3 =============================
4 Examining Process Page Tables
5 =============================
7 pagemap is a new (as of 2.6.25) set of interfaces in the kernel that allow
8 userspace programs to examine the page tables and related information by
9 reading files in ``/proc``.
11 There are four components to pagemap:
13 * ``/proc/pid/pagemap``. This file lets a userspace process find out which
14 physical frame each virtual page is mapped to. It contains one 64-bit
15 value for each virtual page, containing the following data (from
16 ``fs/proc/task_mmu.c``, above pagemap_read):
18 * Bits 0-54 page frame number (PFN) if present
19 * Bits 0-4 swap type if swapped
20 * Bits 5-54 swap offset if swapped
21 * Bit 55 pte is soft-dirty (see
22 :ref:`Documentation/admin-guide/mm/soft-dirty.rst <soft_dirty>`)
23 * Bit 56 page exclusively mapped (since 4.2)
24 * Bit 57 pte is uffd-wp write-protected (since 5.13) (see
25 :ref:`Documentation/admin-guide/mm/userfaultfd.rst <userfaultfd>`)
27 * Bit 61 page is file-page or shared-anon (since 3.5)
31 Since Linux 4.0 only users with the CAP_SYS_ADMIN capability can get PFNs.
32 In 4.0 and 4.1 opens by unprivileged fail with -EPERM. Starting from
33 4.2 the PFN field is zeroed if the user does not have CAP_SYS_ADMIN.
34 Reason: information about PFNs helps in exploiting Rowhammer vulnerability.
36 If the page is not present but in swap, then the PFN contains an
37 encoding of the swap file number and the page's offset into the
38 swap. Unmapped pages return a null PFN. This allows determining
39 precisely which pages are mapped (or in swap) and comparing mapped
40 pages between processes.
42 Efficient users of this interface will use ``/proc/pid/maps`` to
43 determine which areas of memory are actually mapped and llseek to
44 skip over unmapped regions.
46 * ``/proc/kpagecount``. This file contains a 64-bit count of the number of
47 times each page is mapped, indexed by PFN.
49 The page-types tool in the tools/vm directory can be used to query the
50 number of times a page is mapped.
52 * ``/proc/kpageflags``. This file contains a 64-bit set of flags for each
55 The flags are (from ``fs/proc/page.c``, above kpageflags_read):
85 * ``/proc/kpagecgroup``. This file contains a 64-bit inode number of the
86 memory cgroup each page is charged to, indexed by PFN. Only available when
89 Short descriptions to the page flags
90 ====================================
93 page is being locked for exclusive access, e.g. by undergoing read/write IO
95 page is managed by the SLAB/SLOB/SLUB/SLQB kernel memory allocator
96 When compound page is used, SLUB/SLQB will only set this flag on the head
97 page; SLOB will not flag it at all.
99 a free memory block managed by the buddy system allocator
100 The buddy system organizes free memory in blocks of various orders.
101 An order N block has 2^N physically contiguous pages, with the BUDDY flag
102 set for and _only_ for the first page.
104 A compound page with order N consists of 2^N physically contiguous pages.
105 A compound page with order 2 takes the form of "HTTT", where H donates its
106 head page and T donates its tail page(s). The major consumers of compound
107 pages are hugeTLB pages
108 (:ref:`Documentation/admin-guide/mm/hugetlbpage.rst <hugetlbpage>`),
109 the SLUB etc. memory allocators and various device drivers.
110 However in this interface, only huge/giga pages are made visible
113 A compound page tail (see description above).
115 this is an integral part of a HugeTLB page
117 hardware detected memory corruption on this page: don't touch the data!
119 no page frame exists at the requested address
121 identical memory pages dynamically shared between one or more processes
123 contiguous pages which construct transparent hugepages
125 page is logically offline
127 zero page for pfn_zero or huge_zero page
129 page has not been accessed since it was marked idle (see
130 :ref:`Documentation/admin-guide/mm/idle_page_tracking.rst <idle_page_tracking>`).
131 Note that this flag may be stale in case the page was accessed via
132 a PTE. To make sure the flag is up-to-date one has to read
133 ``/sys/kernel/mm/page_idle/bitmap`` first.
135 page is in use as a page table
137 IO related page flags
138 ---------------------
143 page has up-to-date data
144 ie. for file backed page: (in-memory data revision >= on-disk one)
146 page has been written to, hence contains new data
147 i.e. for file backed page: (in-memory data revision > on-disk one)
149 page is being synced to disk
151 LRU related page flags
152 ----------------------
155 page is in one of the LRU lists
157 page is in the active LRU list
159 page is in the unevictable (non-)LRU list It is somehow pinned and
160 not a candidate for LRU page reclaims, e.g. ramfs pages,
161 shmctl(SHM_LOCK) and mlock() memory segments
163 page has been referenced since last LRU list enqueue/requeue
165 page will be reclaimed soon after its pageout IO completed
169 a memory mapped page that is not part of a file
171 page is mapped to swap space, i.e. has an associated swap entry
173 page is backed by swap/RAM
175 The page-types tool in the tools/vm directory can be used to query the
178 Using pagemap to do something useful
179 ====================================
181 The general procedure for using pagemap to find out about a process' memory
182 usage goes like this:
184 1. Read ``/proc/pid/maps`` to determine which parts of the memory space are
186 2. Select the maps you are interested in -- all of them, or a particular
187 library, or the stack or the heap, etc.
188 3. Open ``/proc/pid/pagemap`` and seek to the pages you would like to examine.
189 4. Read a u64 for each page from pagemap.
190 5. Open ``/proc/kpagecount`` and/or ``/proc/kpageflags``. For each PFN you
191 just read, seek to that entry in the file, and read the data you want.
193 For example, to find the "unique set size" (USS), which is the amount of
194 memory that a process is using that is not shared with any other process,
195 you can go through every map in the process, find the PFNs, look those up
196 in kpagecount, and tally up the number of pages that are only referenced
202 Reading from any of the files will return -EINVAL if you are not starting
203 the read on an 8-byte boundary (e.g., if you sought an odd number of bytes
204 into the file), or if the size of the read is not a multiple of 8 bytes.
206 Before Linux 3.11 pagemap bits 55-60 were used for "page-shift" (which is
207 always 12 at most architectures). Since Linux 3.11 their meaning changes
208 after first clear of soft-dirty bits. Since Linux 4.2 they are used for
209 flags unconditionally.