Based on kernel version 3.12. Page generated on 2013-11-13 21:56 EST.
1 Memory Resource Controller 2 3 NOTE: The Memory Resource Controller has generically been referred to as the 4 memory controller in this document. Do not confuse memory controller 5 used here with the memory controller that is used in hardware. 6 7 (For editors) 8 In this document: 9 When we mention a cgroup (cgroupfs's directory) with memory controller, 10 we call it "memory cgroup". When you see git-log and source code, you'll 11 see patch's title and function names tend to use "memcg". 12 In this document, we avoid using it. 13 14 Benefits and Purpose of the memory controller 15 16 The memory controller isolates the memory behaviour of a group of tasks 17 from the rest of the system. The article on LWN  mentions some probable 18 uses of the memory controller. The memory controller can be used to 19 20 a. Isolate an application or a group of applications 21 Memory-hungry applications can be isolated and limited to a smaller 22 amount of memory. 23 b. Create a cgroup with a limited amount of memory; this can be used 24 as a good alternative to booting with mem=XXXX. 25 c. Virtualization solutions can control the amount of memory they want 26 to assign to a virtual machine instance. 27 d. A CD/DVD burner could control the amount of memory used by the 28 rest of the system to ensure that burning does not fail due to lack 29 of available memory. 30 e. There are several other use cases; find one or use the controller just 31 for fun (to learn and hack on the VM subsystem). 32 33 Current Status: linux-2.6.34-mmotm(development version of 2010/April) 34 35 Features: 36 - accounting anonymous pages, file caches, swap caches usage and limiting them. 37 - pages are linked to per-memcg LRU exclusively, and there is no global LRU. 38 - optionally, memory+swap usage can be accounted and limited. 39 - hierarchical accounting 40 - soft limit 41 - moving (recharging) account at moving a task is selectable. 42 - usage threshold notifier 43 - memory pressure notifier 44 - oom-killer disable knob and oom-notifier 45 - Root cgroup has no limit controls. 46 47 Kernel memory support is a work in progress, and the current version provides 48 basically functionality. (See Section 2.7) 49 50 Brief summary of control files. 51 52 tasks # attach a task(thread) and show list of threads 53 cgroup.procs # show list of processes 54 cgroup.event_control # an interface for event_fd() 55 memory.usage_in_bytes # show current res_counter usage for memory 56 (See 5.5 for details) 57 memory.memsw.usage_in_bytes # show current res_counter usage for memory+Swap 58 (See 5.5 for details) 59 memory.limit_in_bytes # set/show limit of memory usage 60 memory.memsw.limit_in_bytes # set/show limit of memory+Swap usage 61 memory.failcnt # show the number of memory usage hits limits 62 memory.memsw.failcnt # show the number of memory+Swap hits limits 63 memory.max_usage_in_bytes # show max memory usage recorded 64 memory.memsw.max_usage_in_bytes # show max memory+Swap usage recorded 65 memory.soft_limit_in_bytes # set/show soft limit of memory usage 66 memory.stat # show various statistics 67 memory.use_hierarchy # set/show hierarchical account enabled 68 memory.force_empty # trigger forced move charge to parent 69 memory.pressure_level # set memory pressure notifications 70 memory.swappiness # set/show swappiness parameter of vmscan 71 (See sysctl's vm.swappiness) 72 memory.move_charge_at_immigrate # set/show controls of moving charges 73 memory.oom_control # set/show oom controls. 74 memory.numa_stat # show the number of memory usage per numa node 75 76 memory.kmem.limit_in_bytes # set/show hard limit for kernel memory 77 memory.kmem.usage_in_bytes # show current kernel memory allocation 78 memory.kmem.failcnt # show the number of kernel memory usage hits limits 79 memory.kmem.max_usage_in_bytes # show max kernel memory usage recorded 80 81 memory.kmem.tcp.limit_in_bytes # set/show hard limit for tcp buf memory 82 memory.kmem.tcp.usage_in_bytes # show current tcp buf memory allocation 83 memory.kmem.tcp.failcnt # show the number of tcp buf memory usage hits limits 84 memory.kmem.tcp.max_usage_in_bytes # show max tcp buf memory usage recorded 85 86 1. History 87 88 The memory controller has a long history. A request for comments for the memory 89 controller was posted by Balbir Singh . At the time the RFC was posted 90 there were several implementations for memory control. The goal of the 91 RFC was to build consensus and agreement for the minimal features required 92 for memory control. The first RSS controller was posted by Balbir Singh 93 in Feb 2007. Pavel Emelianov  has since posted three versions of the 94 RSS controller. At OLS, at the resource management BoF, everyone suggested 95 that we handle both page cache and RSS together. Another request was raised 96 to allow user space handling of OOM. The current memory controller is 97 at version 6; it combines both mapped (RSS) and unmapped Page 98 Cache Control . 99 100 2. Memory Control 101 102 Memory is a unique resource in the sense that it is present in a limited 103 amount. If a task requires a lot of CPU processing, the task can spread 104 its processing over a period of hours, days, months or years, but with 105 memory, the same physical memory needs to be reused to accomplish the task. 106 107 The memory controller implementation has been divided into phases. These 108 are: 109 110 1. Memory controller 111 2. mlock(2) controller 112 3. Kernel user memory accounting and slab control 113 4. user mappings length controller 114 115 The memory controller is the first controller developed. 116 117 2.1. Design 118 119 The core of the design is a counter called the res_counter. The res_counter 120 tracks the current memory usage and limit of the group of processes associated 121 with the controller. Each cgroup has a memory controller specific data 122 structure (mem_cgroup) associated with it. 123 124 2.2. Accounting 125 126 +--------------------+ 127 | mem_cgroup | 128 | (res_counter) | 129 +--------------------+ 130 / ^ \ 131 / | \ 132 +---------------+ | +---------------+ 133 | mm_struct | |.... | mm_struct | 134 | | | | | 135 +---------------+ | +---------------+ 136 | 137 + --------------+ 138 | 139 +---------------+ +------+--------+ 140 | page +----------> page_cgroup| 141 | | | | 142 +---------------+ +---------------+ 143 144 (Figure 1: Hierarchy of Accounting) 145 146 147 Figure 1 shows the important aspects of the controller 148 149 1. Accounting happens per cgroup 150 2. Each mm_struct knows about which cgroup it belongs to 151 3. Each page has a pointer to the page_cgroup, which in turn knows the 152 cgroup it belongs to 153 154 The accounting is done as follows: mem_cgroup_charge_common() is invoked to 155 set up the necessary data structures and check if the cgroup that is being 156 charged is over its limit. If it is, then reclaim is invoked on the cgroup. 157 More details can be found in the reclaim section of this document. 158 If everything goes well, a page meta-data-structure called page_cgroup is 159 updated. page_cgroup has its own LRU on cgroup. 160 (*) page_cgroup structure is allocated at boot/memory-hotplug time. 161 162 2.2.1 Accounting details 163 164 All mapped anon pages (RSS) and cache pages (Page Cache) are accounted. 165 Some pages which are never reclaimable and will not be on the LRU 166 are not accounted. We just account pages under usual VM management. 167 168 RSS pages are accounted at page_fault unless they've already been accounted 169 for earlier. A file page will be accounted for as Page Cache when it's 170 inserted into inode (radix-tree). While it's mapped into the page tables of 171 processes, duplicate accounting is carefully avoided. 172 173 An RSS page is unaccounted when it's fully unmapped. A PageCache page is 174 unaccounted when it's removed from radix-tree. Even if RSS pages are fully 175 unmapped (by kswapd), they may exist as SwapCache in the system until they 176 are really freed. Such SwapCaches are also accounted. 177 A swapped-in page is not accounted until it's mapped. 178 179 Note: The kernel does swapin-readahead and reads multiple swaps at once. 180 This means swapped-in pages may contain pages for other tasks than a task 181 causing page fault. So, we avoid accounting at swap-in I/O. 182 183 At page migration, accounting information is kept. 184 185 Note: we just account pages-on-LRU because our purpose is to control amount 186 of used pages; not-on-LRU pages tend to be out-of-control from VM view. 187 188 2.3 Shared Page Accounting 189 190 Shared pages are accounted on the basis of the first touch approach. The 191 cgroup that first touches a page is accounted for the page. The principle 192 behind this approach is that a cgroup that aggressively uses a shared 193 page will eventually get charged for it (once it is uncharged from 194 the cgroup that brought it in -- this will happen on memory pressure). 195 196 But see section 8.2: when moving a task to another cgroup, its pages may 197 be recharged to the new cgroup, if move_charge_at_immigrate has been chosen. 198 199 Exception: If CONFIG_MEMCG_SWAP is not used. 200 When you do swapoff and make swapped-out pages of shmem(tmpfs) to 201 be backed into memory in force, charges for pages are accounted against the 202 caller of swapoff rather than the users of shmem. 203 204 2.4 Swap Extension (CONFIG_MEMCG_SWAP) 205 206 Swap Extension allows you to record charge for swap. A swapped-in page is 207 charged back to original page allocator if possible. 208 209 When swap is accounted, following files are added. 210 - memory.memsw.usage_in_bytes. 211 - memory.memsw.limit_in_bytes. 212 213 memsw means memory+swap. Usage of memory+swap is limited by 214 memsw.limit_in_bytes. 215 216 Example: Assume a system with 4G of swap. A task which allocates 6G of memory 217 (by mistake) under 2G memory limitation will use all swap. 218 In this case, setting memsw.limit_in_bytes=3G will prevent bad use of swap. 219 By using the memsw limit, you can avoid system OOM which can be caused by swap 220 shortage. 221 222 * why 'memory+swap' rather than swap. 223 The global LRU(kswapd) can swap out arbitrary pages. Swap-out means 224 to move account from memory to swap...there is no change in usage of 225 memory+swap. In other words, when we want to limit the usage of swap without 226 affecting global LRU, memory+swap limit is better than just limiting swap from 227 an OS point of view. 228 229 * What happens when a cgroup hits memory.memsw.limit_in_bytes 230 When a cgroup hits memory.memsw.limit_in_bytes, it's useless to do swap-out 231 in this cgroup. Then, swap-out will not be done by cgroup routine and file 232 caches are dropped. But as mentioned above, global LRU can do swapout memory 233 from it for sanity of the system's memory management state. You can't forbid 234 it by cgroup. 235 236 2.5 Reclaim 237 238 Each cgroup maintains a per cgroup LRU which has the same structure as 239 global VM. When a cgroup goes over its limit, we first try 240 to reclaim memory from the cgroup so as to make space for the new 241 pages that the cgroup has touched. If the reclaim is unsuccessful, 242 an OOM routine is invoked to select and kill the bulkiest task in the 243 cgroup. (See 10. OOM Control below.) 244 245 The reclaim algorithm has not been modified for cgroups, except that 246 pages that are selected for reclaiming come from the per-cgroup LRU 247 list. 248 249 NOTE: Reclaim does not work for the root cgroup, since we cannot set any 250 limits on the root cgroup. 251 252 Note2: When panic_on_oom is set to "2", the whole system will panic. 253 254 When oom event notifier is registered, event will be delivered. 255 (See oom_control section) 256 257 2.6 Locking 258 259 lock_page_cgroup()/unlock_page_cgroup() should not be called under 260 mapping->tree_lock. 261 262 Other lock order is following: 263 PG_locked. 264 mm->page_table_lock 265 zone->lru_lock 266 lock_page_cgroup. 267 In many cases, just lock_page_cgroup() is called. 268 per-zone-per-cgroup LRU (cgroup's private LRU) is just guarded by 269 zone->lru_lock, it has no lock of its own. 270 271 2.7 Kernel Memory Extension (CONFIG_MEMCG_KMEM) 272 273 With the Kernel memory extension, the Memory Controller is able to limit 274 the amount of kernel memory used by the system. Kernel memory is fundamentally 275 different than user memory, since it can't be swapped out, which makes it 276 possible to DoS the system by consuming too much of this precious resource. 277 278 Kernel memory won't be accounted at all until limit on a group is set. This 279 allows for existing setups to continue working without disruption. The limit 280 cannot be set if the cgroup have children, or if there are already tasks in the 281 cgroup. Attempting to set the limit under those conditions will return -EBUSY. 282 When use_hierarchy == 1 and a group is accounted, its children will 283 automatically be accounted regardless of their limit value. 284 285 After a group is first limited, it will be kept being accounted until it 286 is removed. The memory limitation itself, can of course be removed by writing 287 -1 to memory.kmem.limit_in_bytes. In this case, kmem will be accounted, but not 288 limited. 289 290 Kernel memory limits are not imposed for the root cgroup. Usage for the root 291 cgroup may or may not be accounted. The memory used is accumulated into 292 memory.kmem.usage_in_bytes, or in a separate counter when it makes sense. 293 (currently only for tcp). 294 The main "kmem" counter is fed into the main counter, so kmem charges will 295 also be visible from the user counter. 296 297 Currently no soft limit is implemented for kernel memory. It is future work 298 to trigger slab reclaim when those limits are reached. 299 300 2.7.1 Current Kernel Memory resources accounted 301 302 * stack pages: every process consumes some stack pages. By accounting into 303 kernel memory, we prevent new processes from being created when the kernel 304 memory usage is too high. 305 306 * slab pages: pages allocated by the SLAB or SLUB allocator are tracked. A copy 307 of each kmem_cache is created every time the cache is touched by the first time 308 from inside the memcg. The creation is done lazily, so some objects can still be 309 skipped while the cache is being created. All objects in a slab page should 310 belong to the same memcg. This only fails to hold when a task is migrated to a 311 different memcg during the page allocation by the cache. 312 313 * sockets memory pressure: some sockets protocols have memory pressure 314 thresholds. The Memory Controller allows them to be controlled individually 315 per cgroup, instead of globally. 316 317 * tcp memory pressure: sockets memory pressure for the tcp protocol. 318 319 2.7.3 Common use cases 320 321 Because the "kmem" counter is fed to the main user counter, kernel memory can 322 never be limited completely independently of user memory. Say "U" is the user 323 limit, and "K" the kernel limit. There are three possible ways limits can be 324 set: 325 326 U != 0, K = unlimited: 327 This is the standard memcg limitation mechanism already present before kmem 328 accounting. Kernel memory is completely ignored. 329 330 U != 0, K < U: 331 Kernel memory is a subset of the user memory. This setup is useful in 332 deployments where the total amount of memory per-cgroup is overcommited. 333 Overcommiting kernel memory limits is definitely not recommended, since the 334 box can still run out of non-reclaimable memory. 335 In this case, the admin could set up K so that the sum of all groups is 336 never greater than the total memory, and freely set U at the cost of his 337 QoS. 338 339 U != 0, K >= U: 340 Since kmem charges will also be fed to the user counter and reclaim will be 341 triggered for the cgroup for both kinds of memory. This setup gives the 342 admin a unified view of memory, and it is also useful for people who just 343 want to track kernel memory usage. 344 345 3. User Interface 346 347 0. Configuration 348 349 a. Enable CONFIG_CGROUPS 350 b. Enable CONFIG_RESOURCE_COUNTERS 351 c. Enable CONFIG_MEMCG 352 d. Enable CONFIG_MEMCG_SWAP (to use swap extension) 353 d. Enable CONFIG_MEMCG_KMEM (to use kmem extension) 354 355 1. Prepare the cgroups (see cgroups.txt, Why are cgroups needed?) 356 # mount -t tmpfs none /sys/fs/cgroup 357 # mkdir /sys/fs/cgroup/memory 358 # mount -t cgroup none /sys/fs/cgroup/memory -o memory 359 360 2. Make the new group and move bash into it 361 # mkdir /sys/fs/cgroup/memory/0 362 # echo $$ > /sys/fs/cgroup/memory/0/tasks 363 364 Since now we're in the 0 cgroup, we can alter the memory limit: 365 # echo 4M > /sys/fs/cgroup/memory/0/memory.limit_in_bytes 366 367 NOTE: We can use a suffix (k, K, m, M, g or G) to indicate values in kilo, 368 mega or gigabytes. (Here, Kilo, Mega, Giga are Kibibytes, Mebibytes, Gibibytes.) 369 370 NOTE: We can write "-1" to reset the *.limit_in_bytes(unlimited). 371 NOTE: We cannot set limits on the root cgroup any more. 372 373 # cat /sys/fs/cgroup/memory/0/memory.limit_in_bytes 374 4194304 375 376 We can check the usage: 377 # cat /sys/fs/cgroup/memory/0/memory.usage_in_bytes 378 1216512 379 380 A successful write to this file does not guarantee a successful setting of 381 this limit to the value written into the file. This can be due to a 382 number of factors, such as rounding up to page boundaries or the total 383 availability of memory on the system. The user is required to re-read 384 this file after a write to guarantee the value committed by the kernel. 385 386 # echo 1 > memory.limit_in_bytes 387 # cat memory.limit_in_bytes 388 4096 389 390 The memory.failcnt field gives the number of times that the cgroup limit was 391 exceeded. 392 393 The memory.stat file gives accounting information. Now, the number of 394 caches, RSS and Active pages/Inactive pages are shown. 395 396 4. Testing 397 398 For testing features and implementation, see memcg_test.txt. 399 400 Performance test is also important. To see pure memory controller's overhead, 401 testing on tmpfs will give you good numbers of small overheads. 402 Example: do kernel make on tmpfs. 403 404 Page-fault scalability is also important. At measuring parallel 405 page fault test, multi-process test may be better than multi-thread 406 test because it has noise of shared objects/status. 407 408 But the above two are testing extreme situations. 409 Trying usual test under memory controller is always helpful. 410 411 4.1 Troubleshooting 412 413 Sometimes a user might find that the application under a cgroup is 414 terminated by the OOM killer. There are several causes for this: 415 416 1. The cgroup limit is too low (just too low to do anything useful) 417 2. The user is using anonymous memory and swap is turned off or too low 418 419 A sync followed by echo 1 > /proc/sys/vm/drop_caches will help get rid of 420 some of the pages cached in the cgroup (page cache pages). 421 422 To know what happens, disabling OOM_Kill as per "10. OOM Control" (below) and 423 seeing what happens will be helpful. 424 425 4.2 Task migration 426 427 When a task migrates from one cgroup to another, its charge is not 428 carried forward by default. The pages allocated from the original cgroup still 429 remain charged to it, the charge is dropped when the page is freed or 430 reclaimed. 431 432 You can move charges of a task along with task migration. 433 See 8. "Move charges at task migration" 434 435 4.3 Removing a cgroup 436 437 A cgroup can be removed by rmdir, but as discussed in sections 4.1 and 4.2, a 438 cgroup might have some charge associated with it, even though all 439 tasks have migrated away from it. (because we charge against pages, not 440 against tasks.) 441 442 We move the stats to root (if use_hierarchy==0) or parent (if 443 use_hierarchy==1), and no change on the charge except uncharging 444 from the child. 445 446 Charges recorded in swap information is not updated at removal of cgroup. 447 Recorded information is discarded and a cgroup which uses swap (swapcache) 448 will be charged as a new owner of it. 449 450 About use_hierarchy, see Section 6. 451 452 5. Misc. interfaces. 453 454 5.1 force_empty 455 memory.force_empty interface is provided to make cgroup's memory usage empty. 456 You can use this interface only when the cgroup has no tasks. 457 When writing anything to this 458 459 # echo 0 > memory.force_empty 460 461 Almost all pages tracked by this memory cgroup will be unmapped and freed. 462 Some pages cannot be freed because they are locked or in-use. Such pages are 463 moved to parent (if use_hierarchy==1) or root (if use_hierarchy==0) and this 464 cgroup will be empty. 465 466 The typical use case for this interface is before calling rmdir(). 467 Because rmdir() moves all pages to parent, some out-of-use page caches can be 468 moved to the parent. If you want to avoid that, force_empty will be useful. 469 470 Also, note that when memory.kmem.limit_in_bytes is set the charges due to 471 kernel pages will still be seen. This is not considered a failure and the 472 write will still return success. In this case, it is expected that 473 memory.kmem.usage_in_bytes == memory.usage_in_bytes. 474 475 About use_hierarchy, see Section 6. 476 477 5.2 stat file 478 479 memory.stat file includes following statistics 480 481 # per-memory cgroup local status 482 cache - # of bytes of page cache memory. 483 rss - # of bytes of anonymous and swap cache memory (includes 484 transparent hugepages). 485 rss_huge - # of bytes of anonymous transparent hugepages. 486 mapped_file - # of bytes of mapped file (includes tmpfs/shmem) 487 pgpgin - # of charging events to the memory cgroup. The charging 488 event happens each time a page is accounted as either mapped 489 anon page(RSS) or cache page(Page Cache) to the cgroup. 490 pgpgout - # of uncharging events to the memory cgroup. The uncharging 491 event happens each time a page is unaccounted from the cgroup. 492 swap - # of bytes of swap usage 493 writeback - # of bytes of file/anon cache that are queued for syncing to 494 disk. 495 inactive_anon - # of bytes of anonymous and swap cache memory on inactive 496 LRU list. 497 active_anon - # of bytes of anonymous and swap cache memory on active 498 LRU list. 499 inactive_file - # of bytes of file-backed memory on inactive LRU list. 500 active_file - # of bytes of file-backed memory on active LRU list. 501 unevictable - # of bytes of memory that cannot be reclaimed (mlocked etc). 502 503 # status considering hierarchy (see memory.use_hierarchy settings) 504 505 hierarchical_memory_limit - # of bytes of memory limit with regard to hierarchy 506 under which the memory cgroup is 507 hierarchical_memsw_limit - # of bytes of memory+swap limit with regard to 508 hierarchy under which memory cgroup is. 509 510 total_<counter> - # hierarchical version of <counter>, which in 511 addition to the cgroup's own value includes the 512 sum of all hierarchical children's values of 513 <counter>, i.e. total_cache 514 515 # The following additional stats are dependent on CONFIG_DEBUG_VM. 516 517 recent_rotated_anon - VM internal parameter. (see mm/vmscan.c) 518 recent_rotated_file - VM internal parameter. (see mm/vmscan.c) 519 recent_scanned_anon - VM internal parameter. (see mm/vmscan.c) 520 recent_scanned_file - VM internal parameter. (see mm/vmscan.c) 521 522 Memo: 523 recent_rotated means recent frequency of LRU rotation. 524 recent_scanned means recent # of scans to LRU. 525 showing for better debug please see the code for meanings. 526 527 Note: 528 Only anonymous and swap cache memory is listed as part of 'rss' stat. 529 This should not be confused with the true 'resident set size' or the 530 amount of physical memory used by the cgroup. 531 'rss + file_mapped" will give you resident set size of cgroup. 532 (Note: file and shmem may be shared among other cgroups. In that case, 533 file_mapped is accounted only when the memory cgroup is owner of page 534 cache.) 535 536 5.3 swappiness 537 538 Similar to /proc/sys/vm/swappiness, but affecting a hierarchy of groups only. 539 Please note that unlike the global swappiness, memcg knob set to 0 540 really prevents from any swapping even if there is a swap storage 541 available. This might lead to memcg OOM killer if there are no file 542 pages to reclaim. 543 544 Following cgroups' swappiness can't be changed. 545 - root cgroup (uses /proc/sys/vm/swappiness). 546 - a cgroup which uses hierarchy and it has other cgroup(s) below it. 547 - a cgroup which uses hierarchy and not the root of hierarchy. 548 549 5.4 failcnt 550 551 A memory cgroup provides memory.failcnt and memory.memsw.failcnt files. 552 This failcnt(== failure count) shows the number of times that a usage counter 553 hit its limit. When a memory cgroup hits a limit, failcnt increases and 554 memory under it will be reclaimed. 555 556 You can reset failcnt by writing 0 to failcnt file. 557 # echo 0 > .../memory.failcnt 558 559 5.5 usage_in_bytes 560 561 For efficiency, as other kernel components, memory cgroup uses some optimization 562 to avoid unnecessary cacheline false sharing. usage_in_bytes is affected by the 563 method and doesn't show 'exact' value of memory (and swap) usage, it's a fuzz 564 value for efficient access. (Of course, when necessary, it's synchronized.) 565 If you want to know more exact memory usage, you should use RSS+CACHE(+SWAP) 566 value in memory.stat(see 5.2). 567 568 5.6 numa_stat 569 570 This is similar to numa_maps but operates on a per-memcg basis. This is 571 useful for providing visibility into the numa locality information within 572 an memcg since the pages are allowed to be allocated from any physical 573 node. One of the use cases is evaluating application performance by 574 combining this information with the application's CPU allocation. 575 576 We export "total", "file", "anon" and "unevictable" pages per-node for 577 each memcg. The ouput format of memory.numa_stat is: 578 579 total=<total pages> N0=<node 0 pages> N1=<node 1 pages> ... 580 file=<total file pages> N0=<node 0 pages> N1=<node 1 pages> ... 581 anon=<total anon pages> N0=<node 0 pages> N1=<node 1 pages> ... 582 unevictable=<total anon pages> N0=<node 0 pages> N1=<node 1 pages> ... 583 584 And we have total = file + anon + unevictable. 585 586 6. Hierarchy support 587 588 The memory controller supports a deep hierarchy and hierarchical accounting. 589 The hierarchy is created by creating the appropriate cgroups in the 590 cgroup filesystem. Consider for example, the following cgroup filesystem 591 hierarchy 592 593 root 594 / | \ 595 / | \ 596 a b c 597 | \ 598 | \ 599 d e 600 601 In the diagram above, with hierarchical accounting enabled, all memory 602 usage of e, is accounted to its ancestors up until the root (i.e, c and root), 603 that has memory.use_hierarchy enabled. If one of the ancestors goes over its 604 limit, the reclaim algorithm reclaims from the tasks in the ancestor and the 605 children of the ancestor. 606 607 6.1 Enabling hierarchical accounting and reclaim 608 609 A memory cgroup by default disables the hierarchy feature. Support 610 can be enabled by writing 1 to memory.use_hierarchy file of the root cgroup 611 612 # echo 1 > memory.use_hierarchy 613 614 The feature can be disabled by 615 616 # echo 0 > memory.use_hierarchy 617 618 NOTE1: Enabling/disabling will fail if either the cgroup already has other 619 cgroups created below it, or if the parent cgroup has use_hierarchy 620 enabled. 621 622 NOTE2: When panic_on_oom is set to "2", the whole system will panic in 623 case of an OOM event in any cgroup. 624 625 7. Soft limits 626 627 Soft limits allow for greater sharing of memory. The idea behind soft limits 628 is to allow control groups to use as much of the memory as needed, provided 629 630 a. There is no memory contention 631 b. They do not exceed their hard limit 632 633 When the system detects memory contention or low memory, control groups 634 are pushed back to their soft limits. If the soft limit of each control 635 group is very high, they are pushed back as much as possible to make 636 sure that one control group does not starve the others of memory. 637 638 Please note that soft limits is a best-effort feature; it comes with 639 no guarantees, but it does its best to make sure that when memory is 640 heavily contended for, memory is allocated based on the soft limit 641 hints/setup. Currently soft limit based reclaim is set up such that 642 it gets invoked from balance_pgdat (kswapd). 643 644 7.1 Interface 645 646 Soft limits can be setup by using the following commands (in this example we 647 assume a soft limit of 256 MiB) 648 649 # echo 256M > memory.soft_limit_in_bytes 650 651 If we want to change this to 1G, we can at any time use 652 653 # echo 1G > memory.soft_limit_in_bytes 654 655 NOTE1: Soft limits take effect over a long period of time, since they involve 656 reclaiming memory for balancing between memory cgroups 657 NOTE2: It is recommended to set the soft limit always below the hard limit, 658 otherwise the hard limit will take precedence. 659 660 8. Move charges at task migration 661 662 Users can move charges associated with a task along with task migration, that 663 is, uncharge task's pages from the old cgroup and charge them to the new cgroup. 664 This feature is not supported in !CONFIG_MMU environments because of lack of 665 page tables. 666 667 8.1 Interface 668 669 This feature is disabled by default. It can be enabledi (and disabled again) by 670 writing to memory.move_charge_at_immigrate of the destination cgroup. 671 672 If you want to enable it: 673 674 # echo (some positive value) > memory.move_charge_at_immigrate 675 676 Note: Each bits of move_charge_at_immigrate has its own meaning about what type 677 of charges should be moved. See 8.2 for details. 678 Note: Charges are moved only when you move mm->owner, in other words, 679 a leader of a thread group. 680 Note: If we cannot find enough space for the task in the destination cgroup, we 681 try to make space by reclaiming memory. Task migration may fail if we 682 cannot make enough space. 683 Note: It can take several seconds if you move charges much. 684 685 And if you want disable it again: 686 687 # echo 0 > memory.move_charge_at_immigrate 688 689 8.2 Type of charges which can be moved 690 691 Each bit in move_charge_at_immigrate has its own meaning about what type of 692 charges should be moved. But in any case, it must be noted that an account of 693 a page or a swap can be moved only when it is charged to the task's current 694 (old) memory cgroup. 695 696 bit | what type of charges would be moved ? 697 -----+------------------------------------------------------------------------ 698 0 | A charge of an anonymous page (or swap of it) used by the target task. 699 | You must enable Swap Extension (see 2.4) to enable move of swap charges. 700 -----+------------------------------------------------------------------------ 701 1 | A charge of file pages (normal file, tmpfs file (e.g. ipc shared memory) 702 | and swaps of tmpfs file) mmapped by the target task. Unlike the case of 703 | anonymous pages, file pages (and swaps) in the range mmapped by the task 704 | will be moved even if the task hasn't done page fault, i.e. they might 705 | not be the task's "RSS", but other task's "RSS" that maps the same file. 706 | And mapcount of the page is ignored (the page can be moved even if 707 | page_mapcount(page) > 1). You must enable Swap Extension (see 2.4) to 708 | enable move of swap charges. 709 710 8.3 TODO 711 712 - All of moving charge operations are done under cgroup_mutex. It's not good 713 behavior to hold the mutex too long, so we may need some trick. 714 715 9. Memory thresholds 716 717 Memory cgroup implements memory thresholds using the cgroups notification 718 API (see cgroups.txt). It allows to register multiple memory and memsw 719 thresholds and gets notifications when it crosses. 720 721 To register a threshold, an application must: 722 - create an eventfd using eventfd(2); 723 - open memory.usage_in_bytes or memory.memsw.usage_in_bytes; 724 - write string like "<event_fd> <fd of memory.usage_in_bytes> <threshold>" to 725 cgroup.event_control. 726 727 Application will be notified through eventfd when memory usage crosses 728 threshold in any direction. 729 730 It's applicable for root and non-root cgroup. 731 732 10. OOM Control 733 734 memory.oom_control file is for OOM notification and other controls. 735 736 Memory cgroup implements OOM notifier using the cgroup notification 737 API (See cgroups.txt). It allows to register multiple OOM notification 738 delivery and gets notification when OOM happens. 739 740 To register a notifier, an application must: 741 - create an eventfd using eventfd(2) 742 - open memory.oom_control file 743 - write string like "<event_fd> <fd of memory.oom_control>" to 744 cgroup.event_control 745 746 The application will be notified through eventfd when OOM happens. 747 OOM notification doesn't work for the root cgroup. 748 749 You can disable the OOM-killer by writing "1" to memory.oom_control file, as: 750 751 #echo 1 > memory.oom_control 752 753 This operation is only allowed to the top cgroup of a sub-hierarchy. 754 If OOM-killer is disabled, tasks under cgroup will hang/sleep 755 in memory cgroup's OOM-waitqueue when they request accountable memory. 756 757 For running them, you have to relax the memory cgroup's OOM status by 758 * enlarge limit or reduce usage. 759 To reduce usage, 760 * kill some tasks. 761 * move some tasks to other group with account migration. 762 * remove some files (on tmpfs?) 763 764 Then, stopped tasks will work again. 765 766 At reading, current status of OOM is shown. 767 oom_kill_disable 0 or 1 (if 1, oom-killer is disabled) 768 under_oom 0 or 1 (if 1, the memory cgroup is under OOM, tasks may 769 be stopped.) 770 771 11. Memory Pressure 772 773 The pressure level notifications can be used to monitor the memory 774 allocation cost; based on the pressure, applications can implement 775 different strategies of managing their memory resources. The pressure 776 levels are defined as following: 777 778 The "low" level means that the system is reclaiming memory for new 779 allocations. Monitoring this reclaiming activity might be useful for 780 maintaining cache level. Upon notification, the program (typically 781 "Activity Manager") might analyze vmstat and act in advance (i.e. 782 prematurely shutdown unimportant services). 783 784 The "medium" level means that the system is experiencing medium memory 785 pressure, the system might be making swap, paging out active file caches, 786 etc. Upon this event applications may decide to further analyze 787 vmstat/zoneinfo/memcg or internal memory usage statistics and free any 788 resources that can be easily reconstructed or re-read from a disk. 789 790 The "critical" level means that the system is actively thrashing, it is 791 about to out of memory (OOM) or even the in-kernel OOM killer is on its 792 way to trigger. Applications should do whatever they can to help the 793 system. It might be too late to consult with vmstat or any other 794 statistics, so it's advisable to take an immediate action. 795 796 The events are propagated upward until the event is handled, i.e. the 797 events are not pass-through. Here is what this means: for example you have 798 three cgroups: A->B->C. Now you set up an event listener on cgroups A, B 799 and C, and suppose group C experiences some pressure. In this situation, 800 only group C will receive the notification, i.e. groups A and B will not 801 receive it. This is done to avoid excessive "broadcasting" of messages, 802 which disturbs the system and which is especially bad if we are low on 803 memory or thrashing. So, organize the cgroups wisely, or propagate the 804 events manually (or, ask us to implement the pass-through events, 805 explaining why would you need them.) 806 807 The file memory.pressure_level is only used to setup an eventfd. To 808 register a notification, an application must: 809 810 - create an eventfd using eventfd(2); 811 - open memory.pressure_level; 812 - write string like "<event_fd> <fd of memory.pressure_level> <level>" 813 to cgroup.event_control. 814 815 Application will be notified through eventfd when memory pressure is at 816 the specific level (or higher). Read/write operations to 817 memory.pressure_level are no implemented. 818 819 Test: 820 821 Here is a small script example that makes a new cgroup, sets up a 822 memory limit, sets up a notification in the cgroup and then makes child 823 cgroup experience a critical pressure: 824 825 # cd /sys/fs/cgroup/memory/ 826 # mkdir foo 827 # cd foo 828 # cgroup_event_listener memory.pressure_level low & 829 # echo 8000000 > memory.limit_in_bytes 830 # echo 8000000 > memory.memsw.limit_in_bytes 831 # echo $$ > tasks 832 # dd if=/dev/zero | read x 833 834 (Expect a bunch of notifications, and eventually, the oom-killer will 835 trigger.) 836 837 12. TODO 838 839 1. Make per-cgroup scanner reclaim not-shared pages first 840 2. Teach controller to account for shared-pages 841 3. Start reclamation in the background when the limit is 842 not yet hit but the usage is getting closer 843 844 Summary 845 846 Overall, the memory controller has been a stable controller and has been 847 commented and discussed quite extensively in the community. 848 849 References 850 851 1. Singh, Balbir. RFC: Memory Controller, http://lwn.net/Articles/206697/ 852 2. Singh, Balbir. Memory Controller (RSS Control), 853 http://lwn.net/Articles/222762/ 854 3. Emelianov, Pavel. Resource controllers based on process cgroups 855 http://lkml.org/lkml/2007/3/6/198 856 4. Emelianov, Pavel. RSS controller based on process cgroups (v2) 857 http://lkml.org/lkml/2007/4/9/78 858 5. Emelianov, Pavel. RSS controller based on process cgroups (v3) 859 http://lkml.org/lkml/2007/5/30/244 860 6. Menage, Paul. Control Groups v10, http://lwn.net/Articles/236032/ 861 7. Vaidyanathan, Srinivasan, Control Groups: Pagecache accounting and control 862 subsystem (v3), http://lwn.net/Articles/235534/ 863 8. Singh, Balbir. RSS controller v2 test results (lmbench), 864 http://lkml.org/lkml/2007/5/17/232 865 9. Singh, Balbir. RSS controller v2 AIM9 results 866 http://lkml.org/lkml/2007/5/18/1 867 10. Singh, Balbir. Memory controller v6 test results, 868 http://lkml.org/lkml/2007/8/19/36 869 11. Singh, Balbir. Memory controller introduction (v6), 870 http://lkml.org/lkml/2007/8/17/69 871 12. Corbet, Jonathan, Controlling memory use in cgroups, 872 http://lwn.net/Articles/243795/