1==================
2Idle Page Tracking
3==================
4
5Motivation
6==========
7
8The idle page tracking feature allows to track which memory pages are being
9accessed by a workload and which are idle. This information can be useful for
10estimating the workload's working set size, which, in turn, can be taken into
11account when configuring the workload parameters, setting memory cgroup limits,
12or deciding where to place the workload within a compute cluster.
13
14It is enabled by CONFIG_IDLE_PAGE_TRACKING=y.
15
16.. _user_api:
17
18User API
19========
20
21The idle page tracking API is located at ``/sys/kernel/mm/page_idle``.
22Currently, it consists of the only read-write file,
23``/sys/kernel/mm/page_idle/bitmap``.
24
25The file implements a bitmap where each bit corresponds to a memory page. The
26bitmap is represented by an array of 8-byte integers, and the page at PFN #i is
27mapped to bit #i%64 of array element #i/64, byte order is native. When a bit is
28set, the corresponding page is idle.
29
30A page is considered idle if it has not been accessed since it was marked idle
31(for more details on what "accessed" actually means see the :ref:`Implementation
32Details <impl_details>` section).
33To mark a page idle one has to set the bit corresponding to
34the page by writing to the file. A value written to the file is OR-ed with the
35current bitmap value.
36
37Only accesses to user memory pages are tracked. These are pages mapped to a
38process address space, page cache and buffer pages, swap cache pages. For other
39page types (e.g. SLAB pages) an attempt to mark a page idle is silently ignored,
40and hence such pages are never reported idle.
41
42For huge pages the idle flag is set only on the head page, so one has to read
43``/proc/kpageflags`` in order to correctly count idle huge pages.
44
45Reading from or writing to ``/sys/kernel/mm/page_idle/bitmap`` will return
46-EINVAL if you are not starting the read/write on an 8-byte boundary, or
47if the size of the read/write is not a multiple of 8 bytes. Writing to
48this file beyond max PFN will return -ENXIO.
49
50That said, in order to estimate the amount of pages that are not used by a
51workload one should:
52
53 1. Mark all the workload's pages as idle by setting corresponding bits in
54    ``/sys/kernel/mm/page_idle/bitmap``. The pages can be found by reading
55    ``/proc/pid/pagemap`` if the workload is represented by a process, or by
56    filtering out alien pages using ``/proc/kpagecgroup`` in case the workload
57    is placed in a memory cgroup.
58
59 2. Wait until the workload accesses its working set.
60
61 3. Read ``/sys/kernel/mm/page_idle/bitmap`` and count the number of bits set.
62    If one wants to ignore certain types of pages, e.g. mlocked pages since they
63    are not reclaimable, he or she can filter them out using
64    ``/proc/kpageflags``.
65
66The page-types tool in the tools/mm directory can be used to assist in this.
67If the tool is run initially with the appropriate option, it will mark all the
68queried pages as idle.  Subsequent runs of the tool can then show which pages have
69their idle flag cleared in the interim.
70
71See Documentation/admin-guide/mm/pagemap.rst for more information about
72``/proc/pid/pagemap``, ``/proc/kpageflags``, and ``/proc/kpagecgroup``.
73
74.. _impl_details:
75
76Implementation Details
77======================
78
79The kernel internally keeps track of accesses to user memory pages in order to
80reclaim unreferenced pages first on memory shortage conditions. A page is
81considered referenced if it has been recently accessed via a process address
82space, in which case one or more PTEs it is mapped to will have the Accessed bit
83set, or marked accessed explicitly by the kernel (see mark_page_accessed()). The
84latter happens when:
85
86 - a userspace process reads or writes a page using a system call (e.g. read(2)
87   or write(2))
88
89 - a page that is used for storing filesystem buffers is read or written,
90   because a process needs filesystem metadata stored in it (e.g. lists a
91   directory tree)
92
93 - a page is accessed by a device driver using get_user_pages()
94
95When a dirty page is written to swap or disk as a result of memory reclaim or
96exceeding the dirty memory limit, it is not marked referenced.
97
98The idle memory tracking feature adds a new page flag, the Idle flag. This flag
99is set manually, by writing to ``/sys/kernel/mm/page_idle/bitmap`` (see the
100:ref:`User API <user_api>`
101section), and cleared automatically whenever a page is referenced as defined
102above.
103
104When a page is marked idle, the Accessed bit must be cleared in all PTEs it is
105mapped to, otherwise we will not be able to detect accesses to the page coming
106from a process address space. To avoid interference with the reclaimer, which,
107as noted above, uses the Accessed bit to promote actively referenced pages, one
108more page flag is introduced, the Young flag. When the PTE Accessed bit is
109cleared as a result of setting or updating a page's Idle flag, the Young flag
110is set on the page. The reclaimer treats the Young flag as an extra PTE
111Accessed bit and therefore will consider such a page as referenced.
112
113Since the idle memory tracking feature is based on the memory reclaimer logic,
114it only works with pages that are on an LRU list, other pages are silently
115ignored. That means it will ignore a user memory page if it is isolated, but
116since there are usually not many of them, it should not affect the overall
117result noticeably. In order not to stall scanning of the idle page bitmap,
118locked pages may be skipped too.
119