1.. _cgroup-v2:
2
3================
4Control Group v2
5================
6
7:Date: October, 2015
8:Author: Tejun Heo <tj@kernel.org>
9
10This is the authoritative documentation on the design, interface and
11conventions of cgroup v2.  It describes all userland-visible aspects
12of cgroup including core and specific controller behaviors.  All
13future changes must be reflected in this document.  Documentation for
14v1 is available under :ref:`Documentation/admin-guide/cgroup-v1/index.rst <cgroup-v1>`.
15
16.. CONTENTS
17
18   1. Introduction
19     1-1. Terminology
20     1-2. What is cgroup?
21   2. Basic Operations
22     2-1. Mounting
23     2-2. Organizing Processes and Threads
24       2-2-1. Processes
25       2-2-2. Threads
26     2-3. [Un]populated Notification
27     2-4. Controlling Controllers
28       2-4-1. Enabling and Disabling
29       2-4-2. Top-down Constraint
30       2-4-3. No Internal Process Constraint
31     2-5. Delegation
32       2-5-1. Model of Delegation
33       2-5-2. Delegation Containment
34     2-6. Guidelines
35       2-6-1. Organize Once and Control
36       2-6-2. Avoid Name Collisions
37   3. Resource Distribution Models
38     3-1. Weights
39     3-2. Limits
40     3-3. Protections
41     3-4. Allocations
42   4. Interface Files
43     4-1. Format
44     4-2. Conventions
45     4-3. Core Interface Files
46   5. Controllers
47     5-1. CPU
48       5-1-1. CPU Interface Files
49     5-2. Memory
50       5-2-1. Memory Interface Files
51       5-2-2. Usage Guidelines
52       5-2-3. Memory Ownership
53     5-3. IO
54       5-3-1. IO Interface Files
55       5-3-2. Writeback
56       5-3-3. IO Latency
57         5-3-3-1. How IO Latency Throttling Works
58         5-3-3-2. IO Latency Interface Files
59       5-3-4. IO Priority
60     5-4. PID
61       5-4-1. PID Interface Files
62     5-5. Cpuset
63       5.5-1. Cpuset Interface Files
64     5-6. Device
65     5-7. RDMA
66       5-7-1. RDMA Interface Files
67     5-8. HugeTLB
68       5.8-1. HugeTLB Interface Files
69     5-9. Misc
70       5.9-1 Miscellaneous cgroup Interface Files
71       5.9-2 Migration and Ownership
72     5-10. Others
73       5-10-1. perf_event
74     5-N. Non-normative information
75       5-N-1. CPU controller root cgroup process behaviour
76       5-N-2. IO controller root cgroup process behaviour
77   6. Namespace
78     6-1. Basics
79     6-2. The Root and Views
80     6-3. Migration and setns(2)
81     6-4. Interaction with Other Namespaces
82   P. Information on Kernel Programming
83     P-1. Filesystem Support for Writeback
84   D. Deprecated v1 Core Features
85   R. Issues with v1 and Rationales for v2
86     R-1. Multiple Hierarchies
87     R-2. Thread Granularity
88     R-3. Competition Between Inner Nodes and Threads
89     R-4. Other Interface Issues
90     R-5. Controller Issues and Remedies
91       R-5-1. Memory
92
93
94Introduction
95============
96
97Terminology
98-----------
99
100"cgroup" stands for "control group" and is never capitalized.  The
101singular form is used to designate the whole feature and also as a
102qualifier as in "cgroup controllers".  When explicitly referring to
103multiple individual control groups, the plural form "cgroups" is used.
104
105
106What is cgroup?
107---------------
108
109cgroup is a mechanism to organize processes hierarchically and
110distribute system resources along the hierarchy in a controlled and
111configurable manner.
112
113cgroup is largely composed of two parts - the core and controllers.
114cgroup core is primarily responsible for hierarchically organizing
115processes.  A cgroup controller is usually responsible for
116distributing a specific type of system resource along the hierarchy
117although there are utility controllers which serve purposes other than
118resource distribution.
119
120cgroups form a tree structure and every process in the system belongs
121to one and only one cgroup.  All threads of a process belong to the
122same cgroup.  On creation, all processes are put in the cgroup that
123the parent process belongs to at the time.  A process can be migrated
124to another cgroup.  Migration of a process doesn't affect already
125existing descendant processes.
126
127Following certain structural constraints, controllers may be enabled or
128disabled selectively on a cgroup.  All controller behaviors are
129hierarchical - if a controller is enabled on a cgroup, it affects all
130processes which belong to the cgroups consisting the inclusive
131sub-hierarchy of the cgroup.  When a controller is enabled on a nested
132cgroup, it always restricts the resource distribution further.  The
133restrictions set closer to the root in the hierarchy can not be
134overridden from further away.
135
136
137Basic Operations
138================
139
140Mounting
141--------
142
143Unlike v1, cgroup v2 has only single hierarchy.  The cgroup v2
144hierarchy can be mounted with the following mount command::
145
146  # mount -t cgroup2 none $MOUNT_POINT
147
148cgroup2 filesystem has the magic number 0x63677270 ("cgrp").  All
149controllers which support v2 and are not bound to a v1 hierarchy are
150automatically bound to the v2 hierarchy and show up at the root.
151Controllers which are not in active use in the v2 hierarchy can be
152bound to other hierarchies.  This allows mixing v2 hierarchy with the
153legacy v1 multiple hierarchies in a fully backward compatible way.
154
155A controller can be moved across hierarchies only after the controller
156is no longer referenced in its current hierarchy.  Because per-cgroup
157controller states are destroyed asynchronously and controllers may
158have lingering references, a controller may not show up immediately on
159the v2 hierarchy after the final umount of the previous hierarchy.
160Similarly, a controller should be fully disabled to be moved out of
161the unified hierarchy and it may take some time for the disabled
162controller to become available for other hierarchies; furthermore, due
163to inter-controller dependencies, other controllers may need to be
164disabled too.
165
166While useful for development and manual configurations, moving
167controllers dynamically between the v2 and other hierarchies is
168strongly discouraged for production use.  It is recommended to decide
169the hierarchies and controller associations before starting using the
170controllers after system boot.
171
172During transition to v2, system management software might still
173automount the v1 cgroup filesystem and so hijack all controllers
174during boot, before manual intervention is possible. To make testing
175and experimenting easier, the kernel parameter cgroup_no_v1= allows
176disabling controllers in v1 and make them always available in v2.
177
178cgroup v2 currently supports the following mount options.
179
180  nsdelegate
181	Consider cgroup namespaces as delegation boundaries.  This
182	option is system wide and can only be set on mount or modified
183	through remount from the init namespace.  The mount option is
184	ignored on non-init namespace mounts.  Please refer to the
185	Delegation section for details.
186
187  favordynmods
188        Reduce the latencies of dynamic cgroup modifications such as
189        task migrations and controller on/offs at the cost of making
190        hot path operations such as forks and exits more expensive.
191        The static usage pattern of creating a cgroup, enabling
192        controllers, and then seeding it with CLONE_INTO_CGROUP is
193        not affected by this option.
194
195  memory_localevents
196        Only populate memory.events with data for the current cgroup,
197        and not any subtrees. This is legacy behaviour, the default
198        behaviour without this option is to include subtree counts.
199        This option is system wide and can only be set on mount or
200        modified through remount from the init namespace. The mount
201        option is ignored on non-init namespace mounts.
202
203  memory_recursiveprot
204        Recursively apply memory.min and memory.low protection to
205        entire subtrees, without requiring explicit downward
206        propagation into leaf cgroups.  This allows protecting entire
207        subtrees from one another, while retaining free competition
208        within those subtrees.  This should have been the default
209        behavior but is a mount-option to avoid regressing setups
210        relying on the original semantics (e.g. specifying bogusly
211        high 'bypass' protection values at higher tree levels).
212
213
214Organizing Processes and Threads
215--------------------------------
216
217Processes
218~~~~~~~~~
219
220Initially, only the root cgroup exists to which all processes belong.
221A child cgroup can be created by creating a sub-directory::
222
223  # mkdir $CGROUP_NAME
224
225A given cgroup may have multiple child cgroups forming a tree
226structure.  Each cgroup has a read-writable interface file
227"cgroup.procs".  When read, it lists the PIDs of all processes which
228belong to the cgroup one-per-line.  The PIDs are not ordered and the
229same PID may show up more than once if the process got moved to
230another cgroup and then back or the PID got recycled while reading.
231
232A process can be migrated into a cgroup by writing its PID to the
233target cgroup's "cgroup.procs" file.  Only one process can be migrated
234on a single write(2) call.  If a process is composed of multiple
235threads, writing the PID of any thread migrates all threads of the
236process.
237
238When a process forks a child process, the new process is born into the
239cgroup that the forking process belongs to at the time of the
240operation.  After exit, a process stays associated with the cgroup
241that it belonged to at the time of exit until it's reaped; however, a
242zombie process does not appear in "cgroup.procs" and thus can't be
243moved to another cgroup.
244
245A cgroup which doesn't have any children or live processes can be
246destroyed by removing the directory.  Note that a cgroup which doesn't
247have any children and is associated only with zombie processes is
248considered empty and can be removed::
249
250  # rmdir $CGROUP_NAME
251
252"/proc/$PID/cgroup" lists a process's cgroup membership.  If legacy
253cgroup is in use in the system, this file may contain multiple lines,
254one for each hierarchy.  The entry for cgroup v2 is always in the
255format "0::$PATH"::
256
257  # cat /proc/842/cgroup
258  ...
259  0::/test-cgroup/test-cgroup-nested
260
261If the process becomes a zombie and the cgroup it was associated with
262is removed subsequently, " (deleted)" is appended to the path::
263
264  # cat /proc/842/cgroup
265  ...
266  0::/test-cgroup/test-cgroup-nested (deleted)
267
268
269Threads
270~~~~~~~
271
272cgroup v2 supports thread granularity for a subset of controllers to
273support use cases requiring hierarchical resource distribution across
274the threads of a group of processes.  By default, all threads of a
275process belong to the same cgroup, which also serves as the resource
276domain to host resource consumptions which are not specific to a
277process or thread.  The thread mode allows threads to be spread across
278a subtree while still maintaining the common resource domain for them.
279
280Controllers which support thread mode are called threaded controllers.
281The ones which don't are called domain controllers.
282
283Marking a cgroup threaded makes it join the resource domain of its
284parent as a threaded cgroup.  The parent may be another threaded
285cgroup whose resource domain is further up in the hierarchy.  The root
286of a threaded subtree, that is, the nearest ancestor which is not
287threaded, is called threaded domain or thread root interchangeably and
288serves as the resource domain for the entire subtree.
289
290Inside a threaded subtree, threads of a process can be put in
291different cgroups and are not subject to the no internal process
292constraint - threaded controllers can be enabled on non-leaf cgroups
293whether they have threads in them or not.
294
295As the threaded domain cgroup hosts all the domain resource
296consumptions of the subtree, it is considered to have internal
297resource consumptions whether there are processes in it or not and
298can't have populated child cgroups which aren't threaded.  Because the
299root cgroup is not subject to no internal process constraint, it can
300serve both as a threaded domain and a parent to domain cgroups.
301
302The current operation mode or type of the cgroup is shown in the
303"cgroup.type" file which indicates whether the cgroup is a normal
304domain, a domain which is serving as the domain of a threaded subtree,
305or a threaded cgroup.
306
307On creation, a cgroup is always a domain cgroup and can be made
308threaded by writing "threaded" to the "cgroup.type" file.  The
309operation is single direction::
310
311  # echo threaded > cgroup.type
312
313Once threaded, the cgroup can't be made a domain again.  To enable the
314thread mode, the following conditions must be met.
315
316- As the cgroup will join the parent's resource domain.  The parent
317  must either be a valid (threaded) domain or a threaded cgroup.
318
319- When the parent is an unthreaded domain, it must not have any domain
320  controllers enabled or populated domain children.  The root is
321  exempt from this requirement.
322
323Topology-wise, a cgroup can be in an invalid state.  Please consider
324the following topology::
325
326  A (threaded domain) - B (threaded) - C (domain, just created)
327
328C is created as a domain but isn't connected to a parent which can
329host child domains.  C can't be used until it is turned into a
330threaded cgroup.  "cgroup.type" file will report "domain (invalid)" in
331these cases.  Operations which fail due to invalid topology use
332EOPNOTSUPP as the errno.
333
334A domain cgroup is turned into a threaded domain when one of its child
335cgroup becomes threaded or threaded controllers are enabled in the
336"cgroup.subtree_control" file while there are processes in the cgroup.
337A threaded domain reverts to a normal domain when the conditions
338clear.
339
340When read, "cgroup.threads" contains the list of the thread IDs of all
341threads in the cgroup.  Except that the operations are per-thread
342instead of per-process, "cgroup.threads" has the same format and
343behaves the same way as "cgroup.procs".  While "cgroup.threads" can be
344written to in any cgroup, as it can only move threads inside the same
345threaded domain, its operations are confined inside each threaded
346subtree.
347
348The threaded domain cgroup serves as the resource domain for the whole
349subtree, and, while the threads can be scattered across the subtree,
350all the processes are considered to be in the threaded domain cgroup.
351"cgroup.procs" in a threaded domain cgroup contains the PIDs of all
352processes in the subtree and is not readable in the subtree proper.
353However, "cgroup.procs" can be written to from anywhere in the subtree
354to migrate all threads of the matching process to the cgroup.
355
356Only threaded controllers can be enabled in a threaded subtree.  When
357a threaded controller is enabled inside a threaded subtree, it only
358accounts for and controls resource consumptions associated with the
359threads in the cgroup and its descendants.  All consumptions which
360aren't tied to a specific thread belong to the threaded domain cgroup.
361
362Because a threaded subtree is exempt from no internal process
363constraint, a threaded controller must be able to handle competition
364between threads in a non-leaf cgroup and its child cgroups.  Each
365threaded controller defines how such competitions are handled.
366
367
368[Un]populated Notification
369--------------------------
370
371Each non-root cgroup has a "cgroup.events" file which contains
372"populated" field indicating whether the cgroup's sub-hierarchy has
373live processes in it.  Its value is 0 if there is no live process in
374the cgroup and its descendants; otherwise, 1.  poll and [id]notify
375events are triggered when the value changes.  This can be used, for
376example, to start a clean-up operation after all processes of a given
377sub-hierarchy have exited.  The populated state updates and
378notifications are recursive.  Consider the following sub-hierarchy
379where the numbers in the parentheses represent the numbers of processes
380in each cgroup::
381
382  A(4) - B(0) - C(1)
383              \ D(0)
384
385A, B and C's "populated" fields would be 1 while D's 0.  After the one
386process in C exits, B and C's "populated" fields would flip to "0" and
387file modified events will be generated on the "cgroup.events" files of
388both cgroups.
389
390
391Controlling Controllers
392-----------------------
393
394Enabling and Disabling
395~~~~~~~~~~~~~~~~~~~~~~
396
397Each cgroup has a "cgroup.controllers" file which lists all
398controllers available for the cgroup to enable::
399
400  # cat cgroup.controllers
401  cpu io memory
402
403No controller is enabled by default.  Controllers can be enabled and
404disabled by writing to the "cgroup.subtree_control" file::
405
406  # echo "+cpu +memory -io" > cgroup.subtree_control
407
408Only controllers which are listed in "cgroup.controllers" can be
409enabled.  When multiple operations are specified as above, either they
410all succeed or fail.  If multiple operations on the same controller
411are specified, the last one is effective.
412
413Enabling a controller in a cgroup indicates that the distribution of
414the target resource across its immediate children will be controlled.
415Consider the following sub-hierarchy.  The enabled controllers are
416listed in parentheses::
417
418  A(cpu,memory) - B(memory) - C()
419                            \ D()
420
421As A has "cpu" and "memory" enabled, A will control the distribution
422of CPU cycles and memory to its children, in this case, B.  As B has
423"memory" enabled but not "CPU", C and D will compete freely on CPU
424cycles but their division of memory available to B will be controlled.
425
426As a controller regulates the distribution of the target resource to
427the cgroup's children, enabling it creates the controller's interface
428files in the child cgroups.  In the above example, enabling "cpu" on B
429would create the "cpu." prefixed controller interface files in C and
430D.  Likewise, disabling "memory" from B would remove the "memory."
431prefixed controller interface files from C and D.  This means that the
432controller interface files - anything which doesn't start with
433"cgroup." are owned by the parent rather than the cgroup itself.
434
435
436Top-down Constraint
437~~~~~~~~~~~~~~~~~~~
438
439Resources are distributed top-down and a cgroup can further distribute
440a resource only if the resource has been distributed to it from the
441parent.  This means that all non-root "cgroup.subtree_control" files
442can only contain controllers which are enabled in the parent's
443"cgroup.subtree_control" file.  A controller can be enabled only if
444the parent has the controller enabled and a controller can't be
445disabled if one or more children have it enabled.
446
447
448No Internal Process Constraint
449~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
450
451Non-root cgroups can distribute domain resources to their children
452only when they don't have any processes of their own.  In other words,
453only domain cgroups which don't contain any processes can have domain
454controllers enabled in their "cgroup.subtree_control" files.
455
456This guarantees that, when a domain controller is looking at the part
457of the hierarchy which has it enabled, processes are always only on
458the leaves.  This rules out situations where child cgroups compete
459against internal processes of the parent.
460
461The root cgroup is exempt from this restriction.  Root contains
462processes and anonymous resource consumption which can't be associated
463with any other cgroups and requires special treatment from most
464controllers.  How resource consumption in the root cgroup is governed
465is up to each controller (for more information on this topic please
466refer to the Non-normative information section in the Controllers
467chapter).
468
469Note that the restriction doesn't get in the way if there is no
470enabled controller in the cgroup's "cgroup.subtree_control".  This is
471important as otherwise it wouldn't be possible to create children of a
472populated cgroup.  To control resource distribution of a cgroup, the
473cgroup must create children and transfer all its processes to the
474children before enabling controllers in its "cgroup.subtree_control"
475file.
476
477
478Delegation
479----------
480
481Model of Delegation
482~~~~~~~~~~~~~~~~~~~
483
484A cgroup can be delegated in two ways.  First, to a less privileged
485user by granting write access of the directory and its "cgroup.procs",
486"cgroup.threads" and "cgroup.subtree_control" files to the user.
487Second, if the "nsdelegate" mount option is set, automatically to a
488cgroup namespace on namespace creation.
489
490Because the resource control interface files in a given directory
491control the distribution of the parent's resources, the delegatee
492shouldn't be allowed to write to them.  For the first method, this is
493achieved by not granting access to these files.  For the second, the
494kernel rejects writes to all files other than "cgroup.procs" and
495"cgroup.subtree_control" on a namespace root from inside the
496namespace.
497
498The end results are equivalent for both delegation types.  Once
499delegated, the user can build sub-hierarchy under the directory,
500organize processes inside it as it sees fit and further distribute the
501resources it received from the parent.  The limits and other settings
502of all resource controllers are hierarchical and regardless of what
503happens in the delegated sub-hierarchy, nothing can escape the
504resource restrictions imposed by the parent.
505
506Currently, cgroup doesn't impose any restrictions on the number of
507cgroups in or nesting depth of a delegated sub-hierarchy; however,
508this may be limited explicitly in the future.
509
510
511Delegation Containment
512~~~~~~~~~~~~~~~~~~~~~~
513
514A delegated sub-hierarchy is contained in the sense that processes
515can't be moved into or out of the sub-hierarchy by the delegatee.
516
517For delegations to a less privileged user, this is achieved by
518requiring the following conditions for a process with a non-root euid
519to migrate a target process into a cgroup by writing its PID to the
520"cgroup.procs" file.
521
522- The writer must have write access to the "cgroup.procs" file.
523
524- The writer must have write access to the "cgroup.procs" file of the
525  common ancestor of the source and destination cgroups.
526
527The above two constraints ensure that while a delegatee may migrate
528processes around freely in the delegated sub-hierarchy it can't pull
529in from or push out to outside the sub-hierarchy.
530
531For an example, let's assume cgroups C0 and C1 have been delegated to
532user U0 who created C00, C01 under C0 and C10 under C1 as follows and
533all processes under C0 and C1 belong to U0::
534
535  ~~~~~~~~~~~~~ - C0 - C00
536  ~ cgroup    ~      \ C01
537  ~ hierarchy ~
538  ~~~~~~~~~~~~~ - C1 - C10
539
540Let's also say U0 wants to write the PID of a process which is
541currently in C10 into "C00/cgroup.procs".  U0 has write access to the
542file; however, the common ancestor of the source cgroup C10 and the
543destination cgroup C00 is above the points of delegation and U0 would
544not have write access to its "cgroup.procs" files and thus the write
545will be denied with -EACCES.
546
547For delegations to namespaces, containment is achieved by requiring
548that both the source and destination cgroups are reachable from the
549namespace of the process which is attempting the migration.  If either
550is not reachable, the migration is rejected with -ENOENT.
551
552
553Guidelines
554----------
555
556Organize Once and Control
557~~~~~~~~~~~~~~~~~~~~~~~~~
558
559Migrating a process across cgroups is a relatively expensive operation
560and stateful resources such as memory are not moved together with the
561process.  This is an explicit design decision as there often exist
562inherent trade-offs between migration and various hot paths in terms
563of synchronization cost.
564
565As such, migrating processes across cgroups frequently as a means to
566apply different resource restrictions is discouraged.  A workload
567should be assigned to a cgroup according to the system's logical and
568resource structure once on start-up.  Dynamic adjustments to resource
569distribution can be made by changing controller configuration through
570the interface files.
571
572
573Avoid Name Collisions
574~~~~~~~~~~~~~~~~~~~~~
575
576Interface files for a cgroup and its children cgroups occupy the same
577directory and it is possible to create children cgroups which collide
578with interface files.
579
580All cgroup core interface files are prefixed with "cgroup." and each
581controller's interface files are prefixed with the controller name and
582a dot.  A controller's name is composed of lower case alphabets and
583'_'s but never begins with an '_' so it can be used as the prefix
584character for collision avoidance.  Also, interface file names won't
585start or end with terms which are often used in categorizing workloads
586such as job, service, slice, unit or workload.
587
588cgroup doesn't do anything to prevent name collisions and it's the
589user's responsibility to avoid them.
590
591
592Resource Distribution Models
593============================
594
595cgroup controllers implement several resource distribution schemes
596depending on the resource type and expected use cases.  This section
597describes major schemes in use along with their expected behaviors.
598
599
600Weights
601-------
602
603A parent's resource is distributed by adding up the weights of all
604active children and giving each the fraction matching the ratio of its
605weight against the sum.  As only children which can make use of the
606resource at the moment participate in the distribution, this is
607work-conserving.  Due to the dynamic nature, this model is usually
608used for stateless resources.
609
610All weights are in the range [1, 10000] with the default at 100.  This
611allows symmetric multiplicative biases in both directions at fine
612enough granularity while staying in the intuitive range.
613
614As long as the weight is in range, all configuration combinations are
615valid and there is no reason to reject configuration changes or
616process migrations.
617
618"cpu.weight" proportionally distributes CPU cycles to active children
619and is an example of this type.
620
621
622.. _cgroupv2-limits-distributor:
623
624Limits
625------
626
627A child can only consume up to the configured amount of the resource.
628Limits can be over-committed - the sum of the limits of children can
629exceed the amount of resource available to the parent.
630
631Limits are in the range [0, max] and defaults to "max", which is noop.
632
633As limits can be over-committed, all configuration combinations are
634valid and there is no reason to reject configuration changes or
635process migrations.
636
637"io.max" limits the maximum BPS and/or IOPS that a cgroup can consume
638on an IO device and is an example of this type.
639
640.. _cgroupv2-protections-distributor:
641
642Protections
643-----------
644
645A cgroup is protected up to the configured amount of the resource
646as long as the usages of all its ancestors are under their
647protected levels.  Protections can be hard guarantees or best effort
648soft boundaries.  Protections can also be over-committed in which case
649only up to the amount available to the parent is protected among
650children.
651
652Protections are in the range [0, max] and defaults to 0, which is
653noop.
654
655As protections can be over-committed, all configuration combinations
656are valid and there is no reason to reject configuration changes or
657process migrations.
658
659"memory.low" implements best-effort memory protection and is an
660example of this type.
661
662
663Allocations
664-----------
665
666A cgroup is exclusively allocated a certain amount of a finite
667resource.  Allocations can't be over-committed - the sum of the
668allocations of children can not exceed the amount of resource
669available to the parent.
670
671Allocations are in the range [0, max] and defaults to 0, which is no
672resource.
673
674As allocations can't be over-committed, some configuration
675combinations are invalid and should be rejected.  Also, if the
676resource is mandatory for execution of processes, process migrations
677may be rejected.
678
679"cpu.rt.max" hard-allocates realtime slices and is an example of this
680type.
681
682
683Interface Files
684===============
685
686Format
687------
688
689All interface files should be in one of the following formats whenever
690possible::
691
692  New-line separated values
693  (when only one value can be written at once)
694
695	VAL0\n
696	VAL1\n
697	...
698
699  Space separated values
700  (when read-only or multiple values can be written at once)
701
702	VAL0 VAL1 ...\n
703
704  Flat keyed
705
706	KEY0 VAL0\n
707	KEY1 VAL1\n
708	...
709
710  Nested keyed
711
712	KEY0 SUB_KEY0=VAL00 SUB_KEY1=VAL01...
713	KEY1 SUB_KEY0=VAL10 SUB_KEY1=VAL11...
714	...
715
716For a writable file, the format for writing should generally match
717reading; however, controllers may allow omitting later fields or
718implement restricted shortcuts for most common use cases.
719
720For both flat and nested keyed files, only the values for a single key
721can be written at a time.  For nested keyed files, the sub key pairs
722may be specified in any order and not all pairs have to be specified.
723
724
725Conventions
726-----------
727
728- Settings for a single feature should be contained in a single file.
729
730- The root cgroup should be exempt from resource control and thus
731  shouldn't have resource control interface files.
732
733- The default time unit is microseconds.  If a different unit is ever
734  used, an explicit unit suffix must be present.
735
736- A parts-per quantity should use a percentage decimal with at least
737  two digit fractional part - e.g. 13.40.
738
739- If a controller implements weight based resource distribution, its
740  interface file should be named "weight" and have the range [1,
741  10000] with 100 as the default.  The values are chosen to allow
742  enough and symmetric bias in both directions while keeping it
743  intuitive (the default is 100%).
744
745- If a controller implements an absolute resource guarantee and/or
746  limit, the interface files should be named "min" and "max"
747  respectively.  If a controller implements best effort resource
748  guarantee and/or limit, the interface files should be named "low"
749  and "high" respectively.
750
751  In the above four control files, the special token "max" should be
752  used to represent upward infinity for both reading and writing.
753
754- If a setting has a configurable default value and keyed specific
755  overrides, the default entry should be keyed with "default" and
756  appear as the first entry in the file.
757
758  The default value can be updated by writing either "default $VAL" or
759  "$VAL".
760
761  When writing to update a specific override, "default" can be used as
762  the value to indicate removal of the override.  Override entries
763  with "default" as the value must not appear when read.
764
765  For example, a setting which is keyed by major:minor device numbers
766  with integer values may look like the following::
767
768    # cat cgroup-example-interface-file
769    default 150
770    8:0 300
771
772  The default value can be updated by::
773
774    # echo 125 > cgroup-example-interface-file
775
776  or::
777
778    # echo "default 125" > cgroup-example-interface-file
779
780  An override can be set by::
781
782    # echo "8:16 170" > cgroup-example-interface-file
783
784  and cleared by::
785
786    # echo "8:0 default" > cgroup-example-interface-file
787    # cat cgroup-example-interface-file
788    default 125
789    8:16 170
790
791- For events which are not very high frequency, an interface file
792  "events" should be created which lists event key value pairs.
793  Whenever a notifiable event happens, file modified event should be
794  generated on the file.
795
796
797Core Interface Files
798--------------------
799
800All cgroup core files are prefixed with "cgroup."
801
802  cgroup.type
803	A read-write single value file which exists on non-root
804	cgroups.
805
806	When read, it indicates the current type of the cgroup, which
807	can be one of the following values.
808
809	- "domain" : A normal valid domain cgroup.
810
811	- "domain threaded" : A threaded domain cgroup which is
812          serving as the root of a threaded subtree.
813
814	- "domain invalid" : A cgroup which is in an invalid state.
815	  It can't be populated or have controllers enabled.  It may
816	  be allowed to become a threaded cgroup.
817
818	- "threaded" : A threaded cgroup which is a member of a
819          threaded subtree.
820
821	A cgroup can be turned into a threaded cgroup by writing
822	"threaded" to this file.
823
824  cgroup.procs
825	A read-write new-line separated values file which exists on
826	all cgroups.
827
828	When read, it lists the PIDs of all processes which belong to
829	the cgroup one-per-line.  The PIDs are not ordered and the
830	same PID may show up more than once if the process got moved
831	to another cgroup and then back or the PID got recycled while
832	reading.
833
834	A PID can be written to migrate the process associated with
835	the PID to the cgroup.  The writer should match all of the
836	following conditions.
837
838	- It must have write access to the "cgroup.procs" file.
839
840	- It must have write access to the "cgroup.procs" file of the
841	  common ancestor of the source and destination cgroups.
842
843	When delegating a sub-hierarchy, write access to this file
844	should be granted along with the containing directory.
845
846	In a threaded cgroup, reading this file fails with EOPNOTSUPP
847	as all the processes belong to the thread root.  Writing is
848	supported and moves every thread of the process to the cgroup.
849
850  cgroup.threads
851	A read-write new-line separated values file which exists on
852	all cgroups.
853
854	When read, it lists the TIDs of all threads which belong to
855	the cgroup one-per-line.  The TIDs are not ordered and the
856	same TID may show up more than once if the thread got moved to
857	another cgroup and then back or the TID got recycled while
858	reading.
859
860	A TID can be written to migrate the thread associated with the
861	TID to the cgroup.  The writer should match all of the
862	following conditions.
863
864	- It must have write access to the "cgroup.threads" file.
865
866	- The cgroup that the thread is currently in must be in the
867          same resource domain as the destination cgroup.
868
869	- It must have write access to the "cgroup.procs" file of the
870	  common ancestor of the source and destination cgroups.
871
872	When delegating a sub-hierarchy, write access to this file
873	should be granted along with the containing directory.
874
875  cgroup.controllers
876	A read-only space separated values file which exists on all
877	cgroups.
878
879	It shows space separated list of all controllers available to
880	the cgroup.  The controllers are not ordered.
881
882  cgroup.subtree_control
883	A read-write space separated values file which exists on all
884	cgroups.  Starts out empty.
885
886	When read, it shows space separated list of the controllers
887	which are enabled to control resource distribution from the
888	cgroup to its children.
889
890	Space separated list of controllers prefixed with '+' or '-'
891	can be written to enable or disable controllers.  A controller
892	name prefixed with '+' enables the controller and '-'
893	disables.  If a controller appears more than once on the list,
894	the last one is effective.  When multiple enable and disable
895	operations are specified, either all succeed or all fail.
896
897  cgroup.events
898	A read-only flat-keyed file which exists on non-root cgroups.
899	The following entries are defined.  Unless specified
900	otherwise, a value change in this file generates a file
901	modified event.
902
903	  populated
904		1 if the cgroup or its descendants contains any live
905		processes; otherwise, 0.
906	  frozen
907		1 if the cgroup is frozen; otherwise, 0.
908
909  cgroup.max.descendants
910	A read-write single value files.  The default is "max".
911
912	Maximum allowed number of descent cgroups.
913	If the actual number of descendants is equal or larger,
914	an attempt to create a new cgroup in the hierarchy will fail.
915
916  cgroup.max.depth
917	A read-write single value files.  The default is "max".
918
919	Maximum allowed descent depth below the current cgroup.
920	If the actual descent depth is equal or larger,
921	an attempt to create a new child cgroup will fail.
922
923  cgroup.stat
924	A read-only flat-keyed file with the following entries:
925
926	  nr_descendants
927		Total number of visible descendant cgroups.
928
929	  nr_dying_descendants
930		Total number of dying descendant cgroups. A cgroup becomes
931		dying after being deleted by a user. The cgroup will remain
932		in dying state for some time undefined time (which can depend
933		on system load) before being completely destroyed.
934
935		A process can't enter a dying cgroup under any circumstances,
936		a dying cgroup can't revive.
937
938		A dying cgroup can consume system resources not exceeding
939		limits, which were active at the moment of cgroup deletion.
940
941  cgroup.freeze
942	A read-write single value file which exists on non-root cgroups.
943	Allowed values are "0" and "1". The default is "0".
944
945	Writing "1" to the file causes freezing of the cgroup and all
946	descendant cgroups. This means that all belonging processes will
947	be stopped and will not run until the cgroup will be explicitly
948	unfrozen. Freezing of the cgroup may take some time; when this action
949	is completed, the "frozen" value in the cgroup.events control file
950	will be updated to "1" and the corresponding notification will be
951	issued.
952
953	A cgroup can be frozen either by its own settings, or by settings
954	of any ancestor cgroups. If any of ancestor cgroups is frozen, the
955	cgroup will remain frozen.
956
957	Processes in the frozen cgroup can be killed by a fatal signal.
958	They also can enter and leave a frozen cgroup: either by an explicit
959	move by a user, or if freezing of the cgroup races with fork().
960	If a process is moved to a frozen cgroup, it stops. If a process is
961	moved out of a frozen cgroup, it becomes running.
962
963	Frozen status of a cgroup doesn't affect any cgroup tree operations:
964	it's possible to delete a frozen (and empty) cgroup, as well as
965	create new sub-cgroups.
966
967  cgroup.kill
968	A write-only single value file which exists in non-root cgroups.
969	The only allowed value is "1".
970
971	Writing "1" to the file causes the cgroup and all descendant cgroups to
972	be killed. This means that all processes located in the affected cgroup
973	tree will be killed via SIGKILL.
974
975	Killing a cgroup tree will deal with concurrent forks appropriately and
976	is protected against migrations.
977
978	In a threaded cgroup, writing this file fails with EOPNOTSUPP as
979	killing cgroups is a process directed operation, i.e. it affects
980	the whole thread-group.
981
982  cgroup.pressure
983	A read-write single value file that allowed values are "0" and "1".
984	The default is "1".
985
986	Writing "0" to the file will disable the cgroup PSI accounting.
987	Writing "1" to the file will re-enable the cgroup PSI accounting.
988
989	This control attribute is not hierarchical, so disable or enable PSI
990	accounting in a cgroup does not affect PSI accounting in descendants
991	and doesn't need pass enablement via ancestors from root.
992
993	The reason this control attribute exists is that PSI accounts stalls for
994	each cgroup separately and aggregates it at each level of the hierarchy.
995	This may cause non-negligible overhead for some workloads when under
996	deep level of the hierarchy, in which case this control attribute can
997	be used to disable PSI accounting in the non-leaf cgroups.
998
999  irq.pressure
1000	A read-write nested-keyed file.
1001
1002	Shows pressure stall information for IRQ/SOFTIRQ. See
1003	:ref:`Documentation/accounting/psi.rst <psi>` for details.
1004
1005Controllers
1006===========
1007
1008.. _cgroup-v2-cpu:
1009
1010CPU
1011---
1012
1013The "cpu" controllers regulates distribution of CPU cycles.  This
1014controller implements weight and absolute bandwidth limit models for
1015normal scheduling policy and absolute bandwidth allocation model for
1016realtime scheduling policy.
1017
1018In all the above models, cycles distribution is defined only on a temporal
1019base and it does not account for the frequency at which tasks are executed.
1020The (optional) utilization clamping support allows to hint the schedutil
1021cpufreq governor about the minimum desired frequency which should always be
1022provided by a CPU, as well as the maximum desired frequency, which should not
1023be exceeded by a CPU.
1024
1025WARNING: cgroup2 doesn't yet support control of realtime processes and
1026the cpu controller can only be enabled when all RT processes are in
1027the root cgroup.  Be aware that system management software may already
1028have placed RT processes into nonroot cgroups during the system boot
1029process, and these processes may need to be moved to the root cgroup
1030before the cpu controller can be enabled.
1031
1032
1033CPU Interface Files
1034~~~~~~~~~~~~~~~~~~~
1035
1036All time durations are in microseconds.
1037
1038  cpu.stat
1039	A read-only flat-keyed file.
1040	This file exists whether the controller is enabled or not.
1041
1042	It always reports the following three stats:
1043
1044	- usage_usec
1045	- user_usec
1046	- system_usec
1047
1048	and the following three when the controller is enabled:
1049
1050	- nr_periods
1051	- nr_throttled
1052	- throttled_usec
1053	- nr_bursts
1054	- burst_usec
1055
1056  cpu.weight
1057	A read-write single value file which exists on non-root
1058	cgroups.  The default is "100".
1059
1060	The weight in the range [1, 10000].
1061
1062  cpu.weight.nice
1063	A read-write single value file which exists on non-root
1064	cgroups.  The default is "0".
1065
1066	The nice value is in the range [-20, 19].
1067
1068	This interface file is an alternative interface for
1069	"cpu.weight" and allows reading and setting weight using the
1070	same values used by nice(2).  Because the range is smaller and
1071	granularity is coarser for the nice values, the read value is
1072	the closest approximation of the current weight.
1073
1074  cpu.max
1075	A read-write two value file which exists on non-root cgroups.
1076	The default is "max 100000".
1077
1078	The maximum bandwidth limit.  It's in the following format::
1079
1080	  $MAX $PERIOD
1081
1082	which indicates that the group may consume up to $MAX in each
1083	$PERIOD duration.  "max" for $MAX indicates no limit.  If only
1084	one number is written, $MAX is updated.
1085
1086  cpu.max.burst
1087	A read-write single value file which exists on non-root
1088	cgroups.  The default is "0".
1089
1090	The burst in the range [0, $MAX].
1091
1092  cpu.pressure
1093	A read-write nested-keyed file.
1094
1095	Shows pressure stall information for CPU. See
1096	:ref:`Documentation/accounting/psi.rst <psi>` for details.
1097
1098  cpu.uclamp.min
1099        A read-write single value file which exists on non-root cgroups.
1100        The default is "0", i.e. no utilization boosting.
1101
1102        The requested minimum utilization (protection) as a percentage
1103        rational number, e.g. 12.34 for 12.34%.
1104
1105        This interface allows reading and setting minimum utilization clamp
1106        values similar to the sched_setattr(2). This minimum utilization
1107        value is used to clamp the task specific minimum utilization clamp.
1108
1109        The requested minimum utilization (protection) is always capped by
1110        the current value for the maximum utilization (limit), i.e.
1111        `cpu.uclamp.max`.
1112
1113  cpu.uclamp.max
1114        A read-write single value file which exists on non-root cgroups.
1115        The default is "max". i.e. no utilization capping
1116
1117        The requested maximum utilization (limit) as a percentage rational
1118        number, e.g. 98.76 for 98.76%.
1119
1120        This interface allows reading and setting maximum utilization clamp
1121        values similar to the sched_setattr(2). This maximum utilization
1122        value is used to clamp the task specific maximum utilization clamp.
1123
1124
1125
1126Memory
1127------
1128
1129The "memory" controller regulates distribution of memory.  Memory is
1130stateful and implements both limit and protection models.  Due to the
1131intertwining between memory usage and reclaim pressure and the
1132stateful nature of memory, the distribution model is relatively
1133complex.
1134
1135While not completely water-tight, all major memory usages by a given
1136cgroup are tracked so that the total memory consumption can be
1137accounted and controlled to a reasonable extent.  Currently, the
1138following types of memory usages are tracked.
1139
1140- Userland memory - page cache and anonymous memory.
1141
1142- Kernel data structures such as dentries and inodes.
1143
1144- TCP socket buffers.
1145
1146The above list may expand in the future for better coverage.
1147
1148
1149Memory Interface Files
1150~~~~~~~~~~~~~~~~~~~~~~
1151
1152All memory amounts are in bytes.  If a value which is not aligned to
1153PAGE_SIZE is written, the value may be rounded up to the closest
1154PAGE_SIZE multiple when read back.
1155
1156  memory.current
1157	A read-only single value file which exists on non-root
1158	cgroups.
1159
1160	The total amount of memory currently being used by the cgroup
1161	and its descendants.
1162
1163  memory.min
1164	A read-write single value file which exists on non-root
1165	cgroups.  The default is "0".
1166
1167	Hard memory protection.  If the memory usage of a cgroup
1168	is within its effective min boundary, the cgroup's memory
1169	won't be reclaimed under any conditions. If there is no
1170	unprotected reclaimable memory available, OOM killer
1171	is invoked. Above the effective min boundary (or
1172	effective low boundary if it is higher), pages are reclaimed
1173	proportionally to the overage, reducing reclaim pressure for
1174	smaller overages.
1175
1176	Effective min boundary is limited by memory.min values of
1177	all ancestor cgroups. If there is memory.min overcommitment
1178	(child cgroup or cgroups are requiring more protected memory
1179	than parent will allow), then each child cgroup will get
1180	the part of parent's protection proportional to its
1181	actual memory usage below memory.min.
1182
1183	Putting more memory than generally available under this
1184	protection is discouraged and may lead to constant OOMs.
1185
1186	If a memory cgroup is not populated with processes,
1187	its memory.min is ignored.
1188
1189  memory.low
1190	A read-write single value file which exists on non-root
1191	cgroups.  The default is "0".
1192
1193	Best-effort memory protection.  If the memory usage of a
1194	cgroup is within its effective low boundary, the cgroup's
1195	memory won't be reclaimed unless there is no reclaimable
1196	memory available in unprotected cgroups.
1197	Above the effective low	boundary (or
1198	effective min boundary if it is higher), pages are reclaimed
1199	proportionally to the overage, reducing reclaim pressure for
1200	smaller overages.
1201
1202	Effective low boundary is limited by memory.low values of
1203	all ancestor cgroups. If there is memory.low overcommitment
1204	(child cgroup or cgroups are requiring more protected memory
1205	than parent will allow), then each child cgroup will get
1206	the part of parent's protection proportional to its
1207	actual memory usage below memory.low.
1208
1209	Putting more memory than generally available under this
1210	protection is discouraged.
1211
1212  memory.high
1213	A read-write single value file which exists on non-root
1214	cgroups.  The default is "max".
1215
1216	Memory usage throttle limit.  If a cgroup's usage goes
1217	over the high boundary, the processes of the cgroup are
1218	throttled and put under heavy reclaim pressure.
1219
1220	Going over the high limit never invokes the OOM killer and
1221	under extreme conditions the limit may be breached. The high
1222	limit should be used in scenarios where an external process
1223	monitors the limited cgroup to alleviate heavy reclaim
1224	pressure.
1225
1226  memory.max
1227	A read-write single value file which exists on non-root
1228	cgroups.  The default is "max".
1229
1230	Memory usage hard limit.  This is the main mechanism to limit
1231	memory usage of a cgroup.  If a cgroup's memory usage reaches
1232	this limit and can't be reduced, the OOM killer is invoked in
1233	the cgroup. Under certain circumstances, the usage may go
1234	over the limit temporarily.
1235
1236	In default configuration regular 0-order allocations always
1237	succeed unless OOM killer chooses current task as a victim.
1238
1239	Some kinds of allocations don't invoke the OOM killer.
1240	Caller could retry them differently, return into userspace
1241	as -ENOMEM or silently ignore in cases like disk readahead.
1242
1243  memory.reclaim
1244	A write-only nested-keyed file which exists for all cgroups.
1245
1246	This is a simple interface to trigger memory reclaim in the
1247	target cgroup.
1248
1249	This file accepts a single key, the number of bytes to reclaim.
1250	No nested keys are currently supported.
1251
1252	Example::
1253
1254	  echo "1G" > memory.reclaim
1255
1256	The interface can be later extended with nested keys to
1257	configure the reclaim behavior. For example, specify the
1258	type of memory to reclaim from (anon, file, ..).
1259
1260	Please note that the kernel can over or under reclaim from
1261	the target cgroup. If less bytes are reclaimed than the
1262	specified amount, -EAGAIN is returned.
1263
1264	Please note that the proactive reclaim (triggered by this
1265	interface) is not meant to indicate memory pressure on the
1266	memory cgroup. Therefore socket memory balancing triggered by
1267	the memory reclaim normally is not exercised in this case.
1268	This means that the networking layer will not adapt based on
1269	reclaim induced by memory.reclaim.
1270
1271  memory.peak
1272	A read-only single value file which exists on non-root
1273	cgroups.
1274
1275	The max memory usage recorded for the cgroup and its
1276	descendants since the creation of the cgroup.
1277
1278  memory.oom.group
1279	A read-write single value file which exists on non-root
1280	cgroups.  The default value is "0".
1281
1282	Determines whether the cgroup should be treated as
1283	an indivisible workload by the OOM killer. If set,
1284	all tasks belonging to the cgroup or to its descendants
1285	(if the memory cgroup is not a leaf cgroup) are killed
1286	together or not at all. This can be used to avoid
1287	partial kills to guarantee workload integrity.
1288
1289	Tasks with the OOM protection (oom_score_adj set to -1000)
1290	are treated as an exception and are never killed.
1291
1292	If the OOM killer is invoked in a cgroup, it's not going
1293	to kill any tasks outside of this cgroup, regardless
1294	memory.oom.group values of ancestor cgroups.
1295
1296  memory.events
1297	A read-only flat-keyed file which exists on non-root cgroups.
1298	The following entries are defined.  Unless specified
1299	otherwise, a value change in this file generates a file
1300	modified event.
1301
1302	Note that all fields in this file are hierarchical and the
1303	file modified event can be generated due to an event down the
1304	hierarchy. For the local events at the cgroup level see
1305	memory.events.local.
1306
1307	  low
1308		The number of times the cgroup is reclaimed due to
1309		high memory pressure even though its usage is under
1310		the low boundary.  This usually indicates that the low
1311		boundary is over-committed.
1312
1313	  high
1314		The number of times processes of the cgroup are
1315		throttled and routed to perform direct memory reclaim
1316		because the high memory boundary was exceeded.  For a
1317		cgroup whose memory usage is capped by the high limit
1318		rather than global memory pressure, this event's
1319		occurrences are expected.
1320
1321	  max
1322		The number of times the cgroup's memory usage was
1323		about to go over the max boundary.  If direct reclaim
1324		fails to bring it down, the cgroup goes to OOM state.
1325
1326	  oom
1327		The number of time the cgroup's memory usage was
1328		reached the limit and allocation was about to fail.
1329
1330		This event is not raised if the OOM killer is not
1331		considered as an option, e.g. for failed high-order
1332		allocations or if caller asked to not retry attempts.
1333
1334	  oom_kill
1335		The number of processes belonging to this cgroup
1336		killed by any kind of OOM killer.
1337
1338          oom_group_kill
1339                The number of times a group OOM has occurred.
1340
1341  memory.events.local
1342	Similar to memory.events but the fields in the file are local
1343	to the cgroup i.e. not hierarchical. The file modified event
1344	generated on this file reflects only the local events.
1345
1346  memory.stat
1347	A read-only flat-keyed file which exists on non-root cgroups.
1348
1349	This breaks down the cgroup's memory footprint into different
1350	types of memory, type-specific details, and other information
1351	on the state and past events of the memory management system.
1352
1353	All memory amounts are in bytes.
1354
1355	The entries are ordered to be human readable, and new entries
1356	can show up in the middle. Don't rely on items remaining in a
1357	fixed position; use the keys to look up specific values!
1358
1359	If the entry has no per-node counter (or not show in the
1360	memory.numa_stat). We use 'npn' (non-per-node) as the tag
1361	to indicate that it will not show in the memory.numa_stat.
1362
1363	  anon
1364		Amount of memory used in anonymous mappings such as
1365		brk(), sbrk(), and mmap(MAP_ANONYMOUS)
1366
1367	  file
1368		Amount of memory used to cache filesystem data,
1369		including tmpfs and shared memory.
1370
1371	  kernel (npn)
1372		Amount of total kernel memory, including
1373		(kernel_stack, pagetables, percpu, vmalloc, slab) in
1374		addition to other kernel memory use cases.
1375
1376	  kernel_stack
1377		Amount of memory allocated to kernel stacks.
1378
1379	  pagetables
1380                Amount of memory allocated for page tables.
1381
1382	  sec_pagetables
1383		Amount of memory allocated for secondary page tables,
1384		this currently includes KVM mmu allocations on x86
1385		and arm64.
1386
1387	  percpu (npn)
1388		Amount of memory used for storing per-cpu kernel
1389		data structures.
1390
1391	  sock (npn)
1392		Amount of memory used in network transmission buffers
1393
1394	  vmalloc (npn)
1395		Amount of memory used for vmap backed memory.
1396
1397	  shmem
1398		Amount of cached filesystem data that is swap-backed,
1399		such as tmpfs, shm segments, shared anonymous mmap()s
1400
1401	  zswap
1402		Amount of memory consumed by the zswap compression backend.
1403
1404	  zswapped
1405		Amount of application memory swapped out to zswap.
1406
1407	  file_mapped
1408		Amount of cached filesystem data mapped with mmap()
1409
1410	  file_dirty
1411		Amount of cached filesystem data that was modified but
1412		not yet written back to disk
1413
1414	  file_writeback
1415		Amount of cached filesystem data that was modified and
1416		is currently being written back to disk
1417
1418	  swapcached
1419		Amount of swap cached in memory. The swapcache is accounted
1420		against both memory and swap usage.
1421
1422	  anon_thp
1423		Amount of memory used in anonymous mappings backed by
1424		transparent hugepages
1425
1426	  file_thp
1427		Amount of cached filesystem data backed by transparent
1428		hugepages
1429
1430	  shmem_thp
1431		Amount of shm, tmpfs, shared anonymous mmap()s backed by
1432		transparent hugepages
1433
1434	  inactive_anon, active_anon, inactive_file, active_file, unevictable
1435		Amount of memory, swap-backed and filesystem-backed,
1436		on the internal memory management lists used by the
1437		page reclaim algorithm.
1438
1439		As these represent internal list state (eg. shmem pages are on anon
1440		memory management lists), inactive_foo + active_foo may not be equal to
1441		the value for the foo counter, since the foo counter is type-based, not
1442		list-based.
1443
1444	  slab_reclaimable
1445		Part of "slab" that might be reclaimed, such as
1446		dentries and inodes.
1447
1448	  slab_unreclaimable
1449		Part of "slab" that cannot be reclaimed on memory
1450		pressure.
1451
1452	  slab (npn)
1453		Amount of memory used for storing in-kernel data
1454		structures.
1455
1456	  workingset_refault_anon
1457		Number of refaults of previously evicted anonymous pages.
1458
1459	  workingset_refault_file
1460		Number of refaults of previously evicted file pages.
1461
1462	  workingset_activate_anon
1463		Number of refaulted anonymous pages that were immediately
1464		activated.
1465
1466	  workingset_activate_file
1467		Number of refaulted file pages that were immediately activated.
1468
1469	  workingset_restore_anon
1470		Number of restored anonymous pages which have been detected as
1471		an active workingset before they got reclaimed.
1472
1473	  workingset_restore_file
1474		Number of restored file pages which have been detected as an
1475		active workingset before they got reclaimed.
1476
1477	  workingset_nodereclaim
1478		Number of times a shadow node has been reclaimed
1479
1480	  pgscan (npn)
1481		Amount of scanned pages (in an inactive LRU list)
1482
1483	  pgsteal (npn)
1484		Amount of reclaimed pages
1485
1486	  pgscan_kswapd (npn)
1487		Amount of scanned pages by kswapd (in an inactive LRU list)
1488
1489	  pgscan_direct (npn)
1490		Amount of scanned pages directly  (in an inactive LRU list)
1491
1492	  pgscan_khugepaged (npn)
1493		Amount of scanned pages by khugepaged  (in an inactive LRU list)
1494
1495	  pgsteal_kswapd (npn)
1496		Amount of reclaimed pages by kswapd
1497
1498	  pgsteal_direct (npn)
1499		Amount of reclaimed pages directly
1500
1501	  pgsteal_khugepaged (npn)
1502		Amount of reclaimed pages by khugepaged
1503
1504	  pgfault (npn)
1505		Total number of page faults incurred
1506
1507	  pgmajfault (npn)
1508		Number of major page faults incurred
1509
1510	  pgrefill (npn)
1511		Amount of scanned pages (in an active LRU list)
1512
1513	  pgactivate (npn)
1514		Amount of pages moved to the active LRU list
1515
1516	  pgdeactivate (npn)
1517		Amount of pages moved to the inactive LRU list
1518
1519	  pglazyfree (npn)
1520		Amount of pages postponed to be freed under memory pressure
1521
1522	  pglazyfreed (npn)
1523		Amount of reclaimed lazyfree pages
1524
1525	  thp_fault_alloc (npn)
1526		Number of transparent hugepages which were allocated to satisfy
1527		a page fault. This counter is not present when CONFIG_TRANSPARENT_HUGEPAGE
1528                is not set.
1529
1530	  thp_collapse_alloc (npn)
1531		Number of transparent hugepages which were allocated to allow
1532		collapsing an existing range of pages. This counter is not
1533		present when CONFIG_TRANSPARENT_HUGEPAGE is not set.
1534
1535  memory.numa_stat
1536	A read-only nested-keyed file which exists on non-root cgroups.
1537
1538	This breaks down the cgroup's memory footprint into different
1539	types of memory, type-specific details, and other information
1540	per node on the state of the memory management system.
1541
1542	This is useful for providing visibility into the NUMA locality
1543	information within an memcg since the pages are allowed to be
1544	allocated from any physical node. One of the use case is evaluating
1545	application performance by combining this information with the
1546	application's CPU allocation.
1547
1548	All memory amounts are in bytes.
1549
1550	The output format of memory.numa_stat is::
1551
1552	  type N0=<bytes in node 0> N1=<bytes in node 1> ...
1553
1554	The entries are ordered to be human readable, and new entries
1555	can show up in the middle. Don't rely on items remaining in a
1556	fixed position; use the keys to look up specific values!
1557
1558	The entries can refer to the memory.stat.
1559
1560  memory.swap.current
1561	A read-only single value file which exists on non-root
1562	cgroups.
1563
1564	The total amount of swap currently being used by the cgroup
1565	and its descendants.
1566
1567  memory.swap.high
1568	A read-write single value file which exists on non-root
1569	cgroups.  The default is "max".
1570
1571	Swap usage throttle limit.  If a cgroup's swap usage exceeds
1572	this limit, all its further allocations will be throttled to
1573	allow userspace to implement custom out-of-memory procedures.
1574
1575	This limit marks a point of no return for the cgroup. It is NOT
1576	designed to manage the amount of swapping a workload does
1577	during regular operation. Compare to memory.swap.max, which
1578	prohibits swapping past a set amount, but lets the cgroup
1579	continue unimpeded as long as other memory can be reclaimed.
1580
1581	Healthy workloads are not expected to reach this limit.
1582
1583  memory.swap.peak
1584	A read-only single value file which exists on non-root
1585	cgroups.
1586
1587	The max swap usage recorded for the cgroup and its
1588	descendants since the creation of the cgroup.
1589
1590  memory.swap.max
1591	A read-write single value file which exists on non-root
1592	cgroups.  The default is "max".
1593
1594	Swap usage hard limit.  If a cgroup's swap usage reaches this
1595	limit, anonymous memory of the cgroup will not be swapped out.
1596
1597  memory.swap.events
1598	A read-only flat-keyed file which exists on non-root cgroups.
1599	The following entries are defined.  Unless specified
1600	otherwise, a value change in this file generates a file
1601	modified event.
1602
1603	  high
1604		The number of times the cgroup's swap usage was over
1605		the high threshold.
1606
1607	  max
1608		The number of times the cgroup's swap usage was about
1609		to go over the max boundary and swap allocation
1610		failed.
1611
1612	  fail
1613		The number of times swap allocation failed either
1614		because of running out of swap system-wide or max
1615		limit.
1616
1617	When reduced under the current usage, the existing swap
1618	entries are reclaimed gradually and the swap usage may stay
1619	higher than the limit for an extended period of time.  This
1620	reduces the impact on the workload and memory management.
1621
1622  memory.zswap.current
1623	A read-only single value file which exists on non-root
1624	cgroups.
1625
1626	The total amount of memory consumed by the zswap compression
1627	backend.
1628
1629  memory.zswap.max
1630	A read-write single value file which exists on non-root
1631	cgroups.  The default is "max".
1632
1633	Zswap usage hard limit. If a cgroup's zswap pool reaches this
1634	limit, it will refuse to take any more stores before existing
1635	entries fault back in or are written out to disk.
1636
1637  memory.pressure
1638	A read-only nested-keyed file.
1639
1640	Shows pressure stall information for memory. See
1641	:ref:`Documentation/accounting/psi.rst <psi>` for details.
1642
1643
1644Usage Guidelines
1645~~~~~~~~~~~~~~~~
1646
1647"memory.high" is the main mechanism to control memory usage.
1648Over-committing on high limit (sum of high limits > available memory)
1649and letting global memory pressure to distribute memory according to
1650usage is a viable strategy.
1651
1652Because breach of the high limit doesn't trigger the OOM killer but
1653throttles the offending cgroup, a management agent has ample
1654opportunities to monitor and take appropriate actions such as granting
1655more memory or terminating the workload.
1656
1657Determining whether a cgroup has enough memory is not trivial as
1658memory usage doesn't indicate whether the workload can benefit from
1659more memory.  For example, a workload which writes data received from
1660network to a file can use all available memory but can also operate as
1661performant with a small amount of memory.  A measure of memory
1662pressure - how much the workload is being impacted due to lack of
1663memory - is necessary to determine whether a workload needs more
1664memory; unfortunately, memory pressure monitoring mechanism isn't
1665implemented yet.
1666
1667
1668Memory Ownership
1669~~~~~~~~~~~~~~~~
1670
1671A memory area is charged to the cgroup which instantiated it and stays
1672charged to the cgroup until the area is released.  Migrating a process
1673to a different cgroup doesn't move the memory usages that it
1674instantiated while in the previous cgroup to the new cgroup.
1675
1676A memory area may be used by processes belonging to different cgroups.
1677To which cgroup the area will be charged is in-deterministic; however,
1678over time, the memory area is likely to end up in a cgroup which has
1679enough memory allowance to avoid high reclaim pressure.
1680
1681If a cgroup sweeps a considerable amount of memory which is expected
1682to be accessed repeatedly by other cgroups, it may make sense to use
1683POSIX_FADV_DONTNEED to relinquish the ownership of memory areas
1684belonging to the affected files to ensure correct memory ownership.
1685
1686
1687IO
1688--
1689
1690The "io" controller regulates the distribution of IO resources.  This
1691controller implements both weight based and absolute bandwidth or IOPS
1692limit distribution; however, weight based distribution is available
1693only if cfq-iosched is in use and neither scheme is available for
1694blk-mq devices.
1695
1696
1697IO Interface Files
1698~~~~~~~~~~~~~~~~~~
1699
1700  io.stat
1701	A read-only nested-keyed file.
1702
1703	Lines are keyed by $MAJ:$MIN device numbers and not ordered.
1704	The following nested keys are defined.
1705
1706	  ======	=====================
1707	  rbytes	Bytes read
1708	  wbytes	Bytes written
1709	  rios		Number of read IOs
1710	  wios		Number of write IOs
1711	  dbytes	Bytes discarded
1712	  dios		Number of discard IOs
1713	  ======	=====================
1714
1715	An example read output follows::
1716
1717	  8:16 rbytes=1459200 wbytes=314773504 rios=192 wios=353 dbytes=0 dios=0
1718	  8:0 rbytes=90430464 wbytes=299008000 rios=8950 wios=1252 dbytes=50331648 dios=3021
1719
1720  io.cost.qos
1721	A read-write nested-keyed file which exists only on the root
1722	cgroup.
1723
1724	This file configures the Quality of Service of the IO cost
1725	model based controller (CONFIG_BLK_CGROUP_IOCOST) which
1726	currently implements "io.weight" proportional control.  Lines
1727	are keyed by $MAJ:$MIN device numbers and not ordered.  The
1728	line for a given device is populated on the first write for
1729	the device on "io.cost.qos" or "io.cost.model".  The following
1730	nested keys are defined.
1731
1732	  ======	=====================================
1733	  enable	Weight-based control enable
1734	  ctrl		"auto" or "user"
1735	  rpct		Read latency percentile    [0, 100]
1736	  rlat		Read latency threshold
1737	  wpct		Write latency percentile   [0, 100]
1738	  wlat		Write latency threshold
1739	  min		Minimum scaling percentage [1, 10000]
1740	  max		Maximum scaling percentage [1, 10000]
1741	  ======	=====================================
1742
1743	The controller is disabled by default and can be enabled by
1744	setting "enable" to 1.  "rpct" and "wpct" parameters default
1745	to zero and the controller uses internal device saturation
1746	state to adjust the overall IO rate between "min" and "max".
1747
1748	When a better control quality is needed, latency QoS
1749	parameters can be configured.  For example::
1750
1751	  8:16 enable=1 ctrl=auto rpct=95.00 rlat=75000 wpct=95.00 wlat=150000 min=50.00 max=150.0
1752
1753	shows that on sdb, the controller is enabled, will consider
1754	the device saturated if the 95th percentile of read completion
1755	latencies is above 75ms or write 150ms, and adjust the overall
1756	IO issue rate between 50% and 150% accordingly.
1757
1758	The lower the saturation point, the better the latency QoS at
1759	the cost of aggregate bandwidth.  The narrower the allowed
1760	adjustment range between "min" and "max", the more conformant
1761	to the cost model the IO behavior.  Note that the IO issue
1762	base rate may be far off from 100% and setting "min" and "max"
1763	blindly can lead to a significant loss of device capacity or
1764	control quality.  "min" and "max" are useful for regulating
1765	devices which show wide temporary behavior changes - e.g. a
1766	ssd which accepts writes at the line speed for a while and
1767	then completely stalls for multiple seconds.
1768
1769	When "ctrl" is "auto", the parameters are controlled by the
1770	kernel and may change automatically.  Setting "ctrl" to "user"
1771	or setting any of the percentile and latency parameters puts
1772	it into "user" mode and disables the automatic changes.  The
1773	automatic mode can be restored by setting "ctrl" to "auto".
1774
1775  io.cost.model
1776	A read-write nested-keyed file which exists only on the root
1777	cgroup.
1778
1779	This file configures the cost model of the IO cost model based
1780	controller (CONFIG_BLK_CGROUP_IOCOST) which currently
1781	implements "io.weight" proportional control.  Lines are keyed
1782	by $MAJ:$MIN device numbers and not ordered.  The line for a
1783	given device is populated on the first write for the device on
1784	"io.cost.qos" or "io.cost.model".  The following nested keys
1785	are defined.
1786
1787	  =====		================================
1788	  ctrl		"auto" or "user"
1789	  model		The cost model in use - "linear"
1790	  =====		================================
1791
1792	When "ctrl" is "auto", the kernel may change all parameters
1793	dynamically.  When "ctrl" is set to "user" or any other
1794	parameters are written to, "ctrl" become "user" and the
1795	automatic changes are disabled.
1796
1797	When "model" is "linear", the following model parameters are
1798	defined.
1799
1800	  =============	========================================
1801	  [r|w]bps	The maximum sequential IO throughput
1802	  [r|w]seqiops	The maximum 4k sequential IOs per second
1803	  [r|w]randiops	The maximum 4k random IOs per second
1804	  =============	========================================
1805
1806	From the above, the builtin linear model determines the base
1807	costs of a sequential and random IO and the cost coefficient
1808	for the IO size.  While simple, this model can cover most
1809	common device classes acceptably.
1810
1811	The IO cost model isn't expected to be accurate in absolute
1812	sense and is scaled to the device behavior dynamically.
1813
1814	If needed, tools/cgroup/iocost_coef_gen.py can be used to
1815	generate device-specific coefficients.
1816
1817  io.weight
1818	A read-write flat-keyed file which exists on non-root cgroups.
1819	The default is "default 100".
1820
1821	The first line is the default weight applied to devices
1822	without specific override.  The rest are overrides keyed by
1823	$MAJ:$MIN device numbers and not ordered.  The weights are in
1824	the range [1, 10000] and specifies the relative amount IO time
1825	the cgroup can use in relation to its siblings.
1826
1827	The default weight can be updated by writing either "default
1828	$WEIGHT" or simply "$WEIGHT".  Overrides can be set by writing
1829	"$MAJ:$MIN $WEIGHT" and unset by writing "$MAJ:$MIN default".
1830
1831	An example read output follows::
1832
1833	  default 100
1834	  8:16 200
1835	  8:0 50
1836
1837  io.max
1838	A read-write nested-keyed file which exists on non-root
1839	cgroups.
1840
1841	BPS and IOPS based IO limit.  Lines are keyed by $MAJ:$MIN
1842	device numbers and not ordered.  The following nested keys are
1843	defined.
1844
1845	  =====		==================================
1846	  rbps		Max read bytes per second
1847	  wbps		Max write bytes per second
1848	  riops		Max read IO operations per second
1849	  wiops		Max write IO operations per second
1850	  =====		==================================
1851
1852	When writing, any number of nested key-value pairs can be
1853	specified in any order.  "max" can be specified as the value
1854	to remove a specific limit.  If the same key is specified
1855	multiple times, the outcome is undefined.
1856
1857	BPS and IOPS are measured in each IO direction and IOs are
1858	delayed if limit is reached.  Temporary bursts are allowed.
1859
1860	Setting read limit at 2M BPS and write at 120 IOPS for 8:16::
1861
1862	  echo "8:16 rbps=2097152 wiops=120" > io.max
1863
1864	Reading returns the following::
1865
1866	  8:16 rbps=2097152 wbps=max riops=max wiops=120
1867
1868	Write IOPS limit can be removed by writing the following::
1869
1870	  echo "8:16 wiops=max" > io.max
1871
1872	Reading now returns the following::
1873
1874	  8:16 rbps=2097152 wbps=max riops=max wiops=max
1875
1876  io.pressure
1877	A read-only nested-keyed file.
1878
1879	Shows pressure stall information for IO. See
1880	:ref:`Documentation/accounting/psi.rst <psi>` for details.
1881
1882
1883Writeback
1884~~~~~~~~~
1885
1886Page cache is dirtied through buffered writes and shared mmaps and
1887written asynchronously to the backing filesystem by the writeback
1888mechanism.  Writeback sits between the memory and IO domains and
1889regulates the proportion of dirty memory by balancing dirtying and
1890write IOs.
1891
1892The io controller, in conjunction with the memory controller,
1893implements control of page cache writeback IOs.  The memory controller
1894defines the memory domain that dirty memory ratio is calculated and
1895maintained for and the io controller defines the io domain which
1896writes out dirty pages for the memory domain.  Both system-wide and
1897per-cgroup dirty memory states are examined and the more restrictive
1898of the two is enforced.
1899
1900cgroup writeback requires explicit support from the underlying
1901filesystem.  Currently, cgroup writeback is implemented on ext2, ext4,
1902btrfs, f2fs, and xfs.  On other filesystems, all writeback IOs are
1903attributed to the root cgroup.
1904
1905There are inherent differences in memory and writeback management
1906which affects how cgroup ownership is tracked.  Memory is tracked per
1907page while writeback per inode.  For the purpose of writeback, an
1908inode is assigned to a cgroup and all IO requests to write dirty pages
1909from the inode are attributed to that cgroup.
1910
1911As cgroup ownership for memory is tracked per page, there can be pages
1912which are associated with different cgroups than the one the inode is
1913associated with.  These are called foreign pages.  The writeback
1914constantly keeps track of foreign pages and, if a particular foreign
1915cgroup becomes the majority over a certain period of time, switches
1916the ownership of the inode to that cgroup.
1917
1918While this model is enough for most use cases where a given inode is
1919mostly dirtied by a single cgroup even when the main writing cgroup
1920changes over time, use cases where multiple cgroups write to a single
1921inode simultaneously are not supported well.  In such circumstances, a
1922significant portion of IOs are likely to be attributed incorrectly.
1923As memory controller assigns page ownership on the first use and
1924doesn't update it until the page is released, even if writeback
1925strictly follows page ownership, multiple cgroups dirtying overlapping
1926areas wouldn't work as expected.  It's recommended to avoid such usage
1927patterns.
1928
1929The sysctl knobs which affect writeback behavior are applied to cgroup
1930writeback as follows.
1931
1932  vm.dirty_background_ratio, vm.dirty_ratio
1933	These ratios apply the same to cgroup writeback with the
1934	amount of available memory capped by limits imposed by the
1935	memory controller and system-wide clean memory.
1936
1937  vm.dirty_background_bytes, vm.dirty_bytes
1938	For cgroup writeback, this is calculated into ratio against
1939	total available memory and applied the same way as
1940	vm.dirty[_background]_ratio.
1941
1942
1943IO Latency
1944~~~~~~~~~~
1945
1946This is a cgroup v2 controller for IO workload protection.  You provide a group
1947with a latency target, and if the average latency exceeds that target the
1948controller will throttle any peers that have a lower latency target than the
1949protected workload.
1950
1951The limits are only applied at the peer level in the hierarchy.  This means that
1952in the diagram below, only groups A, B, and C will influence each other, and
1953groups D and F will influence each other.  Group G will influence nobody::
1954
1955			[root]
1956		/	   |		\
1957		A	   B		C
1958	       /  \        |
1959	      D    F	   G
1960
1961
1962So the ideal way to configure this is to set io.latency in groups A, B, and C.
1963Generally you do not want to set a value lower than the latency your device
1964supports.  Experiment to find the value that works best for your workload.
1965Start at higher than the expected latency for your device and watch the
1966avg_lat value in io.stat for your workload group to get an idea of the
1967latency you see during normal operation.  Use the avg_lat value as a basis for
1968your real setting, setting at 10-15% higher than the value in io.stat.
1969
1970How IO Latency Throttling Works
1971~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1972
1973io.latency is work conserving; so as long as everybody is meeting their latency
1974target the controller doesn't do anything.  Once a group starts missing its
1975target it begins throttling any peer group that has a higher target than itself.
1976This throttling takes 2 forms:
1977
1978- Queue depth throttling.  This is the number of outstanding IO's a group is
1979  allowed to have.  We will clamp down relatively quickly, starting at no limit
1980  and going all the way down to 1 IO at a time.
1981
1982- Artificial delay induction.  There are certain types of IO that cannot be
1983  throttled without possibly adversely affecting higher priority groups.  This
1984  includes swapping and metadata IO.  These types of IO are allowed to occur
1985  normally, however they are "charged" to the originating group.  If the
1986  originating group is being throttled you will see the use_delay and delay
1987  fields in io.stat increase.  The delay value is how many microseconds that are
1988  being added to any process that runs in this group.  Because this number can
1989  grow quite large if there is a lot of swapping or metadata IO occurring we
1990  limit the individual delay events to 1 second at a time.
1991
1992Once the victimized group starts meeting its latency target again it will start
1993unthrottling any peer groups that were throttled previously.  If the victimized
1994group simply stops doing IO the global counter will unthrottle appropriately.
1995
1996IO Latency Interface Files
1997~~~~~~~~~~~~~~~~~~~~~~~~~~
1998
1999  io.latency
2000	This takes a similar format as the other controllers.
2001
2002		"MAJOR:MINOR target=<target time in microseconds>"
2003
2004  io.stat
2005	If the controller is enabled you will see extra stats in io.stat in
2006	addition to the normal ones.
2007
2008	  depth
2009		This is the current queue depth for the group.
2010
2011	  avg_lat
2012		This is an exponential moving average with a decay rate of 1/exp
2013		bound by the sampling interval.  The decay rate interval can be
2014		calculated by multiplying the win value in io.stat by the
2015		corresponding number of samples based on the win value.
2016
2017	  win
2018		The sampling window size in milliseconds.  This is the minimum
2019		duration of time between evaluation events.  Windows only elapse
2020		with IO activity.  Idle periods extend the most recent window.
2021
2022IO Priority
2023~~~~~~~~~~~
2024
2025A single attribute controls the behavior of the I/O priority cgroup policy,
2026namely the blkio.prio.class attribute. The following values are accepted for
2027that attribute:
2028
2029  no-change
2030	Do not modify the I/O priority class.
2031
2032  promote-to-rt
2033	For requests that have a non-RT I/O priority class, change it into RT.
2034	Also change the priority level of these requests to 4. Do not modify
2035	the I/O priority of requests that have priority class RT.
2036
2037  restrict-to-be
2038	For requests that do not have an I/O priority class or that have I/O
2039	priority class RT, change it into BE. Also change the priority level
2040	of these requests to 0. Do not modify the I/O priority class of
2041	requests that have priority class IDLE.
2042
2043  idle
2044	Change the I/O priority class of all requests into IDLE, the lowest
2045	I/O priority class.
2046
2047  none-to-rt
2048	Deprecated. Just an alias for promote-to-rt.
2049
2050The following numerical values are associated with the I/O priority policies:
2051
2052+----------------+---+
2053| no-change      | 0 |
2054+----------------+---+
2055| rt-to-be       | 2 |
2056+----------------+---+
2057| all-to-idle    | 3 |
2058+----------------+---+
2059
2060The numerical value that corresponds to each I/O priority class is as follows:
2061
2062+-------------------------------+---+
2063| IOPRIO_CLASS_NONE             | 0 |
2064+-------------------------------+---+
2065| IOPRIO_CLASS_RT (real-time)   | 1 |
2066+-------------------------------+---+
2067| IOPRIO_CLASS_BE (best effort) | 2 |
2068+-------------------------------+---+
2069| IOPRIO_CLASS_IDLE             | 3 |
2070+-------------------------------+---+
2071
2072The algorithm to set the I/O priority class for a request is as follows:
2073
2074- If I/O priority class policy is promote-to-rt, change the request I/O
2075  priority class to IOPRIO_CLASS_RT and change the request I/O priority
2076  level to 4.
2077- If I/O priorityt class is not promote-to-rt, translate the I/O priority
2078  class policy into a number, then change the request I/O priority class
2079  into the maximum of the I/O priority class policy number and the numerical
2080  I/O priority class.
2081
2082PID
2083---
2084
2085The process number controller is used to allow a cgroup to stop any
2086new tasks from being fork()'d or clone()'d after a specified limit is
2087reached.
2088
2089The number of tasks in a cgroup can be exhausted in ways which other
2090controllers cannot prevent, thus warranting its own controller.  For
2091example, a fork bomb is likely to exhaust the number of tasks before
2092hitting memory restrictions.
2093
2094Note that PIDs used in this controller refer to TIDs, process IDs as
2095used by the kernel.
2096
2097
2098PID Interface Files
2099~~~~~~~~~~~~~~~~~~~
2100
2101  pids.max
2102	A read-write single value file which exists on non-root
2103	cgroups.  The default is "max".
2104
2105	Hard limit of number of processes.
2106
2107  pids.current
2108	A read-only single value file which exists on all cgroups.
2109
2110	The number of processes currently in the cgroup and its
2111	descendants.
2112
2113Organisational operations are not blocked by cgroup policies, so it is
2114possible to have pids.current > pids.max.  This can be done by either
2115setting the limit to be smaller than pids.current, or attaching enough
2116processes to the cgroup such that pids.current is larger than
2117pids.max.  However, it is not possible to violate a cgroup PID policy
2118through fork() or clone(). These will return -EAGAIN if the creation
2119of a new process would cause a cgroup policy to be violated.
2120
2121
2122Cpuset
2123------
2124
2125The "cpuset" controller provides a mechanism for constraining
2126the CPU and memory node placement of tasks to only the resources
2127specified in the cpuset interface files in a task's current cgroup.
2128This is especially valuable on large NUMA systems where placing jobs
2129on properly sized subsets of the systems with careful processor and
2130memory placement to reduce cross-node memory access and contention
2131can improve overall system performance.
2132
2133The "cpuset" controller is hierarchical.  That means the controller
2134cannot use CPUs or memory nodes not allowed in its parent.
2135
2136
2137Cpuset Interface Files
2138~~~~~~~~~~~~~~~~~~~~~~
2139
2140  cpuset.cpus
2141	A read-write multiple values file which exists on non-root
2142	cpuset-enabled cgroups.
2143
2144	It lists the requested CPUs to be used by tasks within this
2145	cgroup.  The actual list of CPUs to be granted, however, is
2146	subjected to constraints imposed by its parent and can differ
2147	from the requested CPUs.
2148
2149	The CPU numbers are comma-separated numbers or ranges.
2150	For example::
2151
2152	  # cat cpuset.cpus
2153	  0-4,6,8-10
2154
2155	An empty value indicates that the cgroup is using the same
2156	setting as the nearest cgroup ancestor with a non-empty
2157	"cpuset.cpus" or all the available CPUs if none is found.
2158
2159	The value of "cpuset.cpus" stays constant until the next update
2160	and won't be affected by any CPU hotplug events.
2161
2162  cpuset.cpus.effective
2163	A read-only multiple values file which exists on all
2164	cpuset-enabled cgroups.
2165
2166	It lists the onlined CPUs that are actually granted to this
2167	cgroup by its parent.  These CPUs are allowed to be used by
2168	tasks within the current cgroup.
2169
2170	If "cpuset.cpus" is empty, the "cpuset.cpus.effective" file shows
2171	all the CPUs from the parent cgroup that can be available to
2172	be used by this cgroup.  Otherwise, it should be a subset of
2173	"cpuset.cpus" unless none of the CPUs listed in "cpuset.cpus"
2174	can be granted.  In this case, it will be treated just like an
2175	empty "cpuset.cpus".
2176
2177	Its value will be affected by CPU hotplug events.
2178
2179  cpuset.mems
2180	A read-write multiple values file which exists on non-root
2181	cpuset-enabled cgroups.
2182
2183	It lists the requested memory nodes to be used by tasks within
2184	this cgroup.  The actual list of memory nodes granted, however,
2185	is subjected to constraints imposed by its parent and can differ
2186	from the requested memory nodes.
2187
2188	The memory node numbers are comma-separated numbers or ranges.
2189	For example::
2190
2191	  # cat cpuset.mems
2192	  0-1,3
2193
2194	An empty value indicates that the cgroup is using the same
2195	setting as the nearest cgroup ancestor with a non-empty
2196	"cpuset.mems" or all the available memory nodes if none
2197	is found.
2198
2199	The value of "cpuset.mems" stays constant until the next update
2200	and won't be affected by any memory nodes hotplug events.
2201
2202	Setting a non-empty value to "cpuset.mems" causes memory of
2203	tasks within the cgroup to be migrated to the designated nodes if
2204	they are currently using memory outside of the designated nodes.
2205
2206	There is a cost for this memory migration.  The migration
2207	may not be complete and some memory pages may be left behind.
2208	So it is recommended that "cpuset.mems" should be set properly
2209	before spawning new tasks into the cpuset.  Even if there is
2210	a need to change "cpuset.mems" with active tasks, it shouldn't
2211	be done frequently.
2212
2213  cpuset.mems.effective
2214	A read-only multiple values file which exists on all
2215	cpuset-enabled cgroups.
2216
2217	It lists the onlined memory nodes that are actually granted to
2218	this cgroup by its parent. These memory nodes are allowed to
2219	be used by tasks within the current cgroup.
2220
2221	If "cpuset.mems" is empty, it shows all the memory nodes from the
2222	parent cgroup that will be available to be used by this cgroup.
2223	Otherwise, it should be a subset of "cpuset.mems" unless none of
2224	the memory nodes listed in "cpuset.mems" can be granted.  In this
2225	case, it will be treated just like an empty "cpuset.mems".
2226
2227	Its value will be affected by memory nodes hotplug events.
2228
2229  cpuset.cpus.partition
2230	A read-write single value file which exists on non-root
2231	cpuset-enabled cgroups.  This flag is owned by the parent cgroup
2232	and is not delegatable.
2233
2234	It accepts only the following input values when written to.
2235
2236	  ==========	=====================================
2237	  "member"	Non-root member of a partition
2238	  "root"	Partition root
2239	  "isolated"	Partition root without load balancing
2240	  ==========	=====================================
2241
2242	The root cgroup is always a partition root and its state
2243	cannot be changed.  All other non-root cgroups start out as
2244	"member".
2245
2246	When set to "root", the current cgroup is the root of a new
2247	partition or scheduling domain that comprises itself and all
2248	its descendants except those that are separate partition roots
2249	themselves and their descendants.
2250
2251	When set to "isolated", the CPUs in that partition root will
2252	be in an isolated state without any load balancing from the
2253	scheduler.  Tasks placed in such a partition with multiple
2254	CPUs should be carefully distributed and bound to each of the
2255	individual CPUs for optimal performance.
2256
2257	The value shown in "cpuset.cpus.effective" of a partition root
2258	is the CPUs that the partition root can dedicate to a potential
2259	new child partition root. The new child subtracts available
2260	CPUs from its parent "cpuset.cpus.effective".
2261
2262	A partition root ("root" or "isolated") can be in one of the
2263	two possible states - valid or invalid.  An invalid partition
2264	root is in a degraded state where some state information may
2265	be retained, but behaves more like a "member".
2266
2267	All possible state transitions among "member", "root" and
2268	"isolated" are allowed.
2269
2270	On read, the "cpuset.cpus.partition" file can show the following
2271	values.
2272
2273	  =============================	=====================================
2274	  "member"			Non-root member of a partition
2275	  "root"			Partition root
2276	  "isolated"			Partition root without load balancing
2277	  "root invalid (<reason>)"	Invalid partition root
2278	  "isolated invalid (<reason>)"	Invalid isolated partition root
2279	  =============================	=====================================
2280
2281	In the case of an invalid partition root, a descriptive string on
2282	why the partition is invalid is included within parentheses.
2283
2284	For a partition root to become valid, the following conditions
2285	must be met.
2286
2287	1) The "cpuset.cpus" is exclusive with its siblings , i.e. they
2288	   are not shared by any of its siblings (exclusivity rule).
2289	2) The parent cgroup is a valid partition root.
2290	3) The "cpuset.cpus" is not empty and must contain at least
2291	   one of the CPUs from parent's "cpuset.cpus", i.e. they overlap.
2292	4) The "cpuset.cpus.effective" cannot be empty unless there is
2293	   no task associated with this partition.
2294
2295	External events like hotplug or changes to "cpuset.cpus" can
2296	cause a valid partition root to become invalid and vice versa.
2297	Note that a task cannot be moved to a cgroup with empty
2298	"cpuset.cpus.effective".
2299
2300	For a valid partition root with the sibling cpu exclusivity
2301	rule enabled, changes made to "cpuset.cpus" that violate the
2302	exclusivity rule will invalidate the partition as well as its
2303	sibling partitions with conflicting cpuset.cpus values. So
2304	care must be taking in changing "cpuset.cpus".
2305
2306	A valid non-root parent partition may distribute out all its CPUs
2307	to its child partitions when there is no task associated with it.
2308
2309	Care must be taken to change a valid partition root to
2310	"member" as all its child partitions, if present, will become
2311	invalid causing disruption to tasks running in those child
2312	partitions. These inactivated partitions could be recovered if
2313	their parent is switched back to a partition root with a proper
2314	set of "cpuset.cpus".
2315
2316	Poll and inotify events are triggered whenever the state of
2317	"cpuset.cpus.partition" changes.  That includes changes caused
2318	by write to "cpuset.cpus.partition", cpu hotplug or other
2319	changes that modify the validity status of the partition.
2320	This will allow user space agents to monitor unexpected changes
2321	to "cpuset.cpus.partition" without the need to do continuous
2322	polling.
2323
2324
2325Device controller
2326-----------------
2327
2328Device controller manages access to device files. It includes both
2329creation of new device files (using mknod), and access to the
2330existing device files.
2331
2332Cgroup v2 device controller has no interface files and is implemented
2333on top of cgroup BPF. To control access to device files, a user may
2334create bpf programs of type BPF_PROG_TYPE_CGROUP_DEVICE and attach
2335them to cgroups with BPF_CGROUP_DEVICE flag. On an attempt to access a
2336device file, corresponding BPF programs will be executed, and depending
2337on the return value the attempt will succeed or fail with -EPERM.
2338
2339A BPF_PROG_TYPE_CGROUP_DEVICE program takes a pointer to the
2340bpf_cgroup_dev_ctx structure, which describes the device access attempt:
2341access type (mknod/read/write) and device (type, major and minor numbers).
2342If the program returns 0, the attempt fails with -EPERM, otherwise it
2343succeeds.
2344
2345An example of BPF_PROG_TYPE_CGROUP_DEVICE program may be found in
2346tools/testing/selftests/bpf/progs/dev_cgroup.c in the kernel source tree.
2347
2348
2349RDMA
2350----
2351
2352The "rdma" controller regulates the distribution and accounting of
2353RDMA resources.
2354
2355RDMA Interface Files
2356~~~~~~~~~~~~~~~~~~~~
2357
2358  rdma.max
2359	A readwrite nested-keyed file that exists for all the cgroups
2360	except root that describes current configured resource limit
2361	for a RDMA/IB device.
2362
2363	Lines are keyed by device name and are not ordered.
2364	Each line contains space separated resource name and its configured
2365	limit that can be distributed.
2366
2367	The following nested keys are defined.
2368
2369	  ==========	=============================
2370	  hca_handle	Maximum number of HCA Handles
2371	  hca_object 	Maximum number of HCA Objects
2372	  ==========	=============================
2373
2374	An example for mlx4 and ocrdma device follows::
2375
2376	  mlx4_0 hca_handle=2 hca_object=2000
2377	  ocrdma1 hca_handle=3 hca_object=max
2378
2379  rdma.current
2380	A read-only file that describes current resource usage.
2381	It exists for all the cgroup except root.
2382
2383	An example for mlx4 and ocrdma device follows::
2384
2385	  mlx4_0 hca_handle=1 hca_object=20
2386	  ocrdma1 hca_handle=1 hca_object=23
2387
2388HugeTLB
2389-------
2390
2391The HugeTLB controller allows to limit the HugeTLB usage per control group and
2392enforces the controller limit during page fault.
2393
2394HugeTLB Interface Files
2395~~~~~~~~~~~~~~~~~~~~~~~
2396
2397  hugetlb.<hugepagesize>.current
2398	Show current usage for "hugepagesize" hugetlb.  It exists for all
2399	the cgroup except root.
2400
2401  hugetlb.<hugepagesize>.max
2402	Set/show the hard limit of "hugepagesize" hugetlb usage.
2403	The default value is "max".  It exists for all the cgroup except root.
2404
2405  hugetlb.<hugepagesize>.events
2406	A read-only flat-keyed file which exists on non-root cgroups.
2407
2408	  max
2409		The number of allocation failure due to HugeTLB limit
2410
2411  hugetlb.<hugepagesize>.events.local
2412	Similar to hugetlb.<hugepagesize>.events but the fields in the file
2413	are local to the cgroup i.e. not hierarchical. The file modified event
2414	generated on this file reflects only the local events.
2415
2416  hugetlb.<hugepagesize>.numa_stat
2417	Similar to memory.numa_stat, it shows the numa information of the
2418        hugetlb pages of <hugepagesize> in this cgroup.  Only active in
2419        use hugetlb pages are included.  The per-node values are in bytes.
2420
2421Misc
2422----
2423
2424The Miscellaneous cgroup provides the resource limiting and tracking
2425mechanism for the scalar resources which cannot be abstracted like the other
2426cgroup resources. Controller is enabled by the CONFIG_CGROUP_MISC config
2427option.
2428
2429A resource can be added to the controller via enum misc_res_type{} in the
2430include/linux/misc_cgroup.h file and the corresponding name via misc_res_name[]
2431in the kernel/cgroup/misc.c file. Provider of the resource must set its
2432capacity prior to using the resource by calling misc_cg_set_capacity().
2433
2434Once a capacity is set then the resource usage can be updated using charge and
2435uncharge APIs. All of the APIs to interact with misc controller are in
2436include/linux/misc_cgroup.h.
2437
2438Misc Interface Files
2439~~~~~~~~~~~~~~~~~~~~
2440
2441Miscellaneous controller provides 3 interface files. If two misc resources (res_a and res_b) are registered then:
2442
2443  misc.capacity
2444        A read-only flat-keyed file shown only in the root cgroup.  It shows
2445        miscellaneous scalar resources available on the platform along with
2446        their quantities::
2447
2448	  $ cat misc.capacity
2449	  res_a 50
2450	  res_b 10
2451
2452  misc.current
2453        A read-only flat-keyed file shown in the all cgroups.  It shows
2454        the current usage of the resources in the cgroup and its children.::
2455
2456	  $ cat misc.current
2457	  res_a 3
2458	  res_b 0
2459
2460  misc.max
2461        A read-write flat-keyed file shown in the non root cgroups. Allowed
2462        maximum usage of the resources in the cgroup and its children.::
2463
2464	  $ cat misc.max
2465	  res_a max
2466	  res_b 4
2467
2468	Limit can be set by::
2469
2470	  # echo res_a 1 > misc.max
2471
2472	Limit can be set to max by::
2473
2474	  # echo res_a max > misc.max
2475
2476        Limits can be set higher than the capacity value in the misc.capacity
2477        file.
2478
2479  misc.events
2480	A read-only flat-keyed file which exists on non-root cgroups. The
2481	following entries are defined. Unless specified otherwise, a value
2482	change in this file generates a file modified event. All fields in
2483	this file are hierarchical.
2484
2485	  max
2486		The number of times the cgroup's resource usage was
2487		about to go over the max boundary.
2488
2489Migration and Ownership
2490~~~~~~~~~~~~~~~~~~~~~~~
2491
2492A miscellaneous scalar resource is charged to the cgroup in which it is used
2493first, and stays charged to that cgroup until that resource is freed. Migrating
2494a process to a different cgroup does not move the charge to the destination
2495cgroup where the process has moved.
2496
2497Others
2498------
2499
2500perf_event
2501~~~~~~~~~~
2502
2503perf_event controller, if not mounted on a legacy hierarchy, is
2504automatically enabled on the v2 hierarchy so that perf events can
2505always be filtered by cgroup v2 path.  The controller can still be
2506moved to a legacy hierarchy after v2 hierarchy is populated.
2507
2508
2509Non-normative information
2510-------------------------
2511
2512This section contains information that isn't considered to be a part of
2513the stable kernel API and so is subject to change.
2514
2515
2516CPU controller root cgroup process behaviour
2517~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2518
2519When distributing CPU cycles in the root cgroup each thread in this
2520cgroup is treated as if it was hosted in a separate child cgroup of the
2521root cgroup. This child cgroup weight is dependent on its thread nice
2522level.
2523
2524For details of this mapping see sched_prio_to_weight array in
2525kernel/sched/core.c file (values from this array should be scaled
2526appropriately so the neutral - nice 0 - value is 100 instead of 1024).
2527
2528
2529IO controller root cgroup process behaviour
2530~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2531
2532Root cgroup processes are hosted in an implicit leaf child node.
2533When distributing IO resources this implicit child node is taken into
2534account as if it was a normal child cgroup of the root cgroup with a
2535weight value of 200.
2536
2537
2538Namespace
2539=========
2540
2541Basics
2542------
2543
2544cgroup namespace provides a mechanism to virtualize the view of the
2545"/proc/$PID/cgroup" file and cgroup mounts.  The CLONE_NEWCGROUP clone
2546flag can be used with clone(2) and unshare(2) to create a new cgroup
2547namespace.  The process running inside the cgroup namespace will have
2548its "/proc/$PID/cgroup" output restricted to cgroupns root.  The
2549cgroupns root is the cgroup of the process at the time of creation of
2550the cgroup namespace.
2551
2552Without cgroup namespace, the "/proc/$PID/cgroup" file shows the
2553complete path of the cgroup of a process.  In a container setup where
2554a set of cgroups and namespaces are intended to isolate processes the
2555"/proc/$PID/cgroup" file may leak potential system level information
2556to the isolated processes.  For example::
2557
2558  # cat /proc/self/cgroup
2559  0::/batchjobs/container_id1
2560
2561The path '/batchjobs/container_id1' can be considered as system-data
2562and undesirable to expose to the isolated processes.  cgroup namespace
2563can be used to restrict visibility of this path.  For example, before
2564creating a cgroup namespace, one would see::
2565
2566  # ls -l /proc/self/ns/cgroup
2567  lrwxrwxrwx 1 root root 0 2014-07-15 10:37 /proc/self/ns/cgroup -> cgroup:[4026531835]
2568  # cat /proc/self/cgroup
2569  0::/batchjobs/container_id1
2570
2571After unsharing a new namespace, the view changes::
2572
2573  # ls -l /proc/self/ns/cgroup
2574  lrwxrwxrwx 1 root root 0 2014-07-15 10:35 /proc/self/ns/cgroup -> cgroup:[4026532183]
2575  # cat /proc/self/cgroup
2576  0::/
2577
2578When some thread from a multi-threaded process unshares its cgroup
2579namespace, the new cgroupns gets applied to the entire process (all
2580the threads).  This is natural for the v2 hierarchy; however, for the
2581legacy hierarchies, this may be unexpected.
2582
2583A cgroup namespace is alive as long as there are processes inside or
2584mounts pinning it.  When the last usage goes away, the cgroup
2585namespace is destroyed.  The cgroupns root and the actual cgroups
2586remain.
2587
2588
2589The Root and Views
2590------------------
2591
2592The 'cgroupns root' for a cgroup namespace is the cgroup in which the
2593process calling unshare(2) is running.  For example, if a process in
2594/batchjobs/container_id1 cgroup calls unshare, cgroup
2595/batchjobs/container_id1 becomes the cgroupns root.  For the
2596init_cgroup_ns, this is the real root ('/') cgroup.
2597
2598The cgroupns root cgroup does not change even if the namespace creator
2599process later moves to a different cgroup::
2600
2601  # ~/unshare -c # unshare cgroupns in some cgroup
2602  # cat /proc/self/cgroup
2603  0::/
2604  # mkdir sub_cgrp_1
2605  # echo 0 > sub_cgrp_1/cgroup.procs
2606  # cat /proc/self/cgroup
2607  0::/sub_cgrp_1
2608
2609Each process gets its namespace-specific view of "/proc/$PID/cgroup"
2610
2611Processes running inside the cgroup namespace will be able to see
2612cgroup paths (in /proc/self/cgroup) only inside their root cgroup.
2613From within an unshared cgroupns::
2614
2615  # sleep 100000 &
2616  [1] 7353
2617  # echo 7353 > sub_cgrp_1/cgroup.procs
2618  # cat /proc/7353/cgroup
2619  0::/sub_cgrp_1
2620
2621From the initial cgroup namespace, the real cgroup path will be
2622visible::
2623
2624  $ cat /proc/7353/cgroup
2625  0::/batchjobs/container_id1/sub_cgrp_1
2626
2627From a sibling cgroup namespace (that is, a namespace rooted at a
2628different cgroup), the cgroup path relative to its own cgroup
2629namespace root will be shown.  For instance, if PID 7353's cgroup
2630namespace root is at '/batchjobs/container_id2', then it will see::
2631
2632  # cat /proc/7353/cgroup
2633  0::/../container_id2/sub_cgrp_1
2634
2635Note that the relative path always starts with '/' to indicate that
2636its relative to the cgroup namespace root of the caller.
2637
2638
2639Migration and setns(2)
2640----------------------
2641
2642Processes inside a cgroup namespace can move into and out of the
2643namespace root if they have proper access to external cgroups.  For
2644example, from inside a namespace with cgroupns root at
2645/batchjobs/container_id1, and assuming that the global hierarchy is
2646still accessible inside cgroupns::
2647
2648  # cat /proc/7353/cgroup
2649  0::/sub_cgrp_1
2650  # echo 7353 > batchjobs/container_id2/cgroup.procs
2651  # cat /proc/7353/cgroup
2652  0::/../container_id2
2653
2654Note that this kind of setup is not encouraged.  A task inside cgroup
2655namespace should only be exposed to its own cgroupns hierarchy.
2656
2657setns(2) to another cgroup namespace is allowed when:
2658
2659(a) the process has CAP_SYS_ADMIN against its current user namespace
2660(b) the process has CAP_SYS_ADMIN against the target cgroup
2661    namespace's userns
2662
2663No implicit cgroup changes happen with attaching to another cgroup
2664namespace.  It is expected that the someone moves the attaching
2665process under the target cgroup namespace root.
2666
2667
2668Interaction with Other Namespaces
2669---------------------------------
2670
2671Namespace specific cgroup hierarchy can be mounted by a process
2672running inside a non-init cgroup namespace::
2673
2674  # mount -t cgroup2 none $MOUNT_POINT
2675
2676This will mount the unified cgroup hierarchy with cgroupns root as the
2677filesystem root.  The process needs CAP_SYS_ADMIN against its user and
2678mount namespaces.
2679
2680The virtualization of /proc/self/cgroup file combined with restricting
2681the view of cgroup hierarchy by namespace-private cgroupfs mount
2682provides a properly isolated cgroup view inside the container.
2683
2684
2685Information on Kernel Programming
2686=================================
2687
2688This section contains kernel programming information in the areas
2689where interacting with cgroup is necessary.  cgroup core and
2690controllers are not covered.
2691
2692
2693Filesystem Support for Writeback
2694--------------------------------
2695
2696A filesystem can support cgroup writeback by updating
2697address_space_operations->writepage[s]() to annotate bio's using the
2698following two functions.
2699
2700  wbc_init_bio(@wbc, @bio)
2701	Should be called for each bio carrying writeback data and
2702	associates the bio with the inode's owner cgroup and the
2703	corresponding request queue.  This must be called after
2704	a queue (device) has been associated with the bio and
2705	before submission.
2706
2707  wbc_account_cgroup_owner(@wbc, @page, @bytes)
2708	Should be called for each data segment being written out.
2709	While this function doesn't care exactly when it's called
2710	during the writeback session, it's the easiest and most
2711	natural to call it as data segments are added to a bio.
2712
2713With writeback bio's annotated, cgroup support can be enabled per
2714super_block by setting SB_I_CGROUPWB in ->s_iflags.  This allows for
2715selective disabling of cgroup writeback support which is helpful when
2716certain filesystem features, e.g. journaled data mode, are
2717incompatible.
2718
2719wbc_init_bio() binds the specified bio to its cgroup.  Depending on
2720the configuration, the bio may be executed at a lower priority and if
2721the writeback session is holding shared resources, e.g. a journal
2722entry, may lead to priority inversion.  There is no one easy solution
2723for the problem.  Filesystems can try to work around specific problem
2724cases by skipping wbc_init_bio() and using bio_associate_blkg()
2725directly.
2726
2727
2728Deprecated v1 Core Features
2729===========================
2730
2731- Multiple hierarchies including named ones are not supported.
2732
2733- All v1 mount options are not supported.
2734
2735- The "tasks" file is removed and "cgroup.procs" is not sorted.
2736
2737- "cgroup.clone_children" is removed.
2738
2739- /proc/cgroups is meaningless for v2.  Use "cgroup.controllers" file
2740  at the root instead.
2741
2742
2743Issues with v1 and Rationales for v2
2744====================================
2745
2746Multiple Hierarchies
2747--------------------
2748
2749cgroup v1 allowed an arbitrary number of hierarchies and each
2750hierarchy could host any number of controllers.  While this seemed to
2751provide a high level of flexibility, it wasn't useful in practice.
2752
2753For example, as there is only one instance of each controller, utility
2754type controllers such as freezer which can be useful in all
2755hierarchies could only be used in one.  The issue is exacerbated by
2756the fact that controllers couldn't be moved to another hierarchy once
2757hierarchies were populated.  Another issue was that all controllers
2758bound to a hierarchy were forced to have exactly the same view of the
2759hierarchy.  It wasn't possible to vary the granularity depending on
2760the specific controller.
2761
2762In practice, these issues heavily limited which controllers could be
2763put on the same hierarchy and most configurations resorted to putting
2764each controller on its own hierarchy.  Only closely related ones, such
2765as the cpu and cpuacct controllers, made sense to be put on the same
2766hierarchy.  This often meant that userland ended up managing multiple
2767similar hierarchies repeating the same steps on each hierarchy
2768whenever a hierarchy management operation was necessary.
2769
2770Furthermore, support for multiple hierarchies came at a steep cost.
2771It greatly complicated cgroup core implementation but more importantly
2772the support for multiple hierarchies restricted how cgroup could be
2773used in general and what controllers was able to do.
2774
2775There was no limit on how many hierarchies there might be, which meant
2776that a thread's cgroup membership couldn't be described in finite
2777length.  The key might contain any number of entries and was unlimited
2778in length, which made it highly awkward to manipulate and led to
2779addition of controllers which existed only to identify membership,
2780which in turn exacerbated the original problem of proliferating number
2781of hierarchies.
2782
2783Also, as a controller couldn't have any expectation regarding the
2784topologies of hierarchies other controllers might be on, each
2785controller had to assume that all other controllers were attached to
2786completely orthogonal hierarchies.  This made it impossible, or at
2787least very cumbersome, for controllers to cooperate with each other.
2788
2789In most use cases, putting controllers on hierarchies which are
2790completely orthogonal to each other isn't necessary.  What usually is
2791called for is the ability to have differing levels of granularity
2792depending on the specific controller.  In other words, hierarchy may
2793be collapsed from leaf towards root when viewed from specific
2794controllers.  For example, a given configuration might not care about
2795how memory is distributed beyond a certain level while still wanting
2796to control how CPU cycles are distributed.
2797
2798
2799Thread Granularity
2800------------------
2801
2802cgroup v1 allowed threads of a process to belong to different cgroups.
2803This didn't make sense for some controllers and those controllers
2804ended up implementing different ways to ignore such situations but
2805much more importantly it blurred the line between API exposed to
2806individual applications and system management interface.
2807
2808Generally, in-process knowledge is available only to the process
2809itself; thus, unlike service-level organization of processes,
2810categorizing threads of a process requires active participation from
2811the application which owns the target process.
2812
2813cgroup v1 had an ambiguously defined delegation model which got abused
2814in combination with thread granularity.  cgroups were delegated to
2815individual applications so that they can create and manage their own
2816sub-hierarchies and control resource distributions along them.  This
2817effectively raised cgroup to the status of a syscall-like API exposed
2818to lay programs.
2819
2820First of all, cgroup has a fundamentally inadequate interface to be
2821exposed this way.  For a process to access its own knobs, it has to
2822extract the path on the target hierarchy from /proc/self/cgroup,
2823construct the path by appending the name of the knob to the path, open
2824and then read and/or write to it.  This is not only extremely clunky
2825and unusual but also inherently racy.  There is no conventional way to
2826define transaction across the required steps and nothing can guarantee
2827that the process would actually be operating on its own sub-hierarchy.
2828
2829cgroup controllers implemented a number of knobs which would never be
2830accepted as public APIs because they were just adding control knobs to
2831system-management pseudo filesystem.  cgroup ended up with interface
2832knobs which were not properly abstracted or refined and directly
2833revealed kernel internal details.  These knobs got exposed to
2834individual applications through the ill-defined delegation mechanism
2835effectively abusing cgroup as a shortcut to implementing public APIs
2836without going through the required scrutiny.
2837
2838This was painful for both userland and kernel.  Userland ended up with
2839misbehaving and poorly abstracted interfaces and kernel exposing and
2840locked into constructs inadvertently.
2841
2842
2843Competition Between Inner Nodes and Threads
2844-------------------------------------------
2845
2846cgroup v1 allowed threads to be in any cgroups which created an
2847interesting problem where threads belonging to a parent cgroup and its
2848children cgroups competed for resources.  This was nasty as two
2849different types of entities competed and there was no obvious way to
2850settle it.  Different controllers did different things.
2851
2852The cpu controller considered threads and cgroups as equivalents and
2853mapped nice levels to cgroup weights.  This worked for some cases but
2854fell flat when children wanted to be allocated specific ratios of CPU
2855cycles and the number of internal threads fluctuated - the ratios
2856constantly changed as the number of competing entities fluctuated.
2857There also were other issues.  The mapping from nice level to weight
2858wasn't obvious or universal, and there were various other knobs which
2859simply weren't available for threads.
2860
2861The io controller implicitly created a hidden leaf node for each
2862cgroup to host the threads.  The hidden leaf had its own copies of all
2863the knobs with ``leaf_`` prefixed.  While this allowed equivalent
2864control over internal threads, it was with serious drawbacks.  It
2865always added an extra layer of nesting which wouldn't be necessary
2866otherwise, made the interface messy and significantly complicated the
2867implementation.
2868
2869The memory controller didn't have a way to control what happened
2870between internal tasks and child cgroups and the behavior was not
2871clearly defined.  There were attempts to add ad-hoc behaviors and
2872knobs to tailor the behavior to specific workloads which would have
2873led to problems extremely difficult to resolve in the long term.
2874
2875Multiple controllers struggled with internal tasks and came up with
2876different ways to deal with it; unfortunately, all the approaches were
2877severely flawed and, furthermore, the widely different behaviors
2878made cgroup as a whole highly inconsistent.
2879
2880This clearly is a problem which needs to be addressed from cgroup core
2881in a uniform way.
2882
2883
2884Other Interface Issues
2885----------------------
2886
2887cgroup v1 grew without oversight and developed a large number of
2888idiosyncrasies and inconsistencies.  One issue on the cgroup core side
2889was how an empty cgroup was notified - a userland helper binary was
2890forked and executed for each event.  The event delivery wasn't
2891recursive or delegatable.  The limitations of the mechanism also led
2892to in-kernel event delivery filtering mechanism further complicating
2893the interface.
2894
2895Controller interfaces were problematic too.  An extreme example is
2896controllers completely ignoring hierarchical organization and treating
2897all cgroups as if they were all located directly under the root
2898cgroup.  Some controllers exposed a large amount of inconsistent
2899implementation details to userland.
2900
2901There also was no consistency across controllers.  When a new cgroup
2902was created, some controllers defaulted to not imposing extra
2903restrictions while others disallowed any resource usage until
2904explicitly configured.  Configuration knobs for the same type of
2905control used widely differing naming schemes and formats.  Statistics
2906and information knobs were named arbitrarily and used different
2907formats and units even in the same controller.
2908
2909cgroup v2 establishes common conventions where appropriate and updates
2910controllers so that they expose minimal and consistent interfaces.
2911
2912
2913Controller Issues and Remedies
2914------------------------------
2915
2916Memory
2917~~~~~~
2918
2919The original lower boundary, the soft limit, is defined as a limit
2920that is per default unset.  As a result, the set of cgroups that
2921global reclaim prefers is opt-in, rather than opt-out.  The costs for
2922optimizing these mostly negative lookups are so high that the
2923implementation, despite its enormous size, does not even provide the
2924basic desirable behavior.  First off, the soft limit has no
2925hierarchical meaning.  All configured groups are organized in a global
2926rbtree and treated like equal peers, regardless where they are located
2927in the hierarchy.  This makes subtree delegation impossible.  Second,
2928the soft limit reclaim pass is so aggressive that it not just
2929introduces high allocation latencies into the system, but also impacts
2930system performance due to overreclaim, to the point where the feature
2931becomes self-defeating.
2932
2933The memory.low boundary on the other hand is a top-down allocated
2934reserve.  A cgroup enjoys reclaim protection when it's within its
2935effective low, which makes delegation of subtrees possible. It also
2936enjoys having reclaim pressure proportional to its overage when
2937above its effective low.
2938
2939The original high boundary, the hard limit, is defined as a strict
2940limit that can not budge, even if the OOM killer has to be called.
2941But this generally goes against the goal of making the most out of the
2942available memory.  The memory consumption of workloads varies during
2943runtime, and that requires users to overcommit.  But doing that with a
2944strict upper limit requires either a fairly accurate prediction of the
2945working set size or adding slack to the limit.  Since working set size
2946estimation is hard and error prone, and getting it wrong results in
2947OOM kills, most users tend to err on the side of a looser limit and
2948end up wasting precious resources.
2949
2950The memory.high boundary on the other hand can be set much more
2951conservatively.  When hit, it throttles allocations by forcing them
2952into direct reclaim to work off the excess, but it never invokes the
2953OOM killer.  As a result, a high boundary that is chosen too
2954aggressively will not terminate the processes, but instead it will
2955lead to gradual performance degradation.  The user can monitor this
2956and make corrections until the minimal memory footprint that still
2957gives acceptable performance is found.
2958
2959In extreme cases, with many concurrent allocations and a complete
2960breakdown of reclaim progress within the group, the high boundary can
2961be exceeded.  But even then it's mostly better to satisfy the
2962allocation from the slack available in other groups or the rest of the
2963system than killing the group.  Otherwise, memory.max is there to
2964limit this type of spillover and ultimately contain buggy or even
2965malicious applications.
2966
2967Setting the original memory.limit_in_bytes below the current usage was
2968subject to a race condition, where concurrent charges could cause the
2969limit setting to fail. memory.max on the other hand will first set the
2970limit to prevent new charges, and then reclaim and OOM kill until the
2971new limit is met - or the task writing to memory.max is killed.
2972
2973The combined memory+swap accounting and limiting is replaced by real
2974control over swap space.
2975
2976The main argument for a combined memory+swap facility in the original
2977cgroup design was that global or parental pressure would always be
2978able to swap all anonymous memory of a child group, regardless of the
2979child's own (possibly untrusted) configuration.  However, untrusted
2980groups can sabotage swapping by other means - such as referencing its
2981anonymous memory in a tight loop - and an admin can not assume full
2982swappability when overcommitting untrusted jobs.
2983
2984For trusted jobs, on the other hand, a combined counter is not an
2985intuitive userspace interface, and it flies in the face of the idea
2986that cgroup controllers should account and limit specific physical
2987resources.  Swap space is a resource like all others in the system,
2988and that's why unified hierarchy allows distributing it separately.
2989