Based on kernel version 4.0. Page generated on 2015-04-14 21:20 EST.
1 Memory Resource Controller 2 3 NOTE: This document is hopelessly outdated and it asks for a complete 4 rewrite. It still contains a useful information so we are keeping it 5 here but make sure to check the current code if you need a deeper 6 understanding. 7 8 NOTE: The Memory Resource Controller has generically been referred to as the 9 memory controller in this document. Do not confuse memory controller 10 used here with the memory controller that is used in hardware. 11 12 (For editors) 13 In this document: 14 When we mention a cgroup (cgroupfs's directory) with memory controller, 15 we call it "memory cgroup". When you see git-log and source code, you'll 16 see patch's title and function names tend to use "memcg". 17 In this document, we avoid using it. 18 19 Benefits and Purpose of the memory controller 20 21 The memory controller isolates the memory behaviour of a group of tasks 22 from the rest of the system. The article on LWN  mentions some probable 23 uses of the memory controller. The memory controller can be used to 24 25 a. Isolate an application or a group of applications 26 Memory-hungry applications can be isolated and limited to a smaller 27 amount of memory. 28 b. Create a cgroup with a limited amount of memory; this can be used 29 as a good alternative to booting with mem=XXXX. 30 c. Virtualization solutions can control the amount of memory they want 31 to assign to a virtual machine instance. 32 d. A CD/DVD burner could control the amount of memory used by the 33 rest of the system to ensure that burning does not fail due to lack 34 of available memory. 35 e. There are several other use cases; find one or use the controller just 36 for fun (to learn and hack on the VM subsystem). 37 38 Current Status: linux-2.6.34-mmotm(development version of 2010/April) 39 40 Features: 41 - accounting anonymous pages, file caches, swap caches usage and limiting them. 42 - pages are linked to per-memcg LRU exclusively, and there is no global LRU. 43 - optionally, memory+swap usage can be accounted and limited. 44 - hierarchical accounting 45 - soft limit 46 - moving (recharging) account at moving a task is selectable. 47 - usage threshold notifier 48 - memory pressure notifier 49 - oom-killer disable knob and oom-notifier 50 - Root cgroup has no limit controls. 51 52 Kernel memory support is a work in progress, and the current version provides 53 basically functionality. (See Section 2.7) 54 55 Brief summary of control files. 56 57 tasks # attach a task(thread) and show list of threads 58 cgroup.procs # show list of processes 59 cgroup.event_control # an interface for event_fd() 60 memory.usage_in_bytes # show current usage for memory 61 (See 5.5 for details) 62 memory.memsw.usage_in_bytes # show current usage for memory+Swap 63 (See 5.5 for details) 64 memory.limit_in_bytes # set/show limit of memory usage 65 memory.memsw.limit_in_bytes # set/show limit of memory+Swap usage 66 memory.failcnt # show the number of memory usage hits limits 67 memory.memsw.failcnt # show the number of memory+Swap hits limits 68 memory.max_usage_in_bytes # show max memory usage recorded 69 memory.memsw.max_usage_in_bytes # show max memory+Swap usage recorded 70 memory.soft_limit_in_bytes # set/show soft limit of memory usage 71 memory.stat # show various statistics 72 memory.use_hierarchy # set/show hierarchical account enabled 73 memory.force_empty # trigger forced move charge to parent 74 memory.pressure_level # set memory pressure notifications 75 memory.swappiness # set/show swappiness parameter of vmscan 76 (See sysctl's vm.swappiness) 77 memory.move_charge_at_immigrate # set/show controls of moving charges 78 memory.oom_control # set/show oom controls. 79 memory.numa_stat # show the number of memory usage per numa node 80 81 memory.kmem.limit_in_bytes # set/show hard limit for kernel memory 82 memory.kmem.usage_in_bytes # show current kernel memory allocation 83 memory.kmem.failcnt # show the number of kernel memory usage hits limits 84 memory.kmem.max_usage_in_bytes # show max kernel memory usage recorded 85 86 memory.kmem.tcp.limit_in_bytes # set/show hard limit for tcp buf memory 87 memory.kmem.tcp.usage_in_bytes # show current tcp buf memory allocation 88 memory.kmem.tcp.failcnt # show the number of tcp buf memory usage hits limits 89 memory.kmem.tcp.max_usage_in_bytes # show max tcp buf memory usage recorded 90 91 1. History 92 93 The memory controller has a long history. A request for comments for the memory 94 controller was posted by Balbir Singh . At the time the RFC was posted 95 there were several implementations for memory control. The goal of the 96 RFC was to build consensus and agreement for the minimal features required 97 for memory control. The first RSS controller was posted by Balbir Singh 98 in Feb 2007. Pavel Emelianov  has since posted three versions of the 99 RSS controller. At OLS, at the resource management BoF, everyone suggested 100 that we handle both page cache and RSS together. Another request was raised 101 to allow user space handling of OOM. The current memory controller is 102 at version 6; it combines both mapped (RSS) and unmapped Page 103 Cache Control . 104 105 2. Memory Control 106 107 Memory is a unique resource in the sense that it is present in a limited 108 amount. If a task requires a lot of CPU processing, the task can spread 109 its processing over a period of hours, days, months or years, but with 110 memory, the same physical memory needs to be reused to accomplish the task. 111 112 The memory controller implementation has been divided into phases. These 113 are: 114 115 1. Memory controller 116 2. mlock(2) controller 117 3. Kernel user memory accounting and slab control 118 4. user mappings length controller 119 120 The memory controller is the first controller developed. 121 122 2.1. Design 123 124 The core of the design is a counter called the page_counter. The 125 page_counter tracks the current memory usage and limit of the group of 126 processes associated with the controller. Each cgroup has a memory controller 127 specific data structure (mem_cgroup) associated with it. 128 129 2.2. Accounting 130 131 +--------------------+ 132 | mem_cgroup | 133 | (page_counter) | 134 +--------------------+ 135 / ^ \ 136 / | \ 137 +---------------+ | +---------------+ 138 | mm_struct | |.... | mm_struct | 139 | | | | | 140 +---------------+ | +---------------+ 141 | 142 + --------------+ 143 | 144 +---------------+ +------+--------+ 145 | page +----------> page_cgroup| 146 | | | | 147 +---------------+ +---------------+ 148 149 (Figure 1: Hierarchy of Accounting) 150 151 152 Figure 1 shows the important aspects of the controller 153 154 1. Accounting happens per cgroup 155 2. Each mm_struct knows about which cgroup it belongs to 156 3. Each page has a pointer to the page_cgroup, which in turn knows the 157 cgroup it belongs to 158 159 The accounting is done as follows: mem_cgroup_charge_common() is invoked to 160 set up the necessary data structures and check if the cgroup that is being 161 charged is over its limit. If it is, then reclaim is invoked on the cgroup. 162 More details can be found in the reclaim section of this document. 163 If everything goes well, a page meta-data-structure called page_cgroup is 164 updated. page_cgroup has its own LRU on cgroup. 165 (*) page_cgroup structure is allocated at boot/memory-hotplug time. 166 167 2.2.1 Accounting details 168 169 All mapped anon pages (RSS) and cache pages (Page Cache) are accounted. 170 Some pages which are never reclaimable and will not be on the LRU 171 are not accounted. We just account pages under usual VM management. 172 173 RSS pages are accounted at page_fault unless they've already been accounted 174 for earlier. A file page will be accounted for as Page Cache when it's 175 inserted into inode (radix-tree). While it's mapped into the page tables of 176 processes, duplicate accounting is carefully avoided. 177 178 An RSS page is unaccounted when it's fully unmapped. A PageCache page is 179 unaccounted when it's removed from radix-tree. Even if RSS pages are fully 180 unmapped (by kswapd), they may exist as SwapCache in the system until they 181 are really freed. Such SwapCaches are also accounted. 182 A swapped-in page is not accounted until it's mapped. 183 184 Note: The kernel does swapin-readahead and reads multiple swaps at once. 185 This means swapped-in pages may contain pages for other tasks than a task 186 causing page fault. So, we avoid accounting at swap-in I/O. 187 188 At page migration, accounting information is kept. 189 190 Note: we just account pages-on-LRU because our purpose is to control amount 191 of used pages; not-on-LRU pages tend to be out-of-control from VM view. 192 193 2.3 Shared Page Accounting 194 195 Shared pages are accounted on the basis of the first touch approach. The 196 cgroup that first touches a page is accounted for the page. The principle 197 behind this approach is that a cgroup that aggressively uses a shared 198 page will eventually get charged for it (once it is uncharged from 199 the cgroup that brought it in -- this will happen on memory pressure). 200 201 But see section 8.2: when moving a task to another cgroup, its pages may 202 be recharged to the new cgroup, if move_charge_at_immigrate has been chosen. 203 204 Exception: If CONFIG_MEMCG_SWAP is not used. 205 When you do swapoff and make swapped-out pages of shmem(tmpfs) to 206 be backed into memory in force, charges for pages are accounted against the 207 caller of swapoff rather than the users of shmem. 208 209 2.4 Swap Extension (CONFIG_MEMCG_SWAP) 210 211 Swap Extension allows you to record charge for swap. A swapped-in page is 212 charged back to original page allocator if possible. 213 214 When swap is accounted, following files are added. 215 - memory.memsw.usage_in_bytes. 216 - memory.memsw.limit_in_bytes. 217 218 memsw means memory+swap. Usage of memory+swap is limited by 219 memsw.limit_in_bytes. 220 221 Example: Assume a system with 4G of swap. A task which allocates 6G of memory 222 (by mistake) under 2G memory limitation will use all swap. 223 In this case, setting memsw.limit_in_bytes=3G will prevent bad use of swap. 224 By using the memsw limit, you can avoid system OOM which can be caused by swap 225 shortage. 226 227 * why 'memory+swap' rather than swap. 228 The global LRU(kswapd) can swap out arbitrary pages. Swap-out means 229 to move account from memory to swap...there is no change in usage of 230 memory+swap. In other words, when we want to limit the usage of swap without 231 affecting global LRU, memory+swap limit is better than just limiting swap from 232 an OS point of view. 233 234 * What happens when a cgroup hits memory.memsw.limit_in_bytes 235 When a cgroup hits memory.memsw.limit_in_bytes, it's useless to do swap-out 236 in this cgroup. Then, swap-out will not be done by cgroup routine and file 237 caches are dropped. But as mentioned above, global LRU can do swapout memory 238 from it for sanity of the system's memory management state. You can't forbid 239 it by cgroup. 240 241 2.5 Reclaim 242 243 Each cgroup maintains a per cgroup LRU which has the same structure as 244 global VM. When a cgroup goes over its limit, we first try 245 to reclaim memory from the cgroup so as to make space for the new 246 pages that the cgroup has touched. If the reclaim is unsuccessful, 247 an OOM routine is invoked to select and kill the bulkiest task in the 248 cgroup. (See 10. OOM Control below.) 249 250 The reclaim algorithm has not been modified for cgroups, except that 251 pages that are selected for reclaiming come from the per-cgroup LRU 252 list. 253 254 NOTE: Reclaim does not work for the root cgroup, since we cannot set any 255 limits on the root cgroup. 256 257 Note2: When panic_on_oom is set to "2", the whole system will panic. 258 259 When oom event notifier is registered, event will be delivered. 260 (See oom_control section) 261 262 2.6 Locking 263 264 lock_page_cgroup()/unlock_page_cgroup() should not be called under 265 mapping->tree_lock. 266 267 Other lock order is following: 268 PG_locked. 269 mm->page_table_lock 270 zone->lru_lock 271 lock_page_cgroup. 272 In many cases, just lock_page_cgroup() is called. 273 per-zone-per-cgroup LRU (cgroup's private LRU) is just guarded by 274 zone->lru_lock, it has no lock of its own. 275 276 2.7 Kernel Memory Extension (CONFIG_MEMCG_KMEM) 277 278 WARNING: Current implementation lacks reclaim support. That means allocation 279 attempts will fail when close to the limit even if there are plenty of 280 kmem available for reclaim. That makes this option unusable in real 281 life so DO NOT SELECT IT unless for development purposes. 282 283 With the Kernel memory extension, the Memory Controller is able to limit 284 the amount of kernel memory used by the system. Kernel memory is fundamentally 285 different than user memory, since it can't be swapped out, which makes it 286 possible to DoS the system by consuming too much of this precious resource. 287 288 Kernel memory won't be accounted at all until limit on a group is set. This 289 allows for existing setups to continue working without disruption. The limit 290 cannot be set if the cgroup have children, or if there are already tasks in the 291 cgroup. Attempting to set the limit under those conditions will return -EBUSY. 292 When use_hierarchy == 1 and a group is accounted, its children will 293 automatically be accounted regardless of their limit value. 294 295 After a group is first limited, it will be kept being accounted until it 296 is removed. The memory limitation itself, can of course be removed by writing 297 -1 to memory.kmem.limit_in_bytes. In this case, kmem will be accounted, but not 298 limited. 299 300 Kernel memory limits are not imposed for the root cgroup. Usage for the root 301 cgroup may or may not be accounted. The memory used is accumulated into 302 memory.kmem.usage_in_bytes, or in a separate counter when it makes sense. 303 (currently only for tcp). 304 The main "kmem" counter is fed into the main counter, so kmem charges will 305 also be visible from the user counter. 306 307 Currently no soft limit is implemented for kernel memory. It is future work 308 to trigger slab reclaim when those limits are reached. 309 310 2.7.1 Current Kernel Memory resources accounted 311 312 * stack pages: every process consumes some stack pages. By accounting into 313 kernel memory, we prevent new processes from being created when the kernel 314 memory usage is too high. 315 316 * slab pages: pages allocated by the SLAB or SLUB allocator are tracked. A copy 317 of each kmem_cache is created every time the cache is touched by the first time 318 from inside the memcg. The creation is done lazily, so some objects can still be 319 skipped while the cache is being created. All objects in a slab page should 320 belong to the same memcg. This only fails to hold when a task is migrated to a 321 different memcg during the page allocation by the cache. 322 323 * sockets memory pressure: some sockets protocols have memory pressure 324 thresholds. The Memory Controller allows them to be controlled individually 325 per cgroup, instead of globally. 326 327 * tcp memory pressure: sockets memory pressure for the tcp protocol. 328 329 2.7.2 Common use cases 330 331 Because the "kmem" counter is fed to the main user counter, kernel memory can 332 never be limited completely independently of user memory. Say "U" is the user 333 limit, and "K" the kernel limit. There are three possible ways limits can be 334 set: 335 336 U != 0, K = unlimited: 337 This is the standard memcg limitation mechanism already present before kmem 338 accounting. Kernel memory is completely ignored. 339 340 U != 0, K < U: 341 Kernel memory is a subset of the user memory. This setup is useful in 342 deployments where the total amount of memory per-cgroup is overcommited. 343 Overcommiting kernel memory limits is definitely not recommended, since the 344 box can still run out of non-reclaimable memory. 345 In this case, the admin could set up K so that the sum of all groups is 346 never greater than the total memory, and freely set U at the cost of his 347 QoS. 348 349 U != 0, K >= U: 350 Since kmem charges will also be fed to the user counter and reclaim will be 351 triggered for the cgroup for both kinds of memory. This setup gives the 352 admin a unified view of memory, and it is also useful for people who just 353 want to track kernel memory usage. 354 355 3. User Interface 356 357 3.0. Configuration 358 359 a. Enable CONFIG_CGROUPS 360 b. Enable CONFIG_MEMCG 361 c. Enable CONFIG_MEMCG_SWAP (to use swap extension) 362 d. Enable CONFIG_MEMCG_KMEM (to use kmem extension) 363 364 3.1. Prepare the cgroups (see cgroups.txt, Why are cgroups needed?) 365 # mount -t tmpfs none /sys/fs/cgroup 366 # mkdir /sys/fs/cgroup/memory 367 # mount -t cgroup none /sys/fs/cgroup/memory -o memory 368 369 3.2. Make the new group and move bash into it 370 # mkdir /sys/fs/cgroup/memory/0 371 # echo $$ > /sys/fs/cgroup/memory/0/tasks 372 373 Since now we're in the 0 cgroup, we can alter the memory limit: 374 # echo 4M > /sys/fs/cgroup/memory/0/memory.limit_in_bytes 375 376 NOTE: We can use a suffix (k, K, m, M, g or G) to indicate values in kilo, 377 mega or gigabytes. (Here, Kilo, Mega, Giga are Kibibytes, Mebibytes, Gibibytes.) 378 379 NOTE: We can write "-1" to reset the *.limit_in_bytes(unlimited). 380 NOTE: We cannot set limits on the root cgroup any more. 381 382 # cat /sys/fs/cgroup/memory/0/memory.limit_in_bytes 383 4194304 384 385 We can check the usage: 386 # cat /sys/fs/cgroup/memory/0/memory.usage_in_bytes 387 1216512 388 389 A successful write to this file does not guarantee a successful setting of 390 this limit to the value written into the file. This can be due to a 391 number of factors, such as rounding up to page boundaries or the total 392 availability of memory on the system. The user is required to re-read 393 this file after a write to guarantee the value committed by the kernel. 394 395 # echo 1 > memory.limit_in_bytes 396 # cat memory.limit_in_bytes 397 4096 398 399 The memory.failcnt field gives the number of times that the cgroup limit was 400 exceeded. 401 402 The memory.stat file gives accounting information. Now, the number of 403 caches, RSS and Active pages/Inactive pages are shown. 404 405 4. Testing 406 407 For testing features and implementation, see memcg_test.txt. 408 409 Performance test is also important. To see pure memory controller's overhead, 410 testing on tmpfs will give you good numbers of small overheads. 411 Example: do kernel make on tmpfs. 412 413 Page-fault scalability is also important. At measuring parallel 414 page fault test, multi-process test may be better than multi-thread 415 test because it has noise of shared objects/status. 416 417 But the above two are testing extreme situations. 418 Trying usual test under memory controller is always helpful. 419 420 4.1 Troubleshooting 421 422 Sometimes a user might find that the application under a cgroup is 423 terminated by the OOM killer. There are several causes for this: 424 425 1. The cgroup limit is too low (just too low to do anything useful) 426 2. The user is using anonymous memory and swap is turned off or too low 427 428 A sync followed by echo 1 > /proc/sys/vm/drop_caches will help get rid of 429 some of the pages cached in the cgroup (page cache pages). 430 431 To know what happens, disabling OOM_Kill as per "10. OOM Control" (below) and 432 seeing what happens will be helpful. 433 434 4.2 Task migration 435 436 When a task migrates from one cgroup to another, its charge is not 437 carried forward by default. The pages allocated from the original cgroup still 438 remain charged to it, the charge is dropped when the page is freed or 439 reclaimed. 440 441 You can move charges of a task along with task migration. 442 See 8. "Move charges at task migration" 443 444 4.3 Removing a cgroup 445 446 A cgroup can be removed by rmdir, but as discussed in sections 4.1 and 4.2, a 447 cgroup might have some charge associated with it, even though all 448 tasks have migrated away from it. (because we charge against pages, not 449 against tasks.) 450 451 We move the stats to root (if use_hierarchy==0) or parent (if 452 use_hierarchy==1), and no change on the charge except uncharging 453 from the child. 454 455 Charges recorded in swap information is not updated at removal of cgroup. 456 Recorded information is discarded and a cgroup which uses swap (swapcache) 457 will be charged as a new owner of it. 458 459 About use_hierarchy, see Section 6. 460 461 5. Misc. interfaces. 462 463 5.1 force_empty 464 memory.force_empty interface is provided to make cgroup's memory usage empty. 465 When writing anything to this 466 467 # echo 0 > memory.force_empty 468 469 the cgroup will be reclaimed and as many pages reclaimed as possible. 470 471 The typical use case for this interface is before calling rmdir(). 472 Because rmdir() moves all pages to parent, some out-of-use page caches can be 473 moved to the parent. If you want to avoid that, force_empty will be useful. 474 475 Also, note that when memory.kmem.limit_in_bytes is set the charges due to 476 kernel pages will still be seen. This is not considered a failure and the 477 write will still return success. In this case, it is expected that 478 memory.kmem.usage_in_bytes == memory.usage_in_bytes. 479 480 About use_hierarchy, see Section 6. 481 482 5.2 stat file 483 484 memory.stat file includes following statistics 485 486 # per-memory cgroup local status 487 cache - # of bytes of page cache memory. 488 rss - # of bytes of anonymous and swap cache memory (includes 489 transparent hugepages). 490 rss_huge - # of bytes of anonymous transparent hugepages. 491 mapped_file - # of bytes of mapped file (includes tmpfs/shmem) 492 pgpgin - # of charging events to the memory cgroup. The charging 493 event happens each time a page is accounted as either mapped 494 anon page(RSS) or cache page(Page Cache) to the cgroup. 495 pgpgout - # of uncharging events to the memory cgroup. The uncharging 496 event happens each time a page is unaccounted from the cgroup. 497 swap - # of bytes of swap usage 498 writeback - # of bytes of file/anon cache that are queued for syncing to 499 disk. 500 inactive_anon - # of bytes of anonymous and swap cache memory on inactive 501 LRU list. 502 active_anon - # of bytes of anonymous and swap cache memory on active 503 LRU list. 504 inactive_file - # of bytes of file-backed memory on inactive LRU list. 505 active_file - # of bytes of file-backed memory on active LRU list. 506 unevictable - # of bytes of memory that cannot be reclaimed (mlocked etc). 507 508 # status considering hierarchy (see memory.use_hierarchy settings) 509 510 hierarchical_memory_limit - # of bytes of memory limit with regard to hierarchy 511 under which the memory cgroup is 512 hierarchical_memsw_limit - # of bytes of memory+swap limit with regard to 513 hierarchy under which memory cgroup is. 514 515 total_<counter> - # hierarchical version of <counter>, which in 516 addition to the cgroup's own value includes the 517 sum of all hierarchical children's values of 518 <counter>, i.e. total_cache 519 520 # The following additional stats are dependent on CONFIG_DEBUG_VM. 521 522 recent_rotated_anon - VM internal parameter. (see mm/vmscan.c) 523 recent_rotated_file - VM internal parameter. (see mm/vmscan.c) 524 recent_scanned_anon - VM internal parameter. (see mm/vmscan.c) 525 recent_scanned_file - VM internal parameter. (see mm/vmscan.c) 526 527 Memo: 528 recent_rotated means recent frequency of LRU rotation. 529 recent_scanned means recent # of scans to LRU. 530 showing for better debug please see the code for meanings. 531 532 Note: 533 Only anonymous and swap cache memory is listed as part of 'rss' stat. 534 This should not be confused with the true 'resident set size' or the 535 amount of physical memory used by the cgroup. 536 'rss + file_mapped" will give you resident set size of cgroup. 537 (Note: file and shmem may be shared among other cgroups. In that case, 538 file_mapped is accounted only when the memory cgroup is owner of page 539 cache.) 540 541 5.3 swappiness 542 543 Overrides /proc/sys/vm/swappiness for the particular group. The tunable 544 in the root cgroup corresponds to the global swappiness setting. 545 546 Please note that unlike during the global reclaim, limit reclaim 547 enforces that 0 swappiness really prevents from any swapping even if 548 there is a swap storage available. This might lead to memcg OOM killer 549 if there are no file pages to reclaim. 550 551 5.4 failcnt 552 553 A memory cgroup provides memory.failcnt and memory.memsw.failcnt files. 554 This failcnt(== failure count) shows the number of times that a usage counter 555 hit its limit. When a memory cgroup hits a limit, failcnt increases and 556 memory under it will be reclaimed. 557 558 You can reset failcnt by writing 0 to failcnt file. 559 # echo 0 > .../memory.failcnt 560 561 5.5 usage_in_bytes 562 563 For efficiency, as other kernel components, memory cgroup uses some optimization 564 to avoid unnecessary cacheline false sharing. usage_in_bytes is affected by the 565 method and doesn't show 'exact' value of memory (and swap) usage, it's a fuzz 566 value for efficient access. (Of course, when necessary, it's synchronized.) 567 If you want to know more exact memory usage, you should use RSS+CACHE(+SWAP) 568 value in memory.stat(see 5.2). 569 570 5.6 numa_stat 571 572 This is similar to numa_maps but operates on a per-memcg basis. This is 573 useful for providing visibility into the numa locality information within 574 an memcg since the pages are allowed to be allocated from any physical 575 node. One of the use cases is evaluating application performance by 576 combining this information with the application's CPU allocation. 577 578 Each memcg's numa_stat file includes "total", "file", "anon" and "unevictable" 579 per-node page counts including "hierarchical_<counter>" which sums up all 580 hierarchical children's values in addition to the memcg's own value. 581 582 The output format of memory.numa_stat is: 583 584 total=<total pages> N0=<node 0 pages> N1=<node 1 pages> ... 585 file=<total file pages> N0=<node 0 pages> N1=<node 1 pages> ... 586 anon=<total anon pages> N0=<node 0 pages> N1=<node 1 pages> ... 587 unevictable=<total anon pages> N0=<node 0 pages> N1=<node 1 pages> ... 588 hierarchical_<counter>=<counter pages> N0=<node 0 pages> N1=<node 1 pages> ... 589 590 The "total" count is sum of file + anon + unevictable. 591 592 6. Hierarchy support 593 594 The memory controller supports a deep hierarchy and hierarchical accounting. 595 The hierarchy is created by creating the appropriate cgroups in the 596 cgroup filesystem. Consider for example, the following cgroup filesystem 597 hierarchy 598 599 root 600 / | \ 601 / | \ 602 a b c 603 | \ 604 | \ 605 d e 606 607 In the diagram above, with hierarchical accounting enabled, all memory 608 usage of e, is accounted to its ancestors up until the root (i.e, c and root), 609 that has memory.use_hierarchy enabled. If one of the ancestors goes over its 610 limit, the reclaim algorithm reclaims from the tasks in the ancestor and the 611 children of the ancestor. 612 613 6.1 Enabling hierarchical accounting and reclaim 614 615 A memory cgroup by default disables the hierarchy feature. Support 616 can be enabled by writing 1 to memory.use_hierarchy file of the root cgroup 617 618 # echo 1 > memory.use_hierarchy 619 620 The feature can be disabled by 621 622 # echo 0 > memory.use_hierarchy 623 624 NOTE1: Enabling/disabling will fail if either the cgroup already has other 625 cgroups created below it, or if the parent cgroup has use_hierarchy 626 enabled. 627 628 NOTE2: When panic_on_oom is set to "2", the whole system will panic in 629 case of an OOM event in any cgroup. 630 631 7. Soft limits 632 633 Soft limits allow for greater sharing of memory. The idea behind soft limits 634 is to allow control groups to use as much of the memory as needed, provided 635 636 a. There is no memory contention 637 b. They do not exceed their hard limit 638 639 When the system detects memory contention or low memory, control groups 640 are pushed back to their soft limits. If the soft limit of each control 641 group is very high, they are pushed back as much as possible to make 642 sure that one control group does not starve the others of memory. 643 644 Please note that soft limits is a best-effort feature; it comes with 645 no guarantees, but it does its best to make sure that when memory is 646 heavily contended for, memory is allocated based on the soft limit 647 hints/setup. Currently soft limit based reclaim is set up such that 648 it gets invoked from balance_pgdat (kswapd). 649 650 7.1 Interface 651 652 Soft limits can be setup by using the following commands (in this example we 653 assume a soft limit of 256 MiB) 654 655 # echo 256M > memory.soft_limit_in_bytes 656 657 If we want to change this to 1G, we can at any time use 658 659 # echo 1G > memory.soft_limit_in_bytes 660 661 NOTE1: Soft limits take effect over a long period of time, since they involve 662 reclaiming memory for balancing between memory cgroups 663 NOTE2: It is recommended to set the soft limit always below the hard limit, 664 otherwise the hard limit will take precedence. 665 666 8. Move charges at task migration 667 668 Users can move charges associated with a task along with task migration, that 669 is, uncharge task's pages from the old cgroup and charge them to the new cgroup. 670 This feature is not supported in !CONFIG_MMU environments because of lack of 671 page tables. 672 673 8.1 Interface 674 675 This feature is disabled by default. It can be enabled (and disabled again) by 676 writing to memory.move_charge_at_immigrate of the destination cgroup. 677 678 If you want to enable it: 679 680 # echo (some positive value) > memory.move_charge_at_immigrate 681 682 Note: Each bits of move_charge_at_immigrate has its own meaning about what type 683 of charges should be moved. See 8.2 for details. 684 Note: Charges are moved only when you move mm->owner, in other words, 685 a leader of a thread group. 686 Note: If we cannot find enough space for the task in the destination cgroup, we 687 try to make space by reclaiming memory. Task migration may fail if we 688 cannot make enough space. 689 Note: It can take several seconds if you move charges much. 690 691 And if you want disable it again: 692 693 # echo 0 > memory.move_charge_at_immigrate 694 695 8.2 Type of charges which can be moved 696 697 Each bit in move_charge_at_immigrate has its own meaning about what type of 698 charges should be moved. But in any case, it must be noted that an account of 699 a page or a swap can be moved only when it is charged to the task's current 700 (old) memory cgroup. 701 702 bit | what type of charges would be moved ? 703 -----+------------------------------------------------------------------------ 704 0 | A charge of an anonymous page (or swap of it) used by the target task. 705 | You must enable Swap Extension (see 2.4) to enable move of swap charges. 706 -----+------------------------------------------------------------------------ 707 1 | A charge of file pages (normal file, tmpfs file (e.g. ipc shared memory) 708 | and swaps of tmpfs file) mmapped by the target task. Unlike the case of 709 | anonymous pages, file pages (and swaps) in the range mmapped by the task 710 | will be moved even if the task hasn't done page fault, i.e. they might 711 | not be the task's "RSS", but other task's "RSS" that maps the same file. 712 | And mapcount of the page is ignored (the page can be moved even if 713 | page_mapcount(page) > 1). You must enable Swap Extension (see 2.4) to 714 | enable move of swap charges. 715 716 8.3 TODO 717 718 - All of moving charge operations are done under cgroup_mutex. It's not good 719 behavior to hold the mutex too long, so we may need some trick. 720 721 9. Memory thresholds 722 723 Memory cgroup implements memory thresholds using the cgroups notification 724 API (see cgroups.txt). It allows to register multiple memory and memsw 725 thresholds and gets notifications when it crosses. 726 727 To register a threshold, an application must: 728 - create an eventfd using eventfd(2); 729 - open memory.usage_in_bytes or memory.memsw.usage_in_bytes; 730 - write string like "<event_fd> <fd of memory.usage_in_bytes> <threshold>" to 731 cgroup.event_control. 732 733 Application will be notified through eventfd when memory usage crosses 734 threshold in any direction. 735 736 It's applicable for root and non-root cgroup. 737 738 10. OOM Control 739 740 memory.oom_control file is for OOM notification and other controls. 741 742 Memory cgroup implements OOM notifier using the cgroup notification 743 API (See cgroups.txt). It allows to register multiple OOM notification 744 delivery and gets notification when OOM happens. 745 746 To register a notifier, an application must: 747 - create an eventfd using eventfd(2) 748 - open memory.oom_control file 749 - write string like "<event_fd> <fd of memory.oom_control>" to 750 cgroup.event_control 751 752 The application will be notified through eventfd when OOM happens. 753 OOM notification doesn't work for the root cgroup. 754 755 You can disable the OOM-killer by writing "1" to memory.oom_control file, as: 756 757 #echo 1 > memory.oom_control 758 759 If OOM-killer is disabled, tasks under cgroup will hang/sleep 760 in memory cgroup's OOM-waitqueue when they request accountable memory. 761 762 For running them, you have to relax the memory cgroup's OOM status by 763 * enlarge limit or reduce usage. 764 To reduce usage, 765 * kill some tasks. 766 * move some tasks to other group with account migration. 767 * remove some files (on tmpfs?) 768 769 Then, stopped tasks will work again. 770 771 At reading, current status of OOM is shown. 772 oom_kill_disable 0 or 1 (if 1, oom-killer is disabled) 773 under_oom 0 or 1 (if 1, the memory cgroup is under OOM, tasks may 774 be stopped.) 775 776 11. Memory Pressure 777 778 The pressure level notifications can be used to monitor the memory 779 allocation cost; based on the pressure, applications can implement 780 different strategies of managing their memory resources. The pressure 781 levels are defined as following: 782 783 The "low" level means that the system is reclaiming memory for new 784 allocations. Monitoring this reclaiming activity might be useful for 785 maintaining cache level. Upon notification, the program (typically 786 "Activity Manager") might analyze vmstat and act in advance (i.e. 787 prematurely shutdown unimportant services). 788 789 The "medium" level means that the system is experiencing medium memory 790 pressure, the system might be making swap, paging out active file caches, 791 etc. Upon this event applications may decide to further analyze 792 vmstat/zoneinfo/memcg or internal memory usage statistics and free any 793 resources that can be easily reconstructed or re-read from a disk. 794 795 The "critical" level means that the system is actively thrashing, it is 796 about to out of memory (OOM) or even the in-kernel OOM killer is on its 797 way to trigger. Applications should do whatever they can to help the 798 system. It might be too late to consult with vmstat or any other 799 statistics, so it's advisable to take an immediate action. 800 801 The events are propagated upward until the event is handled, i.e. the 802 events are not pass-through. Here is what this means: for example you have 803 three cgroups: A->B->C. Now you set up an event listener on cgroups A, B 804 and C, and suppose group C experiences some pressure. In this situation, 805 only group C will receive the notification, i.e. groups A and B will not 806 receive it. This is done to avoid excessive "broadcasting" of messages, 807 which disturbs the system and which is especially bad if we are low on 808 memory or thrashing. So, organize the cgroups wisely, or propagate the 809 events manually (or, ask us to implement the pass-through events, 810 explaining why would you need them.) 811 812 The file memory.pressure_level is only used to setup an eventfd. To 813 register a notification, an application must: 814 815 - create an eventfd using eventfd(2); 816 - open memory.pressure_level; 817 - write string like "<event_fd> <fd of memory.pressure_level> <level>" 818 to cgroup.event_control. 819 820 Application will be notified through eventfd when memory pressure is at 821 the specific level (or higher). Read/write operations to 822 memory.pressure_level are no implemented. 823 824 Test: 825 826 Here is a small script example that makes a new cgroup, sets up a 827 memory limit, sets up a notification in the cgroup and then makes child 828 cgroup experience a critical pressure: 829 830 # cd /sys/fs/cgroup/memory/ 831 # mkdir foo 832 # cd foo 833 # cgroup_event_listener memory.pressure_level low & 834 # echo 8000000 > memory.limit_in_bytes 835 # echo 8000000 > memory.memsw.limit_in_bytes 836 # echo $$ > tasks 837 # dd if=/dev/zero | read x 838 839 (Expect a bunch of notifications, and eventually, the oom-killer will 840 trigger.) 841 842 12. TODO 843 844 1. Make per-cgroup scanner reclaim not-shared pages first 845 2. Teach controller to account for shared-pages 846 3. Start reclamation in the background when the limit is 847 not yet hit but the usage is getting closer 848 849 Summary 850 851 Overall, the memory controller has been a stable controller and has been 852 commented and discussed quite extensively in the community. 853 854 References 855 856 1. Singh, Balbir. RFC: Memory Controller, http://lwn.net/Articles/206697/ 857 2. Singh, Balbir. Memory Controller (RSS Control), 858 http://lwn.net/Articles/222762/ 859 3. Emelianov, Pavel. Resource controllers based on process cgroups 860 http://lkml.org/lkml/2007/3/6/198 861 4. Emelianov, Pavel. RSS controller based on process cgroups (v2) 862 http://lkml.org/lkml/2007/4/9/78 863 5. Emelianov, Pavel. RSS controller based on process cgroups (v3) 864 http://lkml.org/lkml/2007/5/30/244 865 6. Menage, Paul. Control Groups v10, http://lwn.net/Articles/236032/ 866 7. Vaidyanathan, Srinivasan, Control Groups: Pagecache accounting and control 867 subsystem (v3), http://lwn.net/Articles/235534/ 868 8. Singh, Balbir. RSS controller v2 test results (lmbench), 869 http://lkml.org/lkml/2007/5/17/232 870 9. Singh, Balbir. RSS controller v2 AIM9 results 871 http://lkml.org/lkml/2007/5/18/1 872 10. Singh, Balbir. Memory controller v6 test results, 873 http://lkml.org/lkml/2007/8/19/36 874 11. Singh, Balbir. Memory controller introduction (v6), 875 http://lkml.org/lkml/2007/8/17/69 876 12. Corbet, Jonathan, Controlling memory use in cgroups, 877 http://lwn.net/Articles/243795/