>> 1 .. _userfaultfd:
>> 2
1 =========== 3 ===========
2 Userfaultfd 4 Userfaultfd
3 =========== 5 ===========
4 6
5 Objective 7 Objective
6 ========= 8 =========
7 9
8 Userfaults allow the implementation of on-dema 10 Userfaults allow the implementation of on-demand paging from userland
9 and more generally they allow userland to take 11 and more generally they allow userland to take control of various
10 memory page faults, something otherwise only t 12 memory page faults, something otherwise only the kernel code could do.
11 13
12 For example userfaults allows a proper and mor 14 For example userfaults allows a proper and more optimal implementation
13 of the ``PROT_NONE+SIGSEGV`` trick. 15 of the ``PROT_NONE+SIGSEGV`` trick.
14 16
15 Design 17 Design
16 ====== 18 ======
17 19
18 Userspace creates a new userfaultfd, initializ !! 20 Userfaults are delivered and resolved through the ``userfaultfd`` syscall.
19 regions of virtual memory with it. Then, any p <<
20 region(s) result in a message being delivered <<
21 userspace of the fault. <<
22 21
23 The ``userfaultfd`` (aside from registering an 22 The ``userfaultfd`` (aside from registering and unregistering virtual
24 memory ranges) provides two primary functional 23 memory ranges) provides two primary functionalities:
25 24
26 1) ``read/POLLIN`` protocol to notify a userla 25 1) ``read/POLLIN`` protocol to notify a userland thread of the faults
27 happening 26 happening
28 27
29 2) various ``UFFDIO_*`` ioctls that can manage 28 2) various ``UFFDIO_*`` ioctls that can manage the virtual memory regions
30 registered in the ``userfaultfd`` that allo 29 registered in the ``userfaultfd`` that allows userland to efficiently
31 resolve the userfaults it receives via 1) o 30 resolve the userfaults it receives via 1) or to manage the virtual
32 memory in the background 31 memory in the background
33 32
34 The real advantage of userfaults if compared t 33 The real advantage of userfaults if compared to regular virtual memory
35 management of mremap/mprotect is that the user 34 management of mremap/mprotect is that the userfaults in all their
36 operations never involve heavyweight structure 35 operations never involve heavyweight structures like vmas (in fact the
37 ``userfaultfd`` runtime load never takes the m 36 ``userfaultfd`` runtime load never takes the mmap_lock for writing).
>> 37
38 Vmas are not suitable for page- (or hugepage) 38 Vmas are not suitable for page- (or hugepage) granular fault tracking
39 when dealing with virtual address spaces that 39 when dealing with virtual address spaces that could span
40 Terabytes. Too many vmas would be needed for t 40 Terabytes. Too many vmas would be needed for that.
41 41
42 The ``userfaultfd``, once created, can also be !! 42 The ``userfaultfd`` once opened by invoking the syscall, can also be
43 passed using unix domain sockets to a manager 43 passed using unix domain sockets to a manager process, so the same
44 manager process could handle the userfaults of 44 manager process could handle the userfaults of a multitude of
45 different processes without them being aware a 45 different processes without them being aware about what is going on
46 (well of course unless they later try to use t 46 (well of course unless they later try to use the ``userfaultfd``
47 themselves on the same region the manager is a 47 themselves on the same region the manager is already tracking, which
48 is a corner case that would currently return ` 48 is a corner case that would currently return ``-EBUSY``).
49 49
50 API 50 API
51 === 51 ===
52 52
53 Creating a userfaultfd <<
54 ---------------------- <<
55 <<
56 There are two ways to create a new userfaultfd <<
57 restrict access to this functionality (since h <<
58 handle kernel page faults have been a useful t <<
59 <<
60 The first way, supported since userfaultfd was <<
61 userfaultfd(2) syscall. Access to this is cont <<
62 <<
63 - Any user can always create a userfaultfd whi <<
64 only. Such a userfaultfd can be created usin <<
65 with the flag UFFD_USER_MODE_ONLY. <<
66 <<
67 - In order to also trap kernel page faults for <<
68 process needs the CAP_SYS_PTRACE capability, <<
69 vm.unprivileged_userfaultfd set to 1. By def <<
70 is set to 0. <<
71 <<
72 The second way, added to the kernel more recen <<
73 /dev/userfaultfd and issuing a USERFAULTFD_IOC <<
74 yields equivalent userfaultfds to the userfaul <<
75 <<
76 Unlike userfaultfd(2), access to /dev/userfaul <<
77 filesystem permissions (user/group/mode), whic <<
78 userfaultfd specifically, without also grantin <<
79 the same time (as e.g. granting CAP_SYS_PTRACE <<
80 to /dev/userfaultfd can always create userfaul <<
81 vm.unprivileged_userfaultfd is not considered. <<
82 <<
83 Initializing a userfaultfd <<
84 -------------------------- <<
85 <<
86 When first opened the ``userfaultfd`` must be 53 When first opened the ``userfaultfd`` must be enabled invoking the
87 ``UFFDIO_API`` ioctl specifying a ``uffdio_api 54 ``UFFDIO_API`` ioctl specifying a ``uffdio_api.api`` value set to ``UFFD_API`` (or
88 a later API version) which will specify the `` 55 a later API version) which will specify the ``read/POLLIN`` protocol
89 userland intends to speak on the ``UFFD`` and 56 userland intends to speak on the ``UFFD`` and the ``uffdio_api.features``
90 userland requires. The ``UFFDIO_API`` ioctl if 57 userland requires. The ``UFFDIO_API`` ioctl if successful (i.e. if the
91 requested ``uffdio_api.api`` is spoken also by 58 requested ``uffdio_api.api`` is spoken also by the running kernel and the
92 requested features are going to be enabled) wi 59 requested features are going to be enabled) will return into
93 ``uffdio_api.features`` and ``uffdio_api.ioctl 60 ``uffdio_api.features`` and ``uffdio_api.ioctls`` two 64bit bitmasks of
94 respectively all the available features of the 61 respectively all the available features of the read(2) protocol and
95 the generic ioctl available. 62 the generic ioctl available.
96 63
97 The ``uffdio_api.features`` bitmask returned b 64 The ``uffdio_api.features`` bitmask returned by the ``UFFDIO_API`` ioctl
98 defines what memory types are supported by the 65 defines what memory types are supported by the ``userfaultfd`` and what
99 events, except page fault notifications, may b 66 events, except page fault notifications, may be generated:
100 67
101 - The ``UFFD_FEATURE_EVENT_*`` flags indicate 68 - The ``UFFD_FEATURE_EVENT_*`` flags indicate that various other events
102 other than page faults are supported. These 69 other than page faults are supported. These events are described in more
103 detail below in the `Non-cooperative userfau 70 detail below in the `Non-cooperative userfaultfd`_ section.
104 71
105 - ``UFFD_FEATURE_MISSING_HUGETLBFS`` and ``UFF 72 - ``UFFD_FEATURE_MISSING_HUGETLBFS`` and ``UFFD_FEATURE_MISSING_SHMEM``
106 indicate that the kernel supports ``UFFDIO_R 73 indicate that the kernel supports ``UFFDIO_REGISTER_MODE_MISSING``
107 registrations for hugetlbfs and shared memor 74 registrations for hugetlbfs and shared memory (covering all shmem APIs,
108 i.e. tmpfs, ``IPCSHM``, ``/dev/zero``, ``MAP 75 i.e. tmpfs, ``IPCSHM``, ``/dev/zero``, ``MAP_SHARED``, ``memfd_create``,
109 etc) virtual memory areas, respectively. 76 etc) virtual memory areas, respectively.
110 77
111 - ``UFFD_FEATURE_MINOR_HUGETLBFS`` indicates t 78 - ``UFFD_FEATURE_MINOR_HUGETLBFS`` indicates that the kernel supports
112 ``UFFDIO_REGISTER_MODE_MINOR`` registration 79 ``UFFDIO_REGISTER_MODE_MINOR`` registration for hugetlbfs virtual memory
113 areas. ``UFFD_FEATURE_MINOR_SHMEM`` is the a 80 areas. ``UFFD_FEATURE_MINOR_SHMEM`` is the analogous feature indicating
114 support for shmem virtual memory areas. 81 support for shmem virtual memory areas.
115 82
116 - ``UFFD_FEATURE_MOVE`` indicates that the ker <<
117 existing page contents from userspace. <<
118 <<
119 The userland application should set the featur 83 The userland application should set the feature flags it intends to use
120 when invoking the ``UFFDIO_API`` ioctl, to req 84 when invoking the ``UFFDIO_API`` ioctl, to request that those features be
121 enabled if supported. 85 enabled if supported.
122 86
123 Once the ``userfaultfd`` API has been enabled 87 Once the ``userfaultfd`` API has been enabled the ``UFFDIO_REGISTER``
124 ioctl should be invoked (if present in the ret 88 ioctl should be invoked (if present in the returned ``uffdio_api.ioctls``
125 bitmask) to register a memory range in the ``u 89 bitmask) to register a memory range in the ``userfaultfd`` by setting the
126 uffdio_register structure accordingly. The ``u 90 uffdio_register structure accordingly. The ``uffdio_register.mode``
127 bitmask will specify to the kernel which kind 91 bitmask will specify to the kernel which kind of faults to track for
128 the range. The ``UFFDIO_REGISTER`` ioctl will 92 the range. The ``UFFDIO_REGISTER`` ioctl will return the
129 ``uffdio_register.ioctls`` bitmask of ioctls t 93 ``uffdio_register.ioctls`` bitmask of ioctls that are suitable to resolve
130 userfaults on the range registered. Not all io 94 userfaults on the range registered. Not all ioctls will necessarily be
131 supported for all memory types (e.g. anonymous 95 supported for all memory types (e.g. anonymous memory vs. shmem vs.
132 hugetlbfs), or all types of intercepted faults 96 hugetlbfs), or all types of intercepted faults.
133 97
134 Userland can use the ``uffdio_register.ioctls` 98 Userland can use the ``uffdio_register.ioctls`` to manage the virtual
135 address space in the background (to add or pot 99 address space in the background (to add or potentially also remove
136 memory from the ``userfaultfd`` registered ran 100 memory from the ``userfaultfd`` registered range). This means a userfault
137 could be triggering just before userland maps 101 could be triggering just before userland maps in the background the
138 user-faulted page. 102 user-faulted page.
139 103
140 Resolving Userfaults 104 Resolving Userfaults
141 -------------------- 105 --------------------
142 106
143 There are three basic ways to resolve userfaul 107 There are three basic ways to resolve userfaults:
144 108
145 - ``UFFDIO_COPY`` atomically copies some exist 109 - ``UFFDIO_COPY`` atomically copies some existing page contents from
146 userspace. 110 userspace.
147 111
148 - ``UFFDIO_ZEROPAGE`` atomically zeros the new 112 - ``UFFDIO_ZEROPAGE`` atomically zeros the new page.
149 113
150 - ``UFFDIO_CONTINUE`` maps an existing, previo 114 - ``UFFDIO_CONTINUE`` maps an existing, previously-populated page.
151 115
152 These operations are atomic in the sense that 116 These operations are atomic in the sense that they guarantee nothing can
153 see a half-populated page, since readers will 117 see a half-populated page, since readers will keep userfaulting until the
154 operation has finished. 118 operation has finished.
155 119
156 By default, these wake up userfaults blocked o 120 By default, these wake up userfaults blocked on the range in question.
157 They support a ``UFFDIO_*_MODE_DONTWAKE`` ``mo 121 They support a ``UFFDIO_*_MODE_DONTWAKE`` ``mode`` flag, which indicates
158 that waking will be done separately at some la 122 that waking will be done separately at some later time.
159 123
160 Which ioctl to choose depends on the kind of p 124 Which ioctl to choose depends on the kind of page fault, and what we'd
161 like to do to resolve it: 125 like to do to resolve it:
162 126
163 - For ``UFFDIO_REGISTER_MODE_MISSING`` faults, 127 - For ``UFFDIO_REGISTER_MODE_MISSING`` faults, the fault needs to be
164 resolved by either providing a new page (``U 128 resolved by either providing a new page (``UFFDIO_COPY``), or mapping
165 the zero page (``UFFDIO_ZEROPAGE``). By defa 129 the zero page (``UFFDIO_ZEROPAGE``). By default, the kernel would map
166 the zero page for a missing fault. With user 130 the zero page for a missing fault. With userfaultfd, userspace can
167 decide what content to provide before the fa 131 decide what content to provide before the faulting thread continues.
168 132
169 - For ``UFFDIO_REGISTER_MODE_MINOR`` faults, t 133 - For ``UFFDIO_REGISTER_MODE_MINOR`` faults, there is an existing page (in
170 the page cache). Userspace has the option of 134 the page cache). Userspace has the option of modifying the page's
171 contents before resolving the fault. Once th 135 contents before resolving the fault. Once the contents are correct
172 (modified or not), userspace asks the kernel 136 (modified or not), userspace asks the kernel to map the page and let the
173 faulting thread continue with ``UFFDIO_CONTI 137 faulting thread continue with ``UFFDIO_CONTINUE``.
174 138
175 Notes: 139 Notes:
176 140
177 - You can tell which kind of fault occurred by 141 - You can tell which kind of fault occurred by examining
178 ``pagefault.flags`` within the ``uffd_msg``, 142 ``pagefault.flags`` within the ``uffd_msg``, checking for the
179 ``UFFD_PAGEFAULT_FLAG_*`` flags. 143 ``UFFD_PAGEFAULT_FLAG_*`` flags.
180 144
181 - None of the page-delivering ioctls default t 145 - None of the page-delivering ioctls default to the range that you
182 registered with. You must fill in all field 146 registered with. You must fill in all fields for the appropriate
183 ioctl struct including the range. 147 ioctl struct including the range.
184 148
185 - You get the address of the access that trigg 149 - You get the address of the access that triggered the missing page
186 event out of a struct uffd_msg that you read 150 event out of a struct uffd_msg that you read in the thread from the
187 uffd. You can supply as many pages as you w 151 uffd. You can supply as many pages as you want with these IOCTLs.
188 Keep in mind that unless you used DONTWAKE t 152 Keep in mind that unless you used DONTWAKE then the first of any of
189 those IOCTLs wakes up the faulting thread. 153 those IOCTLs wakes up the faulting thread.
190 154
191 - Be sure to test for all errors including 155 - Be sure to test for all errors including
192 (``pollfd[0].revents & POLLERR``). This can 156 (``pollfd[0].revents & POLLERR``). This can happen, e.g. when ranges
193 supplied were incorrect. 157 supplied were incorrect.
194 158
195 Write Protect Notifications 159 Write Protect Notifications
196 --------------------------- 160 ---------------------------
197 161
198 This is equivalent to (but faster than) using 162 This is equivalent to (but faster than) using mprotect and a SIGSEGV
199 signal handler. 163 signal handler.
200 164
201 Firstly you need to register a range with ``UF 165 Firstly you need to register a range with ``UFFDIO_REGISTER_MODE_WP``.
202 Instead of using mprotect(2) you use 166 Instead of using mprotect(2) you use
203 ``ioctl(uffd, UFFDIO_WRITEPROTECT, struct *uff 167 ``ioctl(uffd, UFFDIO_WRITEPROTECT, struct *uffdio_writeprotect)``
204 while ``mode = UFFDIO_WRITEPROTECT_MODE_WP`` 168 while ``mode = UFFDIO_WRITEPROTECT_MODE_WP``
205 in the struct passed in. The range does not d 169 in the struct passed in. The range does not default to and does not
206 have to be identical to the range you register 170 have to be identical to the range you registered with. You can write
207 protect as many ranges as you like (inside the 171 protect as many ranges as you like (inside the registered range).
208 Then, in the thread reading from uffd the stru 172 Then, in the thread reading from uffd the struct will have
209 ``msg.arg.pagefault.flags & UFFD_PAGEFAULT_FLA 173 ``msg.arg.pagefault.flags & UFFD_PAGEFAULT_FLAG_WP`` set. Now you send
210 ``ioctl(uffd, UFFDIO_WRITEPROTECT, struct *uff 174 ``ioctl(uffd, UFFDIO_WRITEPROTECT, struct *uffdio_writeprotect)``
211 again while ``pagefault.mode`` does not have ` 175 again while ``pagefault.mode`` does not have ``UFFDIO_WRITEPROTECT_MODE_WP``
212 set. This wakes up the thread which will conti 176 set. This wakes up the thread which will continue to run with writes. This
213 allows you to do the bookkeeping about the wri 177 allows you to do the bookkeeping about the write in the uffd reading
214 thread before the ioctl. 178 thread before the ioctl.
215 179
216 If you registered with both ``UFFDIO_REGISTER_ 180 If you registered with both ``UFFDIO_REGISTER_MODE_MISSING`` and
217 ``UFFDIO_REGISTER_MODE_WP`` then you need to t 181 ``UFFDIO_REGISTER_MODE_WP`` then you need to think about the sequence in
218 which you supply a page and undo write protect 182 which you supply a page and undo write protect. Note that there is a
219 difference between writes into a WP area and i 183 difference between writes into a WP area and into a !WP area. The
220 former will have ``UFFD_PAGEFAULT_FLAG_WP`` se 184 former will have ``UFFD_PAGEFAULT_FLAG_WP`` set, the latter
221 ``UFFD_PAGEFAULT_FLAG_WRITE``. The latter did 185 ``UFFD_PAGEFAULT_FLAG_WRITE``. The latter did not fail on protection but
222 you still need to supply a page when ``UFFDIO_ 186 you still need to supply a page when ``UFFDIO_REGISTER_MODE_MISSING`` was
223 used. 187 used.
224 <<
225 Userfaultfd write-protect mode currently behav <<
226 (when e.g. page is missing) over different typ <<
227 <<
228 For anonymous memory, ``ioctl(UFFDIO_WRITEPROT <<
229 (e.g. when pages are missing and not populated <<
230 like shmem and hugetlbfs, none ptes will be wr <<
231 present pte. In other words, there will be a <<
232 message generated when writing to a missing pa <<
233 as long as the page range was write-protected <<
234 not be generated on anonymous memories by defa <<
235 <<
236 If the application wants to be able to write p <<
237 memory, one can pre-populate the memory with e <<
238 newer kernels, one can also detect the feature <<
239 and set the feature bit in advance to make sur <<
240 write protected even upon anonymous memory. <<
241 <<
242 When using ``UFFDIO_REGISTER_MODE_WP`` in comb <<
243 ``UFFDIO_REGISTER_MODE_MISSING`` or ``UFFDIO_R <<
244 resolving missing / minor faults with ``UFFDIO <<
245 respectively, it may be desirable for the new <<
246 write-protected (so future writes will also re <<
247 support a mode flag (``UFFDIO_COPY_MODE_WP`` o <<
248 respectively) to configure the mapping this wa <<
249 <<
250 If the userfaultfd context has ``UFFD_FEATURE_ <<
251 any vma registered with write-protection will <<
252 than the default sync mode. <<
253 <<
254 In async mode, there will be no message genera <<
255 happens, meanwhile the write-protection will b <<
256 the kernel. It can be seen as a more accurate <<
257 tracking and it can be different in a few ways <<
258 <<
259 - The dirty result will not be affected by v <<
260 merging) because the dirty is only tracked <<
261 <<
262 - It supports range operations by default, s <<
263 any range of memory as long as page aligne <<
264 <<
265 - Dirty information will not get lost if the <<
266 various reasons (e.g. during split of a sh <<
267 <<
268 - Due to a reverted meaning of soft-dirty (p <<
269 set; dirty when uffd-wp bit cleared), it h <<
270 some of the memory operations. For exampl <<
271 anonymous (or ``MADV_REMOVE`` on a file ma <<
272 dirtying of memory by dropping uffd-wp bit <<
273 <<
274 The user app can collect the "written/dirty" s <<
275 uffd-wp bit for the pages being interested in <<
276 <<
277 The page will not be under track of uffd-wp as <<
278 explicitly write-protected by ``ioctl(UFFDIO_W <<
279 flag ``UFFDIO_WRITEPROTECT_MODE_WP`` set. Try <<
280 that was tracked by async mode userfaultfd-wp <<
281 <<
282 When userfaultfd-wp async mode is used alone, <<
283 kinds of memory. <<
284 <<
285 Memory Poisioning Emulation <<
286 --------------------------- <<
287 <<
288 In response to a fault (either missing or mino <<
289 take to "resolve" it is to issue a ``UFFDIO_PO <<
290 future faulters to either get a SIGBUS, or in <<
291 receive an MCE as if there were hardware memor <<
292 <<
293 This is used to emulate hardware memory poison <<
294 machine which experiences a real hardware memo <<
295 the VM to another physical machine. Since we w <<
296 transparent to the guest, we want that same ad <<
297 still poisoned, even though it's on a new phys <<
298 doesn't have a memory error in the exact same <<
299 188
300 QEMU/KVM 189 QEMU/KVM
301 ======== 190 ========
302 191
303 QEMU/KVM is using the ``userfaultfd`` syscall 192 QEMU/KVM is using the ``userfaultfd`` syscall to implement postcopy live
304 migration. Postcopy live migration is one form 193 migration. Postcopy live migration is one form of memory
305 externalization consisting of a virtual machin 194 externalization consisting of a virtual machine running with part or
306 all of its memory residing on a different node 195 all of its memory residing on a different node in the cloud. The
307 ``userfaultfd`` abstraction is generic enough 196 ``userfaultfd`` abstraction is generic enough that not a single line of
308 KVM kernel code had to be modified in order to 197 KVM kernel code had to be modified in order to add postcopy live
309 migration to QEMU. 198 migration to QEMU.
310 199
311 Guest async page faults, ``FOLL_NOWAIT`` and a 200 Guest async page faults, ``FOLL_NOWAIT`` and all other ``GUP*`` features work
312 just fine in combination with userfaults. User 201 just fine in combination with userfaults. Userfaults trigger async
313 page faults in the guest scheduler so those gu 202 page faults in the guest scheduler so those guest processes that
314 aren't waiting for userfaults (i.e. network bo 203 aren't waiting for userfaults (i.e. network bound) can keep running in
315 the guest vcpus. 204 the guest vcpus.
316 205
317 It is generally beneficial to run one pass of 206 It is generally beneficial to run one pass of precopy live migration
318 just before starting postcopy live migration, 207 just before starting postcopy live migration, in order to avoid
319 generating userfaults for readonly guest regio 208 generating userfaults for readonly guest regions.
320 209
321 The implementation of postcopy live migration 210 The implementation of postcopy live migration currently uses one
322 single bidirectional socket but in the future 211 single bidirectional socket but in the future two different sockets
323 will be used (to reduce the latency of the use 212 will be used (to reduce the latency of the userfaults to the minimum
324 possible without having to decrease ``/proc/sy 213 possible without having to decrease ``/proc/sys/net/ipv4/tcp_wmem``).
325 214
326 The QEMU in the source node writes all pages t 215 The QEMU in the source node writes all pages that it knows are missing
327 in the destination node, into the socket, and 216 in the destination node, into the socket, and the migration thread of
328 the QEMU running in the destination node runs 217 the QEMU running in the destination node runs ``UFFDIO_COPY|ZEROPAGE``
329 ioctls on the ``userfaultfd`` in order to map 218 ioctls on the ``userfaultfd`` in order to map the received pages into the
330 guest (``UFFDIO_ZEROCOPY`` is used if the sour 219 guest (``UFFDIO_ZEROCOPY`` is used if the source page was a zero page).
331 220
332 A different postcopy thread in the destination 221 A different postcopy thread in the destination node listens with
333 poll() to the ``userfaultfd`` in parallel. Whe 222 poll() to the ``userfaultfd`` in parallel. When a ``POLLIN`` event is
334 generated after a userfault triggers, the post 223 generated after a userfault triggers, the postcopy thread read() from
335 the ``userfaultfd`` and receives the fault add 224 the ``userfaultfd`` and receives the fault address (or ``-EAGAIN`` in case the
336 userfault was already resolved and waken by a 225 userfault was already resolved and waken by a ``UFFDIO_COPY|ZEROPAGE`` run
337 by the parallel QEMU migration thread). 226 by the parallel QEMU migration thread).
338 227
339 After the QEMU postcopy thread (running in the 228 After the QEMU postcopy thread (running in the destination node) gets
340 the userfault address it writes the informatio 229 the userfault address it writes the information about the missing page
341 into the socket. The QEMU source node receives 230 into the socket. The QEMU source node receives the information and
342 roughly "seeks" to that page address and conti 231 roughly "seeks" to that page address and continues sending all
343 remaining missing pages from that new page off 232 remaining missing pages from that new page offset. Soon after that
344 (just the time to flush the tcp_wmem queue thr 233 (just the time to flush the tcp_wmem queue through the network) the
345 migration thread in the QEMU running in the de 234 migration thread in the QEMU running in the destination node will
346 receive the page that triggered the userfault 235 receive the page that triggered the userfault and it'll map it as
347 usual with the ``UFFDIO_COPY|ZEROPAGE`` (witho 236 usual with the ``UFFDIO_COPY|ZEROPAGE`` (without actually knowing if it
348 was spontaneously sent by the source or if it 237 was spontaneously sent by the source or if it was an urgent page
349 requested through a userfault). 238 requested through a userfault).
350 239
351 By the time the userfaults start, the QEMU in 240 By the time the userfaults start, the QEMU in the destination node
352 doesn't need to keep any per-page state bitmap 241 doesn't need to keep any per-page state bitmap relative to the live
353 migration around and a single per-page bitmap 242 migration around and a single per-page bitmap has to be maintained in
354 the QEMU running in the source node to know wh 243 the QEMU running in the source node to know which pages are still
355 missing in the destination node. The bitmap in 244 missing in the destination node. The bitmap in the source node is
356 checked to find which missing pages to send in 245 checked to find which missing pages to send in round robin and we seek
357 over it when receiving incoming userfaults. Af 246 over it when receiving incoming userfaults. After sending each page of
358 course the bitmap is updated accordingly. It's 247 course the bitmap is updated accordingly. It's also useful to avoid
359 sending the same page twice (in case the userf 248 sending the same page twice (in case the userfault is read by the
360 postcopy thread just before ``UFFDIO_COPY|ZERO 249 postcopy thread just before ``UFFDIO_COPY|ZEROPAGE`` runs in the migration
361 thread). 250 thread).
362 251
363 Non-cooperative userfaultfd 252 Non-cooperative userfaultfd
364 =========================== 253 ===========================
365 254
366 When the ``userfaultfd`` is monitored by an ex 255 When the ``userfaultfd`` is monitored by an external manager, the manager
367 must be able to track changes in the process v 256 must be able to track changes in the process virtual memory
368 layout. Userfaultfd can notify the manager abo 257 layout. Userfaultfd can notify the manager about such changes using
369 the same read(2) protocol as for the page faul 258 the same read(2) protocol as for the page fault notifications. The
370 manager has to explicitly enable these events 259 manager has to explicitly enable these events by setting appropriate
371 bits in ``uffdio_api.features`` passed to ``UF 260 bits in ``uffdio_api.features`` passed to ``UFFDIO_API`` ioctl:
372 261
373 ``UFFD_FEATURE_EVENT_FORK`` 262 ``UFFD_FEATURE_EVENT_FORK``
374 enable ``userfaultfd`` hooks for fork( 263 enable ``userfaultfd`` hooks for fork(). When this feature is
375 enabled, the ``userfaultfd`` context o 264 enabled, the ``userfaultfd`` context of the parent process is
376 duplicated into the newly created proc 265 duplicated into the newly created process. The manager
377 receives ``UFFD_EVENT_FORK`` with file 266 receives ``UFFD_EVENT_FORK`` with file descriptor of the new
378 ``userfaultfd`` context in the ``uffd_ 267 ``userfaultfd`` context in the ``uffd_msg.fork``.
379 268
380 ``UFFD_FEATURE_EVENT_REMAP`` 269 ``UFFD_FEATURE_EVENT_REMAP``
381 enable notifications about mremap() ca 270 enable notifications about mremap() calls. When the
382 non-cooperative process moves a virtua 271 non-cooperative process moves a virtual memory area to a
383 different location, the manager will r 272 different location, the manager will receive
384 ``UFFD_EVENT_REMAP``. The ``uffd_msg.r 273 ``UFFD_EVENT_REMAP``. The ``uffd_msg.remap`` will contain the old and
385 new addresses of the area and its orig 274 new addresses of the area and its original length.
386 275
387 ``UFFD_FEATURE_EVENT_REMOVE`` 276 ``UFFD_FEATURE_EVENT_REMOVE``
388 enable notifications about madvise(MAD 277 enable notifications about madvise(MADV_REMOVE) and
389 madvise(MADV_DONTNEED) calls. The even 278 madvise(MADV_DONTNEED) calls. The event ``UFFD_EVENT_REMOVE`` will
390 be generated upon these calls to madvi 279 be generated upon these calls to madvise(). The ``uffd_msg.remove``
391 will contain start and end addresses o 280 will contain start and end addresses of the removed area.
392 281
393 ``UFFD_FEATURE_EVENT_UNMAP`` 282 ``UFFD_FEATURE_EVENT_UNMAP``
394 enable notifications about memory unma 283 enable notifications about memory unmapping. The manager will
395 get ``UFFD_EVENT_UNMAP`` with ``uffd_m 284 get ``UFFD_EVENT_UNMAP`` with ``uffd_msg.remove`` containing start and
396 end addresses of the unmapped area. 285 end addresses of the unmapped area.
397 286
398 Although the ``UFFD_FEATURE_EVENT_REMOVE`` and 287 Although the ``UFFD_FEATURE_EVENT_REMOVE`` and ``UFFD_FEATURE_EVENT_UNMAP``
399 are pretty similar, they quite differ in the a 288 are pretty similar, they quite differ in the action expected from the
400 ``userfaultfd`` manager. In the former case, t 289 ``userfaultfd`` manager. In the former case, the virtual memory is
401 removed, but the area is not, the area remains 290 removed, but the area is not, the area remains monitored by the
402 ``userfaultfd``, and if a page fault occurs in 291 ``userfaultfd``, and if a page fault occurs in that area it will be
403 delivered to the manager. The proper resolutio 292 delivered to the manager. The proper resolution for such page fault is
404 to zeromap the faulting address. However, in t 293 to zeromap the faulting address. However, in the latter case, when an
405 area is unmapped, either explicitly (with munm 294 area is unmapped, either explicitly (with munmap() system call), or
406 implicitly (e.g. during mremap()), the area is 295 implicitly (e.g. during mremap()), the area is removed and in turn the
407 ``userfaultfd`` context for such area disappea 296 ``userfaultfd`` context for such area disappears too and the manager will
408 not get further userland page faults from the 297 not get further userland page faults from the removed area. Still, the
409 notification is required in order to prevent m 298 notification is required in order to prevent manager from using
410 ``UFFDIO_COPY`` on the unmapped area. 299 ``UFFDIO_COPY`` on the unmapped area.
411 300
412 Unlike userland page faults which have to be s 301 Unlike userland page faults which have to be synchronous and require
413 explicit or implicit wakeup, all the events ar 302 explicit or implicit wakeup, all the events are delivered
414 asynchronously and the non-cooperative process 303 asynchronously and the non-cooperative process resumes execution as
415 soon as manager executes read(). The ``userfau 304 soon as manager executes read(). The ``userfaultfd`` manager should
416 carefully synchronize calls to ``UFFDIO_COPY`` 305 carefully synchronize calls to ``UFFDIO_COPY`` with the events
417 processing. To aid the synchronization, the `` 306 processing. To aid the synchronization, the ``UFFDIO_COPY`` ioctl will
418 return ``-ENOSPC`` when the monitored process 307 return ``-ENOSPC`` when the monitored process exits at the time of
419 ``UFFDIO_COPY``, and ``-ENOENT``, when the non 308 ``UFFDIO_COPY``, and ``-ENOENT``, when the non-cooperative process has changed
420 its virtual memory layout simultaneously with 309 its virtual memory layout simultaneously with outstanding ``UFFDIO_COPY``
421 operation. 310 operation.
422 311
423 The current asynchronous model of the event de 312 The current asynchronous model of the event delivery is optimal for
424 single threaded non-cooperative ``userfaultfd` 313 single threaded non-cooperative ``userfaultfd`` manager implementations. A
425 synchronous event delivery model can be added 314 synchronous event delivery model can be added later as a new
426 ``userfaultfd`` feature to facilitate multithr 315 ``userfaultfd`` feature to facilitate multithreading enhancements of the
427 non cooperative manager, for example to allow 316 non cooperative manager, for example to allow ``UFFDIO_COPY`` ioctls to
428 run in parallel to the event reception. Single 317 run in parallel to the event reception. Single threaded
429 implementations should continue to use the cur 318 implementations should continue to use the current async event
430 delivery model instead. 319 delivery model instead.
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.