1================ 2Control Group v2 3================ 4 5:Date: October, 2015 6:Author: Tejun Heo <tj@kernel.org> 7 8This is the authoritative documentation on the design, interface and 9conventions of cgroup v2. It describes all userland-visible aspects 10of cgroup including core and specific controller behaviors. All 11future changes must be reflected in this document. Documentation for 12v1 is available under Documentation/cgroup-v1/. 13 14.. CONTENTS 15 16 1. Introduction 17 1-1. Terminology 18 1-2. What is cgroup? 19 2. Basic Operations 20 2-1. Mounting 21 2-2. Organizing Processes and Threads 22 2-2-1. Processes 23 2-2-2. Threads 24 2-3. [Un]populated Notification 25 2-4. Controlling Controllers 26 2-4-1. Enabling and Disabling 27 2-4-2. Top-down Constraint 28 2-4-3. No Internal Process Constraint 29 2-5. Delegation 30 2-5-1. Model of Delegation 31 2-5-2. Delegation Containment 32 2-6. Guidelines 33 2-6-1. Organize Once and Control 34 2-6-2. Avoid Name Collisions 35 3. Resource Distribution Models 36 3-1. Weights 37 3-2. Limits 38 3-3. Protections 39 3-4. Allocations 40 4. Interface Files 41 4-1. Format 42 4-2. Conventions 43 4-3. Core Interface Files 44 5. Controllers 45 5-1. CPU 46 5-1-1. CPU Interface Files 47 5-2. Memory 48 5-2-1. Memory Interface Files 49 5-2-2. Usage Guidelines 50 5-2-3. Memory Ownership 51 5-3. IO 52 5-3-1. IO Interface Files 53 5-3-2. Writeback 54 5-4. PID 55 5-4-1. PID Interface Files 56 5-5. Device 57 5-6. RDMA 58 5-6-1. RDMA Interface Files 59 5-7. Misc 60 5-7-1. perf_event 61 5-N. Non-normative information 62 5-N-1. CPU controller root cgroup process behaviour 63 5-N-2. IO controller root cgroup process behaviour 64 6. Namespace 65 6-1. Basics 66 6-2. The Root and Views 67 6-3. Migration and setns(2) 68 6-4. Interaction with Other Namespaces 69 P. Information on Kernel Programming 70 P-1. Filesystem Support for Writeback 71 D. Deprecated v1 Core Features 72 R. Issues with v1 and Rationales for v2 73 R-1. Multiple Hierarchies 74 R-2. Thread Granularity 75 R-3. Competition Between Inner Nodes and Threads 76 R-4. Other Interface Issues 77 R-5. Controller Issues and Remedies 78 R-5-1. Memory 79 80 81Introduction 82============ 83 84Terminology 85----------- 86 87"cgroup" stands for "control group" and is never capitalized. The 88singular form is used to designate the whole feature and also as a 89qualifier as in "cgroup controllers". When explicitly referring to 90multiple individual control groups, the plural form "cgroups" is used. 91 92 93What is cgroup? 94--------------- 95 96cgroup is a mechanism to organize processes hierarchically and 97distribute system resources along the hierarchy in a controlled and 98configurable manner. 99 100cgroup is largely composed of two parts - the core and controllers. 101cgroup core is primarily responsible for hierarchically organizing 102processes. A cgroup controller is usually responsible for 103distributing a specific type of system resource along the hierarchy 104although there are utility controllers which serve purposes other than 105resource distribution. 106 107cgroups form a tree structure and every process in the system belongs 108to one and only one cgroup. All threads of a process belong to the 109same cgroup. On creation, all processes are put in the cgroup that 110the parent process belongs to at the time. A process can be migrated 111to another cgroup. Migration of a process doesn't affect already 112existing descendant processes. 113 114Following certain structural constraints, controllers may be enabled or 115disabled selectively on a cgroup. All controller behaviors are 116hierarchical - if a controller is enabled on a cgroup, it affects all 117processes which belong to the cgroups consisting the inclusive 118sub-hierarchy of the cgroup. When a controller is enabled on a nested 119cgroup, it always restricts the resource distribution further. The 120restrictions set closer to the root in the hierarchy can not be 121overridden from further away. 122 123 124Basic Operations 125================ 126 127Mounting 128-------- 129 130Unlike v1, cgroup v2 has only single hierarchy. The cgroup v2 131hierarchy can be mounted with the following mount command:: 132 133 # mount -t cgroup2 none $MOUNT_POINT 134 135cgroup2 filesystem has the magic number 0x63677270 ("cgrp"). All 136controllers which support v2 and are not bound to a v1 hierarchy are 137automatically bound to the v2 hierarchy and show up at the root. 138Controllers which are not in active use in the v2 hierarchy can be 139bound to other hierarchies. This allows mixing v2 hierarchy with the 140legacy v1 multiple hierarchies in a fully backward compatible way. 141 142A controller can be moved across hierarchies only after the controller 143is no longer referenced in its current hierarchy. Because per-cgroup 144controller states are destroyed asynchronously and controllers may 145have lingering references, a controller may not show up immediately on 146the v2 hierarchy after the final umount of the previous hierarchy. 147Similarly, a controller should be fully disabled to be moved out of 148the unified hierarchy and it may take some time for the disabled 149controller to become available for other hierarchies; furthermore, due 150to inter-controller dependencies, other controllers may need to be 151disabled too. 152 153While useful for development and manual configurations, moving 154controllers dynamically between the v2 and other hierarchies is 155strongly discouraged for production use. It is recommended to decide 156the hierarchies and controller associations before starting using the 157controllers after system boot. 158 159During transition to v2, system management software might still 160automount the v1 cgroup filesystem and so hijack all controllers 161during boot, before manual intervention is possible. To make testing 162and experimenting easier, the kernel parameter cgroup_no_v1= allows 163disabling controllers in v1 and make them always available in v2. 164 165cgroup v2 currently supports the following mount options. 166 167 nsdelegate 168 169 Consider cgroup namespaces as delegation boundaries. This 170 option is system wide and can only be set on mount or modified 171 through remount from the init namespace. The mount option is 172 ignored on non-init namespace mounts. Please refer to the 173 Delegation section for details. 174 175 176Organizing Processes and Threads 177-------------------------------- 178 179Processes 180~~~~~~~~~ 181 182Initially, only the root cgroup exists to which all processes belong. 183A child cgroup can be created by creating a sub-directory:: 184 185 # mkdir $CGROUP_NAME 186 187A given cgroup may have multiple child cgroups forming a tree 188structure. Each cgroup has a read-writable interface file 189"cgroup.procs". When read, it lists the PIDs of all processes which 190belong to the cgroup one-per-line. The PIDs are not ordered and the 191same PID may show up more than once if the process got moved to 192another cgroup and then back or the PID got recycled while reading. 193 194A process can be migrated into a cgroup by writing its PID to the 195target cgroup's "cgroup.procs" file. Only one process can be migrated 196on a single write(2) call. If a process is composed of multiple 197threads, writing the PID of any thread migrates all threads of the 198process. 199 200When a process forks a child process, the new process is born into the 201cgroup that the forking process belongs to at the time of the 202operation. After exit, a process stays associated with the cgroup 203that it belonged to at the time of exit until it's reaped; however, a 204zombie process does not appear in "cgroup.procs" and thus can't be 205moved to another cgroup. 206 207A cgroup which doesn't have any children or live processes can be 208destroyed by removing the directory. Note that a cgroup which doesn't 209have any children and is associated only with zombie processes is 210considered empty and can be removed:: 211 212 # rmdir $CGROUP_NAME 213 214"/proc/$PID/cgroup" lists a process's cgroup membership. If legacy 215cgroup is in use in the system, this file may contain multiple lines, 216one for each hierarchy. The entry for cgroup v2 is always in the 217format "0::$PATH":: 218 219 # cat /proc/842/cgroup 220 ... 221 0::/test-cgroup/test-cgroup-nested 222 223If the process becomes a zombie and the cgroup it was associated with 224is removed subsequently, " (deleted)" is appended to the path:: 225 226 # cat /proc/842/cgroup 227 ... 228 0::/test-cgroup/test-cgroup-nested (deleted) 229 230 231Threads 232~~~~~~~ 233 234cgroup v2 supports thread granularity for a subset of controllers to 235support use cases requiring hierarchical resource distribution across 236the threads of a group of processes. By default, all threads of a 237process belong to the same cgroup, which also serves as the resource 238domain to host resource consumptions which are not specific to a 239process or thread. The thread mode allows threads to be spread across 240a subtree while still maintaining the common resource domain for them. 241 242Controllers which support thread mode are called threaded controllers. 243The ones which don't are called domain controllers. 244 245Marking a cgroup threaded makes it join the resource domain of its 246parent as a threaded cgroup. The parent may be another threaded 247cgroup whose resource domain is further up in the hierarchy. The root 248of a threaded subtree, that is, the nearest ancestor which is not 249threaded, is called threaded domain or thread root interchangeably and 250serves as the resource domain for the entire subtree. 251 252Inside a threaded subtree, threads of a process can be put in 253different cgroups and are not subject to the no internal process 254constraint - threaded controllers can be enabled on non-leaf cgroups 255whether they have threads in them or not. 256 257As the threaded domain cgroup hosts all the domain resource 258consumptions of the subtree, it is considered to have internal 259resource consumptions whether there are processes in it or not and 260can't have populated child cgroups which aren't threaded. Because the 261root cgroup is not subject to no internal process constraint, it can 262serve both as a threaded domain and a parent to domain cgroups. 263 264The current operation mode or type of the cgroup is shown in the 265"cgroup.type" file which indicates whether the cgroup is a normal 266domain, a domain which is serving as the domain of a threaded subtree, 267or a threaded cgroup. 268 269On creation, a cgroup is always a domain cgroup and can be made 270threaded by writing "threaded" to the "cgroup.type" file. The 271operation is single direction:: 272 273 # echo threaded > cgroup.type 274 275Once threaded, the cgroup can't be made a domain again. To enable the 276thread mode, the following conditions must be met. 277 278- As the cgroup will join the parent's resource domain. The parent 279 must either be a valid (threaded) domain or a threaded cgroup. 280 281- When the parent is an unthreaded domain, it must not have any domain 282 controllers enabled or populated domain children. The root is 283 exempt from this requirement. 284 285Topology-wise, a cgroup can be in an invalid state. Please consider 286the following topology:: 287 288 A (threaded domain) - B (threaded) - C (domain, just created) 289 290C is created as a domain but isn't connected to a parent which can 291host child domains. C can't be used until it is turned into a 292threaded cgroup. "cgroup.type" file will report "domain (invalid)" in 293these cases. Operations which fail due to invalid topology use 294EOPNOTSUPP as the errno. 295 296A domain cgroup is turned into a threaded domain when one of its child 297cgroup becomes threaded or threaded controllers are enabled in the 298"cgroup.subtree_control" file while there are processes in the cgroup. 299A threaded domain reverts to a normal domain when the conditions 300clear. 301 302When read, "cgroup.threads" contains the list of the thread IDs of all 303threads in the cgroup. Except that the operations are per-thread 304instead of per-process, "cgroup.threads" has the same format and 305behaves the same way as "cgroup.procs". While "cgroup.threads" can be 306written to in any cgroup, as it can only move threads inside the same 307threaded domain, its operations are confined inside each threaded 308subtree. 309 310The threaded domain cgroup serves as the resource domain for the whole 311subtree, and, while the threads can be scattered across the subtree, 312all the processes are considered to be in the threaded domain cgroup. 313"cgroup.procs" in a threaded domain cgroup contains the PIDs of all 314processes in the subtree and is not readable in the subtree proper. 315However, "cgroup.procs" can be written to from anywhere in the subtree 316to migrate all threads of the matching process to the cgroup. 317 318Only threaded controllers can be enabled in a threaded subtree. When 319a threaded controller is enabled inside a threaded subtree, it only 320accounts for and controls resource consumptions associated with the 321threads in the cgroup and its descendants. All consumptions which 322aren't tied to a specific thread belong to the threaded domain cgroup. 323 324Because a threaded subtree is exempt from no internal process 325constraint, a threaded controller must be able to handle competition 326between threads in a non-leaf cgroup and its child cgroups. Each 327threaded controller defines how such competitions are handled. 328 329 330[Un]populated Notification 331-------------------------- 332 333Each non-root cgroup has a "cgroup.events" file which contains 334"populated" field indicating whether the cgroup's sub-hierarchy has 335live processes in it. Its value is 0 if there is no live process in 336the cgroup and its descendants; otherwise, 1. poll and [id]notify 337events are triggered when the value changes. This can be used, for 338example, to start a clean-up operation after all processes of a given 339sub-hierarchy have exited. The populated state updates and 340notifications are recursive. Consider the following sub-hierarchy 341where the numbers in the parentheses represent the numbers of processes 342in each cgroup:: 343 344 A(4) - B(0) - C(1) 345 \ D(0) 346 347A, B and C's "populated" fields would be 1 while D's 0. After the one 348process in C exits, B and C's "populated" fields would flip to "0" and 349file modified events will be generated on the "cgroup.events" files of 350both cgroups. 351 352 353Controlling Controllers 354----------------------- 355 356Enabling and Disabling 357~~~~~~~~~~~~~~~~~~~~~~ 358 359Each cgroup has a "cgroup.controllers" file which lists all 360controllers available for the cgroup to enable:: 361 362 # cat cgroup.controllers 363 cpu io memory 364 365No controller is enabled by default. Controllers can be enabled and 366disabled by writing to the "cgroup.subtree_control" file:: 367 368 # echo "+cpu +memory -io" > cgroup.subtree_control 369 370Only controllers which are listed in "cgroup.controllers" can be 371enabled. When multiple operations are specified as above, either they 372all succeed or fail. If multiple operations on the same controller 373are specified, the last one is effective. 374 375Enabling a controller in a cgroup indicates that the distribution of 376the target resource across its immediate children will be controlled. 377Consider the following sub-hierarchy. The enabled controllers are 378listed in parentheses:: 379 380 A(cpu,memory) - B(memory) - C() 381 \ D() 382 383As A has "cpu" and "memory" enabled, A will control the distribution 384of CPU cycles and memory to its children, in this case, B. As B has 385"memory" enabled but not "CPU", C and D will compete freely on CPU 386cycles but their division of memory available to B will be controlled. 387 388As a controller regulates the distribution of the target resource to 389the cgroup's children, enabling it creates the controller's interface 390files in the child cgroups. In the above example, enabling "cpu" on B 391would create the "cpu." prefixed controller interface files in C and 392D. Likewise, disabling "memory" from B would remove the "memory." 393prefixed controller interface files from C and D. This means that the 394controller interface files - anything which doesn't start with 395"cgroup." are owned by the parent rather than the cgroup itself. 396 397 398Top-down Constraint 399~~~~~~~~~~~~~~~~~~~ 400 401Resources are distributed top-down and a cgroup can further distribute 402a resource only if the resource has been distributed to it from the 403parent. This means that all non-root "cgroup.subtree_control" files 404can only contain controllers which are enabled in the parent's 405"cgroup.subtree_control" file. A controller can be enabled only if 406the parent has the controller enabled and a controller can't be 407disabled if one or more children have it enabled. 408 409 410No Internal Process Constraint 411~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 412 413Non-root cgroups can distribute domain resources to their children 414only when they don't have any processes of their own. In other words, 415only domain cgroups which don't contain any processes can have domain 416controllers enabled in their "cgroup.subtree_control" files. 417 418This guarantees that, when a domain controller is looking at the part 419of the hierarchy which has it enabled, processes are always only on 420the leaves. This rules out situations where child cgroups compete 421against internal processes of the parent. 422 423The root cgroup is exempt from this restriction. Root contains 424processes and anonymous resource consumption which can't be associated 425with any other cgroups and requires special treatment from most 426controllers. How resource consumption in the root cgroup is governed 427is up to each controller (for more information on this topic please 428refer to the Non-normative information section in the Controllers 429chapter). 430 431Note that the restriction doesn't get in the way if there is no 432enabled controller in the cgroup's "cgroup.subtree_control". This is 433important as otherwise it wouldn't be possible to create children of a 434populated cgroup. To control resource distribution of a cgroup, the 435cgroup must create children and transfer all its processes to the 436children before enabling controllers in its "cgroup.subtree_control" 437file. 438 439 440Delegation 441---------- 442 443Model of Delegation 444~~~~~~~~~~~~~~~~~~~ 445 446A cgroup can be delegated in two ways. First, to a less privileged 447user by granting write access of the directory and its "cgroup.procs", 448"cgroup.threads" and "cgroup.subtree_control" files to the user. 449Second, if the "nsdelegate" mount option is set, automatically to a 450cgroup namespace on namespace creation. 451 452Because the resource control interface files in a given directory 453control the distribution of the parent's resources, the delegatee 454shouldn't be allowed to write to them. For the first method, this is 455achieved by not granting access to these files. For the second, the 456kernel rejects writes to all files other than "cgroup.procs" and 457"cgroup.subtree_control" on a namespace root from inside the 458namespace. 459 460The end results are equivalent for both delegation types. Once 461delegated, the user can build sub-hierarchy under the directory, 462organize processes inside it as it sees fit and further distribute the 463resources it received from the parent. The limits and other settings 464of all resource controllers are hierarchical and regardless of what 465happens in the delegated sub-hierarchy, nothing can escape the 466resource restrictions imposed by the parent. 467 468Currently, cgroup doesn't impose any restrictions on the number of 469cgroups in or nesting depth of a delegated sub-hierarchy; however, 470this may be limited explicitly in the future. 471 472 473Delegation Containment 474~~~~~~~~~~~~~~~~~~~~~~ 475 476A delegated sub-hierarchy is contained in the sense that processes 477can't be moved into or out of the sub-hierarchy by the delegatee. 478 479For delegations to a less privileged user, this is achieved by 480requiring the following conditions for a process with a non-root euid 481to migrate a target process into a cgroup by writing its PID to the 482"cgroup.procs" file. 483 484- The writer must have write access to the "cgroup.procs" file. 485 486- The writer must have write access to the "cgroup.procs" file of the 487 common ancestor of the source and destination cgroups. 488 489The above two constraints ensure that while a delegatee may migrate 490processes around freely in the delegated sub-hierarchy it can't pull 491in from or push out to outside the sub-hierarchy. 492 493For an example, let's assume cgroups C0 and C1 have been delegated to 494user U0 who created C00, C01 under C0 and C10 under C1 as follows and 495all processes under C0 and C1 belong to U0:: 496 497 ~~~~~~~~~~~~~ - C0 - C00 498 ~ cgroup ~ \ C01 499 ~ hierarchy ~ 500 ~~~~~~~~~~~~~ - C1 - C10 501 502Let's also say U0 wants to write the PID of a process which is 503currently in C10 into "C00/cgroup.procs". U0 has write access to the 504file; however, the common ancestor of the source cgroup C10 and the 505destination cgroup C00 is above the points of delegation and U0 would 506not have write access to its "cgroup.procs" files and thus the write 507will be denied with -EACCES. 508 509For delegations to namespaces, containment is achieved by requiring 510that both the source and destination cgroups are reachable from the 511namespace of the process which is attempting the migration. If either 512is not reachable, the migration is rejected with -ENOENT. 513 514 515Guidelines 516---------- 517 518Organize Once and Control 519~~~~~~~~~~~~~~~~~~~~~~~~~ 520 521Migrating a process across cgroups is a relatively expensive operation 522and stateful resources such as memory are not moved together with the 523process. This is an explicit design decision as there often exist 524inherent trade-offs between migration and various hot paths in terms 525of synchronization cost. 526 527As such, migrating processes across cgroups frequently as a means to 528apply different resource restrictions is discouraged. A workload 529should be assigned to a cgroup according to the system's logical and 530resource structure once on start-up. Dynamic adjustments to resource 531distribution can be made by changing controller configuration through 532the interface files. 533 534 535Avoid Name Collisions 536~~~~~~~~~~~~~~~~~~~~~ 537 538Interface files for a cgroup and its children cgroups occupy the same 539directory and it is possible to create children cgroups which collide 540with interface files. 541 542All cgroup core interface files are prefixed with "cgroup." and each 543controller's interface files are prefixed with the controller name and 544a dot. A controller's name is composed of lower case alphabets and 545'_'s but never begins with an '_' so it can be used as the prefix 546character for collision avoidance. Also, interface file names won't 547start or end with terms which are often used in categorizing workloads 548such as job, service, slice, unit or workload. 549 550cgroup doesn't do anything to prevent name collisions and it's the 551user's responsibility to avoid them. 552 553 554Resource Distribution Models 555============================ 556 557cgroup controllers implement several resource distribution schemes 558depending on the resource type and expected use cases. This section 559describes major schemes in use along with their expected behaviors. 560 561 562Weights 563------- 564 565A parent's resource is distributed by adding up the weights of all 566active children and giving each the fraction matching the ratio of its 567weight against the sum. As only children which can make use of the 568resource at the moment participate in the distribution, this is 569work-conserving. Due to the dynamic nature, this model is usually 570used for stateless resources. 571 572All weights are in the range [1, 10000] with the default at 100. This 573allows symmetric multiplicative biases in both directions at fine 574enough granularity while staying in the intuitive range. 575 576As long as the weight is in range, all configuration combinations are 577valid and there is no reason to reject configuration changes or 578process migrations. 579 580"cpu.weight" proportionally distributes CPU cycles to active children 581and is an example of this type. 582 583 584Limits 585------ 586 587A child can only consume upto the configured amount of the resource. 588Limits can be over-committed - the sum of the limits of children can 589exceed the amount of resource available to the parent. 590 591Limits are in the range [0, max] and defaults to "max", which is noop. 592 593As limits can be over-committed, all configuration combinations are 594valid and there is no reason to reject configuration changes or 595process migrations. 596 597"io.max" limits the maximum BPS and/or IOPS that a cgroup can consume 598on an IO device and is an example of this type. 599 600 601Protections 602----------- 603 604A cgroup is protected to be allocated upto the configured amount of 605the resource if the usages of all its ancestors are under their 606protected levels. Protections can be hard guarantees or best effort 607soft boundaries. Protections can also be over-committed in which case 608only upto the amount available to the parent is protected among 609children. 610 611Protections are in the range [0, max] and defaults to 0, which is 612noop. 613 614As protections can be over-committed, all configuration combinations 615are valid and there is no reason to reject configuration changes or 616process migrations. 617 618"memory.low" implements best-effort memory protection and is an 619example of this type. 620 621 622Allocations 623----------- 624 625A cgroup is exclusively allocated a certain amount of a finite 626resource. Allocations can't be over-committed - the sum of the 627allocations of children can not exceed the amount of resource 628available to the parent. 629 630Allocations are in the range [0, max] and defaults to 0, which is no 631resource. 632 633As allocations can't be over-committed, some configuration 634combinations are invalid and should be rejected. Also, if the 635resource is mandatory for execution of processes, process migrations 636may be rejected. 637 638"cpu.rt.max" hard-allocates realtime slices and is an example of this 639type. 640 641 642Interface Files 643=============== 644 645Format 646------ 647 648All interface files should be in one of the following formats whenever 649possible:: 650 651 New-line separated values 652 (when only one value can be written at once) 653 654 VAL0\n 655 VAL1\n 656 ... 657 658 Space separated values 659 (when read-only or multiple values can be written at once) 660 661 VAL0 VAL1 ...\n 662 663 Flat keyed 664 665 KEY0 VAL0\n 666 KEY1 VAL1\n 667 ... 668 669 Nested keyed 670 671 KEY0 SUB_KEY0=VAL00 SUB_KEY1=VAL01... 672 KEY1 SUB_KEY0=VAL10 SUB_KEY1=VAL11... 673 ... 674 675For a writable file, the format for writing should generally match 676reading; however, controllers may allow omitting later fields or 677implement restricted shortcuts for most common use cases. 678 679For both flat and nested keyed files, only the values for a single key 680can be written at a time. For nested keyed files, the sub key pairs 681may be specified in any order and not all pairs have to be specified. 682 683 684Conventions 685----------- 686 687- Settings for a single feature should be contained in a single file. 688 689- The root cgroup should be exempt from resource control and thus 690 shouldn't have resource control interface files. Also, 691 informational files on the root cgroup which end up showing global 692 information available elsewhere shouldn't exist. 693 694- If a controller implements weight based resource distribution, its 695 interface file should be named "weight" and have the range [1, 696 10000] with 100 as the default. The values are chosen to allow 697 enough and symmetric bias in both directions while keeping it 698 intuitive (the default is 100%). 699 700- If a controller implements an absolute resource guarantee and/or 701 limit, the interface files should be named "min" and "max" 702 respectively. If a controller implements best effort resource 703 guarantee and/or limit, the interface files should be named "low" 704 and "high" respectively. 705 706 In the above four control files, the special token "max" should be 707 used to represent upward infinity for both reading and writing. 708 709- If a setting has a configurable default value and keyed specific 710 overrides, the default entry should be keyed with "default" and 711 appear as the first entry in the file. 712 713 The default value can be updated by writing either "default $VAL" or 714 "$VAL". 715 716 When writing to update a specific override, "default" can be used as 717 the value to indicate removal of the override. Override entries 718 with "default" as the value must not appear when read. 719 720 For example, a setting which is keyed by major:minor device numbers 721 with integer values may look like the following:: 722 723 # cat cgroup-example-interface-file 724 default 150 725 8:0 300 726 727 The default value can be updated by:: 728 729 # echo 125 > cgroup-example-interface-file 730 731 or:: 732 733 # echo "default 125" > cgroup-example-interface-file 734 735 An override can be set by:: 736 737 # echo "8:16 170" > cgroup-example-interface-file 738 739 and cleared by:: 740 741 # echo "8:0 default" > cgroup-example-interface-file 742 # cat cgroup-example-interface-file 743 default 125 744 8:16 170 745 746- For events which are not very high frequency, an interface file 747 "events" should be created which lists event key value pairs. 748 Whenever a notifiable event happens, file modified event should be 749 generated on the file. 750 751 752Core Interface Files 753-------------------- 754 755All cgroup core files are prefixed with "cgroup." 756 757 cgroup.type 758 759 A read-write single value file which exists on non-root 760 cgroups. 761 762 When read, it indicates the current type of the cgroup, which 763 can be one of the following values. 764 765 - "domain" : A normal valid domain cgroup. 766 767 - "domain threaded" : A threaded domain cgroup which is 768 serving as the root of a threaded subtree. 769 770 - "domain invalid" : A cgroup which is in an invalid state. 771 It can't be populated or have controllers enabled. It may 772 be allowed to become a threaded cgroup. 773 774 - "threaded" : A threaded cgroup which is a member of a 775 threaded subtree. 776 777 A cgroup can be turned into a threaded cgroup by writing 778 "threaded" to this file. 779 780 cgroup.procs 781 A read-write new-line separated values file which exists on 782 all cgroups. 783 784 When read, it lists the PIDs of all processes which belong to 785 the cgroup one-per-line. The PIDs are not ordered and the 786 same PID may show up more than once if the process got moved 787 to another cgroup and then back or the PID got recycled while 788 reading. 789 790 A PID can be written to migrate the process associated with 791 the PID to the cgroup. The writer should match all of the 792 following conditions. 793 794 - It must have write access to the "cgroup.procs" file. 795 796 - It must have write access to the "cgroup.procs" file of the 797 common ancestor of the source and destination cgroups. 798 799 When delegating a sub-hierarchy, write access to this file 800 should be granted along with the containing directory. 801 802 In a threaded cgroup, reading this file fails with EOPNOTSUPP 803 as all the processes belong to the thread root. Writing is 804 supported and moves every thread of the process to the cgroup. 805 806 cgroup.threads 807 A read-write new-line separated values file which exists on 808 all cgroups. 809 810 When read, it lists the TIDs of all threads which belong to 811 the cgroup one-per-line. The TIDs are not ordered and the 812 same TID may show up more than once if the thread got moved to 813 another cgroup and then back or the TID got recycled while 814 reading. 815 816 A TID can be written to migrate the thread associated with the 817 TID to the cgroup. The writer should match all of the 818 following conditions. 819 820 - It must have write access to the "cgroup.threads" file. 821 822 - The cgroup that the thread is currently in must be in the 823 same resource domain as the destination cgroup. 824 825 - It must have write access to the "cgroup.procs" file of the 826 common ancestor of the source and destination cgroups. 827 828 When delegating a sub-hierarchy, write access to this file 829 should be granted along with the containing directory. 830 831 cgroup.controllers 832 A read-only space separated values file which exists on all 833 cgroups. 834 835 It shows space separated list of all controllers available to 836 the cgroup. The controllers are not ordered. 837 838 cgroup.subtree_control 839 A read-write space separated values file which exists on all 840 cgroups. Starts out empty. 841 842 When read, it shows space separated list of the controllers 843 which are enabled to control resource distribution from the 844 cgroup to its children. 845 846 Space separated list of controllers prefixed with '+' or '-' 847 can be written to enable or disable controllers. A controller 848 name prefixed with '+' enables the controller and '-' 849 disables. If a controller appears more than once on the list, 850 the last one is effective. When multiple enable and disable 851 operations are specified, either all succeed or all fail. 852 853 cgroup.events 854 A read-only flat-keyed file which exists on non-root cgroups. 855 The following entries are defined. Unless specified 856 otherwise, a value change in this file generates a file 857 modified event. 858 859 populated 860 1 if the cgroup or its descendants contains any live 861 processes; otherwise, 0. 862 863 cgroup.max.descendants 864 A read-write single value files. The default is "max". 865 866 Maximum allowed number of descent cgroups. 867 If the actual number of descendants is equal or larger, 868 an attempt to create a new cgroup in the hierarchy will fail. 869 870 cgroup.max.depth 871 A read-write single value files. The default is "max". 872 873 Maximum allowed descent depth below the current cgroup. 874 If the actual descent depth is equal or larger, 875 an attempt to create a new child cgroup will fail. 876 877 cgroup.stat 878 A read-only flat-keyed file with the following entries: 879 880 nr_descendants 881 Total number of visible descendant cgroups. 882 883 nr_dying_descendants 884 Total number of dying descendant cgroups. A cgroup becomes 885 dying after being deleted by a user. The cgroup will remain 886 in dying state for some time undefined time (which can depend 887 on system load) before being completely destroyed. 888 889 A process can't enter a dying cgroup under any circumstances, 890 a dying cgroup can't revive. 891 892 A dying cgroup can consume system resources not exceeding 893 limits, which were active at the moment of cgroup deletion. 894 895 896Controllers 897=========== 898 899CPU 900--- 901 902The "cpu" controllers regulates distribution of CPU cycles. This 903controller implements weight and absolute bandwidth limit models for 904normal scheduling policy and absolute bandwidth allocation model for 905realtime scheduling policy. 906 907WARNING: cgroup2 doesn't yet support control of realtime processes and 908the cpu controller can only be enabled when all RT processes are in 909the root cgroup. Be aware that system management software may already 910have placed RT processes into nonroot cgroups during the system boot 911process, and these processes may need to be moved to the root cgroup 912before the cpu controller can be enabled. 913 914 915CPU Interface Files 916~~~~~~~~~~~~~~~~~~~ 917 918All time durations are in microseconds. 919 920 cpu.stat 921 A read-only flat-keyed file which exists on non-root cgroups. 922 This file exists whether the controller is enabled or not. 923 924 It always reports the following three stats: 925 926 - usage_usec 927 - user_usec 928 - system_usec 929 930 and the following three when the controller is enabled: 931 932 - nr_periods 933 - nr_throttled 934 - throttled_usec 935 936 cpu.weight 937 A read-write single value file which exists on non-root 938 cgroups. The default is "100". 939 940 The weight in the range [1, 10000]. 941 942 cpu.weight.nice 943 A read-write single value file which exists on non-root 944 cgroups. The default is "0". 945 946 The nice value is in the range [-20, 19]. 947 948 This interface file is an alternative interface for 949 "cpu.weight" and allows reading and setting weight using the 950 same values used by nice(2). Because the range is smaller and 951 granularity is coarser for the nice values, the read value is 952 the closest approximation of the current weight. 953 954 cpu.max 955 A read-write two value file which exists on non-root cgroups. 956 The default is "max 100000". 957 958 The maximum bandwidth limit. It's in the following format:: 959 960 $MAX $PERIOD 961 962 which indicates that the group may consume upto $MAX in each 963 $PERIOD duration. "max" for $MAX indicates no limit. If only 964 one number is written, $MAX is updated. 965 966 967Memory 968------ 969 970The "memory" controller regulates distribution of memory. Memory is 971stateful and implements both limit and protection models. Due to the 972intertwining between memory usage and reclaim pressure and the 973stateful nature of memory, the distribution model is relatively 974complex. 975 976While not completely water-tight, all major memory usages by a given 977cgroup are tracked so that the total memory consumption can be 978accounted and controlled to a reasonable extent. Currently, the 979following types of memory usages are tracked. 980 981- Userland memory - page cache and anonymous memory. 982 983- Kernel data structures such as dentries and inodes. 984 985- TCP socket buffers. 986 987The above list may expand in the future for better coverage. 988 989 990Memory Interface Files 991~~~~~~~~~~~~~~~~~~~~~~ 992 993All memory amounts are in bytes. If a value which is not aligned to 994PAGE_SIZE is written, the value may be rounded up to the closest 995PAGE_SIZE multiple when read back. 996 997 memory.current 998 A read-only single value file which exists on non-root 999 cgroups. 1000
1001 The total amount of memory currently being used by the cgroup 1002 and its descendants. 1003 1004 memory.low 1005 A read-write single value file which exists on non-root 1006 cgroups. The default is "0". 1007 1008 Best-effort memory protection. If the memory usages of a 1009 cgroup and all its ancestors are below their low boundaries, 1010 the cgroup's memory won't be reclaimed unless memory can be 1011 reclaimed from unprotected cgroups. 1012 1013 Putting more memory than generally available under this 1014 protection is discouraged. 1015 1016 memory.high 1017 A read-write single value file which exists on non-root 1018 cgroups. The default is "max". 1019 1020 Memory usage throttle limit. This is the main mechanism to 1021 control memory usage of a cgroup. If a cgroup's usage goes 1022 over the high boundary, the processes of the cgroup are 1023 throttled and put under heavy reclaim pressure. 1024 1025 Going over the high limit never invokes the OOM killer and 1026 under extreme conditions the limit may be breached. 1027 1028 memory.max 1029 A read-write single value file which exists on non-root 1030 cgroups. The default is "max". 1031 1032 Memory usage hard limit. This is the final protection 1033 mechanism. If a cgroup's memory usage reaches this limit and 1034 can't be reduced, the OOM killer is invoked in the cgroup. 1035 Under certain circumstances, the usage may go over the limit 1036 temporarily. 1037 1038 This is the ultimate protection mechanism. As long as the 1039 high limit is used and monitored properly, this limit's 1040 utility is limited to providing the final safety net. 1041 1042 memory.events 1043 A read-only flat-keyed file which exists on non-root cgroups. 1044 The following entries are defined. Unless specified 1045 otherwise, a value change in this file generates a file 1046 modified event. 1047 1048 low 1049 The number of times the cgroup is reclaimed due to 1050 high memory pressure even though its usage is under 1051 the low boundary. This usually indicates that the low 1052 boundary is over-committed. 1053 1054 high 1055 The number of times processes of the cgroup are 1056 throttled and routed to perform direct memory reclaim 1057 because the high memory boundary was exceeded. For a 1058 cgroup whose memory usage is capped by the high limit 1059 rather than global memory pressure, this event's 1060 occurrences are expected. 1061 1062 max 1063 The number of times the cgroup's memory usage was 1064 about to go over the max boundary. If direct reclaim 1065 fails to bring it down, the cgroup goes to OOM state. 1066 1067 oom 1068 The number of time the cgroup's memory usage was 1069 reached the limit and allocation was about to fail. 1070 1071 Depending on context result could be invocation of OOM 1072 killer and retrying allocation or failing allocation. 1073 1074 Failed allocation in its turn could be returned into 1075 userspace as -ENOMEM or silently ignored in cases like 1076 disk readahead. For now OOM in memory cgroup kills 1077 tasks iff shortage has happened inside page fault. 1078 1079 oom_kill 1080 The number of processes belonging to this cgroup 1081 killed by any kind of OOM killer. 1082 1083 memory.stat 1084 A read-only flat-keyed file which exists on non-root cgroups. 1085 1086 This breaks down the cgroup's memory footprint into different 1087 types of memory, type-specific details, and other information 1088 on the state and past events of the memory management system. 1089 1090 All memory amounts are in bytes. 1091 1092 The entries are ordered to be human readable, and new entries 1093 can show up in the middle. Don't rely on items remaining in a 1094 fixed position; use the keys to look up specific values! 1095 1096 anon 1097 Amount of memory used in anonymous mappings such as 1098 brk(), sbrk(), and mmap(MAP_ANONYMOUS) 1099 1100 file 1101 Amount of memory used to cache filesystem data, 1102 including tmpfs and shared memory. 1103 1104 kernel_stack 1105 Amount of memory allocated to kernel stacks. 1106 1107 slab 1108 Amount of memory used for storing in-kernel data 1109 structures. 1110 1111 sock 1112 Amount of memory used in network transmission buffers 1113 1114 shmem 1115 Amount of cached filesystem data that is swap-backed, 1116 such as tmpfs, shm segments, shared anonymous mmap()s 1117 1118 file_mapped 1119 Amount of cached filesystem data mapped with mmap() 1120 1121 file_dirty 1122 Amount of cached filesystem data that was modified but 1123 not yet written back to disk 1124 1125 file_writeback 1126 Amount of cached filesystem data that was modified and 1127 is currently being written back to disk 1128 1129 inactive_anon, active_anon, inactive_file, active_file, unevictable 1130 Amount of memory, swap-backed and filesystem-backed, 1131 on the internal memory management lists used by the 1132 page reclaim algorithm 1133 1134 slab_reclaimable 1135 Part of "slab" that might be reclaimed, such as 1136 dentries and inodes. 1137 1138 slab_unreclaimable 1139 Part of "slab" that cannot be reclaimed on memory 1140 pressure. 1141 1142 pgfault 1143 Total number of page faults incurred 1144 1145 pgmajfault 1146 Number of major page faults incurred 1147 1148 workingset_refault 1149 1150 Number of refaults of previously evicted pages 1151 1152 workingset_activate 1153 1154 Number of refaulted pages that were immediately activated 1155 1156 workingset_nodereclaim 1157 1158 Number of times a shadow node has been reclaimed 1159 1160 pgrefill 1161 1162 Amount of scanned pages (in an active LRU list) 1163 1164 pgscan 1165 1166 Amount of scanned pages (in an inactive LRU list) 1167 1168 pgsteal 1169 1170 Amount of reclaimed pages 1171 1172 pgactivate 1173 1174 Amount of pages moved to the active LRU list 1175 1176 pgdeactivate 1177 1178 Amount of pages moved to the inactive LRU lis 1179 1180 pglazyfree 1181 1182 Amount of pages postponed to be freed under memory pressure 1183 1184 pglazyfreed 1185 1186 Amount of reclaimed lazyfree pages 1187 1188 memory.swap.current 1189 A read-only single value file which exists on non-root 1190 cgroups. 1191 1192 The total amount of swap currently being used by the cgroup 1193 and its descendants. 1194 1195 memory.swap.max 1196 A read-write single value file which exists on non-root 1197 cgroups. The default is "max". 1198 1199 Swap usage hard limit. If a cgroup's swap usage reaches this 1200 limit, anonymous memory of the cgroup will not be swapped out. 1201 1202 1203Usage Guidelines 1204~~~~~~~~~~~~~~~~ 1205 1206"memory.high" is the main mechanism to control memory usage. 1207Over-committing on high limit (sum of high limits > available memory) 1208and letting global memory pressure to distribute memory according to 1209usage is a viable strategy. 1210 1211Because breach of the high limit doesn't trigger the OOM killer but 1212throttles the offending cgroup, a management agent has ample 1213opportunities to monitor and take appropriate actions such as granting 1214more memory or terminating the workload. 1215 1216Determining whether a cgroup has enough memory is not trivial as 1217memory usage doesn't indicate whether the workload can benefit from 1218more memory. For example, a workload which writes data received from 1219network to a file can use all available memory but can also operate as 1220performant with a small amount of memory. A measure of memory 1221pressure - how much the workload is being impacted due to lack of 1222memory - is necessary to determine whether a workload needs more 1223memory; unfortunately, memory pressure monitoring mechanism isn't 1224implemented yet. 1225 1226 1227Memory Ownership 1228~~~~~~~~~~~~~~~~ 1229 1230A memory area is charged to the cgroup which instantiated it and stays 1231charged to the cgroup until the area is released. Migrating a process 1232to a different cgroup doesn't move the memory usages that it 1233instantiated while in the previous cgroup to the new cgroup. 1234 1235A memory area may be used by processes belonging to different cgroups. 1236To which cgroup the area will be charged is in-deterministic; however, 1237over time, the memory area is likely to end up in a cgroup which has 1238enough memory allowance to avoid high reclaim pressure. 1239 1240If a cgroup sweeps a considerable amount of memory which is expected 1241to be accessed repeatedly by other cgroups, it may make sense to use 1242POSIX_FADV_DONTNEED to relinquish the ownership of memory areas 1243belonging to the affected files to ensure correct memory ownership. 1244 1245 1246IO 1247-- 1248 1249The "io" controller regulates the distribution of IO resources. This 1250controller implements both weight based and absolute bandwidth or IOPS 1251limit distribution; however, weight based distribution is available 1252only if cfq-iosched is in use and neither scheme is available for 1253blk-mq devices. 1254 1255 1256IO Interface Files 1257~~~~~~~~~~~~~~~~~~ 1258 1259 io.stat 1260 A read-only nested-keyed file which exists on non-root 1261 cgroups. 1262 1263 Lines are keyed by $MAJ:$MIN device numbers and not ordered. 1264 The following nested keys are defined. 1265 1266 ====== =================== 1267 rbytes Bytes read 1268 wbytes Bytes written 1269 rios Number of read IOs 1270 wios Number of write IOs 1271 ====== =================== 1272 1273 An example read output follows: 1274 1275 8:16 rbytes=1459200 wbytes=314773504 rios=192 wios=353 1276 8:0 rbytes=90430464 wbytes=299008000 rios=8950 wios=1252 1277 1278 io.weight 1279 A read-write flat-keyed file which exists on non-root cgroups. 1280 The default is "default 100". 1281 1282 The first line is the default weight applied to devices 1283 without specific override. The rest are overrides keyed by 1284 $MAJ:$MIN device numbers and not ordered. The weights are in 1285 the range [1, 10000] and specifies the relative amount IO time 1286 the cgroup can use in relation to its siblings. 1287 1288 The default weight can be updated by writing either "default 1289 $WEIGHT" or simply "$WEIGHT". Overrides can be set by writing 1290 "$MAJ:$MIN $WEIGHT" and unset by writing "$MAJ:$MIN default". 1291 1292 An example read output follows:: 1293 1294 default 100 1295 8:16 200 1296 8:0 50 1297 1298 io.max 1299 A read-write nested-keyed file which exists on non-root 1300 cgroups. 1301 1302 BPS and IOPS based IO limit. Lines are keyed by $MAJ:$MIN 1303 device numbers and not ordered. The following nested keys are 1304 defined. 1305 1306 ===== ================================== 1307 rbps Max read bytes per second 1308 wbps Max write bytes per second 1309 riops Max read IO operations per second 1310 wiops Max write IO operations per second 1311 ===== ================================== 1312 1313 When writing, any number of nested key-value pairs can be 1314 specified in any order. "max" can be specified as the value 1315 to remove a specific limit. If the same key is specified 1316 multiple times, the outcome is undefined. 1317 1318 BPS and IOPS are measured in each IO direction and IOs are 1319 delayed if limit is reached. Temporary bursts are allowed. 1320 1321 Setting read limit at 2M BPS and write at 120 IOPS for 8:16:: 1322 1323 echo "8:16 rbps=2097152 wiops=120" > io.max 1324 1325 Reading returns the following:: 1326 1327 8:16 rbps=2097152 wbps=max riops=max wiops=120 1328 1329 Write IOPS limit can be removed by writing the following:: 1330 1331 echo "8:16 wiops=max" > io.max 1332 1333 Reading now returns the following:: 1334 1335 8:16 rbps=2097152 wbps=max riops=max wiops=max 1336 1337 1338Writeback 1339~~~~~~~~~ 1340 1341Page cache is dirtied through buffered writes and shared mmaps and 1342written asynchronously to the backing filesystem by the writeback 1343mechanism. Writeback sits between the memory and IO domains and 1344regulates the proportion of dirty memory by balancing dirtying and 1345write IOs. 1346 1347The io controller, in conjunction with the memory controller, 1348implements control of page cache writeback IOs. The memory controller 1349defines the memory domain that dirty memory ratio is calculated and 1350maintained for and the io controller defines the io domain which 1351writes out dirty pages for the memory domain. Both system-wide and 1352per-cgroup dirty memory states are examined and the more restrictive 1353of the two is enforced. 1354 1355cgroup writeback requires explicit support from the underlying 1356filesystem. Currently, cgroup writeback is implemented on ext2, ext4 1357and btrfs. On other filesystems, all writeback IOs are attributed to 1358the root cgroup. 1359 1360There are inherent differences in memory and writeback management 1361which affects how cgroup ownership is tracked. Memory is tracked per 1362page while writeback per inode. For the purpose of writeback, an 1363inode is assigned to a cgroup and all IO requests to write dirty pages 1364from the inode are attributed to that cgroup. 1365 1366As cgroup ownership for memory is tracked per page, there can be pages 1367which are associated with different cgroups than the one the inode is 1368associated with. These are called foreign pages. The writeback 1369constantly keeps track of foreign pages and, if a particular foreign 1370cgroup becomes the majority over a certain period of time, switches 1371the ownership of the inode to that cgroup. 1372 1373While this model is enough for most use cases where a given inode is 1374mostly dirtied by a single cgroup even when the main writing cgroup 1375changes over time, use cases where multiple cgroups write to a single 1376inode simultaneously are not supported well. In such circumstances, a 1377significant portion of IOs are likely to be attributed incorrectly. 1378As memory controller assigns page ownership on the first use and 1379doesn't update it until the page is released, even if writeback 1380strictly follows page ownership, multiple cgroups dirtying overlapping 1381areas wouldn't work as expected. It's recommended to avoid such usage 1382patterns. 1383 1384The sysctl knobs which affect writeback behavior are applied to cgroup 1385writeback as follows. 1386 1387 vm.dirty_background_ratio, vm.dirty_ratio 1388 These ratios apply the same to cgroup writeback with the 1389 amount of available memory capped by limits imposed by the 1390 memory controller and system-wide clean memory. 1391 1392 vm.dirty_background_bytes, vm.dirty_bytes 1393 For cgroup writeback, this is calculated into ratio against 1394 total available memory and applied the same way as 1395 vm.dirty[_background]_ratio. 1396 1397 1398PID 1399--- 1400 1401The process number controller is used to allow a cgroup to stop any 1402new tasks from being fork()'d or clone()'d after a specified limit is 1403reached. 1404 1405The number of tasks in a cgroup can be exhausted in ways which other 1406controllers cannot prevent, thus warranting its own controller. For 1407example, a fork bomb is likely to exhaust the number of tasks before 1408hitting memory restrictions. 1409 1410Note that PIDs used in this controller refer to TIDs, process IDs as 1411used by the kernel. 1412 1413 1414PID Interface Files 1415~~~~~~~~~~~~~~~~~~~ 1416 1417 pids.max 1418 A read-write single value file which exists on non-root 1419 cgroups. The default is "max". 1420 1421 Hard limit of number of processes. 1422 1423 pids.current 1424 A read-only single value file which exists on all cgroups. 1425 1426 The number of processes currently in the cgroup and its 1427 descendants. 1428 1429Organisational operations are not blocked by cgroup policies, so it is 1430possible to have pids.current > pids.max. This can be done by either 1431setting the limit to be smaller than pids.current, or attaching enough 1432processes to the cgroup such that pids.current is larger than 1433pids.max. However, it is not possible to violate a cgroup PID policy 1434through fork() or clone(). These will return -EAGAIN if the creation 1435of a new process would cause a cgroup policy to be violated. 1436 1437 1438Device controller 1439----------------- 1440 1441Device controller manages access to device files. It includes both 1442creation of new device files (using mknod), and access to the 1443existing device files. 1444 1445Cgroup v2 device controller has no interface files and is implemented 1446on top of cgroup BPF. To control access to device files, a user may 1447create bpf programs of the BPF_CGROUP_DEVICE type and attach them 1448to cgroups. On an attempt to access a device file, corresponding 1449BPF programs will be executed, and depending on the return value 1450the attempt will succeed or fail with -EPERM. 1451 1452A BPF_CGROUP_DEVICE program takes a pointer to the bpf_cgroup_dev_ctx 1453structure, which describes the device access attempt: access type 1454(mknod/read/write) and device (type, major and minor numbers). 1455If the program returns 0, the attempt fails with -EPERM, otherwise 1456it succeeds. 1457 1458An example of BPF_CGROUP_DEVICE program may be found in the kernel 1459source tree in the tools/testing/selftests/bpf/dev_cgroup.c file. 1460 1461 1462RDMA 1463---- 1464 1465The "rdma" controller regulates the distribution and accounting of 1466of RDMA resources. 1467 1468RDMA Interface Files 1469~~~~~~~~~~~~~~~~~~~~ 1470 1471 rdma.max 1472 A readwrite nested-keyed file that exists for all the cgroups 1473 except root that describes current configured resource limit 1474 for a RDMA/IB device. 1475 1476 Lines are keyed by device name and are not ordered. 1477 Each line contains space separated resource name and its configured 1478 limit that can be distributed. 1479 1480 The following nested keys are defined. 1481 1482 ========== ============================= 1483 hca_handle Maximum number of HCA Handles 1484 hca_object Maximum number of HCA Objects 1485 ========== ============================= 1486 1487 An example for mlx4 and ocrdma device follows:: 1488 1489 mlx4_0 hca_handle=2 hca_object=2000 1490 ocrdma1 hca_handle=3 hca_object=max 1491 1492 rdma.current 1493 A read-only file that describes current resource usage. 1494 It exists for all the cgroup except root. 1495 1496 An example for mlx4 and ocrdma device follows:: 1497 1498 mlx4_0 hca_handle=1 hca_object=20 1499 ocrdma1 hca_handle=1 hca_object=23 1500 1501 1502Misc 1503---- 1504 1505perf_event 1506~~~~~~~~~~ 1507 1508perf_event controller, if not mounted on a legacy hierarchy, is 1509automatically enabled on the v2 hierarchy so that perf events can 1510always be filtered by cgroup v2 path. The controller can still be 1511moved to a legacy hierarchy after v2 hierarchy is populated. 1512 1513 1514Non-normative information 1515------------------------- 1516 1517This section contains information that isn't considered to be a part of 1518the stable kernel API and so is subject to change. 1519 1520 1521CPU controller root cgroup process behaviour 1522~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1523 1524When distributing CPU cycles in the root cgroup each thread in this 1525cgroup is treated as if it was hosted in a separate child cgroup of the 1526root cgroup. This child cgroup weight is dependent on its thread nice 1527level. 1528 1529For details of this mapping see sched_prio_to_weight array in 1530kernel/sched/core.c file (values from this array should be scaled 1531appropriately so the neutral - nice 0 - value is 100 instead of 1024). 1532 1533 1534IO controller root cgroup process behaviour 1535~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1536 1537Root cgroup processes are hosted in an implicit leaf child node. 1538When distributing IO resources this implicit child node is taken into 1539account as if it was a normal child cgroup of the root cgroup with a 1540weight value of 200. 1541 1542 1543Namespace 1544========= 1545 1546Basics 1547------ 1548 1549cgroup namespace provides a mechanism to virtualize the view of the 1550"/proc/$PID/cgroup" file and cgroup mounts. The CLONE_NEWCGROUP clone 1551flag can be used with clone(2) and unshare(2) to create a new cgroup 1552namespace. The process running inside the cgroup namespace will have 1553its "/proc/$PID/cgroup" output restricted to cgroupns root. The 1554cgroupns root is the cgroup of the process at the time of creation of 1555the cgroup namespace. 1556 1557Without cgroup namespace, the "/proc/$PID/cgroup" file shows the 1558complete path of the cgroup of a process. In a container setup where 1559a set of cgroups and namespaces are intended to isolate processes the 1560"/proc/$PID/cgroup" file may leak potential system level information 1561to the isolated processes. For Example:: 1562 1563 # cat /proc/self/cgroup 1564 0::/batchjobs/container_id1 1565 1566The path '/batchjobs/container_id1' can be considered as system-data 1567and undesirable to expose to the isolated processes. cgroup namespace 1568can be used to restrict visibility of this path. For example, before 1569creating a cgroup namespace, one would see:: 1570 1571 # ls -l /proc/self/ns/cgroup 1572 lrwxrwxrwx 1 root root 0 2014-07-15 10:37 /proc/self/ns/cgroup -> cgroup:[4026531835] 1573 # cat /proc/self/cgroup 1574 0::/batchjobs/container_id1 1575 1576After unsharing a new namespace, the view changes:: 1577 1578 # ls -l /proc/self/ns/cgroup 1579 lrwxrwxrwx 1 root root 0 2014-07-15 10:35 /proc/self/ns/cgroup -> cgroup:[4026532183] 1580 # cat /proc/self/cgroup 1581 0::/ 1582 1583When some thread from a multi-threaded process unshares its cgroup 1584namespace, the new cgroupns gets applied to the entire process (all 1585the threads). This is natural for the v2 hierarchy; however, for the 1586legacy hierarchies, this may be unexpected. 1587 1588A cgroup namespace is alive as long as there are processes inside or 1589mounts pinning it. When the last usage goes away, the cgroup 1590namespace is destroyed. The cgroupns root and the actual cgroups 1591remain. 1592 1593 1594The Root and Views 1595------------------ 1596 1597The 'cgroupns root' for a cgroup namespace is the cgroup in which the 1598process calling unshare(2) is running. For example, if a process in 1599/batchjobs/container_id1 cgroup calls unshare, cgroup 1600/batchjobs/container_id1 becomes the cgroupns root. For the 1601init_cgroup_ns, this is the real root ('/') cgroup. 1602 1603The cgroupns root cgroup does not change even if the namespace creator 1604process later moves to a different cgroup:: 1605 1606 # ~/unshare -c # unshare cgroupns in some cgroup 1607 # cat /proc/self/cgroup 1608 0::/ 1609 # mkdir sub_cgrp_1 1610 # echo 0 > sub_cgrp_1/cgroup.procs 1611 # cat /proc/self/cgroup 1612 0::/sub_cgrp_1 1613 1614Each process gets its namespace-specific view of "/proc/$PID/cgroup" 1615 1616Processes running inside the cgroup namespace will be able to see 1617cgroup paths (in /proc/self/cgroup) only inside their root cgroup. 1618From within an unshared cgroupns:: 1619 1620 # sleep 100000 & 1621 [1] 7353 1622 # echo 7353 > sub_cgrp_1/cgroup.procs 1623 # cat /proc/7353/cgroup 1624 0::/sub_cgrp_1 1625 1626From the initial cgroup namespace, the real cgroup path will be 1627visible:: 1628 1629 $ cat /proc/7353/cgroup 1630 0::/batchjobs/container_id1/sub_cgrp_1 1631 1632From a sibling cgroup namespace (that is, a namespace rooted at a 1633different cgroup), the cgroup path relative to its own cgroup 1634namespace root will be shown. For instance, if PID 7353's cgroup 1635namespace root is at '/batchjobs/container_id2', then it will see:: 1636 1637 # cat /proc/7353/cgroup 1638 0::/../container_id2/sub_cgrp_1 1639 1640Note that the relative path always starts with '/' to indicate that 1641its relative to the cgroup namespace root of the caller. 1642 1643 1644Migration and setns(2) 1645---------------------- 1646 1647Processes inside a cgroup namespace can move into and out of the 1648namespace root if they have proper access to external cgroups. For 1649example, from inside a namespace with cgroupns root at 1650/batchjobs/container_id1, and assuming that the global hierarchy is 1651still accessible inside cgroupns:: 1652 1653 # cat /proc/7353/cgroup 1654 0::/sub_cgrp_1 1655 # echo 7353 > batchjobs/container_id2/cgroup.procs 1656 # cat /proc/7353/cgroup 1657 0::/../container_id2 1658 1659Note that this kind of setup is not encouraged. A task inside cgroup 1660namespace should only be exposed to its own cgroupns hierarchy. 1661 1662setns(2) to another cgroup namespace is allowed when: 1663 1664(a) the process has CAP_SYS_ADMIN against its current user namespace 1665(b) the process has CAP_SYS_ADMIN against the target cgroup 1666 namespace's userns 1667 1668No implicit cgroup changes happen with attaching to another cgroup 1669namespace. It is expected that the someone moves the attaching 1670process under the target cgroup namespace root. 1671 1672 1673Interaction with Other Namespaces 1674--------------------------------- 1675 1676Namespace specific cgroup hierarchy can be mounted by a process 1677running inside a non-init cgroup namespace:: 1678 1679 # mount -t cgroup2 none $MOUNT_POINT 1680 1681This will mount the unified cgroup hierarchy with cgroupns root as the 1682filesystem root. The process needs CAP_SYS_ADMIN against its user and 1683mount namespaces. 1684 1685The virtualization of /proc/self/cgroup file combined with restricting 1686the view of cgroup hierarchy by namespace-private cgroupfs mount 1687provides a properly isolated cgroup view inside the container. 1688 1689 1690Information on Kernel Programming 1691================================= 1692 1693This section contains kernel programming information in the areas 1694where interacting with cgroup is necessary. cgroup core and 1695controllers are not covered. 1696 1697 1698Filesystem Support for Writeback 1699-------------------------------- 1700 1701A filesystem can support cgroup writeback by updating 1702address_space_operations->writepage[s]() to annotate bio's using the 1703following two functions. 1704 1705 wbc_init_bio(@wbc, @bio) 1706 Should be called for each bio carrying writeback data and 1707 associates the bio with the inode's owner cgroup. Can be 1708 called anytime between bio allocation and submission. 1709 1710 wbc_account_io(@wbc, @page, @bytes) 1711 Should be called for each data segment being written out. 1712 While this function doesn't care exactly when it's called 1713 during the writeback session, it's the easiest and most 1714 natural to call it as data segments are added to a bio. 1715 1716With writeback bio's annotated, cgroup support can be enabled per 1717super_block by setting SB_I_CGROUPWB in ->s_iflags. This allows for 1718selective disabling of cgroup writeback support which is helpful when 1719certain filesystem features, e.g. journaled data mode, are 1720incompatible. 1721 1722wbc_init_bio() binds the specified bio to its cgroup. Depending on 1723the configuration, the bio may be executed at a lower priority and if 1724the writeback session is holding shared resources, e.g. a journal 1725entry, may lead to priority inversion. There is no one easy solution 1726for the problem. Filesystems can try to work around specific problem 1727cases by skipping wbc_init_bio() or using bio_associate_blkcg() 1728directly. 1729 1730 1731Deprecated v1 Core Features 1732=========================== 1733 1734- Multiple hierarchies including named ones are not supported. 1735 1736- All v1 mount options are not supported. 1737 1738- The "tasks" file is removed and "cgroup.procs" is not sorted. 1739 1740- "cgroup.clone_children" is removed. 1741 1742- /proc/cgroups is meaningless for v2. Use "cgroup.controllers" file 1743 at the root instead. 1744 1745 1746Issues with v1 and Rationales for v2 1747==================================== 1748 1749Multiple Hierarchies 1750-------------------- 1751 1752cgroup v1 allowed an arbitrary number of hierarchies and each 1753hierarchy could host any number of controllers. While this seemed to 1754provide a high level of flexibility, it wasn't useful in practice. 1755 1756For example, as there is only one instance of each controller, utility 1757type controllers such as freezer which can be useful in all 1758hierarchies could only be used in one. The issue is exacerbated by 1759the fact that controllers couldn't be moved to another hierarchy once 1760hierarchies were populated. Another issue was that all controllers 1761bound to a hierarchy were forced to have exactly the same view of the 1762hierarchy. It wasn't possible to vary the granularity depending on 1763the specific controller. 1764 1765In practice, these issues heavily limited which controllers could be 1766put on the same hierarchy and most configurations resorted to putting 1767each controller on its own hierarchy. Only closely related ones, such 1768as the cpu and cpuacct controllers, made sense to be put on the same 1769hierarchy. This often meant that userland ended up managing multiple 1770similar hierarchies repeating the same steps on each hierarchy 1771whenever a hierarchy management operation was necessary. 1772 1773Furthermore, support for multiple hierarchies came at a steep cost. 1774It greatly complicated cgroup core implementation but more importantly 1775the support for multiple hierarchies restricted how cgroup could be 1776used in general and what controllers was able to do. 1777 1778There was no limit on how many hierarchies there might be, which meant 1779that a thread's cgroup membership couldn't be described in finite 1780length. The key might contain any number of entries and was unlimited 1781in length, which made it highly awkward to manipulate and led to 1782addition of controllers which existed only to identify membership, 1783which in turn exacerbated the original problem of proliferating number 1784of hierarchies. 1785 1786Also, as a controller couldn't have any expectation regarding the 1787topologies of hierarchies other controllers might be on, each 1788controller had to assume that all other controllers were attached to 1789completely orthogonal hierarchies. This made it impossible, or at 1790least very cumbersome, for controllers to cooperate with each other. 1791 1792In most use cases, putting controllers on hierarchies which are 1793completely orthogonal to each other isn't necessary. What usually is 1794called for is the ability to have differing levels of granularity 1795depending on the specific controller. In other words, hierarchy may 1796be collapsed from leaf towards root when viewed from specific 1797controllers. For example, a given configuration might not care about 1798how memory is distributed beyond a certain level while still wanting 1799to control how CPU cycles are distributed. 1800 1801 1802Thread Granularity 1803------------------ 1804 1805cgroup v1 allowed threads of a process to belong to different cgroups. 1806This didn't make sense for some controllers and those controllers 1807ended up implementing different ways to ignore such situations but 1808much more importantly it blurred the line between API exposed to 1809individual applications and system management interface. 1810 1811Generally, in-process knowledge is available only to the process 1812itself; thus, unlike service-level organization of processes, 1813categorizing threads of a process requires active participation from 1814the application which owns the target process. 1815 1816cgroup v1 had an ambiguously defined delegation model which got abused 1817in combination with thread granularity. cgroups were delegated to 1818individual applications so that they can create and manage their own 1819sub-hierarchies and control resource distributions along them. This 1820effectively raised cgroup to the status of a syscall-like API exposed 1821to lay programs. 1822 1823First of all, cgroup has a fundamentally inadequate interface to be 1824exposed this way. For a process to access its own knobs, it has to 1825extract the path on the target hierarchy from /proc/self/cgroup, 1826construct the path by appending the name of the knob to the path, open 1827and then read and/or write to it. This is not only extremely clunky 1828and unusual but also inherently racy. There is no conventional way to 1829define transaction across the required steps and nothing can guarantee 1830that the process would actually be operating on its own sub-hierarchy. 1831 1832cgroup controllers implemented a number of knobs which would never be 1833accepted as public APIs because they were just adding control knobs to 1834system-management pseudo filesystem. cgroup ended up with interface 1835knobs which were not properly abstracted or refined and directly 1836revealed kernel internal details. These knobs got exposed to 1837individual applications through the ill-defined delegation mechanism 1838effectively abusing cgroup as a shortcut to implementing public APIs 1839without going through the required scrutiny. 1840 1841This was painful for both userland and kernel. Userland ended up with 1842misbehaving and poorly abstracted interfaces and kernel exposing and 1843locked into constructs inadvertently. 1844 1845 1846Competition Between Inner Nodes and Threads 1847------------------------------------------- 1848 1849cgroup v1 allowed threads to be in any cgroups which created an 1850interesting problem where threads belonging to a parent cgroup and its 1851children cgroups competed for resources. This was nasty as two 1852different types of entities competed and there was no obvious way to 1853settle it. Different controllers did different things. 1854 1855The cpu controller considered threads and cgroups as equivalents and 1856mapped nice levels to cgroup weights. This worked for some cases but 1857fell flat when children wanted to be allocated specific ratios of CPU 1858cycles and the number of internal threads fluctuated - the ratios 1859constantly changed as the number of competing entities fluctuated. 1860There also were other issues. The mapping from nice level to weight 1861wasn't obvious or universal, and there were various other knobs which 1862simply weren't available for threads. 1863 1864The io controller implicitly created a hidden leaf node for each 1865cgroup to host the threads. The hidden leaf had its own copies of all 1866the knobs with ``leaf_`` prefixed. While this allowed equivalent 1867control over internal threads, it was with serious drawbacks. It 1868always added an extra layer of nesting which wouldn't be necessary 1869otherwise, made the interface messy and significantly complicated the 1870implementation. 1871 1872The memory controller didn't have a way to control what happened 1873between internal tasks and child cgroups and the behavior was not 1874clearly defined. There were attempts to add ad-hoc behaviors and 1875knobs to tailor the behavior to specific workloads which would have 1876led to problems extremely difficult to resolve in the long term. 1877 1878Multiple controllers struggled with internal tasks and came up with 1879different ways to deal with it; unfortunately, all the approaches were 1880severely flawed and, furthermore, the widely different behaviors 1881made cgroup as a whole highly inconsistent. 1882 1883This clearly is a problem which needs to be addressed from cgroup core 1884in a uniform way. 1885 1886 1887Other Interface Issues 1888---------------------- 1889 1890cgroup v1 grew without oversight and developed a large number of 1891idiosyncrasies and inconsistencies. One issue on the cgroup core side 1892was how an empty cgroup was notified - a userland helper binary was 1893forked and executed for each event. The event delivery wasn't 1894recursive or delegatable. The limitations of the mechanism also led 1895to in-kernel event delivery filtering mechanism further complicating 1896the interface. 1897 1898Controller interfaces were problematic too. An extreme example is 1899controllers completely ignoring hierarchical organization and treating 1900all cgroups as if they were all located directly under the root 1901cgroup. Some controllers exposed a large amount of inconsistent 1902implementation details to userland. 1903 1904There also was no consistency across controllers. When a new cgroup 1905was created, some controllers defaulted to not imposing extra 1906restrictions while others disallowed any resource usage until 1907explicitly configured. Configuration knobs for the same type of 1908control used widely differing naming schemes and formats. Statistics 1909and information knobs were named arbitrarily and used different 1910formats and units even in the same controller. 1911 1912cgroup v2 establishes common conventions where appropriate and updates 1913controllers so that they expose minimal and consistent interfaces. 1914 1915 1916Controller Issues and Remedies 1917------------------------------ 1918 1919Memory 1920~~~~~~ 1921 1922The original lower boundary, the soft limit, is defined as a limit 1923that is per default unset. As a result, the set of cgroups that 1924global reclaim prefers is opt-in, rather than opt-out. The costs for 1925optimizing these mostly negative lookups are so high that the 1926implementation, despite its enormous size, does not even provide the 1927basic desirable behavior. First off, the soft limit has no 1928hierarchical meaning. All configured groups are organized in a global 1929rbtree and treated like equal peers, regardless where they are located 1930in the hierarchy. This makes subtree delegation impossible. Second, 1931the soft limit reclaim pass is so aggressive that it not just 1932introduces high allocation latencies into the system, but also impacts 1933system performance due to overreclaim, to the point where the feature 1934becomes self-defeating. 1935 1936The memory.low boundary on the other hand is a top-down allocated 1937reserve. A cgroup enjoys reclaim protection when it and all its 1938ancestors are below their low boundaries, which makes delegation of 1939subtrees possible. Secondly, new cgroups have no reserve per default 1940and in the common case most cgroups are eligible for the preferred 1941reclaim pass. This allows the new low boundary to be efficiently 1942implemented with just a minor addition to the generic reclaim code, 1943without the need for out-of-band data structures and reclaim passes. 1944Because the generic reclaim code considers all cgroups except for the 1945ones running low in the preferred first reclaim pass, overreclaim of 1946individual groups is eliminated as well, resulting in much better 1947overall workload performance. 1948 1949The original high boundary, the hard limit, is defined as a strict 1950limit that can not budge, even if the OOM killer has to be called. 1951But this generally goes against the goal of making the most out of the 1952available memory. The memory consumption of workloads varies during 1953runtime, and that requires users to overcommit. But doing that with a 1954strict upper limit requires either a fairly accurate prediction of the 1955working set size or adding slack to the limit. Since working set size 1956estimation is hard and error prone, and getting it wrong results in 1957OOM kills, most users tend to err on the side of a looser limit and 1958end up wasting precious resources. 1959 1960The memory.high boundary on the other hand can be set much more 1961conservatively. When hit, it throttles allocations by forcing them 1962into direct reclaim to work off the excess, but it never invokes the 1963OOM killer. As a result, a high boundary that is chosen too 1964aggressively will not terminate the processes, but instead it will 1965lead to gradual performance degradation. The user can monitor this 1966and make corrections until the minimal memory footprint that still 1967gives acceptable performance is found. 1968 1969In extreme cases, with many concurrent allocations and a complete 1970breakdown of reclaim progress within the group, the high boundary can 1971be exceeded. But even then it's mostly better to satisfy the 1972allocation from the slack available in other groups or the rest of the 1973system than killing the group. Otherwise, memory.max is there to 1974limit this type of spillover and ultimately contain buggy or even 1975malicious applications. 1976 1977Setting the original memory.limit_in_bytes below the current usage was 1978subject to a race condition, where concurrent charges could cause the 1979limit setting to fail. memory.max on the other hand will first set the 1980limit to prevent new charges, and then reclaim and OOM kill until the 1981new limit is met - or the task writing to memory.max is killed. 1982 1983The combined memory+swap accounting and limiting is replaced by real 1984control over swap space. 1985 1986The main argument for a combined memory+swap facility in the original 1987cgroup design was that global or parental pressure would always be 1988able to swap all anonymous memory of a child group, regardless of the 1989child's own (possibly untrusted) configuration. However, untrusted 1990groups can sabotage swapping by other means - such as referencing its 1991anonymous memory in a tight loop - and an admin can not assume full 1992swappability when overcommitting untrusted jobs. 1993 1994For trusted jobs, on the other hand, a combined counter is not an 1995intuitive userspace interface, and it flies in the face of the idea 1996that cgroup controllers should account and limit specific physical 1997resources. Swap space is a resource like all others in the system, 1998and that's why unified hierarchy allows distributing it separately. 1999