diff options
Diffstat (limited to 'Documentation/admin-guide/sysctl')
| -rw-r--r-- | Documentation/admin-guide/sysctl/abi.rst | 73 | ||||
| -rw-r--r-- | Documentation/admin-guide/sysctl/fs.rst | 254 | ||||
| -rw-r--r-- | Documentation/admin-guide/sysctl/kernel.rst | 1586 | ||||
| -rw-r--r-- | Documentation/admin-guide/sysctl/net.rst | 122 | ||||
| -rw-r--r-- | Documentation/admin-guide/sysctl/user.rst | 6 | ||||
| -rw-r--r-- | Documentation/admin-guide/sysctl/vm.rst | 184 |
6 files changed, 1405 insertions, 820 deletions
diff --git a/Documentation/admin-guide/sysctl/abi.rst b/Documentation/admin-guide/sysctl/abi.rst index 599bcde7f0b7..4e6db0a2a4c0 100644 --- a/Documentation/admin-guide/sysctl/abi.rst +++ b/Documentation/admin-guide/sysctl/abi.rst @@ -1,67 +1,34 @@ +.. SPDX-License-Identifier: GPL-2.0+ + ================================ Documentation for /proc/sys/abi/ ================================ -kernel version 2.6.0.test2 +.. See scripts/check-sysctl-docs to keep this up to date: +.. scripts/check-sysctl-docs -vtable="abi" \ +.. Documentation/admin-guide/sysctl/abi.rst \ +.. $(git grep -l register_sysctl_) -Copyright (c) 2003, Fabian Frederick <ffrederick@users.sourceforge.net> +Copyright (c) 2020, Stephen Kitt -For general info: index.rst. +For general info, see Documentation/admin-guide/sysctl/index.rst. ------------------------------------------------------------------------------ -This path is binary emulation relevant aka personality types aka abi. -When a process is executed, it's linked to an exec_domain whose -personality is defined using values available from /proc/sys/abi. -You can find further details about abi in include/linux/personality.h. - -Here are the files featuring in 2.6 kernel: - -- defhandler_coff -- defhandler_elf -- defhandler_lcall7 -- defhandler_libcso -- fake_utsname -- trace - -defhandler_coff ---------------- - -defined value: - PER_SCOSVR3:: - - 0x0003 | STICKY_TIMEOUTS | WHOLE_SECONDS | SHORT_INODE - -defhandler_elf --------------- - -defined value: - PER_LINUX:: - - 0 - -defhandler_lcall7 ------------------ - -defined value : - PER_SVR4:: - - 0x0001 | STICKY_TIMEOUTS | MMAP_PAGE_ZERO, - -defhandler_libsco ------------------ - -defined value: - PER_SVR4:: +The files in ``/proc/sys/abi`` can be used to see and modify +ABI-related settings. - 0x0001 | STICKY_TIMEOUTS | MMAP_PAGE_ZERO, +Currently, these files might (depending on your configuration) +show up in ``/proc/sys/kernel``: -fake_utsname ------------- +.. contents:: :local: -Unused +vsyscall32 (x86) +================ -trace ------ +Determines whether the kernels maps a vDSO page into 32-bit processes; +can be set to 1 to enable, or 0 to disable. Defaults to enabled if +``CONFIG_COMPAT_VDSO`` is set, disabled otherwise. -Unused +This controls the same setting as the ``vdso32`` kernel boot +parameter. diff --git a/Documentation/admin-guide/sysctl/fs.rst b/Documentation/admin-guide/sysctl/fs.rst index 2a45119e3331..47499a1742bd 100644 --- a/Documentation/admin-guide/sysctl/fs.rst +++ b/Documentation/admin-guide/sysctl/fs.rst @@ -2,8 +2,6 @@ Documentation for /proc/sys/fs/ =============================== -kernel version 2.2.10 - Copyright (c) 1998, 1999, Rik van Riel <riel@nl.linux.org> Copyright (c) 2009, Shen Feng<shen@cn.fujitsu.com> @@ -12,159 +10,130 @@ For general info and legal blurb, please look in intro.rst. ------------------------------------------------------------------------------ -This file contains documentation for the sysctl files in -/proc/sys/fs/ and is valid for Linux kernel version 2.2. +This file contains documentation for the sysctl files and directories +in ``/proc/sys/fs/``. The files in this directory can be used to tune and monitor miscellaneous and general things in the operation of the Linux -kernel. Since some of the files _can_ be used to screw up your +kernel. Since some of the files *can* be used to screw up your system, it is advisable to read both documentation and source before actually making adjustments. 1. /proc/sys/fs =============== -Currently, these files are in /proc/sys/fs: - -- aio-max-nr -- aio-nr -- dentry-state -- dquot-max -- dquot-nr -- file-max -- file-nr -- inode-max -- inode-nr -- inode-state -- nr_open -- overflowuid -- overflowgid -- pipe-user-pages-hard -- pipe-user-pages-soft -- protected_fifos -- protected_hardlinks -- protected_regular -- protected_symlinks -- suid_dumpable -- super-max -- super-nr +Currently, these files might (depending on your configuration) +show up in ``/proc/sys/fs``: + +.. contents:: :local: aio-nr & aio-max-nr ------------------- -aio-nr is the running total of the number of events specified on the -io_setup system call for all currently active aio contexts. If aio-nr -reaches aio-max-nr then io_setup will fail with EAGAIN. Note that -raising aio-max-nr does not result in the pre-allocation or re-sizing -of any kernel data structures. +``aio-nr`` shows the current system-wide number of asynchronous io +requests. ``aio-max-nr`` allows you to change the maximum value +``aio-nr`` can grow to. If ``aio-nr`` reaches ``aio-nr-max`` then +``io_setup`` will fail with ``EAGAIN``. Note that raising +``aio-max-nr`` does not result in the +pre-allocation or re-sizing of any kernel data structures. dentry-state ------------ -From linux/include/linux/dcache.h:: +This file shows the values in ``struct dentry_stat_t``, as defined in +``fs/dcache.c``:: struct dentry_stat_t dentry_stat { - int nr_dentry; - int nr_unused; - int age_limit; /* age in seconds */ - int want_pages; /* pages requested by system */ - int nr_negative; /* # of unused negative dentries */ - int dummy; /* Reserved for future use */ + long nr_dentry; + long nr_unused; + long age_limit; /* age in seconds */ + long want_pages; /* pages requested by system */ + long nr_negative; /* # of unused negative dentries */ + long dummy; /* Reserved for future use */ }; Dentries are dynamically allocated and deallocated. -nr_dentry shows the total number of dentries allocated (active -+ unused). nr_unused shows the number of dentries that are not +``nr_dentry`` shows the total number of dentries allocated (active ++ unused). ``nr_unused shows`` the number of dentries that are not actively used, but are saved in the LRU list for future reuse. -Age_limit is the age in seconds after which dcache entries -can be reclaimed when memory is short and want_pages is -nonzero when shrink_dcache_pages() has been called and the +``age_limit`` is the age in seconds after which dcache entries +can be reclaimed when memory is short and ``want_pages`` is +nonzero when ``shrink_dcache_pages()`` has been called and the dcache isn't pruned yet. -nr_negative shows the number of unused dentries that are also +``nr_negative`` shows the number of unused dentries that are also negative dentries which do not map to any files. Instead, they help speeding up rejection of non-existing files provided by the users. -dquot-max & dquot-nr --------------------- - -The file dquot-max shows the maximum number of cached disk -quota entries. - -The file dquot-nr shows the number of allocated disk quota -entries and the number of free disk quota entries. - -If the number of free cached disk quotas is very low and -you have some awesome number of simultaneous system users, -you might want to raise the limit. - - file-max & file-nr ------------------ -The value in file-max denotes the maximum number of file- +The value in ``file-max`` denotes the maximum number of file- handles that the Linux kernel will allocate. When you get lots of error messages about running out of file handles, you might want to increase this limit. Historically,the kernel was able to allocate file handles dynamically, but not to free them again. The three values in -file-nr denote the number of allocated file handles, the number +``file-nr`` denote the number of allocated file handles, the number of allocated but unused file handles, and the maximum number of -file handles. Linux 2.6 always reports 0 as the number of free +file handles. Linux 2.6 and later always reports 0 as the number of free file handles -- this is not an error, it just means that the number of allocated file handles exactly matches the number of used file handles. -Attempts to allocate more file descriptors than file-max are -reported with printk, look for "VFS: file-max limit <number> -reached". +Attempts to allocate more file descriptors than ``file-max`` are +reported with ``printk``, look for:: + VFS: file-max limit <number> reached -nr_open -------- - -This denotes the maximum number of file-handles a process can -allocate. Default value is 1024*1024 (1048576) which should be -enough for most machines. Actual limit depends on RLIMIT_NOFILE -resource limit. +in the kernel logs. -inode-max, inode-nr & inode-state ---------------------------------- +inode-nr & inode-state +---------------------- As with file handles, the kernel allocates the inode structures dynamically, but can't free them yet. -The value in inode-max denotes the maximum number of inode -handlers. This value should be 3-4 times larger than the value -in file-max, since stdin, stdout and network sockets also -need an inode struct to handle them. When you regularly run -out of inodes, you need to increase this value. - -The file inode-nr contains the first two items from -inode-state, so we'll skip to that file... +The file ``inode-nr`` contains the first two items from +``inode-state``, so we'll skip to that file... -Inode-state contains three actual numbers and four dummies. -The actual numbers are, in order of appearance, nr_inodes, -nr_free_inodes and preshrink. +``inode-state`` contains three actual numbers and four dummies. +The actual numbers are, in order of appearance, ``nr_inodes``, +``nr_free_inodes`` and ``preshrink``. -Nr_inodes stands for the number of inodes the system has -allocated, this can be slightly more than inode-max because -Linux allocates them one pageful at a time. +``nr_inodes`` stands for the number of inodes the system has +allocated. -Nr_free_inodes represents the number of free inodes (?) and -preshrink is nonzero when the nr_inodes > inode-max and the +``nr_free_inodes`` represents the number of free inodes (?) and +preshrink is nonzero when the system needs to prune the inode list instead of allocating more. +mount-max +--------- + +This denotes the maximum number of mounts that may exist +in a mount namespace. + + +nr_open +------- + +This denotes the maximum number of file-handles a process can +allocate. Default value is 1024*1024 (1048576) which should be +enough for most machines. Actual limit depends on ``RLIMIT_NOFILE`` +resource limit. + + overflowgid & overflowuid ------------------------- @@ -192,7 +161,7 @@ pipe-user-pages-soft Maximum total number of pages a non-privileged user may allocate for pipes before the pipe size gets limited to a single page. Once this limit is reached, new pipes will be limited to a single page in size for this user in order to -limit total memory usage, and trying to increase them using fcntl() will be +limit total memory usage, and trying to increase them using ``fcntl()`` will be denied until usage goes below the limit again. The default value allows to allocate up to 1024 pipes at their default size. When set to 0, no limit is applied. @@ -207,7 +176,7 @@ file. When set to "0", writing to FIFOs is unrestricted. -When set to "1" don't allow O_CREAT open on FIFOs that we don't own +When set to "1" don't allow ``O_CREAT`` open on FIFOs that we don't own in world writable sticky directories, unless they are owned by the owner of the directory. @@ -221,7 +190,7 @@ protected_hardlinks A long-standing class of security issues is the hardlink-based time-of-check-time-of-use race, most commonly seen in world-writable -directories like /tmp. The common method of exploitation of this flaw +directories like ``/tmp``. The common method of exploitation of this flaw is to cross privilege boundaries when following a given hardlink (i.e. a root process follows a hardlink created by another user). Additionally, on systems without separated partitions, this stops unauthorized users @@ -239,13 +208,13 @@ This protection is based on the restrictions in Openwall and grsecurity. protected_regular ----------------- -This protection is similar to protected_fifos, but it +This protection is similar to `protected_fifos`_, but it avoids writes to an attacker-controlled regular file, where a program expected to create one. When set to "0", writing to regular files is unrestricted. -When set to "1" don't allow O_CREAT open on regular files that we +When set to "1" don't allow ``O_CREAT`` open on regular files that we don't own in world writable sticky directories, unless they are owned by the owner of the directory. @@ -257,11 +226,11 @@ protected_symlinks A long-standing class of security issues is the symlink-based time-of-check-time-of-use race, most commonly seen in world-writable -directories like /tmp. The common method of exploitation of this flaw +directories like ``/tmp``. The common method of exploitation of this flaw is to cross privilege boundaries when following a given symlink (i.e. a root process follows a symlink belonging to another user). For a likely incomplete list of hundreds of examples across the years, please see: -http://cve.mitre.org/cgi-bin/cvekey.cgi?keyword=/tmp +https://cve.mitre.org/cgi-bin/cvekey.cgi?keyword=/tmp When set to "0", symlink following behavior is unrestricted. @@ -272,23 +241,25 @@ follower match, or when the directory owner matches the symlink's owner. This protection is based on the restrictions in Openwall and grsecurity. -suid_dumpable: --------------- +suid_dumpable +------------- This value can be used to query and set the core dump mode for setuid or otherwise protected/tainted binaries. The modes are = ========== =============================================================== -0 (default) traditional behaviour. Any process which has changed +0 (default) Traditional behaviour. Any process which has changed privilege levels or is execute only will not be dumped. -1 (debug) all processes dump core when possible. The core dump is +1 (debug) All processes dump core when possible. The core dump is owned by the current user and no security is applied. This is intended for system debugging situations only. Ptrace is unchecked. This is insecure as it allows regular users to examine the memory contents of privileged processes. -2 (suidsafe) any binary which normally would not be dumped is dumped - anyway, but only if the "core_pattern" kernel sysctl is set to +2 (suidsafe) Any binary which normally would not be dumped is dumped + anyway, but only if the ``core_pattern`` kernel sysctl (see + :ref:`Documentation/admin-guide/sysctl/kernel.rst <core_pattern>`) + is set to either a pipe handler or a fully qualified path. (For more details on this limitation, see CVE-2006-2451.) This mode is appropriate when administrators are attempting to debug @@ -301,36 +272,11 @@ or otherwise protected/tainted binaries. The modes are = ========== =============================================================== -super-max & super-nr --------------------- - -These numbers control the maximum number of superblocks, and -thus the maximum number of mounted filesystems the kernel -can have. You only need to increase super-max if you need to -mount more filesystems than the current value in super-max -allows you to. - - -aio-nr & aio-max-nr -------------------- - -aio-nr shows the current system-wide number of asynchronous io -requests. aio-max-nr allows you to change the maximum value -aio-nr can grow to. - - -mount-max ---------- - -This denotes the maximum number of mounts that may exist -in a mount namespace. - - 2. /proc/sys/fs/binfmt_misc =========================== -Documentation for the files in /proc/sys/fs/binfmt_misc is +Documentation for the files in ``/proc/sys/fs/binfmt_misc`` is in Documentation/admin-guide/binfmt-misc.rst. @@ -343,28 +289,32 @@ creation of a user space library that implements the POSIX message queues API (as noted by the MSG tag in the POSIX 1003.1-2001 version of the System Interfaces specification.) -The "mqueue" filesystem contains values for determining/setting the amount of -resources used by the file system. +The "mqueue" filesystem contains values for determining/setting the +amount of resources used by the file system. -/proc/sys/fs/mqueue/queues_max is a read/write file for setting/getting the -maximum number of message queues allowed on the system. +``/proc/sys/fs/mqueue/queues_max`` is a read/write file for +setting/getting the maximum number of message queues allowed on the +system. -/proc/sys/fs/mqueue/msg_max is a read/write file for setting/getting the -maximum number of messages in a queue value. In fact it is the limiting value -for another (user) limit which is set in mq_open invocation. This attribute of -a queue must be less or equal then msg_max. +``/proc/sys/fs/mqueue/msg_max`` is a read/write file for +setting/getting the maximum number of messages in a queue value. In +fact it is the limiting value for another (user) limit which is set in +``mq_open`` invocation. This attribute of a queue must be less than +or equal to ``msg_max``. -/proc/sys/fs/mqueue/msgsize_max is a read/write file for setting/getting the -maximum message size value (it is every message queue's attribute set during -its creation). +``/proc/sys/fs/mqueue/msgsize_max`` is a read/write file for +setting/getting the maximum message size value (it is an attribute of +every message queue, set during its creation). -/proc/sys/fs/mqueue/msg_default is a read/write file for setting/getting the -default number of messages in a queue value if attr parameter of mq_open(2) is -NULL. If it exceed msg_max, the default value is initialized msg_max. +``/proc/sys/fs/mqueue/msg_default`` is a read/write file for +setting/getting the default number of messages in a queue value if the +``attr`` parameter of ``mq_open(2)`` is ``NULL``. If it exceeds +``msg_max``, the default value is initialized to ``msg_max``. -/proc/sys/fs/mqueue/msgsize_default is a read/write file for setting/getting -the default message size value if attr parameter of mq_open(2) is NULL. If it -exceed msgsize_max, the default value is initialized msgsize_max. +``/proc/sys/fs/mqueue/msgsize_default`` is a read/write file for +setting/getting the default message size value if the ``attr`` +parameter of ``mq_open(2)`` is ``NULL``. If it exceeds +``msgsize_max``, the default value is initialized to ``msgsize_max``. 4. /proc/sys/fs/epoll - Configuration options for the epoll interface ===================================================================== @@ -378,7 +328,7 @@ Every epoll file descriptor can store a number of files to be monitored for event readiness. Each one of these monitored files constitutes a "watch". This configuration option sets the maximum number of "watches" that are allowed for each user. -Each "watch" costs roughly 90 bytes on a 32bit kernel, and roughly 160 bytes -on a 64bit one. -The current default value for max_user_watches is the 1/32 of the available -low memory, divided for the "watch" cost in bytes. +Each "watch" costs roughly 90 bytes on a 32-bit kernel, and roughly 160 bytes +on a 64-bit one. +The current default value for ``max_user_watches`` is 4% of the +available low memory, divided by the "watch" cost in bytes. diff --git a/Documentation/admin-guide/sysctl/kernel.rst b/Documentation/admin-guide/sysctl/kernel.rst index def074807cee..7fd43947832f 100644 --- a/Documentation/admin-guide/sysctl/kernel.rst +++ b/Documentation/admin-guide/sysctl/kernel.rst @@ -2,262 +2,231 @@ Documentation for /proc/sys/kernel/ =================================== -kernel version 2.2.10 +.. See scripts/check-sysctl-docs to keep this up to date + Copyright (c) 1998, 1999, Rik van Riel <riel@nl.linux.org> Copyright (c) 2009, Shen Feng<shen@cn.fujitsu.com> -For general info and legal blurb, please look in index.rst. +For general info and legal blurb, please look in +Documentation/admin-guide/sysctl/index.rst. ------------------------------------------------------------------------------ This file contains documentation for the sysctl files in -/proc/sys/kernel/ and is valid for Linux kernel version 2.2. +``/proc/sys/kernel/``. The files in this directory can be used to tune and monitor miscellaneous and general things in the operation of the Linux -kernel. Since some of the files _can_ be used to screw up your +kernel. Since some of the files *can* be used to screw up your system, it is advisable to read both documentation and source before actually making adjustments. Currently, these files might (depending on your configuration) -show up in /proc/sys/kernel: - -- acct -- acpi_video_flags -- auto_msgmni -- bootloader_type [ X86 only ] -- bootloader_version [ X86 only ] -- cap_last_cap -- core_pattern -- core_pipe_limit -- core_uses_pid -- ctrl-alt-del -- dmesg_restrict -- domainname -- hostname -- hotplug -- hardlockup_all_cpu_backtrace -- hardlockup_panic -- hung_task_panic -- hung_task_check_count -- hung_task_timeout_secs -- hung_task_check_interval_secs -- hung_task_warnings -- hyperv_record_panic_msg -- kexec_load_disabled -- kptr_restrict -- l2cr [ PPC only ] -- modprobe ==> Documentation/debugging-modules.txt -- modules_disabled -- msg_next_id [ sysv ipc ] -- msgmax -- msgmnb -- msgmni -- nmi_watchdog -- osrelease -- ostype -- overflowgid -- overflowuid -- panic -- panic_on_oops -- panic_on_stackoverflow -- panic_on_unrecovered_nmi -- panic_on_warn -- panic_print -- panic_on_rcu_stall -- perf_cpu_time_max_percent -- perf_event_paranoid -- perf_event_max_stack -- perf_event_mlock_kb -- perf_event_max_contexts_per_stack -- pid_max -- powersave-nap [ PPC only ] -- printk -- printk_delay -- printk_ratelimit -- printk_ratelimit_burst -- pty ==> Documentation/filesystems/devpts.txt -- randomize_va_space -- real-root-dev ==> Documentation/admin-guide/initrd.rst -- reboot-cmd [ SPARC only ] -- rtsig-max -- rtsig-nr -- sched_energy_aware -- seccomp/ ==> Documentation/userspace-api/seccomp_filter.rst -- sem -- sem_next_id [ sysv ipc ] -- sg-big-buff [ generic SCSI device (sg) ] -- shm_next_id [ sysv ipc ] -- shm_rmid_forced -- shmall -- shmmax [ sysv ipc ] -- shmmni -- softlockup_all_cpu_backtrace -- soft_watchdog -- stack_erasing -- stop-a [ SPARC only ] -- sysrq ==> Documentation/admin-guide/sysrq.rst -- sysctl_writes_strict -- tainted ==> Documentation/admin-guide/tainted-kernels.rst -- threads-max -- unknown_nmi_panic -- watchdog -- watchdog_thresh -- version - - -acct: -===== +show up in ``/proc/sys/kernel``: + +.. contents:: :local: -highwater lowwater frequency + +acct +==== + +:: + + highwater lowwater frequency If BSD-style process accounting is enabled these values control its behaviour. If free space on filesystem where the log lives -goes below <lowwater>% accounting suspends. If free space gets -above <highwater>% accounting resumes. <Frequency> determines +goes below ``lowwater``\ % accounting suspends. If free space gets +above ``highwater``\ % accounting resumes. ``frequency`` determines how often do we check the amount of free space (value is in seconds). Default: -4 2 30 -That is, suspend accounting if there left <= 2% free; resume it -if we got >=4%; consider information about amount of free space -valid for 30 seconds. +:: -acpi_video_flags: -================= + 4 2 30 -flags +That is, suspend accounting if free space drops below 2%; resume it +if it increases to at least 4%; consider information about amount of +free space valid for 30 seconds. -See Doc*/kernel/power/video.txt, it allows mode of video boot to be -set during run time. +acpi_video_flags +================ -auto_msgmni: -============ +See Documentation/power/video.rst. This allows the video resume mode to be set, +in a similar fashion to the ``acpi_sleep`` kernel parameter, by +combining the following values: + += ======= +1 s3_bios +2 s3_mode +4 s3_beep += ======= + +arch +==== + +The machine hardware name, the same output as ``uname -m`` +(e.g. ``x86_64`` or ``aarch64``). + +auto_msgmni +=========== This variable has no effect and may be removed in future kernel releases. Reading it always returns 0. -Up to Linux 3.17, it enabled/disabled automatic recomputing of msgmni -upon memory add/remove or upon ipc namespace creation/removal. +Up to Linux 3.17, it enabled/disabled automatic recomputing of +`msgmni`_ +upon memory add/remove or upon IPC namespace creation/removal. Echoing "1" into this file enabled msgmni automatic recomputing. -Echoing "0" turned it off. auto_msgmni default value was 1. - +Echoing "0" turned it off. The default value was 1. -bootloader_type: -================ -x86 bootloader identification +bootloader_type (x86 only) +========================== This gives the bootloader type number as indicated by the bootloader, shifted left by 4, and OR'd with the low four bits of the bootloader version. The reason for this encoding is that this used to match the -type_of_loader field in the kernel header; the encoding is kept for +``type_of_loader`` field in the kernel header; the encoding is kept for backwards compatibility. That is, if the full bootloader type number is 0x15 and the full version number is 0x234, this file will contain the value 340 = 0x154. -See the type_of_loader and ext_loader_type fields in -Documentation/x86/boot.rst for additional information. - +See the ``type_of_loader`` and ``ext_loader_type`` fields in +Documentation/arch/x86/boot.rst for additional information. -bootloader_version: -=================== -x86 bootloader version +bootloader_version (x86 only) +============================= The complete bootloader version number. In the example above, this file will contain the value 564 = 0x234. -See the type_of_loader and ext_loader_ver fields in -Documentation/x86/boot.rst for additional information. +See the ``type_of_loader`` and ``ext_loader_ver`` fields in +Documentation/arch/x86/boot.rst for additional information. -cap_last_cap: -============= +bpf_stats_enabled +================= + +Controls whether the kernel should collect statistics on BPF programs +(total time spent running, number of times run...). Enabling +statistics causes a slight reduction in performance on each program +run. The statistics can be seen using ``bpftool``. + += =================================== +0 Don't collect statistics (default). +1 Collect statistics. += =================================== + + +cad_pid +======= + +This is the pid which will be signalled on reboot (notably, by +Ctrl-Alt-Delete). Writing a value to this file which doesn't +correspond to a running process will result in ``-ESRCH``. + +See also `ctrl-alt-del`_. + + +cap_last_cap +============ Highest valid capability of the running kernel. Exports -CAP_LAST_CAP from the kernel. +``CAP_LAST_CAP`` from the kernel. -core_pattern: -============= +.. _core_pattern: + +core_pattern +============ -core_pattern is used to specify a core dumpfile pattern name. +``core_pattern`` is used to specify a core dumpfile pattern name. * max length 127 characters; default value is "core" -* core_pattern is used as a pattern template for the output filename; - certain string patterns (beginning with '%') are substituted with - their actual values. -* backward compatibility with core_uses_pid: +* ``core_pattern`` is used as a pattern template for the output + filename; certain string patterns (beginning with '%') are + substituted with their actual values. +* backward compatibility with ``core_uses_pid``: - If core_pattern does not include "%p" (default does not) - and core_uses_pid is set, then .PID will be appended to + If ``core_pattern`` does not include "%p" (default does not) + and ``core_uses_pid`` is set, then .PID will be appended to the filename. -* corename format specifiers:: - - %<NUL> '%' is dropped - %% output one '%' - %p pid - %P global pid (init PID namespace) - %i tid - %I global tid (init PID namespace) - %u uid (in initial user namespace) - %g gid (in initial user namespace) - %d dump mode, matches PR_SET_DUMPABLE and - /proc/sys/fs/suid_dumpable - %s signal number - %t UNIX time of dump - %h hostname - %e executable filename (may be shortened) - %E executable path - %<OTHER> both are dropped +* corename format specifiers + + ======== ========================================== + %<NUL> '%' is dropped + %% output one '%' + %p pid + %P global pid (init PID namespace) + %i tid + %I global tid (init PID namespace) + %u uid (in initial user namespace) + %g gid (in initial user namespace) + %d dump mode, matches ``PR_SET_DUMPABLE`` and + ``/proc/sys/fs/suid_dumpable`` + %s signal number + %t UNIX time of dump + %h hostname + %e executable filename (may be shortened, could be changed by prctl etc) + %f executable filename + %E executable path + %c maximum size of core file by resource limit RLIMIT_CORE + %C CPU the task ran on + %<OTHER> both are dropped + ======== ========================================== * If the first character of the pattern is a '|', the kernel will treat the rest of the pattern as a command to run. The core dump will be written to the standard input of that program instead of to a file. -core_pipe_limit: -================ +core_pipe_limit +=============== -This sysctl is only applicable when core_pattern is configured to pipe -core files to a user space helper (when the first character of -core_pattern is a '|', see above). When collecting cores via a pipe -to an application, it is occasionally useful for the collecting -application to gather data about the crashing process from its -/proc/pid directory. In order to do this safely, the kernel must wait -for the collecting process to exit, so as not to remove the crashing -processes proc files prematurely. This in turn creates the -possibility that a misbehaving userspace collecting process can block -the reaping of a crashed process simply by never exiting. This sysctl -defends against that. It defines how many concurrent crashing -processes may be piped to user space applications in parallel. If -this value is exceeded, then those crashing processes above that value -are noted via the kernel log and their cores are skipped. 0 is a -special value, indicating that unlimited processes may be captured in -parallel, but that no waiting will take place (i.e. the collecting -process is not guaranteed access to /proc/<crashing pid>/). This -value defaults to 0. - - -core_uses_pid: -============== +This sysctl is only applicable when `core_pattern`_ is configured to +pipe core files to a user space helper (when the first character of +``core_pattern`` is a '|', see above). +When collecting cores via a pipe to an application, it is occasionally +useful for the collecting application to gather data about the +crashing process from its ``/proc/pid`` directory. +In order to do this safely, the kernel must wait for the collecting +process to exit, so as not to remove the crashing processes proc files +prematurely. +This in turn creates the possibility that a misbehaving userspace +collecting process can block the reaping of a crashed process simply +by never exiting. +This sysctl defends against that. +It defines how many concurrent crashing processes may be piped to user +space applications in parallel. +If this value is exceeded, then those crashing processes above that +value are noted via the kernel log and their cores are skipped. +0 is a special value, indicating that unlimited processes may be +captured in parallel, but that no waiting will take place (i.e. the +collecting process is not guaranteed access to ``/proc/<crashing +pid>/``). +This value defaults to 0. + + +core_uses_pid +============= The default coredump filename is "core". By setting -core_uses_pid to 1, the coredump filename becomes core.PID. -If core_pattern does not include "%p" (default does not) -and core_uses_pid is set, then .PID will be appended to +``core_uses_pid`` to 1, the coredump filename becomes core.PID. +If `core_pattern`_ does not include "%p" (default does not) +and ``core_uses_pid`` is set, then .PID will be appended to the filename. -ctrl-alt-del: -============= +ctrl-alt-del +============ When the value in this file is 0, ctrl-alt-del is trapped and -sent to the init(1) program to handle a graceful restart. +sent to the ``init(1)`` program to handle a graceful restart. When, however, the value is > 0, Linux's reaction to a Vulcan Nerve Pinch (tm) will be an immediate reboot, without even syncing its dirty buffers. @@ -269,21 +238,22 @@ Note: to decide what to do with it. -dmesg_restrict: -=============== +dmesg_restrict +============== This toggle indicates whether unprivileged users are prevented -from using dmesg(8) to view messages from the kernel's log buffer. -When dmesg_restrict is set to (0) there are no restrictions. When -dmesg_restrict is set set to (1), users must have CAP_SYSLOG to use -dmesg(8). +from using ``dmesg(8)`` to view messages from the kernel's log +buffer. +When ``dmesg_restrict`` is set to 0 there are no restrictions. +When ``dmesg_restrict`` is set to 1, users must have +``CAP_SYSLOG`` to use ``dmesg(8)``. -The kernel config option CONFIG_SECURITY_DMESG_RESTRICT sets the -default value of dmesg_restrict. +The kernel config option ``CONFIG_SECURITY_DMESG_RESTRICT`` sets the +default value of ``dmesg_restrict``. -domainname & hostname: -====================== +domainname & hostname +===================== These files can be used to set the NIS/YP domainname and the hostname of your box in exactly the same way as the commands @@ -302,167 +272,358 @@ hostname "darkstar" and DNS (Internet Domain Name Server) domainname "frop.org", not to be confused with the NIS (Network Information Service) or YP (Yellow Pages) domainname. These two domain names are in general different. For a detailed discussion -see the hostname(1) man page. +see the ``hostname(1)`` man page. -hardlockup_all_cpu_backtrace: -============================= +firmware_config +=============== + +See Documentation/driver-api/firmware/fallback-mechanisms.rst. + +The entries in this directory allow the firmware loader helper +fallback to be controlled: + +* ``force_sysfs_fallback``, when set to 1, forces the use of the + fallback; +* ``ignore_sysfs_fallback``, when set to 1, ignores any fallback. + + +ftrace_dump_on_oops +=================== + +Determines whether ``ftrace_dump()`` should be called on an oops (or +kernel panic). This will output the contents of the ftrace buffers to +the console. This is very useful for capturing traces that lead to +crashes and outputting them to a serial console. + +======================= =========================================== +0 Disabled (default). +1 Dump buffers of all CPUs. +2(orig_cpu) Dump the buffer of the CPU that triggered the + oops. +<instance> Dump the specific instance buffer on all CPUs. +<instance>=2(orig_cpu) Dump the specific instance buffer on the CPU + that triggered the oops. +======================= =========================================== + +Multiple instance dump is also supported, and instances are separated +by commas. If global buffer also needs to be dumped, please specify +the dump mode (1/2/orig_cpu) first for global buffer. + +So for example to dump "foo" and "bar" instance buffer on all CPUs, +user can:: + + echo "foo,bar" > /proc/sys/kernel/ftrace_dump_on_oops + +To dump global buffer and "foo" instance buffer on all +CPUs along with the "bar" instance buffer on CPU that triggered the +oops, user can:: + + echo "1,foo,bar=2" > /proc/sys/kernel/ftrace_dump_on_oops + +ftrace_enabled, stack_tracer_enabled +==================================== + +See Documentation/trace/ftrace.rst. + + +hardlockup_all_cpu_backtrace +============================ This value controls the hard lockup detector behavior when a hard lockup condition is detected as to whether or not to gather further debug information. If enabled, arch-specific all-CPU stack dumping will be initiated. -0: do nothing. This is the default behavior. += ============================================ +0 Do nothing. This is the default behavior. +1 On detection capture more debug information. += ============================================ -1: on detection capture more debug information. - -hardlockup_panic: -================= +hardlockup_panic +================ This parameter can be used to control whether the kernel panics when a hard lockup is detected. - 0 - don't panic on hard lockup - 1 - panic on hard lockup += =========================== +0 Don't panic on hard lockup. +1 Panic on hard lockup. += =========================== -See Documentation/admin-guide/lockup-watchdogs.rst for more information. This can -also be set using the nmi_watchdog kernel parameter. +See Documentation/admin-guide/lockup-watchdogs.rst for more information. +This can also be set using the nmi_watchdog kernel parameter. -hotplug: -======== +hotplug +======= Path for the hotplug policy agent. -Default value is "/sbin/hotplug". +Default value is ``CONFIG_UEVENT_HELPER_PATH``, which in turn defaults +to the empty string. +This file only exists when ``CONFIG_UEVENT_HELPER`` is enabled. Most +modern systems rely exclusively on the netlink-based uevent source and +don't need this. -hung_task_panic: -================ -Controls the kernel's behavior when a hung task is detected. -This file shows up if CONFIG_DETECT_HUNG_TASK is enabled. +hung_task_all_cpu_backtrace +=========================== -0: continue operation. This is the default behavior. +If this option is set, the kernel will send an NMI to all CPUs to dump +their backtraces when a hung task is detected. This file shows up if +CONFIG_DETECT_HUNG_TASK and CONFIG_SMP are enabled. -1: panic immediately. +0: Won't show all CPUs backtraces when a hung task is detected. +This is the default behavior. +1: Will non-maskably interrupt all CPUs and dump their backtraces when +a hung task is detected. -hung_task_check_count: -====================== + +hung_task_panic +=============== + +Controls the kernel's behavior when a hung task is detected. +This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled. + += ================================================= +0 Continue operation. This is the default behavior. +1 Panic immediately. += ================================================= + + +hung_task_check_count +===================== The upper bound on the number of tasks that are checked. -This file shows up if CONFIG_DETECT_HUNG_TASK is enabled. +This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled. -hung_task_timeout_secs: -======================= +hung_task_timeout_secs +====================== When a task in D state did not get scheduled for more than this value report a warning. -This file shows up if CONFIG_DETECT_HUNG_TASK is enabled. +This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled. -0: means infinite timeout - no checking done. +0 means infinite timeout, no checking is done. -Possible values to set are in range {0..LONG_MAX/HZ}. +Possible values to set are in range {0:``LONG_MAX``/``HZ``}. -hung_task_check_interval_secs: -============================== +hung_task_check_interval_secs +============================= Hung task check interval. If hung task checking is enabled -(see hung_task_timeout_secs), the check is done every -hung_task_check_interval_secs seconds. -This file shows up if CONFIG_DETECT_HUNG_TASK is enabled. +(see `hung_task_timeout_secs`_), the check is done every +``hung_task_check_interval_secs`` seconds. +This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled. -0 (default): means use hung_task_timeout_secs as checking interval. -Possible values to set are in range {0..LONG_MAX/HZ}. +0 (default) means use ``hung_task_timeout_secs`` as checking +interval. +Possible values to set are in range {0:``LONG_MAX``/``HZ``}. -hung_task_warnings: -=================== + +hung_task_warnings +================== The maximum number of warnings to report. During a check interval if a hung task is detected, this value is decreased by 1. When this value reaches 0, no more warnings will be reported. -This file shows up if CONFIG_DETECT_HUNG_TASK is enabled. +This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled. -1: report an infinite number of warnings. -hyperv_record_panic_msg: -======================== +hyperv_record_panic_msg +======================= Controls whether the panic kmsg data should be reported to Hyper-V. -0: do not report panic kmsg data. += ========================================================= +0 Do not report panic kmsg data. +1 Report the panic kmsg data. This is the default behavior. += ========================================================= -1: report the panic kmsg data. This is the default behavior. +ignore-unaligned-usertrap +========================= -kexec_load_disabled: -==================== +On architectures where unaligned accesses cause traps, and where this +feature is supported (``CONFIG_SYSCTL_ARCH_UNALIGN_NO_WARN``; +currently, ``arc`` and ``loongarch``), controls whether all +unaligned traps are logged. + += ============================================================= +0 Log all unaligned accesses. +1 Only warn the first time a process traps. This is the default + setting. += ============================================================= + +See also `unaligned-trap`_. + +io_uring_disabled +================= + +Prevents all processes from creating new io_uring instances. Enabling this +shrinks the kernel's attack surface. -A toggle indicating if the kexec_load syscall has been disabled. This -value defaults to 0 (false: kexec_load enabled), but can be set to 1 -(true: kexec_load disabled). Once true, kexec can no longer be used, and -the toggle cannot be set back to false. This allows a kexec image to be -loaded before disabling the syscall, allowing a system to set up (and -later use) an image without it being altered. Generally used together -with the "modules_disabled" sysctl. += ====================================================================== +0 All processes can create io_uring instances as normal. This is the + default setting. +1 io_uring creation is disabled (io_uring_setup() will fail with + -EPERM) for unprivileged processes not in the io_uring_group group. + Existing io_uring instances can still be used. See the + documentation for io_uring_group for more information. +2 io_uring creation is disabled for all processes. io_uring_setup() + always fails with -EPERM. Existing io_uring instances can still be + used. += ====================================================================== -kptr_restrict: +io_uring_group ============== -This toggle indicates whether restrictions are placed on -exposing kernel addresses via /proc and other interfaces. +When io_uring_disabled is set to 1, a process must either be +privileged (CAP_SYS_ADMIN) or be in the io_uring_group group in order +to create an io_uring instance. If io_uring_group is set to -1 (the +default), only processes with the CAP_SYS_ADMIN capability may create +io_uring instances. -When kptr_restrict is set to 0 (the default) the address is hashed before -printing. (This is the equivalent to %p.) -When kptr_restrict is set to (1), kernel pointers printed using the %pK -format specifier will be replaced with 0's unless the user has CAP_SYSLOG -and effective user and group ids are equal to the real ids. This is -because %pK checks are done at read() time rather than open() time, so -if permissions are elevated between the open() and the read() (e.g via -a setuid binary) then %pK will not leak kernel pointers to unprivileged -users. Note, this is a temporary solution only. The correct long-term -solution is to do the permission checks at open() time. Consider removing -world read permissions from files that use %pK, and using dmesg_restrict -to protect against uses of %pK in dmesg(8) if leaking kernel pointer -values to unprivileged users is a concern. +kexec_load_disabled +=================== -When kptr_restrict is set to (2), kernel pointers printed using -%pK will be replaced with 0's regardless of privileges. +A toggle indicating if the syscalls ``kexec_load`` and +``kexec_file_load`` have been disabled. +This value defaults to 0 (false: ``kexec_*load`` enabled), but can be +set to 1 (true: ``kexec_*load`` disabled). +Once true, kexec can no longer be used, and the toggle cannot be set +back to false. +This allows a kexec image to be loaded before disabling the syscall, +allowing a system to set up (and later use) an image without it being +altered. +Generally used together with the `modules_disabled`_ sysctl. + +kexec_load_limit_panic +====================== +This parameter specifies a limit to the number of times the syscalls +``kexec_load`` and ``kexec_file_load`` can be called with a crash +image. It can only be set with a more restrictive value than the +current one. -l2cr: (PPC only) -================ +== ====================================================== +-1 Unlimited calls to kexec. This is the default setting. +N Number of calls left. +== ====================================================== -This flag controls the L2 cache of G3 processor boards. If -0, the cache is disabled. Enabled if nonzero. +kexec_load_limit_reboot +======================= +Similar functionality as ``kexec_load_limit_panic``, but for a normal +image. -modules_disabled: -================= +kptr_restrict +============= + +This toggle indicates whether restrictions are placed on +exposing kernel addresses via ``/proc`` and other interfaces. + +When ``kptr_restrict`` is set to 0 (the default) the address is hashed +before printing. +(This is the equivalent to %p.) + +When ``kptr_restrict`` is set to 1, kernel pointers printed using the +%pK format specifier will be replaced with 0s unless the user has +``CAP_SYSLOG`` and effective user and group ids are equal to the real +ids. +This is because %pK checks are done at read() time rather than open() +time, so if permissions are elevated between the open() and the read() +(e.g via a setuid binary) then %pK will not leak kernel pointers to +unprivileged users. +Note, this is a temporary solution only. +The correct long-term solution is to do the permission checks at +open() time. +Consider removing world read permissions from files that use %pK, and +using `dmesg_restrict`_ to protect against uses of %pK in ``dmesg(8)`` +if leaking kernel pointer values to unprivileged users is a concern. + +When ``kptr_restrict`` is set to 2, kernel pointers printed using +%pK will be replaced with 0s regardless of privileges. + + +modprobe +======== + +The full path to the usermode helper for autoloading kernel modules, +by default ``CONFIG_MODPROBE_PATH``, which in turn defaults to +"/sbin/modprobe". This binary is executed when the kernel requests a +module. For example, if userspace passes an unknown filesystem type +to mount(), then the kernel will automatically request the +corresponding filesystem module by executing this usermode helper. +This usermode helper should insert the needed module into the kernel. + +This sysctl only affects module autoloading. It has no effect on the +ability to explicitly insert modules. + +This sysctl can be used to debug module loading requests:: + + echo '#! /bin/sh' > /tmp/modprobe + echo 'echo "$@" >> /tmp/modprobe.log' >> /tmp/modprobe + echo 'exec /sbin/modprobe "$@"' >> /tmp/modprobe + chmod a+x /tmp/modprobe + echo /tmp/modprobe > /proc/sys/kernel/modprobe + +Alternatively, if this sysctl is set to the empty string, then module +autoloading is completely disabled. The kernel will not try to +execute a usermode helper at all, nor will it call the +kernel_module_request LSM hook. + +If CONFIG_STATIC_USERMODEHELPER=y is set in the kernel configuration, +then the configured static usermode helper overrides this sysctl, +except that the empty string is still accepted to completely disable +module autoloading as described above. + +modules_disabled +================ A toggle value indicating if modules are allowed to be loaded in an otherwise modular kernel. This toggle defaults to off (0), but can be set true (1). Once true, modules can be neither loaded nor unloaded, and the toggle cannot be set back -to false. Generally used with the "kexec_load_disabled" toggle. +to false. Generally used with the `kexec_load_disabled`_ toggle. + + +.. _msgmni: + +msgmax, msgmnb, and msgmni +========================== +``msgmax`` is the maximum size of an IPC message, in bytes. 8192 by +default (``MSGMAX``). -msg_next_id, sem_next_id, and shm_next_id: -========================================== +``msgmnb`` is the maximum size of an IPC queue, in bytes. 16384 by +default (``MSGMNB``). + +``msgmni`` is the maximum number of IPC queues. 32000 by default +(``MSGMNI``). + +All of these parameters are set per ipc namespace. The maximum number of bytes +in POSIX message queues is limited by ``RLIMIT_MSGQUEUE``. This limit is +respected hierarchically in the each user namespace. + +msg_next_id, sem_next_id, and shm_next_id (System V IPC) +======================================================== These three toggles allows to specify desired id for next allocated IPC object: message, semaphore or shared memory respectively. By default they are equal to -1, which means generic allocation logic. -Possible values to set are in range {0..INT_MAX}. +Possible values to set are in range {0:``INT_MAX``}. Notes: 1) kernel doesn't guarantee, that new object will have desired id. So, @@ -472,15 +633,24 @@ Notes: fails, it is undefined if the value remains unmodified or is reset to -1. -nmi_watchdog: -============= +ngroups_max +=========== + +Maximum number of supplementary groups, _i.e._ the maximum size which +``setgroups`` will accept. Exports ``NGROUPS_MAX`` from the kernel. + + + +nmi_watchdog +============ This parameter can be used to control the NMI watchdog (i.e. the hard lockup detector) on x86 systems. -0 - disable the hard lockup detector - -1 - enable the hard lockup detector += ================================= +0 Disable the hard lockup detector. +1 Enable the hard lockup detector. += ================================= The hard lockup detector monitors each CPU for its ability to respond to timer interrupts. The mechanism utilizes CPU performance counter registers @@ -492,73 +662,91 @@ in a KVM virtual machine. This default can be overridden by adding:: nmi_watchdog=1 -to the guest kernel command line (see Documentation/admin-guide/kernel-parameters.rst). +to the guest kernel command line (see +Documentation/admin-guide/kernel-parameters.rst). -numa_balancing: -=============== +nmi_wd_lpm_factor (PPC only) +============================ + +Factor to apply to the NMI watchdog timeout (only when ``nmi_watchdog`` is +set to 1). This factor represents the percentage added to +``watchdog_thresh`` when calculating the NMI watchdog timeout during an +LPM. The soft lockup timeout is not impacted. -Enables/disables automatic page fault based NUMA memory -balancing. Memory is moved automatically to nodes -that access it often. +A value of 0 means no change. The default value is 200 meaning the NMI +watchdog is set to 30s (based on ``watchdog_thresh`` equal to 10). + + +numa_balancing +============== -Enables/disables automatic NUMA memory balancing. On NUMA machines, there -is a performance penalty if remote memory is accessed by a CPU. When this -feature is enabled the kernel samples what task thread is accessing memory -by periodically unmapping pages and later trapping a page fault. At the -time of the page fault, it is determined if the data being accessed should -be migrated to a local memory node. +Enables/disables and configures automatic page fault based NUMA memory +balancing. Memory is moved automatically to nodes that access it often. +The value to set can be the result of ORing the following: + += ================================= +0 NUMA_BALANCING_DISABLED +1 NUMA_BALANCING_NORMAL +2 NUMA_BALANCING_MEMORY_TIERING += ================================= + +Or NUMA_BALANCING_NORMAL to optimize page placement among different +NUMA nodes to reduce remote accessing. On NUMA machines, there is a +performance penalty if remote memory is accessed by a CPU. When this +feature is enabled the kernel samples what task thread is accessing +memory by periodically unmapping pages and later trapping a page +fault. At the time of the page fault, it is determined if the data +being accessed should be migrated to a local memory node. The unmapping of pages and trapping faults incur additional overhead that ideally is offset by improved memory locality but there is no universal guarantee. If the target workload is already bound to NUMA nodes then this -feature should be disabled. Otherwise, if the system overhead from the -feature is too high then the rate the kernel samples for NUMA hinting -faults may be controlled by the numa_balancing_scan_period_min_ms, -numa_balancing_scan_delay_ms, numa_balancing_scan_period_max_ms, -numa_balancing_scan_size_mb, and numa_balancing_settle_count sysctls. +feature should be disabled. -numa_balancing_scan_period_min_ms, numa_balancing_scan_delay_ms, numa_balancing_scan_period_max_ms, numa_balancing_scan_size_mb -=============================================================================================================================== +Or NUMA_BALANCING_MEMORY_TIERING to optimize page placement among +different types of memory (represented as different NUMA nodes) to +place the hot pages in the fast memory. This is implemented based on +unmapping and page fault too. +numa_balancing_promote_rate_limit_MBps +====================================== -Automatic NUMA balancing scans tasks address space and unmaps pages to -detect if pages are properly placed or if the data should be migrated to a -memory node local to where the task is running. Every "scan delay" the task -scans the next "scan size" number of pages in its address space. When the -end of the address space is reached the scanner restarts from the beginning. +Too high promotion/demotion throughput between different memory types +may hurt application latency. This can be used to rate limit the +promotion throughput. The per-node max promotion throughput in MB/s +will be limited to be no more than the set value. -In combination, the "scan delay" and "scan size" determine the scan rate. -When "scan delay" decreases, the scan rate increases. The scan delay and -hence the scan rate of every task is adaptive and depends on historical -behaviour. If pages are properly placed then the scan delay increases, -otherwise the scan delay decreases. The "scan size" is not adaptive but -the higher the "scan size", the higher the scan rate. +A rule of thumb is to set this to less than 1/10 of the PMEM node +write bandwidth. -Higher scan rates incur higher system overhead as page faults must be -trapped and potentially data must be migrated. However, the higher the scan -rate, the more quickly a tasks memory is migrated to a local node if the -workload pattern changes and minimises performance impact due to remote -memory accesses. These sysctls control the thresholds for scan delays and -the number of pages scanned. +oops_all_cpu_backtrace +====================== -numa_balancing_scan_period_min_ms is the minimum time in milliseconds to -scan a tasks virtual memory. It effectively controls the maximum scanning -rate for each task. +If this option is set, the kernel will send an NMI to all CPUs to dump +their backtraces when an oops event occurs. It should be used as a last +resort in case a panic cannot be triggered (to protect VMs running, for +example) or kdump can't be collected. This file shows up if CONFIG_SMP +is enabled. -numa_balancing_scan_delay_ms is the starting "scan delay" used for a task -when it initially forks. +0: Won't show all CPUs backtraces when an oops is detected. +This is the default behavior. -numa_balancing_scan_period_max_ms is the maximum time in milliseconds to -scan a tasks virtual memory. It effectively controls the minimum scanning -rate for each task. +1: Will non-maskably interrupt all CPUs and dump their backtraces when +an oops event is detected. -numa_balancing_scan_size_mb is how many megabytes worth of pages are -scanned for a given scan. +oops_limit +========== + +Number of kernel oopses after which the kernel should panic when +``panic_on_oops`` is not set. Setting this to 0 disables checking +the count. Setting this to 1 has the same effect as setting +``panic_on_oops=1``. The default value is 10000. -osrelease, ostype & version: -============================ + +osrelease, ostype & version +=========================== :: @@ -569,15 +757,16 @@ osrelease, ostype & version: # cat version #5 Wed Feb 25 21:49:24 MET 1998 -The files osrelease and ostype should be clear enough. Version +The files ``osrelease`` and ``ostype`` should be clear enough. +``version`` needs a little more clarification however. The '#5' means that this is the fifth kernel built from this source base and the date behind it indicates the time the kernel was built. The only way to tune these values is to rebuild the kernel :-) -overflowgid & overflowuid: -========================== +overflowgid & overflowuid +========================= if your architecture did not always support 32-bit UIDs (i.e. arm, i386, m68k, sh, and sparc32), a fixed UID and GID will be returned to @@ -588,108 +777,129 @@ These sysctls allow you to change the value of the fixed UID and GID. The default is 65534. +panic +===== + +The value in this file determines the behaviour of the kernel on a panic: -====== -The value in this file represents the number of seconds the kernel -waits before rebooting on a panic. When you use the software watchdog, -the recommended setting is 60. +* if zero, the kernel will loop forever; +* if negative, the kernel will reboot immediately; +* if positive, the kernel will reboot after the corresponding number + of seconds. +When you use the software watchdog, the recommended setting is 60. -panic_on_io_nmi: -================ + +panic_on_io_nmi +=============== Controls the kernel's behavior when a CPU receives an NMI caused by an IO error. -0: try to continue operation (default) += ================================================================== +0 Try to continue operation (default). +1 Panic immediately. The IO error triggered an NMI. This indicates a + serious system condition which could result in IO data corruption. + Rather than continuing, panicking might be a better choice. Some + servers issue this sort of NMI when the dump button is pushed, + and you can use this option to take a crash dump. += ================================================================== -1: panic immediately. The IO error triggered an NMI. This indicates a - serious system condition which could result in IO data corruption. - Rather than continuing, panicking might be a better choice. Some - servers issue this sort of NMI when the dump button is pushed, - and you can use this option to take a crash dump. - -panic_on_oops: -============== +panic_on_oops +============= Controls the kernel's behaviour when an oops or BUG is encountered. -0: try to continue operation - -1: panic immediately. If the `panic` sysctl is also non-zero then the - machine will be rebooted. += =================================================================== +0 Try to continue operation. +1 Panic immediately. If the `panic` sysctl is also non-zero then the + machine will be rebooted. += =================================================================== -panic_on_stackoverflow: -======================= +panic_on_stackoverflow +====================== Controls the kernel's behavior when detecting the overflows of kernel, IRQ and exception stacks except a user stack. -This file shows up if CONFIG_DEBUG_STACKOVERFLOW is enabled. - -0: try to continue operation. +This file shows up if ``CONFIG_DEBUG_STACKOVERFLOW`` is enabled. -1: panic immediately. += ========================== +0 Try to continue operation. +1 Panic immediately. += ========================== -panic_on_unrecovered_nmi: -========================= +panic_on_unrecovered_nmi +======================== The default Linux behaviour on an NMI of either memory or unknown is to continue operation. For many environments such as scientific computing it is preferable that the box is taken out and the error dealt with than an uncorrected parity/ECC error get propagated. -A small number of systems do generate NMI's for bizarre random reasons +A small number of systems do generate NMIs for bizarre random reasons such as power management so the default is off. That sysctl works like the existing panic controls already in that directory. -panic_on_warn: -============== +panic_on_warn +============= Calls panic() in the WARN() path when set to 1. This is useful to avoid a kernel rebuild when attempting to kdump at the location of a WARN(). -0: only WARN(), default behaviour. += ================================================ +0 Only WARN(), default behaviour. +1 Call panic() after printing out WARN() location. += ================================================ -1: call panic() after printing out WARN() location. - -panic_print: -============ +panic_print +=========== Bitmask for printing system info when panic happens. User can chose combination of the following bits: -===== ======================================== +===== ============================================ bit 0 print all tasks info bit 1 print system memory info bit 2 print timer info -bit 3 print locks info if CONFIG_LOCKDEP is on +bit 3 print locks info if ``CONFIG_LOCKDEP`` is on bit 4 print ftrace buffer -===== ======================================== +bit 5 print all printk messages in buffer +bit 6 print all CPUs backtrace (if available in the arch) +bit 7 print only tasks in uninterruptible (blocked) state +===== ============================================ So for example to print tasks and memory info on panic, user can:: echo 3 > /proc/sys/kernel/panic_print -panic_on_rcu_stall: -=================== +panic_on_rcu_stall +================== When set to 1, calls panic() after RCU stall detection messages. This is useful to define the root cause of RCU stalls using a vmcore. -0: do not panic() when RCU stall takes place, default behavior. += ============================================================ +0 Do not panic() when RCU stall takes place, default behavior. +1 panic() after printing RCU stall messages. += ============================================================ -1: panic() after printing RCU stall messages. +max_rcu_stall_to_panic +====================== +When ``panic_on_rcu_stall`` is set to 1, this value determines the +number of times that RCU can stall before panic() is called. -perf_cpu_time_max_percent: -========================== +When ``panic_on_rcu_stall`` is set to 0, this value is has no effect. + +perf_cpu_time_max_percent +========================= Hints to the kernel how much CPU time it should be allowed to use to handle perf sampling events. If the perf subsystem @@ -702,171 +912,241 @@ unexpectedly take too long to execute, the NMIs can become stacked up next to each other so much that nothing else is allowed to execute. -0: - disable the mechanism. Do not monitor or correct perf's - sampling rate no matter how CPU time it takes. +===== ======================================================== +0 Disable the mechanism. Do not monitor or correct perf's + sampling rate no matter how CPU time it takes. -1-100: - attempt to throttle perf's sample rate to this - percentage of CPU. Note: the kernel calculates an - "expected" length of each sample event. 100 here means - 100% of that expected length. Even if this is set to - 100, you may still see sample throttling if this - length is exceeded. Set to 0 if you truly do not care - how much CPU is consumed. +1-100 Attempt to throttle perf's sample rate to this + percentage of CPU. Note: the kernel calculates an + "expected" length of each sample event. 100 here means + 100% of that expected length. Even if this is set to + 100, you may still see sample throttling if this + length is exceeded. Set to 0 if you truly do not care + how much CPU is consumed. +===== ======================================================== -perf_event_paranoid: -==================== +perf_event_paranoid +=================== Controls use of the performance events system by unprivileged -users (without CAP_SYS_ADMIN). The default value is 2. +users (without CAP_PERFMON). The default value is 2. + +For backward compatibility reasons access to system performance +monitoring and observability remains open for CAP_SYS_ADMIN +privileged processes but CAP_SYS_ADMIN usage for secure system +performance monitoring and observability operations is discouraged +with respect to CAP_PERFMON use cases. === ================================================================== - -1 Allow use of (almost) all events by all users + -1 Allow use of (almost) all events by all users. - Ignore mlock limit after perf_event_mlock_kb without CAP_IPC_LOCK + Ignore mlock limit after perf_event_mlock_kb without + ``CAP_IPC_LOCK``. ->=0 Disallow ftrace function tracepoint by users without CAP_SYS_ADMIN +>=0 Disallow ftrace function tracepoint by users without + ``CAP_PERFMON``. - Disallow raw tracepoint access by users without CAP_SYS_ADMIN + Disallow raw tracepoint access by users without ``CAP_PERFMON``. ->=1 Disallow CPU event access by users without CAP_SYS_ADMIN +>=1 Disallow CPU event access by users without ``CAP_PERFMON``. ->=2 Disallow kernel profiling by users without CAP_SYS_ADMIN +>=2 Disallow kernel profiling by users without ``CAP_PERFMON``. === ================================================================== -perf_event_max_stack: -===================== +perf_event_max_stack +==================== -Controls maximum number of stack frames to copy for (attr.sample_type & -PERF_SAMPLE_CALLCHAIN) configured events, for instance, when using -'perf record -g' or 'perf trace --call-graph fp'. +Controls maximum number of stack frames to copy for (``attr.sample_type & +PERF_SAMPLE_CALLCHAIN``) configured events, for instance, when using +'``perf record -g``' or '``perf trace --call-graph fp``'. This can only be done when no events are in use that have callchains -enabled, otherwise writing to this file will return -EBUSY. +enabled, otherwise writing to this file will return ``-EBUSY``. The default value is 127. -perf_event_mlock_kb: -==================== +perf_event_mlock_kb +=================== -Control size of per-cpu ring buffer not counted agains mlock limit. +Control size of per-cpu ring buffer not counted against mlock limit. The default value is 512 + 1 page -perf_event_max_contexts_per_stack: -================================== +perf_event_max_contexts_per_stack +================================= Controls maximum number of stack frame context entries for -(attr.sample_type & PERF_SAMPLE_CALLCHAIN) configured events, for -instance, when using 'perf record -g' or 'perf trace --call-graph fp'. +(``attr.sample_type & PERF_SAMPLE_CALLCHAIN``) configured events, for +instance, when using '``perf record -g``' or '``perf trace --call-graph fp``'. This can only be done when no events are in use that have callchains -enabled, otherwise writing to this file will return -EBUSY. +enabled, otherwise writing to this file will return ``-EBUSY``. The default value is 8. -pid_max: -======== +perf_user_access (arm64 and riscv only) +======================================= + +Controls user space access for reading perf event counters. + +arm64 +===== + +The default value is 0 (access disabled). + +When set to 1, user space can read performance monitor counter registers +directly. + +See Documentation/arch/arm64/perf.rst for more information. + +riscv +===== + +When set to 0, user space access is disabled. + +The default value is 1, user space can read performance monitor counter +registers through perf, any direct access without perf intervention will trigger +an illegal instruction. + +When set to 2, which enables legacy mode (user space has direct access to cycle +and insret CSRs only). Note that this legacy value is deprecated and will be +removed once all user space applications are fixed. + +Note that the time CSR is always directly accessible to all modes. + +pid_max +======= PID allocation wrap value. When the kernel's next PID value reaches this value, it wraps back to a minimum PID value. -PIDs of value pid_max or larger are not allocated. +PIDs of value ``pid_max`` or larger are not allocated. -ns_last_pid: -============ +ns_last_pid +=========== The last pid allocated in the current (the one task using this sysctl lives in) pid namespace. When selecting a pid for a next task on fork kernel tries to allocate a number starting from this one. -powersave-nap: (PPC only) -========================= +powersave-nap (PPC only) +======================== If set, Linux-PPC will use the 'nap' mode of powersaving, otherwise the 'doze' mode will be used. + ============================================================== -printk: -======= +printk +====== -The four values in printk denote: console_loglevel, -default_message_loglevel, minimum_console_loglevel and -default_console_loglevel respectively. +The four values in printk denote: ``console_loglevel``, +``default_message_loglevel``, ``minimum_console_loglevel`` and +``default_console_loglevel`` respectively. These values influence printk() behavior when printing or -logging error messages. See 'man 2 syslog' for more info on +logging error messages. See '``man 2 syslog``' for more info on the different loglevels. -- console_loglevel: - messages with a higher priority than - this will be printed to the console -- default_message_loglevel: - messages without an explicit priority - will be printed with this priority -- minimum_console_loglevel: - minimum (highest) value to which - console_loglevel can be set -- default_console_loglevel: - default value for console_loglevel +======================== ===================================== +console_loglevel messages with a higher priority than + this will be printed to the console +default_message_loglevel messages without an explicit priority + will be printed with this priority +minimum_console_loglevel minimum (highest) value to which + console_loglevel can be set +default_console_loglevel default value for console_loglevel +======================== ===================================== -printk_delay: -============= +printk_delay +============ -Delay each printk message in printk_delay milliseconds +Delay each printk message in ``printk_delay`` milliseconds Value from 0 - 10000 is allowed. -printk_ratelimit: -================= +printk_ratelimit +================ -Some warning messages are rate limited. printk_ratelimit specifies +Some warning messages are rate limited. ``printk_ratelimit`` specifies the minimum length of time between these messages (in seconds). The default value is 5 seconds. A value of 0 will disable rate limiting. -printk_ratelimit_burst: -======================= +printk_ratelimit_burst +====================== -While long term we enforce one message per printk_ratelimit +While long term we enforce one message per `printk_ratelimit`_ seconds, we do allow a burst of messages to pass through. -printk_ratelimit_burst specifies the number of messages we can +``printk_ratelimit_burst`` specifies the number of messages we can send before ratelimiting kicks in. The default value is 10 messages. -printk_devkmsg: -=============== - -Control the logging to /dev/kmsg from userspace: - -ratelimit: - default, ratelimited +printk_devkmsg +============== -on: unlimited logging to /dev/kmsg from userspace +Control the logging to ``/dev/kmsg`` from userspace: -off: logging to /dev/kmsg disabled +========= ============================================= +ratelimit default, ratelimited +on unlimited logging to /dev/kmsg from userspace +off logging to /dev/kmsg disabled +========= ============================================= -The kernel command line parameter printk.devkmsg= overrides this and is +The kernel command line parameter ``printk.devkmsg=`` overrides this and is a one-time setting until next reboot: once set, it cannot be changed by this sysctl interface anymore. +============================================================== -randomize_va_space: -=================== + +pty +=== + +See Documentation/filesystems/devpts.rst. + + +random +====== + +This is a directory, with the following entries: + +* ``boot_id``: a UUID generated the first time this is retrieved, and + unvarying after that; + +* ``uuid``: a UUID generated every time this is retrieved (this can + thus be used to generate UUIDs at will); + +* ``entropy_avail``: the pool's entropy count, in bits; + +* ``poolsize``: the entropy pool size, in bits; + +* ``urandom_min_reseed_secs``: obsolete (used to determine the minimum + number of seconds between urandom pool reseeding). This file is + writable for compatibility purposes, but writing to it has no effect + on any RNG behavior; + +* ``write_wakeup_threshold``: when the entropy count drops below this + (as a number of bits), processes waiting to write to ``/dev/random`` + are woken up. This file is writable for compatibility purposes, but + writing to it has no effect on any RNG behavior. + + +randomize_va_space +================== This option can be used to select the type of process address space randomization that is used in the system, for architectures @@ -881,10 +1161,10 @@ that support this feature. This, among other things, implies that shared libraries will be loaded to random addresses. Also for PIE-linked binaries, the location of code start is randomized. This is the default if the - CONFIG_COMPAT_BRK option is enabled. + ``CONFIG_COMPAT_BRK`` option is enabled. 2 Additionally enable heap randomization. This is the default if - CONFIG_COMPAT_BRK is disabled. + ``CONFIG_COMPAT_BRK`` is disabled. There are a few legacy applications out there (such as some ancient versions of libc.so.5 from 1996) that assume that brk area starts @@ -894,109 +1174,185 @@ that support this feature. systems it is safe to choose full randomization. Systems with ancient and/or broken binaries should be configured - with CONFIG_COMPAT_BRK enabled, which excludes the heap from process + with ``CONFIG_COMPAT_BRK`` enabled, which excludes the heap from process address space randomization. == =========================================================================== -reboot-cmd: (Sparc only) -======================== - -??? This seems to be a way to give an argument to the Sparc -ROM/Flash boot loader. Maybe to tell it what to do after -rebooting. ??? +real-root-dev +============= +See Documentation/admin-guide/initrd.rst. -rtsig-max & rtsig-nr: -===================== -The file rtsig-max can be used to tune the maximum number -of POSIX realtime (queued) signals that can be outstanding -in the system. +reboot-cmd (SPARC only) +======================= -rtsig-nr shows the number of RT signals currently queued. +??? This seems to be a way to give an argument to the Sparc +ROM/Flash boot loader. Maybe to tell it what to do after +rebooting. ??? -sched_energy_aware: -=================== +sched_energy_aware +================== Enables/disables Energy Aware Scheduling (EAS). EAS starts automatically on platforms where it can run (that is, platforms with asymmetric CPU topologies and having an Energy Model available). If your platform happens to meet the requirements for EAS but you do not want to use it, change -this value to 0. +this value to 0. On Non-EAS platforms, write operation fails and +read doesn't return anything. +task_delayacct +=============== -sched_schedstats: -================= +Enables/disables task delay accounting (see +Documentation/accounting/delay-accounting.rst. Enabling this feature incurs +a small amount of overhead in the scheduler but is useful for debugging +and performance tuning. It is required by some tools such as iotop. + +sched_schedstats +================ Enables/disables scheduler statistics. Enabling this feature incurs a small amount of overhead in the scheduler but is useful for debugging and performance tuning. +sched_util_clamp_min +==================== -sg-big-buff: -============ +Max allowed *minimum* utilization. + +Default value is 1024, which is the maximum possible value. + +It means that any requested uclamp.min value cannot be greater than +sched_util_clamp_min, i.e., it is restricted to the range +[0:sched_util_clamp_min]. + +sched_util_clamp_max +==================== + +Max allowed *maximum* utilization. + +Default value is 1024, which is the maximum possible value. + +It means that any requested uclamp.max value cannot be greater than +sched_util_clamp_max, i.e., it is restricted to the range +[0:sched_util_clamp_max]. + +sched_util_clamp_min_rt_default +=============================== + +By default Linux is tuned for performance. Which means that RT tasks always run +at the highest frequency and most capable (highest capacity) CPU (in +heterogeneous systems). + +Uclamp achieves this by setting the requested uclamp.min of all RT tasks to +1024 by default, which effectively boosts the tasks to run at the highest +frequency and biases them to run on the biggest CPU. + +This knob allows admins to change the default behavior when uclamp is being +used. In battery powered devices particularly, running at the maximum +capacity and frequency will increase energy consumption and shorten the battery +life. + +This knob is only effective for RT tasks which the user hasn't modified their +requested uclamp.min value via sched_setattr() syscall. + +This knob will not escape the range constraint imposed by sched_util_clamp_min +defined above. + +For example if + + sched_util_clamp_min_rt_default = 800 + sched_util_clamp_min = 600 + +Then the boost will be clamped to 600 because 800 is outside of the permissible +range of [0:600]. This could happen for instance if a powersave mode will +restrict all boosts temporarily by modifying sched_util_clamp_min. As soon as +this restriction is lifted, the requested sched_util_clamp_min_rt_default +will take effect. + +seccomp +======= + +See Documentation/userspace-api/seccomp_filter.rst. + + +sg-big-buff +=========== This file shows the size of the generic SCSI (sg) buffer. You can't tune it just yet, but you could change it on -compile time by editing include/scsi/sg.h and changing -the value of SG_BIG_BUFF. +compile time by editing ``include/scsi/sg.h`` and changing +the value of ``SG_BIG_BUFF``. There shouldn't be any reason to change this value. If you can come up with one, you probably know what you are doing anyway :) -shmall: -======= +shmall +====== -This parameter sets the total amount of shared memory pages that -can be used system wide. Hence, SHMALL should always be at least -ceil(shmmax/PAGE_SIZE). +This parameter sets the total amount of shared memory pages that can be used +inside ipc namespace. The shared memory pages counting occurs for each ipc +namespace separately and is not inherited. Hence, ``shmall`` should always be at +least ``ceil(shmmax/PAGE_SIZE)``. -If you are not sure what the default PAGE_SIZE is on your Linux -system, you can run the following command: +If you are not sure what the default ``PAGE_SIZE`` is on your Linux +system, you can run the following command:: # getconf PAGE_SIZE +To reduce or disable the ability to allocate shared memory, you must create a +new ipc namespace, set this parameter to the required value and prohibit the +creation of a new ipc namespace in the current user namespace or cgroups can +be used. -shmmax: -======= +shmmax +====== This value can be used to query and set the run time limit on the maximum shared memory segment size that can be created. Shared memory segments up to 1Gb are now supported in the -kernel. This value defaults to SHMMAX. +kernel. This value defaults to ``SHMMAX``. -shm_rmid_forced: -================ +shmmni +====== + +This value determines the maximum number of shared memory segments. +4096 by default (``SHMMNI``). + + +shm_rmid_forced +=============== Linux lets you set resource limits, including how much memory one -process can consume, via setrlimit(2). Unfortunately, shared memory +process can consume, via ``setrlimit(2)``. Unfortunately, shared memory segments are allowed to exist without association with any process, and thus might not be counted against any resource limits. If enabled, shared memory segments are automatically destroyed when their attach count becomes zero after a detach or a process termination. It will also destroy segments that were created, but never attached to, on exit -from the process. The only use left for IPC_RMID is to immediately +from the process. The only use left for ``IPC_RMID`` is to immediately destroy an unattached segment. Of course, this breaks the way things are defined, so some applications might stop working. Note that this feature will do you no good unless you also configure your resource -limits (in particular, RLIMIT_AS and RLIMIT_NPROC). Most systems don't +limits (in particular, ``RLIMIT_AS`` and ``RLIMIT_NPROC``). Most systems don't need this. Note that if you change this from 0 to 1, already created segments without users and with a dead originative process will be destroyed. -sysctl_writes_strict: -===================== +sysctl_writes_strict +==================== Control how file position affects the behavior of updating sysctl values -via the /proc/sys interface: +via the ``/proc/sys`` interface: == ====================================================================== -1 Legacy per-write sysctl value handling, with no printk warnings. @@ -1013,8 +1369,8 @@ via the /proc/sys interface: == ====================================================================== -softlockup_all_cpu_backtrace: -============================= +softlockup_all_cpu_backtrace +============================ This value controls the soft lockup detector thread's behavior when a soft lockup condition is detected as to whether or not @@ -1024,43 +1380,103 @@ be issued an NMI and instructed to capture stack trace. This feature is only applicable for architectures which support NMI. -0: do nothing. This is the default behavior. += ============================================ +0 Do nothing. This is the default behavior. +1 On detection capture more debug information. += ============================================ -1: on detection capture more debug information. +softlockup_panic +================= -soft_watchdog: -============== +This parameter can be used to control whether the kernel panics +when a soft lockup is detected. + += ============================================ +0 Don't panic on soft lockup. +1 Panic on soft lockup. += ============================================ + +This can also be set using the softlockup_panic kernel parameter. -This parameter can be used to control the soft lockup detector. - 0 - disable the soft lockup detector +soft_watchdog +============= + +This parameter can be used to control the soft lockup detector. - 1 - enable the soft lockup detector += ================================= +0 Disable the soft lockup detector. +1 Enable the soft lockup detector. += ================================= The soft lockup detector monitors CPUs for threads that are hogging the CPUs -without rescheduling voluntarily, and thus prevent the 'watchdog/N' threads -from running. The mechanism depends on the CPUs ability to respond to timer -interrupts which are needed for the 'watchdog/N' threads to be woken up by -the watchdog timer function, otherwise the NMI watchdog - if enabled - can -detect a hard lockup condition. +without rescheduling voluntarily, and thus prevent the 'migration/N' threads +from running, causing the watchdog work fail to execute. The mechanism depends +on the CPUs ability to respond to timer interrupts which are needed for the +watchdog work to be queued by the watchdog timer function, otherwise the NMI +watchdog — if enabled — can detect a hard lockup condition. -stack_erasing: -============== +split_lock_mitigate (x86 only) +============================== + +On x86, each "split lock" imposes a system-wide performance penalty. On larger +systems, large numbers of split locks from unprivileged users can result in +denials of service to well-behaved and potentially more important users. + +The kernel mitigates these bad users by detecting split locks and imposing +penalties: forcing them to wait and only allowing one core to execute split +locks at a time. + +These mitigations can make those bad applications unbearably slow. Setting +split_lock_mitigate=0 may restore some application performance, but will also +increase system exposure to denial of service attacks from split lock users. + += =================================================================== +0 Disable the mitigation mode - just warns the split lock on kernel log + and exposes the system to denials of service from the split lockers. +1 Enable the mitigation mode (this is the default) - penalizes the split + lockers with intentional performance degradation. += =================================================================== + + +stack_erasing +============= This parameter can be used to control kernel stack erasing at the end -of syscalls for kernels built with CONFIG_GCC_PLUGIN_STACKLEAK. +of syscalls for kernels built with ``CONFIG_GCC_PLUGIN_STACKLEAK``. That erasing reduces the information which kernel stack leak bugs can reveal and blocks some uninitialized stack variable attacks. The tradeoff is the performance impact: on a single CPU system kernel compilation sees a 1% slowdown, other systems and workloads may vary. - 0: kernel stack erasing is disabled, STACKLEAK_METRICS are not updated. += ==================================================================== +0 Kernel stack erasing is disabled, STACKLEAK_METRICS are not updated. +1 Kernel stack erasing is enabled (default), it is performed before + returning to the userspace at the end of syscalls. += ==================================================================== + + +stop-a (SPARC only) +=================== + +Controls Stop-A: + += ==================================== +0 Stop-A has no effect. +1 Stop-A breaks to the PROM (default). += ==================================== + +Stop-A is always enabled on a panic, so that the user can return to +the boot PROM. - 1: kernel stack erasing is enabled (default), it is performed before - returning to the userspace at the end of syscalls. + +sysrq +===== + +See Documentation/admin-guide/sysrq.rst. tainted @@ -1072,7 +1488,7 @@ ORed together. The letters are seen in "Tainted" line of Oops reports. ====== ===== ============================================================== 1 `(P)` proprietary module was loaded 2 `(F)` module was force loaded - 4 `(S)` SMP kernel oops on an officially SMP incapable processor + 4 `(S)` kernel running on an out of specification system 8 `(R)` module was force unloaded 16 `(M)` processor reported a Machine Check Exception (MCE) 32 `(B)` bad page referenced or some unexpected page flags @@ -1092,28 +1508,79 @@ ORed together. The letters are seen in "Tainted" line of Oops reports. See Documentation/admin-guide/tainted-kernels.rst for more information. +Note: + writes to this sysctl interface will fail with ``EINVAL`` if the kernel is + booted with the command line option ``panic_on_taint=<bitmask>,nousertaint`` + and any of the ORed together values being written to ``tainted`` match with + the bitmask declared on panic_on_taint. + See Documentation/admin-guide/kernel-parameters.rst for more details on + that particular kernel command line option and its optional + ``nousertaint`` switch. -threads-max: -============ +threads-max +=========== This value controls the maximum number of threads that can be created -using fork(). +using ``fork()``. During initialization the kernel sets this value such that even if the maximum number of threads is created, the thread structures occupy only a part (1/8th) of the available RAM pages. -The minimum value that can be written to threads-max is 1. +The minimum value that can be written to ``threads-max`` is 1. -The maximum value that can be written to threads-max is given by the -constant FUTEX_TID_MASK (0x3fffffff). +The maximum value that can be written to ``threads-max`` is given by the +constant ``FUTEX_TID_MASK`` (0x3fffffff). -If a value outside of this range is written to threads-max an error -EINVAL occurs. +If a value outside of this range is written to ``threads-max`` an +``EINVAL`` error occurs. -unknown_nmi_panic: -================== +traceoff_on_warning +=================== + +When set, disables tracing (see Documentation/trace/ftrace.rst) when a +``WARN()`` is hit. + + +tracepoint_printk +================= + +When tracepoints are sent to printk() (enabled by the ``tp_printk`` +boot parameter), this entry provides runtime control:: + + echo 0 > /proc/sys/kernel/tracepoint_printk + +will stop tracepoints from being sent to printk(), and:: + + echo 1 > /proc/sys/kernel/tracepoint_printk + +will send them to printk() again. + +This only works if the kernel was booted with ``tp_printk`` enabled. + +See Documentation/admin-guide/kernel-parameters.rst and +Documentation/trace/boottime-trace.rst. + + +unaligned-trap +============== + +On architectures where unaligned accesses cause traps, and where this +feature is supported (``CONFIG_SYSCTL_ARCH_UNALIGN_ALLOW``; currently, +``arc``, ``parisc`` and ``loongarch``), controls whether unaligned traps +are caught and emulated (instead of failing). + += ======================================================== +0 Do not emulate unaligned accesses. +1 Emulate unaligned accesses. This is the default setting. += ======================================================== + +See also `ignore-unaligned-usertrap`_. + + +unknown_nmi_panic +================= The value in this file affects behavior of handling NMI. When the value is non-zero, unknown NMI is trapped and then panic occurs. At @@ -1123,37 +1590,70 @@ NMI switch that most IA32 servers have fires unknown NMI up, for example. If a system hangs up, try pressing the NMI switch. -watchdog: -========= +unprivileged_bpf_disabled +========================= + +Writing 1 to this entry will disable unprivileged calls to ``bpf()``; +once disabled, calling ``bpf()`` without ``CAP_SYS_ADMIN`` or ``CAP_BPF`` +will return ``-EPERM``. Once set to 1, this can't be cleared from the +running kernel anymore. + +Writing 2 to this entry will also disable unprivileged calls to ``bpf()``, +however, an admin can still change this setting later on, if needed, by +writing 0 or 1 to this entry. + +If ``BPF_UNPRIV_DEFAULT_OFF`` is enabled in the kernel config, then this +entry will default to 2 instead of 0. + += ============================================================= +0 Unprivileged calls to ``bpf()`` are enabled +1 Unprivileged calls to ``bpf()`` are disabled without recovery +2 Unprivileged calls to ``bpf()`` are disabled += ============================================================= -This parameter can be used to disable or enable the soft lockup detector -_and_ the NMI watchdog (i.e. the hard lockup detector) at the same time. - 0 - disable both lockup detectors +warn_limit +========== - 1 - enable both lockup detectors +Number of kernel warnings after which the kernel should panic when +``panic_on_warn`` is not set. Setting this to 0 disables checking +the warning count. Setting this to 1 has the same effect as setting +``panic_on_warn=1``. The default value is 0. + + +watchdog +======== + +This parameter can be used to disable or enable the soft lockup detector +*and* the NMI watchdog (i.e. the hard lockup detector) at the same time. + += ============================== +0 Disable both lockup detectors. +1 Enable both lockup detectors. += ============================== The soft lockup detector and the NMI watchdog can also be disabled or -enabled individually, using the soft_watchdog and nmi_watchdog parameters. -If the watchdog parameter is read, for example by executing:: +enabled individually, using the ``soft_watchdog`` and ``nmi_watchdog`` +parameters. +If the ``watchdog`` parameter is read, for example by executing:: cat /proc/sys/kernel/watchdog -the output of this command (0 or 1) shows the logical OR of soft_watchdog -and nmi_watchdog. +the output of this command (0 or 1) shows the logical OR of +``soft_watchdog`` and ``nmi_watchdog``. -watchdog_cpumask: -================= +watchdog_cpumask +================ This value can be used to control on which cpus the watchdog may run. -The default cpumask is all possible cores, but if NO_HZ_FULL is +The default cpumask is all possible cores, but if ``NO_HZ_FULL`` is enabled in the kernel config, and cores are specified with the -nohz_full= boot argument, those cores are excluded by default. +``nohz_full=`` boot argument, those cores are excluded by default. Offline cores can be included in this mask, and if the core is later brought online, the watchdog will be started based on the mask value. -Typically this value would only be touched in the nohz_full case +Typically this value would only be touched in the ``nohz_full`` case to re-enable cores that by default were not running the watchdog, if a kernel lockup was suspected on those cores. @@ -1164,12 +1664,12 @@ might say:: echo 0,2-4 > /proc/sys/kernel/watchdog_cpumask -watchdog_thresh: -================ +watchdog_thresh +=============== This value can be used to control the frequency of hrtimer and NMI events and the soft and hard lockup thresholds. The default threshold is 10 seconds. -The softlockup threshold is (2 * watchdog_thresh). Setting this +The softlockup threshold is (``2 * watchdog_thresh``). Setting this tunable to zero will disable lockup detection altogether. diff --git a/Documentation/admin-guide/sysctl/net.rst b/Documentation/admin-guide/sysctl/net.rst index 287b98708a40..7250c0542828 100644 --- a/Documentation/admin-guide/sysctl/net.rst +++ b/Documentation/admin-guide/sysctl/net.rst @@ -31,17 +31,18 @@ see only some of them, depending on your kernel's configuration. Table : Subdirectories in /proc/sys/net - ========= =================== = ========== ================== + ========= =================== = ========== =================== Directory Content Directory Content - ========= =================== = ========== ================== - core General parameter appletalk Appletalk protocol - unix Unix domain sockets netrom NET/ROM - 802 E802 protocol ax25 AX25 - ethernet Ethernet protocol rose X.25 PLP layer + ========= =================== = ========== =================== + 802 E802 protocol mptcp Multipath TCP + appletalk Appletalk protocol netfilter Network Filter + ax25 AX25 netrom NET/ROM + bridge Bridging rose X.25 PLP layer + core General parameter tipc TIPC + ethernet Ethernet protocol unix Unix domain sockets ipv4 IP version 4 x25 X.25 protocol - bridge Bridging decnet DEC net - ipv6 IP version 6 tipc TIPC - ========= =================== = ========== ================== + ipv6 IP version 6 + ========= =================== = ========== =================== 1. /proc/sys/net/core - Network core options ============================================ @@ -64,15 +65,17 @@ two flavors of JITs, the newer eBPF JIT currently supported on: - arm64 - arm32 - ppc64 + - ppc32 - sparc64 - mips64 - s390x - - riscv + - riscv64 + - riscv32 + - loongarch64 And the older cBPF JIT supported on the following archs: - mips - - ppc - sparc eBPF JITs are a superset of cBPF JITs, meaning the kernel will @@ -100,6 +103,9 @@ Values: - 1 - enable JIT hardening for unprivileged users only - 2 - enable JIT hardening for all users +where "privileged user" in this context means a process having +CAP_BPF or CAP_SYS_ADMIN in the root user name space. + bpf_jit_kallsyms ---------------- @@ -200,6 +206,11 @@ Will increase power usage. Default: 0 (off) +mem_pcpu_rsv +------------ + +Per-cpu reserved forward alloc cache size in page units. Default 1MB per CPU. + rmem_default ------------ @@ -210,6 +221,12 @@ rmem_max The maximum receive socket buffer size in bytes. +rps_default_mask +---------------- + +The default RPS CPU mask used on newly created network devices. An empty +mask means RPS disabled by default. + tstamp_allow_data ----------------- Allow processes to receive tx timestamps looped together with the original @@ -270,7 +287,7 @@ poll cycle or the number of packets processed reaches netdev_budget. netdev_max_backlog ------------------ -Maximum number of packets, queued on the INPUT side, when the interface +Maximum number of packets, queued on the INPUT side, when the interface receives packets faster than kernel can process them. netdev_rss_key @@ -310,21 +327,52 @@ permit to distribute the load on several cpus. If set to 1 (default), timestamps are sampled as soon as possible, before queueing. +netdev_unregister_timeout_secs +------------------------------ + +Unregister network device timeout in seconds. +This option controls the timeout (in seconds) used to issue a warning while +waiting for a network device refcount to drop to 0 during device +unregistration. A lower value may be useful during bisection to detect +a leaked reference faster. A larger value may be useful to prevent false +warnings on slow/loaded systems. +Default value is 10, minimum 1, maximum 3600. + +skb_defer_max +------------- + +Max size (in skbs) of the per-cpu list of skbs being freed +by the cpu which allocated them. Used by TCP stack so far. + +Default: 64 + optmem_max ---------- Maximum ancillary buffer size allowed per socket. Ancillary data is a sequence -of struct cmsghdr structures with appended data. +of struct cmsghdr structures with appended data. TCP tx zerocopy also uses +optmem_max as a limit for its internal structures. + +Default : 128 KB fb_tunnels_only_for_init_net ---------------------------- Controls if fallback tunnels (like tunl0, gre0, gretap0, erspan0, -sit0, ip6tnl0, ip6gre0) are automatically created when a new -network namespace is created, if corresponding tunnel is present -in initial network namespace. -If set to 1, these devices are not automatically created, and -user space is responsible for creating them if needed. +sit0, ip6tnl0, ip6gre0) are automatically created. There are 3 possibilities +(a) value = 0; respective fallback tunnels are created when module is +loaded in every net namespaces (backward compatible behavior). +(b) value = 1; [kcmd value: initns] respective fallback tunnels are +created only in init net namespace and every other net namespace will +not have them. +(c) value = 2; [kcmd value: none] fallback tunnels are not created +when a module is loaded in any of the net namespace. Setting value to +"2" is pointless after boot if these modules are built-in, so there is +a kernel command-line option that can change this default. Please refer to +Documentation/admin-guide/kernel-parameters.txt for additional details. + +Not creating fallback tunnels gives control to userspace to create +whatever is needed only and avoid creating devices which are redundant. Default : 0 (for compatibility reasons) @@ -338,10 +386,42 @@ settings from init_net and for IPv6 we reset all settings to default. If set to 1, both IPv4 and IPv6 settings are forced to inherit from current ones in init_net. If set to 2, both IPv4 and IPv6 settings are -forced to reset to their default values. +forced to reset to their default values. If set to 3, both IPv4 and IPv6 +settings are forced to inherit from current ones in the netns where this +new netns has been created. Default : 0 (for compatibility reasons) +txrehash +-------- + +Controls default hash rethink behaviour on socket when SO_TXREHASH option is set +to SOCK_TXREHASH_DEFAULT (i. e. not overridden by setsockopt). + +If set to 1 (default), hash rethink is performed on listening socket. +If set to 0, hash rethink is not performed. + +gro_normal_batch +---------------- + +Maximum number of the segments to batch up on output of GRO. When a packet +exits GRO, either as a coalesced superframe or as an original packet which +GRO has decided not to coalesce, it is placed on a per-NAPI list. This +list is then passed to the stack when the number of segments reaches the +gro_normal_batch limit. + +high_order_alloc_disable +------------------------ + +By default the allocator for page frags tries to use high order pages (order-3 +on x86). While the default behavior gives good results in most cases, some users +might have hit a contention in page allocations/freeing. This was especially +true on older kernels (< 5.14) when high-order pages were not stored on per-cpu +lists. This allows to opt-in for order-0 allocation instead but is now mostly of +historical importance. + +Default: 0 + 2. /proc/sys/net/unix - Parameters for Unix domain sockets ---------------------------------------------------------- @@ -352,8 +432,8 @@ socket's buffer. It will not take effect unless PF_UNIX flag is specified. 3. /proc/sys/net/ipv4 - IPV4 settings ------------------------------------- -Please see: Documentation/networking/ip-sysctl.txt and ipvs-sysctl.txt for -descriptions of these entries. +Please see: Documentation/networking/ip-sysctl.rst and +Documentation/admin-guide/sysctl/net.rst for descriptions of these entries. 4. Appletalk diff --git a/Documentation/admin-guide/sysctl/user.rst b/Documentation/admin-guide/sysctl/user.rst index 650eaa03f15e..c45824589339 100644 --- a/Documentation/admin-guide/sysctl/user.rst +++ b/Documentation/admin-guide/sysctl/user.rst @@ -65,6 +65,12 @@ max_pid_namespaces The maximum number of pid namespaces that any user in the current user namespace may create. +max_time_namespaces +=================== + + The maximum number of time namespaces that any user in the current + user namespace may create. + max_user_namespaces =================== diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentation/admin-guide/sysctl/vm.rst index 64aeee1009ca..c59889de122b 100644 --- a/Documentation/admin-guide/sysctl/vm.rst +++ b/Documentation/admin-guide/sysctl/vm.rst @@ -25,8 +25,8 @@ files can be found in mm/swap.c. Currently, these files are in /proc/sys/vm: - admin_reserve_kbytes -- block_dump - compact_memory +- compaction_proactiveness - compact_unevictable_allowed - dirty_background_bytes - dirty_background_ratio @@ -37,6 +37,7 @@ Currently, these files are in /proc/sys/vm: - dirty_writeback_centisecs - drop_caches - extfrag_threshold +- highmem_is_dirtyable - hugetlb_shm_group - laptop_mode - legacy_va_layout @@ -61,8 +62,9 @@ Currently, these files are in /proc/sys/vm: - overcommit_memory - overcommit_ratio - page-cluster +- page_lock_unfairness - panic_on_oom -- percpu_pagelist_fraction +- percpu_pagelist_high_fraction - stat_interval - stat_refresh - numa_stat @@ -104,13 +106,6 @@ On x86_64 this is about 128MB. Changing this takes effect whenever an application requests memory. -block_dump -========== - -block_dump enables block I/O debugging when set to a nonzero value. More -information on block I/O debugging is in Documentation/admin-guide/laptops/laptop-mode.rst. - - compact_memory ============== @@ -119,6 +114,22 @@ all zones are compacted such that free memory is available in contiguous blocks where possible. This can be important for example in the allocation of huge pages although processes will also directly compact memory as required. +compaction_proactiveness +======================== + +This tunable takes a value in the range [0, 100] with a default value of +20. This tunable determines how aggressively compaction is done in the +background. Write of a non zero value to this tunable will immediately +trigger the proactive compaction. Setting it to 0 disables proactive compaction. + +Note that compaction has a non-trivial system-wide impact as pages +belonging to different processes are moved around, which could also lead +to latency spikes in unsuspecting applications. The kernel employs +various heuristics to avoid wasting CPU cycles if it detects that +proactive compaction is not being effective. + +Be careful when setting it to extreme values like 100, as that may +cause excessive background compaction activity. compact_unevictable_allowed =========================== @@ -128,6 +139,9 @@ allowed to examine the unevictable lru (mlocked pages) for pages to compact. This should be used on systems where stalls for minor page faults are an acceptable trade for large contiguous free memory. Set to 0 to prevent compaction from moving pages that are unevictable. Default value is 1. +On CONFIG_PREEMPT_RT the default value is 0 in order to avoid a page fault, due +to compaction, which would block the task from becoming active until the fault +is resolved. dirty_background_bytes @@ -342,7 +356,7 @@ The lowmem_reserve_ratio is an array. You can see them by reading this file:: But, these values are not used directly. The kernel calculates # of protection pages for each zones from them. These are shown as array of protection pages -in /proc/zoneinfo like followings. (This is an example of x86-64 box). +in /proc/zoneinfo like the following. (This is an example of x86-64 box). Each zone has an array of protection pages like this:: Node 0, zone DMA @@ -408,7 +422,7 @@ While most applications need less than a thousand maps, certain programs, particularly malloc debuggers, may consume lots of them, e.g., up to one or two maps per allocation. -The default value is 65536. +The default value is 65530. memory_failure_early_kill: @@ -419,7 +433,7 @@ a 2bit error in a memory module) is detected in the background by hardware that cannot be handled by the kernel. 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 uptodate copy of the data it will kill to prevent any data +no other up-to-date copy of the data it will kill to prevent any data corruptions from propagating. 1: Kill all processes that have the corrupted and not reloadable page mapped @@ -548,6 +562,43 @@ Change the minimum size of the hugepage pool. See Documentation/admin-guide/mm/hugetlbpage.rst +hugetlb_optimize_vmemmap +======================== + +This knob is not available when the size of 'struct page' (a structure defined +in include/linux/mm_types.h) is not power of two (an unusual system config could +result in this). + +Enable (set to 1) or disable (set to 0) HugeTLB Vmemmap Optimization (HVO). + +Once enabled, the vmemmap pages of subsequent allocation of HugeTLB pages from +buddy allocator will be optimized (7 pages per 2MB HugeTLB page and 4095 pages +per 1GB HugeTLB page), whereas already allocated HugeTLB pages will not be +optimized. When those optimized HugeTLB pages are freed from the HugeTLB pool +to the buddy allocator, the vmemmap pages representing that range needs to be +remapped again and the vmemmap pages discarded earlier need to be rellocated +again. If your use case is that HugeTLB pages are allocated 'on the fly' (e.g. +never explicitly allocating HugeTLB pages with 'nr_hugepages' but only set +'nr_overcommit_hugepages', those overcommitted HugeTLB pages are allocated 'on +the fly') instead of being pulled from the HugeTLB pool, you should weigh the +benefits of memory savings against the more overhead (~2x slower than before) +of allocation or freeing HugeTLB pages between the HugeTLB pool and the buddy +allocator. Another behavior to note is that if the system is under heavy memory +pressure, it could prevent the user from freeing HugeTLB pages from the HugeTLB +pool to the buddy allocator since the allocation of vmemmap pages could be +failed, you have to retry later if your system encounter this situation. + +Once disabled, the vmemmap pages of subsequent allocation of HugeTLB pages from +buddy allocator will not be optimized meaning the extra overhead at allocation +time from buddy allocator disappears, whereas already optimized HugeTLB pages +will not be affected. If you want to make sure there are no optimized HugeTLB +pages, you can set "nr_hugepages" to 0 first and then disable this. Note that +writing 0 to nr_hugepages will make any "in use" HugeTLB pages become surplus +pages. So, those surplus pages are still optimized until they are no longer +in use. You would need to wait for those surplus pages to be released before +there are no optimized pages in the system. + + nr_hugepages_mempolicy ====================== @@ -580,7 +631,7 @@ trimming of allocations is initiated. The default value is 1. -See Documentation/nommu-mmap.txt for more information. +See Documentation/admin-guide/mm/nommu-mmap.rst for more information. numa_zonelist_order @@ -691,8 +742,8 @@ overcommit_memory This value contains a flag that enables memory overcommitment. -When this flag is 0, the kernel attempts to estimate the amount -of free memory left when userspace requests more memory. +When this flag is 0, the kernel compares the userspace memory request +size against total memory plus swap and rejects obvious overcommits. When this flag is 1, the kernel pretends there is always enough memory until it actually runs out. @@ -707,7 +758,7 @@ and don't use much of it. The default value is 0. -See Documentation/vm/overcommit-accounting.rst and +See Documentation/mm/overcommit-accounting.rst and mm/util.c::__vm_enough_memory() for more information. @@ -741,6 +792,14 @@ extra faults and I/O delays for following faults if they would have been part of that consecutive pages readahead would have brought in. +page_lock_unfairness +==================== + +This value determines the number of times that the page lock can be +stolen from under a waiter. After the lock is stolen the number of times +specified in this file (default is 5), the "fair lock handoff" semantics +will apply, and the waiter will only be awakened if the lock can be taken. + panic_on_oom ============ @@ -770,22 +829,24 @@ panic_on_oom=2+kdump gives you very strong tool to investigate why oom happens. You can get snapshot. -percpu_pagelist_fraction -======================== +percpu_pagelist_high_fraction +============================= -This is the fraction of pages at most (high mark pcp->high) in each zone that -are allocated for each per cpu page list. The min value for this is 8. It -means that we don't allow more than 1/8th of pages in each zone to be -allocated in any single per_cpu_pagelist. This entry only changes the value -of hot per cpu pagelists. User can specify a number like 100 to allocate -1/100th of each zone to each per cpu page list. +This is the fraction of pages in each zone that are can be stored to +per-cpu page lists. It is an upper boundary that is divided depending +on the number of online CPUs. The min value for this is 8 which means +that we do not allow more than 1/8th of pages in each zone to be stored +on per-cpu page lists. This entry only changes the value of hot per-cpu +page lists. A user can specify a number like 100 to allocate 1/100th of +each zone between per-cpu lists. -The batch value of each per cpu pagelist is also updated as a result. It is -set to pcp->high/4. The upper limit of batch is (PAGE_SHIFT * 8) +The batch value of each per-cpu page list remains the same regardless of +the value of the high fraction so allocation latencies are unaffected. -The initial value is zero. Kernel does not use this value at boot time to set -the high water marks for each per cpu page list. If the user writes '0' to this -sysctl, it will revert to this default behavior. +The initial value is zero. Kernel uses this value to set the high pcp->high +mark based on the low watermark for the zone and the number of local +online CPUs. If the user writes '0' to this sysctl, it will revert to +this default behavior. stat_interval @@ -828,25 +889,46 @@ tooling to work, you can do:: swappiness ========== -This control is used to define how aggressive the kernel will swap -memory pages. Higher values will increase aggressiveness, lower values -decrease the amount of swap. A value of 0 instructs the kernel not to -initiate swap until the amount of free and file-backed pages is less -than the high water mark in a zone. +This control is used to define the rough relative IO cost of swapping +and filesystem paging, as a value between 0 and 200. At 100, the VM +assumes equal IO cost and will thus apply memory pressure to the page +cache and swap-backed pages equally; lower values signify more +expensive swap IO, higher values indicates cheaper. + +Keep in mind that filesystem IO patterns under memory pressure tend to +be more efficient than swap's random IO. An optimal value will require +experimentation and will also be workload-dependent. The default value is 60. +For in-memory swap, like zram or zswap, as well as hybrid setups that +have swap on faster devices than the filesystem, values beyond 100 can +be considered. For example, if the random IO against the swap device +is on average 2x faster than IO from the filesystem, swappiness should +be 133 (x + 2x = 200, 2x = 133.33). + +At 0, the kernel will not initiate swap until the amount of free and +file-backed pages is less than the high watermark in a zone. + unprivileged_userfaultfd ======================== -This flag controls whether unprivileged users can use the userfaultfd -system calls. Set this to 1 to allow unprivileged users to use the -userfaultfd system calls, or set this to 0 to restrict userfaultfd to only -privileged users (with SYS_CAP_PTRACE capability). +This flag controls the mode in which unprivileged users can use the +userfaultfd system calls. Set this to 0 to restrict unprivileged users +to handle page faults in user mode only. In this case, users without +SYS_CAP_PTRACE must pass UFFD_USER_MODE_ONLY in order for userfaultfd to +succeed. Prohibiting use of userfaultfd for handling faults from kernel +mode may make certain vulnerabilities more difficult to exploit. -The default value is 1. +Set this to 1 to allow unprivileged users to use the userfaultfd system +calls without any restrictions. + +The default value is 0. +Another way to control permissions for userfaultfd is to use +/dev/userfaultfd instead of userfaultfd(2). See +Documentation/admin-guide/mm/userfaultfd.rst. user_reserve_kbytes =================== @@ -898,12 +980,12 @@ allocations, THP and hugetlbfs pages. To make it sensible with respect to the watermark_scale_factor parameter, the unit is in fractions of 10,000. The default value of -15,000 on !DISCONTIGMEM configurations means that up to 150% of the high -watermark will be reclaimed in the event of a pageblock being mixed due -to fragmentation. The level of reclaim is determined by the number of -fragmentation events that occurred in the recent past. If this value is -smaller than a pageblock then a pageblocks worth of pages will be reclaimed -(e.g. 2MB on 64-bit x86). A boost factor of 0 will disable the feature. +15,000 means that up to 150% of the high watermark will be reclaimed in the +event of a pageblock being mixed due to fragmentation. The level of reclaim +is determined by the number of fragmentation events that occurred in the +recent past. If this value is smaller than a pageblock then a pageblocks +worth of pages will be reclaimed (e.g. 2MB on 64-bit x86). A boost factor +of 0 will disable the feature. watermark_scale_factor @@ -915,7 +997,7 @@ how much memory needs to be free before kswapd goes back to sleep. The unit is in fractions of 10,000. The default value of 10 means the distances between watermarks are 0.1% of the available memory in the -node/system. The maximum value is 1000, or 10% of memory. +node/system. The maximum value is 3000, or 30% of memory. A high rate of threads entering direct reclaim (allocstall) or kswapd going to sleep prematurely (kswapd_low_wmark_hit_quickly) can indicate @@ -945,11 +1027,11 @@ that benefit from having their data cached, zone_reclaim_mode should be left disabled as the caching effect is likely to be more important than data locality. -zone_reclaim may be enabled if it's known that the workload is partitioned -such that each partition fits within a NUMA node and that accessing remote -memory would cause a measurable performance reduction. The page allocator -will then reclaim easily reusable pages (those page cache pages that are -currently not used) before allocating off node pages. +Consider enabling one or more zone_reclaim mode bits if it's known that the +workload is partitioned such that each partition fits within a NUMA node +and that accessing remote memory would cause a measurable performance +reduction. The page allocator will take additional actions before +allocating off node pages. Allowing zone reclaim to write out pages stops processes that are writing large amounts of data from dirtying pages on other nodes. Zone |