xref: /linux/Documentation/admin-guide/mm/pagemap.rst (revision 00c010e130e58301db2ea0cec1eadc931e1cb8cf)
1=============================
2Examining Process Page Tables
3=============================
4
5pagemap is a new (as of 2.6.25) set of interfaces in the kernel that allow
6userspace programs to examine the page tables and related information by
7reading files in ``/proc``.
8
9There are four components to pagemap:
10
11 * ``/proc/pid/pagemap``.  This file lets a userspace process find out which
12   physical frame each virtual page is mapped to.  It contains one 64-bit
13   value for each virtual page, containing the following data (from
14   ``fs/proc/task_mmu.c``, above pagemap_read):
15
16    * Bits 0-54  page frame number (PFN) if present
17    * Bits 0-4   swap type if swapped
18    * Bits 5-54  swap offset if swapped
19    * Bit  55    pte is soft-dirty (see
20      Documentation/admin-guide/mm/soft-dirty.rst)
21    * Bit  56    page exclusively mapped (since 4.2)
22    * Bit  57    pte is uffd-wp write-protected (since 5.13) (see
23      Documentation/admin-guide/mm/userfaultfd.rst)
24    * Bit  58    pte is a guard region (since 6.15) (see madvise (2) man page)
25    * Bits 59-60 zero
26    * Bit  61    page is file-page or shared-anon (since 3.5)
27    * Bit  62    page swapped
28    * Bit  63    page present
29
30   Since Linux 4.0 only users with the CAP_SYS_ADMIN capability can get PFNs.
31   In 4.0 and 4.1 opens by unprivileged fail with -EPERM.  Starting from
32   4.2 the PFN field is zeroed if the user does not have CAP_SYS_ADMIN.
33   Reason: information about PFNs helps in exploiting Rowhammer vulnerability.
34
35   If the page is not present but in swap, then the PFN contains an
36   encoding of the swap file number and the page's offset into the
37   swap. Unmapped pages return a null PFN. This allows determining
38   precisely which pages are mapped (or in swap) and comparing mapped
39   pages between processes.
40
41   Traditionally, bit 56 indicates that a page is mapped exactly once and bit
42   56 is clear when a page is mapped multiple times, even when mapped in the
43   same process multiple times. In some kernel configurations, the semantics
44   for pages part of a larger allocation (e.g., THP) can differ: bit 56 is set
45   if all pages part of the corresponding large allocation are *certainly*
46   mapped in the same process, even if the page is mapped multiple times in that
47   process. Bit 56 is clear when any page page of the larger allocation
48   is *maybe* mapped in a different process. In some cases, a large allocation
49   might be treated as "maybe mapped by multiple processes" even though this
50   is no longer the case.
51
52   Efficient users of this interface will use ``/proc/pid/maps`` to
53   determine which areas of memory are actually mapped and llseek to
54   skip over unmapped regions.
55
56 * ``/proc/kpagecount``.  This file contains a 64-bit count of the number of
57   times each page is mapped, indexed by PFN. Some kernel configurations do
58   not track the precise number of times a page part of a larger allocation
59   (e.g., THP) is mapped. In these configurations, the average number of
60   mappings per page in this larger allocation is returned instead. However,
61   if any page of the large allocation is mapped, the returned value will
62   be at least 1.
63
64The page-types tool in the tools/mm directory can be used to query the
65number of times a page is mapped.
66
67 * ``/proc/kpageflags``.  This file contains a 64-bit set of flags for each
68   page, indexed by PFN.
69
70   The flags are (from ``fs/proc/page.c``, above kpageflags_read):
71
72    0. LOCKED
73    1. ERROR
74    2. REFERENCED
75    3. UPTODATE
76    4. DIRTY
77    5. LRU
78    6. ACTIVE
79    7. SLAB
80    8. WRITEBACK
81    9. RECLAIM
82    10. BUDDY
83    11. MMAP
84    12. ANON
85    13. SWAPCACHE
86    14. SWAPBACKED
87    15. COMPOUND_HEAD
88    16. COMPOUND_TAIL
89    17. HUGE
90    18. UNEVICTABLE
91    19. HWPOISON
92    20. NOPAGE
93    21. KSM
94    22. THP
95    23. OFFLINE
96    24. ZERO_PAGE
97    25. IDLE
98    26. PGTABLE
99
100 * ``/proc/kpagecgroup``.  This file contains a 64-bit inode number of the
101   memory cgroup each page is charged to, indexed by PFN. Only available when
102   CONFIG_MEMCG is set.
103
104Short descriptions to the page flags
105====================================
106
1070 - LOCKED
108   The page is being locked for exclusive access, e.g. by undergoing read/write
109   IO.
1107 - SLAB
111   The page is managed by the SLAB/SLUB kernel memory allocator.
112   When compound page is used, either will only set this flag on the head
113   page.
11410 - BUDDY
115    A free memory block managed by the buddy system allocator.
116    The buddy system organizes free memory in blocks of various orders.
117    An order N block has 2^N physically contiguous pages, with the BUDDY flag
118    set for and _only_ for the first page.
11915 - COMPOUND_HEAD
120    A compound page with order N consists of 2^N physically contiguous pages.
121    A compound page with order 2 takes the form of "HTTT", where H donates its
122    head page and T donates its tail page(s).  The major consumers of compound
123    pages are hugeTLB pages (Documentation/admin-guide/mm/hugetlbpage.rst),
124    the SLUB etc.  memory allocators and various device drivers.
125    However in this interface, only huge/giga pages are made visible
126    to end users.
12716 - COMPOUND_TAIL
128    A compound page tail (see description above).
12917 - HUGE
130    This is an integral part of a HugeTLB page.
13119 - HWPOISON
132    Hardware detected memory corruption on this page: don't touch the data!
13320 - NOPAGE
134    No page frame exists at the requested address.
13521 - KSM
136    Identical memory pages dynamically shared between one or more processes.
13722 - THP
138    Contiguous pages which construct THP of any size and mapped by any granularity.
13923 - OFFLINE
140    The page is logically offline.
14124 - ZERO_PAGE
142    Zero page for pfn_zero or huge_zero page.
14325 - IDLE
144    The page has not been accessed since it was marked idle (see
145    Documentation/admin-guide/mm/idle_page_tracking.rst).
146    Note that this flag may be stale in case the page was accessed via
147    a PTE. To make sure the flag is up-to-date one has to read
148    ``/sys/kernel/mm/page_idle/bitmap`` first.
14926 - PGTABLE
150    The page is in use as a page table.
151
152IO related page flags
153---------------------
154
1551 - ERROR
156   IO error occurred.
1573 - UPTODATE
158   The page has up-to-date data.
159   ie. for file backed page: (in-memory data revision >= on-disk one)
1604 - DIRTY
161   The page has been written to, hence contains new data.
162   i.e. for file backed page: (in-memory data revision >  on-disk one)
1638 - WRITEBACK
164   The page is being synced to disk.
165
166LRU related page flags
167----------------------
168
1695 - LRU
170   The page is in one of the LRU lists.
1716 - ACTIVE
172   The page is in the active LRU list.
17318 - UNEVICTABLE
174   The page is in the unevictable (non-)LRU list It is somehow pinned and
175   not a candidate for LRU page reclaims, e.g. ramfs pages,
176   shmctl(SHM_LOCK) and mlock() memory segments.
1772 - REFERENCED
178   The page has been referenced since last LRU list enqueue/requeue.
1799 - RECLAIM
180   The page will be reclaimed soon after its pageout IO completed.
18111 - MMAP
182   A memory mapped page.
18312 - ANON
184   A memory mapped page that is not part of a file.
18513 - SWAPCACHE
186   The page is mapped to swap space, i.e. has an associated swap entry.
18714 - SWAPBACKED
188   The page is backed by swap/RAM.
189
190The page-types tool in the tools/mm directory can be used to query the
191above flags.
192
193Exceptions for Shared Memory
194============================
195
196Page table entries for shared pages are cleared when the pages are zapped or
197swapped out. This makes swapped out pages indistinguishable from never-allocated
198ones.
199
200In kernel space, the swap location can still be retrieved from the page cache.
201However, values stored only on the normal PTE get lost irretrievably when the
202page is swapped out (i.e. SOFT_DIRTY).
203
204In user space, whether the page is present, swapped or none can be deduced with
205the help of lseek and/or mincore system calls.
206
207lseek() can differentiate between accessed pages (present or swapped out) and
208holes (none/non-allocated) by specifying the SEEK_DATA flag on the file where
209the pages are backed. For anonymous shared pages, the file can be found in
210``/proc/pid/map_files/``.
211
212mincore() can differentiate between pages in memory (present, including swap
213cache) and out of memory (swapped out or none/non-allocated).
214
215Other notes
216===========
217
218Reading from any of the files will return -EINVAL if you are not starting
219the read on an 8-byte boundary (e.g., if you sought an odd number of bytes
220into the file), or if the size of the read is not a multiple of 8 bytes.
221
222Before Linux 3.11 pagemap bits 55-60 were used for "page-shift" (which is
223always 12 at most architectures). Since Linux 3.11 their meaning changes
224after first clear of soft-dirty bits. Since Linux 4.2 they are used for
225flags unconditionally.
226
227Pagemap Scan IOCTL
228==================
229
230The ``PAGEMAP_SCAN`` IOCTL on the pagemap file can be used to get or optionally
231clear the info about page table entries. The following operations are supported
232in this IOCTL:
233
234- Scan the address range and get the memory ranges matching the provided criteria.
235  This is performed when the output buffer is specified.
236- Write-protect the pages. The ``PM_SCAN_WP_MATCHING`` is used to write-protect
237  the pages of interest. The ``PM_SCAN_CHECK_WPASYNC`` aborts the operation if
238  non-Async Write Protected pages are found. The ``PM_SCAN_WP_MATCHING`` can be
239  used with or without ``PM_SCAN_CHECK_WPASYNC``.
240- Both of those operations can be combined into one atomic operation where we can
241  get and write protect the pages as well.
242
243Following flags about pages are currently supported:
244
245- ``PAGE_IS_WPALLOWED`` - Page has async-write-protection enabled
246- ``PAGE_IS_WRITTEN`` - Page has been written to from the time it was write protected
247- ``PAGE_IS_FILE`` - Page is file backed
248- ``PAGE_IS_PRESENT`` - Page is present in the memory
249- ``PAGE_IS_SWAPPED`` - Page is in swapped
250- ``PAGE_IS_PFNZERO`` - Page has zero PFN
251- ``PAGE_IS_HUGE`` - Page is PMD-mapped THP or Hugetlb backed
252- ``PAGE_IS_SOFT_DIRTY`` - Page is soft-dirty
253- ``PAGE_IS_GUARD`` - Page is a part of a guard region
254
255The ``struct pm_scan_arg`` is used as the argument of the IOCTL.
256
257 1. The size of the ``struct pm_scan_arg`` must be specified in the ``size``
258    field. This field will be helpful in recognizing the structure if extensions
259    are done later.
260 2. The flags can be specified in the ``flags`` field. The ``PM_SCAN_WP_MATCHING``
261    and ``PM_SCAN_CHECK_WPASYNC`` are the only added flags at this time. The get
262    operation is optionally performed depending upon if the output buffer is
263    provided or not.
264 3. The range is specified through ``start`` and ``end``.
265 4. The walk can abort before visiting the complete range such as the user buffer
266    can get full etc. The walk ending address is specified in``end_walk``.
267 5. The output buffer of ``struct page_region`` array and size is specified in
268    ``vec`` and ``vec_len``.
269 6. The optional maximum requested pages are specified in the ``max_pages``.
270 7. The masks are specified in ``category_mask``, ``category_anyof_mask``,
271    ``category_inverted`` and ``return_mask``.
272
273Find pages which have been written and WP them as well::
274
275   struct pm_scan_arg arg = {
276   .size = sizeof(arg),
277   .flags = PM_SCAN_CHECK_WPASYNC | PM_SCAN_CHECK_WPASYNC,
278   ..
279   .category_mask = PAGE_IS_WRITTEN,
280   .return_mask = PAGE_IS_WRITTEN,
281   };
282
283Find pages which have been written, are file backed, not swapped and either
284present or huge::
285
286   struct pm_scan_arg arg = {
287   .size = sizeof(arg),
288   .flags = 0,
289   ..
290   .category_mask = PAGE_IS_WRITTEN | PAGE_IS_SWAPPED,
291   .category_inverted = PAGE_IS_SWAPPED,
292   .category_anyof_mask = PAGE_IS_PRESENT | PAGE_IS_HUGE,
293   .return_mask = PAGE_IS_WRITTEN | PAGE_IS_SWAPPED |
294                  PAGE_IS_PRESENT | PAGE_IS_HUGE,
295   };
296
297The ``PAGE_IS_WRITTEN`` flag can be considered as a better-performing alternative
298of soft-dirty flag. It doesn't get affected by VMA merging of the kernel and hence
299the user can find the true soft-dirty pages in case of normal pages. (There may
300still be extra dirty pages reported for THP or Hugetlb pages.)
301
302"PAGE_IS_WRITTEN" category is used with uffd write protect-enabled ranges to
303implement memory dirty tracking in userspace:
304
305 1. The userfaultfd file descriptor is created with ``userfaultfd`` syscall.
306 2. The ``UFFD_FEATURE_WP_UNPOPULATED`` and ``UFFD_FEATURE_WP_ASYNC`` features
307    are set by ``UFFDIO_API`` IOCTL.
308 3. The memory range is registered with ``UFFDIO_REGISTER_MODE_WP`` mode
309    through ``UFFDIO_REGISTER`` IOCTL.
310 4. Then any part of the registered memory or the whole memory region must
311    be write protected using ``PAGEMAP_SCAN`` IOCTL with flag ``PM_SCAN_WP_MATCHING``
312    or the ``UFFDIO_WRITEPROTECT`` IOCTL can be used. Both of these perform the
313    same operation. The former is better in terms of performance.
314 5. Now the ``PAGEMAP_SCAN`` IOCTL can be used to either just find pages which
315    have been written to since they were last marked and/or optionally write protect
316    the pages as well.
317