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 output can only be used by Xen PV guests.
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 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1762 Format: { strict | legacy | 2008 | relaxed }
1765 Choose which programs will be accepted for execution
1766 based on the IEEE 754 NaN encoding(s) supported by
1767 the FPU and the NaN encoding requested with the value
1768 of an ELF file header flag individually set by each
1769 binary. Hardware implementations are permitted to
1770 support either or both of the legacy and the 2008 NaN
1773 Available settings are as follows:
1774 strict accept binaries that request a NaN encoding
1775 supported by the FPU
1776 legacy only accept legacy-NaN binaries, if supported
1778 2008 only accept 2008-NaN binaries, if supported
1780 relaxed accept any binaries regardless of whether
1781 supported by the FPU
1783 The FPU emulator is always able to support both NaN
1784 encodings, so if no FPU hardware is present or it has
1785 been disabled with 'nofpu', then the settings of
1786 'legacy' and '2008' strap the emulator accordingly,
1787 'relaxed' straps the emulator for both legacy-NaN and
1788 2008-NaN, whereas 'strict' enables legacy-NaN only on
1789 legacy processors and both NaN encodings on MIPS32 or
1792 The setting for ABS.fmt/NEG.fmt instruction execution
1793 mode generally follows that for the NaN encoding,
1794 except where unsupported by hardware.
1796 ignore_loglevel [KNL]
1797 Ignore loglevel setting - this will print /all/
1798 kernel messages to the console. Useful for debugging.
1799 We also add it as printk module parameter, so users
1800 could change it dynamically, usually by
1801 /sys/module/printk/parameters/ignore_loglevel.
1804 Ignore RLIMIT_DATA setting for data mappings,
1805 print warning at first misuse. Can be changed via
1806 /sys/module/kernel/parameters/ignore_rlimit_data.
1808 ihash_entries= [KNL]
1809 Set number of hash buckets for inode cache.
1811 ima_appraise= [IMA] appraise integrity measurements
1812 Format: { "off" | "enforce" | "fix" | "log" }
1815 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1816 The builtin appraise policy appraises all files
1819 ima_canonical_fmt [IMA]
1820 Use the canonical format for the binary runtime
1821 measurements, instead of host native format.
1824 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1828 The list of supported hash algorithms is defined
1829 in crypto/hash_info.h.
1832 The builtin policies to load during IMA setup.
1833 Format: "tcb | appraise_tcb | secure_boot |
1834 fail_securely | critical_data"
1836 The "tcb" policy measures all programs exec'd, files
1837 mmap'd for exec, and all files opened with the read
1838 mode bit set by either the effective uid (euid=0) or
1841 The "appraise_tcb" policy appraises the integrity of
1842 all files owned by root.
1844 The "secure_boot" policy appraises the integrity
1845 of files (eg. kexec kernel image, kernel modules,
1846 firmware, policy, etc) based on file signatures.
1848 The "fail_securely" policy forces file signature
1849 verification failure also on privileged mounted
1850 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1853 The "critical_data" policy measures kernel integrity
1856 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1857 Load a policy which meets the needs of the Trusted
1858 Computing Base. This means IMA will measure all
1859 programs exec'd, files mmap'd for exec, and all files
1860 opened for read by uid=0.
1863 Select one of defined IMA measurements template formats.
1864 Formats: { "ima" | "ima-ng" | "ima-sig" }
1868 [IMA] Define a custom template format.
1869 Format: { "field1|...|fieldN" }
1871 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1872 Format: <min_file_size>
1873 Set the minimal file size for using asynchronous hash.
1874 If left unspecified, ahash usage is disabled.
1876 ahash performance varies for different data sizes on
1877 different crypto accelerators. This option can be used
1878 to achieve the best performance for a particular HW.
1880 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1882 Set hashing buffer size. Default: 4k.
1884 ahash performance varies for different chunk sizes on
1885 different crypto accelerators. This option can be used
1886 to achieve best performance for particular HW.
1890 Run specified binary instead of /sbin/init as init
1893 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1894 for working out where the kernel is dying during
1897 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1898 initcall functions. Useful for debugging built-in
1899 modules and initcalls.
1901 initramfs_async= [KNL]
1904 This parameter controls whether the initramfs
1905 image is unpacked asynchronously, concurrently
1906 with devices being probed and
1907 initialized. This should normally just work,
1908 but as a debugging aid, one can get the
1909 historical behaviour of the initramfs
1910 unpacking being completed before device_ and
1913 initrd= [BOOT] Specify the location of the initial ramdisk
1915 initrdmem= [KNL] Specify a physical address and size from which to
1916 load the initrd. If an initrd is compiled in or
1917 specified in the bootparams, it takes priority over this
1919 Format: ss[KMG],nn[KMG]
1922 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
1925 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
1927 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
1929 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
1931 init_pkru= [X86] Specify the default memory protection keys rights
1932 register contents for all processes. 0x55555554 by
1933 default (disallow access to all but pkey 0). Can
1934 override in debugfs after boot.
1936 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1939 int_pln_enable [X86] Enable power limit notification interrupt
1941 integrity_audit=[IMA]
1942 Format: { "0" | "1" }
1943 0 -- basic integrity auditing messages. (Default)
1944 1 -- additional integrity auditing messages.
1946 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1948 Enable intel iommu driver.
1950 Disable intel iommu driver.
1951 igfx_off [Default Off]
1952 By default, gfx is mapped as normal device. If a gfx
1953 device has a dedicated DMAR unit, the DMAR unit is
1954 bypassed by not enabling DMAR with this option. In
1955 this case, gfx device will use physical address for
1957 strict [Default Off]
1958 Deprecated, equivalent to iommu.strict=1.
1959 sp_off [Default Off]
1960 By default, super page will be supported if Intel IOMMU
1961 has the capability. With this option, super page will
1964 Enable the Intel IOMMU scalable mode if the hardware
1965 advertises that it has support for the scalable mode
1968 Disallow use of the Intel IOMMU scalable mode.
1969 tboot_noforce [Default Off]
1970 Do not force the Intel IOMMU enabled under tboot.
1971 By default, tboot will force Intel IOMMU on, which
1972 could harm performance of some high-throughput
1973 devices like 40GBit network cards, even if identity
1975 Note that using this option lowers the security
1976 provided by tboot because it makes the system
1977 vulnerable to DMA attacks.
1979 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1980 0 disables intel_idle and fall back on acpi_idle.
1981 1 to 9 specify maximum depth of C-state.
1985 Do not enable intel_pstate as the default
1986 scaling driver for the supported processors
1988 Use intel_pstate as a scaling driver, but configure it
1989 to work with generic cpufreq governors (instead of
1990 enabling its internal governor). This mode cannot be
1991 used along with the hardware-managed P-states (HWP)
1994 Enable intel_pstate on systems that prohibit it by default
1995 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1996 instead of acpi-cpufreq may disable platform features, such
1997 as thermal controls and power capping, that rely on ACPI
1998 P-States information being indicated to OSPM and therefore
1999 should be used with caution. This option does not work with
2000 processors that aren't supported by the intel_pstate driver
2001 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2003 Do not enable hardware P state control (HWP)
2006 Only load intel_pstate on systems which support
2007 hardware P state control (HWP) if available.
2009 Enforce ACPI _PPC performance limits. If the Fixed ACPI
2010 Description Table, specifies preferred power management
2011 profile as "Enterprise Server" or "Performance Server",
2012 then this feature is turned on by default.
2014 Allow per-logical-CPU P-State performance control limits using
2015 cpufreq sysfs interface
2017 intremap= [X86-64, Intel-IOMMU]
2018 on enable Interrupt Remapping (default)
2019 off disable Interrupt Remapping
2020 nosid disable Source ID checking
2022 BIOS x2APIC opt-out request will be ignored
2023 nopost disable Interrupt Posting
2025 iomem= Disable strict checking of access to MMIO memory
2026 strict regions from userspace.
2041 nobypass [PPC/POWERNV]
2042 Disable IOMMU bypass, using IOMMU for PCI devices.
2044 iommu.forcedac= [ARM64, X86] Control IOVA allocation for PCI devices.
2045 Format: { "0" | "1" }
2046 0 - Try to allocate a 32-bit DMA address first, before
2047 falling back to the full range if needed.
2048 1 - Allocate directly from the full usable range,
2049 forcing Dual Address Cycle for PCI cards supporting
2050 greater than 32-bit addressing.
2052 iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour
2053 Format: { "0" | "1" }
2055 Request that DMA unmap operations use deferred
2056 invalidation of hardware TLBs, for increased
2057 throughput at the cost of reduced device isolation.
2058 Will fall back to strict mode if not supported by
2059 the relevant IOMMU driver.
2061 DMA unmap operations invalidate IOMMU hardware TLBs
2063 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}.
2064 Note: on x86, strict mode specified via one of the
2065 legacy driver-specific options takes precedence.
2068 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
2069 Format: { "0" | "1" }
2070 0 - Use IOMMU translation for DMA.
2071 1 - Bypass the IOMMU for DMA.
2072 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2074 io7= [HW] IO7 for Marvel-based Alpha systems
2075 See comment before marvel_specify_io7 in
2076 arch/alpha/kernel/core_marvel.c.
2078 io_delay= [X86] I/O delay method
2080 Standard port 0x80 based delay
2082 Alternate port 0xed based delay (needed on some systems)
2084 Simple two microseconds delay
2089 See Documentation/admin-guide/nfs/nfsroot.rst.
2091 ipcmni_extend [KNL] Extend the maximum number of unique System V
2092 IPC identifiers from 32,768 to 16,777,216.
2094 irqaffinity= [SMP] Set the default irq affinity mask
2095 The argument is a cpu list, as described above.
2097 irqchip.gicv2_force_probe=
2100 Force the kernel to look for the second 4kB page
2101 of a GICv2 controller even if the memory range
2102 exposed by the device tree is too small.
2104 irqchip.gicv3_nolpi=
2106 Force the kernel to ignore the availability of
2107 LPIs (and by consequence ITSs). Intended for system
2108 that use the kernel as a bootloader, and thus want
2109 to let secondary kernels in charge of setting up
2112 irqchip.gicv3_pseudo_nmi= [ARM64]
2113 Enables support for pseudo-NMIs in the kernel. This
2114 requires the kernel to be built with
2115 CONFIG_ARM64_PSEUDO_NMI.
2118 When an interrupt is not handled search all handlers
2119 for it. Intended to get systems with badly broken
2123 When an interrupt is not handled search all handlers
2124 for it. Also check all handlers each timer
2125 interrupt. Intended to get systems with badly broken
2129 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2131 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2132 [Deprecated - use cpusets instead]
2133 Format: [flag-list,]<cpu-list>
2135 Specify one or more CPUs to isolate from disturbances
2136 specified in the flag list (default: domain):
2139 Disable the tick when a single task runs.
2141 A residual 1Hz tick is offloaded to workqueues, which you
2142 need to affine to housekeeping through the global
2143 workqueue's affinity configured via the
2144 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2145 by using the 'domain' flag described below.
2147 NOTE: by default the global workqueue runs on all CPUs,
2148 so to protect individual CPUs the 'cpumask' file has to
2149 be configured manually after bootup.
2152 Isolate from the general SMP balancing and scheduling
2153 algorithms. Note that performing domain isolation this way
2154 is irreversible: it's not possible to bring back a CPU to
2155 the domains once isolated through isolcpus. It's strongly
2156 advised to use cpusets instead to disable scheduler load
2157 balancing through the "cpuset.sched_load_balance" file.
2158 It offers a much more flexible interface where CPUs can
2159 move in and out of an isolated set anytime.
2161 You can move a process onto or off an "isolated" CPU via
2162 the CPU affinity syscalls or cpuset.
2163 <cpu number> begins at 0 and the maximum value is
2164 "number of CPUs in system - 1".
2168 Isolate from being targeted by managed interrupts
2169 which have an interrupt mask containing isolated
2170 CPUs. The affinity of managed interrupts is
2171 handled by the kernel and cannot be changed via
2172 the /proc/irq/* interfaces.
2174 This isolation is best effort and only effective
2175 if the automatically assigned interrupt mask of a
2176 device queue contains isolated and housekeeping
2177 CPUs. If housekeeping CPUs are online then such
2178 interrupts are directed to the housekeeping CPU
2179 so that IO submitted on the housekeeping CPU
2180 cannot disturb the isolated CPU.
2182 If a queue's affinity mask contains only isolated
2183 CPUs then this parameter has no effect on the
2184 interrupt routing decision, though interrupts are
2185 only delivered when tasks running on those
2186 isolated CPUs submit IO. IO submitted on
2187 housekeeping CPUs has no influence on those
2190 The format of <cpu-list> is described above.
2194 ivrs_ioapic [HW,X86-64]
2195 Provide an override to the IOAPIC-ID<->DEVICE-ID
2196 mapping provided in the IVRS ACPI table. For
2197 example, to map IOAPIC-ID decimal 10 to
2198 PCI device 00:14.0 write the parameter as:
2199 ivrs_ioapic[10]=00:14.0
2201 ivrs_hpet [HW,X86-64]
2202 Provide an override to the HPET-ID<->DEVICE-ID
2203 mapping provided in the IVRS ACPI table. For
2204 example, to map HPET-ID decimal 0 to
2205 PCI device 00:14.0 write the parameter as:
2206 ivrs_hpet[0]=00:14.0
2208 ivrs_acpihid [HW,X86-64]
2209 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2210 mapping provided in the IVRS ACPI table. For
2211 example, to map UART-HID:UID AMD0020:0 to
2212 PCI device 00:14.5 write the parameter as:
2213 ivrs_acpihid[00:14.5]=AMD0020:0
2215 js= [HW,JOY] Analog joystick
2216 See Documentation/input/joydev/joystick.rst.
2219 When CONFIG_RANDOMIZE_BASE is set, this disables
2220 kernel and module base offset ASLR (Address Space
2221 Layout Randomization).
2224 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2225 report on every invalid memory access. Without this
2226 parameter KASAN will print report only for the first
2231 kernelcore= [KNL,X86,IA-64,PPC]
2232 Format: nn[KMGTPE] | nn% | "mirror"
2233 This parameter specifies the amount of memory usable by
2234 the kernel for non-movable allocations. The requested
2235 amount is spread evenly throughout all nodes in the
2236 system as ZONE_NORMAL. The remaining memory is used for
2237 movable memory in its own zone, ZONE_MOVABLE. In the
2238 event, a node is too small to have both ZONE_NORMAL and
2239 ZONE_MOVABLE, kernelcore memory will take priority and
2240 other nodes will have a larger ZONE_MOVABLE.
2242 ZONE_MOVABLE is used for the allocation of pages that
2243 may be reclaimed or moved by the page migration
2244 subsystem. Note that allocations like PTEs-from-HighMem
2245 still use the HighMem zone if it exists, and the Normal
2246 zone if it does not.
2248 It is possible to specify the exact amount of memory in
2249 the form of "nn[KMGTPE]", a percentage of total system
2250 memory in the form of "nn%", or "mirror". If "mirror"
2251 option is specified, mirrored (reliable) memory is used
2252 for non-movable allocations and remaining memory is used
2253 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2254 are exclusive, so you cannot specify multiple forms.
2256 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2257 Format: <Controller#>[,poll interval]
2258 The controller # is the number of the ehci usb debug
2259 port as it is probed via PCI. The poll interval is
2260 optional and is the number seconds in between
2261 each poll cycle to the debug port in case you need
2262 the functionality for interrupting the kernel with
2263 gdb or control-c on the dbgp connection. When
2264 not using this parameter you use sysrq-g to break into
2265 the kernel debugger.
2267 kgdboc= [KGDB,HW] kgdb over consoles.
2268 Requires a tty driver that supports console polling,
2269 or a supported polling keyboard driver (non-usb).
2270 Serial only format: <serial_device>[,baud]
2271 keyboard only format: kbd
2272 keyboard and serial format: kbd,<serial_device>[,baud]
2273 Optional Kernel mode setting:
2274 kms, kbd format: kms,kbd
2275 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2277 kgdboc_earlycon= [KGDB,HW]
2278 If the boot console provides the ability to read
2279 characters and can work in polling mode, you can use
2280 this parameter to tell kgdb to use it as a backend
2281 until the normal console is registered. Intended to
2282 be used together with the kgdboc parameter which
2283 specifies the normal console to transition to.
2285 The name of the early console should be specified
2286 as the value of this parameter. Note that the name of
2287 the early console might be different than the tty
2288 name passed to kgdboc. It's OK to leave the value
2289 blank and the first boot console that implements
2290 read() will be picked.
2292 kgdbwait [KGDB] Stop kernel execution and enter the
2293 kernel debugger at the earliest opportunity.
2295 kmac= [MIPS] Korina ethernet MAC address.
2296 Configure the RouterBoard 532 series on-chip
2297 Ethernet adapter MAC address.
2299 kmemleak= [KNL] Boot-time kmemleak enable/disable
2300 Valid arguments: on, off
2302 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2305 kprobe_event=[probe-list]
2306 [FTRACE] Add kprobe events and enable at boot time.
2307 The probe-list is a semicolon delimited list of probe
2308 definitions. Each definition is same as kprobe_events
2309 interface, but the parameters are comma delimited.
2310 For example, to add a kprobe event on vfs_read with
2311 arg1 and arg2, add to the command line;
2313 kprobe_event=p,vfs_read,$arg1,$arg2
2315 See also Documentation/trace/kprobetrace.rst "Kernel
2316 Boot Parameter" section.
2318 kpti= [ARM64] Control page table isolation of user
2319 and kernel address spaces.
2320 Default: enabled on cores which need mitigation.
2324 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2325 Default is 0 (don't ignore, but inject #GP)
2327 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2328 Default is false (don't support).
2330 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
2335 [KVM] Controls the software workaround for the
2336 X86_BUG_ITLB_MULTIHIT bug.
2337 force : Always deploy workaround.
2338 off : Never deploy workaround.
2339 auto : Deploy workaround based on the presence of
2340 X86_BUG_ITLB_MULTIHIT.
2344 If the software workaround is enabled for the host,
2345 guests do need not to enable it for nested guests.
2347 kvm.nx_huge_pages_recovery_ratio=
2348 [KVM] Controls how many 4KiB pages are periodically zapped
2349 back to huge pages. 0 disables the recovery, otherwise if
2350 the value is N KVM will zap 1/Nth of the 4KiB pages every
2351 minute. The default is 60.
2353 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2354 Default is 1 (enabled)
2356 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2358 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2361 [KVM,ARM] Select one of KVM/arm64's modes of operation.
2363 nvhe: Standard nVHE-based mode, without support for
2366 protected: nVHE-based mode with support for guests whose
2367 state is kept private from the host.
2368 Not valid if the kernel is running in EL2.
2370 Defaults to VHE/nVHE based on hardware support.
2372 kvm-arm.vgic_v3_group0_trap=
2373 [KVM,ARM] Trap guest accesses to GICv3 group-0
2376 kvm-arm.vgic_v3_group1_trap=
2377 [KVM,ARM] Trap guest accesses to GICv3 group-1
2380 kvm-arm.vgic_v3_common_trap=
2381 [KVM,ARM] Trap guest accesses to GICv3 common
2384 kvm-arm.vgic_v4_enable=
2385 [KVM,ARM] Allow use of GICv4 for direct injection of
2388 kvm_cma_resv_ratio=n [PPC]
2389 Reserves given percentage from system memory area for
2390 contiguous memory allocation for KVM hash pagetable
2392 By default it reserves 5% of total system memory.
2396 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2397 (virtualized MMU) support on capable Intel chips.
2398 Default is 1 (enabled)
2400 kvm-intel.emulate_invalid_guest_state=
2401 [KVM,Intel] Enable emulation of invalid guest states
2402 Default is 0 (disabled)
2404 kvm-intel.flexpriority=
2405 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2406 Default is 1 (enabled)
2409 [KVM,Intel] Enable VMX nesting (nVMX).
2410 Default is 0 (disabled)
2412 kvm-intel.unrestricted_guest=
2413 [KVM,Intel] Disable unrestricted guest feature
2414 (virtualized real and unpaged mode) on capable
2415 Intel chips. Default is 1 (enabled)
2417 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2420 Valid arguments: never, cond, always
2422 always: L1D cache flush on every VMENTER.
2423 cond: Flush L1D on VMENTER only when the code between
2424 VMEXIT and VMENTER can leak host memory.
2425 never: Disables the mitigation
2427 Default is cond (do L1 cache flush in specific instances)
2429 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2430 feature (tagged TLBs) on capable Intel chips.
2431 Default is 1 (enabled)
2433 l1d_flush= [X86,INTEL]
2434 Control mitigation for L1D based snooping vulnerability.
2436 Certain CPUs are vulnerable to an exploit against CPU
2437 internal buffers which can forward information to a
2438 disclosure gadget under certain conditions.
2440 In vulnerable processors, the speculatively
2441 forwarded data can be used in a cache side channel
2442 attack, to access data to which the attacker does
2443 not have direct access.
2445 This parameter controls the mitigation. The
2448 on - enable the interface for the mitigation
2450 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2453 The kernel PTE inversion protection is unconditionally
2454 enabled and cannot be disabled.
2457 Provides all available mitigations for the
2458 L1TF vulnerability. Disables SMT and
2459 enables all mitigations in the
2460 hypervisors, i.e. unconditional L1D flush.
2462 SMT control and L1D flush control via the
2463 sysfs interface is still possible after
2464 boot. Hypervisors will issue a warning
2465 when the first VM is started in a
2466 potentially insecure configuration,
2467 i.e. SMT enabled or L1D flush disabled.
2470 Same as 'full', but disables SMT and L1D
2471 flush runtime control. Implies the
2472 'nosmt=force' command line option.
2473 (i.e. sysfs control of SMT is disabled.)
2476 Leaves SMT enabled and enables the default
2477 hypervisor mitigation, i.e. conditional
2480 SMT control and L1D flush control via the
2481 sysfs interface is still possible after
2482 boot. Hypervisors will issue a warning
2483 when the first VM is started in a
2484 potentially insecure configuration,
2485 i.e. SMT enabled or L1D flush disabled.
2489 Disables SMT and enables the default
2490 hypervisor mitigation.
2492 SMT control and L1D flush control via the
2493 sysfs interface is still possible after
2494 boot. Hypervisors will issue a warning
2495 when the first VM is started in a
2496 potentially insecure configuration,
2497 i.e. SMT enabled or L1D flush disabled.
2500 Same as 'flush', but hypervisors will not
2501 warn when a VM is started in a potentially
2502 insecure configuration.
2505 Disables hypervisor mitigations and doesn't
2507 It also drops the swap size and available
2508 RAM limit restriction on both hypervisor and
2513 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2519 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2522 lapic= [X86,APIC] Do not use TSC deadline
2523 value for LAPIC timer one-shot implementation. Default
2524 back to the programmable timer unit in the LAPIC.
2525 Format: notscdeadline
2527 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2530 libata.dma= [LIBATA] DMA control
2531 libata.dma=0 Disable all PATA and SATA DMA
2532 libata.dma=1 PATA and SATA Disk DMA only
2533 libata.dma=2 ATAPI (CDROM) DMA only
2534 libata.dma=4 Compact Flash DMA only
2535 Combinations also work, so libata.dma=3 enables DMA
2536 for disks and CDROMs, but not CFs.
2538 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2539 libata.ignore_hpa=0 keep BIOS limits (default)
2540 libata.ignore_hpa=1 ignore limits, using full disk
2542 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2546 libata.force= [LIBATA] Force configurations. The format is comma-
2547 separated list of "[ID:]VAL" where ID is
2548 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2549 matching port, link or device. Basically, it matches
2550 the ATA ID string printed on console by libata. If
2551 the whole ID part is omitted, the last PORT and DEVICE
2552 values are used. If ID hasn't been specified yet, the
2553 configuration applies to all ports, links and devices.
2555 If only DEVICE is omitted, the parameter applies to
2556 the port and all links and devices behind it. DEVICE
2557 number of 0 either selects the first device or the
2558 first fan-out link behind PMP device. It does not
2559 select the host link. DEVICE number of 15 selects the
2560 host link and device attached to it.
2562 The VAL specifies the configuration to force. As long
2563 as there's no ambiguity shortcut notation is allowed.
2564 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2565 The following configurations can be forced.
2567 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2568 Any ID with matching PORT is used.
2570 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2572 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2573 udma[/][16,25,33,44,66,100,133] notation is also
2576 * [no]ncq: Turn on or off NCQ.
2578 * [no]ncqtrim: Turn off queued DSM TRIM.
2580 * nohrst, nosrst, norst: suppress hard, soft
2583 * rstonce: only attempt one reset during
2584 hot-unplug link recovery
2586 * dump_id: dump IDENTIFY data.
2588 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2590 * disable: Disable this device.
2592 If there are multiple matching configurations changing
2593 the same attribute, the last one is used.
2595 memblock=debug [KNL] Enable memblock debug messages.
2597 load_ramdisk= [RAM] [Deprecated]
2599 lockd.nlm_grace_period=P [NFS] Assign grace period.
2602 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2605 lockd.nlm_timeout=T [NFS] Assign timeout value.
2608 lockd.nlm_udpport=M [NFS] Assign UDP port.
2611 lockdown= [SECURITY]
2612 { integrity | confidentiality }
2613 Enable the kernel lockdown feature. If set to
2614 integrity, kernel features that allow userland to
2615 modify the running kernel are disabled. If set to
2616 confidentiality, kernel features that allow userland
2617 to extract confidential information from the kernel
2620 locktorture.nreaders_stress= [KNL]
2621 Set the number of locking read-acquisition kthreads.
2622 Defaults to being automatically set based on the
2623 number of online CPUs.
2625 locktorture.nwriters_stress= [KNL]
2626 Set the number of locking write-acquisition kthreads.
2628 locktorture.onoff_holdoff= [KNL]
2629 Set time (s) after boot for CPU-hotplug testing.
2631 locktorture.onoff_interval= [KNL]
2632 Set time (s) between CPU-hotplug operations, or
2633 zero to disable CPU-hotplug testing.
2635 locktorture.shuffle_interval= [KNL]
2636 Set task-shuffle interval (jiffies). Shuffling
2637 tasks allows some CPUs to go into dyntick-idle
2638 mode during the locktorture test.
2640 locktorture.shutdown_secs= [KNL]
2641 Set time (s) after boot system shutdown. This
2642 is useful for hands-off automated testing.
2644 locktorture.stat_interval= [KNL]
2645 Time (s) between statistics printk()s.
2647 locktorture.stutter= [KNL]
2648 Time (s) to stutter testing, for example,
2649 specifying five seconds causes the test to run for
2650 five seconds, wait for five seconds, and so on.
2651 This tests the locking primitive's ability to
2652 transition abruptly to and from idle.
2654 locktorture.torture_type= [KNL]
2655 Specify the locking implementation to test.
2657 locktorture.verbose= [KNL]
2658 Enable additional printk() statements.
2660 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2663 loglevel= All Kernel Messages with a loglevel smaller than the
2664 console loglevel will be printed to the console. It can
2665 also be changed with klogd or other programs. The
2666 loglevels are defined as follows:
2668 0 (KERN_EMERG) system is unusable
2669 1 (KERN_ALERT) action must be taken immediately
2670 2 (KERN_CRIT) critical conditions
2671 3 (KERN_ERR) error conditions
2672 4 (KERN_WARNING) warning conditions
2673 5 (KERN_NOTICE) normal but significant condition
2674 6 (KERN_INFO) informational
2675 7 (KERN_DEBUG) debug-level messages
2677 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2678 in bytes. n must be a power of two and greater
2679 than the minimal size. The minimal size is defined
2680 by LOG_BUF_SHIFT kernel config parameter. There is
2681 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2682 that allows to increase the default size depending on
2683 the number of CPUs. See init/Kconfig for more details.
2685 logo.nologo [FB] Disables display of the built-in Linux logo.
2686 This may be used to provide more screen space for
2687 kernel log messages and is useful when debugging
2688 kernel boot problems.
2690 lp=0 [LP] Specify parallel ports to use, e.g,
2691 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2692 lp=reset first parallel port). 'lp=0' disables the
2693 lp=auto printer driver. 'lp=reset' (which can be
2694 specified in addition to the ports) causes
2695 attached printers to be reset. Using
2696 lp=port1,port2,... specifies the parallel ports
2697 to associate lp devices with, starting with
2698 lp0. A port specification may be 'none' to skip
2699 that lp device, or a parport name such as
2700 'parport0'. Specifying 'lp=auto' instead of a
2701 port specification list means that device IDs
2702 from each port should be examined, to see if
2703 an IEEE 1284-compliant printer is attached; if
2704 so, the driver will manage that printer.
2705 See also header of drivers/char/lp.c.
2708 Sets loops_per_jiffy to given constant, thus avoiding
2709 time-consuming boot-time autodetection (up to 250 ms per
2710 CPU). 0 enables autodetection (default). To determine
2711 the correct value for your kernel, boot with normal
2712 autodetection and see what value is printed. Note that
2713 on SMP systems the preset will be applied to all CPUs,
2714 which is likely to cause problems if your CPUs need
2715 significantly divergent settings. An incorrect value
2716 will cause delays in the kernel to be wrong, leading to
2717 unpredictable I/O errors and other breakage. Although
2718 unlikely, in the extreme case this might damage your
2722 Format: <io>,<irq>,<dma>
2724 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2727 [SECURITY] Choose order of LSM initialization. This
2728 overrides CONFIG_LSM, and the "security=" parameter.
2730 machvec= [IA-64] Force the use of a particular machine-vector
2731 (machvec) in a generic kernel.
2732 Example: machvec=hpzx1
2734 machtype= [Loongson] Share the same kernel image file between
2735 different yeeloong laptops.
2736 Example: machtype=lemote-yeeloong-2f-7inch
2738 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2739 than or equal to this physical address is ignored.
2741 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2742 will bring up during bootup. maxcpus=n : n >= 0 limits
2743 the kernel to bring up 'n' processors. Surely after
2744 bootup you can bring up the other plugged cpu by executing
2745 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2746 only takes effect during system bootup.
2747 While n=0 is a special case, it is equivalent to "nosmp",
2748 which also disables the IO APIC.
2750 max_loop= [LOOP] The number of loop block devices that get
2751 (loop.max_loop) unconditionally pre-created at init time. The default
2752 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2753 of statically allocating a predefined number, loop
2754 devices can be requested on-demand with the
2755 /dev/loop-control interface.
2757 mce [X86-32] Machine Check Exception
2759 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2761 md= [HW] RAID subsystems devices and level
2762 See Documentation/admin-guide/md.rst.
2765 Format: <first>,<last>
2766 Specifies range of consoles to be captured by the MDA.
2769 Control mitigation for the Micro-architectural Data
2770 Sampling (MDS) vulnerability.
2772 Certain CPUs are vulnerable to an exploit against CPU
2773 internal buffers which can forward information to a
2774 disclosure gadget under certain conditions.
2776 In vulnerable processors, the speculatively
2777 forwarded data can be used in a cache side channel
2778 attack, to access data to which the attacker does
2779 not have direct access.
2781 This parameter controls the MDS mitigation. The
2784 full - Enable MDS mitigation on vulnerable CPUs
2785 full,nosmt - Enable MDS mitigation and disable
2786 SMT on vulnerable CPUs
2787 off - Unconditionally disable MDS mitigation
2789 On TAA-affected machines, mds=off can be prevented by
2790 an active TAA mitigation as both vulnerabilities are
2791 mitigated with the same mechanism so in order to disable
2792 this mitigation, you need to specify tsx_async_abort=off
2795 Not specifying this option is equivalent to
2798 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2800 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2801 Amount of memory to be used in cases as follows:
2804 2 when the kernel is not able to see the whole system memory;
2805 3 memory that lies after 'mem=' boundary is excluded from
2806 the hypervisor, then assigned to KVM guests.
2808 [X86] Work as limiting max address. Use together
2809 with memmap= to avoid physical address space collisions.
2810 Without memmap= PCI devices could be placed at addresses
2811 belonging to unused RAM.
2813 Note that this only takes effects during boot time since
2814 in above case 3, memory may need be hot added after boot
2815 if system memory of hypervisor is not sufficient.
2817 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2821 [KNL,SH] Allow user to override the default size for
2822 per-device physically contiguous DMA buffers.
2824 memhp_default_state=online/offline
2825 [KNL] Set the initial state for the memory hotplug
2826 onlining policy. If not specified, the default value is
2827 set according to the
2828 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2830 See Documentation/admin-guide/mm/memory-hotplug.rst.
2832 memmap=exactmap [KNL,X86] Enable setting of an exact
2833 E820 memory map, as specified by the user.
2834 Such memmap=exactmap lines can be constructed based on
2835 BIOS output or other requirements. See the memmap=nn@ss
2838 memmap=nn[KMG]@ss[KMG]
2839 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory.
2840 Region of memory to be used is from ss to ss+nn.
2841 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2842 which limits max address to nn[KMG].
2843 Multiple different regions can be specified,
2846 memmap=100M@2G,100M#3G,1G!1024G
2848 memmap=nn[KMG]#ss[KMG]
2849 [KNL,ACPI] Mark specific memory as ACPI data.
2850 Region of memory to be marked is from ss to ss+nn.
2852 memmap=nn[KMG]$ss[KMG]
2853 [KNL,ACPI] Mark specific memory as reserved.
2854 Region of memory to be reserved is from ss to ss+nn.
2855 Example: Exclude memory from 0x18690000-0x1869ffff
2856 memmap=64K$0x18690000
2858 memmap=0x10000$0x18690000
2859 Some bootloaders may need an escape character before '$',
2860 like Grub2, otherwise '$' and the following number
2863 memmap=nn[KMG]!ss[KMG]
2864 [KNL,X86] Mark specific memory as protected.
2865 Region of memory to be used, from ss to ss+nn.
2866 The memory region may be marked as e820 type 12 (0xc)
2867 and is NVDIMM or ADR memory.
2869 memmap=<size>%<offset>-<oldtype>+<newtype>
2870 [KNL,ACPI] Convert memory within the specified region
2871 from <oldtype> to <newtype>. If "-<oldtype>" is left
2872 out, the whole region will be marked as <newtype>,
2873 even if previously unavailable. If "+<newtype>" is left
2874 out, matching memory will be removed. Types are
2875 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2876 3 = ACPI, 12 = PRAM.
2878 memory_corruption_check=0/1 [X86]
2879 Some BIOSes seem to corrupt the first 64k of
2880 memory when doing things like suspend/resume.
2881 Setting this option will scan the memory
2882 looking for corruption. Enabling this will
2883 both detect corruption and prevent the kernel
2884 from using the memory being corrupted.
2885 However, its intended as a diagnostic tool; if
2886 repeatable BIOS-originated corruption always
2887 affects the same memory, you can use memmap=
2888 to prevent the kernel from using that memory.
2890 memory_corruption_check_size=size [X86]
2891 By default it checks for corruption in the low
2892 64k, making this memory unavailable for normal
2893 use. Use this parameter to scan for
2894 corruption in more or less memory.
2896 memory_corruption_check_period=seconds [X86]
2897 By default it checks for corruption every 60
2898 seconds. Use this parameter to check at some
2899 other rate. 0 disables periodic checking.
2901 memory_hotplug.memmap_on_memory
2902 [KNL,X86,ARM] Boolean flag to enable this feature.
2903 Format: {on | off (default)}
2904 When enabled, runtime hotplugged memory will
2905 allocate its internal metadata (struct pages)
2906 from the hotadded memory which will allow to
2907 hotadd a lot of memory without requiring
2908 additional memory to do so.
2909 This feature is disabled by default because it
2910 has some implication on large (e.g. GB)
2911 allocations in some configurations (e.g. small
2913 The state of the flag can be read in
2914 /sys/module/memory_hotplug/parameters/memmap_on_memory.
2915 Note that even when enabled, there are a few cases where
2916 the feature is not effective.
2918 This is not compatible with hugetlb_free_vmemmap. If
2919 both parameters are enabled, hugetlb_free_vmemmap takes
2920 precedence over memory_hotplug.memmap_on_memory.
2922 memtest= [KNL,X86,ARM,PPC,RISCV] Enable memtest
2924 default : 0 <disable>
2925 Specifies the number of memtest passes to be
2926 performed. Each pass selects another test
2927 pattern from a given set of patterns. Memtest
2928 fills the memory with this pattern, validates
2929 memory contents and reserves bad memory
2930 regions that are detected.
2932 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2933 Valid arguments: on, off
2934 Default (depends on kernel configuration option):
2935 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2936 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2937 mem_encrypt=on: Activate SME
2938 mem_encrypt=off: Do not activate SME
2940 Refer to Documentation/virt/kvm/amd-memory-encryption.rst
2941 for details on when memory encryption can be activated.
2943 mem_sleep_default= [SUSPEND] Default system suspend mode:
2944 s2idle - Suspend-To-Idle
2945 shallow - Power-On Suspend or equivalent (if supported)
2946 deep - Suspend-To-RAM or equivalent (if supported)
2947 See Documentation/admin-guide/pm/sleep-states.rst.
2949 meye.*= [HW] Set MotionEye Camera parameters
2950 See Documentation/admin-guide/media/meye.rst.
2952 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2953 Multi-Function General Purpose Timers on AMD Geode
2956 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2957 the BIOS has incorrectly applied a workaround. TinyBIOS
2958 version 0.98 is known to be affected, 0.99 fixes the
2959 problem by letting the user disable the workaround.
2963 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2964 physical address is ignored.
2966 mini2440= [ARM,HW,KNL]
2967 Format:[0..2][b][c][t]
2969 MINI2440 configuration specification:
2970 0 - The attached screen is the 3.5" TFT
2971 1 - The attached screen is the 7" TFT
2972 2 - The VGA Shield is attached (1024x768)
2973 Leaving out the screen size parameter will not load
2974 the TFT driver, and the framebuffer will be left
2976 b - Enable backlight. The TFT backlight pin will be
2977 linked to the kernel VESA blanking code and a GPIO
2978 LED. This parameter is not necessary when using the
2980 c - Enable the s3c camera interface.
2981 t - Reserved for enabling touchscreen support. The
2982 touchscreen support is not enabled in the mainstream
2983 kernel as of 2.6.30, a preliminary port can be found
2984 in the "bleeding edge" mini2440 support kernel at
2985 https://repo.or.cz/w/linux-2.6/mini2440.git
2988 [X86,PPC,S390,ARM64] Control optional mitigations for
2989 CPU vulnerabilities. This is a set of curated,
2990 arch-independent options, each of which is an
2991 aggregation of existing arch-specific options.
2994 Disable all optional CPU mitigations. This
2995 improves system performance, but it may also
2996 expose users to several CPU vulnerabilities.
2997 Equivalent to: nopti [X86,PPC]
2999 nospectre_v1 [X86,PPC]
3001 nospectre_v2 [X86,PPC,S390,ARM64]
3002 spectre_v2_user=off [X86]
3003 spec_store_bypass_disable=off [X86,PPC]
3004 ssbd=force-off [ARM64]
3007 tsx_async_abort=off [X86]
3008 kvm.nx_huge_pages=off [X86]
3009 no_entry_flush [PPC]
3010 no_uaccess_flush [PPC]
3013 This does not have any effect on
3014 kvm.nx_huge_pages when
3015 kvm.nx_huge_pages=force.
3018 Mitigate all CPU vulnerabilities, but leave SMT
3019 enabled, even if it's vulnerable. This is for
3020 users who don't want to be surprised by SMT
3021 getting disabled across kernel upgrades, or who
3022 have other ways of avoiding SMT-based attacks.
3023 Equivalent to: (default behavior)
3026 Mitigate all CPU vulnerabilities, disabling SMT
3027 if needed. This is for users who always want to
3028 be fully mitigated, even if it means losing SMT.
3029 Equivalent to: l1tf=flush,nosmt [X86]
3030 mds=full,nosmt [X86]
3031 tsx_async_abort=full,nosmt [X86]
3034 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
3035 parameter allows control of the logging verbosity for
3036 the additional memory initialisation checks. A value
3037 of 0 disables mminit logging and a level of 4 will
3038 log everything. Information is printed at KERN_DEBUG
3039 so loglevel=8 may also need to be specified.
3042 [KNL] When CONFIG_MODULE_SIG is set, this means that
3043 modules without (valid) signatures will fail to load.
3044 Note that if CONFIG_MODULE_SIG_FORCE is set, that
3045 is always true, so this option does nothing.
3047 module_blacklist= [KNL] Do not load a comma-separated list of
3048 modules. Useful for debugging problem modules.
3051 [MOUSE] Maximum time between finger touching and
3052 leaving touchpad surface for touch to be considered
3053 a tap and be reported as a left button click (for
3054 touchpads working in absolute mode only).
3056 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
3057 reporting absolute coordinates, such as tablets
3058 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
3059 reporting absolute coordinates, such as tablets
3061 movablecore= [KNL,X86,IA-64,PPC]
3062 Format: nn[KMGTPE] | nn%
3063 This parameter is the complement to kernelcore=, it
3064 specifies the amount of memory used for migratable
3065 allocations. If both kernelcore and movablecore is
3066 specified, then kernelcore will be at *least* the
3067 specified value but may be more. If movablecore on its
3068 own is specified, the administrator must be careful
3069 that the amount of memory usable for all allocations
3072 movable_node [KNL] Boot-time switch to make hotplugable memory
3073 NUMA nodes to be movable. This means that the memory
3074 of such nodes will be usable only for movable
3075 allocations which rules out almost all kernel
3076 allocations. Use with caution!
3078 MTD_Partition= [MTD]
3079 Format: <name>,<region-number>,<size>,<offset>
3081 MTD_Region= [MTD] Format:
3082 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
3085 See drivers/mtd/parsers/cmdlinepart.c
3087 multitce=off [PPC] This parameter disables the use of the pSeries
3088 firmware feature for updating multiple TCE entries
3091 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
3093 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
3095 boundary - index of last SLC block on Flex-OneNAND.
3096 The remaining blocks are configured as MLC blocks.
3097 lock - Configure if Flex-OneNAND boundary should be locked.
3098 Once locked, the boundary cannot be changed.
3099 1 indicates lock status, 0 indicates unlock status.
3102 ARM/S3C2412 JIVE boot control
3104 See arch/arm/mach-s3c/mach-jive.c
3106 mtouchusb.raw_coordinates=
3107 [HW] Make the MicroTouch USB driver use raw coordinates
3108 ('y', default) or cooked coordinates ('n')
3110 mtrr_chunk_size=nn[KMG] [X86]
3111 used for mtrr cleanup. It is largest continuous chunk
3112 that could hold holes aka. UC entries.
3114 mtrr_gran_size=nn[KMG] [X86]
3115 Used for mtrr cleanup. It is granularity of mtrr block.
3117 Large value could prevent small alignment from
3120 mtrr_spare_reg_nr=n [X86]
3122 Range: 0,7 : spare reg number
3124 Used for mtrr cleanup. It is spare mtrr entries number.
3125 Set to 2 or more if your graphical card needs more.
3127 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
3129 netdev= [NET] Network devices parameters
3130 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
3131 Note that mem_start is often overloaded to mean
3132 something different and driver-specific.
3133 This usage is only documented in each driver source
3137 [NETFILTER] Enable connection tracking flow accounting
3138 0 to disable accounting
3139 1 to enable accounting
3142 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3143 See Documentation/admin-guide/nfs/nfsroot.rst.
3145 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3146 See Documentation/admin-guide/nfs/nfsroot.rst.
3148 nfsrootdebug [NFS] enable nfsroot debugging messages.
3149 See Documentation/admin-guide/nfs/nfsroot.rst.
3151 nfs.callback_nr_threads=
3152 [NFSv4] set the total number of threads that the
3153 NFS client will assign to service NFSv4 callback
3156 nfs.callback_tcpport=
3157 [NFS] set the TCP port on which the NFSv4 callback
3158 channel should listen.
3161 [NFS] sets the pathname to the program which is used
3162 to update the NFS client cache entries.
3164 nfs.cache_getent_timeout=
3165 [NFS] sets the timeout after which an attempt to
3166 update a cache entry is deemed to have failed.
3168 nfs.idmap_cache_timeout=
3169 [NFS] set the maximum lifetime for idmapper cache
3173 [NFS] enable 64-bit inode numbers.
3174 If zero, the NFS client will fake up a 32-bit inode
3175 number for the readdir() and stat() syscalls instead
3176 of returning the full 64-bit number.
3177 The default is to return 64-bit inode numbers.
3179 nfs.max_session_cb_slots=
3180 [NFSv4.1] Sets the maximum number of session
3181 slots the client will assign to the callback
3182 channel. This determines the maximum number of
3183 callbacks the client will process in parallel for
3184 a particular server.
3186 nfs.max_session_slots=
3187 [NFSv4.1] Sets the maximum number of session slots
3188 the client will attempt to negotiate with the server.
3189 This limits the number of simultaneous RPC requests
3190 that the client can send to the NFSv4.1 server.
3191 Note that there is little point in setting this
3192 value higher than the max_tcp_slot_table_limit.
3194 nfs.nfs4_disable_idmapping=
3195 [NFSv4] When set to the default of '1', this option
3196 ensures that both the RPC level authentication
3197 scheme and the NFS level operations agree to use
3198 numeric uids/gids if the mount is using the
3199 'sec=sys' security flavour. In effect it is
3200 disabling idmapping, which can make migration from
3201 legacy NFSv2/v3 systems to NFSv4 easier.
3202 Servers that do not support this mode of operation
3203 will be autodetected by the client, and it will fall
3204 back to using the idmapper.
3205 To turn off this behaviour, set the value to '0'.
3207 [NFS4] Specify an additional fixed unique ident-
3208 ification string that NFSv4 clients can insert into
3209 their nfs_client_id4 string. This is typically a
3210 UUID that is generated at system install time.
3212 nfs.send_implementation_id =
3213 [NFSv4.1] Send client implementation identification
3214 information in exchange_id requests.
3215 If zero, no implementation identification information
3217 The default is to send the implementation identification
3220 nfs.recover_lost_locks =
3221 [NFSv4] Attempt to recover locks that were lost due
3222 to a lease timeout on the server. Please note that
3223 doing this risks data corruption, since there are
3224 no guarantees that the file will remain unchanged
3225 after the locks are lost.
3226 If you want to enable the kernel legacy behaviour of
3227 attempting to recover these locks, then set this
3229 The default parameter value of '0' causes the kernel
3230 not to attempt recovery of lost locks.
3232 nfs4.layoutstats_timer =
3233 [NFSv4.2] Change the rate at which the kernel sends
3234 layoutstats to the pNFS metadata server.
3236 Setting this to value to 0 causes the kernel to use
3237 whatever value is the default set by the layout
3238 driver. A non-zero value sets the minimum interval
3239 in seconds between layoutstats transmissions.
3241 nfsd.nfs4_disable_idmapping=
3242 [NFSv4] When set to the default of '1', the NFSv4
3243 server will return only numeric uids and gids to
3244 clients using auth_sys, and will accept numeric uids
3245 and gids from such clients. This is intended to ease
3246 migration from NFSv2/v3.
3248 nmi_backtrace.backtrace_idle [KNL]
3249 Dump stacks even of idle CPUs in response to an
3250 NMI stack-backtrace request.
3252 nmi_debug= [KNL,SH] Specify one or more actions to take
3253 when a NMI is triggered.
3254 Format: [state][,regs][,debounce][,die]
3256 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3257 Format: [panic,][nopanic,][num]
3259 0 - turn hardlockup detector in nmi_watchdog off
3260 1 - turn hardlockup detector in nmi_watchdog on
3261 When panic is specified, panic when an NMI watchdog
3262 timeout occurs (or 'nopanic' to not panic on an NMI
3263 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3264 To disable both hard and soft lockup detectors,
3265 please see 'nowatchdog'.
3266 This is useful when you use a panic=... timeout and
3267 need the box quickly up again.
3269 These settings can be accessed at runtime via
3270 the nmi_watchdog and hardlockup_panic sysctls.
3272 netpoll.carrier_timeout=
3273 [NET] Specifies amount of time (in seconds) that
3274 netpoll should wait for a carrier. By default netpoll
3277 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3278 emulation library even if a 387 maths coprocessor
3281 no5lvl [X86-64] Disable 5-level paging mode. Forces
3282 kernel to use 4-level paging instead.
3284 nofsgsbase [X86] Disables FSGSBASE instructions.
3287 [HW] Never suspend the console
3288 Disable suspending of consoles during suspend and
3289 hibernate operations. Once disabled, debugging
3290 messages can reach various consoles while the rest
3291 of the system is being put to sleep (ie, while
3292 debugging driver suspend/resume hooks). This may
3293 not work reliably with all consoles, but is known
3294 to work with serial and VGA consoles.
3295 To facilitate more flexible debugging, we also add
3296 console_suspend, a printk module parameter to control
3297 it. Users could use console_suspend (usually
3298 /sys/module/printk/parameters/console_suspend) to
3299 turn on/off it dynamically.
3301 novmcoredd [KNL,KDUMP]
3302 Disable device dump. Device dump allows drivers to
3303 append dump data to vmcore so you can collect driver
3304 specified debug info. Drivers can append the data
3305 without any limit and this data is stored in memory,
3306 so this may cause significant memory stress. Disabling
3307 device dump can help save memory but the driver debug
3308 data will be no longer available. This parameter
3309 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3312 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3313 caches in the slab allocator. Saves per-node memory,
3314 but will impact performance.
3318 noaltinstr [S390] Disables alternative instructions patching
3319 (CPU alternatives feature).
3321 noapic [SMP,APIC] Tells the kernel to not make use of any
3322 IOAPICs that may be present in the system.
3324 noautogroup Disable scheduler automatic task group creation.
3326 nobats [PPC] Do not use BATs for mapping kernel lowmem
3327 on "Classic" PPC cores.
3331 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
3333 delayacct [KNL] Enable per-task delay accounting
3335 nodsp [SH] Disable hardware DSP at boot time.
3337 noefi Disable EFI runtime services support.
3339 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3344 On X86-32 available only on PAE configured kernels.
3345 noexec=on: enable non-executable mappings (default)
3346 noexec=off: disable non-executable mappings
3349 Disable SMAP (Supervisor Mode Access Prevention)
3350 even if it is supported by processor.
3353 Disable SMEP (Supervisor Mode Execution Prevention)
3354 even if it is supported by processor.
3357 This affects only 32-bit executables.
3358 noexec32=on: enable non-executable mappings (default)
3359 read doesn't imply executable mappings
3360 noexec32=off: disable non-executable mappings
3361 read implies executable mappings
3363 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3365 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3366 register save and restore. The kernel will only save
3367 legacy floating-point registers on task switch.
3369 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings.
3371 nohugevmalloc [PPC] Disable kernel huge vmalloc mappings.
3373 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3374 Equivalent to smt=1.
3376 [KNL,X86] Disable symmetric multithreading (SMT).
3377 nosmt=force: Force disable SMT, cannot be undone
3378 via the sysfs control file.
3380 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3381 (bounds check bypass). With this option data leaks are
3382 possible in the system.
3384 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3385 the Spectre variant 2 (indirect branch prediction)
3386 vulnerability. System may allow data leaks with this
3389 nospec_store_bypass_disable
3390 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3393 [PPC] Don't flush the L1-D cache after accessing user data.
3395 noxsave [BUGS=X86] Disables x86 extended register state save
3396 and restore using xsave. The kernel will fallback to
3397 enabling legacy floating-point and sse state.
3399 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3400 register states. The kernel will fall back to use
3401 xsave to save the states. By using this parameter,
3402 performance of saving the states is degraded because
3403 xsave doesn't support modified optimization while
3404 xsaveopt supports it on xsaveopt enabled systems.
3406 noxsaves [X86] Disables xsaves and xrstors used in saving and
3407 restoring x86 extended register state in compacted
3408 form of xsave area. The kernel will fall back to use
3409 xsaveopt and xrstor to save and restore the states
3410 in standard form of xsave area. By using this
3411 parameter, xsave area per process might occupy more
3412 memory on xsaves enabled systems.
3414 nohlt [ARM,ARM64,MICROBLAZE,SH] Forces the kernel to busy wait
3415 in do_idle() and not use the arch_cpu_idle()
3416 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3417 to be effective. This is useful on platforms where the
3418 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3419 correctly or when doing power measurements to evalute
3420 the impact of the sleep instructions. This is also
3421 useful when using JTAG debugger.
3423 no_file_caps Tells the kernel not to honor file capabilities. The
3424 only way then for a file to be executed with privilege
3425 is to be setuid root or executed by root.
3427 nohalt [IA-64] Tells the kernel not to use the power saving
3428 function PAL_HALT_LIGHT when idle. This increases
3429 power-consumption. On the positive side, it reduces
3430 interrupt wake-up latency, which may improve performance
3431 in certain environments such as networked servers or
3435 Force pointers printed to the console or buffers to be
3436 unhashed. By default, when a pointer is printed via %p
3437 format string, that pointer is "hashed", i.e. obscured
3438 by hashing the pointer value. This is a security feature
3439 that hides actual kernel addresses from unprivileged
3440 users, but it also makes debugging the kernel more
3441 difficult since unequal pointers can no longer be
3442 compared. However, if this command-line option is
3443 specified, then all normal pointers will have their true
3444 value printed. Pointers printed via %pK may still be
3445 hashed. This option should only be specified when
3446 debugging the kernel. Please do not use on production
3449 nohibernate [HIBERNATION] Disable hibernation and resume.
3451 nohz= [KNL] Boottime enable/disable dynamic ticks
3452 Valid arguments: on, off
3455 nohz_full= [KNL,BOOT,SMP,ISOL]
3456 The argument is a cpu list, as described above.
3457 In kernels built with CONFIG_NO_HZ_FULL=y, set
3458 the specified list of CPUs whose tick will be stopped
3459 whenever possible. The boot CPU will be forced outside
3460 the range to maintain the timekeeping. Any CPUs
3461 in this list will have their RCU callbacks offloaded,
3462 just as if they had also been called out in the
3463 rcu_nocbs= boot parameter.
3465 noiotrap [SH] Disables trapped I/O port accesses.
3467 noirqdebug [X86-32] Disables the code which attempts to detect and
3468 disable unhandled interrupt sources.
3470 no_timer_check [X86,APIC] Disables the code which tests for
3471 broken timer IRQ sources.
3473 noisapnp [ISAPNP] Disables ISA PnP code.
3475 noinitrd [RAM] Tells the kernel not to load any configured
3478 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3480 [Deprecated - use intremap=off]
3484 noinvpcid [X86] Disable the INVPCID cpu feature.
3486 nojitter [IA-64] Disables jitter checking for ITC timers.
3488 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3490 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3494 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3495 clock and use the default one.
3497 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time
3498 accounting. steal time is computed, but won't
3499 influence scheduler behaviour
3501 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3503 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3505 noltlbs [PPC] Do not use large page/tlb entries for kernel
3506 lowmem mapping on PPC40x and PPC8xx
3508 nomca [IA-64] Disable machine check abort handling
3510 nomce [X86-32] Disable Machine Check Exception
3512 nomfgpt [X86-32] Disable Multi-Function General Purpose
3513 Timer usage (for AMD Geode machines).
3515 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3516 shutdown the other cpus. Instead use the REBOOT_VECTOR
3519 nomodule Disable module load
3521 nopat [X86] Disable PAT (page attribute table extension of
3522 pagetables) support.
3524 nopcid [X86-64] Disable the PCID cpu feature.
3526 norandmaps Don't use address space randomization. Equivalent to
3527 echo 0 > /proc/sys/kernel/randomize_va_space
3529 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3530 with UP alternatives
3532 nordrand [X86] Disable kernel use of the RDRAND and
3533 RDSEED instructions even if they are supported
3534 by the processor. RDRAND and RDSEED are still
3535 available to user space applications.
3537 noresume [SWSUSP] Disables resume and restores original swap
3540 no-scroll [VGA] Disables scrollback.
3541 This is required for the Braillex ib80-piezo Braille
3542 reader made by F.H. Papenmeier (Germany).
3546 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
3548 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3550 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3551 and disable the IO APIC. legacy for "maxcpus=0".
3553 nosoftlockup [KNL] Disable the soft-lockup detector.
3555 nosync [HW,M68K] Disables sync negotiation for all devices.
3557 nowatchdog [KNL] Disable both lockup detectors, i.e.
3558 soft-lockup and NMI watchdog (hard-lockup).
3562 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3564 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
3565 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
3566 Some features depend on CPU0. Known dependencies are:
3567 1. Resume from suspend/hibernate depends on CPU0.
3568 Suspend/hibernate will fail if CPU0 is offline and you
3569 need to online CPU0 before suspend/hibernate.
3570 2. PIC interrupts also depend on CPU0. CPU0 can't be
3571 removed if a PIC interrupt is detected.
3572 It's said poweroff/reboot may depend on CPU0 on some
3573 machines although I haven't seen such issues so far
3574 after CPU0 is offline on a few tested machines.
3575 If the dependencies are under your control, you can
3576 turn on cpu0_hotplug.
3578 nps_mtm_hs_ctr= [KNL,ARC]
3579 This parameter sets the maximum duration, in
3580 cycles, each HW thread of the CTOP can run
3581 without interruptions, before HW switches it.
3582 The actual maximum duration is 16 times this
3584 Format: integer between 1 and 255
3587 nptcg= [IA-64] Override max number of concurrent global TLB
3588 purges which is reported from either PAL_VM_SUMMARY or
3591 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3592 could support. nr_cpus=n : n >= 1 limits the kernel to
3593 support 'n' processors. It could be larger than the
3594 number of already plugged CPU during bootup, later in
3595 runtime you can physically add extra cpu until it reaches
3596 n. So during boot up some boot time memory for per-cpu
3597 variables need be pre-allocated for later physical cpu
3600 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3602 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86] Disable NUMA, Only
3603 set up a single NUMA node spanning all memory.
3605 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
3607 Allowed values are enable and disable
3609 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3610 'node', 'default' can be specified
3611 This can be set from sysctl after boot.
3612 See Documentation/admin-guide/sysctl/vm.rst for details.
3614 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3615 See Documentation/core-api/debugging-via-ohci1394.rst for more
3618 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3619 Rather than timing out after 20 ms if an EC
3620 command is not properly ACKed, override the length
3621 of the timeout. We have interrupts disabled while
3622 waiting for the ACK, so if this is set too high
3623 interrupts *may* be lost!
3625 omap_mux= [OMAP] Override bootloader pin multiplexing.
3626 Format: <mux_mode0.mode_name=value>...
3627 For example, to override I2C bus2:
3628 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3630 oops=panic Always panic on oopses. Default is to just kill the
3631 process, but there is a small probability of
3632 deadlocking the machine.
3633 This will also cause panics on machine check exceptions.
3634 Useful together with panic=30 to trigger a reboot.
3637 [KNL] Boolean flag to control whether the page allocator
3638 should randomize its free lists. The randomization may
3639 be automatically enabled if the kernel detects it is
3640 running on a platform with a direct-mapped memory-side
3641 cache, and this parameter can be used to
3642 override/disable that behavior. The state of the flag
3643 can be read from sysfs at:
3644 /sys/module/page_alloc/parameters/shuffle.
3646 page_owner= [KNL] Boot-time page_owner enabling option.
3647 Storage of the information about who allocated
3648 each page is disabled in default. With this switch,
3650 on: enable the feature
3652 page_poison= [KNL] Boot-time parameter changing the state of
3653 poisoning on the buddy allocator, available with
3654 CONFIG_PAGE_POISONING=y.
3655 off: turn off poisoning (default)
3656 on: turn on poisoning
3658 page_reporting.page_reporting_order=
3659 [KNL] Minimal page reporting order
3661 Adjust the minimal page reporting order. The page
3662 reporting is disabled when it exceeds (MAX_ORDER-1).
3664 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3665 timeout > 0: seconds before rebooting
3666 timeout = 0: wait forever
3667 timeout < 0: reboot immediately
3670 panic_print= Bitmask for printing system info when panic happens.
3671 User can chose combination of the following bits:
3672 bit 0: print all tasks info
3673 bit 1: print system memory info
3674 bit 2: print timer info
3675 bit 3: print locks info if CONFIG_LOCKDEP is on
3676 bit 4: print ftrace buffer
3677 bit 5: print all printk messages in buffer
3679 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3680 Format: <hex>[,nousertaint]
3681 Hexadecimal bitmask representing the set of TAINT flags
3682 that will cause the kernel to panic when add_taint() is
3683 called with any of the flags in this set.
3684 The optional switch "nousertaint" can be utilized to
3685 prevent userspace forced crashes by writing to sysctl
3686 /proc/sys/kernel/tainted any flagset matching with the
3687 bitmask set on panic_on_taint.
3688 See Documentation/admin-guide/tainted-kernels.rst for
3689 extra details on the taint flags that users can pick
3690 to compose the bitmask to assign to panic_on_taint.
3692 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3695 crash_kexec_post_notifiers
3696 Run kdump after running panic-notifiers and dumping
3697 kmsg. This only for the users who doubt kdump always
3698 succeeds in any situation.
3699 Note that this also increases risks of kdump failure,
3700 because some panic notifiers can make the crashed
3701 kernel more unstable.
3703 parkbd.port= [HW] Parallel port number the keyboard adapter is
3704 connected to, default is 0.
3706 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3707 0 for XT, 1 for AT (default is AT).
3710 parport= [HW,PPT] Specify parallel ports. 0 disables.
3711 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3712 Use 'auto' to force the driver to use any
3713 IRQ/DMA settings detected (the default is to
3714 ignore detected IRQ/DMA settings because of
3715 possible conflicts). You can specify the base
3716 address, IRQ, and DMA settings; IRQ and DMA
3717 should be numbers, or 'auto' (for using detected
3718 settings on that particular port), or 'nofifo'
3719 (to avoid using a FIFO even if it is detected).
3720 Parallel ports are assigned in the order they
3721 are specified on the command line, starting
3724 parport_init_mode= [HW,PPT]
3725 Configure VIA parallel port to operate in
3726 a specific mode. This is necessary on Pegasos
3727 computer where firmware has no options for setting
3728 up parallel port mode and sets it to spp.
3729 Currently this function knows 686a and 8231 chips.
3730 Format: [spp|ps2|epp|ecp|ecpepp]
3732 pata_legacy.all= [HW,LIBATA]
3734 Set to non-zero to probe primary and secondary ISA
3735 port ranges on PCI systems where no PCI PATA device
3736 has been found at either range. Disabled by default.
3738 pata_legacy.autospeed= [HW,LIBATA]
3740 Set to non-zero if a chip is present that snoops speed
3741 changes. Disabled by default.
3743 pata_legacy.ht6560a= [HW,LIBATA]
3745 Set to 1, 2, or 3 for HT 6560A on the primary channel,
3746 the secondary channel, or both channels respectively.
3747 Disabled by default.
3749 pata_legacy.ht6560b= [HW,LIBATA]
3751 Set to 1, 2, or 3 for HT 6560B on the primary channel,
3752 the secondary channel, or both channels respectively.
3753 Disabled by default.
3755 pata_legacy.iordy_mask= [HW,LIBATA]
3757 IORDY enable mask. Set individual bits to allow IORDY
3758 for the respective channel. Bit 0 is for the first
3759 legacy channel handled by this driver, bit 1 is for
3760 the second channel, and so on. The sequence will often
3761 correspond to the primary legacy channel, the secondary
3762 legacy channel, and so on, but the handling of a PCI
3763 bus and the use of other driver options may interfere
3764 with the sequence. By default IORDY is allowed across
3767 pata_legacy.opti82c46x= [HW,LIBATA]
3769 Set to 1, 2, or 3 for Opti 82c611A on the primary
3770 channel, the secondary channel, or both channels
3771 respectively. Disabled by default.
3773 pata_legacy.opti82c611a= [HW,LIBATA]
3775 Set to 1, 2, or 3 for Opti 82c465MV on the primary
3776 channel, the secondary channel, or both channels
3777 respectively. Disabled by default.
3779 pata_legacy.pio_mask= [HW,LIBATA]
3781 PIO mode mask for autospeed devices. Set individual
3782 bits to allow the use of the respective PIO modes.
3783 Bit 0 is for mode 0, bit 1 is for mode 1, and so on.
3784 All modes allowed by default.
3786 pata_legacy.probe_all= [HW,LIBATA]
3788 Set to non-zero to probe tertiary and further ISA
3789 port ranges on PCI systems. Disabled by default.
3791 pata_legacy.probe_mask= [HW,LIBATA]
3793 Probe mask for legacy ISA PATA ports. Depending on
3794 platform configuration and the use of other driver
3795 options up to 6 legacy ports are supported: 0x1f0,
3796 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing
3797 of individual ports can be disabled by setting the
3798 corresponding bits in the mask to 1. Bit 0 is for
3799 the first port in the list above (0x1f0), and so on.
3800 By default all supported ports are probed.
3802 pata_legacy.qdi= [HW,LIBATA]
3804 Set to non-zero to probe QDI controllers. By default
3805 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
3807 pata_legacy.winbond= [HW,LIBATA]
3809 Set to non-zero to probe Winbond controllers. Use
3810 the standard I/O port (0x130) if 1, otherwise the
3811 value given is the I/O port to use (typically 0x1b0).
3812 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
3815 pata_platform.pio_mask= [HW,LIBATA]
3817 Supported PIO mode mask. Set individual bits to allow
3818 the use of the respective PIO modes. Bit 0 is for
3819 mode 0, bit 1 is for mode 1, and so on. Mode 0 only
3823 Halt all CPUs after the first oops has been printed for
3824 the specified number of seconds. This is to be used if
3825 your oopses keep scrolling off the screen.
3830 See header of drivers/block/paride/pcd.c.
3831 See also Documentation/admin-guide/blockdev/paride.rst.
3833 pci=option[,option...] [PCI] various PCI subsystem options.
3835 Some options herein operate on a specific device
3836 or a set of devices (<pci_dev>). These are
3837 specified in one of the following formats:
3839 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
3840 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
3842 Note: the first format specifies a PCI
3843 bus/device/function address which may change
3844 if new hardware is inserted, if motherboard
3845 firmware changes, or due to changes caused
3846 by other kernel parameters. If the
3847 domain is left unspecified, it is
3848 taken to be zero. Optionally, a path
3849 to a device through multiple device/function
3850 addresses can be specified after the base
3851 address (this is more robust against
3852 renumbering issues). The second format
3853 selects devices using IDs from the
3854 configuration space which may match multiple
3855 devices in the system.
3857 earlydump dump PCI config space before the kernel
3859 off [X86] don't probe for the PCI bus
3860 bios [X86-32] force use of PCI BIOS, don't access
3861 the hardware directly. Use this if your machine
3862 has a non-standard PCI host bridge.
3863 nobios [X86-32] disallow use of PCI BIOS, only direct
3864 hardware access methods are allowed. Use this
3865 if you experience crashes upon bootup and you
3866 suspect they are caused by the BIOS.
3867 conf1 [X86] Force use of PCI Configuration Access
3868 Mechanism 1 (config address in IO port 0xCF8,
3869 data in IO port 0xCFC, both 32-bit).
3870 conf2 [X86] Force use of PCI Configuration Access
3871 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3872 the function, IO port 0xCFA, also 8-bit, sets
3873 bus number. The config space is then accessed
3874 through ports 0xC000-0xCFFF).
3875 See http://wiki.osdev.org/PCI for more info
3876 on the configuration access mechanisms.
3877 noaer [PCIE] If the PCIEAER kernel config parameter is
3878 enabled, this kernel boot option can be used to
3879 disable the use of PCIE advanced error reporting.
3880 nodomains [PCI] Disable support for multiple PCI
3881 root domains (aka PCI segments, in ACPI-speak).
3882 nommconf [X86] Disable use of MMCONFIG for PCI
3884 check_enable_amd_mmconf [X86] check for and enable
3885 properly configured MMIO access to PCI
3886 config space on AMD family 10h CPU
3887 nomsi [MSI] If the PCI_MSI kernel config parameter is
3888 enabled, this kernel boot option can be used to
3889 disable the use of MSI interrupts system-wide.
3890 noioapicquirk [APIC] Disable all boot interrupt quirks.
3891 Safety option to keep boot IRQs enabled. This
3892 should never be necessary.
3893 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3894 primary IO-APIC for bridges that cannot disable
3895 boot IRQs. This fixes a source of spurious IRQs
3896 when the system masks IRQs.
3897 noioapicreroute [APIC] Disable workaround that uses the
3898 boot IRQ equivalent of an IRQ that connects to
3899 a chipset where boot IRQs cannot be disabled.
3900 The opposite of ioapicreroute.
3901 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3902 routing table. These calls are known to be buggy
3903 on several machines and they hang the machine
3904 when used, but on other computers it's the only
3905 way to get the interrupt routing table. Try
3906 this option if the kernel is unable to allocate
3907 IRQs or discover secondary PCI buses on your
3909 rom [X86] Assign address space to expansion ROMs.
3910 Use with caution as certain devices share
3911 address decoders between ROMs and other
3913 norom [X86] Do not assign address space to
3914 expansion ROMs that do not already have
3915 BIOS assigned address ranges.
3916 nobar [X86] Do not assign address space to the
3917 BARs that weren't assigned by the BIOS.
3918 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3919 assigned automatically to PCI devices. You can
3920 make the kernel exclude IRQs of your ISA cards
3922 pirqaddr=0xAAAAA [X86] Specify the physical address
3923 of the PIRQ table (normally generated
3924 by the BIOS) if it is outside the
3925 F0000h-100000h range.
3926 lastbus=N [X86] Scan all buses thru bus #N. Can be
3927 useful if the kernel is unable to find your
3928 secondary buses and you want to tell it
3929 explicitly which ones they are.
3930 assign-busses [X86] Always assign all PCI bus
3931 numbers ourselves, overriding
3932 whatever the firmware may have done.
3933 usepirqmask [X86] Honor the possible IRQ mask stored
3934 in the BIOS $PIR table. This is needed on
3935 some systems with broken BIOSes, notably
3936 some HP Pavilion N5400 and Omnibook XE3
3937 notebooks. This will have no effect if ACPI
3938 IRQ routing is enabled.
3939 noacpi [X86] Do not use ACPI for IRQ routing
3940 or for PCI scanning.
3941 use_crs [X86] Use PCI host bridge window information
3942 from ACPI. On BIOSes from 2008 or later, this
3943 is enabled by default. If you need to use this,
3944 please report a bug.
3945 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3946 If you need to use this, please report a bug.
3947 routeirq Do IRQ routing for all PCI devices.
3948 This is normally done in pci_enable_device(),
3949 so this option is a temporary workaround
3950 for broken drivers that don't call it.
3951 skip_isa_align [X86] do not align io start addr, so can
3952 handle more pci cards
3953 noearly [X86] Don't do any early type 1 scanning.
3954 This might help on some broken boards which
3955 machine check when some devices' config space
3956 is read. But various workarounds are disabled
3957 and some IOMMU drivers will not work.
3958 bfsort Sort PCI devices into breadth-first order.
3959 This sorting is done to get a device
3960 order compatible with older (<= 2.4) kernels.
3961 nobfsort Don't sort PCI devices into breadth-first order.
3962 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3963 tuning and use the BIOS-configured MPS defaults.
3964 pcie_bus_safe Set every device's MPS to the largest value
3965 supported by all devices below the root complex.
3966 pcie_bus_perf Set device MPS to the largest allowable MPS
3967 based on its parent bus. Also set MRRS (Max
3968 Read Request Size) to the largest supported
3969 value (no larger than the MPS that the device
3970 or bus can support) for best performance.
3971 pcie_bus_peer2peer Set every device's MPS to 128B, which
3972 every device is guaranteed to support. This
3973 configuration allows peer-to-peer DMA between
3974 any pair of devices, possibly at the cost of
3975 reduced performance. This also guarantees
3976 that hot-added devices will work.
3977 cbiosize=nn[KMG] The fixed amount of bus space which is
3978 reserved for the CardBus bridge's IO window.
3979 The default value is 256 bytes.
3980 cbmemsize=nn[KMG] The fixed amount of bus space which is
3981 reserved for the CardBus bridge's memory
3982 window. The default value is 64 megabytes.
3985 [<order of align>@]<pci_dev>[; ...]
3986 Specifies alignment and device to reassign
3987 aligned memory resources. How to
3988 specify the device is described above.
3989 If <order of align> is not specified,
3990 PAGE_SIZE is used as alignment.
3991 A PCI-PCI bridge can be specified if resource
3992 windows need to be expanded.
3993 To specify the alignment for several
3994 instances of a device, the PCI vendor,
3995 device, subvendor, and subdevice may be
3996 specified, e.g., 12@pci:8086:9c22:103c:198f
3997 for 4096-byte alignment.
3998 ecrc= Enable/disable PCIe ECRC (transaction layer
3999 end-to-end CRC checking).
4000 bios: Use BIOS/firmware settings. This is the
4004 hpiosize=nn[KMG] The fixed amount of bus space which is
4005 reserved for hotplug bridge's IO window.
4006 Default size is 256 bytes.
4007 hpmmiosize=nn[KMG] The fixed amount of bus space which is
4008 reserved for hotplug bridge's MMIO window.
4009 Default size is 2 megabytes.
4010 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
4011 reserved for hotplug bridge's MMIO_PREF window.
4012 Default size is 2 megabytes.
4013 hpmemsize=nn[KMG] The fixed amount of bus space which is
4014 reserved for hotplug bridge's MMIO and
4016 Default size is 2 megabytes.
4017 hpbussize=nn The minimum amount of additional bus numbers
4018 reserved for buses below a hotplug bridge.
4020 realloc= Enable/disable reallocating PCI bridge resources
4021 if allocations done by BIOS are too small to
4022 accommodate resources required by all child
4024 off: Turn realloc off
4026 realloc same as realloc=on
4027 noari do not use PCIe ARI.
4028 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
4029 do not use PCIe ATS (and IOMMU device IOTLB).
4030 pcie_scan_all Scan all possible PCIe devices. Otherwise we
4031 only look for one device below a PCIe downstream
4033 big_root_window Try to add a big 64bit memory window to the PCIe
4034 root complex on AMD CPUs. Some GFX hardware
4035 can resize a BAR to allow access to all VRAM.
4036 Adding the window is slightly risky (it may
4037 conflict with unreported devices), so this
4039 disable_acs_redir=<pci_dev>[; ...]
4040 Specify one or more PCI devices (in the format
4041 specified above) separated by semicolons.
4042 Each device specified will have the PCI ACS
4043 redirect capabilities forced off which will
4044 allow P2P traffic between devices through
4045 bridges without forcing it upstream. Note:
4046 this removes isolation between devices and
4047 may put more devices in an IOMMU group.
4048 force_floating [S390] Force usage of floating interrupts.
4049 nomio [S390] Do not use MIO instructions.
4050 norid [S390] ignore the RID field and force use of
4051 one PCI domain per PCI function
4053 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
4056 force Enable ASPM even on devices that claim not to support it.
4057 WARNING: Forcing ASPM on may cause system lockups.
4059 pcie_ports= [PCIE] PCIe port services handling:
4060 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
4061 even if the platform doesn't give the OS permission to
4062 use them. This may cause conflicts if the platform
4063 also tries to use these services.
4064 dpc-native Use native PCIe service for DPC only. May
4065 cause conflicts if firmware uses AER or DPC.
4066 compat Disable native PCIe services (PME, AER, DPC, PCIe
4069 pcie_port_pm= [PCIE] PCIe port power management handling:
4070 off Disable power management of all PCIe ports
4071 force Forcibly enable power management of all PCIe ports
4073 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
4074 nomsi Do not use MSI for native PCIe PME signaling (this makes
4075 all PCIe root ports use INTx for all services).
4077 pcmv= [HW,PCMCIA] BadgePAD 4
4081 Keep all power-domains already enabled by bootloader on,
4082 even if no driver has claimed them. This is useful
4083 for debug and development, but should not be
4084 needed on a platform with proper driver support.
4087 See Documentation/admin-guide/blockdev/paride.rst.
4089 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
4092 See arch/parisc/kernel/pdc_chassis.c
4094 percpu_alloc= Select which percpu first chunk allocator to use.
4095 Currently supported values are "embed" and "page".
4096 Archs may support subset or none of the selections.
4097 See comments in mm/percpu.c for details on each
4098 allocator. This parameter is primarily for debugging
4099 and performance comparison.
4102 See Documentation/admin-guide/blockdev/paride.rst.
4105 See Documentation/admin-guide/blockdev/paride.rst.
4107 pirq= [SMP,APIC] Manual mp-table setup
4108 See Documentation/x86/i386/IO-APIC.rst.
4110 plip= [PPT,NET] Parallel port network link
4111 Format: { parport<nr> | timid | 0 }
4112 See also Documentation/admin-guide/parport.rst.
4114 pmtmr= [X86] Manual setup of pmtmr I/O Port.
4115 Override pmtimer IOPort with a hex value.
4118 pm_debug_messages [SUSPEND,KNL]
4119 Enable suspend/resume debug messages during boot up.
4122 Enable PNP debug messages (depends on the
4123 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4124 via /sys/module/pnp/parameters/debug. We always show
4125 current resource usage; turning this on also shows
4126 possible settings and some assignment information.
4132 { on | off | curr | res | no-curr | no-res }
4135 [ISAPNP] Exclude IRQs for the autoconfiguration
4138 [ISAPNP] Exclude DMAs for the autoconfiguration
4140 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
4141 Ranges are in pairs (I/O port base and size).
4144 [ISAPNP] Exclude memory regions for the
4146 Ranges are in pairs (memory base and size).
4148 ports= [IP_VS_FTP] IPVS ftp helper module
4150 Up to 8 (IP_VS_APP_MAX_PORTS) ports
4152 Format: <port>,<port>....
4154 powersave=off [PPC] This option disables power saving features.
4155 It specifically disables cpuidle and sets the
4156 platform machine description specific power_save
4157 function to NULL. On Idle the CPU just reduces
4160 ppc_strict_facility_enable
4161 [PPC] This option catches any kernel floating point,
4162 Altivec, VSX and SPE outside of regions specifically
4163 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
4164 There is some performance impact when enabling this.
4168 Disable Hardware Transactional Memory
4171 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
4172 none - Limited to cond_resched() calls
4173 voluntary - Limited to cond_resched() and might_sleep() calls
4174 full - Any section that isn't explicitly preempt disabled
4175 can be preempted anytime.
4177 print-fatal-signals=
4178 [KNL] debug: print fatal signals
4180 If enabled, warn about various signal handling
4181 related application anomalies: too many signals,
4182 too many POSIX.1 timers, fatal signals causing a
4185 If you hit the warning due to signal overflow,
4186 you might want to try "ulimit -i unlimited".
4190 printk.always_kmsg_dump=
4191 Trigger kmsg_dump for cases other than kernel oops or
4193 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4196 printk.console_no_auto_verbose=
4197 Disable console loglevel raise on oops, panic
4198 or lockdep-detected issues (only if lock debug is on).
4199 With an exception to setups with low baudrate on
4200 serial console, keeping this 0 is a good choice
4201 in order to provide more debug information.
4203 default: 0 (auto_verbose is enabled)
4205 printk.devkmsg={on,off,ratelimit}
4206 Control writing to /dev/kmsg.
4207 on - unlimited logging to /dev/kmsg from userspace
4208 off - logging to /dev/kmsg disabled
4209 ratelimit - ratelimit the logging
4212 printk.time= Show timing data prefixed to each printk message line
4213 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4215 processor.max_cstate= [HW,ACPI]
4216 Limit processor to maximum C-state
4217 max_cstate=9 overrides any DMI blacklist limit.
4219 processor.nocst [HW,ACPI]
4220 Ignore the _CST method to determine C-states,
4221 instead using the legacy FADT method
4223 profile= [KNL] Enable kernel profiling via /proc/profile
4224 Format: [<profiletype>,]<number>
4225 Param: <profiletype>: "schedule", "sleep", or "kvm"
4226 [defaults to kernel profiling]
4227 Param: "schedule" - profile schedule points.
4228 Param: "sleep" - profile D-state sleeping (millisecs).
4229 Requires CONFIG_SCHEDSTATS
4230 Param: "kvm" - profile VM exits.
4231 Param: <number> - step/bucket size as a power of 2 for
4232 statistical time based profiling.
4234 prompt_ramdisk= [RAM] [Deprecated]
4236 prot_virt= [S390] enable hosting protected virtual machines
4237 isolated from the hypervisor (if hardware supports
4241 psi= [KNL] Enable or disable pressure stall information
4245 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
4246 probe for; one of (bare|imps|exps|lifebook|any).
4247 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4249 psmouse.resetafter= [HW,MOUSE]
4250 Try to reset the device after so many bad packets
4253 [HW,MOUSE] Set desired mouse resolution, in dpi.
4254 psmouse.smartscroll=
4255 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4256 0 = disabled, 1 = enabled (default).
4258 pstore.backend= Specify the name of the pstore backend to use
4261 See Documentation/admin-guide/blockdev/paride.rst.
4263 pti= [X86-64] Control Page Table Isolation of user and
4264 kernel address spaces. Disabling this feature
4265 removes hardening, but improves performance of
4266 system calls and interrupts.
4268 on - unconditionally enable
4269 off - unconditionally disable
4270 auto - kernel detects whether your CPU model is
4271 vulnerable to issues that PTI mitigates
4273 Not specifying this option is equivalent to pti=auto.
4276 Equivalent to pti=off
4279 [KNL] Number of legacy pty's. Overwrites compiled-in
4282 quiet [KNL] Disable most log messages
4287 See Documentation/admin-guide/md.rst.
4289 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4290 See Documentation/admin-guide/blockdev/ramdisk.rst.
4292 ramdisk_start= [RAM] RAM disk image start address
4294 random.trust_cpu={on,off}
4295 [KNL] Enable or disable trusting the use of the
4296 CPU's random number generator (if available) to
4297 fully seed the kernel's CRNG. Default is controlled
4298 by CONFIG_RANDOM_TRUST_CPU.
4300 randomize_kstack_offset=
4301 [KNL] Enable or disable kernel stack offset
4302 randomization, which provides roughly 5 bits of
4303 entropy, frustrating memory corruption attacks
4304 that depend on stack address determinism or
4305 cross-syscall address exposures. This is only
4306 available on architectures that have defined
4307 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
4308 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4309 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
4311 ras=option[,option,...] [KNL] RAS-specific options
4314 Disable the Correctable Errors Collector,
4315 see CONFIG_RAS_CEC help text.
4318 The argument is a cpu list, as described above.
4320 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
4321 the specified list of CPUs to be no-callback CPUs.
4322 Invocation of these CPUs' RCU callbacks will be
4323 offloaded to "rcuox/N" kthreads created for that
4324 purpose, where "x" is "p" for RCU-preempt, and
4325 "s" for RCU-sched, and "N" is the CPU number.
4326 This reduces OS jitter on the offloaded CPUs,
4327 which can be useful for HPC and real-time
4328 workloads. It can also improve energy efficiency
4329 for asymmetric multiprocessors.
4332 Rather than requiring that offloaded CPUs
4333 (specified by rcu_nocbs= above) explicitly
4334 awaken the corresponding "rcuoN" kthreads,
4335 make these kthreads poll for callbacks.
4336 This improves the real-time response for the
4337 offloaded CPUs by relieving them of the need to
4338 wake up the corresponding kthread, but degrades
4339 energy efficiency by requiring that the kthreads
4340 periodically wake up to do the polling.
4342 rcutree.blimit= [KNL]
4343 Set maximum number of finished RCU callbacks to
4344 process in one batch.
4346 rcutree.dump_tree= [KNL]
4347 Dump the structure of the rcu_node combining tree
4348 out at early boot. This is used for diagnostic
4349 purposes, to verify correct tree setup.
4351 rcutree.gp_cleanup_delay= [KNL]
4352 Set the number of jiffies to delay each step of
4353 RCU grace-period cleanup.
4355 rcutree.gp_init_delay= [KNL]
4356 Set the number of jiffies to delay each step of
4357 RCU grace-period initialization.
4359 rcutree.gp_preinit_delay= [KNL]
4360 Set the number of jiffies to delay each step of
4361 RCU grace-period pre-initialization, that is,
4362 the propagation of recent CPU-hotplug changes up
4363 the rcu_node combining tree.
4365 rcutree.use_softirq= [KNL]
4366 If set to zero, move all RCU_SOFTIRQ processing to
4367 per-CPU rcuc kthreads. Defaults to a non-zero
4368 value, meaning that RCU_SOFTIRQ is used by default.
4369 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4371 But note that CONFIG_PREEMPT_RT=y kernels disable
4372 this kernel boot parameter, forcibly setting it
4375 rcutree.rcu_fanout_exact= [KNL]
4376 Disable autobalancing of the rcu_node combining
4377 tree. This is used by rcutorture, and might
4378 possibly be useful for architectures having high
4379 cache-to-cache transfer latencies.
4381 rcutree.rcu_fanout_leaf= [KNL]
4382 Change the number of CPUs assigned to each
4383 leaf rcu_node structure. Useful for very
4384 large systems, which will choose the value 64,
4385 and for NUMA systems with large remote-access
4386 latencies, which will choose a value aligned
4387 with the appropriate hardware boundaries.
4389 rcutree.rcu_min_cached_objs= [KNL]
4390 Minimum number of objects which are cached and
4391 maintained per one CPU. Object size is equal
4392 to PAGE_SIZE. The cache allows to reduce the
4393 pressure to page allocator, also it makes the
4394 whole algorithm to behave better in low memory
4397 rcutree.rcu_delay_page_cache_fill_msec= [KNL]
4398 Set the page-cache refill delay (in milliseconds)
4399 in response to low-memory conditions. The range
4400 of permitted values is in the range 0:100000.
4402 rcutree.jiffies_till_first_fqs= [KNL]
4403 Set delay from grace-period initialization to
4404 first attempt to force quiescent states.
4405 Units are jiffies, minimum value is zero,
4406 and maximum value is HZ.
4408 rcutree.jiffies_till_next_fqs= [KNL]
4409 Set delay between subsequent attempts to force
4410 quiescent states. Units are jiffies, minimum
4411 value is one, and maximum value is HZ.
4413 rcutree.jiffies_till_sched_qs= [KNL]
4414 Set required age in jiffies for a
4415 given grace period before RCU starts
4416 soliciting quiescent-state help from
4417 rcu_note_context_switch() and cond_resched().
4418 If not specified, the kernel will calculate
4419 a value based on the most recent settings
4420 of rcutree.jiffies_till_first_fqs
4421 and rcutree.jiffies_till_next_fqs.
4422 This calculated value may be viewed in
4423 rcutree.jiffies_to_sched_qs. Any attempt to set
4424 rcutree.jiffies_to_sched_qs will be cheerfully
4427 rcutree.kthread_prio= [KNL,BOOT]
4428 Set the SCHED_FIFO priority of the RCU per-CPU
4429 kthreads (rcuc/N). This value is also used for
4430 the priority of the RCU boost threads (rcub/N)
4431 and for the RCU grace-period kthreads (rcu_bh,
4432 rcu_preempt, and rcu_sched). If RCU_BOOST is
4433 set, valid values are 1-99 and the default is 1
4434 (the least-favored priority). Otherwise, when
4435 RCU_BOOST is not set, valid values are 0-99 and
4436 the default is zero (non-realtime operation).
4438 rcutree.rcu_nocb_gp_stride= [KNL]
4439 Set the number of NOCB callback kthreads in
4440 each group, which defaults to the square root
4441 of the number of CPUs. Larger numbers reduce
4442 the wakeup overhead on the global grace-period
4443 kthread, but increases that same overhead on
4444 each group's NOCB grace-period kthread.
4446 rcutree.qhimark= [KNL]
4447 Set threshold of queued RCU callbacks beyond which
4448 batch limiting is disabled.
4450 rcutree.qlowmark= [KNL]
4451 Set threshold of queued RCU callbacks below which
4452 batch limiting is re-enabled.
4454 rcutree.qovld= [KNL]
4455 Set threshold of queued RCU callbacks beyond which
4456 RCU's force-quiescent-state scan will aggressively
4457 enlist help from cond_resched() and sched IPIs to
4458 help CPUs more quickly reach quiescent states.
4459 Set to less than zero to make this be set based
4460 on rcutree.qhimark at boot time and to zero to
4461 disable more aggressive help enlistment.
4463 rcutree.rcu_idle_gp_delay= [KNL]
4464 Set wakeup interval for idle CPUs that have
4465 RCU callbacks (RCU_FAST_NO_HZ=y).
4467 rcutree.rcu_kick_kthreads= [KNL]
4468 Cause the grace-period kthread to get an extra
4469 wake_up() if it sleeps three times longer than
4470 it should at force-quiescent-state time.
4471 This wake_up() will be accompanied by a
4472 WARN_ONCE() splat and an ftrace_dump().
4474 rcutree.rcu_unlock_delay= [KNL]
4475 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4476 this specifies an rcu_read_unlock()-time delay
4477 in microseconds. This defaults to zero.
4478 Larger delays increase the probability of
4479 catching RCU pointer leaks, that is, buggy use
4480 of RCU-protected pointers after the relevant
4481 rcu_read_unlock() has completed.
4483 rcutree.sysrq_rcu= [KNL]
4484 Commandeer a sysrq key to dump out Tree RCU's
4485 rcu_node tree with an eye towards determining
4486 why a new grace period has not yet started.
4488 rcuscale.gp_async= [KNL]
4489 Measure performance of asynchronous
4490 grace-period primitives such as call_rcu().
4492 rcuscale.gp_async_max= [KNL]
4493 Specify the maximum number of outstanding
4494 callbacks per writer thread. When a writer
4495 thread exceeds this limit, it invokes the
4496 corresponding flavor of rcu_barrier() to allow
4497 previously posted callbacks to drain.
4499 rcuscale.gp_exp= [KNL]
4500 Measure performance of expedited synchronous
4501 grace-period primitives.
4503 rcuscale.holdoff= [KNL]
4504 Set test-start holdoff period. The purpose of
4505 this parameter is to delay the start of the
4506 test until boot completes in order to avoid
4509 rcuscale.kfree_rcu_test= [KNL]
4510 Set to measure performance of kfree_rcu() flooding.
4512 rcuscale.kfree_rcu_test_double= [KNL]
4513 Test the double-argument variant of kfree_rcu().
4514 If this parameter has the same value as
4515 rcuscale.kfree_rcu_test_single, both the single-
4516 and double-argument variants are tested.
4518 rcuscale.kfree_rcu_test_single= [KNL]
4519 Test the single-argument variant of kfree_rcu().
4520 If this parameter has the same value as
4521 rcuscale.kfree_rcu_test_double, both the single-
4522 and double-argument variants are tested.
4524 rcuscale.kfree_nthreads= [KNL]
4525 The number of threads running loops of kfree_rcu().
4527 rcuscale.kfree_alloc_num= [KNL]
4528 Number of allocations and frees done in an iteration.
4530 rcuscale.kfree_loops= [KNL]
4531 Number of loops doing rcuscale.kfree_alloc_num number
4532 of allocations and frees.
4534 rcuscale.nreaders= [KNL]
4535 Set number of RCU readers. The value -1 selects
4536 N, where N is the number of CPUs. A value
4537 "n" less than -1 selects N-n+1, where N is again
4538 the number of CPUs. For example, -2 selects N
4539 (the number of CPUs), -3 selects N+1, and so on.
4540 A value of "n" less than or equal to -N selects
4543 rcuscale.nwriters= [KNL]
4544 Set number of RCU writers. The values operate
4545 the same as for rcuscale.nreaders.
4546 N, where N is the number of CPUs
4548 rcuscale.perf_type= [KNL]
4549 Specify the RCU implementation to test.
4551 rcuscale.shutdown= [KNL]
4552 Shut the system down after performance tests
4553 complete. This is useful for hands-off automated
4556 rcuscale.verbose= [KNL]
4557 Enable additional printk() statements.
4559 rcuscale.writer_holdoff= [KNL]
4560 Write-side holdoff between grace periods,
4561 in microseconds. The default of zero says
4564 rcutorture.fqs_duration= [KNL]
4565 Set duration of force_quiescent_state bursts
4568 rcutorture.fqs_holdoff= [KNL]
4569 Set holdoff time within force_quiescent_state bursts
4572 rcutorture.fqs_stutter= [KNL]
4573 Set wait time between force_quiescent_state bursts
4576 rcutorture.fwd_progress= [KNL]
4577 Enable RCU grace-period forward-progress testing
4578 for the types of RCU supporting this notion.
4580 rcutorture.fwd_progress_div= [KNL]
4581 Specify the fraction of a CPU-stall-warning
4582 period to do tight-loop forward-progress testing.
4584 rcutorture.fwd_progress_holdoff= [KNL]
4585 Number of seconds to wait between successive
4586 forward-progress tests.
4588 rcutorture.fwd_progress_need_resched= [KNL]
4589 Enclose cond_resched() calls within checks for
4590 need_resched() during tight-loop forward-progress
4593 rcutorture.gp_cond= [KNL]
4594 Use conditional/asynchronous update-side
4595 primitives, if available.
4597 rcutorture.gp_exp= [KNL]
4598 Use expedited update-side primitives, if available.
4600 rcutorture.gp_normal= [KNL]
4601 Use normal (non-expedited) asynchronous
4602 update-side primitives, if available.
4604 rcutorture.gp_sync= [KNL]
4605 Use normal (non-expedited) synchronous
4606 update-side primitives, if available. If all
4607 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4608 rcutorture.gp_normal=, and rcutorture.gp_sync=
4609 are zero, rcutorture acts as if is interpreted
4610 they are all non-zero.
4612 rcutorture.irqreader= [KNL]
4613 Run RCU readers from irq handlers, or, more
4614 accurately, from a timer handler. Not all RCU
4615 flavors take kindly to this sort of thing.
4617 rcutorture.leakpointer= [KNL]
4618 Leak an RCU-protected pointer out of the reader.
4619 This can of course result in splats, and is
4620 intended to test the ability of things like
4621 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
4624 rcutorture.n_barrier_cbs= [KNL]
4625 Set callbacks/threads for rcu_barrier() testing.
4627 rcutorture.nfakewriters= [KNL]
4628 Set number of concurrent RCU writers. These just
4629 stress RCU, they don't participate in the actual
4630 test, hence the "fake".
4632 rcutorture.nocbs_nthreads= [KNL]
4633 Set number of RCU callback-offload togglers.
4634 Zero (the default) disables toggling.
4636 rcutorture.nocbs_toggle= [KNL]
4637 Set the delay in milliseconds between successive
4638 callback-offload toggling attempts.
4640 rcutorture.nreaders= [KNL]
4641 Set number of RCU readers. The value -1 selects
4642 N-1, where N is the number of CPUs. A value
4643 "n" less than -1 selects N-n-2, where N is again
4644 the number of CPUs. For example, -2 selects N
4645 (the number of CPUs), -3 selects N+1, and so on.
4647 rcutorture.object_debug= [KNL]
4648 Enable debug-object double-call_rcu() testing.
4650 rcutorture.onoff_holdoff= [KNL]
4651 Set time (s) after boot for CPU-hotplug testing.
4653 rcutorture.onoff_interval= [KNL]
4654 Set time (jiffies) between CPU-hotplug operations,
4655 or zero to disable CPU-hotplug testing.
4657 rcutorture.read_exit= [KNL]
4658 Set the number of read-then-exit kthreads used
4659 to test the interaction of RCU updaters and
4660 task-exit processing.
4662 rcutorture.read_exit_burst= [KNL]
4663 The number of times in a given read-then-exit
4664 episode that a set of read-then-exit kthreads
4667 rcutorture.read_exit_delay= [KNL]
4668 The delay, in seconds, between successive
4669 read-then-exit testing episodes.
4671 rcutorture.shuffle_interval= [KNL]
4672 Set task-shuffle interval (s). Shuffling tasks
4673 allows some CPUs to go into dyntick-idle mode
4674 during the rcutorture test.
4676 rcutorture.shutdown_secs= [KNL]
4677 Set time (s) after boot system shutdown. This
4678 is useful for hands-off automated testing.
4680 rcutorture.stall_cpu= [KNL]
4681 Duration of CPU stall (s) to test RCU CPU stall
4682 warnings, zero to disable.
4684 rcutorture.stall_cpu_block= [KNL]
4685 Sleep while stalling if set. This will result
4686 in warnings from preemptible RCU in addition
4687 to any other stall-related activity.
4689 rcutorture.stall_cpu_holdoff= [KNL]
4690 Time to wait (s) after boot before inducing stall.
4692 rcutorture.stall_cpu_irqsoff= [KNL]
4693 Disable interrupts while stalling if set.
4695 rcutorture.stall_gp_kthread= [KNL]
4696 Duration (s) of forced sleep within RCU
4697 grace-period kthread to test RCU CPU stall
4698 warnings, zero to disable. If both stall_cpu
4699 and stall_gp_kthread are specified, the
4700 kthread is starved first, then the CPU.
4702 rcutorture.stat_interval= [KNL]
4703 Time (s) between statistics printk()s.
4705 rcutorture.stutter= [KNL]
4706 Time (s) to stutter testing, for example, specifying
4707 five seconds causes the test to run for five seconds,
4708 wait for five seconds, and so on. This tests RCU's
4709 ability to transition abruptly to and from idle.
4711 rcutorture.test_boost= [KNL]
4712 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
4713 "Maybe" means test if the RCU implementation
4714 under test support RCU priority boosting.
4716 rcutorture.test_boost_duration= [KNL]
4717 Duration (s) of each individual boost test.
4719 rcutorture.test_boost_interval= [KNL]
4720 Interval (s) between each boost test.
4722 rcutorture.test_no_idle_hz= [KNL]
4723 Test RCU's dyntick-idle handling. See also the
4724 rcutorture.shuffle_interval parameter.
4726 rcutorture.torture_type= [KNL]
4727 Specify the RCU implementation to test.
4729 rcutorture.verbose= [KNL]
4730 Enable additional printk() statements.
4732 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
4733 Dump ftrace buffer after reporting RCU CPU
4736 rcupdate.rcu_cpu_stall_suppress= [KNL]
4737 Suppress RCU CPU stall warning messages.
4739 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
4740 Suppress RCU CPU stall warning messages and
4741 rcutorture writer stall warnings that occur
4742 during early boot, that is, during the time
4743 before the init task is spawned.
4745 rcupdate.rcu_cpu_stall_timeout= [KNL]
4746 Set timeout for RCU CPU stall warning messages.
4748 rcupdate.rcu_expedited= [KNL]
4749 Use expedited grace-period primitives, for
4750 example, synchronize_rcu_expedited() instead
4751 of synchronize_rcu(). This reduces latency,
4752 but can increase CPU utilization, degrade
4753 real-time latency, and degrade energy efficiency.
4754 No effect on CONFIG_TINY_RCU kernels.
4756 rcupdate.rcu_normal= [KNL]
4757 Use only normal grace-period primitives,
4758 for example, synchronize_rcu() instead of
4759 synchronize_rcu_expedited(). This improves
4760 real-time latency, CPU utilization, and
4761 energy efficiency, but can expose users to
4762 increased grace-period latency. This parameter
4763 overrides rcupdate.rcu_expedited. No effect on
4764 CONFIG_TINY_RCU kernels.
4766 rcupdate.rcu_normal_after_boot= [KNL]
4767 Once boot has completed (that is, after
4768 rcu_end_inkernel_boot() has been invoked), use
4769 only normal grace-period primitives. No effect
4770 on CONFIG_TINY_RCU kernels.
4772 But note that CONFIG_PREEMPT_RT=y kernels enables
4773 this kernel boot parameter, forcibly setting
4774 it to the value one, that is, converting any
4775 post-boot attempt at an expedited RCU grace
4776 period to instead use normal non-expedited
4777 grace-period processing.
4779 rcupdate.rcu_task_ipi_delay= [KNL]
4780 Set time in jiffies during which RCU tasks will
4781 avoid sending IPIs, starting with the beginning
4782 of a given grace period. Setting a large
4783 number avoids disturbing real-time workloads,
4784 but lengthens grace periods.
4786 rcupdate.rcu_task_stall_timeout= [KNL]
4787 Set timeout in jiffies for RCU task stall warning
4788 messages. Disable with a value less than or equal
4791 rcupdate.rcu_self_test= [KNL]
4792 Run the RCU early boot self tests
4796 Run specified binary instead of /init from the ramdisk,
4797 used for early userspace startup. See initrd.
4800 force - Override the decision by the kernel to hide the
4801 advertisement of RDRAND support (this affects
4802 certain AMD processors because of buggy BIOS
4803 support, specifically around the suspend/resume
4807 Turn on/off individual RDT features. List is:
4808 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
4810 E.g. to turn on cmt and turn off mba use:
4814 Format (x86 or x86_64):
4815 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \
4817 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
4819 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
4820 (prefix with 'panic_' to set mode for panic
4822 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
4823 reboot_force is either force or not specified,
4824 reboot_cpu is s[mp]#### with #### being the processor
4825 to be used for rebooting.
4827 refscale.holdoff= [KNL]
4828 Set test-start holdoff period. The purpose of
4829 this parameter is to delay the start of the
4830 test until boot completes in order to avoid
4833 refscale.loops= [KNL]
4834 Set the number of loops over the synchronization
4835 primitive under test. Increasing this number
4836 reduces noise due to loop start/end overhead,
4837 but the default has already reduced the per-pass
4838 noise to a handful of picoseconds on ca. 2020
4841 refscale.nreaders= [KNL]
4842 Set number of readers. The default value of -1
4843 selects N, where N is roughly 75% of the number
4844 of CPUs. A value of zero is an interesting choice.
4846 refscale.nruns= [KNL]
4847 Set number of runs, each of which is dumped onto
4850 refscale.readdelay= [KNL]
4851 Set the read-side critical-section duration,
4852 measured in microseconds.
4854 refscale.scale_type= [KNL]
4855 Specify the read-protection implementation to test.
4857 refscale.shutdown= [KNL]
4858 Shut down the system at the end of the performance
4859 test. This defaults to 1 (shut it down) when
4860 refscale is built into the kernel and to 0 (leave
4861 it running) when refscale is built as a module.
4863 refscale.verbose= [KNL]
4864 Enable additional printk() statements.
4866 refscale.verbose_batched= [KNL]
4867 Batch the additional printk() statements. If zero
4868 (the default) or negative, print everything. Otherwise,
4869 print every Nth verbose statement, where N is the value
4873 [KNL, SMP] Set scheduler's default relax_domain_level.
4874 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
4876 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
4877 Format: <base1>,<size1>[,<base2>,<size2>,...]
4878 Reserve I/O ports or memory so the kernel won't use
4879 them. If <base> is less than 0x10000, the region
4880 is assumed to be I/O ports; otherwise it is memory.
4882 reservetop= [X86-32]
4884 Reserves a hole at the top of the kernel virtual
4887 reset_devices [KNL] Force drivers to reset the underlying device
4888 during initialization.
4891 Specify the partition device for software suspend
4893 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
4895 resume_offset= [SWSUSP]
4896 Specify the offset from the beginning of the partition
4897 given by "resume=" at which the swap header is located,
4898 in <PAGE_SIZE> units (needed only for swap files).
4899 See Documentation/power/swsusp-and-swap-files.rst
4901 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
4902 read the resume files
4904 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
4905 Useful for devices that are detected asynchronously
4906 (e.g. USB and MMC devices).
4908 hibernate= [HIBERNATION]
4909 noresume Don't check if there's a hibernation image
4910 present during boot.
4911 nocompress Don't compress/decompress hibernation images.
4912 no Disable hibernation and resume.
4913 protect_image Turn on image protection during restoration
4914 (that will set all pages holding image data
4915 during restoration read-only).
4917 retain_initrd [RAM] Keep initrd memory after extraction
4919 rfkill.default_state=
4920 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
4921 etc. communication is blocked by default.
4924 rfkill.master_switch_mode=
4925 0 The "airplane mode" button does nothing.
4926 1 The "airplane mode" button toggles between everything
4927 blocked and the previous configuration.
4928 2 The "airplane mode" button toggles between everything
4929 blocked and everything unblocked.
4931 rhash_entries= [KNL,NET]
4932 Set number of hash buckets for route cache
4935 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
4938 ro [KNL] Mount root device read-only on boot
4941 on Mark read-only kernel memory as read-only (default).
4942 off Leave read-only kernel memory writable for debugging.
4945 Enable the uart passthrough on the designated usb port
4946 on Rockchip SoCs. When active, the signals of the
4947 debug-uart get routed to the D+ and D- pins of the usb
4948 port and the regular usb controller gets disabled.
4950 root= [KNL] Root filesystem
4951 See name_to_dev_t comment in init/do_mounts.c.
4953 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
4954 mount the root filesystem
4956 rootflags= [KNL] Set root filesystem mount option string
4958 rootfstype= [KNL] Set root filesystem type
4960 rootwait [KNL] Wait (indefinitely) for root device to show up.
4961 Useful for devices that are detected asynchronously
4962 (e.g. USB and MMC devices).
4964 rproc_mem=nn[KMG][@address]
4965 [KNL,ARM,CMA] Remoteproc physical memory block.
4966 Memory area to be used by remote processor image,
4969 rw [KNL] Mount root device read-write on boot
4971 S [KNL] Run init in single mode
4973 s390_iommu= [HW,S390]
4974 Set s390 IOTLB flushing mode
4976 With strict flushing every unmap operation will result in
4977 an IOTLB flush. Default is lazy flushing before reuse,
4981 See drivers/net/irda/sa1100_ir.c.
4983 sched_verbose [KNL] Enables verbose scheduler debug messages.
4985 schedstats= [KNL,X86] Enable or disable scheduled statistics.
4986 Allowed values are enable and disable. This feature
4987 incurs a small amount of overhead in the scheduler
4988 but is useful for debugging and performance tuning.
4990 sched_thermal_decay_shift=
4991 [KNL, SMP] Set a decay shift for scheduler thermal
4992 pressure signal. Thermal pressure signal follows the
4993 default decay period of other scheduler pelt
4994 signals(usually 32 ms but configurable). Setting
4995 sched_thermal_decay_shift will left shift the decay
4996 period for the thermal pressure signal by the shift
4998 i.e. with the default pelt decay period of 32 ms
4999 sched_thermal_decay_shift thermal pressure decay pr
5003 Format: integer between 0 and 10
5006 scftorture.holdoff= [KNL]
5007 Number of seconds to hold off before starting
5008 test. Defaults to zero for module insertion and
5009 to 10 seconds for built-in smp_call_function()
5012 scftorture.longwait= [KNL]
5013 Request ridiculously long waits randomly selected
5014 up to the chosen limit in seconds. Zero (the
5015 default) disables this feature. Please note
5016 that requesting even small non-zero numbers of
5017 seconds can result in RCU CPU stall warnings,
5018 softlockup complaints, and so on.
5020 scftorture.nthreads= [KNL]
5021 Number of kthreads to spawn to invoke the
5022 smp_call_function() family of functions.
5023 The default of -1 specifies a number of kthreads
5024 equal to the number of CPUs.
5026 scftorture.onoff_holdoff= [KNL]
5027 Number seconds to wait after the start of the
5028 test before initiating CPU-hotplug operations.
5030 scftorture.onoff_interval= [KNL]
5031 Number seconds to wait between successive
5032 CPU-hotplug operations. Specifying zero (which
5033 is the default) disables CPU-hotplug operations.
5035 scftorture.shutdown_secs= [KNL]
5036 The number of seconds following the start of the
5037 test after which to shut down the system. The
5038 default of zero avoids shutting down the system.
5039 Non-zero values are useful for automated tests.
5041 scftorture.stat_interval= [KNL]
5042 The number of seconds between outputting the
5043 current test statistics to the console. A value
5044 of zero disables statistics output.
5046 scftorture.stutter_cpus= [KNL]
5047 The number of jiffies to wait between each change
5048 to the set of CPUs under test.
5050 scftorture.use_cpus_read_lock= [KNL]
5051 Use use_cpus_read_lock() instead of the default
5052 preempt_disable() to disable CPU hotplug
5053 while invoking one of the smp_call_function*()
5056 scftorture.verbose= [KNL]
5057 Enable additional printk() statements.
5059 scftorture.weight_single= [KNL]
5060 The probability weighting to use for the
5061 smp_call_function_single() function with a zero
5062 "wait" parameter. A value of -1 selects the
5063 default if all other weights are -1. However,
5064 if at least one weight has some other value, a
5065 value of -1 will instead select a weight of zero.
5067 scftorture.weight_single_wait= [KNL]
5068 The probability weighting to use for the
5069 smp_call_function_single() function with a
5070 non-zero "wait" parameter. See weight_single.
5072 scftorture.weight_many= [KNL]
5073 The probability weighting to use for the
5074 smp_call_function_many() function with a zero
5075 "wait" parameter. See weight_single.
5076 Note well that setting a high probability for
5077 this weighting can place serious IPI load
5080 scftorture.weight_many_wait= [KNL]
5081 The probability weighting to use for the
5082 smp_call_function_many() function with a
5083 non-zero "wait" parameter. See weight_single
5086 scftorture.weight_all= [KNL]
5087 The probability weighting to use for the
5088 smp_call_function_all() function with a zero
5089 "wait" parameter. See weight_single and
5092 scftorture.weight_all_wait= [KNL]
5093 The probability weighting to use for the
5094 smp_call_function_all() function with a
5095 non-zero "wait" parameter. See weight_single
5098 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
5099 xtime_lock contention on larger systems, and/or RCU lock
5100 contention on all systems with CONFIG_MAXSMP set.
5101 Format: { "0" | "1" }
5102 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
5104 Note: increases power consumption, thus should only be
5105 enabled if running jitter sensitive (HPC/RT) workloads.
5107 security= [SECURITY] Choose a legacy "major" security module to
5108 enable at boot. This has been deprecated by the
5111 selinux= [SELINUX] Disable or enable SELinux at boot time.
5112 Format: { "0" | "1" }
5113 See security/selinux/Kconfig help text.
5118 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
5119 Format: { "0" | "1" }
5120 See security/apparmor/Kconfig help text
5123 Default value is set via kernel config option.
5125 serialnumber [BUGS=X86-32]
5128 Maximal number of shapers.
5136 Enable merging of slabs with similar size when the
5137 kernel is built without CONFIG_SLAB_MERGE_DEFAULT.
5140 Disable merging of slabs with similar size. May be
5141 necessary if there is some reason to distinguish
5142 allocs to different slabs, especially in hardened
5143 environments where the risk of heap overflows and
5144 layout control by attackers can usually be
5145 frustrated by disabling merging. This will reduce
5146 most of the exposure of a heap attack to a single
5147 cache (risks via metadata attacks are mostly
5148 unchanged). Debug options disable merging on their
5150 For more information see Documentation/vm/slub.rst.
5152 slab_max_order= [MM, SLAB]
5153 Determines the maximum allowed order for slabs.
5154 A high setting may cause OOMs due to memory
5155 fragmentation. Defaults to 1 for systems with
5156 more than 32MB of RAM, 0 otherwise.
5158 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
5159 Enabling slub_debug allows one to determine the
5160 culprit if slab objects become corrupted. Enabling
5161 slub_debug can create guard zones around objects and
5162 may poison objects when not in use. Also tracks the
5163 last alloc / free. For more information see
5164 Documentation/vm/slub.rst.
5166 slub_max_order= [MM, SLUB]
5167 Determines the maximum allowed order for slabs.
5168 A high setting may cause OOMs due to memory
5169 fragmentation. For more information see
5170 Documentation/vm/slub.rst.
5172 slub_min_objects= [MM, SLUB]
5173 The minimum number of objects per slab. SLUB will
5174 increase the slab order up to slub_max_order to
5175 generate a sufficiently large slab able to contain
5176 the number of objects indicated. The higher the number
5177 of objects the smaller the overhead of tracking slabs
5178 and the less frequently locks need to be acquired.
5179 For more information see Documentation/vm/slub.rst.
5181 slub_min_order= [MM, SLUB]
5182 Determines the minimum page order for slabs. Must be
5183 lower than slub_max_order.
5184 For more information see Documentation/vm/slub.rst.
5186 slub_merge [MM, SLUB]
5187 Same with slab_merge.
5189 slub_nomerge [MM, SLUB]
5190 Same with slab_nomerge. This is supported for legacy.
5191 See slab_nomerge for more information.
5194 Format: <io1>[,<io2>[,...,<io8>]]
5196 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
5197 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
5198 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
5199 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
5200 smsc-ircc2.ircc_irq= [HW] IRQ line
5201 smsc-ircc2.ircc_dma= [HW] DMA channel
5202 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
5203 0: Toshiba Satellite 1800 (GP data pin select)
5204 1: Fast pin select (default)
5207 smt [KNL,S390] Set the maximum number of threads (logical
5208 CPUs) to use per physical CPU on systems capable of
5209 symmetric multithreading (SMT). Will be capped to the
5210 actual hardware limit.
5212 Default: -1 (no limit)
5215 [KNL] Should the soft-lockup detector generate panics.
5218 A value of 1 instructs the soft-lockup detector
5219 to panic the machine when a soft-lockup occurs. It is
5220 also controlled by the kernel.softlockup_panic sysctl
5221 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
5222 respective build-time switch to that functionality.
5224 softlockup_all_cpu_backtrace=
5225 [KNL] Should the soft-lockup detector generate
5226 backtraces on all cpus.
5229 sonypi.*= [HW] Sony Programmable I/O Control Device driver
5230 See Documentation/admin-guide/laptops/sonypi.rst
5232 spectre_v2= [X86] Control mitigation of Spectre variant 2
5233 (indirect branch speculation) vulnerability.
5234 The default operation protects the kernel from
5237 on - unconditionally enable, implies
5239 off - unconditionally disable, implies
5241 auto - kernel detects whether your CPU model is
5244 Selecting 'on' will, and 'auto' may, choose a
5245 mitigation method at run time according to the
5246 CPU, the available microcode, the setting of the
5247 CONFIG_RETPOLINE configuration option, and the
5248 compiler with which the kernel was built.
5250 Selecting 'on' will also enable the mitigation
5251 against user space to user space task attacks.
5253 Selecting 'off' will disable both the kernel and
5254 the user space protections.
5256 Specific mitigations can also be selected manually:
5258 retpoline - replace indirect branches
5259 retpoline,generic - google's original retpoline
5260 retpoline,amd - AMD-specific minimal thunk
5262 Not specifying this option is equivalent to
5266 [X86] Control mitigation of Spectre variant 2
5267 (indirect branch speculation) vulnerability between
5270 on - Unconditionally enable mitigations. Is
5271 enforced by spectre_v2=on
5273 off - Unconditionally disable mitigations. Is
5274 enforced by spectre_v2=off
5276 prctl - Indirect branch speculation is enabled,
5277 but mitigation can be enabled via prctl
5278 per thread. The mitigation control state
5279 is inherited on fork.
5282 - Like "prctl" above, but only STIBP is
5283 controlled per thread. IBPB is issued
5284 always when switching between different user
5288 - Same as "prctl" above, but all seccomp
5289 threads will enable the mitigation unless
5290 they explicitly opt out.
5293 - Like "seccomp" above, but only STIBP is
5294 controlled per thread. IBPB is issued
5295 always when switching between different
5296 user space processes.
5298 auto - Kernel selects the mitigation depending on
5299 the available CPU features and vulnerability.
5302 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl"
5304 Not specifying this option is equivalent to
5305 spectre_v2_user=auto.
5307 spec_store_bypass_disable=
5308 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
5309 (Speculative Store Bypass vulnerability)
5311 Certain CPUs are vulnerable to an exploit against a
5312 a common industry wide performance optimization known
5313 as "Speculative Store Bypass" in which recent stores
5314 to the same memory location may not be observed by
5315 later loads during speculative execution. The idea
5316 is that such stores are unlikely and that they can
5317 be detected prior to instruction retirement at the
5318 end of a particular speculation execution window.
5320 In vulnerable processors, the speculatively forwarded
5321 store can be used in a cache side channel attack, for
5322 example to read memory to which the attacker does not
5323 directly have access (e.g. inside sandboxed code).
5325 This parameter controls whether the Speculative Store
5326 Bypass optimization is used.
5328 On x86 the options are:
5330 on - Unconditionally disable Speculative Store Bypass
5331 off - Unconditionally enable Speculative Store Bypass
5332 auto - Kernel detects whether the CPU model contains an
5333 implementation of Speculative Store Bypass and
5334 picks the most appropriate mitigation. If the
5335 CPU is not vulnerable, "off" is selected. If the
5336 CPU is vulnerable the default mitigation is
5337 architecture and Kconfig dependent. See below.
5338 prctl - Control Speculative Store Bypass per thread
5339 via prctl. Speculative Store Bypass is enabled
5340 for a process by default. The state of the control
5341 is inherited on fork.
5342 seccomp - Same as "prctl" above, but all seccomp threads
5343 will disable SSB unless they explicitly opt out.
5345 Default mitigations:
5346 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
5348 On powerpc the options are:
5350 on,auto - On Power8 and Power9 insert a store-forwarding
5351 barrier on kernel entry and exit. On Power7
5352 perform a software flush on kernel entry and
5356 Not specifying this option is equivalent to
5357 spec_store_bypass_disable=auto.
5359 spia_io_base= [HW,MTD]
5365 [X86] Enable split lock detection or bus lock detection
5367 When enabled (and if hardware support is present), atomic
5368 instructions that access data across cache line
5369 boundaries will result in an alignment check exception
5370 for split lock detection or a debug exception for
5375 warn - the kernel will emit rate-limited warnings
5376 about applications triggering the #AC
5377 exception or the #DB exception. This mode is
5378 the default on CPUs that support split lock
5379 detection or bus lock detection. Default
5380 behavior is by #AC if both features are
5381 enabled in hardware.
5383 fatal - the kernel will send SIGBUS to applications
5384 that trigger the #AC exception or the #DB
5385 exception. Default behavior is by #AC if
5386 both features are enabled in hardware.
5389 Set system wide rate limit to N bus locks
5390 per second for bus lock detection.
5393 N/A for split lock detection.
5396 If an #AC exception is hit in the kernel or in
5397 firmware (i.e. not while executing in user mode)
5398 the kernel will oops in either "warn" or "fatal"
5401 #DB exception for bus lock is triggered only when
5405 Control the Special Register Buffer Data Sampling
5408 Certain CPUs are vulnerable to an MDS-like
5409 exploit which can leak bits from the random
5412 By default, this issue is mitigated by
5413 microcode. However, the microcode fix can cause
5414 the RDRAND and RDSEED instructions to become
5415 much slower. Among other effects, this will
5416 result in reduced throughput from /dev/urandom.
5418 The microcode mitigation can be disabled with
5419 the following option:
5421 off: Disable mitigation and remove
5422 performance impact to RDRAND and RDSEED
5424 srcutree.counter_wrap_check [KNL]
5425 Specifies how frequently to check for
5426 grace-period sequence counter wrap for the
5427 srcu_data structure's ->srcu_gp_seq_needed field.
5428 The greater the number of bits set in this kernel
5429 parameter, the less frequently counter wrap will
5430 be checked for. Note that the bottom two bits
5433 srcutree.exp_holdoff [KNL]
5434 Specifies how many nanoseconds must elapse
5435 since the end of the last SRCU grace period for
5436 a given srcu_struct until the next normal SRCU
5437 grace period will be considered for automatic
5438 expediting. Set to zero to disable automatic
5442 Speculative Store Bypass Disable control
5444 On CPUs that are vulnerable to the Speculative
5445 Store Bypass vulnerability and offer a
5446 firmware based mitigation, this parameter
5447 indicates how the mitigation should be used:
5449 force-on: Unconditionally enable mitigation for
5450 for both kernel and userspace
5451 force-off: Unconditionally disable mitigation for
5452 for both kernel and userspace
5453 kernel: Always enable mitigation in the
5454 kernel, and offer a prctl interface
5455 to allow userspace to register its
5456 interest in being mitigated too.
5458 stack_guard_gap= [MM]
5459 override the default stack gap protection. The value
5460 is in page units and it defines how many pages prior
5461 to (for stacks growing down) resp. after (for stacks
5462 growing up) the main stack are reserved for no other
5463 mapping. Default value is 256 pages.
5465 stack_depot_disable= [KNL]
5466 Setting this to true through kernel command line will
5467 disable the stack depot thereby saving the static memory
5468 consumed by the stack hash table. By default this is set
5472 Enabled the stack tracer on boot up.
5474 stacktrace_filter=[function-list]
5475 [FTRACE] Limit the functions that the stack tracer
5476 will trace at boot up. function-list is a comma-separated
5477 list of functions. This list can be changed at run
5478 time by the stack_trace_filter file in the debugfs
5479 tracing directory. Note, this enables stack tracing
5480 and the stacktrace above is not needed.
5484 Set the STI (builtin display/keyboard on the HP-PARISC
5485 machines) console (graphic card) which should be used
5486 as the initial boot-console.
5487 See also comment in drivers/video/console/sticore.c.
5490 See comment in drivers/video/console/sticore.c.
5493 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
5495 sunrpc.min_resvport=
5496 sunrpc.max_resvport=
5498 SunRPC servers often require that client requests
5499 originate from a privileged port (i.e. a port in the
5500 range 0 < portnr < 1024).
5501 An administrator who wishes to reserve some of these
5502 ports for other uses may adjust the range that the
5503 kernel's sunrpc client considers to be privileged
5504 using these two parameters to set the minimum and
5505 maximum port values.
5507 sunrpc.svc_rpc_per_connection_limit=
5509 Limit the number of requests that the server will
5510 process in parallel from a single connection.
5511 The default value is 0 (no limit).
5515 Control how the NFS server code allocates CPUs to
5516 service thread pools. Depending on how many NICs
5517 you have and where their interrupts are bound, this
5518 option will affect which CPUs will do NFS serving.
5519 Note: this parameter cannot be changed while the
5520 NFS server is running.
5522 auto the server chooses an appropriate mode
5523 automatically using heuristics
5524 global a single global pool contains all CPUs
5525 percpu one pool for each CPU
5526 pernode one pool for each NUMA node (equivalent
5527 to global on non-NUMA machines)
5529 sunrpc.tcp_slot_table_entries=
5530 sunrpc.udp_slot_table_entries=
5532 Sets the upper limit on the number of simultaneous
5533 RPC calls that can be sent from the client to a
5534 server. Increasing these values may allow you to
5535 improve throughput, but will also increase the
5536 amount of memory reserved for use by the client.
5538 suspend.pm_test_delay=
5540 Sets the number of seconds to remain in a suspend test
5541 mode before resuming the system (see
5542 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
5543 is set. Default value is 5.
5546 Format: { on | off | y | n | 1 | 0 }
5547 This parameter controls use of the Protected
5548 Execution Facility on pSeries.
5551 [KNL] Enable accounting of swap in memory resource
5552 controller if no parameter or 1 is given or disable
5553 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
5555 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
5556 Format: { <int> | force | noforce }
5557 <int> -- Number of I/O TLB slabs
5558 force -- force using of bounce buffers even if they
5559 wouldn't be automatically used by the kernel
5560 noforce -- Never use bounce buffers (for debugging)
5565 Set a sysctl parameter, right before loading the init
5566 process, as if the value was written to the respective
5567 /proc/sys/... file. Both '.' and '/' are recognized as
5568 separators. Unrecognized parameters and invalid values
5569 are reported in the kernel log. Sysctls registered
5570 later by a loaded module cannot be set this way.
5571 Example: sysctl.vm.swappiness=40
5573 sysfs.deprecated=0|1 [KNL]
5574 Enable/disable old style sysfs layout for old udev
5575 on older distributions. When this option is enabled
5576 very new udev will not work anymore. When this option
5577 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
5578 in older udev will not work anymore.
5579 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
5580 the kernel configuration.
5582 sysrq_always_enabled
5584 Ignore sysrq setting - this boot parameter will
5585 neutralize any effect of /proc/sys/kernel/sysrq.
5586 Useful for debugging.
5588 tcpmhash_entries= [KNL,NET]
5589 Set the number of tcp_metrics_hash slots.
5590 Default value is 8192 or 16384 depending on total
5591 ram pages. This is used to specify the TCP metrics
5592 cache size. See Documentation/networking/ip-sysctl.rst
5593 "tcp_no_metrics_save" section for more details.
5597 test_suspend= [SUSPEND][,N]
5598 Specify "mem" (for Suspend-to-RAM) or "standby" (for
5599 standby suspend) or "freeze" (for suspend type freeze)
5600 as the system sleep state during system startup with
5601 the optional capability to repeat N number of times.
5602 The system is woken from this state using a
5603 wakeup-capable RTC alarm.
5605 thash_entries= [KNL,NET]
5606 Set number of hash buckets for TCP connection
5608 thermal.act= [HW,ACPI]
5609 -1: disable all active trip points in all thermal zones
5610 <degrees C>: override all lowest active trip points
5612 thermal.crt= [HW,ACPI]
5613 -1: disable all critical trip points in all thermal zones
5614 <degrees C>: override all critical trip points
5616 thermal.nocrt= [HW,ACPI]
5617 Set to disable actions on ACPI thermal zone
5618 critical and hot trip points.
5620 thermal.off= [HW,ACPI]
5621 1: disable ACPI thermal control
5623 thermal.psv= [HW,ACPI]
5624 -1: disable all passive trip points
5625 <degrees C>: override all passive trip points to this
5628 thermal.tzp= [HW,ACPI]
5629 Specify global default ACPI thermal zone polling rate
5630 <deci-seconds>: poll all this frequency
5631 0: no polling (default)
5634 Force threading of all interrupt handlers except those
5635 marked explicitly IRQF_NO_THREAD.
5639 Specify if the kernel should make use of the cpu
5640 topology information if the hardware supports this.
5641 The scheduler will make use of this information and
5642 e.g. base its process migration decisions on it.
5645 topology_updates= [KNL, PPC, NUMA]
5647 Specify if the kernel should ignore (off)
5648 topology updates sent by the hypervisor to this
5651 torture.disable_onoff_at_boot= [KNL]
5652 Prevent the CPU-hotplug component of torturing
5653 until after init has spawned.
5655 torture.ftrace_dump_at_shutdown= [KNL]
5656 Dump the ftrace buffer at torture-test shutdown,
5657 even if there were no errors. This can be a
5658 very costly operation when many torture tests
5659 are running concurrently, especially on systems
5660 with rotating-rust storage.
5662 torture.verbose_sleep_frequency= [KNL]
5663 Specifies how many verbose printk()s should be
5664 emitted between each sleep. The default of zero
5665 disables verbose-printk() sleeping.
5667 torture.verbose_sleep_duration= [KNL]
5668 Duration of each verbose-printk() sleep in jiffies.
5672 tpm_suspend_pcr=[HW,TPM]
5673 Format: integer pcr id
5674 Specify that at suspend time, the tpm driver
5675 should extend the specified pcr with zeros,
5676 as a workaround for some chips which fail to
5677 flush the last written pcr on TPM_SaveState.
5678 This will guarantee that all the other pcrs
5681 trace_buf_size=nn[KMG]
5682 [FTRACE] will set tracing buffer size on each cpu.
5684 trace_event=[event-list]
5685 [FTRACE] Set and start specified trace events in order
5686 to facilitate early boot debugging. The event-list is a
5687 comma-separated list of trace events to enable. See
5688 also Documentation/trace/events.rst
5690 trace_options=[option-list]
5691 [FTRACE] Enable or disable tracer options at boot.
5692 The option-list is a comma delimited list of options
5693 that can be enabled or disabled just as if you were
5694 to echo the option name into
5696 /sys/kernel/debug/tracing/trace_options
5698 For example, to enable stacktrace option (to dump the
5699 stack trace of each event), add to the command line:
5701 trace_options=stacktrace
5703 See also Documentation/trace/ftrace.rst "trace options"
5707 Have the tracepoints sent to printk as well as the
5708 tracing ring buffer. This is useful for early boot up
5709 where the system hangs or reboots and does not give the
5710 option for reading the tracing buffer or performing a
5711 ftrace_dump_on_oops.
5713 To turn off having tracepoints sent to printk,
5714 echo 0 > /proc/sys/kernel/tracepoint_printk
5715 Note, echoing 1 into this file without the
5716 tracepoint_printk kernel cmdline option has no effect.
5718 The tp_printk_stop_on_boot (see below) can also be used
5719 to stop the printing of events to console at
5724 Having tracepoints sent to printk() and activating high
5725 frequency tracepoints such as irq or sched, can cause
5726 the system to live lock.
5728 tp_printk_stop_on_boot[FTRACE]
5729 When tp_printk (above) is set, it can cause a lot of noise
5730 on the console. It may be useful to only include the
5731 printing of events during boot up, as user space may
5732 make the system inoperable.
5734 This command line option will stop the printing of events
5735 to console at the late_initcall_sync() time frame.
5738 [FTRACE] enable this option to disable tracing when a
5739 warning is hit. This turns off "tracing_on". Tracing can
5740 be enabled again by echoing '1' into the "tracing_on"
5741 file located in /sys/kernel/debug/tracing/
5743 This option is useful, as it disables the trace before
5744 the WARNING dump is called, which prevents the trace to
5745 be filled with content caused by the warning output.
5747 This option can also be set at run time via the sysctl
5748 option: kernel/traceoff_on_warning
5750 transparent_hugepage=
5752 Format: [always|madvise|never]
5753 Can be used to control the default behavior of the system
5754 with respect to transparent hugepages.
5755 See Documentation/admin-guide/mm/transhuge.rst
5758 trusted.source= [KEYS]
5760 This parameter identifies the trust source as a backend
5761 for trusted keys implementation. Supported trust
5765 If not specified then it defaults to iterating through
5766 the trust source list starting with TPM and assigns the
5767 first trust source as a backend which is initialized
5768 successfully during iteration.
5770 tsc= Disable clocksource stability checks for TSC.
5772 [x86] reliable: mark tsc clocksource as reliable, this
5773 disables clocksource verification at runtime, as well
5774 as the stability checks done at bootup. Used to enable
5775 high-resolution timer mode on older hardware, and in
5776 virtualized environment.
5777 [x86] noirqtime: Do not use TSC to do irq accounting.
5778 Used to run time disable IRQ_TIME_ACCOUNTING on any
5779 platforms where RDTSC is slow and this accounting
5781 [x86] unstable: mark the TSC clocksource as unstable, this
5782 marks the TSC unconditionally unstable at bootup and
5783 avoids any further wobbles once the TSC watchdog notices.
5784 [x86] nowatchdog: disable clocksource watchdog. Used
5785 in situations with strict latency requirements (where
5786 interruptions from clocksource watchdog are not
5789 tsc_early_khz= [X86] Skip early TSC calibration and use the given
5790 value instead. Useful when the early TSC frequency discovery
5791 procedure is not reliable, such as on overclocked systems
5792 with CPUID.16h support and partial CPUID.15h support.
5793 Format: <unsigned int>
5795 tsx= [X86] Control Transactional Synchronization
5796 Extensions (TSX) feature in Intel processors that
5797 support TSX control.
5799 This parameter controls the TSX feature. The options are:
5801 on - Enable TSX on the system. Although there are
5802 mitigations for all known security vulnerabilities,
5803 TSX has been known to be an accelerator for
5804 several previous speculation-related CVEs, and
5805 so there may be unknown security risks associated
5806 with leaving it enabled.
5808 off - Disable TSX on the system. (Note that this
5809 option takes effect only on newer CPUs which are
5810 not vulnerable to MDS, i.e., have
5811 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
5812 the new IA32_TSX_CTRL MSR through a microcode
5813 update. This new MSR allows for the reliable
5814 deactivation of the TSX functionality.)
5816 auto - Disable TSX if X86_BUG_TAA is present,
5817 otherwise enable TSX on the system.
5819 Not specifying this option is equivalent to tsx=off.
5821 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5824 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
5825 Abort (TAA) vulnerability.
5827 Similar to Micro-architectural Data Sampling (MDS)
5828 certain CPUs that support Transactional
5829 Synchronization Extensions (TSX) are vulnerable to an
5830 exploit against CPU internal buffers which can forward
5831 information to a disclosure gadget under certain
5834 In vulnerable processors, the speculatively forwarded
5835 data can be used in a cache side channel attack, to
5836 access data to which the attacker does not have direct
5839 This parameter controls the TAA mitigation. The
5842 full - Enable TAA mitigation on vulnerable CPUs
5845 full,nosmt - Enable TAA mitigation and disable SMT on
5846 vulnerable CPUs. If TSX is disabled, SMT
5847 is not disabled because CPU is not
5848 vulnerable to cross-thread TAA attacks.
5849 off - Unconditionally disable TAA mitigation
5851 On MDS-affected machines, tsx_async_abort=off can be
5852 prevented by an active MDS mitigation as both vulnerabilities
5853 are mitigated with the same mechanism so in order to disable
5854 this mitigation, you need to specify mds=off too.
5856 Not specifying this option is equivalent to
5857 tsx_async_abort=full. On CPUs which are MDS affected
5858 and deploy MDS mitigation, TAA mitigation is not
5859 required and doesn't provide any additional
5863 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5865 turbografx.map[2|3]= [HW,JOY]
5866 TurboGraFX parallel port interface
5868 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
5869 See also Documentation/input/devices/joystick-parport.rst
5871 udbg-immortal [PPC] When debugging early kernel crashes that
5872 happen after console_init() and before a proper
5873 console driver takes over, this boot options might
5874 help "seeing" what's going on.
5876 uhash_entries= [KNL,NET]
5877 Set number of hash buckets for UDP/UDP-Lite connections
5880 [USB] Ignore overcurrent events (default N).
5881 Some badly-designed motherboards generate lots of
5882 bogus events, for ports that aren't wired to
5883 anything. Set this parameter to avoid log spamming.
5884 Note that genuine overcurrent events won't be
5888 [X86] Cause panic on unknown NMI.
5890 usbcore.authorized_default=
5891 [USB] Default USB device authorization:
5892 (default -1 = authorized except for wireless USB,
5893 0 = not authorized, 1 = authorized, 2 = authorized
5894 if device connected to internal port)
5896 usbcore.autosuspend=
5897 [USB] The autosuspend time delay (in seconds) used
5898 for newly-detected USB devices (default 2). This
5899 is the time required before an idle device will be
5900 autosuspended. Devices for which the delay is set
5901 to a negative value won't be autosuspended at all.
5903 usbcore.usbfs_snoop=
5904 [USB] Set to log all usbfs traffic (default 0 = off).
5906 usbcore.usbfs_snoop_max=
5907 [USB] Maximum number of bytes to snoop in each URB
5910 usbcore.blinkenlights=
5911 [USB] Set to cycle leds on hubs (default 0 = off).
5913 usbcore.old_scheme_first=
5914 [USB] Start with the old device initialization
5915 scheme (default 0 = off).
5917 usbcore.usbfs_memory_mb=
5918 [USB] Memory limit (in MB) for buffers allocated by
5919 usbfs (default = 16, 0 = max = 2047).
5921 usbcore.use_both_schemes=
5922 [USB] Try the other device initialization scheme
5923 if the first one fails (default 1 = enabled).
5925 usbcore.initial_descriptor_timeout=
5926 [USB] Specifies timeout for the initial 64-byte
5927 USB_REQ_GET_DESCRIPTOR request in milliseconds
5928 (default 5000 = 5.0 seconds).
5930 usbcore.nousb [USB] Disable the USB subsystem
5933 [USB] A list of quirk entries to augment the built-in
5934 usb core quirk list. List entries are separated by
5935 commas. Each entry has the form
5936 VendorID:ProductID:Flags. The IDs are 4-digit hex
5937 numbers and Flags is a set of letters. Each letter
5938 will change the built-in quirk; setting it if it is
5939 clear and clearing it if it is set. The letters have
5940 the following meanings:
5941 a = USB_QUIRK_STRING_FETCH_255 (string
5942 descriptors must not be fetched using
5944 b = USB_QUIRK_RESET_RESUME (device can't resume
5945 correctly so reset it instead);
5946 c = USB_QUIRK_NO_SET_INTF (device can't handle
5947 Set-Interface requests);
5948 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
5949 handle its Configuration or Interface
5951 e = USB_QUIRK_RESET (device can't be reset
5952 (e.g morph devices), don't use reset);
5953 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
5954 more interface descriptions than the
5955 bNumInterfaces count, and can't handle
5956 talking to these interfaces);
5957 g = USB_QUIRK_DELAY_INIT (device needs a pause
5958 during initialization, after we read
5959 the device descriptor);
5960 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
5961 high speed and super speed interrupt
5962 endpoints, the USB 2.0 and USB 3.0 spec
5963 require the interval in microframes (1
5964 microframe = 125 microseconds) to be
5965 calculated as interval = 2 ^
5967 Devices with this quirk report their
5968 bInterval as the result of this
5969 calculation instead of the exponent
5970 variable used in the calculation);
5971 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
5972 handle device_qualifier descriptor
5974 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
5975 generates spurious wakeup, ignore
5976 remote wakeup capability);
5977 k = USB_QUIRK_NO_LPM (device can't handle Link
5979 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
5980 (Device reports its bInterval as linear
5981 frames instead of the USB 2.0
5983 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
5984 to be disconnected before suspend to
5985 prevent spurious wakeup);
5986 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
5987 pause after every control message);
5988 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
5989 delay after resetting its port);
5990 Example: quirks=0781:5580:bk,0a5c:5834:gij
5993 [USBHID] The interval which mice are to be polled at.
5996 [USBHID] The interval which joysticks are to be polled at.
5999 [USBHID] The interval which keyboards are to be polled at.
6001 usb-storage.delay_use=
6002 [UMS] The delay in seconds before a new device is
6003 scanned for Logical Units (default 1).
6006 [UMS] A list of quirks entries to supplement or
6007 override the built-in unusual_devs list. List
6008 entries are separated by commas. Each entry has
6009 the form VID:PID:Flags where VID and PID are Vendor
6010 and Product ID values (4-digit hex numbers) and
6011 Flags is a set of characters, each corresponding
6012 to a common usb-storage quirk flag as follows:
6013 a = SANE_SENSE (collect more than 18 bytes
6014 of sense data, not on uas);
6015 b = BAD_SENSE (don't collect more than 18
6016 bytes of sense data, not on uas);
6017 c = FIX_CAPACITY (decrease the reported
6018 device capacity by one sector);
6019 d = NO_READ_DISC_INFO (don't use
6020 READ_DISC_INFO command, not on uas);
6021 e = NO_READ_CAPACITY_16 (don't use
6022 READ_CAPACITY_16 command);
6023 f = NO_REPORT_OPCODES (don't use report opcodes
6025 g = MAX_SECTORS_240 (don't transfer more than
6026 240 sectors at a time, uas only);
6027 h = CAPACITY_HEURISTICS (decrease the
6028 reported device capacity by one
6029 sector if the number is odd);
6030 i = IGNORE_DEVICE (don't bind to this
6032 j = NO_REPORT_LUNS (don't use report luns
6034 k = NO_SAME (do not use WRITE_SAME, uas only)
6035 l = NOT_LOCKABLE (don't try to lock and
6036 unlock ejectable media, not on uas);
6037 m = MAX_SECTORS_64 (don't transfer more
6038 than 64 sectors = 32 KB at a time,
6040 n = INITIAL_READ10 (force a retry of the
6041 initial READ(10) command, not on uas);
6042 o = CAPACITY_OK (accept the capacity
6043 reported by the device, not on uas);
6044 p = WRITE_CACHE (the device cache is ON
6045 by default, not on uas);
6046 r = IGNORE_RESIDUE (the device reports
6047 bogus residue values, not on uas);
6048 s = SINGLE_LUN (the device has only one
6050 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
6051 commands, uas only);
6052 u = IGNORE_UAS (don't bind to the uas driver);
6053 w = NO_WP_DETECT (don't test whether the
6054 medium is write-protected).
6055 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
6056 even if the device claims no cache,
6058 Example: quirks=0419:aaf5:rl,0421:0433:rc
6060 user_debug= [KNL,ARM]
6062 See arch/arm/Kconfig.debug help text.
6063 1 - undefined instruction events
6065 4 - invalid data aborts
6068 Example: user_debug=31
6071 [X86] Flags controlling user PTE allocations.
6073 nohigh = do not allocate PTE pages in
6074 HIGHMEM regardless of setting
6078 On X86_32, this is an alias for vdso32=. Otherwise:
6080 vdso=1: enable VDSO (the default)
6081 vdso=0: disable VDSO mapping
6083 vdso32= [X86] Control the 32-bit vDSO
6084 vdso32=1: enable 32-bit VDSO
6085 vdso32=0 or vdso32=2: disable 32-bit VDSO
6087 See the help text for CONFIG_COMPAT_VDSO for more
6088 details. If CONFIG_COMPAT_VDSO is set, the default is
6089 vdso32=0; otherwise, the default is vdso32=1.
6091 For compatibility with older kernels, vdso32=2 is an
6094 Try vdso32=0 if you encounter an error that says:
6095 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
6098 vector=percpu: enable percpu vector domain
6100 video= [FB] Frame buffer configuration
6101 See Documentation/fb/modedb.rst.
6103 video.brightness_switch_enabled= [0,1]
6104 If set to 1, on receiving an ACPI notify event
6105 generated by hotkey, video driver will adjust brightness
6106 level and then send out the event to user space through
6107 the allocated input device; If set to 0, video driver
6108 will only send out the event without touching backlight
6113 [VMMIO] Memory mapped virtio (platform) device.
6115 <size>@<baseaddr>:<irq>[:<id>]
6117 <size> := size (can use standard suffixes
6119 <baseaddr> := physical base address
6120 <irq> := interrupt number (as passed to
6122 <id> := (optional) platform device id
6124 virtio_mmio.device=1K@0x100b0000:48:7
6126 Can be used multiple times for multiple devices.
6128 vga= [BOOT,X86-32] Select a particular video mode
6129 See Documentation/x86/boot.rst and
6130 Documentation/admin-guide/svga.rst.
6131 Use vga=ask for menu.
6132 This is actually a boot loader parameter; the value is
6133 passed to the kernel using a special protocol.
6135 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
6136 May slow down system boot speed, especially when
6137 enabled on systems with a large amount of memory.
6138 All options are enabled by default, and this
6139 interface is meant to allow for selectively
6140 enabling or disabling specific virtual memory
6143 Available options are:
6144 P Enable page structure init time poisoning
6145 - Disable all of the above options
6147 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
6148 size of <nn>. This can be used to increase the
6149 minimum size (128MB on x86). It can also be used to
6150 decrease the size and leave more room for directly
6153 vmcp_cma=nn[MG] [KNL,S390]
6154 Sets the memory size reserved for contiguous memory
6155 allocations for the vmcp device driver.
6157 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
6160 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
6163 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
6167 Controls the behavior of vsyscalls (i.e. calls to
6168 fixed addresses of 0xffffffffff600x00 from legacy
6169 code). Most statically-linked binaries and older
6170 versions of glibc use these calls. Because these
6171 functions are at fixed addresses, they make nice
6172 targets for exploits that can control RIP.
6174 emulate [default] Vsyscalls turn into traps and are
6175 emulated reasonably safely. The vsyscall
6178 xonly Vsyscalls turn into traps and are
6179 emulated reasonably safely. The vsyscall
6180 page is not readable.
6182 none Vsyscalls don't work at all. This makes
6183 them quite hard to use for exploits but
6184 might break your system.
6186 vt.color= [VT] Default text color.
6187 Format: 0xYX, X = foreground, Y = background.
6188 Default: 0x07 = light gray on black.
6190 vt.cur_default= [VT] Default cursor shape.
6191 Format: 0xCCBBAA, where AA, BB, and CC are the same as
6192 the parameters of the <Esc>[?A;B;Cc escape sequence;
6193 see VGA-softcursor.txt. Default: 2 = underline.
6195 vt.default_blu= [VT]
6196 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
6197 Change the default blue palette of the console.
6198 This is a 16-member array composed of values
6201 vt.default_grn= [VT]
6202 Format: <green0>,<green1>,<green2>,...,<green15>
6203 Change the default green palette of the console.
6204 This is a 16-member array composed of values
6207 vt.default_red= [VT]
6208 Format: <red0>,<red1>,<red2>,...,<red15>
6209 Change the default red palette of the console.
6210 This is a 16-member array composed of values
6216 Set system-wide default UTF-8 mode for all tty's.
6217 Default is 1, i.e. UTF-8 mode is enabled for all
6218 newly opened terminals.
6220 vt.global_cursor_default=
6223 Set system-wide default for whether a cursor
6224 is shown on new VTs. Default is -1,
6225 i.e. cursors will be created by default unless
6226 overridden by individual drivers. 0 will hide
6227 cursors, 1 will display them.
6229 vt.italic= [VT] Default color for italic text; 0-15.
6232 vt.underline= [VT] Default color for underlined text; 0-15.
6235 watchdog timers [HW,WDT] For information on watchdog timers,
6236 see Documentation/watchdog/watchdog-parameters.rst
6237 or other driver-specific files in the
6238 Documentation/watchdog/ directory.
6242 Set the hard lockup detector stall duration
6243 threshold in seconds. The soft lockup detector
6244 threshold is set to twice the value. A value of 0
6245 disables both lockup detectors. Default is 10
6248 workqueue.watchdog_thresh=
6249 If CONFIG_WQ_WATCHDOG is configured, workqueue can
6250 warn stall conditions and dump internal state to
6251 help debugging. 0 disables workqueue stall
6252 detection; otherwise, it's the stall threshold
6253 duration in seconds. The default value is 30 and
6254 it can be updated at runtime by writing to the
6255 corresponding sysfs file.
6257 workqueue.disable_numa
6258 By default, all work items queued to unbound
6259 workqueues are affine to the NUMA nodes they're
6260 issued on, which results in better behavior in
6261 general. If NUMA affinity needs to be disabled for
6262 whatever reason, this option can be used. Note
6263 that this also can be controlled per-workqueue for
6264 workqueues visible under /sys/bus/workqueue/.
6266 workqueue.power_efficient
6267 Per-cpu workqueues are generally preferred because
6268 they show better performance thanks to cache
6269 locality; unfortunately, per-cpu workqueues tend to
6270 be more power hungry than unbound workqueues.
6272 Enabling this makes the per-cpu workqueues which
6273 were observed to contribute significantly to power
6274 consumption unbound, leading to measurably lower
6275 power usage at the cost of small performance
6278 The default value of this parameter is determined by
6279 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
6281 workqueue.debug_force_rr_cpu
6282 Workqueue used to implicitly guarantee that work
6283 items queued without explicit CPU specified are put
6284 on the local CPU. This guarantee is no longer true
6285 and while local CPU is still preferred work items
6286 may be put on foreign CPUs. This debug option
6287 forces round-robin CPU selection to flush out
6288 usages which depend on the now broken guarantee.
6289 When enabled, memory and cache locality will be
6292 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
6293 default x2apic cluster mode on platforms
6296 xen_512gb_limit [KNL,X86-64,XEN]
6297 Restricts the kernel running paravirtualized under Xen
6298 to use only up to 512 GB of RAM. The reason to do so is
6299 crash analysis tools and Xen tools for doing domain
6300 save/restore/migration must be enabled to handle larger
6303 xen_emul_unplug= [HW,X86,XEN]
6304 Unplug Xen emulated devices
6305 Format: [unplug0,][unplug1]
6306 ide-disks -- unplug primary master IDE devices
6307 aux-ide-disks -- unplug non-primary-master IDE devices
6308 nics -- unplug network devices
6309 all -- unplug all emulated devices (NICs and IDE disks)
6310 unnecessary -- unplugging emulated devices is
6311 unnecessary even if the host did not respond to
6313 never -- do not unplug even if version check succeeds
6315 xen_legacy_crash [X86,XEN]
6316 Crash from Xen panic notifier, without executing late
6317 panic() code such as dumping handler.
6319 xen_nopvspin [X86,XEN]
6320 Disables the qspinlock slowpath using Xen PV optimizations.
6321 This parameter is obsoleted by "nopvspin" parameter, which
6322 has equivalent effect for XEN platform.
6325 Disables the PV optimizations forcing the HVM guest to
6326 run as generic HVM guest with no PV drivers.
6327 This option is obsoleted by the "nopv" option, which
6328 has equivalent effect for XEN platform.
6330 xen_no_vector_callback
6331 [KNL,X86,XEN] Disable the vector callback for Xen
6332 event channel interrupts.
6334 xen_scrub_pages= [XEN]
6335 Boolean option to control scrubbing pages before giving them back
6336 to Xen, for use by other domains. Can be also changed at runtime
6337 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
6338 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
6340 xen_timer_slop= [X86-64,XEN]
6341 Set the timer slop (in nanoseconds) for the virtual Xen
6342 timers (default is 100000). This adjusts the minimum
6343 delta of virtualized Xen timers, where lower values
6344 improve timer resolution at the expense of processing
6345 more timer interrupts.
6347 xen.event_eoi_delay= [XEN]
6348 How long to delay EOI handling in case of event
6349 storms (jiffies). Default is 10.
6351 xen.event_loop_timeout= [XEN]
6352 After which time (jiffies) the event handling loop
6353 should start to delay EOI handling. Default is 2.
6355 xen.fifo_events= [XEN]
6356 Boolean parameter to disable using fifo event handling
6357 even if available. Normally fifo event handling is
6358 preferred over the 2-level event handling, as it is
6359 fairer and the number of possible event channels is
6360 much higher. Default is on (use fifo events).
6362 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
6363 Disables the PV optimizations forcing the guest to run
6364 as generic guest with no PV drivers. Currently support
6365 XEN HVM, KVM, HYPER_V and VMWARE guest.
6367 nopvspin [X86,XEN,KVM]
6368 Disables the qspinlock slow path using PV optimizations
6369 which allow the hypervisor to 'idle' the guest on lock
6372 xirc2ps_cs= [NET,PCMCIA]
6374 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
6377 By default on POWER9 and above, the kernel will
6378 natively use the XIVE interrupt controller. This option
6379 allows the fallback firmware mode to be used:
6381 off Fallback to firmware control of XIVE interrupt
6382 controller on both pseries and powernv
6383 platforms. Only useful on POWER9 and above.
6385 xhci-hcd.quirks [USB,KNL]
6386 A hex value specifying bitmask with supplemental xhci
6387 host controller quirks. Meaning of each bit can be
6388 consulted in header drivers/usb/host/xhci.h.
6391 Format: { early | on | rw | ro | off }
6392 Controls if xmon debugger is enabled. Default is off.
6393 Passing only "xmon" is equivalent to "xmon=early".
6394 early Call xmon as early as possible on boot; xmon
6395 debugger is called from setup_arch().
6396 on xmon debugger hooks will be installed so xmon
6397 is only called on a kernel crash. Default mode,
6398 i.e. either "ro" or "rw" mode, is controlled
6399 with CONFIG_XMON_DEFAULT_RO_MODE.
6400 rw xmon debugger hooks will be installed so xmon
6401 is called only on a kernel crash, mode is write,
6402 meaning SPR registers, memory and, other data
6403 can be written using xmon commands.
6404 ro same as "rw" option above but SPR registers,
6405 memory, and other data can't be written using
6407 off xmon is disabled.