Lines Matching +full:keep +full:- +full:power +full:- +full:in +full:- +full:suspend
5 .. See scripts/check-sysctl-docs to keep this up to date
12 For general info and legal blurb, please look in
13 Documentation/admin-guide/sysctl/index.rst.
15 ------------------------------------------------------------------------------
17 This file contains documentation for the sysctl files in
20 The files in this directory can be used to tune and monitor
21 miscellaneous and general things in the operation of the Linux
27 show up in ``/proc/sys/kernel``:
39 If BSD-style process accounting is enabled these values control
43 how often do we check the amount of free space (value is in
50 That is, suspend accounting if free space drops below 2%; resume it
58 See Documentation/power/video.rst. This allows the video resume mode to be set,
59 in a similar fashion to the ``acpi_sleep`` kernel parameter, by
71 The machine hardware name, the same output as ``uname -m``
77 This variable has no effect and may be removed in future kernel
92 ``type_of_loader`` field in the kernel header; the encoding is kept for
97 See the ``type_of_loader`` and ``ext_loader_type`` fields in
104 The complete bootloader version number. In the example above, this
107 See the ``type_of_loader`` and ``ext_loader_ver`` fields in
116 statistics causes a slight reduction in performance on each program
129 Ctrl-Alt-Delete). Writing a value to this file which doesn't
130 correspond to a running process will result in ``-ESRCH``.
132 See also `ctrl-alt-del`_.
168 %u uid (in initial user namespace)
169 %g gid (in initial user namespace)
197 In order to do this safely, the kernel must wait for the collecting
200 This in turn creates the possibility that a misbehaving userspace
205 space applications in parallel.
209 captured in parallel, but that no waiting will take place (i.e. the
218 The default coredump writes VMAs in address order. By setting
223 in the smaller VMAs.
236 ctrl-alt-del
239 When the value in this file is 0, ctrl-alt-del is trapped and
246 when a program (like dosemu) has the keyboard in 'raw'
247 mode, the ctrl-alt-del is intercepted by the program before it
270 hostname of your box in exactly the same way as the commands
285 domain names are in general different. For a detailed discussion
303 See Documentation/driver-api/firmware/fallback-mechanisms.rst.
305 The entries in this directory allow the firmware loader helper
357 debug information. If enabled, arch-specific all-CPU stack dumping
377 See Documentation/admin-guide/lockup-watchdogs.rst for more information.
385 Default value is ``CONFIG_UEVENT_HELPER_PATH``, which in turn defaults
389 modern systems rely exclusively on the netlink-based uevent source and
403 1: Will non-maskably interrupt all CPUs and dump their backtraces when
429 When a task in D state did not get scheduled
435 Possible values to set are in range {0:``LONG_MAX``/``HZ``}.
449 Possible values to set are in range {0:``LONG_MAX``/``HZ``}.
460 -1: report an infinite number of warnings.
466 Controls whether the panic kmsg data should be reported to Hyper-V.
474 ignore-unaligned-usertrap
488 See also `unaligned-trap`_.
500 -EPERM) for unprivileged processes not in the io_uring_group group.
504 always fails with -EPERM. Existing io_uring instances can still be
513 privileged (CAP_SYS_ADMIN) or be in the io_uring_group group in order
514 to create an io_uring instance. If io_uring_group is set to -1 (the
542 -1 Unlimited calls to kexec. This is the default setting.
571 The correct long-term solution is to do the permission checks at
574 using `dmesg_restrict`_ to protect against uses of %pK in ``dmesg(8)``
585 by default ``CONFIG_MODPROBE_PATH``, which in turn defaults to
608 If CONFIG_STATIC_USERMODEHELPER=y is set in the kernel configuration,
617 in an otherwise modular kernel. This toggle defaults to off
628 ``msgmax`` is the maximum size of an IPC message, in bytes. 8192 by
631 ``msgmnb`` is the maximum size of an IPC queue, in bytes. 16384 by
638 in POSIX message queues is limited by ``RLIMIT_MSGQUEUE``. This limit is
639 respected hierarchically in the each user namespace.
647 By default they are equal to -1, which means generic allocation logic.
648 Possible values to set are in range {0:``INT_MAX``}.
653 2) Toggle with non-default value will be set back to -1 by kernel after
655 fails, it is undefined if the value remains unmodified or is reset to -1.
679 that are programmed to generate Non-Maskable Interrupts (NMIs) periodically
683 in a KVM virtual machine. This default can be overridden by adding::
688 Documentation/admin-guide/kernel-parameters.rst).
731 place the hot pages in the fast memory. This is implemented based on
739 promotion throughput. The per-node max promotion throughput in MB/s
750 resort in case a panic cannot be triggered (to protect VMs running, for
757 1: Will non-maskably interrupt all CPUs and dump their backtraces when
787 The only way to tune these values is to rebuild the kernel :-)
793 if your architecture did not always support 32-bit UIDs (i.e. arm,
795 applications that use the old 16-bit UID/GID system calls, if the
805 The value in this file determines the behaviour of the kernel on a
825 serious system condition which could result in IO data corruption.
839 1 Panic immediately. If the `panic` sysctl is also non-zero then the
866 such as power management so the default is off. That sysctl works like
867 the existing panic controls already in that directory.
873 Calls panic() in the WARN() path when set to 1. This is useful to avoid
894 bit 5 print all printk messages in buffer
895 bit 6 print all CPUs backtrace (if available in the arch)
896 bit 7 print only tasks in uninterruptible (blocked) state
932 Some perf sampling happens in NMIs. If these samples
941 1-100 Attempt to throttle perf's sample rate to this
964 -1 Allow use of (almost) all events by all users.
985 '``perf record -g``' or '``perf trace --call-graph fp``'.
987 This can only be done when no events are in use that have callchains
988 enabled, otherwise writing to this file will return ``-EBUSY``.
996 Control size of per-cpu ring buffer not counted against mlock limit.
1006 instance, when using '``perf record -g``' or '``perf trace --call-graph fp``'.
1008 This can only be done when no events are in use that have callchains
1009 enabled, otherwise writing to this file will return ``-EBUSY``.
1055 The last pid allocated in the current (the one task using this sysctl
1056 lives in) pid namespace. When selecting a pid for a next task on fork
1060 powersave-nap (PPC only)
1063 If set, Linux-PPC will use the 'nap' mode of powersaving,
1072 The four values in printk denote: ``console_loglevel``,
1094 Delay each printk message in ``printk_delay`` milliseconds
1096 Value from 0 - 10000 is allowed.
1103 the minimum length of time between these messages (in seconds).
1115 send before ratelimiting kicks in.
1132 a one-time setting until next reboot: once set, it cannot be changed by
1155 * ``entropy_avail``: the pool's entropy count, in bits;
1157 * ``poolsize``: the entropy pool size, in bits;
1174 space randomization that is used in the system, for architectures
1184 loaded to random addresses. Also for PIE-linked binaries, the
1195 non-legacy applications that would be broken this way, so for most
1204 real-root-dev
1207 See Documentation/admin-guide/initrd.rst.
1210 reboot-cmd (SPARC only)
1226 this value to 0. On Non-EAS platforms, write operation fails and
1233 Documentation/accounting/delay-accounting.rst. Enabling this feature incurs
1234 a small amount of overhead in the scheduler but is useful for debugging
1241 incurs a small amount of overhead in the scheduler but is
1270 at the highest frequency and most capable (highest capacity) CPU (in
1278 used. In battery powered devices particularly, running at the maximum
1302 See Documentation/userspace-api/seccomp_filter.rst.
1305 sg-big-buff
1333 creation of a new ipc namespace in the current user namespace or cgroups can
1341 Shared memory segments up to 1Gb are now supported in the
1366 limits (in particular, ``RLIMIT_AS`` and ``RLIMIT_NPROC``). Most systems don't
1380 -1 Legacy per-write sysctl value handling, with no printk warnings.
1390 be fully contained in the buffer sent in the write syscall.
1446 On x86, each "split lock" imposes a system-wide performance penalty. On larger
1447 systems, large numbers of split locks from unprivileged users can result in
1448 denials of service to well-behaved and potentially more important users.
1459 0 Disable the mitigation mode - just warns the split lock on kernel log
1461 1 Enable the mitigation mode (this is the default) - penalizes the split
1484 stop-a (SPARC only)
1487 Controls Stop-A:
1490 0 Stop-A has no effect.
1491 1 Stop-A breaks to the PROM (default).
1494 Stop-A is always enabled on a panic, so that the user can return to
1501 See Documentation/admin-guide/sysrq.rst.
1507 Non-zero if the kernel has been tainted. Numeric values, which can be
1508 ORed together. The letters are seen in "Tainted" line of Oops reports.
1522 2048 `(I)` workaround for bug in platform firmware applied
1523 4096 `(O)` externally-built ("out-of-tree") module was loaded
1531 See Documentation/admin-guide/tainted-kernels.rst for more information.
1538 See Documentation/admin-guide/kernel-parameters.rst for more details on
1542 threads-max
1552 The minimum value that can be written to ``threads-max`` is 1.
1554 The maximum value that can be written to ``threads-max`` is given by the
1557 If a value outside of this range is written to ``threads-max`` an
1584 See Documentation/admin-guide/kernel-parameters.rst and
1585 Documentation/trace/boottime-trace.rst.
1588 unaligned-trap
1601 See also `ignore-unaligned-usertrap`_.
1607 The value in this file affects behavior of handling NMI. When the
1608 value is non-zero, unknown NMI is trapped and then panic occurs. At
1620 will return ``-EPERM``. Once set to 1, this can't be cleared from the
1627 If ``BPF_UNPRIV_DEFAULT_OFF`` is enabled in the kernel config, then this
1673 enabled in the kernel config, and cores are specified with the
1675 Offline cores can be included in this mask, and if the core is later
1678 Typically this value would only be touched in the ``nohz_full`` case
1679 to re-enable cores that by default were not running the watchdog,
1686 echo 0,2-4 > /proc/sys/kernel/watchdog_cpumask