1.. SPDX-License-Identifier: GPL-2.0
2
3======
4Design
5======
6
7
8.. _damon_design_execution_model_and_data_structures:
9
10Execution Model and Data Structures
11===================================
12
13The monitoring-related information including the monitoring request
14specification and DAMON-based operation schemes are stored in a data structure
15called DAMON ``context``.  DAMON executes each context with a kernel thread
16called ``kdamond``.  Multiple kdamonds could run in parallel, for different
17types of monitoring.
18
19To know how user-space can do the configurations and start/stop DAMON, refer to
20:ref:`DAMON sysfs interface <sysfs_interface>` documentation.
21
22
23Overall Architecture
24====================
25
26DAMON subsystem is configured with three layers including
27
28- :ref:`Operations Set <damon_operations_set>`: Implements fundamental
29  operations for DAMON that depends on the given monitoring target
30  address-space and available set of software/hardware primitives,
31- :ref:`Core <damon_core_logic>`: Implements core logics including monitoring
32  overhead/accuracy control and access-aware system operations on top of the
33  operations set layer, and
34- :ref:`Modules <damon_modules>`: Implements kernel modules for various
35  purposes that provides interfaces for the user space, on top of the core
36  layer.
37
38
39.. _damon_operations_set:
40
41Operations Set Layer
42====================
43
44.. _damon_design_configurable_operations_set:
45
46For data access monitoring and additional low level work, DAMON needs a set of
47implementations for specific operations that are dependent on and optimized for
48the given target address space.  For example, below two operations for access
49monitoring are address-space dependent.
50
511. Identification of the monitoring target address range for the address space.
522. Access check of specific address range in the target space.
53
54DAMON consolidates these implementations in a layer called DAMON Operations
55Set, and defines the interface between it and the upper layer.  The upper layer
56is dedicated for DAMON's core logics including the mechanism for control of the
57monitoring accruracy and the overhead.
58
59Hence, DAMON can easily be extended for any address space and/or available
60hardware features by configuring the core logic to use the appropriate
61operations set.  If there is no available operations set for a given purpose, a
62new operations set can be implemented following the interface between the
63layers.
64
65For example, physical memory, virtual memory, swap space, those for specific
66processes, NUMA nodes, files, and backing memory devices would be supportable.
67Also, if some architectures or devices support special optimized access check
68features, those will be easily configurable.
69
70DAMON currently provides below three operation sets.  Below two subsections
71describe how those work.
72
73 - vaddr: Monitor virtual address spaces of specific processes
74 - fvaddr: Monitor fixed virtual address ranges
75 - paddr: Monitor the physical address space of the system
76
77To know how user-space can do the configuration via :ref:`DAMON sysfs interface
78<sysfs_interface>`, refer to :ref:`operations <sysfs_context>` file part of the
79documentation.
80
81
82 .. _damon_design_vaddr_target_regions_construction:
83
84VMA-based Target Address Range Construction
85-------------------------------------------
86
87A mechanism of ``vaddr`` DAMON operations set that automatically initializes
88and updates the monitoring target address regions so that entire memory
89mappings of the target processes can be covered.
90
91This mechanism is only for the ``vaddr`` operations set.  In cases of
92``fvaddr`` and ``paddr`` operation sets, users are asked to manually set the
93monitoring target address ranges.
94
95Only small parts in the super-huge virtual address space of the processes are
96mapped to the physical memory and accessed.  Thus, tracking the unmapped
97address regions is just wasteful.  However, because DAMON can deal with some
98level of noise using the adaptive regions adjustment mechanism, tracking every
99mapping is not strictly required but could even incur a high overhead in some
100cases.  That said, too huge unmapped areas inside the monitoring target should
101be removed to not take the time for the adaptive mechanism.
102
103For the reason, this implementation converts the complex mappings to three
104distinct regions that cover every mapped area of the address space.  The two
105gaps between the three regions are the two biggest unmapped areas in the given
106address space.  The two biggest unmapped areas would be the gap between the
107heap and the uppermost mmap()-ed region, and the gap between the lowermost
108mmap()-ed region and the stack in most of the cases.  Because these gaps are
109exceptionally huge in usual address spaces, excluding these will be sufficient
110to make a reasonable trade-off.  Below shows this in detail::
111
112    <heap>
113    <BIG UNMAPPED REGION 1>
114    <uppermost mmap()-ed region>
115    (small mmap()-ed regions and munmap()-ed regions)
116    <lowermost mmap()-ed region>
117    <BIG UNMAPPED REGION 2>
118    <stack>
119
120
121PTE Accessed-bit Based Access Check
122-----------------------------------
123
124Both of the implementations for physical and virtual address spaces use PTE
125Accessed-bit for basic access checks.  Only one difference is the way of
126finding the relevant PTE Accessed bit(s) from the address.  While the
127implementation for the virtual address walks the page table for the target task
128of the address, the implementation for the physical address walks every page
129table having a mapping to the address.  In this way, the implementations find
130and clear the bit(s) for next sampling target address and checks whether the
131bit(s) set again after one sampling period.  This could disturb other kernel
132subsystems using the Accessed bits, namely Idle page tracking and the reclaim
133logic.  DAMON does nothing to avoid disturbing Idle page tracking, so handling
134the interference is the responsibility of sysadmins.  However, it solves the
135conflict with the reclaim logic using ``PG_idle`` and ``PG_young`` page flags,
136as Idle page tracking does.
137
138
139.. _damon_core_logic:
140
141Core Logics
142===========
143
144.. _damon_design_monitoring:
145
146Monitoring
147----------
148
149Below four sections describe each of the DAMON core mechanisms and the five
150monitoring attributes, ``sampling interval``, ``aggregation interval``,
151``update interval``, ``minimum number of regions``, and ``maximum number of
152regions``.
153
154To know how user-space can set the attributes via :ref:`DAMON sysfs interface
155<sysfs_interface>`, refer to :ref:`monitoring_attrs <sysfs_monitoring_attrs>`
156part of the documentation.
157
158
159Access Frequency Monitoring
160~~~~~~~~~~~~~~~~~~~~~~~~~~~
161
162The output of DAMON says what pages are how frequently accessed for a given
163duration.  The resolution of the access frequency is controlled by setting
164``sampling interval`` and ``aggregation interval``.  In detail, DAMON checks
165access to each page per ``sampling interval`` and aggregates the results.  In
166other words, counts the number of the accesses to each page.  After each
167``aggregation interval`` passes, DAMON calls callback functions that previously
168registered by users so that users can read the aggregated results and then
169clears the results.  This can be described in below simple pseudo-code::
170
171    while monitoring_on:
172        for page in monitoring_target:
173            if accessed(page):
174                nr_accesses[page] += 1
175        if time() % aggregation_interval == 0:
176            for callback in user_registered_callbacks:
177                callback(monitoring_target, nr_accesses)
178            for page in monitoring_target:
179                nr_accesses[page] = 0
180        sleep(sampling interval)
181
182The monitoring overhead of this mechanism will arbitrarily increase as the
183size of the target workload grows.
184
185
186.. _damon_design_region_based_sampling:
187
188Region Based Sampling
189~~~~~~~~~~~~~~~~~~~~~
190
191To avoid the unbounded increase of the overhead, DAMON groups adjacent pages
192that assumed to have the same access frequencies into a region.  As long as the
193assumption (pages in a region have the same access frequencies) is kept, only
194one page in the region is required to be checked.  Thus, for each ``sampling
195interval``, DAMON randomly picks one page in each region, waits for one
196``sampling interval``, checks whether the page is accessed meanwhile, and
197increases the access frequency counter of the region if so.  The counter is
198called ``nr_accesses`` of the region.  Therefore, the monitoring overhead is
199controllable by setting the number of regions.  DAMON allows users to set the
200minimum and the maximum number of regions for the trade-off.
201
202This scheme, however, cannot preserve the quality of the output if the
203assumption is not guaranteed.
204
205
206.. _damon_design_adaptive_regions_adjustment:
207
208Adaptive Regions Adjustment
209~~~~~~~~~~~~~~~~~~~~~~~~~~~
210
211Even somehow the initial monitoring target regions are well constructed to
212fulfill the assumption (pages in same region have similar access frequencies),
213the data access pattern can be dynamically changed.  This will result in low
214monitoring quality.  To keep the assumption as much as possible, DAMON
215adaptively merges and splits each region based on their access frequency.
216
217For each ``aggregation interval``, it compares the access frequencies
218(``nr_accesses``) of adjacent regions.  If the difference is small, and if the
219sum of the two regions' sizes is smaller than the size of total regions divided
220by the ``minimum number of regions``, DAMON merges the two regions.  If the
221resulting number of total regions is still higher than ``maximum number of
222regions``, it repeats the merging with increasing access frequenceis difference
223threshold until the upper-limit of the number of regions is met, or the
224threshold becomes higher than possible maximum value (``aggregation interval``
225divided by ``sampling interval``).   Then, after it reports and clears the
226aggregated access frequency of each region, it splits each region into two or
227three regions if the total number of regions will not exceed the user-specified
228maximum number of regions after the split.
229
230In this way, DAMON provides its best-effort quality and minimal overhead while
231keeping the bounds users set for their trade-off.
232
233
234.. _damon_design_age_tracking:
235
236Age Tracking
237~~~~~~~~~~~~
238
239By analyzing the monitoring results, users can also find how long the current
240access pattern of a region has maintained.  That could be used for good
241understanding of the access pattern.  For example, page placement algorithm
242utilizing both the frequency and the recency could be implemented using that.
243To make such access pattern maintained period analysis easier, DAMON maintains
244yet another counter called ``age`` in each region.  For each ``aggregation
245interval``, DAMON checks if the region's size and access frequency
246(``nr_accesses``) has significantly changed.  If so, the counter is reset to
247zero.  Otherwise, the counter is increased.
248
249
250Dynamic Target Space Updates Handling
251~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
252
253The monitoring target address range could dynamically changed.  For example,
254virtual memory could be dynamically mapped and unmapped.  Physical memory could
255be hot-plugged.
256
257As the changes could be quite frequent in some cases, DAMON allows the
258monitoring operations to check dynamic changes including memory mapping changes
259and applies it to monitoring operations-related data structures such as the
260abstracted monitoring target memory area only for each of a user-specified time
261interval (``update interval``).
262
263User-space can get the monitoring results via DAMON sysfs interface and/or
264tracepoints.  For more details, please refer to the documentations for
265:ref:`DAMOS tried regions <sysfs_schemes_tried_regions>` and :ref:`tracepoint`,
266respectively.
267
268
269.. _damon_design_monitoring_params_tuning_guide:
270
271Monitoring Parameters Tuning Guide
272~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
273
274In short, set ``aggregation interval`` to capture meaningful amount of accesses
275for the purpose.  The amount of accesses can be measured using ``nr_accesses``
276and ``age`` of regions in the aggregated monitoring results snapshot.  The
277default value of the interval, ``100ms``, turns out to be too short in many
278cases.  Set ``sampling interval`` proportional to ``aggregation interval``.  By
279default, ``1/20`` is recommended as the ratio.
280
281``Aggregation interval`` should be set as the time interval that the workload
282can make an amount of accesses for the monitoring purpose, within the interval.
283If the interval is too short, only small number of accesses are captured.  As a
284result, the monitoring results look everything is samely accessed only rarely.
285For many purposes, that would be useless.  If it is too long, however, the time
286to converge regions with the :ref:`regions adjustment mechanism
287<damon_design_adaptive_regions_adjustment>` can be too long, depending on the
288time scale of the given purpose.  This could happen if the workload is actually
289making only rare accesses but the user thinks the amount of accesses for the
290monitoring purpose too high.  For such cases, the target amount of access to
291capture per ``aggregation interval`` should carefully reconsidered.  Also, note
292that the captured amount of accesses is represented with not only
293``nr_accesses``, but also ``age``.  For example, even if every region on the
294monitoring results show zero ``nr_accesses``, regions could still be
295distinguished using ``age`` values as the recency information.
296
297Hence the optimum value of ``aggregation interval`` depends on the access
298intensiveness of the workload.  The user should tune the interval based on the
299amount of access that captured on each aggregated snapshot of the monitoring
300results.
301
302Note that the default value of the interval is 100 milliseconds, which is too
303short in many cases, especially on large systems.
304
305``Sampling interval`` defines the resolution of each aggregation.  If it is set
306too large, monitoring results will look like every region was samely rarely
307accessed, or samely frequently accessed.  That is, regions become
308undistinguishable based on access pattern, and therefore the results will be
309useless in many use cases.  If ``sampling interval`` is too small, it will not
310degrade the resolution, but will increase the monitoring overhead.  If it is
311appropriate enough to provide a resolution of the monitoring results that
312sufficient for the given purpose, it shouldn't be unnecessarily further
313lowered.  It is recommended to be set proportional to ``aggregation interval``.
314By default, the ratio is set as ``1/20``, and it is still recommended.
315
316Refer to below documents for an example tuning based on the above guide.
317
318.. toctree::
319   :maxdepth: 1
320
321   monitoring_intervals_tuning_example
322
323
324.. _damon_design_damos:
325
326Operation Schemes
327-----------------
328
329One common purpose of data access monitoring is access-aware system efficiency
330optimizations.  For example,
331
332    paging out memory regions that are not accessed for more than two minutes
333
334or
335
336    using THP for memory regions that are larger than 2 MiB and showing a high
337    access frequency for more than one minute.
338
339One straightforward approach for such schemes would be profile-guided
340optimizations.  That is, getting data access monitoring results of the
341workloads or the system using DAMON, finding memory regions of special
342characteristics by profiling the monitoring results, and making system
343operation changes for the regions.  The changes could be made by modifying or
344providing advice to the software (the application and/or the kernel), or
345reconfiguring the hardware.  Both offline and online approaches could be
346available.
347
348Among those, providing advice to the kernel at runtime would be flexible and
349effective, and therefore widely be used.   However, implementing such schemes
350could impose unnecessary redundancy and inefficiency.  The profiling could be
351redundant if the type of interest is common.  Exchanging the information
352including monitoring results and operation advice between kernel and user
353spaces could be inefficient.
354
355To allow users to reduce such redundancy and inefficiencies by offloading the
356works, DAMON provides a feature called Data Access Monitoring-based Operation
357Schemes (DAMOS).  It lets users specify their desired schemes at a high
358level.  For such specifications, DAMON starts monitoring, finds regions having
359the access pattern of interest, and applies the user-desired operation actions
360to the regions, for every user-specified time interval called
361``apply_interval``.
362
363To know how user-space can set ``apply_interval`` via :ref:`DAMON sysfs
364interface <sysfs_interface>`, refer to :ref:`apply_interval_us <sysfs_scheme>`
365part of the documentation.
366
367
368.. _damon_design_damos_action:
369
370Operation Action
371~~~~~~~~~~~~~~~~
372
373The management action that the users desire to apply to the regions of their
374interest.  For example, paging out, prioritizing for next reclamation victim
375selection, advising ``khugepaged`` to collapse or split, or doing nothing but
376collecting statistics of the regions.
377
378The list of supported actions is defined in DAMOS, but the implementation of
379each action is in the DAMON operations set layer because the implementation
380normally depends on the monitoring target address space.  For example, the code
381for paging specific virtual address ranges out would be different from that for
382physical address ranges.  And the monitoring operations implementation sets are
383not mandated to support all actions of the list.  Hence, the availability of
384specific DAMOS action depends on what operations set is selected to be used
385together.
386
387The list of the supported actions, their meaning, and DAMON operations sets
388that supports each action are as below.
389
390 - ``willneed``: Call ``madvise()`` for the region with ``MADV_WILLNEED``.
391   Supported by ``vaddr`` and ``fvaddr`` operations set.
392 - ``cold``: Call ``madvise()`` for the region with ``MADV_COLD``.
393   Supported by ``vaddr`` and ``fvaddr`` operations set.
394 - ``pageout``: Reclaim the region.
395   Supported by ``vaddr``, ``fvaddr`` and ``paddr`` operations set.
396 - ``hugepage``: Call ``madvise()`` for the region with ``MADV_HUGEPAGE``.
397   Supported by ``vaddr`` and ``fvaddr`` operations set.
398 - ``nohugepage``: Call ``madvise()`` for the region with ``MADV_NOHUGEPAGE``.
399   Supported by ``vaddr`` and ``fvaddr`` operations set.
400 - ``lru_prio``: Prioritize the region on its LRU lists.
401   Supported by ``paddr`` operations set.
402 - ``lru_deprio``: Deprioritize the region on its LRU lists.
403   Supported by ``paddr`` operations set.
404 - ``migrate_hot``: Migrate the regions prioritizing warmer regions.
405   Supported by ``paddr`` operations set.
406 - ``migrate_cold``: Migrate the regions prioritizing colder regions.
407   Supported by ``paddr`` operations set.
408 - ``stat``: Do nothing but count the statistics.
409   Supported by all operations sets.
410
411Applying the actions except ``stat`` to a region is considered as changing the
412region's characteristics.  Hence, DAMOS resets the age of regions when any such
413actions are applied to those.
414
415To know how user-space can set the action via :ref:`DAMON sysfs interface
416<sysfs_interface>`, refer to :ref:`action <sysfs_scheme>` part of the
417documentation.
418
419
420.. _damon_design_damos_access_pattern:
421
422Target Access Pattern
423~~~~~~~~~~~~~~~~~~~~~
424
425The access pattern of the schemes' interest.  The patterns are constructed with
426the properties that DAMON's monitoring results provide, specifically the size,
427the access frequency, and the age.  Users can describe their access pattern of
428interest by setting minimum and maximum values of the three properties.  If a
429region's three properties are in the ranges, DAMOS classifies it as one of the
430regions that the scheme is having an interest in.
431
432To know how user-space can set the access pattern via :ref:`DAMON sysfs
433interface <sysfs_interface>`, refer to :ref:`access_pattern
434<sysfs_access_pattern>` part of the documentation.
435
436
437.. _damon_design_damos_quotas:
438
439Quotas
440~~~~~~
441
442DAMOS upper-bound overhead control feature.  DAMOS could incur high overhead if
443the target access pattern is not properly tuned.  For example, if a huge memory
444region having the access pattern of interest is found, applying the scheme's
445action to all pages of the huge region could consume unacceptably large system
446resources.  Preventing such issues by tuning the access pattern could be
447challenging, especially if the access patterns of the workloads are highly
448dynamic.
449
450To mitigate that situation, DAMOS provides an upper-bound overhead control
451feature called quotas.  It lets users specify an upper limit of time that DAMOS
452can use for applying the action, and/or a maximum bytes of memory regions that
453the action can be applied within a user-specified time duration.
454
455To know how user-space can set the basic quotas via :ref:`DAMON sysfs interface
456<sysfs_interface>`, refer to :ref:`quotas <sysfs_quotas>` part of the
457documentation.
458
459
460.. _damon_design_damos_quotas_prioritization:
461
462Prioritization
463^^^^^^^^^^^^^^
464
465A mechanism for making a good decision under the quotas.  When the action
466cannot be applied to all regions of interest due to the quotas, DAMOS
467prioritizes regions and applies the action to only regions having high enough
468priorities so that it will not exceed the quotas.
469
470The prioritization mechanism should be different for each action.  For example,
471rarely accessed (colder) memory regions would be prioritized for page-out
472scheme action.  In contrast, the colder regions would be deprioritized for huge
473page collapse scheme action.  Hence, the prioritization mechanisms for each
474action are implemented in each DAMON operations set, together with the actions.
475
476Though the implementation is up to the DAMON operations set, it would be common
477to calculate the priority using the access pattern properties of the regions.
478Some users would want the mechanisms to be personalized for their specific
479case.  For example, some users would want the mechanism to weigh the recency
480(``age``) more than the access frequency (``nr_accesses``).  DAMOS allows users
481to specify the weight of each access pattern property and passes the
482information to the underlying mechanism.  Nevertheless, how and even whether
483the weight will be respected are up to the underlying prioritization mechanism
484implementation.
485
486To know how user-space can set the prioritization weights via :ref:`DAMON sysfs
487interface <sysfs_interface>`, refer to :ref:`weights <sysfs_quotas>` part of
488the documentation.
489
490
491.. _damon_design_damos_quotas_auto_tuning:
492
493Aim-oriented Feedback-driven Auto-tuning
494^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
495
496Automatic feedback-driven quota tuning.  Instead of setting the absolute quota
497value, users can specify the metric of their interest, and what target value
498they want the metric value to be.  DAMOS then automatically tunes the
499aggressiveness (the quota) of the corresponding scheme.  For example, if DAMOS
500is under achieving the goal, DAMOS automatically increases the quota.  If DAMOS
501is over achieving the goal, it decreases the quota.
502
503The goal can be specified with three parameters, namely ``target_metric``,
504``target_value``, and ``current_value``.  The auto-tuning mechanism tries to
505make ``current_value`` of ``target_metric`` be same to ``target_value``.
506Currently, two ``target_metric`` are provided.
507
508- ``user_input``: User-provided value.  Users could use any metric that they
509  has interest in for the value.  Use space main workload's latency or
510  throughput, system metrics like free memory ratio or memory pressure stall
511  time (PSI) could be examples.  Note that users should explicitly set
512  ``current_value`` on their own in this case.  In other words, users should
513  repeatedly provide the feedback.
514- ``some_mem_psi_us``: System-wide ``some`` memory pressure stall information
515  in microseconds that measured from last quota reset to next quota reset.
516  DAMOS does the measurement on its own, so only ``target_value`` need to be
517  set by users at the initial time.  In other words, DAMOS does self-feedback.
518
519To know how user-space can set the tuning goal metric, the target value, and/or
520the current value via :ref:`DAMON sysfs interface <sysfs_interface>`, refer to
521:ref:`quota goals <sysfs_schemes_quota_goals>` part of the documentation.
522
523
524.. _damon_design_damos_watermarks:
525
526Watermarks
527~~~~~~~~~~
528
529Conditional DAMOS (de)activation automation.  Users might want DAMOS to run
530only under certain situations.  For example, when a sufficient amount of free
531memory is guaranteed, running a scheme for proactive reclamation would only
532consume unnecessary system resources.  To avoid such consumption, the user would
533need to manually monitor some metrics such as free memory ratio, and turn
534DAMON/DAMOS on or off.
535
536DAMOS allows users to offload such works using three watermarks.  It allows the
537users to configure the metric of their interest, and three watermark values,
538namely high, middle, and low.  If the value of the metric becomes above the
539high watermark or below the low watermark, the scheme is deactivated.  If the
540metric becomes below the mid watermark but above the low watermark, the scheme
541is activated.  If all schemes are deactivated by the watermarks, the monitoring
542is also deactivated.  In this case, the DAMON worker thread only periodically
543checks the watermarks and therefore incurs nearly zero overhead.
544
545To know how user-space can set the watermarks via :ref:`DAMON sysfs interface
546<sysfs_interface>`, refer to :ref:`watermarks <sysfs_watermarks>` part of the
547documentation.
548
549
550.. _damon_design_damos_filters:
551
552Filters
553~~~~~~~
554
555Non-access pattern-based target memory regions filtering.  If users run
556self-written programs or have good profiling tools, they could know something
557more than the kernel, such as future access patterns or some special
558requirements for specific types of memory. For example, some users may know
559only anonymous pages can impact their program's performance.  They can also
560have a list of latency-critical processes.
561
562To let users optimize DAMOS schemes with such special knowledge, DAMOS provides
563a feature called DAMOS filters.  The feature allows users to set an arbitrary
564number of filters for each scheme.  Each filter specifies
565
566- a type of memory (``type``),
567- whether it is for the memory of the type or all except the type
568  (``matching``), and
569- whether it is to allow (include) or reject (exclude) applying
570  the scheme's action to the memory (``allow``).
571
572When multiple filters are installed, each filter is evaluated in the installed
573order.  If a part of memory is matched to one of the filter, next filters are
574ignored.  If the memory passes through the filters evaluation stage because it
575is not matched to any of the filters, applying the scheme's action to it is
576allowed, same to the behavior when no filter exists.
577
578For example, let's assume 1) a filter for allowing anonymous pages and 2)
579another filter for rejecting young pages are installed in the order.  If a page
580of a region that eligible to apply the scheme's action is an anonymous page,
581the scheme's action will be applied to the page regardless of whether it is
582young or not, since it matches with the first allow-filter.  If the page is
583not anonymous but young, the scheme's action will not be applied, since the
584second reject-filter blocks it.  If the page is neither anonymous nor young,
585the page will pass through the filters evaluation stage since there is no
586matching filter, and the action will be applied to the page.
587
588Note that the action can equally be applied to memory that either explicitly
589filter-allowed or filters evaluation stage passed.  It means that installing
590allow-filters at the end of the list makes no practical change but only
591filters-checking overhead.
592
593For efficient handling of filters, some types of filters are handled by the
594core layer, while others are handled by operations set.  In the latter case,
595hence, support of the filter types depends on the DAMON operations set.  In
596case of the core layer-handled filters, the memory regions that excluded by the
597filter are not counted as the scheme has tried to the region.  In contrast, if
598a memory regions is filtered by an operations set layer-handled filter, it is
599counted as the scheme has tried.  This difference affects the statistics.
600
601Below ``type`` of filters are currently supported.
602
603- anonymous page
604    - Applied to pages that containing data that not stored in files.
605    - Handled by operations set layer.  Supported by only ``paddr`` set.
606- memory cgroup
607    - Applied to pages that belonging to a given cgroup.
608    - Handled by operations set layer.  Supported by only ``paddr`` set.
609- young page
610    - Applied to pages that are accessed after the last access check from the
611      scheme.
612    - Handled by operations set layer.  Supported by only ``paddr`` set.
613- address range
614    - Applied to pages that belonging to a given address range.
615    - Handled by the core logic.
616- DAMON monitoring target
617    - Applied to pages that belonging to a given DAMON monitoring target.
618    - Handled by the core logic.
619
620To know how user-space can set the watermarks via :ref:`DAMON sysfs interface
621<sysfs_interface>`, refer to :ref:`filters <sysfs_filters>` part of the
622documentation.
623
624.. _damon_design_damos_stat:
625
626Statistics
627~~~~~~~~~~
628
629The statistics of DAMOS behaviors that designed to help monitoring, tuning and
630debugging of DAMOS.
631
632DAMOS accounts below statistics for each scheme, from the beginning of the
633scheme's execution.
634
635- ``nr_tried``: Total number of regions that the scheme is tried to be applied.
636- ``sz_trtied``: Total size of regions that the scheme is tried to be applied.
637- ``sz_ops_filter_passed``: Total bytes that passed operations set
638  layer-handled DAMOS filters.
639- ``nr_applied``: Total number of regions that the scheme is applied.
640- ``sz_applied``: Total size of regions that the scheme is applied.
641- ``qt_exceeds``: Total number of times the quota of the scheme has exceeded.
642
643"A scheme is tried to be applied to a region" means DAMOS core logic determined
644the region is eligible to apply the scheme's :ref:`action
645<damon_design_damos_action>`.  The :ref:`access pattern
646<damon_design_damos_access_pattern>`, :ref:`quotas
647<damon_design_damos_quotas>`, :ref:`watermarks
648<damon_design_damos_watermarks>`, and :ref:`filters
649<damon_design_damos_filters>` that handled on core logic could affect this.
650The core logic will only ask the underlying :ref:`operation set
651<damon_operations_set>` to do apply the action to the region, so whether the
652action is really applied or not is unclear.  That's why it is called "tried".
653
654"A scheme is applied to a region" means the :ref:`operation set
655<damon_operations_set>` has applied the action to at least a part of the
656region.  The :ref:`filters <damon_design_damos_filters>` that handled by the
657operation set, and the types of the :ref:`action <damon_design_damos_action>`
658and the pages of the region can affect this.  For example, if a filter is set
659to exclude anonymous pages and the region has only anonymous pages, or if the
660action is ``pageout`` while all pages of the region are unreclaimable, applying
661the action to the region will fail.
662
663To know how user-space can read the stats via :ref:`DAMON sysfs interface
664<sysfs_interface>`, refer to :ref:s`stats <sysfs_stats>` part of the
665documentation.
666
667Regions Walking
668~~~~~~~~~~~~~~~
669
670DAMOS feature allowing users access each region that a DAMOS action has just
671applied.  Using this feature, DAMON :ref:`API <damon_design_api>` allows users
672access full properties of the regions including the access monitoring results
673and amount of the region's internal memory that passed the DAMOS filters.
674:ref:`DAMON sysfs interface <sysfs_interface>` also allows users read the data
675via special :ref:`files <sysfs_schemes_tried_regions>`.
676
677.. _damon_design_api:
678
679Application Programming Interface
680---------------------------------
681
682The programming interface for kernel space data access-aware applications.
683DAMON is a framework, so it does nothing by itself.  Instead, it only helps
684other kernel components such as subsystems and modules building their data
685access-aware applications using DAMON's core features.  For this, DAMON exposes
686its all features to other kernel components via its application programming
687interface, namely ``include/linux/damon.h``.  Please refer to the API
688:doc:`document </mm/damon/api>` for details of the interface.
689
690
691.. _damon_modules:
692
693Modules
694=======
695
696Because the core of DAMON is a framework for kernel components, it doesn't
697provide any direct interface for the user space.  Such interfaces should be
698implemented by each DAMON API user kernel components, instead.  DAMON subsystem
699itself implements such DAMON API user modules, which are supposed to be used
700for general purpose DAMON control and special purpose data access-aware system
701operations, and provides stable application binary interfaces (ABI) for the
702user space.  The user space can build their efficient data access-aware
703applications using the interfaces.
704
705
706General Purpose User Interface Modules
707--------------------------------------
708
709DAMON modules that provide user space ABIs for general purpose DAMON usage in
710runtime.
711
712Like many other ABIs, the modules create files on pseudo file systems like
713'sysfs', allow users to specify their requests to and get the answers from
714DAMON by writing to and reading from the files.  As a response to such I/O,
715DAMON user interface modules control DAMON and retrieve the results as user
716requested via the DAMON API, and return the results to the user-space.
717
718The ABIs are designed to be used for user space applications development,
719rather than human beings' fingers.  Human users are recommended to use such
720user space tools.  One such Python-written user space tool is available at
721Github (https://github.com/damonitor/damo), Pypi
722(https://pypistats.org/packages/damo), and Fedora
723(https://packages.fedoraproject.org/pkgs/python-damo/damo/).
724
725Currently, one module for this type, namely 'DAMON sysfs interface' is
726available.  Please refer to the ABI :ref:`doc <sysfs_interface>` for details of
727the interfaces.
728
729
730Special-Purpose Access-aware Kernel Modules
731-------------------------------------------
732
733DAMON modules that provide user space ABI for specific purpose DAMON usage.
734
735DAMON user interface modules are for full control of all DAMON features in
736runtime.  For each special-purpose system-wide data access-aware system
737operations such as proactive reclamation or LRU lists balancing, the interfaces
738could be simplified by removing unnecessary knobs for the specific purpose, and
739extended for boot-time and even compile time control.  Default values of DAMON
740control parameters for the usage would also need to be optimized for the
741purpose.
742
743To support such cases, yet more DAMON API user kernel modules that provide more
744simple and optimized user space interfaces are available.  Currently, two
745modules for proactive reclamation and LRU lists manipulation are provided.  For
746more detail, please read the usage documents for those
747(:doc:`/admin-guide/mm/damon/reclaim` and
748:doc:`/admin-guide/mm/damon/lru_sort`).
749