Lines Matching +full:sub +full:- +full:controllers
1 .. _cgroup-v2:
11 conventions of cgroup v2. It describes all userland-visible aspects
14 v1 is available under :ref:`Documentation/admin-guide/cgroup-v1/index.rst <cgroup-v1>`.
19 1-1. Terminology
20 1-2. What is cgroup?
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
38 3-1. Weights
39 3-2. Limits
40 3-3. Protections
41 3-4. Allocations
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
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
83 P-1. Filesystem Support for Writeback
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
98 -----------
102 qualifier as in "cgroup controllers". When explicitly referring to
107 ---------------
113 cgroup is largely composed of two parts - the core and controllers.
117 although there are utility controllers which serve purposes other than
127 Following certain structural constraints, controllers may be enabled or
129 hierarchical - if a controller is enabled on a cgroup, it affects all
131 sub-hierarchy of the cgroup. When a controller is enabled on a nested
141 --------
146 # mount -t cgroup2 none $MOUNT_POINT
149 controllers which support v2 and are not bound to a v1 hierarchy are
151 Controllers which are not in active use in the v2 hierarchy can be
156 is no longer referenced in its current hierarchy. Because per-cgroup
157 controller states are destroyed asynchronously and controllers may
163 to inter-controller dependencies, other controllers may need to be
167 controllers dynamically between the v2 and other hierarchies is
170 controllers after system boot.
173 automount the v1 cgroup filesystem and so hijack all controllers
176 disabling controllers in v1 and make them always available in v2.
184 ignored on non-init namespace mounts. Please refer to the
192 controllers, and then seeding it with CLONE_INTO_CGROUP is
201 option is ignored on non-init namespace mounts.
209 behavior but is a mount-option to avoid regressing setups
223 controller. The pre-allocated pool does not belong to anyone.
243 The option restores v1-like behavior of pids.events:max, that is only
251 --------------------------------
257 A child cgroup can be created by creating a sub-directory::
262 structure. Each cgroup has a read-writable interface file
264 belong to the cgroup one-per-line. The PIDs are not ordered and the
295 0::/test-cgroup/test-cgroup-nested
302 0::/test-cgroup/test-cgroup-nested (deleted)
308 cgroup v2 supports thread granularity for a subset of controllers to
316 Controllers which support thread mode are called threaded controllers.
317 The ones which don't are called domain controllers.
328 constraint - threaded controllers can be enabled on non-leaf cgroups
352 - As the cgroup will join the parent's resource domain. The parent
355 - When the parent is an unthreaded domain, it must not have any domain
356 controllers enabled or populated domain children. The root is
359 Topology-wise, a cgroup can be in an invalid state. Please consider
362 A (threaded domain) - B (threaded) - C (domain, just created)
371 cgroup becomes threaded or threaded controllers are enabled in the
377 threads in the cgroup. Except that the operations are per-thread
378 instead of per-process, "cgroup.threads" has the same format and
392 Only threaded controllers can be enabled in a threaded subtree. When
400 between threads in a non-leaf cgroup and its child cgroups. Each
403 Currently, the following controllers are threaded and can be enabled
406 - cpu
407 - cpuset
408 - perf_event
409 - pids
412 --------------------------
414 Each non-root cgroup has a "cgroup.events" file which contains
415 "populated" field indicating whether the cgroup's sub-hierarchy has
419 example, to start a clean-up operation after all processes of a given
420 sub-hierarchy have exited. The populated state updates and
421 notifications are recursive. Consider the following sub-hierarchy
425 A(4) - B(0) - C(1)
434 Controlling Controllers
435 -----------------------
440 Each cgroup has a "cgroup.controllers" file which lists all
441 controllers available for the cgroup to enable::
443 # cat cgroup.controllers
446 No controller is enabled by default. Controllers can be enabled and
449 # echo "+cpu +memory -io" > cgroup.subtree_control
451 Only controllers which are listed in "cgroup.controllers" can be
458 Consider the following sub-hierarchy. The enabled controllers are
461 A(cpu,memory) - B(memory) - C()
475 controller interface files - anything which doesn't start with
479 Top-down Constraint
482 Resources are distributed top-down and a cgroup can further distribute
484 parent. This means that all non-root "cgroup.subtree_control" files
485 can only contain controllers which are enabled in the parent's
494 Non-root cgroups can distribute domain resources to their children
497 controllers enabled in their "cgroup.subtree_control" files.
507 controllers. How resource consumption in the root cgroup is governed
509 refer to the Non-normative information section in the Controllers
517 children before enabling controllers in its "cgroup.subtree_control"
522 ----------
544 delegated, the user can build sub-hierarchy under the directory,
547 of all resource controllers are hierarchical and regardless of what
548 happens in the delegated sub-hierarchy, nothing can escape the
552 cgroups in or nesting depth of a delegated sub-hierarchy; however,
559 A delegated sub-hierarchy is contained in the sense that processes
560 can't be moved into or out of the sub-hierarchy by the delegatee.
563 requiring the following conditions for a process with a non-root euid
567 - The writer must have write access to the "cgroup.procs" file.
569 - The writer must have write access to the "cgroup.procs" file of the
573 processes around freely in the delegated sub-hierarchy it can't pull
574 in from or push out to outside the sub-hierarchy.
580 ~~~~~~~~~~~~~ - C0 - C00
583 ~~~~~~~~~~~~~ - C1 - C10
590 will be denied with -EACCES.
595 is not reachable, the migration is rejected with -ENOENT.
599 ----------
607 inherent trade-offs between migration and various hot paths in terms
613 resource structure once on start-up. Dynamic adjustments to resource
640 cgroup controllers implement several resource distribution schemes
646 -------
652 work-conserving. Due to the dynamic nature, this model is usually
667 .. _cgroupv2-limits-distributor:
670 ------
673 Limits can be over-committed - the sum of the limits of children can
678 As limits can be over-committed, all configuration combinations are
685 .. _cgroupv2-protections-distributor:
688 -----------
693 soft boundaries. Protections can also be over-committed in which case
700 As protections can be over-committed, all configuration combinations
704 "memory.low" implements best-effort memory protection and is an
709 -----------
712 resource. Allocations can't be over-committed - the sum of the
719 As allocations can't be over-committed, some configuration
724 "cpu.rt.max" hard-allocates realtime slices and is an example of this
732 ------
737 New-line separated values
745 (when read-only or multiple values can be written at once)
762 reading; however, controllers may allow omitting later fields or
766 can be written at a time. For nested keyed files, the sub key pairs
771 -----------
773 - Settings for a single feature should be contained in a single file.
775 - The root cgroup should be exempt from resource control and thus
778 - The default time unit is microseconds. If a different unit is ever
781 - A parts-per quantity should use a percentage decimal with at least
782 two digit fractional part - e.g. 13.40.
784 - If a controller implements weight based resource distribution, its
790 - If a controller implements an absolute resource guarantee and/or
799 - If a setting has a configurable default value and keyed specific
813 # cat cgroup-example-interface-file
819 # echo 125 > cgroup-example-interface-file
823 # echo "default 125" > cgroup-example-interface-file
827 # echo "8:16 170" > cgroup-example-interface-file
831 # echo "8:0 default" > cgroup-example-interface-file
832 # cat cgroup-example-interface-file
836 - For events which are not very high frequency, an interface file
843 --------------------
848 A read-write single value file which exists on non-root
854 - "domain" : A normal valid domain cgroup.
856 - "domain threaded" : A threaded domain cgroup which is
859 - "domain invalid" : A cgroup which is in an invalid state.
860 It can't be populated or have controllers enabled. It may
863 - "threaded" : A threaded cgroup which is a member of a
870 A read-write new-line separated values file which exists on
874 the cgroup one-per-line. The PIDs are not ordered and the
883 - It must have write access to the "cgroup.procs" file.
885 - It must have write access to the "cgroup.procs" file of the
888 When delegating a sub-hierarchy, write access to this file
896 A read-write new-line separated values file which exists on
900 the cgroup one-per-line. The TIDs are not ordered and the
909 - It must have write access to the "cgroup.threads" file.
911 - The cgroup that the thread is currently in must be in the
914 - It must have write access to the "cgroup.procs" file of the
917 When delegating a sub-hierarchy, write access to this file
920 cgroup.controllers
921 A read-only space separated values file which exists on all
924 It shows space separated list of all controllers available to
925 the cgroup. The controllers are not ordered.
928 A read-write space separated values file which exists on all
931 When read, it shows space separated list of the controllers
935 Space separated list of controllers prefixed with '+' or '-'
936 can be written to enable or disable controllers. A controller
937 name prefixed with '+' enables the controller and '-'
943 A read-only flat-keyed file which exists on non-root cgroups.
955 A read-write single value files. The default is "max".
962 A read-write single value files. The default is "max".
969 A read-only flat-keyed file with the following entries:
995 A read-write single value file which exists on non-root cgroups.
1018 create new sub-cgroups.
1021 A write-only single value file which exists in non-root cgroups.
1033 the whole thread-group.
1036 A read-write single value file that allowed values are "0" and "1".
1040 Writing "1" to the file will re-enable the cgroup PSI accounting.
1048 This may cause non-negligible overhead for some workloads when under
1050 be used to disable PSI accounting in the non-leaf cgroups.
1053 A read-write nested-keyed file.
1058 Controllers chapter
1061 .. _cgroup-v2-cpu:
1064 ---
1066 The "cpu" controllers regulates distribution of CPU cycles. This
1095 A read-only flat-keyed file.
1100 - usage_usec
1101 - user_usec
1102 - system_usec
1106 - nr_periods
1107 - nr_throttled
1108 - throttled_usec
1109 - nr_bursts
1110 - burst_usec
1113 A read-write single value file which exists on non-root
1123 A read-write single value file which exists on non-root
1126 The nice value is in the range [-20, 19].
1135 A read-write two value file which exists on non-root cgroups.
1147 A read-write single value file which exists on non-root
1153 A read-write nested-keyed file.
1159 A read-write single value file which exists on non-root cgroups.
1174 A read-write single value file which exists on non-root cgroups.
1185 A read-write single value file which exists on non-root cgroups.
1188 This is the cgroup analog of the per-task SCHED_IDLE sched policy.
1197 ------
1205 While not completely water-tight, all major memory usages by a given
1210 - Userland memory - page cache and anonymous memory.
1212 - Kernel data structures such as dentries and inodes.
1214 - TCP socket buffers.
1227 A read-only single value file which exists on non-root
1234 A read-write single value file which exists on non-root
1260 A read-write single value file which exists on non-root
1263 Best-effort memory protection. If the memory usage of a
1283 A read-write single value file which exists on non-root
1297 A read-write single value file which exists on non-root
1306 In default configuration regular 0-order allocations always
1311 as -ENOMEM or silently ignore in cases like disk readahead.
1314 A write-only nested-keyed file which exists for all cgroups.
1325 specified amount, -EAGAIN is returned.
1346 A read-write single value file which exists on non-root cgroups.
1351 A write of any non-empty string to this file resets it to the
1356 A read-write single value file which exists on non-root
1366 Tasks with the OOM protection (oom_score_adj set to -1000)
1374 A read-only flat-keyed file which exists on non-root cgroups.
1388 boundary is over-committed.
1408 considered as an option, e.g. for failed high-order
1424 A read-only flat-keyed file which exists on non-root cgroups.
1427 types of memory, type-specific details, and other information
1436 If the entry has no per-node counter (or not show in the
1437 memory.numa_stat). We use 'npn' (non-per-node) as the tag
1465 Amount of memory used for storing per-cpu kernel
1475 Amount of cached filesystem data that is swap-backed,
1512 Amount of memory, swap-backed and filesystem-backed,
1518 the value for the foo counter, since the foo counter is type-based, not
1519 list-based.
1530 Amount of memory used for storing in-kernel data
1608 Number of zero-filled pages swapped out with I/O skipped due to the
1659 A read-only nested-keyed file which exists on non-root cgroups.
1662 types of memory, type-specific details, and other information
1684 A read-only single value file which exists on non-root
1691 A read-write single value file which exists on non-root
1696 allow userspace to implement custom out-of-memory procedures.
1707 A read-write single value file which exists on non-root cgroups.
1712 A write of any non-empty string to this file resets it to the
1717 A read-write single value file which exists on non-root
1724 A read-only flat-keyed file which exists on non-root cgroups.
1740 because of running out of swap system-wide or max
1749 A read-only single value file which exists on non-root
1756 A read-write single value file which exists on non-root
1764 A read-write single value file. The default value is "1".
1782 A read-only nested-keyed file.
1792 Over-committing on high limit (sum of high limits > available memory)
1806 pressure - how much the workload is being impacted due to lack of
1807 memory - is necessary to determine whether a workload needs more
1821 To which cgroup the area will be charged is in-deterministic; however,
1832 --
1837 only if cfq-iosched is in use and neither scheme is available for
1838 blk-mq devices.
1845 A read-only nested-keyed file.
1865 A read-write nested-keyed file which exists only on the root
1877 enable Weight-based control enable
1909 devices which show wide temporary behavior changes - e.g. a
1920 A read-write nested-keyed file which exists only on the root
1933 model The cost model in use - "linear"
1959 generate device-specific coefficients.
1962 A read-write flat-keyed file which exists on non-root cgroups.
1982 A read-write nested-keyed file which exists on non-root
1996 When writing, any number of nested key-value pairs can be
2021 A read-only nested-keyed file.
2040 writes out dirty pages for the memory domain. Both system-wide and
2041 per-cgroup dirty memory states are examined and the more restrictive
2079 memory controller and system-wide clean memory.
2112 your real setting, setting at 10-15% higher than the value in io.stat.
2122 - Queue depth throttling. This is the number of outstanding IO's a group is
2126 - Artificial delay induction. There are certain types of IO that cannot be
2144 This takes a similar format as the other controllers.
2173 no-change
2176 promote-to-rt
2177 For requests that have a non-RT I/O priority class, change it into RT.
2181 restrict-to-be
2191 none-to-rt
2192 Deprecated. Just an alias for promote-to-rt.
2196 +----------------+---+
2197 | no-change | 0 |
2198 +----------------+---+
2199 | promote-to-rt | 1 |
2200 +----------------+---+
2201 | restrict-to-be | 2 |
2202 +----------------+---+
2204 +----------------+---+
2208 +-------------------------------+---+
2210 +-------------------------------+---+
2211 | IOPRIO_CLASS_RT (real-time) | 1 |
2212 +-------------------------------+---+
2214 +-------------------------------+---+
2216 +-------------------------------+---+
2220 - If I/O priority class policy is promote-to-rt, change the request I/O
2223 - If I/O priority class policy is not promote-to-rt, translate the I/O priority
2229 ---
2236 controllers cannot prevent, thus warranting its own controller. For
2248 A read-write single value file which exists on non-root
2254 A read-only single value file which exists on non-root cgroups.
2260 A read-only single value file which exists on non-root cgroups.
2266 A read-only flat-keyed file which exists on non-root cgroups. Unless
2284 through fork() or clone(). These will return -EAGAIN if the creation
2289 ------
2296 memory placement to reduce cross-node memory access and contention
2307 A read-write multiple values file which exists on non-root
2308 cpuset-enabled cgroups.
2315 The CPU numbers are comma-separated numbers or ranges.
2319 0-4,6,8-10
2322 setting as the nearest cgroup ancestor with a non-empty
2329 A read-only multiple values file which exists on all
2330 cpuset-enabled cgroups.
2346 A read-write multiple values file which exists on non-root
2347 cpuset-enabled cgroups.
2354 The memory node numbers are comma-separated numbers or ranges.
2358 0-1,3
2361 setting as the nearest cgroup ancestor with a non-empty
2368 Setting a non-empty value to "cpuset.mems" causes memory of
2380 A read-only multiple values file which exists on all
2381 cpuset-enabled cgroups.
2396 A read-write multiple values file which exists on non-root
2397 cpuset-enabled cgroups.
2430 A read-only multiple values file which exists on all non-root
2431 cpuset-enabled cgroups.
2443 A read-only and root cgroup only multiple values file.
2450 A read-write single value file which exists on non-root
2451 cpuset-enabled cgroups. This flag is owned by the parent cgroup
2457 "member" Non-root member of a partition
2462 A cpuset partition is a collection of cpuset-enabled cgroups with
2469 There are two types of partitions - local and remote. A local
2485 be changed. All other non-root cgroups start out as "member".
2498 two possible states - valid or invalid. An invalid partition
2509 "member" Non-root member of a partition
2536 A valid non-root parent partition may distribute out all its CPUs
2555 A user can pre-configure certain CPUs to an isolated state
2562 -----------------
2573 on the return value the attempt will succeed or fail with -EPERM.
2578 If the program returns 0, the attempt fails with -EPERM, otherwise it
2586 ----
2595 A readwrite nested-keyed file that exists for all the cgroups
2616 A read-only file that describes current resource usage.
2625 -------
2642 A read-only flat-keyed file which exists on non-root cgroups.
2655 use hugetlb pages are included. The per-node values are in bytes.
2658 ----
2680 A read-only flat-keyed file shown only in the root cgroup. It shows
2689 A read-only flat-keyed file shown in the all cgroups. It shows
2697 A read-only flat-keyed file shown in all cgroups. It shows the
2706 A read-write flat-keyed file shown in the non root cgroups. Allowed
2725 A read-only flat-keyed file which exists on non-root cgroups. The
2748 ------
2759 Non-normative information
2760 -------------------------
2776 appropriately so the neutral - nice 0 - value is 100 instead of 1024).
2792 ------
2811 The path '/batchjobs/container_id1' can be considered as system-data
2816 # ls -l /proc/self/ns/cgroup
2817 lrwxrwxrwx 1 root root 0 2014-07-15 10:37 /proc/self/ns/cgroup -> cgroup:[4026531835]
2823 # ls -l /proc/self/ns/cgroup
2824 lrwxrwxrwx 1 root root 0 2014-07-15 10:35 /proc/self/ns/cgroup -> cgroup:[4026532183]
2828 When some thread from a multi-threaded process unshares its cgroup
2840 ------------------
2851 # ~/unshare -c # unshare cgroupns in some cgroup
2859 Each process gets its namespace-specific view of "/proc/$PID/cgroup"
2890 ----------------------
2919 ---------------------------------
2922 running inside a non-init cgroup namespace::
2924 # mount -t cgroup2 none $MOUNT_POINT
2931 the view of cgroup hierarchy by namespace-private cgroupfs mount
2940 controllers are not covered.
2944 --------------------------------
2947 address_space_operations->writepage[s]() to annotate bio's using the
2964 super_block by setting SB_I_CGROUPWB in ->s_iflags. This allows for
2981 - Multiple hierarchies including named ones are not supported.
2983 - All v1 mount options are not supported.
2985 - The "tasks" file is removed and "cgroup.procs" is not sorted.
2987 - "cgroup.clone_children" is removed.
2989 - /proc/cgroups is meaningless for v2. Use "cgroup.controllers" or
2997 --------------------
3000 hierarchy could host any number of controllers. While this seemed to
3004 type controllers such as freezer which can be useful in all
3006 the fact that controllers couldn't be moved to another hierarchy once
3007 hierarchies were populated. Another issue was that all controllers
3012 In practice, these issues heavily limited which controllers could be
3015 as the cpu and cpuacct controllers, made sense to be put on the same
3023 used in general and what controllers was able to do.
3029 addition of controllers which existed only to identify membership,
3034 topologies of hierarchies other controllers might be on, each
3035 controller had to assume that all other controllers were attached to
3037 least very cumbersome, for controllers to cooperate with each other.
3039 In most use cases, putting controllers on hierarchies which are
3044 controllers. For example, a given configuration might not care about
3050 ------------------
3053 This didn't make sense for some controllers and those controllers
3058 Generally, in-process knowledge is available only to the process
3059 itself; thus, unlike service-level organization of processes,
3066 sub-hierarchies and control resource distributions along them. This
3067 effectively raised cgroup to the status of a syscall-like API exposed
3077 that the process would actually be operating on its own sub-hierarchy.
3079 cgroup controllers implemented a number of knobs which would never be
3081 system-management pseudo filesystem. cgroup ended up with interface
3084 individual applications through the ill-defined delegation mechanism
3094 -------------------------------------------
3100 settle it. Different controllers did different things.
3105 cycles and the number of internal threads fluctuated - the ratios
3121 clearly defined. There were attempts to add ad-hoc behaviors and
3125 Multiple controllers struggled with internal tasks and came up with
3135 ----------------------
3139 was how an empty cgroup was notified - a userland helper binary was
3142 to in-kernel event delivery filtering mechanism further complicating
3146 controllers completely ignoring hierarchical organization and treating
3148 cgroup. Some controllers exposed a large amount of inconsistent
3151 There also was no consistency across controllers. When a new cgroup
3152 was created, some controllers defaulted to not imposing extra
3160 controllers so that they expose minimal and consistent interfaces.
3164 ------------------------------
3171 global reclaim prefers is opt-in, rather than opt-out. The costs for
3181 becomes self-defeating.
3183 The memory.low boundary on the other hand is a top-down allocated
3221 new limit is met - or the task writing to memory.max is killed.
3230 groups can sabotage swapping by other means - such as referencing its
3231 anonymous memory in a tight loop - and an admin can not assume full
3236 that cgroup controllers should account and limit specific physical