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]
27 If set to vendor, prefer vendor specific driver
28 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
29 of the ACPI video.ko driver.
31 acpi_force_32bit_fadt_addr
32 force FADT to use 32 bit addresses rather than the
33 64 bit X_* addresses. Some firmware have broken 64
34 bit addresses for force ACPI ignore these and use
35 the older legacy 32 bit addresses.
37 acpica_no_return_repair [HW, ACPI]
38 Disable AML predefined validation mechanism
39 This mechanism can repair the evaluation result to make
40 the return objects more ACPI specification compliant.
41 This option is useful for developers to identify the
42 root cause of an AML interpreter issue when the issue
43 has something to do with the repair mechanism.
45 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
46 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
48 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
49 debug output. Bits in debug_layer correspond to a
50 _COMPONENT in an ACPI source file, e.g.,
51 #define _COMPONENT ACPI_PCI_COMPONENT
52 Bits in debug_level correspond to a level in
53 ACPI_DEBUG_PRINT statements, e.g.,
54 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
55 The debug_level mask defaults to "info". See
56 Documentation/firmware-guide/acpi/debug.rst for more information about
57 debug layers and levels.
59 Enable processor driver info messages:
60 acpi.debug_layer=0x20000000
61 Enable PCI/PCI interrupt routing info messages:
62 acpi.debug_layer=0x400000
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
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_rsdp= [ACPI,EFI,KEXEC]
140 Pass the RSDP address to the kernel, mostly used
141 on machines running EFI runtime service to boot the
142 second kernel for kdump.
144 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
145 Format: To spoof as Windows 98: ="Microsoft Windows"
147 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
148 of 2 which is mandated by ACPI 6) as the supported ACPI
149 specification revision (when using this switch, it may
150 be necessary to carry out a cold reboot _twice_ in a
151 row to make it take effect on the platform firmware).
153 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
154 acpi_osi="string1" # add string1
155 acpi_osi="!string2" # remove string2
156 acpi_osi=!* # remove all strings
157 acpi_osi=! # disable all built-in OS vendor
159 acpi_osi=!! # enable all built-in OS vendor
161 acpi_osi= # disable all strings
163 'acpi_osi=!' can be used in combination with single or
164 multiple 'acpi_osi="string1"' to support specific OS
165 vendor string(s). Note that such command can only
166 affect the default state of the OS vendor strings, thus
167 it cannot affect the default state of the feature group
168 strings and the current state of the OS vendor strings,
169 specifying it multiple times through kernel command line
170 is meaningless. This command is useful when one do not
171 care about the state of the feature group strings which
172 should be controlled by the OSPM.
174 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
175 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
176 can make '_OSI("Windows 2000")' TRUE.
178 'acpi_osi=' cannot be used in combination with other
179 'acpi_osi=' command lines, the _OSI method will not
180 exist in the ACPI namespace. NOTE that such command can
181 only affect the _OSI support state, thus specifying it
182 multiple times through kernel command line is also
185 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
188 'acpi_osi=!*' can be used in combination with single or
189 multiple 'acpi_osi="string1"' to support specific
190 string(s). Note that such command can affect the
191 current state of both the OS vendor strings and the
192 feature group strings, thus specifying it multiple times
193 through kernel command line is meaningful. But it may
194 still not able to affect the final state of a string if
195 there are quirks related to this string. This command
196 is useful when one want to control the state of the
197 feature group strings to debug BIOS issues related to
200 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
201 '_OSI("Module Device")' FALSE.
202 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
203 '_OSI("Module Device")' TRUE.
204 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
206 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
208 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
209 they all will make '_OSI("Windows 2000")' TRUE.
212 Override the pmtimer bug detection: force the kernel
213 to assume that this machine's pmtimer latches its value
214 and always returns good values.
216 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
217 Format: { level | edge | high | low }
219 acpi_skip_timer_override [HW,ACPI]
220 Recognize and ignore IRQ0/pin2 Interrupt Override.
221 For broken nForce2 BIOS resulting in XT-PIC timer.
223 acpi_sleep= [HW,ACPI] Sleep options
224 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
225 old_ordering, nonvs, sci_force_enable, nobl }
226 See Documentation/power/video.rst for information on
228 s3_beep is for debugging; it makes the PC's speaker beep
229 as soon as the kernel's real-mode entry point is called.
230 s4_nohwsig prevents ACPI hardware signature from being
231 used during resume from hibernation.
232 old_ordering causes the ACPI 1.0 ordering of the _PTS
233 control method, with respect to putting devices into
234 low power states, to be enforced (the ACPI 2.0 ordering
235 of _PTS is used by default).
236 nonvs prevents the kernel from saving/restoring the
237 ACPI NVS memory during suspend/hibernation and resume.
238 sci_force_enable causes the kernel to set SCI_EN directly
239 on resume from S1/S3 (which is against the ACPI spec,
240 but some broken systems don't work without it).
241 nobl causes the internal blacklist of systems known to
242 behave incorrectly in some ways with respect to system
243 suspend and resume to be ignored (use wisely).
245 acpi_use_timer_override [HW,ACPI]
246 Use timer override. For some broken Nvidia NF5 boards
247 that require a timer override, but don't have HPET
249 add_efi_memmap [EFI; X86] Include EFI memory map in
250 kernel's map of available physical RAM.
253 { off | try_unsupported }
254 off: disable AGP support
255 try_unsupported: try to drive unsupported chipsets
256 (may crash computer or cause data corruption)
259 See Documentation/sound/alsa-configuration.rst
262 Allow the default userspace alignment fault handler
263 behaviour to be specified. Bit 0 enables warnings,
264 bit 1 enables fixups, and bit 2 sends a segfault.
266 align_va_addr= [X86-64]
267 Align virtual addresses by clearing slice [14:12] when
268 allocating a VMA at process creation time. This option
269 gives you up to 3% performance improvement on AMD F15h
270 machines (where it is enabled by default) for a
271 CPU-intensive style benchmark, and it can vary highly in
272 a microbenchmark depending on workload and compiler.
274 32: only for 32-bit processes
275 64: only for 64-bit processes
276 on: enable for both 32- and 64-bit processes
277 off: disable for both 32- and 64-bit processes
279 alloc_snapshot [FTRACE]
280 Allocate the ftrace snapshot buffer on boot up when the
281 main buffer is allocated. This is handy if debugging
282 and you need to use tracing_snapshot() on boot up, and
283 do not want to use tracing_snapshot_alloc() as it needs
284 to be done where GFP_KERNEL allocations are allowed.
286 amd_iommu= [HW,X86-64]
287 Pass parameters to the AMD IOMMU driver in the system.
289 fullflush - enable flushing of IO/TLB entries when
290 they are unmapped. Otherwise they are
291 flushed before they will be reused, which
293 off - do not initialize any AMD IOMMU found in
295 force_isolation - Force device isolation for all
296 devices. The IOMMU driver is not
297 allowed anymore to lift isolation
298 requirements as needed. This option
299 does not override iommu=pt
301 amd_iommu_dump= [HW,X86-64]
302 Enable AMD IOMMU driver option to dump the ACPI table
303 for AMD IOMMU. With this option enabled, AMD IOMMU
304 driver will print ACPI tables for AMD IOMMU during
305 IOMMU initialization.
307 amd_iommu_intr= [HW,X86-64]
308 Specifies one of the following AMD IOMMU interrupt
310 legacy - Use legacy interrupt remapping mode.
311 vapic - Use virtual APIC mode, which allows IOMMU
312 to inject interrupts directly into guest.
313 This mode requires kvm-amd.avic=1.
314 (Default when IOMMU HW support is present.)
316 amijoy.map= [HW,JOY] Amiga joystick support
317 Map of devices attached to JOY0DAT and JOY1DAT
319 See also Documentation/input/joydev/joystick.rst
321 analog.map= [HW,JOY] Analog joystick and gamepad support
322 Specifies type or capabilities of an analog joystick
323 connected to one of 16 gameports
324 Format: <type1>,<type2>,..<type16>
327 Power management functions (SPARCstation-4/5 + deriv.)
329 Disable APC CPU standby support. SPARCstation-Fox does
330 not play well with APC CPU idle - disable it if you have
331 APC and your system crashes randomly.
333 apic= [APIC,X86] Advanced Programmable Interrupt Controller
334 Change the output verbosity while booting
335 Format: { quiet (default) | verbose | debug }
336 Change the amount of debugging information output
337 when initialising the APIC and IO-APIC components.
338 For X86-32, this can also be used to specify an APIC
340 Format: apic=driver_name
341 Examples: apic=bigsmp
343 apic_extnmi= [APIC,X86] External NMI delivery setting
344 Format: { bsp (default) | all | none }
345 bsp: External NMI is delivered only to CPU 0
346 all: External NMIs are broadcast to all CPUs as a
348 none: External NMI is masked for all CPUs. This is
349 useful so that a dump capture kernel won't be
353 See Documentation/networking/ipv6.txt.
355 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
356 Limit apic dumping. The parameter defines the maximal
357 number of local apics being dumped. Also it is possible
358 to set it to "all" by meaning -- no limit here.
359 Format: { 1 (default) | 2 | ... | all }.
360 The parameter valid if only apic=debug or
361 apic=verbose is specified.
362 Example: apic=debug show_lapic=all
364 apm= [APM] Advanced Power Management
365 See header of arch/x86/kernel/apm_32.c.
367 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
368 Format: <io>,<irq>,<nodeID>
372 atarimouse= [HW,MOUSE] Atari Mouse
374 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
375 EzKey and similar keyboards
377 atkbd.reset= [HW] Reset keyboard during initialization
379 atkbd.set= [HW] Select keyboard code set
380 Format: <int> (2 = AT (default), 3 = PS/2)
382 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
385 atkbd.softraw= [HW] Choose between synthetic and real raw mode
386 Format: <bool> (0 = real, 1 = synthetic (default))
388 atkbd.softrepeat= [HW]
389 Use software keyboard repeat
391 audit= [KNL] Enable the audit sub-system
392 Format: { "0" | "1" | "off" | "on" }
393 0 | off - kernel audit is disabled and can not be
394 enabled until the next reboot
395 unset - kernel audit is initialized but disabled and
396 will be fully enabled by the userspace auditd.
397 1 | on - kernel audit is initialized and partially
398 enabled, storing at most audit_backlog_limit
399 messages in RAM until it is fully enabled by the
403 audit_backlog_limit= [KNL] Set the audit queue size limit.
404 Format: <int> (must be >=0)
407 bau= [X86_UV] Enable the BAU on SGI UV. The default
408 behavior is to disable the BAU (i.e. bau=0).
409 Format: { "0" | "1" }
412 unset - Disable the BAU.
414 baycom_epp= [HW,AX25]
417 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
419 See header of drivers/net/hamradio/baycom_par.c.
421 baycom_ser_fdx= [HW,AX25]
422 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
423 Format: <io>,<irq>,<mode>[,<baud>]
424 See header of drivers/net/hamradio/baycom_ser_fdx.c.
426 baycom_ser_hdx= [HW,AX25]
427 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
428 Format: <io>,<irq>,<mode>
429 See header of drivers/net/hamradio/baycom_ser_hdx.c.
431 blkdevparts= Manual partition parsing of block device(s) for
432 embedded devices based on command line input.
433 See Documentation/block/cmdline-partition.rst
435 boot_delay= Milliseconds to delay each printk during boot.
436 Values larger than 10 seconds (10000) are changed to
441 Disable BERT OS support on buggy BIOSes.
443 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
444 bttv.radio= Most important insmod options are available as
446 bttv.pll= See Documentation/media/v4l-drivers/bttv.rst
449 bulk_remove=off [PPC] This parameter disables the use of the pSeries
450 firmware feature for flushing multiple hpte entries
453 c101= [NET] Moxa C101 synchronous serial card
455 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
456 Sometimes CPU hardware bugs make them report the cache
457 size incorrectly. The kernel will attempt work arounds
458 to fix known problems, but for some CPUs it is not
459 possible to determine what the correct size should be.
460 This option provides an override for these situations.
463 [NET] Specifies amount of time (in seconds) that
464 the kernel should wait for a network carrier. By default
465 it waits 120 seconds.
467 ca_keys= [KEYS] This parameter identifies a specific key(s) on
468 the system trusted keyring to be used for certificate
470 format: { id:<keyid> | builtin }
472 cca= [MIPS] Override the kernel pages' cache coherency
473 algorithm. Accepted values range from 0 to 7
474 inclusive. See arch/mips/include/asm/pgtable-bits.h
475 for platform specific values (SB1, Loongson3 and
478 ccw_timeout_log [S390]
479 See Documentation/s390/common_io.rst for details.
481 cgroup_disable= [KNL] Disable a particular controller
482 Format: {name of the controller(s) to disable}
483 The effects of cgroup_disable=foo are:
484 - foo isn't auto-mounted if you mount all cgroups in
486 - foo isn't visible as an individually mountable
488 {Currently only "memory" controller deal with this and
489 cut the overhead, others just disable the usage. So
490 only cgroup_disable=memory is actually worthy}
492 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
493 Format: { { controller | "all" | "named" }
494 [,{ controller | "all" | "named" }...] }
495 Like cgroup_disable, but only applies to cgroup v1;
496 the blacklisted controllers remain available in cgroup2.
497 "all" blacklists all controllers and "named" disables
498 named mounts. Specifying both "all" and "named" disables
501 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
503 nosocket -- Disable socket memory accounting.
504 nokmem -- Disable kernel memory accounting.
506 checkreqprot [SELINUX] Set initial checkreqprot flag value.
507 Format: { "0" | "1" }
508 See security/selinux/Kconfig help text.
509 0 -- check protection applied by kernel (includes
510 any implied execute protection).
511 1 -- check protection requested by application.
512 Default value is set via a kernel config option.
513 Value can be changed at runtime via
514 /selinux/checkreqprot.
517 See Documentation/s390/common_io.rst for details.
520 Prevents the clock framework from automatically gating
521 clocks that have not been explicitly enabled by a Linux
522 device driver but are enabled in hardware at reset or
523 by the bootloader/firmware. Note that this does not
524 force such clocks to be always-on nor does it reserve
525 those clocks in any way. This parameter is useful for
526 debug and development, but should not be needed on a
527 platform with proper driver support. For more
528 information, see Documentation/driver-api/clk.rst.
530 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
532 Forces specified clocksource (if available) to be used
533 when calculating gettimeofday(). If specified
534 clocksource is not available, it defaults to PIT.
535 Format: { pit | tsc | cyclone | pmtmr }
537 clocksource= Override the default clocksource
539 Override the default clocksource and use the clocksource
540 with the name specified.
541 Some clocksource names to choose from, depending on
543 [all] jiffies (this is the base, fallback clocksource)
545 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
546 pxa_timer,timer3,32k_counter,timer0_1
547 [X86-32] pit,hpet,tsc;
548 scx200_hrt on Geode; cyclone on IBM x440
556 clocksource.arm_arch_timer.evtstrm=
559 Enable/disable the eventstream feature of the ARM
560 architected timer so that code using WFE-based polling
561 loops can be debugged more effectively on production
564 clearcpuid=BITNUM [X86]
565 Disable CPUID feature X for the kernel. See
566 arch/x86/include/asm/cpufeatures.h for the valid bit
567 numbers. Note the Linux specific bits are not necessarily
568 stable over kernel options, but the vendor specific
570 Also note that user programs calling CPUID directly
571 or using the feature without checking anything
572 will still see it. This just prevents it from
573 being used by the kernel or shown in /proc/cpuinfo.
574 Also note the kernel might malfunction if you disable
577 cma=nn[MG]@[start[MG][-end[MG]]]
579 Sets the size of kernel global memory area for
580 contiguous memory allocations and optionally the
581 placement constraint by the physical address range of
582 memory allocations. A value of 0 disables CMA
583 altogether. For more information, see
584 include/linux/dma-contiguous.h
586 cmo_free_hint= [PPC] Format: { yes | no }
587 Specify whether pages are marked as being inactive
588 when they are freed. This is used in CMO environments
589 to determine OS memory pressure for page stealing by
593 coherent_pool=nn[KMG] [ARM,KNL]
594 Sets the size of memory pool for coherent, atomic dma
595 allocations, by default set to 256K.
597 com20020= [HW,NET] ARCnet - COM20020 chipset
599 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
601 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
605 ARCnet - COM90xx chipset (memory-mapped buffers)
606 Format: <io>[,<irq>[,<memstart>]]
608 condev= [HW,S390] console device
611 console= [KNL] Output console device and options.
613 tty<n> Use the virtual console device <n>.
617 Use the specified serial port. The options are of
618 the form "bbbbpnf", where "bbbb" is the baud rate,
619 "p" is parity ("n", "o", or "e"), "n" is number of
620 bits, and "f" is flow control ("r" for RTS or
621 omit it). Default is "9600n8".
623 See Documentation/admin-guide/serial-console.rst for more
625 Documentation/networking/netconsole.txt for an
628 uart[8250],io,<addr>[,options]
629 uart[8250],mmio,<addr>[,options]
630 uart[8250],mmio16,<addr>[,options]
631 uart[8250],mmio32,<addr>[,options]
632 uart[8250],0x<addr>[,options]
633 Start an early, polled-mode console on the 8250/16550
634 UART at the specified I/O port or MMIO address,
635 switching to the matching ttyS device later.
636 MMIO inter-register address stride is either 8-bit
637 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
638 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
639 to be equivalent to 'mmio'. 'options' are specified in
640 the same format described for ttyS above; if unspecified,
641 the h/w is not re-initialized.
643 hvc<n> Use the hypervisor console device <n>. This is for
644 both Xen and PowerPC hypervisors.
646 If the device connected to the port is not a TTY but a braille
647 device, prepend "brl," before the device type, for instance
649 For now, only VisioBraille is supported.
652 [KNL] Change console messages format
654 By default we print messages on consoles in
655 "[time stamp] text\n" format (time stamp may not be
656 printed, depending on CONFIG_PRINTK_TIME or
657 `printk_time' param).
659 Switch to syslog format: "<%u>[time stamp] text\n"
660 IOW, each message will have a facility and loglevel
661 prefix. The format is similar to one used by syslog()
662 syscall, or to executing "dmesg -S --raw" or to reading
665 consoleblank= [KNL] The console blank (screen saver) timeout in
666 seconds. A value of 0 disables the blank timer.
670 [KNL] Change the default value for
671 /proc/<pid>/coredump_filter.
672 See also Documentation/filesystems/proc.txt.
674 coresight_cpu_debug.enable
677 Enable/disable the CPU sampling based debugging.
678 0: default value, disable debugging
679 1: enable debugging at boot time
681 cpuidle.off=1 [CPU_IDLE]
682 disable the cpuidle sub-system
685 [CPU_IDLE] Name of the cpuidle governor to use.
687 cpufreq.off=1 [CPU_FREQ]
688 disable the cpufreq sub-system
691 [X86] Delay for N microsec between assert and de-assert
692 of APIC INIT to start processors. This delay occurs
693 on every CPU online, such as boot, and resume from suspend.
696 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
698 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
700 crashkernel=size[KMG][@offset[KMG]]
701 [KNL] Using kexec, Linux can switch to a 'crash kernel'
702 upon panic. This parameter reserves the physical
703 memory region [offset, offset + size] for that kernel
704 image. If '@offset' is omitted, then a suitable offset
705 is selected automatically.
706 [KNL, x86_64] select a region under 4G first, and
707 fall back to reserve region above 4G when '@offset'
708 hasn't been specified.
709 See Documentation/admin-guide/kdump/kdump.rst for further details.
711 crashkernel=range1:size1[,range2:size2,...][@offset]
712 [KNL] Same as above, but depends on the memory
713 in the running system. The syntax of range is
714 start-[end] where start and end are both
715 a memory unit (amount[KMG]). See also
716 Documentation/admin-guide/kdump/kdump.rst for an example.
718 crashkernel=size[KMG],high
719 [KNL, x86_64] range could be above 4G. Allow kernel
720 to allocate physical memory region from top, so could
721 be above 4G if system have more than 4G ram installed.
722 Otherwise memory region will be allocated below 4G, if
724 It will be ignored if crashkernel=X is specified.
725 crashkernel=size[KMG],low
726 [KNL, x86_64] range under 4G. When crashkernel=X,high
727 is passed, kernel could allocate physical memory region
728 above 4G, that cause second kernel crash on system
729 that require some amount of low memory, e.g. swiotlb
730 requires at least 64M+32K low memory, also enough extra
731 low memory is needed to make sure DMA buffers for 32-bit
732 devices won't run out. Kernel would try to allocate at
733 at least 256M below 4G automatically.
734 This one let user to specify own low range under 4G
735 for second kernel instead.
736 0: to disable low allocation.
737 It will be ignored when crashkernel=X,high is not used
738 or memory reserved is below 4G.
741 [KNL] Disable crypto self-tests
746 cs89x0_media= [HW,NET]
747 Format: { rj45 | aui | bnc }
750 See header of drivers/s390/block/dasd_devmap.c.
752 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
753 (one device per port)
754 Format: <port#>,<type>
755 See also Documentation/input/devices/joystick-parport.rst
757 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
759 Documentation/admin-guide/dynamic-debug-howto.rst for
760 details. Deprecated, see dyndbg.
762 debug [KNL] Enable kernel debugging (events log level).
765 [KNL] Enable printing [hashed] pointers early in the
766 boot sequence. If enabled, we use a weak hash instead
767 of siphash to hash pointers. Use this option if you are
768 seeing instances of '(___ptrval___)') and need to see a
769 value (hashed pointer) instead. Cryptographically
770 insecure, please do not use on production kernels.
773 [KNL] verbose self-tests
775 Print debugging info while doing the locking API
777 We default to 0 (no extra messages), setting it to
778 1 will print _a lot_ more information - normally
779 only useful to kernel developers.
781 debug_objects [KNL] Enable object debugging
784 [KNL] Disable object debugging
786 debug_guardpage_minorder=
787 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
788 parameter allows control of the order of pages that will
789 be intentionally kept free (and hence protected) by the
790 buddy allocator. Bigger value increase the probability
791 of catching random memory corruption, but reduce the
792 amount of memory for normal system use. The maximum
793 possible value is MAX_ORDER/2. Setting this parameter
794 to 1 or 2 should be enough to identify most random
795 memory corruption problems caused by bugs in kernel or
796 driver code when a CPU writes to (or reads from) a
797 random memory location. Note that there exists a class
798 of memory corruptions problems caused by buggy H/W or
799 F/W or by drivers badly programing DMA (basically when
800 memory is written at bus level and the CPU MMU is
801 bypassed) which are not detectable by
802 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
803 tracking down these problems.
806 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
807 enables the feature at boot time. By default, it is
808 disabled and the system will work mostly the same as a
809 kernel built without CONFIG_DEBUG_PAGEALLOC.
810 Note: to get most of debug_pagealloc error reports, it's
811 useful to also enable the page_owner functionality.
812 on: enable the feature
814 debugpat [X86] Enable PAT debugging
816 decnet.addr= [HW,NET]
817 Format: <area>[,<node>]
818 See also Documentation/networking/decnet.txt.
821 [same as hugepagesz=] The size of the default
822 HugeTLB page size. This is the size represented by
823 the legacy /proc/ hugepages APIs, used for SHM, and
824 default size when mounting hugetlbfs filesystems.
825 Defaults to the default architecture's huge page size
828 deferred_probe_timeout=
829 [KNL] Debugging option to set a timeout in seconds for
830 deferred probe to give up waiting on dependencies to
831 probe. Only specific dependencies (subsystems or
832 drivers) that have opted in will be ignored. A timeout of 0
833 will timeout at the end of initcalls. This option will also
834 dump out devices still on the deferred probe list after
838 Set number of hash buckets for dentry cache.
840 disable_1tb_segments [PPC]
841 Disables the use of 1TB hash page table segments. This
842 causes the kernel to fall back to 256MB segments which
843 can be useful when debugging issues that require an SLB
847 See Documentation/networking/ipv6.txt.
850 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
851 hardening is enabled for this boot. Hardened
852 usercopy checking is used to protect the kernel
853 from reading or writing beyond known memory
854 allocation boundaries as a proactive defense
855 against bounds-checking flaws in the kernel's
856 copy_to_user()/copy_from_user() interface.
857 on Perform hardened usercopy checks (default).
858 off Disable hardened usercopy checks.
861 Disable RADIX MMU mode on POWER9
864 Disable TLBIE instruction. Currently does not work
865 with KVM, with HASH MMU, or with coherent accelerators.
867 disable_cpu_apicid= [X86,APIC,SMP]
869 The number of initial APIC ID for the
870 corresponding CPU to be disabled at boot,
871 mostly used for the kdump 2nd kernel to
872 disable BSP to wake up multiple CPUs without
873 causing system reset or hang due to sending
876 perf_v4_pmi= [X86,INTEL]
878 Disable Intel PMU counter freezing feature.
879 The feature only exists starting from
880 Arch Perfmon v4 (Skylake and newer).
882 disable_ddw [PPC/PSERIES]
883 Disable Dynamic DMA Window support. Use this if
884 to workaround buggy firmware.
887 See Documentation/networking/ipv6.txt.
889 disable_mtrr_cleanup [X86]
890 The kernel tries to adjust MTRR layout from continuous
891 to discrete, to make X server driver able to add WB
892 entry later. This parameter disables that.
894 disable_mtrr_trim [X86, Intel and AMD only]
895 By default the kernel will trim any uncacheable
896 memory out of your available memory pool based on
897 MTRR settings. This parameter disables that behavior,
898 possibly causing your machine to run very slowly.
900 disable_timer_pin_1 [X86]
901 Disable PIN 1 of APIC timer
902 Can be useful to work around chipset bugs.
904 dis_ucode_ldr [X86] Disable the microcode loader.
906 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
907 this option disables the debugging code at boot.
909 dma_debug_entries=<number>
910 This option allows to tune the number of preallocated
911 entries for DMA-API debugging code. One entry is
912 required per DMA-API allocation. Use this if the
913 DMA-API debugging code disables itself because the
914 architectural default is too low.
916 dma_debug_driver=<driver_name>
917 With this option the DMA-API debugging driver
918 filter feature can be enabled at boot time. Just
919 pass the driver to filter for as the parameter.
920 The filter can be disabled or changed to another
921 driver later using sysfs.
923 driver_async_probe= [KNL]
924 List of driver names to be probed asynchronously.
925 Format: <driver_name1>,<driver_name2>...
927 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
928 Broken monitors, graphic adapters, KVMs and EDIDless
929 panels may send no or incorrect EDID data sets.
930 This parameter allows to specify an EDID data sets
931 in the /lib/firmware directory that are used instead.
932 Generic built-in EDID data sets are used, if one of
933 edid/1024x768.bin, edid/1280x1024.bin,
934 edid/1680x1050.bin, or edid/1920x1080.bin is given
935 and no file with the same name exists. Details and
936 instructions how to build your own EDID data are
937 available in Documentation/driver-api/edid.rst. An EDID
938 data set will only be used for a particular connector,
939 if its name and a colon are prepended to the EDID
940 name. Each connector may use a unique EDID data
941 set by separating the files with a comma. An EDID
942 data set with no connector name will be used for
943 any connectors not explicitly specified.
948 Format: {"off" | "known"}
949 Control how the dt_cpu_ftrs device-tree binding is
950 used for CPU feature discovery and setup (if it
952 off: Do not use it, fall back to legacy cpu table.
953 known: Do not pass through unknown features to guests
954 or userspace, only those that the kernel is aware of.
956 dump_apple_properties [X86]
957 Dump name and content of EFI device properties on
958 x86 Macs. Useful for driver authors to determine
959 what data is available or for reverse-engineering.
961 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
962 module.dyndbg[="val"]
963 Enable debug messages at boot time. See
964 Documentation/admin-guide/dynamic-debug-howto.rst
967 nompx [X86] Disables Intel Memory Protection Extensions.
968 See Documentation/x86/intel_mpx.rst for more
969 information about the feature.
971 nopku [X86] Disable Memory Protection Keys CPU feature found
974 module.async_probe [KNL]
975 Enable asynchronous probe on this module.
977 early_ioremap_debug [KNL]
978 Enable debug messages in early_ioremap support. This
979 is useful for tracking down temporary early mappings
980 which are not unmapped.
982 earlycon= [KNL] Output early console device and options.
984 When used with no options, the early console is
985 determined by stdout-path property in device tree's
986 chosen node or the ACPI SPCR table if supported by
989 cdns,<addr>[,options]
990 Start an early, polled-mode console on a Cadence
991 (xuartps) serial port at the specified address. Only
992 supported option is baud rate. If baud rate is not
993 specified, the serial port must already be setup and
996 uart[8250],io,<addr>[,options]
997 uart[8250],mmio,<addr>[,options]
998 uart[8250],mmio32,<addr>[,options]
999 uart[8250],mmio32be,<addr>[,options]
1000 uart[8250],0x<addr>[,options]
1001 Start an early, polled-mode console on the 8250/16550
1002 UART at the specified I/O port or MMIO address.
1003 MMIO inter-register address stride is either 8-bit
1004 (mmio) or 32-bit (mmio32 or mmio32be).
1005 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1006 to be equivalent to 'mmio'. 'options' are specified
1007 in the same format described for "console=ttyS<n>"; if
1008 unspecified, the h/w is not initialized.
1012 Start an early, polled-mode console on a pl011 serial
1013 port at the specified address. The pl011 serial port
1014 must already be setup and configured. Options are not
1015 yet supported. If 'mmio32' is specified, then only
1016 the driver will use only 32-bit accessors to read/write
1017 the device registers.
1020 Start an early, polled-mode console on a meson serial
1021 port at the specified address. The serial port must
1022 already be setup and configured. Options are not yet
1026 Start an early, polled-mode console on an msm serial
1027 port at the specified address. The serial port
1028 must already be setup and configured. Options are not
1031 msm_serial_dm,<addr>
1032 Start an early, polled-mode console on an msm serial
1033 dm port at the specified address. The serial port
1034 must already be setup and configured. Options are not
1038 Start an early, polled-mode console on a serial port
1039 of an Actions Semi SoC, such as S500 or S900, at the
1040 specified address. The serial port must already be
1041 setup and configured. Options are not yet supported.
1044 Start an early, polled-mode console on a serial port
1045 of an RDA Micro SoC, such as RDA8810PL, at the
1046 specified address. The serial port must already be
1047 setup and configured. Options are not yet supported.
1050 Use RISC-V SBI (Supervisor Binary Interface) for early
1053 smh Use ARM semihosting calls for early console.
1061 Use early console provided by serial driver available
1062 on Samsung SoCs, requires selecting proper type and
1063 a correct base address of the selected UART port. The
1064 serial port must already be setup and configured.
1065 Options are not yet supported.
1068 Start an early, polled-mode console on a lantiq serial
1069 (lqasc) port at the specified address. The serial port
1070 must already be setup and configured. Options are not
1075 Use early console provided by Freescale LP UART driver
1076 found on Freescale Vybrid and QorIQ LS1021A processors.
1077 A valid base address must be provided, and the serial
1078 port must already be setup and configured.
1081 Start an early, polled-mode console on the
1082 Armada 3700 serial port at the specified
1083 address. The serial port must already be setup
1084 and configured. Options are not yet supported.
1087 Start an early, polled-mode console on a Qualcomm
1088 Generic Interface (GENI) based serial port at the
1089 specified address. The serial port must already be
1090 setup and configured. Options are not yet supported.
1093 Start an early, unaccelerated console on the EFI
1094 memory mapped framebuffer (if available). On cache
1095 coherent non-x86 systems that use system memory for
1096 the framebuffer, pass the 'ram' option so that it is
1097 mapped with the correct attributes.
1100 Use early console provided by Freescale LINFlexD UART
1101 serial driver for NXP S32V234 SoCs. A valid base
1102 address must be provided, and the serial port must
1103 already be setup and configured.
1105 earlyprintk= [X86,SH,ARM,M68k,S390]
1109 earlyprintk=serial[,ttySn[,baudrate]]
1110 earlyprintk=serial[,0x...[,baudrate]]
1111 earlyprintk=ttySn[,baudrate]
1112 earlyprintk=dbgp[debugController#]
1113 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1114 earlyprintk=xdbc[xhciController#]
1116 earlyprintk is useful when the kernel crashes before
1117 the normal console is initialized. It is not enabled by
1118 default because it has some cosmetic problems.
1120 Append ",keep" to not disable it when the real console
1123 Only one of vga, efi, serial, or usb debug port can
1126 Currently only ttyS0 and ttyS1 may be specified by
1127 name. Other I/O ports may be explicitly specified
1128 on some architectures (x86 and arm at least) by
1129 replacing ttySn with an I/O port address, like this:
1130 earlyprintk=serial,0x1008,115200
1131 You can find the port for a given device in
1132 /proc/tty/driver/serial:
1133 2: uart:ST16650V2 port:00001008 irq:18 ...
1135 Interaction with the standard serial driver is not
1138 The VGA and EFI output is eventually overwritten by
1141 The xen output can only be used by Xen PV guests.
1143 The sclp output can only be used on s390.
1145 The optional "force" to "pciserial" enables use of a
1146 PCI device even when its classcode is not of the
1149 edac_report= [HW,EDAC] Control how to report EDAC event
1150 Format: {"on" | "off" | "force"}
1151 on: enable EDAC to report H/W event. May be overridden
1152 by other higher priority error reporting module.
1153 off: disable H/W event reporting through EDAC.
1154 force: enforce the use of EDAC to report H/W event.
1157 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1160 This is designed to be used in conjunction with
1161 the boot argument: earlyprintk=vga
1164 Format: {"off" | "on" | "skip[mbr]"}
1167 Format: { "old_map", "nochunk", "noruntime", "debug",
1169 old_map [X86-64]: switch to the old ioremap-based EFI
1170 runtime services mapping. 32-bit still uses this one by
1172 nochunk: disable reading files in "chunks" in the EFI
1173 boot stub, as chunking can cause problems with some
1174 firmware implementations.
1175 noruntime : disable EFI runtime services support
1176 debug: enable misc debug output
1177 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1178 attribute may cause the kernel to reserve the
1179 memory range for a memory mapping driver to
1180 claim. Specify efi=nosoftreserve to disable this
1181 reservation and treat the memory by its base type
1182 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1184 efi_no_storage_paranoia [EFI; X86]
1185 Using this parameter you can use more than 50% of
1186 your efi variable storage. Use this parameter only if
1187 you are really sure that your UEFI does sane gc and
1188 fulfills the spec otherwise your board may brick.
1190 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1191 Add arbitrary attribute to specific memory range by
1192 updating original EFI memory map.
1193 Region of memory which aa attribute is added to is
1196 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1197 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1198 attribute is added to range 0x100000000-0x180000000 and
1199 0x10a0000000-0x1120000000.
1201 If efi_fake_mem=8G@9G:0x40000 is specified, the
1202 EFI_MEMORY_SP(0x40000) attribute is added to
1203 range 0x240000000-0x43fffffff.
1205 Using this parameter you can do debugging of EFI memmap
1206 related features. For example, you can do debugging of
1207 Address Range Mirroring feature even if your box
1208 doesn't support it, or mark specific memory as
1211 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1212 that is to be dynamically loaded by Linux. If there are
1213 multiple variables with the same name but with different
1214 vendor GUIDs, all of them will be loaded. See
1215 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1218 eisa_irq_edge= [PARISC,HW]
1219 See header of drivers/parisc/eisa.c.
1222 See comment before function elanfreq_setup() in
1223 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1225 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1226 Specifies physical address of start of kernel core
1227 image elf header and optionally the size. Generally
1228 kexec loader will pass this option to capture kernel.
1229 See Documentation/admin-guide/kdump/kdump.rst for details.
1231 enable_mtrr_cleanup [X86]
1232 The kernel tries to adjust MTRR layout from continuous
1233 to discrete, to make X server driver able to add WB
1234 entry later. This parameter enables that.
1236 enable_timer_pin_1 [X86]
1237 Enable PIN 1 of APIC timer
1238 Can be useful to work around chipset bugs
1239 (in particular on some ATI chipsets).
1240 The kernel tries to set a reasonable default.
1242 enforcing [SELINUX] Set initial enforcing status.
1244 See security/selinux/Kconfig help text.
1245 0 -- permissive (log only, no denials).
1246 1 -- enforcing (deny and log).
1248 Value can be changed at runtime via /selinux/enforce.
1251 Disable Error Record Serialization Table (ERST)
1254 ether= [HW,NET] Ethernet cards parameters
1255 This option is obsoleted by the "netdev=" option, which
1256 has equivalent usage. See its documentation for details.
1260 Permit 'security.evm' to be updated regardless of
1261 current integrity status.
1265 fail_make_request=[KNL]
1266 General fault injection mechanism.
1267 Format: <interval>,<probability>,<space>,<times>
1268 See also Documentation/fault-injection/.
1271 See Documentation/admin-guide/blockdev/floppy.rst.
1273 force_pal_cache_flush
1274 [IA-64] Avoid check_sal_cache_flush which may hang on
1275 buggy SAL_CACHE_FLUSH implementations. Using this
1276 parameter will force ia64_sal_cache_flush to call
1277 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1280 Forcefully enable Physical Address Extension (PAE).
1281 Many Pentium M systems disable PAE but may have a
1282 functionally usable PAE implementation.
1283 Warning: use of this parameter will taint the kernel
1284 and may cause unknown problems.
1287 [FTRACE] will set and start the specified tracer
1288 as early as possible in order to facilitate early
1291 ftrace_dump_on_oops[=orig_cpu]
1292 [FTRACE] will dump the trace buffers on oops.
1293 If no parameter is passed, ftrace will dump
1294 buffers of all CPUs, but if you pass orig_cpu, it will
1295 dump only the buffer of the CPU that triggered the
1298 ftrace_filter=[function-list]
1299 [FTRACE] Limit the functions traced by the function
1300 tracer at boot up. function-list is a comma separated
1301 list of functions. This list can be changed at run
1302 time by the set_ftrace_filter file in the debugfs
1305 ftrace_notrace=[function-list]
1306 [FTRACE] Do not trace the functions specified in
1307 function-list. This list can be changed at run time
1308 by the set_ftrace_notrace file in the debugfs
1311 ftrace_graph_filter=[function-list]
1312 [FTRACE] Limit the top level callers functions traced
1313 by the function graph tracer at boot up.
1314 function-list is a comma separated list of functions
1315 that can be changed at run time by the
1316 set_graph_function file in the debugfs tracing directory.
1318 ftrace_graph_notrace=[function-list]
1319 [FTRACE] Do not trace from the functions specified in
1320 function-list. This list is a comma separated list of
1321 functions that can be changed at run time by the
1322 set_graph_notrace file in the debugfs tracing directory.
1324 ftrace_graph_max_depth=<uint>
1325 [FTRACE] Used with the function graph tracer. This is
1326 the max depth it will trace into a function. This value
1327 can be changed at run time by the max_graph_depth file
1328 in the tracefs tracing directory. default: 0 (no limit)
1331 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1332 support via parallel port (up to 5 devices per port)
1333 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1334 See also Documentation/input/devices/joystick-parport.rst
1338 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1342 gcov_persist= [GCOV] When non-zero (default), profiling data for
1343 kernel modules is saved and remains accessible via
1344 debugfs, even when the module is unloaded/reloaded.
1345 When zero, profiling data is discarded and associated
1346 debugfs files are removed at module unload time.
1348 goldfish [X86] Enable the goldfish android emulator platform.
1349 Don't use this when you are not running on the
1352 gpt [EFI] Forces disk with valid GPT signature but
1353 invalid Protective MBR to be treated as GPT. If the
1354 primary GPT is corrupted, it enables the backup/alternate
1355 GPT to be used instead.
1357 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1358 the "Enable 0" bit of the configuration register.
1361 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1362 the "Enable 0" bit of the configuration register.
1365 grcan.select= [HW] Select which physical interface to use.
1368 grcan.txsize= [HW] Sets the size of the tx buffer.
1369 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1371 grcan.rxsize= [HW] Sets the size of the rx buffer.
1372 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1375 gpio-mockup.gpio_mockup_ranges
1376 [HW] Sets the ranges of gpiochip of for this device.
1377 Format: <start1>,<end1>,<start2>,<end2>...
1379 hardlockup_all_cpu_backtrace=
1380 [KNL] Should the hard-lockup detector generate
1381 backtraces on all cpus.
1384 hashdist= [KNL,NUMA] Large hashes allocated during boot
1385 are distributed across NUMA nodes. Defaults on
1386 for 64-bit NUMA, off otherwise.
1387 Format: 0 | 1 (for off | on)
1389 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1391 hd= [EIDE] (E)IDE hard drive subsystem geometry
1392 Format: <cyl>,<head>,<sect>
1395 Disable Hardware Error Source Table (HEST) support;
1396 corresponding firmware-first mode error processing
1397 logic will be disabled.
1399 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1400 size of <nn>. This works even on boxes that have no
1401 highmem otherwise. This also works to reduce highmem
1402 size on bigger boxes.
1404 highres= [KNL] Enable/disable high resolution timer mode.
1405 Valid parameters: "on", "off"
1410 hpet= [X86-32,HPET] option to control HPET usage
1411 Format: { enable (default) | disable | force |
1413 disable: disable HPET and use PIT instead
1414 force: allow force enabled of undocumented chips (ICH4,
1416 verbose: show contents of HPET registers during setup
1418 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1419 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1421 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1422 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
1423 On x86-64 and powerpc, this option can be specified
1424 multiple times interleaved with hugepages= to reserve
1425 huge pages of different sizes. Valid pages sizes on
1426 x86-64 are 2M (when the CPU supports "pse") and 1G
1427 (when the CPU supports the "pdpe1gb" cpuinfo flag).
1430 [KNL] Should the hung task detector generate panics.
1433 A nonzero value instructs the kernel to panic when a
1434 hung task is detected. The default value is controlled
1435 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1436 option. The value selected by this boot parameter can
1437 be changed later by the kernel.hung_task_panic sysctl.
1439 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1440 terminal devices. Valid values: 0..8
1441 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1442 If specified, z/VM IUCV HVC accepts connections
1443 from listed z/VM user IDs only.
1445 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1446 which allow the hypervisor to 'idle' the
1447 guest on lock contention.
1450 Do not unregister boot console at start. This is only
1451 useful for debugging when something happens in the window
1452 between unregistering the boot console and initializing
1455 i2c_bus= [HW] Override the default board specific I2C bus speed
1456 or register an additional I2C bus that is not
1457 registered from board initialization code.
1461 i8042.debug [HW] Toggle i8042 debug mode
1462 i8042.unmask_kbd_data
1463 [HW] Enable printing of interrupt data from the KBD port
1464 (disabled by default, and as a pre-condition
1465 requires that i8042.debug=1 be enabled)
1466 i8042.direct [HW] Put keyboard port into non-translated mode
1467 i8042.dumbkbd [HW] Pretend that controller can only read data from
1468 keyboard and cannot control its state
1469 (Don't attempt to blink the leds)
1470 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1471 i8042.nokbd [HW] Don't check/create keyboard port
1472 i8042.noloop [HW] Disable the AUX Loopback command while probing
1474 i8042.nomux [HW] Don't check presence of an active multiplexing
1476 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1478 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1479 i8042.reset [HW] Reset the controller during init, cleanup and
1480 suspend-to-ram transitions, only during s2r
1481 transitions, or never reset
1482 Format: { 1 | Y | y | 0 | N | n }
1483 1, Y, y: always reset controller
1484 0, N, n: don't ever reset controller
1485 Default: only on s2r transitions on x86; most other
1486 architectures force reset to be always executed
1487 i8042.unlock [HW] Unlock (ignore) the keylock
1488 i8042.kbdreset [HW] Reset device connected to KBD port
1492 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1493 indicates that the driver is running on unsupported
1495 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1496 does not match list of supported models.
1498 [HW] Report power status in /proc/i8k
1499 (disabled by default)
1500 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1503 i915.invert_brightness=
1504 [DRM] Invert the sense of the variable that is used to
1505 set the brightness of the panel backlight. Normally a
1506 brightness value of 0 indicates backlight switched off,
1507 and the maximum of the brightness value sets the backlight
1508 to maximum brightness. If this parameter is set to 0
1509 (default) and the machine requires it, or this parameter
1510 is set to 1, a brightness value of 0 sets the backlight
1511 to maximum brightness, and the maximum of the brightness
1512 value switches the backlight off.
1513 -1 -- never invert brightness
1514 0 -- machine default
1515 1 -- force brightness inversion
1518 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1520 ide-core.nodma= [HW] (E)IDE subsystem
1521 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1522 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1523 .cdrom .chs .ignore_cable are additional options
1524 See Documentation/ide/ide.rst.
1526 ide-generic.probe-mask= [HW] (E)IDE subsystem
1528 Probe mask for legacy ISA IDE ports. Depending on
1529 platform up to 6 ports are supported, enabled by
1530 setting corresponding bits in the mask to 1. The
1531 default value is 0x0, which has a special meaning.
1532 On systems that have PCI, it triggers scanning the
1533 PCI bus for the first and the second port, which
1534 are then probed. On systems without PCI the value
1535 of 0x0 enables probing the two first ports as if it
1538 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1539 Claim all unknown PCI IDE storage controllers.
1542 Format: idle=poll, idle=halt, idle=nomwait
1543 Poll forces a polling idle loop that can slightly
1544 improve the performance of waking up a idle CPU, but
1545 will use a lot of power and make the system run hot.
1547 idle=halt: Halt is forced to be used for CPU idle.
1548 In such case C2/C3 won't be used again.
1549 idle=nomwait: Disable mwait for CPU C-states
1551 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1552 Format: { strict | legacy | 2008 | relaxed }
1555 Choose which programs will be accepted for execution
1556 based on the IEEE 754 NaN encoding(s) supported by
1557 the FPU and the NaN encoding requested with the value
1558 of an ELF file header flag individually set by each
1559 binary. Hardware implementations are permitted to
1560 support either or both of the legacy and the 2008 NaN
1563 Available settings are as follows:
1564 strict accept binaries that request a NaN encoding
1565 supported by the FPU
1566 legacy only accept legacy-NaN binaries, if supported
1568 2008 only accept 2008-NaN binaries, if supported
1570 relaxed accept any binaries regardless of whether
1571 supported by the FPU
1573 The FPU emulator is always able to support both NaN
1574 encodings, so if no FPU hardware is present or it has
1575 been disabled with 'nofpu', then the settings of
1576 'legacy' and '2008' strap the emulator accordingly,
1577 'relaxed' straps the emulator for both legacy-NaN and
1578 2008-NaN, whereas 'strict' enables legacy-NaN only on
1579 legacy processors and both NaN encodings on MIPS32 or
1582 The setting for ABS.fmt/NEG.fmt instruction execution
1583 mode generally follows that for the NaN encoding,
1584 except where unsupported by hardware.
1586 ignore_loglevel [KNL]
1587 Ignore loglevel setting - this will print /all/
1588 kernel messages to the console. Useful for debugging.
1589 We also add it as printk module parameter, so users
1590 could change it dynamically, usually by
1591 /sys/module/printk/parameters/ignore_loglevel.
1594 Ignore RLIMIT_DATA setting for data mappings,
1595 print warning at first misuse. Can be changed via
1596 /sys/module/kernel/parameters/ignore_rlimit_data.
1598 ihash_entries= [KNL]
1599 Set number of hash buckets for inode cache.
1601 ima_appraise= [IMA] appraise integrity measurements
1602 Format: { "off" | "enforce" | "fix" | "log" }
1605 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1606 The builtin appraise policy appraises all files
1609 ima_canonical_fmt [IMA]
1610 Use the canonical format for the binary runtime
1611 measurements, instead of host native format.
1614 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1618 The list of supported hash algorithms is defined
1619 in crypto/hash_info.h.
1622 The builtin policies to load during IMA setup.
1623 Format: "tcb | appraise_tcb | secure_boot |
1626 The "tcb" policy measures all programs exec'd, files
1627 mmap'd for exec, and all files opened with the read
1628 mode bit set by either the effective uid (euid=0) or
1631 The "appraise_tcb" policy appraises the integrity of
1632 all files owned by root.
1634 The "secure_boot" policy appraises the integrity
1635 of files (eg. kexec kernel image, kernel modules,
1636 firmware, policy, etc) based on file signatures.
1638 The "fail_securely" policy forces file signature
1639 verification failure also on privileged mounted
1640 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1643 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1644 Load a policy which meets the needs of the Trusted
1645 Computing Base. This means IMA will measure all
1646 programs exec'd, files mmap'd for exec, and all files
1647 opened for read by uid=0.
1650 Select one of defined IMA measurements template formats.
1651 Formats: { "ima" | "ima-ng" | "ima-sig" }
1655 [IMA] Define a custom template format.
1656 Format: { "field1|...|fieldN" }
1658 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1659 Format: <min_file_size>
1660 Set the minimal file size for using asynchronous hash.
1661 If left unspecified, ahash usage is disabled.
1663 ahash performance varies for different data sizes on
1664 different crypto accelerators. This option can be used
1665 to achieve the best performance for a particular HW.
1667 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1669 Set hashing buffer size. Default: 4k.
1671 ahash performance varies for different chunk sizes on
1672 different crypto accelerators. This option can be used
1673 to achieve best performance for particular HW.
1677 Run specified binary instead of /sbin/init as init
1680 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1681 for working out where the kernel is dying during
1684 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1685 initcall functions. Useful for debugging built-in
1686 modules and initcalls.
1688 initrd= [BOOT] Specify the location of the initial ramdisk
1690 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
1693 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
1695 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
1697 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
1699 init_pkru= [x86] Specify the default memory protection keys rights
1700 register contents for all processes. 0x55555554 by
1701 default (disallow access to all but pkey 0). Can
1702 override in debugfs after boot.
1704 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1707 int_pln_enable [x86] Enable power limit notification interrupt
1709 integrity_audit=[IMA]
1710 Format: { "0" | "1" }
1711 0 -- basic integrity auditing messages. (Default)
1712 1 -- additional integrity auditing messages.
1714 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1716 Enable intel iommu driver.
1718 Disable intel iommu driver.
1719 igfx_off [Default Off]
1720 By default, gfx is mapped as normal device. If a gfx
1721 device has a dedicated DMAR unit, the DMAR unit is
1722 bypassed by not enabling DMAR with this option. In
1723 this case, gfx device will use physical address for
1726 With this option iommu will not optimize to look
1727 for io virtual address below 32-bit forcing dual
1728 address cycle on pci bus for cards supporting greater
1729 than 32-bit addressing. The default is to look
1730 for translation below 32-bit and if not available
1731 then look in the higher range.
1732 strict [Default Off]
1733 With this option on every unmap_single operation will
1734 result in a hardware IOTLB flush operation as opposed
1735 to batching them for performance.
1736 sp_off [Default Off]
1737 By default, super page will be supported if Intel IOMMU
1738 has the capability. With this option, super page will
1741 By default, scalable mode will be disabled even if the
1742 hardware advertises that it has support for the scalable
1743 mode translation. With this option set, scalable mode
1744 will be used on hardware which claims to support it.
1745 tboot_noforce [Default Off]
1746 Do not force the Intel IOMMU enabled under tboot.
1747 By default, tboot will force Intel IOMMU on, which
1748 could harm performance of some high-throughput
1749 devices like 40GBit network cards, even if identity
1751 Note that using this option lowers the security
1752 provided by tboot because it makes the system
1753 vulnerable to DMA attacks.
1754 nobounce [Default off]
1755 Disable bounce buffer for unstrusted devices such as
1756 the Thunderbolt devices. This will treat the untrusted
1757 devices as the trusted ones, hence might expose security
1758 risks of DMA attacks.
1760 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1761 0 disables intel_idle and fall back on acpi_idle.
1762 1 to 9 specify maximum depth of C-state.
1766 Do not enable intel_pstate as the default
1767 scaling driver for the supported processors
1769 Use intel_pstate as a scaling driver, but configure it
1770 to work with generic cpufreq governors (instead of
1771 enabling its internal governor). This mode cannot be
1772 used along with the hardware-managed P-states (HWP)
1775 Enable intel_pstate on systems that prohibit it by default
1776 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1777 instead of acpi-cpufreq may disable platform features, such
1778 as thermal controls and power capping, that rely on ACPI
1779 P-States information being indicated to OSPM and therefore
1780 should be used with caution. This option does not work with
1781 processors that aren't supported by the intel_pstate driver
1782 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1784 Do not enable hardware P state control (HWP)
1787 Only load intel_pstate on systems which support
1788 hardware P state control (HWP) if available.
1790 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1791 Description Table, specifies preferred power management
1792 profile as "Enterprise Server" or "Performance Server",
1793 then this feature is turned on by default.
1795 Allow per-logical-CPU P-State performance control limits using
1796 cpufreq sysfs interface
1798 intremap= [X86-64, Intel-IOMMU]
1799 on enable Interrupt Remapping (default)
1800 off disable Interrupt Remapping
1801 nosid disable Source ID checking
1803 BIOS x2APIC opt-out request will be ignored
1804 nopost disable Interrupt Posting
1806 iomem= Disable strict checking of access to MMIO memory
1807 strict regions from userspace.
1822 nobypass [PPC/POWERNV]
1823 Disable IOMMU bypass, using IOMMU for PCI devices.
1825 iommu.strict= [ARM64] Configure TLB invalidation behaviour
1826 Format: { "0" | "1" }
1828 Request that DMA unmap operations use deferred
1829 invalidation of hardware TLBs, for increased
1830 throughput at the cost of reduced device isolation.
1831 Will fall back to strict mode if not supported by
1832 the relevant IOMMU driver.
1833 1 - Strict mode (default).
1834 DMA unmap operations invalidate IOMMU hardware TLBs
1838 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
1839 Format: { "0" | "1" }
1840 0 - Use IOMMU translation for DMA.
1841 1 - Bypass the IOMMU for DMA.
1842 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
1844 io7= [HW] IO7 for Marvel based alpha systems
1845 See comment before marvel_specify_io7 in
1846 arch/alpha/kernel/core_marvel.c.
1848 io_delay= [X86] I/O delay method
1850 Standard port 0x80 based delay
1852 Alternate port 0xed based delay (needed on some systems)
1854 Simple two microseconds delay
1859 See Documentation/filesystems/nfs/nfsroot.txt.
1861 ipcmni_extend [KNL] Extend the maximum number of unique System V
1862 IPC identifiers from 32,768 to 16,777,216.
1864 irqaffinity= [SMP] Set the default irq affinity mask
1865 The argument is a cpu list, as described above.
1867 irqchip.gicv2_force_probe=
1870 Force the kernel to look for the second 4kB page
1871 of a GICv2 controller even if the memory range
1872 exposed by the device tree is too small.
1874 irqchip.gicv3_nolpi=
1876 Force the kernel to ignore the availability of
1877 LPIs (and by consequence ITSs). Intended for system
1878 that use the kernel as a bootloader, and thus want
1879 to let secondary kernels in charge of setting up
1882 irqchip.gicv3_pseudo_nmi= [ARM64]
1883 Enables support for pseudo-NMIs in the kernel. This
1884 requires the kernel to be built with
1885 CONFIG_ARM64_PSEUDO_NMI.
1888 When an interrupt is not handled search all handlers
1889 for it. Intended to get systems with badly broken
1893 When an interrupt is not handled search all handlers
1894 for it. Also check all handlers each timer
1895 interrupt. Intended to get systems with badly broken
1899 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1901 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
1902 [Deprecated - use cpusets instead]
1903 Format: [flag-list,]<cpu-list>
1905 Specify one or more CPUs to isolate from disturbances
1906 specified in the flag list (default: domain):
1909 Disable the tick when a single task runs.
1911 A residual 1Hz tick is offloaded to workqueues, which you
1912 need to affine to housekeeping through the global
1913 workqueue's affinity configured via the
1914 /sys/devices/virtual/workqueue/cpumask sysfs file, or
1915 by using the 'domain' flag described below.
1917 NOTE: by default the global workqueue runs on all CPUs,
1918 so to protect individual CPUs the 'cpumask' file has to
1919 be configured manually after bootup.
1922 Isolate from the general SMP balancing and scheduling
1923 algorithms. Note that performing domain isolation this way
1924 is irreversible: it's not possible to bring back a CPU to
1925 the domains once isolated through isolcpus. It's strongly
1926 advised to use cpusets instead to disable scheduler load
1927 balancing through the "cpuset.sched_load_balance" file.
1928 It offers a much more flexible interface where CPUs can
1929 move in and out of an isolated set anytime.
1931 You can move a process onto or off an "isolated" CPU via
1932 the CPU affinity syscalls or cpuset.
1933 <cpu number> begins at 0 and the maximum value is
1934 "number of CPUs in system - 1".
1936 The format of <cpu-list> is described above.
1942 ivrs_ioapic [HW,X86_64]
1943 Provide an override to the IOAPIC-ID<->DEVICE-ID
1944 mapping provided in the IVRS ACPI table. For
1945 example, to map IOAPIC-ID decimal 10 to
1946 PCI device 00:14.0 write the parameter as:
1947 ivrs_ioapic[10]=00:14.0
1949 ivrs_hpet [HW,X86_64]
1950 Provide an override to the HPET-ID<->DEVICE-ID
1951 mapping provided in the IVRS ACPI table. For
1952 example, to map HPET-ID decimal 0 to
1953 PCI device 00:14.0 write the parameter as:
1954 ivrs_hpet[0]=00:14.0
1956 ivrs_acpihid [HW,X86_64]
1957 Provide an override to the ACPI-HID:UID<->DEVICE-ID
1958 mapping provided in the IVRS ACPI table. For
1959 example, to map UART-HID:UID AMD0020:0 to
1960 PCI device 00:14.5 write the parameter as:
1961 ivrs_acpihid[00:14.5]=AMD0020:0
1963 js= [HW,JOY] Analog joystick
1964 See Documentation/input/joydev/joystick.rst.
1967 When CONFIG_RANDOMIZE_BASE is set, this disables
1968 kernel and module base offset ASLR (Address Space
1969 Layout Randomization).
1972 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
1973 report on every invalid memory access. Without this
1974 parameter KASAN will print report only for the first
1979 kernelcore= [KNL,X86,IA-64,PPC]
1980 Format: nn[KMGTPE] | nn% | "mirror"
1981 This parameter specifies the amount of memory usable by
1982 the kernel for non-movable allocations. The requested
1983 amount is spread evenly throughout all nodes in the
1984 system as ZONE_NORMAL. The remaining memory is used for
1985 movable memory in its own zone, ZONE_MOVABLE. In the
1986 event, a node is too small to have both ZONE_NORMAL and
1987 ZONE_MOVABLE, kernelcore memory will take priority and
1988 other nodes will have a larger ZONE_MOVABLE.
1990 ZONE_MOVABLE is used for the allocation of pages that
1991 may be reclaimed or moved by the page migration
1992 subsystem. Note that allocations like PTEs-from-HighMem
1993 still use the HighMem zone if it exists, and the Normal
1994 zone if it does not.
1996 It is possible to specify the exact amount of memory in
1997 the form of "nn[KMGTPE]", a percentage of total system
1998 memory in the form of "nn%", or "mirror". If "mirror"
1999 option is specified, mirrored (reliable) memory is used
2000 for non-movable allocations and remaining memory is used
2001 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2002 are exclusive, so you cannot specify multiple forms.
2004 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2005 Format: <Controller#>[,poll interval]
2006 The controller # is the number of the ehci usb debug
2007 port as it is probed via PCI. The poll interval is
2008 optional and is the number seconds in between
2009 each poll cycle to the debug port in case you need
2010 the functionality for interrupting the kernel with
2011 gdb or control-c on the dbgp connection. When
2012 not using this parameter you use sysrq-g to break into
2013 the kernel debugger.
2015 kgdboc= [KGDB,HW] kgdb over consoles.
2016 Requires a tty driver that supports console polling,
2017 or a supported polling keyboard driver (non-usb).
2018 Serial only format: <serial_device>[,baud]
2019 keyboard only format: kbd
2020 keyboard and serial format: kbd,<serial_device>[,baud]
2021 Optional Kernel mode setting:
2022 kms, kbd format: kms,kbd
2023 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2025 kgdbwait [KGDB] Stop kernel execution and enter the
2026 kernel debugger at the earliest opportunity.
2028 kmac= [MIPS] korina ethernet MAC address.
2029 Configure the RouterBoard 532 series on-chip
2030 Ethernet adapter MAC address.
2032 kmemleak= [KNL] Boot-time kmemleak enable/disable
2033 Valid arguments: on, off
2035 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2038 kprobe_event=[probe-list]
2039 [FTRACE] Add kprobe events and enable at boot time.
2040 The probe-list is a semicolon delimited list of probe
2041 definitions. Each definition is same as kprobe_events
2042 interface, but the parameters are comma delimited.
2043 For example, to add a kprobe event on vfs_read with
2044 arg1 and arg2, add to the command line;
2046 kprobe_event=p,vfs_read,$arg1,$arg2
2048 See also Documentation/trace/kprobetrace.rst "Kernel
2049 Boot Parameter" section.
2051 kpti= [ARM64] Control page table isolation of user
2052 and kernel address spaces.
2053 Default: enabled on cores which need mitigation.
2057 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2058 Default is 0 (don't ignore, but inject #GP)
2060 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2061 Default is false (don't support).
2063 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
2068 [KVM] Controls the software workaround for the
2069 X86_BUG_ITLB_MULTIHIT bug.
2070 force : Always deploy workaround.
2071 off : Never deploy workaround.
2072 auto : Deploy workaround based on the presence of
2073 X86_BUG_ITLB_MULTIHIT.
2077 If the software workaround is enabled for the host,
2078 guests do need not to enable it for nested guests.
2080 kvm.nx_huge_pages_recovery_ratio=
2081 [KVM] Controls how many 4KiB pages are periodically zapped
2082 back to huge pages. 0 disables the recovery, otherwise if
2083 the value is N KVM will zap 1/Nth of the 4KiB pages every
2084 minute. The default is 60.
2086 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2087 Default is 1 (enabled)
2089 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2091 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2093 kvm-arm.vgic_v3_group0_trap=
2094 [KVM,ARM] Trap guest accesses to GICv3 group-0
2097 kvm-arm.vgic_v3_group1_trap=
2098 [KVM,ARM] Trap guest accesses to GICv3 group-1
2101 kvm-arm.vgic_v3_common_trap=
2102 [KVM,ARM] Trap guest accesses to GICv3 common
2105 kvm-arm.vgic_v4_enable=
2106 [KVM,ARM] Allow use of GICv4 for direct injection of
2109 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2110 (virtualized MMU) support on capable Intel chips.
2111 Default is 1 (enabled)
2113 kvm-intel.emulate_invalid_guest_state=
2114 [KVM,Intel] Enable emulation of invalid guest states
2115 Default is 0 (disabled)
2117 kvm-intel.flexpriority=
2118 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2119 Default is 1 (enabled)
2122 [KVM,Intel] Enable VMX nesting (nVMX).
2123 Default is 0 (disabled)
2125 kvm-intel.unrestricted_guest=
2126 [KVM,Intel] Disable unrestricted guest feature
2127 (virtualized real and unpaged mode) on capable
2128 Intel chips. Default is 1 (enabled)
2130 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2133 Valid arguments: never, cond, always
2135 always: L1D cache flush on every VMENTER.
2136 cond: Flush L1D on VMENTER only when the code between
2137 VMEXIT and VMENTER can leak host memory.
2138 never: Disables the mitigation
2140 Default is cond (do L1 cache flush in specific instances)
2142 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2143 feature (tagged TLBs) on capable Intel chips.
2144 Default is 1 (enabled)
2146 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2149 The kernel PTE inversion protection is unconditionally
2150 enabled and cannot be disabled.
2153 Provides all available mitigations for the
2154 L1TF vulnerability. Disables SMT and
2155 enables all mitigations in the
2156 hypervisors, i.e. unconditional L1D flush.
2158 SMT control and L1D flush control via the
2159 sysfs interface is still possible after
2160 boot. Hypervisors will issue a warning
2161 when the first VM is started in a
2162 potentially insecure configuration,
2163 i.e. SMT enabled or L1D flush disabled.
2166 Same as 'full', but disables SMT and L1D
2167 flush runtime control. Implies the
2168 'nosmt=force' command line option.
2169 (i.e. sysfs control of SMT is disabled.)
2172 Leaves SMT enabled and enables the default
2173 hypervisor mitigation, i.e. conditional
2176 SMT control and L1D flush control via the
2177 sysfs interface is still possible after
2178 boot. Hypervisors will issue a warning
2179 when the first VM is started in a
2180 potentially insecure configuration,
2181 i.e. SMT enabled or L1D flush disabled.
2185 Disables SMT and enables the default
2186 hypervisor mitigation.
2188 SMT control and L1D flush control via the
2189 sysfs interface is still possible after
2190 boot. Hypervisors will issue a warning
2191 when the first VM is started in a
2192 potentially insecure configuration,
2193 i.e. SMT enabled or L1D flush disabled.
2196 Same as 'flush', but hypervisors will not
2197 warn when a VM is started in a potentially
2198 insecure configuration.
2201 Disables hypervisor mitigations and doesn't
2203 It also drops the swap size and available
2204 RAM limit restriction on both hypervisor and
2209 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2215 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2218 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
2219 value for LAPIC timer one-shot implementation. Default
2220 back to the programmable timer unit in the LAPIC.
2222 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2225 libata.dma= [LIBATA] DMA control
2226 libata.dma=0 Disable all PATA and SATA DMA
2227 libata.dma=1 PATA and SATA Disk DMA only
2228 libata.dma=2 ATAPI (CDROM) DMA only
2229 libata.dma=4 Compact Flash DMA only
2230 Combinations also work, so libata.dma=3 enables DMA
2231 for disks and CDROMs, but not CFs.
2233 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2234 libata.ignore_hpa=0 keep BIOS limits (default)
2235 libata.ignore_hpa=1 ignore limits, using full disk
2237 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2241 libata.force= [LIBATA] Force configurations. The format is comma
2242 separated list of "[ID:]VAL" where ID is
2243 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2244 matching port, link or device. Basically, it matches
2245 the ATA ID string printed on console by libata. If
2246 the whole ID part is omitted, the last PORT and DEVICE
2247 values are used. If ID hasn't been specified yet, the
2248 configuration applies to all ports, links and devices.
2250 If only DEVICE is omitted, the parameter applies to
2251 the port and all links and devices behind it. DEVICE
2252 number of 0 either selects the first device or the
2253 first fan-out link behind PMP device. It does not
2254 select the host link. DEVICE number of 15 selects the
2255 host link and device attached to it.
2257 The VAL specifies the configuration to force. As long
2258 as there's no ambiguity shortcut notation is allowed.
2259 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2260 The following configurations can be forced.
2262 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2263 Any ID with matching PORT is used.
2265 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2267 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2268 udma[/][16,25,33,44,66,100,133] notation is also
2271 * [no]ncq: Turn on or off NCQ.
2273 * [no]ncqtrim: Turn off queued DSM TRIM.
2275 * nohrst, nosrst, norst: suppress hard, soft
2278 * rstonce: only attempt one reset during
2279 hot-unplug link recovery
2281 * dump_id: dump IDENTIFY data.
2283 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2285 * disable: Disable this device.
2287 If there are multiple matching configurations changing
2288 the same attribute, the last one is used.
2290 memblock=debug [KNL] Enable memblock debug messages.
2292 load_ramdisk= [RAM] List of ramdisks to load from floppy
2293 See Documentation/admin-guide/blockdev/ramdisk.rst.
2295 lockd.nlm_grace_period=P [NFS] Assign grace period.
2298 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2301 lockd.nlm_timeout=T [NFS] Assign timeout value.
2304 lockd.nlm_udpport=M [NFS] Assign UDP port.
2307 lockdown= [SECURITY]
2308 { integrity | confidentiality }
2309 Enable the kernel lockdown feature. If set to
2310 integrity, kernel features that allow userland to
2311 modify the running kernel are disabled. If set to
2312 confidentiality, kernel features that allow userland
2313 to extract confidential information from the kernel
2316 locktorture.nreaders_stress= [KNL]
2317 Set the number of locking read-acquisition kthreads.
2318 Defaults to being automatically set based on the
2319 number of online CPUs.
2321 locktorture.nwriters_stress= [KNL]
2322 Set the number of locking write-acquisition kthreads.
2324 locktorture.onoff_holdoff= [KNL]
2325 Set time (s) after boot for CPU-hotplug testing.
2327 locktorture.onoff_interval= [KNL]
2328 Set time (s) between CPU-hotplug operations, or
2329 zero to disable CPU-hotplug testing.
2331 locktorture.shuffle_interval= [KNL]
2332 Set task-shuffle interval (jiffies). Shuffling
2333 tasks allows some CPUs to go into dyntick-idle
2334 mode during the locktorture test.
2336 locktorture.shutdown_secs= [KNL]
2337 Set time (s) after boot system shutdown. This
2338 is useful for hands-off automated testing.
2340 locktorture.stat_interval= [KNL]
2341 Time (s) between statistics printk()s.
2343 locktorture.stutter= [KNL]
2344 Time (s) to stutter testing, for example,
2345 specifying five seconds causes the test to run for
2346 five seconds, wait for five seconds, and so on.
2347 This tests the locking primitive's ability to
2348 transition abruptly to and from idle.
2350 locktorture.torture_type= [KNL]
2351 Specify the locking implementation to test.
2353 locktorture.verbose= [KNL]
2354 Enable additional printk() statements.
2356 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2359 loglevel= All Kernel Messages with a loglevel smaller than the
2360 console loglevel will be printed to the console. It can
2361 also be changed with klogd or other programs. The
2362 loglevels are defined as follows:
2364 0 (KERN_EMERG) system is unusable
2365 1 (KERN_ALERT) action must be taken immediately
2366 2 (KERN_CRIT) critical conditions
2367 3 (KERN_ERR) error conditions
2368 4 (KERN_WARNING) warning conditions
2369 5 (KERN_NOTICE) normal but significant condition
2370 6 (KERN_INFO) informational
2371 7 (KERN_DEBUG) debug-level messages
2373 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2374 in bytes. n must be a power of two and greater
2375 than the minimal size. The minimal size is defined
2376 by LOG_BUF_SHIFT kernel config parameter. There is
2377 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2378 that allows to increase the default size depending on
2379 the number of CPUs. See init/Kconfig for more details.
2381 logo.nologo [FB] Disables display of the built-in Linux logo.
2382 This may be used to provide more screen space for
2383 kernel log messages and is useful when debugging
2384 kernel boot problems.
2386 lp=0 [LP] Specify parallel ports to use, e.g,
2387 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2388 lp=reset first parallel port). 'lp=0' disables the
2389 lp=auto printer driver. 'lp=reset' (which can be
2390 specified in addition to the ports) causes
2391 attached printers to be reset. Using
2392 lp=port1,port2,... specifies the parallel ports
2393 to associate lp devices with, starting with
2394 lp0. A port specification may be 'none' to skip
2395 that lp device, or a parport name such as
2396 'parport0'. Specifying 'lp=auto' instead of a
2397 port specification list means that device IDs
2398 from each port should be examined, to see if
2399 an IEEE 1284-compliant printer is attached; if
2400 so, the driver will manage that printer.
2401 See also header of drivers/char/lp.c.
2404 Sets loops_per_jiffy to given constant, thus avoiding
2405 time-consuming boot-time autodetection (up to 250 ms per
2406 CPU). 0 enables autodetection (default). To determine
2407 the correct value for your kernel, boot with normal
2408 autodetection and see what value is printed. Note that
2409 on SMP systems the preset will be applied to all CPUs,
2410 which is likely to cause problems if your CPUs need
2411 significantly divergent settings. An incorrect value
2412 will cause delays in the kernel to be wrong, leading to
2413 unpredictable I/O errors and other breakage. Although
2414 unlikely, in the extreme case this might damage your
2418 Format: <io>,<irq>,<dma>
2420 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2423 [SECURITY] Choose order of LSM initialization. This
2424 overrides CONFIG_LSM, and the "security=" parameter.
2426 machvec= [IA-64] Force the use of a particular machine-vector
2427 (machvec) in a generic kernel.
2428 Example: machvec=hpzx1
2430 machtype= [Loongson] Share the same kernel image file between different
2432 Example: machtype=lemote-yeeloong-2f-7inch
2434 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2435 than or equal to this physical address is ignored.
2437 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2438 will bring up during bootup. maxcpus=n : n >= 0 limits
2439 the kernel to bring up 'n' processors. Surely after
2440 bootup you can bring up the other plugged cpu by executing
2441 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2442 only takes effect during system bootup.
2443 While n=0 is a special case, it is equivalent to "nosmp",
2444 which also disables the IO APIC.
2446 max_loop= [LOOP] The number of loop block devices that get
2447 (loop.max_loop) unconditionally pre-created at init time. The default
2448 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2449 of statically allocating a predefined number, loop
2450 devices can be requested on-demand with the
2451 /dev/loop-control interface.
2453 mce [X86-32] Machine Check Exception
2455 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2457 md= [HW] RAID subsystems devices and level
2458 See Documentation/admin-guide/md.rst.
2461 Format: <first>,<last>
2462 Specifies range of consoles to be captured by the MDA.
2465 Control mitigation for the Micro-architectural Data
2466 Sampling (MDS) vulnerability.
2468 Certain CPUs are vulnerable to an exploit against CPU
2469 internal buffers which can forward information to a
2470 disclosure gadget under certain conditions.
2472 In vulnerable processors, the speculatively
2473 forwarded data can be used in a cache side channel
2474 attack, to access data to which the attacker does
2475 not have direct access.
2477 This parameter controls the MDS mitigation. The
2480 full - Enable MDS mitigation on vulnerable CPUs
2481 full,nosmt - Enable MDS mitigation and disable
2482 SMT on vulnerable CPUs
2483 off - Unconditionally disable MDS mitigation
2485 On TAA-affected machines, mds=off can be prevented by
2486 an active TAA mitigation as both vulnerabilities are
2487 mitigated with the same mechanism so in order to disable
2488 this mitigation, you need to specify tsx_async_abort=off
2491 Not specifying this option is equivalent to
2494 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2496 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2497 Amount of memory to be used when the kernel is not able
2498 to see the whole system memory or for test.
2499 [X86] Work as limiting max address. Use together
2500 with memmap= to avoid physical address space collisions.
2501 Without memmap= PCI devices could be placed at addresses
2502 belonging to unused RAM.
2504 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2508 [KNL,SH] Allow user to override the default size for
2509 per-device physically contiguous DMA buffers.
2511 memhp_default_state=online/offline
2512 [KNL] Set the initial state for the memory hotplug
2513 onlining policy. If not specified, the default value is
2514 set according to the
2515 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2517 See Documentation/admin-guide/mm/memory-hotplug.rst.
2519 memmap=exactmap [KNL,X86] Enable setting of an exact
2520 E820 memory map, as specified by the user.
2521 Such memmap=exactmap lines can be constructed based on
2522 BIOS output or other requirements. See the memmap=nn@ss
2525 memmap=nn[KMG]@ss[KMG]
2526 [KNL] Force usage of a specific region of memory.
2527 Region of memory to be used is from ss to ss+nn.
2528 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2529 which limits max address to nn[KMG].
2530 Multiple different regions can be specified,
2533 memmap=100M@2G,100M#3G,1G!1024G
2535 memmap=nn[KMG]#ss[KMG]
2536 [KNL,ACPI] Mark specific memory as ACPI data.
2537 Region of memory to be marked is from ss to ss+nn.
2539 memmap=nn[KMG]$ss[KMG]
2540 [KNL,ACPI] Mark specific memory as reserved.
2541 Region of memory to be reserved is from ss to ss+nn.
2542 Example: Exclude memory from 0x18690000-0x1869ffff
2543 memmap=64K$0x18690000
2545 memmap=0x10000$0x18690000
2546 Some bootloaders may need an escape character before '$',
2547 like Grub2, otherwise '$' and the following number
2550 memmap=nn[KMG]!ss[KMG]
2551 [KNL,X86] Mark specific memory as protected.
2552 Region of memory to be used, from ss to ss+nn.
2553 The memory region may be marked as e820 type 12 (0xc)
2554 and is NVDIMM or ADR memory.
2556 memmap=<size>%<offset>-<oldtype>+<newtype>
2557 [KNL,ACPI] Convert memory within the specified region
2558 from <oldtype> to <newtype>. If "-<oldtype>" is left
2559 out, the whole region will be marked as <newtype>,
2560 even if previously unavailable. If "+<newtype>" is left
2561 out, matching memory will be removed. Types are
2562 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2563 3 = ACPI, 12 = PRAM.
2565 memory_corruption_check=0/1 [X86]
2566 Some BIOSes seem to corrupt the first 64k of
2567 memory when doing things like suspend/resume.
2568 Setting this option will scan the memory
2569 looking for corruption. Enabling this will
2570 both detect corruption and prevent the kernel
2571 from using the memory being corrupted.
2572 However, its intended as a diagnostic tool; if
2573 repeatable BIOS-originated corruption always
2574 affects the same memory, you can use memmap=
2575 to prevent the kernel from using that memory.
2577 memory_corruption_check_size=size [X86]
2578 By default it checks for corruption in the low
2579 64k, making this memory unavailable for normal
2580 use. Use this parameter to scan for
2581 corruption in more or less memory.
2583 memory_corruption_check_period=seconds [X86]
2584 By default it checks for corruption every 60
2585 seconds. Use this parameter to check at some
2586 other rate. 0 disables periodic checking.
2588 memtest= [KNL,X86,ARM,PPC] Enable memtest
2590 default : 0 <disable>
2591 Specifies the number of memtest passes to be
2592 performed. Each pass selects another test
2593 pattern from a given set of patterns. Memtest
2594 fills the memory with this pattern, validates
2595 memory contents and reserves bad memory
2596 regions that are detected.
2598 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2599 Valid arguments: on, off
2600 Default (depends on kernel configuration option):
2601 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2602 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2603 mem_encrypt=on: Activate SME
2604 mem_encrypt=off: Do not activate SME
2606 Refer to Documentation/virt/kvm/amd-memory-encryption.rst
2607 for details on when memory encryption can be activated.
2609 mem_sleep_default= [SUSPEND] Default system suspend mode:
2610 s2idle - Suspend-To-Idle
2611 shallow - Power-On Suspend or equivalent (if supported)
2612 deep - Suspend-To-RAM or equivalent (if supported)
2613 See Documentation/admin-guide/pm/sleep-states.rst.
2615 meye.*= [HW] Set MotionEye Camera parameters
2616 See Documentation/media/v4l-drivers/meye.rst.
2618 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2619 Multi-Function General Purpose Timers on AMD Geode
2622 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2623 the BIOS has incorrectly applied a workaround. TinyBIOS
2624 version 0.98 is known to be affected, 0.99 fixes the
2625 problem by letting the user disable the workaround.
2629 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2630 physical address is ignored.
2632 mini2440= [ARM,HW,KNL]
2633 Format:[0..2][b][c][t]
2635 MINI2440 configuration specification:
2636 0 - The attached screen is the 3.5" TFT
2637 1 - The attached screen is the 7" TFT
2638 2 - The VGA Shield is attached (1024x768)
2639 Leaving out the screen size parameter will not load
2640 the TFT driver, and the framebuffer will be left
2642 b - Enable backlight. The TFT backlight pin will be
2643 linked to the kernel VESA blanking code and a GPIO
2644 LED. This parameter is not necessary when using the
2646 c - Enable the s3c camera interface.
2647 t - Reserved for enabling touchscreen support. The
2648 touchscreen support is not enabled in the mainstream
2649 kernel as of 2.6.30, a preliminary port can be found
2650 in the "bleeding edge" mini2440 support kernel at
2651 http://repo.or.cz/w/linux-2.6/mini2440.git
2654 [X86,PPC,S390,ARM64] Control optional mitigations for
2655 CPU vulnerabilities. This is a set of curated,
2656 arch-independent options, each of which is an
2657 aggregation of existing arch-specific options.
2660 Disable all optional CPU mitigations. This
2661 improves system performance, but it may also
2662 expose users to several CPU vulnerabilities.
2663 Equivalent to: nopti [X86,PPC]
2665 nospectre_v1 [X86,PPC]
2667 nospectre_v2 [X86,PPC,S390,ARM64]
2668 spectre_v2_user=off [X86]
2669 spec_store_bypass_disable=off [X86,PPC]
2670 ssbd=force-off [ARM64]
2673 tsx_async_abort=off [X86]
2674 kvm.nx_huge_pages=off [X86]
2677 This does not have any effect on
2678 kvm.nx_huge_pages when
2679 kvm.nx_huge_pages=force.
2682 Mitigate all CPU vulnerabilities, but leave SMT
2683 enabled, even if it's vulnerable. This is for
2684 users who don't want to be surprised by SMT
2685 getting disabled across kernel upgrades, or who
2686 have other ways of avoiding SMT-based attacks.
2687 Equivalent to: (default behavior)
2690 Mitigate all CPU vulnerabilities, disabling SMT
2691 if needed. This is for users who always want to
2692 be fully mitigated, even if it means losing SMT.
2693 Equivalent to: l1tf=flush,nosmt [X86]
2694 mds=full,nosmt [X86]
2695 tsx_async_abort=full,nosmt [X86]
2698 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2699 parameter allows control of the logging verbosity for
2700 the additional memory initialisation checks. A value
2701 of 0 disables mminit logging and a level of 4 will
2702 log everything. Information is printed at KERN_DEBUG
2703 so loglevel=8 may also need to be specified.
2706 [KNL] When CONFIG_MODULE_SIG is set, this means that
2707 modules without (valid) signatures will fail to load.
2708 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2709 is always true, so this option does nothing.
2711 module_blacklist= [KNL] Do not load a comma-separated list of
2712 modules. Useful for debugging problem modules.
2715 [MOUSE] Maximum time between finger touching and
2716 leaving touchpad surface for touch to be considered
2717 a tap and be reported as a left button click (for
2718 touchpads working in absolute mode only).
2720 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2721 reporting absolute coordinates, such as tablets
2722 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2723 reporting absolute coordinates, such as tablets
2725 movablecore= [KNL,X86,IA-64,PPC]
2726 Format: nn[KMGTPE] | nn%
2727 This parameter is the complement to kernelcore=, it
2728 specifies the amount of memory used for migratable
2729 allocations. If both kernelcore and movablecore is
2730 specified, then kernelcore will be at *least* the
2731 specified value but may be more. If movablecore on its
2732 own is specified, the administrator must be careful
2733 that the amount of memory usable for all allocations
2736 movable_node [KNL] Boot-time switch to make hotplugable memory
2737 NUMA nodes to be movable. This means that the memory
2738 of such nodes will be usable only for movable
2739 allocations which rules out almost all kernel
2740 allocations. Use with caution!
2742 MTD_Partition= [MTD]
2743 Format: <name>,<region-number>,<size>,<offset>
2745 MTD_Region= [MTD] Format:
2746 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2749 See drivers/mtd/cmdlinepart.c.
2751 multitce=off [PPC] This parameter disables the use of the pSeries
2752 firmware feature for updating multiple TCE entries
2755 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2757 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2759 boundary - index of last SLC block on Flex-OneNAND.
2760 The remaining blocks are configured as MLC blocks.
2761 lock - Configure if Flex-OneNAND boundary should be locked.
2762 Once locked, the boundary cannot be changed.
2763 1 indicates lock status, 0 indicates unlock status.
2766 ARM/S3C2412 JIVE boot control
2768 See arch/arm/mach-s3c2412/mach-jive.c
2770 mtouchusb.raw_coordinates=
2771 [HW] Make the MicroTouch USB driver use raw coordinates
2772 ('y', default) or cooked coordinates ('n')
2774 mtrr_chunk_size=nn[KMG] [X86]
2775 used for mtrr cleanup. It is largest continuous chunk
2776 that could hold holes aka. UC entries.
2778 mtrr_gran_size=nn[KMG] [X86]
2779 Used for mtrr cleanup. It is granularity of mtrr block.
2781 Large value could prevent small alignment from
2784 mtrr_spare_reg_nr=n [X86]
2786 Range: 0,7 : spare reg number
2788 Used for mtrr cleanup. It is spare mtrr entries number.
2789 Set to 2 or more if your graphical card needs more.
2791 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2793 netdev= [NET] Network devices parameters
2794 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2795 Note that mem_start is often overloaded to mean
2796 something different and driver-specific.
2797 This usage is only documented in each driver source
2801 [NETFILTER] Enable connection tracking flow accounting
2802 0 to disable accounting
2803 1 to enable accounting
2806 nfsaddrs= [NFS] Deprecated. Use ip= instead.
2807 See Documentation/filesystems/nfs/nfsroot.txt.
2809 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
2810 See Documentation/filesystems/nfs/nfsroot.txt.
2812 nfsrootdebug [NFS] enable nfsroot debugging messages.
2813 See Documentation/filesystems/nfs/nfsroot.txt.
2815 nfs.callback_nr_threads=
2816 [NFSv4] set the total number of threads that the
2817 NFS client will assign to service NFSv4 callback
2820 nfs.callback_tcpport=
2821 [NFS] set the TCP port on which the NFSv4 callback
2822 channel should listen.
2825 [NFS] sets the pathname to the program which is used
2826 to update the NFS client cache entries.
2828 nfs.cache_getent_timeout=
2829 [NFS] sets the timeout after which an attempt to
2830 update a cache entry is deemed to have failed.
2832 nfs.idmap_cache_timeout=
2833 [NFS] set the maximum lifetime for idmapper cache
2837 [NFS] enable 64-bit inode numbers.
2838 If zero, the NFS client will fake up a 32-bit inode
2839 number for the readdir() and stat() syscalls instead
2840 of returning the full 64-bit number.
2841 The default is to return 64-bit inode numbers.
2843 nfs.max_session_cb_slots=
2844 [NFSv4.1] Sets the maximum number of session
2845 slots the client will assign to the callback
2846 channel. This determines the maximum number of
2847 callbacks the client will process in parallel for
2848 a particular server.
2850 nfs.max_session_slots=
2851 [NFSv4.1] Sets the maximum number of session slots
2852 the client will attempt to negotiate with the server.
2853 This limits the number of simultaneous RPC requests
2854 that the client can send to the NFSv4.1 server.
2855 Note that there is little point in setting this
2856 value higher than the max_tcp_slot_table_limit.
2858 nfs.nfs4_disable_idmapping=
2859 [NFSv4] When set to the default of '1', this option
2860 ensures that both the RPC level authentication
2861 scheme and the NFS level operations agree to use
2862 numeric uids/gids if the mount is using the
2863 'sec=sys' security flavour. In effect it is
2864 disabling idmapping, which can make migration from
2865 legacy NFSv2/v3 systems to NFSv4 easier.
2866 Servers that do not support this mode of operation
2867 will be autodetected by the client, and it will fall
2868 back to using the idmapper.
2869 To turn off this behaviour, set the value to '0'.
2871 [NFS4] Specify an additional fixed unique ident-
2872 ification string that NFSv4 clients can insert into
2873 their nfs_client_id4 string. This is typically a
2874 UUID that is generated at system install time.
2876 nfs.send_implementation_id =
2877 [NFSv4.1] Send client implementation identification
2878 information in exchange_id requests.
2879 If zero, no implementation identification information
2881 The default is to send the implementation identification
2884 nfs.recover_lost_locks =
2885 [NFSv4] Attempt to recover locks that were lost due
2886 to a lease timeout on the server. Please note that
2887 doing this risks data corruption, since there are
2888 no guarantees that the file will remain unchanged
2889 after the locks are lost.
2890 If you want to enable the kernel legacy behaviour of
2891 attempting to recover these locks, then set this
2893 The default parameter value of '0' causes the kernel
2894 not to attempt recovery of lost locks.
2896 nfs4.layoutstats_timer =
2897 [NFSv4.2] Change the rate at which the kernel sends
2898 layoutstats to the pNFS metadata server.
2900 Setting this to value to 0 causes the kernel to use
2901 whatever value is the default set by the layout
2902 driver. A non-zero value sets the minimum interval
2903 in seconds between layoutstats transmissions.
2905 nfsd.nfs4_disable_idmapping=
2906 [NFSv4] When set to the default of '1', the NFSv4
2907 server will return only numeric uids and gids to
2908 clients using auth_sys, and will accept numeric uids
2909 and gids from such clients. This is intended to ease
2910 migration from NFSv2/v3.
2912 nmi_debug= [KNL,SH] Specify one or more actions to take
2913 when a NMI is triggered.
2914 Format: [state][,regs][,debounce][,die]
2916 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
2917 Format: [panic,][nopanic,][num]
2919 0 - turn hardlockup detector in nmi_watchdog off
2920 1 - turn hardlockup detector in nmi_watchdog on
2921 When panic is specified, panic when an NMI watchdog
2922 timeout occurs (or 'nopanic' to not panic on an NMI
2923 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
2924 To disable both hard and soft lockup detectors,
2925 please see 'nowatchdog'.
2926 This is useful when you use a panic=... timeout and
2927 need the box quickly up again.
2929 These settings can be accessed at runtime via
2930 the nmi_watchdog and hardlockup_panic sysctls.
2932 netpoll.carrier_timeout=
2933 [NET] Specifies amount of time (in seconds) that
2934 netpoll should wait for a carrier. By default netpoll
2937 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
2938 emulation library even if a 387 maths coprocessor
2941 no5lvl [X86-64] Disable 5-level paging mode. Forces
2942 kernel to use 4-level paging instead.
2945 [HW] Never suspend the console
2946 Disable suspending of consoles during suspend and
2947 hibernate operations. Once disabled, debugging
2948 messages can reach various consoles while the rest
2949 of the system is being put to sleep (ie, while
2950 debugging driver suspend/resume hooks). This may
2951 not work reliably with all consoles, but is known
2952 to work with serial and VGA consoles.
2953 To facilitate more flexible debugging, we also add
2954 console_suspend, a printk module parameter to control
2955 it. Users could use console_suspend (usually
2956 /sys/module/printk/parameters/console_suspend) to
2957 turn on/off it dynamically.
2959 novmcoredd [KNL,KDUMP]
2960 Disable device dump. Device dump allows drivers to
2961 append dump data to vmcore so you can collect driver
2962 specified debug info. Drivers can append the data
2963 without any limit and this data is stored in memory,
2964 so this may cause significant memory stress. Disabling
2965 device dump can help save memory but the driver debug
2966 data will be no longer available. This parameter
2967 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
2970 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
2971 caches in the slab allocator. Saves per-node memory,
2972 but will impact performance.
2976 noaltinstr [S390] Disables alternative instructions patching
2977 (CPU alternatives feature).
2979 noapic [SMP,APIC] Tells the kernel to not make use of any
2980 IOAPICs that may be present in the system.
2982 noautogroup Disable scheduler automatic task group creation.
2984 nobats [PPC] Do not use BATs for mapping kernel lowmem
2985 on "Classic" PPC cores.
2989 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
2991 nodelayacct [KNL] Disable per-task delay accounting
2993 nodsp [SH] Disable hardware DSP at boot time.
2995 noefi Disable EFI runtime services support.
3000 On X86-32 available only on PAE configured kernels.
3001 noexec=on: enable non-executable mappings (default)
3002 noexec=off: disable non-executable mappings
3005 Disable SMAP (Supervisor Mode Access Prevention)
3006 even if it is supported by processor.
3009 Disable SMEP (Supervisor Mode Execution Prevention)
3010 even if it is supported by processor.
3013 This affects only 32-bit executables.
3014 noexec32=on: enable non-executable mappings (default)
3015 read doesn't imply executable mappings
3016 noexec32=off: disable non-executable mappings
3017 read implies executable mappings
3019 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3021 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3022 register save and restore. The kernel will only save
3023 legacy floating-point registers on task switch.
3025 nohugeiomap [KNL,x86,PPC] Disable kernel huge I/O mappings.
3027 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3028 Equivalent to smt=1.
3030 [KNL,x86] Disable symmetric multithreading (SMT).
3031 nosmt=force: Force disable SMT, cannot be undone
3032 via the sysfs control file.
3034 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3035 (bounds check bypass). With this option data leaks are
3036 possible in the system.
3038 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3039 the Spectre variant 2 (indirect branch prediction)
3040 vulnerability. System may allow data leaks with this
3043 nospec_store_bypass_disable
3044 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3046 noxsave [BUGS=X86] Disables x86 extended register state save
3047 and restore using xsave. The kernel will fallback to
3048 enabling legacy floating-point and sse state.
3050 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3051 register states. The kernel will fall back to use
3052 xsave to save the states. By using this parameter,
3053 performance of saving the states is degraded because
3054 xsave doesn't support modified optimization while
3055 xsaveopt supports it on xsaveopt enabled systems.
3057 noxsaves [X86] Disables xsaves and xrstors used in saving and
3058 restoring x86 extended register state in compacted
3059 form of xsave area. The kernel will fall back to use
3060 xsaveopt and xrstor to save and restore the states
3061 in standard form of xsave area. By using this
3062 parameter, xsave area per process might occupy more
3063 memory on xsaves enabled systems.
3065 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
3066 wfi(ARM) instruction doesn't work correctly and not to
3067 use it. This is also useful when using JTAG debugger.
3069 no_file_caps Tells the kernel not to honor file capabilities. The
3070 only way then for a file to be executed with privilege
3071 is to be setuid root or executed by root.
3073 nohalt [IA-64] Tells the kernel not to use the power saving
3074 function PAL_HALT_LIGHT when idle. This increases
3075 power-consumption. On the positive side, it reduces
3076 interrupt wake-up latency, which may improve performance
3077 in certain environments such as networked servers or
3080 nohibernate [HIBERNATION] Disable hibernation and resume.
3082 nohz= [KNL] Boottime enable/disable dynamic ticks
3083 Valid arguments: on, off
3086 nohz_full= [KNL,BOOT,SMP,ISOL]
3087 The argument is a cpu list, as described above.
3088 In kernels built with CONFIG_NO_HZ_FULL=y, set
3089 the specified list of CPUs whose tick will be stopped
3090 whenever possible. The boot CPU will be forced outside
3091 the range to maintain the timekeeping. Any CPUs
3092 in this list will have their RCU callbacks offloaded,
3093 just as if they had also been called out in the
3094 rcu_nocbs= boot parameter.
3096 noiotrap [SH] Disables trapped I/O port accesses.
3098 noirqdebug [X86-32] Disables the code which attempts to detect and
3099 disable unhandled interrupt sources.
3101 no_timer_check [X86,APIC] Disables the code which tests for
3102 broken timer IRQ sources.
3104 noisapnp [ISAPNP] Disables ISA PnP code.
3106 noinitrd [RAM] Tells the kernel not to load any configured
3109 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3111 [Deprecated - use intremap=off]
3115 noinvpcid [X86] Disable the INVPCID cpu feature.
3117 nojitter [IA-64] Disables jitter checking for ITC timers.
3119 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3121 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3125 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3126 clock and use the default one.
3128 no-steal-acc [X86,KVM,ARM64] Disable paravirtualized steal time
3129 accounting. steal time is computed, but won't
3130 influence scheduler behaviour
3132 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3134 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3136 noltlbs [PPC] Do not use large page/tlb entries for kernel
3137 lowmem mapping on PPC40x and PPC8xx
3139 nomca [IA-64] Disable machine check abort handling
3141 nomce [X86-32] Disable Machine Check Exception
3143 nomfgpt [X86-32] Disable Multi-Function General Purpose
3144 Timer usage (for AMD Geode machines).
3146 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3147 shutdown the other cpus. Instead use the REBOOT_VECTOR
3150 nomodule Disable module load
3152 nopat [X86] Disable PAT (page attribute table extension of
3153 pagetables) support.
3155 nopcid [X86-64] Disable the PCID cpu feature.
3157 norandmaps Don't use address space randomization. Equivalent to
3158 echo 0 > /proc/sys/kernel/randomize_va_space
3160 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3161 with UP alternatives
3163 nordrand [X86] Disable kernel use of the RDRAND and
3164 RDSEED instructions even if they are supported
3165 by the processor. RDRAND and RDSEED are still
3166 available to user space applications.
3168 noresume [SWSUSP] Disables resume and restores original swap
3171 no-scroll [VGA] Disables scrollback.
3172 This is required for the Braillex ib80-piezo Braille
3173 reader made by F.H. Papenmeier (Germany).
3177 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
3179 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3180 and disable the IO APIC. legacy for "maxcpus=0".
3182 nosoftlockup [KNL] Disable the soft-lockup detector.
3184 nosync [HW,M68K] Disables sync negotiation for all devices.
3186 nowatchdog [KNL] Disable both lockup detectors, i.e.
3187 soft-lockup and NMI watchdog (hard-lockup).
3191 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3193 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
3194 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
3195 Some features depend on CPU0. Known dependencies are:
3196 1. Resume from suspend/hibernate depends on CPU0.
3197 Suspend/hibernate will fail if CPU0 is offline and you
3198 need to online CPU0 before suspend/hibernate.
3199 2. PIC interrupts also depend on CPU0. CPU0 can't be
3200 removed if a PIC interrupt is detected.
3201 It's said poweroff/reboot may depend on CPU0 on some
3202 machines although I haven't seen such issues so far
3203 after CPU0 is offline on a few tested machines.
3204 If the dependencies are under your control, you can
3205 turn on cpu0_hotplug.
3207 nps_mtm_hs_ctr= [KNL,ARC]
3208 This parameter sets the maximum duration, in
3209 cycles, each HW thread of the CTOP can run
3210 without interruptions, before HW switches it.
3211 The actual maximum duration is 16 times this
3213 Format: integer between 1 and 255
3216 nptcg= [IA-64] Override max number of concurrent global TLB
3217 purges which is reported from either PAL_VM_SUMMARY or
3220 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3221 could support. nr_cpus=n : n >= 1 limits the kernel to
3222 support 'n' processors. It could be larger than the
3223 number of already plugged CPU during bootup, later in
3224 runtime you can physically add extra cpu until it reaches
3225 n. So during boot up some boot time memory for per-cpu
3226 variables need be pre-allocated for later physical cpu
3229 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3231 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
3232 Allowed values are enable and disable
3234 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3235 'node', 'default' can be specified
3236 This can be set from sysctl after boot.
3237 See Documentation/admin-guide/sysctl/vm.rst for details.
3239 of_devlink [OF, KNL] Create device links between consumer and
3240 supplier devices by scanning the devictree to infer the
3241 consumer/supplier relationships. A consumer device
3242 will not be probed until all the supplier devices have
3243 probed successfully.
3245 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3246 See Documentation/debugging-via-ohci1394.txt for more
3249 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3250 Rather than timing out after 20 ms if an EC
3251 command is not properly ACKed, override the length
3252 of the timeout. We have interrupts disabled while
3253 waiting for the ACK, so if this is set too high
3254 interrupts *may* be lost!
3256 omap_mux= [OMAP] Override bootloader pin multiplexing.
3257 Format: <mux_mode0.mode_name=value>...
3258 For example, to override I2C bus2:
3259 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3261 oprofile.timer= [HW]
3262 Use timer interrupt instead of performance counters
3264 oprofile.cpu_type= Force an oprofile cpu type
3265 This might be useful if you have an older oprofile
3266 userland or if you want common events.
3267 Format: { arch_perfmon }
3268 arch_perfmon: [X86] Force use of architectural
3269 perfmon on Intel CPUs instead of the
3270 CPU specific event set.
3271 timer: [X86] Force use of architectural NMI
3272 timer mode (see also oprofile.timer
3273 for generic hr timer mode)
3275 oops=panic Always panic on oopses. Default is to just kill the
3276 process, but there is a small probability of
3277 deadlocking the machine.
3278 This will also cause panics on machine check exceptions.
3279 Useful together with panic=30 to trigger a reboot.
3282 [KNL] Boolean flag to control whether the page allocator
3283 should randomize its free lists. The randomization may
3284 be automatically enabled if the kernel detects it is
3285 running on a platform with a direct-mapped memory-side
3286 cache, and this parameter can be used to
3287 override/disable that behavior. The state of the flag
3288 can be read from sysfs at:
3289 /sys/module/page_alloc/parameters/shuffle.
3291 page_owner= [KNL] Boot-time page_owner enabling option.
3292 Storage of the information about who allocated
3293 each page is disabled in default. With this switch,
3295 on: enable the feature
3297 page_poison= [KNL] Boot-time parameter changing the state of
3298 poisoning on the buddy allocator, available with
3299 CONFIG_PAGE_POISONING=y.
3300 off: turn off poisoning (default)
3301 on: turn on poisoning
3303 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3304 timeout > 0: seconds before rebooting
3305 timeout = 0: wait forever
3306 timeout < 0: reboot immediately
3309 panic_print= Bitmask for printing system info when panic happens.
3310 User can chose combination of the following bits:
3311 bit 0: print all tasks info
3312 bit 1: print system memory info
3313 bit 2: print timer info
3314 bit 3: print locks info if CONFIG_LOCKDEP is on
3315 bit 4: print ftrace buffer
3316 bit 5: print all printk messages in buffer
3318 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3321 crash_kexec_post_notifiers
3322 Run kdump after running panic-notifiers and dumping
3323 kmsg. This only for the users who doubt kdump always
3324 succeeds in any situation.
3325 Note that this also increases risks of kdump failure,
3326 because some panic notifiers can make the crashed
3327 kernel more unstable.
3329 parkbd.port= [HW] Parallel port number the keyboard adapter is
3330 connected to, default is 0.
3332 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3333 0 for XT, 1 for AT (default is AT).
3336 parport= [HW,PPT] Specify parallel ports. 0 disables.
3337 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3338 Use 'auto' to force the driver to use any
3339 IRQ/DMA settings detected (the default is to
3340 ignore detected IRQ/DMA settings because of
3341 possible conflicts). You can specify the base
3342 address, IRQ, and DMA settings; IRQ and DMA
3343 should be numbers, or 'auto' (for using detected
3344 settings on that particular port), or 'nofifo'
3345 (to avoid using a FIFO even if it is detected).
3346 Parallel ports are assigned in the order they
3347 are specified on the command line, starting
3350 parport_init_mode= [HW,PPT]
3351 Configure VIA parallel port to operate in
3352 a specific mode. This is necessary on Pegasos
3353 computer where firmware has no options for setting
3354 up parallel port mode and sets it to spp.
3355 Currently this function knows 686a and 8231 chips.
3356 Format: [spp|ps2|epp|ecp|ecpepp]
3359 Halt all CPUs after the first oops has been printed for
3360 the specified number of seconds. This is to be used if
3361 your oopses keep scrolling off the screen.
3366 See header of drivers/block/paride/pcd.c.
3367 See also Documentation/admin-guide/blockdev/paride.rst.
3369 pci=option[,option...] [PCI] various PCI subsystem options.
3371 Some options herein operate on a specific device
3372 or a set of devices (<pci_dev>). These are
3373 specified in one of the following formats:
3375 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
3376 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
3378 Note: the first format specifies a PCI
3379 bus/device/function address which may change
3380 if new hardware is inserted, if motherboard
3381 firmware changes, or due to changes caused
3382 by other kernel parameters. If the
3383 domain is left unspecified, it is
3384 taken to be zero. Optionally, a path
3385 to a device through multiple device/function
3386 addresses can be specified after the base
3387 address (this is more robust against
3388 renumbering issues). The second format
3389 selects devices using IDs from the
3390 configuration space which may match multiple
3391 devices in the system.
3393 earlydump dump PCI config space before the kernel
3395 off [X86] don't probe for the PCI bus
3396 bios [X86-32] force use of PCI BIOS, don't access
3397 the hardware directly. Use this if your machine
3398 has a non-standard PCI host bridge.
3399 nobios [X86-32] disallow use of PCI BIOS, only direct
3400 hardware access methods are allowed. Use this
3401 if you experience crashes upon bootup and you
3402 suspect they are caused by the BIOS.
3403 conf1 [X86] Force use of PCI Configuration Access
3404 Mechanism 1 (config address in IO port 0xCF8,
3405 data in IO port 0xCFC, both 32-bit).
3406 conf2 [X86] Force use of PCI Configuration Access
3407 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3408 the function, IO port 0xCFA, also 8-bit, sets
3409 bus number. The config space is then accessed
3410 through ports 0xC000-0xCFFF).
3411 See http://wiki.osdev.org/PCI for more info
3412 on the configuration access mechanisms.
3413 noaer [PCIE] If the PCIEAER kernel config parameter is
3414 enabled, this kernel boot option can be used to
3415 disable the use of PCIE advanced error reporting.
3416 nodomains [PCI] Disable support for multiple PCI
3417 root domains (aka PCI segments, in ACPI-speak).
3418 nommconf [X86] Disable use of MMCONFIG for PCI
3420 check_enable_amd_mmconf [X86] check for and enable
3421 properly configured MMIO access to PCI
3422 config space on AMD family 10h CPU
3423 nomsi [MSI] If the PCI_MSI kernel config parameter is
3424 enabled, this kernel boot option can be used to
3425 disable the use of MSI interrupts system-wide.
3426 noioapicquirk [APIC] Disable all boot interrupt quirks.
3427 Safety option to keep boot IRQs enabled. This
3428 should never be necessary.
3429 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3430 primary IO-APIC for bridges that cannot disable
3431 boot IRQs. This fixes a source of spurious IRQs
3432 when the system masks IRQs.
3433 noioapicreroute [APIC] Disable workaround that uses the
3434 boot IRQ equivalent of an IRQ that connects to
3435 a chipset where boot IRQs cannot be disabled.
3436 The opposite of ioapicreroute.
3437 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3438 routing table. These calls are known to be buggy
3439 on several machines and they hang the machine
3440 when used, but on other computers it's the only
3441 way to get the interrupt routing table. Try
3442 this option if the kernel is unable to allocate
3443 IRQs or discover secondary PCI buses on your
3445 rom [X86] Assign address space to expansion ROMs.
3446 Use with caution as certain devices share
3447 address decoders between ROMs and other
3449 norom [X86] Do not assign address space to
3450 expansion ROMs that do not already have
3451 BIOS assigned address ranges.
3452 nobar [X86] Do not assign address space to the
3453 BARs that weren't assigned by the BIOS.
3454 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3455 assigned automatically to PCI devices. You can
3456 make the kernel exclude IRQs of your ISA cards
3458 pirqaddr=0xAAAAA [X86] Specify the physical address
3459 of the PIRQ table (normally generated
3460 by the BIOS) if it is outside the
3461 F0000h-100000h range.
3462 lastbus=N [X86] Scan all buses thru bus #N. Can be
3463 useful if the kernel is unable to find your
3464 secondary buses and you want to tell it
3465 explicitly which ones they are.
3466 assign-busses [X86] Always assign all PCI bus
3467 numbers ourselves, overriding
3468 whatever the firmware may have done.
3469 usepirqmask [X86] Honor the possible IRQ mask stored
3470 in the BIOS $PIR table. This is needed on
3471 some systems with broken BIOSes, notably
3472 some HP Pavilion N5400 and Omnibook XE3
3473 notebooks. This will have no effect if ACPI
3474 IRQ routing is enabled.
3475 noacpi [X86] Do not use ACPI for IRQ routing
3476 or for PCI scanning.
3477 use_crs [X86] Use PCI host bridge window information
3478 from ACPI. On BIOSes from 2008 or later, this
3479 is enabled by default. If you need to use this,
3480 please report a bug.
3481 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3482 If you need to use this, please report a bug.
3483 routeirq Do IRQ routing for all PCI devices.
3484 This is normally done in pci_enable_device(),
3485 so this option is a temporary workaround
3486 for broken drivers that don't call it.
3487 skip_isa_align [X86] do not align io start addr, so can
3488 handle more pci cards
3489 noearly [X86] Don't do any early type 1 scanning.
3490 This might help on some broken boards which
3491 machine check when some devices' config space
3492 is read. But various workarounds are disabled
3493 and some IOMMU drivers will not work.
3494 bfsort Sort PCI devices into breadth-first order.
3495 This sorting is done to get a device
3496 order compatible with older (<= 2.4) kernels.
3497 nobfsort Don't sort PCI devices into breadth-first order.
3498 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3499 tuning and use the BIOS-configured MPS defaults.
3500 pcie_bus_safe Set every device's MPS to the largest value
3501 supported by all devices below the root complex.
3502 pcie_bus_perf Set device MPS to the largest allowable MPS
3503 based on its parent bus. Also set MRRS (Max
3504 Read Request Size) to the largest supported
3505 value (no larger than the MPS that the device
3506 or bus can support) for best performance.
3507 pcie_bus_peer2peer Set every device's MPS to 128B, which
3508 every device is guaranteed to support. This
3509 configuration allows peer-to-peer DMA between
3510 any pair of devices, possibly at the cost of
3511 reduced performance. This also guarantees
3512 that hot-added devices will work.
3513 cbiosize=nn[KMG] The fixed amount of bus space which is
3514 reserved for the CardBus bridge's IO window.
3515 The default value is 256 bytes.
3516 cbmemsize=nn[KMG] The fixed amount of bus space which is
3517 reserved for the CardBus bridge's memory
3518 window. The default value is 64 megabytes.
3521 [<order of align>@]<pci_dev>[; ...]
3522 Specifies alignment and device to reassign
3523 aligned memory resources. How to
3524 specify the device is described above.
3525 If <order of align> is not specified,
3526 PAGE_SIZE is used as alignment.
3527 A PCI-PCI bridge can be specified if resource
3528 windows need to be expanded.
3529 To specify the alignment for several
3530 instances of a device, the PCI vendor,
3531 device, subvendor, and subdevice may be
3532 specified, e.g., 12@pci:8086:9c22:103c:198f
3533 for 4096-byte alignment.
3534 ecrc= Enable/disable PCIe ECRC (transaction layer
3535 end-to-end CRC checking).
3536 bios: Use BIOS/firmware settings. This is the
3540 hpiosize=nn[KMG] The fixed amount of bus space which is
3541 reserved for hotplug bridge's IO window.
3542 Default size is 256 bytes.
3543 hpmmiosize=nn[KMG] The fixed amount of bus space which is
3544 reserved for hotplug bridge's MMIO window.
3545 Default size is 2 megabytes.
3546 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
3547 reserved for hotplug bridge's MMIO_PREF window.
3548 Default size is 2 megabytes.
3549 hpmemsize=nn[KMG] The fixed amount of bus space which is
3550 reserved for hotplug bridge's MMIO and
3552 Default size is 2 megabytes.
3553 hpbussize=nn The minimum amount of additional bus numbers
3554 reserved for buses below a hotplug bridge.
3556 realloc= Enable/disable reallocating PCI bridge resources
3557 if allocations done by BIOS are too small to
3558 accommodate resources required by all child
3560 off: Turn realloc off
3562 realloc same as realloc=on
3563 noari do not use PCIe ARI.
3564 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
3565 do not use PCIe ATS (and IOMMU device IOTLB).
3566 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3567 only look for one device below a PCIe downstream
3569 big_root_window Try to add a big 64bit memory window to the PCIe
3570 root complex on AMD CPUs. Some GFX hardware
3571 can resize a BAR to allow access to all VRAM.
3572 Adding the window is slightly risky (it may
3573 conflict with unreported devices), so this
3575 disable_acs_redir=<pci_dev>[; ...]
3576 Specify one or more PCI devices (in the format
3577 specified above) separated by semicolons.
3578 Each device specified will have the PCI ACS
3579 redirect capabilities forced off which will
3580 allow P2P traffic between devices through
3581 bridges without forcing it upstream. Note:
3582 this removes isolation between devices and
3583 may put more devices in an IOMMU group.
3584 force_floating [S390] Force usage of floating interrupts.
3585 nomio [S390] Do not use MIO instructions.
3587 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3590 force Enable ASPM even on devices that claim not to support it.
3591 WARNING: Forcing ASPM on may cause system lockups.
3593 pcie_ports= [PCIE] PCIe port services handling:
3594 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
3595 even if the platform doesn't give the OS permission to
3596 use them. This may cause conflicts if the platform
3597 also tries to use these services.
3598 dpc-native Use native PCIe service for DPC only. May
3599 cause conflicts if firmware uses AER or DPC.
3600 compat Disable native PCIe services (PME, AER, DPC, PCIe
3603 pcie_port_pm= [PCIE] PCIe port power management handling:
3604 off Disable power management of all PCIe ports
3605 force Forcibly enable power management of all PCIe ports
3607 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3608 nomsi Do not use MSI for native PCIe PME signaling (this makes
3609 all PCIe root ports use INTx for all services).
3611 pcmv= [HW,PCMCIA] BadgePAD 4
3615 Keep all power-domains already enabled by bootloader on,
3616 even if no driver has claimed them. This is useful
3617 for debug and development, but should not be
3618 needed on a platform with proper driver support.
3621 See Documentation/admin-guide/blockdev/paride.rst.
3623 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3626 See arch/parisc/kernel/pdc_chassis.c
3628 percpu_alloc= Select which percpu first chunk allocator to use.
3629 Currently supported values are "embed" and "page".
3630 Archs may support subset or none of the selections.
3631 See comments in mm/percpu.c for details on each
3632 allocator. This parameter is primarily for debugging
3633 and performance comparison.
3636 See Documentation/admin-guide/blockdev/paride.rst.
3639 See Documentation/admin-guide/blockdev/paride.rst.
3641 pirq= [SMP,APIC] Manual mp-table setup
3642 See Documentation/x86/i386/IO-APIC.rst.
3644 plip= [PPT,NET] Parallel port network link
3645 Format: { parport<nr> | timid | 0 }
3646 See also Documentation/admin-guide/parport.rst.
3648 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3649 Override pmtimer IOPort with a hex value.
3653 Enable PNP debug messages (depends on the
3654 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3655 via /sys/module/pnp/parameters/debug. We always show
3656 current resource usage; turning this on also shows
3657 possible settings and some assignment information.
3663 { on | off | curr | res | no-curr | no-res }
3666 [ISAPNP] Exclude IRQs for the autoconfiguration
3669 [ISAPNP] Exclude DMAs for the autoconfiguration
3671 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3672 Ranges are in pairs (I/O port base and size).
3675 [ISAPNP] Exclude memory regions for the
3677 Ranges are in pairs (memory base and size).
3679 ports= [IP_VS_FTP] IPVS ftp helper module
3681 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3683 Format: <port>,<port>....
3685 powersave=off [PPC] This option disables power saving features.
3686 It specifically disables cpuidle and sets the
3687 platform machine description specific power_save
3688 function to NULL. On Idle the CPU just reduces
3691 ppc_strict_facility_enable
3692 [PPC] This option catches any kernel floating point,
3693 Altivec, VSX and SPE outside of regions specifically
3694 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3695 There is some performance impact when enabling this.
3699 Disable Hardware Transactional Memory
3701 print-fatal-signals=
3702 [KNL] debug: print fatal signals
3704 If enabled, warn about various signal handling
3705 related application anomalies: too many signals,
3706 too many POSIX.1 timers, fatal signals causing a
3709 If you hit the warning due to signal overflow,
3710 you might want to try "ulimit -i unlimited".
3714 printk.always_kmsg_dump=
3715 Trigger kmsg_dump for cases other than kernel oops or
3717 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3720 printk.devkmsg={on,off,ratelimit}
3721 Control writing to /dev/kmsg.
3722 on - unlimited logging to /dev/kmsg from userspace
3723 off - logging to /dev/kmsg disabled
3724 ratelimit - ratelimit the logging
3727 printk.time= Show timing data prefixed to each printk message line
3728 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3730 processor.max_cstate= [HW,ACPI]
3731 Limit processor to maximum C-state
3732 max_cstate=9 overrides any DMI blacklist limit.
3734 processor.nocst [HW,ACPI]
3735 Ignore the _CST method to determine C-states,
3736 instead using the legacy FADT method
3738 profile= [KNL] Enable kernel profiling via /proc/profile
3739 Format: [<profiletype>,]<number>
3740 Param: <profiletype>: "schedule", "sleep", or "kvm"
3741 [defaults to kernel profiling]
3742 Param: "schedule" - profile schedule points.
3743 Param: "sleep" - profile D-state sleeping (millisecs).
3744 Requires CONFIG_SCHEDSTATS
3745 Param: "kvm" - profile VM exits.
3746 Param: <number> - step/bucket size as a power of 2 for
3747 statistical time based profiling.
3749 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3751 See Documentation/admin-guide/blockdev/ramdisk.rst.
3753 psi= [KNL] Enable or disable pressure stall information
3757 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3758 probe for; one of (bare|imps|exps|lifebook|any).
3759 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3761 psmouse.resetafter= [HW,MOUSE]
3762 Try to reset the device after so many bad packets
3765 [HW,MOUSE] Set desired mouse resolution, in dpi.
3766 psmouse.smartscroll=
3767 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
3768 0 = disabled, 1 = enabled (default).
3770 pstore.backend= Specify the name of the pstore backend to use
3773 See Documentation/admin-guide/blockdev/paride.rst.
3775 pti= [X86_64] Control Page Table Isolation of user and
3776 kernel address spaces. Disabling this feature
3777 removes hardening, but improves performance of
3778 system calls and interrupts.
3780 on - unconditionally enable
3781 off - unconditionally disable
3782 auto - kernel detects whether your CPU model is
3783 vulnerable to issues that PTI mitigates
3785 Not specifying this option is equivalent to pti=auto.
3788 Equivalent to pti=off
3791 [KNL] Number of legacy pty's. Overwrites compiled-in
3794 quiet [KNL] Disable most log messages
3799 See Documentation/admin-guide/md.rst.
3801 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
3802 See Documentation/admin-guide/blockdev/ramdisk.rst.
3804 random.trust_cpu={on,off}
3805 [KNL] Enable or disable trusting the use of the
3806 CPU's random number generator (if available) to
3807 fully seed the kernel's CRNG. Default is controlled
3808 by CONFIG_RANDOM_TRUST_CPU.
3810 ras=option[,option,...] [KNL] RAS-specific options
3813 Disable the Correctable Errors Collector,
3814 see CONFIG_RAS_CEC help text.
3817 The argument is a cpu list, as described above,
3818 except that the string "all" can be used to
3819 specify every CPU on the system.
3821 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
3822 the specified list of CPUs to be no-callback CPUs.
3823 Invocation of these CPUs' RCU callbacks will be
3824 offloaded to "rcuox/N" kthreads created for that
3825 purpose, where "x" is "p" for RCU-preempt, and
3826 "s" for RCU-sched, and "N" is the CPU number.
3827 This reduces OS jitter on the offloaded CPUs,
3828 which can be useful for HPC and real-time
3829 workloads. It can also improve energy efficiency
3830 for asymmetric multiprocessors.
3833 Rather than requiring that offloaded CPUs
3834 (specified by rcu_nocbs= above) explicitly
3835 awaken the corresponding "rcuoN" kthreads,
3836 make these kthreads poll for callbacks.
3837 This improves the real-time response for the
3838 offloaded CPUs by relieving them of the need to
3839 wake up the corresponding kthread, but degrades
3840 energy efficiency by requiring that the kthreads
3841 periodically wake up to do the polling.
3843 rcutree.blimit= [KNL]
3844 Set maximum number of finished RCU callbacks to
3845 process in one batch.
3847 rcutree.dump_tree= [KNL]
3848 Dump the structure of the rcu_node combining tree
3849 out at early boot. This is used for diagnostic
3850 purposes, to verify correct tree setup.
3852 rcutree.gp_cleanup_delay= [KNL]
3853 Set the number of jiffies to delay each step of
3854 RCU grace-period cleanup.
3856 rcutree.gp_init_delay= [KNL]
3857 Set the number of jiffies to delay each step of
3858 RCU grace-period initialization.
3860 rcutree.gp_preinit_delay= [KNL]
3861 Set the number of jiffies to delay each step of
3862 RCU grace-period pre-initialization, that is,
3863 the propagation of recent CPU-hotplug changes up
3864 the rcu_node combining tree.
3866 rcutree.use_softirq= [KNL]
3867 If set to zero, move all RCU_SOFTIRQ processing to
3868 per-CPU rcuc kthreads. Defaults to a non-zero
3869 value, meaning that RCU_SOFTIRQ is used by default.
3870 Specify rcutree.use_softirq=0 to use rcuc kthreads.
3872 rcutree.rcu_fanout_exact= [KNL]
3873 Disable autobalancing of the rcu_node combining
3874 tree. This is used by rcutorture, and might
3875 possibly be useful for architectures having high
3876 cache-to-cache transfer latencies.
3878 rcutree.rcu_fanout_leaf= [KNL]
3879 Change the number of CPUs assigned to each
3880 leaf rcu_node structure. Useful for very
3881 large systems, which will choose the value 64,
3882 and for NUMA systems with large remote-access
3883 latencies, which will choose a value aligned
3884 with the appropriate hardware boundaries.
3886 rcutree.jiffies_till_first_fqs= [KNL]
3887 Set delay from grace-period initialization to
3888 first attempt to force quiescent states.
3889 Units are jiffies, minimum value is zero,
3890 and maximum value is HZ.
3892 rcutree.jiffies_till_next_fqs= [KNL]
3893 Set delay between subsequent attempts to force
3894 quiescent states. Units are jiffies, minimum
3895 value is one, and maximum value is HZ.
3897 rcutree.jiffies_till_sched_qs= [KNL]
3898 Set required age in jiffies for a
3899 given grace period before RCU starts
3900 soliciting quiescent-state help from
3901 rcu_note_context_switch() and cond_resched().
3902 If not specified, the kernel will calculate
3903 a value based on the most recent settings
3904 of rcutree.jiffies_till_first_fqs
3905 and rcutree.jiffies_till_next_fqs.
3906 This calculated value may be viewed in
3907 rcutree.jiffies_to_sched_qs. Any attempt to set
3908 rcutree.jiffies_to_sched_qs will be cheerfully
3911 rcutree.kthread_prio= [KNL,BOOT]
3912 Set the SCHED_FIFO priority of the RCU per-CPU
3913 kthreads (rcuc/N). This value is also used for
3914 the priority of the RCU boost threads (rcub/N)
3915 and for the RCU grace-period kthreads (rcu_bh,
3916 rcu_preempt, and rcu_sched). If RCU_BOOST is
3917 set, valid values are 1-99 and the default is 1
3918 (the least-favored priority). Otherwise, when
3919 RCU_BOOST is not set, valid values are 0-99 and
3920 the default is zero (non-realtime operation).
3922 rcutree.rcu_nocb_gp_stride= [KNL]
3923 Set the number of NOCB callback kthreads in
3924 each group, which defaults to the square root
3925 of the number of CPUs. Larger numbers reduce
3926 the wakeup overhead on the global grace-period
3927 kthread, but increases that same overhead on
3928 each group's NOCB grace-period kthread.
3930 rcutree.qhimark= [KNL]
3931 Set threshold of queued RCU callbacks beyond which
3932 batch limiting is disabled.
3934 rcutree.qlowmark= [KNL]
3935 Set threshold of queued RCU callbacks below which
3936 batch limiting is re-enabled.
3938 rcutree.rcu_idle_gp_delay= [KNL]
3939 Set wakeup interval for idle CPUs that have
3940 RCU callbacks (RCU_FAST_NO_HZ=y).
3942 rcutree.rcu_idle_lazy_gp_delay= [KNL]
3943 Set wakeup interval for idle CPUs that have
3944 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
3945 Lazy RCU callbacks are those which RCU can
3946 prove do nothing more than free memory.
3948 rcutree.rcu_kick_kthreads= [KNL]
3949 Cause the grace-period kthread to get an extra
3950 wake_up() if it sleeps three times longer than
3951 it should at force-quiescent-state time.
3952 This wake_up() will be accompanied by a
3953 WARN_ONCE() splat and an ftrace_dump().
3955 rcutree.sysrq_rcu= [KNL]
3956 Commandeer a sysrq key to dump out Tree RCU's
3957 rcu_node tree with an eye towards determining
3958 why a new grace period has not yet started.
3960 rcuperf.gp_async= [KNL]
3961 Measure performance of asynchronous
3962 grace-period primitives such as call_rcu().
3964 rcuperf.gp_async_max= [KNL]
3965 Specify the maximum number of outstanding
3966 callbacks per writer thread. When a writer
3967 thread exceeds this limit, it invokes the
3968 corresponding flavor of rcu_barrier() to allow
3969 previously posted callbacks to drain.
3971 rcuperf.gp_exp= [KNL]
3972 Measure performance of expedited synchronous
3973 grace-period primitives.
3975 rcuperf.holdoff= [KNL]
3976 Set test-start holdoff period. The purpose of
3977 this parameter is to delay the start of the
3978 test until boot completes in order to avoid
3981 rcuperf.nreaders= [KNL]
3982 Set number of RCU readers. The value -1 selects
3983 N, where N is the number of CPUs. A value
3984 "n" less than -1 selects N-n+1, where N is again
3985 the number of CPUs. For example, -2 selects N
3986 (the number of CPUs), -3 selects N+1, and so on.
3987 A value of "n" less than or equal to -N selects
3990 rcuperf.nwriters= [KNL]
3991 Set number of RCU writers. The values operate
3992 the same as for rcuperf.nreaders.
3993 N, where N is the number of CPUs
3995 rcuperf.perf_type= [KNL]
3996 Specify the RCU implementation to test.
3998 rcuperf.shutdown= [KNL]
3999 Shut the system down after performance tests
4000 complete. This is useful for hands-off automated
4003 rcuperf.verbose= [KNL]
4004 Enable additional printk() statements.
4006 rcuperf.writer_holdoff= [KNL]
4007 Write-side holdoff between grace periods,
4008 in microseconds. The default of zero says
4011 rcutorture.fqs_duration= [KNL]
4012 Set duration of force_quiescent_state bursts
4015 rcutorture.fqs_holdoff= [KNL]
4016 Set holdoff time within force_quiescent_state bursts
4019 rcutorture.fqs_stutter= [KNL]
4020 Set wait time between force_quiescent_state bursts
4023 rcutorture.fwd_progress= [KNL]
4024 Enable RCU grace-period forward-progress testing
4025 for the types of RCU supporting this notion.
4027 rcutorture.fwd_progress_div= [KNL]
4028 Specify the fraction of a CPU-stall-warning
4029 period to do tight-loop forward-progress testing.
4031 rcutorture.fwd_progress_holdoff= [KNL]
4032 Number of seconds to wait between successive
4033 forward-progress tests.
4035 rcutorture.fwd_progress_need_resched= [KNL]
4036 Enclose cond_resched() calls within checks for
4037 need_resched() during tight-loop forward-progress
4040 rcutorture.gp_cond= [KNL]
4041 Use conditional/asynchronous update-side
4042 primitives, if available.
4044 rcutorture.gp_exp= [KNL]
4045 Use expedited update-side primitives, if available.
4047 rcutorture.gp_normal= [KNL]
4048 Use normal (non-expedited) asynchronous
4049 update-side primitives, if available.
4051 rcutorture.gp_sync= [KNL]
4052 Use normal (non-expedited) synchronous
4053 update-side primitives, if available. If all
4054 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4055 rcutorture.gp_normal=, and rcutorture.gp_sync=
4056 are zero, rcutorture acts as if is interpreted
4057 they are all non-zero.
4059 rcutorture.n_barrier_cbs= [KNL]
4060 Set callbacks/threads for rcu_barrier() testing.
4062 rcutorture.nfakewriters= [KNL]
4063 Set number of concurrent RCU writers. These just
4064 stress RCU, they don't participate in the actual
4065 test, hence the "fake".
4067 rcutorture.nreaders= [KNL]
4068 Set number of RCU readers. The value -1 selects
4069 N-1, where N is the number of CPUs. A value
4070 "n" less than -1 selects N-n-2, where N is again
4071 the number of CPUs. For example, -2 selects N
4072 (the number of CPUs), -3 selects N+1, and so on.
4074 rcutorture.object_debug= [KNL]
4075 Enable debug-object double-call_rcu() testing.
4077 rcutorture.onoff_holdoff= [KNL]
4078 Set time (s) after boot for CPU-hotplug testing.
4080 rcutorture.onoff_interval= [KNL]
4081 Set time (jiffies) between CPU-hotplug operations,
4082 or zero to disable CPU-hotplug testing.
4084 rcutorture.shuffle_interval= [KNL]
4085 Set task-shuffle interval (s). Shuffling tasks
4086 allows some CPUs to go into dyntick-idle mode
4087 during the rcutorture test.
4089 rcutorture.shutdown_secs= [KNL]
4090 Set time (s) after boot system shutdown. This
4091 is useful for hands-off automated testing.
4093 rcutorture.stall_cpu= [KNL]
4094 Duration of CPU stall (s) to test RCU CPU stall
4095 warnings, zero to disable.
4097 rcutorture.stall_cpu_holdoff= [KNL]
4098 Time to wait (s) after boot before inducing stall.
4100 rcutorture.stall_cpu_irqsoff= [KNL]
4101 Disable interrupts while stalling if set.
4103 rcutorture.stat_interval= [KNL]
4104 Time (s) between statistics printk()s.
4106 rcutorture.stutter= [KNL]
4107 Time (s) to stutter testing, for example, specifying
4108 five seconds causes the test to run for five seconds,
4109 wait for five seconds, and so on. This tests RCU's
4110 ability to transition abruptly to and from idle.
4112 rcutorture.test_boost= [KNL]
4113 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
4114 "Maybe" means test if the RCU implementation
4115 under test support RCU priority boosting.
4117 rcutorture.test_boost_duration= [KNL]
4118 Duration (s) of each individual boost test.
4120 rcutorture.test_boost_interval= [KNL]
4121 Interval (s) between each boost test.
4123 rcutorture.test_no_idle_hz= [KNL]
4124 Test RCU's dyntick-idle handling. See also the
4125 rcutorture.shuffle_interval parameter.
4127 rcutorture.torture_type= [KNL]
4128 Specify the RCU implementation to test.
4130 rcutorture.verbose= [KNL]
4131 Enable additional printk() statements.
4133 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
4134 Dump ftrace buffer after reporting RCU CPU
4137 rcupdate.rcu_cpu_stall_suppress= [KNL]
4138 Suppress RCU CPU stall warning messages.
4140 rcupdate.rcu_cpu_stall_timeout= [KNL]
4141 Set timeout for RCU CPU stall warning messages.
4143 rcupdate.rcu_expedited= [KNL]
4144 Use expedited grace-period primitives, for
4145 example, synchronize_rcu_expedited() instead
4146 of synchronize_rcu(). This reduces latency,
4147 but can increase CPU utilization, degrade
4148 real-time latency, and degrade energy efficiency.
4149 No effect on CONFIG_TINY_RCU kernels.
4151 rcupdate.rcu_normal= [KNL]
4152 Use only normal grace-period primitives,
4153 for example, synchronize_rcu() instead of
4154 synchronize_rcu_expedited(). This improves
4155 real-time latency, CPU utilization, and
4156 energy efficiency, but can expose users to
4157 increased grace-period latency. This parameter
4158 overrides rcupdate.rcu_expedited. No effect on
4159 CONFIG_TINY_RCU kernels.
4161 rcupdate.rcu_normal_after_boot= [KNL]
4162 Once boot has completed (that is, after
4163 rcu_end_inkernel_boot() has been invoked), use
4164 only normal grace-period primitives. No effect
4165 on CONFIG_TINY_RCU kernels.
4167 rcupdate.rcu_task_stall_timeout= [KNL]
4168 Set timeout in jiffies for RCU task stall warning
4169 messages. Disable with a value less than or equal
4172 rcupdate.rcu_self_test= [KNL]
4173 Run the RCU early boot self tests
4177 Run specified binary instead of /init from the ramdisk,
4178 used for early userspace startup. See initrd.
4181 force - Override the decision by the kernel to hide the
4182 advertisement of RDRAND support (this affects
4183 certain AMD processors because of buggy BIOS
4184 support, specifically around the suspend/resume
4188 Turn on/off individual RDT features. List is:
4189 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
4191 E.g. to turn on cmt and turn off mba use:
4195 Format (x86 or x86_64):
4196 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
4198 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
4200 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
4201 (prefix with 'panic_' to set mode for panic
4203 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
4204 reboot_force is either force or not specified,
4205 reboot_cpu is s[mp]#### with #### being the processor
4206 to be used for rebooting.
4209 [KNL, SMP] Set scheduler's default relax_domain_level.
4210 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
4212 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
4213 Format: <base1>,<size1>[,<base2>,<size2>,...]
4214 Reserve I/O ports or memory so the kernel won't use
4215 them. If <base> is less than 0x10000, the region
4216 is assumed to be I/O ports; otherwise it is memory.
4218 reservetop= [X86-32]
4220 Reserves a hole at the top of the kernel virtual
4225 Set the amount of memory to reserve for BIOS at
4226 the bottom of the address space.
4228 reset_devices [KNL] Force drivers to reset the underlying device
4229 during initialization.
4232 Specify the partition device for software suspend
4234 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
4236 resume_offset= [SWSUSP]
4237 Specify the offset from the beginning of the partition
4238 given by "resume=" at which the swap header is located,
4239 in <PAGE_SIZE> units (needed only for swap files).
4240 See Documentation/power/swsusp-and-swap-files.rst
4242 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
4243 read the resume files
4245 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
4246 Useful for devices that are detected asynchronously
4247 (e.g. USB and MMC devices).
4249 hibernate= [HIBERNATION]
4250 noresume Don't check if there's a hibernation image
4251 present during boot.
4252 nocompress Don't compress/decompress hibernation images.
4253 no Disable hibernation and resume.
4254 protect_image Turn on image protection during restoration
4255 (that will set all pages holding image data
4256 during restoration read-only).
4258 retain_initrd [RAM] Keep initrd memory after extraction
4260 rfkill.default_state=
4261 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
4262 etc. communication is blocked by default.
4265 rfkill.master_switch_mode=
4266 0 The "airplane mode" button does nothing.
4267 1 The "airplane mode" button toggles between everything
4268 blocked and the previous configuration.
4269 2 The "airplane mode" button toggles between everything
4270 blocked and everything unblocked.
4272 rhash_entries= [KNL,NET]
4273 Set number of hash buckets for route cache
4276 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
4279 ro [KNL] Mount root device read-only on boot
4282 on Mark read-only kernel memory as read-only (default).
4283 off Leave read-only kernel memory writable for debugging.
4286 Enable the uart passthrough on the designated usb port
4287 on Rockchip SoCs. When active, the signals of the
4288 debug-uart get routed to the D+ and D- pins of the usb
4289 port and the regular usb controller gets disabled.
4291 root= [KNL] Root filesystem
4292 See name_to_dev_t comment in init/do_mounts.c.
4294 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
4295 mount the root filesystem
4297 rootflags= [KNL] Set root filesystem mount option string
4299 rootfstype= [KNL] Set root filesystem type
4301 rootwait [KNL] Wait (indefinitely) for root device to show up.
4302 Useful for devices that are detected asynchronously
4303 (e.g. USB and MMC devices).
4305 rproc_mem=nn[KMG][@address]
4306 [KNL,ARM,CMA] Remoteproc physical memory block.
4307 Memory area to be used by remote processor image,
4310 rw [KNL] Mount root device read-write on boot
4312 S [KNL] Run init in single mode
4314 s390_iommu= [HW,S390]
4315 Set s390 IOTLB flushing mode
4317 With strict flushing every unmap operation will result in
4318 an IOTLB flush. Default is lazy flushing before reuse,
4322 See drivers/net/irda/sa1100_ir.c.
4324 sbni= [NET] Granch SBNI12 leased line adapter
4326 sched_debug [KNL] Enables verbose scheduler debug messages.
4328 schedstats= [KNL,X86] Enable or disable scheduled statistics.
4329 Allowed values are enable and disable. This feature
4330 incurs a small amount of overhead in the scheduler
4331 but is useful for debugging and performance tuning.
4333 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
4334 xtime_lock contention on larger systems, and/or RCU lock
4335 contention on all systems with CONFIG_MAXSMP set.
4336 Format: { "0" | "1" }
4337 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
4339 Note: increases power consumption, thus should only be
4340 enabled if running jitter sensitive (HPC/RT) workloads.
4342 security= [SECURITY] Choose a legacy "major" security module to
4343 enable at boot. This has been deprecated by the
4346 selinux= [SELINUX] Disable or enable SELinux at boot time.
4347 Format: { "0" | "1" }
4348 See security/selinux/Kconfig help text.
4351 Default value is set via kernel config option.
4352 If enabled at boot time, /selinux/disable can be used
4353 later to disable prior to initial policy load.
4355 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
4356 Format: { "0" | "1" }
4357 See security/apparmor/Kconfig help text
4360 Default value is set via kernel config option.
4362 serialnumber [BUGS=X86-32]
4365 Maximal number of shapers.
4373 Disable merging of slabs with similar size. May be
4374 necessary if there is some reason to distinguish
4375 allocs to different slabs, especially in hardened
4376 environments where the risk of heap overflows and
4377 layout control by attackers can usually be
4378 frustrated by disabling merging. This will reduce
4379 most of the exposure of a heap attack to a single
4380 cache (risks via metadata attacks are mostly
4381 unchanged). Debug options disable merging on their
4383 For more information see Documentation/vm/slub.rst.
4385 slab_max_order= [MM, SLAB]
4386 Determines the maximum allowed order for slabs.
4387 A high setting may cause OOMs due to memory
4388 fragmentation. Defaults to 1 for systems with
4389 more than 32MB of RAM, 0 otherwise.
4391 slub_debug[=options[,slabs]] [MM, SLUB]
4392 Enabling slub_debug allows one to determine the
4393 culprit if slab objects become corrupted. Enabling
4394 slub_debug can create guard zones around objects and
4395 may poison objects when not in use. Also tracks the
4396 last alloc / free. For more information see
4397 Documentation/vm/slub.rst.
4399 slub_memcg_sysfs= [MM, SLUB]
4400 Determines whether to enable sysfs directories for
4401 memory cgroup sub-caches. 1 to enable, 0 to disable.
4402 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON.
4403 Enabling this can lead to a very high number of debug
4404 directories and files being created under
4407 slub_max_order= [MM, SLUB]
4408 Determines the maximum allowed order for slabs.
4409 A high setting may cause OOMs due to memory
4410 fragmentation. For more information see
4411 Documentation/vm/slub.rst.
4413 slub_min_objects= [MM, SLUB]
4414 The minimum number of objects per slab. SLUB will
4415 increase the slab order up to slub_max_order to
4416 generate a sufficiently large slab able to contain
4417 the number of objects indicated. The higher the number
4418 of objects the smaller the overhead of tracking slabs
4419 and the less frequently locks need to be acquired.
4420 For more information see Documentation/vm/slub.rst.
4422 slub_min_order= [MM, SLUB]
4423 Determines the minimum page order for slabs. Must be
4424 lower than slub_max_order.
4425 For more information see Documentation/vm/slub.rst.
4427 slub_nomerge [MM, SLUB]
4428 Same with slab_nomerge. This is supported for legacy.
4429 See slab_nomerge for more information.
4432 Format: <io1>[,<io2>[,...,<io8>]]
4434 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
4435 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
4436 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
4437 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
4438 smsc-ircc2.ircc_irq= [HW] IRQ line
4439 smsc-ircc2.ircc_dma= [HW] DMA channel
4440 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
4441 0: Toshiba Satellite 1800 (GP data pin select)
4442 1: Fast pin select (default)
4445 smt [KNL,S390] Set the maximum number of threads (logical
4446 CPUs) to use per physical CPU on systems capable of
4447 symmetric multithreading (SMT). Will be capped to the
4448 actual hardware limit.
4450 Default: -1 (no limit)
4453 [KNL] Should the soft-lockup detector generate panics.
4456 A nonzero value instructs the soft-lockup detector
4457 to panic the machine when a soft-lockup occurs. This
4458 is also controlled by CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC
4459 which is the respective build-time switch to that
4462 softlockup_all_cpu_backtrace=
4463 [KNL] Should the soft-lockup detector generate
4464 backtraces on all cpus.
4467 sonypi.*= [HW] Sony Programmable I/O Control Device driver
4468 See Documentation/admin-guide/laptops/sonypi.rst
4470 spectre_v2= [X86] Control mitigation of Spectre variant 2
4471 (indirect branch speculation) vulnerability.
4472 The default operation protects the kernel from
4475 on - unconditionally enable, implies
4477 off - unconditionally disable, implies
4479 auto - kernel detects whether your CPU model is
4482 Selecting 'on' will, and 'auto' may, choose a
4483 mitigation method at run time according to the
4484 CPU, the available microcode, the setting of the
4485 CONFIG_RETPOLINE configuration option, and the
4486 compiler with which the kernel was built.
4488 Selecting 'on' will also enable the mitigation
4489 against user space to user space task attacks.
4491 Selecting 'off' will disable both the kernel and
4492 the user space protections.
4494 Specific mitigations can also be selected manually:
4496 retpoline - replace indirect branches
4497 retpoline,generic - google's original retpoline
4498 retpoline,amd - AMD-specific minimal thunk
4500 Not specifying this option is equivalent to
4504 [X86] Control mitigation of Spectre variant 2
4505 (indirect branch speculation) vulnerability between
4508 on - Unconditionally enable mitigations. Is
4509 enforced by spectre_v2=on
4511 off - Unconditionally disable mitigations. Is
4512 enforced by spectre_v2=off
4514 prctl - Indirect branch speculation is enabled,
4515 but mitigation can be enabled via prctl
4516 per thread. The mitigation control state
4517 is inherited on fork.
4520 - Like "prctl" above, but only STIBP is
4521 controlled per thread. IBPB is issued
4522 always when switching between different user
4526 - Same as "prctl" above, but all seccomp
4527 threads will enable the mitigation unless
4528 they explicitly opt out.
4531 - Like "seccomp" above, but only STIBP is
4532 controlled per thread. IBPB is issued
4533 always when switching between different
4534 user space processes.
4536 auto - Kernel selects the mitigation depending on
4537 the available CPU features and vulnerability.
4540 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl"
4542 Not specifying this option is equivalent to
4543 spectre_v2_user=auto.
4545 spec_store_bypass_disable=
4546 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
4547 (Speculative Store Bypass vulnerability)
4549 Certain CPUs are vulnerable to an exploit against a
4550 a common industry wide performance optimization known
4551 as "Speculative Store Bypass" in which recent stores
4552 to the same memory location may not be observed by
4553 later loads during speculative execution. The idea
4554 is that such stores are unlikely and that they can
4555 be detected prior to instruction retirement at the
4556 end of a particular speculation execution window.
4558 In vulnerable processors, the speculatively forwarded
4559 store can be used in a cache side channel attack, for
4560 example to read memory to which the attacker does not
4561 directly have access (e.g. inside sandboxed code).
4563 This parameter controls whether the Speculative Store
4564 Bypass optimization is used.
4566 On x86 the options are:
4568 on - Unconditionally disable Speculative Store Bypass
4569 off - Unconditionally enable Speculative Store Bypass
4570 auto - Kernel detects whether the CPU model contains an
4571 implementation of Speculative Store Bypass and
4572 picks the most appropriate mitigation. If the
4573 CPU is not vulnerable, "off" is selected. If the
4574 CPU is vulnerable the default mitigation is
4575 architecture and Kconfig dependent. See below.
4576 prctl - Control Speculative Store Bypass per thread
4577 via prctl. Speculative Store Bypass is enabled
4578 for a process by default. The state of the control
4579 is inherited on fork.
4580 seccomp - Same as "prctl" above, but all seccomp threads
4581 will disable SSB unless they explicitly opt out.
4583 Default mitigations:
4584 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
4586 On powerpc the options are:
4588 on,auto - On Power8 and Power9 insert a store-forwarding
4589 barrier on kernel entry and exit. On Power7
4590 perform a software flush on kernel entry and
4594 Not specifying this option is equivalent to
4595 spec_store_bypass_disable=auto.
4597 spia_io_base= [HW,MTD]
4602 srcutree.counter_wrap_check [KNL]
4603 Specifies how frequently to check for
4604 grace-period sequence counter wrap for the
4605 srcu_data structure's ->srcu_gp_seq_needed field.
4606 The greater the number of bits set in this kernel
4607 parameter, the less frequently counter wrap will
4608 be checked for. Note that the bottom two bits
4611 srcutree.exp_holdoff [KNL]
4612 Specifies how many nanoseconds must elapse
4613 since the end of the last SRCU grace period for
4614 a given srcu_struct until the next normal SRCU
4615 grace period will be considered for automatic
4616 expediting. Set to zero to disable automatic
4620 Speculative Store Bypass Disable control
4622 On CPUs that are vulnerable to the Speculative
4623 Store Bypass vulnerability and offer a
4624 firmware based mitigation, this parameter
4625 indicates how the mitigation should be used:
4627 force-on: Unconditionally enable mitigation for
4628 for both kernel and userspace
4629 force-off: Unconditionally disable mitigation for
4630 for both kernel and userspace
4631 kernel: Always enable mitigation in the
4632 kernel, and offer a prctl interface
4633 to allow userspace to register its
4634 interest in being mitigated too.
4636 stack_guard_gap= [MM]
4637 override the default stack gap protection. The value
4638 is in page units and it defines how many pages prior
4639 to (for stacks growing down) resp. after (for stacks
4640 growing up) the main stack are reserved for no other
4641 mapping. Default value is 256 pages.
4644 Enabled the stack tracer on boot up.
4646 stacktrace_filter=[function-list]
4647 [FTRACE] Limit the functions that the stack tracer
4648 will trace at boot up. function-list is a comma separated
4649 list of functions. This list can be changed at run
4650 time by the stack_trace_filter file in the debugfs
4651 tracing directory. Note, this enables stack tracing
4652 and the stacktrace above is not needed.
4656 Set the STI (builtin display/keyboard on the HP-PARISC
4657 machines) console (graphic card) which should be used
4658 as the initial boot-console.
4659 See also comment in drivers/video/console/sticore.c.
4662 See comment in drivers/video/console/sticore.c.
4665 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
4667 sunrpc.min_resvport=
4668 sunrpc.max_resvport=
4670 SunRPC servers often require that client requests
4671 originate from a privileged port (i.e. a port in the
4672 range 0 < portnr < 1024).
4673 An administrator who wishes to reserve some of these
4674 ports for other uses may adjust the range that the
4675 kernel's sunrpc client considers to be privileged
4676 using these two parameters to set the minimum and
4677 maximum port values.
4679 sunrpc.svc_rpc_per_connection_limit=
4681 Limit the number of requests that the server will
4682 process in parallel from a single connection.
4683 The default value is 0 (no limit).
4687 Control how the NFS server code allocates CPUs to
4688 service thread pools. Depending on how many NICs
4689 you have and where their interrupts are bound, this
4690 option will affect which CPUs will do NFS serving.
4691 Note: this parameter cannot be changed while the
4692 NFS server is running.
4694 auto the server chooses an appropriate mode
4695 automatically using heuristics
4696 global a single global pool contains all CPUs
4697 percpu one pool for each CPU
4698 pernode one pool for each NUMA node (equivalent
4699 to global on non-NUMA machines)
4701 sunrpc.tcp_slot_table_entries=
4702 sunrpc.udp_slot_table_entries=
4704 Sets the upper limit on the number of simultaneous
4705 RPC calls that can be sent from the client to a
4706 server. Increasing these values may allow you to
4707 improve throughput, but will also increase the
4708 amount of memory reserved for use by the client.
4710 suspend.pm_test_delay=
4712 Sets the number of seconds to remain in a suspend test
4713 mode before resuming the system (see
4714 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
4715 is set. Default value is 5.
4718 Format: { on | off | y | n | 1 | 0 }
4719 This parameter controls use of the Protected
4720 Execution Facility on pSeries.
4723 [KNL] Enable accounting of swap in memory resource
4724 controller if no parameter or 1 is given or disable
4725 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
4727 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
4728 Format: { <int> | force | noforce }
4729 <int> -- Number of I/O TLB slabs
4730 force -- force using of bounce buffers even if they
4731 wouldn't be automatically used by the kernel
4732 noforce -- Never use bounce buffers (for debugging)
4736 sysfs.deprecated=0|1 [KNL]
4737 Enable/disable old style sysfs layout for old udev
4738 on older distributions. When this option is enabled
4739 very new udev will not work anymore. When this option
4740 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
4741 in older udev will not work anymore.
4742 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
4743 the kernel configuration.
4745 sysrq_always_enabled
4747 Ignore sysrq setting - this boot parameter will
4748 neutralize any effect of /proc/sys/kernel/sysrq.
4749 Useful for debugging.
4751 tcpmhash_entries= [KNL,NET]
4752 Set the number of tcp_metrics_hash slots.
4753 Default value is 8192 or 16384 depending on total
4754 ram pages. This is used to specify the TCP metrics
4755 cache size. See Documentation/networking/ip-sysctl.txt
4756 "tcp_no_metrics_save" section for more details.
4760 test_suspend= [SUSPEND][,N]
4761 Specify "mem" (for Suspend-to-RAM) or "standby" (for
4762 standby suspend) or "freeze" (for suspend type freeze)
4763 as the system sleep state during system startup with
4764 the optional capability to repeat N number of times.
4765 The system is woken from this state using a
4766 wakeup-capable RTC alarm.
4768 thash_entries= [KNL,NET]
4769 Set number of hash buckets for TCP connection
4771 thermal.act= [HW,ACPI]
4772 -1: disable all active trip points in all thermal zones
4773 <degrees C>: override all lowest active trip points
4775 thermal.crt= [HW,ACPI]
4776 -1: disable all critical trip points in all thermal zones
4777 <degrees C>: override all critical trip points
4779 thermal.nocrt= [HW,ACPI]
4780 Set to disable actions on ACPI thermal zone
4781 critical and hot trip points.
4783 thermal.off= [HW,ACPI]
4784 1: disable ACPI thermal control
4786 thermal.psv= [HW,ACPI]
4787 -1: disable all passive trip points
4788 <degrees C>: override all passive trip points to this
4791 thermal.tzp= [HW,ACPI]
4792 Specify global default ACPI thermal zone polling rate
4793 <deci-seconds>: poll all this frequency
4794 0: no polling (default)
4797 Force threading of all interrupt handlers except those
4798 marked explicitly IRQF_NO_THREAD.
4802 Specify if the kernel should make use of the cpu
4803 topology information if the hardware supports this.
4804 The scheduler will make use of this information and
4805 e.g. base its process migration decisions on it.
4808 topology_updates= [KNL, PPC, NUMA]
4810 Specify if the kernel should ignore (off)
4811 topology updates sent by the hypervisor to this
4816 tpm_suspend_pcr=[HW,TPM]
4817 Format: integer pcr id
4818 Specify that at suspend time, the tpm driver
4819 should extend the specified pcr with zeros,
4820 as a workaround for some chips which fail to
4821 flush the last written pcr on TPM_SaveState.
4822 This will guarantee that all the other pcrs
4825 trace_buf_size=nn[KMG]
4826 [FTRACE] will set tracing buffer size on each cpu.
4828 trace_event=[event-list]
4829 [FTRACE] Set and start specified trace events in order
4830 to facilitate early boot debugging. The event-list is a
4831 comma separated list of trace events to enable. See
4832 also Documentation/trace/events.rst
4834 trace_options=[option-list]
4835 [FTRACE] Enable or disable tracer options at boot.
4836 The option-list is a comma delimited list of options
4837 that can be enabled or disabled just as if you were
4838 to echo the option name into
4840 /sys/kernel/debug/tracing/trace_options
4842 For example, to enable stacktrace option (to dump the
4843 stack trace of each event), add to the command line:
4845 trace_options=stacktrace
4847 See also Documentation/trace/ftrace.rst "trace options"
4851 Have the tracepoints sent to printk as well as the
4852 tracing ring buffer. This is useful for early boot up
4853 where the system hangs or reboots and does not give the
4854 option for reading the tracing buffer or performing a
4855 ftrace_dump_on_oops.
4857 To turn off having tracepoints sent to printk,
4858 echo 0 > /proc/sys/kernel/tracepoint_printk
4859 Note, echoing 1 into this file without the
4860 tracepoint_printk kernel cmdline option has no effect.
4864 Having tracepoints sent to printk() and activating high
4865 frequency tracepoints such as irq or sched, can cause
4866 the system to live lock.
4869 [FTRACE] enable this option to disable tracing when a
4870 warning is hit. This turns off "tracing_on". Tracing can
4871 be enabled again by echoing '1' into the "tracing_on"
4872 file located in /sys/kernel/debug/tracing/
4874 This option is useful, as it disables the trace before
4875 the WARNING dump is called, which prevents the trace to
4876 be filled with content caused by the warning output.
4878 This option can also be set at run time via the sysctl
4879 option: kernel/traceoff_on_warning
4881 transparent_hugepage=
4883 Format: [always|madvise|never]
4884 Can be used to control the default behavior of the system
4885 with respect to transparent hugepages.
4886 See Documentation/admin-guide/mm/transhuge.rst
4889 tsc= Disable clocksource stability checks for TSC.
4891 [x86] reliable: mark tsc clocksource as reliable, this
4892 disables clocksource verification at runtime, as well
4893 as the stability checks done at bootup. Used to enable
4894 high-resolution timer mode on older hardware, and in
4895 virtualized environment.
4896 [x86] noirqtime: Do not use TSC to do irq accounting.
4897 Used to run time disable IRQ_TIME_ACCOUNTING on any
4898 platforms where RDTSC is slow and this accounting
4900 [x86] unstable: mark the TSC clocksource as unstable, this
4901 marks the TSC unconditionally unstable at bootup and
4902 avoids any further wobbles once the TSC watchdog notices.
4903 [x86] nowatchdog: disable clocksource watchdog. Used
4904 in situations with strict latency requirements (where
4905 interruptions from clocksource watchdog are not
4908 tsx= [X86] Control Transactional Synchronization
4909 Extensions (TSX) feature in Intel processors that
4910 support TSX control.
4912 This parameter controls the TSX feature. The options are:
4914 on - Enable TSX on the system. Although there are
4915 mitigations for all known security vulnerabilities,
4916 TSX has been known to be an accelerator for
4917 several previous speculation-related CVEs, and
4918 so there may be unknown security risks associated
4919 with leaving it enabled.
4921 off - Disable TSX on the system. (Note that this
4922 option takes effect only on newer CPUs which are
4923 not vulnerable to MDS, i.e., have
4924 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
4925 the new IA32_TSX_CTRL MSR through a microcode
4926 update. This new MSR allows for the reliable
4927 deactivation of the TSX functionality.)
4929 auto - Disable TSX if X86_BUG_TAA is present,
4930 otherwise enable TSX on the system.
4932 Not specifying this option is equivalent to tsx=off.
4934 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
4937 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
4938 Abort (TAA) vulnerability.
4940 Similar to Micro-architectural Data Sampling (MDS)
4941 certain CPUs that support Transactional
4942 Synchronization Extensions (TSX) are vulnerable to an
4943 exploit against CPU internal buffers which can forward
4944 information to a disclosure gadget under certain
4947 In vulnerable processors, the speculatively forwarded
4948 data can be used in a cache side channel attack, to
4949 access data to which the attacker does not have direct
4952 This parameter controls the TAA mitigation. The
4955 full - Enable TAA mitigation on vulnerable CPUs
4958 full,nosmt - Enable TAA mitigation and disable SMT on
4959 vulnerable CPUs. If TSX is disabled, SMT
4960 is not disabled because CPU is not
4961 vulnerable to cross-thread TAA attacks.
4962 off - Unconditionally disable TAA mitigation
4964 On MDS-affected machines, tsx_async_abort=off can be
4965 prevented by an active MDS mitigation as both vulnerabilities
4966 are mitigated with the same mechanism so in order to disable
4967 this mitigation, you need to specify mds=off too.
4969 Not specifying this option is equivalent to
4970 tsx_async_abort=full. On CPUs which are MDS affected
4971 and deploy MDS mitigation, TAA mitigation is not
4972 required and doesn't provide any additional
4976 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
4978 turbografx.map[2|3]= [HW,JOY]
4979 TurboGraFX parallel port interface
4981 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
4982 See also Documentation/input/devices/joystick-parport.rst
4984 udbg-immortal [PPC] When debugging early kernel crashes that
4985 happen after console_init() and before a proper
4986 console driver takes over, this boot options might
4987 help "seeing" what's going on.
4989 uhash_entries= [KNL,NET]
4990 Set number of hash buckets for UDP/UDP-Lite connections
4993 [USB] Ignore overcurrent events (default N).
4994 Some badly-designed motherboards generate lots of
4995 bogus events, for ports that aren't wired to
4996 anything. Set this parameter to avoid log spamming.
4997 Note that genuine overcurrent events won't be
5001 [X86] Cause panic on unknown NMI.
5003 usbcore.authorized_default=
5004 [USB] Default USB device authorization:
5005 (default -1 = authorized except for wireless USB,
5006 0 = not authorized, 1 = authorized, 2 = authorized
5007 if device connected to internal port)
5009 usbcore.autosuspend=
5010 [USB] The autosuspend time delay (in seconds) used
5011 for newly-detected USB devices (default 2). This
5012 is the time required before an idle device will be
5013 autosuspended. Devices for which the delay is set
5014 to a negative value won't be autosuspended at all.
5016 usbcore.usbfs_snoop=
5017 [USB] Set to log all usbfs traffic (default 0 = off).
5019 usbcore.usbfs_snoop_max=
5020 [USB] Maximum number of bytes to snoop in each URB
5023 usbcore.blinkenlights=
5024 [USB] Set to cycle leds on hubs (default 0 = off).
5026 usbcore.old_scheme_first=
5027 [USB] Start with the old device initialization
5028 scheme, applies only to low and full-speed devices
5031 usbcore.usbfs_memory_mb=
5032 [USB] Memory limit (in MB) for buffers allocated by
5033 usbfs (default = 16, 0 = max = 2047).
5035 usbcore.use_both_schemes=
5036 [USB] Try the other device initialization scheme
5037 if the first one fails (default 1 = enabled).
5039 usbcore.initial_descriptor_timeout=
5040 [USB] Specifies timeout for the initial 64-byte
5041 USB_REQ_GET_DESCRIPTOR request in milliseconds
5042 (default 5000 = 5.0 seconds).
5044 usbcore.nousb [USB] Disable the USB subsystem
5047 [USB] A list of quirk entries to augment the built-in
5048 usb core quirk list. List entries are separated by
5049 commas. Each entry has the form
5050 VendorID:ProductID:Flags. The IDs are 4-digit hex
5051 numbers and Flags is a set of letters. Each letter
5052 will change the built-in quirk; setting it if it is
5053 clear and clearing it if it is set. The letters have
5054 the following meanings:
5055 a = USB_QUIRK_STRING_FETCH_255 (string
5056 descriptors must not be fetched using
5058 b = USB_QUIRK_RESET_RESUME (device can't resume
5059 correctly so reset it instead);
5060 c = USB_QUIRK_NO_SET_INTF (device can't handle
5061 Set-Interface requests);
5062 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
5063 handle its Configuration or Interface
5065 e = USB_QUIRK_RESET (device can't be reset
5066 (e.g morph devices), don't use reset);
5067 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
5068 more interface descriptions than the
5069 bNumInterfaces count, and can't handle
5070 talking to these interfaces);
5071 g = USB_QUIRK_DELAY_INIT (device needs a pause
5072 during initialization, after we read
5073 the device descriptor);
5074 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
5075 high speed and super speed interrupt
5076 endpoints, the USB 2.0 and USB 3.0 spec
5077 require the interval in microframes (1
5078 microframe = 125 microseconds) to be
5079 calculated as interval = 2 ^
5081 Devices with this quirk report their
5082 bInterval as the result of this
5083 calculation instead of the exponent
5084 variable used in the calculation);
5085 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
5086 handle device_qualifier descriptor
5088 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
5089 generates spurious wakeup, ignore
5090 remote wakeup capability);
5091 k = USB_QUIRK_NO_LPM (device can't handle Link
5093 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
5094 (Device reports its bInterval as linear
5095 frames instead of the USB 2.0
5097 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
5098 to be disconnected before suspend to
5099 prevent spurious wakeup);
5100 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
5101 pause after every control message);
5102 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
5103 delay after resetting its port);
5104 Example: quirks=0781:5580:bk,0a5c:5834:gij
5107 [USBHID] The interval which mice are to be polled at.
5110 [USBHID] The interval which joysticks are to be polled at.
5113 [USBHID] The interval which keyboards are to be polled at.
5115 usb-storage.delay_use=
5116 [UMS] The delay in seconds before a new device is
5117 scanned for Logical Units (default 1).
5120 [UMS] A list of quirks entries to supplement or
5121 override the built-in unusual_devs list. List
5122 entries are separated by commas. Each entry has
5123 the form VID:PID:Flags where VID and PID are Vendor
5124 and Product ID values (4-digit hex numbers) and
5125 Flags is a set of characters, each corresponding
5126 to a common usb-storage quirk flag as follows:
5127 a = SANE_SENSE (collect more than 18 bytes
5128 of sense data, not on uas);
5129 b = BAD_SENSE (don't collect more than 18
5130 bytes of sense data, not on uas);
5131 c = FIX_CAPACITY (decrease the reported
5132 device capacity by one sector);
5133 d = NO_READ_DISC_INFO (don't use
5134 READ_DISC_INFO command, not on uas);
5135 e = NO_READ_CAPACITY_16 (don't use
5136 READ_CAPACITY_16 command);
5137 f = NO_REPORT_OPCODES (don't use report opcodes
5139 g = MAX_SECTORS_240 (don't transfer more than
5140 240 sectors at a time, uas only);
5141 h = CAPACITY_HEURISTICS (decrease the
5142 reported device capacity by one
5143 sector if the number is odd);
5144 i = IGNORE_DEVICE (don't bind to this
5146 j = NO_REPORT_LUNS (don't use report luns
5148 l = NOT_LOCKABLE (don't try to lock and
5149 unlock ejectable media, not on uas);
5150 m = MAX_SECTORS_64 (don't transfer more
5151 than 64 sectors = 32 KB at a time,
5153 n = INITIAL_READ10 (force a retry of the
5154 initial READ(10) command, not on uas);
5155 o = CAPACITY_OK (accept the capacity
5156 reported by the device, not on uas);
5157 p = WRITE_CACHE (the device cache is ON
5158 by default, not on uas);
5159 r = IGNORE_RESIDUE (the device reports
5160 bogus residue values, not on uas);
5161 s = SINGLE_LUN (the device has only one
5163 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
5164 commands, uas only);
5165 u = IGNORE_UAS (don't bind to the uas driver);
5166 w = NO_WP_DETECT (don't test whether the
5167 medium is write-protected).
5168 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
5169 even if the device claims no cache,
5171 Example: quirks=0419:aaf5:rl,0421:0433:rc
5173 user_debug= [KNL,ARM]
5175 See arch/arm/Kconfig.debug help text.
5176 1 - undefined instruction events
5178 4 - invalid data aborts
5181 Example: user_debug=31
5184 [X86] Flags controlling user PTE allocations.
5186 nohigh = do not allocate PTE pages in
5187 HIGHMEM regardless of setting
5191 On X86_32, this is an alias for vdso32=. Otherwise:
5193 vdso=1: enable VDSO (the default)
5194 vdso=0: disable VDSO mapping
5196 vdso32= [X86] Control the 32-bit vDSO
5197 vdso32=1: enable 32-bit VDSO
5198 vdso32=0 or vdso32=2: disable 32-bit VDSO
5200 See the help text for CONFIG_COMPAT_VDSO for more
5201 details. If CONFIG_COMPAT_VDSO is set, the default is
5202 vdso32=0; otherwise, the default is vdso32=1.
5204 For compatibility with older kernels, vdso32=2 is an
5207 Try vdso32=0 if you encounter an error that says:
5208 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
5211 vector=percpu: enable percpu vector domain
5213 video= [FB] Frame buffer configuration
5214 See Documentation/fb/modedb.rst.
5216 video.brightness_switch_enabled= [0,1]
5217 If set to 1, on receiving an ACPI notify event
5218 generated by hotkey, video driver will adjust brightness
5219 level and then send out the event to user space through
5220 the allocated input device; If set to 0, video driver
5221 will only send out the event without touching backlight
5226 [VMMIO] Memory mapped virtio (platform) device.
5228 <size>@<baseaddr>:<irq>[:<id>]
5230 <size> := size (can use standard suffixes
5232 <baseaddr> := physical base address
5233 <irq> := interrupt number (as passed to
5235 <id> := (optional) platform device id
5237 virtio_mmio.device=1K@0x100b0000:48:7
5239 Can be used multiple times for multiple devices.
5241 vga= [BOOT,X86-32] Select a particular video mode
5242 See Documentation/x86/boot.rst and
5243 Documentation/admin-guide/svga.rst.
5244 Use vga=ask for menu.
5245 This is actually a boot loader parameter; the value is
5246 passed to the kernel using a special protocol.
5248 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
5249 May slow down system boot speed, especially when
5250 enabled on systems with a large amount of memory.
5251 All options are enabled by default, and this
5252 interface is meant to allow for selectively
5253 enabling or disabling specific virtual memory
5256 Available options are:
5257 P Enable page structure init time poisoning
5258 - Disable all of the above options
5260 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
5261 size of <nn>. This can be used to increase the
5262 minimum size (128MB on x86). It can also be used to
5263 decrease the size and leave more room for directly
5266 vmcp_cma=nn[MG] [KNL,S390]
5267 Sets the memory size reserved for contiguous memory
5268 allocations for the vmcp device driver.
5270 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
5273 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
5276 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
5280 Controls the behavior of vsyscalls (i.e. calls to
5281 fixed addresses of 0xffffffffff600x00 from legacy
5282 code). Most statically-linked binaries and older
5283 versions of glibc use these calls. Because these
5284 functions are at fixed addresses, they make nice
5285 targets for exploits that can control RIP.
5287 emulate [default] Vsyscalls turn into traps and are
5288 emulated reasonably safely. The vsyscall
5291 xonly Vsyscalls turn into traps and are
5292 emulated reasonably safely. The vsyscall
5293 page is not readable.
5295 none Vsyscalls don't work at all. This makes
5296 them quite hard to use for exploits but
5297 might break your system.
5299 vt.color= [VT] Default text color.
5300 Format: 0xYX, X = foreground, Y = background.
5301 Default: 0x07 = light gray on black.
5303 vt.cur_default= [VT] Default cursor shape.
5304 Format: 0xCCBBAA, where AA, BB, and CC are the same as
5305 the parameters of the <Esc>[?A;B;Cc escape sequence;
5306 see VGA-softcursor.txt. Default: 2 = underline.
5308 vt.default_blu= [VT]
5309 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
5310 Change the default blue palette of the console.
5311 This is a 16-member array composed of values
5314 vt.default_grn= [VT]
5315 Format: <green0>,<green1>,<green2>,...,<green15>
5316 Change the default green palette of the console.
5317 This is a 16-member array composed of values
5320 vt.default_red= [VT]
5321 Format: <red0>,<red1>,<red2>,...,<red15>
5322 Change the default red palette of the console.
5323 This is a 16-member array composed of values
5329 Set system-wide default UTF-8 mode for all tty's.
5330 Default is 1, i.e. UTF-8 mode is enabled for all
5331 newly opened terminals.
5333 vt.global_cursor_default=
5336 Set system-wide default for whether a cursor
5337 is shown on new VTs. Default is -1,
5338 i.e. cursors will be created by default unless
5339 overridden by individual drivers. 0 will hide
5340 cursors, 1 will display them.
5342 vt.italic= [VT] Default color for italic text; 0-15.
5345 vt.underline= [VT] Default color for underlined text; 0-15.
5348 watchdog timers [HW,WDT] For information on watchdog timers,
5349 see Documentation/watchdog/watchdog-parameters.rst
5350 or other driver-specific files in the
5351 Documentation/watchdog/ directory.
5355 Set the hard lockup detector stall duration
5356 threshold in seconds. The soft lockup detector
5357 threshold is set to twice the value. A value of 0
5358 disables both lockup detectors. Default is 10
5361 workqueue.watchdog_thresh=
5362 If CONFIG_WQ_WATCHDOG is configured, workqueue can
5363 warn stall conditions and dump internal state to
5364 help debugging. 0 disables workqueue stall
5365 detection; otherwise, it's the stall threshold
5366 duration in seconds. The default value is 30 and
5367 it can be updated at runtime by writing to the
5368 corresponding sysfs file.
5370 workqueue.disable_numa
5371 By default, all work items queued to unbound
5372 workqueues are affine to the NUMA nodes they're
5373 issued on, which results in better behavior in
5374 general. If NUMA affinity needs to be disabled for
5375 whatever reason, this option can be used. Note
5376 that this also can be controlled per-workqueue for
5377 workqueues visible under /sys/bus/workqueue/.
5379 workqueue.power_efficient
5380 Per-cpu workqueues are generally preferred because
5381 they show better performance thanks to cache
5382 locality; unfortunately, per-cpu workqueues tend to
5383 be more power hungry than unbound workqueues.
5385 Enabling this makes the per-cpu workqueues which
5386 were observed to contribute significantly to power
5387 consumption unbound, leading to measurably lower
5388 power usage at the cost of small performance
5391 The default value of this parameter is determined by
5392 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
5394 workqueue.debug_force_rr_cpu
5395 Workqueue used to implicitly guarantee that work
5396 items queued without explicit CPU specified are put
5397 on the local CPU. This guarantee is no longer true
5398 and while local CPU is still preferred work items
5399 may be put on foreign CPUs. This debug option
5400 forces round-robin CPU selection to flush out
5401 usages which depend on the now broken guarantee.
5402 When enabled, memory and cache locality will be
5405 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
5406 default x2apic cluster mode on platforms
5409 x86_intel_mid_timer= [X86-32,APBT]
5410 Choose timer option for x86 Intel MID platform.
5411 Two valid options are apbt timer only and lapic timer
5412 plus one apbt timer for broadcast timer.
5413 x86_intel_mid_timer=apbt_only | lapic_and_apbt
5415 xen_512gb_limit [KNL,X86-64,XEN]
5416 Restricts the kernel running paravirtualized under Xen
5417 to use only up to 512 GB of RAM. The reason to do so is
5418 crash analysis tools and Xen tools for doing domain
5419 save/restore/migration must be enabled to handle larger
5422 xen_emul_unplug= [HW,X86,XEN]
5423 Unplug Xen emulated devices
5424 Format: [unplug0,][unplug1]
5425 ide-disks -- unplug primary master IDE devices
5426 aux-ide-disks -- unplug non-primary-master IDE devices
5427 nics -- unplug network devices
5428 all -- unplug all emulated devices (NICs and IDE disks)
5429 unnecessary -- unplugging emulated devices is
5430 unnecessary even if the host did not respond to
5432 never -- do not unplug even if version check succeeds
5434 xen_legacy_crash [X86,XEN]
5435 Crash from Xen panic notifier, without executing late
5436 panic() code such as dumping handler.
5438 xen_nopvspin [X86,XEN]
5439 Disables the ticketlock slowpath using Xen PV
5443 Disables the PV optimizations forcing the HVM guest to
5444 run as generic HVM guest with no PV drivers.
5445 This option is obsoleted by the "nopv" option, which
5446 has equivalent effect for XEN platform.
5448 xen_scrub_pages= [XEN]
5449 Boolean option to control scrubbing pages before giving them back
5450 to Xen, for use by other domains. Can be also changed at runtime
5451 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
5452 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
5454 xen_timer_slop= [X86-64,XEN]
5455 Set the timer slop (in nanoseconds) for the virtual Xen
5456 timers (default is 100000). This adjusts the minimum
5457 delta of virtualized Xen timers, where lower values
5458 improve timer resolution at the expense of processing
5459 more timer interrupts.
5461 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
5462 Disables the PV optimizations forcing the guest to run
5463 as generic guest with no PV drivers. Currently support
5464 XEN HVM, KVM, HYPER_V and VMWARE guest.
5466 xirc2ps_cs= [NET,PCMCIA]
5468 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
5471 By default on POWER9 and above, the kernel will
5472 natively use the XIVE interrupt controller. This option
5473 allows the fallback firmware mode to be used:
5475 off Fallback to firmware control of XIVE interrupt
5476 controller on both pseries and powernv
5477 platforms. Only useful on POWER9 and above.
5479 xhci-hcd.quirks [USB,KNL]
5480 A hex value specifying bitmask with supplemental xhci
5481 host controller quirks. Meaning of each bit can be
5482 consulted in header drivers/usb/host/xhci.h.
5485 Format: { early | on | rw | ro | off }
5486 Controls if xmon debugger is enabled. Default is off.
5487 Passing only "xmon" is equivalent to "xmon=early".
5488 early Call xmon as early as possible on boot; xmon
5489 debugger is called from setup_arch().
5490 on xmon debugger hooks will be installed so xmon
5491 is only called on a kernel crash. Default mode,
5492 i.e. either "ro" or "rw" mode, is controlled
5493 with CONFIG_XMON_DEFAULT_RO_MODE.
5494 rw xmon debugger hooks will be installed so xmon
5495 is called only on a kernel crash, mode is write,
5496 meaning SPR registers, memory and, other data
5497 can be written using xmon commands.
5498 ro same as "rw" option above but SPR registers,
5499 memory, and other data can't be written using
5501 off xmon is disabled.