# proc_sys_vm - virtual memory subsystem - man(5) - [phpMan] [_proc_sys_vm_(5)](https://www.chedong.com/phpMan.php/man/procsysvm/5/markdown) File Formats Manual [_proc_sys_vm_(5)](https://www.chedong.com/phpMan.php/man/procsysvm/5/markdown) ## NAME /proc/sys/vm/ - virtual memory subsystem ## DESCRIPTION _/proc/sys/vm/_ This directory contains files for memory management tuning, buffer, and cache manage‐ ment. _/proc/sys/vm/admin_reserve_kbytes_ (since Linux 3.10) This file defines the amount of free memory (in KiB) on the system that should be re‐ served for users with the capability **CAP_SYS_ADMIN**. The default value in this file is the minimum of [3% of free pages, 8MiB] expressed as KiB. The default is intended to provide enough for the superuser to log in and kill a process, if necessary, under the default overcommit 'guess' mode (i.e., 0 in _/proc/sys/vm/overcommit_memory_). Systems running in "overcommit never" mode (i.e., 2 in _/proc/sys/vm/overcommit_memory_) should increase the value in this file to account for the full virtual memory size of the programs used to recover (e.g., [**login**(1)](https://www.chedong.com/phpMan.php/man/login/1/markdown) [**ssh**(1)](https://www.chedong.com/phpMan.php/man/ssh/1/markdown), and [**top**(1)](https://www.chedong.com/phpMan.php/man/top/1/markdown)) Otherwise, the supe‐ ruser may not be able to log in to recover the system. For example, on x86-64 a suit‐ able value is 131072 (128MiB reserved). Changing the value in this file takes effect whenever an application requests memory. _/proc/sys/vm/compact_memory_ (since Linux 2.6.35) When 1 is written to this file, all zones are compacted such that free memory is available in contiguous blocks where possible. The effect of this action can be seen by examining _/proc/buddyinfo_. Present only if the kernel was configured with **CONFIG_COMPACTION**. _/proc/sys/vm/drop_caches_ (since Linux 2.6.16) Writing to this file causes the kernel to drop clean caches, dentries, and inodes from memory, causing that memory to become free. This can be useful for memory management testing and performing reproducible filesystem benchmarks. Because writing to this file causes the benefits of caching to be lost, it can degrade overall system perfor‐ mance. To free pagecache, use: echo 1 > /proc/sys/vm/drop_caches To free dentries and inodes, use: echo 2 > /proc/sys/vm/drop_caches To free pagecache, dentries, and inodes, use: echo 3 > /proc/sys/vm/drop_caches Because writing to this file is a nondestructive operation and dirty objects are not freeable, the user should run [**sync**(1)](https://www.chedong.com/phpMan.php/man/sync/1/markdown) first. _/proc/sys/vm/sysctl_hugetlb_shm_group_ (since Linux 2.6.7) This writable file contains a group ID that is allowed to allocate memory using huge pages. If a process has a filesystem group ID or any supplementary group ID that matches this group ID, then it can make huge-page allocations without holding the **CAP_IPC_LOCK **capability; see [**memfd_create**(2)](https://www.chedong.com/phpMan.php/man/memfdcreate/2/markdown), [**mmap**(2)](https://www.chedong.com/phpMan.php/man/mmap/2/markdown), and [**shmget**(2)](https://www.chedong.com/phpMan.php/man/shmget/2/markdown). _/proc/sys/vm/legacy_va_layout_ (since Linux 2.6.9) If nonzero, this disables the new 32-bit memory-mapping layout; the kernel will use the legacy (2.4) layout for all processes. _/proc/sys/vm/memory_failure_early_kill_ (since Linux 2.6.32) Control how to kill processes when an uncorrected memory error (typically a 2-bit er‐ ror in a memory module) that cannot be handled by the kernel is detected in the back‐ ground by hardware. In some cases (like the page still having a valid copy on disk), the kernel will handle the failure transparently without affecting any applications. But if there is no other up-to-date copy of the data, it will kill processes to pre‐ vent any data corruptions from propagating. The file has one of the following values: **1 **Kill all processes that have the corrupted-and-not-reloadable page mapped as soon as the corruption is detected. Note that this is not supported for a few types of pages, such as kernel internally allocated data or the swap cache, but works for the majority of user pages. **0 **Unmap the corrupted page from all processes and kill a process only if it tries to access the page. The kill is performed using a **SIGBUS **signal with _si_code_ set to **BUS_MCEERR_AO**. Processes can handle this if they want to; see [**sigaction**(2)](https://www.chedong.com/phpMan.php/man/sigaction/2/markdown) for more details. This feature is active only on architectures/platforms with advanced machine check handling and depends on the hardware capabilities. Applications can override the _memory_failure_early_kill_ setting individually with the [**prctl**(2)](https://www.chedong.com/phpMan.php/man/prctl/2/markdown) **PR_MCE_KILL **operation. Present only if the kernel was configured with **CONFIG_MEMORY_FAILURE**. _/proc/sys/vm/memory_failure_recovery_ (since Linux 2.6.32) Enable memory failure recovery (when supported by the platform). **1 **Attempt recovery. **0 **Always panic on a memory failure. Present only if the kernel was configured with **CONFIG_MEMORY_FAILURE**. _/proc/sys/vm/oom_dump_tasks_ (since Linux 2.6.25) Enables a system-wide task dump (excluding kernel threads) to be produced when the kernel performs an OOM-killing. The dump includes the following information for each task (thread, process): thread ID, real user ID, thread group ID (process ID), virtual memory size, resident set size, the CPU that the task is scheduled on, oom_adj score (see the description of _/proc/_pid_/oom_adj_), and command name. This is helpful to de‐ termine why the OOM-killer was invoked and to identify the rogue task that caused it. If this contains the value zero, this information is suppressed. On very large sys‐ tems with thousands of tasks, it may not be feasible to dump the memory state informa‐ tion for each one. Such systems should not be forced to incur a performance penalty in OOM situations when the information may not be desired. If this is set to nonzero, this information is shown whenever the OOM-killer actually kills a memory-hogging task. The default value is 0. _/proc/sys/vm/oom_kill_allocating_task_ (since Linux 2.6.24) This enables or disables killing the OOM-triggering task in out-of-memory situations. If this is set to zero, the OOM-killer will scan through the entire tasklist and se‐ lect a task based on heuristics to kill. This normally selects a rogue memory-hogging task that frees up a large amount of memory when killed. If this is set to nonzero, the OOM-killer simply kills the task that triggered the out-of-memory condition. This avoids a possibly expensive tasklist scan. If _/proc/sys/vm/panic_on_oom_ is nonzero, it takes precedence over whatever value is used in _/proc/sys/vm/oom_kill_allocating_task_. The default value is 0. _/proc/sys/vm/overcommit_kbytes_ (since Linux 3.14) This writable file provides an alternative to _/proc/sys/vm/overcommit_ratio_ for con‐ trolling the _CommitLimit_ when _/proc/sys/vm/overcommit_memory_ has the value 2. It al‐ lows the amount of memory overcommitting to be specified as an absolute value (in kB), rather than as a percentage, as is done with _overcommit_ratio_. This allows for finer- grained control of _CommitLimit_ on systems with extremely large memory sizes. Only one of _overcommit_kbytes_ or _overcommit_ratio_ can have an effect: if _overcom‐_ _mit_kbytes_ has a nonzero value, then it is used to calculate _CommitLimit_, otherwise _overcommit_ratio_ is used. Writing a value to either of these files causes the value in the other file to be set to zero. _/proc/sys/vm/overcommit_memory_ This file contains the kernel virtual memory accounting mode. Values are: 0: heuristic overcommit (this is the default) 1: always overcommit, never check 2: always check, never overcommit In mode 0, calls of [**mmap**(2)](https://www.chedong.com/phpMan.php/man/mmap/2/markdown) with **MAP_NORESERVE **are not checked, and the default check is very weak, leading to the risk of getting a process "OOM-killed". In mode 1, the kernel pretends there is always enough memory, until memory actually runs out. One use case for this mode is scientific computing applications that employ large sparse arrays. Before Linux 2.6.0, any nonzero value implies mode 1. In mode 2 (available since Linux 2.6), the total virtual address space that can be al‐ located (_CommitLimit_ in _/proc/meminfo_) is calculated as CommitLimit = (total_RAM - total_huge_TLB) * overcommit_ratio / 100 + total_swap where: • _total_RAM_ is the total amount of RAM on the system; • _total_huge_TLB_ is the amount of memory set aside for huge pages; • _overcommit_ratio_ is the value in _/proc/sys/vm/overcommit_ratio_; and • _total_swap_ is the amount of swap space. For example, on a system with 16 GB of physical RAM, 16 GB of swap, no space dedicated to huge pages, and an _overcommit_ratio_ of 50, this formula yields a _CommitLimit_ of 24 GB. Since Linux 3.14, if the value in _/proc/sys/vm/overcommit_kbytes_ is nonzero, then _Com‐_ _mitLimit_ is instead calculated as: CommitLimit = overcommit_kbytes + total_swap See also the description of _/proc/sys/vm/admin_reserve_kbytes_ and _/proc/sys/vm/user_reserve_kbytes_. _/proc/sys/vm/overcommit_ratio_ (since Linux 2.6.0) This writable file defines a percentage by which memory can be overcommitted. The de‐ fault value in the file is 50. See the description of _/proc/sys/vm/overcommit_memory_. _/proc/sys/vm/panic_on_oom_ (since Linux 2.6.18) This enables or disables a kernel panic in an out-of-memory situation. If this file is set to the value 0, the kernel's OOM-killer will kill some rogue process. Usually, the OOM-killer is able to kill a rogue process and the system will survive. If this file is set to the value 1, then the kernel normally panics when out-of-memory happens. However, if a process limits allocations to certain nodes using memory poli‐ cies ([**mbind**(2)](https://www.chedong.com/phpMan.php/man/mbind/2/markdown) **MPOL_BIND**) or cpusets ([**cpuset**(7)](https://www.chedong.com/phpMan.php/man/cpuset/7/markdown)) and those nodes reach memory exhaus‐ tion status, one process may be killed by the OOM-killer. No panic occurs in this case: because other nodes' memory may be free, this means the system as a whole may not have reached an out-of-memory situation yet. If this file is set to the value 2, the kernel always panics when an out-of-memory condition occurs. The default value is 0. 1 and 2 are for failover of clustering. Select either ac‐ cording to your policy of failover. _/proc/sys/vm/swappiness_ The value in this file controls how aggressively the kernel will swap memory pages. Higher values increase aggressiveness, lower values decrease aggressiveness. The de‐ fault value is 60. _/proc/sys/vm/user_reserve_kbytes_ (since Linux 3.10) Specifies an amount of memory (in KiB) to reserve for user processes. This is in‐ tended to prevent a user from starting a single memory hogging process, such that they cannot recover (kill the hog). The value in this file has an effect only when _/proc/sys/vm/overcommit_memory_ is set to 2 ("overcommit never" mode). In this case, the system reserves an amount of memory that is the minimum of [3% of current process size, _user_reserve_kbytes_]. The default value in this file is the minimum of [3% of free pages, 128MiB] expressed as KiB. If the value in this file is set to zero, then a user will be allowed to allocate all free memory with a single process (minus the amount reserved by _/proc/sys/vm/admin_re‐_ _serve_kbytes_). Any subsequent attempts to execute a command will result in "fork: Cannot allocate memory". Changing the value in this file takes effect whenever an application requests memory. _/proc/sys/vm/unprivileged_userfaultfd_ (since Linux 5.2) This (writable) file exposes a flag that controls whether unprivileged processes are allowed to employ [**userfaultfd**(2)](https://www.chedong.com/phpMan.php/man/userfaultfd/2/markdown). If this file has the value 1, then unprivileged processes may use [**userfaultfd**(2)](https://www.chedong.com/phpMan.php/man/userfaultfd/2/markdown). If this file has the value 0, then only processes that have the **CAP_SYS_PTRACE **capability may employ [**userfaultfd**(2)](https://www.chedong.com/phpMan.php/man/userfaultfd/2/markdown). The default value in this file is 1. ## SEE ALSO [**proc**(5)](https://www.chedong.com/phpMan.php/man/proc/5/markdown), [**proc_sys**(5)](https://www.chedong.com/phpMan.php/man/procsys/5/markdown) Linux man-pages 6.7 2023-09-30 [_proc_sys_vm_(5)](https://www.chedong.com/phpMan.php/man/procsysvm/5/markdown)