Lines Matching +full:fan +full:- +full:stop +full:- +full:to +full:- +full:start +full:- +full:us
4 By default, unaccepted memory is accepted lazily to
9 accept_memory=eager can be used to accept all memory
16 force -- enable ACPI if default was off
17 on -- enable ACPI but allow fallback to DT [arm64,riscv64]
18 off -- disable ACPI if default was on
19 noirq -- do not use ACPI for IRQ routing
20 strict -- Be less tolerant of platforms that are not
22 rsdt -- prefer RSDT over (default) XSDT
23 copy_dsdt -- copy DSDT to memory
24 nocmcff -- Disable firmware first mode for corrected
26 source to check if firmware has set the FF flag. This
28 nospcr -- disable console in ACPI SPCR table as
45 If set to vendor, prefer vendor-specific driver
48 If set to video, use the ACPI video.ko driver.
49 If set to native, use the device's native backlight mode.
50 If set to none, disable the ACPI backlight interface.
53 force FADT to use 32 bit addresses rather than the
60 This mechanism can repair the evaluation result to make
62 This option is useful for developers to identify the
64 has something to do with the repair mechanism.
69 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
70 debug output. Bits in debug_layer correspond to a
73 Bits in debug_level correspond to a level in
76 The debug_level mask defaults to "info". See
77 Documentation/firmware-guide/acpi/debug.rst for more information about
82 Enable AML "Debug" output, i.e., stores to the Debug
85 Enable all messages related to ACPI hardware:
90 if you need to capture more output.
99 strict (default): access to resources claimed by ACPI
100 is denied; legacy drivers trying to access reserved
101 resources will fail to bind to device using them.
102 lax: access to resources claimed by ACPI is allowed;
103 legacy drivers trying to access reserved resources
110 By default, this is disabled due to x86 early mapping
129 Due to the existence of _Lxx/_Exx, some GPEs triggered
133 This facility can be used to prevent such uncontrolled
135 Format: <byte> or <bitmap-list>
138 Disable auto-serialization of AML methods
139 AML control methods that contain the opcodes to create
141 auto-serialization feature.
143 This option allows to turn off the feature.
156 tables to /sys/firmware/acpi/tables/dynamic.
159 Ignore the ACPI-based watchdog interface (WDAT) and let
163 Pass the RSDP address to the kernel, mostly used
164 on machines running EFI runtime service to boot the
168 Format: To spoof as Windows 98: ="Microsoft Windows"
170 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
173 be necessary to carry out a cold reboot _twice_ in a
174 row to make it take effect on the platform firmware).
180 acpi_osi=! # disable all built-in OS vendor
182 acpi_osi=!! # enable all built-in OS vendor
187 multiple 'acpi_osi="string1"' to support specific OS
198 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
212 multiple 'acpi_osi="string1"' to support specific
217 still not able to affect the final state of a string if
218 there are quirks related to this string. This command
219 is useful when one want to control the state of the
220 feature group strings to debug BIOS issues related to
228 equivalent to
236 to assume that this machine's pmtimer latches its value
244 For broken nForce2 BIOS resulting in XT-PIC timer.
253 as soon as the kernel's real-mode entry point is called.
254 s4_hwsig causes the kernel to check the ACPI hardware
256 refuse to resume if it has changed. This complies with
259 on docking. So the default behaviour is to allow resume
265 control method, with respect to putting devices into
266 low power states, to be enforced (the ACPI 2.0 ordering
270 sci_force_enable causes the kernel to set SCI_EN directly
273 nobl causes the internal blacklist of systems known to
274 behave incorrectly in some ways with respect to system
275 suspend and resume to be ignored (use wisely).
287 try_unsupported: try to drive unsupported chipsets
291 See Documentation/sound/alsa-configuration.rst
295 behaviour to be specified. Bit 0 enables warnings,
298 align_va_addr= [X86-64]
301 gives you up to 3% performance improvement on AMD F15h
303 CPU-intensive style benchmark, and it can vary highly in
306 32: only for 32-bit processes
307 64: only for 64-bit processes
308 on: enable for both 32- and 64-bit processes
309 off: disable for both 32- and 64-bit processes
314 and you need to use tracing_snapshot() on boot up, and
315 do not want to use tracing_snapshot_alloc() as it needs
316 to be done where GFP_KERNEL allocations are allowed.
319 Allow execve() of 32-bit applications and setting of the
321 subset of the CPUs support 32-bit EL0. When this
322 parameter is present, the set of CPUs supporting 32-bit
324 and hot-unplug operations may be restricted.
326 See Documentation/arch/arm64/asymmetric-32bit.rst for more
329 amd_iommu= [HW,X86-64]
330 Pass parameters to the AMD IOMMU driver in the system.
332 fullflush - Deprecated, equivalent to iommu.strict=1
333 off - do not initialize any AMD IOMMU found in
335 force_isolation - Force device isolation for all
337 allowed anymore to lift isolation
340 force_enable - Force enable the IOMMU on platforms known
341 to be buggy with IOMMU enabled. Use this
343 pgtbl_v1 - Use v1 page table for DMA-API (Default).
344 pgtbl_v2 - Use v2 page table for DMA-API.
345 irtcachedis - Disable Interrupt Remapping Table (IRT) caching.
346 nohugepages - Limit page-sizes used for v1 page-tables
347 to 4 KiB.
348 v2_pgsizes_only - Limit page-sizes used for v1 page-tables
349 to 4KiB/2Mib/1GiB.
352 amd_iommu_dump= [HW,X86-64]
353 Enable AMD IOMMU driver option to dump the ACPI table
358 amd_iommu_intr= [HW,X86-64]
361 legacy - Use legacy interrupt remapping mode.
362 vapic - Use virtual APIC mode, which allows IOMMU
363 to inject interrupts directly into guest.
364 This mode requires kvm-amd.avic=1.
375 tries to match the same performance level if it is
379 driver provides a hint to the hardware if software wants
380 to bias toward performance (0x0) or energy efficiency (0xff)
381 to the CPPC firmware. then CPPC power algorithm will
388 to the current workload.
393 Disable amd-pstate preferred core.
396 Map of devices attached to JOY0DAT and JOY1DAT
402 connected to one of 16 gameports
406 Power management functions (SPARCstation-4/5 + deriv.)
408 Disable APC CPU standby support. SPARCstation-Fox does
409 not play well with APC CPU idle - disable it if you have
412 apic [APIC,X86-64] Use IO-APIC. Default.
418 when initialising the APIC and IO-APIC components.
419 For X86-32, this can also be used to specify an APIC
426 bsp: External NMI is delivered only to CPU 0
427 all: External NMIs are broadcast to all CPUs as a
446 0 -- disable.
447 1 -- enable.
450 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
498 audit= [KNL] Enable the audit sub-system
500 0 | off - kernel audit is disabled and can not be
502 unset - kernel audit is initialized but disabled and
504 1 | on - kernel audit is initialized and partially
515 behavior is to disable the BAU (i.e. bau=0).
517 0 - Disable the BAU.
518 1 - Enable the BAU.
519 unset - Disable the BAU.
540 Control the ability to open a mounted block device
542 the FS. This was implemented as a means to prevent
554 Disable BGRT to avoid flickering OEM logo.
558 See Documentation/block/cmdline-partition.rst
561 Milliseconds to delay each printk during boot.
563 and you may also have to specify "lpj=". Boot_delay
569 Extended command line options can be added to an initrd
570 and this will cause the kernel to look for it.
572 See Documentation/admin-guide/bootconfig.rst
577 bttv.pll= See Documentation/admin-guide/media/bttv.rst
586 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
589 to fix known problems, but for some CPUs it is not
590 possible to determine what the correct size should be.
599 the system trusted keyring to be used for certificate
604 algorithm. Accepted values range from 0 to 7
605 inclusive. See arch/mips/include/asm/pgtable-bits.h
613 Format: {name of the controller(s) or feature(s) to disable}
615 - foo isn't auto-mounted if you mount all cgroups in
617 - foo isn't visible as an individually mountable
619 - if foo is an optional feature then the feature is
625 Specifying "pressure" disables per-cgroup pressure
631 Like cgroup_disable, but only applies to cgroup v1;
639 Defaults to the value of CONFIG_CGROUP_FAVOR_DYNMODS.
641 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
643 nosocket -- Disable socket memory accounting.
644 nokmem -- Disable kernel memory accounting.
645 nobpf -- Disable BPF memory accounting.
650 0 -- check protection applied by kernel (includes
652 1 -- check protection requested by application.
656 Setting checkreqprot to 1 is deprecated.
664 numbers X. Note the Linux-specific bits are not necessarily
665 stable over kernel options, but the vendor-specific
685 force such clocks to be always-on nor does it reserve
689 information, see Documentation/driver-api/clk.rst.
691 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
693 Forces specified clocksource (if available) to be used
695 clocksource is not available, it defaults to PIT.
702 Some clocksource names to choose from, depending on
708 [X86-32] pit,hpet,tsc;
715 [X86-64] hpet,tsc
721 architected timer so that code using WFE-based polling
728 are marked unstable due to excessive skew.
729 A negative value says to check all CPUs, while
730 zero says not to check any. Values larger than
731 nr_cpu_ids are silently truncated to nr_cpu_ids.
735 clocksource-wdtest.holdoff= [KNL]
738 Defaults to zero when built as a module and to
741 cma=nn[MG]@[start[MG][-end[MG]]]
752 Sets the size of kernel per-numa memory area for
754 per-numa CMA altogether. And If this option is not
756 With per-numa CMA enabled, DMA users on node nid will
757 first try to allocate buffer from the pernuma area
759 they will fallback to the global default memory area.
768 first try to allocate buffer from the numa area
770 they will fallback to the global default memory area.
775 to determine OS memory pressure for page stealing by
781 allocations, by default set to 256K.
783 com20020= [HW,NET] ARCnet - COM20020 chipset
787 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
791 ARCnet - COM90xx chipset (memory-mapped buffers)
799 When set to true, drop data on the 3215 console when
802 x3270) does not have to enter the clear key for the
803 console output to advance and the kernel to continue.
804 This leads to a much faster boot time when a 3270
820 See Documentation/admin-guide/serial-console.rst for more
832 The mapping of the serial ports to the tty instances
835 $ ls -d /sys/bus/serial-base/devices/*:*.*/tty/*
836 /sys/bus/serial-base/devices/00:04:0.0/tty/ttyS0
841 is ready. The use of an earlycon parameter in addition to
849 Start an early, polled-mode console on the 8250/16550
851 switching to the matching ttyS device later.
852 MMIO inter-register address stride is either 8-bit
853 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
855 to be equivalent to 'mmio'. 'options' are specified in
857 the h/w is not re-initialized.
863 Use to disable console output, i.e., to have kernel
868 If the device connected to the port is not a TTY but a braille
881 Switch to syslog format: "<%u>[time stamp] text\n"
883 prefix. The format is similar to one used by syslog()
884 syscall, or to executing "dmesg -S --raw" or to reading
889 Defaults to 0.
908 disable the cpuidle sub-system
911 [CPU_IDLE] Name of the cpuidle governor to use.
914 disable the cpufreq sub-system
918 policy to use. This governor must be registered in the
922 [X86,EARLY] Delay for N microsec between assert and de-assert
923 of APIC INIT to start processors. This delay occurs
934 Only jump to kdump kernel after running the panic
943 like Hyper-V, PowerPC (fadump) and AMD SEV-SNP.
946 [KNL,EARLY] Using kexec, Linux can switch to a 'crash kernel'
951 [KNL, X86-64, ARM64, RISCV, LoongArch] Select a region
952 under 4G first, and fall back to reserve region above
954 See Documentation/admin-guide/kdump/kdump.rst for further details.
959 start-[end] where start and end are both
961 Documentation/admin-guide/kdump/kdump.rst for an example.
964 [KNL, X86-64, ARM64, RISCV, LoongArch] range could be
966 Allow kernel to allocate physical memory region from top,
972 [KNL, X86-64, ARM64, RISCV, LoongArch] range under 4G.
977 enough extra low memory is needed to make sure DMA buffers
978 for 32-bit devices won't run out. Kernel would try to allocate
981 --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB)
982 --> arm64: 128MiB
983 --> riscv: 128MiB
984 --> loongarch: 128MiB
987 0: to disable low allocation.
992 [KNL] Disable crypto self-tests
1000 csdlock_debug= [KNL] Enable or disable debug add-ons of cross-CPU
1002 additional debug data is printed to the console
1004 CPU is pinged again in order to try to resolve
1015 See also Documentation/input/devices/joystick-parport.rst
1022 of siphash to hash pointers. Use this option if you are
1023 seeing instances of '(___ptrval___)') and need to see a
1028 [KNL] verbose locking self-tests
1031 self-tests.
1032 Bitmask for the various LOCKTYPE_ tests. Defaults to 0
1033 (no extra messages), setting it to -1 (all bits set)
1034 will print _a_lot_ more information - normally only
1035 useful to lockdep developers.
1047 parameter to 1 or 2 should be enough to identify most
1049 kernel or driver code when a CPU writes to (or reads
1063 Note: to get most of debug_pagealloc error reports, it's
1064 useful to also enable the page_owner functionality.
1067 debugfs= [KNL,EARLY] This parameter enables what is exposed to
1069 Format: { on, no-mount, off }
1071 no-mount:
1073 access APIs and a crashkernel can be used to read
1074 its content. There is nothing to mount.
1076 get a -EPERM as result when trying to register files
1080 Default value is set in build-time with a kernel configuration.
1089 filesystems. If not specified, defaults to the
1092 Documentation/admin-guide/mm/hugetlbpage.rst.
1096 [KNL] Debugging option to set a timeout in seconds for
1097 deferred probe to give up waiting on dependencies to
1106 delayacct [KNL] Enable per-task delay accounting
1127 [HW] Factor to multiply fan speed with.
1130 [HW] Maximum configurable fan speed.
1149 causes the kernel to fall back to 256MB segments which
1151 miss to occur.
1165 to workaround buggy firmware.
1171 The kernel tries to adjust MTRR layout from continuous
1172 to discrete, to make X server driver able to add WB
1179 possibly causing your machine to run very slowly.
1183 Can be useful to work around chipset bugs.
1191 This option allows to tune the number of preallocated
1192 entries for DMA-API debugging code. One entry is
1193 required per DMA-API allocation. Use this if the
1194 DMA-API debugging code disables itself because the
1198 With this option the DMA-API debugging driver
1200 pass the driver to filter for as the parameter.
1201 The filter can be disabled or changed to another
1207 vulnerability which may allow userspace to infer
1218 are enabled. In order to disable RFDS mitigation all
1219 VERW based mitigations need to be disabled.
1222 Documentation/admin-guide/hw-vuln/reg-file-data-sampling.rst
1225 List of driver names to be probed asynchronously. *
1234 This parameter allows to specify an EDID data sets
1237 connector, if its name and a colon are prepended to
1247 Control how the dt_cpu_ftrs device-tree binding is
1250 off: Do not use it, fall back to legacy cpu table.
1251 known: Do not pass through unknown features to guests
1256 x86 Macs. Useful for driver authors to determine
1257 what data is available or for reverse-engineering.
1262 Documentation/admin-guide/dynamic-debug-howto.rst
1273 determined by stdout-path property in device tree's
1278 Start an early, polled-mode console on a Cadence
1289 Start an early, polled-mode console on the 8250/16550
1291 MMIO inter-register address stride is either 8-bit
1292 (mmio) or 32-bit (mmio32 or mmio32be).
1294 to be equivalent to 'mmio'. 'options' are specified
1298 to 'BASE_BAUD' * 16.
1302 Start an early, polled-mode console on a pl011 serial
1306 the driver will use only 32-bit accessors to read/write
1310 Start an early console on a litex serial port at the
1315 Start an early, polled-mode console on a meson serial
1321 Start an early, polled-mode console on an msm serial
1327 Start an early, polled-mode console on an msm serial
1333 Start an early, polled-mode console on a serial port
1339 Start an early, polled-mode console on a serial port
1345 Use RISC-V SBI (Supervisor Binary Interface) for early
1363 Start an early, polled-mode console on a lantiq serial
1377 Start an early, polled-mode, output-only console on the
1382 Start an early, polled-mode console on the
1388 Start an early, polled-mode console on a Qualcomm
1394 Start an early, unaccelerated console on the EFI
1396 coherent non-x86 systems that use system memory for
1422 Append ",keep" to not disable it when the real console
1449 The optional "force" to "pciserial" enables use of a
1453 edac_report= [HW,EDAC] Control how to report EDAC event
1455 on: enable EDAC to report H/W event. May be overridden
1458 force: enforce the use of EDAC to report H/W event.
1476 attribute may cause the kernel to reserve the
1477 memory range for a memory mapping driver to
1478 claim. Specify efi=nosoftreserve to disable this
1492 that is to be dynamically loaded by Linux. If there are
1495 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1504 This is designed to be used in conjunction with
1512 elanfreq= [X86-32]
1517 Specifies physical address of start of kernel core
1519 kexec loader will pass this option to capture kernel.
1520 See Documentation/admin-guide/kdump/kdump.rst for details.
1523 The kernel tries to adjust MTRR layout from continuous
1524 to discrete, to make X server driver able to add WB
1529 Can be useful to work around chipset bugs
1531 The kernel tries to set a reasonable default.
1536 0 -- permissive (log only, no denials).
1537 1 -- enforcing (deny and log).
1552 Permit 'security.evm' to be updated regardless of
1555 early_page_ext [KNL,EARLY] Enforces page_ext initialization to earlier
1558 might be disabled to achieve that (e.g. parallelized
1570 See also Documentation/fault-injection/.
1574 See Documentation/admin-guide/sysctl/net.rst for
1578 See Documentation/admin-guide/blockdev/floppy.rst.
1580 forcepae [X86-32]
1587 fred= [X86-64]
1594 [FTRACE] will set and start the specified tracer
1595 as early as possible in order to facilitate early
1603 boot up that is likely to be overridden by user space
1604 start up functionality.
1612 The above will cause the "foo" tracing instance to trigger
1625 oops by passing 2 or orig_cpu to it.
1638 ftrace_filter=[function-list]
1640 tracer at boot up. function-list is a comma-separated
1645 ftrace_notrace=[function-list]
1647 function-list. This list can be changed at run time
1651 ftrace_graph_filter=[function-list]
1654 function-list is a comma-separated list of functions
1658 ftrace_graph_notrace=[function-list]
1660 function-list. This list is a comma-separated list of
1671 devices by scanning the firmware to infer the
1680 off -- Don't create device links from firmware info.
1681 permissive -- Create device links from firmware info
1684 on -- Create device links from firmware info and use it
1685 to enforce probe and suspend/resume ordering.
1686 rpm -- Like "on", but also use to order runtime PM.
1695 probing, this parameter controls what to do with
1699 strict -- Default. Continue waiting on consumers to
1701 timeout -- Give up waiting on consumers and call
1709 support via parallel port (up to 5 devices per port)
1711 See also Documentation/input/devices/joystick-parport.rst
1715 gart_fix_e820= [X86-64,EARLY] disable the fix e820 for K8 GART
1724 allows unprivileged speculative access to data which was
1732 force: Disable AVX to mitigate systems without
1734 mitigation is present. Known to cause crashes in
1741 gcov_persist= [GCOV] When non-zero (default), profiling data for
1751 gpio-mockup.gpio_mockup_ranges
1754 gpio-mockup.gpio_mockup_named_lines
1758 invalid Protective MBR to be treated as GPT. If the
1760 GPT to be used instead.
1770 grcan.select= [HW] Select which physical interface to use.
1783 usercopy checking is used to protect the kernel
1786 against bounds-checking flaws in the kernel's
1792 [KNL] Should the hard-lockup detector generate
1798 for 64-bit NUMA, off otherwise.
1806 corresponding firmware-first mode error processing
1816 during restoration read-only).
1818 hibernate.compressor= [HIBERNATION] Compression algorithm to be
1823 lzo: Select LZO compression algorithm to
1826 lz4: Select LZ4 compression algorithm to
1829 highmem=nn[KMG] [KNL,BOOT,EARLY] forces the highmem zone to have an exact
1831 highmem otherwise. This also works to reduce highmem
1844 Using this parameter to set the hostname makes it
1845 possible to ensure the hostname is correctly set before
1853 hpet= [X86-32,HPET] option to control HPET usage
1861 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1864 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1866 the number of pages of hugepagesz to be allocated.
1868 line, it specifies the number of pages to allocate for
1870 number of pages to allocate per-node can be specified.
1871 See also Documentation/admin-guide/mm/hugetlbpage.rst.
1877 conjunction with hugepages (above) to allocate huge
1882 Documentation/admin-guide/mm/hugetlbpage.rst.
1893 boot-time allocation of gigantic hugepages is skipped.
1899 Allows heavy hugetlb users to free up some more
1919 A value of 1 instructs the kernel to panic when a
1921 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1927 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1933 which allow the hypervisor to 'idle' the guest
1943 Set device-properties for ACPI-enumerated I2C-attached
1944 touchscreen, to e.g. fix coordinates of upside-down
1952 Omit "=<val>" entirely Set a boolean device-property
1953 Unsigned number Set a u32 device-property
1954 Anything else Set a string device-property
1957 i2c_touchscreen_props=GDIX1001:touchscreen-inverted-x:
1958 touchscreen-inverted-y
1960 i2c_touchscreen_props=MSSL1680:touchscreen-size-x=1920:
1961 touchscreen-size-y=1080:touchscreen-inverted-y:
1962 firmware-name=gsl1680-vendor-model.fw:silead,home-button
1967 (disabled by default, and as a pre-condition
1969 i8042.direct [HW] Put keyboard port into non-translated mode
1972 (Don't attempt to blink the leds)
1979 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1983 suspend-to-ram transitions, only during s2r
1989 architectures force reset to be always executed
1991 i8042.kbdreset [HW] Reset device connected to KBD port
1998 [DRM] Invert the sense of the variable that is used to
2002 to maximum brightness. If this parameter is set to 0
2004 is set to 1, a brightness value of 0 sets the backlight
2005 to maximum brightness, and the maximum of the brightness
2007 -1 -- never invert brightness
2008 0 -- machine default
2009 1 -- force brightness inversion
2011 ia32_emulation= [X86-64]
2013 When true, allows loading 32-bit programs and executing 32-bit
2027 to get slightly better performance in multiprocessor
2035 idle=halt: Halt is forced to be used for CPU idle.
2038 idle=nomwait: Disable mwait for CPU C-states
2043 support for the idxd driver. By default it is set to
2049 for the device. By default it is set to false (0).
2059 binary. Hardware implementations are permitted to
2066 legacy only accept legacy-NaN binaries, if supported
2068 2008 only accept 2008-NaN binaries, if supported
2075 The FPU emulator is always able to support both NaN
2079 'relaxed' straps the emulator for both legacy-NaN and
2080 2008-NaN, whereas 'strict' enables legacy-NaN only on
2089 Ignore loglevel setting - this will print /all/
2090 kernel messages to the console. Useful for debugging.
2124 The builtin policies to load during IMA setup.
2156 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" |
2157 "ima-sigv2" }
2158 Default: "ima-ng"
2171 to achieve the best performance for a particular HW.
2179 to achieve best performance for particular HW.
2190 initcall_blacklist= [KNL] Do not execute a comma-separated list of
2191 initcall functions. Useful for debugging built-in
2208 initrdmem= [KNL,EARLY] Specify a physical address and size from which to
2226 default (disallow access to all but pkey 0). Can
2236 0 -- basic integrity auditing messages. (Default)
2237 1 -- additional integrity auditing messages.
2251 Deprecated, equivalent to iommu.strict=1.
2265 could harm performance of some high-throughput
2270 vulnerable to DMA attacks.
2274 1 to 9 specify maximum depth of C-state.
2281 Use intel_pstate driver to bypass the scaling
2283 algorithms for p-state selection. There are two
2284 P-state selection algorithms provided by
2287 on whether or not the hardware managed P-states
2292 to work with generic cpufreq governors (instead of
2294 used along with the hardware-managed P-states (HWP)
2298 in favor of acpi-cpufreq. Forcing the intel_pstate driver
2299 instead of acpi-cpufreq may disable platform features, such
2301 P-States information being indicated to OSPM and therefore
2304 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2317 Allow per-logical-CPU P-State performance control limits using
2320 intremap= [X86-64,Intel-IOMMU,EARLY]
2325 BIOS x2APIC opt-out request will be ignored
2330 iomem= Disable strict checking of access to MMIO memory
2356 Intel machines). This can be used to prevent the usage
2368 AMD Gart HW IOMMU-specific options:
2387 Do scatter-gather (SG) merging. Implies "force"
2391 Don't do scatter-gather (SG) merging.
2394 Ask the IOMMU not to touch the aperture for AGP.
2404 0 - Try to allocate a 32-bit DMA address first, before
2405 falling back to the full range if needed.
2406 1 - Allocate directly from the full usable range,
2408 greater than 32-bit addressing.
2412 0 - Lazy mode.
2416 Will fall back to strict mode if not supported by
2418 1 - Strict mode.
2421 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}.
2423 legacy driver-specific options takes precedence.
2426 [ARM64,X86,EARLY] Configure DMA to bypass the IOMMU by default.
2428 0 - Use IOMMU translation for DMA.
2429 1 - Bypass the IOMMU for DMA.
2430 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2432 io7= [HW] IO7 for Marvel-based Alpha systems
2447 See Documentation/admin-guide/nfs/nfsroot.rst.
2450 IPC identifiers from 32,768 to 16,777,216.
2460 Start IPE with success auditing enabled, emitting
2470 Force the kernel to look for the second 4kB page
2476 Force the kernel to ignore the availability of
2479 to let secondary kernels in charge of setting up
2483 Enables support for pseudo-NMIs in the kernel. This
2484 requires the kernel to be built with
2489 for it. Intended to get systems with badly broken
2495 interrupt. Intended to get systems with badly broken
2502 [Deprecated - use cpusets instead]
2503 Format: [flag-list,]<cpu-list>
2505 Specify one or more CPUs to isolate from disturbances
2511 offloaded. This is equivalent to the nohz_full parameter.
2513 A residual 1Hz tick is offloaded to workqueues, which you
2514 need to affine to housekeeping through the global
2520 so to protect individual CPUs the 'cpumask' file has to
2526 is irreversible: it's not possible to bring back a CPU to
2528 advised to use cpusets instead to disable scheduler load
2536 "number of CPUs in system - 1".
2550 interrupts are directed to the housekeeping CPU
2562 The format of <cpu-list> is described above.
2566 ivrs_ioapic [HW,X86-64]
2567 Provide an override to the IOAPIC-ID<->DEVICE-ID
2571 For example, to map IOAPIC-ID decimal 10 to
2577 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0
2580 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and
2584 ivrs_hpet [HW,X86-64]
2585 Provide an override to the HPET-ID<->DEVICE-ID
2589 For example, to map HPET-ID decimal 10 to
2595 * To map HPET-ID decimal 0 to PCI device 00:14.0
2598 * To map HPET-ID decimal 10 to PCI segment 0x1 and
2602 ivrs_acpihid [HW,X86-64]
2603 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2607 For example, to map UART-HID:UID AMD0020:0 to
2613 * To map UART-HID:UID AMD0020:0 to PCI segment is 0,
2616 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and
2624 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2630 Do not unregister boot console at start. This is only
2640 the kernel for non-movable allocations. The requested
2644 event, a node is too small to have both ZONE_NORMAL and
2650 subsystem. Note that allocations like PTEs-from-HighMem
2654 It is possible to specify the exact amount of memory in
2658 for non-movable allocations and remaining memory is used
2667 each poll cycle to the debug port in case you need
2669 gdb or control-c on the dbgp connection. When
2670 not using this parameter you use sysrq-g to break into
2675 or a supported polling keyboard driver (non-usb).
2684 If the boot console provides the ability to read
2686 this parameter to tell kgdb to use it as a backend
2687 until the normal console is registered. Intended to
2689 specifies the normal console to transition to.
2694 name passed to kgdboc. It's OK to leave the value
2698 kgdbwait [KGDB,EARLY] Stop kernel execution and enter the
2702 Configure the RouterBoard 532 series on-chip
2705 kmemleak= [KNL,EARLY] Boot-time kmemleak enable/disable
2711 kprobe_event=[probe-list]
2713 The probe-list is a semicolon delimited list of probe
2716 For example, to add a kprobe event on vfs_read with
2717 arg1 and arg2, add to the command line;
2731 CONFIG_KUNIT to be set to be fully enabled. The
2736 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2740 [KVM,X86] Controls whether or not KVM will try to
2742 Eager page splitting reduces interruptions to vCPU
2743 execution by eliminating the write-protection faults
2745 required to split huge pages lazily.
2748 only to a small region of VM memory may benefit from
2749 disabling eager page splitting to allow huge pages to
2770 virtualization on-demand when creating and destroying
2778 is that doing so may interfere with using out-of-tree
2779 hypervisors that want to "own" virtualization hardware.
2795 guests do need not to enable it for nested guests.
2799 back to huge pages. 0 disables the recovery, otherwise if
2805 back to huge pages. If the value is a non-zero N, KVM will
2810 kvm-amd.nested= [KVM,AMD] Control nested virtualization feature in
2813 kvm-amd.npt= [KVM,AMD] Control KVM's use of Nested Page Tables,
2814 a.k.a. Two-Dimensional Page Tables. Default is 1
2818 kvm-arm.mode=
2824 nvhe: Standard nVHE-based mode, without support for
2831 nested: VHE-based mode with support for nested
2835 Defaults to VHE/nVHE based on hardware support. Setting
2836 mode to "protected" will disable kexec and hibernation
2837 for the host. To force nVHE on VHE hardware, add
2838 "arm64_sw.hvhe=0 id_aa64mmfr1.vh=0" to the
2839 command-line.
2843 kvm-arm.vgic_v3_group0_trap=
2844 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-0
2847 kvm-arm.vgic_v3_group1_trap=
2848 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-1
2851 kvm-arm.vgic_v3_common_trap=
2852 [KVM,ARM,EARLY] Trap guest accesses to GICv3 common
2855 kvm-arm.vgic_v4_enable=
2859 kvm-arm.wfe_trap_policy=
2860 [KVM,ARM] Control when to set WFE instruction trap for
2868 kvm-arm.wfi_trap_policy=
2869 [KVM,ARM] Control when to set WFI instruction trap for
2885 kvm-intel.ept= [KVM,Intel] Control KVM's use of Extended Page Tables,
2886 a.k.a. Two-Dimensional Page Tables. Default is 1
2890 kvm-intel.emulate_invalid_guest_state=
2891 [KVM,Intel] Control whether to emulate invalid guest
2892 state. Ignored if kvm-intel.enable_unrestricted_guest=1,
2894 guests. This param doesn't apply to nested guests (L2),
2898 kvm-intel.flexpriority=
2903 kvm-intel.nested=
2907 kvm-intel.unrestricted_guest=
2913 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2914 CVE-2018-3620.
2925 kvm-intel.vpid= [KVM,Intel] Control KVM's use of Virtual Processor
2933 Certain CPUs are vulnerable to an exploit against CPU
2934 internal buffers which can forward information to a
2939 attack, to access data to which the attacker does
2945 on - enable the interface for the mitigation
3010 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
3016 lapic [X86-32,APIC,EARLY] Enable the local APIC even if BIOS
3020 value for LAPIC timer one-shot implementation. Default
3021 back to the programmable timer unit in the LAPIC.
3043 libata.force= [LIBATA] Force configurations. The format is a comma-
3050 to all ports, links and devices.
3052 If only DEVICE is omitted, the parameter applies to
3055 first fan-out link behind PMP device. It does not
3057 host link and device attached to it.
3059 The VAL specifies the configuration to force. As long
3069 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
3076 * rstonce: only attempt one reset during hot-unplug
3103 READ LOG DMA EXT command to access logs.
3105 * [no]iddevlog: Enable or disable access to the
3108 * [no]logdir: Enable or disable access to the general
3111 * max_sec_128: Set transfer size limit to 128 sectors.
3113 * max_sec_1024: Set or clear transfer size limit to
3116 * max_sec_lba48: Set or clear transfer size limit to
3119 * external: Mark port as external (hotplug-capable).
3152 Enable the kernel lockdown feature. If set to
3153 integrity, kernel features that allow userland to
3154 modify the running kernel are disabled. If set to
3156 to extract confidential information from the kernel
3165 Specify the list of CPUs to which the readers are
3166 to be bound.
3169 Specify the list of CPUs to which the writers are
3170 to be bound.
3173 Specify the number of self-propagating call_rcu()
3174 chains to set up. These are used to ensure that
3176 in progress at any given time. Defaults to 0,
3181 occasional long-duration lock hold time. Defaults
3182 to 100 milliseconds. Select 0 to disable.
3186 locktorture is to exercise, up to a limit of 8
3187 (MAX_NESTED_LOCKS). Specify zero to disable.
3192 Set the number of locking read-acquisition kthreads.
3193 Defaults to being automatically set based on the
3197 Set the number of locking write-acquisition kthreads.
3200 Set time (s) after boot for CPU-hotplug testing.
3203 Set time (s) between CPU-hotplug operations, or
3204 zero to disable CPU-hotplug testing.
3207 Do periodic testing of real-time lock priority
3208 boosting. Select 0 to disable, 1 to boost
3209 only rt_mutex, and 2 to boost unconditionally.
3210 Defaults to 2, which might seem to be an
3212 non-real-time spinlocks, due to their disabling
3213 of preemption. Note that non-realtime mutexes
3226 Set task-shuffle interval (jiffies). Shuffling
3227 tasks allows some CPUs to go into dyntick-idle
3232 is useful for hands-off automated testing.
3238 Time (s) to stutter testing, for example,
3239 specifying five seconds causes the test to run for
3241 This tests the locking primitive's ability to
3242 transition abruptly to and from idle.
3245 Specify the locking implementation to test.
3251 Run the write-side locktorture kthreads at
3252 sched_set_fifo() real-time priority.
3259 console loglevel will be printed to the console. It can
3270 7 (KERN_DEBUG) debug-level messages
3278 parameter that allows to increase the default size
3282 logo.nologo [FB] Disables display of the built-in Linux logo.
3283 This may be used to provide more screen space for
3287 lp=0 [LP] Specify parallel ports to use, e.g,
3291 specified in addition to the ports) causes
3292 attached printers to be reset. Using
3294 to associate lp devices with, starting with
3295 lp0. A port specification may be 'none' to skip
3299 from each port should be examined, to see if
3300 an IEEE 1284-compliant printer is attached; if
3305 Sets loops_per_jiffy to given constant, thus avoiding
3306 time-consuming boot-time autodetection (up to 250 ms per
3307 CPU). 0 enables autodetection (default). To determine
3310 on SMP systems the preset will be applied to all CPUs,
3311 which is likely to cause problems if your CPUs need
3313 will cause delays in the kernel to be wrong, leading to
3326 Example: machtype=lemote-yeeloong-2f-7inch
3330 the kernel to bring up 'n' processors. Surely after
3334 While n=0 is a special case, it is equivalent to "nosmp",
3338 (loop.max_loop) unconditionally pre-created at init time. The default
3341 devices can be requested on-demand with the
3342 /dev/loop-control interface.
3344 mce= [X86-{32,64}]
3358 with due to the shared banks, i.e. you might get
3381 do not opt-in to Local MCE delivery. Use legacy method
3382 to broadcast MCEs.
3389 If your BIOS doesn't do that it's a good idea to
3390 enable though to make sure you log even machine check
3398 sets the time in us to wait for other CPUs on machine
3399 checks. 0 to disable.
3402 don't overwrite the bios-set CMCI threshold. This boot
3405 always sets the CMCI threshold to 1. Enabling this may
3411 force-enable recoverable machine check code paths
3417 See Documentation/admin-guide/md.rst.
3421 Specifies range of consoles to be captured by the MDA.
3424 Control mitigation for the Micro-architectural Data
3427 Certain CPUs are vulnerable to an exploit against CPU
3428 internal buffers which can forward information to a
3433 attack, to access data to which the attacker does
3439 full - Enable MDS mitigation on vulnerable CPUs
3440 full,nosmt - Enable MDS mitigation and disable
3442 off - Unconditionally disable MDS mitigation
3444 On TAA-affected machines, mds=off can be prevented by
3446 mitigated with the same mechanism so in order to disable
3447 this mitigation, you need to specify tsx_async_abort=off
3450 Not specifying this option is equivalent to
3453 For details see: Documentation/admin-guide/hw-vuln/mds.rst
3459 of memory Amount of memory to be used in cases
3463 2 when the kernel is not able to see the whole system memory;
3465 the hypervisor, then assigned to KVM guests.
3466 4 to limit the memory available for kdump kernel.
3468 [ARC,MICROBLAZE] - the limit applies only to low memory,
3471 [ARM64] - only limits memory covered by the linear
3475 with memmap= to avoid physical address space collisions.
3477 belonging to unused RAM.
3484 [ARM,MIPS,EARLY] - override the memory layout
3491 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
3497 [KNL,SH] Allow user to override the default size for
3498 per-device physically contiguous DMA buffers.
3503 set according to the
3506 See Documentation/admin-guide/mm/memory-hotplug.rst.
3516 Region of memory to be used is from ss to ss+nn.
3517 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
3518 which limits max address to nn[KMG].
3526 Region of memory to be marked is from ss to ss+nn.
3530 Region of memory to be reserved is from ss to ss+nn.
3531 Example: Exclude memory from 0x18690000-0x1869ffff
3541 Region of memory to be used, from ss to ss+nn.
3545 memmap=<size>%<offset>-<oldtype>+<newtype>
3547 from <oldtype> to <newtype>. If "-<oldtype>" is left
3555 Some BIOSes seem to corrupt the first 64k of
3562 repeatable BIOS-originated corruption always
3564 to prevent the kernel from using that memory.
3569 use. Use this parameter to scan for
3574 seconds. Use this parameter to check at some
3578 [KNL,X86,ARM] Boolean flag to enable this feature.
3584 hotadded memory which will allow to hotadd a
3586 memory to do so.
3599 Specifies the number of memtest passes to be
3606 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
3612 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst
3616 s2idle - Suspend-To-Idle
3617 shallow - Power-On Suspend or equivalent (if supported)
3618 deep - Suspend-To-RAM or equivalent (if supported)
3619 See Documentation/admin-guide/pm/sleep-states.rst.
3621 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
3623 version 0.98 is known to be affected, 0.99 fixes the
3637 0 - The attached screen is the 3.5" TFT
3638 1 - The attached screen is the 7" TFT
3639 2 - The VGA Shield is attached (1024x768)
3643 b - Enable backlight. The TFT backlight pin will be
3644 linked to the kernel VESA blanking code and a GPIO
3647 c - Enable the s3c camera interface.
3648 t - Reserved for enabling touchscreen support. The
3652 https://repo.or.cz/w/linux-2.6/mini2440.git
3657 arch-independent options, each of which is an
3658 aggregation of existing arch-specific options.
3666 expose users to several CPU vulnerabilities.
3667 Equivalent to: if nokaslr then kpti=0 [ARM64]
3687 ssbd=force-off [ARM64]
3698 users who don't want to be surprised by SMT
3700 have other ways of avoiding SMT-based attacks.
3701 Equivalent to: (default behavior)
3705 if needed. This is for users who always want to
3707 Equivalent to: l1tf=flush,nosmt [X86]
3719 so loglevel=8 may also need to be specified.
3729 Therefore, similar to MDS and TAA, the mitigation
3730 is to clear the affected CPU buffers.
3735 full - Enable mitigation on vulnerable CPUs
3737 full,nosmt - Enable mitigation and disable SMT on
3740 off - Unconditionally disable mitigation
3745 mitigated with the same mechanism so in order to
3746 disable this mitigation, you need to specify
3749 Not specifying this option is equivalent to
3753 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst
3763 [KNL] When set to true, modules will use async probing
3764 by default. To enable/disable async probing for a
3779 modules without (valid) signatures will fail to load.
3783 module_blacklist= [KNL] Do not load a comma-separated list of
3788 leaving touchpad surface for touch to be considered
3799 This parameter is the complement to kernelcore=, it
3808 movable_node [KNL,EARLY] Boot-time switch to make hotplugable memory
3809 NUMA nodes to be movable. This means that the memory
3815 Format: <name>,<region-number>,<size>,<offset>
3818 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
3828 Enable printing debug information related to MTRR
3846 Set to 2 or more if your graphical card needs more.
3856 Note that mem_start is often overloaded to mean
3857 something different and driver-specific.
3868 0 to disable accounting
3869 1 to enable accounting
3873 [NFS] sets the pathname to the program which is used
3874 to update the NFS client cache entries.
3877 [NFS] sets the timeout after which an attempt to
3878 update a cache entry is deemed to have failed.
3882 NFS client will assign to service NFSv4 callback
3897 [NFS] enable 64-bit inode numbers.
3898 If zero, the NFS client will fake up a 32-bit inode
3900 of returning the full 64-bit number.
3901 The default is to return 64-bit inode numbers.
3909 slots the client will assign to the callback
3916 the client will attempt to negotiate with the server.
3918 that the client can send to the NFSv4.1 server.
3923 [NFSv4] When set to the default of '1', this option
3925 scheme and the NFS level operations agree to use
3929 legacy NFSv2/v3 systems to NFSv4 easier.
3932 back to using the idmapper.
3933 To turn off this behaviour, set the value to '0'.
3936 [NFS4] Specify an additional fixed unique ident-
3942 [NFSv4] Attempt to recover locks that were lost due
3943 to a lease timeout on the server. Please note that
3947 If you want to enable the kernel legacy behaviour of
3948 attempting to recover these locks, then set this
3949 parameter to '1'.
3951 not to attempt recovery of lost locks.
3958 The default is to send the implementation identification
3963 layoutstats to the pNFS metadata server.
3965 Setting this to value to 0 causes the kernel to use
3967 driver. A non-zero value sets the minimum interval
3971 [NFSv4.2] When set to 1, the server will support
3972 server-to-server copies for which this server is
3976 [NFSv4] When set to the default of '1', the NFSv4
3977 server will return only numeric uids and gids to
3979 and gids from such clients. This is intended to ease
3984 server-to-server copy, knfsd temporarily mounts
3991 See Documentation/admin-guide/nfs/nfsroot.rst.
3993 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3994 See Documentation/admin-guide/nfs/nfsroot.rst.
3997 See Documentation/admin-guide/nfs/nfsroot.rst.
4000 Dump stacks even of idle CPUs in response to an
4001 NMI stack-backtrace request.
4003 nmi_debug= [KNL,SH] Specify one or more actions to take
4010 0 - turn hardlockup detector in nmi_watchdog off
4011 1 - turn hardlockup detector in nmi_watchdog on
4012 rNNN - configure the watchdog with raw perf event 0xNNN
4015 timeout occurs (or 'nopanic' to not panic on an NMI
4017 To disable both hard and soft lockup detectors,
4025 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
4029 no4lvl [RISCV,EARLY] Disable 4-level and 5-level paging modes.
4030 Forces kernel to use 3-level paging instead.
4032 no5lvl [X86-64,RISCV,EARLY] Disable 5-level paging mode. Forces
4033 kernel to use 4-level paging instead.
4037 noapic [SMP,APIC,EARLY] Tells the kernel to not make use of any
4051 of the system is being put to sleep (ie, while
4054 to work with serial and VGA consoles.
4055 To facilitate more flexible debugging, we also add
4056 console_suspend, a printk module parameter to control
4058 /sys/module/printk/parameters/console_suspend) to
4068 no_entry_flush [PPC,EARLY] Don't flush the L1-D cache when entering the kernel.
4070 noexec32 [X86-64]
4071 This affects only 32-bit executables.
4072 noexec32=on: enable non-executable mappings (default)
4074 noexec32=off: disable non-executable mappings
4077 no_file_caps Tells the kernel not to honor file capabilities. The
4078 only way then for a file to be executed with privilege
4079 is to be setuid root or executed by root.
4085 nofxsr [BUGS=X86-32] Disables x86 floating point extended
4087 legacy floating-point registers on task switch.
4093 Force pointers printed to the console or buffers to be
4100 compared. However, if this command-line option is
4108 nohlt [ARM,ARM64,MICROBLAZE,MIPS,PPC,RISCV,SH] Forces the kernel to
4111 to be effective. This is useful on platforms where the
4113 correctly or when doing power measurements to evaluate
4132 the range to maintain the timekeeping. Any CPUs
4140 noinitrd [RAM] Tells the kernel not to load any configured
4143 nointremap [X86-64,Intel-IOMMU,EARLY] Do not enable interrupt
4145 [Deprecated - use intremap=off]
4151 noirqdebug [X86-32] Disables the code which attempts to detect and
4161 no-kvmapf [X86,KVM,EARLY] Disable paravirtualized asynchronous page
4164 no-kvmclock [X86,KVM,EARLY] Disable paravirtualized KVM clock driver
4166 nolapic [X86-32,APIC,EARLY] Do not enable or use the local APIC.
4168 nolapic_timer [X86-32,APIC,EARLY] Do not use the local APIC timer.
4170 nomce [X86-32] Disable Machine Check Exception
4172 nomfgpt [X86-32] Disable Multi-Function General Purpose
4178 not load if they could possibly displace the pre-
4181 perform display-mode changes or accelerated rendering.
4187 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
4194 nopcid [X86-64,EARLY] Disable the PCID cpu feature.
4199 nopti [X86-64,EARLY]
4200 Equivalent to pti=off
4203 Disables the PV optimizations forcing the guest to run
4209 which allow the hypervisor to 'idle' the guest on lock
4212 norandmaps Don't use address space randomization. Equivalent to
4215 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
4221 no-scroll [VGA] Disables scrollback.
4222 This is required for the Braillex ib80-piezo Braille
4225 nosgx [X86-64,SGX,EARLY] Disables Intel SGX kernel support.
4235 nosmp [SMP,EARLY] Tells an SMP kernel to act as a UP kernel,
4239 Equivalent to smt=1.
4245 nosoftlockup [KNL] Disable the soft-lockup detector.
4251 nospectre_bhb [ARM64,EARLY] Disable all mitigations for Spectre-BHB (branch
4264 no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES,RISCV,LOONGARCH,EARLY]
4271 timer IRQ sources, i.e., the IO-APIC timer. This can
4276 [PPC,EARLY] Don't flush the L1-D cache after accessing user data.
4279 Disable device dump. Device dump allows drivers to
4280 append dump data to vmcore so you can collect driver
4289 no-vmw-sched-clock
4294 soft-lockup and NMI watchdog (hard-lockup).
4298 nox2apic [X86-64,APIC,EARLY] Do not enable x2APIC mode.
4305 and restore using xsave. The kernel will fallback to
4306 enabling legacy floating-point and sse state.
4309 register states. The kernel will fall back to use
4310 xsave to save the states. By using this parameter,
4317 form of xsave area. The kernel will fall back to use
4318 xsaveopt and xrstor to save and restore the states
4324 could support. nr_cpus=n : n >= 1 limits the kernel to
4328 n. So during boot up some boot time memory for per-cpu
4329 variables need be pre-allocated for later physical cpu
4332 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
4356 soft-reserved memory partitioning.
4365 See Documentation/admin-guide/sysctl/vm.rst for details.
4368 See Documentation/core-api/debugging-via-ohci1394.rst for more
4380 For example, to override I2C bus2:
4383 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
4387 boundary - index of last SLC block on Flex-OneNAND.
4389 lock - Configure if Flex-OneNAND boundary should be locked.
4394 Always panic on oopses. Default is to just kill the
4398 Useful together with panic=30 to trigger a reboot.
4401 [KNL] Boolean flag to control whether the page allocator
4403 used to enable/disable page randomization. The state of
4408 page_owner= [KNL,EARLY] Boot-time page_owner enabling option.
4414 page_poison= [KNL,EARLY] Boot-time parameter changing the state of
4436 that will cause the kernel to panic when add_taint() is
4438 The optional switch "nousertaint" can be utilized to
4439 prevent userspace forced crashes by writing to sysctl
4442 See Documentation/admin-guide/tainted-kernels.rst for
4444 to compose the bitmask to assign to panic_on_taint.
4446 panic_on_warn=1 panic() instead of WARN(). Useful to cause kdump
4461 Use this option carefully, maybe worth to setup a
4465 connected to, default is 0.
4473 Use 'auto' to force the driver to use any
4474 IRQ/DMA settings detected (the default is to
4480 (to avoid using a FIFO even if it is detected).
4486 Configure VIA parallel port to operate in
4489 up parallel port mode and sets it to spp.
4495 Set to non-zero to probe primary and secondary ISA
4501 Set to non-zero if a chip is present that snoops speed
4506 Set to 1, 2, or 3 for HT 6560A on the primary channel,
4512 Set to 1, 2, or 3 for HT 6560B on the primary channel,
4518 IORDY enable mask. Set individual bits to allow IORDY
4522 correspond to the primary legacy channel, the secondary
4530 Set to 1, 2, or 3 for Opti 82c611A on the primary
4536 Set to 1, 2, or 3 for Opti 82c465MV on the primary
4543 bits to allow the use of the respective PIO modes.
4549 Set to non-zero to probe tertiary and further ISA
4556 options up to 6 legacy ports are supported: 0x1f0,
4559 corresponding bits in the mask to 1. Bit 0 is for
4565 Set to non-zero to probe QDI controllers. By default
4566 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
4570 Set to non-zero to probe Winbond controllers. Use
4572 value given is the I/O port to use (typically 0x1b0).
4573 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
4578 Supported PIO mode mask. Set individual bits to allow
4585 the specified number of seconds. This is to be used if
4602 firmware changes, or due to changes caused
4605 taken to be zero. Optionally, a path
4606 to a device through multiple device/function
4617 bios [X86-32] force use of PCI BIOS, don't access
4619 has a non-standard PCI host bridge.
4620 nobios [X86-32] disallow use of PCI BIOS, only direct
4626 data in IO port 0xCFC, both 32-bit).
4628 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
4629 the function, IO port 0xCFA, also 8-bit, sets
4631 through ports 0xC000-0xCFFF).
4635 enabled, this kernel boot option can be used to
4638 root domains (aka PCI segments, in ACPI-speak).
4642 properly configured MMIO access to PCI
4645 enabled, this kernel boot option can be used to
4646 disable the use of MSI interrupts system-wide.
4648 Safety option to keep boot IRQs enabled. This
4650 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
4651 primary IO-APIC for bridges that cannot disable
4655 boot IRQ equivalent of an IRQ that connects to
4658 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
4659 routing table. These calls are known to be buggy
4662 way to get the interrupt routing table. Try
4663 this option if the kernel is unable to allocate
4666 rom [X86] Assign address space to expansion ROMs.
4670 norom [X86] Do not assign address space to
4673 nobar [X86] Do not assign address space to the
4675 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
4676 assigned automatically to PCI devices. You can
4682 F0000h-100000h range.
4684 useful if the kernel is unable to find your
4685 secondary buses and you want to tell it
4687 assign-busses [X86] Always assign all PCI bus
4700 is enabled by default. If you need to use this,
4703 If you need to use this, please report a bug.
4704 use_e820 [X86] Use E820 reservations to exclude parts of
4707 If you need to use this, please report a bug to
4708 <linux-[email protected]>.
4711 hardware. If you need to use this, please report
4712 a bug to <linux-[email protected]>.
4717 skip_isa_align [X86] do not align io start addr, so can
4724 bfsort Sort PCI devices into breadth-first order.
4725 This sorting is done to get a device
4727 nobfsort Don't sort PCI devices into breadth-first order.
4729 tuning and use the BIOS-configured MPS defaults.
4730 pcie_bus_safe Set every device's MPS to the largest value
4732 pcie_bus_perf Set device MPS to the largest allowable MPS
4734 Read Request Size) to the largest supported
4737 pcie_bus_peer2peer Set every device's MPS to 128B, which
4738 every device is guaranteed to support. This
4739 configuration allows peer-to-peer DMA between
4742 that hot-added devices will work.
4752 Specifies alignment and device to reassign
4753 aligned memory resources. How to
4757 A PCI-PCI bridge can be specified if resource
4758 windows need to be expanded.
4759 To specify the alignment for several
4763 for 4096-byte alignment.
4765 end-to-end CRC checking). Only effective if
4789 if allocations done by BIOS are too small to
4796 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
4801 big_root_window Try to add a big 64bit memory window to the PCIe
4803 can resize a BAR to allow access to all VRAM.
4827 bit-0 : ACS Source Validation
4828 bit-1 : ACS Translation Blocking
4829 bit-2 : ACS P2P Request Redirect
4830 bit-3 : ACS P2P Completion Redirect
4831 bit-4 : ACS Upstream Forwarding
4832 bit-5 : ACS P2P Egress Control
4833 bit-6 : ACS Direct Translated P2P
4841 ACS to enable P2P Request Redirect, disable
4843 Validation unchanged from whatever power-up
4844 or firmware set it to.
4854 to disable PCIe TLP Processing Hints support
4855 system-wide.
4861 force Enable ASPM even on devices that claim not to support it.
4866 even if the platform doesn't give the OS permission to
4868 also tries to use these services.
4869 dpc-native Use native PCIe service for DPC only. May
4886 Keep all power-domains already enabled by bootloader on,
4897 Select which percpu first chunk allocator to use.
4904 pirq= [SMP,APIC] Manual mp-table setup
4905 See Documentation/arch/x86/i386/IO-APIC.rst.
4909 See also Documentation/admin-guide/parport.rst.
4918 PMU counters by setting MMCR0 to 0 (the FC bit is
4919 cleared). If a number is given, then MMCR1 is set to
4928 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4937 { on | off | curr | res | no-curr | no-res }
4955 Up to 8 (IP_VS_APP_MAX_PORTS) ports
4967 function to NULL. On Idle the CPU just reduces
4982 none - Limited to cond_resched() calls
4983 voluntary - Limited to cond_resched() and might_sleep() calls
4984 full - Any section that isn't explicitly preempt disabled
4988 lazy - Scheduler controlled. Similar to full but instead
4990 one HZ tick time to yield itself before the
4992 task returns to user space.
4994 print-fatal-signals=
5000 coredump - etc.
5002 If you hit the warning due to signal overflow,
5003 you might want to try "ulimit -i unlimited".
5015 or lockdep-detected issues (only if lock debug is on).
5016 With an exception to setups with low baudrate on
5018 in order to provide more debug information.
5023 Control writing to /dev/kmsg.
5024 on - unlimited logging to /dev/kmsg from userspace
5025 off - logging to /dev/kmsg disabled
5026 ratelimit - ratelimit the logging
5029 printk.time= Show timing data prefixed to each printk message line
5034 Traditionally /proc/pid/mem allows memory permissions to be
5035 overridden without restrictions. This option may be set to
5037 - 'always': traditional behavior always allows mem overrides.
5038 - 'ptrace': only allow mem overrides for active ptracers.
5039 - 'never': never allow mem overrides.
5043 Limit processor to maximum C-state
5047 Ignore the _CST method to determine C-states,
5053 [defaults to kernel profiling]
5054 Param: "schedule" - profile schedule points.
5055 Param: "kvm" - profile VM exits.
5056 Param: <number> - step/bucket size as a power of 2 for
5072 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
5077 Try to reset the device after so many bad packets
5085 pstore.backend= Specify the name of the pstore backend to use
5087 pti= [X86-64] Control Page Table Isolation of user and
5092 on - unconditionally enable
5093 off - unconditionally disable
5094 auto - kernel detects whether your CPU model is
5095 vulnerable to issues that PTI mitigates
5097 Not specifying this option is equivalent to pti=auto.
5100 [KNL] Number of legacy pty's. Overwrites compiled-in
5112 See Documentation/admin-guide/md.rst.
5115 See Documentation/admin-guide/blockdev/ramdisk.rst.
5117 ramdisk_start= [RAM] RAM disk image start address
5121 random number generator (if available) to
5126 passed by the bootloader (if available) to
5134 cross-syscall address exposures. This is only
5140 ras=option[,option,...] [KNL] RAS-specific options
5146 rcu_nocbs[=cpu-list]
5151 enable the no-callback CPU mode, which prevents
5154 callbacks will instead be offloaded to "rcuox/N"
5156 "p" for RCU-preempt, "s" for RCU-sched, and "g"
5160 and real-time workloads. It can also improve
5164 list of CPUs is set to no-callback mode from boot.
5167 arguments are omitted, no CPU will be set to
5168 no-callback mode from boot but the mode may be
5179 This improves the real-time response for the
5180 offloaded CPUs by relieving them of the need to
5183 periodically wake up to do the polling.
5186 Set maximum number of finished RCU callbacks to
5190 Do only a one-line RCU CPU stall warning when
5191 there is an ongoing too-long CSD-lock wait.
5194 Request a call to rcu_barrier(). This is
5197 If triggered before the RCU grace-period machinery
5203 purposes, to verify correct tree setup.
5206 Set the number of jiffies to delay each step of
5207 RCU grace-period cleanup.
5210 Set the number of jiffies to delay each step of
5211 RCU grace-period initialization.
5214 Set the number of jiffies to delay each step of
5215 RCU grace-period pre-initialization, that is,
5216 the propagation of recent CPU-hotplug changes up
5220 Set delay from grace-period initialization to
5221 first attempt to force quiescent states.
5226 Set delay between subsequent attempts to force
5233 soliciting quiescent-state help from
5240 rcutree.jiffies_to_sched_qs. Any attempt to set
5245 Set the SCHED_FIFO priority of the RCU per-CPU
5248 and for the RCU grace-period kthreads (rcu_bh,
5250 set, valid values are 1-99 and the default is 1
5251 (the least-favored priority). Otherwise, when
5252 RCU_BOOST is not set, valid values are 0-99 and
5253 the default is zero (non-realtime operation).
5258 On callback-offloaded (rcu_nocbs) CPUs,
5261 use of the ->nocb_bypass list. However, in the
5262 common non-flooded case, RCU queues directly to
5263 the main ->cblist in order to avoid the extra
5264 overhead of the ->nocb_bypass list and its lock.
5266 a single jiffy, RCU pre-queues the callbacks into
5267 the ->nocb_bypass queue. The definition of "too
5271 On callback-offloaded (rcu_nocbs) CPUs, avoid
5274 Defaults to zero. Large values will be capped
5276 to the nearest value representable by jiffies.
5284 batch limiting is re-enabled.
5288 RCU's force-quiescent-state scan will aggressively
5289 enlist help from cond_resched() and sched IPIs to
5291 Set to less than zero to make this be set based
5292 on rcutree.qhimark at boot time and to zero to
5296 Set the page-cache refill delay (in milliseconds)
5297 in response to low-memory conditions. The range
5301 Set the shift-right count to use to compute
5302 the callback-invocation batch limit bl from
5307 order to allow the CPU to do other work.
5309 Please note that this callback-invocation batch
5310 limit applies only to non-offloaded callback
5319 cache-to-cache transfer latencies.
5322 Change the number of CPUs assigned to each
5325 and for NUMA systems with large remote-access
5332 to PAGE_SIZE. The cache allows to reduce the
5333 pressure to page allocator, also it makes the
5334 whole algorithm to behave better in low memory
5339 each group, which defaults to the square root
5341 the wakeup overhead on the global grace-period
5343 each group's NOCB grace-period kthread.
5346 Cause the grace-period kthread to get an extra
5348 it should at force-quiescent-state time.
5354 callbacks to the specified number of nanoseconds.
5356 every 32 callbacks in order to limit the pain
5361 this specifies an rcu_read_unlock()-time delay
5362 in microseconds. This defaults to zero.
5365 of RCU-protected pointers after the relevant
5369 Commandeer a sysrq key to dump out Tree RCU's
5374 If set to zero, move all RCU_SOFTIRQ processing to
5375 per-CPU rcuc kthreads. Defaults to a non-zero
5377 Specify rcutree.use_softirq=0 to use rcuc kthreads.
5381 to zero.
5384 To save power, batch RCU callbacks and flush after
5395 How to enable it:
5404 grace-period primitives such as call_rcu().
5410 corresponding flavor of rcu_barrier() to allow
5411 previously posted callbacks to drain.
5415 grace-period primitives.
5418 Set test-start holdoff period. The purpose of
5419 this parameter is to delay the start of the
5420 test until boot completes in order to avoid
5430 Defaults to 1.
5433 Set to measure performance of kfree_rcu() flooding.
5436 Test the double-argument variant of kfree_rcu().
5438 rcuscale.kfree_rcu_test_single, both the single-
5439 and double-argument variants are tested.
5442 Test the single-argument variant of kfree_rcu().
5444 rcuscale.kfree_rcu_test_double, both the single-
5445 and double-argument variants are tested.
5459 does not affect the data-collection interval,
5464 Set number of RCU readers. The value -1 selects
5466 "n" less than -1 selects N-n+1, where N is again
5467 the number of CPUs. For example, -2 selects N
5468 (the number of CPUs), -3 selects N+1, and so on.
5469 A value of "n" less than or equal to -N selects
5478 Specify the RCU implementation to test.
5482 complete. This is useful for hands-off automated
5489 Write-side holdoff between grace periods,
5494 Additional write-side holdoff between grace
5511 Specifies the number of kthreads to be used
5512 for RCU grace-period forward-progress testing
5514 Defaults to 1 kthread, values less than zero or
5516 of CPUs to be used.
5519 Specify the fraction of a CPU-stall-warning
5520 period to do tight-loop forward-progress testing.
5523 Number of seconds to wait between successive
5524 forward-progress tests.
5528 need_resched() during tight-loop forward-progress
5532 Use conditional/asynchronous update-side
5533 normal-grace-period primitives, if available.
5536 Use conditional/asynchronous update-side
5537 expedited-grace-period primitives, if available.
5540 Use conditional/asynchronous update-side
5541 normal-grace-period primitives that also take
5546 Use conditional/asynchronous update-side
5547 expedited-grace-period primitives that also take
5556 be randomly selected to nanosecond granularity up
5557 to this wait interval. Defaults to 16 jiffies,
5566 interval will be randomly selected to nanosecond
5567 granularity up to this wait interval. Defaults to
5571 Use expedited update-side primitives, if available.
5574 Use normal (non-expedited) asynchronous
5575 update-side primitives, if available.
5578 Use polled update-side normal-grace-period
5582 Use polled update-side expedited-grace-period
5586 Use polled update-side normal-grace-period
5591 Use polled update-side expedited-grace-period
5600 be randomly selected to nanosecond granularity up
5601 to this wait interval. Defaults to 16 jiffies,
5610 interval will be randomly selected to nanosecond
5611 granularity up to this wait interval. Defaults to
5615 Use normal (non-expedited) synchronous
5616 update-side primitives, if available. If all
5620 they are all non-zero.
5625 flavors take kindly to this sort of thing.
5628 Leak an RCU-protected pointer out of the reader.
5630 intended to test the ability of things like
5631 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
5643 Set number of RCU callback-offload togglers.
5648 callback-offload toggling attempts.
5651 Set number of RCU readers. The value -1 selects
5652 N-1, where N is the number of CPUs. A value
5653 "n" less than -1 selects N-n-2, where N is again
5654 the number of CPUs. For example, -2 selects N
5655 (the number of CPUs), -3 selects N+1, and so on.
5658 Enable debug-object double-call_rcu() testing.
5661 Set time (s) after boot for CPU-hotplug testing.
5664 Set time (jiffies) between CPU-hotplug operations,
5665 or zero to disable CPU-hotplug testing.
5669 by a high-priority FIFO real-time task. Set to
5670 zero (the default) to disable. The CPUs to
5677 Set interval (in milliseconds, defaulting to one
5678 second) between preemptions by a high-priority
5679 FIFO real-time task. This delay is mediated
5680 by an hrtimer and is further fuzzed to avoid
5684 The number of times in a given read-then-exit
5685 episode that a set of read-then-exit kthreads
5690 read-then-exit testing episodes.
5693 A bit mask indicating which readers to use.
5695 are entered from low-order bit up, and are
5697 0x1 bit is normal readers, 0x2 NMI-safe readers,
5698 and 0x4 light-weight readers.
5701 Set task-shuffle interval (s). Shuffling tasks
5702 allows some CPUs to go into dyntick-idle mode
5707 is useful for hands-off automated testing.
5710 Duration of CPU stall (s) to test RCU CPU stall
5711 warnings, zero to disable.
5715 in warnings from preemptible RCU in addition to
5716 any other stall-related activity. Note that
5719 cause the CPU to pass through a quiescent state.
5722 in scheduling-while-atomic splats.
5728 Time to wait (s) after boot before inducing stall.
5735 Number of times to repeat the stall sequence,
5741 grace-period kthread to test RCU CPU stall
5742 warnings, zero to disable. If both stall_cpu
5750 Time (s) to stutter testing, for example, specifying
5751 five seconds causes the test to run for five seconds,
5753 ability to transition abruptly to and from idle.
5767 Test RCU's dyntick-idle handling. See also the
5771 Specify the RCU implementation to test.
5805 adjusted to an arch timer tick resolution.
5806 Setting this to zero causes the value from
5807 rcupdate.rcu_cpu_stall_timeout to be used (after
5808 conversion from seconds to milliseconds).
5822 Use expedited grace-period primitives, for
5826 real-time latency, and degrade energy efficiency.
5830 Use only normal grace-period primitives,
5833 real-time latency, CPU utilization, and
5834 energy efficiency, but can expose users to
5835 increased grace-period latency. This parameter
5842 only normal grace-period primitives. No effect
5847 it to the value one, that is, converting any
5848 post-boot attempt at an expedited RCU grace
5849 period to instead use normal non-expedited
5850 grace-period processing.
5855 the RCU Tasks flavors to collapse back to using
5858 set to the default value of -1.
5861 Set the minimum number of callback-queuing-time
5862 lock-contention events per jiffy required to
5863 cause the RCU Tasks flavors to switch to per-CPU
5865 when rcupdate.rcu_task_enqueue_lim is set to
5866 the default value of -1.
5869 Set the number of callback queues to use for the
5871 of -1 allows this to be automatically (and
5879 number avoids disturbing real-time workloads,
5884 cancel laziness on that CPU. Use -1 to disable
5886 doing so increases the danger of OOM due to
5892 of the problem for those not patient enough to
5894 only printed prior to the stall-warning message
5896 less than or equal to zero. Defaults to ten
5904 to one through ten, inclusive. It defaults to
5914 than or equal to zero. Defaults to ten minutes.
5942 force - Override the decision by the kernel to hide the
5952 E.g. to turn on cmt and turn off mba use:
5962 (prefix with 'panic_' to set mode for panic
5967 to be used for rebooting.
5981 There are some built-in platform specific "quirks"
5982 - you may see: "reboot: <name> series board detected.
5986 built-in quirk table, and use the generic default
5995 Don't stop other CPUs on reboot. This can make reboot
6002 Use a write to the PCI config space register 0xcf9 to
6019 Set test-start holdoff period. The purpose of
6020 this parameter is to delay the start of the
6021 test until boot completes in order to avoid
6025 Number of data elements to use for the forms of
6033 reduces noise due to loop start/end overhead,
6034 but the default has already reduced the per-pass
6035 noise to a handful of picoseconds on ca. 2020
6039 Set number of readers. The default value of -1
6048 Set the read-side critical-section duration,
6052 Specify the read-protection implementation to test.
6056 test. This defaults to 1 (shut it down) when
6057 refscale is built into the kernel and to 0 (leave
6078 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
6080 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
6084 is assumed to be I/O ports; otherwise it is memory.
6089 other subsystems can use to access it. This is typically
6091 line will try to reserve the same physical memory on
6092 soft reboots. Note, it is not guaranteed to be the same
6098 Any subsystem using this feature must add a way to verify
6103 The format is size:align:label for example, to request
6108 reservetop= [X86-32,EARLY]
6113 reset_devices [KNL] Force drivers to reset the underlying device
6125 See Documentation/power/swsusp-and-swap-files.rst
6127 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
6130 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
6141 AMD-based UNRET and IBPB mitigations alone do not stop
6143 sibling threads. For that reason, STIBP is used on pro-
6147 off - no mitigation
6148 auto - automatically select a migitation
6149 auto,nosmt - automatically select a mitigation,
6153 ibpb - On AMD, mitigate short speculation
6158 ibpb,nosmt - Like "ibpb" above but will disable SMT
6162 unret - Force enable untrained return thunks,
6163 only effective on AMD f15h-f17h based
6165 unret,nosmt - Like unret, but will disable SMT when STIBP
6170 time according to the CPU.
6172 Not specifying this option is equivalent to retbleed=auto.
6192 falling back to detecting extension support by parsing
6197 ro [KNL] Mount root device read-only on boot
6200 on Mark read-only kernel memory as read-only (default).
6201 off Leave read-only kernel memory writable for debugging.
6202 full Mark read-only kernel memory and aliases as read-only
6209 debug-uart get routed to the D+ and D- pins of the usb
6215 block/early-lookup.c for details.
6220 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
6227 rootwait [KNL] Wait (indefinitely) for root device to show up.
6231 rootwait= [KNL] Maximum time (in seconds) to wait for root device
6232 to show up before attempting to mount the root
6237 Memory area to be used by remote processor image,
6240 rw [KNL] Mount root device read-write on boot
6249 reuse, which is faster. Deprecated, equivalent to
6292 Number of seconds to hold off before starting
6293 test. Defaults to zero for module insertion and
6294 to 10 seconds for built-in smp_call_function()
6299 up to the chosen limit in seconds. Zero (the
6301 that requesting even small non-zero numbers of
6306 Number of kthreads to spawn to invoke the
6308 The default of -1 specifies a number of kthreads
6309 equal to the number of CPUs.
6312 Number seconds to wait after the start of the
6313 test before initiating CPU-hotplug operations.
6316 Number seconds to wait between successive
6317 CPU-hotplug operations. Specifying zero (which
6318 is the default) disables CPU-hotplug operations.
6321 The number of seconds following the start of the
6322 test after which to shut down the system. The
6324 Non-zero values are useful for automated tests.
6328 current test statistics to the console. A value
6332 The number of jiffies to wait between each change
6333 to the set of CPUs under test.
6337 preempt_disable() to disable CPU hotplug
6345 The probability weighting to use for the
6347 "wait" parameter. A value of -1 selects the
6348 default if all other weights are -1. However,
6350 value of -1 will instead select a weight of zero.
6353 The probability weighting to use for the
6355 non-zero "wait" parameter. See weight_single.
6358 The probability weighting to use for the
6366 The probability weighting to use for the
6368 non-zero "wait" parameter. See weight_single
6372 The probability weighting to use for the
6378 The probability weighting to use for the
6380 non-zero "wait" parameter. See weight_single
6387 skew_tick= [KNL,EARLY] Offset the periodic timer tick per cpu to mitigate
6391 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
6392 1 -- enable.
6396 security= [SECURITY] Choose a legacy "major" security module to
6403 0 -- disable.
6404 1 -- enable.
6407 serialnumber [BUGS=X86-32]
6409 sev=option[,option...] [X86-64]
6415 Do not enable SEV-SNP (applies to host/hypervisor
6417 in memory accesses when users do not want to run
6418 SEV-SNP guests.
6426 to set it to "all" by meaning -- no limit here.
6433 Enabling slab_debug allows one to determine the
6443 A high setting may cause OOMs due to memory
6455 increase the slab order up to slab_max_order to
6456 generate a sufficiently large slab able to contain
6459 and the less frequently locks need to be acquired.
6465 lower or equal to slab_max_order. For more information see
6471 necessary if there is some reason to distinguish
6472 allocs to different slabs, especially in hardened
6476 most of the exposure of a heap attack to a single
6486 policies to be applied at the folio level when
6501 for a CPU to release the CSD lock. This is
6504 of time. Defaults to 5,000 milliseconds, and
6507 using the csdlock_debug- kernel parameter.
6512 system. By default, let CSD-lock acquisition
6514 for this value provides a 5-minute timeout.
6516 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
6517 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
6518 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
6519 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
6520 smsc-ircc2.ircc_irq= [HW] IRQ line
6521 smsc-ircc2.ircc_dma= [HW] DMA channel
6522 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
6528 (logical CPUs) to use per physical CPU on systems
6530 be capped to the actual hardware limit.
6532 Default: -1 (no limit)
6535 [KNL] Should the soft-lockup detector generate panics.
6538 A value of 1 instructs the soft-lockup detector
6539 to panic the machine when a soft-lockup occurs. It is
6542 respective build-time switch to that functionality.
6545 [KNL] Should the soft-lockup detector generate
6550 See Documentation/admin-guide/laptops/sonypi.rst
6557 on - (default) Enable the HW or SW mitigation as
6560 vmexit - On systems which don't have the HW mitigation
6563 protected from VM-originated BHI attacks, but
6564 may still be vulnerable to syscall attacks.
6565 off - Disable the mitigation.
6572 on - unconditionally enable, implies
6574 off - unconditionally disable, implies
6576 auto - kernel detects whether your CPU model is
6580 mitigation method at run time according to the
6586 against user space to user space task attacks.
6593 retpoline - replace indirect branches
6594 retpoline,generic - Retpolines
6595 retpoline,lfence - LFENCE; indirect branch
6596 retpoline,amd - alias for retpoline,lfence
6597 eibrs - Enhanced/Auto IBRS
6598 eibrs,retpoline - Enhanced/Auto IBRS + Retpolines
6599 eibrs,lfence - Enhanced/Auto IBRS + LFENCE
6600 ibrs - use IBRS to protect kernel
6602 Not specifying this option is equivalent to
6610 on - Unconditionally enable mitigations. Is
6613 off - Unconditionally disable mitigations. Is
6616 prctl - Indirect branch speculation is enabled,
6622 - Like "prctl" above, but only STIBP is
6628 - Same as "prctl" above, but all seccomp
6633 - Like "seccomp" above, but only STIBP is
6638 auto - Kernel selects the mitigation depending on
6643 Not specifying this option is equivalent to
6649 off - Disable mitigation
6650 microcode - Enable microcode mitigation only
6651 safe-ret - Enable sw-only safe RET mitigation (default)
6652 ibpb - Enable mitigation by issuing IBPB on
6654 ibpb-vmexit - Issue IBPB only on VMEXIT
6655 (cloud-specific mitigation)
6661 Certain CPUs are vulnerable to an exploit against a
6664 to the same memory location may not be observed by
6667 be detected prior to instruction retirement at the
6672 example to read memory to which the attacker does not
6680 on - Unconditionally disable Speculative Store Bypass
6681 off - Unconditionally enable Speculative Store Bypass
6682 auto - Kernel detects whether the CPU model contains an
6688 prctl - Control Speculative Store Bypass per thread
6692 seccomp - Same as "prctl" above, but all seccomp threads
6700 on,auto - On Power8 and Power9 insert a store-forwarding
6704 off - No action.
6706 Not specifying this option is equivalent to
6718 off - not enabled
6720 warn - the kernel will emit rate-limited warnings
6728 fatal - the kernel will send SIGBUS to applications
6733 ratelimit:N -
6734 Set system wide rate limit to N bus locks
6753 Certain CPUs are vulnerable to an MDS-like
6759 the RDRAND and RDSEED instructions to become
6767 performance impact to RDRAND and RDSEED
6773 This kernel-boot parameter defaults to 128,
6774 but takes effect only when the low-order four
6775 bits of srcutree.convert_to_big is equal to 3
6780 srcu_struct structure will be converted to big
6785 2: When rcutorture decides to.
6791 instead of the compile-time CONFIG_NR_CPUS.
6794 Specifies how frequently to check for
6795 grace-period sequence counter wrap for the
6796 srcu_data structure's ->srcu_gp_seq_needed field.
6807 expediting. Set to zero to disable automatic
6811 Specifies the number of no-delay instances
6818 Specifies the per-grace-period phase, number of
6819 non-sleeping polls of readers. Beyond this limit,
6825 Specifies number of microseconds of non-sleeping
6826 delay between each non-sleeping poll of readers.
6829 Specifies the number of update-side contention
6832 structure to big form. Note that the value of
6834 set for contention-based conversions to occur.
6839 On CPUs that are vulnerable to the Speculative
6844 force-on: Unconditionally enable mitigation for
6846 force-off: Unconditionally disable mitigation for
6850 to allow userspace to register its
6856 to (for stacks growing down) resp. after (for stacks
6861 Setting this to true through kernel command line will
6864 to false.
6869 stacktrace_filter=[function-list]
6871 will trace at boot up. function-list is a comma-separated
6879 Set the STI (builtin display/keyboard on the HP-PARISC
6881 as the initial boot-console.
6896 be used to filter out binaries which have
6901 page table to increase the rate of hash page table
6906 them frequently to increase the rate of SLB faults
6915 An administrator who wishes to reserve some of these
6917 kernel's sunrpc client considers to be privileged
6918 using these two parameters to set the minimum and
6929 Control how the NFS server code allocates CPUs to
6941 to global on non-NUMA machines)
6947 RPC calls that can be sent from the client to a
6948 server. Increasing these values may allow you to
6954 Sets the number of seconds to remain in a suspend test
6966 <int> -- Number of I/O TLB slabs
6967 <int> -- Second integer after comma. Number of swiotlb
6969 to a power of 2.
6970 force -- force using of bounce buffers even if they
6972 noforce -- Never use bounce buffers (for debugging)
6978 process, as if the value was written to the respective
6987 Ignore sysrq setting - this boot parameter will
6994 ram pages. This is used to specify the TCP metrics
6995 cache size. See Documentation/networking/ip-sysctl.rst
7002 Specify "mem" (for Suspend-to-RAM) or "standby" (for
7005 the optional capability to repeat N number of times.
7007 wakeup-capable RTC alarm.
7013 -1: disable all active trip points in all thermal zones
7017 -1: disable all critical trip points in all thermal zones
7024 -1: disable all passive trip points
7025 <degrees C>: override all passive trip points to this
7030 <deci-seconds>: poll all this frequency
7034 Format: <size>[KMG],<size>[KMG]:<state>;<size>[KMG]-<size>[KMG]:<state>
7037 to anonymous transparent hugepages.
7039 See Documentation/admin-guide/mm/transhuge.rst for more
7047 Format: <size>[KMG],<size>[KMG]:<policy>;<size>[KMG]-<size>[KMG]:<policy>
7053 See Documentation/admin-guide/mm/transhuge.rst for more
7065 Prevent the CPU-hotplug component of torturing
7069 Dump the ftrace buffer at torture-test shutdown,
7073 with rotating-rust storage.
7078 disables verbose-printk() sleeping.
7081 Duration of each verbose-printk() sleep in jiffies.
7096 as a workaround for some chips which fail to
7103 for the FIFO interface. By default it is set to false
7106 https://trustedcomputinggroup.org/resource/pc-client-platform-tpm-profile-ptp-specification/
7109 Have the tracepoints sent to printk as well as the
7115 To turn off having tracepoints sent to printk,
7121 to stop the printing of events to console at
7126 Having tracepoints sent to printk() and activating high
7128 the system to live lock.
7132 on the console. It may be useful to only include the
7136 This command line option will stop the printing of events
7137 to console at the late_initcall_sync() time frame.
7144 local - Use the per CPU time stamp counter
7148 global - Event time stamps are synchronize across
7151 counter - Simple counting of events (1, 2, ..)
7152 note, some counts may be skipped due to the
7155 uptime - Use jiffies as the time stamp.
7156 perf - Use the same clock that perf uses.
7157 mono - Use ktime_get_mono_fast_ns() for time stamps.
7158 mono_raw - Use ktime_get_raw_fast_ns() for time
7160 boot - Use ktime_get_boot_fast_ns() for time stamps.
7164 trace_event=[event-list]
7165 [FTRACE] Set and start specified trace events in order
7166 to facilitate early boot debugging. The event-list is a
7167 comma-separated list of trace events to enable. See
7170 To enable modules, use :mod: keyword:
7178 trace_instance=[instance-info]
7198 Flags can be added to the instance to modify its behavior when it is
7203 traceoff - Have the tracing instance tracing disabled after it is created.
7204 traceprintk - Have trace_printk() write into this trace instance
7225 and place the ring buffer in this memory. Note that due to KASLR, the
7234 it is recommend to disable tracing so that events from a previous boot do not
7241 is set up to not wipe memory. For instance, CONFIG_RESET_ATTACK_MITIGATION
7249 trace_options=[option-list]
7251 The option-list is a comma delimited list of options
7253 to echo the option name into
7257 For example, to enable stacktrace option (to dump the
7258 stack trace of each event), add to the command line:
7265 trace_trigger=[trigger-list]
7285 [FTRACE] enable this option to disable tracing when a
7291 the WARNING dump is called, which prevents the trace to
7300 Can be used to control the default behavior of the system
7301 with respect to transparent hugepages.
7302 See Documentation/admin-guide/mm/transhuge.rst
7307 Can be used to control the hugepage allocation policy for
7309 See Documentation/admin-guide/mm/transhuge.rst
7314 Can be used to control the default hugepage allocation policy
7316 See Documentation/admin-guide/mm/transhuge.rst
7324 - "tpm"
7325 - "tee"
7326 - "caam"
7327 - "dcp"
7328 If not specified then it defaults to iterating through
7335 The RNG used to generate key material for trusted keys.
7337 - "kernel"
7338 - the same value as trusted.source: "tpm" or "tee"
7339 - "default"
7341 the RNG's choice is left to each individual trust source.
7344 This is intended to be used in combination with
7349 This is intended to be used in combination with
7359 as the stability checks done at bootup. Used to enable
7360 high-resolution timer mode on older hardware, and in
7362 [x86] noirqtime: Do not use TSC to do irq accounting.
7363 Used to run time disable IRQ_TIME_ACCOUNTING on any
7378 which to check other HW timers (HPET or PM timer), but
7396 on - Enable TSX on the system. Although there are
7398 TSX has been known to be an accelerator for
7399 several previous speculation-related CVEs, and
7403 off - Disable TSX on the system. (Note that this
7405 not vulnerable to MDS, i.e., have
7411 auto - Disable TSX if X86_BUG_TAA is present,
7414 Not specifying this option is equivalent to tsx=off.
7416 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
7422 Similar to Micro-architectural Data Sampling (MDS)
7424 Synchronization Extensions (TSX) are vulnerable to an
7426 information to a disclosure gadget under certain
7430 data can be used in a cache side channel attack, to
7431 access data to which the attacker does not have direct
7437 full - Enable TAA mitigation on vulnerable CPUs
7440 full,nosmt - Enable TAA mitigation and disable SMT on
7443 vulnerable to cross-thread TAA attacks.
7444 off - Unconditionally disable TAA mitigation
7446 On MDS-affected machines, tsx_async_abort=off can be
7448 are mitigated with the same mechanism so in order to disable
7449 this mitigation, you need to specify mds=off too.
7451 Not specifying this option is equivalent to
7458 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
7464 See also Documentation/input/devices/joystick-parport.rst
7466 udbg-immortal [PPC] When debugging early kernel crashes that
7472 Set number of hash buckets for UDP/UDP-Lite connections
7474 uhci-hcd.ignore_oc=
7476 Some badly-designed motherboards generate lots of
7477 bogus events, for ports that aren't wired to
7478 anything. Set this parameter to avoid log spamming.
7485 unwind_debug [X86-64,EARLY]
7493 (default -1 = authorized (same as 1),
7495 if device connected to internal port)
7499 for newly-detected USB devices (default 2). This
7502 to a negative value won't be autosuspended at all.
7505 [USB] Set to log all usbfs traffic (default 0 = off).
7508 [USB] Maximum number of bytes to snoop in each URB
7512 [USB] Set to cycle leds on hubs (default 0 = off).
7515 [USB] Start with the old device initialization
7527 [USB] Specifies timeout for the initial 64-byte
7534 [USB] A list of quirk entries to augment the built-in
7537 VendorID:ProductID:Flags. The IDs are 4-digit hex
7539 will change the built-in quirk; setting it if it is
7544 a 255-byte read);
7548 Set-Interface requests);
7557 talking to these interfaces);
7565 microframe = 125 microseconds) to be
7567 (bInterval-1).
7585 to be disconnected before suspend to
7593 request from 5000 ms to 500 ms);
7597 [USBHID] The interval which mice are to be polled at.
7600 [USBHID] The interval which joysticks are to be polled at.
7603 [USBHID] The interval which keyboards are to be polled at.
7605 usb-storage.delay_use=
7612 usb-storage.quirks=
7613 [UMS] A list of quirks entries to supplement or
7614 override the built-in unusual_devs list. List
7617 and Product ID values (4-digit hex numbers) and
7619 to a common usb-storage quirk flag as follows:
7637 i = IGNORE_DEVICE (don't bind to this
7642 l = NOT_LOCKABLE (don't try to lock and
7659 u = IGNORE_UAS (don't bind to the uas driver);
7661 medium is write-protected).
7670 1 - undefined instruction events
7671 2 - system calls
7672 4 - invalid data aborts
7673 8 - SIGSEGV faults
7674 16 - SIGBUS faults
7690 vdso32= [X86] Control the 32-bit vDSO
7691 vdso32=1: enable 32-bit VDSO
7692 vdso32=0 or vdso32=2: disable 32-bit VDSO
7702 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
7709 If set to 1, on receiving an ACPI notify event
7711 level and then send out the event to user space through
7712 the allocated input device. If set to 0, video driver
7725 <irq> := interrupt number (as passed to
7733 vga= [BOOT,X86-32] Select a particular video mode
7735 Documentation/admin-guide/svga.rst.
7738 passed to the kernel using a special protocol.
7744 interface is meant to allow for selectively
7750 - Disable all of the above options
7752 vmalloc=nn[KMG] [KNL,BOOT,EARLY] Forces the vmalloc area to have an
7753 exact size of <nn>. This can be used to increase
7755 It can also be used to decrease the size and leave more room
7774 vsyscall= [X86-64,EARLY]
7775 Controls the behavior of vsyscalls (i.e. calls to
7777 code). Most statically-linked binaries and older
7791 them quite hard to use for exploits but
7801 see vga-softcursor.rst. Default: 2 = underline.
7806 This is a 16-member array composed of values
7807 ranging from 0-255.
7812 This is a 16-member array composed of values
7813 ranging from 0-255.
7818 This is a 16-member array composed of values
7819 ranging from 0-255.
7824 Set system-wide default UTF-8 mode for all tty's.
7825 Default is 1, i.e. UTF-8 mode is enabled for all
7830 Format=<-1|0|1>
7831 Set system-wide default for whether a cursor
7832 is shown on new VTs. Default is -1,
7837 vt.italic= [VT] Default color for italic text; 0-15.
7840 vt.underline= [VT] Default color for underlined text; 0-15.
7844 see Documentation/watchdog/watchdog-parameters.rst
7845 or other driver-specific files in the
7852 threshold is set to twice the value. A value of 0
7857 [KNL,SMP] Specify to constrain one or some CPUs
7858 to use in unbound workqueues.
7859 Format: <cpu-list>
7865 warn stall conditions and dump internal state to
7869 it can be updated at runtime by writing to the
7875 stall to trigger panic.
7880 Per-cpu work items which run for longer than this
7882 and excluded from concurrency management to prevent
7883 them from noticeably delaying other per-cpu work
7894 intensive_threshold_us repeatedly. In order to prevent
7895 spurious warnings, start printing only after a work
7901 Per-cpu workqueues are generally preferred because
7902 they show better performance thanks to cache
7903 locality; unfortunately, per-cpu workqueues tend to
7906 Enabling this makes the per-cpu workqueues which
7907 were observed to contribute significantly to power
7908 consumption unbound, leading to measurably lower
7916 Select the default affinity scope to use for unbound
7920 Documentation/core-api/workqueue.rst.
7922 This can be changed after boot by writing to the
7928 Workqueue used to implicitly guarantee that work
7933 forces round-robin CPU selection to flush out
7941 on - Enable writecombine, use WUC for ioremap_wc()
7942 off - Disable writecombine, use SUC for ioremap_wc()
7944 x2apic_phys [X86-64,APIC,EARLY] Use x2apic physical mode instead of
7948 xen_512gb_limit [KNL,X86-64,XEN]
7950 to use only up to 512 GB of RAM. The reason to do so is
7952 save/restore/migration must be enabled to handle larger
7958 ide-disks -- unplug primary master IDE devices
7959 aux-ide-disks -- unplug non-primary-master IDE devices
7960 nics -- unplug network devices
7961 all -- unplug all emulated devices (NICs and IDE disks)
7962 unnecessary -- unplugging emulated devices is
7963 unnecessary even if the host did not respond to
7965 never -- do not unplug even if version check succeeds
7979 Select whether to always use non-faulting (safe) MSR
7984 Disables the PV optimizations forcing the HVM guest to
7994 Boolean option to control scrubbing pages before giving them back
7995 to Xen, for use by other domains. Can be also changed at runtime
7999 xen_timer_slop= [X86-64,XEN,EARLY]
8007 The time (in seconds) to wait before giving up to boot
8008 in case initial ballooning fails to free enough memory.
8014 How long to delay EOI handling in case of event
8019 should start to delay EOI handling. Default is 2.
8022 Boolean parameter to disable using fifo event handling
8024 preferred over the 2-level event handling, as it is
8035 allows the fallback firmware mode to be used:
8037 off Fallback to firmware control of XIVE interrupt
8041 xive.store-eoi=off [PPC]
8044 is active. This option allows the XIVE driver to use
8047 xhci-hcd.quirks [USB,KNL]
8055 Passing only "xmon" is equivalent to "xmon=early".