docs/vm: idle_page_tracking.txt: convert to ReST format

Signed-off-by: Mike Rapoport <rppt@linux.vnet.ibm.com>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
This commit is contained in:
Mike Rapoport 2018-03-21 21:22:26 +02:00 committed by Jonathan Corbet
parent b53ba58845
commit e3f2025a57

View File

@ -1,4 +1,11 @@
MOTIVATION .. _idle_page_tracking:
==================
Idle Page Tracking
==================
Motivation
==========
The idle page tracking feature allows to track which memory pages are being The idle page tracking feature allows to track which memory pages are being
accessed by a workload and which are idle. This information can be useful for accessed by a workload and which are idle. This information can be useful for
@ -8,10 +15,14 @@ or deciding where to place the workload within a compute cluster.
It is enabled by CONFIG_IDLE_PAGE_TRACKING=y. It is enabled by CONFIG_IDLE_PAGE_TRACKING=y.
USER API .. _user_api:
The idle page tracking API is located at /sys/kernel/mm/page_idle. Currently, User API
it consists of the only read-write file, /sys/kernel/mm/page_idle/bitmap. ========
The idle page tracking API is located at ``/sys/kernel/mm/page_idle``.
Currently, it consists of the only read-write file,
``/sys/kernel/mm/page_idle/bitmap``.
The file implements a bitmap where each bit corresponds to a memory page. The The file implements a bitmap where each bit corresponds to a memory page. The
bitmap is represented by an array of 8-byte integers, and the page at PFN #i is bitmap is represented by an array of 8-byte integers, and the page at PFN #i is
@ -19,8 +30,9 @@ mapped to bit #i%64 of array element #i/64, byte order is native. When a bit is
set, the corresponding page is idle. set, the corresponding page is idle.
A page is considered idle if it has not been accessed since it was marked idle A page is considered idle if it has not been accessed since it was marked idle
(for more details on what "accessed" actually means see the IMPLEMENTATION (for more details on what "accessed" actually means see the :ref:`Implementation
DETAILS section). To mark a page idle one has to set the bit corresponding to Details <impl_details>` section).
To mark a page idle one has to set the bit corresponding to
the page by writing to the file. A value written to the file is OR-ed with the the page by writing to the file. A value written to the file is OR-ed with the
current bitmap value. current bitmap value.
@ -30,9 +42,9 @@ page types (e.g. SLAB pages) an attempt to mark a page idle is silently ignored,
and hence such pages are never reported idle. and hence such pages are never reported idle.
For huge pages the idle flag is set only on the head page, so one has to read For huge pages the idle flag is set only on the head page, so one has to read
/proc/kpageflags in order to correctly count idle huge pages. ``/proc/kpageflags`` in order to correctly count idle huge pages.
Reading from or writing to /sys/kernel/mm/page_idle/bitmap will return Reading from or writing to ``/sys/kernel/mm/page_idle/bitmap`` will return
-EINVAL if you are not starting the read/write on an 8-byte boundary, or -EINVAL if you are not starting the read/write on an 8-byte boundary, or
if the size of the read/write is not a multiple of 8 bytes. Writing to if the size of the read/write is not a multiple of 8 bytes. Writing to
this file beyond max PFN will return -ENXIO. this file beyond max PFN will return -ENXIO.
@ -41,21 +53,25 @@ That said, in order to estimate the amount of pages that are not used by a
workload one should: workload one should:
1. Mark all the workload's pages as idle by setting corresponding bits in 1. Mark all the workload's pages as idle by setting corresponding bits in
/sys/kernel/mm/page_idle/bitmap. The pages can be found by reading ``/sys/kernel/mm/page_idle/bitmap``. The pages can be found by reading
/proc/pid/pagemap if the workload is represented by a process, or by ``/proc/pid/pagemap`` if the workload is represented by a process, or by
filtering out alien pages using /proc/kpagecgroup in case the workload is filtering out alien pages using ``/proc/kpagecgroup`` in case the workload
placed in a memory cgroup. is placed in a memory cgroup.
2. Wait until the workload accesses its working set. 2. Wait until the workload accesses its working set.
3. Read /sys/kernel/mm/page_idle/bitmap and count the number of bits set. If 3. Read ``/sys/kernel/mm/page_idle/bitmap`` and count the number of bits set.
one wants to ignore certain types of pages, e.g. mlocked pages since they If one wants to ignore certain types of pages, e.g. mlocked pages since they
are not reclaimable, he or she can filter them out using /proc/kpageflags. are not reclaimable, he or she can filter them out using
``/proc/kpageflags``.
See Documentation/vm/pagemap.txt for more information about /proc/pid/pagemap, See Documentation/vm/pagemap.txt for more information about
/proc/kpageflags, and /proc/kpagecgroup. ``/proc/pid/pagemap``, ``/proc/kpageflags``, and ``/proc/kpagecgroup``.
IMPLEMENTATION DETAILS .. _impl_details:
Implementation Details
======================
The kernel internally keeps track of accesses to user memory pages in order to The kernel internally keeps track of accesses to user memory pages in order to
reclaim unreferenced pages first on memory shortage conditions. A page is reclaim unreferenced pages first on memory shortage conditions. A page is
@ -77,7 +93,8 @@ When a dirty page is written to swap or disk as a result of memory reclaim or
exceeding the dirty memory limit, it is not marked referenced. exceeding the dirty memory limit, it is not marked referenced.
The idle memory tracking feature adds a new page flag, the Idle flag. This flag The idle memory tracking feature adds a new page flag, the Idle flag. This flag
is set manually, by writing to /sys/kernel/mm/page_idle/bitmap (see the USER API is set manually, by writing to ``/sys/kernel/mm/page_idle/bitmap`` (see the
:ref:`User API <user_api>`
section), and cleared automatically whenever a page is referenced as defined section), and cleared automatically whenever a page is referenced as defined
above. above.