Lines Matching refs:memory
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
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.
52 - memory pressure notifier
56 Kernel memory support is a work in progress, and the current version provides
58 <cgroup-v1-memory-kernel-extension>`)
68 memory.usage_in_bytes show current usage for memory
70 memory.memsw.usage_in_bytes show current usage for memory+Swap
72 memory.limit_in_bytes set/show limit of memory usage
73 memory.memsw.limit_in_bytes set/show limit of memory+Swap usage
74 memory.failcnt show the number of memory usage hits limits
75 memory.memsw.failcnt show the number of memory+Swap hits limits
76 memory.max_usage_in_bytes show max memory usage recorded
77 memory.memsw.max_usage_in_bytes show max memory+Swap usage recorded
78 memory.soft_limit_in_bytes This knob is deprecated and has no effect.
81 memory.stat show various statistics
82 memory.use_hierarchy set/show hierarchical account enabled
85 memory.force_empty trigger forced page reclaim
86 memory.pressure_level set memory pressure notifications
89 memory.swappiness set/show swappiness parameter of vmscan
92 memory.move_charge_at_immigrate This knob is deprecated.
93 memory.oom_control set/show oom controls.
96 memory.numa_stat show the number of memory usage per numa
98 memory.kmem.limit_in_bytes Deprecated knob to set and read the kernel
99 memory hard limit. Kernel hard limit is not
103 Kernel memory is still charged and reported
104 by memory.kmem.usage_in_bytes.
105 memory.kmem.usage_in_bytes show current kernel memory allocation
106 memory.kmem.failcnt show the number of kernel memory usage
108 memory.kmem.max_usage_in_bytes show max kernel memory usage recorded
110 memory.kmem.tcp.limit_in_bytes set/show hard limit for tcp buf memory
113 memory.kmem.tcp.usage_in_bytes show current tcp buf memory allocation
116 memory.kmem.tcp.failcnt show the number of tcp buf memory usage
120 memory.kmem.tcp.max_usage_in_bytes show max tcp buf memory usage recorded
128 The memory controller has a long history. A request for comments for the memory
130 there were several implementations for memory control. The goal of the
132 for memory control. The first RSS controller was posted by Balbir Singh [2]_
136 raised to allow user space handling of OOM. The current memory controller is
146 memory, the same physical memory needs to be reused to accomplish the task.
148 The memory controller implementation has been divided into phases. These
153 3. Kernel user memory accounting and slab control
156 The memory controller is the first controller developed.
162 page_counter tracks the current memory usage and limit of the group of
163 processes associated with the controller. Each cgroup has a memory controller
205 (*) page_cgroup structure is allocated at boot/memory-hotplug time.
241 the cgroup that brought it in -- this will happen on memory pressure).
251 - memory.memsw.usage_in_bytes.
252 - memory.memsw.limit_in_bytes.
254 memsw means memory+swap. Usage of memory+swap is limited by
257 Example: Assume a system with 4G of swap. A task which allocates 6G of memory
258 (by mistake) under 2G memory limitation will use all swap.
263 2.4.1 why 'memory+swap' rather than swap
267 to move account from memory to swap...there is no change in usage of
268 memory+swap. In other words, when we want to limit the usage of swap without
269 affecting global LRU, memory+swap limit is better than just limiting swap from
272 2.4.2. What happens when a cgroup hits memory.memsw.limit_in_bytes
275 When a cgroup hits memory.memsw.limit_in_bytes, it's useless to do swap-out
277 caches are dropped. But as mentioned above, global LRU can do swapout memory
278 from it for sanity of the system's memory management state. You can't forbid
286 to reclaim memory from the cgroup so as to make space for the new
289 cgroup. (See :ref:`10. OOM Control <cgroup-v1-memory-oom-control>` below.)
303 (See :ref:`oom_control <cgroup-v1-memory-oom-control>` section)
319 .. _cgroup-v1-memory-kernel-extension:
324 With the Kernel memory extension, the Memory Controller is able to limit
325 the amount of kernel memory used by the system. Kernel memory is fundamentally
326 different than user memory, since it can't be swapped out, which makes it
329 Kernel memory accounting is enabled for all memory cgroups by default. But
330 it can be disabled system-wide by passing cgroup.memory=nokmem to the kernel
331 at boot time. In this case, kernel memory will not be accounted at all.
333 Kernel memory limits are not imposed for the root cgroup. Usage for the root
334 cgroup may or may not be accounted. The memory used is accumulated into
335 memory.kmem.usage_in_bytes, or in a separate counter when it makes sense.
346 kernel memory, we prevent new processes from being created when the kernel
347 memory usage is too high.
357 sockets memory pressure:
358 some sockets protocols have memory pressure
362 tcp memory pressure:
363 sockets memory pressure for the tcp protocol.
368 Because the "kmem" counter is fed to the main user counter, kernel memory can
369 never be limited completely independently of user memory. Say "U" is the user
375 accounting. Kernel memory is completely ignored.
378 Kernel memory is a subset of the user memory. This setup is useful in
379 deployments where the total amount of memory per-cgroup is overcommitted.
380 Overcommitting kernel memory limits is definitely not recommended, since the
381 box can still run out of non-reclaimable memory.
383 never greater than the total memory, and freely set U at the cost of his
387 In the current implementation, memory reclaim will NOT be triggered for
393 triggered for the cgroup for both kinds of memory. This setup gives the
394 admin a unified view of memory, and it is also useful for people who just
395 want to track kernel memory usage.
407 # mkdir /sys/fs/cgroup/memory
408 # mount -t cgroup none /sys/fs/cgroup/memory -o memory
412 # mkdir /sys/fs/cgroup/memory/0
413 # echo $$ > /sys/fs/cgroup/memory/0/tasks
415 4. Since now we're in the 0 cgroup, we can alter the memory limit::
417 # echo 4M > /sys/fs/cgroup/memory/0/memory.limit_in_bytes
421 # cat /sys/fs/cgroup/memory/0/memory.limit_in_bytes
438 # cat /sys/fs/cgroup/memory/0/memory.usage_in_bytes
444 availability of memory on the system. The user is required to re-read
447 # echo 1 > memory.limit_in_bytes
448 # cat memory.limit_in_bytes
451 The memory.failcnt field gives the number of times that the cgroup limit was
454 The memory.stat file gives accounting information. Now, the number of
462 Performance test is also important. To see pure memory controller's overhead,
471 Trying usual test under memory controller is always helpful.
473 .. _cgroup-v1-memory-test-troubleshoot:
482 2. The user is using anonymous memory and swap is turned off or too low
488 <cgroup-v1-memory-oom-control>` (below) and seeing what happens will be
491 .. _cgroup-v1-memory-test-task-migration:
502 See :ref:`8. "Move charges at task migration" <cgroup-v1-memory-move-charges>`
508 <cgroup-v1-memory-test-troubleshoot>` and :ref:`4.2
509 <cgroup-v1-memory-test-task-migration>`, a cgroup might have some charge
525 memory.force_empty interface is provided to make cgroup's memory usage empty.
528 # echo 0 > memory.force_empty
535 memory pressure happens. If you want to avoid that, force_empty will be useful.
540 memory.stat file includes following statistics:
542 * per-memory cgroup local status
545 cache # of bytes of page cache memory.
546 rss # of bytes of anonymous and swap cache memory (includes
550 pgpgin # of charging events to the memory cgroup. The charging
553 pgpgout # of uncharging events to the memory cgroup. The uncharging
557 swapcached # of bytes of swap cached in memory
561 inactive_anon # of bytes of anonymous and swap cache memory on inactive
563 active_anon # of bytes of anonymous and swap cache memory on active
565 inactive_file # of bytes of file-backed memory and MADV_FREE anonymous
566 memory (LazyFree pages) on inactive LRU list.
567 active_file # of bytes of file-backed memory on active LRU list.
568 unevictable # of bytes of memory that cannot be reclaimed (mlocked etc).
571 * status considering hierarchy (see memory.use_hierarchy settings):
574 hierarchical_memory_limit # of bytes of memory limit with regard to
576 under which the memory cgroup is
577 hierarchical_memsw_limit # of bytes of memory+swap limit with regard to
578 hierarchy under which memory cgroup is.
601 Only anonymous and swap cache memory is listed as part of 'rss' stat.
603 amount of physical memory used by the cgroup.
609 only some, but not all that memory is mapped.
612 mapped_file is accounted only when the memory cgroup is owner of page
629 A memory cgroup provides memory.failcnt and memory.memsw.failcnt files.
631 hit its limit. When a memory cgroup hits a limit, failcnt increases and
632 memory under it will be reclaimed.
636 # echo 0 > .../memory.failcnt
641 For efficiency, as other kernel components, memory cgroup uses some optimization
643 method and doesn't show 'exact' value of memory (and swap) usage, it's a fuzz
645 If you want to know more exact memory usage, you should use RSS+CACHE(+SWAP)
646 value in memory.stat(see 5.2).
661 The output format of memory.numa_stat is::
674 The memory controller supports a deep hierarchy and hierarchical accounting.
687 In the diagram above, with hierarchical accounting enabled, all memory
699 For compatibility reasons writing 1 to memory.use_hierarchy will always pass::
701 # echo 1 > memory.use_hierarchy
708 Writing to memory.soft_limit_in_bytes has no effect and reading it will
711 Use memory.low and memory.min in cgroup v2 instead.
713 .. _cgroup-v1-memory-move-charges:
720 Reading memory.move_charge_at_immigrate will always return 0 and writing
726 Memory cgroup implements memory thresholds using the cgroups notification
727 API (see cgroups.txt). It allows to register multiple memory and memsw
733 - open memory.usage_in_bytes or memory.memsw.usage_in_bytes;
734 - write string like "<event_fd> <fd of memory.usage_in_bytes> <threshold>" to
737 Application will be notified through eventfd when memory usage crosses
742 .. _cgroup-v1-memory-oom-control:
749 memory.oom_control file is for OOM notification and other controls.
758 - open memory.oom_control file
759 - write string like "<event_fd> <fd of memory.oom_control>" to
765 You can disable the OOM-killer by writing "1" to memory.oom_control file, as:
767 #echo 1 > memory.oom_control
770 in memory cgroup's OOM-waitqueue when they request accountable memory.
772 For running them, you have to relax the memory cgroup's OOM status by
789 (if 1, the memory cgroup is under OOM, tasks may be stopped.)
799 The pressure level notifications can be used to monitor the memory
801 different strategies of managing their memory resources. The pressure
804 The "low" level means that the system is reclaiming memory for new
810 The "medium" level means that the system is experiencing medium memory
813 vmstat/zoneinfo/memcg or internal memory usage statistics and free any
817 about to out of memory (OOM) or even the in-kernel OOM killer is on its
828 especially bad if we are low on memory or thrashing. Group B, will receive
840 example, groups A, B, and C will receive notification of memory pressure.
843 memory pressure is experienced in the memcg for which the notification is
845 registered for "local" notification and the group experiences memory
856 The file memory.pressure_level is only used to setup an eventfd. To
860 - open memory.pressure_level;
861 - write string as "<event_fd> <fd of memory.pressure_level> <level[,mode]>"
864 Application will be notified through eventfd when memory pressure is at
866 memory.pressure_level are no implemented.
871 memory limit, sets up a notification in the cgroup and then makes child
874 # cd /sys/fs/cgroup/memory/
877 # cgroup_event_listener memory.pressure_level low,hierarchy &
878 # echo 8000000 > memory.limit_in_bytes
879 # echo 8000000 > memory.memsw.limit_in_bytes
897 Overall, the memory controller has been a stable controller and has been
925 .. [12] Corbet, Jonathan, Controlling memory use in cgroups,