; } |
a64fb3cd6 mm: audit/fix non... |
3429 |
subsys_initcall(init_reserve_notifier); |