linux/Documentation/cgroups/cgroups.txt
<<
>>
Prefs
   1                                CGROUPS
   2                                -------
   3
   4Written by Paul Menage <menage@google.com> based on
   5Documentation/cgroups/cpusets.txt
   6
   7Original copyright statements from cpusets.txt:
   8Portions Copyright (C) 2004 BULL SA.
   9Portions Copyright (c) 2004-2006 Silicon Graphics, Inc.
  10Modified by Paul Jackson <pj@sgi.com>
  11Modified by Christoph Lameter <clameter@sgi.com>
  12
  13CONTENTS:
  14=========
  15
  161. Control Groups
  17  1.1 What are cgroups ?
  18  1.2 Why are cgroups needed ?
  19  1.3 How are cgroups implemented ?
  20  1.4 What does notify_on_release do ?
  21  1.5 How do I use cgroups ?
  222. Usage Examples and Syntax
  23  2.1 Basic Usage
  24  2.2 Attaching processes
  253. Kernel API
  26  3.1 Overview
  27  3.2 Synchronization
  28  3.3 Subsystem API
  294. Questions
  30
  311. Control Groups
  32=================
  33
  341.1 What are cgroups ?
  35----------------------
  36
  37Control Groups provide a mechanism for aggregating/partitioning sets of
  38tasks, and all their future children, into hierarchical groups with
  39specialized behaviour.
  40
  41Definitions:
  42
  43A *cgroup* associates a set of tasks with a set of parameters for one
  44or more subsystems.
  45
  46A *subsystem* is a module that makes use of the task grouping
  47facilities provided by cgroups to treat groups of tasks in
  48particular ways. A subsystem is typically a "resource controller" that
  49schedules a resource or applies per-cgroup limits, but it may be
  50anything that wants to act on a group of processes, e.g. a
  51virtualization subsystem.
  52
  53A *hierarchy* is a set of cgroups arranged in a tree, such that
  54every task in the system is in exactly one of the cgroups in the
  55hierarchy, and a set of subsystems; each subsystem has system-specific
  56state attached to each cgroup in the hierarchy.  Each hierarchy has
  57an instance of the cgroup virtual filesystem associated with it.
  58
  59At any one time there may be multiple active hierarchies of task
  60cgroups. Each hierarchy is a partition of all tasks in the system.
  61
  62User level code may create and destroy cgroups by name in an
  63instance of the cgroup virtual file system, specify and query to
  64which cgroup a task is assigned, and list the task pids assigned to
  65a cgroup. Those creations and assignments only affect the hierarchy
  66associated with that instance of the cgroup file system.
  67
  68On their own, the only use for cgroups is for simple job
  69tracking. The intention is that other subsystems hook into the generic
  70cgroup support to provide new attributes for cgroups, such as
  71accounting/limiting the resources which processes in a cgroup can
  72access. For example, cpusets (see Documentation/cgroups/cpusets.txt) allows
  73you to associate a set of CPUs and a set of memory nodes with the
  74tasks in each cgroup.
  75
  761.2 Why are cgroups needed ?
  77----------------------------
  78
  79There are multiple efforts to provide process aggregations in the
  80Linux kernel, mainly for resource tracking purposes. Such efforts
  81include cpusets, CKRM/ResGroups, UserBeanCounters, and virtual server
  82namespaces. These all require the basic notion of a
  83grouping/partitioning of processes, with newly forked processes ending
  84in the same group (cgroup) as their parent process.
  85
  86The kernel cgroup patch provides the minimum essential kernel
  87mechanisms required to efficiently implement such groups. It has
  88minimal impact on the system fast paths, and provides hooks for
  89specific subsystems such as cpusets to provide additional behaviour as
  90desired.
  91
  92Multiple hierarchy support is provided to allow for situations where
  93the division of tasks into cgroups is distinctly different for
  94different subsystems - having parallel hierarchies allows each
  95hierarchy to be a natural division of tasks, without having to handle
  96complex combinations of tasks that would be present if several
  97unrelated subsystems needed to be forced into the same tree of
  98cgroups.
  99
 100At one extreme, each resource controller or subsystem could be in a
 101separate hierarchy; at the other extreme, all subsystems
 102would be attached to the same hierarchy.
 103
 104As an example of a scenario (originally proposed by vatsa@in.ibm.com)
 105that can benefit from multiple hierarchies, consider a large
 106university server with various users - students, professors, system
 107tasks etc. The resource planning for this server could be along the
 108following lines:
 109
 110       CPU :           Top cpuset
 111                       /       \
 112               CPUSet1         CPUSet2
 113                  |              |
 114               (Profs)         (Students)
 115
 116               In addition (system tasks) are attached to topcpuset (so
 117               that they can run anywhere) with a limit of 20%
 118
 119       Memory : Professors (50%), students (30%), system (20%)
 120
 121       Disk : Prof (50%), students (30%), system (20%)
 122
 123       Network : WWW browsing (20%), Network File System (60%), others (20%)
 124                               / \
 125                       Prof (15%) students (5%)
 126
 127Browsers like Firefox/Lynx go into the WWW network class, while (k)nfsd go
 128into NFS network class.
 129
 130At the same time Firefox/Lynx will share an appropriate CPU/Memory class
 131depending on who launched it (prof/student).
 132
 133With the ability to classify tasks differently for different resources
 134(by putting those resource subsystems in different hierarchies) then
 135the admin can easily set up a script which receives exec notifications
 136and depending on who is launching the browser he can
 137
 138       # echo browser_pid > /mnt/<restype>/<userclass>/tasks
 139
 140With only a single hierarchy, he now would potentially have to create
 141a separate cgroup for every browser launched and associate it with
 142approp network and other resource class.  This may lead to
 143proliferation of such cgroups.
 144
 145Also lets say that the administrator would like to give enhanced network
 146access temporarily to a student's browser (since it is night and the user
 147wants to do online gaming :))  OR give one of the students simulation
 148apps enhanced CPU power,
 149
 150With ability to write pids directly to resource classes, it's just a
 151matter of :
 152
 153       # echo pid > /mnt/network/<new_class>/tasks
 154       (after some time)
 155       # echo pid > /mnt/network/<orig_class>/tasks
 156
 157Without this ability, he would have to split the cgroup into
 158multiple separate ones and then associate the new cgroups with the
 159new resource classes.
 160
 161
 162
 1631.3 How are cgroups implemented ?
 164---------------------------------
 165
 166Control Groups extends the kernel as follows:
 167
 168 - Each task in the system has a reference-counted pointer to a
 169   css_set.
 170
 171 - A css_set contains a set of reference-counted pointers to
 172   cgroup_subsys_state objects, one for each cgroup subsystem
 173   registered in the system. There is no direct link from a task to
 174   the cgroup of which it's a member in each hierarchy, but this
 175   can be determined by following pointers through the
 176   cgroup_subsys_state objects. This is because accessing the
 177   subsystem state is something that's expected to happen frequently
 178   and in performance-critical code, whereas operations that require a
 179   task's actual cgroup assignments (in particular, moving between
 180   cgroups) are less common. A linked list runs through the cg_list
 181   field of each task_struct using the css_set, anchored at
 182   css_set->tasks.
 183
 184 - A cgroup hierarchy filesystem can be mounted  for browsing and
 185   manipulation from user space.
 186
 187 - You can list all the tasks (by pid) attached to any cgroup.
 188
 189The implementation of cgroups requires a few, simple hooks
 190into the rest of the kernel, none in performance critical paths:
 191
 192 - in init/main.c, to initialize the root cgroups and initial
 193   css_set at system boot.
 194
 195 - in fork and exit, to attach and detach a task from its css_set.
 196
 197In addition a new file system, of type "cgroup" may be mounted, to
 198enable browsing and modifying the cgroups presently known to the
 199kernel.  When mounting a cgroup hierarchy, you may specify a
 200comma-separated list of subsystems to mount as the filesystem mount
 201options.  By default, mounting the cgroup filesystem attempts to
 202mount a hierarchy containing all registered subsystems.
 203
 204If an active hierarchy with exactly the same set of subsystems already
 205exists, it will be reused for the new mount. If no existing hierarchy
 206matches, and any of the requested subsystems are in use in an existing
 207hierarchy, the mount will fail with -EBUSY. Otherwise, a new hierarchy
 208is activated, associated with the requested subsystems.
 209
 210It's not currently possible to bind a new subsystem to an active
 211cgroup hierarchy, or to unbind a subsystem from an active cgroup
 212hierarchy. This may be possible in future, but is fraught with nasty
 213error-recovery issues.
 214
 215When a cgroup filesystem is unmounted, if there are any
 216child cgroups created below the top-level cgroup, that hierarchy
 217will remain active even though unmounted; if there are no
 218child cgroups then the hierarchy will be deactivated.
 219
 220No new system calls are added for cgroups - all support for
 221querying and modifying cgroups is via this cgroup file system.
 222
 223Each task under /proc has an added file named 'cgroup' displaying,
 224for each active hierarchy, the subsystem names and the cgroup name
 225as the path relative to the root of the cgroup file system.
 226
 227Each cgroup is represented by a directory in the cgroup file system
 228containing the following files describing that cgroup:
 229
 230 - tasks: list of tasks (by pid) attached to that cgroup
 231 - notify_on_release flag: run the release agent on exit?
 232 - release_agent: the path to use for release notifications (this file
 233   exists in the top cgroup only)
 234
 235Other subsystems such as cpusets may add additional files in each
 236cgroup dir.
 237
 238New cgroups are created using the mkdir system call or shell
 239command.  The properties of a cgroup, such as its flags, are
 240modified by writing to the appropriate file in that cgroups
 241directory, as listed above.
 242
 243The named hierarchical structure of nested cgroups allows partitioning
 244a large system into nested, dynamically changeable, "soft-partitions".
 245
 246The attachment of each task, automatically inherited at fork by any
 247children of that task, to a cgroup allows organizing the work load
 248on a system into related sets of tasks.  A task may be re-attached to
 249any other cgroup, if allowed by the permissions on the necessary
 250cgroup file system directories.
 251
 252When a task is moved from one cgroup to another, it gets a new
 253css_set pointer - if there's an already existing css_set with the
 254desired collection of cgroups then that group is reused, else a new
 255css_set is allocated. The appropriate existing css_set is located by
 256looking into a hash table.
 257
 258To allow access from a cgroup to the css_sets (and hence tasks)
 259that comprise it, a set of cg_cgroup_link objects form a lattice;
 260each cg_cgroup_link is linked into a list of cg_cgroup_links for
 261a single cgroup on its cgrp_link_list field, and a list of
 262cg_cgroup_links for a single css_set on its cg_link_list.
 263
 264Thus the set of tasks in a cgroup can be listed by iterating over
 265each css_set that references the cgroup, and sub-iterating over
 266each css_set's task set.
 267
 268The use of a Linux virtual file system (vfs) to represent the
 269cgroup hierarchy provides for a familiar permission and name space
 270for cgroups, with a minimum of additional kernel code.
 271
 2721.4 What does notify_on_release do ?
 273------------------------------------
 274
 275If the notify_on_release flag is enabled (1) in a cgroup, then
 276whenever the last task in the cgroup leaves (exits or attaches to
 277some other cgroup) and the last child cgroup of that cgroup
 278is removed, then the kernel runs the command specified by the contents
 279of the "release_agent" file in that hierarchy's root directory,
 280supplying the pathname (relative to the mount point of the cgroup
 281file system) of the abandoned cgroup.  This enables automatic
 282removal of abandoned cgroups.  The default value of
 283notify_on_release in the root cgroup at system boot is disabled
 284(0).  The default value of other cgroups at creation is the current
 285value of their parents notify_on_release setting. The default value of
 286a cgroup hierarchy's release_agent path is empty.
 287
 2881.5 How do I use cgroups ?
 289--------------------------
 290
 291To start a new job that is to be contained within a cgroup, using
 292the "cpuset" cgroup subsystem, the steps are something like:
 293
 294 1) mkdir /dev/cgroup
 295 2) mount -t cgroup -ocpuset cpuset /dev/cgroup
 296 3) Create the new cgroup by doing mkdir's and write's (or echo's) in
 297    the /dev/cgroup virtual file system.
 298 4) Start a task that will be the "founding father" of the new job.
 299 5) Attach that task to the new cgroup by writing its pid to the
 300    /dev/cgroup tasks file for that cgroup.
 301 6) fork, exec or clone the job tasks from this founding father task.
 302
 303For example, the following sequence of commands will setup a cgroup
 304named "Charlie", containing just CPUs 2 and 3, and Memory Node 1,
 305and then start a subshell 'sh' in that cgroup:
 306
 307  mount -t cgroup cpuset -ocpuset /dev/cgroup
 308  cd /dev/cgroup
 309  mkdir Charlie
 310  cd Charlie
 311  /bin/echo 2-3 > cpuset.cpus
 312  /bin/echo 1 > cpuset.mems
 313  /bin/echo $$ > tasks
 314  sh
 315  # The subshell 'sh' is now running in cgroup Charlie
 316  # The next line should display '/Charlie'
 317  cat /proc/self/cgroup
 318
 3192. Usage Examples and Syntax
 320============================
 321
 3222.1 Basic Usage
 323---------------
 324
 325Creating, modifying, using the cgroups can be done through the cgroup
 326virtual filesystem.
 327
 328To mount a cgroup hierarchy with all available subsystems, type:
 329# mount -t cgroup xxx /dev/cgroup
 330
 331The "xxx" is not interpreted by the cgroup code, but will appear in
 332/proc/mounts so may be any useful identifying string that you like.
 333
 334To mount a cgroup hierarchy with just the cpuset and numtasks
 335subsystems, type:
 336# mount -t cgroup -o cpuset,memory hier1 /dev/cgroup
 337
 338To change the set of subsystems bound to a mounted hierarchy, just
 339remount with different options:
 340# mount -o remount,cpuset,ns hier1 /dev/cgroup
 341
 342Now memory is removed from the hierarchy and ns is added.
 343
 344Note this will add ns to the hierarchy but won't remove memory or
 345cpuset, because the new options are appended to the old ones:
 346# mount -o remount,ns /dev/cgroup
 347
 348To Specify a hierarchy's release_agent:
 349# mount -t cgroup -o cpuset,release_agent="/sbin/cpuset_release_agent" \
 350  xxx /dev/cgroup
 351
 352Note that specifying 'release_agent' more than once will return failure.
 353
 354Note that changing the set of subsystems is currently only supported
 355when the hierarchy consists of a single (root) cgroup. Supporting
 356the ability to arbitrarily bind/unbind subsystems from an existing
 357cgroup hierarchy is intended to be implemented in the future.
 358
 359Then under /dev/cgroup you can find a tree that corresponds to the
 360tree of the cgroups in the system. For instance, /dev/cgroup
 361is the cgroup that holds the whole system.
 362
 363If you want to change the value of release_agent:
 364# echo "/sbin/new_release_agent" > /dev/cgroup/release_agent
 365
 366It can also be changed via remount.
 367
 368If you want to create a new cgroup under /dev/cgroup:
 369# cd /dev/cgroup
 370# mkdir my_cgroup
 371
 372Now you want to do something with this cgroup.
 373# cd my_cgroup
 374
 375In this directory you can find several files:
 376# ls
 377notify_on_release tasks
 378(plus whatever files added by the attached subsystems)
 379
 380Now attach your shell to this cgroup:
 381# /bin/echo $$ > tasks
 382
 383You can also create cgroups inside your cgroup by using mkdir in this
 384directory.
 385# mkdir my_sub_cs
 386
 387To remove a cgroup, just use rmdir:
 388# rmdir my_sub_cs
 389
 390This will fail if the cgroup is in use (has cgroups inside, or
 391has processes attached, or is held alive by other subsystem-specific
 392reference).
 393
 3942.2 Attaching processes
 395-----------------------
 396
 397# /bin/echo PID > tasks
 398
 399Note that it is PID, not PIDs. You can only attach ONE task at a time.
 400If you have several tasks to attach, you have to do it one after another:
 401
 402# /bin/echo PID1 > tasks
 403# /bin/echo PID2 > tasks
 404        ...
 405# /bin/echo PIDn > tasks
 406
 407You can attach the current shell task by echoing 0:
 408
 409# echo 0 > tasks
 410
 4113. Kernel API
 412=============
 413
 4143.1 Overview
 415------------
 416
 417Each kernel subsystem that wants to hook into the generic cgroup
 418system needs to create a cgroup_subsys object. This contains
 419various methods, which are callbacks from the cgroup system, along
 420with a subsystem id which will be assigned by the cgroup system.
 421
 422Other fields in the cgroup_subsys object include:
 423
 424- subsys_id: a unique array index for the subsystem, indicating which
 425  entry in cgroup->subsys[] this subsystem should be managing.
 426
 427- name: should be initialized to a unique subsystem name. Should be
 428  no longer than MAX_CGROUP_TYPE_NAMELEN.
 429
 430- early_init: indicate if the subsystem needs early initialization
 431  at system boot.
 432
 433Each cgroup object created by the system has an array of pointers,
 434indexed by subsystem id; this pointer is entirely managed by the
 435subsystem; the generic cgroup code will never touch this pointer.
 436
 4373.2 Synchronization
 438-------------------
 439
 440There is a global mutex, cgroup_mutex, used by the cgroup
 441system. This should be taken by anything that wants to modify a
 442cgroup. It may also be taken to prevent cgroups from being
 443modified, but more specific locks may be more appropriate in that
 444situation.
 445
 446See kernel/cgroup.c for more details.
 447
 448Subsystems can take/release the cgroup_mutex via the functions
 449cgroup_lock()/cgroup_unlock().
 450
 451Accessing a task's cgroup pointer may be done in the following ways:
 452- while holding cgroup_mutex
 453- while holding the task's alloc_lock (via task_lock())
 454- inside an rcu_read_lock() section via rcu_dereference()
 455
 4563.3 Subsystem API
 457-----------------
 458
 459Each subsystem should:
 460
 461- add an entry in linux/cgroup_subsys.h
 462- define a cgroup_subsys object called <name>_subsys
 463
 464Each subsystem may export the following methods. The only mandatory
 465methods are create/destroy. Any others that are null are presumed to
 466be successful no-ops.
 467
 468struct cgroup_subsys_state *create(struct cgroup_subsys *ss,
 469                                   struct cgroup *cgrp)
 470(cgroup_mutex held by caller)
 471
 472Called to create a subsystem state object for a cgroup. The
 473subsystem should allocate its subsystem state object for the passed
 474cgroup, returning a pointer to the new object on success or a
 475negative error code. On success, the subsystem pointer should point to
 476a structure of type cgroup_subsys_state (typically embedded in a
 477larger subsystem-specific object), which will be initialized by the
 478cgroup system. Note that this will be called at initialization to
 479create the root subsystem state for this subsystem; this case can be
 480identified by the passed cgroup object having a NULL parent (since
 481it's the root of the hierarchy) and may be an appropriate place for
 482initialization code.
 483
 484void destroy(struct cgroup_subsys *ss, struct cgroup *cgrp)
 485(cgroup_mutex held by caller)
 486
 487The cgroup system is about to destroy the passed cgroup; the subsystem
 488should do any necessary cleanup and free its subsystem state
 489object. By the time this method is called, the cgroup has already been
 490unlinked from the file system and from the child list of its parent;
 491cgroup->parent is still valid. (Note - can also be called for a
 492newly-created cgroup if an error occurs after this subsystem's
 493create() method has been called for the new cgroup).
 494
 495int pre_destroy(struct cgroup_subsys *ss, struct cgroup *cgrp);
 496
 497Called before checking the reference count on each subsystem. This may
 498be useful for subsystems which have some extra references even if
 499there are not tasks in the cgroup. If pre_destroy() returns error code,
 500rmdir() will fail with it. From this behavior, pre_destroy() can be
 501called multiple times against a cgroup.
 502
 503int can_attach(struct cgroup_subsys *ss, struct cgroup *cgrp,
 504               struct task_struct *task)
 505(cgroup_mutex held by caller)
 506
 507Called prior to moving a task into a cgroup; if the subsystem
 508returns an error, this will abort the attach operation.  If a NULL
 509task is passed, then a successful result indicates that *any*
 510unspecified task can be moved into the cgroup. Note that this isn't
 511called on a fork. If this method returns 0 (success) then this should
 512remain valid while the caller holds cgroup_mutex.
 513
 514void attach(struct cgroup_subsys *ss, struct cgroup *cgrp,
 515            struct cgroup *old_cgrp, struct task_struct *task)
 516(cgroup_mutex held by caller)
 517
 518Called after the task has been attached to the cgroup, to allow any
 519post-attachment activity that requires memory allocations or blocking.
 520
 521void fork(struct cgroup_subsy *ss, struct task_struct *task)
 522
 523Called when a task is forked into a cgroup.
 524
 525void exit(struct cgroup_subsys *ss, struct task_struct *task)
 526
 527Called during task exit.
 528
 529int populate(struct cgroup_subsys *ss, struct cgroup *cgrp)
 530(cgroup_mutex held by caller)
 531
 532Called after creation of a cgroup to allow a subsystem to populate
 533the cgroup directory with file entries.  The subsystem should make
 534calls to cgroup_add_file() with objects of type cftype (see
 535include/linux/cgroup.h for details).  Note that although this
 536method can return an error code, the error code is currently not
 537always handled well.
 538
 539void post_clone(struct cgroup_subsys *ss, struct cgroup *cgrp)
 540(cgroup_mutex held by caller)
 541
 542Called at the end of cgroup_clone() to do any parameter
 543initialization which might be required before a task could attach.  For
 544example in cpusets, no task may attach before 'cpus' and 'mems' are set
 545up.
 546
 547void bind(struct cgroup_subsys *ss, struct cgroup *root)
 548(cgroup_mutex and ss->hierarchy_mutex held by caller)
 549
 550Called when a cgroup subsystem is rebound to a different hierarchy
 551and root cgroup. Currently this will only involve movement between
 552the default hierarchy (which never has sub-cgroups) and a hierarchy
 553that is being created/destroyed (and hence has no sub-cgroups).
 554
 5554. Questions
 556============
 557
 558Q: what's up with this '/bin/echo' ?
 559A: bash's builtin 'echo' command does not check calls to write() against
 560   errors. If you use it in the cgroup file system, you won't be
 561   able to tell whether a command succeeded or failed.
 562
 563Q: When I attach processes, only the first of the line gets really attached !
 564A: We can only return one error code per call to write(). So you should also
 565   put only ONE pid.
 566
 567
lxr.linux.no kindly hosted by Redpill Linpro AS, provider of Linux consulting and operations services since 1995.