1 Memory Resource Controller
3 NOTE: The Memory Resource Controller has been generically been referred
4 to as the memory controller in this document. Do not confuse memory controller
5 used here with the memory controller that is used in hardware.
9 a. Enable control of Anonymous, Page Cache (mapped and unmapped) and
10 Swap Cache memory pages.
11 b. The infrastructure allows easy addition of other types of memory to control
12 c. Provides *zero overhead* for non memory controller users
13 d. Provides a double LRU: global memory pressure causes reclaim from the
14 global LRU; a cgroup on hitting a limit, reclaims from the per
17 Benefits and Purpose of the memory controller
19 The memory controller isolates the memory behaviour of a group of tasks
20 from the rest of the system. The article on LWN [12] mentions some probable
21 uses of the memory controller. The memory controller can be used to
23 a. Isolate an application or a group of applications
24 Memory hungry applications can be isolated and limited to a smaller
26 b. Create a cgroup with limited amount of memory, this can be used
27 as a good alternative to booting with mem=XXXX.
28 c. Virtualization solutions can control the amount of memory they want
29 to assign to a virtual machine instance.
30 d. A CD/DVD burner could control the amount of memory used by the
31 rest of the system to ensure that burning does not fail due to lack
33 e. There are several other use cases, find one or use the controller just
34 for fun (to learn and hack on the VM subsystem).
38 The memory controller has a long history. A request for comments for the memory
39 controller was posted by Balbir Singh [1]. At the time the RFC was posted
40 there were several implementations for memory control. The goal of the
41 RFC was to build consensus and agreement for the minimal features required
42 for memory control. The first RSS controller was posted by Balbir Singh[2]
43 in Feb 2007. Pavel Emelianov [3][4][5] has since posted three versions of the
44 RSS controller. At OLS, at the resource management BoF, everyone suggested
45 that we handle both page cache and RSS together. Another request was raised
46 to allow user space handling of OOM. The current memory controller is
47 at version 6; it combines both mapped (RSS) and unmapped Page
52 Memory is a unique resource in the sense that it is present in a limited
53 amount. If a task requires a lot of CPU processing, the task can spread
54 its processing over a period of hours, days, months or years, but with
55 memory, the same physical memory needs to be reused to accomplish the task.
57 The memory controller implementation has been divided into phases. These
61 2. mlock(2) controller
62 3. Kernel user memory accounting and slab control
63 4. user mappings length controller
65 The memory controller is the first controller developed.
69 The core of the design is a counter called the res_counter. The res_counter
70 tracks the current memory usage and limit of the group of processes associated
71 with the controller. Each cgroup has a memory controller specific data
72 structure (mem_cgroup) associated with it.
76 +--------------------+
79 +--------------------+
82 +---------------+ | +---------------+
83 | mm_struct | |.... | mm_struct |
85 +---------------+ | +---------------+
89 +---------------+ +------+--------+
90 | page +----------> page_cgroup|
92 +---------------+ +---------------+
94 (Figure 1: Hierarchy of Accounting)
97 Figure 1 shows the important aspects of the controller
99 1. Accounting happens per cgroup
100 2. Each mm_struct knows about which cgroup it belongs to
101 3. Each page has a pointer to the page_cgroup, which in turn knows the
104 The accounting is done as follows: mem_cgroup_charge() is invoked to setup
105 the necessary data structures and check if the cgroup that is being charged
106 is over its limit. If it is then reclaim is invoked on the cgroup.
107 More details can be found in the reclaim section of this document.
108 If everything goes well, a page meta-data-structure called page_cgroup is
109 allocated and associated with the page. This routine also adds the page to
112 2.2.1 Accounting details
114 All mapped anon pages (RSS) and cache pages (Page Cache) are accounted.
115 (some pages which never be reclaimable and will not be on global LRU
116 are not accounted. we just accounts pages under usual vm management.)
118 RSS pages are accounted at page_fault unless they've already been accounted
119 for earlier. A file page will be accounted for as Page Cache when it's
120 inserted into inode (radix-tree). While it's mapped into the page tables of
121 processes, duplicate accounting is carefully avoided.
123 A RSS page is unaccounted when it's fully unmapped. A PageCache page is
124 unaccounted when it's removed from radix-tree.
126 At page migration, accounting information is kept.
128 Note: we just account pages-on-lru because our purpose is to control amount
129 of used pages. not-on-lru pages are tend to be out-of-control from vm view.
131 2.3 Shared Page Accounting
133 Shared pages are accounted on the basis of the first touch approach. The
134 cgroup that first touches a page is accounted for the page. The principle
135 behind this approach is that a cgroup that aggressively uses a shared
136 page will eventually get charged for it (once it is uncharged from
137 the cgroup that brought it in -- this will happen on memory pressure).
139 Exception: If CONFIG_CGROUP_CGROUP_MEM_RES_CTLR_SWAP is not used..
140 When you do swapoff and make swapped-out pages of shmem(tmpfs) to
141 be backed into memory in force, charges for pages are accounted against the
142 caller of swapoff rather than the users of shmem.
145 2.4 Swap Extension (CONFIG_CGROUP_MEM_RES_CTLR_SWAP)
146 Swap Extension allows you to record charge for swap. A swapped-in page is
147 charged back to original page allocator if possible.
149 When swap is accounted, following files are added.
150 - memory.memsw.usage_in_bytes.
151 - memory.memsw.limit_in_bytes.
153 usage of mem+swap is limited by memsw.limit_in_bytes.
155 * why 'mem+swap' rather than swap.
156 The global LRU(kswapd) can swap out arbitrary pages. Swap-out means
157 to move account from memory to swap...there is no change in usage of
158 mem+swap. In other words, when we want to limit the usage of swap without
159 affecting global LRU, mem+swap limit is better than just limiting swap from
162 * What happens when a cgroup hits memory.memsw.limit_in_bytes
163 When a cgroup his memory.memsw.limit_in_bytes, it's useless to do swap-out
164 in this cgroup. Then, swap-out will not be done by cgroup routine and file
165 caches are dropped. But as mentioned above, global LRU can do swapout memory
166 from it for sanity of the system's memory management state. You can't forbid
171 Each cgroup maintains a per cgroup LRU that consists of an active
172 and inactive list. When a cgroup goes over its limit, we first try
173 to reclaim memory from the cgroup so as to make space for the new
174 pages that the cgroup has touched. If the reclaim is unsuccessful,
175 an OOM routine is invoked to select and kill the bulkiest task in the
178 The reclaim algorithm has not been modified for cgroups, except that
179 pages that are selected for reclaiming come from the per cgroup LRU
184 The memory controller uses the following hierarchy
186 1. zone->lru_lock is used for selecting pages to be isolated
187 2. mem->per_zone->lru_lock protects the per cgroup LRU (per zone)
188 3. lock_page_cgroup() is used to protect page->page_cgroup
194 a. Enable CONFIG_CGROUPS
195 b. Enable CONFIG_RESOURCE_COUNTERS
196 c. Enable CONFIG_CGROUP_MEM_RES_CTLR
198 1. Prepare the cgroups
200 # mount -t cgroup none /cgroups -o memory
202 2. Make the new group and move bash into it
204 # echo $$ > /cgroups/0/tasks
206 Since now we're in the 0 cgroup,
207 We can alter the memory limit:
208 # echo 4M > /cgroups/0/memory.limit_in_bytes
210 NOTE: We can use a suffix (k, K, m, M, g or G) to indicate values in kilo,
213 # cat /cgroups/0/memory.limit_in_bytes
216 NOTE: The interface has now changed to display the usage in bytes
219 We can check the usage:
220 # cat /cgroups/0/memory.usage_in_bytes
223 A successful write to this file does not guarantee a successful set of
224 this limit to the value written into the file. This can be due to a
225 number of factors, such as rounding up to page boundaries or the total
226 availability of memory on the system. The user is required to re-read
227 this file after a write to guarantee the value committed by the kernel.
229 # echo 1 > memory.limit_in_bytes
230 # cat memory.limit_in_bytes
233 The memory.failcnt field gives the number of times that the cgroup limit was
236 The memory.stat file gives accounting information. Now, the number of
237 caches, RSS and Active pages/Inactive pages are shown.
241 Balbir posted lmbench, AIM9, LTP and vmmstress results [10] and [11].
242 Apart from that v6 has been tested with several applications and regular
243 daily use. The controller has also been tested on the PPC64, x86_64 and
248 Sometimes a user might find that the application under a cgroup is
249 terminated. There are several causes for this:
251 1. The cgroup limit is too low (just too low to do anything useful)
252 2. The user is using anonymous memory and swap is turned off or too low
254 A sync followed by echo 1 > /proc/sys/vm/drop_caches will help get rid of
255 some of the pages cached in the cgroup (page cache pages).
259 When a task migrates from one cgroup to another, it's charge is not
260 carried forward. The pages allocated from the original cgroup still
261 remain charged to it, the charge is dropped when the page is freed or
264 4.3 Removing a cgroup
266 A cgroup can be removed by rmdir, but as discussed in sections 4.1 and 4.2, a
267 cgroup might have some charge associated with it, even though all
268 tasks have migrated away from it.
269 Such charges are freed(at default) or moved to its parent. When moved,
270 both of RSS and CACHES are moved to parent.
271 If both of them are busy, rmdir() returns -EBUSY. See 5.1 Also.
273 Charges recorded in swap information is not updated at removal of cgroup.
274 Recorded information is discarded and a cgroup which uses swap (swapcache)
275 will be charged as a new owner of it.
281 memory.force_empty interface is provided to make cgroup's memory usage empty.
282 You can use this interface only when the cgroup has no tasks.
283 When writing anything to this
285 # echo 0 > memory.force_empty
287 Almost all pages tracked by this memcg will be unmapped and freed. Some of
288 pages cannot be freed because it's locked or in-use. Such pages are moved
289 to parent and this cgroup will be empty. But this may return -EBUSY in
292 Typical use case of this interface is that calling this before rmdir().
293 Because rmdir() moves all pages to parent, some out-of-use page caches can be
294 moved to the parent. If you want to avoid that, force_empty will be useful.
298 memory.stat file includes following statistics
300 cache - # of bytes of page cache memory.
301 rss - # of bytes of anonymous and swap cache memory.
302 pgpgin - # of pages paged in (equivalent to # of charging events).
303 pgpgout - # of pages paged out (equivalent to # of uncharging events).
304 active_anon - # of bytes of anonymous and swap cache memory on active
306 inactive_anon - # of bytes of anonymous memory and swap cache memory on
308 active_file - # of bytes of file-backed memory on active lru list.
309 inactive_file - # of bytes of file-backed memory on inactive lru list.
310 unevictable - # of bytes of memory that cannot be reclaimed (mlocked etc).
312 The following additional stats are dependent on CONFIG_DEBUG_VM.
314 inactive_ratio - VM internal parameter. (see mm/page_alloc.c)
315 recent_rotated_anon - VM internal parameter. (see mm/vmscan.c)
316 recent_rotated_file - VM internal parameter. (see mm/vmscan.c)
317 recent_scanned_anon - VM internal parameter. (see mm/vmscan.c)
318 recent_scanned_file - VM internal parameter. (see mm/vmscan.c)
321 recent_rotated means recent frequency of lru rotation.
322 recent_scanned means recent # of scans to lru.
323 showing for better debug please see the code for meanings.
326 Only anonymous and swap cache memory is listed as part of 'rss' stat.
327 This should not be confused with the true 'resident set size' or the
328 amount of physical memory used by the cgroup. Per-cgroup rss
329 accounting is not done yet.
332 Similar to /proc/sys/vm/swappiness, but affecting a hierarchy of groups only.
334 Following cgroups' swapiness can't be changed.
335 - root cgroup (uses /proc/sys/vm/swappiness).
336 - a cgroup which uses hierarchy and it has child cgroup.
337 - a cgroup which uses hierarchy and not the root of hierarchy.
342 The memory controller supports a deep hierarchy and hierarchical accounting.
343 The hierarchy is created by creating the appropriate cgroups in the
344 cgroup filesystem. Consider for example, the following cgroup filesystem
355 In the diagram above, with hierarchical accounting enabled, all memory
356 usage of e, is accounted to its ancestors up until the root (i.e, c and root),
357 that has memory.use_hierarchy enabled. If one of the ancestors goes over its
358 limit, the reclaim algorithm reclaims from the tasks in the ancestor and the
359 children of the ancestor.
361 6.1 Enabling hierarchical accounting and reclaim
363 The memory controller by default disables the hierarchy feature. Support
364 can be enabled by writing 1 to memory.use_hierarchy file of the root cgroup
366 # echo 1 > memory.use_hierarchy
368 The feature can be disabled by
370 # echo 0 > memory.use_hierarchy
372 NOTE1: Enabling/disabling will fail if the cgroup already has other
373 cgroups created below it.
375 NOTE2: This feature can be enabled/disabled per subtree.
379 1. Add support for accounting huge pages (as a separate controller)
380 2. Make per-cgroup scanner reclaim not-shared pages first
381 3. Teach controller to account for shared-pages
382 4. Start reclamation in the background when the limit is
383 not yet hit but the usage is getting closer
387 Overall, the memory controller has been a stable controller and has been
388 commented and discussed quite extensively in the community.
392 1. Singh, Balbir. RFC: Memory Controller, http://lwn.net/Articles/206697/
393 2. Singh, Balbir. Memory Controller (RSS Control),
394 http://lwn.net/Articles/222762/
395 3. Emelianov, Pavel. Resource controllers based on process cgroups
396 http://lkml.org/lkml/2007/3/6/198
397 4. Emelianov, Pavel. RSS controller based on process cgroups (v2)
398 http://lkml.org/lkml/2007/4/9/78
399 5. Emelianov, Pavel. RSS controller based on process cgroups (v3)
400 http://lkml.org/lkml/2007/5/30/244
401 6. Menage, Paul. Control Groups v10, http://lwn.net/Articles/236032/
402 7. Vaidyanathan, Srinivasan, Control Groups: Pagecache accounting and control
403 subsystem (v3), http://lwn.net/Articles/235534/
404 8. Singh, Balbir. RSS controller v2 test results (lmbench),
405 http://lkml.org/lkml/2007/5/17/232
406 9. Singh, Balbir. RSS controller v2 AIM9 results
407 http://lkml.org/lkml/2007/5/18/1
408 10. Singh, Balbir. Memory controller v6 test results,
409 http://lkml.org/lkml/2007/8/19/36
410 11. Singh, Balbir. Memory controller introduction (v6),
411 http://lkml.org/lkml/2007/8/17/69
412 12. Corbet, Jonathan, Controlling memory use in cgroups,
413 http://lwn.net/Articles/243795/