Blame view
mm/memory.c
107 KB
1da177e4c
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 |
/* * linux/mm/memory.c * * Copyright (C) 1991, 1992, 1993, 1994 Linus Torvalds */ /* * demand-loading started 01.12.91 - seems it is high on the list of * things wanted, and it should be easy to implement. - Linus */ /* * Ok, demand-loading was easy, shared pages a little bit tricker. Shared * pages started 02.12.91, seems to work. - Linus. * * Tested sharing by executing about 30 /bin/sh: under the old kernel it * would have taken more than the 6M I have free, but it worked well as * far as I could see. * * Also corrected some "invalidate()"s - I wasn't doing enough of them. */ /* * Real VM (paging to/from disk) started 18.12.91. Much more work and * thought has to go into this. Oh, well.. * 19.12.91 - works, somewhat. Sometimes I get faults, don't know why. * Found it. Everything seems to work now. * 20.12.91 - Ok, making the swap-device changeable like the root. */ /* * 05.04.94 - Multi-page memory management added for v1.1. * Idea by Alex Bligh (alex@cconcepts.co.uk) * * 16.07.99 - Support of BIGMEM added by Gerhard Wichert, Siemens AG * (Gerhard.Wichert@pdb.siemens.de) * * Aug/Sep 2004 Changed to four level page tables (Andi Kleen) */ #include <linux/kernel_stat.h> #include <linux/mm.h> #include <linux/hugetlb.h> #include <linux/mman.h> #include <linux/swap.h> #include <linux/highmem.h> #include <linux/pagemap.h> |
9a8408951
|
48 |
#include <linux/ksm.h> |
1da177e4c
|
49 |
#include <linux/rmap.h> |
b95f1b31b
|
50 |
#include <linux/export.h> |
0ff922452
|
51 |
#include <linux/delayacct.h> |
1da177e4c
|
52 |
#include <linux/init.h> |
edc79b2a4
|
53 |
#include <linux/writeback.h> |
8a9f3ccd2
|
54 |
#include <linux/memcontrol.h> |
cddb8a5c1
|
55 |
#include <linux/mmu_notifier.h> |
3dc147414
|
56 57 58 |
#include <linux/kallsyms.h> #include <linux/swapops.h> #include <linux/elf.h> |
5a0e3ad6a
|
59 |
#include <linux/gfp.h> |
1da177e4c
|
60 |
|
6952b61de
|
61 |
#include <asm/io.h> |
1da177e4c
|
62 63 64 65 66 |
#include <asm/pgalloc.h> #include <asm/uaccess.h> #include <asm/tlb.h> #include <asm/tlbflush.h> #include <asm/pgtable.h> |
42b777281
|
67 |
#include "internal.h" |
d41dee369
|
68 |
#ifndef CONFIG_NEED_MULTIPLE_NODES |
1da177e4c
|
69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 |
/* use the per-pgdat data instead for discontigmem - mbligh */ unsigned long max_mapnr; struct page *mem_map; EXPORT_SYMBOL(max_mapnr); EXPORT_SYMBOL(mem_map); #endif unsigned long num_physpages; /* * A number of key systems in x86 including ioremap() rely on the assumption * that high_memory defines the upper bound on direct map memory, then end * of ZONE_NORMAL. Under CONFIG_DISCONTIG this means that max_low_pfn and * highstart_pfn must be the same; there must be no gap between ZONE_NORMAL * and ZONE_HIGHMEM. */ void * high_memory; |
1da177e4c
|
86 87 88 |
EXPORT_SYMBOL(num_physpages); EXPORT_SYMBOL(high_memory); |
1da177e4c
|
89 |
|
32a932332
|
90 91 92 93 94 95 96 97 98 99 100 101 |
/* * Randomize the address space (stacks, mmaps, brk, etc.). * * ( When CONFIG_COMPAT_BRK=y we exclude brk from randomization, * as ancient (libc5 based) binaries can segfault. ) */ int randomize_va_space __read_mostly = #ifdef CONFIG_COMPAT_BRK 1; #else 2; #endif |
a62eaf151
|
102 103 104 105 |
static int __init disable_randmaps(char *s) { randomize_va_space = 0; |
9b41046cd
|
106 |
return 1; |
a62eaf151
|
107 108 |
} __setup("norandmaps", disable_randmaps); |
62eede62d
|
109 |
unsigned long zero_pfn __read_mostly; |
03f6462a3
|
110 |
unsigned long highest_memmap_pfn __read_mostly; |
a13ea5b75
|
111 112 113 114 115 116 117 118 119 120 |
/* * CONFIG_MMU architectures set up ZERO_PAGE in their paging_init() */ static int __init init_zero_pfn(void) { zero_pfn = page_to_pfn(ZERO_PAGE(0)); return 0; } core_initcall(init_zero_pfn); |
a62eaf151
|
121 |
|
d559db086
|
122 |
|
34e55232e
|
123 |
#if defined(SPLIT_RSS_COUNTING) |
a3a2e76c7
|
124 |
static void __sync_task_rss_stat(struct task_struct *task, struct mm_struct *mm) |
34e55232e
|
125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 |
{ int i; for (i = 0; i < NR_MM_COUNTERS; i++) { if (task->rss_stat.count[i]) { add_mm_counter(mm, i, task->rss_stat.count[i]); task->rss_stat.count[i] = 0; } } task->rss_stat.events = 0; } static void add_mm_counter_fast(struct mm_struct *mm, int member, int val) { struct task_struct *task = current; if (likely(task->mm == mm)) task->rss_stat.count[member] += val; else add_mm_counter(mm, member, val); } #define inc_mm_counter_fast(mm, member) add_mm_counter_fast(mm, member, 1) #define dec_mm_counter_fast(mm, member) add_mm_counter_fast(mm, member, -1) /* sync counter once per 64 page faults */ #define TASK_RSS_EVENTS_THRESH (64) static void check_sync_rss_stat(struct task_struct *task) { if (unlikely(task != current)) return; if (unlikely(task->rss_stat.events++ > TASK_RSS_EVENTS_THRESH)) __sync_task_rss_stat(task, task->mm); } unsigned long get_mm_counter(struct mm_struct *mm, int member) { long val = 0; /* * Don't use task->mm here...for avoiding to use task_get_mm().. * The caller must guarantee task->mm is not invalid. */ val = atomic_long_read(&mm->rss_stat.count[member]); /* * counter is updated in asynchronous manner and may go to minus. * But it's never be expected number for users. */ if (val < 0) return 0; return (unsigned long)val; } void sync_mm_rss(struct task_struct *task, struct mm_struct *mm) { __sync_task_rss_stat(task, mm); } |
9547d01bf
|
181 |
#else /* SPLIT_RSS_COUNTING */ |
34e55232e
|
182 183 184 185 186 187 188 |
#define inc_mm_counter_fast(mm, member) inc_mm_counter(mm, member) #define dec_mm_counter_fast(mm, member) dec_mm_counter(mm, member) static void check_sync_rss_stat(struct task_struct *task) { } |
9547d01bf
|
189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 |
#endif /* SPLIT_RSS_COUNTING */ #ifdef HAVE_GENERIC_MMU_GATHER static int tlb_next_batch(struct mmu_gather *tlb) { struct mmu_gather_batch *batch; batch = tlb->active; if (batch->next) { tlb->active = batch->next; return 1; } batch = (void *)__get_free_pages(GFP_NOWAIT | __GFP_NOWARN, 0); if (!batch) return 0; batch->next = NULL; batch->nr = 0; batch->max = MAX_GATHER_BATCH; tlb->active->next = batch; tlb->active = batch; return 1; } /* tlb_gather_mmu * Called to initialize an (on-stack) mmu_gather structure for page-table * tear-down from @mm. The @fullmm argument is used when @mm is without * users and we're going to destroy the full address space (exit/execve). */ void tlb_gather_mmu(struct mmu_gather *tlb, struct mm_struct *mm, bool fullmm) { tlb->mm = mm; tlb->fullmm = fullmm; tlb->need_flush = 0; tlb->fast_mode = (num_possible_cpus() == 1); tlb->local.next = NULL; tlb->local.nr = 0; tlb->local.max = ARRAY_SIZE(tlb->__pages); tlb->active = &tlb->local; #ifdef CONFIG_HAVE_RCU_TABLE_FREE tlb->batch = NULL; #endif } void tlb_flush_mmu(struct mmu_gather *tlb) { struct mmu_gather_batch *batch; if (!tlb->need_flush) return; tlb->need_flush = 0; tlb_flush(tlb); #ifdef CONFIG_HAVE_RCU_TABLE_FREE tlb_table_flush(tlb); |
34e55232e
|
249 |
#endif |
9547d01bf
|
250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 |
if (tlb_fast_mode(tlb)) return; for (batch = &tlb->local; batch; batch = batch->next) { free_pages_and_swap_cache(batch->pages, batch->nr); batch->nr = 0; } tlb->active = &tlb->local; } /* tlb_finish_mmu * Called at the end of the shootdown operation to free up any resources * that were required. */ void tlb_finish_mmu(struct mmu_gather *tlb, unsigned long start, unsigned long end) { struct mmu_gather_batch *batch, *next; tlb_flush_mmu(tlb); /* keep the page table cache within bounds */ check_pgt_cache(); for (batch = tlb->local.next; batch; batch = next) { next = batch->next; free_pages((unsigned long)batch, 0); } tlb->local.next = NULL; } /* __tlb_remove_page * Must perform the equivalent to __free_pte(pte_get_and_clear(ptep)), while * handling the additional races in SMP caused by other CPUs caching valid * mappings in their TLBs. Returns the number of free page slots left. * When out of page slots we must call tlb_flush_mmu(). */ int __tlb_remove_page(struct mmu_gather *tlb, struct page *page) { struct mmu_gather_batch *batch; |
f21760b15
|
289 |
VM_BUG_ON(!tlb->need_flush); |
9547d01bf
|
290 291 292 293 294 295 296 297 298 299 300 |
if (tlb_fast_mode(tlb)) { free_page_and_swap_cache(page); return 1; /* avoid calling tlb_flush_mmu() */ } batch = tlb->active; batch->pages[batch->nr++] = page; if (batch->nr == batch->max) { if (!tlb_next_batch(tlb)) return 0; |
0b43c3aab
|
301 |
batch = tlb->active; |
9547d01bf
|
302 303 304 305 306 307 308 |
} VM_BUG_ON(batch->nr > batch->max); return batch->max - batch->nr; } #endif /* HAVE_GENERIC_MMU_GATHER */ |
267239116
|
309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 |
#ifdef CONFIG_HAVE_RCU_TABLE_FREE /* * See the comment near struct mmu_table_batch. */ static void tlb_remove_table_smp_sync(void *arg) { /* Simply deliver the interrupt */ } static void tlb_remove_table_one(void *table) { /* * This isn't an RCU grace period and hence the page-tables cannot be * assumed to be actually RCU-freed. * * It is however sufficient for software page-table walkers that rely on * IRQ disabling. See the comment near struct mmu_table_batch. */ smp_call_function(tlb_remove_table_smp_sync, NULL, 1); __tlb_remove_table(table); } static void tlb_remove_table_rcu(struct rcu_head *head) { struct mmu_table_batch *batch; int i; batch = container_of(head, struct mmu_table_batch, rcu); for (i = 0; i < batch->nr; i++) __tlb_remove_table(batch->tables[i]); free_page((unsigned long)batch); } void tlb_table_flush(struct mmu_gather *tlb) { struct mmu_table_batch **batch = &tlb->batch; if (*batch) { call_rcu_sched(&(*batch)->rcu, tlb_remove_table_rcu); *batch = NULL; } } void tlb_remove_table(struct mmu_gather *tlb, void *table) { struct mmu_table_batch **batch = &tlb->batch; tlb->need_flush = 1; /* * When there's less then two users of this mm there cannot be a * concurrent page-table walk. */ if (atomic_read(&tlb->mm->mm_users) < 2) { __tlb_remove_table(table); return; } if (*batch == NULL) { *batch = (struct mmu_table_batch *)__get_free_page(GFP_NOWAIT | __GFP_NOWARN); if (*batch == NULL) { tlb_remove_table_one(table); return; } (*batch)->nr = 0; } (*batch)->tables[(*batch)->nr++] = table; if ((*batch)->nr == MAX_TABLE_BATCH) tlb_table_flush(tlb); } |
9547d01bf
|
383 |
#endif /* CONFIG_HAVE_RCU_TABLE_FREE */ |
267239116
|
384 |
|
1da177e4c
|
385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 |
/* * If a p?d_bad entry is found while walking page tables, report * the error, before resetting entry to p?d_none. Usually (but * very seldom) called out from the p?d_none_or_clear_bad macros. */ void pgd_clear_bad(pgd_t *pgd) { pgd_ERROR(*pgd); pgd_clear(pgd); } void pud_clear_bad(pud_t *pud) { pud_ERROR(*pud); pud_clear(pud); } void pmd_clear_bad(pmd_t *pmd) { pmd_ERROR(*pmd); pmd_clear(pmd); } /* * Note: this doesn't free the actual pages themselves. That * has been handled earlier when unmapping all the memory regions. */ |
9e1b32caa
|
413 414 |
static void free_pte_range(struct mmu_gather *tlb, pmd_t *pmd, unsigned long addr) |
1da177e4c
|
415 |
{ |
2f569afd9
|
416 |
pgtable_t token = pmd_pgtable(*pmd); |
e0da382c9
|
417 |
pmd_clear(pmd); |
9e1b32caa
|
418 |
pte_free_tlb(tlb, token, addr); |
e0da382c9
|
419 |
tlb->mm->nr_ptes--; |
1da177e4c
|
420 |
} |
e0da382c9
|
421 422 423 |
static inline void free_pmd_range(struct mmu_gather *tlb, pud_t *pud, unsigned long addr, unsigned long end, unsigned long floor, unsigned long ceiling) |
1da177e4c
|
424 425 426 |
{ pmd_t *pmd; unsigned long next; |
e0da382c9
|
427 |
unsigned long start; |
1da177e4c
|
428 |
|
e0da382c9
|
429 |
start = addr; |
1da177e4c
|
430 |
pmd = pmd_offset(pud, addr); |
1da177e4c
|
431 432 433 434 |
do { next = pmd_addr_end(addr, end); if (pmd_none_or_clear_bad(pmd)) continue; |
9e1b32caa
|
435 |
free_pte_range(tlb, pmd, addr); |
1da177e4c
|
436 |
} while (pmd++, addr = next, addr != end); |
e0da382c9
|
437 438 439 440 441 442 443 |
start &= PUD_MASK; if (start < floor) return; if (ceiling) { ceiling &= PUD_MASK; if (!ceiling) return; |
1da177e4c
|
444 |
} |
e0da382c9
|
445 446 447 448 449 |
if (end - 1 > ceiling - 1) return; pmd = pmd_offset(pud, start); pud_clear(pud); |
9e1b32caa
|
450 |
pmd_free_tlb(tlb, pmd, start); |
1da177e4c
|
451 |
} |
e0da382c9
|
452 453 454 |
static inline void free_pud_range(struct mmu_gather *tlb, pgd_t *pgd, unsigned long addr, unsigned long end, unsigned long floor, unsigned long ceiling) |
1da177e4c
|
455 456 457 |
{ pud_t *pud; unsigned long next; |
e0da382c9
|
458 |
unsigned long start; |
1da177e4c
|
459 |
|
e0da382c9
|
460 |
start = addr; |
1da177e4c
|
461 |
pud = pud_offset(pgd, addr); |
1da177e4c
|
462 463 464 465 |
do { next = pud_addr_end(addr, end); if (pud_none_or_clear_bad(pud)) continue; |
e0da382c9
|
466 |
free_pmd_range(tlb, pud, addr, next, floor, ceiling); |
1da177e4c
|
467 |
} while (pud++, addr = next, addr != end); |
e0da382c9
|
468 469 470 471 472 473 474 |
start &= PGDIR_MASK; if (start < floor) return; if (ceiling) { ceiling &= PGDIR_MASK; if (!ceiling) return; |
1da177e4c
|
475 |
} |
e0da382c9
|
476 477 478 479 480 |
if (end - 1 > ceiling - 1) return; pud = pud_offset(pgd, start); pgd_clear(pgd); |
9e1b32caa
|
481 |
pud_free_tlb(tlb, pud, start); |
1da177e4c
|
482 483 484 |
} /* |
e0da382c9
|
485 486 |
* This function frees user-level page tables of a process. * |
1da177e4c
|
487 488 |
* Must be called with pagetable lock held. */ |
42b777281
|
489 |
void free_pgd_range(struct mmu_gather *tlb, |
e0da382c9
|
490 491 |
unsigned long addr, unsigned long end, unsigned long floor, unsigned long ceiling) |
1da177e4c
|
492 493 494 |
{ pgd_t *pgd; unsigned long next; |
e0da382c9
|
495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 |
/* * The next few lines have given us lots of grief... * * Why are we testing PMD* at this top level? Because often * there will be no work to do at all, and we'd prefer not to * go all the way down to the bottom just to discover that. * * Why all these "- 1"s? Because 0 represents both the bottom * of the address space and the top of it (using -1 for the * top wouldn't help much: the masks would do the wrong thing). * The rule is that addr 0 and floor 0 refer to the bottom of * the address space, but end 0 and ceiling 0 refer to the top * Comparisons need to use "end - 1" and "ceiling - 1" (though * that end 0 case should be mythical). * * Wherever addr is brought up or ceiling brought down, we must * be careful to reject "the opposite 0" before it confuses the * subsequent tests. But what about where end is brought down * by PMD_SIZE below? no, end can't go down to 0 there. * * Whereas we round start (addr) and ceiling down, by different * masks at different levels, in order to test whether a table * now has no other vmas using it, so can be freed, we don't * bother to round floor or end up - the tests don't need that. */ |
1da177e4c
|
521 |
|
e0da382c9
|
522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 |
addr &= PMD_MASK; if (addr < floor) { addr += PMD_SIZE; if (!addr) return; } if (ceiling) { ceiling &= PMD_MASK; if (!ceiling) return; } if (end - 1 > ceiling - 1) end -= PMD_SIZE; if (addr > end - 1) return; |
42b777281
|
537 |
pgd = pgd_offset(tlb->mm, addr); |
1da177e4c
|
538 539 540 541 |
do { next = pgd_addr_end(addr, end); if (pgd_none_or_clear_bad(pgd)) continue; |
42b777281
|
542 |
free_pud_range(tlb, pgd, addr, next, floor, ceiling); |
1da177e4c
|
543 |
} while (pgd++, addr = next, addr != end); |
e0da382c9
|
544 |
} |
42b777281
|
545 |
void free_pgtables(struct mmu_gather *tlb, struct vm_area_struct *vma, |
3bf5ee956
|
546 |
unsigned long floor, unsigned long ceiling) |
e0da382c9
|
547 548 549 550 |
{ while (vma) { struct vm_area_struct *next = vma->vm_next; unsigned long addr = vma->vm_start; |
8f4f8c164
|
551 |
/* |
25d9e2d15
|
552 553 |
* Hide vma from rmap and truncate_pagecache before freeing * pgtables |
8f4f8c164
|
554 |
*/ |
5beb49305
|
555 |
unlink_anon_vmas(vma); |
8f4f8c164
|
556 |
unlink_file_vma(vma); |
9da61aef0
|
557 |
if (is_vm_hugetlb_page(vma)) { |
3bf5ee956
|
558 |
hugetlb_free_pgd_range(tlb, addr, vma->vm_end, |
e0da382c9
|
559 |
floor, next? next->vm_start: ceiling); |
3bf5ee956
|
560 561 562 563 564 |
} else { /* * Optimization: gather nearby vmas into one call down */ while (next && next->vm_start <= vma->vm_end + PMD_SIZE |
4866920b9
|
565 |
&& !is_vm_hugetlb_page(next)) { |
3bf5ee956
|
566 567 |
vma = next; next = vma->vm_next; |
5beb49305
|
568 |
unlink_anon_vmas(vma); |
8f4f8c164
|
569 |
unlink_file_vma(vma); |
3bf5ee956
|
570 571 572 573 |
} free_pgd_range(tlb, addr, vma->vm_end, floor, next? next->vm_start: ceiling); } |
e0da382c9
|
574 575 |
vma = next; } |
1da177e4c
|
576 |
} |
8ac1f8320
|
577 578 |
int __pte_alloc(struct mm_struct *mm, struct vm_area_struct *vma, pmd_t *pmd, unsigned long address) |
1da177e4c
|
579 |
{ |
2f569afd9
|
580 |
pgtable_t new = pte_alloc_one(mm, address); |
8ac1f8320
|
581 |
int wait_split_huge_page; |
1bb3630e8
|
582 583 |
if (!new) return -ENOMEM; |
362a61ad6
|
584 585 586 587 588 589 590 591 592 593 594 595 596 597 |
/* * Ensure all pte setup (eg. pte page lock and page clearing) are * visible before the pte is made visible to other CPUs by being * put into page tables. * * The other side of the story is the pointer chasing in the page * table walking code (when walking the page table without locking; * ie. most of the time). Fortunately, these data accesses consist * of a chain of data-dependent loads, meaning most CPUs (alpha * being the notable exception) will already guarantee loads are * seen in-order. See the alpha page table accessors for the * smp_read_barrier_depends() barriers in page table walking code. */ smp_wmb(); /* Could be smp_wmb__xxx(before|after)_spin_lock */ |
c74df32c7
|
598 |
spin_lock(&mm->page_table_lock); |
8ac1f8320
|
599 600 |
wait_split_huge_page = 0; if (likely(pmd_none(*pmd))) { /* Has another populated it ? */ |
1da177e4c
|
601 |
mm->nr_ptes++; |
1da177e4c
|
602 |
pmd_populate(mm, pmd, new); |
2f569afd9
|
603 |
new = NULL; |
8ac1f8320
|
604 605 |
} else if (unlikely(pmd_trans_splitting(*pmd))) wait_split_huge_page = 1; |
c74df32c7
|
606 |
spin_unlock(&mm->page_table_lock); |
2f569afd9
|
607 608 |
if (new) pte_free(mm, new); |
8ac1f8320
|
609 610 |
if (wait_split_huge_page) wait_split_huge_page(vma->anon_vma, pmd); |
1bb3630e8
|
611 |
return 0; |
1da177e4c
|
612 |
} |
1bb3630e8
|
613 |
int __pte_alloc_kernel(pmd_t *pmd, unsigned long address) |
1da177e4c
|
614 |
{ |
1bb3630e8
|
615 616 617 |
pte_t *new = pte_alloc_one_kernel(&init_mm, address); if (!new) return -ENOMEM; |
362a61ad6
|
618 |
smp_wmb(); /* See comment in __pte_alloc */ |
1bb3630e8
|
619 |
spin_lock(&init_mm.page_table_lock); |
8ac1f8320
|
620 |
if (likely(pmd_none(*pmd))) { /* Has another populated it ? */ |
1bb3630e8
|
621 |
pmd_populate_kernel(&init_mm, pmd, new); |
2f569afd9
|
622 |
new = NULL; |
8ac1f8320
|
623 624 |
} else VM_BUG_ON(pmd_trans_splitting(*pmd)); |
1bb3630e8
|
625 |
spin_unlock(&init_mm.page_table_lock); |
2f569afd9
|
626 627 |
if (new) pte_free_kernel(&init_mm, new); |
1bb3630e8
|
628 |
return 0; |
1da177e4c
|
629 |
} |
d559db086
|
630 631 632 633 634 635 |
static inline void init_rss_vec(int *rss) { memset(rss, 0, sizeof(int) * NR_MM_COUNTERS); } static inline void add_mm_rss_vec(struct mm_struct *mm, int *rss) |
ae8597623
|
636 |
{ |
d559db086
|
637 |
int i; |
34e55232e
|
638 639 |
if (current->mm == mm) sync_mm_rss(current, mm); |
d559db086
|
640 641 642 |
for (i = 0; i < NR_MM_COUNTERS; i++) if (rss[i]) add_mm_counter(mm, i, rss[i]); |
ae8597623
|
643 |
} |
1da177e4c
|
644 |
/* |
6aab341e0
|
645 646 647 |
* This function is called to print an error when a bad pte * is found. For example, we might have a PFN-mapped pte in * a region that doesn't allow it. |
b5810039a
|
648 649 650 |
* * The calling function must still handle the error. */ |
3dc147414
|
651 652 |
static void print_bad_pte(struct vm_area_struct *vma, unsigned long addr, pte_t pte, struct page *page) |
b5810039a
|
653 |
{ |
3dc147414
|
654 655 656 657 658 |
pgd_t *pgd = pgd_offset(vma->vm_mm, addr); pud_t *pud = pud_offset(pgd, addr); pmd_t *pmd = pmd_offset(pud, addr); struct address_space *mapping; pgoff_t index; |
d936cf9b3
|
659 660 661 662 663 664 665 666 667 668 669 670 671 672 |
static unsigned long resume; static unsigned long nr_shown; static unsigned long nr_unshown; /* * Allow a burst of 60 reports, then keep quiet for that minute; * or allow a steady drip of one report per second. */ if (nr_shown == 60) { if (time_before(jiffies, resume)) { nr_unshown++; return; } if (nr_unshown) { |
1e9e63650
|
673 674 675 |
printk(KERN_ALERT "BUG: Bad page map: %lu messages suppressed ", |
d936cf9b3
|
676 677 678 679 680 681 682 |
nr_unshown); nr_unshown = 0; } nr_shown = 0; } if (nr_shown++ == 0) resume = jiffies + 60 * HZ; |
3dc147414
|
683 684 685 |
mapping = vma->vm_file ? vma->vm_file->f_mapping : NULL; index = linear_page_index(vma, addr); |
1e9e63650
|
686 687 688 |
printk(KERN_ALERT "BUG: Bad page map in process %s pte:%08llx pmd:%08llx ", |
3dc147414
|
689 690 |
current->comm, (long long)pte_val(pte), (long long)pmd_val(*pmd)); |
718a38211
|
691 692 |
if (page) dump_page(page); |
1e9e63650
|
693 |
printk(KERN_ALERT |
3dc147414
|
694 695 696 697 698 699 700 |
"addr:%p vm_flags:%08lx anon_vma:%p mapping:%p index:%lx ", (void *)addr, vma->vm_flags, vma->anon_vma, mapping, index); /* * Choose text because data symbols depend on CONFIG_KALLSYMS_ALL=y */ if (vma->vm_ops) |
1e9e63650
|
701 702 |
print_symbol(KERN_ALERT "vma->vm_ops->fault: %s ", |
3dc147414
|
703 704 |
(unsigned long)vma->vm_ops->fault); if (vma->vm_file && vma->vm_file->f_op) |
1e9e63650
|
705 706 |
print_symbol(KERN_ALERT "vma->vm_file->f_op->mmap: %s ", |
3dc147414
|
707 |
(unsigned long)vma->vm_file->f_op->mmap); |
b5810039a
|
708 |
dump_stack(); |
3dc147414
|
709 |
add_taint(TAINT_BAD_PAGE); |
b5810039a
|
710 |
} |
ca16d140a
|
711 |
static inline int is_cow_mapping(vm_flags_t flags) |
67121172f
|
712 713 714 |
{ return (flags & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE; } |
62eede62d
|
715 716 717 718 719 720 721 722 723 724 725 726 727 |
#ifndef is_zero_pfn static inline int is_zero_pfn(unsigned long pfn) { return pfn == zero_pfn; } #endif #ifndef my_zero_pfn static inline unsigned long my_zero_pfn(unsigned long addr) { return zero_pfn; } #endif |
b5810039a
|
728 |
/* |
7e675137a
|
729 |
* vm_normal_page -- This function gets the "struct page" associated with a pte. |
6aab341e0
|
730 |
* |
7e675137a
|
731 732 733 |
* "Special" mappings do not wish to be associated with a "struct page" (either * it doesn't exist, or it exists but they don't want to touch it). In this * case, NULL is returned here. "Normal" mappings do have a struct page. |
b379d7901
|
734 |
* |
7e675137a
|
735 736 737 738 739 740 741 742 |
* There are 2 broad cases. Firstly, an architecture may define a pte_special() * pte bit, in which case this function is trivial. Secondly, an architecture * may not have a spare pte bit, which requires a more complicated scheme, * described below. * * A raw VM_PFNMAP mapping (ie. one that is not COWed) is always considered a * special mapping (even if there are underlying and valid "struct pages"). * COWed pages of a VM_PFNMAP are always normal. |
6aab341e0
|
743 |
* |
b379d7901
|
744 745 |
* The way we recognize COWed pages within VM_PFNMAP mappings is through the * rules set up by "remap_pfn_range()": the vma will have the VM_PFNMAP bit |
7e675137a
|
746 747 |
* set, and the vm_pgoff will point to the first PFN mapped: thus every special * mapping will always honor the rule |
6aab341e0
|
748 749 750 |
* * pfn_of_page == vma->vm_pgoff + ((addr - vma->vm_start) >> PAGE_SHIFT) * |
7e675137a
|
751 752 753 754 755 756 |
* And for normal mappings this is false. * * This restricts such mappings to be a linear translation from virtual address * to pfn. To get around this restriction, we allow arbitrary mappings so long * as the vma is not a COW mapping; in that case, we know that all ptes are * special (because none can have been COWed). |
b379d7901
|
757 |
* |
b379d7901
|
758 |
* |
7e675137a
|
759 |
* In order to support COW of arbitrary special mappings, we have VM_MIXEDMAP. |
b379d7901
|
760 761 762 763 764 765 766 767 768 |
* * VM_MIXEDMAP mappings can likewise contain memory with or without "struct * page" backing, however the difference is that _all_ pages with a struct * page (that is, those where pfn_valid is true) are refcounted and considered * normal pages by the VM. The disadvantage is that pages are refcounted * (which can be slower and simply not an option for some PFNMAP users). The * advantage is that we don't have to follow the strict linearity rule of * PFNMAP mappings in order to support COWable mappings. * |
ee498ed73
|
769 |
*/ |
7e675137a
|
770 771 772 773 774 775 776 |
#ifdef __HAVE_ARCH_PTE_SPECIAL # define HAVE_PTE_SPECIAL 1 #else # define HAVE_PTE_SPECIAL 0 #endif struct page *vm_normal_page(struct vm_area_struct *vma, unsigned long addr, pte_t pte) |
ee498ed73
|
777 |
{ |
22b31eec6
|
778 |
unsigned long pfn = pte_pfn(pte); |
7e675137a
|
779 780 |
if (HAVE_PTE_SPECIAL) { |
22b31eec6
|
781 782 |
if (likely(!pte_special(pte))) goto check_pfn; |
a13ea5b75
|
783 784 |
if (vma->vm_flags & (VM_PFNMAP | VM_MIXEDMAP)) return NULL; |
62eede62d
|
785 |
if (!is_zero_pfn(pfn)) |
22b31eec6
|
786 |
print_bad_pte(vma, addr, pte, NULL); |
7e675137a
|
787 788 789 790 |
return NULL; } /* !HAVE_PTE_SPECIAL case follows: */ |
b379d7901
|
791 792 793 794 795 796 |
if (unlikely(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP))) { if (vma->vm_flags & VM_MIXEDMAP) { if (!pfn_valid(pfn)) return NULL; goto out; } else { |
7e675137a
|
797 798 |
unsigned long off; off = (addr - vma->vm_start) >> PAGE_SHIFT; |
b379d7901
|
799 800 801 802 803 |
if (pfn == vma->vm_pgoff + off) return NULL; if (!is_cow_mapping(vma->vm_flags)) return NULL; } |
6aab341e0
|
804 |
} |
62eede62d
|
805 806 |
if (is_zero_pfn(pfn)) return NULL; |
22b31eec6
|
807 808 809 810 811 |
check_pfn: if (unlikely(pfn > highest_memmap_pfn)) { print_bad_pte(vma, addr, pte, NULL); return NULL; } |
6aab341e0
|
812 813 |
/* |
7e675137a
|
814 |
* NOTE! We still have PageReserved() pages in the page tables. |
7e675137a
|
815 |
* eg. VDSO mappings can cause them to exist. |
6aab341e0
|
816 |
*/ |
b379d7901
|
817 |
out: |
6aab341e0
|
818 |
return pfn_to_page(pfn); |
ee498ed73
|
819 820 821 |
} /* |
1da177e4c
|
822 823 824 |
* copy one vm_area from one task to the other. Assumes the page tables * already present in the new task to be cleared in the whole range * covered by this vma. |
1da177e4c
|
825 |
*/ |
570a335b8
|
826 |
static inline unsigned long |
1da177e4c
|
827 |
copy_one_pte(struct mm_struct *dst_mm, struct mm_struct *src_mm, |
b5810039a
|
828 |
pte_t *dst_pte, pte_t *src_pte, struct vm_area_struct *vma, |
8c1037627
|
829 |
unsigned long addr, int *rss) |
1da177e4c
|
830 |
{ |
b5810039a
|
831 |
unsigned long vm_flags = vma->vm_flags; |
1da177e4c
|
832 833 |
pte_t pte = *src_pte; struct page *page; |
1da177e4c
|
834 835 836 837 |
/* pte contains position in swap or file, so copy. */ if (unlikely(!pte_present(pte))) { if (!pte_file(pte)) { |
0697212a4
|
838 |
swp_entry_t entry = pte_to_swp_entry(pte); |
570a335b8
|
839 840 |
if (swap_duplicate(entry) < 0) return entry.val; |
1da177e4c
|
841 842 843 |
/* make sure dst_mm is on swapoff's mmlist. */ if (unlikely(list_empty(&dst_mm->mmlist))) { spin_lock(&mmlist_lock); |
f412ac08c
|
844 845 846 |
if (list_empty(&dst_mm->mmlist)) list_add(&dst_mm->mmlist, &src_mm->mmlist); |
1da177e4c
|
847 848 |
spin_unlock(&mmlist_lock); } |
b084d4353
|
849 850 851 |
if (likely(!non_swap_entry(entry))) rss[MM_SWAPENTS]++; else if (is_write_migration_entry(entry) && |
0697212a4
|
852 853 854 855 856 857 858 859 860 |
is_cow_mapping(vm_flags)) { /* * COW mappings require pages in both parent * and child to be set to read. */ make_migration_entry_read(&entry); pte = swp_entry_to_pte(entry); set_pte_at(src_mm, addr, src_pte, pte); } |
1da177e4c
|
861 |
} |
ae8597623
|
862 |
goto out_set_pte; |
1da177e4c
|
863 |
} |
1da177e4c
|
864 865 866 867 |
/* * If it's a COW mapping, write protect it both * in the parent and the child */ |
67121172f
|
868 |
if (is_cow_mapping(vm_flags)) { |
1da177e4c
|
869 |
ptep_set_wrprotect(src_mm, addr, src_pte); |
3dc907951
|
870 |
pte = pte_wrprotect(pte); |
1da177e4c
|
871 872 873 874 875 876 877 878 879 |
} /* * If it's a shared mapping, mark it clean in * the child */ if (vm_flags & VM_SHARED) pte = pte_mkclean(pte); pte = pte_mkold(pte); |
6aab341e0
|
880 881 882 883 |
page = vm_normal_page(vma, addr, pte); if (page) { get_page(page); |
21333b2b6
|
884 |
page_dup_rmap(page); |
d559db086
|
885 886 887 888 |
if (PageAnon(page)) rss[MM_ANONPAGES]++; else rss[MM_FILEPAGES]++; |
6aab341e0
|
889 |
} |
ae8597623
|
890 891 892 |
out_set_pte: set_pte_at(dst_mm, addr, dst_pte, pte); |
570a335b8
|
893 |
return 0; |
1da177e4c
|
894 |
} |
71e3aac07
|
895 896 897 |
int copy_pte_range(struct mm_struct *dst_mm, struct mm_struct *src_mm, pmd_t *dst_pmd, pmd_t *src_pmd, struct vm_area_struct *vma, unsigned long addr, unsigned long end) |
1da177e4c
|
898 |
{ |
c36987e2e
|
899 |
pte_t *orig_src_pte, *orig_dst_pte; |
1da177e4c
|
900 |
pte_t *src_pte, *dst_pte; |
c74df32c7
|
901 |
spinlock_t *src_ptl, *dst_ptl; |
e040f218b
|
902 |
int progress = 0; |
d559db086
|
903 |
int rss[NR_MM_COUNTERS]; |
570a335b8
|
904 |
swp_entry_t entry = (swp_entry_t){0}; |
1da177e4c
|
905 906 |
again: |
d559db086
|
907 |
init_rss_vec(rss); |
c74df32c7
|
908 |
dst_pte = pte_alloc_map_lock(dst_mm, dst_pmd, addr, &dst_ptl); |
1da177e4c
|
909 910 |
if (!dst_pte) return -ENOMEM; |
ece0e2b64
|
911 |
src_pte = pte_offset_map(src_pmd, addr); |
4c21e2f24
|
912 |
src_ptl = pte_lockptr(src_mm, src_pmd); |
f20dc5f7c
|
913 |
spin_lock_nested(src_ptl, SINGLE_DEPTH_NESTING); |
c36987e2e
|
914 915 |
orig_src_pte = src_pte; orig_dst_pte = dst_pte; |
6606c3e0d
|
916 |
arch_enter_lazy_mmu_mode(); |
1da177e4c
|
917 |
|
1da177e4c
|
918 919 920 921 922 |
do { /* * We are holding two locks at this point - either of them * could generate latencies in another task on another CPU. */ |
e040f218b
|
923 924 925 |
if (progress >= 32) { progress = 0; if (need_resched() || |
95c354fe9
|
926 |
spin_needbreak(src_ptl) || spin_needbreak(dst_ptl)) |
e040f218b
|
927 928 |
break; } |
1da177e4c
|
929 930 931 932 |
if (pte_none(*src_pte)) { progress++; continue; } |
570a335b8
|
933 934 935 936 |
entry.val = copy_one_pte(dst_mm, src_mm, dst_pte, src_pte, vma, addr, rss); if (entry.val) break; |
1da177e4c
|
937 938 |
progress += 8; } while (dst_pte++, src_pte++, addr += PAGE_SIZE, addr != end); |
1da177e4c
|
939 |
|
6606c3e0d
|
940 |
arch_leave_lazy_mmu_mode(); |
c74df32c7
|
941 |
spin_unlock(src_ptl); |
ece0e2b64
|
942 |
pte_unmap(orig_src_pte); |
d559db086
|
943 |
add_mm_rss_vec(dst_mm, rss); |
c36987e2e
|
944 |
pte_unmap_unlock(orig_dst_pte, dst_ptl); |
c74df32c7
|
945 |
cond_resched(); |
570a335b8
|
946 947 948 949 950 951 |
if (entry.val) { if (add_swap_count_continuation(entry, GFP_KERNEL) < 0) return -ENOMEM; progress = 0; } |
1da177e4c
|
952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 |
if (addr != end) goto again; return 0; } static inline int copy_pmd_range(struct mm_struct *dst_mm, struct mm_struct *src_mm, pud_t *dst_pud, pud_t *src_pud, struct vm_area_struct *vma, unsigned long addr, unsigned long end) { pmd_t *src_pmd, *dst_pmd; unsigned long next; dst_pmd = pmd_alloc(dst_mm, dst_pud, addr); if (!dst_pmd) return -ENOMEM; src_pmd = pmd_offset(src_pud, addr); do { next = pmd_addr_end(addr, end); |
71e3aac07
|
970 971 |
if (pmd_trans_huge(*src_pmd)) { int err; |
14d1a55cd
|
972 |
VM_BUG_ON(next-addr != HPAGE_PMD_SIZE); |
71e3aac07
|
973 974 975 976 977 978 979 980 |
err = copy_huge_pmd(dst_mm, src_mm, dst_pmd, src_pmd, addr, vma); if (err == -ENOMEM) return -ENOMEM; if (!err) continue; /* fall through */ } |
1da177e4c
|
981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 |
if (pmd_none_or_clear_bad(src_pmd)) continue; if (copy_pte_range(dst_mm, src_mm, dst_pmd, src_pmd, vma, addr, next)) return -ENOMEM; } while (dst_pmd++, src_pmd++, addr = next, addr != end); return 0; } static inline int copy_pud_range(struct mm_struct *dst_mm, struct mm_struct *src_mm, pgd_t *dst_pgd, pgd_t *src_pgd, struct vm_area_struct *vma, unsigned long addr, unsigned long end) { pud_t *src_pud, *dst_pud; unsigned long next; dst_pud = pud_alloc(dst_mm, dst_pgd, addr); if (!dst_pud) return -ENOMEM; src_pud = pud_offset(src_pgd, addr); do { next = pud_addr_end(addr, end); if (pud_none_or_clear_bad(src_pud)) continue; if (copy_pmd_range(dst_mm, src_mm, dst_pud, src_pud, vma, addr, next)) return -ENOMEM; } while (dst_pud++, src_pud++, addr = next, addr != end); return 0; } int copy_page_range(struct mm_struct *dst_mm, struct mm_struct *src_mm, struct vm_area_struct *vma) { pgd_t *src_pgd, *dst_pgd; unsigned long next; unsigned long addr = vma->vm_start; unsigned long end = vma->vm_end; |
cddb8a5c1
|
1019 |
int ret; |
1da177e4c
|
1020 |
|
d992895ba
|
1021 1022 1023 1024 1025 1026 |
/* * Don't copy ptes where a page fault will fill them correctly. * Fork becomes much lighter when there are big shared or private * readonly mappings. The tradeoff is that copy_page_range is more * efficient than faulting. */ |
4d7672b46
|
1027 |
if (!(vma->vm_flags & (VM_HUGETLB|VM_NONLINEAR|VM_PFNMAP|VM_INSERTPAGE))) { |
d992895ba
|
1028 1029 1030 |
if (!vma->anon_vma) return 0; } |
1da177e4c
|
1031 1032 |
if (is_vm_hugetlb_page(vma)) return copy_hugetlb_page_range(dst_mm, src_mm, vma); |
34801ba9b
|
1033 |
if (unlikely(is_pfn_mapping(vma))) { |
2ab640379
|
1034 1035 1036 1037 1038 1039 1040 1041 |
/* * We do not free on error cases below as remove_vma * gets called on error from higher level routine */ ret = track_pfn_vma_copy(vma); if (ret) return ret; } |
cddb8a5c1
|
1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 |
/* * We need to invalidate the secondary MMU mappings only when * there could be a permission downgrade on the ptes of the * parent mm. And a permission downgrade will only happen if * is_cow_mapping() returns true. */ if (is_cow_mapping(vma->vm_flags)) mmu_notifier_invalidate_range_start(src_mm, addr, end); ret = 0; |
1da177e4c
|
1052 1053 1054 1055 1056 1057 |
dst_pgd = pgd_offset(dst_mm, addr); src_pgd = pgd_offset(src_mm, addr); do { next = pgd_addr_end(addr, end); if (pgd_none_or_clear_bad(src_pgd)) continue; |
cddb8a5c1
|
1058 1059 1060 1061 1062 |
if (unlikely(copy_pud_range(dst_mm, src_mm, dst_pgd, src_pgd, vma, addr, next))) { ret = -ENOMEM; break; } |
1da177e4c
|
1063 |
} while (dst_pgd++, src_pgd++, addr = next, addr != end); |
cddb8a5c1
|
1064 1065 1066 1067 1068 |
if (is_cow_mapping(vma->vm_flags)) mmu_notifier_invalidate_range_end(src_mm, vma->vm_start, end); return ret; |
1da177e4c
|
1069 |
} |
51c6f666f
|
1070 |
static unsigned long zap_pte_range(struct mmu_gather *tlb, |
b5810039a
|
1071 |
struct vm_area_struct *vma, pmd_t *pmd, |
1da177e4c
|
1072 |
unsigned long addr, unsigned long end, |
97a894136
|
1073 |
struct zap_details *details) |
1da177e4c
|
1074 |
{ |
b5810039a
|
1075 |
struct mm_struct *mm = tlb->mm; |
d16dfc550
|
1076 |
int force_flush = 0; |
d559db086
|
1077 |
int rss[NR_MM_COUNTERS]; |
97a894136
|
1078 |
spinlock_t *ptl; |
5f1a19070
|
1079 |
pte_t *start_pte; |
97a894136
|
1080 |
pte_t *pte; |
d559db086
|
1081 |
|
d16dfc550
|
1082 |
again: |
e303297e6
|
1083 |
init_rss_vec(rss); |
5f1a19070
|
1084 1085 |
start_pte = pte_offset_map_lock(mm, pmd, addr, &ptl); pte = start_pte; |
6606c3e0d
|
1086 |
arch_enter_lazy_mmu_mode(); |
1da177e4c
|
1087 1088 |
do { pte_t ptent = *pte; |
51c6f666f
|
1089 |
if (pte_none(ptent)) { |
1da177e4c
|
1090 |
continue; |
51c6f666f
|
1091 |
} |
6f5e6b9e6
|
1092 |
|
1da177e4c
|
1093 |
if (pte_present(ptent)) { |
ee498ed73
|
1094 |
struct page *page; |
51c6f666f
|
1095 |
|
6aab341e0
|
1096 |
page = vm_normal_page(vma, addr, ptent); |
1da177e4c
|
1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 |
if (unlikely(details) && page) { /* * unmap_shared_mapping_pages() wants to * invalidate cache without truncating: * unmap shared but keep private pages. */ if (details->check_mapping && details->check_mapping != page->mapping) continue; /* * Each page->index must be checked when * invalidating or truncating nonlinear. */ if (details->nonlinear_vma && (page->index < details->first_index || page->index > details->last_index)) continue; } |
b5810039a
|
1115 |
ptent = ptep_get_and_clear_full(mm, addr, pte, |
a600388d2
|
1116 |
tlb->fullmm); |
1da177e4c
|
1117 1118 1119 1120 1121 1122 |
tlb_remove_tlb_entry(tlb, pte, addr); if (unlikely(!page)) continue; if (unlikely(details) && details->nonlinear_vma && linear_page_index(details->nonlinear_vma, addr) != page->index) |
b5810039a
|
1123 |
set_pte_at(mm, addr, pte, |
1da177e4c
|
1124 |
pgoff_to_pte(page->index)); |
1da177e4c
|
1125 |
if (PageAnon(page)) |
d559db086
|
1126 |
rss[MM_ANONPAGES]--; |
6237bcd94
|
1127 1128 1129 |
else { if (pte_dirty(ptent)) set_page_dirty(page); |
4917e5d04
|
1130 1131 |
if (pte_young(ptent) && likely(!VM_SequentialReadHint(vma))) |
bf3f3bc5e
|
1132 |
mark_page_accessed(page); |
d559db086
|
1133 |
rss[MM_FILEPAGES]--; |
6237bcd94
|
1134 |
} |
edc315fd2
|
1135 |
page_remove_rmap(page); |
3dc147414
|
1136 1137 |
if (unlikely(page_mapcount(page) < 0)) print_bad_pte(vma, addr, ptent, page); |
d16dfc550
|
1138 1139 1140 |
force_flush = !__tlb_remove_page(tlb, page); if (force_flush) break; |
1da177e4c
|
1141 1142 1143 1144 1145 1146 1147 1148 |
continue; } /* * If details->check_mapping, we leave swap entries; * if details->nonlinear_vma, we leave file entries. */ if (unlikely(details)) continue; |
2509ef26d
|
1149 1150 1151 |
if (pte_file(ptent)) { if (unlikely(!(vma->vm_flags & VM_NONLINEAR))) print_bad_pte(vma, addr, ptent, NULL); |
b084d4353
|
1152 1153 1154 1155 1156 1157 1158 1159 |
} else { swp_entry_t entry = pte_to_swp_entry(ptent); if (!non_swap_entry(entry)) rss[MM_SWAPENTS]--; if (unlikely(!free_swap_and_cache(entry))) print_bad_pte(vma, addr, ptent, NULL); } |
9888a1cae
|
1160 |
pte_clear_not_present_full(mm, addr, pte, tlb->fullmm); |
97a894136
|
1161 |
} while (pte++, addr += PAGE_SIZE, addr != end); |
ae8597623
|
1162 |
|
d559db086
|
1163 |
add_mm_rss_vec(mm, rss); |
6606c3e0d
|
1164 |
arch_leave_lazy_mmu_mode(); |
5f1a19070
|
1165 |
pte_unmap_unlock(start_pte, ptl); |
51c6f666f
|
1166 |
|
d16dfc550
|
1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 |
/* * mmu_gather ran out of room to batch pages, we break out of * the PTE lock to avoid doing the potential expensive TLB invalidate * and page-free while holding it. */ if (force_flush) { force_flush = 0; tlb_flush_mmu(tlb); if (addr != end) goto again; } |
51c6f666f
|
1178 |
return addr; |
1da177e4c
|
1179 |
} |
51c6f666f
|
1180 |
static inline unsigned long zap_pmd_range(struct mmu_gather *tlb, |
b5810039a
|
1181 |
struct vm_area_struct *vma, pud_t *pud, |
1da177e4c
|
1182 |
unsigned long addr, unsigned long end, |
97a894136
|
1183 |
struct zap_details *details) |
1da177e4c
|
1184 1185 1186 1187 1188 1189 1190 |
{ pmd_t *pmd; unsigned long next; pmd = pmd_offset(pud, addr); do { next = pmd_addr_end(addr, end); |
71e3aac07
|
1191 |
if (pmd_trans_huge(*pmd)) { |
14d1a55cd
|
1192 1193 |
if (next-addr != HPAGE_PMD_SIZE) { VM_BUG_ON(!rwsem_is_locked(&tlb->mm->mmap_sem)); |
71e3aac07
|
1194 |
split_huge_page_pmd(vma->vm_mm, pmd); |
f21760b15
|
1195 |
} else if (zap_huge_pmd(tlb, vma, pmd, addr)) |
71e3aac07
|
1196 |
continue; |
71e3aac07
|
1197 1198 |
/* fall through */ } |
97a894136
|
1199 |
if (pmd_none_or_clear_bad(pmd)) |
1da177e4c
|
1200 |
continue; |
97a894136
|
1201 1202 1203 |
next = zap_pte_range(tlb, vma, pmd, addr, next, details); cond_resched(); } while (pmd++, addr = next, addr != end); |
51c6f666f
|
1204 1205 |
return addr; |
1da177e4c
|
1206 |
} |
51c6f666f
|
1207 |
static inline unsigned long zap_pud_range(struct mmu_gather *tlb, |
b5810039a
|
1208 |
struct vm_area_struct *vma, pgd_t *pgd, |
1da177e4c
|
1209 |
unsigned long addr, unsigned long end, |
97a894136
|
1210 |
struct zap_details *details) |
1da177e4c
|
1211 1212 1213 1214 1215 1216 1217 |
{ pud_t *pud; unsigned long next; pud = pud_offset(pgd, addr); do { next = pud_addr_end(addr, end); |
97a894136
|
1218 |
if (pud_none_or_clear_bad(pud)) |
1da177e4c
|
1219 |
continue; |
97a894136
|
1220 1221 |
next = zap_pmd_range(tlb, vma, pud, addr, next, details); } while (pud++, addr = next, addr != end); |
51c6f666f
|
1222 1223 |
return addr; |
1da177e4c
|
1224 |
} |
51c6f666f
|
1225 1226 |
static unsigned long unmap_page_range(struct mmu_gather *tlb, struct vm_area_struct *vma, |
1da177e4c
|
1227 |
unsigned long addr, unsigned long end, |
97a894136
|
1228 |
struct zap_details *details) |
1da177e4c
|
1229 1230 1231 1232 1233 1234 1235 1236 |
{ pgd_t *pgd; unsigned long next; if (details && !details->check_mapping && !details->nonlinear_vma) details = NULL; BUG_ON(addr >= end); |
569b846df
|
1237 |
mem_cgroup_uncharge_start(); |
1da177e4c
|
1238 1239 1240 1241 |
tlb_start_vma(tlb, vma); pgd = pgd_offset(vma->vm_mm, addr); do { next = pgd_addr_end(addr, end); |
97a894136
|
1242 |
if (pgd_none_or_clear_bad(pgd)) |
1da177e4c
|
1243 |
continue; |
97a894136
|
1244 1245 |
next = zap_pud_range(tlb, vma, pgd, addr, next, details); } while (pgd++, addr = next, addr != end); |
1da177e4c
|
1246 |
tlb_end_vma(tlb, vma); |
569b846df
|
1247 |
mem_cgroup_uncharge_end(); |
51c6f666f
|
1248 1249 |
return addr; |
1da177e4c
|
1250 |
} |
1da177e4c
|
1251 1252 |
/** * unmap_vmas - unmap a range of memory covered by a list of vma's |
0164f69d0
|
1253 |
* @tlb: address of the caller's struct mmu_gather |
1da177e4c
|
1254 1255 1256 1257 1258 1259 |
* @vma: the starting vma * @start_addr: virtual address at which to start unmapping * @end_addr: virtual address at which to end unmapping * @nr_accounted: Place number of unmapped pages in vm-accountable vma's here * @details: details of nonlinear truncation or shared cache invalidation * |
ee39b37b2
|
1260 |
* Returns the end address of the unmapping (restart addr if interrupted). |
1da177e4c
|
1261 |
* |
508034a32
|
1262 |
* Unmap all pages in the vma list. |
1da177e4c
|
1263 |
* |
1da177e4c
|
1264 1265 1266 1267 1268 1269 1270 1271 1272 |
* Only addresses between `start' and `end' will be unmapped. * * The VMA list must be sorted in ascending virtual address order. * * unmap_vmas() assumes that the caller will flush the whole unmapped address * range after unmap_vmas() returns. So the only responsibility here is to * ensure that any thus-far unmapped pages are flushed before unmap_vmas() * drops the lock and schedules. */ |
d16dfc550
|
1273 |
unsigned long unmap_vmas(struct mmu_gather *tlb, |
1da177e4c
|
1274 1275 1276 1277 |
struct vm_area_struct *vma, unsigned long start_addr, unsigned long end_addr, unsigned long *nr_accounted, struct zap_details *details) { |
ee39b37b2
|
1278 |
unsigned long start = start_addr; |
cddb8a5c1
|
1279 |
struct mm_struct *mm = vma->vm_mm; |
1da177e4c
|
1280 |
|
cddb8a5c1
|
1281 |
mmu_notifier_invalidate_range_start(mm, start_addr, end_addr); |
1da177e4c
|
1282 |
for ( ; vma && vma->vm_start < end_addr; vma = vma->vm_next) { |
1da177e4c
|
1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 |
unsigned long end; start = max(vma->vm_start, start_addr); if (start >= vma->vm_end) continue; end = min(vma->vm_end, end_addr); if (end <= vma->vm_start) continue; if (vma->vm_flags & VM_ACCOUNT) *nr_accounted += (end - start) >> PAGE_SHIFT; |
34801ba9b
|
1294 |
if (unlikely(is_pfn_mapping(vma))) |
2ab640379
|
1295 |
untrack_pfn_vma(vma, 0, 0); |
1da177e4c
|
1296 |
while (start != end) { |
51c6f666f
|
1297 |
if (unlikely(is_vm_hugetlb_page(vma))) { |
a137e1cc6
|
1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 |
/* * It is undesirable to test vma->vm_file as it * should be non-null for valid hugetlb area. * However, vm_file will be NULL in the error * cleanup path of do_mmap_pgoff. When * hugetlbfs ->mmap method fails, * do_mmap_pgoff() nullifies vma->vm_file * before calling this function to clean up. * Since no pte has actually been setup, it is * safe to do nothing in this case. */ |
97a894136
|
1309 |
if (vma->vm_file) |
a137e1cc6
|
1310 |
unmap_hugepage_range(vma, start, end, NULL); |
a137e1cc6
|
1311 |
|
51c6f666f
|
1312 1313 |
start = end; } else |
97a894136
|
1314 |
start = unmap_page_range(tlb, vma, start, end, details); |
1da177e4c
|
1315 1316 |
} } |
97a894136
|
1317 |
|
cddb8a5c1
|
1318 |
mmu_notifier_invalidate_range_end(mm, start_addr, end_addr); |
ee39b37b2
|
1319 |
return start; /* which is now the end (or restart) address */ |
1da177e4c
|
1320 1321 1322 1323 1324 1325 1326 1327 1328 |
} /** * zap_page_range - remove user pages in a given range * @vma: vm_area_struct holding the applicable pages * @address: starting address of pages to zap * @size: number of bytes to zap * @details: details of nonlinear truncation or shared cache invalidation */ |
ee39b37b2
|
1329 |
unsigned long zap_page_range(struct vm_area_struct *vma, unsigned long address, |
1da177e4c
|
1330 1331 1332 |
unsigned long size, struct zap_details *details) { struct mm_struct *mm = vma->vm_mm; |
d16dfc550
|
1333 |
struct mmu_gather tlb; |
1da177e4c
|
1334 1335 |
unsigned long end = address + size; unsigned long nr_accounted = 0; |
1da177e4c
|
1336 |
lru_add_drain(); |
d16dfc550
|
1337 |
tlb_gather_mmu(&tlb, mm, 0); |
365e9c87a
|
1338 |
update_hiwater_rss(mm); |
508034a32
|
1339 |
end = unmap_vmas(&tlb, vma, address, end, &nr_accounted, details); |
d16dfc550
|
1340 |
tlb_finish_mmu(&tlb, address, end); |
ee39b37b2
|
1341 |
return end; |
1da177e4c
|
1342 |
} |
c627f9cc0
|
1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 |
/** * zap_vma_ptes - remove ptes mapping the vma * @vma: vm_area_struct holding ptes to be zapped * @address: starting address of pages to zap * @size: number of bytes to zap * * This function only unmaps ptes assigned to VM_PFNMAP vmas. * * The entire address range must be fully contained within the vma. * * Returns 0 if successful. */ int zap_vma_ptes(struct vm_area_struct *vma, unsigned long address, unsigned long size) { if (address < vma->vm_start || address + size > vma->vm_end || !(vma->vm_flags & VM_PFNMAP)) return -1; zap_page_range(vma, address, size, NULL); return 0; } EXPORT_SYMBOL_GPL(zap_vma_ptes); |
142762bd8
|
1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 |
/** * follow_page - look up a page descriptor from a user-virtual address * @vma: vm_area_struct mapping @address * @address: virtual address to look up * @flags: flags modifying lookup behaviour * * @flags can have FOLL_ flags set, defined in <linux/mm.h> * * Returns the mapped (struct page *), %NULL if no mapping exists, or * an error pointer if there is a mapping to something not represented * by a page descriptor (see also vm_normal_page()). |
1da177e4c
|
1376 |
*/ |
6aab341e0
|
1377 |
struct page *follow_page(struct vm_area_struct *vma, unsigned long address, |
deceb6cd1
|
1378 |
unsigned int flags) |
1da177e4c
|
1379 1380 1381 1382 1383 |
{ pgd_t *pgd; pud_t *pud; pmd_t *pmd; pte_t *ptep, pte; |
deceb6cd1
|
1384 |
spinlock_t *ptl; |
1da177e4c
|
1385 |
struct page *page; |
6aab341e0
|
1386 |
struct mm_struct *mm = vma->vm_mm; |
1da177e4c
|
1387 |
|
deceb6cd1
|
1388 1389 1390 1391 1392 |
page = follow_huge_addr(mm, address, flags & FOLL_WRITE); if (!IS_ERR(page)) { BUG_ON(flags & FOLL_GET); goto out; } |
1da177e4c
|
1393 |
|
deceb6cd1
|
1394 |
page = NULL; |
1da177e4c
|
1395 1396 |
pgd = pgd_offset(mm, address); if (pgd_none(*pgd) || unlikely(pgd_bad(*pgd))) |
deceb6cd1
|
1397 |
goto no_page_table; |
1da177e4c
|
1398 1399 |
pud = pud_offset(pgd, address); |
ceb868796
|
1400 |
if (pud_none(*pud)) |
deceb6cd1
|
1401 |
goto no_page_table; |
8a07651ee
|
1402 |
if (pud_huge(*pud) && vma->vm_flags & VM_HUGETLB) { |
ceb868796
|
1403 1404 1405 1406 1407 1408 |
BUG_ON(flags & FOLL_GET); page = follow_huge_pud(mm, address, pud, flags & FOLL_WRITE); goto out; } if (unlikely(pud_bad(*pud))) goto no_page_table; |
1da177e4c
|
1409 |
pmd = pmd_offset(pud, address); |
aeed5fce3
|
1410 |
if (pmd_none(*pmd)) |
deceb6cd1
|
1411 |
goto no_page_table; |
71e3aac07
|
1412 |
if (pmd_huge(*pmd) && vma->vm_flags & VM_HUGETLB) { |
deceb6cd1
|
1413 1414 |
BUG_ON(flags & FOLL_GET); page = follow_huge_pmd(mm, address, pmd, flags & FOLL_WRITE); |
1da177e4c
|
1415 |
goto out; |
deceb6cd1
|
1416 |
} |
71e3aac07
|
1417 |
if (pmd_trans_huge(*pmd)) { |
500d65d47
|
1418 1419 1420 1421 |
if (flags & FOLL_SPLIT) { split_huge_page_pmd(mm, pmd); goto split_fallthrough; } |
71e3aac07
|
1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 |
spin_lock(&mm->page_table_lock); if (likely(pmd_trans_huge(*pmd))) { if (unlikely(pmd_trans_splitting(*pmd))) { spin_unlock(&mm->page_table_lock); wait_split_huge_page(vma->anon_vma, pmd); } else { page = follow_trans_huge_pmd(mm, address, pmd, flags); spin_unlock(&mm->page_table_lock); goto out; } } else spin_unlock(&mm->page_table_lock); /* fall through */ } |
500d65d47
|
1437 |
split_fallthrough: |
aeed5fce3
|
1438 1439 |
if (unlikely(pmd_bad(*pmd))) goto no_page_table; |
deceb6cd1
|
1440 |
ptep = pte_offset_map_lock(mm, pmd, address, &ptl); |
1da177e4c
|
1441 1442 |
pte = *ptep; |
deceb6cd1
|
1443 |
if (!pte_present(pte)) |
89f5b7da2
|
1444 |
goto no_page; |
deceb6cd1
|
1445 1446 |
if ((flags & FOLL_WRITE) && !pte_write(pte)) goto unlock; |
a13ea5b75
|
1447 |
|
6aab341e0
|
1448 |
page = vm_normal_page(vma, address, pte); |
a13ea5b75
|
1449 1450 |
if (unlikely(!page)) { if ((flags & FOLL_DUMP) || |
62eede62d
|
1451 |
!is_zero_pfn(pte_pfn(pte))) |
a13ea5b75
|
1452 1453 1454 |
goto bad_page; page = pte_page(pte); } |
1da177e4c
|
1455 |
|
deceb6cd1
|
1456 |
if (flags & FOLL_GET) |
70b50f94f
|
1457 |
get_page_foll(page); |
deceb6cd1
|
1458 1459 1460 1461 |
if (flags & FOLL_TOUCH) { if ((flags & FOLL_WRITE) && !pte_dirty(pte) && !PageDirty(page)) set_page_dirty(page); |
bd775c42e
|
1462 1463 1464 1465 1466 |
/* * pte_mkyoung() would be more correct here, but atomic care * is needed to avoid losing the dirty bit: it is easier to use * mark_page_accessed(). */ |
deceb6cd1
|
1467 1468 |
mark_page_accessed(page); } |
a1fde08c7
|
1469 |
if ((flags & FOLL_MLOCK) && (vma->vm_flags & VM_LOCKED)) { |
110d74a92
|
1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 |
/* * The preliminary mapping check is mainly to avoid the * pointless overhead of lock_page on the ZERO_PAGE * which might bounce very badly if there is contention. * * If the page is already locked, we don't need to * handle it now - vmscan will handle it later if and * when it attempts to reclaim the page. */ if (page->mapping && trylock_page(page)) { lru_add_drain(); /* push cached pages to LRU */ /* * Because we lock page here and migration is * blocked by the pte's page reference, we need * only check for file-cache page truncation. */ if (page->mapping) mlock_vma_page(page); unlock_page(page); } } |
deceb6cd1
|
1491 1492 |
unlock: pte_unmap_unlock(ptep, ptl); |
1da177e4c
|
1493 |
out: |
deceb6cd1
|
1494 |
return page; |
1da177e4c
|
1495 |
|
89f5b7da2
|
1496 1497 1498 1499 1500 1501 1502 1503 |
bad_page: pte_unmap_unlock(ptep, ptl); return ERR_PTR(-EFAULT); no_page: pte_unmap_unlock(ptep, ptl); if (!pte_none(pte)) return page; |
8e4b9a607
|
1504 |
|
deceb6cd1
|
1505 1506 1507 |
no_page_table: /* * When core dumping an enormous anonymous area that nobody |
8e4b9a607
|
1508 1509 1510 1511 1512 |
* has touched so far, we don't want to allocate unnecessary pages or * page tables. Return error instead of NULL to skip handle_mm_fault, * then get_dump_page() will return NULL to leave a hole in the dump. * But we can only make this optimization where a hole would surely * be zero-filled if handle_mm_fault() actually did handle it. |
deceb6cd1
|
1513 |
*/ |
8e4b9a607
|
1514 1515 1516 |
if ((flags & FOLL_DUMP) && (!vma->vm_ops || !vma->vm_ops->fault)) return ERR_PTR(-EFAULT); |
deceb6cd1
|
1517 |
return page; |
1da177e4c
|
1518 |
} |
95042f9eb
|
1519 1520 |
static inline int stack_guard_page(struct vm_area_struct *vma, unsigned long addr) { |
a09a79f66
|
1521 1522 |
return stack_guard_page_start(vma, addr) || stack_guard_page_end(vma, addr+PAGE_SIZE); |
95042f9eb
|
1523 |
} |
0014bd990
|
1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 |
/** * __get_user_pages() - pin user pages in memory * @tsk: task_struct of target task * @mm: mm_struct of target mm * @start: starting user address * @nr_pages: number of pages from start to pin * @gup_flags: flags modifying pin behaviour * @pages: array that receives pointers to the pages pinned. * Should be at least nr_pages long. Or NULL, if caller * only intends to ensure the pages are faulted in. * @vmas: array of pointers to vmas corresponding to each page. * Or NULL if the caller does not require them. * @nonblocking: whether waiting for disk IO or mmap_sem contention * * Returns number of pages pinned. This may be fewer than the number * requested. If nr_pages is 0 or negative, returns 0. If no pages * were pinned, returns -errno. Each page returned must be released * with a put_page() call when it is finished with. vmas will only * remain valid while mmap_sem is held. * * Must be called with mmap_sem held for read or write. * * __get_user_pages walks a process's page tables and takes a reference to * each struct page that each user address corresponds to at a given * instant. That is, it takes the page that would be accessed if a user * thread accesses the given user virtual address at that instant. * * This does not guarantee that the page exists in the user mappings when * __get_user_pages returns, and there may even be a completely different * page there in some cases (eg. if mmapped pagecache has been invalidated * and subsequently re faulted). However it does guarantee that the page * won't be freed completely. And mostly callers simply care that the page * contains data that was valid *at some point in time*. Typically, an IO * or similar operation cannot guarantee anything stronger anyway because * locks can't be held over the syscall boundary. * * If @gup_flags & FOLL_WRITE == 0, the page must not be written to. If * the page is written to, set_page_dirty (or set_page_dirty_lock, as * appropriate) must be called after the page is finished with, and * before put_page is called. * * If @nonblocking != NULL, __get_user_pages will not wait for disk IO * or mmap_sem contention, and if waiting is needed to pin all pages, * *@nonblocking will be set to 0. * * In most cases, get_user_pages or get_user_pages_fast should be used * instead of __get_user_pages. __get_user_pages should be used only if * you need some special @gup_flags. */ |
b291f0003
|
1573 |
int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm, |
58fa879e1
|
1574 |
unsigned long start, int nr_pages, unsigned int gup_flags, |
53a7706d5
|
1575 1576 |
struct page **pages, struct vm_area_struct **vmas, int *nonblocking) |
1da177e4c
|
1577 1578 |
{ int i; |
58fa879e1
|
1579 |
unsigned long vm_flags; |
1da177e4c
|
1580 |
|
9d73777e5
|
1581 |
if (nr_pages <= 0) |
900cf086f
|
1582 |
return 0; |
58fa879e1
|
1583 1584 |
VM_BUG_ON(!!pages != !!(gup_flags & FOLL_GET)); |
1da177e4c
|
1585 1586 |
/* * Require read or write permissions. |
58fa879e1
|
1587 |
* If FOLL_FORCE is set, we only require the "MAY" flags. |
1da177e4c
|
1588 |
*/ |
58fa879e1
|
1589 1590 1591 1592 |
vm_flags = (gup_flags & FOLL_WRITE) ? (VM_WRITE | VM_MAYWRITE) : (VM_READ | VM_MAYREAD); vm_flags &= (gup_flags & FOLL_FORCE) ? (VM_MAYREAD | VM_MAYWRITE) : (VM_READ | VM_WRITE); |
1da177e4c
|
1593 1594 1595 |
i = 0; do { |
deceb6cd1
|
1596 |
struct vm_area_struct *vma; |
1da177e4c
|
1597 1598 |
vma = find_extend_vma(mm, start); |
e7f22e207
|
1599 |
if (!vma && in_gate_area(mm, start)) { |
1da177e4c
|
1600 |
unsigned long pg = start & PAGE_MASK; |
1da177e4c
|
1601 1602 1603 1604 |
pgd_t *pgd; pud_t *pud; pmd_t *pmd; pte_t *pte; |
b291f0003
|
1605 1606 |
/* user gate pages are read-only */ |
58fa879e1
|
1607 |
if (gup_flags & FOLL_WRITE) |
1da177e4c
|
1608 1609 1610 1611 1612 1613 1614 1615 1616 |
return i ? : -EFAULT; if (pg > TASK_SIZE) pgd = pgd_offset_k(pg); else pgd = pgd_offset_gate(mm, pg); BUG_ON(pgd_none(*pgd)); pud = pud_offset(pgd, pg); BUG_ON(pud_none(*pud)); pmd = pmd_offset(pud, pg); |
690dbe1ce
|
1617 1618 |
if (pmd_none(*pmd)) return i ? : -EFAULT; |
f66055ab6
|
1619 |
VM_BUG_ON(pmd_trans_huge(*pmd)); |
1da177e4c
|
1620 |
pte = pte_offset_map(pmd, pg); |
690dbe1ce
|
1621 1622 1623 1624 |
if (pte_none(*pte)) { pte_unmap(pte); return i ? : -EFAULT; } |
95042f9eb
|
1625 |
vma = get_gate_vma(mm); |
1da177e4c
|
1626 |
if (pages) { |
de51257aa
|
1627 |
struct page *page; |
95042f9eb
|
1628 |
page = vm_normal_page(vma, start, *pte); |
de51257aa
|
1629 1630 1631 1632 1633 1634 1635 1636 1637 |
if (!page) { if (!(gup_flags & FOLL_DUMP) && is_zero_pfn(pte_pfn(*pte))) page = pte_page(*pte); else { pte_unmap(pte); return i ? : -EFAULT; } } |
6aab341e0
|
1638 |
pages[i] = page; |
de51257aa
|
1639 |
get_page(page); |
1da177e4c
|
1640 1641 |
} pte_unmap(pte); |
95042f9eb
|
1642 |
goto next_page; |
1da177e4c
|
1643 |
} |
b291f0003
|
1644 1645 |
if (!vma || (vma->vm_flags & (VM_IO | VM_PFNMAP)) || |
1c3aff1ce
|
1646 |
!(vm_flags & vma->vm_flags)) |
1da177e4c
|
1647 |
return i ? : -EFAULT; |
2a15efc95
|
1648 1649 |
if (is_vm_hugetlb_page(vma)) { i = follow_hugetlb_page(mm, vma, pages, vmas, |
58fa879e1
|
1650 |
&start, &nr_pages, i, gup_flags); |
2a15efc95
|
1651 1652 |
continue; } |
deceb6cd1
|
1653 |
|
1da177e4c
|
1654 |
do { |
08ef47293
|
1655 |
struct page *page; |
58fa879e1
|
1656 |
unsigned int foll_flags = gup_flags; |
1da177e4c
|
1657 |
|
462e00cc7
|
1658 |
/* |
4779280d1
|
1659 |
* If we have a pending SIGKILL, don't keep faulting |
1c3aff1ce
|
1660 |
* pages and potentially allocating memory. |
462e00cc7
|
1661 |
*/ |
1c3aff1ce
|
1662 |
if (unlikely(fatal_signal_pending(current))) |
4779280d1
|
1663 |
return i ? i : -ERESTARTSYS; |
462e00cc7
|
1664 |
|
deceb6cd1
|
1665 |
cond_resched(); |
6aab341e0
|
1666 |
while (!(page = follow_page(vma, start, foll_flags))) { |
deceb6cd1
|
1667 |
int ret; |
53a7706d5
|
1668 |
unsigned int fault_flags = 0; |
a09a79f66
|
1669 1670 1671 1672 1673 |
/* For mlock, just skip the stack guard page. */ if (foll_flags & FOLL_MLOCK) { if (stack_guard_page(vma, start)) goto next_page; } |
53a7706d5
|
1674 1675 1676 1677 |
if (foll_flags & FOLL_WRITE) fault_flags |= FAULT_FLAG_WRITE; if (nonblocking) fault_flags |= FAULT_FLAG_ALLOW_RETRY; |
318b275fb
|
1678 1679 |
if (foll_flags & FOLL_NOWAIT) fault_flags |= (FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_RETRY_NOWAIT); |
d06063cc2
|
1680 |
|
d26ed650d
|
1681 |
ret = handle_mm_fault(mm, vma, start, |
53a7706d5
|
1682 |
fault_flags); |
d26ed650d
|
1683 |
|
83c54070e
|
1684 1685 1686 |
if (ret & VM_FAULT_ERROR) { if (ret & VM_FAULT_OOM) return i ? i : -ENOMEM; |
69ebb83e1
|
1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 |
if (ret & (VM_FAULT_HWPOISON | VM_FAULT_HWPOISON_LARGE)) { if (i) return i; else if (gup_flags & FOLL_HWPOISON) return -EHWPOISON; else return -EFAULT; } if (ret & VM_FAULT_SIGBUS) |
83c54070e
|
1697 1698 1699 |
return i ? i : -EFAULT; BUG(); } |
e7f22e207
|
1700 1701 1702 1703 1704 1705 1706 |
if (tsk) { if (ret & VM_FAULT_MAJOR) tsk->maj_flt++; else tsk->min_flt++; } |
83c54070e
|
1707 |
|
53a7706d5
|
1708 |
if (ret & VM_FAULT_RETRY) { |
318b275fb
|
1709 1710 |
if (nonblocking) *nonblocking = 0; |
53a7706d5
|
1711 1712 |
return i; } |
a68d2ebc1
|
1713 |
/* |
83c54070e
|
1714 1715 1716 1717 |
* The VM_FAULT_WRITE bit tells us that * do_wp_page has broken COW when necessary, * even if maybe_mkwrite decided not to set * pte_write. We can thus safely do subsequent |
878b63ac8
|
1718 1719 1720 1721 1722 1723 |
* page lookups as if they were reads. But only * do so when looping for pte_write is futile: * in some cases userspace may also be wanting * to write to the gotten user page, which a * read fault here might prevent (a readonly * page might get reCOWed by userspace write). |
a68d2ebc1
|
1724 |
*/ |
878b63ac8
|
1725 1726 |
if ((ret & VM_FAULT_WRITE) && !(vma->vm_flags & VM_WRITE)) |
deceb6cd1
|
1727 |
foll_flags &= ~FOLL_WRITE; |
83c54070e
|
1728 |
|
7f7bbbe50
|
1729 |
cond_resched(); |
1da177e4c
|
1730 |
} |
89f5b7da2
|
1731 1732 |
if (IS_ERR(page)) return i ? i : PTR_ERR(page); |
1da177e4c
|
1733 |
if (pages) { |
08ef47293
|
1734 |
pages[i] = page; |
03beb0766
|
1735 |
|
a6f36be32
|
1736 |
flush_anon_page(vma, page, start); |
08ef47293
|
1737 |
flush_dcache_page(page); |
1da177e4c
|
1738 |
} |
95042f9eb
|
1739 |
next_page: |
1da177e4c
|
1740 1741 1742 1743 |
if (vmas) vmas[i] = vma; i++; start += PAGE_SIZE; |
9d73777e5
|
1744 1745 1746 |
nr_pages--; } while (nr_pages && start < vma->vm_end); } while (nr_pages); |
1da177e4c
|
1747 1748 |
return i; } |
0014bd990
|
1749 |
EXPORT_SYMBOL(__get_user_pages); |
b291f0003
|
1750 |
|
2efaca927
|
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 |
/* * fixup_user_fault() - manually resolve a user page fault * @tsk: the task_struct to use for page fault accounting, or * NULL if faults are not to be recorded. * @mm: mm_struct of target mm * @address: user address * @fault_flags:flags to pass down to handle_mm_fault() * * This is meant to be called in the specific scenario where for locking reasons * we try to access user memory in atomic context (within a pagefault_disable() * section), this returns -EFAULT, and we want to resolve the user fault before * trying again. * * Typically this is meant to be used by the futex code. * * The main difference with get_user_pages() is that this function will * unconditionally call handle_mm_fault() which will in turn perform all the * necessary SW fixup of the dirty and young bits in the PTE, while * handle_mm_fault() only guarantees to update these in the struct page. * * This is important for some architectures where those bits also gate the * access permission to the page because they are maintained in software. On * such architectures, gup() will not be enough to make a subsequent access * succeed. * * This should be called with the mm_sem held for read. */ int fixup_user_fault(struct task_struct *tsk, struct mm_struct *mm, unsigned long address, unsigned int fault_flags) { struct vm_area_struct *vma; int ret; vma = find_extend_vma(mm, address); if (!vma || address < vma->vm_start) return -EFAULT; ret = handle_mm_fault(mm, vma, address, fault_flags); if (ret & VM_FAULT_ERROR) { if (ret & VM_FAULT_OOM) return -ENOMEM; if (ret & (VM_FAULT_HWPOISON | VM_FAULT_HWPOISON_LARGE)) return -EHWPOISON; if (ret & VM_FAULT_SIGBUS) return -EFAULT; BUG(); } if (tsk) { if (ret & VM_FAULT_MAJOR) tsk->maj_flt++; else tsk->min_flt++; } return 0; } /* |
d2bf6be8a
|
1808 |
* get_user_pages() - pin user pages in memory |
e7f22e207
|
1809 1810 |
* @tsk: the task_struct to use for page fault accounting, or * NULL if faults are not to be recorded. |
d2bf6be8a
|
1811 1812 |
* @mm: mm_struct of target mm * @start: starting user address |
9d73777e5
|
1813 |
* @nr_pages: number of pages from start to pin |
d2bf6be8a
|
1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 |
* @write: whether pages will be written to by the caller * @force: whether to force write access even if user mapping is * readonly. This will result in the page being COWed even * in MAP_SHARED mappings. You do not want this. * @pages: array that receives pointers to the pages pinned. * Should be at least nr_pages long. Or NULL, if caller * only intends to ensure the pages are faulted in. * @vmas: array of pointers to vmas corresponding to each page. * Or NULL if the caller does not require them. * * Returns number of pages pinned. This may be fewer than the number |
9d73777e5
|
1825 |
* requested. If nr_pages is 0 or negative, returns 0. If no pages |
d2bf6be8a
|
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 |
* were pinned, returns -errno. Each page returned must be released * with a put_page() call when it is finished with. vmas will only * remain valid while mmap_sem is held. * * Must be called with mmap_sem held for read or write. * * get_user_pages walks a process's page tables and takes a reference to * each struct page that each user address corresponds to at a given * instant. That is, it takes the page that would be accessed if a user * thread accesses the given user virtual address at that instant. * * This does not guarantee that the page exists in the user mappings when * get_user_pages returns, and there may even be a completely different * page there in some cases (eg. if mmapped pagecache has been invalidated * and subsequently re faulted). However it does guarantee that the page * won't be freed completely. And mostly callers simply care that the page * contains data that was valid *at some point in time*. Typically, an IO * or similar operation cannot guarantee anything stronger anyway because * locks can't be held over the syscall boundary. * * If write=0, the page must not be written to. If the page is written to, * set_page_dirty (or set_page_dirty_lock, as appropriate) must be called * after the page is finished with, and before put_page is called. * * get_user_pages is typically used for fewer-copy IO operations, to get a * handle on the memory by some means other than accesses via the user virtual * addresses. The pages may be submitted for DMA to devices or accessed via * their kernel linear mapping (via the kmap APIs). Care should be taken to * use the correct cache flushing APIs. * * See also get_user_pages_fast, for performance critical applications. */ |
b291f0003
|
1858 |
int get_user_pages(struct task_struct *tsk, struct mm_struct *mm, |
9d73777e5
|
1859 |
unsigned long start, int nr_pages, int write, int force, |
b291f0003
|
1860 1861 |
struct page **pages, struct vm_area_struct **vmas) { |
58fa879e1
|
1862 |
int flags = FOLL_TOUCH; |
b291f0003
|
1863 |
|
58fa879e1
|
1864 1865 |
if (pages) flags |= FOLL_GET; |
b291f0003
|
1866 |
if (write) |
58fa879e1
|
1867 |
flags |= FOLL_WRITE; |
b291f0003
|
1868 |
if (force) |
58fa879e1
|
1869 |
flags |= FOLL_FORCE; |
b291f0003
|
1870 |
|
53a7706d5
|
1871 1872 |
return __get_user_pages(tsk, mm, start, nr_pages, flags, pages, vmas, NULL); |
b291f0003
|
1873 |
} |
1da177e4c
|
1874 |
EXPORT_SYMBOL(get_user_pages); |
f3e8fccd0
|
1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 |
/** * get_dump_page() - pin user page in memory while writing it to core dump * @addr: user address * * Returns struct page pointer of user page pinned for dump, * to be freed afterwards by page_cache_release() or put_page(). * * Returns NULL on any kind of failure - a hole must then be inserted into * the corefile, to preserve alignment with its headers; and also returns * NULL wherever the ZERO_PAGE, or an anonymous pte_none, has been found - * allowing a hole to be left in the corefile to save diskspace. * * Called without mmap_sem, but after all other threads have been killed. */ #ifdef CONFIG_ELF_CORE struct page *get_dump_page(unsigned long addr) { struct vm_area_struct *vma; struct page *page; if (__get_user_pages(current, current->mm, addr, 1, |
53a7706d5
|
1896 1897 |
FOLL_FORCE | FOLL_DUMP | FOLL_GET, &page, &vma, NULL) < 1) |
f3e8fccd0
|
1898 |
return NULL; |
f3e8fccd0
|
1899 1900 1901 1902 |
flush_cache_page(vma, addr, page_to_pfn(page)); return page; } #endif /* CONFIG_ELF_CORE */ |
25ca1d6c0
|
1903 |
pte_t *__get_locked_pte(struct mm_struct *mm, unsigned long addr, |
920c7a5d0
|
1904 |
spinlock_t **ptl) |
c9cfcddfd
|
1905 1906 1907 1908 |
{ pgd_t * pgd = pgd_offset(mm, addr); pud_t * pud = pud_alloc(mm, pgd, addr); if (pud) { |
49c91fb01
|
1909 |
pmd_t * pmd = pmd_alloc(mm, pud, addr); |
f66055ab6
|
1910 1911 |
if (pmd) { VM_BUG_ON(pmd_trans_huge(*pmd)); |
c9cfcddfd
|
1912 |
return pte_alloc_map_lock(mm, pmd, addr, ptl); |
f66055ab6
|
1913 |
} |
c9cfcddfd
|
1914 1915 1916 |
} return NULL; } |
1da177e4c
|
1917 |
/* |
238f58d89
|
1918 1919 1920 1921 1922 1923 |
* This is the old fallback for page remapping. * * For historical reasons, it only allows reserved pages. Only * old drivers should use this, and they needed to mark their * pages reserved for the old functions anyway. */ |
423bad600
|
1924 1925 |
static int insert_page(struct vm_area_struct *vma, unsigned long addr, struct page *page, pgprot_t prot) |
238f58d89
|
1926 |
{ |
423bad600
|
1927 |
struct mm_struct *mm = vma->vm_mm; |
238f58d89
|
1928 |
int retval; |
c9cfcddfd
|
1929 |
pte_t *pte; |
8a9f3ccd2
|
1930 |
spinlock_t *ptl; |
238f58d89
|
1931 |
retval = -EINVAL; |
a145dd411
|
1932 |
if (PageAnon(page)) |
5b4e655e9
|
1933 |
goto out; |
238f58d89
|
1934 1935 |
retval = -ENOMEM; flush_dcache_page(page); |
c9cfcddfd
|
1936 |
pte = get_locked_pte(mm, addr, &ptl); |
238f58d89
|
1937 |
if (!pte) |
5b4e655e9
|
1938 |
goto out; |
238f58d89
|
1939 1940 1941 1942 1943 1944 |
retval = -EBUSY; if (!pte_none(*pte)) goto out_unlock; /* Ok, finally just insert the thing.. */ get_page(page); |
34e55232e
|
1945 |
inc_mm_counter_fast(mm, MM_FILEPAGES); |
238f58d89
|
1946 1947 1948 1949 |
page_add_file_rmap(page); set_pte_at(mm, addr, pte, mk_pte(page, prot)); retval = 0; |
8a9f3ccd2
|
1950 1951 |
pte_unmap_unlock(pte, ptl); return retval; |
238f58d89
|
1952 1953 1954 1955 1956 |
out_unlock: pte_unmap_unlock(pte, ptl); out: return retval; } |
bfa5bf6d6
|
1957 1958 1959 1960 1961 1962 |
/** * vm_insert_page - insert single page into user vma * @vma: user vma to map to * @addr: target user address of this page * @page: source kernel page * |
a145dd411
|
1963 1964 1965 1966 1967 1968 |
* This allows drivers to insert individual pages they've allocated * into a user vma. * * The page has to be a nice clean _individual_ kernel allocation. * If you allocate a compound page, you need to have marked it as * such (__GFP_COMP), or manually just split the page up yourself |
8dfcc9ba2
|
1969 |
* (see split_page()). |
a145dd411
|
1970 1971 1972 1973 1974 1975 1976 1977 1978 |
* * NOTE! Traditionally this was done with "remap_pfn_range()" which * took an arbitrary page protection parameter. This doesn't allow * that. Your vma protection will have to be set up correctly, which * means that if you want a shared writable mapping, you'd better * ask for a shared writable mapping! * * The page does not need to be reserved. */ |
423bad600
|
1979 1980 |
int vm_insert_page(struct vm_area_struct *vma, unsigned long addr, struct page *page) |
a145dd411
|
1981 1982 1983 1984 1985 |
{ if (addr < vma->vm_start || addr >= vma->vm_end) return -EFAULT; if (!page_count(page)) return -EINVAL; |
4d7672b46
|
1986 |
vma->vm_flags |= VM_INSERTPAGE; |
423bad600
|
1987 |
return insert_page(vma, addr, page, vma->vm_page_prot); |
a145dd411
|
1988 |
} |
e3c3374fb
|
1989 |
EXPORT_SYMBOL(vm_insert_page); |
a145dd411
|
1990 |
|
423bad600
|
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 |
static int insert_pfn(struct vm_area_struct *vma, unsigned long addr, unsigned long pfn, pgprot_t prot) { struct mm_struct *mm = vma->vm_mm; int retval; pte_t *pte, entry; spinlock_t *ptl; retval = -ENOMEM; pte = get_locked_pte(mm, addr, &ptl); if (!pte) goto out; retval = -EBUSY; if (!pte_none(*pte)) goto out_unlock; /* Ok, finally just insert the thing.. */ entry = pte_mkspecial(pfn_pte(pfn, prot)); set_pte_at(mm, addr, pte, entry); |
4b3073e1c
|
2010 |
update_mmu_cache(vma, addr, pte); /* XXX: why not for insert_page? */ |
423bad600
|
2011 2012 2013 2014 2015 2016 2017 |
retval = 0; out_unlock: pte_unmap_unlock(pte, ptl); out: return retval; } |
e0dc0d8f4
|
2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 |
/** * vm_insert_pfn - insert single pfn into user vma * @vma: user vma to map to * @addr: target user address of this page * @pfn: source kernel pfn * * Similar to vm_inert_page, this allows drivers to insert individual pages * they've allocated into a user vma. Same comments apply. * * This function should only be called from a vm_ops->fault handler, and * in that case the handler should return NULL. |
0d71d10a4
|
2029 2030 2031 2032 2033 |
* * vma cannot be a COW mapping. * * As this is called only for pages that do not currently exist, we * do not need to flush old virtual caches or the TLB. |
e0dc0d8f4
|
2034 2035 |
*/ int vm_insert_pfn(struct vm_area_struct *vma, unsigned long addr, |
423bad600
|
2036 |
unsigned long pfn) |
e0dc0d8f4
|
2037 |
{ |
2ab640379
|
2038 |
int ret; |
e4b866ed1
|
2039 |
pgprot_t pgprot = vma->vm_page_prot; |
7e675137a
|
2040 2041 2042 2043 2044 2045 |
/* * Technically, architectures with pte_special can avoid all these * restrictions (same for remap_pfn_range). However we would like * consistency in testing and feature parity among all, so we should * try to keep these invariants in place for everybody. */ |
b379d7901
|
2046 2047 2048 2049 2050 |
BUG_ON(!(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP))); BUG_ON((vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP)) == (VM_PFNMAP|VM_MIXEDMAP)); BUG_ON((vma->vm_flags & VM_PFNMAP) && is_cow_mapping(vma->vm_flags)); BUG_ON((vma->vm_flags & VM_MIXEDMAP) && pfn_valid(pfn)); |
e0dc0d8f4
|
2051 |
|
423bad600
|
2052 2053 |
if (addr < vma->vm_start || addr >= vma->vm_end) return -EFAULT; |
e4b866ed1
|
2054 |
if (track_pfn_vma_new(vma, &pgprot, pfn, PAGE_SIZE)) |
2ab640379
|
2055 |
return -EINVAL; |
e4b866ed1
|
2056 |
ret = insert_pfn(vma, addr, pfn, pgprot); |
2ab640379
|
2057 2058 2059 2060 2061 |
if (ret) untrack_pfn_vma(vma, pfn, PAGE_SIZE); return ret; |
423bad600
|
2062 2063 |
} EXPORT_SYMBOL(vm_insert_pfn); |
e0dc0d8f4
|
2064 |
|
423bad600
|
2065 2066 2067 2068 |
int vm_insert_mixed(struct vm_area_struct *vma, unsigned long addr, unsigned long pfn) { BUG_ON(!(vma->vm_flags & VM_MIXEDMAP)); |
e0dc0d8f4
|
2069 |
|
423bad600
|
2070 2071 |
if (addr < vma->vm_start || addr >= vma->vm_end) return -EFAULT; |
e0dc0d8f4
|
2072 |
|
423bad600
|
2073 2074 2075 2076 |
/* * If we don't have pte special, then we have to use the pfn_valid() * based VM_MIXEDMAP scheme (see vm_normal_page), and thus we *must* * refcount the page if pfn_valid is true (hence insert_page rather |
62eede62d
|
2077 2078 |
* than insert_pfn). If a zero_pfn were inserted into a VM_MIXEDMAP * without pte special, it would there be refcounted as a normal page. |
423bad600
|
2079 2080 2081 2082 2083 2084 2085 2086 |
*/ if (!HAVE_PTE_SPECIAL && pfn_valid(pfn)) { struct page *page; page = pfn_to_page(pfn); return insert_page(vma, addr, page, vma->vm_page_prot); } return insert_pfn(vma, addr, pfn, vma->vm_page_prot); |
e0dc0d8f4
|
2087 |
} |
423bad600
|
2088 |
EXPORT_SYMBOL(vm_insert_mixed); |
e0dc0d8f4
|
2089 |
|
a145dd411
|
2090 |
/* |
1da177e4c
|
2091 2092 2093 2094 2095 2096 2097 2098 2099 |
* maps a range of physical memory into the requested pages. the old * mappings are removed. any references to nonexistent pages results * in null mappings (currently treated as "copy-on-access") */ static int remap_pte_range(struct mm_struct *mm, pmd_t *pmd, unsigned long addr, unsigned long end, unsigned long pfn, pgprot_t prot) { pte_t *pte; |
c74df32c7
|
2100 |
spinlock_t *ptl; |
1da177e4c
|
2101 |
|
c74df32c7
|
2102 |
pte = pte_alloc_map_lock(mm, pmd, addr, &ptl); |
1da177e4c
|
2103 2104 |
if (!pte) return -ENOMEM; |
6606c3e0d
|
2105 |
arch_enter_lazy_mmu_mode(); |
1da177e4c
|
2106 2107 |
do { BUG_ON(!pte_none(*pte)); |
7e675137a
|
2108 |
set_pte_at(mm, addr, pte, pte_mkspecial(pfn_pte(pfn, prot))); |
1da177e4c
|
2109 2110 |
pfn++; } while (pte++, addr += PAGE_SIZE, addr != end); |
6606c3e0d
|
2111 |
arch_leave_lazy_mmu_mode(); |
c74df32c7
|
2112 |
pte_unmap_unlock(pte - 1, ptl); |
1da177e4c
|
2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 |
return 0; } static inline int remap_pmd_range(struct mm_struct *mm, pud_t *pud, unsigned long addr, unsigned long end, unsigned long pfn, pgprot_t prot) { pmd_t *pmd; unsigned long next; pfn -= addr >> PAGE_SHIFT; pmd = pmd_alloc(mm, pud, addr); if (!pmd) return -ENOMEM; |
f66055ab6
|
2127 |
VM_BUG_ON(pmd_trans_huge(*pmd)); |
1da177e4c
|
2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 |
do { next = pmd_addr_end(addr, end); if (remap_pte_range(mm, pmd, addr, next, pfn + (addr >> PAGE_SHIFT), prot)) return -ENOMEM; } while (pmd++, addr = next, addr != end); return 0; } static inline int remap_pud_range(struct mm_struct *mm, pgd_t *pgd, unsigned long addr, unsigned long end, unsigned long pfn, pgprot_t prot) { pud_t *pud; unsigned long next; pfn -= addr >> PAGE_SHIFT; pud = pud_alloc(mm, pgd, addr); if (!pud) return -ENOMEM; do { next = pud_addr_end(addr, end); if (remap_pmd_range(mm, pud, addr, next, pfn + (addr >> PAGE_SHIFT), prot)) return -ENOMEM; } while (pud++, addr = next, addr != end); return 0; } |
bfa5bf6d6
|
2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 |
/** * remap_pfn_range - remap kernel memory to userspace * @vma: user vma to map to * @addr: target user address to start at * @pfn: physical address of kernel memory * @size: size of map area * @prot: page protection flags for this mapping * * Note: this is only safe if the mm semaphore is held when called. */ |
1da177e4c
|
2166 2167 2168 2169 2170 |
int remap_pfn_range(struct vm_area_struct *vma, unsigned long addr, unsigned long pfn, unsigned long size, pgprot_t prot) { pgd_t *pgd; unsigned long next; |
2d15cab85
|
2171 |
unsigned long end = addr + PAGE_ALIGN(size); |
1da177e4c
|
2172 2173 2174 2175 2176 2177 2178 2179 |
struct mm_struct *mm = vma->vm_mm; int err; /* * Physically remapped pages are special. Tell the * rest of the world about it: * VM_IO tells people not to look at these pages * (accesses can have side effects). |
0b14c179a
|
2180 2181 2182 2183 2184 |
* VM_RESERVED is specified all over the place, because * in 2.4 it kept swapout's vma scan off this vma; but * in 2.6 the LRU scan won't even find its pages, so this * flag means no more than count its pages in reserved_vm, * and omit it from core dump, even when VM_IO turned off. |
6aab341e0
|
2185 2186 2187 |
* VM_PFNMAP tells the core MM that the base pages are just * raw PFN mappings, and do not have a "struct page" associated * with them. |
fb155c161
|
2188 2189 2190 2191 |
* * There's a horrible special case to handle copy-on-write * behaviour that some programs depend on. We mark the "original" * un-COW'ed pages by matching them up with "vma->vm_pgoff". |
1da177e4c
|
2192 |
*/ |
4bb9c5c02
|
2193 |
if (addr == vma->vm_start && end == vma->vm_end) { |
fb155c161
|
2194 |
vma->vm_pgoff = pfn; |
895791dac
|
2195 |
vma->vm_flags |= VM_PFN_AT_MMAP; |
4bb9c5c02
|
2196 |
} else if (is_cow_mapping(vma->vm_flags)) |
3c8bb73ac
|
2197 |
return -EINVAL; |
fb155c161
|
2198 |
|
6aab341e0
|
2199 |
vma->vm_flags |= VM_IO | VM_RESERVED | VM_PFNMAP; |
1da177e4c
|
2200 |
|
e4b866ed1
|
2201 |
err = track_pfn_vma_new(vma, &prot, pfn, PAGE_ALIGN(size)); |
a36706131
|
2202 2203 2204 2205 2206 2207 |
if (err) { /* * To indicate that track_pfn related cleanup is not * needed from higher level routine calling unmap_vmas */ vma->vm_flags &= ~(VM_IO | VM_RESERVED | VM_PFNMAP); |
895791dac
|
2208 |
vma->vm_flags &= ~VM_PFN_AT_MMAP; |
2ab640379
|
2209 |
return -EINVAL; |
a36706131
|
2210 |
} |
2ab640379
|
2211 |
|
1da177e4c
|
2212 2213 2214 2215 |
BUG_ON(addr >= end); pfn -= addr >> PAGE_SHIFT; pgd = pgd_offset(mm, addr); flush_cache_range(vma, addr, end); |
1da177e4c
|
2216 2217 2218 2219 2220 2221 2222 |
do { next = pgd_addr_end(addr, end); err = remap_pud_range(mm, pgd, addr, next, pfn + (addr >> PAGE_SHIFT), prot); if (err) break; } while (pgd++, addr = next, addr != end); |
2ab640379
|
2223 2224 2225 |
if (err) untrack_pfn_vma(vma, pfn, PAGE_ALIGN(size)); |
1da177e4c
|
2226 2227 2228 |
return err; } EXPORT_SYMBOL(remap_pfn_range); |
aee16b3ce
|
2229 2230 2231 2232 2233 2234 |
static int apply_to_pte_range(struct mm_struct *mm, pmd_t *pmd, unsigned long addr, unsigned long end, pte_fn_t fn, void *data) { pte_t *pte; int err; |
2f569afd9
|
2235 |
pgtable_t token; |
949099148
|
2236 |
spinlock_t *uninitialized_var(ptl); |
aee16b3ce
|
2237 2238 2239 2240 2241 2242 2243 2244 |
pte = (mm == &init_mm) ? pte_alloc_kernel(pmd, addr) : pte_alloc_map_lock(mm, pmd, addr, &ptl); if (!pte) return -ENOMEM; BUG_ON(pmd_huge(*pmd)); |
38e0edb15
|
2245 |
arch_enter_lazy_mmu_mode(); |
2f569afd9
|
2246 |
token = pmd_pgtable(*pmd); |
aee16b3ce
|
2247 2248 |
do { |
c36987e2e
|
2249 |
err = fn(pte++, token, addr, data); |
aee16b3ce
|
2250 2251 |
if (err) break; |
c36987e2e
|
2252 |
} while (addr += PAGE_SIZE, addr != end); |
aee16b3ce
|
2253 |
|
38e0edb15
|
2254 |
arch_leave_lazy_mmu_mode(); |
aee16b3ce
|
2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 |
if (mm != &init_mm) pte_unmap_unlock(pte-1, ptl); return err; } static int apply_to_pmd_range(struct mm_struct *mm, pud_t *pud, unsigned long addr, unsigned long end, pte_fn_t fn, void *data) { pmd_t *pmd; unsigned long next; int err; |
ceb868796
|
2267 |
BUG_ON(pud_huge(*pud)); |
aee16b3ce
|
2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 |
pmd = pmd_alloc(mm, pud, addr); if (!pmd) return -ENOMEM; do { next = pmd_addr_end(addr, end); err = apply_to_pte_range(mm, pmd, addr, next, fn, data); if (err) break; } while (pmd++, addr = next, addr != end); return err; } static int apply_to_pud_range(struct mm_struct *mm, pgd_t *pgd, unsigned long addr, unsigned long end, pte_fn_t fn, void *data) { pud_t *pud; unsigned long next; int err; pud = pud_alloc(mm, pgd, addr); if (!pud) return -ENOMEM; do { next = pud_addr_end(addr, end); err = apply_to_pmd_range(mm, pud, addr, next, fn, data); if (err) break; } while (pud++, addr = next, addr != end); return err; } /* * Scan a region of virtual memory, filling in page tables as necessary * and calling a provided function on each leaf page table. */ int apply_to_page_range(struct mm_struct *mm, unsigned long addr, unsigned long size, pte_fn_t fn, void *data) { pgd_t *pgd; unsigned long next; |
57250a5bf
|
2309 |
unsigned long end = addr + size; |
aee16b3ce
|
2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 |
int err; BUG_ON(addr >= end); pgd = pgd_offset(mm, addr); do { next = pgd_addr_end(addr, end); err = apply_to_pud_range(mm, pgd, addr, next, fn, data); if (err) break; } while (pgd++, addr = next, addr != end); |
57250a5bf
|
2320 |
|
aee16b3ce
|
2321 2322 2323 |
return err; } EXPORT_SYMBOL_GPL(apply_to_page_range); |
1da177e4c
|
2324 |
/* |
8f4e2101f
|
2325 2326 2327 |
* handle_pte_fault chooses page fault handler according to an entry * which was read non-atomically. Before making any commitment, on * those architectures or configurations (e.g. i386 with PAE) which |
a335b2e17
|
2328 |
* might give a mix of unmatched parts, do_swap_page and do_nonlinear_fault |
8f4e2101f
|
2329 2330 |
* must check under lock before unmapping the pte and proceeding * (but do_wp_page is only called after already making such a check; |
a335b2e17
|
2331 |
* and do_anonymous_page can safely check later on). |
8f4e2101f
|
2332 |
*/ |
4c21e2f24
|
2333 |
static inline int pte_unmap_same(struct mm_struct *mm, pmd_t *pmd, |
8f4e2101f
|
2334 2335 2336 2337 2338 |
pte_t *page_table, pte_t orig_pte) { int same = 1; #if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT) if (sizeof(pte_t) > sizeof(unsigned long)) { |
4c21e2f24
|
2339 2340 |
spinlock_t *ptl = pte_lockptr(mm, pmd); spin_lock(ptl); |
8f4e2101f
|
2341 |
same = pte_same(*page_table, orig_pte); |
4c21e2f24
|
2342 |
spin_unlock(ptl); |
8f4e2101f
|
2343 2344 2345 2346 2347 |
} #endif pte_unmap(page_table); return same; } |
9de455b20
|
2348 |
static inline void cow_user_page(struct page *dst, struct page *src, unsigned long va, struct vm_area_struct *vma) |
6aab341e0
|
2349 2350 2351 2352 2353 2354 2355 2356 2357 |
{ /* * If the source page was a PFN mapping, we don't have * a "struct page" for it. We do a best-effort copy by * just copying from the original user address. If that * fails, we just zero-fill it. Live with it. */ if (unlikely(!src)) { void *kaddr = kmap_atomic(dst, KM_USER0); |
5d2a2dbbc
|
2358 2359 2360 2361 2362 2363 2364 2365 2366 |
void __user *uaddr = (void __user *)(va & PAGE_MASK); /* * This really shouldn't fail, because the page is there * in the page tables. But it might just be unreadable, * in which case we just give up and fill the result with * zeroes. */ if (__copy_from_user_inatomic(kaddr, uaddr, PAGE_SIZE)) |
3ecb01df3
|
2367 |
clear_page(kaddr); |
6aab341e0
|
2368 |
kunmap_atomic(kaddr, KM_USER0); |
c4ec7b0de
|
2369 |
flush_dcache_page(dst); |
0ed361dec
|
2370 2371 |
} else copy_user_highpage(dst, src, va, vma); |
6aab341e0
|
2372 |
} |
1da177e4c
|
2373 |
/* |
1da177e4c
|
2374 2375 2376 2377 |
* This routine handles present pages, when users try to write * to a shared page. It is done by copying the page to a new address * and decrementing the shared-page counter for the old page. * |
1da177e4c
|
2378 2379 2380 2381 2382 2383 2384 2385 2386 |
* Note that this routine assumes that the protection checks have been * done by the caller (the low-level page fault routine in most cases). * Thus we can safely just mark it writable once we've done any necessary * COW. * * We also mark the page dirty at this point even though the page will * change only once the write actually happens. This avoids a few races, * and potentially makes it more efficient. * |
8f4e2101f
|
2387 2388 2389 |
* We enter with non-exclusive mmap_sem (to exclude vma changes, * but allow concurrent faults), with pte both mapped and locked. * We return with mmap_sem still held, but pte unmapped and unlocked. |
1da177e4c
|
2390 |
*/ |
65500d234
|
2391 2392 |
static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, pte_t *page_table, pmd_t *pmd, |
8f4e2101f
|
2393 |
spinlock_t *ptl, pte_t orig_pte) |
e6219ec81
|
2394 |
__releases(ptl) |
1da177e4c
|
2395 |
{ |
e5bbe4dfc
|
2396 |
struct page *old_page, *new_page; |
1da177e4c
|
2397 |
pte_t entry; |
b009c024f
|
2398 |
int ret = 0; |
a200ee182
|
2399 |
int page_mkwrite = 0; |
d08b3851d
|
2400 |
struct page *dirty_page = NULL; |
1da177e4c
|
2401 |
|
6aab341e0
|
2402 |
old_page = vm_normal_page(vma, address, orig_pte); |
251b97f55
|
2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 |
if (!old_page) { /* * VM_MIXEDMAP !pfn_valid() case * * We should not cow pages in a shared writeable mapping. * Just mark the pages writable as we can't do any dirty * accounting on raw pfn maps. */ if ((vma->vm_flags & (VM_WRITE|VM_SHARED)) == (VM_WRITE|VM_SHARED)) goto reuse; |
6aab341e0
|
2414 |
goto gotten; |
251b97f55
|
2415 |
} |
1da177e4c
|
2416 |
|
d08b3851d
|
2417 |
/* |
ee6a64578
|
2418 2419 |
* Take out anonymous pages first, anonymous shared vmas are * not dirty accountable. |
d08b3851d
|
2420 |
*/ |
9a8408951
|
2421 |
if (PageAnon(old_page) && !PageKsm(old_page)) { |
ab967d860
|
2422 2423 2424 2425 2426 2427 2428 2429 |
if (!trylock_page(old_page)) { page_cache_get(old_page); pte_unmap_unlock(page_table, ptl); lock_page(old_page); page_table = pte_offset_map_lock(mm, pmd, address, &ptl); if (!pte_same(*page_table, orig_pte)) { unlock_page(old_page); |
ab967d860
|
2430 2431 2432 |
goto unlock; } page_cache_release(old_page); |
ee6a64578
|
2433 |
} |
b009c024f
|
2434 |
if (reuse_swap_page(old_page)) { |
c44b67432
|
2435 2436 2437 2438 2439 2440 |
/* * The page is all ours. Move it to our anon_vma so * the rmap code will not search our parent or siblings. * Protected against the rmap code by the page lock. */ page_move_anon_rmap(old_page, vma, address); |
b009c024f
|
2441 2442 2443 |
unlock_page(old_page); goto reuse; } |
ab967d860
|
2444 |
unlock_page(old_page); |
ee6a64578
|
2445 |
} else if (unlikely((vma->vm_flags & (VM_WRITE|VM_SHARED)) == |
d08b3851d
|
2446 |
(VM_WRITE|VM_SHARED))) { |
ee6a64578
|
2447 2448 2449 2450 2451 |
/* * Only catch write-faults on shared writable pages, * read-only shared pages can get COWed by * get_user_pages(.write=1, .force=1). */ |
9637a5efd
|
2452 |
if (vma->vm_ops && vma->vm_ops->page_mkwrite) { |
c2ec175c3
|
2453 2454 2455 2456 2457 2458 2459 2460 |
struct vm_fault vmf; int tmp; vmf.virtual_address = (void __user *)(address & PAGE_MASK); vmf.pgoff = old_page->index; vmf.flags = FAULT_FLAG_WRITE|FAULT_FLAG_MKWRITE; vmf.page = old_page; |
9637a5efd
|
2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 |
/* * Notify the address space that the page is about to * become writable so that it can prohibit this or wait * for the page to get into an appropriate state. * * We do this without the lock held, so that it can * sleep if it needs to. */ page_cache_get(old_page); pte_unmap_unlock(page_table, ptl); |
c2ec175c3
|
2471 2472 2473 2474 |
tmp = vma->vm_ops->page_mkwrite(vma, &vmf); if (unlikely(tmp & (VM_FAULT_ERROR | VM_FAULT_NOPAGE))) { ret = tmp; |
9637a5efd
|
2475 |
goto unwritable_page; |
c2ec175c3
|
2476 |
} |
b827e496c
|
2477 2478 2479 2480 2481 2482 2483 2484 2485 |
if (unlikely(!(tmp & VM_FAULT_LOCKED))) { lock_page(old_page); if (!old_page->mapping) { ret = 0; /* retry the fault */ unlock_page(old_page); goto unwritable_page; } } else VM_BUG_ON(!PageLocked(old_page)); |
9637a5efd
|
2486 |
|
9637a5efd
|
2487 2488 2489 2490 2491 2492 2493 2494 |
/* * Since we dropped the lock we need to revalidate * the PTE as someone else may have changed it. If * they did, we just return, as we can count on the * MMU to tell us if they didn't also make it writable. */ page_table = pte_offset_map_lock(mm, pmd, address, &ptl); |
b827e496c
|
2495 2496 |
if (!pte_same(*page_table, orig_pte)) { unlock_page(old_page); |
9637a5efd
|
2497 |
goto unlock; |
b827e496c
|
2498 |
} |
a200ee182
|
2499 2500 |
page_mkwrite = 1; |
1da177e4c
|
2501 |
} |
d08b3851d
|
2502 2503 |
dirty_page = old_page; get_page(dirty_page); |
9637a5efd
|
2504 |
|
251b97f55
|
2505 |
reuse: |
9637a5efd
|
2506 2507 2508 |
flush_cache_page(vma, address, pte_pfn(orig_pte)); entry = pte_mkyoung(orig_pte); entry = maybe_mkwrite(pte_mkdirty(entry), vma); |
954ffcb35
|
2509 |
if (ptep_set_access_flags(vma, address, page_table, entry,1)) |
4b3073e1c
|
2510 |
update_mmu_cache(vma, address, page_table); |
72ddc8f72
|
2511 |
pte_unmap_unlock(page_table, ptl); |
9637a5efd
|
2512 |
ret |= VM_FAULT_WRITE; |
72ddc8f72
|
2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 |
if (!dirty_page) return ret; /* * Yes, Virginia, this is actually required to prevent a race * with clear_page_dirty_for_io() from clearing the page dirty * bit after it clear all dirty ptes, but before a racing * do_wp_page installs a dirty pte. * |
a335b2e17
|
2523 |
* __do_fault is protected similarly. |
72ddc8f72
|
2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 |
*/ if (!page_mkwrite) { wait_on_page_locked(dirty_page); set_page_dirty_balance(dirty_page, page_mkwrite); } put_page(dirty_page); if (page_mkwrite) { struct address_space *mapping = dirty_page->mapping; set_page_dirty(dirty_page); unlock_page(dirty_page); page_cache_release(dirty_page); if (mapping) { /* * Some device drivers do not set page.mapping * but still dirty their pages */ balance_dirty_pages_ratelimited(mapping); } } /* file_update_time outside page_lock */ if (vma->vm_file) file_update_time(vma->vm_file); return ret; |
1da177e4c
|
2550 |
} |
1da177e4c
|
2551 2552 2553 2554 |
/* * Ok, we need to copy. Oh, well.. */ |
b5810039a
|
2555 |
page_cache_get(old_page); |
920fc356f
|
2556 |
gotten: |
8f4e2101f
|
2557 |
pte_unmap_unlock(page_table, ptl); |
1da177e4c
|
2558 2559 |
if (unlikely(anon_vma_prepare(vma))) |
65500d234
|
2560 |
goto oom; |
a13ea5b75
|
2561 |
|
62eede62d
|
2562 |
if (is_zero_pfn(pte_pfn(orig_pte))) { |
a13ea5b75
|
2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 |
new_page = alloc_zeroed_user_highpage_movable(vma, address); if (!new_page) goto oom; } else { new_page = alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma, address); if (!new_page) goto oom; cow_user_page(new_page, old_page, address, vma); } __SetPageUptodate(new_page); |
2c26fdd70
|
2573 |
if (mem_cgroup_newpage_charge(new_page, mm, GFP_KERNEL)) |
8a9f3ccd2
|
2574 |
goto oom_free_new; |
1da177e4c
|
2575 2576 2577 |
/* * Re-check the pte - we dropped the lock */ |
8f4e2101f
|
2578 |
page_table = pte_offset_map_lock(mm, pmd, address, &ptl); |
65500d234
|
2579 |
if (likely(pte_same(*page_table, orig_pte))) { |
920fc356f
|
2580 |
if (old_page) { |
920fc356f
|
2581 |
if (!PageAnon(old_page)) { |
34e55232e
|
2582 2583 |
dec_mm_counter_fast(mm, MM_FILEPAGES); inc_mm_counter_fast(mm, MM_ANONPAGES); |
920fc356f
|
2584 2585 |
} } else |
34e55232e
|
2586 |
inc_mm_counter_fast(mm, MM_ANONPAGES); |
eca351336
|
2587 |
flush_cache_page(vma, address, pte_pfn(orig_pte)); |
65500d234
|
2588 2589 |
entry = mk_pte(new_page, vma->vm_page_prot); entry = maybe_mkwrite(pte_mkdirty(entry), vma); |
4ce072f1f
|
2590 2591 2592 2593 2594 2595 |
/* * Clear the pte entry and flush it first, before updating the * pte with the new entry. This will avoid a race condition * seen in the presence of one thread doing SMC and another * thread doing COW. */ |
828502d30
|
2596 |
ptep_clear_flush(vma, address, page_table); |
9617d95e6
|
2597 |
page_add_new_anon_rmap(new_page, vma, address); |
828502d30
|
2598 2599 2600 2601 2602 2603 |
/* * We call the notify macro here because, when using secondary * mmu page tables (such as kvm shadow page tables), we want the * new page to be mapped directly into the secondary page table. */ set_pte_at_notify(mm, address, page_table, entry); |
4b3073e1c
|
2604 |
update_mmu_cache(vma, address, page_table); |
945754a17
|
2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 |
if (old_page) { /* * Only after switching the pte to the new page may * we remove the mapcount here. Otherwise another * process may come and find the rmap count decremented * before the pte is switched to the new page, and * "reuse" the old page writing into it while our pte * here still points into it and can be read by other * threads. * * The critical issue is to order this * page_remove_rmap with the ptp_clear_flush above. * Those stores are ordered by (if nothing else,) * the barrier present in the atomic_add_negative * in page_remove_rmap. * * Then the TLB flush in ptep_clear_flush ensures that * no process can access the old page before the * decremented mapcount is visible. And the old page * cannot be reused until after the decremented * mapcount is visible. So transitively, TLBs to * old page will be flushed before it can be reused. */ |
edc315fd2
|
2628 |
page_remove_rmap(old_page); |
945754a17
|
2629 |
} |
1da177e4c
|
2630 2631 |
/* Free the old page.. */ new_page = old_page; |
f33ea7f40
|
2632 |
ret |= VM_FAULT_WRITE; |
8a9f3ccd2
|
2633 2634 |
} else mem_cgroup_uncharge_page(new_page); |
920fc356f
|
2635 2636 |
if (new_page) page_cache_release(new_page); |
65500d234
|
2637 |
unlock: |
8f4e2101f
|
2638 |
pte_unmap_unlock(page_table, ptl); |
e15f8c01a
|
2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 |
if (old_page) { /* * Don't let another task, with possibly unlocked vma, * keep the mlocked page. */ if ((ret & VM_FAULT_WRITE) && (vma->vm_flags & VM_LOCKED)) { lock_page(old_page); /* LRU manipulation */ munlock_vma_page(old_page); unlock_page(old_page); } page_cache_release(old_page); } |
f33ea7f40
|
2651 |
return ret; |
8a9f3ccd2
|
2652 |
oom_free_new: |
6dbf6d3bb
|
2653 |
page_cache_release(new_page); |
65500d234
|
2654 |
oom: |
b827e496c
|
2655 2656 2657 2658 2659 |
if (old_page) { if (page_mkwrite) { unlock_page(old_page); page_cache_release(old_page); } |
920fc356f
|
2660 |
page_cache_release(old_page); |
b827e496c
|
2661 |
} |
1da177e4c
|
2662 |
return VM_FAULT_OOM; |
9637a5efd
|
2663 2664 2665 |
unwritable_page: page_cache_release(old_page); |
c2ec175c3
|
2666 |
return ret; |
1da177e4c
|
2667 |
} |
97a894136
|
2668 |
static void unmap_mapping_range_vma(struct vm_area_struct *vma, |
1da177e4c
|
2669 2670 2671 |
unsigned long start_addr, unsigned long end_addr, struct zap_details *details) { |
97a894136
|
2672 |
zap_page_range(vma, start_addr, end_addr - start_addr, details); |
1da177e4c
|
2673 2674 2675 2676 2677 2678 2679 2680 |
} static inline void unmap_mapping_range_tree(struct prio_tree_root *root, struct zap_details *details) { struct vm_area_struct *vma; struct prio_tree_iter iter; pgoff_t vba, vea, zba, zea; |
1da177e4c
|
2681 2682 |
vma_prio_tree_foreach(vma, &iter, root, details->first_index, details->last_index) { |
1da177e4c
|
2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 |
vba = vma->vm_pgoff; vea = vba + ((vma->vm_end - vma->vm_start) >> PAGE_SHIFT) - 1; /* Assume for now that PAGE_CACHE_SHIFT == PAGE_SHIFT */ zba = details->first_index; if (zba < vba) zba = vba; zea = details->last_index; if (zea > vea) zea = vea; |
97a894136
|
2693 |
unmap_mapping_range_vma(vma, |
1da177e4c
|
2694 2695 |
((zba - vba) << PAGE_SHIFT) + vma->vm_start, ((zea - vba + 1) << PAGE_SHIFT) + vma->vm_start, |
97a894136
|
2696 |
details); |
1da177e4c
|
2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 |
} } static inline void unmap_mapping_range_list(struct list_head *head, struct zap_details *details) { struct vm_area_struct *vma; /* * In nonlinear VMAs there is no correspondence between virtual address * offset and file offset. So we must perform an exhaustive search * across *all* the pages in each nonlinear VMA, not just the pages * whose virtual address lies outside the file truncation point. */ |
1da177e4c
|
2711 |
list_for_each_entry(vma, head, shared.vm_set.list) { |
1da177e4c
|
2712 |
details->nonlinear_vma = vma; |
97a894136
|
2713 |
unmap_mapping_range_vma(vma, vma->vm_start, vma->vm_end, details); |
1da177e4c
|
2714 2715 2716 2717 |
} } /** |
72fd4a35a
|
2718 |
* unmap_mapping_range - unmap the portion of all mmaps in the specified address_space corresponding to the specified page range in the underlying file. |
3d41088fa
|
2719 |
* @mapping: the address space containing mmaps to be unmapped. |
1da177e4c
|
2720 2721 |
* @holebegin: byte in first page to unmap, relative to the start of * the underlying file. This will be rounded down to a PAGE_SIZE |
25d9e2d15
|
2722 |
* boundary. Note that this is different from truncate_pagecache(), which |
1da177e4c
|
2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 |
* must keep the partial page. In contrast, we must get rid of * partial pages. * @holelen: size of prospective hole in bytes. This will be rounded * up to a PAGE_SIZE boundary. A holelen of zero truncates to the * end of the file. * @even_cows: 1 when truncating a file, unmap even private COWed pages; * but 0 when invalidating pagecache, don't throw away private data. */ void unmap_mapping_range(struct address_space *mapping, loff_t const holebegin, loff_t const holelen, int even_cows) { struct zap_details details; pgoff_t hba = holebegin >> PAGE_SHIFT; pgoff_t hlen = (holelen + PAGE_SIZE - 1) >> PAGE_SHIFT; /* Check for overflow. */ if (sizeof(holelen) > sizeof(hlen)) { long long holeend = (holebegin + holelen + PAGE_SIZE - 1) >> PAGE_SHIFT; if (holeend & ~(long long)ULONG_MAX) hlen = ULONG_MAX - hba + 1; } details.check_mapping = even_cows? NULL: mapping; details.nonlinear_vma = NULL; details.first_index = hba; details.last_index = hba + hlen - 1; if (details.last_index < details.first_index) details.last_index = ULONG_MAX; |
1da177e4c
|
2752 |
|
1da177e4c
|
2753 |
|
3d48ae45e
|
2754 |
mutex_lock(&mapping->i_mmap_mutex); |
1da177e4c
|
2755 2756 2757 2758 |
if (unlikely(!prio_tree_empty(&mapping->i_mmap))) unmap_mapping_range_tree(&mapping->i_mmap, &details); if (unlikely(!list_empty(&mapping->i_mmap_nonlinear))) unmap_mapping_range_list(&mapping->i_mmap_nonlinear, &details); |
3d48ae45e
|
2759 |
mutex_unlock(&mapping->i_mmap_mutex); |
1da177e4c
|
2760 2761 |
} EXPORT_SYMBOL(unmap_mapping_range); |
1da177e4c
|
2762 |
/* |
8f4e2101f
|
2763 2764 2765 |
* We enter with non-exclusive mmap_sem (to exclude vma changes, * but allow concurrent faults), and pte mapped but not yet locked. * We return with mmap_sem still held, but pte unmapped and unlocked. |
1da177e4c
|
2766 |
*/ |
65500d234
|
2767 2768 |
static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, pte_t *page_table, pmd_t *pmd, |
30c9f3a9f
|
2769 |
unsigned int flags, pte_t orig_pte) |
1da177e4c
|
2770 |
{ |
8f4e2101f
|
2771 |
spinlock_t *ptl; |
4969c1192
|
2772 |
struct page *page, *swapcache = NULL; |
65500d234
|
2773 |
swp_entry_t entry; |
1da177e4c
|
2774 |
pte_t pte; |
d065bd810
|
2775 |
int locked; |
56039efa1
|
2776 |
struct mem_cgroup *ptr; |
ad8c2ee80
|
2777 |
int exclusive = 0; |
83c54070e
|
2778 |
int ret = 0; |
1da177e4c
|
2779 |
|
4c21e2f24
|
2780 |
if (!pte_unmap_same(mm, pmd, page_table, orig_pte)) |
8f4e2101f
|
2781 |
goto out; |
65500d234
|
2782 2783 |
entry = pte_to_swp_entry(orig_pte); |
d1737fdbe
|
2784 2785 2786 2787 2788 2789 2790 |
if (unlikely(non_swap_entry(entry))) { if (is_migration_entry(entry)) { migration_entry_wait(mm, pmd, address); } else if (is_hwpoison_entry(entry)) { ret = VM_FAULT_HWPOISON; } else { print_bad_pte(vma, address, orig_pte, NULL); |
d99be1a8e
|
2791 |
ret = VM_FAULT_SIGBUS; |
d1737fdbe
|
2792 |
} |
0697212a4
|
2793 2794 |
goto out; } |
0ff922452
|
2795 |
delayacct_set_flag(DELAYACCT_PF_SWAPIN); |
1da177e4c
|
2796 2797 |
page = lookup_swap_cache(entry); if (!page) { |
a5c9b696e
|
2798 |
grab_swap_token(mm); /* Contend for token _before_ read-in */ |
02098feaa
|
2799 2800 |
page = swapin_readahead(entry, GFP_HIGHUSER_MOVABLE, vma, address); |
1da177e4c
|
2801 2802 |
if (!page) { /* |
8f4e2101f
|
2803 2804 |
* Back out if somebody else faulted in this pte * while we released the pte lock. |
1da177e4c
|
2805 |
*/ |
8f4e2101f
|
2806 |
page_table = pte_offset_map_lock(mm, pmd, address, &ptl); |
1da177e4c
|
2807 2808 |
if (likely(pte_same(*page_table, orig_pte))) ret = VM_FAULT_OOM; |
0ff922452
|
2809 |
delayacct_clear_flag(DELAYACCT_PF_SWAPIN); |
65500d234
|
2810 |
goto unlock; |
1da177e4c
|
2811 2812 2813 2814 |
} /* Had to read the page from swap area: Major fault */ ret = VM_FAULT_MAJOR; |
f8891e5e1
|
2815 |
count_vm_event(PGMAJFAULT); |
456f998ec
|
2816 |
mem_cgroup_count_vm_event(mm, PGMAJFAULT); |
d1737fdbe
|
2817 |
} else if (PageHWPoison(page)) { |
71f72525d
|
2818 2819 2820 2821 |
/* * hwpoisoned dirty swapcache pages are kept for killing * owner processes (which may be unknown at hwpoison time) */ |
d1737fdbe
|
2822 2823 |
ret = VM_FAULT_HWPOISON; delayacct_clear_flag(DELAYACCT_PF_SWAPIN); |
4779cb31c
|
2824 |
goto out_release; |
1da177e4c
|
2825 |
} |
d065bd810
|
2826 |
locked = lock_page_or_retry(page, mm, flags); |
073e587ec
|
2827 |
delayacct_clear_flag(DELAYACCT_PF_SWAPIN); |
d065bd810
|
2828 2829 2830 2831 |
if (!locked) { ret |= VM_FAULT_RETRY; goto out_release; } |
073e587ec
|
2832 |
|
4969c1192
|
2833 |
/* |
31c4a3d3a
|
2834 2835 2836 2837 |
* Make sure try_to_free_swap or reuse_swap_page or swapoff did not * release the swapcache from under us. The page pin, and pte_same * test below, are not enough to exclude that. Even if it is still * swapcache, we need to check that the page's swap has not changed. |
4969c1192
|
2838 |
*/ |
31c4a3d3a
|
2839 |
if (unlikely(!PageSwapCache(page) || page_private(page) != entry.val)) |
4969c1192
|
2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 |
goto out_page; if (ksm_might_need_to_copy(page, vma, address)) { swapcache = page; page = ksm_does_need_to_copy(page, vma, address); if (unlikely(!page)) { ret = VM_FAULT_OOM; page = swapcache; swapcache = NULL; goto out_page; } |
5ad646880
|
2852 |
} |
2c26fdd70
|
2853 |
if (mem_cgroup_try_charge_swapin(mm, page, GFP_KERNEL, &ptr)) { |
8a9f3ccd2
|
2854 |
ret = VM_FAULT_OOM; |
bc43f75cd
|
2855 |
goto out_page; |
8a9f3ccd2
|
2856 |
} |
1da177e4c
|
2857 |
/* |
8f4e2101f
|
2858 |
* Back out if somebody else already faulted in this pte. |
1da177e4c
|
2859 |
*/ |
8f4e2101f
|
2860 |
page_table = pte_offset_map_lock(mm, pmd, address, &ptl); |
9e9bef07c
|
2861 |
if (unlikely(!pte_same(*page_table, orig_pte))) |
b81074800
|
2862 |
goto out_nomap; |
b81074800
|
2863 2864 2865 2866 |
if (unlikely(!PageUptodate(page))) { ret = VM_FAULT_SIGBUS; goto out_nomap; |
1da177e4c
|
2867 |
} |
8c7c6e34a
|
2868 2869 2870 2871 2872 2873 2874 2875 |
/* * The page isn't present yet, go ahead with the fault. * * Be careful about the sequence of operations here. * To get its accounting right, reuse_swap_page() must be called * while the page is counted on swap but not yet in mapcount i.e. * before page_add_anon_rmap() and swap_free(); try_to_free_swap() * must be called after the swap_free(), or it will never succeed. |
03f3c4336
|
2876 2877 2878 2879 |
* Because delete_from_swap_page() may be called by reuse_swap_page(), * mem_cgroup_commit_charge_swapin() may not be able to find swp_entry * in page->private. In this case, a record in swap_cgroup is silently * discarded at swap_free(). |
8c7c6e34a
|
2880 |
*/ |
1da177e4c
|
2881 |
|
34e55232e
|
2882 |
inc_mm_counter_fast(mm, MM_ANONPAGES); |
b084d4353
|
2883 |
dec_mm_counter_fast(mm, MM_SWAPENTS); |
1da177e4c
|
2884 |
pte = mk_pte(page, vma->vm_page_prot); |
30c9f3a9f
|
2885 |
if ((flags & FAULT_FLAG_WRITE) && reuse_swap_page(page)) { |
1da177e4c
|
2886 |
pte = maybe_mkwrite(pte_mkdirty(pte), vma); |
30c9f3a9f
|
2887 |
flags &= ~FAULT_FLAG_WRITE; |
9a5b489b8
|
2888 |
ret |= VM_FAULT_WRITE; |
ad8c2ee80
|
2889 |
exclusive = 1; |
1da177e4c
|
2890 |
} |
1da177e4c
|
2891 2892 |
flush_icache_page(vma, page); set_pte_at(mm, address, page_table, pte); |
ad8c2ee80
|
2893 |
do_page_add_anon_rmap(page, vma, address, exclusive); |
03f3c4336
|
2894 2895 |
/* It's better to call commit-charge after rmap is established */ mem_cgroup_commit_charge_swapin(page, ptr); |
1da177e4c
|
2896 |
|
c475a8ab6
|
2897 |
swap_free(entry); |
b291f0003
|
2898 |
if (vm_swap_full() || (vma->vm_flags & VM_LOCKED) || PageMlocked(page)) |
a2c43eed8
|
2899 |
try_to_free_swap(page); |
c475a8ab6
|
2900 |
unlock_page(page); |
4969c1192
|
2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 |
if (swapcache) { /* * Hold the lock to avoid the swap entry to be reused * until we take the PT lock for the pte_same() check * (to avoid false positives from pte_same). For * further safety release the lock after the swap_free * so that the swap count won't change under a * parallel locked swapcache. */ unlock_page(swapcache); page_cache_release(swapcache); } |
c475a8ab6
|
2913 |
|
30c9f3a9f
|
2914 |
if (flags & FAULT_FLAG_WRITE) { |
61469f1d5
|
2915 2916 2917 |
ret |= do_wp_page(mm, vma, address, page_table, pmd, ptl, pte); if (ret & VM_FAULT_ERROR) ret &= VM_FAULT_ERROR; |
1da177e4c
|
2918 2919 2920 2921 |
goto out; } /* No need to invalidate - it was non-present before */ |
4b3073e1c
|
2922 |
update_mmu_cache(vma, address, page_table); |
65500d234
|
2923 |
unlock: |
8f4e2101f
|
2924 |
pte_unmap_unlock(page_table, ptl); |
1da177e4c
|
2925 2926 |
out: return ret; |
b81074800
|
2927 |
out_nomap: |
7a81b88cb
|
2928 |
mem_cgroup_cancel_charge_swapin(ptr); |
8f4e2101f
|
2929 |
pte_unmap_unlock(page_table, ptl); |
bc43f75cd
|
2930 |
out_page: |
b81074800
|
2931 |
unlock_page(page); |
4779cb31c
|
2932 |
out_release: |
b81074800
|
2933 |
page_cache_release(page); |
4969c1192
|
2934 2935 2936 2937 |
if (swapcache) { unlock_page(swapcache); page_cache_release(swapcache); } |
65500d234
|
2938 |
return ret; |
1da177e4c
|
2939 2940 2941 |
} /* |
8ca3eb080
|
2942 2943 |
* This is like a special single-page "expand_{down|up}wards()", * except we must first make sure that 'address{-|+}PAGE_SIZE' |
320b2b8de
|
2944 |
* doesn't hit another vma. |
320b2b8de
|
2945 2946 2947 2948 2949 |
*/ static inline int check_stack_guard_page(struct vm_area_struct *vma, unsigned long address) { address &= PAGE_MASK; if ((vma->vm_flags & VM_GROWSDOWN) && address == vma->vm_start) { |
0e8e50e20
|
2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 |
struct vm_area_struct *prev = vma->vm_prev; /* * Is there a mapping abutting this one below? * * That's only ok if it's the same stack mapping * that has gotten split.. */ if (prev && prev->vm_end == address) return prev->vm_flags & VM_GROWSDOWN ? 0 : -ENOMEM; |
320b2b8de
|
2960 |
|
d05f3169c
|
2961 |
expand_downwards(vma, address - PAGE_SIZE); |
320b2b8de
|
2962 |
} |
8ca3eb080
|
2963 2964 2965 2966 2967 2968 2969 2970 2971 |
if ((vma->vm_flags & VM_GROWSUP) && address + PAGE_SIZE == vma->vm_end) { struct vm_area_struct *next = vma->vm_next; /* As VM_GROWSDOWN but s/below/above/ */ if (next && next->vm_start == address + PAGE_SIZE) return next->vm_flags & VM_GROWSUP ? 0 : -ENOMEM; expand_upwards(vma, address + PAGE_SIZE); } |
320b2b8de
|
2972 2973 2974 2975 |
return 0; } /* |
8f4e2101f
|
2976 2977 2978 |
* We enter with non-exclusive mmap_sem (to exclude vma changes, * but allow concurrent faults), and pte mapped but not yet locked. * We return with mmap_sem still held, but pte unmapped and unlocked. |
1da177e4c
|
2979 |
*/ |
65500d234
|
2980 2981 |
static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, pte_t *page_table, pmd_t *pmd, |
30c9f3a9f
|
2982 |
unsigned int flags) |
1da177e4c
|
2983 |
{ |
8f4e2101f
|
2984 2985 |
struct page *page; spinlock_t *ptl; |
1da177e4c
|
2986 |
pte_t entry; |
1da177e4c
|
2987 |
|
11ac55247
|
2988 2989 2990 2991 |
pte_unmap(page_table); /* Check if we need to add a guard page to the stack */ if (check_stack_guard_page(vma, address) < 0) |
320b2b8de
|
2992 |
return VM_FAULT_SIGBUS; |
11ac55247
|
2993 |
/* Use the zero-page for reads */ |
62eede62d
|
2994 2995 2996 |
if (!(flags & FAULT_FLAG_WRITE)) { entry = pte_mkspecial(pfn_pte(my_zero_pfn(address), vma->vm_page_prot)); |
11ac55247
|
2997 |
page_table = pte_offset_map_lock(mm, pmd, address, &ptl); |
a13ea5b75
|
2998 2999 3000 3001 |
if (!pte_none(*page_table)) goto unlock; goto setpte; } |
557ed1fa2
|
3002 |
/* Allocate our own private page. */ |
557ed1fa2
|
3003 3004 3005 3006 3007 |
if (unlikely(anon_vma_prepare(vma))) goto oom; page = alloc_zeroed_user_highpage_movable(vma, address); if (!page) goto oom; |
0ed361dec
|
3008 |
__SetPageUptodate(page); |
8f4e2101f
|
3009 |
|
2c26fdd70
|
3010 |
if (mem_cgroup_newpage_charge(page, mm, GFP_KERNEL)) |
8a9f3ccd2
|
3011 |
goto oom_free_page; |
557ed1fa2
|
3012 |
entry = mk_pte(page, vma->vm_page_prot); |
1ac0cb5d0
|
3013 3014 |
if (vma->vm_flags & VM_WRITE) entry = pte_mkwrite(pte_mkdirty(entry)); |
1da177e4c
|
3015 |
|
557ed1fa2
|
3016 |
page_table = pte_offset_map_lock(mm, pmd, address, &ptl); |
1c2fb7a4c
|
3017 |
if (!pte_none(*page_table)) |
557ed1fa2
|
3018 |
goto release; |
9ba692948
|
3019 |
|
34e55232e
|
3020 |
inc_mm_counter_fast(mm, MM_ANONPAGES); |
557ed1fa2
|
3021 |
page_add_new_anon_rmap(page, vma, address); |
a13ea5b75
|
3022 |
setpte: |
65500d234
|
3023 |
set_pte_at(mm, address, page_table, entry); |
1da177e4c
|
3024 3025 |
/* No need to invalidate - it was non-present before */ |
4b3073e1c
|
3026 |
update_mmu_cache(vma, address, page_table); |
65500d234
|
3027 |
unlock: |
8f4e2101f
|
3028 |
pte_unmap_unlock(page_table, ptl); |
83c54070e
|
3029 |
return 0; |
8f4e2101f
|
3030 |
release: |
8a9f3ccd2
|
3031 |
mem_cgroup_uncharge_page(page); |
8f4e2101f
|
3032 3033 |
page_cache_release(page); goto unlock; |
8a9f3ccd2
|
3034 |
oom_free_page: |
6dbf6d3bb
|
3035 |
page_cache_release(page); |
65500d234
|
3036 |
oom: |
1da177e4c
|
3037 3038 3039 3040 |
return VM_FAULT_OOM; } /* |
54cb8821d
|
3041 |
* __do_fault() tries to create a new page mapping. It aggressively |
1da177e4c
|
3042 |
* tries to share with existing pages, but makes a separate copy if |
54cb8821d
|
3043 3044 |
* the FAULT_FLAG_WRITE is set in the flags parameter in order to avoid * the next page fault. |
1da177e4c
|
3045 3046 3047 3048 |
* * As this is called only for pages that do not currently exist, we * do not need to flush old virtual caches or the TLB. * |
8f4e2101f
|
3049 |
* We enter with non-exclusive mmap_sem (to exclude vma changes, |
16abfa086
|
3050 |
* but allow concurrent faults), and pte neither mapped nor locked. |
8f4e2101f
|
3051 |
* We return with mmap_sem still held, but pte unmapped and unlocked. |
1da177e4c
|
3052 |
*/ |
54cb8821d
|
3053 |
static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma, |
16abfa086
|
3054 |
unsigned long address, pmd_t *pmd, |
54cb8821d
|
3055 |
pgoff_t pgoff, unsigned int flags, pte_t orig_pte) |
1da177e4c
|
3056 |
{ |
16abfa086
|
3057 |
pte_t *page_table; |
8f4e2101f
|
3058 |
spinlock_t *ptl; |
d0217ac04
|
3059 |
struct page *page; |
1d65f86db
|
3060 |
struct page *cow_page; |
1da177e4c
|
3061 |
pte_t entry; |
1da177e4c
|
3062 |
int anon = 0; |
d08b3851d
|
3063 |
struct page *dirty_page = NULL; |
d0217ac04
|
3064 3065 |
struct vm_fault vmf; int ret; |
a200ee182
|
3066 |
int page_mkwrite = 0; |
54cb8821d
|
3067 |
|
1d65f86db
|
3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 |
/* * If we do COW later, allocate page befor taking lock_page() * on the file cache page. This will reduce lock holding time. */ if ((flags & FAULT_FLAG_WRITE) && !(vma->vm_flags & VM_SHARED)) { if (unlikely(anon_vma_prepare(vma))) return VM_FAULT_OOM; cow_page = alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma, address); if (!cow_page) return VM_FAULT_OOM; if (mem_cgroup_newpage_charge(cow_page, mm, GFP_KERNEL)) { page_cache_release(cow_page); return VM_FAULT_OOM; } } else cow_page = NULL; |
d0217ac04
|
3087 3088 3089 3090 |
vmf.virtual_address = (void __user *)(address & PAGE_MASK); vmf.pgoff = pgoff; vmf.flags = flags; vmf.page = NULL; |
1da177e4c
|
3091 |
|
3c18ddd16
|
3092 |
ret = vma->vm_ops->fault(vma, &vmf); |
d065bd810
|
3093 3094 |
if (unlikely(ret & (VM_FAULT_ERROR | VM_FAULT_NOPAGE | VM_FAULT_RETRY))) |
1d65f86db
|
3095 |
goto uncharge_out; |
1da177e4c
|
3096 |
|
a3b947eac
|
3097 3098 3099 |
if (unlikely(PageHWPoison(vmf.page))) { if (ret & VM_FAULT_LOCKED) unlock_page(vmf.page); |
1d65f86db
|
3100 3101 |
ret = VM_FAULT_HWPOISON; goto uncharge_out; |
a3b947eac
|
3102 |
} |
d00806b18
|
3103 |
/* |
d0217ac04
|
3104 |
* For consistency in subsequent calls, make the faulted page always |
d00806b18
|
3105 3106 |
* locked. */ |
83c54070e
|
3107 |
if (unlikely(!(ret & VM_FAULT_LOCKED))) |
d0217ac04
|
3108 |
lock_page(vmf.page); |
54cb8821d
|
3109 |
else |
d0217ac04
|
3110 |
VM_BUG_ON(!PageLocked(vmf.page)); |
d00806b18
|
3111 |
|
1da177e4c
|
3112 3113 3114 |
/* * Should we do an early C-O-W break? */ |
d0217ac04
|
3115 |
page = vmf.page; |
54cb8821d
|
3116 |
if (flags & FAULT_FLAG_WRITE) { |
9637a5efd
|
3117 |
if (!(vma->vm_flags & VM_SHARED)) { |
1d65f86db
|
3118 |
page = cow_page; |
54cb8821d
|
3119 |
anon = 1; |
d0217ac04
|
3120 |
copy_user_highpage(page, vmf.page, address, vma); |
0ed361dec
|
3121 |
__SetPageUptodate(page); |
9637a5efd
|
3122 |
} else { |
54cb8821d
|
3123 3124 |
/* * If the page will be shareable, see if the backing |
9637a5efd
|
3125 |
* address space wants to know that the page is about |
54cb8821d
|
3126 3127 |
* to become writable */ |
696761476
|
3128 |
if (vma->vm_ops->page_mkwrite) { |
c2ec175c3
|
3129 |
int tmp; |
696761476
|
3130 |
unlock_page(page); |
b827e496c
|
3131 |
vmf.flags = FAULT_FLAG_WRITE|FAULT_FLAG_MKWRITE; |
c2ec175c3
|
3132 3133 3134 3135 |
tmp = vma->vm_ops->page_mkwrite(vma, &vmf); if (unlikely(tmp & (VM_FAULT_ERROR | VM_FAULT_NOPAGE))) { ret = tmp; |
b827e496c
|
3136 |
goto unwritable_page; |
d0217ac04
|
3137 |
} |
b827e496c
|
3138 3139 3140 3141 3142 3143 3144 3145 3146 |
if (unlikely(!(tmp & VM_FAULT_LOCKED))) { lock_page(page); if (!page->mapping) { ret = 0; /* retry the fault */ unlock_page(page); goto unwritable_page; } } else VM_BUG_ON(!PageLocked(page)); |
a200ee182
|
3147 |
page_mkwrite = 1; |
9637a5efd
|
3148 3149 |
} } |
54cb8821d
|
3150 |
|
1da177e4c
|
3151 |
} |
8f4e2101f
|
3152 |
page_table = pte_offset_map_lock(mm, pmd, address, &ptl); |
1da177e4c
|
3153 3154 3155 3156 3157 3158 |
/* * This silly early PAGE_DIRTY setting removes a race * due to the bad i386 page protection. But it's valid * for other architectures too. * |
30c9f3a9f
|
3159 |
* Note that if FAULT_FLAG_WRITE is set, we either now have |
1da177e4c
|
3160 3161 3162 3163 3164 |
* an exclusive copy of the page, or this is a shared mapping, * so we can make it writable and dirty to avoid having to * handle that later. */ /* Only go through if we didn't race with anybody else... */ |
1c2fb7a4c
|
3165 |
if (likely(pte_same(*page_table, orig_pte))) { |
d00806b18
|
3166 3167 |
flush_icache_page(vma, page); entry = mk_pte(page, vma->vm_page_prot); |
54cb8821d
|
3168 |
if (flags & FAULT_FLAG_WRITE) |
1da177e4c
|
3169 |
entry = maybe_mkwrite(pte_mkdirty(entry), vma); |
1da177e4c
|
3170 |
if (anon) { |
34e55232e
|
3171 |
inc_mm_counter_fast(mm, MM_ANONPAGES); |
64d6519dd
|
3172 |
page_add_new_anon_rmap(page, vma, address); |
f57e88a8d
|
3173 |
} else { |
34e55232e
|
3174 |
inc_mm_counter_fast(mm, MM_FILEPAGES); |
d00806b18
|
3175 |
page_add_file_rmap(page); |
54cb8821d
|
3176 |
if (flags & FAULT_FLAG_WRITE) { |
d00806b18
|
3177 |
dirty_page = page; |
d08b3851d
|
3178 3179 |
get_page(dirty_page); } |
4294621f4
|
3180 |
} |
64d6519dd
|
3181 |
set_pte_at(mm, address, page_table, entry); |
d00806b18
|
3182 3183 |
/* no need to invalidate: a not-present page won't be cached */ |
4b3073e1c
|
3184 |
update_mmu_cache(vma, address, page_table); |
1da177e4c
|
3185 |
} else { |
1d65f86db
|
3186 3187 |
if (cow_page) mem_cgroup_uncharge_page(cow_page); |
d00806b18
|
3188 3189 3190 |
if (anon) page_cache_release(page); else |
54cb8821d
|
3191 |
anon = 1; /* no anon but release faulted_page */ |
1da177e4c
|
3192 |
} |
8f4e2101f
|
3193 |
pte_unmap_unlock(page_table, ptl); |
d00806b18
|
3194 |
|
b827e496c
|
3195 3196 |
if (dirty_page) { struct address_space *mapping = page->mapping; |
8f7b3d156
|
3197 |
|
b827e496c
|
3198 3199 3200 |
if (set_page_dirty(dirty_page)) page_mkwrite = 1; unlock_page(dirty_page); |
d08b3851d
|
3201 |
put_page(dirty_page); |
b827e496c
|
3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 |
if (page_mkwrite && mapping) { /* * Some device drivers do not set page.mapping but still * dirty their pages */ balance_dirty_pages_ratelimited(mapping); } /* file_update_time outside page_lock */ if (vma->vm_file) file_update_time(vma->vm_file); } else { unlock_page(vmf.page); if (anon) page_cache_release(vmf.page); |
d08b3851d
|
3217 |
} |
d00806b18
|
3218 |
|
83c54070e
|
3219 |
return ret; |
b827e496c
|
3220 3221 3222 3223 |
unwritable_page: page_cache_release(page); return ret; |
1d65f86db
|
3224 3225 3226 3227 3228 3229 3230 |
uncharge_out: /* fs's fault handler get error */ if (cow_page) { mem_cgroup_uncharge_page(cow_page); page_cache_release(cow_page); } return ret; |
54cb8821d
|
3231 |
} |
d00806b18
|
3232 |
|
54cb8821d
|
3233 3234 |
static int do_linear_fault(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, pte_t *page_table, pmd_t *pmd, |
30c9f3a9f
|
3235 |
unsigned int flags, pte_t orig_pte) |
54cb8821d
|
3236 3237 |
{ pgoff_t pgoff = (((address & PAGE_MASK) |
0da7e01f5
|
3238 |
- vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff; |
54cb8821d
|
3239 |
|
16abfa086
|
3240 3241 |
pte_unmap(page_table); return __do_fault(mm, vma, address, pmd, pgoff, flags, orig_pte); |
54cb8821d
|
3242 |
} |
f4b81804a
|
3243 |
/* |
1da177e4c
|
3244 3245 3246 |
* Fault of a previously existing named mapping. Repopulate the pte * from the encoded file_pte if possible. This enables swappable * nonlinear vmas. |
8f4e2101f
|
3247 3248 3249 3250 |
* * We enter with non-exclusive mmap_sem (to exclude vma changes, * but allow concurrent faults), and pte mapped but not yet locked. * We return with mmap_sem still held, but pte unmapped and unlocked. |
1da177e4c
|
3251 |
*/ |
d0217ac04
|
3252 |
static int do_nonlinear_fault(struct mm_struct *mm, struct vm_area_struct *vma, |
65500d234
|
3253 |
unsigned long address, pte_t *page_table, pmd_t *pmd, |
30c9f3a9f
|
3254 |
unsigned int flags, pte_t orig_pte) |
1da177e4c
|
3255 |
{ |
65500d234
|
3256 |
pgoff_t pgoff; |
1da177e4c
|
3257 |
|
30c9f3a9f
|
3258 |
flags |= FAULT_FLAG_NONLINEAR; |
4c21e2f24
|
3259 |
if (!pte_unmap_same(mm, pmd, page_table, orig_pte)) |
83c54070e
|
3260 |
return 0; |
1da177e4c
|
3261 |
|
2509ef26d
|
3262 |
if (unlikely(!(vma->vm_flags & VM_NONLINEAR))) { |
65500d234
|
3263 3264 3265 |
/* * Page table corrupted: show pte and kill process. */ |
3dc147414
|
3266 |
print_bad_pte(vma, address, orig_pte, NULL); |
d99be1a8e
|
3267 |
return VM_FAULT_SIGBUS; |
65500d234
|
3268 |
} |
65500d234
|
3269 3270 |
pgoff = pte_to_pgoff(orig_pte); |
16abfa086
|
3271 |
return __do_fault(mm, vma, address, pmd, pgoff, flags, orig_pte); |
1da177e4c
|
3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 |
} /* * These routines also need to handle stuff like marking pages dirty * and/or accessed for architectures that don't do it in hardware (most * RISC architectures). The early dirtying is also good on the i386. * * There is also a hook called "update_mmu_cache()" that architectures * with external mmu caches can use to update those (ie the Sparc or * PowerPC hashed page tables that act as extended TLBs). * |
c74df32c7
|
3283 3284 3285 |
* We enter with non-exclusive mmap_sem (to exclude vma changes, * but allow concurrent faults), and pte mapped but not yet locked. * We return with mmap_sem still held, but pte unmapped and unlocked. |
1da177e4c
|
3286 |
*/ |
71e3aac07
|
3287 3288 3289 |
int handle_pte_fault(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, pte_t *pte, pmd_t *pmd, unsigned int flags) |
1da177e4c
|
3290 3291 |
{ pte_t entry; |
8f4e2101f
|
3292 |
spinlock_t *ptl; |
1da177e4c
|
3293 |
|
8dab5241d
|
3294 |
entry = *pte; |
1da177e4c
|
3295 |
if (!pte_present(entry)) { |
65500d234
|
3296 |
if (pte_none(entry)) { |
f4b81804a
|
3297 |
if (vma->vm_ops) { |
3c18ddd16
|
3298 |
if (likely(vma->vm_ops->fault)) |
54cb8821d
|
3299 |
return do_linear_fault(mm, vma, address, |
30c9f3a9f
|
3300 |
pte, pmd, flags, entry); |
f4b81804a
|
3301 3302 |
} return do_anonymous_page(mm, vma, address, |
30c9f3a9f
|
3303 |
pte, pmd, flags); |
65500d234
|
3304 |
} |
1da177e4c
|
3305 |
if (pte_file(entry)) |
d0217ac04
|
3306 |
return do_nonlinear_fault(mm, vma, address, |
30c9f3a9f
|
3307 |
pte, pmd, flags, entry); |
65500d234
|
3308 |
return do_swap_page(mm, vma, address, |
30c9f3a9f
|
3309 |
pte, pmd, flags, entry); |
1da177e4c
|
3310 |
} |
4c21e2f24
|
3311 |
ptl = pte_lockptr(mm, pmd); |
8f4e2101f
|
3312 3313 3314 |
spin_lock(ptl); if (unlikely(!pte_same(*pte, entry))) goto unlock; |
30c9f3a9f
|
3315 |
if (flags & FAULT_FLAG_WRITE) { |
1da177e4c
|
3316 |
if (!pte_write(entry)) |
8f4e2101f
|
3317 3318 |
return do_wp_page(mm, vma, address, pte, pmd, ptl, entry); |
1da177e4c
|
3319 3320 3321 |
entry = pte_mkdirty(entry); } entry = pte_mkyoung(entry); |
30c9f3a9f
|
3322 |
if (ptep_set_access_flags(vma, address, pte, entry, flags & FAULT_FLAG_WRITE)) { |
4b3073e1c
|
3323 |
update_mmu_cache(vma, address, pte); |
1a44e1490
|
3324 3325 3326 3327 3328 3329 3330 |
} else { /* * This is needed only for protection faults but the arch code * is not yet telling us if this is a protection fault or not. * This still avoids useless tlb flushes for .text page faults * with threads. */ |
30c9f3a9f
|
3331 |
if (flags & FAULT_FLAG_WRITE) |
61c77326d
|
3332 |
flush_tlb_fix_spurious_fault(vma, address); |
1a44e1490
|
3333 |
} |
8f4e2101f
|
3334 3335 |
unlock: pte_unmap_unlock(pte, ptl); |
83c54070e
|
3336 |
return 0; |
1da177e4c
|
3337 3338 3339 3340 3341 |
} /* * By the time we get here, we already hold the mm semaphore */ |
83c54070e
|
3342 |
int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma, |
d06063cc2
|
3343 |
unsigned long address, unsigned int flags) |
1da177e4c
|
3344 3345 3346 3347 3348 3349 3350 |
{ pgd_t *pgd; pud_t *pud; pmd_t *pmd; pte_t *pte; __set_current_state(TASK_RUNNING); |
f8891e5e1
|
3351 |
count_vm_event(PGFAULT); |
456f998ec
|
3352 |
mem_cgroup_count_vm_event(mm, PGFAULT); |
1da177e4c
|
3353 |
|
34e55232e
|
3354 3355 |
/* do counter updates before entering really critical section. */ check_sync_rss_stat(current); |
ac9b9c667
|
3356 |
if (unlikely(is_vm_hugetlb_page(vma))) |
30c9f3a9f
|
3357 |
return hugetlb_fault(mm, vma, address, flags); |
1da177e4c
|
3358 |
|
1da177e4c
|
3359 |
pgd = pgd_offset(mm, address); |
1da177e4c
|
3360 3361 |
pud = pud_alloc(mm, pgd, address); if (!pud) |
c74df32c7
|
3362 |
return VM_FAULT_OOM; |
1da177e4c
|
3363 3364 |
pmd = pmd_alloc(mm, pud, address); if (!pmd) |
c74df32c7
|
3365 |
return VM_FAULT_OOM; |
71e3aac07
|
3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 |
if (pmd_none(*pmd) && transparent_hugepage_enabled(vma)) { if (!vma->vm_ops) return do_huge_pmd_anonymous_page(mm, vma, address, pmd, flags); } else { pmd_t orig_pmd = *pmd; barrier(); if (pmd_trans_huge(orig_pmd)) { if (flags & FAULT_FLAG_WRITE && !pmd_write(orig_pmd) && !pmd_trans_splitting(orig_pmd)) return do_huge_pmd_wp_page(mm, vma, address, pmd, orig_pmd); return 0; } } /* * Use __pte_alloc instead of pte_alloc_map, because we can't * run pte_offset_map on the pmd, if an huge pmd could * materialize from under us from a different thread. */ |
cc03638df
|
3388 |
if (unlikely(pmd_none(*pmd)) && __pte_alloc(mm, vma, pmd, address)) |
c74df32c7
|
3389 |
return VM_FAULT_OOM; |
71e3aac07
|
3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 |
/* if an huge pmd materialized from under us just retry later */ if (unlikely(pmd_trans_huge(*pmd))) return 0; /* * A regular pmd is established and it can't morph into a huge pmd * from under us anymore at this point because we hold the mmap_sem * read mode and khugepaged takes it in write mode. So now it's * safe to run pte_offset_map(). */ pte = pte_offset_map(pmd, address); |
1da177e4c
|
3400 |
|
30c9f3a9f
|
3401 |
return handle_pte_fault(mm, vma, address, pte, pmd, flags); |
1da177e4c
|
3402 3403 3404 3405 3406 |
} #ifndef __PAGETABLE_PUD_FOLDED /* * Allocate page upper directory. |
872fec16d
|
3407 |
* We've already handled the fast-path in-line. |
1da177e4c
|
3408 |
*/ |
1bb3630e8
|
3409 |
int __pud_alloc(struct mm_struct *mm, pgd_t *pgd, unsigned long address) |
1da177e4c
|
3410 |
{ |
c74df32c7
|
3411 3412 |
pud_t *new = pud_alloc_one(mm, address); if (!new) |
1bb3630e8
|
3413 |
return -ENOMEM; |
1da177e4c
|
3414 |
|
362a61ad6
|
3415 |
smp_wmb(); /* See comment in __pte_alloc */ |
872fec16d
|
3416 |
spin_lock(&mm->page_table_lock); |
1bb3630e8
|
3417 |
if (pgd_present(*pgd)) /* Another has populated it */ |
5e5419734
|
3418 |
pud_free(mm, new); |
1bb3630e8
|
3419 3420 |
else pgd_populate(mm, pgd, new); |
c74df32c7
|
3421 |
spin_unlock(&mm->page_table_lock); |
1bb3630e8
|
3422 |
return 0; |
1da177e4c
|
3423 3424 3425 3426 3427 3428 |
} #endif /* __PAGETABLE_PUD_FOLDED */ #ifndef __PAGETABLE_PMD_FOLDED /* * Allocate page middle directory. |
872fec16d
|
3429 |
* We've already handled the fast-path in-line. |
1da177e4c
|
3430 |
*/ |
1bb3630e8
|
3431 |
int __pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address) |
1da177e4c
|
3432 |
{ |
c74df32c7
|
3433 3434 |
pmd_t *new = pmd_alloc_one(mm, address); if (!new) |
1bb3630e8
|
3435 |
return -ENOMEM; |
1da177e4c
|
3436 |
|
362a61ad6
|
3437 |
smp_wmb(); /* See comment in __pte_alloc */ |
872fec16d
|
3438 |
spin_lock(&mm->page_table_lock); |
1da177e4c
|
3439 |
#ifndef __ARCH_HAS_4LEVEL_HACK |
1bb3630e8
|
3440 |
if (pud_present(*pud)) /* Another has populated it */ |
5e5419734
|
3441 |
pmd_free(mm, new); |
1bb3630e8
|
3442 3443 |
else pud_populate(mm, pud, new); |
1da177e4c
|
3444 |
#else |
1bb3630e8
|
3445 |
if (pgd_present(*pud)) /* Another has populated it */ |
5e5419734
|
3446 |
pmd_free(mm, new); |
1bb3630e8
|
3447 3448 |
else pgd_populate(mm, pud, new); |
1da177e4c
|
3449 |
#endif /* __ARCH_HAS_4LEVEL_HACK */ |
c74df32c7
|
3450 |
spin_unlock(&mm->page_table_lock); |
1bb3630e8
|
3451 |
return 0; |
e0f39591c
|
3452 |
} |
1da177e4c
|
3453 3454 3455 3456 3457 3458 3459 3460 3461 |
#endif /* __PAGETABLE_PMD_FOLDED */ int make_pages_present(unsigned long addr, unsigned long end) { int ret, len, write; struct vm_area_struct * vma; vma = find_vma(current->mm, addr); if (!vma) |
a477097d9
|
3462 |
return -ENOMEM; |
5ecfda041
|
3463 3464 3465 3466 3467 3468 |
/* * We want to touch writable mappings with a write fault in order * to break COW, except for shared mappings because these don't COW * and we would not want to dirty them for nothing. */ write = (vma->vm_flags & (VM_WRITE | VM_SHARED)) == VM_WRITE; |
5bcb28b13
|
3469 3470 |
BUG_ON(addr >= end); BUG_ON(end > vma->vm_end); |
68e116a3b
|
3471 |
len = DIV_ROUND_UP(end, PAGE_SIZE) - addr/PAGE_SIZE; |
1da177e4c
|
3472 3473 |
ret = get_user_pages(current, current->mm, addr, len, write, 0, NULL, NULL); |
c11d69d8c
|
3474 |
if (ret < 0) |
1da177e4c
|
3475 |
return ret; |
9978ad583
|
3476 |
return ret == len ? 0 : -EFAULT; |
1da177e4c
|
3477 |
} |
1da177e4c
|
3478 3479 3480 |
#if !defined(__HAVE_ARCH_GATE_AREA) #if defined(AT_SYSINFO_EHDR) |
5ce7852cd
|
3481 |
static struct vm_area_struct gate_vma; |
1da177e4c
|
3482 3483 3484 3485 3486 3487 |
static int __init gate_vma_init(void) { gate_vma.vm_mm = NULL; gate_vma.vm_start = FIXADDR_USER_START; gate_vma.vm_end = FIXADDR_USER_END; |
b6558c4a2
|
3488 3489 |
gate_vma.vm_flags = VM_READ | VM_MAYREAD | VM_EXEC | VM_MAYEXEC; gate_vma.vm_page_prot = __P101; |
f47aef55d
|
3490 3491 3492 3493 3494 3495 3496 |
/* * Make sure the vDSO gets into every core dump. * Dumping its contents makes post-mortem fully interpretable later * without matching up the same kernel and hardware config to see * what PC values meant. */ gate_vma.vm_flags |= VM_ALWAYSDUMP; |
1da177e4c
|
3497 3498 3499 3500 |
return 0; } __initcall(gate_vma_init); #endif |
31db58b3a
|
3501 |
struct vm_area_struct *get_gate_vma(struct mm_struct *mm) |
1da177e4c
|
3502 3503 3504 3505 3506 3507 3508 |
{ #ifdef AT_SYSINFO_EHDR return &gate_vma; #else return NULL; #endif } |
cae5d3903
|
3509 |
int in_gate_area_no_mm(unsigned long addr) |
1da177e4c
|
3510 3511 3512 3513 3514 3515 3516 3517 3518 |
{ #ifdef AT_SYSINFO_EHDR if ((addr >= FIXADDR_USER_START) && (addr < FIXADDR_USER_END)) return 1; #endif return 0; } #endif /* __HAVE_ARCH_GATE_AREA */ |
0ec76a110
|
3519 |
|
1b36ba815
|
3520 |
static int __follow_pte(struct mm_struct *mm, unsigned long address, |
f8ad0f499
|
3521 3522 3523 3524 3525 3526 3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 |
pte_t **ptepp, spinlock_t **ptlp) { pgd_t *pgd; pud_t *pud; pmd_t *pmd; pte_t *ptep; pgd = pgd_offset(mm, address); if (pgd_none(*pgd) || unlikely(pgd_bad(*pgd))) goto out; pud = pud_offset(pgd, address); if (pud_none(*pud) || unlikely(pud_bad(*pud))) goto out; pmd = pmd_offset(pud, address); |
f66055ab6
|
3537 |
VM_BUG_ON(pmd_trans_huge(*pmd)); |
f8ad0f499
|
3538 3539 3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 |
if (pmd_none(*pmd) || unlikely(pmd_bad(*pmd))) goto out; /* We cannot handle huge page PFN maps. Luckily they don't exist. */ if (pmd_huge(*pmd)) goto out; ptep = pte_offset_map_lock(mm, pmd, address, ptlp); if (!ptep) goto out; if (!pte_present(*ptep)) goto unlock; *ptepp = ptep; return 0; unlock: pte_unmap_unlock(ptep, *ptlp); out: return -EINVAL; } |
1b36ba815
|
3557 3558 3559 3560 3561 3562 3563 3564 3565 3566 |
static inline int follow_pte(struct mm_struct *mm, unsigned long address, pte_t **ptepp, spinlock_t **ptlp) { int res; /* (void) is needed to make gcc happy */ (void) __cond_lock(*ptlp, !(res = __follow_pte(mm, address, ptepp, ptlp))); return res; } |
3b6748e2d
|
3567 3568 3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 |
/** * follow_pfn - look up PFN at a user virtual address * @vma: memory mapping * @address: user virtual address * @pfn: location to store found PFN * * Only IO mappings and raw PFN mappings are allowed. * * Returns zero and the pfn at @pfn on success, -ve otherwise. */ int follow_pfn(struct vm_area_struct *vma, unsigned long address, unsigned long *pfn) { int ret = -EINVAL; spinlock_t *ptl; pte_t *ptep; if (!(vma->vm_flags & (VM_IO | VM_PFNMAP))) return ret; ret = follow_pte(vma->vm_mm, address, &ptep, &ptl); if (ret) return ret; *pfn = pte_pfn(*ptep); pte_unmap_unlock(ptep, ptl); return 0; } EXPORT_SYMBOL(follow_pfn); |
28b2ee20c
|
3595 |
#ifdef CONFIG_HAVE_IOREMAP_PROT |
d87fe6607
|
3596 3597 3598 |
int follow_phys(struct vm_area_struct *vma, unsigned long address, unsigned int flags, unsigned long *prot, resource_size_t *phys) |
28b2ee20c
|
3599 |
{ |
03668a4de
|
3600 |
int ret = -EINVAL; |
28b2ee20c
|
3601 3602 |
pte_t *ptep, pte; spinlock_t *ptl; |
28b2ee20c
|
3603 |
|
d87fe6607
|
3604 3605 |
if (!(vma->vm_flags & (VM_IO | VM_PFNMAP))) goto out; |
28b2ee20c
|
3606 |
|
03668a4de
|
3607 |
if (follow_pte(vma->vm_mm, address, &ptep, &ptl)) |
d87fe6607
|
3608 |
goto out; |
28b2ee20c
|
3609 |
pte = *ptep; |
03668a4de
|
3610 |
|
28b2ee20c
|
3611 3612 |
if ((flags & FOLL_WRITE) && !pte_write(pte)) goto unlock; |
28b2ee20c
|
3613 3614 |
*prot = pgprot_val(pte_pgprot(pte)); |
03668a4de
|
3615 |
*phys = (resource_size_t)pte_pfn(pte) << PAGE_SHIFT; |
28b2ee20c
|
3616 |
|
03668a4de
|
3617 |
ret = 0; |
28b2ee20c
|
3618 3619 3620 |
unlock: pte_unmap_unlock(ptep, ptl); out: |
d87fe6607
|
3621 |
return ret; |
28b2ee20c
|
3622 3623 3624 3625 3626 3627 3628 |
} int generic_access_phys(struct vm_area_struct *vma, unsigned long addr, void *buf, int len, int write) { resource_size_t phys_addr; unsigned long prot = 0; |
2bc7273b0
|
3629 |
void __iomem *maddr; |
28b2ee20c
|
3630 |
int offset = addr & (PAGE_SIZE-1); |
d87fe6607
|
3631 |
if (follow_phys(vma, addr, write, &prot, &phys_addr)) |
28b2ee20c
|
3632 3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 |
return -EINVAL; maddr = ioremap_prot(phys_addr, PAGE_SIZE, prot); if (write) memcpy_toio(maddr + offset, buf, len); else memcpy_fromio(buf, maddr + offset, len); iounmap(maddr); return len; } #endif |
0ec76a110
|
3644 |
/* |
206cb6365
|
3645 3646 |
* Access another process' address space as given in mm. If non-NULL, use the * given task for page fault accounting. |
0ec76a110
|
3647 |
*/ |
206cb6365
|
3648 3649 |
static int __access_remote_vm(struct task_struct *tsk, struct mm_struct *mm, unsigned long addr, void *buf, int len, int write) |
0ec76a110
|
3650 |
{ |
0ec76a110
|
3651 |
struct vm_area_struct *vma; |
0ec76a110
|
3652 |
void *old_buf = buf; |
0ec76a110
|
3653 |
down_read(&mm->mmap_sem); |
183ff22bb
|
3654 |
/* ignore errors, just check how much was successfully transferred */ |
0ec76a110
|
3655 3656 3657 |
while (len) { int bytes, ret, offset; void *maddr; |
28b2ee20c
|
3658 |
struct page *page = NULL; |
0ec76a110
|
3659 3660 3661 |
ret = get_user_pages(tsk, mm, addr, 1, write, 1, &page, &vma); |
28b2ee20c
|
3662 3663 3664 3665 3666 3667 3668 |
if (ret <= 0) { /* * Check if this is a VM_IO | VM_PFNMAP VMA, which * we can access using slightly different code. */ #ifdef CONFIG_HAVE_IOREMAP_PROT vma = find_vma(mm, addr); |
fe936dfc2
|
3669 |
if (!vma || vma->vm_start > addr) |
28b2ee20c
|
3670 3671 3672 3673 3674 3675 3676 3677 |
break; if (vma->vm_ops && vma->vm_ops->access) ret = vma->vm_ops->access(vma, addr, buf, len, write); if (ret <= 0) #endif break; bytes = ret; |
0ec76a110
|
3678 |
} else { |
28b2ee20c
|
3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 |
bytes = len; offset = addr & (PAGE_SIZE-1); if (bytes > PAGE_SIZE-offset) bytes = PAGE_SIZE-offset; maddr = kmap(page); if (write) { copy_to_user_page(vma, page, addr, maddr + offset, buf, bytes); set_page_dirty_lock(page); } else { copy_from_user_page(vma, page, addr, buf, maddr + offset, bytes); } kunmap(page); page_cache_release(page); |
0ec76a110
|
3695 |
} |
0ec76a110
|
3696 3697 3698 3699 3700 |
len -= bytes; buf += bytes; addr += bytes; } up_read(&mm->mmap_sem); |
0ec76a110
|
3701 3702 3703 |
return buf - old_buf; } |
03252919b
|
3704 |
|
5ddd36b9c
|
3705 |
/** |
ae91dbfc9
|
3706 |
* access_remote_vm - access another process' address space |
5ddd36b9c
|
3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 |
* @mm: the mm_struct of the target address space * @addr: start address to access * @buf: source or destination buffer * @len: number of bytes to transfer * @write: whether the access is a write * * The caller must hold a reference on @mm. */ int access_remote_vm(struct mm_struct *mm, unsigned long addr, void *buf, int len, int write) { return __access_remote_vm(NULL, mm, addr, buf, len, write); } |
03252919b
|
3720 |
/* |
206cb6365
|
3721 3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735 3736 3737 3738 3739 |
* Access another process' address space. * Source/target buffer must be kernel space, * Do not walk the page table directly, use get_user_pages */ int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write) { struct mm_struct *mm; int ret; mm = get_task_mm(tsk); if (!mm) return 0; ret = __access_remote_vm(tsk, mm, addr, buf, len, write); mmput(mm); return ret; } |
03252919b
|
3740 3741 3742 3743 3744 3745 3746 |
/* * Print the name of a VMA. */ void print_vma_addr(char *prefix, unsigned long ip) { struct mm_struct *mm = current->mm; struct vm_area_struct *vma; |
e8bff74af
|
3747 3748 3749 3750 3751 3752 |
/* * Do not print if we are in atomic * contexts (in exception stacks, etc.): */ if (preempt_count()) return; |
03252919b
|
3753 3754 3755 3756 3757 3758 3759 |
down_read(&mm->mmap_sem); vma = find_vma(mm, ip); if (vma && vma->vm_file) { struct file *f = vma->vm_file; char *buf = (char *)__get_free_page(GFP_KERNEL); if (buf) { char *p, *s; |
cf28b4863
|
3760 |
p = d_path(&f->f_path, buf, PAGE_SIZE); |
03252919b
|
3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 |
if (IS_ERR(p)) p = "?"; s = strrchr(p, '/'); if (s) p = s+1; printk("%s%s[%lx+%lx]", prefix, p, vma->vm_start, vma->vm_end - vma->vm_start); free_page((unsigned long)buf); } } up_read(¤t->mm->mmap_sem); } |
3ee1afa30
|
3774 3775 3776 3777 |
#ifdef CONFIG_PROVE_LOCKING void might_fault(void) { |
95156f005
|
3778 3779 3780 3781 3782 3783 3784 3785 |
/* * Some code (nfs/sunrpc) uses socket ops on kernel memory while * holding the mmap_sem, this is safe because kernel memory doesn't * get paged out, therefore we'll never actually fault, and the * below annotations will generate false positives. */ if (segment_eq(get_fs(), KERNEL_DS)) return; |
3ee1afa30
|
3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 |
might_sleep(); /* * it would be nicer only to annotate paths which are not under * pagefault_disable, however that requires a larger audit and * providing helpers like get_user_atomic. */ if (!in_atomic() && current->mm) might_lock_read(¤t->mm->mmap_sem); } EXPORT_SYMBOL(might_fault); #endif |
47ad8475c
|
3797 3798 3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 |
#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLBFS) static void clear_gigantic_page(struct page *page, unsigned long addr, unsigned int pages_per_huge_page) { int i; struct page *p = page; might_sleep(); for (i = 0; i < pages_per_huge_page; i++, p = mem_map_next(p, page, i)) { cond_resched(); clear_user_highpage(p, addr + i * PAGE_SIZE); } } void clear_huge_page(struct page *page, unsigned long addr, unsigned int pages_per_huge_page) { int i; if (unlikely(pages_per_huge_page > MAX_ORDER_NR_PAGES)) { clear_gigantic_page(page, addr, pages_per_huge_page); return; } might_sleep(); for (i = 0; i < pages_per_huge_page; i++) { cond_resched(); clear_user_highpage(page + i, addr + i * PAGE_SIZE); } } static void copy_user_gigantic_page(struct page *dst, struct page *src, unsigned long addr, struct vm_area_struct *vma, unsigned int pages_per_huge_page) { int i; struct page *dst_base = dst; struct page *src_base = src; for (i = 0; i < pages_per_huge_page; ) { cond_resched(); copy_user_highpage(dst, src, addr + i*PAGE_SIZE, vma); i++; dst = mem_map_next(dst, dst_base, i); src = mem_map_next(src, src_base, i); } } void copy_user_huge_page(struct page *dst, struct page *src, unsigned long addr, struct vm_area_struct *vma, unsigned int pages_per_huge_page) { int i; if (unlikely(pages_per_huge_page > MAX_ORDER_NR_PAGES)) { copy_user_gigantic_page(dst, src, addr, vma, pages_per_huge_page); return; } might_sleep(); for (i = 0; i < pages_per_huge_page; i++) { cond_resched(); copy_user_highpage(dst + i, src + i, addr + i*PAGE_SIZE, vma); } } #endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLBFS */ |