Lines Matching full:memory
2 Memory Resource Controller
12 The Memory Resource Controller has generically been referred to as the
13 memory controller in this document. Do not confuse memory controller
14 used here with the memory controller that is used in hardware.
17 When we mention a cgroup (cgroupfs's directory) with memory controller,
18 we call it "memory cgroup". When you see git-log and source code, you'll
22 Benefits and Purpose of the memory controller
25 The memory controller isolates the memory behaviour of a group of tasks
27 uses of the memory controller. The memory controller can be used to
30 Memory-hungry applications can be isolated and limited to a smaller
31 amount of memory.
32 b. Create a cgroup with a limited amount of memory; this can be used
34 c. Virtualization solutions can control the amount of memory they want
36 d. A CD/DVD burner could control the amount of memory used by the
38 of available memory.
48 - optionally, memory+swap usage can be accounted and limited.
53 - memory pressure notifier
57 Kernel memory support is a work in progress, and the current version provides
67 memory.usage_in_bytes show current usage for memory
69 memory.memsw.usage_in_bytes show current usage for memory+Swap
71 memory.limit_in_bytes set/show limit of memory usage
72 memory.memsw.limit_in_bytes set/show limit of memory+Swap usage
73 memory.failcnt show the number of memory usage hits limits
74 memory.memsw.failcnt show the number of memory+Swap hits limits
75 memory.max_usage_in_bytes show max memory usage recorded
76 memory.memsw.max_usage_in_bytes show max memory+Swap usage recorded
77 memory.soft_limit_in_bytes set/show soft limit of memory usage
78 memory.stat show various statistics
79 memory.use_hierarchy set/show hierarchical account enabled
80 memory.force_empty trigger forced page reclaim
81 memory.pressure_level set memory pressure notifications
82 memory.swappiness set/show swappiness parameter of vmscan
84 memory.move_charge_at_immigrate set/show controls of moving charges
85 memory.oom_control set/show oom controls.
86 memory.numa_stat show the number of memory usage per numa
88 memory.kmem.limit_in_bytes set/show hard limit for kernel memory
92 memory.kmem.usage_in_bytes show current kernel memory allocation
93 memory.kmem.failcnt show the number of kernel memory usage
95 memory.kmem.max_usage_in_bytes show max kernel memory usage recorded
97 memory.kmem.tcp.limit_in_bytes set/show hard limit for tcp buf memory
98 memory.kmem.tcp.usage_in_bytes show current tcp buf memory allocation
99 memory.kmem.tcp.failcnt show the number of tcp buf memory usage
101 memory.kmem.tcp.max_usage_in_bytes show max tcp buf memory usage recorded
107 The memory controller has a long history. A request for comments for the memory
109 there were several implementations for memory control. The goal of the
111 for memory control. The first RSS controller was posted by Balbir Singh[2]
115 to allow user space handling of OOM. The current memory controller is
119 2. Memory Control
122 Memory is a unique resource in the sense that it is present in a limited
125 memory, the same physical memory needs to be reused to accomplish the task.
127 The memory controller implementation has been divided into phases. These
130 1. Memory controller
132 3. Kernel user memory accounting and slab control
135 The memory controller is the first controller developed.
141 page_counter tracks the current memory usage and limit of the group of
142 processes associated with the controller. Each cgroup has a memory controller
184 (*) page_cgroup structure is allocated at boot/memory-hotplug time.
220 the cgroup that brought it in -- this will happen on memory pressure).
227 be backed into memory in force, charges for pages are accounted against the
238 - memory.memsw.usage_in_bytes.
239 - memory.memsw.limit_in_bytes.
241 memsw means memory+swap. Usage of memory+swap is limited by
244 Example: Assume a system with 4G of swap. A task which allocates 6G of memory
245 (by mistake) under 2G memory limitation will use all swap.
250 **why 'memory+swap' rather than swap**
253 to move account from memory to swap...there is no change in usage of
254 memory+swap. In other words, when we want to limit the usage of swap without
255 affecting global LRU, memory+swap limit is better than just limiting swap from
258 **What happens when a cgroup hits memory.memsw.limit_in_bytes**
260 When a cgroup hits memory.memsw.limit_in_bytes, it's useless to do swap-out
262 caches are dropped. But as mentioned above, global LRU can do swapout memory
263 from it for sanity of the system's memory management state. You can't forbid
271 to reclaim memory from the cgroup so as to make space for the new
308 2.7 Kernel Memory Extension (CONFIG_MEMCG_KMEM)
311 With the Kernel memory extension, the Memory Controller is able to limit
312 the amount of kernel memory used by the system. Kernel memory is fundamentally
313 different than user memory, since it can't be swapped out, which makes it
316 Kernel memory accounting is enabled for all memory cgroups by default. But
317 it can be disabled system-wide by passing cgroup.memory=nokmem to the kernel
318 at boot time. In this case, kernel memory will not be accounted at all.
320 Kernel memory limits are not imposed for the root cgroup. Usage for the root
321 cgroup may or may not be accounted. The memory used is accumulated into
322 memory.kmem.usage_in_bytes, or in a separate counter when it makes sense.
328 Currently no soft limit is implemented for kernel memory. It is future work
331 2.7.1 Current Kernel Memory resources accounted
336 kernel memory, we prevent new processes from being created when the kernel
337 memory usage is too high.
347 sockets memory pressure:
348 some sockets protocols have memory pressure
349 thresholds. The Memory Controller allows them to be controlled individually
352 tcp memory pressure:
353 sockets memory pressure for the tcp protocol.
358 Because the "kmem" counter is fed to the main user counter, kernel memory can
359 never be limited completely independently of user memory. Say "U" is the user
365 accounting. Kernel memory is completely ignored.
368 Kernel memory is a subset of the user memory. This setup is useful in
369 deployments where the total amount of memory per-cgroup is overcommited.
370 Overcommiting kernel memory limits is definitely not recommended, since the
371 box can still run out of non-reclaimable memory.
373 never greater than the total memory, and freely set U at the cost of his
377 In the current implementation, memory reclaim will NOT be
383 triggered for the cgroup for both kinds of memory. This setup gives the
384 admin a unified view of memory, and it is also useful for people who just
385 want to track kernel memory usage.
404 # mkdir /sys/fs/cgroup/memory
405 # mount -t cgroup none /sys/fs/cgroup/memory -o memory
409 # mkdir /sys/fs/cgroup/memory/0
410 # echo $$ > /sys/fs/cgroup/memory/0/tasks
412 Since now we're in the 0 cgroup, we can alter the memory limit::
414 # echo 4M > /sys/fs/cgroup/memory/0/memory.limit_in_bytes
429 # cat /sys/fs/cgroup/memory/0/memory.limit_in_bytes
434 # cat /sys/fs/cgroup/memory/0/memory.usage_in_bytes
440 availability of memory on the system. The user is required to re-read
443 # echo 1 > memory.limit_in_bytes
444 # cat memory.limit_in_bytes
447 The memory.failcnt field gives the number of times that the cgroup limit was
450 The memory.stat file gives accounting information. Now, the number of
458 Performance test is also important. To see pure memory controller's overhead,
467 Trying usual test under memory controller is always helpful.
476 2. The user is using anonymous memory and swap is turned off or too low
518 memory.force_empty interface is provided to make cgroup's memory usage empty.
521 # echo 0 > memory.force_empty
528 memory pressure happens. If you want to avoid that, force_empty will be useful.
530 Also, note that when memory.kmem.limit_in_bytes is set the charges due to
533 memory.kmem.usage_in_bytes == memory.usage_in_bytes.
540 memory.stat file includes following statistics
542 per-memory cgroup local status
546 cache # of bytes of page cache memory.
547 rss # of bytes of anonymous and swap cache memory (includes
551 pgpgin # of charging events to the memory cgroup. The charging
554 pgpgout # of uncharging events to the memory cgroup. The uncharging
560 inactive_anon # of bytes of anonymous and swap cache memory on inactive
562 active_anon # of bytes of anonymous and swap cache memory on active
564 inactive_file # of bytes of file-backed memory on inactive LRU list.
565 active_file # of bytes of file-backed memory on active LRU list.
566 unevictable # of bytes of memory that cannot be reclaimed (mlocked etc).
569 status considering hierarchy (see memory.use_hierarchy settings)
573 hierarchical_memory_limit # of bytes of memory limit with regard to hierarchy
574 under which the memory cgroup is
575 hierarchical_memsw_limit # of bytes of memory+swap limit with regard to
576 hierarchy under which memory cgroup is.
600 Only anonymous and swap cache memory is listed as part of 'rss' stat.
602 amount of physical memory used by the cgroup.
607 mapped_file is accounted only when the memory cgroup is owner of page
624 A memory cgroup provides memory.failcnt and memory.memsw.failcnt files.
626 hit its limit. When a memory cgroup hits a limit, failcnt increases and
627 memory under it will be reclaimed.
631 # echo 0 > .../memory.failcnt
636 For efficiency, as other kernel components, memory cgroup uses some optimization
638 method and doesn't show 'exact' value of memory (and swap) usage, it's a fuzz
640 If you want to know more exact memory usage, you should use RSS+CACHE(+SWAP)
641 value in memory.stat(see 5.2).
656 The output format of memory.numa_stat is::
669 The memory controller supports a deep hierarchy and hierarchical accounting.
682 In the diagram above, with hierarchical accounting enabled, all memory
684 that has memory.use_hierarchy enabled. If one of the ancestors goes over its
691 A memory cgroup by default disables the hierarchy feature. Support
692 can be enabled by writing 1 to memory.use_hierarchy file of the root cgroup::
694 # echo 1 > memory.use_hierarchy
698 # echo 0 > memory.use_hierarchy
712 Soft limits allow for greater sharing of memory. The idea behind soft limits
713 is to allow control groups to use as much of the memory as needed, provided
715 a. There is no memory contention
718 When the system detects memory contention or low memory, control groups
721 sure that one control group does not starve the others of memory.
724 no guarantees, but it does its best to make sure that when memory is
725 heavily contended for, memory is allocated based on the soft limit
735 # echo 256M > memory.soft_limit_in_bytes
739 # echo 1G > memory.soft_limit_in_bytes
743 reclaiming memory for balancing between memory cgroups
760 writing to memory.move_charge_at_immigrate of the destination cgroup.
764 # echo (some positive value) > memory.move_charge_at_immigrate
774 try to make space by reclaiming memory. Task migration may fail if we
781 # echo 0 > memory.move_charge_at_immigrate
789 (old) memory cgroup.
797 | 1 | A charge of file pages (normal file, tmpfs file (e.g. ipc shared memory) |
813 9. Memory thresholds
816 Memory cgroup implements memory thresholds using the cgroups notification
817 API (see cgroups.txt). It allows to register multiple memory and memsw
823 - open memory.usage_in_bytes or memory.memsw.usage_in_bytes;
824 - write string like "<event_fd> <fd of memory.usage_in_bytes> <threshold>" to
827 Application will be notified through eventfd when memory usage crosses
835 memory.oom_control file is for OOM notification and other controls.
837 Memory cgroup implements OOM notifier using the cgroup notification
844 - open memory.oom_control file
845 - write string like "<event_fd> <fd of memory.oom_control>" to
851 You can disable the OOM-killer by writing "1" to memory.oom_control file, as:
853 #echo 1 > memory.oom_control
856 in memory cgroup's OOM-waitqueue when they request accountable memory.
858 For running them, you have to relax the memory cgroup's OOM status by
875 (if 1, the memory cgroup is under OOM, tasks may be stopped.)
877 11. Memory Pressure
880 The pressure level notifications can be used to monitor the memory
882 different strategies of managing their memory resources. The pressure
885 The "low" level means that the system is reclaiming memory for new
891 The "medium" level means that the system is experiencing medium memory
894 vmstat/zoneinfo/memcg or internal memory usage statistics and free any
898 about to out of memory (OOM) or even the in-kernel OOM killer is on its
909 especially bad if we are low on memory or thrashing. Group B, will receive
921 example, groups A, B, and C will receive notification of memory pressure.
924 memory pressure is experienced in the memcg for which the notification is
926 registered for "local" notification and the group experiences memory
937 The file memory.pressure_level is only used to setup an eventfd. To
941 - open memory.pressure_level;
942 - write string as "<event_fd> <fd of memory.pressure_level> <level[,mode]>"
945 Application will be notified through eventfd when memory pressure is at
947 memory.pressure_level are no implemented.
952 memory limit, sets up a notification in the cgroup and then makes child
955 # cd /sys/fs/cgroup/memory/
958 # cgroup_event_listener memory.pressure_level low,hierarchy &
959 # echo 8000000 > memory.limit_in_bytes
960 # echo 8000000 > memory.memsw.limit_in_bytes
978 Overall, the memory controller has been a stable controller and has been
984 1. Singh, Balbir. RFC: Memory Controller, http://lwn.net/Articles/206697/
985 2. Singh, Balbir. Memory Controller (RSS Control),
1000 10. Singh, Balbir. Memory controller v6 test results,
1002 11. Singh, Balbir. Memory controller introduction (v6),
1004 12. Corbet, Jonathan, Controlling memory use in cgroups,