Blame view
fs/userfaultfd.c
51.3 KB
20c8ccb19
|
1 |
// SPDX-License-Identifier: GPL-2.0-only |
86039bd3b
|
2 3 4 5 6 7 8 |
/* * fs/userfaultfd.c * * Copyright (C) 2007 Davide Libenzi <davidel@xmailserver.org> * Copyright (C) 2008-2009 Red Hat, Inc. * Copyright (C) 2015 Red Hat, Inc. * |
86039bd3b
|
9 10 11 |
* Some part derived from fs/eventfd.c (anon inode setup) and * mm/ksm.c (mm hashing). */ |
9cd75c3cd
|
12 |
#include <linux/list.h> |
86039bd3b
|
13 |
#include <linux/hashtable.h> |
174cd4b1e
|
14 |
#include <linux/sched/signal.h> |
6e84f3152
|
15 |
#include <linux/sched/mm.h> |
86039bd3b
|
16 17 18 19 20 21 22 23 24 25 26 27 |
#include <linux/mm.h> #include <linux/poll.h> #include <linux/slab.h> #include <linux/seq_file.h> #include <linux/file.h> #include <linux/bug.h> #include <linux/anon_inodes.h> #include <linux/syscalls.h> #include <linux/userfaultfd_k.h> #include <linux/mempolicy.h> #include <linux/ioctl.h> #include <linux/security.h> |
cab350afc
|
28 |
#include <linux/hugetlb.h> |
86039bd3b
|
29 |
|
cefdca0a8
|
30 |
int sysctl_unprivileged_userfaultfd __read_mostly = 1; |
3004ec9ca
|
31 |
static struct kmem_cache *userfaultfd_ctx_cachep __read_mostly; |
86039bd3b
|
32 33 34 35 |
enum userfaultfd_state { UFFD_STATE_WAIT_API, UFFD_STATE_RUNNING, }; |
3004ec9ca
|
36 37 38 |
/* * Start with fault_pending_wqh and fault_wqh so they're more likely * to be in the same cacheline. |
cbcfa130a
|
39 40 41 42 43 44 45 46 47 48 |
* * Locking order: * fd_wqh.lock * fault_pending_wqh.lock * fault_wqh.lock * event_wqh.lock * * To avoid deadlocks, IRQs must be disabled when taking any of the above locks, * since fd_wqh.lock is taken by aio_poll() while it's holding a lock that's * also taken in IRQ context. |
3004ec9ca
|
49 |
*/ |
86039bd3b
|
50 |
struct userfaultfd_ctx { |
15b726ef0
|
51 52 53 |
/* waitqueue head for the pending (i.e. not read) userfaults */ wait_queue_head_t fault_pending_wqh; /* waitqueue head for the userfaults */ |
86039bd3b
|
54 55 56 |
wait_queue_head_t fault_wqh; /* waitqueue head for the pseudo fd to wakeup poll/read */ wait_queue_head_t fd_wqh; |
9cd75c3cd
|
57 58 |
/* waitqueue head for events */ wait_queue_head_t event_wqh; |
2c5b7e1be
|
59 60 |
/* a refile sequence protected by fault_pending_wqh lock */ struct seqcount refile_seq; |
3004ec9ca
|
61 |
/* pseudo fd refcounting */ |
ca8804206
|
62 |
refcount_t refcount; |
86039bd3b
|
63 64 |
/* userfaultfd syscall flags */ unsigned int flags; |
9cd75c3cd
|
65 66 |
/* features requested from the userspace */ unsigned int features; |
86039bd3b
|
67 68 69 70 |
/* state machine */ enum userfaultfd_state state; /* released */ bool released; |
df2cc96e7
|
71 72 |
/* memory mappings are changing because of non-cooperative event */ bool mmap_changing; |
86039bd3b
|
73 74 75 |
/* mm with one ore more vmas attached to this userfaultfd_ctx */ struct mm_struct *mm; }; |
893e26e61
|
76 77 78 79 80 |
struct userfaultfd_fork_ctx { struct userfaultfd_ctx *orig; struct userfaultfd_ctx *new; struct list_head list; }; |
897ab3e0c
|
81 82 83 84 85 86 |
struct userfaultfd_unmap_ctx { struct userfaultfd_ctx *ctx; unsigned long start; unsigned long end; struct list_head list; }; |
86039bd3b
|
87 |
struct userfaultfd_wait_queue { |
a9b85f941
|
88 |
struct uffd_msg msg; |
ac6424b98
|
89 |
wait_queue_entry_t wq; |
86039bd3b
|
90 |
struct userfaultfd_ctx *ctx; |
15a77c6fe
|
91 |
bool waken; |
86039bd3b
|
92 93 94 95 96 97 |
}; struct userfaultfd_wake_range { unsigned long start; unsigned long len; }; |
ac6424b98
|
98 |
static int userfaultfd_wake_function(wait_queue_entry_t *wq, unsigned mode, |
86039bd3b
|
99 100 101 102 103 104 105 106 107 |
int wake_flags, void *key) { struct userfaultfd_wake_range *range = key; int ret; struct userfaultfd_wait_queue *uwq; unsigned long start, len; uwq = container_of(wq, struct userfaultfd_wait_queue, wq); ret = 0; |
86039bd3b
|
108 109 110 |
/* len == 0 means wake all */ start = range->start; len = range->len; |
a9b85f941
|
111 112 |
if (len && (start > uwq->msg.arg.pagefault.address || start + len <= uwq->msg.arg.pagefault.address)) |
86039bd3b
|
113 |
goto out; |
15a77c6fe
|
114 115 |
WRITE_ONCE(uwq->waken, true); /* |
a9668cd6e
|
116 117 |
* The Program-Order guarantees provided by the scheduler * ensure uwq->waken is visible before the task is woken. |
15a77c6fe
|
118 |
*/ |
86039bd3b
|
119 |
ret = wake_up_state(wq->private, mode); |
a9668cd6e
|
120 |
if (ret) { |
86039bd3b
|
121 122 123 |
/* * Wake only once, autoremove behavior. * |
a9668cd6e
|
124 125 126 127 128 129 130 |
* After the effect of list_del_init is visible to the other * CPUs, the waitqueue may disappear from under us, see the * !list_empty_careful() in handle_userfault(). * * try_to_wake_up() has an implicit smp_mb(), and the * wq->private is read before calling the extern function * "wake_up_state" (which in turns calls try_to_wake_up). |
86039bd3b
|
131 |
*/ |
2055da973
|
132 |
list_del_init(&wq->entry); |
a9668cd6e
|
133 |
} |
86039bd3b
|
134 135 136 137 138 139 140 141 |
out: return ret; } /** * userfaultfd_ctx_get - Acquires a reference to the internal userfaultfd * context. * @ctx: [in] Pointer to the userfaultfd context. |
86039bd3b
|
142 143 144 |
*/ static void userfaultfd_ctx_get(struct userfaultfd_ctx *ctx) { |
ca8804206
|
145 |
refcount_inc(&ctx->refcount); |
86039bd3b
|
146 147 148 149 150 151 152 153 154 155 156 157 |
} /** * userfaultfd_ctx_put - Releases a reference to the internal userfaultfd * context. * @ctx: [in] Pointer to userfaultfd context. * * The userfaultfd context reference must have been previously acquired either * with userfaultfd_ctx_get() or userfaultfd_ctx_fdget(). */ static void userfaultfd_ctx_put(struct userfaultfd_ctx *ctx) { |
ca8804206
|
158 |
if (refcount_dec_and_test(&ctx->refcount)) { |
86039bd3b
|
159 160 161 162 |
VM_BUG_ON(spin_is_locked(&ctx->fault_pending_wqh.lock)); VM_BUG_ON(waitqueue_active(&ctx->fault_pending_wqh)); VM_BUG_ON(spin_is_locked(&ctx->fault_wqh.lock)); VM_BUG_ON(waitqueue_active(&ctx->fault_wqh)); |
9cd75c3cd
|
163 164 |
VM_BUG_ON(spin_is_locked(&ctx->event_wqh.lock)); VM_BUG_ON(waitqueue_active(&ctx->event_wqh)); |
86039bd3b
|
165 166 |
VM_BUG_ON(spin_is_locked(&ctx->fd_wqh.lock)); VM_BUG_ON(waitqueue_active(&ctx->fd_wqh)); |
d2005e3f4
|
167 |
mmdrop(ctx->mm); |
3004ec9ca
|
168 |
kmem_cache_free(userfaultfd_ctx_cachep, ctx); |
86039bd3b
|
169 170 |
} } |
a9b85f941
|
171 |
static inline void msg_init(struct uffd_msg *msg) |
86039bd3b
|
172 |
{ |
a9b85f941
|
173 174 175 176 177 178 179 180 181 182 |
BUILD_BUG_ON(sizeof(struct uffd_msg) != 32); /* * Must use memset to zero out the paddings or kernel data is * leaked to userland. */ memset(msg, 0, sizeof(struct uffd_msg)); } static inline struct uffd_msg userfault_msg(unsigned long address, unsigned int flags, |
9d4ac9348
|
183 184 |
unsigned long reason, unsigned int features) |
a9b85f941
|
185 186 187 188 189 |
{ struct uffd_msg msg; msg_init(&msg); msg.event = UFFD_EVENT_PAGEFAULT; msg.arg.pagefault.address = address; |
86039bd3b
|
190 191 |
if (flags & FAULT_FLAG_WRITE) /* |
a4605a61d
|
192 |
* If UFFD_FEATURE_PAGEFAULT_FLAG_WP was set in the |
a9b85f941
|
193 194 195 196 |
* uffdio_api.features and UFFD_PAGEFAULT_FLAG_WRITE * was not set in a UFFD_EVENT_PAGEFAULT, it means it * was a read fault, otherwise if set it means it's * a write fault. |
86039bd3b
|
197 |
*/ |
a9b85f941
|
198 |
msg.arg.pagefault.flags |= UFFD_PAGEFAULT_FLAG_WRITE; |
86039bd3b
|
199 200 |
if (reason & VM_UFFD_WP) /* |
a9b85f941
|
201 202 203 204 205 |
* If UFFD_FEATURE_PAGEFAULT_FLAG_WP was set in the * uffdio_api.features and UFFD_PAGEFAULT_FLAG_WP was * not set in a UFFD_EVENT_PAGEFAULT, it means it was * a missing fault, otherwise if set it means it's a * write protect fault. |
86039bd3b
|
206 |
*/ |
a9b85f941
|
207 |
msg.arg.pagefault.flags |= UFFD_PAGEFAULT_FLAG_WP; |
9d4ac9348
|
208 |
if (features & UFFD_FEATURE_THREAD_ID) |
a36985d31
|
209 |
msg.arg.pagefault.feat.ptid = task_pid_vnr(current); |
a9b85f941
|
210 |
return msg; |
86039bd3b
|
211 |
} |
369cd2121
|
212 213 214 215 216 217 |
#ifdef CONFIG_HUGETLB_PAGE /* * Same functionality as userfaultfd_must_wait below with modifications for * hugepmd ranges. */ static inline bool userfaultfd_huge_must_wait(struct userfaultfd_ctx *ctx, |
7868a2087
|
218 |
struct vm_area_struct *vma, |
369cd2121
|
219 220 221 222 223 |
unsigned long address, unsigned long flags, unsigned long reason) { struct mm_struct *mm = ctx->mm; |
1e2c04362
|
224 |
pte_t *ptep, pte; |
369cd2121
|
225 226 227 |
bool ret = true; VM_BUG_ON(!rwsem_is_locked(&mm->mmap_sem)); |
1e2c04362
|
228 229 230 |
ptep = huge_pte_offset(mm, address, vma_mmu_pagesize(vma)); if (!ptep) |
369cd2121
|
231 232 233 |
goto out; ret = false; |
1e2c04362
|
234 |
pte = huge_ptep_get(ptep); |
369cd2121
|
235 236 237 238 239 |
/* * Lockless access: we're in a wait_event so it's ok if it * changes under us. */ |
1e2c04362
|
240 |
if (huge_pte_none(pte)) |
369cd2121
|
241 |
ret = true; |
1e2c04362
|
242 |
if (!huge_pte_write(pte) && (reason & VM_UFFD_WP)) |
369cd2121
|
243 244 245 246 247 248 |
ret = true; out: return ret; } #else static inline bool userfaultfd_huge_must_wait(struct userfaultfd_ctx *ctx, |
7868a2087
|
249 |
struct vm_area_struct *vma, |
369cd2121
|
250 251 252 253 254 255 256 |
unsigned long address, unsigned long flags, unsigned long reason) { return false; /* should never get here */ } #endif /* CONFIG_HUGETLB_PAGE */ |
86039bd3b
|
257 |
/* |
8d2afd96c
|
258 259 260 261 262 263 264 265 266 267 268 269 270 |
* Verify the pagetables are still not ok after having reigstered into * the fault_pending_wqh to avoid userland having to UFFDIO_WAKE any * userfault that has already been resolved, if userfaultfd_read and * UFFDIO_COPY|ZEROPAGE are being run simultaneously on two different * threads. */ static inline bool userfaultfd_must_wait(struct userfaultfd_ctx *ctx, unsigned long address, unsigned long flags, unsigned long reason) { struct mm_struct *mm = ctx->mm; pgd_t *pgd; |
c2febafc6
|
271 |
p4d_t *p4d; |
8d2afd96c
|
272 273 274 275 276 277 278 279 280 281 |
pud_t *pud; pmd_t *pmd, _pmd; pte_t *pte; bool ret = true; VM_BUG_ON(!rwsem_is_locked(&mm->mmap_sem)); pgd = pgd_offset(mm, address); if (!pgd_present(*pgd)) goto out; |
c2febafc6
|
282 283 284 285 |
p4d = p4d_offset(pgd, address); if (!p4d_present(*p4d)) goto out; pud = pud_offset(p4d, address); |
8d2afd96c
|
286 287 288 289 290 291 292 293 294 295 296 297 |
if (!pud_present(*pud)) goto out; pmd = pmd_offset(pud, address); /* * READ_ONCE must function as a barrier with narrower scope * and it must be equivalent to: * _pmd = *pmd; barrier(); * * This is to deal with the instability (as in * pmd_trans_unstable) of the pmd. */ _pmd = READ_ONCE(*pmd); |
a365ac09d
|
298 |
if (pmd_none(_pmd)) |
8d2afd96c
|
299 300 301 |
goto out; ret = false; |
a365ac09d
|
302 303 |
if (!pmd_present(_pmd)) goto out; |
8d2afd96c
|
304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 |
if (pmd_trans_huge(_pmd)) goto out; /* * the pmd is stable (as in !pmd_trans_unstable) so we can re-read it * and use the standard pte_offset_map() instead of parsing _pmd. */ pte = pte_offset_map(pmd, address); /* * Lockless access: we're in a wait_event so it's ok if it * changes under us. */ if (pte_none(*pte)) ret = true; pte_unmap(pte); out: return ret; } /* |
86039bd3b
|
325 326 327 328 329 330 331 332 333 334 335 336 337 338 |
* The locking rules involved in returning VM_FAULT_RETRY depending on * FAULT_FLAG_ALLOW_RETRY, FAULT_FLAG_RETRY_NOWAIT and * FAULT_FLAG_KILLABLE are not straightforward. The "Caution" * recommendation in __lock_page_or_retry is not an understatement. * * If FAULT_FLAG_ALLOW_RETRY is set, the mmap_sem must be released * before returning VM_FAULT_RETRY only if FAULT_FLAG_RETRY_NOWAIT is * not set. * * If FAULT_FLAG_ALLOW_RETRY is set but FAULT_FLAG_KILLABLE is not * set, VM_FAULT_RETRY can still be returned if and only if there are * fatal_signal_pending()s, and the mmap_sem must be released before * returning it. */ |
2b7403035
|
339 |
vm_fault_t handle_userfault(struct vm_fault *vmf, unsigned long reason) |
86039bd3b
|
340 |
{ |
82b0f8c39
|
341 |
struct mm_struct *mm = vmf->vma->vm_mm; |
86039bd3b
|
342 343 |
struct userfaultfd_ctx *ctx; struct userfaultfd_wait_queue uwq; |
2b7403035
|
344 |
vm_fault_t ret = VM_FAULT_SIGBUS; |
dfa37dc3f
|
345 |
bool must_wait, return_to_userland; |
15a77c6fe
|
346 |
long blocking_state; |
86039bd3b
|
347 |
|
64c2b2030
|
348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 |
/* * We don't do userfault handling for the final child pid update. * * We also don't do userfault handling during * coredumping. hugetlbfs has the special * follow_hugetlb_page() to skip missing pages in the * FOLL_DUMP case, anon memory also checks for FOLL_DUMP with * the no_page_table() helper in follow_page_mask(), but the * shmem_vm_ops->fault method is invoked even during * coredumping without mmap_sem and it ends up here. */ if (current->flags & (PF_EXITING|PF_DUMPCORE)) goto out; /* * Coredumping runs without mmap_sem so we can only check that * the mmap_sem is held, if PF_DUMPCORE was not set. */ WARN_ON_ONCE(!rwsem_is_locked(&mm->mmap_sem)); |
82b0f8c39
|
367 |
ctx = vmf->vma->vm_userfaultfd_ctx.ctx; |
86039bd3b
|
368 |
if (!ctx) |
ba85c702e
|
369 |
goto out; |
86039bd3b
|
370 371 372 373 374 |
BUG_ON(ctx->mm != mm); VM_BUG_ON(reason & ~(VM_UFFD_MISSING|VM_UFFD_WP)); VM_BUG_ON(!(reason & VM_UFFD_MISSING) ^ !!(reason & VM_UFFD_WP)); |
2d6d6f5a0
|
375 376 |
if (ctx->features & UFFD_FEATURE_SIGBUS) goto out; |
86039bd3b
|
377 378 379 380 381 |
/* * If it's already released don't get it. This avoids to loop * in __get_user_pages if userfaultfd_release waits on the * caller of handle_userfault to release the mmap_sem. */ |
6aa7de059
|
382 |
if (unlikely(READ_ONCE(ctx->released))) { |
656710a60
|
383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 |
/* * Don't return VM_FAULT_SIGBUS in this case, so a non * cooperative manager can close the uffd after the * last UFFDIO_COPY, without risking to trigger an * involuntary SIGBUS if the process was starting the * userfaultfd while the userfaultfd was still armed * (but after the last UFFDIO_COPY). If the uffd * wasn't already closed when the userfault reached * this point, that would normally be solved by * userfaultfd_must_wait returning 'false'. * * If we were to return VM_FAULT_SIGBUS here, the non * cooperative manager would be instead forced to * always call UFFDIO_UNREGISTER before it can safely * close the uffd. */ ret = VM_FAULT_NOPAGE; |
ba85c702e
|
400 |
goto out; |
656710a60
|
401 |
} |
86039bd3b
|
402 403 404 405 406 407 408 409 410 411 412 |
/* * Check that we can return VM_FAULT_RETRY. * * NOTE: it should become possible to return VM_FAULT_RETRY * even if FAULT_FLAG_TRIED is set without leading to gup() * -EBUSY failures, if the userfaultfd is to be extended for * VM_UFFD_WP tracking and we intend to arm the userfault * without first stopping userland access to the memory. For * VM_UFFD_MISSING userfaults this is enough for now. */ |
82b0f8c39
|
413 |
if (unlikely(!(vmf->flags & FAULT_FLAG_ALLOW_RETRY))) { |
86039bd3b
|
414 415 416 417 418 |
/* * Validate the invariant that nowait must allow retry * to be sure not to return SIGBUS erroneously on * nowait invocations. */ |
82b0f8c39
|
419 |
BUG_ON(vmf->flags & FAULT_FLAG_RETRY_NOWAIT); |
86039bd3b
|
420 421 422 |
#ifdef CONFIG_DEBUG_VM if (printk_ratelimit()) { printk(KERN_WARNING |
82b0f8c39
|
423 424 425 |
"FAULT_FLAG_ALLOW_RETRY missing %x ", vmf->flags); |
86039bd3b
|
426 427 428 |
dump_stack(); } #endif |
ba85c702e
|
429 |
goto out; |
86039bd3b
|
430 431 432 433 434 435 |
} /* * Handle nowait, not much to do other than tell it to retry * and wait. */ |
ba85c702e
|
436 |
ret = VM_FAULT_RETRY; |
82b0f8c39
|
437 |
if (vmf->flags & FAULT_FLAG_RETRY_NOWAIT) |
ba85c702e
|
438 |
goto out; |
86039bd3b
|
439 440 441 |
/* take the reference before dropping the mmap_sem */ userfaultfd_ctx_get(ctx); |
86039bd3b
|
442 443 |
init_waitqueue_func_entry(&uwq.wq, userfaultfd_wake_function); uwq.wq.private = current; |
9d4ac9348
|
444 445 |
uwq.msg = userfault_msg(vmf->address, vmf->flags, reason, ctx->features); |
86039bd3b
|
446 |
uwq.ctx = ctx; |
15a77c6fe
|
447 |
uwq.waken = false; |
86039bd3b
|
448 |
|
bae473a42
|
449 |
return_to_userland = |
82b0f8c39
|
450 |
(vmf->flags & (FAULT_FLAG_USER|FAULT_FLAG_KILLABLE)) == |
dfa37dc3f
|
451 |
(FAULT_FLAG_USER|FAULT_FLAG_KILLABLE); |
15a77c6fe
|
452 453 |
blocking_state = return_to_userland ? TASK_INTERRUPTIBLE : TASK_KILLABLE; |
dfa37dc3f
|
454 |
|
cbcfa130a
|
455 |
spin_lock_irq(&ctx->fault_pending_wqh.lock); |
86039bd3b
|
456 457 458 459 |
/* * After the __add_wait_queue the uwq is visible to userland * through poll/read(). */ |
15b726ef0
|
460 461 462 463 464 465 |
__add_wait_queue(&ctx->fault_pending_wqh, &uwq.wq); /* * The smp_mb() after __set_current_state prevents the reads * following the spin_unlock to happen before the list_add in * __add_wait_queue. */ |
15a77c6fe
|
466 |
set_current_state(blocking_state); |
cbcfa130a
|
467 |
spin_unlock_irq(&ctx->fault_pending_wqh.lock); |
86039bd3b
|
468 |
|
369cd2121
|
469 470 471 472 |
if (!is_vm_hugetlb_page(vmf->vma)) must_wait = userfaultfd_must_wait(ctx, vmf->address, vmf->flags, reason); else |
7868a2087
|
473 474 |
must_wait = userfaultfd_huge_must_wait(ctx, vmf->vma, vmf->address, |
369cd2121
|
475 |
vmf->flags, reason); |
8d2afd96c
|
476 |
up_read(&mm->mmap_sem); |
6aa7de059
|
477 |
if (likely(must_wait && !READ_ONCE(ctx->released) && |
dfa37dc3f
|
478 479 |
(return_to_userland ? !signal_pending(current) : !fatal_signal_pending(current)))) { |
a9a08845e
|
480 |
wake_up_poll(&ctx->fd_wqh, EPOLLIN); |
86039bd3b
|
481 |
schedule(); |
ba85c702e
|
482 |
ret |= VM_FAULT_MAJOR; |
15a77c6fe
|
483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 |
/* * False wakeups can orginate even from rwsem before * up_read() however userfaults will wait either for a * targeted wakeup on the specific uwq waitqueue from * wake_userfault() or for signals or for uffd * release. */ while (!READ_ONCE(uwq.waken)) { /* * This needs the full smp_store_mb() * guarantee as the state write must be * visible to other CPUs before reading * uwq.waken from other CPUs. */ set_current_state(blocking_state); if (READ_ONCE(uwq.waken) || READ_ONCE(ctx->released) || (return_to_userland ? signal_pending(current) : fatal_signal_pending(current))) break; schedule(); } |
ba85c702e
|
506 |
} |
86039bd3b
|
507 |
|
ba85c702e
|
508 |
__set_current_state(TASK_RUNNING); |
15b726ef0
|
509 |
|
dfa37dc3f
|
510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 |
if (return_to_userland) { if (signal_pending(current) && !fatal_signal_pending(current)) { /* * If we got a SIGSTOP or SIGCONT and this is * a normal userland page fault, just let * userland return so the signal will be * handled and gdb debugging works. The page * fault code immediately after we return from * this function is going to release the * mmap_sem and it's not depending on it * (unlike gup would if we were not to return * VM_FAULT_RETRY). * * If a fatal signal is pending we still take * the streamlined VM_FAULT_RETRY failure path * and there's no need to retake the mmap_sem * in such case. */ down_read(&mm->mmap_sem); |
6bbc4a414
|
530 |
ret = VM_FAULT_NOPAGE; |
dfa37dc3f
|
531 532 |
} } |
15b726ef0
|
533 534 535 536 537 538 539 540 541 542 543 544 545 |
/* * Here we race with the list_del; list_add in * userfaultfd_ctx_read(), however because we don't ever run * list_del_init() to refile across the two lists, the prev * and next pointers will never point to self. list_add also * would never let any of the two pointers to point to * self. So list_empty_careful won't risk to see both pointers * pointing to self at any time during the list refile. The * only case where list_del_init() is called is the full * removal in the wake function and there we don't re-list_add * and it's fine not to block on the spinlock. The uwq on this * kernel stack can be released after the list_del_init. */ |
2055da973
|
546 |
if (!list_empty_careful(&uwq.wq.entry)) { |
cbcfa130a
|
547 |
spin_lock_irq(&ctx->fault_pending_wqh.lock); |
15b726ef0
|
548 549 550 551 |
/* * No need of list_del_init(), the uwq on the stack * will be freed shortly anyway. */ |
2055da973
|
552 |
list_del(&uwq.wq.entry); |
cbcfa130a
|
553 |
spin_unlock_irq(&ctx->fault_pending_wqh.lock); |
86039bd3b
|
554 |
} |
86039bd3b
|
555 556 557 558 559 560 |
/* * ctx may go away after this if the userfault pseudo fd is * already released. */ userfaultfd_ctx_put(ctx); |
ba85c702e
|
561 562 |
out: return ret; |
86039bd3b
|
563 |
} |
8c9e7bb7a
|
564 565 |
static void userfaultfd_event_wait_completion(struct userfaultfd_ctx *ctx, struct userfaultfd_wait_queue *ewq) |
9cd75c3cd
|
566 |
{ |
0cbb4b4f4
|
567 |
struct userfaultfd_ctx *release_new_ctx; |
9a69a829f
|
568 569 |
if (WARN_ON_ONCE(current->flags & PF_EXITING)) goto out; |
9cd75c3cd
|
570 571 572 |
ewq->ctx = ctx; init_waitqueue_entry(&ewq->wq, current); |
0cbb4b4f4
|
573 |
release_new_ctx = NULL; |
9cd75c3cd
|
574 |
|
cbcfa130a
|
575 |
spin_lock_irq(&ctx->event_wqh.lock); |
9cd75c3cd
|
576 577 578 579 580 581 582 583 584 |
/* * After the __add_wait_queue the uwq is visible to userland * through poll/read(). */ __add_wait_queue(&ctx->event_wqh, &ewq->wq); for (;;) { set_current_state(TASK_KILLABLE); if (ewq->msg.event == 0) break; |
6aa7de059
|
585 |
if (READ_ONCE(ctx->released) || |
9cd75c3cd
|
586 |
fatal_signal_pending(current)) { |
384632e67
|
587 588 589 590 591 592 |
/* * &ewq->wq may be queued in fork_event, but * __remove_wait_queue ignores the head * parameter. It would be a problem if it * didn't. */ |
9cd75c3cd
|
593 |
__remove_wait_queue(&ctx->event_wqh, &ewq->wq); |
7eb76d457
|
594 595 596 597 598 599 |
if (ewq->msg.event == UFFD_EVENT_FORK) { struct userfaultfd_ctx *new; new = (struct userfaultfd_ctx *) (unsigned long) ewq->msg.arg.reserved.reserved1; |
0cbb4b4f4
|
600 |
release_new_ctx = new; |
7eb76d457
|
601 |
} |
9cd75c3cd
|
602 603 |
break; } |
cbcfa130a
|
604 |
spin_unlock_irq(&ctx->event_wqh.lock); |
9cd75c3cd
|
605 |
|
a9a08845e
|
606 |
wake_up_poll(&ctx->fd_wqh, EPOLLIN); |
9cd75c3cd
|
607 |
schedule(); |
cbcfa130a
|
608 |
spin_lock_irq(&ctx->event_wqh.lock); |
9cd75c3cd
|
609 610 |
} __set_current_state(TASK_RUNNING); |
cbcfa130a
|
611 |
spin_unlock_irq(&ctx->event_wqh.lock); |
9cd75c3cd
|
612 |
|
0cbb4b4f4
|
613 614 615 616 617 618 |
if (release_new_ctx) { struct vm_area_struct *vma; struct mm_struct *mm = release_new_ctx->mm; /* the various vma->vm_userfaultfd_ctx still points to it */ down_write(&mm->mmap_sem); |
04f5866e4
|
619 620 |
/* no task can run (and in turn coredump) yet */ VM_WARN_ON(!mmget_still_valid(mm)); |
0cbb4b4f4
|
621 |
for (vma = mm->mmap; vma; vma = vma->vm_next) |
31e810aa1
|
622 |
if (vma->vm_userfaultfd_ctx.ctx == release_new_ctx) { |
0cbb4b4f4
|
623 |
vma->vm_userfaultfd_ctx = NULL_VM_UFFD_CTX; |
31e810aa1
|
624 625 |
vma->vm_flags &= ~(VM_UFFD_WP | VM_UFFD_MISSING); } |
0cbb4b4f4
|
626 627 628 629 |
up_write(&mm->mmap_sem); userfaultfd_ctx_put(release_new_ctx); } |
9cd75c3cd
|
630 631 632 633 |
/* * ctx may go away after this if the userfault pseudo fd is * already released. */ |
9a69a829f
|
634 |
out: |
df2cc96e7
|
635 |
WRITE_ONCE(ctx->mmap_changing, false); |
9cd75c3cd
|
636 |
userfaultfd_ctx_put(ctx); |
9cd75c3cd
|
637 638 639 640 641 642 643 644 645 |
} static void userfaultfd_event_complete(struct userfaultfd_ctx *ctx, struct userfaultfd_wait_queue *ewq) { ewq->msg.event = 0; wake_up_locked(&ctx->event_wqh); __remove_wait_queue(&ctx->event_wqh, &ewq->wq); } |
893e26e61
|
646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 |
int dup_userfaultfd(struct vm_area_struct *vma, struct list_head *fcs) { struct userfaultfd_ctx *ctx = NULL, *octx; struct userfaultfd_fork_ctx *fctx; octx = vma->vm_userfaultfd_ctx.ctx; if (!octx || !(octx->features & UFFD_FEATURE_EVENT_FORK)) { vma->vm_userfaultfd_ctx = NULL_VM_UFFD_CTX; vma->vm_flags &= ~(VM_UFFD_WP | VM_UFFD_MISSING); return 0; } list_for_each_entry(fctx, fcs, list) if (fctx->orig == octx) { ctx = fctx->new; break; } if (!ctx) { fctx = kmalloc(sizeof(*fctx), GFP_KERNEL); if (!fctx) return -ENOMEM; ctx = kmem_cache_alloc(userfaultfd_ctx_cachep, GFP_KERNEL); if (!ctx) { kfree(fctx); return -ENOMEM; } |
ca8804206
|
674 |
refcount_set(&ctx->refcount, 1); |
893e26e61
|
675 676 677 678 |
ctx->flags = octx->flags; ctx->state = UFFD_STATE_RUNNING; ctx->features = octx->features; ctx->released = false; |
df2cc96e7
|
679 |
ctx->mmap_changing = false; |
893e26e61
|
680 |
ctx->mm = vma->vm_mm; |
00bb31fa4
|
681 |
mmgrab(ctx->mm); |
893e26e61
|
682 683 |
userfaultfd_ctx_get(octx); |
df2cc96e7
|
684 |
WRITE_ONCE(octx->mmap_changing, true); |
893e26e61
|
685 686 687 688 689 690 691 692 |
fctx->orig = octx; fctx->new = ctx; list_add_tail(&fctx->list, fcs); } vma->vm_userfaultfd_ctx.ctx = ctx; return 0; } |
8c9e7bb7a
|
693 |
static void dup_fctx(struct userfaultfd_fork_ctx *fctx) |
893e26e61
|
694 695 696 697 698 699 700 701 |
{ struct userfaultfd_ctx *ctx = fctx->orig; struct userfaultfd_wait_queue ewq; msg_init(&ewq.msg); ewq.msg.event = UFFD_EVENT_FORK; ewq.msg.arg.reserved.reserved1 = (unsigned long)fctx->new; |
8c9e7bb7a
|
702 |
userfaultfd_event_wait_completion(ctx, &ewq); |
893e26e61
|
703 704 705 706 |
} void dup_userfaultfd_complete(struct list_head *fcs) { |
893e26e61
|
707 708 709 |
struct userfaultfd_fork_ctx *fctx, *n; list_for_each_entry_safe(fctx, n, fcs, list) { |
8c9e7bb7a
|
710 |
dup_fctx(fctx); |
893e26e61
|
711 712 713 714 |
list_del(&fctx->list); kfree(fctx); } } |
72f87654c
|
715 716 717 718 719 720 |
void mremap_userfaultfd_prep(struct vm_area_struct *vma, struct vm_userfaultfd_ctx *vm_ctx) { struct userfaultfd_ctx *ctx; ctx = vma->vm_userfaultfd_ctx.ctx; |
3cfd22be0
|
721 722 723 724 725 |
if (!ctx) return; if (ctx->features & UFFD_FEATURE_EVENT_REMAP) { |
72f87654c
|
726 727 |
vm_ctx->ctx = ctx; userfaultfd_ctx_get(ctx); |
df2cc96e7
|
728 |
WRITE_ONCE(ctx->mmap_changing, true); |
3cfd22be0
|
729 730 731 732 |
} else { /* Drop uffd context if remap feature not enabled */ vma->vm_userfaultfd_ctx = NULL_VM_UFFD_CTX; vma->vm_flags &= ~(VM_UFFD_WP | VM_UFFD_MISSING); |
72f87654c
|
733 734 |
} } |
90794bf19
|
735 |
void mremap_userfaultfd_complete(struct vm_userfaultfd_ctx *vm_ctx, |
72f87654c
|
736 737 738 |
unsigned long from, unsigned long to, unsigned long len) { |
90794bf19
|
739 |
struct userfaultfd_ctx *ctx = vm_ctx->ctx; |
72f87654c
|
740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 |
struct userfaultfd_wait_queue ewq; if (!ctx) return; if (to & ~PAGE_MASK) { userfaultfd_ctx_put(ctx); return; } msg_init(&ewq.msg); ewq.msg.event = UFFD_EVENT_REMAP; ewq.msg.arg.remap.from = from; ewq.msg.arg.remap.to = to; ewq.msg.arg.remap.len = len; userfaultfd_event_wait_completion(ctx, &ewq); } |
70ccb92fd
|
759 |
bool userfaultfd_remove(struct vm_area_struct *vma, |
d811914d8
|
760 |
unsigned long start, unsigned long end) |
05ce77249
|
761 762 763 764 765 766 |
{ struct mm_struct *mm = vma->vm_mm; struct userfaultfd_ctx *ctx; struct userfaultfd_wait_queue ewq; ctx = vma->vm_userfaultfd_ctx.ctx; |
d811914d8
|
767 |
if (!ctx || !(ctx->features & UFFD_FEATURE_EVENT_REMOVE)) |
70ccb92fd
|
768 |
return true; |
05ce77249
|
769 770 |
userfaultfd_ctx_get(ctx); |
df2cc96e7
|
771 |
WRITE_ONCE(ctx->mmap_changing, true); |
05ce77249
|
772 |
up_read(&mm->mmap_sem); |
05ce77249
|
773 |
msg_init(&ewq.msg); |
d811914d8
|
774 775 776 |
ewq.msg.event = UFFD_EVENT_REMOVE; ewq.msg.arg.remove.start = start; ewq.msg.arg.remove.end = end; |
05ce77249
|
777 778 |
userfaultfd_event_wait_completion(ctx, &ewq); |
70ccb92fd
|
779 |
return false; |
05ce77249
|
780 |
} |
897ab3e0c
|
781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 |
static bool has_unmap_ctx(struct userfaultfd_ctx *ctx, struct list_head *unmaps, unsigned long start, unsigned long end) { struct userfaultfd_unmap_ctx *unmap_ctx; list_for_each_entry(unmap_ctx, unmaps, list) if (unmap_ctx->ctx == ctx && unmap_ctx->start == start && unmap_ctx->end == end) return true; return false; } int userfaultfd_unmap_prep(struct vm_area_struct *vma, unsigned long start, unsigned long end, struct list_head *unmaps) { for ( ; vma && vma->vm_start < end; vma = vma->vm_next) { struct userfaultfd_unmap_ctx *unmap_ctx; struct userfaultfd_ctx *ctx = vma->vm_userfaultfd_ctx.ctx; if (!ctx || !(ctx->features & UFFD_FEATURE_EVENT_UNMAP) || has_unmap_ctx(ctx, unmaps, start, end)) continue; unmap_ctx = kzalloc(sizeof(*unmap_ctx), GFP_KERNEL); if (!unmap_ctx) return -ENOMEM; userfaultfd_ctx_get(ctx); |
df2cc96e7
|
811 |
WRITE_ONCE(ctx->mmap_changing, true); |
897ab3e0c
|
812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 |
unmap_ctx->ctx = ctx; unmap_ctx->start = start; unmap_ctx->end = end; list_add_tail(&unmap_ctx->list, unmaps); } return 0; } void userfaultfd_unmap_complete(struct mm_struct *mm, struct list_head *uf) { struct userfaultfd_unmap_ctx *ctx, *n; struct userfaultfd_wait_queue ewq; list_for_each_entry_safe(ctx, n, uf, list) { msg_init(&ewq.msg); ewq.msg.event = UFFD_EVENT_UNMAP; ewq.msg.arg.remove.start = ctx->start; ewq.msg.arg.remove.end = ctx->end; userfaultfd_event_wait_completion(ctx->ctx, &ewq); list_del(&ctx->list); kfree(ctx); } } |
86039bd3b
|
839 840 841 842 843 844 845 846 |
static int userfaultfd_release(struct inode *inode, struct file *file) { struct userfaultfd_ctx *ctx = file->private_data; struct mm_struct *mm = ctx->mm; struct vm_area_struct *vma, *prev; /* len == 0 means wake all */ struct userfaultfd_wake_range range = { .len = 0, }; unsigned long new_flags; |
46d0b24c5
|
847 |
bool still_valid; |
86039bd3b
|
848 |
|
6aa7de059
|
849 |
WRITE_ONCE(ctx->released, true); |
86039bd3b
|
850 |
|
d2005e3f4
|
851 852 |
if (!mmget_not_zero(mm)) goto wakeup; |
86039bd3b
|
853 854 855 856 857 858 859 860 861 |
/* * Flush page faults out of all CPUs. NOTE: all page faults * must be retried without returning VM_FAULT_SIGBUS if * userfaultfd_ctx_get() succeeds but vma->vma_userfault_ctx * changes while handle_userfault released the mmap_sem. So * it's critical that released is set to true (above), before * taking the mmap_sem for writing. */ down_write(&mm->mmap_sem); |
46d0b24c5
|
862 |
still_valid = mmget_still_valid(mm); |
86039bd3b
|
863 864 865 866 867 868 869 870 871 872 |
prev = NULL; for (vma = mm->mmap; vma; vma = vma->vm_next) { cond_resched(); BUG_ON(!!vma->vm_userfaultfd_ctx.ctx ^ !!(vma->vm_flags & (VM_UFFD_MISSING | VM_UFFD_WP))); if (vma->vm_userfaultfd_ctx.ctx != ctx) { prev = vma; continue; } new_flags = vma->vm_flags & ~(VM_UFFD_MISSING | VM_UFFD_WP); |
46d0b24c5
|
873 874 875 876 877 878 879 880 881 882 883 |
if (still_valid) { prev = vma_merge(mm, prev, vma->vm_start, vma->vm_end, new_flags, vma->anon_vma, vma->vm_file, vma->vm_pgoff, vma_policy(vma), NULL_VM_UFFD_CTX); if (prev) vma = prev; else prev = vma; } |
86039bd3b
|
884 885 886 887 |
vma->vm_flags = new_flags; vma->vm_userfaultfd_ctx = NULL_VM_UFFD_CTX; } up_write(&mm->mmap_sem); |
d2005e3f4
|
888 889 |
mmput(mm); wakeup: |
86039bd3b
|
890 |
/* |
15b726ef0
|
891 |
* After no new page faults can wait on this fault_*wqh, flush |
86039bd3b
|
892 |
* the last page faults that may have been already waiting on |
15b726ef0
|
893 |
* the fault_*wqh. |
86039bd3b
|
894 |
*/ |
cbcfa130a
|
895 |
spin_lock_irq(&ctx->fault_pending_wqh.lock); |
ac5be6b47
|
896 |
__wake_up_locked_key(&ctx->fault_pending_wqh, TASK_NORMAL, &range); |
c430d1e84
|
897 |
__wake_up(&ctx->fault_wqh, TASK_NORMAL, 1, &range); |
cbcfa130a
|
898 |
spin_unlock_irq(&ctx->fault_pending_wqh.lock); |
86039bd3b
|
899 |
|
5a18b64e3
|
900 901 |
/* Flush pending events that may still wait on event_wqh */ wake_up_all(&ctx->event_wqh); |
a9a08845e
|
902 |
wake_up_poll(&ctx->fd_wqh, EPOLLHUP); |
86039bd3b
|
903 904 905 |
userfaultfd_ctx_put(ctx); return 0; } |
15b726ef0
|
906 |
/* fault_pending_wqh.lock must be hold by the caller */ |
6dcc27fd3
|
907 908 |
static inline struct userfaultfd_wait_queue *find_userfault_in( wait_queue_head_t *wqh) |
86039bd3b
|
909 |
{ |
ac6424b98
|
910 |
wait_queue_entry_t *wq; |
15b726ef0
|
911 |
struct userfaultfd_wait_queue *uwq; |
86039bd3b
|
912 |
|
456a73789
|
913 |
lockdep_assert_held(&wqh->lock); |
86039bd3b
|
914 |
|
15b726ef0
|
915 |
uwq = NULL; |
6dcc27fd3
|
916 |
if (!waitqueue_active(wqh)) |
15b726ef0
|
917 918 |
goto out; /* walk in reverse to provide FIFO behavior to read userfaults */ |
2055da973
|
919 |
wq = list_last_entry(&wqh->head, typeof(*wq), entry); |
15b726ef0
|
920 921 922 |
uwq = container_of(wq, struct userfaultfd_wait_queue, wq); out: return uwq; |
86039bd3b
|
923 |
} |
6dcc27fd3
|
924 925 926 927 928 929 |
static inline struct userfaultfd_wait_queue *find_userfault( struct userfaultfd_ctx *ctx) { return find_userfault_in(&ctx->fault_pending_wqh); } |
86039bd3b
|
930 |
|
9cd75c3cd
|
931 932 933 934 935 |
static inline struct userfaultfd_wait_queue *find_userfault_evt( struct userfaultfd_ctx *ctx) { return find_userfault_in(&ctx->event_wqh); } |
076ccb76e
|
936 |
static __poll_t userfaultfd_poll(struct file *file, poll_table *wait) |
86039bd3b
|
937 938 |
{ struct userfaultfd_ctx *ctx = file->private_data; |
076ccb76e
|
939 |
__poll_t ret; |
86039bd3b
|
940 941 942 943 944 |
poll_wait(file, &ctx->fd_wqh, wait); switch (ctx->state) { case UFFD_STATE_WAIT_API: |
a9a08845e
|
945 |
return EPOLLERR; |
86039bd3b
|
946 |
case UFFD_STATE_RUNNING: |
ba85c702e
|
947 948 949 950 951 |
/* * poll() never guarantees that read won't block. * userfaults can be waken before they're read(). */ if (unlikely(!(file->f_flags & O_NONBLOCK))) |
a9a08845e
|
952 |
return EPOLLERR; |
15b726ef0
|
953 954 955 956 957 958 959 960 961 962 963 964 965 |
/* * lockless access to see if there are pending faults * __pollwait last action is the add_wait_queue but * the spin_unlock would allow the waitqueue_active to * pass above the actual list_add inside * add_wait_queue critical section. So use a full * memory barrier to serialize the list_add write of * add_wait_queue() with the waitqueue_active read * below. */ ret = 0; smp_mb(); if (waitqueue_active(&ctx->fault_pending_wqh)) |
a9a08845e
|
966 |
ret = EPOLLIN; |
9cd75c3cd
|
967 |
else if (waitqueue_active(&ctx->event_wqh)) |
a9a08845e
|
968 |
ret = EPOLLIN; |
9cd75c3cd
|
969 |
|
86039bd3b
|
970 971 |
return ret; default: |
8474901a3
|
972 |
WARN_ON_ONCE(1); |
a9a08845e
|
973 |
return EPOLLERR; |
86039bd3b
|
974 975 |
} } |
893e26e61
|
976 977 978 979 980 981 982 |
static const struct file_operations userfaultfd_fops; static int resolve_userfault_fork(struct userfaultfd_ctx *ctx, struct userfaultfd_ctx *new, struct uffd_msg *msg) { int fd; |
893e26e61
|
983 |
|
284cd241a
|
984 985 |
fd = anon_inode_getfd("[userfaultfd]", &userfaultfd_fops, new, O_RDWR | (new->flags & UFFD_SHARED_FCNTL_FLAGS)); |
893e26e61
|
986 987 |
if (fd < 0) return fd; |
893e26e61
|
988 989 |
msg->arg.reserved.reserved1 = 0; msg->arg.fork.ufd = fd; |
893e26e61
|
990 991 |
return 0; } |
86039bd3b
|
992 |
static ssize_t userfaultfd_ctx_read(struct userfaultfd_ctx *ctx, int no_wait, |
a9b85f941
|
993 |
struct uffd_msg *msg) |
86039bd3b
|
994 995 996 |
{ ssize_t ret; DECLARE_WAITQUEUE(wait, current); |
15b726ef0
|
997 |
struct userfaultfd_wait_queue *uwq; |
893e26e61
|
998 999 1000 1001 1002 1003 1004 1005 1006 |
/* * Handling fork event requires sleeping operations, so * we drop the event_wqh lock, then do these ops, then * lock it back and wake up the waiter. While the lock is * dropped the ewq may go away so we keep track of it * carefully. */ LIST_HEAD(fork_event); struct userfaultfd_ctx *fork_nctx = NULL; |
86039bd3b
|
1007 |
|
15b726ef0
|
1008 |
/* always take the fd_wqh lock before the fault_pending_wqh lock */ |
ae62c16e1
|
1009 |
spin_lock_irq(&ctx->fd_wqh.lock); |
86039bd3b
|
1010 1011 1012 |
__add_wait_queue(&ctx->fd_wqh, &wait); for (;;) { set_current_state(TASK_INTERRUPTIBLE); |
15b726ef0
|
1013 1014 1015 |
spin_lock(&ctx->fault_pending_wqh.lock); uwq = find_userfault(ctx); if (uwq) { |
86039bd3b
|
1016 |
/* |
2c5b7e1be
|
1017 1018 1019 1020 1021 1022 1023 1024 1025 |
* Use a seqcount to repeat the lockless check * in wake_userfault() to avoid missing * wakeups because during the refile both * waitqueue could become empty if this is the * only userfault. */ write_seqcount_begin(&ctx->refile_seq); /* |
15b726ef0
|
1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 |
* The fault_pending_wqh.lock prevents the uwq * to disappear from under us. * * Refile this userfault from * fault_pending_wqh to fault_wqh, it's not * pending anymore after we read it. * * Use list_del() by hand (as * userfaultfd_wake_function also uses * list_del_init() by hand) to be sure nobody * changes __remove_wait_queue() to use * list_del_init() in turn breaking the * !list_empty_careful() check in |
2055da973
|
1039 |
* handle_userfault(). The uwq->wq.head list |
15b726ef0
|
1040 1041 1042 1043 1044 |
* must never be empty at any time during the * refile, or the waitqueue could disappear * from under us. The "wait_queue_head_t" * parameter of __remove_wait_queue() is unused * anyway. |
86039bd3b
|
1045 |
*/ |
2055da973
|
1046 |
list_del(&uwq->wq.entry); |
c430d1e84
|
1047 |
add_wait_queue(&ctx->fault_wqh, &uwq->wq); |
15b726ef0
|
1048 |
|
2c5b7e1be
|
1049 |
write_seqcount_end(&ctx->refile_seq); |
a9b85f941
|
1050 1051 |
/* careful to always initialize msg if ret == 0 */ *msg = uwq->msg; |
15b726ef0
|
1052 |
spin_unlock(&ctx->fault_pending_wqh.lock); |
86039bd3b
|
1053 1054 1055 |
ret = 0; break; } |
15b726ef0
|
1056 |
spin_unlock(&ctx->fault_pending_wqh.lock); |
9cd75c3cd
|
1057 1058 1059 1060 1061 |
spin_lock(&ctx->event_wqh.lock); uwq = find_userfault_evt(ctx); if (uwq) { *msg = uwq->msg; |
893e26e61
|
1062 1063 1064 1065 |
if (uwq->msg.event == UFFD_EVENT_FORK) { fork_nctx = (struct userfaultfd_ctx *) (unsigned long) uwq->msg.arg.reserved.reserved1; |
2055da973
|
1066 |
list_move(&uwq->wq.entry, &fork_event); |
384632e67
|
1067 1068 1069 1070 1071 1072 |
/* * fork_nctx can be freed as soon as * we drop the lock, unless we take a * reference on it. */ userfaultfd_ctx_get(fork_nctx); |
893e26e61
|
1073 1074 1075 1076 |
spin_unlock(&ctx->event_wqh.lock); ret = 0; break; } |
9cd75c3cd
|
1077 1078 1079 1080 1081 1082 |
userfaultfd_event_complete(ctx, uwq); spin_unlock(&ctx->event_wqh.lock); ret = 0; break; } spin_unlock(&ctx->event_wqh.lock); |
86039bd3b
|
1083 1084 1085 1086 1087 1088 1089 1090 |
if (signal_pending(current)) { ret = -ERESTARTSYS; break; } if (no_wait) { ret = -EAGAIN; break; } |
ae62c16e1
|
1091 |
spin_unlock_irq(&ctx->fd_wqh.lock); |
86039bd3b
|
1092 |
schedule(); |
ae62c16e1
|
1093 |
spin_lock_irq(&ctx->fd_wqh.lock); |
86039bd3b
|
1094 1095 1096 |
} __remove_wait_queue(&ctx->fd_wqh, &wait); __set_current_state(TASK_RUNNING); |
ae62c16e1
|
1097 |
spin_unlock_irq(&ctx->fd_wqh.lock); |
86039bd3b
|
1098 |
|
893e26e61
|
1099 1100 |
if (!ret && msg->event == UFFD_EVENT_FORK) { ret = resolve_userfault_fork(ctx, fork_nctx, msg); |
cbcfa130a
|
1101 |
spin_lock_irq(&ctx->event_wqh.lock); |
384632e67
|
1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 |
if (!list_empty(&fork_event)) { /* * The fork thread didn't abort, so we can * drop the temporary refcount. */ userfaultfd_ctx_put(fork_nctx); uwq = list_first_entry(&fork_event, typeof(*uwq), wq.entry); /* * If fork_event list wasn't empty and in turn * the event wasn't already released by fork * (the event is allocated on fork kernel * stack), put the event back to its place in * the event_wq. fork_event head will be freed * as soon as we return so the event cannot * stay queued there no matter the current * "ret" value. */ list_del(&uwq->wq.entry); __add_wait_queue(&ctx->event_wqh, &uwq->wq); |
893e26e61
|
1124 |
|
384632e67
|
1125 1126 1127 1128 1129 1130 |
/* * Leave the event in the waitqueue and report * error to userland if we failed to resolve * the userfault fork. */ if (likely(!ret)) |
893e26e61
|
1131 |
userfaultfd_event_complete(ctx, uwq); |
384632e67
|
1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 |
} else { /* * Here the fork thread aborted and the * refcount from the fork thread on fork_nctx * has already been released. We still hold * the reference we took before releasing the * lock above. If resolve_userfault_fork * failed we've to drop it because the * fork_nctx has to be freed in such case. If * it succeeded we'll hold it because the new * uffd references it. */ if (ret) userfaultfd_ctx_put(fork_nctx); |
893e26e61
|
1146 |
} |
cbcfa130a
|
1147 |
spin_unlock_irq(&ctx->event_wqh.lock); |
893e26e61
|
1148 |
} |
86039bd3b
|
1149 1150 1151 1152 1153 1154 1155 1156 |
return ret; } static ssize_t userfaultfd_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { struct userfaultfd_ctx *ctx = file->private_data; ssize_t _ret, ret = 0; |
a9b85f941
|
1157 |
struct uffd_msg msg; |
86039bd3b
|
1158 1159 1160 1161 |
int no_wait = file->f_flags & O_NONBLOCK; if (ctx->state == UFFD_STATE_WAIT_API) return -EINVAL; |
86039bd3b
|
1162 1163 |
for (;;) { |
a9b85f941
|
1164 |
if (count < sizeof(msg)) |
86039bd3b
|
1165 |
return ret ? ret : -EINVAL; |
a9b85f941
|
1166 |
_ret = userfaultfd_ctx_read(ctx, no_wait, &msg); |
86039bd3b
|
1167 1168 |
if (_ret < 0) return ret ? ret : _ret; |
a9b85f941
|
1169 |
if (copy_to_user((__u64 __user *) buf, &msg, sizeof(msg))) |
86039bd3b
|
1170 |
return ret ? ret : -EFAULT; |
a9b85f941
|
1171 1172 1173 |
ret += sizeof(msg); buf += sizeof(msg); count -= sizeof(msg); |
86039bd3b
|
1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 |
/* * Allow to read more than one fault at time but only * block if waiting for the very first one. */ no_wait = O_NONBLOCK; } } static void __wake_userfault(struct userfaultfd_ctx *ctx, struct userfaultfd_wake_range *range) { |
cbcfa130a
|
1185 |
spin_lock_irq(&ctx->fault_pending_wqh.lock); |
86039bd3b
|
1186 |
/* wake all in the range and autoremove */ |
15b726ef0
|
1187 |
if (waitqueue_active(&ctx->fault_pending_wqh)) |
ac5be6b47
|
1188 |
__wake_up_locked_key(&ctx->fault_pending_wqh, TASK_NORMAL, |
15b726ef0
|
1189 1190 |
range); if (waitqueue_active(&ctx->fault_wqh)) |
c430d1e84
|
1191 |
__wake_up(&ctx->fault_wqh, TASK_NORMAL, 1, range); |
cbcfa130a
|
1192 |
spin_unlock_irq(&ctx->fault_pending_wqh.lock); |
86039bd3b
|
1193 1194 1195 1196 1197 |
} static __always_inline void wake_userfault(struct userfaultfd_ctx *ctx, struct userfaultfd_wake_range *range) { |
2c5b7e1be
|
1198 1199 |
unsigned seq; bool need_wakeup; |
86039bd3b
|
1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 |
/* * To be sure waitqueue_active() is not reordered by the CPU * before the pagetable update, use an explicit SMP memory * barrier here. PT lock release or up_read(mmap_sem) still * have release semantics that can allow the * waitqueue_active() to be reordered before the pte update. */ smp_mb(); /* * Use waitqueue_active because it's very frequent to * change the address space atomically even if there are no * userfaults yet. So we take the spinlock only when we're * sure we've userfaults to wake. */ |
2c5b7e1be
|
1215 1216 1217 1218 1219 1220 1221 |
do { seq = read_seqcount_begin(&ctx->refile_seq); need_wakeup = waitqueue_active(&ctx->fault_pending_wqh) || waitqueue_active(&ctx->fault_wqh); cond_resched(); } while (read_seqcount_retry(&ctx->refile_seq, seq)); if (need_wakeup) |
86039bd3b
|
1222 1223 1224 1225 |
__wake_userfault(ctx, range); } static __always_inline int validate_range(struct mm_struct *mm, |
7d0325749
|
1226 |
__u64 *start, __u64 len) |
86039bd3b
|
1227 1228 |
{ __u64 task_size = mm->task_size; |
7d0325749
|
1229 1230 1231 |
*start = untagged_addr(*start); if (*start & ~PAGE_MASK) |
86039bd3b
|
1232 1233 1234 1235 1236 |
return -EINVAL; if (len & ~PAGE_MASK) return -EINVAL; if (!len) return -EINVAL; |
7d0325749
|
1237 |
if (*start < mmap_min_addr) |
86039bd3b
|
1238 |
return -EINVAL; |
7d0325749
|
1239 |
if (*start >= task_size) |
86039bd3b
|
1240 |
return -EINVAL; |
7d0325749
|
1241 |
if (len > task_size - *start) |
86039bd3b
|
1242 1243 1244 |
return -EINVAL; return 0; } |
ba6907db6
|
1245 1246 |
static inline bool vma_can_userfault(struct vm_area_struct *vma) { |
cac673292
|
1247 1248 |
return vma_is_anonymous(vma) || is_vm_hugetlb_page(vma) || vma_is_shmem(vma); |
ba6907db6
|
1249 |
} |
86039bd3b
|
1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 |
static int userfaultfd_register(struct userfaultfd_ctx *ctx, unsigned long arg) { struct mm_struct *mm = ctx->mm; struct vm_area_struct *vma, *prev, *cur; int ret; struct uffdio_register uffdio_register; struct uffdio_register __user *user_uffdio_register; unsigned long vm_flags, new_flags; bool found; |
ce53e8e6f
|
1260 |
bool basic_ioctls; |
86039bd3b
|
1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 |
unsigned long start, end, vma_end; user_uffdio_register = (struct uffdio_register __user *) arg; ret = -EFAULT; if (copy_from_user(&uffdio_register, user_uffdio_register, sizeof(uffdio_register)-sizeof(__u64))) goto out; ret = -EINVAL; if (!uffdio_register.mode) goto out; if (uffdio_register.mode & ~(UFFDIO_REGISTER_MODE_MISSING| UFFDIO_REGISTER_MODE_WP)) goto out; vm_flags = 0; if (uffdio_register.mode & UFFDIO_REGISTER_MODE_MISSING) vm_flags |= VM_UFFD_MISSING; if (uffdio_register.mode & UFFDIO_REGISTER_MODE_WP) { vm_flags |= VM_UFFD_WP; /* * FIXME: remove the below error constraint by * implementing the wprotect tracking mode. */ ret = -EINVAL; goto out; } |
7d0325749
|
1288 |
ret = validate_range(mm, &uffdio_register.range.start, |
86039bd3b
|
1289 1290 1291 1292 1293 1294 |
uffdio_register.range.len); if (ret) goto out; start = uffdio_register.range.start; end = start + uffdio_register.range.len; |
d2005e3f4
|
1295 1296 1297 |
ret = -ENOMEM; if (!mmget_not_zero(mm)) goto out; |
86039bd3b
|
1298 |
down_write(&mm->mmap_sem); |
04f5866e4
|
1299 1300 |
if (!mmget_still_valid(mm)) goto out_unlock; |
86039bd3b
|
1301 |
vma = find_vma_prev(mm, start, &prev); |
86039bd3b
|
1302 1303 1304 1305 1306 1307 1308 1309 1310 |
if (!vma) goto out_unlock; /* check that there's at least one vma in the range */ ret = -EINVAL; if (vma->vm_start >= end) goto out_unlock; /* |
cab350afc
|
1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 |
* If the first vma contains huge pages, make sure start address * is aligned to huge page size. */ if (is_vm_hugetlb_page(vma)) { unsigned long vma_hpagesize = vma_kernel_pagesize(vma); if (start & (vma_hpagesize - 1)) goto out_unlock; } /* |
86039bd3b
|
1322 |
* Search for not compatible vmas. |
86039bd3b
|
1323 1324 |
*/ found = false; |
ce53e8e6f
|
1325 |
basic_ioctls = false; |
86039bd3b
|
1326 1327 1328 1329 1330 1331 1332 1333 |
for (cur = vma; cur && cur->vm_start < end; cur = cur->vm_next) { cond_resched(); BUG_ON(!!cur->vm_userfaultfd_ctx.ctx ^ !!(cur->vm_flags & (VM_UFFD_MISSING | VM_UFFD_WP))); /* check not compatible vmas */ ret = -EINVAL; |
ba6907db6
|
1334 |
if (!vma_can_userfault(cur)) |
86039bd3b
|
1335 |
goto out_unlock; |
29ec90660
|
1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 |
/* * UFFDIO_COPY will fill file holes even without * PROT_WRITE. This check enforces that if this is a * MAP_SHARED, the process has write permission to the backing * file. If VM_MAYWRITE is set it also enforces that on a * MAP_SHARED vma: there is no F_WRITE_SEAL and no further * F_WRITE_SEAL can be taken until the vma is destroyed. */ ret = -EPERM; if (unlikely(!(cur->vm_flags & VM_MAYWRITE))) goto out_unlock; |
cab350afc
|
1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 |
/* * If this vma contains ending address, and huge pages * check alignment. */ if (is_vm_hugetlb_page(cur) && end <= cur->vm_end && end > cur->vm_start) { unsigned long vma_hpagesize = vma_kernel_pagesize(cur); ret = -EINVAL; if (end & (vma_hpagesize - 1)) goto out_unlock; } |
86039bd3b
|
1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 |
/* * Check that this vma isn't already owned by a * different userfaultfd. We can't allow more than one * userfaultfd to own a single vma simultaneously or we * wouldn't know which one to deliver the userfaults to. */ ret = -EBUSY; if (cur->vm_userfaultfd_ctx.ctx && cur->vm_userfaultfd_ctx.ctx != ctx) goto out_unlock; |
cab350afc
|
1372 1373 1374 |
/* * Note vmas containing huge pages */ |
ce53e8e6f
|
1375 1376 |
if (is_vm_hugetlb_page(cur)) basic_ioctls = true; |
cab350afc
|
1377 |
|
86039bd3b
|
1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 |
found = true; } BUG_ON(!found); if (vma->vm_start < start) prev = vma; ret = 0; do { cond_resched(); |
ba6907db6
|
1388 |
BUG_ON(!vma_can_userfault(vma)); |
86039bd3b
|
1389 1390 |
BUG_ON(vma->vm_userfaultfd_ctx.ctx && vma->vm_userfaultfd_ctx.ctx != ctx); |
29ec90660
|
1391 |
WARN_ON(!(vma->vm_flags & VM_MAYWRITE)); |
86039bd3b
|
1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 |
/* * Nothing to do: this vma is already registered into this * userfaultfd and with the right tracking mode too. */ if (vma->vm_userfaultfd_ctx.ctx == ctx && (vma->vm_flags & vm_flags) == vm_flags) goto skip; if (vma->vm_start > start) start = vma->vm_start; vma_end = min(end, vma->vm_end); new_flags = (vma->vm_flags & ~vm_flags) | vm_flags; prev = vma_merge(mm, prev, start, vma_end, new_flags, vma->anon_vma, vma->vm_file, vma->vm_pgoff, vma_policy(vma), ((struct vm_userfaultfd_ctx){ ctx })); if (prev) { vma = prev; goto next; } if (vma->vm_start < start) { ret = split_vma(mm, vma, start, 1); if (ret) break; } if (vma->vm_end > end) { ret = split_vma(mm, vma, end, 0); if (ret) break; } next: /* * In the vma_merge() successful mprotect-like case 8: * the next vma was merged into the current one and * the current one has not been updated yet. */ vma->vm_flags = new_flags; vma->vm_userfaultfd_ctx.ctx = ctx; skip: prev = vma; start = vma->vm_end; vma = vma->vm_next; } while (vma && vma->vm_start < end); out_unlock: up_write(&mm->mmap_sem); |
d2005e3f4
|
1440 |
mmput(mm); |
86039bd3b
|
1441 1442 1443 1444 1445 1446 |
if (!ret) { /* * Now that we scanned all vmas we can already tell * userland which ioctls methods are guaranteed to * succeed on this range. */ |
ce53e8e6f
|
1447 |
if (put_user(basic_ioctls ? UFFD_API_RANGE_IOCTLS_BASIC : |
cab350afc
|
1448 |
UFFD_API_RANGE_IOCTLS, |
86039bd3b
|
1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 |
&user_uffdio_register->ioctls)) ret = -EFAULT; } out: return ret; } static int userfaultfd_unregister(struct userfaultfd_ctx *ctx, unsigned long arg) { struct mm_struct *mm = ctx->mm; struct vm_area_struct *vma, *prev, *cur; int ret; struct uffdio_range uffdio_unregister; unsigned long new_flags; bool found; unsigned long start, end, vma_end; const void __user *buf = (void __user *)arg; ret = -EFAULT; if (copy_from_user(&uffdio_unregister, buf, sizeof(uffdio_unregister))) goto out; |
7d0325749
|
1471 |
ret = validate_range(mm, &uffdio_unregister.start, |
86039bd3b
|
1472 1473 1474 1475 1476 1477 |
uffdio_unregister.len); if (ret) goto out; start = uffdio_unregister.start; end = start + uffdio_unregister.len; |
d2005e3f4
|
1478 1479 1480 |
ret = -ENOMEM; if (!mmget_not_zero(mm)) goto out; |
86039bd3b
|
1481 |
down_write(&mm->mmap_sem); |
04f5866e4
|
1482 1483 |
if (!mmget_still_valid(mm)) goto out_unlock; |
86039bd3b
|
1484 |
vma = find_vma_prev(mm, start, &prev); |
86039bd3b
|
1485 1486 1487 1488 1489 1490 1491 1492 1493 |
if (!vma) goto out_unlock; /* check that there's at least one vma in the range */ ret = -EINVAL; if (vma->vm_start >= end) goto out_unlock; /* |
cab350afc
|
1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 |
* If the first vma contains huge pages, make sure start address * is aligned to huge page size. */ if (is_vm_hugetlb_page(vma)) { unsigned long vma_hpagesize = vma_kernel_pagesize(vma); if (start & (vma_hpagesize - 1)) goto out_unlock; } /* |
86039bd3b
|
1505 |
* Search for not compatible vmas. |
86039bd3b
|
1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 |
*/ found = false; ret = -EINVAL; for (cur = vma; cur && cur->vm_start < end; cur = cur->vm_next) { cond_resched(); BUG_ON(!!cur->vm_userfaultfd_ctx.ctx ^ !!(cur->vm_flags & (VM_UFFD_MISSING | VM_UFFD_WP))); /* * Check not compatible vmas, not strictly required * here as not compatible vmas cannot have an * userfaultfd_ctx registered on them, but this * provides for more strict behavior to notice * unregistration errors. */ |
ba6907db6
|
1522 |
if (!vma_can_userfault(cur)) |
86039bd3b
|
1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 |
goto out_unlock; found = true; } BUG_ON(!found); if (vma->vm_start < start) prev = vma; ret = 0; do { cond_resched(); |
ba6907db6
|
1535 |
BUG_ON(!vma_can_userfault(vma)); |
86039bd3b
|
1536 1537 1538 1539 1540 1541 1542 |
/* * Nothing to do: this vma is already registered into this * userfaultfd and with the right tracking mode too. */ if (!vma->vm_userfaultfd_ctx.ctx) goto skip; |
01e881f5a
|
1543 |
WARN_ON(!(vma->vm_flags & VM_MAYWRITE)); |
86039bd3b
|
1544 1545 1546 |
if (vma->vm_start > start) start = vma->vm_start; vma_end = min(end, vma->vm_end); |
09fa5296a
|
1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 |
if (userfaultfd_missing(vma)) { /* * Wake any concurrent pending userfault while * we unregister, so they will not hang * permanently and it avoids userland to call * UFFDIO_WAKE explicitly. */ struct userfaultfd_wake_range range; range.start = start; range.len = vma_end - start; wake_userfault(vma->vm_userfaultfd_ctx.ctx, &range); } |
86039bd3b
|
1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 |
new_flags = vma->vm_flags & ~(VM_UFFD_MISSING | VM_UFFD_WP); prev = vma_merge(mm, prev, start, vma_end, new_flags, vma->anon_vma, vma->vm_file, vma->vm_pgoff, vma_policy(vma), NULL_VM_UFFD_CTX); if (prev) { vma = prev; goto next; } if (vma->vm_start < start) { ret = split_vma(mm, vma, start, 1); if (ret) break; } if (vma->vm_end > end) { ret = split_vma(mm, vma, end, 0); if (ret) break; } next: /* * In the vma_merge() successful mprotect-like case 8: * the next vma was merged into the current one and * the current one has not been updated yet. */ vma->vm_flags = new_flags; vma->vm_userfaultfd_ctx = NULL_VM_UFFD_CTX; skip: prev = vma; start = vma->vm_end; vma = vma->vm_next; } while (vma && vma->vm_start < end); out_unlock: up_write(&mm->mmap_sem); |
d2005e3f4
|
1594 |
mmput(mm); |
86039bd3b
|
1595 1596 1597 1598 1599 |
out: return ret; } /* |
ba85c702e
|
1600 1601 |
* userfaultfd_wake may be used in combination with the * UFFDIO_*_MODE_DONTWAKE to wakeup userfaults in batches. |
86039bd3b
|
1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 |
*/ static int userfaultfd_wake(struct userfaultfd_ctx *ctx, unsigned long arg) { int ret; struct uffdio_range uffdio_wake; struct userfaultfd_wake_range range; const void __user *buf = (void __user *)arg; ret = -EFAULT; if (copy_from_user(&uffdio_wake, buf, sizeof(uffdio_wake))) goto out; |
7d0325749
|
1614 |
ret = validate_range(ctx->mm, &uffdio_wake.start, uffdio_wake.len); |
86039bd3b
|
1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 |
if (ret) goto out; range.start = uffdio_wake.start; range.len = uffdio_wake.len; /* * len == 0 means wake all and we don't want to wake all here, * so check it again to be sure. */ VM_BUG_ON(!range.len); wake_userfault(ctx, &range); ret = 0; out: return ret; } |
ad465cae9
|
1633 1634 1635 1636 1637 1638 1639 1640 1641 |
static int userfaultfd_copy(struct userfaultfd_ctx *ctx, unsigned long arg) { __s64 ret; struct uffdio_copy uffdio_copy; struct uffdio_copy __user *user_uffdio_copy; struct userfaultfd_wake_range range; user_uffdio_copy = (struct uffdio_copy __user *) arg; |
df2cc96e7
|
1642 1643 1644 |
ret = -EAGAIN; if (READ_ONCE(ctx->mmap_changing)) goto out; |
ad465cae9
|
1645 1646 1647 1648 1649 |
ret = -EFAULT; if (copy_from_user(&uffdio_copy, user_uffdio_copy, /* don't copy "copy" last field */ sizeof(uffdio_copy)-sizeof(__s64))) goto out; |
7d0325749
|
1650 |
ret = validate_range(ctx->mm, &uffdio_copy.dst, uffdio_copy.len); |
ad465cae9
|
1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 |
if (ret) goto out; /* * double check for wraparound just in case. copy_from_user() * will later check uffdio_copy.src + uffdio_copy.len to fit * in the userland range. */ ret = -EINVAL; if (uffdio_copy.src + uffdio_copy.len <= uffdio_copy.src) goto out; if (uffdio_copy.mode & ~UFFDIO_COPY_MODE_DONTWAKE) goto out; |
d2005e3f4
|
1663 1664 |
if (mmget_not_zero(ctx->mm)) { ret = mcopy_atomic(ctx->mm, uffdio_copy.dst, uffdio_copy.src, |
df2cc96e7
|
1665 |
uffdio_copy.len, &ctx->mmap_changing); |
d2005e3f4
|
1666 |
mmput(ctx->mm); |
96333187a
|
1667 |
} else { |
e86b298be
|
1668 |
return -ESRCH; |
d2005e3f4
|
1669 |
} |
ad465cae9
|
1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 |
if (unlikely(put_user(ret, &user_uffdio_copy->copy))) return -EFAULT; if (ret < 0) goto out; BUG_ON(!ret); /* len == 0 would wake all */ range.len = ret; if (!(uffdio_copy.mode & UFFDIO_COPY_MODE_DONTWAKE)) { range.start = uffdio_copy.dst; wake_userfault(ctx, &range); } ret = range.len == uffdio_copy.len ? 0 : -EAGAIN; out: return ret; } static int userfaultfd_zeropage(struct userfaultfd_ctx *ctx, unsigned long arg) { __s64 ret; struct uffdio_zeropage uffdio_zeropage; struct uffdio_zeropage __user *user_uffdio_zeropage; struct userfaultfd_wake_range range; user_uffdio_zeropage = (struct uffdio_zeropage __user *) arg; |
df2cc96e7
|
1695 1696 1697 |
ret = -EAGAIN; if (READ_ONCE(ctx->mmap_changing)) goto out; |
ad465cae9
|
1698 1699 1700 1701 1702 |
ret = -EFAULT; if (copy_from_user(&uffdio_zeropage, user_uffdio_zeropage, /* don't copy "zeropage" last field */ sizeof(uffdio_zeropage)-sizeof(__s64))) goto out; |
7d0325749
|
1703 |
ret = validate_range(ctx->mm, &uffdio_zeropage.range.start, |
ad465cae9
|
1704 1705 1706 1707 1708 1709 |
uffdio_zeropage.range.len); if (ret) goto out; ret = -EINVAL; if (uffdio_zeropage.mode & ~UFFDIO_ZEROPAGE_MODE_DONTWAKE) goto out; |
d2005e3f4
|
1710 1711 |
if (mmget_not_zero(ctx->mm)) { ret = mfill_zeropage(ctx->mm, uffdio_zeropage.range.start, |
df2cc96e7
|
1712 1713 |
uffdio_zeropage.range.len, &ctx->mmap_changing); |
d2005e3f4
|
1714 |
mmput(ctx->mm); |
9d95aa4ba
|
1715 |
} else { |
e86b298be
|
1716 |
return -ESRCH; |
d2005e3f4
|
1717 |
} |
ad465cae9
|
1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 |
if (unlikely(put_user(ret, &user_uffdio_zeropage->zeropage))) return -EFAULT; if (ret < 0) goto out; /* len == 0 would wake all */ BUG_ON(!ret); range.len = ret; if (!(uffdio_zeropage.mode & UFFDIO_ZEROPAGE_MODE_DONTWAKE)) { range.start = uffdio_zeropage.range.start; wake_userfault(ctx, &range); } ret = range.len == uffdio_zeropage.range.len ? 0 : -EAGAIN; out: return ret; } |
9cd75c3cd
|
1733 1734 1735 1736 1737 1738 1739 |
static inline unsigned int uffd_ctx_features(__u64 user_features) { /* * For the current set of features the bits just coincide */ return (unsigned int)user_features; } |
86039bd3b
|
1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 |
/* * userland asks for a certain API version and we return which bits * and ioctl commands are implemented in this kernel for such API * version or -EINVAL if unknown. */ static int userfaultfd_api(struct userfaultfd_ctx *ctx, unsigned long arg) { struct uffdio_api uffdio_api; void __user *buf = (void __user *)arg; int ret; |
656031445
|
1751 |
__u64 features; |
86039bd3b
|
1752 1753 1754 1755 1756 |
ret = -EINVAL; if (ctx->state != UFFD_STATE_WAIT_API) goto out; ret = -EFAULT; |
a9b85f941
|
1757 |
if (copy_from_user(&uffdio_api, buf, sizeof(uffdio_api))) |
86039bd3b
|
1758 |
goto out; |
656031445
|
1759 |
features = uffdio_api.features; |
2176441fd
|
1760 1761 1762 1763 1764 1765 |
ret = -EINVAL; if (uffdio_api.api != UFFD_API || (features & ~UFFD_API_FEATURES)) goto err_out; ret = -EPERM; if ((features & UFFD_FEATURE_EVENT_FORK) && !capable(CAP_SYS_PTRACE)) goto err_out; |
656031445
|
1766 1767 |
/* report all available features and ioctls to userland */ uffdio_api.features = UFFD_API_FEATURES; |
86039bd3b
|
1768 1769 1770 1771 1772 |
uffdio_api.ioctls = UFFD_API_IOCTLS; ret = -EFAULT; if (copy_to_user(buf, &uffdio_api, sizeof(uffdio_api))) goto out; ctx->state = UFFD_STATE_RUNNING; |
656031445
|
1773 1774 |
/* only enable the requested features for this uffd context */ ctx->features = uffd_ctx_features(features); |
86039bd3b
|
1775 1776 1777 |
ret = 0; out: return ret; |
2176441fd
|
1778 1779 1780 1781 1782 |
err_out: memset(&uffdio_api, 0, sizeof(uffdio_api)); if (copy_to_user(buf, &uffdio_api, sizeof(uffdio_api))) ret = -EFAULT; goto out; |
86039bd3b
|
1783 1784 1785 1786 1787 1788 1789 |
} static long userfaultfd_ioctl(struct file *file, unsigned cmd, unsigned long arg) { int ret = -EINVAL; struct userfaultfd_ctx *ctx = file->private_data; |
e6485a47b
|
1790 1791 |
if (cmd != UFFDIO_API && ctx->state == UFFD_STATE_WAIT_API) return -EINVAL; |
86039bd3b
|
1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 |
switch(cmd) { case UFFDIO_API: ret = userfaultfd_api(ctx, arg); break; case UFFDIO_REGISTER: ret = userfaultfd_register(ctx, arg); break; case UFFDIO_UNREGISTER: ret = userfaultfd_unregister(ctx, arg); break; case UFFDIO_WAKE: ret = userfaultfd_wake(ctx, arg); break; |
ad465cae9
|
1805 1806 1807 1808 1809 1810 |
case UFFDIO_COPY: ret = userfaultfd_copy(ctx, arg); break; case UFFDIO_ZEROPAGE: ret = userfaultfd_zeropage(ctx, arg); break; |
86039bd3b
|
1811 1812 1813 1814 1815 1816 1817 1818 |
} return ret; } #ifdef CONFIG_PROC_FS static void userfaultfd_show_fdinfo(struct seq_file *m, struct file *f) { struct userfaultfd_ctx *ctx = f->private_data; |
ac6424b98
|
1819 |
wait_queue_entry_t *wq; |
86039bd3b
|
1820 |
unsigned long pending = 0, total = 0; |
cbcfa130a
|
1821 |
spin_lock_irq(&ctx->fault_pending_wqh.lock); |
2055da973
|
1822 |
list_for_each_entry(wq, &ctx->fault_pending_wqh.head, entry) { |
15b726ef0
|
1823 1824 1825 |
pending++; total++; } |
2055da973
|
1826 |
list_for_each_entry(wq, &ctx->fault_wqh.head, entry) { |
86039bd3b
|
1827 1828 |
total++; } |
cbcfa130a
|
1829 |
spin_unlock_irq(&ctx->fault_pending_wqh.lock); |
86039bd3b
|
1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 |
/* * If more protocols will be added, there will be all shown * separated by a space. Like this: * protocols: aa:... bb:... */ seq_printf(m, "pending:\t%lu total:\t%lu API:\t%Lx:%x:%Lx ", |
045098e94
|
1840 |
pending, total, UFFD_API, ctx->features, |
86039bd3b
|
1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 |
UFFD_API_IOCTLS|UFFD_API_RANGE_IOCTLS); } #endif static const struct file_operations userfaultfd_fops = { #ifdef CONFIG_PROC_FS .show_fdinfo = userfaultfd_show_fdinfo, #endif .release = userfaultfd_release, .poll = userfaultfd_poll, .read = userfaultfd_read, .unlocked_ioctl = userfaultfd_ioctl, .compat_ioctl = userfaultfd_ioctl, .llseek = noop_llseek, }; |
3004ec9ca
|
1856 1857 1858 1859 1860 1861 |
static void init_once_userfaultfd_ctx(void *mem) { struct userfaultfd_ctx *ctx = (struct userfaultfd_ctx *) mem; init_waitqueue_head(&ctx->fault_pending_wqh); init_waitqueue_head(&ctx->fault_wqh); |
9cd75c3cd
|
1862 |
init_waitqueue_head(&ctx->event_wqh); |
3004ec9ca
|
1863 |
init_waitqueue_head(&ctx->fd_wqh); |
2c5b7e1be
|
1864 |
seqcount_init(&ctx->refile_seq); |
3004ec9ca
|
1865 |
} |
284cd241a
|
1866 |
SYSCALL_DEFINE1(userfaultfd, int, flags) |
86039bd3b
|
1867 |
{ |
86039bd3b
|
1868 |
struct userfaultfd_ctx *ctx; |
284cd241a
|
1869 |
int fd; |
86039bd3b
|
1870 |
|
cefdca0a8
|
1871 1872 |
if (!sysctl_unprivileged_userfaultfd && !capable(CAP_SYS_PTRACE)) return -EPERM; |
86039bd3b
|
1873 1874 1875 1876 1877 |
BUG_ON(!current->mm); /* Check the UFFD_* constants for consistency. */ BUILD_BUG_ON(UFFD_CLOEXEC != O_CLOEXEC); BUILD_BUG_ON(UFFD_NONBLOCK != O_NONBLOCK); |
86039bd3b
|
1878 |
if (flags & ~UFFD_SHARED_FCNTL_FLAGS) |
284cd241a
|
1879 |
return -EINVAL; |
86039bd3b
|
1880 |
|
3004ec9ca
|
1881 |
ctx = kmem_cache_alloc(userfaultfd_ctx_cachep, GFP_KERNEL); |
86039bd3b
|
1882 |
if (!ctx) |
284cd241a
|
1883 |
return -ENOMEM; |
86039bd3b
|
1884 |
|
ca8804206
|
1885 |
refcount_set(&ctx->refcount, 1); |
86039bd3b
|
1886 |
ctx->flags = flags; |
9cd75c3cd
|
1887 |
ctx->features = 0; |
86039bd3b
|
1888 1889 |
ctx->state = UFFD_STATE_WAIT_API; ctx->released = false; |
df2cc96e7
|
1890 |
ctx->mmap_changing = false; |
86039bd3b
|
1891 1892 |
ctx->mm = current->mm; /* prevent the mm struct to be freed */ |
f1f100764
|
1893 |
mmgrab(ctx->mm); |
86039bd3b
|
1894 |
|
284cd241a
|
1895 1896 1897 |
fd = anon_inode_getfd("[userfaultfd]", &userfaultfd_fops, ctx, O_RDWR | (flags & UFFD_SHARED_FCNTL_FLAGS)); if (fd < 0) { |
d2005e3f4
|
1898 |
mmdrop(ctx->mm); |
3004ec9ca
|
1899 |
kmem_cache_free(userfaultfd_ctx_cachep, ctx); |
c03e946fd
|
1900 |
} |
86039bd3b
|
1901 |
return fd; |
86039bd3b
|
1902 |
} |
3004ec9ca
|
1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 |
static int __init userfaultfd_init(void) { userfaultfd_ctx_cachep = kmem_cache_create("userfaultfd_ctx_cache", sizeof(struct userfaultfd_ctx), 0, SLAB_HWCACHE_ALIGN|SLAB_PANIC, init_once_userfaultfd_ctx); return 0; } __initcall(userfaultfd_init); |