1 acpi= [HW,ACPI,X86,ARM64]
2 Advanced Configuration and Power Interface
3 Format: { force | on | off | strict | noirq | rsdt |
5 force -- enable ACPI if default was off
6 on -- enable ACPI but allow fallback to DT [arm64]
7 off -- disable ACPI if default was on
8 noirq -- do not use ACPI for IRQ routing
9 strict -- Be less tolerant of platforms that are not
10 strictly ACPI specification compliant.
11 rsdt -- prefer RSDT over (default) XSDT
12 copy_dsdt -- copy DSDT to memory
13 For ARM64, ONLY "acpi=off", "acpi=on" or "acpi=force"
16 See also Documentation/power/runtime_pm.rst, pci=noacpi
18 acpi_apic_instance= [ACPI, IOAPIC]
20 2: use 2nd APIC table, if available
21 1,0: use 1st APIC table
24 acpi_backlight= [HW,ACPI]
25 { vendor | video | native | none }
26 If set to vendor, prefer vendor-specific driver
27 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
28 of the ACPI video.ko driver.
29 If set to video, use the ACPI video.ko driver.
30 If set to native, use the device's native backlight mode.
31 If set to none, disable the ACPI backlight interface.
33 acpi_force_32bit_fadt_addr
34 force FADT to use 32 bit addresses rather than the
35 64 bit X_* addresses. Some firmware have broken 64
36 bit addresses for force ACPI ignore these and use
37 the older legacy 32 bit addresses.
39 acpica_no_return_repair [HW, ACPI]
40 Disable AML predefined validation mechanism
41 This mechanism can repair the evaluation result to make
42 the return objects more ACPI specification compliant.
43 This option is useful for developers to identify the
44 root cause of an AML interpreter issue when the issue
45 has something to do with the repair mechanism.
47 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
48 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
50 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
51 debug output. Bits in debug_layer correspond to a
52 _COMPONENT in an ACPI source file, e.g.,
53 #define _COMPONENT ACPI_EVENTS
54 Bits in debug_level correspond to a level in
55 ACPI_DEBUG_PRINT statements, e.g.,
56 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
57 The debug_level mask defaults to "info". See
58 Documentation/firmware-guide/acpi/debug.rst for more information about
59 debug layers and levels.
61 Enable processor driver info messages:
62 acpi.debug_layer=0x20000000
63 Enable AML "Debug" output, i.e., stores to the Debug
64 object while interpreting AML:
65 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
66 Enable all messages related to ACPI hardware:
67 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
69 Some values produce so much output that the system is
70 unusable. The "log_buf_len" parameter may be useful
71 if you need to capture more output.
73 acpi_enforce_resources= [ACPI]
75 Check for resource conflicts between native drivers
76 and ACPI OperationRegions (SystemIO and SystemMemory
77 only). IO ports and memory declared in ACPI might be
78 used by the ACPI subsystem in arbitrary AML code and
79 can interfere with legacy drivers.
80 strict (default): access to resources claimed by ACPI
81 is denied; legacy drivers trying to access reserved
82 resources will fail to bind to device using them.
83 lax: access to resources claimed by ACPI is allowed;
84 legacy drivers trying to access reserved resources
85 will bind successfully but a warning message is logged.
86 no: ACPI OperationRegions are not marked as reserved,
87 no further checks are performed.
89 acpi_force_table_verification [HW,ACPI]
90 Enable table checksum verification during early stage.
91 By default, this is disabled due to x86 early mapping
94 acpi_irq_balance [HW,ACPI]
95 ACPI will balance active IRQs
98 acpi_irq_nobalance [HW,ACPI]
99 ACPI will not move active IRQs (default)
102 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
103 Format: <irq>,<irq>...
105 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
107 Format: <irq>,<irq>...
109 acpi_mask_gpe= [HW,ACPI]
110 Due to the existence of _Lxx/_Exx, some GPEs triggered
111 by unsupported hardware/firmware features can result in
112 GPE floodings that cannot be automatically disabled by
114 This facility can be used to prevent such uncontrolled
116 Format: <byte> or <bitmap-list>
118 acpi_no_auto_serialize [HW,ACPI]
119 Disable auto-serialization of AML methods
120 AML control methods that contain the opcodes to create
121 named objects will be marked as "Serialized" by the
122 auto-serialization feature.
123 This feature is enabled by default.
124 This option allows to turn off the feature.
126 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
129 acpi_no_static_ssdt [HW,ACPI]
130 Disable installation of static SSDTs at early boot time
131 By default, SSDTs contained in the RSDT/XSDT will be
132 installed automatically and they will appear under
133 /sys/firmware/acpi/tables.
134 This option turns off this feature.
135 Note that specifying this option does not affect
136 dynamic table installation which will install SSDT
137 tables to /sys/firmware/acpi/tables/dynamic.
139 acpi_no_watchdog [HW,ACPI,WDT]
140 Ignore the ACPI-based watchdog interface (WDAT) and let
141 a native driver control the watchdog device instead.
143 acpi_rsdp= [ACPI,EFI,KEXEC]
144 Pass the RSDP address to the kernel, mostly used
145 on machines running EFI runtime service to boot the
146 second kernel for kdump.
148 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
149 Format: To spoof as Windows 98: ="Microsoft Windows"
151 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
152 of 2 which is mandated by ACPI 6) as the supported ACPI
153 specification revision (when using this switch, it may
154 be necessary to carry out a cold reboot _twice_ in a
155 row to make it take effect on the platform firmware).
157 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
158 acpi_osi="string1" # add string1
159 acpi_osi="!string2" # remove string2
160 acpi_osi=!* # remove all strings
161 acpi_osi=! # disable all built-in OS vendor
163 acpi_osi=!! # enable all built-in OS vendor
165 acpi_osi= # disable all strings
167 'acpi_osi=!' can be used in combination with single or
168 multiple 'acpi_osi="string1"' to support specific OS
169 vendor string(s). Note that such command can only
170 affect the default state of the OS vendor strings, thus
171 it cannot affect the default state of the feature group
172 strings and the current state of the OS vendor strings,
173 specifying it multiple times through kernel command line
174 is meaningless. This command is useful when one do not
175 care about the state of the feature group strings which
176 should be controlled by the OSPM.
178 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
179 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
180 can make '_OSI("Windows 2000")' TRUE.
182 'acpi_osi=' cannot be used in combination with other
183 'acpi_osi=' command lines, the _OSI method will not
184 exist in the ACPI namespace. NOTE that such command can
185 only affect the _OSI support state, thus specifying it
186 multiple times through kernel command line is also
189 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
192 'acpi_osi=!*' can be used in combination with single or
193 multiple 'acpi_osi="string1"' to support specific
194 string(s). Note that such command can affect the
195 current state of both the OS vendor strings and the
196 feature group strings, thus specifying it multiple times
197 through kernel command line is meaningful. But it may
198 still not able to affect the final state of a string if
199 there are quirks related to this string. This command
200 is useful when one want to control the state of the
201 feature group strings to debug BIOS issues related to
204 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
205 '_OSI("Module Device")' FALSE.
206 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
207 '_OSI("Module Device")' TRUE.
208 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
210 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
212 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
213 they all will make '_OSI("Windows 2000")' TRUE.
216 Override the pmtimer bug detection: force the kernel
217 to assume that this machine's pmtimer latches its value
218 and always returns good values.
220 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
221 Format: { level | edge | high | low }
223 acpi_skip_timer_override [HW,ACPI]
224 Recognize and ignore IRQ0/pin2 Interrupt Override.
225 For broken nForce2 BIOS resulting in XT-PIC timer.
227 acpi_sleep= [HW,ACPI] Sleep options
228 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
229 old_ordering, nonvs, sci_force_enable, nobl }
230 See Documentation/power/video.rst for information on
232 s3_beep is for debugging; it makes the PC's speaker beep
233 as soon as the kernel's real-mode entry point is called.
234 s4_nohwsig prevents ACPI hardware signature from being
235 used during resume from hibernation.
236 old_ordering causes the ACPI 1.0 ordering of the _PTS
237 control method, with respect to putting devices into
238 low power states, to be enforced (the ACPI 2.0 ordering
239 of _PTS is used by default).
240 nonvs prevents the kernel from saving/restoring the
241 ACPI NVS memory during suspend/hibernation and resume.
242 sci_force_enable causes the kernel to set SCI_EN directly
243 on resume from S1/S3 (which is against the ACPI spec,
244 but some broken systems don't work without it).
245 nobl causes the internal blacklist of systems known to
246 behave incorrectly in some ways with respect to system
247 suspend and resume to be ignored (use wisely).
249 acpi_use_timer_override [HW,ACPI]
250 Use timer override. For some broken Nvidia NF5 boards
251 that require a timer override, but don't have HPET
253 add_efi_memmap [EFI; X86] Include EFI memory map in
254 kernel's map of available physical RAM.
257 { off | try_unsupported }
258 off: disable AGP support
259 try_unsupported: try to drive unsupported chipsets
260 (may crash computer or cause data corruption)
263 See Documentation/sound/alsa-configuration.rst
266 Allow the default userspace alignment fault handler
267 behaviour to be specified. Bit 0 enables warnings,
268 bit 1 enables fixups, and bit 2 sends a segfault.
270 align_va_addr= [X86-64]
271 Align virtual addresses by clearing slice [14:12] when
272 allocating a VMA at process creation time. This option
273 gives you up to 3% performance improvement on AMD F15h
274 machines (where it is enabled by default) for a
275 CPU-intensive style benchmark, and it can vary highly in
276 a microbenchmark depending on workload and compiler.
278 32: only for 32-bit processes
279 64: only for 64-bit processes
280 on: enable for both 32- and 64-bit processes
281 off: disable for both 32- and 64-bit processes
283 alloc_snapshot [FTRACE]
284 Allocate the ftrace snapshot buffer on boot up when the
285 main buffer is allocated. This is handy if debugging
286 and you need to use tracing_snapshot() on boot up, and
287 do not want to use tracing_snapshot_alloc() as it needs
288 to be done where GFP_KERNEL allocations are allowed.
290 allow_mismatched_32bit_el0 [ARM64]
291 Allow execve() of 32-bit applications and setting of the
292 PER_LINUX32 personality on systems where only a strict
293 subset of the CPUs support 32-bit EL0. When this
294 parameter is present, the set of CPUs supporting 32-bit
295 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0
296 and hot-unplug operations may be restricted.
298 See Documentation/arm64/asymmetric-32bit.rst for more
301 amd_iommu= [HW,X86-64]
302 Pass parameters to the AMD IOMMU driver in the system.
304 fullflush - Deprecated, equivalent to iommu.strict=1
305 off - do not initialize any AMD IOMMU found in
307 force_isolation - Force device isolation for all
308 devices. The IOMMU driver is not
309 allowed anymore to lift isolation
310 requirements as needed. This option
311 does not override iommu=pt
312 force_enable - Force enable the IOMMU on platforms known
313 to be buggy with IOMMU enabled. Use this
316 amd_iommu_dump= [HW,X86-64]
317 Enable AMD IOMMU driver option to dump the ACPI table
318 for AMD IOMMU. With this option enabled, AMD IOMMU
319 driver will print ACPI tables for AMD IOMMU during
320 IOMMU initialization.
322 amd_iommu_intr= [HW,X86-64]
323 Specifies one of the following AMD IOMMU interrupt
325 legacy - Use legacy interrupt remapping mode.
326 vapic - Use virtual APIC mode, which allows IOMMU
327 to inject interrupts directly into guest.
328 This mode requires kvm-amd.avic=1.
329 (Default when IOMMU HW support is present.)
331 amijoy.map= [HW,JOY] Amiga joystick support
332 Map of devices attached to JOY0DAT and JOY1DAT
334 See also Documentation/input/joydev/joystick.rst
336 analog.map= [HW,JOY] Analog joystick and gamepad support
337 Specifies type or capabilities of an analog joystick
338 connected to one of 16 gameports
339 Format: <type1>,<type2>,..<type16>
342 Power management functions (SPARCstation-4/5 + deriv.)
344 Disable APC CPU standby support. SPARCstation-Fox does
345 not play well with APC CPU idle - disable it if you have
346 APC and your system crashes randomly.
348 apic= [APIC,X86] Advanced Programmable Interrupt Controller
349 Change the output verbosity while booting
350 Format: { quiet (default) | verbose | debug }
351 Change the amount of debugging information output
352 when initialising the APIC and IO-APIC components.
353 For X86-32, this can also be used to specify an APIC
355 Format: apic=driver_name
356 Examples: apic=bigsmp
358 apic_extnmi= [APIC,X86] External NMI delivery setting
359 Format: { bsp (default) | all | none }
360 bsp: External NMI is delivered only to CPU 0
361 all: External NMIs are broadcast to all CPUs as a
363 none: External NMI is masked for all CPUs. This is
364 useful so that a dump capture kernel won't be
368 See Documentation/networking/ipv6.rst.
370 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
371 Limit apic dumping. The parameter defines the maximal
372 number of local apics being dumped. Also it is possible
373 to set it to "all" by meaning -- no limit here.
374 Format: { 1 (default) | 2 | ... | all }.
375 The parameter valid if only apic=debug or
376 apic=verbose is specified.
377 Example: apic=debug show_lapic=all
379 apm= [APM] Advanced Power Management
380 See header of arch/x86/kernel/apm_32.c.
382 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
383 Format: <io>,<irq>,<nodeID>
385 arm64.nobti [ARM64] Unconditionally disable Branch Target
386 Identification support
388 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication
391 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension
396 atarimouse= [HW,MOUSE] Atari Mouse
398 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
399 EzKey and similar keyboards
401 atkbd.reset= [HW] Reset keyboard during initialization
403 atkbd.set= [HW] Select keyboard code set
404 Format: <int> (2 = AT (default), 3 = PS/2)
406 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
409 atkbd.softraw= [HW] Choose between synthetic and real raw mode
410 Format: <bool> (0 = real, 1 = synthetic (default))
412 atkbd.softrepeat= [HW]
413 Use software keyboard repeat
415 audit= [KNL] Enable the audit sub-system
416 Format: { "0" | "1" | "off" | "on" }
417 0 | off - kernel audit is disabled and can not be
418 enabled until the next reboot
419 unset - kernel audit is initialized but disabled and
420 will be fully enabled by the userspace auditd.
421 1 | on - kernel audit is initialized and partially
422 enabled, storing at most audit_backlog_limit
423 messages in RAM until it is fully enabled by the
427 audit_backlog_limit= [KNL] Set the audit queue size limit.
428 Format: <int> (must be >=0)
431 bau= [X86_UV] Enable the BAU on SGI UV. The default
432 behavior is to disable the BAU (i.e. bau=0).
433 Format: { "0" | "1" }
436 unset - Disable the BAU.
438 baycom_epp= [HW,AX25]
441 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
443 See header of drivers/net/hamradio/baycom_par.c.
445 baycom_ser_fdx= [HW,AX25]
446 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
447 Format: <io>,<irq>,<mode>[,<baud>]
448 See header of drivers/net/hamradio/baycom_ser_fdx.c.
450 baycom_ser_hdx= [HW,AX25]
451 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
452 Format: <io>,<irq>,<mode>
453 See header of drivers/net/hamradio/baycom_ser_hdx.c.
455 blkdevparts= Manual partition parsing of block device(s) for
456 embedded devices based on command line input.
457 See Documentation/block/cmdline-partition.rst
459 boot_delay= Milliseconds to delay each printk during boot.
460 Values larger than 10 seconds (10000) are changed to
465 Extended command line options can be added to an initrd
466 and this will cause the kernel to look for it.
468 See Documentation/admin-guide/bootconfig.rst
471 Disable BERT OS support on buggy BIOSes.
473 bgrt_disable [ACPI][X86]
474 Disable BGRT to avoid flickering OEM logo.
476 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
477 bttv.radio= Most important insmod options are available as
479 bttv.pll= See Documentation/admin-guide/media/bttv.rst
482 bulk_remove=off [PPC] This parameter disables the use of the pSeries
483 firmware feature for flushing multiple hpte entries
486 c101= [NET] Moxa C101 synchronous serial card
488 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
489 Sometimes CPU hardware bugs make them report the cache
490 size incorrectly. The kernel will attempt work arounds
491 to fix known problems, but for some CPUs it is not
492 possible to determine what the correct size should be.
493 This option provides an override for these situations.
496 [NET] Specifies amount of time (in seconds) that
497 the kernel should wait for a network carrier. By default
498 it waits 120 seconds.
500 ca_keys= [KEYS] This parameter identifies a specific key(s) on
501 the system trusted keyring to be used for certificate
503 format: { id:<keyid> | builtin }
505 cca= [MIPS] Override the kernel pages' cache coherency
506 algorithm. Accepted values range from 0 to 7
507 inclusive. See arch/mips/include/asm/pgtable-bits.h
508 for platform specific values (SB1, Loongson3 and
511 ccw_timeout_log [S390]
512 See Documentation/s390/common_io.rst for details.
514 cgroup_disable= [KNL] Disable a particular controller or optional feature
515 Format: {name of the controller(s) or feature(s) to disable}
516 The effects of cgroup_disable=foo are:
517 - foo isn't auto-mounted if you mount all cgroups in
519 - foo isn't visible as an individually mountable
521 - if foo is an optional feature then the feature is
522 disabled and corresponding cgroup files are not
524 {Currently only "memory" controller deal with this and
525 cut the overhead, others just disable the usage. So
526 only cgroup_disable=memory is actually worthy}
527 Specifying "pressure" disables per-cgroup pressure
528 stall information accounting feature
530 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
531 Format: { { controller | "all" | "named" }
532 [,{ controller | "all" | "named" }...] }
533 Like cgroup_disable, but only applies to cgroup v1;
534 the blacklisted controllers remain available in cgroup2.
535 "all" blacklists all controllers and "named" disables
536 named mounts. Specifying both "all" and "named" disables
539 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
541 nosocket -- Disable socket memory accounting.
542 nokmem -- Disable kernel memory accounting.
544 checkreqprot [SELINUX] Set initial checkreqprot flag value.
545 Format: { "0" | "1" }
546 See security/selinux/Kconfig help text.
547 0 -- check protection applied by kernel (includes
548 any implied execute protection).
549 1 -- check protection requested by application.
550 Default value is set via a kernel config option.
551 Value can be changed at runtime via
552 /sys/fs/selinux/checkreqprot.
553 Setting checkreqprot to 1 is deprecated.
556 See Documentation/s390/common_io.rst for details.
559 Prevents the clock framework from automatically gating
560 clocks that have not been explicitly enabled by a Linux
561 device driver but are enabled in hardware at reset or
562 by the bootloader/firmware. Note that this does not
563 force such clocks to be always-on nor does it reserve
564 those clocks in any way. This parameter is useful for
565 debug and development, but should not be needed on a
566 platform with proper driver support. For more
567 information, see Documentation/driver-api/clk.rst.
569 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
571 Forces specified clocksource (if available) to be used
572 when calculating gettimeofday(). If specified
573 clocksource is not available, it defaults to PIT.
574 Format: { pit | tsc | cyclone | pmtmr }
576 clocksource= Override the default clocksource
578 Override the default clocksource and use the clocksource
579 with the name specified.
580 Some clocksource names to choose from, depending on
582 [all] jiffies (this is the base, fallback clocksource)
584 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
585 pxa_timer,timer3,32k_counter,timer0_1
586 [X86-32] pit,hpet,tsc;
587 scx200_hrt on Geode; cyclone on IBM x440
595 clocksource.arm_arch_timer.evtstrm=
598 Enable/disable the eventstream feature of the ARM
599 architected timer so that code using WFE-based polling
600 loops can be debugged more effectively on production
603 clocksource.max_cswd_read_retries= [KNL]
604 Number of clocksource_watchdog() retries due to
605 external delays before the clock will be marked
606 unstable. Defaults to three retries, that is,
607 four attempts to read the clock under test.
609 clocksource.verify_n_cpus= [KNL]
610 Limit the number of CPUs checked for clocksources
611 marked with CLOCK_SOURCE_VERIFY_PERCPU that
612 are marked unstable due to excessive skew.
613 A negative value says to check all CPUs, while
614 zero says not to check any. Values larger than
615 nr_cpu_ids are silently truncated to nr_cpu_ids.
616 The actual CPUs are chosen randomly, with
617 no replacement if the same CPU is chosen twice.
619 clocksource-wdtest.holdoff= [KNL]
620 Set the time in seconds that the clocksource
621 watchdog test waits before commencing its tests.
622 Defaults to zero when built as a module and to
623 10 seconds when built into the kernel.
625 clearcpuid=BITNUM[,BITNUM...] [X86]
626 Disable CPUID feature X for the kernel. See
627 arch/x86/include/asm/cpufeatures.h for the valid bit
628 numbers. Note the Linux specific bits are not necessarily
629 stable over kernel options, but the vendor specific
631 Also note that user programs calling CPUID directly
632 or using the feature without checking anything
633 will still see it. This just prevents it from
634 being used by the kernel or shown in /proc/cpuinfo.
635 Also note the kernel might malfunction if you disable
638 cma=nn[MG]@[start[MG][-end[MG]]]
640 Sets the size of kernel global memory area for
641 contiguous memory allocations and optionally the
642 placement constraint by the physical address range of
643 memory allocations. A value of 0 disables CMA
644 altogether. For more information, see
645 kernel/dma/contiguous.c
649 Sets the size of kernel per-numa memory area for
650 contiguous memory allocations. A value of 0 disables
651 per-numa CMA altogether. And If this option is not
652 specificed, the default value is 0.
653 With per-numa CMA enabled, DMA users on node nid will
654 first try to allocate buffer from the pernuma area
655 which is located in node nid, if the allocation fails,
656 they will fallback to the global default memory area.
658 cmo_free_hint= [PPC] Format: { yes | no }
659 Specify whether pages are marked as being inactive
660 when they are freed. This is used in CMO environments
661 to determine OS memory pressure for page stealing by
665 coherent_pool=nn[KMG] [ARM,KNL]
666 Sets the size of memory pool for coherent, atomic dma
667 allocations, by default set to 256K.
669 com20020= [HW,NET] ARCnet - COM20020 chipset
671 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
673 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
677 ARCnet - COM90xx chipset (memory-mapped buffers)
678 Format: <io>[,<irq>[,<memstart>]]
680 condev= [HW,S390] console device
683 console= [KNL] Output console device and options.
685 tty<n> Use the virtual console device <n>.
689 Use the specified serial port. The options are of
690 the form "bbbbpnf", where "bbbb" is the baud rate,
691 "p" is parity ("n", "o", or "e"), "n" is number of
692 bits, and "f" is flow control ("r" for RTS or
693 omit it). Default is "9600n8".
695 See Documentation/admin-guide/serial-console.rst for more
697 Documentation/networking/netconsole.rst for an
700 uart[8250],io,<addr>[,options]
701 uart[8250],mmio,<addr>[,options]
702 uart[8250],mmio16,<addr>[,options]
703 uart[8250],mmio32,<addr>[,options]
704 uart[8250],0x<addr>[,options]
705 Start an early, polled-mode console on the 8250/16550
706 UART at the specified I/O port or MMIO address,
707 switching to the matching ttyS device later.
708 MMIO inter-register address stride is either 8-bit
709 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
710 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
711 to be equivalent to 'mmio'. 'options' are specified in
712 the same format described for ttyS above; if unspecified,
713 the h/w is not re-initialized.
715 hvc<n> Use the hypervisor console device <n>. This is for
716 both Xen and PowerPC hypervisors.
718 If the device connected to the port is not a TTY but a braille
719 device, prepend "brl," before the device type, for instance
721 For now, only VisioBraille is supported.
724 [KNL] Change console messages format
726 By default we print messages on consoles in
727 "[time stamp] text\n" format (time stamp may not be
728 printed, depending on CONFIG_PRINTK_TIME or
729 `printk_time' param).
731 Switch to syslog format: "<%u>[time stamp] text\n"
732 IOW, each message will have a facility and loglevel
733 prefix. The format is similar to one used by syslog()
734 syscall, or to executing "dmesg -S --raw" or to reading
737 consoleblank= [KNL] The console blank (screen saver) timeout in
738 seconds. A value of 0 disables the blank timer.
742 [KNL] Change the default value for
743 /proc/<pid>/coredump_filter.
744 See also Documentation/filesystems/proc.rst.
746 coresight_cpu_debug.enable
749 Enable/disable the CPU sampling based debugging.
750 0: default value, disable debugging
751 1: enable debugging at boot time
753 cpuidle.off=1 [CPU_IDLE]
754 disable the cpuidle sub-system
757 [CPU_IDLE] Name of the cpuidle governor to use.
759 cpufreq.off=1 [CPU_FREQ]
760 disable the cpufreq sub-system
762 cpufreq.default_governor=
763 [CPU_FREQ] Name of the default cpufreq governor or
764 policy to use. This governor must be registered in the
765 kernel before the cpufreq driver probes.
768 [X86] Delay for N microsec between assert and de-assert
769 of APIC INIT to start processors. This delay occurs
770 on every CPU online, such as boot, and resume from suspend.
773 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
775 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
777 crashkernel=size[KMG][@offset[KMG]]
778 [KNL] Using kexec, Linux can switch to a 'crash kernel'
779 upon panic. This parameter reserves the physical
780 memory region [offset, offset + size] for that kernel
781 image. If '@offset' is omitted, then a suitable offset
782 is selected automatically.
783 [KNL, X86-64] Select a region under 4G first, and
784 fall back to reserve region above 4G when '@offset'
785 hasn't been specified.
786 See Documentation/admin-guide/kdump/kdump.rst for further details.
788 crashkernel=range1:size1[,range2:size2,...][@offset]
789 [KNL] Same as above, but depends on the memory
790 in the running system. The syntax of range is
791 start-[end] where start and end are both
792 a memory unit (amount[KMG]). See also
793 Documentation/admin-guide/kdump/kdump.rst for an example.
795 crashkernel=size[KMG],high
796 [KNL, X86-64] range could be above 4G. Allow kernel
797 to allocate physical memory region from top, so could
798 be above 4G if system have more than 4G ram installed.
799 Otherwise memory region will be allocated below 4G, if
801 It will be ignored if crashkernel=X is specified.
802 crashkernel=size[KMG],low
803 [KNL, X86-64] range under 4G. When crashkernel=X,high
804 is passed, kernel could allocate physical memory region
805 above 4G, that cause second kernel crash on system
806 that require some amount of low memory, e.g. swiotlb
807 requires at least 64M+32K low memory, also enough extra
808 low memory is needed to make sure DMA buffers for 32-bit
809 devices won't run out. Kernel would try to allocate at
810 at least 256M below 4G automatically.
811 This one let user to specify own low range under 4G
812 for second kernel instead.
813 0: to disable low allocation.
814 It will be ignored when crashkernel=X,high is not used
815 or memory reserved is below 4G.
818 [KNL] Disable crypto self-tests
823 cs89x0_media= [HW,NET]
824 Format: { rj45 | aui | bnc }
826 csdlock_debug= [KNL] Enable debug add-ons of cross-CPU function call
827 handling. When switched on, additional debug data is
828 printed to the console in case a hanging CPU is
829 detected, and that CPU is pinged again in order to try
830 to resolve the hang situation.
831 0: disable csdlock debugging (default)
832 1: enable basic csdlock debugging (minor impact)
833 ext: enable extended csdlock debugging (more impact,
837 See header of drivers/s390/block/dasd_devmap.c.
839 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
840 (one device per port)
841 Format: <port#>,<type>
842 See also Documentation/input/devices/joystick-parport.rst
844 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
846 Documentation/admin-guide/dynamic-debug-howto.rst for
847 details. Deprecated, see dyndbg.
849 debug [KNL] Enable kernel debugging (events log level).
852 [KNL] Enable printing [hashed] pointers early in the
853 boot sequence. If enabled, we use a weak hash instead
854 of siphash to hash pointers. Use this option if you are
855 seeing instances of '(___ptrval___)') and need to see a
856 value (hashed pointer) instead. Cryptographically
857 insecure, please do not use on production kernels.
860 [KNL] verbose locking self-tests
862 Print debugging info while doing the locking API
864 Bitmask for the various LOCKTYPE_ tests. Defaults to 0
865 (no extra messages), setting it to -1 (all bits set)
866 will print _a_lot_ more information - normally only
867 useful to lockdep developers.
869 debug_objects [KNL] Enable object debugging
872 [KNL] Disable object debugging
874 debug_guardpage_minorder=
875 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
876 parameter allows control of the order of pages that will
877 be intentionally kept free (and hence protected) by the
878 buddy allocator. Bigger value increase the probability
879 of catching random memory corruption, but reduce the
880 amount of memory for normal system use. The maximum
881 possible value is MAX_ORDER/2. Setting this parameter
882 to 1 or 2 should be enough to identify most random
883 memory corruption problems caused by bugs in kernel or
884 driver code when a CPU writes to (or reads from) a
885 random memory location. Note that there exists a class
886 of memory corruptions problems caused by buggy H/W or
887 F/W or by drivers badly programing DMA (basically when
888 memory is written at bus level and the CPU MMU is
889 bypassed) which are not detectable by
890 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
891 tracking down these problems.
894 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
895 enables the feature at boot time. By default, it is
896 disabled and the system will work mostly the same as a
897 kernel built without CONFIG_DEBUG_PAGEALLOC.
898 Note: to get most of debug_pagealloc error reports, it's
899 useful to also enable the page_owner functionality.
900 on: enable the feature
902 debugfs= [KNL] This parameter enables what is exposed to userspace
903 and debugfs internal clients.
904 Format: { on, no-mount, off }
905 on: All functions are enabled.
907 Filesystem is not registered but kernel clients can
908 access APIs and a crashkernel can be used to read
909 its content. There is nothing to mount.
910 off: Filesystem is not registered and clients
911 get a -EPERM as result when trying to register files
912 or directories within debugfs.
913 This is equivalent of the runtime functionality if
914 debugfs was not enabled in the kernel at all.
915 Default value is set in build-time with a kernel configuration.
917 debugpat [X86] Enable PAT debugging
919 decnet.addr= [HW,NET]
920 Format: <area>[,<node>]
921 See also Documentation/networking/decnet.rst.
924 [HW] The size of the default HugeTLB page. This is
925 the size represented by the legacy /proc/ hugepages
926 APIs. In addition, this is the default hugetlb size
927 used for shmget(), mmap() and mounting hugetlbfs
928 filesystems. If not specified, defaults to the
929 architecture's default huge page size. Huge page
930 sizes are architecture dependent. See also
931 Documentation/admin-guide/mm/hugetlbpage.rst.
934 deferred_probe_timeout=
935 [KNL] Debugging option to set a timeout in seconds for
936 deferred probe to give up waiting on dependencies to
937 probe. Only specific dependencies (subsystems or
938 drivers) that have opted in will be ignored. A timeout of 0
939 will timeout at the end of initcalls. This option will also
940 dump out devices still on the deferred probe list after
944 Format: { on | off | def_only | inf_only | always }
945 on: s390 zlib hardware support for compression on
946 level 1 and decompression (default)
947 off: No s390 zlib hardware support
948 def_only: s390 zlib hardware support for deflate
949 only (compression on level 1)
950 inf_only: s390 zlib hardware support for inflate
952 always: Same as 'on' but ignores the selected compression
953 level always using hardware support (used for debugging)
956 Set number of hash buckets for dentry cache.
958 disable_1tb_segments [PPC]
959 Disables the use of 1TB hash page table segments. This
960 causes the kernel to fall back to 256MB segments which
961 can be useful when debugging issues that require an SLB
965 Limits the number of kernel SLB entries, and flushes
966 them frequently to increase the rate of SLB faults
970 See Documentation/networking/ipv6.rst.
973 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
974 hardening is enabled for this boot. Hardened
975 usercopy checking is used to protect the kernel
976 from reading or writing beyond known memory
977 allocation boundaries as a proactive defense
978 against bounds-checking flaws in the kernel's
979 copy_to_user()/copy_from_user() interface.
980 on Perform hardened usercopy checks (default).
981 off Disable hardened usercopy checks.
984 Disable RADIX MMU mode on POWER9
986 radix_hcall_invalidate=on [PPC/PSERIES]
987 Disable RADIX GTSE feature and use hcall for TLB
991 Disable TLBIE instruction. Currently does not work
992 with KVM, with HASH MMU, or with coherent accelerators.
994 disable_cpu_apicid= [X86,APIC,SMP]
996 The number of initial APIC ID for the
997 corresponding CPU to be disabled at boot,
998 mostly used for the kdump 2nd kernel to
999 disable BSP to wake up multiple CPUs without
1000 causing system reset or hang due to sending
1001 INIT from AP to BSP.
1003 disable_ddw [PPC/PSERIES]
1004 Disable Dynamic DMA Window support. Use this
1005 to workaround buggy firmware.
1007 disable_ipv6= [IPV6]
1008 See Documentation/networking/ipv6.rst.
1010 disable_mtrr_cleanup [X86]
1011 The kernel tries to adjust MTRR layout from continuous
1012 to discrete, to make X server driver able to add WB
1013 entry later. This parameter disables that.
1015 disable_mtrr_trim [X86, Intel and AMD only]
1016 By default the kernel will trim any uncacheable
1017 memory out of your available memory pool based on
1018 MTRR settings. This parameter disables that behavior,
1019 possibly causing your machine to run very slowly.
1021 disable_timer_pin_1 [X86]
1022 Disable PIN 1 of APIC timer
1023 Can be useful to work around chipset bugs.
1025 dis_ucode_ldr [X86] Disable the microcode loader.
1027 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
1028 this option disables the debugging code at boot.
1030 dma_debug_entries=<number>
1031 This option allows to tune the number of preallocated
1032 entries for DMA-API debugging code. One entry is
1033 required per DMA-API allocation. Use this if the
1034 DMA-API debugging code disables itself because the
1035 architectural default is too low.
1037 dma_debug_driver=<driver_name>
1038 With this option the DMA-API debugging driver
1039 filter feature can be enabled at boot time. Just
1040 pass the driver to filter for as the parameter.
1041 The filter can be disabled or changed to another
1042 driver later using sysfs.
1044 driver_async_probe= [KNL]
1045 List of driver names to be probed asynchronously.
1046 Format: <driver_name1>,<driver_name2>...
1048 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
1049 Broken monitors, graphic adapters, KVMs and EDIDless
1050 panels may send no or incorrect EDID data sets.
1051 This parameter allows to specify an EDID data sets
1052 in the /lib/firmware directory that are used instead.
1053 Generic built-in EDID data sets are used, if one of
1054 edid/1024x768.bin, edid/1280x1024.bin,
1055 edid/1680x1050.bin, or edid/1920x1080.bin is given
1056 and no file with the same name exists. Details and
1057 instructions how to build your own EDID data are
1058 available in Documentation/admin-guide/edid.rst. An EDID
1059 data set will only be used for a particular connector,
1060 if its name and a colon are prepended to the EDID
1061 name. Each connector may use a unique EDID data
1062 set by separating the files with a comma. An EDID
1063 data set with no connector name will be used for
1064 any connectors not explicitly specified.
1069 Format: {"off" | "known"}
1070 Control how the dt_cpu_ftrs device-tree binding is
1071 used for CPU feature discovery and setup (if it
1073 off: Do not use it, fall back to legacy cpu table.
1074 known: Do not pass through unknown features to guests
1075 or userspace, only those that the kernel is aware of.
1077 dump_apple_properties [X86]
1078 Dump name and content of EFI device properties on
1079 x86 Macs. Useful for driver authors to determine
1080 what data is available or for reverse-engineering.
1082 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1083 <module>.dyndbg[="val"]
1084 Enable debug messages at boot time. See
1085 Documentation/admin-guide/dynamic-debug-howto.rst
1088 nopku [X86] Disable Memory Protection Keys CPU feature found
1091 <module>.async_probe [KNL]
1092 Enable asynchronous probe on this module.
1094 early_ioremap_debug [KNL]
1095 Enable debug messages in early_ioremap support. This
1096 is useful for tracking down temporary early mappings
1097 which are not unmapped.
1099 earlycon= [KNL] Output early console device and options.
1101 When used with no options, the early console is
1102 determined by stdout-path property in device tree's
1103 chosen node or the ACPI SPCR table if supported by
1106 cdns,<addr>[,options]
1107 Start an early, polled-mode console on a Cadence
1108 (xuartps) serial port at the specified address. Only
1109 supported option is baud rate. If baud rate is not
1110 specified, the serial port must already be setup and
1113 uart[8250],io,<addr>[,options]
1114 uart[8250],mmio,<addr>[,options]
1115 uart[8250],mmio32,<addr>[,options]
1116 uart[8250],mmio32be,<addr>[,options]
1117 uart[8250],0x<addr>[,options]
1118 Start an early, polled-mode console on the 8250/16550
1119 UART at the specified I/O port or MMIO address.
1120 MMIO inter-register address stride is either 8-bit
1121 (mmio) or 32-bit (mmio32 or mmio32be).
1122 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1123 to be equivalent to 'mmio'. 'options' are specified
1124 in the same format described for "console=ttyS<n>"; if
1125 unspecified, the h/w is not initialized.
1129 Start an early, polled-mode console on a pl011 serial
1130 port at the specified address. The pl011 serial port
1131 must already be setup and configured. Options are not
1132 yet supported. If 'mmio32' is specified, then only
1133 the driver will use only 32-bit accessors to read/write
1134 the device registers.
1137 Start an early console on a litex serial port at the
1138 specified address. The serial port must already be
1139 setup and configured. Options are not yet supported.
1142 Start an early, polled-mode console on a meson serial
1143 port at the specified address. The serial port must
1144 already be setup and configured. Options are not yet
1148 Start an early, polled-mode console on an msm serial
1149 port at the specified address. The serial port
1150 must already be setup and configured. Options are not
1153 msm_serial_dm,<addr>
1154 Start an early, polled-mode console on an msm serial
1155 dm port at the specified address. The serial port
1156 must already be setup and configured. Options are not
1160 Start an early, polled-mode console on a serial port
1161 of an Actions Semi SoC, such as S500 or S900, at the
1162 specified address. The serial port must already be
1163 setup and configured. Options are not yet supported.
1166 Start an early, polled-mode console on a serial port
1167 of an RDA Micro SoC, such as RDA8810PL, at the
1168 specified address. The serial port must already be
1169 setup and configured. Options are not yet supported.
1172 Use RISC-V SBI (Supervisor Binary Interface) for early
1175 smh Use ARM semihosting calls for early console.
1183 Use early console provided by serial driver available
1184 on Samsung SoCs, requires selecting proper type and
1185 a correct base address of the selected UART port. The
1186 serial port must already be setup and configured.
1187 Options are not yet supported.
1190 Start an early, polled-mode console on a lantiq serial
1191 (lqasc) port at the specified address. The serial port
1192 must already be setup and configured. Options are not
1197 Use early console provided by Freescale LP UART driver
1198 found on Freescale Vybrid and QorIQ LS1021A processors.
1199 A valid base address must be provided, and the serial
1200 port must already be setup and configured.
1204 Start an early, polled-mode, output-only console on the
1205 Freescale i.MX UART at the specified address. The UART
1206 must already be setup and configured.
1209 Start an early, polled-mode console on the
1210 Armada 3700 serial port at the specified
1211 address. The serial port must already be setup
1212 and configured. Options are not yet supported.
1215 Start an early, polled-mode console on a Qualcomm
1216 Generic Interface (GENI) based serial port at the
1217 specified address. The serial port must already be
1218 setup and configured. Options are not yet supported.
1221 Start an early, unaccelerated console on the EFI
1222 memory mapped framebuffer (if available). On cache
1223 coherent non-x86 systems that use system memory for
1224 the framebuffer, pass the 'ram' option so that it is
1225 mapped with the correct attributes.
1228 Use early console provided by Freescale LINFlexD UART
1229 serial driver for NXP S32V234 SoCs. A valid base
1230 address must be provided, and the serial port must
1231 already be setup and configured.
1233 earlyprintk= [X86,SH,ARM,M68k,S390]
1237 earlyprintk=serial[,ttySn[,baudrate]]
1238 earlyprintk=serial[,0x...[,baudrate]]
1239 earlyprintk=ttySn[,baudrate]
1240 earlyprintk=dbgp[debugController#]
1241 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1242 earlyprintk=xdbc[xhciController#]
1244 earlyprintk is useful when the kernel crashes before
1245 the normal console is initialized. It is not enabled by
1246 default because it has some cosmetic problems.
1248 Append ",keep" to not disable it when the real console
1251 Only one of vga, efi, serial, or usb debug port can
1254 Currently only ttyS0 and ttyS1 may be specified by
1255 name. Other I/O ports may be explicitly specified
1256 on some architectures (x86 and arm at least) by
1257 replacing ttySn with an I/O port address, like this:
1258 earlyprintk=serial,0x1008,115200
1259 You can find the port for a given device in
1260 /proc/tty/driver/serial:
1261 2: uart:ST16650V2 port:00001008 irq:18 ...
1263 Interaction with the standard serial driver is not
1266 The VGA and EFI output is eventually overwritten by
1269 The xen option can only be used in Xen domains.
1271 The sclp output can only be used on s390.
1273 The optional "force" to "pciserial" enables use of a
1274 PCI device even when its classcode is not of the
1277 edac_report= [HW,EDAC] Control how to report EDAC event
1278 Format: {"on" | "off" | "force"}
1279 on: enable EDAC to report H/W event. May be overridden
1280 by other higher priority error reporting module.
1281 off: disable H/W event reporting through EDAC.
1282 force: enforce the use of EDAC to report H/W event.
1285 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1288 This is designed to be used in conjunction with
1289 the boot argument: earlyprintk=vga
1291 This parameter works in place of the kgdboc parameter
1292 but can only be used if the backing tty is available
1293 very early in the boot process. For early debugging
1294 via a serial port see kgdboc_earlycon instead.
1297 Format: {"off" | "on" | "skip[mbr]"}
1300 Format: { "debug", "disable_early_pci_dma",
1301 "nochunk", "noruntime", "nosoftreserve",
1302 "novamap", "no_disable_early_pci_dma" }
1303 debug: enable misc debug output.
1304 disable_early_pci_dma: disable the busmaster bit on all
1305 PCI bridges while in the EFI boot stub.
1306 nochunk: disable reading files in "chunks" in the EFI
1307 boot stub, as chunking can cause problems with some
1308 firmware implementations.
1309 noruntime : disable EFI runtime services support
1310 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1311 attribute may cause the kernel to reserve the
1312 memory range for a memory mapping driver to
1313 claim. Specify efi=nosoftreserve to disable this
1314 reservation and treat the memory by its base type
1315 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1316 novamap: do not call SetVirtualAddressMap().
1317 no_disable_early_pci_dma: Leave the busmaster bit set
1318 on all PCI bridges while in the EFI boot stub
1320 efi_no_storage_paranoia [EFI; X86]
1321 Using this parameter you can use more than 50% of
1322 your efi variable storage. Use this parameter only if
1323 you are really sure that your UEFI does sane gc and
1324 fulfills the spec otherwise your board may brick.
1326 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1327 Add arbitrary attribute to specific memory range by
1328 updating original EFI memory map.
1329 Region of memory which aa attribute is added to is
1332 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1333 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1334 attribute is added to range 0x100000000-0x180000000 and
1335 0x10a0000000-0x1120000000.
1337 If efi_fake_mem=8G@9G:0x40000 is specified, the
1338 EFI_MEMORY_SP(0x40000) attribute is added to
1339 range 0x240000000-0x43fffffff.
1341 Using this parameter you can do debugging of EFI memmap
1342 related features. For example, you can do debugging of
1343 Address Range Mirroring feature even if your box
1344 doesn't support it, or mark specific memory as
1347 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1348 that is to be dynamically loaded by Linux. If there are
1349 multiple variables with the same name but with different
1350 vendor GUIDs, all of them will be loaded. See
1351 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1354 eisa_irq_edge= [PARISC,HW]
1355 See header of drivers/parisc/eisa.c.
1358 See comment before function elanfreq_setup() in
1359 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1361 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1362 Specifies physical address of start of kernel core
1363 image elf header and optionally the size. Generally
1364 kexec loader will pass this option to capture kernel.
1365 See Documentation/admin-guide/kdump/kdump.rst for details.
1367 enable_mtrr_cleanup [X86]
1368 The kernel tries to adjust MTRR layout from continuous
1369 to discrete, to make X server driver able to add WB
1370 entry later. This parameter enables that.
1372 enable_timer_pin_1 [X86]
1373 Enable PIN 1 of APIC timer
1374 Can be useful to work around chipset bugs
1375 (in particular on some ATI chipsets).
1376 The kernel tries to set a reasonable default.
1378 enforcing [SELINUX] Set initial enforcing status.
1380 See security/selinux/Kconfig help text.
1381 0 -- permissive (log only, no denials).
1382 1 -- enforcing (deny and log).
1384 Value can be changed at runtime via
1385 /sys/fs/selinux/enforce.
1388 Disable Error Record Serialization Table (ERST)
1391 ether= [HW,NET] Ethernet cards parameters
1392 This option is obsoleted by the "netdev=" option, which
1393 has equivalent usage. See its documentation for details.
1397 Permit 'security.evm' to be updated regardless of
1398 current integrity status.
1403 fail_make_request=[KNL]
1404 General fault injection mechanism.
1405 Format: <interval>,<probability>,<space>,<times>
1406 See also Documentation/fault-injection/.
1409 Format: { initns | none }
1410 See Documentation/admin-guide/sysctl/net.rst for
1411 fb_tunnels_only_for_init_ns
1414 See Documentation/admin-guide/blockdev/floppy.rst.
1416 force_pal_cache_flush
1417 [IA-64] Avoid check_sal_cache_flush which may hang on
1418 buggy SAL_CACHE_FLUSH implementations. Using this
1419 parameter will force ia64_sal_cache_flush to call
1420 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1423 Forcefully enable Physical Address Extension (PAE).
1424 Many Pentium M systems disable PAE but may have a
1425 functionally usable PAE implementation.
1426 Warning: use of this parameter will taint the kernel
1427 and may cause unknown problems.
1430 [FTRACE] will set and start the specified tracer
1431 as early as possible in order to facilitate early
1434 ftrace_dump_on_oops[=orig_cpu]
1435 [FTRACE] will dump the trace buffers on oops.
1436 If no parameter is passed, ftrace will dump
1437 buffers of all CPUs, but if you pass orig_cpu, it will
1438 dump only the buffer of the CPU that triggered the
1441 ftrace_filter=[function-list]
1442 [FTRACE] Limit the functions traced by the function
1443 tracer at boot up. function-list is a comma-separated
1444 list of functions. This list can be changed at run
1445 time by the set_ftrace_filter file in the debugfs
1448 ftrace_notrace=[function-list]
1449 [FTRACE] Do not trace the functions specified in
1450 function-list. This list can be changed at run time
1451 by the set_ftrace_notrace file in the debugfs
1454 ftrace_graph_filter=[function-list]
1455 [FTRACE] Limit the top level callers functions traced
1456 by the function graph tracer at boot up.
1457 function-list is a comma-separated list of functions
1458 that can be changed at run time by the
1459 set_graph_function file in the debugfs tracing directory.
1461 ftrace_graph_notrace=[function-list]
1462 [FTRACE] Do not trace from the functions specified in
1463 function-list. This list is a comma-separated list of
1464 functions that can be changed at run time by the
1465 set_graph_notrace file in the debugfs tracing directory.
1467 ftrace_graph_max_depth=<uint>
1468 [FTRACE] Used with the function graph tracer. This is
1469 the max depth it will trace into a function. This value
1470 can be changed at run time by the max_graph_depth file
1471 in the tracefs tracing directory. default: 0 (no limit)
1473 fw_devlink= [KNL] Create device links between consumer and supplier
1474 devices by scanning the firmware to infer the
1475 consumer/supplier relationships. This feature is
1476 especially useful when drivers are loaded as modules as
1477 it ensures proper ordering of tasks like device probing
1478 (suppliers first, then consumers), supplier boot state
1479 clean up (only after all consumers have probed),
1480 suspend/resume & runtime PM (consumers first, then
1482 Format: { off | permissive | on | rpm }
1483 off -- Don't create device links from firmware info.
1484 permissive -- Create device links from firmware info
1485 but use it only for ordering boot state clean
1486 up (sync_state() calls).
1487 on -- Create device links from firmware info and use it
1488 to enforce probe and suspend/resume ordering.
1489 rpm -- Like "on", but also use to order runtime PM.
1491 fw_devlink.strict=<bool>
1492 [KNL] Treat all inferred dependencies as mandatory
1493 dependencies. This only applies for fw_devlink=on|rpm.
1497 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1498 support via parallel port (up to 5 devices per port)
1499 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1500 See also Documentation/input/devices/joystick-parport.rst
1504 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART
1508 gcov_persist= [GCOV] When non-zero (default), profiling data for
1509 kernel modules is saved and remains accessible via
1510 debugfs, even when the module is unloaded/reloaded.
1511 When zero, profiling data is discarded and associated
1512 debugfs files are removed at module unload time.
1514 goldfish [X86] Enable the goldfish android emulator platform.
1515 Don't use this when you are not running on the
1518 gpio-mockup.gpio_mockup_ranges
1519 [HW] Sets the ranges of gpiochip of for this device.
1520 Format: <start1>,<end1>,<start2>,<end2>...
1521 gpio-mockup.gpio_mockup_named_lines
1522 [HW] Let the driver know GPIO lines should be named.
1524 gpt [EFI] Forces disk with valid GPT signature but
1525 invalid Protective MBR to be treated as GPT. If the
1526 primary GPT is corrupted, it enables the backup/alternate
1527 GPT to be used instead.
1529 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1530 the "Enable 0" bit of the configuration register.
1533 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1534 the "Enable 0" bit of the configuration register.
1537 grcan.select= [HW] Select which physical interface to use.
1540 grcan.txsize= [HW] Sets the size of the tx buffer.
1541 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1543 grcan.rxsize= [HW] Sets the size of the rx buffer.
1544 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1547 hardlockup_all_cpu_backtrace=
1548 [KNL] Should the hard-lockup detector generate
1549 backtraces on all cpus.
1552 hashdist= [KNL,NUMA] Large hashes allocated during boot
1553 are distributed across NUMA nodes. Defaults on
1554 for 64-bit NUMA, off otherwise.
1555 Format: 0 | 1 (for off | on)
1557 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1559 hd= [EIDE] (E)IDE hard drive subsystem geometry
1560 Format: <cyl>,<head>,<sect>
1563 Disable Hardware Error Source Table (HEST) support;
1564 corresponding firmware-first mode error processing
1565 logic will be disabled.
1567 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1568 size of <nn>. This works even on boxes that have no
1569 highmem otherwise. This also works to reduce highmem
1570 size on bigger boxes.
1572 highres= [KNL] Enable/disable high resolution timer mode.
1573 Valid parameters: "on", "off"
1578 hpet= [X86-32,HPET] option to control HPET usage
1579 Format: { enable (default) | disable | force |
1581 disable: disable HPET and use PIT instead
1582 force: allow force enabled of undocumented chips (ICH4,
1584 verbose: show contents of HPET registers during setup
1586 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1587 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1589 hugetlb_cma= [HW,CMA] The size of a CMA area used for allocation
1590 of gigantic hugepages.
1593 Reserve a CMA area of given size and allocate gigantic
1594 hugepages using the CMA allocator. If enabled, the
1595 boot-time allocation of gigantic hugepages is skipped.
1597 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1598 If this follows hugepagesz (below), it specifies
1599 the number of pages of hugepagesz to be allocated.
1600 If this is the first HugeTLB parameter on the command
1601 line, it specifies the number of pages to allocate for
1602 the default huge page size. See also
1603 Documentation/admin-guide/mm/hugetlbpage.rst.
1607 [HW] The size of the HugeTLB pages. This is used in
1608 conjunction with hugepages (above) to allocate huge
1609 pages of a specific size at boot. The pair
1610 hugepagesz=X hugepages=Y can be specified once for
1611 each supported huge page size. Huge page sizes are
1612 architecture dependent. See also
1613 Documentation/admin-guide/mm/hugetlbpage.rst.
1616 hugetlb_free_vmemmap=
1617 [KNL] Reguires CONFIG_HUGETLB_PAGE_FREE_VMEMMAP
1619 Allows heavy hugetlb users to free up some more
1620 memory (6 * PAGE_SIZE for each 2MB hugetlb page).
1621 Format: { on | off (default) }
1623 on: enable the feature
1624 off: disable the feature
1626 Built with CONFIG_HUGETLB_PAGE_FREE_VMEMMAP_DEFAULT_ON=y,
1629 This is not compatible with memory_hotplug.memmap_on_memory.
1630 If both parameters are enabled, hugetlb_free_vmemmap takes
1631 precedence over memory_hotplug.memmap_on_memory.
1634 [KNL] Should the hung task detector generate panics.
1637 A value of 1 instructs the kernel to panic when a
1638 hung task is detected. The default value is controlled
1639 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1640 option. The value selected by this boot parameter can
1641 be changed later by the kernel.hung_task_panic sysctl.
1643 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1644 terminal devices. Valid values: 0..8
1645 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1646 If specified, z/VM IUCV HVC accepts connections
1647 from listed z/VM user IDs only.
1649 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1650 which allow the hypervisor to 'idle' the
1651 guest on lock contention.
1654 Do not unregister boot console at start. This is only
1655 useful for debugging when something happens in the window
1656 between unregistering the boot console and initializing
1659 i2c_bus= [HW] Override the default board specific I2C bus speed
1660 or register an additional I2C bus that is not
1661 registered from board initialization code.
1665 i8042.debug [HW] Toggle i8042 debug mode
1666 i8042.unmask_kbd_data
1667 [HW] Enable printing of interrupt data from the KBD port
1668 (disabled by default, and as a pre-condition
1669 requires that i8042.debug=1 be enabled)
1670 i8042.direct [HW] Put keyboard port into non-translated mode
1671 i8042.dumbkbd [HW] Pretend that controller can only read data from
1672 keyboard and cannot control its state
1673 (Don't attempt to blink the leds)
1674 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1675 i8042.nokbd [HW] Don't check/create keyboard port
1676 i8042.noloop [HW] Disable the AUX Loopback command while probing
1678 i8042.nomux [HW] Don't check presence of an active multiplexing
1680 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1682 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1683 i8042.reset [HW] Reset the controller during init, cleanup and
1684 suspend-to-ram transitions, only during s2r
1685 transitions, or never reset
1686 Format: { 1 | Y | y | 0 | N | n }
1687 1, Y, y: always reset controller
1688 0, N, n: don't ever reset controller
1689 Default: only on s2r transitions on x86; most other
1690 architectures force reset to be always executed
1691 i8042.unlock [HW] Unlock (ignore) the keylock
1692 i8042.kbdreset [HW] Reset device connected to KBD port
1696 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1697 indicates that the driver is running on unsupported
1699 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1700 does not match list of supported models.
1702 [HW] Report power status in /proc/i8k
1703 (disabled by default)
1704 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1707 i915.invert_brightness=
1708 [DRM] Invert the sense of the variable that is used to
1709 set the brightness of the panel backlight. Normally a
1710 brightness value of 0 indicates backlight switched off,
1711 and the maximum of the brightness value sets the backlight
1712 to maximum brightness. If this parameter is set to 0
1713 (default) and the machine requires it, or this parameter
1714 is set to 1, a brightness value of 0 sets the backlight
1715 to maximum brightness, and the maximum of the brightness
1716 value switches the backlight off.
1717 -1 -- never invert brightness
1718 0 -- machine default
1719 1 -- force brightness inversion
1722 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1724 ide-core.nodma= [HW] (E)IDE subsystem
1725 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1726 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1727 .cdrom .chs .ignore_cable are additional options
1728 See Documentation/ide/ide.rst.
1730 ide-generic.probe-mask= [HW] (E)IDE subsystem
1732 Probe mask for legacy ISA IDE ports. Depending on
1733 platform up to 6 ports are supported, enabled by
1734 setting corresponding bits in the mask to 1. The
1735 default value is 0x0, which has a special meaning.
1736 On systems that have PCI, it triggers scanning the
1737 PCI bus for the first and the second port, which
1738 are then probed. On systems without PCI the value
1739 of 0x0 enables probing the two first ports as if it
1742 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1743 Claim all unknown PCI IDE storage controllers.
1746 Format: idle=poll, idle=halt, idle=nomwait
1747 Poll forces a polling idle loop that can slightly
1748 improve the performance of waking up a idle CPU, but
1749 will use a lot of power and make the system run hot.
1751 idle=halt: Halt is forced to be used for CPU idle.
1752 In such case C2/C3 won't be used again.
1753 idle=nomwait: Disable mwait for CPU C-states
1757 Allow force disabling of Shared Virtual Memory (SVA)
1758 support for the idxd driver. By default it is set to
1761 idxd.tc_override= [HW]
1763 Allow override of default traffic class configuration
1764 for the device. By default it is set to false (0).
1766 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1767 Format: { strict | legacy | 2008 | relaxed }
1770 Choose which programs will be accepted for execution
1771 based on the IEEE 754 NaN encoding(s) supported by
1772 the FPU and the NaN encoding requested with the value
1773 of an ELF file header flag individually set by each
1774 binary. Hardware implementations are permitted to
1775 support either or both of the legacy and the 2008 NaN
1778 Available settings are as follows:
1779 strict accept binaries that request a NaN encoding
1780 supported by the FPU
1781 legacy only accept legacy-NaN binaries, if supported
1783 2008 only accept 2008-NaN binaries, if supported
1785 relaxed accept any binaries regardless of whether
1786 supported by the FPU
1788 The FPU emulator is always able to support both NaN
1789 encodings, so if no FPU hardware is present or it has
1790 been disabled with 'nofpu', then the settings of
1791 'legacy' and '2008' strap the emulator accordingly,
1792 'relaxed' straps the emulator for both legacy-NaN and
1793 2008-NaN, whereas 'strict' enables legacy-NaN only on
1794 legacy processors and both NaN encodings on MIPS32 or
1797 The setting for ABS.fmt/NEG.fmt instruction execution
1798 mode generally follows that for the NaN encoding,
1799 except where unsupported by hardware.
1801 ignore_loglevel [KNL]
1802 Ignore loglevel setting - this will print /all/
1803 kernel messages to the console. Useful for debugging.
1804 We also add it as printk module parameter, so users
1805 could change it dynamically, usually by
1806 /sys/module/printk/parameters/ignore_loglevel.
1809 Ignore RLIMIT_DATA setting for data mappings,
1810 print warning at first misuse. Can be changed via
1811 /sys/module/kernel/parameters/ignore_rlimit_data.
1813 ihash_entries= [KNL]
1814 Set number of hash buckets for inode cache.
1816 ima_appraise= [IMA] appraise integrity measurements
1817 Format: { "off" | "enforce" | "fix" | "log" }
1820 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1821 The builtin appraise policy appraises all files
1824 ima_canonical_fmt [IMA]
1825 Use the canonical format for the binary runtime
1826 measurements, instead of host native format.
1829 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1833 The list of supported hash algorithms is defined
1834 in crypto/hash_info.h.
1837 The builtin policies to load during IMA setup.
1838 Format: "tcb | appraise_tcb | secure_boot |
1839 fail_securely | critical_data"
1841 The "tcb" policy measures all programs exec'd, files
1842 mmap'd for exec, and all files opened with the read
1843 mode bit set by either the effective uid (euid=0) or
1846 The "appraise_tcb" policy appraises the integrity of
1847 all files owned by root.
1849 The "secure_boot" policy appraises the integrity
1850 of files (eg. kexec kernel image, kernel modules,
1851 firmware, policy, etc) based on file signatures.
1853 The "fail_securely" policy forces file signature
1854 verification failure also on privileged mounted
1855 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1858 The "critical_data" policy measures kernel integrity
1861 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1862 Load a policy which meets the needs of the Trusted
1863 Computing Base. This means IMA will measure all
1864 programs exec'd, files mmap'd for exec, and all files
1865 opened for read by uid=0.
1868 Select one of defined IMA measurements template formats.
1869 Formats: { "ima" | "ima-ng" | "ima-sig" }
1873 [IMA] Define a custom template format.
1874 Format: { "field1|...|fieldN" }
1876 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1877 Format: <min_file_size>
1878 Set the minimal file size for using asynchronous hash.
1879 If left unspecified, ahash usage is disabled.
1881 ahash performance varies for different data sizes on
1882 different crypto accelerators. This option can be used
1883 to achieve the best performance for a particular HW.
1885 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1887 Set hashing buffer size. Default: 4k.
1889 ahash performance varies for different chunk sizes on
1890 different crypto accelerators. This option can be used
1891 to achieve best performance for particular HW.
1895 Run specified binary instead of /sbin/init as init
1898 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1899 for working out where the kernel is dying during
1902 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1903 initcall functions. Useful for debugging built-in
1904 modules and initcalls.
1906 initramfs_async= [KNL]
1909 This parameter controls whether the initramfs
1910 image is unpacked asynchronously, concurrently
1911 with devices being probed and
1912 initialized. This should normally just work,
1913 but as a debugging aid, one can get the
1914 historical behaviour of the initramfs
1915 unpacking being completed before device_ and
1918 initrd= [BOOT] Specify the location of the initial ramdisk
1920 initrdmem= [KNL] Specify a physical address and size from which to
1921 load the initrd. If an initrd is compiled in or
1922 specified in the bootparams, it takes priority over this
1924 Format: ss[KMG],nn[KMG]
1927 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
1930 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
1932 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
1934 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
1936 init_pkru= [X86] Specify the default memory protection keys rights
1937 register contents for all processes. 0x55555554 by
1938 default (disallow access to all but pkey 0). Can
1939 override in debugfs after boot.
1941 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1944 int_pln_enable [X86] Enable power limit notification interrupt
1946 integrity_audit=[IMA]
1947 Format: { "0" | "1" }
1948 0 -- basic integrity auditing messages. (Default)
1949 1 -- additional integrity auditing messages.
1951 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1953 Enable intel iommu driver.
1955 Disable intel iommu driver.
1956 igfx_off [Default Off]
1957 By default, gfx is mapped as normal device. If a gfx
1958 device has a dedicated DMAR unit, the DMAR unit is
1959 bypassed by not enabling DMAR with this option. In
1960 this case, gfx device will use physical address for
1962 strict [Default Off]
1963 Deprecated, equivalent to iommu.strict=1.
1964 sp_off [Default Off]
1965 By default, super page will be supported if Intel IOMMU
1966 has the capability. With this option, super page will
1969 Enable the Intel IOMMU scalable mode if the hardware
1970 advertises that it has support for the scalable mode
1973 Disallow use of the Intel IOMMU scalable mode.
1974 tboot_noforce [Default Off]
1975 Do not force the Intel IOMMU enabled under tboot.
1976 By default, tboot will force Intel IOMMU on, which
1977 could harm performance of some high-throughput
1978 devices like 40GBit network cards, even if identity
1980 Note that using this option lowers the security
1981 provided by tboot because it makes the system
1982 vulnerable to DMA attacks.
1984 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1985 0 disables intel_idle and fall back on acpi_idle.
1986 1 to 9 specify maximum depth of C-state.
1990 Do not enable intel_pstate as the default
1991 scaling driver for the supported processors
1993 Use intel_pstate as a scaling driver, but configure it
1994 to work with generic cpufreq governors (instead of
1995 enabling its internal governor). This mode cannot be
1996 used along with the hardware-managed P-states (HWP)
1999 Enable intel_pstate on systems that prohibit it by default
2000 in favor of acpi-cpufreq. Forcing the intel_pstate driver
2001 instead of acpi-cpufreq may disable platform features, such
2002 as thermal controls and power capping, that rely on ACPI
2003 P-States information being indicated to OSPM and therefore
2004 should be used with caution. This option does not work with
2005 processors that aren't supported by the intel_pstate driver
2006 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2008 Do not enable hardware P state control (HWP)
2011 Only load intel_pstate on systems which support
2012 hardware P state control (HWP) if available.
2014 Enforce ACPI _PPC performance limits. If the Fixed ACPI
2015 Description Table, specifies preferred power management
2016 profile as "Enterprise Server" or "Performance Server",
2017 then this feature is turned on by default.
2019 Allow per-logical-CPU P-State performance control limits using
2020 cpufreq sysfs interface
2022 intremap= [X86-64, Intel-IOMMU]
2023 on enable Interrupt Remapping (default)
2024 off disable Interrupt Remapping
2025 nosid disable Source ID checking
2027 BIOS x2APIC opt-out request will be ignored
2028 nopost disable Interrupt Posting
2030 iomem= Disable strict checking of access to MMIO memory
2031 strict regions from userspace.
2046 nobypass [PPC/POWERNV]
2047 Disable IOMMU bypass, using IOMMU for PCI devices.
2049 iommu.forcedac= [ARM64, X86] Control IOVA allocation for PCI devices.
2050 Format: { "0" | "1" }
2051 0 - Try to allocate a 32-bit DMA address first, before
2052 falling back to the full range if needed.
2053 1 - Allocate directly from the full usable range,
2054 forcing Dual Address Cycle for PCI cards supporting
2055 greater than 32-bit addressing.
2057 iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour
2058 Format: { "0" | "1" }
2060 Request that DMA unmap operations use deferred
2061 invalidation of hardware TLBs, for increased
2062 throughput at the cost of reduced device isolation.
2063 Will fall back to strict mode if not supported by
2064 the relevant IOMMU driver.
2066 DMA unmap operations invalidate IOMMU hardware TLBs
2068 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}.
2069 Note: on x86, strict mode specified via one of the
2070 legacy driver-specific options takes precedence.
2073 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
2074 Format: { "0" | "1" }
2075 0 - Use IOMMU translation for DMA.
2076 1 - Bypass the IOMMU for DMA.
2077 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2079 io7= [HW] IO7 for Marvel-based Alpha systems
2080 See comment before marvel_specify_io7 in
2081 arch/alpha/kernel/core_marvel.c.
2083 io_delay= [X86] I/O delay method
2085 Standard port 0x80 based delay
2087 Alternate port 0xed based delay (needed on some systems)
2089 Simple two microseconds delay
2094 See Documentation/admin-guide/nfs/nfsroot.rst.
2096 ipcmni_extend [KNL] Extend the maximum number of unique System V
2097 IPC identifiers from 32,768 to 16,777,216.
2099 irqaffinity= [SMP] Set the default irq affinity mask
2100 The argument is a cpu list, as described above.
2102 irqchip.gicv2_force_probe=
2105 Force the kernel to look for the second 4kB page
2106 of a GICv2 controller even if the memory range
2107 exposed by the device tree is too small.
2109 irqchip.gicv3_nolpi=
2111 Force the kernel to ignore the availability of
2112 LPIs (and by consequence ITSs). Intended for system
2113 that use the kernel as a bootloader, and thus want
2114 to let secondary kernels in charge of setting up
2117 irqchip.gicv3_pseudo_nmi= [ARM64]
2118 Enables support for pseudo-NMIs in the kernel. This
2119 requires the kernel to be built with
2120 CONFIG_ARM64_PSEUDO_NMI.
2123 When an interrupt is not handled search all handlers
2124 for it. Intended to get systems with badly broken
2128 When an interrupt is not handled search all handlers
2129 for it. Also check all handlers each timer
2130 interrupt. Intended to get systems with badly broken
2134 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2136 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2137 [Deprecated - use cpusets instead]
2138 Format: [flag-list,]<cpu-list>
2140 Specify one or more CPUs to isolate from disturbances
2141 specified in the flag list (default: domain):
2144 Disable the tick when a single task runs.
2146 A residual 1Hz tick is offloaded to workqueues, which you
2147 need to affine to housekeeping through the global
2148 workqueue's affinity configured via the
2149 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2150 by using the 'domain' flag described below.
2152 NOTE: by default the global workqueue runs on all CPUs,
2153 so to protect individual CPUs the 'cpumask' file has to
2154 be configured manually after bootup.
2157 Isolate from the general SMP balancing and scheduling
2158 algorithms. Note that performing domain isolation this way
2159 is irreversible: it's not possible to bring back a CPU to
2160 the domains once isolated through isolcpus. It's strongly
2161 advised to use cpusets instead to disable scheduler load
2162 balancing through the "cpuset.sched_load_balance" file.
2163 It offers a much more flexible interface where CPUs can
2164 move in and out of an isolated set anytime.
2166 You can move a process onto or off an "isolated" CPU via
2167 the CPU affinity syscalls or cpuset.
2168 <cpu number> begins at 0 and the maximum value is
2169 "number of CPUs in system - 1".
2173 Isolate from being targeted by managed interrupts
2174 which have an interrupt mask containing isolated
2175 CPUs. The affinity of managed interrupts is
2176 handled by the kernel and cannot be changed via
2177 the /proc/irq/* interfaces.
2179 This isolation is best effort and only effective
2180 if the automatically assigned interrupt mask of a
2181 device queue contains isolated and housekeeping
2182 CPUs. If housekeeping CPUs are online then such
2183 interrupts are directed to the housekeeping CPU
2184 so that IO submitted on the housekeeping CPU
2185 cannot disturb the isolated CPU.
2187 If a queue's affinity mask contains only isolated
2188 CPUs then this parameter has no effect on the
2189 interrupt routing decision, though interrupts are
2190 only delivered when tasks running on those
2191 isolated CPUs submit IO. IO submitted on
2192 housekeeping CPUs has no influence on those
2195 The format of <cpu-list> is described above.
2199 ivrs_ioapic [HW,X86-64]
2200 Provide an override to the IOAPIC-ID<->DEVICE-ID
2201 mapping provided in the IVRS ACPI table. For
2202 example, to map IOAPIC-ID decimal 10 to
2203 PCI device 00:14.0 write the parameter as:
2204 ivrs_ioapic[10]=00:14.0
2206 ivrs_hpet [HW,X86-64]
2207 Provide an override to the HPET-ID<->DEVICE-ID
2208 mapping provided in the IVRS ACPI table. For
2209 example, to map HPET-ID decimal 0 to
2210 PCI device 00:14.0 write the parameter as:
2211 ivrs_hpet[0]=00:14.0
2213 ivrs_acpihid [HW,X86-64]
2214 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2215 mapping provided in the IVRS ACPI table. For
2216 example, to map UART-HID:UID AMD0020:0 to
2217 PCI device 00:14.5 write the parameter as:
2218 ivrs_acpihid[00:14.5]=AMD0020:0
2220 js= [HW,JOY] Analog joystick
2221 See Documentation/input/joydev/joystick.rst.
2224 When CONFIG_RANDOMIZE_BASE is set, this disables
2225 kernel and module base offset ASLR (Address Space
2226 Layout Randomization).
2229 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2230 report on every invalid memory access. Without this
2231 parameter KASAN will print report only for the first
2236 kernelcore= [KNL,X86,IA-64,PPC]
2237 Format: nn[KMGTPE] | nn% | "mirror"
2238 This parameter specifies the amount of memory usable by
2239 the kernel for non-movable allocations. The requested
2240 amount is spread evenly throughout all nodes in the
2241 system as ZONE_NORMAL. The remaining memory is used for
2242 movable memory in its own zone, ZONE_MOVABLE. In the
2243 event, a node is too small to have both ZONE_NORMAL and
2244 ZONE_MOVABLE, kernelcore memory will take priority and
2245 other nodes will have a larger ZONE_MOVABLE.
2247 ZONE_MOVABLE is used for the allocation of pages that
2248 may be reclaimed or moved by the page migration
2249 subsystem. Note that allocations like PTEs-from-HighMem
2250 still use the HighMem zone if it exists, and the Normal
2251 zone if it does not.
2253 It is possible to specify the exact amount of memory in
2254 the form of "nn[KMGTPE]", a percentage of total system
2255 memory in the form of "nn%", or "mirror". If "mirror"
2256 option is specified, mirrored (reliable) memory is used
2257 for non-movable allocations and remaining memory is used
2258 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2259 are exclusive, so you cannot specify multiple forms.
2261 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2262 Format: <Controller#>[,poll interval]
2263 The controller # is the number of the ehci usb debug
2264 port as it is probed via PCI. The poll interval is
2265 optional and is the number seconds in between
2266 each poll cycle to the debug port in case you need
2267 the functionality for interrupting the kernel with
2268 gdb or control-c on the dbgp connection. When
2269 not using this parameter you use sysrq-g to break into
2270 the kernel debugger.
2272 kgdboc= [KGDB,HW] kgdb over consoles.
2273 Requires a tty driver that supports console polling,
2274 or a supported polling keyboard driver (non-usb).
2275 Serial only format: <serial_device>[,baud]
2276 keyboard only format: kbd
2277 keyboard and serial format: kbd,<serial_device>[,baud]
2278 Optional Kernel mode setting:
2279 kms, kbd format: kms,kbd
2280 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2282 kgdboc_earlycon= [KGDB,HW]
2283 If the boot console provides the ability to read
2284 characters and can work in polling mode, you can use
2285 this parameter to tell kgdb to use it as a backend
2286 until the normal console is registered. Intended to
2287 be used together with the kgdboc parameter which
2288 specifies the normal console to transition to.
2290 The name of the early console should be specified
2291 as the value of this parameter. Note that the name of
2292 the early console might be different than the tty
2293 name passed to kgdboc. It's OK to leave the value
2294 blank and the first boot console that implements
2295 read() will be picked.
2297 kgdbwait [KGDB] Stop kernel execution and enter the
2298 kernel debugger at the earliest opportunity.
2300 kmac= [MIPS] Korina ethernet MAC address.
2301 Configure the RouterBoard 532 series on-chip
2302 Ethernet adapter MAC address.
2304 kmemleak= [KNL] Boot-time kmemleak enable/disable
2305 Valid arguments: on, off
2307 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2310 kprobe_event=[probe-list]
2311 [FTRACE] Add kprobe events and enable at boot time.
2312 The probe-list is a semicolon delimited list of probe
2313 definitions. Each definition is same as kprobe_events
2314 interface, but the parameters are comma delimited.
2315 For example, to add a kprobe event on vfs_read with
2316 arg1 and arg2, add to the command line;
2318 kprobe_event=p,vfs_read,$arg1,$arg2
2320 See also Documentation/trace/kprobetrace.rst "Kernel
2321 Boot Parameter" section.
2323 kpti= [ARM64] Control page table isolation of user
2324 and kernel address spaces.
2325 Default: enabled on cores which need mitigation.
2329 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2330 Default is 0 (don't ignore, but inject #GP)
2332 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2333 Default is false (don't support).
2335 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
2340 [KVM] Controls the software workaround for the
2341 X86_BUG_ITLB_MULTIHIT bug.
2342 force : Always deploy workaround.
2343 off : Never deploy workaround.
2344 auto : Deploy workaround based on the presence of
2345 X86_BUG_ITLB_MULTIHIT.
2349 If the software workaround is enabled for the host,
2350 guests do need not to enable it for nested guests.
2352 kvm.nx_huge_pages_recovery_ratio=
2353 [KVM] Controls how many 4KiB pages are periodically zapped
2354 back to huge pages. 0 disables the recovery, otherwise if
2355 the value is N KVM will zap 1/Nth of the 4KiB pages every
2356 period (see below). The default is 60.
2358 kvm.nx_huge_pages_recovery_period_ms=
2359 [KVM] Controls the time period at which KVM zaps 4KiB pages
2360 back to huge pages. If the value is a non-zero N, KVM will
2361 zap a portion (see ratio above) of the pages every N msecs.
2362 If the value is 0 (the default), KVM will pick a period based
2363 on the ratio, such that a page is zapped after 1 hour on average.
2365 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2366 Default is 1 (enabled)
2368 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2370 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2373 [KVM,ARM] Select one of KVM/arm64's modes of operation.
2375 none: Forcefully disable KVM.
2377 nvhe: Standard nVHE-based mode, without support for
2380 protected: nVHE-based mode with support for guests whose
2381 state is kept private from the host.
2382 Not valid if the kernel is running in EL2.
2384 Defaults to VHE/nVHE based on hardware support. Setting
2385 mode to "protected" will disable kexec and hibernation
2388 kvm-arm.vgic_v3_group0_trap=
2389 [KVM,ARM] Trap guest accesses to GICv3 group-0
2392 kvm-arm.vgic_v3_group1_trap=
2393 [KVM,ARM] Trap guest accesses to GICv3 group-1
2396 kvm-arm.vgic_v3_common_trap=
2397 [KVM,ARM] Trap guest accesses to GICv3 common
2400 kvm-arm.vgic_v4_enable=
2401 [KVM,ARM] Allow use of GICv4 for direct injection of
2404 kvm_cma_resv_ratio=n [PPC]
2405 Reserves given percentage from system memory area for
2406 contiguous memory allocation for KVM hash pagetable
2408 By default it reserves 5% of total system memory.
2412 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2413 (virtualized MMU) support on capable Intel chips.
2414 Default is 1 (enabled)
2416 kvm-intel.emulate_invalid_guest_state=
2417 [KVM,Intel] Enable emulation of invalid guest states
2418 Default is 0 (disabled)
2420 kvm-intel.flexpriority=
2421 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2422 Default is 1 (enabled)
2425 [KVM,Intel] Enable VMX nesting (nVMX).
2426 Default is 0 (disabled)
2428 kvm-intel.unrestricted_guest=
2429 [KVM,Intel] Disable unrestricted guest feature
2430 (virtualized real and unpaged mode) on capable
2431 Intel chips. Default is 1 (enabled)
2433 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2436 Valid arguments: never, cond, always
2438 always: L1D cache flush on every VMENTER.
2439 cond: Flush L1D on VMENTER only when the code between
2440 VMEXIT and VMENTER can leak host memory.
2441 never: Disables the mitigation
2443 Default is cond (do L1 cache flush in specific instances)
2445 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2446 feature (tagged TLBs) on capable Intel chips.
2447 Default is 1 (enabled)
2449 l1d_flush= [X86,INTEL]
2450 Control mitigation for L1D based snooping vulnerability.
2452 Certain CPUs are vulnerable to an exploit against CPU
2453 internal buffers which can forward information to a
2454 disclosure gadget under certain conditions.
2456 In vulnerable processors, the speculatively
2457 forwarded data can be used in a cache side channel
2458 attack, to access data to which the attacker does
2459 not have direct access.
2461 This parameter controls the mitigation. The
2464 on - enable the interface for the mitigation
2466 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2469 The kernel PTE inversion protection is unconditionally
2470 enabled and cannot be disabled.
2473 Provides all available mitigations for the
2474 L1TF vulnerability. Disables SMT and
2475 enables all mitigations in the
2476 hypervisors, i.e. unconditional L1D flush.
2478 SMT control and L1D flush control via the
2479 sysfs interface is still possible after
2480 boot. Hypervisors will issue a warning
2481 when the first VM is started in a
2482 potentially insecure configuration,
2483 i.e. SMT enabled or L1D flush disabled.
2486 Same as 'full', but disables SMT and L1D
2487 flush runtime control. Implies the
2488 'nosmt=force' command line option.
2489 (i.e. sysfs control of SMT is disabled.)
2492 Leaves SMT enabled and enables the default
2493 hypervisor mitigation, i.e. conditional
2496 SMT control and L1D flush control via the
2497 sysfs interface is still possible after
2498 boot. Hypervisors will issue a warning
2499 when the first VM is started in a
2500 potentially insecure configuration,
2501 i.e. SMT enabled or L1D flush disabled.
2505 Disables SMT and enables the default
2506 hypervisor mitigation.
2508 SMT control and L1D flush control via the
2509 sysfs interface is still possible after
2510 boot. Hypervisors will issue a warning
2511 when the first VM is started in a
2512 potentially insecure configuration,
2513 i.e. SMT enabled or L1D flush disabled.
2516 Same as 'flush', but hypervisors will not
2517 warn when a VM is started in a potentially
2518 insecure configuration.
2521 Disables hypervisor mitigations and doesn't
2523 It also drops the swap size and available
2524 RAM limit restriction on both hypervisor and
2529 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2535 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2538 lapic= [X86,APIC] Do not use TSC deadline
2539 value for LAPIC timer one-shot implementation. Default
2540 back to the programmable timer unit in the LAPIC.
2541 Format: notscdeadline
2543 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2546 libata.dma= [LIBATA] DMA control
2547 libata.dma=0 Disable all PATA and SATA DMA
2548 libata.dma=1 PATA and SATA Disk DMA only
2549 libata.dma=2 ATAPI (CDROM) DMA only
2550 libata.dma=4 Compact Flash DMA only
2551 Combinations also work, so libata.dma=3 enables DMA
2552 for disks and CDROMs, but not CFs.
2554 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2555 libata.ignore_hpa=0 keep BIOS limits (default)
2556 libata.ignore_hpa=1 ignore limits, using full disk
2558 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2562 libata.force= [LIBATA] Force configurations. The format is comma-
2563 separated list of "[ID:]VAL" where ID is
2564 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2565 matching port, link or device. Basically, it matches
2566 the ATA ID string printed on console by libata. If
2567 the whole ID part is omitted, the last PORT and DEVICE
2568 values are used. If ID hasn't been specified yet, the
2569 configuration applies to all ports, links and devices.
2571 If only DEVICE is omitted, the parameter applies to
2572 the port and all links and devices behind it. DEVICE
2573 number of 0 either selects the first device or the
2574 first fan-out link behind PMP device. It does not
2575 select the host link. DEVICE number of 15 selects the
2576 host link and device attached to it.
2578 The VAL specifies the configuration to force. As long
2579 as there's no ambiguity shortcut notation is allowed.
2580 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2581 The following configurations can be forced.
2583 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2584 Any ID with matching PORT is used.
2586 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2588 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2589 udma[/][16,25,33,44,66,100,133] notation is also
2592 * [no]ncq: Turn on or off NCQ.
2594 * [no]ncqtrim: Turn off queued DSM TRIM.
2596 * nohrst, nosrst, norst: suppress hard, soft
2599 * rstonce: only attempt one reset during
2600 hot-unplug link recovery
2602 * dump_id: dump IDENTIFY data.
2604 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2606 * disable: Disable this device.
2608 If there are multiple matching configurations changing
2609 the same attribute, the last one is used.
2611 memblock=debug [KNL] Enable memblock debug messages.
2613 load_ramdisk= [RAM] [Deprecated]
2615 lockd.nlm_grace_period=P [NFS] Assign grace period.
2618 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2621 lockd.nlm_timeout=T [NFS] Assign timeout value.
2624 lockd.nlm_udpport=M [NFS] Assign UDP port.
2627 lockdown= [SECURITY]
2628 { integrity | confidentiality }
2629 Enable the kernel lockdown feature. If set to
2630 integrity, kernel features that allow userland to
2631 modify the running kernel are disabled. If set to
2632 confidentiality, kernel features that allow userland
2633 to extract confidential information from the kernel
2636 locktorture.nreaders_stress= [KNL]
2637 Set the number of locking read-acquisition kthreads.
2638 Defaults to being automatically set based on the
2639 number of online CPUs.
2641 locktorture.nwriters_stress= [KNL]
2642 Set the number of locking write-acquisition kthreads.
2644 locktorture.onoff_holdoff= [KNL]
2645 Set time (s) after boot for CPU-hotplug testing.
2647 locktorture.onoff_interval= [KNL]
2648 Set time (s) between CPU-hotplug operations, or
2649 zero to disable CPU-hotplug testing.
2651 locktorture.shuffle_interval= [KNL]
2652 Set task-shuffle interval (jiffies). Shuffling
2653 tasks allows some CPUs to go into dyntick-idle
2654 mode during the locktorture test.
2656 locktorture.shutdown_secs= [KNL]
2657 Set time (s) after boot system shutdown. This
2658 is useful for hands-off automated testing.
2660 locktorture.stat_interval= [KNL]
2661 Time (s) between statistics printk()s.
2663 locktorture.stutter= [KNL]
2664 Time (s) to stutter testing, for example,
2665 specifying five seconds causes the test to run for
2666 five seconds, wait for five seconds, and so on.
2667 This tests the locking primitive's ability to
2668 transition abruptly to and from idle.
2670 locktorture.torture_type= [KNL]
2671 Specify the locking implementation to test.
2673 locktorture.verbose= [KNL]
2674 Enable additional printk() statements.
2676 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2679 loglevel= All Kernel Messages with a loglevel smaller than the
2680 console loglevel will be printed to the console. It can
2681 also be changed with klogd or other programs. The
2682 loglevels are defined as follows:
2684 0 (KERN_EMERG) system is unusable
2685 1 (KERN_ALERT) action must be taken immediately
2686 2 (KERN_CRIT) critical conditions
2687 3 (KERN_ERR) error conditions
2688 4 (KERN_WARNING) warning conditions
2689 5 (KERN_NOTICE) normal but significant condition
2690 6 (KERN_INFO) informational
2691 7 (KERN_DEBUG) debug-level messages
2693 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2694 in bytes. n must be a power of two and greater
2695 than the minimal size. The minimal size is defined
2696 by LOG_BUF_SHIFT kernel config parameter. There is
2697 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2698 that allows to increase the default size depending on
2699 the number of CPUs. See init/Kconfig for more details.
2701 logo.nologo [FB] Disables display of the built-in Linux logo.
2702 This may be used to provide more screen space for
2703 kernel log messages and is useful when debugging
2704 kernel boot problems.
2706 lp=0 [LP] Specify parallel ports to use, e.g,
2707 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2708 lp=reset first parallel port). 'lp=0' disables the
2709 lp=auto printer driver. 'lp=reset' (which can be
2710 specified in addition to the ports) causes
2711 attached printers to be reset. Using
2712 lp=port1,port2,... specifies the parallel ports
2713 to associate lp devices with, starting with
2714 lp0. A port specification may be 'none' to skip
2715 that lp device, or a parport name such as
2716 'parport0'. Specifying 'lp=auto' instead of a
2717 port specification list means that device IDs
2718 from each port should be examined, to see if
2719 an IEEE 1284-compliant printer is attached; if
2720 so, the driver will manage that printer.
2721 See also header of drivers/char/lp.c.
2724 Sets loops_per_jiffy to given constant, thus avoiding
2725 time-consuming boot-time autodetection (up to 250 ms per
2726 CPU). 0 enables autodetection (default). To determine
2727 the correct value for your kernel, boot with normal
2728 autodetection and see what value is printed. Note that
2729 on SMP systems the preset will be applied to all CPUs,
2730 which is likely to cause problems if your CPUs need
2731 significantly divergent settings. An incorrect value
2732 will cause delays in the kernel to be wrong, leading to
2733 unpredictable I/O errors and other breakage. Although
2734 unlikely, in the extreme case this might damage your
2738 Format: <io>,<irq>,<dma>
2740 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2743 [SECURITY] Choose order of LSM initialization. This
2744 overrides CONFIG_LSM, and the "security=" parameter.
2746 machvec= [IA-64] Force the use of a particular machine-vector
2747 (machvec) in a generic kernel.
2748 Example: machvec=hpzx1
2750 machtype= [Loongson] Share the same kernel image file between
2751 different yeeloong laptops.
2752 Example: machtype=lemote-yeeloong-2f-7inch
2754 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2755 than or equal to this physical address is ignored.
2757 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2758 will bring up during bootup. maxcpus=n : n >= 0 limits
2759 the kernel to bring up 'n' processors. Surely after
2760 bootup you can bring up the other plugged cpu by executing
2761 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2762 only takes effect during system bootup.
2763 While n=0 is a special case, it is equivalent to "nosmp",
2764 which also disables the IO APIC.
2766 max_loop= [LOOP] The number of loop block devices that get
2767 (loop.max_loop) unconditionally pre-created at init time. The default
2768 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2769 of statically allocating a predefined number, loop
2770 devices can be requested on-demand with the
2771 /dev/loop-control interface.
2773 mce [X86-32] Machine Check Exception
2775 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2777 md= [HW] RAID subsystems devices and level
2778 See Documentation/admin-guide/md.rst.
2781 Format: <first>,<last>
2782 Specifies range of consoles to be captured by the MDA.
2785 Control mitigation for the Micro-architectural Data
2786 Sampling (MDS) vulnerability.
2788 Certain CPUs are vulnerable to an exploit against CPU
2789 internal buffers which can forward information to a
2790 disclosure gadget under certain conditions.
2792 In vulnerable processors, the speculatively
2793 forwarded data can be used in a cache side channel
2794 attack, to access data to which the attacker does
2795 not have direct access.
2797 This parameter controls the MDS mitigation. The
2800 full - Enable MDS mitigation on vulnerable CPUs
2801 full,nosmt - Enable MDS mitigation and disable
2802 SMT on vulnerable CPUs
2803 off - Unconditionally disable MDS mitigation
2805 On TAA-affected machines, mds=off can be prevented by
2806 an active TAA mitigation as both vulnerabilities are
2807 mitigated with the same mechanism so in order to disable
2808 this mitigation, you need to specify tsx_async_abort=off
2811 Not specifying this option is equivalent to
2814 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2816 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2817 Amount of memory to be used in cases as follows:
2820 2 when the kernel is not able to see the whole system memory;
2821 3 memory that lies after 'mem=' boundary is excluded from
2822 the hypervisor, then assigned to KVM guests.
2824 [X86] Work as limiting max address. Use together
2825 with memmap= to avoid physical address space collisions.
2826 Without memmap= PCI devices could be placed at addresses
2827 belonging to unused RAM.
2829 Note that this only takes effects during boot time since
2830 in above case 3, memory may need be hot added after boot
2831 if system memory of hypervisor is not sufficient.
2833 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2837 [KNL,SH] Allow user to override the default size for
2838 per-device physically contiguous DMA buffers.
2840 memhp_default_state=online/offline
2841 [KNL] Set the initial state for the memory hotplug
2842 onlining policy. If not specified, the default value is
2843 set according to the
2844 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2846 See Documentation/admin-guide/mm/memory-hotplug.rst.
2848 memmap=exactmap [KNL,X86] Enable setting of an exact
2849 E820 memory map, as specified by the user.
2850 Such memmap=exactmap lines can be constructed based on
2851 BIOS output or other requirements. See the memmap=nn@ss
2854 memmap=nn[KMG]@ss[KMG]
2855 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory.
2856 Region of memory to be used is from ss to ss+nn.
2857 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2858 which limits max address to nn[KMG].
2859 Multiple different regions can be specified,
2862 memmap=100M@2G,100M#3G,1G!1024G
2864 memmap=nn[KMG]#ss[KMG]
2865 [KNL,ACPI] Mark specific memory as ACPI data.
2866 Region of memory to be marked is from ss to ss+nn.
2868 memmap=nn[KMG]$ss[KMG]
2869 [KNL,ACPI] Mark specific memory as reserved.
2870 Region of memory to be reserved is from ss to ss+nn.
2871 Example: Exclude memory from 0x18690000-0x1869ffff
2872 memmap=64K$0x18690000
2874 memmap=0x10000$0x18690000
2875 Some bootloaders may need an escape character before '$',
2876 like Grub2, otherwise '$' and the following number
2879 memmap=nn[KMG]!ss[KMG]
2880 [KNL,X86] Mark specific memory as protected.
2881 Region of memory to be used, from ss to ss+nn.
2882 The memory region may be marked as e820 type 12 (0xc)
2883 and is NVDIMM or ADR memory.
2885 memmap=<size>%<offset>-<oldtype>+<newtype>
2886 [KNL,ACPI] Convert memory within the specified region
2887 from <oldtype> to <newtype>. If "-<oldtype>" is left
2888 out, the whole region will be marked as <newtype>,
2889 even if previously unavailable. If "+<newtype>" is left
2890 out, matching memory will be removed. Types are
2891 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2892 3 = ACPI, 12 = PRAM.
2894 memory_corruption_check=0/1 [X86]
2895 Some BIOSes seem to corrupt the first 64k of
2896 memory when doing things like suspend/resume.
2897 Setting this option will scan the memory
2898 looking for corruption. Enabling this will
2899 both detect corruption and prevent the kernel
2900 from using the memory being corrupted.
2901 However, its intended as a diagnostic tool; if
2902 repeatable BIOS-originated corruption always
2903 affects the same memory, you can use memmap=
2904 to prevent the kernel from using that memory.
2906 memory_corruption_check_size=size [X86]
2907 By default it checks for corruption in the low
2908 64k, making this memory unavailable for normal
2909 use. Use this parameter to scan for
2910 corruption in more or less memory.
2912 memory_corruption_check_period=seconds [X86]
2913 By default it checks for corruption every 60
2914 seconds. Use this parameter to check at some
2915 other rate. 0 disables periodic checking.
2917 memory_hotplug.memmap_on_memory
2918 [KNL,X86,ARM] Boolean flag to enable this feature.
2919 Format: {on | off (default)}
2920 When enabled, runtime hotplugged memory will
2921 allocate its internal metadata (struct pages)
2922 from the hotadded memory which will allow to
2923 hotadd a lot of memory without requiring
2924 additional memory to do so.
2925 This feature is disabled by default because it
2926 has some implication on large (e.g. GB)
2927 allocations in some configurations (e.g. small
2929 The state of the flag can be read in
2930 /sys/module/memory_hotplug/parameters/memmap_on_memory.
2931 Note that even when enabled, there are a few cases where
2932 the feature is not effective.
2934 This is not compatible with hugetlb_free_vmemmap. If
2935 both parameters are enabled, hugetlb_free_vmemmap takes
2936 precedence over memory_hotplug.memmap_on_memory.
2938 memtest= [KNL,X86,ARM,PPC,RISCV] Enable memtest
2940 default : 0 <disable>
2941 Specifies the number of memtest passes to be
2942 performed. Each pass selects another test
2943 pattern from a given set of patterns. Memtest
2944 fills the memory with this pattern, validates
2945 memory contents and reserves bad memory
2946 regions that are detected.
2948 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2949 Valid arguments: on, off
2950 Default (depends on kernel configuration option):
2951 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2952 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2953 mem_encrypt=on: Activate SME
2954 mem_encrypt=off: Do not activate SME
2956 Refer to Documentation/virt/kvm/amd-memory-encryption.rst
2957 for details on when memory encryption can be activated.
2959 mem_sleep_default= [SUSPEND] Default system suspend mode:
2960 s2idle - Suspend-To-Idle
2961 shallow - Power-On Suspend or equivalent (if supported)
2962 deep - Suspend-To-RAM or equivalent (if supported)
2963 See Documentation/admin-guide/pm/sleep-states.rst.
2965 meye.*= [HW] Set MotionEye Camera parameters
2966 See Documentation/admin-guide/media/meye.rst.
2968 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2969 Multi-Function General Purpose Timers on AMD Geode
2972 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2973 the BIOS has incorrectly applied a workaround. TinyBIOS
2974 version 0.98 is known to be affected, 0.99 fixes the
2975 problem by letting the user disable the workaround.
2979 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2980 physical address is ignored.
2982 mini2440= [ARM,HW,KNL]
2983 Format:[0..2][b][c][t]
2985 MINI2440 configuration specification:
2986 0 - The attached screen is the 3.5" TFT
2987 1 - The attached screen is the 7" TFT
2988 2 - The VGA Shield is attached (1024x768)
2989 Leaving out the screen size parameter will not load
2990 the TFT driver, and the framebuffer will be left
2992 b - Enable backlight. The TFT backlight pin will be
2993 linked to the kernel VESA blanking code and a GPIO
2994 LED. This parameter is not necessary when using the
2996 c - Enable the s3c camera interface.
2997 t - Reserved for enabling touchscreen support. The
2998 touchscreen support is not enabled in the mainstream
2999 kernel as of 2.6.30, a preliminary port can be found
3000 in the "bleeding edge" mini2440 support kernel at
3001 https://repo.or.cz/w/linux-2.6/mini2440.git
3004 [X86,PPC,S390,ARM64] Control optional mitigations for
3005 CPU vulnerabilities. This is a set of curated,
3006 arch-independent options, each of which is an
3007 aggregation of existing arch-specific options.
3010 Disable all optional CPU mitigations. This
3011 improves system performance, but it may also
3012 expose users to several CPU vulnerabilities.
3013 Equivalent to: nopti [X86,PPC]
3015 nospectre_v1 [X86,PPC]
3017 nospectre_v2 [X86,PPC,S390,ARM64]
3018 spectre_v2_user=off [X86]
3019 spec_store_bypass_disable=off [X86,PPC]
3020 ssbd=force-off [ARM64]
3023 tsx_async_abort=off [X86]
3024 kvm.nx_huge_pages=off [X86]
3025 no_entry_flush [PPC]
3026 no_uaccess_flush [PPC]
3029 This does not have any effect on
3030 kvm.nx_huge_pages when
3031 kvm.nx_huge_pages=force.
3034 Mitigate all CPU vulnerabilities, but leave SMT
3035 enabled, even if it's vulnerable. This is for
3036 users who don't want to be surprised by SMT
3037 getting disabled across kernel upgrades, or who
3038 have other ways of avoiding SMT-based attacks.
3039 Equivalent to: (default behavior)
3042 Mitigate all CPU vulnerabilities, disabling SMT
3043 if needed. This is for users who always want to
3044 be fully mitigated, even if it means losing SMT.
3045 Equivalent to: l1tf=flush,nosmt [X86]
3046 mds=full,nosmt [X86]
3047 tsx_async_abort=full,nosmt [X86]
3050 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
3051 parameter allows control of the logging verbosity for
3052 the additional memory initialisation checks. A value
3053 of 0 disables mminit logging and a level of 4 will
3054 log everything. Information is printed at KERN_DEBUG
3055 so loglevel=8 may also need to be specified.
3058 [KNL] When CONFIG_MODULE_SIG is set, this means that
3059 modules without (valid) signatures will fail to load.
3060 Note that if CONFIG_MODULE_SIG_FORCE is set, that
3061 is always true, so this option does nothing.
3063 module_blacklist= [KNL] Do not load a comma-separated list of
3064 modules. Useful for debugging problem modules.
3067 [MOUSE] Maximum time between finger touching and
3068 leaving touchpad surface for touch to be considered
3069 a tap and be reported as a left button click (for
3070 touchpads working in absolute mode only).
3072 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
3073 reporting absolute coordinates, such as tablets
3074 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
3075 reporting absolute coordinates, such as tablets
3077 movablecore= [KNL,X86,IA-64,PPC]
3078 Format: nn[KMGTPE] | nn%
3079 This parameter is the complement to kernelcore=, it
3080 specifies the amount of memory used for migratable
3081 allocations. If both kernelcore and movablecore is
3082 specified, then kernelcore will be at *least* the
3083 specified value but may be more. If movablecore on its
3084 own is specified, the administrator must be careful
3085 that the amount of memory usable for all allocations
3088 movable_node [KNL] Boot-time switch to make hotplugable memory
3089 NUMA nodes to be movable. This means that the memory
3090 of such nodes will be usable only for movable
3091 allocations which rules out almost all kernel
3092 allocations. Use with caution!
3094 MTD_Partition= [MTD]
3095 Format: <name>,<region-number>,<size>,<offset>
3097 MTD_Region= [MTD] Format:
3098 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
3101 See drivers/mtd/parsers/cmdlinepart.c
3103 multitce=off [PPC] This parameter disables the use of the pSeries
3104 firmware feature for updating multiple TCE entries
3107 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
3109 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
3111 boundary - index of last SLC block on Flex-OneNAND.
3112 The remaining blocks are configured as MLC blocks.
3113 lock - Configure if Flex-OneNAND boundary should be locked.
3114 Once locked, the boundary cannot be changed.
3115 1 indicates lock status, 0 indicates unlock status.
3118 ARM/S3C2412 JIVE boot control
3120 See arch/arm/mach-s3c/mach-jive.c
3122 mtouchusb.raw_coordinates=
3123 [HW] Make the MicroTouch USB driver use raw coordinates
3124 ('y', default) or cooked coordinates ('n')
3126 mtrr_chunk_size=nn[KMG] [X86]
3127 used for mtrr cleanup. It is largest continuous chunk
3128 that could hold holes aka. UC entries.
3130 mtrr_gran_size=nn[KMG] [X86]
3131 Used for mtrr cleanup. It is granularity of mtrr block.
3133 Large value could prevent small alignment from
3136 mtrr_spare_reg_nr=n [X86]
3138 Range: 0,7 : spare reg number
3140 Used for mtrr cleanup. It is spare mtrr entries number.
3141 Set to 2 or more if your graphical card needs more.
3143 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
3145 netdev= [NET] Network devices parameters
3146 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
3147 Note that mem_start is often overloaded to mean
3148 something different and driver-specific.
3149 This usage is only documented in each driver source
3153 [NETFILTER] Enable connection tracking flow accounting
3154 0 to disable accounting
3155 1 to enable accounting
3158 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3159 See Documentation/admin-guide/nfs/nfsroot.rst.
3161 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3162 See Documentation/admin-guide/nfs/nfsroot.rst.
3164 nfsrootdebug [NFS] enable nfsroot debugging messages.
3165 See Documentation/admin-guide/nfs/nfsroot.rst.
3167 nfs.callback_nr_threads=
3168 [NFSv4] set the total number of threads that the
3169 NFS client will assign to service NFSv4 callback
3172 nfs.callback_tcpport=
3173 [NFS] set the TCP port on which the NFSv4 callback
3174 channel should listen.
3177 [NFS] sets the pathname to the program which is used
3178 to update the NFS client cache entries.
3180 nfs.cache_getent_timeout=
3181 [NFS] sets the timeout after which an attempt to
3182 update a cache entry is deemed to have failed.
3184 nfs.idmap_cache_timeout=
3185 [NFS] set the maximum lifetime for idmapper cache
3189 [NFS] enable 64-bit inode numbers.
3190 If zero, the NFS client will fake up a 32-bit inode
3191 number for the readdir() and stat() syscalls instead
3192 of returning the full 64-bit number.
3193 The default is to return 64-bit inode numbers.
3195 nfs.max_session_cb_slots=
3196 [NFSv4.1] Sets the maximum number of session
3197 slots the client will assign to the callback
3198 channel. This determines the maximum number of
3199 callbacks the client will process in parallel for
3200 a particular server.
3202 nfs.max_session_slots=
3203 [NFSv4.1] Sets the maximum number of session slots
3204 the client will attempt to negotiate with the server.
3205 This limits the number of simultaneous RPC requests
3206 that the client can send to the NFSv4.1 server.
3207 Note that there is little point in setting this
3208 value higher than the max_tcp_slot_table_limit.
3210 nfs.nfs4_disable_idmapping=
3211 [NFSv4] When set to the default of '1', this option
3212 ensures that both the RPC level authentication
3213 scheme and the NFS level operations agree to use
3214 numeric uids/gids if the mount is using the
3215 'sec=sys' security flavour. In effect it is
3216 disabling idmapping, which can make migration from
3217 legacy NFSv2/v3 systems to NFSv4 easier.
3218 Servers that do not support this mode of operation
3219 will be autodetected by the client, and it will fall
3220 back to using the idmapper.
3221 To turn off this behaviour, set the value to '0'.
3223 [NFS4] Specify an additional fixed unique ident-
3224 ification string that NFSv4 clients can insert into
3225 their nfs_client_id4 string. This is typically a
3226 UUID that is generated at system install time.
3228 nfs.send_implementation_id =
3229 [NFSv4.1] Send client implementation identification
3230 information in exchange_id requests.
3231 If zero, no implementation identification information
3233 The default is to send the implementation identification
3236 nfs.recover_lost_locks =
3237 [NFSv4] Attempt to recover locks that were lost due
3238 to a lease timeout on the server. Please note that
3239 doing this risks data corruption, since there are
3240 no guarantees that the file will remain unchanged
3241 after the locks are lost.
3242 If you want to enable the kernel legacy behaviour of
3243 attempting to recover these locks, then set this
3245 The default parameter value of '0' causes the kernel
3246 not to attempt recovery of lost locks.
3248 nfs4.layoutstats_timer =
3249 [NFSv4.2] Change the rate at which the kernel sends
3250 layoutstats to the pNFS metadata server.
3252 Setting this to value to 0 causes the kernel to use
3253 whatever value is the default set by the layout
3254 driver. A non-zero value sets the minimum interval
3255 in seconds between layoutstats transmissions.
3257 nfsd.nfs4_disable_idmapping=
3258 [NFSv4] When set to the default of '1', the NFSv4
3259 server will return only numeric uids and gids to
3260 clients using auth_sys, and will accept numeric uids
3261 and gids from such clients. This is intended to ease
3262 migration from NFSv2/v3.
3264 nmi_backtrace.backtrace_idle [KNL]
3265 Dump stacks even of idle CPUs in response to an
3266 NMI stack-backtrace request.
3268 nmi_debug= [KNL,SH] Specify one or more actions to take
3269 when a NMI is triggered.
3270 Format: [state][,regs][,debounce][,die]
3272 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3273 Format: [panic,][nopanic,][num]
3275 0 - turn hardlockup detector in nmi_watchdog off
3276 1 - turn hardlockup detector in nmi_watchdog on
3277 When panic is specified, panic when an NMI watchdog
3278 timeout occurs (or 'nopanic' to not panic on an NMI
3279 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3280 To disable both hard and soft lockup detectors,
3281 please see 'nowatchdog'.
3282 This is useful when you use a panic=... timeout and
3283 need the box quickly up again.
3285 These settings can be accessed at runtime via
3286 the nmi_watchdog and hardlockup_panic sysctls.
3288 netpoll.carrier_timeout=
3289 [NET] Specifies amount of time (in seconds) that
3290 netpoll should wait for a carrier. By default netpoll
3293 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3294 emulation library even if a 387 maths coprocessor
3297 no5lvl [X86-64] Disable 5-level paging mode. Forces
3298 kernel to use 4-level paging instead.
3300 nofsgsbase [X86] Disables FSGSBASE instructions.
3303 [HW] Never suspend the console
3304 Disable suspending of consoles during suspend and
3305 hibernate operations. Once disabled, debugging
3306 messages can reach various consoles while the rest
3307 of the system is being put to sleep (ie, while
3308 debugging driver suspend/resume hooks). This may
3309 not work reliably with all consoles, but is known
3310 to work with serial and VGA consoles.
3311 To facilitate more flexible debugging, we also add
3312 console_suspend, a printk module parameter to control
3313 it. Users could use console_suspend (usually
3314 /sys/module/printk/parameters/console_suspend) to
3315 turn on/off it dynamically.
3317 novmcoredd [KNL,KDUMP]
3318 Disable device dump. Device dump allows drivers to
3319 append dump data to vmcore so you can collect driver
3320 specified debug info. Drivers can append the data
3321 without any limit and this data is stored in memory,
3322 so this may cause significant memory stress. Disabling
3323 device dump can help save memory but the driver debug
3324 data will be no longer available. This parameter
3325 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3328 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3329 caches in the slab allocator. Saves per-node memory,
3330 but will impact performance.
3334 noaltinstr [S390] Disables alternative instructions patching
3335 (CPU alternatives feature).
3337 noapic [SMP,APIC] Tells the kernel to not make use of any
3338 IOAPICs that may be present in the system.
3340 noautogroup Disable scheduler automatic task group creation.
3342 nobats [PPC] Do not use BATs for mapping kernel lowmem
3343 on "Classic" PPC cores.
3347 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
3349 delayacct [KNL] Enable per-task delay accounting
3351 nodsp [SH] Disable hardware DSP at boot time.
3353 noefi Disable EFI runtime services support.
3355 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3360 On X86-32 available only on PAE configured kernels.
3361 noexec=on: enable non-executable mappings (default)
3362 noexec=off: disable non-executable mappings
3365 Disable SMAP (Supervisor Mode Access Prevention)
3366 even if it is supported by processor.
3369 Disable SMEP (Supervisor Mode Execution Prevention)
3370 even if it is supported by processor.
3373 This affects only 32-bit executables.
3374 noexec32=on: enable non-executable mappings (default)
3375 read doesn't imply executable mappings
3376 noexec32=off: disable non-executable mappings
3377 read implies executable mappings
3379 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3381 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3382 register save and restore. The kernel will only save
3383 legacy floating-point registers on task switch.
3385 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings.
3387 nohugevmalloc [PPC] Disable kernel huge vmalloc mappings.
3389 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3390 Equivalent to smt=1.
3392 [KNL,X86] Disable symmetric multithreading (SMT).
3393 nosmt=force: Force disable SMT, cannot be undone
3394 via the sysfs control file.
3396 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3397 (bounds check bypass). With this option data leaks are
3398 possible in the system.
3400 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3401 the Spectre variant 2 (indirect branch prediction)
3402 vulnerability. System may allow data leaks with this
3405 nospec_store_bypass_disable
3406 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3409 [PPC] Don't flush the L1-D cache after accessing user data.
3411 noxsave [BUGS=X86] Disables x86 extended register state save
3412 and restore using xsave. The kernel will fallback to
3413 enabling legacy floating-point and sse state.
3415 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3416 register states. The kernel will fall back to use
3417 xsave to save the states. By using this parameter,
3418 performance of saving the states is degraded because
3419 xsave doesn't support modified optimization while
3420 xsaveopt supports it on xsaveopt enabled systems.
3422 noxsaves [X86] Disables xsaves and xrstors used in saving and
3423 restoring x86 extended register state in compacted
3424 form of xsave area. The kernel will fall back to use
3425 xsaveopt and xrstor to save and restore the states
3426 in standard form of xsave area. By using this
3427 parameter, xsave area per process might occupy more
3428 memory on xsaves enabled systems.
3430 nohlt [ARM,ARM64,MICROBLAZE,SH] Forces the kernel to busy wait
3431 in do_idle() and not use the arch_cpu_idle()
3432 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3433 to be effective. This is useful on platforms where the
3434 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3435 correctly or when doing power measurements to evalute
3436 the impact of the sleep instructions. This is also
3437 useful when using JTAG debugger.
3439 no_file_caps Tells the kernel not to honor file capabilities. The
3440 only way then for a file to be executed with privilege
3441 is to be setuid root or executed by root.
3443 nohalt [IA-64] Tells the kernel not to use the power saving
3444 function PAL_HALT_LIGHT when idle. This increases
3445 power-consumption. On the positive side, it reduces
3446 interrupt wake-up latency, which may improve performance
3447 in certain environments such as networked servers or
3451 Force pointers printed to the console or buffers to be
3452 unhashed. By default, when a pointer is printed via %p
3453 format string, that pointer is "hashed", i.e. obscured
3454 by hashing the pointer value. This is a security feature
3455 that hides actual kernel addresses from unprivileged
3456 users, but it also makes debugging the kernel more
3457 difficult since unequal pointers can no longer be
3458 compared. However, if this command-line option is
3459 specified, then all normal pointers will have their true
3460 value printed. Pointers printed via %pK may still be
3461 hashed. This option should only be specified when
3462 debugging the kernel. Please do not use on production
3465 nohibernate [HIBERNATION] Disable hibernation and resume.
3467 nohz= [KNL] Boottime enable/disable dynamic ticks
3468 Valid arguments: on, off
3471 nohz_full= [KNL,BOOT,SMP,ISOL]
3472 The argument is a cpu list, as described above.
3473 In kernels built with CONFIG_NO_HZ_FULL=y, set
3474 the specified list of CPUs whose tick will be stopped
3475 whenever possible. The boot CPU will be forced outside
3476 the range to maintain the timekeeping. Any CPUs
3477 in this list will have their RCU callbacks offloaded,
3478 just as if they had also been called out in the
3479 rcu_nocbs= boot parameter.
3481 noiotrap [SH] Disables trapped I/O port accesses.
3483 noirqdebug [X86-32] Disables the code which attempts to detect and
3484 disable unhandled interrupt sources.
3486 no_timer_check [X86,APIC] Disables the code which tests for
3487 broken timer IRQ sources.
3489 noisapnp [ISAPNP] Disables ISA PnP code.
3491 noinitrd [RAM] Tells the kernel not to load any configured
3494 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3496 [Deprecated - use intremap=off]
3500 noinvpcid [X86] Disable the INVPCID cpu feature.
3502 nojitter [IA-64] Disables jitter checking for ITC timers.
3504 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3506 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3510 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3511 clock and use the default one.
3513 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time
3514 accounting. steal time is computed, but won't
3515 influence scheduler behaviour
3517 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3519 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3521 noltlbs [PPC] Do not use large page/tlb entries for kernel
3522 lowmem mapping on PPC40x and PPC8xx
3524 nomca [IA-64] Disable machine check abort handling
3526 nomce [X86-32] Disable Machine Check Exception
3528 nomfgpt [X86-32] Disable Multi-Function General Purpose
3529 Timer usage (for AMD Geode machines).
3531 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3532 shutdown the other cpus. Instead use the REBOOT_VECTOR
3535 nomodule Disable module load
3537 nopat [X86] Disable PAT (page attribute table extension of
3538 pagetables) support.
3540 nopcid [X86-64] Disable the PCID cpu feature.
3542 norandmaps Don't use address space randomization. Equivalent to
3543 echo 0 > /proc/sys/kernel/randomize_va_space
3545 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3546 with UP alternatives
3548 nordrand [X86] Disable kernel use of the RDRAND and
3549 RDSEED instructions even if they are supported
3550 by the processor. RDRAND and RDSEED are still
3551 available to user space applications.
3553 noresume [SWSUSP] Disables resume and restores original swap
3556 no-scroll [VGA] Disables scrollback.
3557 This is required for the Braillex ib80-piezo Braille
3558 reader made by F.H. Papenmeier (Germany).
3562 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
3564 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3566 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3567 and disable the IO APIC. legacy for "maxcpus=0".
3569 nosoftlockup [KNL] Disable the soft-lockup detector.
3571 nosync [HW,M68K] Disables sync negotiation for all devices.
3573 nowatchdog [KNL] Disable both lockup detectors, i.e.
3574 soft-lockup and NMI watchdog (hard-lockup).
3578 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3580 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
3581 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
3582 Some features depend on CPU0. Known dependencies are:
3583 1. Resume from suspend/hibernate depends on CPU0.
3584 Suspend/hibernate will fail if CPU0 is offline and you
3585 need to online CPU0 before suspend/hibernate.
3586 2. PIC interrupts also depend on CPU0. CPU0 can't be
3587 removed if a PIC interrupt is detected.
3588 It's said poweroff/reboot may depend on CPU0 on some
3589 machines although I haven't seen such issues so far
3590 after CPU0 is offline on a few tested machines.
3591 If the dependencies are under your control, you can
3592 turn on cpu0_hotplug.
3594 nps_mtm_hs_ctr= [KNL,ARC]
3595 This parameter sets the maximum duration, in
3596 cycles, each HW thread of the CTOP can run
3597 without interruptions, before HW switches it.
3598 The actual maximum duration is 16 times this
3600 Format: integer between 1 and 255
3603 nptcg= [IA-64] Override max number of concurrent global TLB
3604 purges which is reported from either PAL_VM_SUMMARY or
3607 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3608 could support. nr_cpus=n : n >= 1 limits the kernel to
3609 support 'n' processors. It could be larger than the
3610 number of already plugged CPU during bootup, later in
3611 runtime you can physically add extra cpu until it reaches
3612 n. So during boot up some boot time memory for per-cpu
3613 variables need be pre-allocated for later physical cpu
3616 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3618 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86] Disable NUMA, Only
3619 set up a single NUMA node spanning all memory.
3621 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
3623 Allowed values are enable and disable
3625 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3626 'node', 'default' can be specified
3627 This can be set from sysctl after boot.
3628 See Documentation/admin-guide/sysctl/vm.rst for details.
3630 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3631 See Documentation/core-api/debugging-via-ohci1394.rst for more
3634 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3635 Rather than timing out after 20 ms if an EC
3636 command is not properly ACKed, override the length
3637 of the timeout. We have interrupts disabled while
3638 waiting for the ACK, so if this is set too high
3639 interrupts *may* be lost!
3641 omap_mux= [OMAP] Override bootloader pin multiplexing.
3642 Format: <mux_mode0.mode_name=value>...
3643 For example, to override I2C bus2:
3644 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3646 oops=panic Always panic on oopses. Default is to just kill the
3647 process, but there is a small probability of
3648 deadlocking the machine.
3649 This will also cause panics on machine check exceptions.
3650 Useful together with panic=30 to trigger a reboot.
3653 [KNL] Boolean flag to control whether the page allocator
3654 should randomize its free lists. The randomization may
3655 be automatically enabled if the kernel detects it is
3656 running on a platform with a direct-mapped memory-side
3657 cache, and this parameter can be used to
3658 override/disable that behavior. The state of the flag
3659 can be read from sysfs at:
3660 /sys/module/page_alloc/parameters/shuffle.
3662 page_owner= [KNL] Boot-time page_owner enabling option.
3663 Storage of the information about who allocated
3664 each page is disabled in default. With this switch,
3666 on: enable the feature
3668 page_poison= [KNL] Boot-time parameter changing the state of
3669 poisoning on the buddy allocator, available with
3670 CONFIG_PAGE_POISONING=y.
3671 off: turn off poisoning (default)
3672 on: turn on poisoning
3674 page_reporting.page_reporting_order=
3675 [KNL] Minimal page reporting order
3677 Adjust the minimal page reporting order. The page
3678 reporting is disabled when it exceeds (MAX_ORDER-1).
3680 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3681 timeout > 0: seconds before rebooting
3682 timeout = 0: wait forever
3683 timeout < 0: reboot immediately
3686 panic_print= Bitmask for printing system info when panic happens.
3687 User can chose combination of the following bits:
3688 bit 0: print all tasks info
3689 bit 1: print system memory info
3690 bit 2: print timer info
3691 bit 3: print locks info if CONFIG_LOCKDEP is on
3692 bit 4: print ftrace buffer
3693 bit 5: print all printk messages in buffer
3695 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3696 Format: <hex>[,nousertaint]
3697 Hexadecimal bitmask representing the set of TAINT flags
3698 that will cause the kernel to panic when add_taint() is
3699 called with any of the flags in this set.
3700 The optional switch "nousertaint" can be utilized to
3701 prevent userspace forced crashes by writing to sysctl
3702 /proc/sys/kernel/tainted any flagset matching with the
3703 bitmask set on panic_on_taint.
3704 See Documentation/admin-guide/tainted-kernels.rst for
3705 extra details on the taint flags that users can pick
3706 to compose the bitmask to assign to panic_on_taint.
3708 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3711 crash_kexec_post_notifiers
3712 Run kdump after running panic-notifiers and dumping
3713 kmsg. This only for the users who doubt kdump always
3714 succeeds in any situation.
3715 Note that this also increases risks of kdump failure,
3716 because some panic notifiers can make the crashed
3717 kernel more unstable.
3719 parkbd.port= [HW] Parallel port number the keyboard adapter is
3720 connected to, default is 0.
3722 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3723 0 for XT, 1 for AT (default is AT).
3726 parport= [HW,PPT] Specify parallel ports. 0 disables.
3727 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3728 Use 'auto' to force the driver to use any
3729 IRQ/DMA settings detected (the default is to
3730 ignore detected IRQ/DMA settings because of
3731 possible conflicts). You can specify the base
3732 address, IRQ, and DMA settings; IRQ and DMA
3733 should be numbers, or 'auto' (for using detected
3734 settings on that particular port), or 'nofifo'
3735 (to avoid using a FIFO even if it is detected).
3736 Parallel ports are assigned in the order they
3737 are specified on the command line, starting
3740 parport_init_mode= [HW,PPT]
3741 Configure VIA parallel port to operate in
3742 a specific mode. This is necessary on Pegasos
3743 computer where firmware has no options for setting
3744 up parallel port mode and sets it to spp.
3745 Currently this function knows 686a and 8231 chips.
3746 Format: [spp|ps2|epp|ecp|ecpepp]
3748 pata_legacy.all= [HW,LIBATA]
3750 Set to non-zero to probe primary and secondary ISA
3751 port ranges on PCI systems where no PCI PATA device
3752 has been found at either range. Disabled by default.
3754 pata_legacy.autospeed= [HW,LIBATA]
3756 Set to non-zero if a chip is present that snoops speed
3757 changes. Disabled by default.
3759 pata_legacy.ht6560a= [HW,LIBATA]
3761 Set to 1, 2, or 3 for HT 6560A on the primary channel,
3762 the secondary channel, or both channels respectively.
3763 Disabled by default.
3765 pata_legacy.ht6560b= [HW,LIBATA]
3767 Set to 1, 2, or 3 for HT 6560B on the primary channel,
3768 the secondary channel, or both channels respectively.
3769 Disabled by default.
3771 pata_legacy.iordy_mask= [HW,LIBATA]
3773 IORDY enable mask. Set individual bits to allow IORDY
3774 for the respective channel. Bit 0 is for the first
3775 legacy channel handled by this driver, bit 1 is for
3776 the second channel, and so on. The sequence will often
3777 correspond to the primary legacy channel, the secondary
3778 legacy channel, and so on, but the handling of a PCI
3779 bus and the use of other driver options may interfere
3780 with the sequence. By default IORDY is allowed across
3783 pata_legacy.opti82c46x= [HW,LIBATA]
3785 Set to 1, 2, or 3 for Opti 82c611A on the primary
3786 channel, the secondary channel, or both channels
3787 respectively. Disabled by default.
3789 pata_legacy.opti82c611a= [HW,LIBATA]
3791 Set to 1, 2, or 3 for Opti 82c465MV on the primary
3792 channel, the secondary channel, or both channels
3793 respectively. Disabled by default.
3795 pata_legacy.pio_mask= [HW,LIBATA]
3797 PIO mode mask for autospeed devices. Set individual
3798 bits to allow the use of the respective PIO modes.
3799 Bit 0 is for mode 0, bit 1 is for mode 1, and so on.
3800 All modes allowed by default.
3802 pata_legacy.probe_all= [HW,LIBATA]
3804 Set to non-zero to probe tertiary and further ISA
3805 port ranges on PCI systems. Disabled by default.
3807 pata_legacy.probe_mask= [HW,LIBATA]
3809 Probe mask for legacy ISA PATA ports. Depending on
3810 platform configuration and the use of other driver
3811 options up to 6 legacy ports are supported: 0x1f0,
3812 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing
3813 of individual ports can be disabled by setting the
3814 corresponding bits in the mask to 1. Bit 0 is for
3815 the first port in the list above (0x1f0), and so on.
3816 By default all supported ports are probed.
3818 pata_legacy.qdi= [HW,LIBATA]
3820 Set to non-zero to probe QDI controllers. By default
3821 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
3823 pata_legacy.winbond= [HW,LIBATA]
3825 Set to non-zero to probe Winbond controllers. Use
3826 the standard I/O port (0x130) if 1, otherwise the
3827 value given is the I/O port to use (typically 0x1b0).
3828 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
3831 pata_platform.pio_mask= [HW,LIBATA]
3833 Supported PIO mode mask. Set individual bits to allow
3834 the use of the respective PIO modes. Bit 0 is for
3835 mode 0, bit 1 is for mode 1, and so on. Mode 0 only
3839 Halt all CPUs after the first oops has been printed for
3840 the specified number of seconds. This is to be used if
3841 your oopses keep scrolling off the screen.
3846 See header of drivers/block/paride/pcd.c.
3847 See also Documentation/admin-guide/blockdev/paride.rst.
3849 pci=option[,option...] [PCI] various PCI subsystem options.
3851 Some options herein operate on a specific device
3852 or a set of devices (<pci_dev>). These are
3853 specified in one of the following formats:
3855 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
3856 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
3858 Note: the first format specifies a PCI
3859 bus/device/function address which may change
3860 if new hardware is inserted, if motherboard
3861 firmware changes, or due to changes caused
3862 by other kernel parameters. If the
3863 domain is left unspecified, it is
3864 taken to be zero. Optionally, a path
3865 to a device through multiple device/function
3866 addresses can be specified after the base
3867 address (this is more robust against
3868 renumbering issues). The second format
3869 selects devices using IDs from the
3870 configuration space which may match multiple
3871 devices in the system.
3873 earlydump dump PCI config space before the kernel
3875 off [X86] don't probe for the PCI bus
3876 bios [X86-32] force use of PCI BIOS, don't access
3877 the hardware directly. Use this if your machine
3878 has a non-standard PCI host bridge.
3879 nobios [X86-32] disallow use of PCI BIOS, only direct
3880 hardware access methods are allowed. Use this
3881 if you experience crashes upon bootup and you
3882 suspect they are caused by the BIOS.
3883 conf1 [X86] Force use of PCI Configuration Access
3884 Mechanism 1 (config address in IO port 0xCF8,
3885 data in IO port 0xCFC, both 32-bit).
3886 conf2 [X86] Force use of PCI Configuration Access
3887 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3888 the function, IO port 0xCFA, also 8-bit, sets
3889 bus number. The config space is then accessed
3890 through ports 0xC000-0xCFFF).
3891 See http://wiki.osdev.org/PCI for more info
3892 on the configuration access mechanisms.
3893 noaer [PCIE] If the PCIEAER kernel config parameter is
3894 enabled, this kernel boot option can be used to
3895 disable the use of PCIE advanced error reporting.
3896 nodomains [PCI] Disable support for multiple PCI
3897 root domains (aka PCI segments, in ACPI-speak).
3898 nommconf [X86] Disable use of MMCONFIG for PCI
3900 check_enable_amd_mmconf [X86] check for and enable
3901 properly configured MMIO access to PCI
3902 config space on AMD family 10h CPU
3903 nomsi [MSI] If the PCI_MSI kernel config parameter is
3904 enabled, this kernel boot option can be used to
3905 disable the use of MSI interrupts system-wide.
3906 noioapicquirk [APIC] Disable all boot interrupt quirks.
3907 Safety option to keep boot IRQs enabled. This
3908 should never be necessary.
3909 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3910 primary IO-APIC for bridges that cannot disable
3911 boot IRQs. This fixes a source of spurious IRQs
3912 when the system masks IRQs.
3913 noioapicreroute [APIC] Disable workaround that uses the
3914 boot IRQ equivalent of an IRQ that connects to
3915 a chipset where boot IRQs cannot be disabled.
3916 The opposite of ioapicreroute.
3917 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3918 routing table. These calls are known to be buggy
3919 on several machines and they hang the machine
3920 when used, but on other computers it's the only
3921 way to get the interrupt routing table. Try
3922 this option if the kernel is unable to allocate
3923 IRQs or discover secondary PCI buses on your
3925 rom [X86] Assign address space to expansion ROMs.
3926 Use with caution as certain devices share
3927 address decoders between ROMs and other
3929 norom [X86] Do not assign address space to
3930 expansion ROMs that do not already have
3931 BIOS assigned address ranges.
3932 nobar [X86] Do not assign address space to the
3933 BARs that weren't assigned by the BIOS.
3934 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3935 assigned automatically to PCI devices. You can
3936 make the kernel exclude IRQs of your ISA cards
3938 pirqaddr=0xAAAAA [X86] Specify the physical address
3939 of the PIRQ table (normally generated
3940 by the BIOS) if it is outside the
3941 F0000h-100000h range.
3942 lastbus=N [X86] Scan all buses thru bus #N. Can be
3943 useful if the kernel is unable to find your
3944 secondary buses and you want to tell it
3945 explicitly which ones they are.
3946 assign-busses [X86] Always assign all PCI bus
3947 numbers ourselves, overriding
3948 whatever the firmware may have done.
3949 usepirqmask [X86] Honor the possible IRQ mask stored
3950 in the BIOS $PIR table. This is needed on
3951 some systems with broken BIOSes, notably
3952 some HP Pavilion N5400 and Omnibook XE3
3953 notebooks. This will have no effect if ACPI
3954 IRQ routing is enabled.
3955 noacpi [X86] Do not use ACPI for IRQ routing
3956 or for PCI scanning.
3957 use_crs [X86] Use PCI host bridge window information
3958 from ACPI. On BIOSes from 2008 or later, this
3959 is enabled by default. If you need to use this,
3960 please report a bug.
3961 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3962 If you need to use this, please report a bug.
3963 routeirq Do IRQ routing for all PCI devices.
3964 This is normally done in pci_enable_device(),
3965 so this option is a temporary workaround
3966 for broken drivers that don't call it.
3967 skip_isa_align [X86] do not align io start addr, so can
3968 handle more pci cards
3969 noearly [X86] Don't do any early type 1 scanning.
3970 This might help on some broken boards which
3971 machine check when some devices' config space
3972 is read. But various workarounds are disabled
3973 and some IOMMU drivers will not work.
3974 bfsort Sort PCI devices into breadth-first order.
3975 This sorting is done to get a device
3976 order compatible with older (<= 2.4) kernels.
3977 nobfsort Don't sort PCI devices into breadth-first order.
3978 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3979 tuning and use the BIOS-configured MPS defaults.
3980 pcie_bus_safe Set every device's MPS to the largest value
3981 supported by all devices below the root complex.
3982 pcie_bus_perf Set device MPS to the largest allowable MPS
3983 based on its parent bus. Also set MRRS (Max
3984 Read Request Size) to the largest supported
3985 value (no larger than the MPS that the device
3986 or bus can support) for best performance.
3987 pcie_bus_peer2peer Set every device's MPS to 128B, which
3988 every device is guaranteed to support. This
3989 configuration allows peer-to-peer DMA between
3990 any pair of devices, possibly at the cost of
3991 reduced performance. This also guarantees
3992 that hot-added devices will work.
3993 cbiosize=nn[KMG] The fixed amount of bus space which is
3994 reserved for the CardBus bridge's IO window.
3995 The default value is 256 bytes.
3996 cbmemsize=nn[KMG] The fixed amount of bus space which is
3997 reserved for the CardBus bridge's memory
3998 window. The default value is 64 megabytes.
4001 [<order of align>@]<pci_dev>[; ...]
4002 Specifies alignment and device to reassign
4003 aligned memory resources. How to
4004 specify the device is described above.
4005 If <order of align> is not specified,
4006 PAGE_SIZE is used as alignment.
4007 A PCI-PCI bridge can be specified if resource
4008 windows need to be expanded.
4009 To specify the alignment for several
4010 instances of a device, the PCI vendor,
4011 device, subvendor, and subdevice may be
4012 specified, e.g., 12@pci:8086:9c22:103c:198f
4013 for 4096-byte alignment.
4014 ecrc= Enable/disable PCIe ECRC (transaction layer
4015 end-to-end CRC checking).
4016 bios: Use BIOS/firmware settings. This is the
4020 hpiosize=nn[KMG] The fixed amount of bus space which is
4021 reserved for hotplug bridge's IO window.
4022 Default size is 256 bytes.
4023 hpmmiosize=nn[KMG] The fixed amount of bus space which is
4024 reserved for hotplug bridge's MMIO window.
4025 Default size is 2 megabytes.
4026 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
4027 reserved for hotplug bridge's MMIO_PREF window.
4028 Default size is 2 megabytes.
4029 hpmemsize=nn[KMG] The fixed amount of bus space which is
4030 reserved for hotplug bridge's MMIO and
4032 Default size is 2 megabytes.
4033 hpbussize=nn The minimum amount of additional bus numbers
4034 reserved for buses below a hotplug bridge.
4036 realloc= Enable/disable reallocating PCI bridge resources
4037 if allocations done by BIOS are too small to
4038 accommodate resources required by all child
4040 off: Turn realloc off
4042 realloc same as realloc=on
4043 noari do not use PCIe ARI.
4044 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
4045 do not use PCIe ATS (and IOMMU device IOTLB).
4046 pcie_scan_all Scan all possible PCIe devices. Otherwise we
4047 only look for one device below a PCIe downstream
4049 big_root_window Try to add a big 64bit memory window to the PCIe
4050 root complex on AMD CPUs. Some GFX hardware
4051 can resize a BAR to allow access to all VRAM.
4052 Adding the window is slightly risky (it may
4053 conflict with unreported devices), so this
4055 disable_acs_redir=<pci_dev>[; ...]
4056 Specify one or more PCI devices (in the format
4057 specified above) separated by semicolons.
4058 Each device specified will have the PCI ACS
4059 redirect capabilities forced off which will
4060 allow P2P traffic between devices through
4061 bridges without forcing it upstream. Note:
4062 this removes isolation between devices and
4063 may put more devices in an IOMMU group.
4064 force_floating [S390] Force usage of floating interrupts.
4065 nomio [S390] Do not use MIO instructions.
4066 norid [S390] ignore the RID field and force use of
4067 one PCI domain per PCI function
4069 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
4072 force Enable ASPM even on devices that claim not to support it.
4073 WARNING: Forcing ASPM on may cause system lockups.
4075 pcie_ports= [PCIE] PCIe port services handling:
4076 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
4077 even if the platform doesn't give the OS permission to
4078 use them. This may cause conflicts if the platform
4079 also tries to use these services.
4080 dpc-native Use native PCIe service for DPC only. May
4081 cause conflicts if firmware uses AER or DPC.
4082 compat Disable native PCIe services (PME, AER, DPC, PCIe
4085 pcie_port_pm= [PCIE] PCIe port power management handling:
4086 off Disable power management of all PCIe ports
4087 force Forcibly enable power management of all PCIe ports
4089 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
4090 nomsi Do not use MSI for native PCIe PME signaling (this makes
4091 all PCIe root ports use INTx for all services).
4093 pcmv= [HW,PCMCIA] BadgePAD 4
4097 Keep all power-domains already enabled by bootloader on,
4098 even if no driver has claimed them. This is useful
4099 for debug and development, but should not be
4100 needed on a platform with proper driver support.
4103 See Documentation/admin-guide/blockdev/paride.rst.
4105 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
4108 See arch/parisc/kernel/pdc_chassis.c
4110 percpu_alloc= Select which percpu first chunk allocator to use.
4111 Currently supported values are "embed" and "page".
4112 Archs may support subset or none of the selections.
4113 See comments in mm/percpu.c for details on each
4114 allocator. This parameter is primarily for debugging
4115 and performance comparison.
4118 See Documentation/admin-guide/blockdev/paride.rst.
4121 See Documentation/admin-guide/blockdev/paride.rst.
4123 pirq= [SMP,APIC] Manual mp-table setup
4124 See Documentation/x86/i386/IO-APIC.rst.
4126 plip= [PPT,NET] Parallel port network link
4127 Format: { parport<nr> | timid | 0 }
4128 See also Documentation/admin-guide/parport.rst.
4130 pmtmr= [X86] Manual setup of pmtmr I/O Port.
4131 Override pmtimer IOPort with a hex value.
4134 pm_debug_messages [SUSPEND,KNL]
4135 Enable suspend/resume debug messages during boot up.
4138 Enable PNP debug messages (depends on the
4139 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4140 via /sys/module/pnp/parameters/debug. We always show
4141 current resource usage; turning this on also shows
4142 possible settings and some assignment information.
4148 { on | off | curr | res | no-curr | no-res }
4151 [ISAPNP] Exclude IRQs for the autoconfiguration
4154 [ISAPNP] Exclude DMAs for the autoconfiguration
4156 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
4157 Ranges are in pairs (I/O port base and size).
4160 [ISAPNP] Exclude memory regions for the
4162 Ranges are in pairs (memory base and size).
4164 ports= [IP_VS_FTP] IPVS ftp helper module
4166 Up to 8 (IP_VS_APP_MAX_PORTS) ports
4168 Format: <port>,<port>....
4170 powersave=off [PPC] This option disables power saving features.
4171 It specifically disables cpuidle and sets the
4172 platform machine description specific power_save
4173 function to NULL. On Idle the CPU just reduces
4176 ppc_strict_facility_enable
4177 [PPC] This option catches any kernel floating point,
4178 Altivec, VSX and SPE outside of regions specifically
4179 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
4180 There is some performance impact when enabling this.
4184 Disable Hardware Transactional Memory
4187 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
4188 none - Limited to cond_resched() calls
4189 voluntary - Limited to cond_resched() and might_sleep() calls
4190 full - Any section that isn't explicitly preempt disabled
4191 can be preempted anytime.
4193 print-fatal-signals=
4194 [KNL] debug: print fatal signals
4196 If enabled, warn about various signal handling
4197 related application anomalies: too many signals,
4198 too many POSIX.1 timers, fatal signals causing a
4201 If you hit the warning due to signal overflow,
4202 you might want to try "ulimit -i unlimited".
4206 printk.always_kmsg_dump=
4207 Trigger kmsg_dump for cases other than kernel oops or
4209 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4212 printk.console_no_auto_verbose=
4213 Disable console loglevel raise on oops, panic
4214 or lockdep-detected issues (only if lock debug is on).
4215 With an exception to setups with low baudrate on
4216 serial console, keeping this 0 is a good choice
4217 in order to provide more debug information.
4219 default: 0 (auto_verbose is enabled)
4221 printk.devkmsg={on,off,ratelimit}
4222 Control writing to /dev/kmsg.
4223 on - unlimited logging to /dev/kmsg from userspace
4224 off - logging to /dev/kmsg disabled
4225 ratelimit - ratelimit the logging
4228 printk.time= Show timing data prefixed to each printk message line
4229 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4231 processor.max_cstate= [HW,ACPI]
4232 Limit processor to maximum C-state
4233 max_cstate=9 overrides any DMI blacklist limit.
4235 processor.nocst [HW,ACPI]
4236 Ignore the _CST method to determine C-states,
4237 instead using the legacy FADT method
4239 profile= [KNL] Enable kernel profiling via /proc/profile
4240 Format: [<profiletype>,]<number>
4241 Param: <profiletype>: "schedule", "sleep", or "kvm"
4242 [defaults to kernel profiling]
4243 Param: "schedule" - profile schedule points.
4244 Param: "sleep" - profile D-state sleeping (millisecs).
4245 Requires CONFIG_SCHEDSTATS
4246 Param: "kvm" - profile VM exits.
4247 Param: <number> - step/bucket size as a power of 2 for
4248 statistical time based profiling.
4250 prompt_ramdisk= [RAM] [Deprecated]
4252 prot_virt= [S390] enable hosting protected virtual machines
4253 isolated from the hypervisor (if hardware supports
4257 psi= [KNL] Enable or disable pressure stall information
4261 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
4262 probe for; one of (bare|imps|exps|lifebook|any).
4263 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4265 psmouse.resetafter= [HW,MOUSE]
4266 Try to reset the device after so many bad packets
4269 [HW,MOUSE] Set desired mouse resolution, in dpi.
4270 psmouse.smartscroll=
4271 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4272 0 = disabled, 1 = enabled (default).
4274 pstore.backend= Specify the name of the pstore backend to use
4277 See Documentation/admin-guide/blockdev/paride.rst.
4279 pti= [X86-64] Control Page Table Isolation of user and
4280 kernel address spaces. Disabling this feature
4281 removes hardening, but improves performance of
4282 system calls and interrupts.
4284 on - unconditionally enable
4285 off - unconditionally disable
4286 auto - kernel detects whether your CPU model is
4287 vulnerable to issues that PTI mitigates
4289 Not specifying this option is equivalent to pti=auto.
4292 Equivalent to pti=off
4295 [KNL] Number of legacy pty's. Overwrites compiled-in
4298 quiet [KNL] Disable most log messages
4303 See Documentation/admin-guide/md.rst.
4305 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4306 See Documentation/admin-guide/blockdev/ramdisk.rst.
4308 ramdisk_start= [RAM] RAM disk image start address
4310 random.trust_cpu={on,off}
4311 [KNL] Enable or disable trusting the use of the
4312 CPU's random number generator (if available) to
4313 fully seed the kernel's CRNG. Default is controlled
4314 by CONFIG_RANDOM_TRUST_CPU.
4316 randomize_kstack_offset=
4317 [KNL] Enable or disable kernel stack offset
4318 randomization, which provides roughly 5 bits of
4319 entropy, frustrating memory corruption attacks
4320 that depend on stack address determinism or
4321 cross-syscall address exposures. This is only
4322 available on architectures that have defined
4323 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
4324 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4325 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
4327 ras=option[,option,...] [KNL] RAS-specific options
4330 Disable the Correctable Errors Collector,
4331 see CONFIG_RAS_CEC help text.
4334 The argument is a cpu list, as described above.
4336 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
4337 the specified list of CPUs to be no-callback CPUs.
4338 Invocation of these CPUs' RCU callbacks will be
4339 offloaded to "rcuox/N" kthreads created for that
4340 purpose, where "x" is "p" for RCU-preempt, and
4341 "s" for RCU-sched, and "N" is the CPU number.
4342 This reduces OS jitter on the offloaded CPUs,
4343 which can be useful for HPC and real-time
4344 workloads. It can also improve energy efficiency
4345 for asymmetric multiprocessors.
4348 Rather than requiring that offloaded CPUs
4349 (specified by rcu_nocbs= above) explicitly
4350 awaken the corresponding "rcuoN" kthreads,
4351 make these kthreads poll for callbacks.
4352 This improves the real-time response for the
4353 offloaded CPUs by relieving them of the need to
4354 wake up the corresponding kthread, but degrades
4355 energy efficiency by requiring that the kthreads
4356 periodically wake up to do the polling.
4358 rcutree.blimit= [KNL]
4359 Set maximum number of finished RCU callbacks to
4360 process in one batch.
4362 rcutree.dump_tree= [KNL]
4363 Dump the structure of the rcu_node combining tree
4364 out at early boot. This is used for diagnostic
4365 purposes, to verify correct tree setup.
4367 rcutree.gp_cleanup_delay= [KNL]
4368 Set the number of jiffies to delay each step of
4369 RCU grace-period cleanup.
4371 rcutree.gp_init_delay= [KNL]
4372 Set the number of jiffies to delay each step of
4373 RCU grace-period initialization.
4375 rcutree.gp_preinit_delay= [KNL]
4376 Set the number of jiffies to delay each step of
4377 RCU grace-period pre-initialization, that is,
4378 the propagation of recent CPU-hotplug changes up
4379 the rcu_node combining tree.
4381 rcutree.use_softirq= [KNL]
4382 If set to zero, move all RCU_SOFTIRQ processing to
4383 per-CPU rcuc kthreads. Defaults to a non-zero
4384 value, meaning that RCU_SOFTIRQ is used by default.
4385 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4387 But note that CONFIG_PREEMPT_RT=y kernels disable
4388 this kernel boot parameter, forcibly setting it
4391 rcutree.rcu_fanout_exact= [KNL]
4392 Disable autobalancing of the rcu_node combining
4393 tree. This is used by rcutorture, and might
4394 possibly be useful for architectures having high
4395 cache-to-cache transfer latencies.
4397 rcutree.rcu_fanout_leaf= [KNL]
4398 Change the number of CPUs assigned to each
4399 leaf rcu_node structure. Useful for very
4400 large systems, which will choose the value 64,
4401 and for NUMA systems with large remote-access
4402 latencies, which will choose a value aligned
4403 with the appropriate hardware boundaries.
4405 rcutree.rcu_min_cached_objs= [KNL]
4406 Minimum number of objects which are cached and
4407 maintained per one CPU. Object size is equal
4408 to PAGE_SIZE. The cache allows to reduce the
4409 pressure to page allocator, also it makes the
4410 whole algorithm to behave better in low memory
4413 rcutree.rcu_delay_page_cache_fill_msec= [KNL]
4414 Set the page-cache refill delay (in milliseconds)
4415 in response to low-memory conditions. The range
4416 of permitted values is in the range 0:100000.
4418 rcutree.jiffies_till_first_fqs= [KNL]
4419 Set delay from grace-period initialization to
4420 first attempt to force quiescent states.
4421 Units are jiffies, minimum value is zero,
4422 and maximum value is HZ.
4424 rcutree.jiffies_till_next_fqs= [KNL]
4425 Set delay between subsequent attempts to force
4426 quiescent states. Units are jiffies, minimum
4427 value is one, and maximum value is HZ.
4429 rcutree.jiffies_till_sched_qs= [KNL]
4430 Set required age in jiffies for a
4431 given grace period before RCU starts
4432 soliciting quiescent-state help from
4433 rcu_note_context_switch() and cond_resched().
4434 If not specified, the kernel will calculate
4435 a value based on the most recent settings
4436 of rcutree.jiffies_till_first_fqs
4437 and rcutree.jiffies_till_next_fqs.
4438 This calculated value may be viewed in
4439 rcutree.jiffies_to_sched_qs. Any attempt to set
4440 rcutree.jiffies_to_sched_qs will be cheerfully
4443 rcutree.kthread_prio= [KNL,BOOT]
4444 Set the SCHED_FIFO priority of the RCU per-CPU
4445 kthreads (rcuc/N). This value is also used for
4446 the priority of the RCU boost threads (rcub/N)
4447 and for the RCU grace-period kthreads (rcu_bh,
4448 rcu_preempt, and rcu_sched). If RCU_BOOST is
4449 set, valid values are 1-99 and the default is 1
4450 (the least-favored priority). Otherwise, when
4451 RCU_BOOST is not set, valid values are 0-99 and
4452 the default is zero (non-realtime operation).
4454 rcutree.rcu_nocb_gp_stride= [KNL]
4455 Set the number of NOCB callback kthreads in
4456 each group, which defaults to the square root
4457 of the number of CPUs. Larger numbers reduce
4458 the wakeup overhead on the global grace-period
4459 kthread, but increases that same overhead on
4460 each group's NOCB grace-period kthread.
4462 rcutree.qhimark= [KNL]
4463 Set threshold of queued RCU callbacks beyond which
4464 batch limiting is disabled.
4466 rcutree.qlowmark= [KNL]
4467 Set threshold of queued RCU callbacks below which
4468 batch limiting is re-enabled.
4470 rcutree.qovld= [KNL]
4471 Set threshold of queued RCU callbacks beyond which
4472 RCU's force-quiescent-state scan will aggressively
4473 enlist help from cond_resched() and sched IPIs to
4474 help CPUs more quickly reach quiescent states.
4475 Set to less than zero to make this be set based
4476 on rcutree.qhimark at boot time and to zero to
4477 disable more aggressive help enlistment.
4479 rcutree.rcu_idle_gp_delay= [KNL]
4480 Set wakeup interval for idle CPUs that have
4481 RCU callbacks (RCU_FAST_NO_HZ=y).
4483 rcutree.rcu_kick_kthreads= [KNL]
4484 Cause the grace-period kthread to get an extra
4485 wake_up() if it sleeps three times longer than
4486 it should at force-quiescent-state time.
4487 This wake_up() will be accompanied by a
4488 WARN_ONCE() splat and an ftrace_dump().
4490 rcutree.rcu_unlock_delay= [KNL]
4491 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4492 this specifies an rcu_read_unlock()-time delay
4493 in microseconds. This defaults to zero.
4494 Larger delays increase the probability of
4495 catching RCU pointer leaks, that is, buggy use
4496 of RCU-protected pointers after the relevant
4497 rcu_read_unlock() has completed.
4499 rcutree.sysrq_rcu= [KNL]
4500 Commandeer a sysrq key to dump out Tree RCU's
4501 rcu_node tree with an eye towards determining
4502 why a new grace period has not yet started.
4504 rcuscale.gp_async= [KNL]
4505 Measure performance of asynchronous
4506 grace-period primitives such as call_rcu().
4508 rcuscale.gp_async_max= [KNL]
4509 Specify the maximum number of outstanding
4510 callbacks per writer thread. When a writer
4511 thread exceeds this limit, it invokes the
4512 corresponding flavor of rcu_barrier() to allow
4513 previously posted callbacks to drain.
4515 rcuscale.gp_exp= [KNL]
4516 Measure performance of expedited synchronous
4517 grace-period primitives.
4519 rcuscale.holdoff= [KNL]
4520 Set test-start holdoff period. The purpose of
4521 this parameter is to delay the start of the
4522 test until boot completes in order to avoid
4525 rcuscale.kfree_rcu_test= [KNL]
4526 Set to measure performance of kfree_rcu() flooding.
4528 rcuscale.kfree_rcu_test_double= [KNL]
4529 Test the double-argument variant of kfree_rcu().
4530 If this parameter has the same value as
4531 rcuscale.kfree_rcu_test_single, both the single-
4532 and double-argument variants are tested.
4534 rcuscale.kfree_rcu_test_single= [KNL]
4535 Test the single-argument variant of kfree_rcu().
4536 If this parameter has the same value as
4537 rcuscale.kfree_rcu_test_double, both the single-
4538 and double-argument variants are tested.
4540 rcuscale.kfree_nthreads= [KNL]
4541 The number of threads running loops of kfree_rcu().
4543 rcuscale.kfree_alloc_num= [KNL]
4544 Number of allocations and frees done in an iteration.
4546 rcuscale.kfree_loops= [KNL]
4547 Number of loops doing rcuscale.kfree_alloc_num number
4548 of allocations and frees.
4550 rcuscale.nreaders= [KNL]
4551 Set number of RCU readers. The value -1 selects
4552 N, where N is the number of CPUs. A value
4553 "n" less than -1 selects N-n+1, where N is again
4554 the number of CPUs. For example, -2 selects N
4555 (the number of CPUs), -3 selects N+1, and so on.
4556 A value of "n" less than or equal to -N selects
4559 rcuscale.nwriters= [KNL]
4560 Set number of RCU writers. The values operate
4561 the same as for rcuscale.nreaders.
4562 N, where N is the number of CPUs
4564 rcuscale.perf_type= [KNL]
4565 Specify the RCU implementation to test.
4567 rcuscale.shutdown= [KNL]
4568 Shut the system down after performance tests
4569 complete. This is useful for hands-off automated
4572 rcuscale.verbose= [KNL]
4573 Enable additional printk() statements.
4575 rcuscale.writer_holdoff= [KNL]
4576 Write-side holdoff between grace periods,
4577 in microseconds. The default of zero says
4580 rcutorture.fqs_duration= [KNL]
4581 Set duration of force_quiescent_state bursts
4584 rcutorture.fqs_holdoff= [KNL]
4585 Set holdoff time within force_quiescent_state bursts
4588 rcutorture.fqs_stutter= [KNL]
4589 Set wait time between force_quiescent_state bursts
4592 rcutorture.fwd_progress= [KNL]
4593 Enable RCU grace-period forward-progress testing
4594 for the types of RCU supporting this notion.
4596 rcutorture.fwd_progress_div= [KNL]
4597 Specify the fraction of a CPU-stall-warning
4598 period to do tight-loop forward-progress testing.
4600 rcutorture.fwd_progress_holdoff= [KNL]
4601 Number of seconds to wait between successive
4602 forward-progress tests.
4604 rcutorture.fwd_progress_need_resched= [KNL]
4605 Enclose cond_resched() calls within checks for
4606 need_resched() during tight-loop forward-progress
4609 rcutorture.gp_cond= [KNL]
4610 Use conditional/asynchronous update-side
4611 primitives, if available.
4613 rcutorture.gp_exp= [KNL]
4614 Use expedited update-side primitives, if available.
4616 rcutorture.gp_normal= [KNL]
4617 Use normal (non-expedited) asynchronous
4618 update-side primitives, if available.
4620 rcutorture.gp_sync= [KNL]
4621 Use normal (non-expedited) synchronous
4622 update-side primitives, if available. If all
4623 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4624 rcutorture.gp_normal=, and rcutorture.gp_sync=
4625 are zero, rcutorture acts as if is interpreted
4626 they are all non-zero.
4628 rcutorture.irqreader= [KNL]
4629 Run RCU readers from irq handlers, or, more
4630 accurately, from a timer handler. Not all RCU
4631 flavors take kindly to this sort of thing.
4633 rcutorture.leakpointer= [KNL]
4634 Leak an RCU-protected pointer out of the reader.
4635 This can of course result in splats, and is
4636 intended to test the ability of things like
4637 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
4640 rcutorture.n_barrier_cbs= [KNL]
4641 Set callbacks/threads for rcu_barrier() testing.
4643 rcutorture.nfakewriters= [KNL]
4644 Set number of concurrent RCU writers. These just
4645 stress RCU, they don't participate in the actual
4646 test, hence the "fake".
4648 rcutorture.nocbs_nthreads= [KNL]
4649 Set number of RCU callback-offload togglers.
4650 Zero (the default) disables toggling.
4652 rcutorture.nocbs_toggle= [KNL]
4653 Set the delay in milliseconds between successive
4654 callback-offload toggling attempts.
4656 rcutorture.nreaders= [KNL]
4657 Set number of RCU readers. The value -1 selects
4658 N-1, where N is the number of CPUs. A value
4659 "n" less than -1 selects N-n-2, where N is again
4660 the number of CPUs. For example, -2 selects N
4661 (the number of CPUs), -3 selects N+1, and so on.
4663 rcutorture.object_debug= [KNL]
4664 Enable debug-object double-call_rcu() testing.
4666 rcutorture.onoff_holdoff= [KNL]
4667 Set time (s) after boot for CPU-hotplug testing.
4669 rcutorture.onoff_interval= [KNL]
4670 Set time (jiffies) between CPU-hotplug operations,
4671 or zero to disable CPU-hotplug testing.
4673 rcutorture.read_exit= [KNL]
4674 Set the number of read-then-exit kthreads used
4675 to test the interaction of RCU updaters and
4676 task-exit processing.
4678 rcutorture.read_exit_burst= [KNL]
4679 The number of times in a given read-then-exit
4680 episode that a set of read-then-exit kthreads
4683 rcutorture.read_exit_delay= [KNL]
4684 The delay, in seconds, between successive
4685 read-then-exit testing episodes.
4687 rcutorture.shuffle_interval= [KNL]
4688 Set task-shuffle interval (s). Shuffling tasks
4689 allows some CPUs to go into dyntick-idle mode
4690 during the rcutorture test.
4692 rcutorture.shutdown_secs= [KNL]
4693 Set time (s) after boot system shutdown. This
4694 is useful for hands-off automated testing.
4696 rcutorture.stall_cpu= [KNL]
4697 Duration of CPU stall (s) to test RCU CPU stall
4698 warnings, zero to disable.
4700 rcutorture.stall_cpu_block= [KNL]
4701 Sleep while stalling if set. This will result
4702 in warnings from preemptible RCU in addition
4703 to any other stall-related activity.
4705 rcutorture.stall_cpu_holdoff= [KNL]
4706 Time to wait (s) after boot before inducing stall.
4708 rcutorture.stall_cpu_irqsoff= [KNL]
4709 Disable interrupts while stalling if set.
4711 rcutorture.stall_gp_kthread= [KNL]
4712 Duration (s) of forced sleep within RCU
4713 grace-period kthread to test RCU CPU stall
4714 warnings, zero to disable. If both stall_cpu
4715 and stall_gp_kthread are specified, the
4716 kthread is starved first, then the CPU.
4718 rcutorture.stat_interval= [KNL]
4719 Time (s) between statistics printk()s.
4721 rcutorture.stutter= [KNL]
4722 Time (s) to stutter testing, for example, specifying
4723 five seconds causes the test to run for five seconds,
4724 wait for five seconds, and so on. This tests RCU's
4725 ability to transition abruptly to and from idle.
4727 rcutorture.test_boost= [KNL]
4728 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
4729 "Maybe" means test if the RCU implementation
4730 under test support RCU priority boosting.
4732 rcutorture.test_boost_duration= [KNL]
4733 Duration (s) of each individual boost test.
4735 rcutorture.test_boost_interval= [KNL]
4736 Interval (s) between each boost test.
4738 rcutorture.test_no_idle_hz= [KNL]
4739 Test RCU's dyntick-idle handling. See also the
4740 rcutorture.shuffle_interval parameter.
4742 rcutorture.torture_type= [KNL]
4743 Specify the RCU implementation to test.
4745 rcutorture.verbose= [KNL]
4746 Enable additional printk() statements.
4748 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
4749 Dump ftrace buffer after reporting RCU CPU
4752 rcupdate.rcu_cpu_stall_suppress= [KNL]
4753 Suppress RCU CPU stall warning messages.
4755 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
4756 Suppress RCU CPU stall warning messages and
4757 rcutorture writer stall warnings that occur
4758 during early boot, that is, during the time
4759 before the init task is spawned.
4761 rcupdate.rcu_cpu_stall_timeout= [KNL]
4762 Set timeout for RCU CPU stall warning messages.
4764 rcupdate.rcu_expedited= [KNL]
4765 Use expedited grace-period primitives, for
4766 example, synchronize_rcu_expedited() instead
4767 of synchronize_rcu(). This reduces latency,
4768 but can increase CPU utilization, degrade
4769 real-time latency, and degrade energy efficiency.
4770 No effect on CONFIG_TINY_RCU kernels.
4772 rcupdate.rcu_normal= [KNL]
4773 Use only normal grace-period primitives,
4774 for example, synchronize_rcu() instead of
4775 synchronize_rcu_expedited(). This improves
4776 real-time latency, CPU utilization, and
4777 energy efficiency, but can expose users to
4778 increased grace-period latency. This parameter
4779 overrides rcupdate.rcu_expedited. No effect on
4780 CONFIG_TINY_RCU kernels.
4782 rcupdate.rcu_normal_after_boot= [KNL]
4783 Once boot has completed (that is, after
4784 rcu_end_inkernel_boot() has been invoked), use
4785 only normal grace-period primitives. No effect
4786 on CONFIG_TINY_RCU kernels.
4788 But note that CONFIG_PREEMPT_RT=y kernels enables
4789 this kernel boot parameter, forcibly setting
4790 it to the value one, that is, converting any
4791 post-boot attempt at an expedited RCU grace
4792 period to instead use normal non-expedited
4793 grace-period processing.
4795 rcupdate.rcu_task_ipi_delay= [KNL]
4796 Set time in jiffies during which RCU tasks will
4797 avoid sending IPIs, starting with the beginning
4798 of a given grace period. Setting a large
4799 number avoids disturbing real-time workloads,
4800 but lengthens grace periods.
4802 rcupdate.rcu_task_stall_timeout= [KNL]
4803 Set timeout in jiffies for RCU task stall warning
4804 messages. Disable with a value less than or equal
4807 rcupdate.rcu_self_test= [KNL]
4808 Run the RCU early boot self tests
4812 Run specified binary instead of /init from the ramdisk,
4813 used for early userspace startup. See initrd.
4816 force - Override the decision by the kernel to hide the
4817 advertisement of RDRAND support (this affects
4818 certain AMD processors because of buggy BIOS
4819 support, specifically around the suspend/resume
4823 Turn on/off individual RDT features. List is:
4824 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
4826 E.g. to turn on cmt and turn off mba use:
4830 Format (x86 or x86_64):
4831 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \
4833 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
4835 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
4836 (prefix with 'panic_' to set mode for panic
4838 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
4839 reboot_force is either force or not specified,
4840 reboot_cpu is s[mp]#### with #### being the processor
4841 to be used for rebooting.
4843 refscale.holdoff= [KNL]
4844 Set test-start holdoff period. The purpose of
4845 this parameter is to delay the start of the
4846 test until boot completes in order to avoid
4849 refscale.loops= [KNL]
4850 Set the number of loops over the synchronization
4851 primitive under test. Increasing this number
4852 reduces noise due to loop start/end overhead,
4853 but the default has already reduced the per-pass
4854 noise to a handful of picoseconds on ca. 2020
4857 refscale.nreaders= [KNL]
4858 Set number of readers. The default value of -1
4859 selects N, where N is roughly 75% of the number
4860 of CPUs. A value of zero is an interesting choice.
4862 refscale.nruns= [KNL]
4863 Set number of runs, each of which is dumped onto
4866 refscale.readdelay= [KNL]
4867 Set the read-side critical-section duration,
4868 measured in microseconds.
4870 refscale.scale_type= [KNL]
4871 Specify the read-protection implementation to test.
4873 refscale.shutdown= [KNL]
4874 Shut down the system at the end of the performance
4875 test. This defaults to 1 (shut it down) when
4876 refscale is built into the kernel and to 0 (leave
4877 it running) when refscale is built as a module.
4879 refscale.verbose= [KNL]
4880 Enable additional printk() statements.
4882 refscale.verbose_batched= [KNL]
4883 Batch the additional printk() statements. If zero
4884 (the default) or negative, print everything. Otherwise,
4885 print every Nth verbose statement, where N is the value
4889 [KNL, SMP] Set scheduler's default relax_domain_level.
4890 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
4892 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
4893 Format: <base1>,<size1>[,<base2>,<size2>,...]
4894 Reserve I/O ports or memory so the kernel won't use
4895 them. If <base> is less than 0x10000, the region
4896 is assumed to be I/O ports; otherwise it is memory.
4898 reservetop= [X86-32]
4900 Reserves a hole at the top of the kernel virtual
4903 reset_devices [KNL] Force drivers to reset the underlying device
4904 during initialization.
4907 Specify the partition device for software suspend
4909 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
4911 resume_offset= [SWSUSP]
4912 Specify the offset from the beginning of the partition
4913 given by "resume=" at which the swap header is located,
4914 in <PAGE_SIZE> units (needed only for swap files).
4915 See Documentation/power/swsusp-and-swap-files.rst
4917 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
4918 read the resume files
4920 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
4921 Useful for devices that are detected asynchronously
4922 (e.g. USB and MMC devices).
4924 hibernate= [HIBERNATION]
4925 noresume Don't check if there's a hibernation image
4926 present during boot.
4927 nocompress Don't compress/decompress hibernation images.
4928 no Disable hibernation and resume.
4929 protect_image Turn on image protection during restoration
4930 (that will set all pages holding image data
4931 during restoration read-only).
4933 retain_initrd [RAM] Keep initrd memory after extraction
4935 rfkill.default_state=
4936 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
4937 etc. communication is blocked by default.
4940 rfkill.master_switch_mode=
4941 0 The "airplane mode" button does nothing.
4942 1 The "airplane mode" button toggles between everything
4943 blocked and the previous configuration.
4944 2 The "airplane mode" button toggles between everything
4945 blocked and everything unblocked.
4947 rhash_entries= [KNL,NET]
4948 Set number of hash buckets for route cache
4951 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
4954 ro [KNL] Mount root device read-only on boot
4957 on Mark read-only kernel memory as read-only (default).
4958 off Leave read-only kernel memory writable for debugging.
4961 Enable the uart passthrough on the designated usb port
4962 on Rockchip SoCs. When active, the signals of the
4963 debug-uart get routed to the D+ and D- pins of the usb
4964 port and the regular usb controller gets disabled.
4966 root= [KNL] Root filesystem
4967 See name_to_dev_t comment in init/do_mounts.c.
4969 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
4970 mount the root filesystem
4972 rootflags= [KNL] Set root filesystem mount option string
4974 rootfstype= [KNL] Set root filesystem type
4976 rootwait [KNL] Wait (indefinitely) for root device to show up.
4977 Useful for devices that are detected asynchronously
4978 (e.g. USB and MMC devices).
4980 rproc_mem=nn[KMG][@address]
4981 [KNL,ARM,CMA] Remoteproc physical memory block.
4982 Memory area to be used by remote processor image,
4985 rw [KNL] Mount root device read-write on boot
4987 S [KNL] Run init in single mode
4989 s390_iommu= [HW,S390]
4990 Set s390 IOTLB flushing mode
4992 With strict flushing every unmap operation will result in
4993 an IOTLB flush. Default is lazy flushing before reuse,
4997 See drivers/net/irda/sa1100_ir.c.
4999 sched_verbose [KNL] Enables verbose scheduler debug messages.
5001 schedstats= [KNL,X86] Enable or disable scheduled statistics.
5002 Allowed values are enable and disable. This feature
5003 incurs a small amount of overhead in the scheduler
5004 but is useful for debugging and performance tuning.
5006 sched_thermal_decay_shift=
5007 [KNL, SMP] Set a decay shift for scheduler thermal
5008 pressure signal. Thermal pressure signal follows the
5009 default decay period of other scheduler pelt
5010 signals(usually 32 ms but configurable). Setting
5011 sched_thermal_decay_shift will left shift the decay
5012 period for the thermal pressure signal by the shift
5014 i.e. with the default pelt decay period of 32 ms
5015 sched_thermal_decay_shift thermal pressure decay pr
5019 Format: integer between 0 and 10
5022 scftorture.holdoff= [KNL]
5023 Number of seconds to hold off before starting
5024 test. Defaults to zero for module insertion and
5025 to 10 seconds for built-in smp_call_function()
5028 scftorture.longwait= [KNL]
5029 Request ridiculously long waits randomly selected
5030 up to the chosen limit in seconds. Zero (the
5031 default) disables this feature. Please note
5032 that requesting even small non-zero numbers of
5033 seconds can result in RCU CPU stall warnings,
5034 softlockup complaints, and so on.
5036 scftorture.nthreads= [KNL]
5037 Number of kthreads to spawn to invoke the
5038 smp_call_function() family of functions.
5039 The default of -1 specifies a number of kthreads
5040 equal to the number of CPUs.
5042 scftorture.onoff_holdoff= [KNL]
5043 Number seconds to wait after the start of the
5044 test before initiating CPU-hotplug operations.
5046 scftorture.onoff_interval= [KNL]
5047 Number seconds to wait between successive
5048 CPU-hotplug operations. Specifying zero (which
5049 is the default) disables CPU-hotplug operations.
5051 scftorture.shutdown_secs= [KNL]
5052 The number of seconds following the start of the
5053 test after which to shut down the system. The
5054 default of zero avoids shutting down the system.
5055 Non-zero values are useful for automated tests.
5057 scftorture.stat_interval= [KNL]
5058 The number of seconds between outputting the
5059 current test statistics to the console. A value
5060 of zero disables statistics output.
5062 scftorture.stutter_cpus= [KNL]
5063 The number of jiffies to wait between each change
5064 to the set of CPUs under test.
5066 scftorture.use_cpus_read_lock= [KNL]
5067 Use use_cpus_read_lock() instead of the default
5068 preempt_disable() to disable CPU hotplug
5069 while invoking one of the smp_call_function*()
5072 scftorture.verbose= [KNL]
5073 Enable additional printk() statements.
5075 scftorture.weight_single= [KNL]
5076 The probability weighting to use for the
5077 smp_call_function_single() function with a zero
5078 "wait" parameter. A value of -1 selects the
5079 default if all other weights are -1. However,
5080 if at least one weight has some other value, a
5081 value of -1 will instead select a weight of zero.
5083 scftorture.weight_single_wait= [KNL]
5084 The probability weighting to use for the
5085 smp_call_function_single() function with a
5086 non-zero "wait" parameter. See weight_single.
5088 scftorture.weight_many= [KNL]
5089 The probability weighting to use for the
5090 smp_call_function_many() function with a zero
5091 "wait" parameter. See weight_single.
5092 Note well that setting a high probability for
5093 this weighting can place serious IPI load
5096 scftorture.weight_many_wait= [KNL]
5097 The probability weighting to use for the
5098 smp_call_function_many() function with a
5099 non-zero "wait" parameter. See weight_single
5102 scftorture.weight_all= [KNL]
5103 The probability weighting to use for the
5104 smp_call_function_all() function with a zero
5105 "wait" parameter. See weight_single and
5108 scftorture.weight_all_wait= [KNL]
5109 The probability weighting to use for the
5110 smp_call_function_all() function with a
5111 non-zero "wait" parameter. See weight_single
5114 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
5115 xtime_lock contention on larger systems, and/or RCU lock
5116 contention on all systems with CONFIG_MAXSMP set.
5117 Format: { "0" | "1" }
5118 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
5120 Note: increases power consumption, thus should only be
5121 enabled if running jitter sensitive (HPC/RT) workloads.
5123 security= [SECURITY] Choose a legacy "major" security module to
5124 enable at boot. This has been deprecated by the
5127 selinux= [SELINUX] Disable or enable SELinux at boot time.
5128 Format: { "0" | "1" }
5129 See security/selinux/Kconfig help text.
5134 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
5135 Format: { "0" | "1" }
5136 See security/apparmor/Kconfig help text
5139 Default value is set via kernel config option.
5141 serialnumber [BUGS=X86-32]
5144 Maximal number of shapers.
5152 Enable merging of slabs with similar size when the
5153 kernel is built without CONFIG_SLAB_MERGE_DEFAULT.
5156 Disable merging of slabs with similar size. May be
5157 necessary if there is some reason to distinguish
5158 allocs to different slabs, especially in hardened
5159 environments where the risk of heap overflows and
5160 layout control by attackers can usually be
5161 frustrated by disabling merging. This will reduce
5162 most of the exposure of a heap attack to a single
5163 cache (risks via metadata attacks are mostly
5164 unchanged). Debug options disable merging on their
5166 For more information see Documentation/vm/slub.rst.
5168 slab_max_order= [MM, SLAB]
5169 Determines the maximum allowed order for slabs.
5170 A high setting may cause OOMs due to memory
5171 fragmentation. Defaults to 1 for systems with
5172 more than 32MB of RAM, 0 otherwise.
5174 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
5175 Enabling slub_debug allows one to determine the
5176 culprit if slab objects become corrupted. Enabling
5177 slub_debug can create guard zones around objects and
5178 may poison objects when not in use. Also tracks the
5179 last alloc / free. For more information see
5180 Documentation/vm/slub.rst.
5182 slub_max_order= [MM, SLUB]
5183 Determines the maximum allowed order for slabs.
5184 A high setting may cause OOMs due to memory
5185 fragmentation. For more information see
5186 Documentation/vm/slub.rst.
5188 slub_min_objects= [MM, SLUB]
5189 The minimum number of objects per slab. SLUB will
5190 increase the slab order up to slub_max_order to
5191 generate a sufficiently large slab able to contain
5192 the number of objects indicated. The higher the number
5193 of objects the smaller the overhead of tracking slabs
5194 and the less frequently locks need to be acquired.
5195 For more information see Documentation/vm/slub.rst.
5197 slub_min_order= [MM, SLUB]
5198 Determines the minimum page order for slabs. Must be
5199 lower than slub_max_order.
5200 For more information see Documentation/vm/slub.rst.
5202 slub_merge [MM, SLUB]
5203 Same with slab_merge.
5205 slub_nomerge [MM, SLUB]
5206 Same with slab_nomerge. This is supported for legacy.
5207 See slab_nomerge for more information.
5210 Format: <io1>[,<io2>[,...,<io8>]]
5212 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
5213 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
5214 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
5215 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
5216 smsc-ircc2.ircc_irq= [HW] IRQ line
5217 smsc-ircc2.ircc_dma= [HW] DMA channel
5218 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
5219 0: Toshiba Satellite 1800 (GP data pin select)
5220 1: Fast pin select (default)
5223 smt [KNL,S390] Set the maximum number of threads (logical
5224 CPUs) to use per physical CPU on systems capable of
5225 symmetric multithreading (SMT). Will be capped to the
5226 actual hardware limit.
5228 Default: -1 (no limit)
5231 [KNL] Should the soft-lockup detector generate panics.
5234 A value of 1 instructs the soft-lockup detector
5235 to panic the machine when a soft-lockup occurs. It is
5236 also controlled by the kernel.softlockup_panic sysctl
5237 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
5238 respective build-time switch to that functionality.
5240 softlockup_all_cpu_backtrace=
5241 [KNL] Should the soft-lockup detector generate
5242 backtraces on all cpus.
5245 sonypi.*= [HW] Sony Programmable I/O Control Device driver
5246 See Documentation/admin-guide/laptops/sonypi.rst
5248 spectre_v2= [X86] Control mitigation of Spectre variant 2
5249 (indirect branch speculation) vulnerability.
5250 The default operation protects the kernel from
5253 on - unconditionally enable, implies
5255 off - unconditionally disable, implies
5257 auto - kernel detects whether your CPU model is
5260 Selecting 'on' will, and 'auto' may, choose a
5261 mitigation method at run time according to the
5262 CPU, the available microcode, the setting of the
5263 CONFIG_RETPOLINE configuration option, and the
5264 compiler with which the kernel was built.
5266 Selecting 'on' will also enable the mitigation
5267 against user space to user space task attacks.
5269 Selecting 'off' will disable both the kernel and
5270 the user space protections.
5272 Specific mitigations can also be selected manually:
5274 retpoline - replace indirect branches
5275 retpoline,generic - google's original retpoline
5276 retpoline,amd - AMD-specific minimal thunk
5278 Not specifying this option is equivalent to
5282 [X86] Control mitigation of Spectre variant 2
5283 (indirect branch speculation) vulnerability between
5286 on - Unconditionally enable mitigations. Is
5287 enforced by spectre_v2=on
5289 off - Unconditionally disable mitigations. Is
5290 enforced by spectre_v2=off
5292 prctl - Indirect branch speculation is enabled,
5293 but mitigation can be enabled via prctl
5294 per thread. The mitigation control state
5295 is inherited on fork.
5298 - Like "prctl" above, but only STIBP is
5299 controlled per thread. IBPB is issued
5300 always when switching between different user
5304 - Same as "prctl" above, but all seccomp
5305 threads will enable the mitigation unless
5306 they explicitly opt out.
5309 - Like "seccomp" above, but only STIBP is
5310 controlled per thread. IBPB is issued
5311 always when switching between different
5312 user space processes.
5314 auto - Kernel selects the mitigation depending on
5315 the available CPU features and vulnerability.
5317 Default mitigation: "prctl"
5319 Not specifying this option is equivalent to
5320 spectre_v2_user=auto.
5322 spec_store_bypass_disable=
5323 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
5324 (Speculative Store Bypass vulnerability)
5326 Certain CPUs are vulnerable to an exploit against a
5327 a common industry wide performance optimization known
5328 as "Speculative Store Bypass" in which recent stores
5329 to the same memory location may not be observed by
5330 later loads during speculative execution. The idea
5331 is that such stores are unlikely and that they can
5332 be detected prior to instruction retirement at the
5333 end of a particular speculation execution window.
5335 In vulnerable processors, the speculatively forwarded
5336 store can be used in a cache side channel attack, for
5337 example to read memory to which the attacker does not
5338 directly have access (e.g. inside sandboxed code).
5340 This parameter controls whether the Speculative Store
5341 Bypass optimization is used.
5343 On x86 the options are:
5345 on - Unconditionally disable Speculative Store Bypass
5346 off - Unconditionally enable Speculative Store Bypass
5347 auto - Kernel detects whether the CPU model contains an
5348 implementation of Speculative Store Bypass and
5349 picks the most appropriate mitigation. If the
5350 CPU is not vulnerable, "off" is selected. If the
5351 CPU is vulnerable the default mitigation is
5352 architecture and Kconfig dependent. See below.
5353 prctl - Control Speculative Store Bypass per thread
5354 via prctl. Speculative Store Bypass is enabled
5355 for a process by default. The state of the control
5356 is inherited on fork.
5357 seccomp - Same as "prctl" above, but all seccomp threads
5358 will disable SSB unless they explicitly opt out.
5360 Default mitigations:
5363 On powerpc the options are:
5365 on,auto - On Power8 and Power9 insert a store-forwarding
5366 barrier on kernel entry and exit. On Power7
5367 perform a software flush on kernel entry and
5371 Not specifying this option is equivalent to
5372 spec_store_bypass_disable=auto.
5374 spia_io_base= [HW,MTD]
5380 [X86] Enable split lock detection or bus lock detection
5382 When enabled (and if hardware support is present), atomic
5383 instructions that access data across cache line
5384 boundaries will result in an alignment check exception
5385 for split lock detection or a debug exception for
5390 warn - the kernel will emit rate-limited warnings
5391 about applications triggering the #AC
5392 exception or the #DB exception. This mode is
5393 the default on CPUs that support split lock
5394 detection or bus lock detection. Default
5395 behavior is by #AC if both features are
5396 enabled in hardware.
5398 fatal - the kernel will send SIGBUS to applications
5399 that trigger the #AC exception or the #DB
5400 exception. Default behavior is by #AC if
5401 both features are enabled in hardware.
5404 Set system wide rate limit to N bus locks
5405 per second for bus lock detection.
5408 N/A for split lock detection.
5411 If an #AC exception is hit in the kernel or in
5412 firmware (i.e. not while executing in user mode)
5413 the kernel will oops in either "warn" or "fatal"
5416 #DB exception for bus lock is triggered only when
5420 Control the Special Register Buffer Data Sampling
5423 Certain CPUs are vulnerable to an MDS-like
5424 exploit which can leak bits from the random
5427 By default, this issue is mitigated by
5428 microcode. However, the microcode fix can cause
5429 the RDRAND and RDSEED instructions to become
5430 much slower. Among other effects, this will
5431 result in reduced throughput from /dev/urandom.
5433 The microcode mitigation can be disabled with
5434 the following option:
5436 off: Disable mitigation and remove
5437 performance impact to RDRAND and RDSEED
5439 srcutree.counter_wrap_check [KNL]
5440 Specifies how frequently to check for
5441 grace-period sequence counter wrap for the
5442 srcu_data structure's ->srcu_gp_seq_needed field.
5443 The greater the number of bits set in this kernel
5444 parameter, the less frequently counter wrap will
5445 be checked for. Note that the bottom two bits
5448 srcutree.exp_holdoff [KNL]
5449 Specifies how many nanoseconds must elapse
5450 since the end of the last SRCU grace period for
5451 a given srcu_struct until the next normal SRCU
5452 grace period will be considered for automatic
5453 expediting. Set to zero to disable automatic
5457 Speculative Store Bypass Disable control
5459 On CPUs that are vulnerable to the Speculative
5460 Store Bypass vulnerability and offer a
5461 firmware based mitigation, this parameter
5462 indicates how the mitigation should be used:
5464 force-on: Unconditionally enable mitigation for
5465 for both kernel and userspace
5466 force-off: Unconditionally disable mitigation for
5467 for both kernel and userspace
5468 kernel: Always enable mitigation in the
5469 kernel, and offer a prctl interface
5470 to allow userspace to register its
5471 interest in being mitigated too.
5473 stack_guard_gap= [MM]
5474 override the default stack gap protection. The value
5475 is in page units and it defines how many pages prior
5476 to (for stacks growing down) resp. after (for stacks
5477 growing up) the main stack are reserved for no other
5478 mapping. Default value is 256 pages.
5480 stack_depot_disable= [KNL]
5481 Setting this to true through kernel command line will
5482 disable the stack depot thereby saving the static memory
5483 consumed by the stack hash table. By default this is set
5487 Enabled the stack tracer on boot up.
5489 stacktrace_filter=[function-list]
5490 [FTRACE] Limit the functions that the stack tracer
5491 will trace at boot up. function-list is a comma-separated
5492 list of functions. This list can be changed at run
5493 time by the stack_trace_filter file in the debugfs
5494 tracing directory. Note, this enables stack tracing
5495 and the stacktrace above is not needed.
5499 Set the STI (builtin display/keyboard on the HP-PARISC
5500 machines) console (graphic card) which should be used
5501 as the initial boot-console.
5502 See also comment in drivers/video/console/sticore.c.
5505 See comment in drivers/video/console/sticore.c.
5508 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
5513 Enable or disable strict sigaltstack size checks
5514 against the required signal frame size which
5515 depends on the supported FPU features. This can
5516 be used to filter out binaries which have
5517 not yet been made aware of AT_MINSIGSTKSZ.
5519 sunrpc.min_resvport=
5520 sunrpc.max_resvport=
5522 SunRPC servers often require that client requests
5523 originate from a privileged port (i.e. a port in the
5524 range 0 < portnr < 1024).
5525 An administrator who wishes to reserve some of these
5526 ports for other uses may adjust the range that the
5527 kernel's sunrpc client considers to be privileged
5528 using these two parameters to set the minimum and
5529 maximum port values.
5531 sunrpc.svc_rpc_per_connection_limit=
5533 Limit the number of requests that the server will
5534 process in parallel from a single connection.
5535 The default value is 0 (no limit).
5539 Control how the NFS server code allocates CPUs to
5540 service thread pools. Depending on how many NICs
5541 you have and where their interrupts are bound, this
5542 option will affect which CPUs will do NFS serving.
5543 Note: this parameter cannot be changed while the
5544 NFS server is running.
5546 auto the server chooses an appropriate mode
5547 automatically using heuristics
5548 global a single global pool contains all CPUs
5549 percpu one pool for each CPU
5550 pernode one pool for each NUMA node (equivalent
5551 to global on non-NUMA machines)
5553 sunrpc.tcp_slot_table_entries=
5554 sunrpc.udp_slot_table_entries=
5556 Sets the upper limit on the number of simultaneous
5557 RPC calls that can be sent from the client to a
5558 server. Increasing these values may allow you to
5559 improve throughput, but will also increase the
5560 amount of memory reserved for use by the client.
5562 suspend.pm_test_delay=
5564 Sets the number of seconds to remain in a suspend test
5565 mode before resuming the system (see
5566 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
5567 is set. Default value is 5.
5570 Format: { on | off | y | n | 1 | 0 }
5571 This parameter controls use of the Protected
5572 Execution Facility on pSeries.
5575 [KNL] Enable accounting of swap in memory resource
5576 controller if no parameter or 1 is given or disable
5577 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
5579 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
5580 Format: { <int> | force | noforce }
5581 <int> -- Number of I/O TLB slabs
5582 force -- force using of bounce buffers even if they
5583 wouldn't be automatically used by the kernel
5584 noforce -- Never use bounce buffers (for debugging)
5589 Set a sysctl parameter, right before loading the init
5590 process, as if the value was written to the respective
5591 /proc/sys/... file. Both '.' and '/' are recognized as
5592 separators. Unrecognized parameters and invalid values
5593 are reported in the kernel log. Sysctls registered
5594 later by a loaded module cannot be set this way.
5595 Example: sysctl.vm.swappiness=40
5597 sysfs.deprecated=0|1 [KNL]
5598 Enable/disable old style sysfs layout for old udev
5599 on older distributions. When this option is enabled
5600 very new udev will not work anymore. When this option
5601 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
5602 in older udev will not work anymore.
5603 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
5604 the kernel configuration.
5606 sysrq_always_enabled
5608 Ignore sysrq setting - this boot parameter will
5609 neutralize any effect of /proc/sys/kernel/sysrq.
5610 Useful for debugging.
5612 tcpmhash_entries= [KNL,NET]
5613 Set the number of tcp_metrics_hash slots.
5614 Default value is 8192 or 16384 depending on total
5615 ram pages. This is used to specify the TCP metrics
5616 cache size. See Documentation/networking/ip-sysctl.rst
5617 "tcp_no_metrics_save" section for more details.
5621 test_suspend= [SUSPEND][,N]
5622 Specify "mem" (for Suspend-to-RAM) or "standby" (for
5623 standby suspend) or "freeze" (for suspend type freeze)
5624 as the system sleep state during system startup with
5625 the optional capability to repeat N number of times.
5626 The system is woken from this state using a
5627 wakeup-capable RTC alarm.
5629 thash_entries= [KNL,NET]
5630 Set number of hash buckets for TCP connection
5632 thermal.act= [HW,ACPI]
5633 -1: disable all active trip points in all thermal zones
5634 <degrees C>: override all lowest active trip points
5636 thermal.crt= [HW,ACPI]
5637 -1: disable all critical trip points in all thermal zones
5638 <degrees C>: override all critical trip points
5640 thermal.nocrt= [HW,ACPI]
5641 Set to disable actions on ACPI thermal zone
5642 critical and hot trip points.
5644 thermal.off= [HW,ACPI]
5645 1: disable ACPI thermal control
5647 thermal.psv= [HW,ACPI]
5648 -1: disable all passive trip points
5649 <degrees C>: override all passive trip points to this
5652 thermal.tzp= [HW,ACPI]
5653 Specify global default ACPI thermal zone polling rate
5654 <deci-seconds>: poll all this frequency
5655 0: no polling (default)
5658 Force threading of all interrupt handlers except those
5659 marked explicitly IRQF_NO_THREAD.
5663 Specify if the kernel should make use of the cpu
5664 topology information if the hardware supports this.
5665 The scheduler will make use of this information and
5666 e.g. base its process migration decisions on it.
5669 topology_updates= [KNL, PPC, NUMA]
5671 Specify if the kernel should ignore (off)
5672 topology updates sent by the hypervisor to this
5675 torture.disable_onoff_at_boot= [KNL]
5676 Prevent the CPU-hotplug component of torturing
5677 until after init has spawned.
5679 torture.ftrace_dump_at_shutdown= [KNL]
5680 Dump the ftrace buffer at torture-test shutdown,
5681 even if there were no errors. This can be a
5682 very costly operation when many torture tests
5683 are running concurrently, especially on systems
5684 with rotating-rust storage.
5686 torture.verbose_sleep_frequency= [KNL]
5687 Specifies how many verbose printk()s should be
5688 emitted between each sleep. The default of zero
5689 disables verbose-printk() sleeping.
5691 torture.verbose_sleep_duration= [KNL]
5692 Duration of each verbose-printk() sleep in jiffies.
5696 tpm_suspend_pcr=[HW,TPM]
5697 Format: integer pcr id
5698 Specify that at suspend time, the tpm driver
5699 should extend the specified pcr with zeros,
5700 as a workaround for some chips which fail to
5701 flush the last written pcr on TPM_SaveState.
5702 This will guarantee that all the other pcrs
5705 trace_buf_size=nn[KMG]
5706 [FTRACE] will set tracing buffer size on each cpu.
5708 trace_event=[event-list]
5709 [FTRACE] Set and start specified trace events in order
5710 to facilitate early boot debugging. The event-list is a
5711 comma-separated list of trace events to enable. See
5712 also Documentation/trace/events.rst
5714 trace_options=[option-list]
5715 [FTRACE] Enable or disable tracer options at boot.
5716 The option-list is a comma delimited list of options
5717 that can be enabled or disabled just as if you were
5718 to echo the option name into
5720 /sys/kernel/debug/tracing/trace_options
5722 For example, to enable stacktrace option (to dump the
5723 stack trace of each event), add to the command line:
5725 trace_options=stacktrace
5727 See also Documentation/trace/ftrace.rst "trace options"
5731 Have the tracepoints sent to printk as well as the
5732 tracing ring buffer. This is useful for early boot up
5733 where the system hangs or reboots and does not give the
5734 option for reading the tracing buffer or performing a
5735 ftrace_dump_on_oops.
5737 To turn off having tracepoints sent to printk,
5738 echo 0 > /proc/sys/kernel/tracepoint_printk
5739 Note, echoing 1 into this file without the
5740 tracepoint_printk kernel cmdline option has no effect.
5742 The tp_printk_stop_on_boot (see below) can also be used
5743 to stop the printing of events to console at
5748 Having tracepoints sent to printk() and activating high
5749 frequency tracepoints such as irq or sched, can cause
5750 the system to live lock.
5752 tp_printk_stop_on_boot[FTRACE]
5753 When tp_printk (above) is set, it can cause a lot of noise
5754 on the console. It may be useful to only include the
5755 printing of events during boot up, as user space may
5756 make the system inoperable.
5758 This command line option will stop the printing of events
5759 to console at the late_initcall_sync() time frame.
5762 [FTRACE] enable this option to disable tracing when a
5763 warning is hit. This turns off "tracing_on". Tracing can
5764 be enabled again by echoing '1' into the "tracing_on"
5765 file located in /sys/kernel/debug/tracing/
5767 This option is useful, as it disables the trace before
5768 the WARNING dump is called, which prevents the trace to
5769 be filled with content caused by the warning output.
5771 This option can also be set at run time via the sysctl
5772 option: kernel/traceoff_on_warning
5774 transparent_hugepage=
5776 Format: [always|madvise|never]
5777 Can be used to control the default behavior of the system
5778 with respect to transparent hugepages.
5779 See Documentation/admin-guide/mm/transhuge.rst
5782 trusted.source= [KEYS]
5784 This parameter identifies the trust source as a backend
5785 for trusted keys implementation. Supported trust
5789 If not specified then it defaults to iterating through
5790 the trust source list starting with TPM and assigns the
5791 first trust source as a backend which is initialized
5792 successfully during iteration.
5794 tsc= Disable clocksource stability checks for TSC.
5796 [x86] reliable: mark tsc clocksource as reliable, this
5797 disables clocksource verification at runtime, as well
5798 as the stability checks done at bootup. Used to enable
5799 high-resolution timer mode on older hardware, and in
5800 virtualized environment.
5801 [x86] noirqtime: Do not use TSC to do irq accounting.
5802 Used to run time disable IRQ_TIME_ACCOUNTING on any
5803 platforms where RDTSC is slow and this accounting
5805 [x86] unstable: mark the TSC clocksource as unstable, this
5806 marks the TSC unconditionally unstable at bootup and
5807 avoids any further wobbles once the TSC watchdog notices.
5808 [x86] nowatchdog: disable clocksource watchdog. Used
5809 in situations with strict latency requirements (where
5810 interruptions from clocksource watchdog are not
5813 tsc_early_khz= [X86] Skip early TSC calibration and use the given
5814 value instead. Useful when the early TSC frequency discovery
5815 procedure is not reliable, such as on overclocked systems
5816 with CPUID.16h support and partial CPUID.15h support.
5817 Format: <unsigned int>
5819 tsx= [X86] Control Transactional Synchronization
5820 Extensions (TSX) feature in Intel processors that
5821 support TSX control.
5823 This parameter controls the TSX feature. The options are:
5825 on - Enable TSX on the system. Although there are
5826 mitigations for all known security vulnerabilities,
5827 TSX has been known to be an accelerator for
5828 several previous speculation-related CVEs, and
5829 so there may be unknown security risks associated
5830 with leaving it enabled.
5832 off - Disable TSX on the system. (Note that this
5833 option takes effect only on newer CPUs which are
5834 not vulnerable to MDS, i.e., have
5835 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
5836 the new IA32_TSX_CTRL MSR through a microcode
5837 update. This new MSR allows for the reliable
5838 deactivation of the TSX functionality.)
5840 auto - Disable TSX if X86_BUG_TAA is present,
5841 otherwise enable TSX on the system.
5843 Not specifying this option is equivalent to tsx=off.
5845 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5848 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
5849 Abort (TAA) vulnerability.
5851 Similar to Micro-architectural Data Sampling (MDS)
5852 certain CPUs that support Transactional
5853 Synchronization Extensions (TSX) are vulnerable to an
5854 exploit against CPU internal buffers which can forward
5855 information to a disclosure gadget under certain
5858 In vulnerable processors, the speculatively forwarded
5859 data can be used in a cache side channel attack, to
5860 access data to which the attacker does not have direct
5863 This parameter controls the TAA mitigation. The
5866 full - Enable TAA mitigation on vulnerable CPUs
5869 full,nosmt - Enable TAA mitigation and disable SMT on
5870 vulnerable CPUs. If TSX is disabled, SMT
5871 is not disabled because CPU is not
5872 vulnerable to cross-thread TAA attacks.
5873 off - Unconditionally disable TAA mitigation
5875 On MDS-affected machines, tsx_async_abort=off can be
5876 prevented by an active MDS mitigation as both vulnerabilities
5877 are mitigated with the same mechanism so in order to disable
5878 this mitigation, you need to specify mds=off too.
5880 Not specifying this option is equivalent to
5881 tsx_async_abort=full. On CPUs which are MDS affected
5882 and deploy MDS mitigation, TAA mitigation is not
5883 required and doesn't provide any additional
5887 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5889 turbografx.map[2|3]= [HW,JOY]
5890 TurboGraFX parallel port interface
5892 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
5893 See also Documentation/input/devices/joystick-parport.rst
5895 udbg-immortal [PPC] When debugging early kernel crashes that
5896 happen after console_init() and before a proper
5897 console driver takes over, this boot options might
5898 help "seeing" what's going on.
5900 uhash_entries= [KNL,NET]
5901 Set number of hash buckets for UDP/UDP-Lite connections
5904 [USB] Ignore overcurrent events (default N).
5905 Some badly-designed motherboards generate lots of
5906 bogus events, for ports that aren't wired to
5907 anything. Set this parameter to avoid log spamming.
5908 Note that genuine overcurrent events won't be
5912 [X86] Cause panic on unknown NMI.
5914 usbcore.authorized_default=
5915 [USB] Default USB device authorization:
5916 (default -1 = authorized except for wireless USB,
5917 0 = not authorized, 1 = authorized, 2 = authorized
5918 if device connected to internal port)
5920 usbcore.autosuspend=
5921 [USB] The autosuspend time delay (in seconds) used
5922 for newly-detected USB devices (default 2). This
5923 is the time required before an idle device will be
5924 autosuspended. Devices for which the delay is set
5925 to a negative value won't be autosuspended at all.
5927 usbcore.usbfs_snoop=
5928 [USB] Set to log all usbfs traffic (default 0 = off).
5930 usbcore.usbfs_snoop_max=
5931 [USB] Maximum number of bytes to snoop in each URB
5934 usbcore.blinkenlights=
5935 [USB] Set to cycle leds on hubs (default 0 = off).
5937 usbcore.old_scheme_first=
5938 [USB] Start with the old device initialization
5939 scheme (default 0 = off).
5941 usbcore.usbfs_memory_mb=
5942 [USB] Memory limit (in MB) for buffers allocated by
5943 usbfs (default = 16, 0 = max = 2047).
5945 usbcore.use_both_schemes=
5946 [USB] Try the other device initialization scheme
5947 if the first one fails (default 1 = enabled).
5949 usbcore.initial_descriptor_timeout=
5950 [USB] Specifies timeout for the initial 64-byte
5951 USB_REQ_GET_DESCRIPTOR request in milliseconds
5952 (default 5000 = 5.0 seconds).
5954 usbcore.nousb [USB] Disable the USB subsystem
5957 [USB] A list of quirk entries to augment the built-in
5958 usb core quirk list. List entries are separated by
5959 commas. Each entry has the form
5960 VendorID:ProductID:Flags. The IDs are 4-digit hex
5961 numbers and Flags is a set of letters. Each letter
5962 will change the built-in quirk; setting it if it is
5963 clear and clearing it if it is set. The letters have
5964 the following meanings:
5965 a = USB_QUIRK_STRING_FETCH_255 (string
5966 descriptors must not be fetched using
5968 b = USB_QUIRK_RESET_RESUME (device can't resume
5969 correctly so reset it instead);
5970 c = USB_QUIRK_NO_SET_INTF (device can't handle
5971 Set-Interface requests);
5972 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
5973 handle its Configuration or Interface
5975 e = USB_QUIRK_RESET (device can't be reset
5976 (e.g morph devices), don't use reset);
5977 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
5978 more interface descriptions than the
5979 bNumInterfaces count, and can't handle
5980 talking to these interfaces);
5981 g = USB_QUIRK_DELAY_INIT (device needs a pause
5982 during initialization, after we read
5983 the device descriptor);
5984 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
5985 high speed and super speed interrupt
5986 endpoints, the USB 2.0 and USB 3.0 spec
5987 require the interval in microframes (1
5988 microframe = 125 microseconds) to be
5989 calculated as interval = 2 ^
5991 Devices with this quirk report their
5992 bInterval as the result of this
5993 calculation instead of the exponent
5994 variable used in the calculation);
5995 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
5996 handle device_qualifier descriptor
5998 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
5999 generates spurious wakeup, ignore
6000 remote wakeup capability);
6001 k = USB_QUIRK_NO_LPM (device can't handle Link
6003 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
6004 (Device reports its bInterval as linear
6005 frames instead of the USB 2.0
6007 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
6008 to be disconnected before suspend to
6009 prevent spurious wakeup);
6010 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
6011 pause after every control message);
6012 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
6013 delay after resetting its port);
6014 Example: quirks=0781:5580:bk,0a5c:5834:gij
6017 [USBHID] The interval which mice are to be polled at.
6020 [USBHID] The interval which joysticks are to be polled at.
6023 [USBHID] The interval which keyboards are to be polled at.
6025 usb-storage.delay_use=
6026 [UMS] The delay in seconds before a new device is
6027 scanned for Logical Units (default 1).
6030 [UMS] A list of quirks entries to supplement or
6031 override the built-in unusual_devs list. List
6032 entries are separated by commas. Each entry has
6033 the form VID:PID:Flags where VID and PID are Vendor
6034 and Product ID values (4-digit hex numbers) and
6035 Flags is a set of characters, each corresponding
6036 to a common usb-storage quirk flag as follows:
6037 a = SANE_SENSE (collect more than 18 bytes
6038 of sense data, not on uas);
6039 b = BAD_SENSE (don't collect more than 18
6040 bytes of sense data, not on uas);
6041 c = FIX_CAPACITY (decrease the reported
6042 device capacity by one sector);
6043 d = NO_READ_DISC_INFO (don't use
6044 READ_DISC_INFO command, not on uas);
6045 e = NO_READ_CAPACITY_16 (don't use
6046 READ_CAPACITY_16 command);
6047 f = NO_REPORT_OPCODES (don't use report opcodes
6049 g = MAX_SECTORS_240 (don't transfer more than
6050 240 sectors at a time, uas only);
6051 h = CAPACITY_HEURISTICS (decrease the
6052 reported device capacity by one
6053 sector if the number is odd);
6054 i = IGNORE_DEVICE (don't bind to this
6056 j = NO_REPORT_LUNS (don't use report luns
6058 k = NO_SAME (do not use WRITE_SAME, uas only)
6059 l = NOT_LOCKABLE (don't try to lock and
6060 unlock ejectable media, not on uas);
6061 m = MAX_SECTORS_64 (don't transfer more
6062 than 64 sectors = 32 KB at a time,
6064 n = INITIAL_READ10 (force a retry of the
6065 initial READ(10) command, not on uas);
6066 o = CAPACITY_OK (accept the capacity
6067 reported by the device, not on uas);
6068 p = WRITE_CACHE (the device cache is ON
6069 by default, not on uas);
6070 r = IGNORE_RESIDUE (the device reports
6071 bogus residue values, not on uas);
6072 s = SINGLE_LUN (the device has only one
6074 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
6075 commands, uas only);
6076 u = IGNORE_UAS (don't bind to the uas driver);
6077 w = NO_WP_DETECT (don't test whether the
6078 medium is write-protected).
6079 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
6080 even if the device claims no cache,
6082 Example: quirks=0419:aaf5:rl,0421:0433:rc
6084 user_debug= [KNL,ARM]
6086 See arch/arm/Kconfig.debug help text.
6087 1 - undefined instruction events
6089 4 - invalid data aborts
6092 Example: user_debug=31
6095 [X86] Flags controlling user PTE allocations.
6097 nohigh = do not allocate PTE pages in
6098 HIGHMEM regardless of setting
6102 On X86_32, this is an alias for vdso32=. Otherwise:
6104 vdso=1: enable VDSO (the default)
6105 vdso=0: disable VDSO mapping
6107 vdso32= [X86] Control the 32-bit vDSO
6108 vdso32=1: enable 32-bit VDSO
6109 vdso32=0 or vdso32=2: disable 32-bit VDSO
6111 See the help text for CONFIG_COMPAT_VDSO for more
6112 details. If CONFIG_COMPAT_VDSO is set, the default is
6113 vdso32=0; otherwise, the default is vdso32=1.
6115 For compatibility with older kernels, vdso32=2 is an
6118 Try vdso32=0 if you encounter an error that says:
6119 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
6122 vector=percpu: enable percpu vector domain
6124 video= [FB] Frame buffer configuration
6125 See Documentation/fb/modedb.rst.
6127 video.brightness_switch_enabled= [0,1]
6128 If set to 1, on receiving an ACPI notify event
6129 generated by hotkey, video driver will adjust brightness
6130 level and then send out the event to user space through
6131 the allocated input device; If set to 0, video driver
6132 will only send out the event without touching backlight
6137 [VMMIO] Memory mapped virtio (platform) device.
6139 <size>@<baseaddr>:<irq>[:<id>]
6141 <size> := size (can use standard suffixes
6143 <baseaddr> := physical base address
6144 <irq> := interrupt number (as passed to
6146 <id> := (optional) platform device id
6148 virtio_mmio.device=1K@0x100b0000:48:7
6150 Can be used multiple times for multiple devices.
6152 vga= [BOOT,X86-32] Select a particular video mode
6153 See Documentation/x86/boot.rst and
6154 Documentation/admin-guide/svga.rst.
6155 Use vga=ask for menu.
6156 This is actually a boot loader parameter; the value is
6157 passed to the kernel using a special protocol.
6159 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
6160 May slow down system boot speed, especially when
6161 enabled on systems with a large amount of memory.
6162 All options are enabled by default, and this
6163 interface is meant to allow for selectively
6164 enabling or disabling specific virtual memory
6167 Available options are:
6168 P Enable page structure init time poisoning
6169 - Disable all of the above options
6171 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
6172 size of <nn>. This can be used to increase the
6173 minimum size (128MB on x86). It can also be used to
6174 decrease the size and leave more room for directly
6177 vmcp_cma=nn[MG] [KNL,S390]
6178 Sets the memory size reserved for contiguous memory
6179 allocations for the vmcp device driver.
6181 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
6184 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
6187 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
6191 Controls the behavior of vsyscalls (i.e. calls to
6192 fixed addresses of 0xffffffffff600x00 from legacy
6193 code). Most statically-linked binaries and older
6194 versions of glibc use these calls. Because these
6195 functions are at fixed addresses, they make nice
6196 targets for exploits that can control RIP.
6198 emulate [default] Vsyscalls turn into traps and are
6199 emulated reasonably safely. The vsyscall
6202 xonly Vsyscalls turn into traps and are
6203 emulated reasonably safely. The vsyscall
6204 page is not readable.
6206 none Vsyscalls don't work at all. This makes
6207 them quite hard to use for exploits but
6208 might break your system.
6210 vt.color= [VT] Default text color.
6211 Format: 0xYX, X = foreground, Y = background.
6212 Default: 0x07 = light gray on black.
6214 vt.cur_default= [VT] Default cursor shape.
6215 Format: 0xCCBBAA, where AA, BB, and CC are the same as
6216 the parameters of the <Esc>[?A;B;Cc escape sequence;
6217 see VGA-softcursor.txt. Default: 2 = underline.
6219 vt.default_blu= [VT]
6220 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
6221 Change the default blue palette of the console.
6222 This is a 16-member array composed of values
6225 vt.default_grn= [VT]
6226 Format: <green0>,<green1>,<green2>,...,<green15>
6227 Change the default green palette of the console.
6228 This is a 16-member array composed of values
6231 vt.default_red= [VT]
6232 Format: <red0>,<red1>,<red2>,...,<red15>
6233 Change the default red palette of the console.
6234 This is a 16-member array composed of values
6240 Set system-wide default UTF-8 mode for all tty's.
6241 Default is 1, i.e. UTF-8 mode is enabled for all
6242 newly opened terminals.
6244 vt.global_cursor_default=
6247 Set system-wide default for whether a cursor
6248 is shown on new VTs. Default is -1,
6249 i.e. cursors will be created by default unless
6250 overridden by individual drivers. 0 will hide
6251 cursors, 1 will display them.
6253 vt.italic= [VT] Default color for italic text; 0-15.
6256 vt.underline= [VT] Default color for underlined text; 0-15.
6259 watchdog timers [HW,WDT] For information on watchdog timers,
6260 see Documentation/watchdog/watchdog-parameters.rst
6261 or other driver-specific files in the
6262 Documentation/watchdog/ directory.
6266 Set the hard lockup detector stall duration
6267 threshold in seconds. The soft lockup detector
6268 threshold is set to twice the value. A value of 0
6269 disables both lockup detectors. Default is 10
6272 workqueue.watchdog_thresh=
6273 If CONFIG_WQ_WATCHDOG is configured, workqueue can
6274 warn stall conditions and dump internal state to
6275 help debugging. 0 disables workqueue stall
6276 detection; otherwise, it's the stall threshold
6277 duration in seconds. The default value is 30 and
6278 it can be updated at runtime by writing to the
6279 corresponding sysfs file.
6281 workqueue.disable_numa
6282 By default, all work items queued to unbound
6283 workqueues are affine to the NUMA nodes they're
6284 issued on, which results in better behavior in
6285 general. If NUMA affinity needs to be disabled for
6286 whatever reason, this option can be used. Note
6287 that this also can be controlled per-workqueue for
6288 workqueues visible under /sys/bus/workqueue/.
6290 workqueue.power_efficient
6291 Per-cpu workqueues are generally preferred because
6292 they show better performance thanks to cache
6293 locality; unfortunately, per-cpu workqueues tend to
6294 be more power hungry than unbound workqueues.
6296 Enabling this makes the per-cpu workqueues which
6297 were observed to contribute significantly to power
6298 consumption unbound, leading to measurably lower
6299 power usage at the cost of small performance
6302 The default value of this parameter is determined by
6303 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
6305 workqueue.debug_force_rr_cpu
6306 Workqueue used to implicitly guarantee that work
6307 items queued without explicit CPU specified are put
6308 on the local CPU. This guarantee is no longer true
6309 and while local CPU is still preferred work items
6310 may be put on foreign CPUs. This debug option
6311 forces round-robin CPU selection to flush out
6312 usages which depend on the now broken guarantee.
6313 When enabled, memory and cache locality will be
6316 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
6317 default x2apic cluster mode on platforms
6320 xen_512gb_limit [KNL,X86-64,XEN]
6321 Restricts the kernel running paravirtualized under Xen
6322 to use only up to 512 GB of RAM. The reason to do so is
6323 crash analysis tools and Xen tools for doing domain
6324 save/restore/migration must be enabled to handle larger
6327 xen_emul_unplug= [HW,X86,XEN]
6328 Unplug Xen emulated devices
6329 Format: [unplug0,][unplug1]
6330 ide-disks -- unplug primary master IDE devices
6331 aux-ide-disks -- unplug non-primary-master IDE devices
6332 nics -- unplug network devices
6333 all -- unplug all emulated devices (NICs and IDE disks)
6334 unnecessary -- unplugging emulated devices is
6335 unnecessary even if the host did not respond to
6337 never -- do not unplug even if version check succeeds
6339 xen_legacy_crash [X86,XEN]
6340 Crash from Xen panic notifier, without executing late
6341 panic() code such as dumping handler.
6343 xen_nopvspin [X86,XEN]
6344 Disables the qspinlock slowpath using Xen PV optimizations.
6345 This parameter is obsoleted by "nopvspin" parameter, which
6346 has equivalent effect for XEN platform.
6349 Disables the PV optimizations forcing the HVM guest to
6350 run as generic HVM guest with no PV drivers.
6351 This option is obsoleted by the "nopv" option, which
6352 has equivalent effect for XEN platform.
6354 xen_no_vector_callback
6355 [KNL,X86,XEN] Disable the vector callback for Xen
6356 event channel interrupts.
6358 xen_scrub_pages= [XEN]
6359 Boolean option to control scrubbing pages before giving them back
6360 to Xen, for use by other domains. Can be also changed at runtime
6361 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
6362 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
6364 xen_timer_slop= [X86-64,XEN]
6365 Set the timer slop (in nanoseconds) for the virtual Xen
6366 timers (default is 100000). This adjusts the minimum
6367 delta of virtualized Xen timers, where lower values
6368 improve timer resolution at the expense of processing
6369 more timer interrupts.
6371 xen.event_eoi_delay= [XEN]
6372 How long to delay EOI handling in case of event
6373 storms (jiffies). Default is 10.
6375 xen.event_loop_timeout= [XEN]
6376 After which time (jiffies) the event handling loop
6377 should start to delay EOI handling. Default is 2.
6379 xen.fifo_events= [XEN]
6380 Boolean parameter to disable using fifo event handling
6381 even if available. Normally fifo event handling is
6382 preferred over the 2-level event handling, as it is
6383 fairer and the number of possible event channels is
6384 much higher. Default is on (use fifo events).
6386 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
6387 Disables the PV optimizations forcing the guest to run
6388 as generic guest with no PV drivers. Currently support
6389 XEN HVM, KVM, HYPER_V and VMWARE guest.
6391 nopvspin [X86,XEN,KVM]
6392 Disables the qspinlock slow path using PV optimizations
6393 which allow the hypervisor to 'idle' the guest on lock
6396 xirc2ps_cs= [NET,PCMCIA]
6398 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
6401 By default on POWER9 and above, the kernel will
6402 natively use the XIVE interrupt controller. This option
6403 allows the fallback firmware mode to be used:
6405 off Fallback to firmware control of XIVE interrupt
6406 controller on both pseries and powernv
6407 platforms. Only useful on POWER9 and above.
6409 xhci-hcd.quirks [USB,KNL]
6410 A hex value specifying bitmask with supplemental xhci
6411 host controller quirks. Meaning of each bit can be
6412 consulted in header drivers/usb/host/xhci.h.
6415 Format: { early | on | rw | ro | off }
6416 Controls if xmon debugger is enabled. Default is off.
6417 Passing only "xmon" is equivalent to "xmon=early".
6418 early Call xmon as early as possible on boot; xmon
6419 debugger is called from setup_arch().
6420 on xmon debugger hooks will be installed so xmon
6421 is only called on a kernel crash. Default mode,
6422 i.e. either "ro" or "rw" mode, is controlled
6423 with CONFIG_XMON_DEFAULT_RO_MODE.
6424 rw xmon debugger hooks will be installed so xmon
6425 is called only on a kernel crash, mode is write,
6426 meaning SPR registers, memory and, other data
6427 can be written using xmon commands.
6428 ro same as "rw" option above but SPR registers,
6429 memory, and other data can't be written using
6431 off xmon is disabled.