Blame view
mm/userfaultfd.c
17.6 KB
20c8ccb19
|
1 |
// SPDX-License-Identifier: GPL-2.0-only |
c1a4de99f
|
2 3 4 5 |
/* * mm/userfaultfd.c * * Copyright (C) 2015 Red Hat, Inc. |
c1a4de99f
|
6 7 8 |
*/ #include <linux/mm.h> |
174cd4b1e
|
9 |
#include <linux/sched/signal.h> |
c1a4de99f
|
10 11 12 13 14 15 |
#include <linux/pagemap.h> #include <linux/rmap.h> #include <linux/swap.h> #include <linux/swapops.h> #include <linux/userfaultfd_k.h> #include <linux/mmu_notifier.h> |
60d4d2d2b
|
16 |
#include <linux/hugetlb.h> |
26071cedc
|
17 |
#include <linux/shmem_fs.h> |
c1a4de99f
|
18 19 |
#include <asm/tlbflush.h> #include "internal.h" |
643aa36ea
|
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 48 |
static __always_inline struct vm_area_struct *find_dst_vma(struct mm_struct *dst_mm, unsigned long dst_start, unsigned long len) { /* * Make sure that the dst range is both valid and fully within a * single existing vma. */ struct vm_area_struct *dst_vma; dst_vma = find_vma(dst_mm, dst_start); if (!dst_vma) return NULL; if (dst_start < dst_vma->vm_start || dst_start + len > dst_vma->vm_end) return NULL; /* * Check the vma is registered in uffd, this is required to * enforce the VM_MAYWRITE check done at uffd registration * time. */ if (!dst_vma->vm_userfaultfd_ctx.ctx) return NULL; return dst_vma; } |
c1a4de99f
|
49 50 51 52 |
static int mcopy_atomic_pte(struct mm_struct *dst_mm, pmd_t *dst_pmd, struct vm_area_struct *dst_vma, unsigned long dst_addr, |
b6ebaedb4
|
53 |
unsigned long src_addr, |
72981e0e7
|
54 55 |
struct page **pagep, bool wp_copy) |
c1a4de99f
|
56 |
{ |
c1a4de99f
|
57 58 |
pte_t _dst_pte, *dst_pte; spinlock_t *ptl; |
c1a4de99f
|
59 60 |
void *page_kaddr; int ret; |
b6ebaedb4
|
61 |
struct page *page; |
e2a50c1f6
|
62 63 |
pgoff_t offset, max_off; struct inode *inode; |
c1a4de99f
|
64 |
|
b6ebaedb4
|
65 66 67 68 69 70 71 72 73 74 75 |
if (!*pagep) { ret = -ENOMEM; page = alloc_page_vma(GFP_HIGHUSER_MOVABLE, dst_vma, dst_addr); if (!page) goto out; page_kaddr = kmap_atomic(page); ret = copy_from_user(page_kaddr, (const void __user *) src_addr, PAGE_SIZE); kunmap_atomic(page_kaddr); |
c1e8d7c6a
|
76 |
/* fallback to copy_from_user outside mmap_lock */ |
b6ebaedb4
|
77 |
if (unlikely(ret)) { |
9e368259a
|
78 |
ret = -ENOENT; |
b6ebaedb4
|
79 80 81 82 83 84 85 86 |
*pagep = page; /* don't free the page */ goto out; } } else { page = *pagep; *pagep = NULL; } |
c1a4de99f
|
87 88 89 |
/* * The memory barrier inside __SetPageUptodate makes sure that |
f4f5329d4
|
90 |
* preceding stores to the page contents become visible before |
c1a4de99f
|
91 92 93 94 95 |
* the set_pte_at() write. */ __SetPageUptodate(page); ret = -ENOMEM; |
d9eb1ea2b
|
96 |
if (mem_cgroup_charge(page, dst_mm, GFP_KERNEL)) |
c1a4de99f
|
97 |
goto out_release; |
72981e0e7
|
98 |
_dst_pte = pte_mkdirty(mk_pte(page, dst_vma->vm_page_prot)); |
292924b26
|
99 100 101 102 103 104 |
if (dst_vma->vm_flags & VM_WRITE) { if (wp_copy) _dst_pte = pte_mkuffd_wp(_dst_pte); else _dst_pte = pte_mkwrite(_dst_pte); } |
c1a4de99f
|
105 |
|
c1a4de99f
|
106 |
dst_pte = pte_offset_map_lock(dst_mm, dst_pmd, dst_addr, &ptl); |
e2a50c1f6
|
107 108 109 110 111 112 113 114 115 116 |
if (dst_vma->vm_file) { /* the shmem MAP_PRIVATE case requires checking the i_size */ inode = dst_vma->vm_file->f_inode; offset = linear_page_index(dst_vma, dst_addr); max_off = DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE); ret = -EFAULT; if (unlikely(offset >= max_off)) goto out_release_uncharge_unlock; } ret = -EEXIST; |
c1a4de99f
|
117 118 119 120 |
if (!pte_none(*dst_pte)) goto out_release_uncharge_unlock; inc_mm_counter(dst_mm, MM_ANONPAGES); |
be5d0a74c
|
121 |
page_add_new_anon_rmap(page, dst_vma, dst_addr, false); |
b518154e5
|
122 |
lru_cache_add_inactive_or_unevictable(page, dst_vma); |
c1a4de99f
|
123 124 125 126 127 128 129 130 131 132 133 134 |
set_pte_at(dst_mm, dst_addr, dst_pte, _dst_pte); /* No need to invalidate - it was non-present before */ update_mmu_cache(dst_vma, dst_addr, dst_pte); pte_unmap_unlock(dst_pte, ptl); ret = 0; out: return ret; out_release_uncharge_unlock: pte_unmap_unlock(dst_pte, ptl); |
c1a4de99f
|
135 |
out_release: |
09cbfeaf1
|
136 |
put_page(page); |
c1a4de99f
|
137 |
goto out; |
c1a4de99f
|
138 139 140 141 142 143 144 145 146 147 |
} static int mfill_zeropage_pte(struct mm_struct *dst_mm, pmd_t *dst_pmd, struct vm_area_struct *dst_vma, unsigned long dst_addr) { pte_t _dst_pte, *dst_pte; spinlock_t *ptl; int ret; |
e2a50c1f6
|
148 149 |
pgoff_t offset, max_off; struct inode *inode; |
c1a4de99f
|
150 151 152 |
_dst_pte = pte_mkspecial(pfn_pte(my_zero_pfn(dst_addr), dst_vma->vm_page_prot)); |
c1a4de99f
|
153 |
dst_pte = pte_offset_map_lock(dst_mm, dst_pmd, dst_addr, &ptl); |
e2a50c1f6
|
154 155 156 157 158 159 160 161 162 163 |
if (dst_vma->vm_file) { /* the shmem MAP_PRIVATE case requires checking the i_size */ inode = dst_vma->vm_file->f_inode; offset = linear_page_index(dst_vma, dst_addr); max_off = DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE); ret = -EFAULT; if (unlikely(offset >= max_off)) goto out_unlock; } ret = -EEXIST; |
c1a4de99f
|
164 165 166 167 168 169 170 171 172 173 174 175 176 177 |
if (!pte_none(*dst_pte)) goto out_unlock; set_pte_at(dst_mm, dst_addr, dst_pte, _dst_pte); /* No need to invalidate - it was non-present before */ update_mmu_cache(dst_vma, dst_addr, dst_pte); ret = 0; out_unlock: pte_unmap_unlock(dst_pte, ptl); return ret; } static pmd_t *mm_alloc_pmd(struct mm_struct *mm, unsigned long address) { pgd_t *pgd; |
c2febafc6
|
178 |
p4d_t *p4d; |
c1a4de99f
|
179 |
pud_t *pud; |
c1a4de99f
|
180 181 |
pgd = pgd_offset(mm, address); |
c2febafc6
|
182 183 184 185 186 187 188 189 190 191 192 193 |
p4d = p4d_alloc(mm, pgd, address); if (!p4d) return NULL; pud = pud_alloc(mm, p4d, address); if (!pud) return NULL; /* * Note that we didn't run this because the pmd was * missing, the *pmd may be already established and in * turn it may also be a trans_huge_pmd. */ return pmd_alloc(mm, pud, address); |
c1a4de99f
|
194 |
} |
60d4d2d2b
|
195 196 197 |
#ifdef CONFIG_HUGETLB_PAGE /* * __mcopy_atomic processing for HUGETLB vmas. Note that this routine is |
c1e8d7c6a
|
198 |
* called with mmap_lock held, it will release mmap_lock before returning. |
60d4d2d2b
|
199 200 201 202 203 204 205 206 |
*/ static __always_inline ssize_t __mcopy_atomic_hugetlb(struct mm_struct *dst_mm, struct vm_area_struct *dst_vma, unsigned long dst_start, unsigned long src_start, unsigned long len, bool zeropage) { |
1c9e8def4
|
207 208 |
int vm_alloc_shared = dst_vma->vm_flags & VM_SHARED; int vm_shared = dst_vma->vm_flags & VM_SHARED; |
60d4d2d2b
|
209 210 211 212 213 |
ssize_t err; pte_t *dst_pte; unsigned long src_addr, dst_addr; long copied; struct page *page; |
60d4d2d2b
|
214 215 216 217 218 219 220 221 222 223 224 225 |
unsigned long vma_hpagesize; pgoff_t idx; u32 hash; struct address_space *mapping; /* * There is no default zero huge page for all huge page sizes as * supported by hugetlb. A PMD_SIZE huge pages may exist as used * by THP. Since we can not reliably insert a zero page, this * feature is not supported. */ if (zeropage) { |
d8ed45c5d
|
226 |
mmap_read_unlock(dst_mm); |
60d4d2d2b
|
227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 |
return -EINVAL; } src_addr = src_start; dst_addr = dst_start; copied = 0; page = NULL; vma_hpagesize = vma_kernel_pagesize(dst_vma); /* * Validate alignment based on huge page size */ err = -EINVAL; if (dst_start & (vma_hpagesize - 1) || len & (vma_hpagesize - 1)) goto out_unlock; retry: /* |
c1e8d7c6a
|
245 |
* On routine entry dst_vma is set. If we had to drop mmap_lock and |
60d4d2d2b
|
246 247 248 |
* retry, dst_vma will be set to NULL and we must lookup again. */ if (!dst_vma) { |
27d02568f
|
249 |
err = -ENOENT; |
643aa36ea
|
250 |
dst_vma = find_dst_vma(dst_mm, dst_start, len); |
60d4d2d2b
|
251 252 |
if (!dst_vma || !is_vm_hugetlb_page(dst_vma)) goto out_unlock; |
1c9e8def4
|
253 |
|
27d02568f
|
254 255 256 |
err = -EINVAL; if (vma_hpagesize != vma_kernel_pagesize(dst_vma)) goto out_unlock; |
1c9e8def4
|
257 |
vm_shared = dst_vma->vm_flags & VM_SHARED; |
60d4d2d2b
|
258 |
} |
60d4d2d2b
|
259 |
/* |
1c9e8def4
|
260 |
* If not shared, ensure the dst_vma has a anon_vma. |
60d4d2d2b
|
261 262 |
*/ err = -ENOMEM; |
1c9e8def4
|
263 264 265 266 |
if (!vm_shared) { if (unlikely(anon_vma_prepare(dst_vma))) goto out_unlock; } |
60d4d2d2b
|
267 |
|
60d4d2d2b
|
268 269 270 271 |
while (src_addr < src_start + len) { pte_t dst_pteval; BUG_ON(dst_addr >= dst_start + len); |
60d4d2d2b
|
272 273 |
/* |
c0d0381ad
|
274 275 276 277 |
* Serialize via i_mmap_rwsem and hugetlb_fault_mutex. * i_mmap_rwsem ensures the dst_pte remains valid even * in the case of shared pmds. fault mutex prevents * races with other faulting threads. |
60d4d2d2b
|
278 |
*/ |
ddeaab32a
|
279 |
mapping = dst_vma->vm_file->f_mapping; |
c0d0381ad
|
280 281 |
i_mmap_lock_read(mapping); idx = linear_page_index(dst_vma, dst_addr); |
188b04a7d
|
282 |
hash = hugetlb_fault_mutex_hash(mapping, idx); |
60d4d2d2b
|
283 284 285 |
mutex_lock(&hugetlb_fault_mutex_table[hash]); err = -ENOMEM; |
4fb07ee65
|
286 |
dst_pte = huge_pte_alloc(dst_mm, dst_addr, vma_hpagesize); |
60d4d2d2b
|
287 288 |
if (!dst_pte) { mutex_unlock(&hugetlb_fault_mutex_table[hash]); |
c0d0381ad
|
289 |
i_mmap_unlock_read(mapping); |
60d4d2d2b
|
290 291 292 293 294 295 296 |
goto out_unlock; } err = -EEXIST; dst_pteval = huge_ptep_get(dst_pte); if (!huge_pte_none(dst_pteval)) { mutex_unlock(&hugetlb_fault_mutex_table[hash]); |
c0d0381ad
|
297 |
i_mmap_unlock_read(mapping); |
60d4d2d2b
|
298 299 300 301 302 303 304 |
goto out_unlock; } err = hugetlb_mcopy_atomic_pte(dst_mm, dst_pte, dst_vma, dst_addr, src_addr, &page); mutex_unlock(&hugetlb_fault_mutex_table[hash]); |
c0d0381ad
|
305 |
i_mmap_unlock_read(mapping); |
1c9e8def4
|
306 |
vm_alloc_shared = vm_shared; |
60d4d2d2b
|
307 308 |
cond_resched(); |
9e368259a
|
309 |
if (unlikely(err == -ENOENT)) { |
d8ed45c5d
|
310 |
mmap_read_unlock(dst_mm); |
60d4d2d2b
|
311 312 313 314 |
BUG_ON(!page); err = copy_huge_page_from_user(page, (const void __user *)src_addr, |
4fb07ee65
|
315 316 |
vma_hpagesize / PAGE_SIZE, true); |
60d4d2d2b
|
317 318 319 320 |
if (unlikely(err)) { err = -EFAULT; goto out; } |
d8ed45c5d
|
321 |
mmap_read_lock(dst_mm); |
60d4d2d2b
|
322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 |
dst_vma = NULL; goto retry; } else BUG_ON(page); if (!err) { dst_addr += vma_hpagesize; src_addr += vma_hpagesize; copied += vma_hpagesize; if (fatal_signal_pending(current)) err = -EINTR; } if (err) break; } out_unlock: |
d8ed45c5d
|
341 |
mmap_read_unlock(dst_mm); |
60d4d2d2b
|
342 |
out: |
21205bf8f
|
343 344 345 |
if (page) { /* * We encountered an error and are about to free a newly |
1c9e8def4
|
346 347 348 349 350 351 |
* allocated huge page. * * Reservation handling is very subtle, and is different for * private and shared mappings. See the routine * restore_reserve_on_error for details. Unfortunately, we * can not call restore_reserve_on_error now as it would |
c1e8d7c6a
|
352 |
* require holding mmap_lock. |
1c9e8def4
|
353 354 355 356 357 |
* * If a reservation for the page existed in the reservation * map of a private mapping, the map was modified to indicate * the reservation was consumed when the page was allocated. * We clear the PagePrivate flag now so that the global |
21205bf8f
|
358 359 360 |
* reserve count will not be incremented in free_huge_page. * The reservation map will still indicate the reservation * was consumed and possibly prevent later page allocation. |
1c9e8def4
|
361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 |
* This is better than leaking a global reservation. If no * reservation existed, it is still safe to clear PagePrivate * as no adjustments to reservation counts were made during * allocation. * * The reservation map for shared mappings indicates which * pages have reservations. When a huge page is allocated * for an address with a reservation, no change is made to * the reserve map. In this case PagePrivate will be set * to indicate that the global reservation count should be * incremented when the page is freed. This is the desired * behavior. However, when a huge page is allocated for an * address without a reservation a reservation entry is added * to the reservation map, and PagePrivate will not be set. * When the page is freed, the global reserve count will NOT * be incremented and it will appear as though we have leaked * reserved page. In this case, set PagePrivate so that the * global reserve count will be incremented to match the * reservation map entry which was created. * * Note that vm_alloc_shared is based on the flags of the vma * for which the page was originally allocated. dst_vma could * be different or NULL on error. |
21205bf8f
|
384 |
*/ |
1c9e8def4
|
385 386 387 388 |
if (vm_alloc_shared) SetPagePrivate(page); else ClearPagePrivate(page); |
60d4d2d2b
|
389 |
put_page(page); |
21205bf8f
|
390 |
} |
60d4d2d2b
|
391 392 393 394 395 396 397 398 399 400 401 402 403 404 |
BUG_ON(copied < 0); BUG_ON(err > 0); BUG_ON(!copied && !err); return copied ? copied : err; } #else /* !CONFIG_HUGETLB_PAGE */ /* fail at build time if gcc attempts to use this */ extern ssize_t __mcopy_atomic_hugetlb(struct mm_struct *dst_mm, struct vm_area_struct *dst_vma, unsigned long dst_start, unsigned long src_start, unsigned long len, bool zeropage); #endif /* CONFIG_HUGETLB_PAGE */ |
3217d3c79
|
405 406 407 408 409 410 |
static __always_inline ssize_t mfill_atomic_pte(struct mm_struct *dst_mm, pmd_t *dst_pmd, struct vm_area_struct *dst_vma, unsigned long dst_addr, unsigned long src_addr, struct page **page, |
72981e0e7
|
411 412 |
bool zeropage, bool wp_copy) |
3217d3c79
|
413 414 |
{ ssize_t err; |
5b51072e9
|
415 416 417 418 419 420 421 422 423 424 425 |
/* * The normal page fault path for a shmem will invoke the * fault, fill the hole in the file and COW it right away. The * result generates plain anonymous memory. So when we are * asked to fill an hole in a MAP_PRIVATE shmem mapping, we'll * generate anonymous memory directly without actually filling * the hole. For the MAP_PRIVATE case the robustness check * only happens in the pagetable (to verify it's still none) * and not in the radix tree. */ if (!(dst_vma->vm_flags & VM_SHARED)) { |
3217d3c79
|
426 427 |
if (!zeropage) err = mcopy_atomic_pte(dst_mm, dst_pmd, dst_vma, |
72981e0e7
|
428 429 |
dst_addr, src_addr, page, wp_copy); |
3217d3c79
|
430 431 432 433 |
else err = mfill_zeropage_pte(dst_mm, dst_pmd, dst_vma, dst_addr); } else { |
72981e0e7
|
434 |
VM_WARN_ON_ONCE(wp_copy); |
8fb44e540
|
435 |
if (!zeropage) |
3217d3c79
|
436 437 438 |
err = shmem_mcopy_atomic_pte(dst_mm, dst_pmd, dst_vma, dst_addr, src_addr, page); |
8fb44e540
|
439 440 441 |
else err = shmem_mfill_zeropage_pte(dst_mm, dst_pmd, dst_vma, dst_addr); |
3217d3c79
|
442 443 444 445 |
} return err; } |
c1a4de99f
|
446 447 448 449 |
static __always_inline ssize_t __mcopy_atomic(struct mm_struct *dst_mm, unsigned long dst_start, unsigned long src_start, unsigned long len, |
df2cc96e7
|
450 |
bool zeropage, |
72981e0e7
|
451 452 |
bool *mmap_changing, __u64 mode) |
c1a4de99f
|
453 454 455 456 457 |
{ struct vm_area_struct *dst_vma; ssize_t err; pmd_t *dst_pmd; unsigned long src_addr, dst_addr; |
b6ebaedb4
|
458 459 |
long copied; struct page *page; |
72981e0e7
|
460 |
bool wp_copy; |
c1a4de99f
|
461 462 463 464 465 466 467 468 469 470 |
/* * Sanitize the command parameters: */ BUG_ON(dst_start & ~PAGE_MASK); BUG_ON(len & ~PAGE_MASK); /* Does the address range wrap, or is the span zero-sized? */ BUG_ON(src_start + len <= src_start); BUG_ON(dst_start + len <= dst_start); |
b6ebaedb4
|
471 472 473 474 475 |
src_addr = src_start; dst_addr = dst_start; copied = 0; page = NULL; retry: |
d8ed45c5d
|
476 |
mmap_read_lock(dst_mm); |
c1a4de99f
|
477 478 |
/* |
df2cc96e7
|
479 480 481 482 483 484 485 486 487 |
* If memory mappings are changing because of non-cooperative * operation (e.g. mremap) running in parallel, bail out and * request the user to retry later */ err = -EAGAIN; if (mmap_changing && READ_ONCE(*mmap_changing)) goto out_unlock; /* |
c1a4de99f
|
488 489 490 |
* Make sure the vma is not shared, that the dst range is * both valid and fully within a single existing vma. */ |
27d02568f
|
491 |
err = -ENOENT; |
643aa36ea
|
492 |
dst_vma = find_dst_vma(dst_mm, dst_start, len); |
26071cedc
|
493 494 |
if (!dst_vma) goto out_unlock; |
c1a4de99f
|
495 |
|
27d02568f
|
496 497 498 499 500 501 502 503 |
err = -EINVAL; /* * shmem_zero_setup is invoked in mmap for MAP_ANONYMOUS|MAP_SHARED but * it will overwrite vm_ops, so vma_is_anonymous must return false. */ if (WARN_ON_ONCE(vma_is_anonymous(dst_vma) && dst_vma->vm_flags & VM_SHARED)) goto out_unlock; |
c1a4de99f
|
504 |
/* |
72981e0e7
|
505 506 507 508 509 510 511 512 |
* validate 'mode' now that we know the dst_vma: don't allow * a wrprotect copy if the userfaultfd didn't register as WP. */ wp_copy = mode & UFFDIO_COPY_MODE_WP; if (wp_copy && !(dst_vma->vm_flags & VM_UFFD_WP)) goto out_unlock; /* |
60d4d2d2b
|
513 514 515 516 517 |
* If this is a HUGETLB vma, pass off to appropriate routine */ if (is_vm_hugetlb_page(dst_vma)) return __mcopy_atomic_hugetlb(dst_mm, dst_vma, dst_start, src_start, len, zeropage); |
26071cedc
|
518 |
if (!vma_is_anonymous(dst_vma) && !vma_is_shmem(dst_vma)) |
b6ebaedb4
|
519 |
goto out_unlock; |
c1a4de99f
|
520 521 522 523 524 525 526 |
/* * Ensure the dst_vma has a anon_vma or this page * would get a NULL anon_vma when moved in the * dst_vma. */ err = -ENOMEM; |
5b51072e9
|
527 528 |
if (!(dst_vma->vm_flags & VM_SHARED) && unlikely(anon_vma_prepare(dst_vma))) |
b6ebaedb4
|
529 |
goto out_unlock; |
c1a4de99f
|
530 |
|
b6ebaedb4
|
531 |
while (src_addr < src_start + len) { |
c1a4de99f
|
532 |
pmd_t dst_pmdval; |
b6ebaedb4
|
533 |
|
c1a4de99f
|
534 |
BUG_ON(dst_addr >= dst_start + len); |
b6ebaedb4
|
535 |
|
c1a4de99f
|
536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 |
dst_pmd = mm_alloc_pmd(dst_mm, dst_addr); if (unlikely(!dst_pmd)) { err = -ENOMEM; break; } dst_pmdval = pmd_read_atomic(dst_pmd); /* * If the dst_pmd is mapped as THP don't * override it and just be strict. */ if (unlikely(pmd_trans_huge(dst_pmdval))) { err = -EEXIST; break; } if (unlikely(pmd_none(dst_pmdval)) && |
4cf589249
|
552 |
unlikely(__pte_alloc(dst_mm, dst_pmd))) { |
c1a4de99f
|
553 554 555 556 557 558 559 560 561 562 563 |
err = -ENOMEM; break; } /* If an huge pmd materialized from under us fail */ if (unlikely(pmd_trans_huge(*dst_pmd))) { err = -EFAULT; break; } BUG_ON(pmd_none(*dst_pmd)); BUG_ON(pmd_trans_huge(*dst_pmd)); |
3217d3c79
|
564 |
err = mfill_atomic_pte(dst_mm, dst_pmd, dst_vma, dst_addr, |
72981e0e7
|
565 |
src_addr, &page, zeropage, wp_copy); |
c1a4de99f
|
566 |
cond_resched(); |
9e368259a
|
567 |
if (unlikely(err == -ENOENT)) { |
b6ebaedb4
|
568 |
void *page_kaddr; |
d8ed45c5d
|
569 |
mmap_read_unlock(dst_mm); |
b6ebaedb4
|
570 571 572 573 574 575 576 577 578 579 580 581 582 583 |
BUG_ON(!page); page_kaddr = kmap(page); err = copy_from_user(page_kaddr, (const void __user *) src_addr, PAGE_SIZE); kunmap(page); if (unlikely(err)) { err = -EFAULT; goto out; } goto retry; } else BUG_ON(page); |
c1a4de99f
|
584 585 586 587 588 589 590 591 592 593 594 |
if (!err) { dst_addr += PAGE_SIZE; src_addr += PAGE_SIZE; copied += PAGE_SIZE; if (fatal_signal_pending(current)) err = -EINTR; } if (err) break; } |
b6ebaedb4
|
595 |
out_unlock: |
d8ed45c5d
|
596 |
mmap_read_unlock(dst_mm); |
b6ebaedb4
|
597 598 |
out: if (page) |
09cbfeaf1
|
599 |
put_page(page); |
c1a4de99f
|
600 601 602 603 604 605 606 |
BUG_ON(copied < 0); BUG_ON(err > 0); BUG_ON(!copied && !err); return copied ? copied : err; } ssize_t mcopy_atomic(struct mm_struct *dst_mm, unsigned long dst_start, |
df2cc96e7
|
607 |
unsigned long src_start, unsigned long len, |
72981e0e7
|
608 |
bool *mmap_changing, __u64 mode) |
c1a4de99f
|
609 |
{ |
df2cc96e7
|
610 |
return __mcopy_atomic(dst_mm, dst_start, src_start, len, false, |
72981e0e7
|
611 |
mmap_changing, mode); |
c1a4de99f
|
612 613 614 |
} ssize_t mfill_zeropage(struct mm_struct *dst_mm, unsigned long start, |
df2cc96e7
|
615 |
unsigned long len, bool *mmap_changing) |
c1a4de99f
|
616 |
{ |
72981e0e7
|
617 |
return __mcopy_atomic(dst_mm, start, 0, len, true, mmap_changing, 0); |
c1a4de99f
|
618 |
} |
ffd057939
|
619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 |
int mwriteprotect_range(struct mm_struct *dst_mm, unsigned long start, unsigned long len, bool enable_wp, bool *mmap_changing) { struct vm_area_struct *dst_vma; pgprot_t newprot; int err; /* * Sanitize the command parameters: */ BUG_ON(start & ~PAGE_MASK); BUG_ON(len & ~PAGE_MASK); /* Does the address range wrap, or is the span zero-sized? */ BUG_ON(start + len <= start); |
d8ed45c5d
|
635 |
mmap_read_lock(dst_mm); |
ffd057939
|
636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 |
/* * If memory mappings are changing because of non-cooperative * operation (e.g. mremap) running in parallel, bail out and * request the user to retry later */ err = -EAGAIN; if (mmap_changing && READ_ONCE(*mmap_changing)) goto out_unlock; err = -ENOENT; dst_vma = find_dst_vma(dst_mm, start, len); /* * Make sure the vma is not shared, that the dst range is * both valid and fully within a single existing vma. */ if (!dst_vma || (dst_vma->vm_flags & VM_SHARED)) goto out_unlock; if (!userfaultfd_wp(dst_vma)) goto out_unlock; if (!vma_is_anonymous(dst_vma)) goto out_unlock; if (enable_wp) newprot = vm_get_page_prot(dst_vma->vm_flags & ~(VM_WRITE)); else newprot = vm_get_page_prot(dst_vma->vm_flags); change_protection(dst_vma, start, start + len, newprot, enable_wp ? MM_CP_UFFD_WP : MM_CP_UFFD_WP_RESOLVE); err = 0; out_unlock: |
d8ed45c5d
|
669 |
mmap_read_unlock(dst_mm); |
ffd057939
|
670 671 |
return err; } |