admin-guide/sysctl/kernel.rst

5 .. See scripts/check-sysctl-docs to keep this up to date
13 Documentation/admin-guide/sysctl/index.rst.
15 ------------------------------------------------------------------------------
39 If BSD-style process accounting is enabled these values control
71 The machine hardware name, the same output as ``uname -m``
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`_.
178 	%c		maximum size of core file by resource limit RLIMIT_CORE
219 ``core_sort_vma`` to 1, VMAs will be written from smallest size
220 to largest size. This is known to break at least elfutils, but
236 ctrl-alt-del
239 When the value in this file is 0, ctrl-alt-del is trapped and
247   mode, the ctrl-alt-del is intercepted by the program before it
292 See Documentation/driver-api/firmware/fallback-mechanisms.rst.
346 debug information. If enabled, arch-specific all-CPU stack dumping
366 See Documentation/admin-guide/lockup-watchdogs.rst for more information.
378 modern systems rely exclusively on the netlink-based uevent source and
392 1: Will non-maskably interrupt all CPUs and dump their backtraces when
458 -1: report an infinite number of warnings.
464 Controls whether the panic kmsg data should be reported to Hyper-V.
472 ignore-unaligned-usertrap
486 See also `unaligned-trap`_.
498   -EPERM) for unprivileged processes not in the io_uring_group group.
502   always fails with -EPERM. Existing io_uring instances can still be
512 to create an io_uring instance.  If io_uring_group is set to -1 (the
540 -1 Unlimited calls to kexec. This is the default setting.
569 The correct long-term solution is to do the permission checks at
626 ``msgmax`` is the maximum size of an IPC message, in bytes. 8192 by
629 ``msgmnb`` is the maximum size of an IPC queue, in bytes. 16384 by
645 By default they are equal to -1, which means generic allocation logic.
651   2) Toggle with non-default value will be set back to -1 by kernel after
653      fails, it is undefined if the value remains unmodified or is reset to -1.
659 Maximum number of supplementary groups, _i.e._ the maximum size which
677 that are programmed to generate Non-Maskable Interrupts (NMIs) periodically
686 Documentation/admin-guide/kernel-parameters.rst).
737 promotion throughput.  The per-node max promotion throughput in MB/s
755 1: Will non-maskably interrupt all CPUs and dump their backtraces when
785 The only way to tune these values is to rebuild the kernel :-)
791 if your architecture did not always support 32-bit UIDs (i.e. arm,
793 applications that use the old 16-bit UID/GID system calls, if the
837 1 Panic immediately.  If the `panic` sysctl is also non-zero then the
939 1-100 Attempt to throttle perf's sample rate to this
962  -1  Allow use of (almost) all events by all users.
983 '``perf record -g``' or '``perf trace --call-graph fp``'.
986 enabled, otherwise writing to this file will return ``-EBUSY``.
994 Control size of per-cpu ring buffer not counted against mlock limit.
1004 instance, when using '``perf record -g``' or '``perf trace --call-graph fp``'.
1007 enabled, otherwise writing to this file will return ``-EBUSY``.
1012 perf_user_access (arm64 and riscv only)
1027 riscv  chapter
1058 powersave-nap (PPC only)
1061 If set, Linux-PPC will use the 'nap' mode of powersaving,
1094 Value from 0 - 10000 is allowed.
1130 a one-time setting until next reboot: once set, it cannot be changed by
1155 * ``poolsize``: the entropy pool size, in bits;
1182     loaded to random addresses.  Also for PIE-linked binaries, the
1183     location of code start is randomized.  This is the default if the
1191     just after the end of the code+bss.  These applications break when
1193     non-legacy applications that would be broken this way, so for most
1202 real-root-dev
1205 See Documentation/admin-guide/initrd.rst.
1208 reboot-cmd (SPARC only)
1224 this value to 0. On Non-EAS platforms, write operation fails and
1231 Documentation/accounting/delay-accounting.rst. Enabling this feature incurs
1300 See Documentation/userspace-api/seccomp_filter.rst.
1303 sg-big-buff
1306 This file shows the size of the generic SCSI (sg) buffer.
1338 on the maximum shared memory segment size that can be created.
1378   -1   Legacy per-write sysctl value handling, with no printk warnings.
1444 On x86, each "split lock" imposes a system-wide performance penalty. On larger
1446 denials of service to well-behaved and potentially more important users.
1457 0 Disable the mitigation mode - just warns the split lock on kernel log
1459 1 Enable the mitigation mode (this is the default) - penalizes the split
1482 stop-a (SPARC only)
1485 Controls Stop-A:
1488 0 Stop-A has no effect.
1489 1 Stop-A breaks to the PROM (default).
1492 Stop-A is always enabled on a panic, so that the user can return to
1499 See Documentation/admin-guide/sysrq.rst.
1505 Non-zero if the kernel has been tainted. Numeric values, which can be
1521   4096  `(O)`  externally-built ("out-of-tree") module was loaded
1529 See Documentation/admin-guide/tainted-kernels.rst for more information.
1536   See Documentation/admin-guide/kernel-parameters.rst for more details on
1540 threads-max
1550 The minimum value that can be written to ``threads-max`` is 1.
1552 The maximum value that can be written to ``threads-max`` is given by the
1555 If a value outside of this range is written to ``threads-max`` an
1561 When set to a non-zero value, attempt to migrate timers away from idle cpus to
1589 See Documentation/admin-guide/kernel-parameters.rst and
1590 Documentation/trace/boottime-trace.rst.
1593 unaligned-trap
1606 See also `ignore-unaligned-usertrap`_.
1613 value is non-zero, unknown NMI is trapped and then panic occurs. At
1625 will return ``-EPERM``. Once set to 1, this can't be cleared from the
1684 to re-enable cores that by default were not running the watchdog,
1691   echo 0,2-4 > /proc/sys/kernel/watchdog_cpumask