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.txt, 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/acpi/debug.txt 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.txt 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/alsa-parameters.txt
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 whilst 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" } (0 = disabled, 1 = enabled)
393 0 - kernel audit is disabled and can not be enabled
394 until the next reboot
395 unset - kernel audit is initialized but disabled and
396 will be fully enabled by the userspace auditd.
397 1 - kernel audit is initialized and partially enabled,
398 storing at most audit_backlog_limit messages in
399 RAM until it is fully enabled by the userspace
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.txt
435 boot_delay= Milliseconds to delay each printk during boot.
436 Values larger than 10 seconds (10000) are changed to
440 bootmem_debug [KNL] Enable bootmem allocator debug messages.
443 Disable BERT OS support on buggy BIOSes.
445 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
446 bttv.radio= Most important insmod options are available as
448 bttv.pll= See Documentation/media/v4l-drivers/bttv.rst
451 bulk_remove=off [PPC] This parameter disables the use of the pSeries
452 firmware feature for flushing multiple hpte entries
455 c101= [NET] Moxa C101 synchronous serial card
457 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
458 Sometimes CPU hardware bugs make them report the cache
459 size incorrectly. The kernel will attempt work arounds
460 to fix known problems, but for some CPUs it is not
461 possible to determine what the correct size should be.
462 This option provides an override for these situations.
464 ca_keys= [KEYS] This parameter identifies a specific key(s) on
465 the system trusted keyring to be used for certificate
467 format: { id:<keyid> | builtin }
469 cca= [MIPS] Override the kernel pages' cache coherency
470 algorithm. Accepted values range from 0 to 7
471 inclusive. See arch/mips/include/asm/pgtable-bits.h
472 for platform specific values (SB1, Loongson3 and
475 ccw_timeout_log [S390]
476 See Documentation/s390/CommonIO for details.
478 cgroup_disable= [KNL] Disable a particular controller
479 Format: {name of the controller(s) to disable}
480 The effects of cgroup_disable=foo are:
481 - foo isn't auto-mounted if you mount all cgroups in
483 - foo isn't visible as an individually mountable
485 {Currently only "memory" controller deal with this and
486 cut the overhead, others just disable the usage. So
487 only cgroup_disable=memory is actually worthy}
489 cgroup_no_v1= [KNL] Disable one, multiple, all cgroup controllers in v1
490 Format: { controller[,controller...] | "all" }
491 Like cgroup_disable, but only applies to cgroup v1;
492 the blacklisted controllers remain available in cgroup2.
494 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
496 nosocket -- Disable socket memory accounting.
497 nokmem -- Disable kernel memory accounting.
499 checkreqprot [SELINUX] Set initial checkreqprot flag value.
500 Format: { "0" | "1" }
501 See security/selinux/Kconfig help text.
502 0 -- check protection applied by kernel (includes
503 any implied execute protection).
504 1 -- check protection requested by application.
505 Default value is set via a kernel config option.
506 Value can be changed at runtime via
507 /selinux/checkreqprot.
510 See Documentation/s390/CommonIO for details.
513 Prevents the clock framework from automatically gating
514 clocks that have not been explicitly enabled by a Linux
515 device driver but are enabled in hardware at reset or
516 by the bootloader/firmware. Note that this does not
517 force such clocks to be always-on nor does it reserve
518 those clocks in any way. This parameter is useful for
519 debug and development, but should not be needed on a
520 platform with proper driver support. For more
521 information, see Documentation/clk.txt.
523 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
525 Forces specified clocksource (if available) to be used
526 when calculating gettimeofday(). If specified
527 clocksource is not available, it defaults to PIT.
528 Format: { pit | tsc | cyclone | pmtmr }
530 clocksource= Override the default clocksource
532 Override the default clocksource and use the clocksource
533 with the name specified.
534 Some clocksource names to choose from, depending on
536 [all] jiffies (this is the base, fallback clocksource)
538 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
539 pxa_timer,timer3,32k_counter,timer0_1
540 [X86-32] pit,hpet,tsc;
541 scx200_hrt on Geode; cyclone on IBM x440
549 clocksource.arm_arch_timer.evtstrm=
552 Enable/disable the eventstream feature of the ARM
553 architected timer so that code using WFE-based polling
554 loops can be debugged more effectively on production
557 clearcpuid=BITNUM [X86]
558 Disable CPUID feature X for the kernel. See
559 arch/x86/include/asm/cpufeatures.h for the valid bit
560 numbers. Note the Linux specific bits are not necessarily
561 stable over kernel options, but the vendor specific
563 Also note that user programs calling CPUID directly
564 or using the feature without checking anything
565 will still see it. This just prevents it from
566 being used by the kernel or shown in /proc/cpuinfo.
567 Also note the kernel might malfunction if you disable
570 cma=nn[MG]@[start[MG][-end[MG]]]
572 Sets the size of kernel global memory area for
573 contiguous memory allocations and optionally the
574 placement constraint by the physical address range of
575 memory allocations. A value of 0 disables CMA
576 altogether. For more information, see
577 include/linux/dma-contiguous.h
579 cmo_free_hint= [PPC] Format: { yes | no }
580 Specify whether pages are marked as being inactive
581 when they are freed. This is used in CMO environments
582 to determine OS memory pressure for page stealing by
586 coherent_pool=nn[KMG] [ARM,KNL]
587 Sets the size of memory pool for coherent, atomic dma
588 allocations, by default set to 256K.
590 code_bytes [X86] How many bytes of object code to print
595 com20020= [HW,NET] ARCnet - COM20020 chipset
597 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
599 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
603 ARCnet - COM90xx chipset (memory-mapped buffers)
604 Format: <io>[,<irq>[,<memstart>]]
606 condev= [HW,S390] console device
609 console= [KNL] Output console device and options.
611 tty<n> Use the virtual console device <n>.
615 Use the specified serial port. The options are of
616 the form "bbbbpnf", where "bbbb" is the baud rate,
617 "p" is parity ("n", "o", or "e"), "n" is number of
618 bits, and "f" is flow control ("r" for RTS or
619 omit it). Default is "9600n8".
621 See Documentation/admin-guide/serial-console.rst for more
623 Documentation/networking/netconsole.txt for an
626 uart[8250],io,<addr>[,options]
627 uart[8250],mmio,<addr>[,options]
628 uart[8250],mmio16,<addr>[,options]
629 uart[8250],mmio32,<addr>[,options]
630 uart[8250],0x<addr>[,options]
631 Start an early, polled-mode console on the 8250/16550
632 UART at the specified I/O port or MMIO address,
633 switching to the matching ttyS device later.
634 MMIO inter-register address stride is either 8-bit
635 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
636 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
637 to be equivalent to 'mmio'. 'options' are specified in
638 the same format described for ttyS above; if unspecified,
639 the h/w is not re-initialized.
641 hvc<n> Use the hypervisor console device <n>. This is for
642 both Xen and PowerPC hypervisors.
644 If the device connected to the port is not a TTY but a braille
645 device, prepend "brl," before the device type, for instance
647 For now, only VisioBraille is supported.
650 [KNL] Change console messages format
652 By default we print messages on consoles in
653 "[time stamp] text\n" format (time stamp may not be
654 printed, depending on CONFIG_PRINTK_TIME or
655 `printk_time' param).
657 Switch to syslog format: "<%u>[time stamp] text\n"
658 IOW, each message will have a facility and loglevel
659 prefix. The format is similar to one used by syslog()
660 syscall, or to executing "dmesg -S --raw" or to reading
663 consoleblank= [KNL] The console blank (screen saver) timeout in
664 seconds. A value of 0 disables the blank timer.
668 [KNL] Change the default value for
669 /proc/<pid>/coredump_filter.
670 See also Documentation/filesystems/proc.txt.
672 coresight_cpu_debug.enable
675 Enable/disable the CPU sampling based debugging.
676 0: default value, disable debugging
677 1: enable debugging at boot time
679 cpuidle.off=1 [CPU_IDLE]
680 disable the cpuidle sub-system
682 cpufreq.off=1 [CPU_FREQ]
683 disable the cpufreq sub-system
686 [X86] Delay for N microsec between assert and de-assert
687 of APIC INIT to start processors. This delay occurs
688 on every CPU online, such as boot, and resume from suspend.
691 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
693 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
695 crashkernel=size[KMG][@offset[KMG]]
696 [KNL] Using kexec, Linux can switch to a 'crash kernel'
697 upon panic. This parameter reserves the physical
698 memory region [offset, offset + size] for that kernel
699 image. If '@offset' is omitted, then a suitable offset
700 is selected automatically. Check
701 Documentation/kdump/kdump.txt for further details.
703 crashkernel=range1:size1[,range2:size2,...][@offset]
704 [KNL] Same as above, but depends on the memory
705 in the running system. The syntax of range is
706 start-[end] where start and end are both
707 a memory unit (amount[KMG]). See also
708 Documentation/kdump/kdump.txt for an example.
710 crashkernel=size[KMG],high
711 [KNL, x86_64] range could be above 4G. Allow kernel
712 to allocate physical memory region from top, so could
713 be above 4G if system have more than 4G ram installed.
714 Otherwise memory region will be allocated below 4G, if
716 It will be ignored if crashkernel=X is specified.
717 crashkernel=size[KMG],low
718 [KNL, x86_64] range under 4G. When crashkernel=X,high
719 is passed, kernel could allocate physical memory region
720 above 4G, that cause second kernel crash on system
721 that require some amount of low memory, e.g. swiotlb
722 requires at least 64M+32K low memory, also enough extra
723 low memory is needed to make sure DMA buffers for 32-bit
724 devices won't run out. Kernel would try to allocate at
725 at least 256M below 4G automatically.
726 This one let user to specify own low range under 4G
727 for second kernel instead.
728 0: to disable low allocation.
729 It will be ignored when crashkernel=X,high is not used
730 or memory reserved is below 4G.
733 [KNL] Disable crypto self-tests
738 cs89x0_media= [HW,NET]
739 Format: { rj45 | aui | bnc }
742 See header of drivers/s390/block/dasd_devmap.c.
744 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
745 (one device per port)
746 Format: <port#>,<type>
747 See also Documentation/input/devices/joystick-parport.rst
749 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
751 Documentation/admin-guide/dynamic-debug-howto.rst for
752 details. Deprecated, see dyndbg.
754 debug [KNL] Enable kernel debugging (events log level).
757 [KNL] verbose self-tests
759 Print debugging info while doing the locking API
761 We default to 0 (no extra messages), setting it to
762 1 will print _a lot_ more information - normally
763 only useful to kernel developers.
765 debug_objects [KNL] Enable object debugging
768 [KNL] Disable object debugging
770 debug_guardpage_minorder=
771 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
772 parameter allows control of the order of pages that will
773 be intentionally kept free (and hence protected) by the
774 buddy allocator. Bigger value increase the probability
775 of catching random memory corruption, but reduce the
776 amount of memory for normal system use. The maximum
777 possible value is MAX_ORDER/2. Setting this parameter
778 to 1 or 2 should be enough to identify most random
779 memory corruption problems caused by bugs in kernel or
780 driver code when a CPU writes to (or reads from) a
781 random memory location. Note that there exists a class
782 of memory corruptions problems caused by buggy H/W or
783 F/W or by drivers badly programing DMA (basically when
784 memory is written at bus level and the CPU MMU is
785 bypassed) which are not detectable by
786 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
787 tracking down these problems.
790 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
791 parameter enables the feature at boot time. In
792 default, it is disabled. We can avoid allocating huge
793 chunk of memory for debug pagealloc if we don't enable
794 it at boot time and the system will work mostly same
795 with the kernel built without CONFIG_DEBUG_PAGEALLOC.
796 on: enable the feature
798 debugpat [X86] Enable PAT debugging
800 decnet.addr= [HW,NET]
801 Format: <area>[,<node>]
802 See also Documentation/networking/decnet.txt.
805 [same as hugepagesz=] The size of the default
806 HugeTLB page size. This is the size represented by
807 the legacy /proc/ hugepages APIs, used for SHM, and
808 default size when mounting hugetlbfs filesystems.
809 Defaults to the default architecture's huge page size
813 Set number of hash buckets for dentry cache.
815 disable_1tb_segments [PPC]
816 Disables the use of 1TB hash page table segments. This
817 causes the kernel to fall back to 256MB segments which
818 can be useful when debugging issues that require an SLB
822 See Documentation/networking/ipv6.txt.
825 Disable RADIX MMU mode on POWER9
827 disable_cpu_apicid= [X86,APIC,SMP]
829 The number of initial APIC ID for the
830 corresponding CPU to be disabled at boot,
831 mostly used for the kdump 2nd kernel to
832 disable BSP to wake up multiple CPUs without
833 causing system reset or hang due to sending
836 disable_ddw [PPC/PSERIES]
837 Disable Dynamic DMA Window support. Use this if
838 to workaround buggy firmware.
841 See Documentation/networking/ipv6.txt.
843 disable_mtrr_cleanup [X86]
844 The kernel tries to adjust MTRR layout from continuous
845 to discrete, to make X server driver able to add WB
846 entry later. This parameter disables that.
848 disable_mtrr_trim [X86, Intel and AMD only]
849 By default the kernel will trim any uncacheable
850 memory out of your available memory pool based on
851 MTRR settings. This parameter disables that behavior,
852 possibly causing your machine to run very slowly.
854 disable_timer_pin_1 [X86]
855 Disable PIN 1 of APIC timer
856 Can be useful to work around chipset bugs.
858 dis_ucode_ldr [X86] Disable the microcode loader.
860 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
861 this option disables the debugging code at boot.
863 dma_debug_entries=<number>
864 This option allows to tune the number of preallocated
865 entries for DMA-API debugging code. One entry is
866 required per DMA-API allocation. Use this if the
867 DMA-API debugging code disables itself because the
868 architectural default is too low.
870 dma_debug_driver=<driver_name>
871 With this option the DMA-API debugging driver
872 filter feature can be enabled at boot time. Just
873 pass the driver to filter for as the parameter.
874 The filter can be disabled or changed to another
875 driver later using sysfs.
877 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
878 Broken monitors, graphic adapters, KVMs and EDIDless
879 panels may send no or incorrect EDID data sets.
880 This parameter allows to specify an EDID data sets
881 in the /lib/firmware directory that are used instead.
882 Generic built-in EDID data sets are used, if one of
883 edid/1024x768.bin, edid/1280x1024.bin,
884 edid/1680x1050.bin, or edid/1920x1080.bin is given
885 and no file with the same name exists. Details and
886 instructions how to build your own EDID data are
887 available in Documentation/EDID/HOWTO.txt. An EDID
888 data set will only be used for a particular connector,
889 if its name and a colon are prepended to the EDID
890 name. Each connector may use a unique EDID data
891 set by separating the files with a comma. An EDID
892 data set with no connector name will be used for
893 any connectors not explicitly specified.
898 Format: {"off" | "known"}
899 Control how the dt_cpu_ftrs device-tree binding is
900 used for CPU feature discovery and setup (if it
902 off: Do not use it, fall back to legacy cpu table.
903 known: Do not pass through unknown features to guests
904 or userspace, only those that the kernel is aware of.
906 dump_apple_properties [X86]
907 Dump name and content of EFI device properties on
908 x86 Macs. Useful for driver authors to determine
909 what data is available or for reverse-engineering.
911 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
912 module.dyndbg[="val"]
913 Enable debug messages at boot time. See
914 Documentation/admin-guide/dynamic-debug-howto.rst
917 nompx [X86] Disables Intel Memory Protection Extensions.
918 See Documentation/x86/intel_mpx.txt for more
919 information about the feature.
921 nopku [X86] Disable Memory Protection Keys CPU feature found
924 module.async_probe [KNL]
925 Enable asynchronous probe on this module.
927 early_ioremap_debug [KNL]
928 Enable debug messages in early_ioremap support. This
929 is useful for tracking down temporary early mappings
930 which are not unmapped.
932 earlycon= [KNL] Output early console device and options.
934 When used with no options, the early console is
935 determined by the stdout-path property in device
938 cdns,<addr>[,options]
939 Start an early, polled-mode console on a Cadence
940 (xuartps) serial port at the specified address. Only
941 supported option is baud rate. If baud rate is not
942 specified, the serial port must already be setup and
945 uart[8250],io,<addr>[,options]
946 uart[8250],mmio,<addr>[,options]
947 uart[8250],mmio32,<addr>[,options]
948 uart[8250],mmio32be,<addr>[,options]
949 uart[8250],0x<addr>[,options]
950 Start an early, polled-mode console on the 8250/16550
951 UART at the specified I/O port or MMIO address.
952 MMIO inter-register address stride is either 8-bit
953 (mmio) or 32-bit (mmio32 or mmio32be).
954 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
955 to be equivalent to 'mmio'. 'options' are specified
956 in the same format described for "console=ttyS<n>"; if
957 unspecified, the h/w is not initialized.
961 Start an early, polled-mode console on a pl011 serial
962 port at the specified address. The pl011 serial port
963 must already be setup and configured. Options are not
964 yet supported. If 'mmio32' is specified, then only
965 the driver will use only 32-bit accessors to read/write
966 the device registers.
969 Start an early, polled-mode console on a meson serial
970 port at the specified address. The serial port must
971 already be setup and configured. Options are not yet
975 Start an early, polled-mode console on an msm serial
976 port at the specified address. The serial port
977 must already be setup and configured. Options are not
981 Start an early, polled-mode console on an msm serial
982 dm port at the specified address. The serial port
983 must already be setup and configured. Options are not
987 Start an early, polled-mode console on a serial port
988 of an Actions Semi SoC, such as S500 or S900, at the
989 specified address. The serial port must already be
990 setup and configured. Options are not yet supported.
992 smh Use ARM semihosting calls for early console.
1000 Use early console provided by serial driver available
1001 on Samsung SoCs, requires selecting proper type and
1002 a correct base address of the selected UART port. The
1003 serial port must already be setup and configured.
1004 Options are not yet supported.
1007 Start an early, polled-mode console on a lantiq serial
1008 (lqasc) port at the specified address. The serial port
1009 must already be setup and configured. Options are not
1014 Use early console provided by Freescale LP UART driver
1015 found on Freescale Vybrid and QorIQ LS1021A processors.
1016 A valid base address must be provided, and the serial
1017 port must already be setup and configured.
1020 Start an early, polled-mode console on the
1021 Armada 3700 serial port at the specified
1022 address. The serial port must already be setup
1023 and configured. Options are not yet supported.
1025 earlyprintk= [X86,SH,BLACKFIN,ARM,M68k,S390]
1030 earlyprintk=serial[,ttySn[,baudrate]]
1031 earlyprintk=serial[,0x...[,baudrate]]
1032 earlyprintk=ttySn[,baudrate]
1033 earlyprintk=dbgp[debugController#]
1034 earlyprintk=pciserial,bus:device.function[,baudrate]
1035 earlyprintk=xdbc[xhciController#]
1037 earlyprintk is useful when the kernel crashes before
1038 the normal console is initialized. It is not enabled by
1039 default because it has some cosmetic problems.
1041 Append ",keep" to not disable it when the real console
1044 Only one of vga, efi, serial, or usb debug port can
1047 Currently only ttyS0 and ttyS1 may be specified by
1048 name. Other I/O ports may be explicitly specified
1049 on some architectures (x86 and arm at least) by
1050 replacing ttySn with an I/O port address, like this:
1051 earlyprintk=serial,0x1008,115200
1052 You can find the port for a given device in
1053 /proc/tty/driver/serial:
1054 2: uart:ST16650V2 port:00001008 irq:18 ...
1056 Interaction with the standard serial driver is not
1059 The VGA and EFI output is eventually overwritten by
1062 The xen output can only be used by Xen PV guests.
1064 The sclp output can only be used on s390.
1066 edac_report= [HW,EDAC] Control how to report EDAC event
1067 Format: {"on" | "off" | "force"}
1068 on: enable EDAC to report H/W event. May be overridden
1069 by other higher priority error reporting module.
1070 off: disable H/W event reporting through EDAC.
1071 force: enforce the use of EDAC to report H/W event.
1074 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1077 This is designed to be used in conjunction with
1078 the boot argument: earlyprintk=vga
1081 Format: {"off" | "on" | "skip[mbr]"}
1084 Format: { "old_map", "nochunk", "noruntime", "debug" }
1085 old_map [X86-64]: switch to the old ioremap-based EFI
1086 runtime services mapping. 32-bit still uses this one by
1088 nochunk: disable reading files in "chunks" in the EFI
1089 boot stub, as chunking can cause problems with some
1090 firmware implementations.
1091 noruntime : disable EFI runtime services support
1092 debug: enable misc debug output
1094 efi_no_storage_paranoia [EFI; X86]
1095 Using this parameter you can use more than 50% of
1096 your efi variable storage. Use this parameter only if
1097 you are really sure that your UEFI does sane gc and
1098 fulfills the spec otherwise your board may brick.
1100 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1101 Add arbitrary attribute to specific memory range by
1102 updating original EFI memory map.
1103 Region of memory which aa attribute is added to is
1105 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1106 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1107 attribute is added to range 0x100000000-0x180000000 and
1108 0x10a0000000-0x1120000000.
1110 Using this parameter you can do debugging of EFI memmap
1111 related feature. For example, you can do debugging of
1112 Address Range Mirroring feature even if your box
1115 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1116 that is to be dynamically loaded by Linux. If there are
1117 multiple variables with the same name but with different
1118 vendor GUIDs, all of them will be loaded. See
1119 Documentation/acpi/ssdt-overlays.txt for details.
1122 eisa_irq_edge= [PARISC,HW]
1123 See header of drivers/parisc/eisa.c.
1126 See comment before function elanfreq_setup() in
1127 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1130 Format: {"cfq" | "deadline" | "noop"}
1131 See Documentation/block/cfq-iosched.txt and
1132 Documentation/block/deadline-iosched.txt for details.
1134 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1135 Specifies physical address of start of kernel core
1136 image elf header and optionally the size. Generally
1137 kexec loader will pass this option to capture kernel.
1138 See Documentation/kdump/kdump.txt for details.
1140 enable_mtrr_cleanup [X86]
1141 The kernel tries to adjust MTRR layout from continuous
1142 to discrete, to make X server driver able to add WB
1143 entry later. This parameter enables that.
1145 enable_timer_pin_1 [X86]
1146 Enable PIN 1 of APIC timer
1147 Can be useful to work around chipset bugs
1148 (in particular on some ATI chipsets).
1149 The kernel tries to set a reasonable default.
1151 enforcing [SELINUX] Set initial enforcing status.
1153 See security/selinux/Kconfig help text.
1154 0 -- permissive (log only, no denials).
1155 1 -- enforcing (deny and log).
1157 Value can be changed at runtime via /selinux/enforce.
1160 Disable Error Record Serialization Table (ERST)
1163 ether= [HW,NET] Ethernet cards parameters
1164 This option is obsoleted by the "netdev=" option, which
1165 has equivalent usage. See its documentation for details.
1169 Permit 'security.evm' to be updated regardless of
1170 current integrity status.
1174 fail_make_request=[KNL]
1175 General fault injection mechanism.
1176 Format: <interval>,<probability>,<space>,<times>
1177 See also Documentation/fault-injection/.
1180 See Documentation/blockdev/floppy.txt.
1182 force_pal_cache_flush
1183 [IA-64] Avoid check_sal_cache_flush which may hang on
1184 buggy SAL_CACHE_FLUSH implementations. Using this
1185 parameter will force ia64_sal_cache_flush to call
1186 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1189 Forcefully enable Physical Address Extension (PAE).
1190 Many Pentium M systems disable PAE but may have a
1191 functionally usable PAE implementation.
1192 Warning: use of this parameter will taint the kernel
1193 and may cause unknown problems.
1196 [FTRACE] will set and start the specified tracer
1197 as early as possible in order to facilitate early
1200 ftrace_dump_on_oops[=orig_cpu]
1201 [FTRACE] will dump the trace buffers on oops.
1202 If no parameter is passed, ftrace will dump
1203 buffers of all CPUs, but if you pass orig_cpu, it will
1204 dump only the buffer of the CPU that triggered the
1207 ftrace_filter=[function-list]
1208 [FTRACE] Limit the functions traced by the function
1209 tracer at boot up. function-list is a comma separated
1210 list of functions. This list can be changed at run
1211 time by the set_ftrace_filter file in the debugfs
1214 ftrace_notrace=[function-list]
1215 [FTRACE] Do not trace the functions specified in
1216 function-list. This list can be changed at run time
1217 by the set_ftrace_notrace file in the debugfs
1220 ftrace_graph_filter=[function-list]
1221 [FTRACE] Limit the top level callers functions traced
1222 by the function graph tracer at boot up.
1223 function-list is a comma separated list of functions
1224 that can be changed at run time by the
1225 set_graph_function file in the debugfs tracing directory.
1227 ftrace_graph_notrace=[function-list]
1228 [FTRACE] Do not trace from the functions specified in
1229 function-list. This list is a comma separated list of
1230 functions that can be changed at run time by the
1231 set_graph_notrace file in the debugfs tracing directory.
1233 ftrace_graph_max_depth=<uint>
1234 [FTRACE] Used with the function graph tracer. This is
1235 the max depth it will trace into a function. This value
1236 can be changed at run time by the max_graph_depth file
1237 in the tracefs tracing directory. default: 0 (no limit)
1240 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1241 support via parallel port (up to 5 devices per port)
1242 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1243 See also Documentation/input/devices/joystick-parport.rst
1247 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1251 gcov_persist= [GCOV] When non-zero (default), profiling data for
1252 kernel modules is saved and remains accessible via
1253 debugfs, even when the module is unloaded/reloaded.
1254 When zero, profiling data is discarded and associated
1255 debugfs files are removed at module unload time.
1257 goldfish [X86] Enable the goldfish android emulator platform.
1258 Don't use this when you are not running on the
1261 gpt [EFI] Forces disk with valid GPT signature but
1262 invalid Protective MBR to be treated as GPT. If the
1263 primary GPT is corrupted, it enables the backup/alternate
1264 GPT to be used instead.
1266 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1267 the "Enable 0" bit of the configuration register.
1270 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1271 the "Enable 0" bit of the configuration register.
1274 grcan.select= [HW] Select which physical interface to use.
1277 grcan.txsize= [HW] Sets the size of the tx buffer.
1278 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1280 grcan.rxsize= [HW] Sets the size of the rx buffer.
1281 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1284 gpio-mockup.gpio_mockup_ranges
1285 [HW] Sets the ranges of gpiochip of for this device.
1286 Format: <start1>,<end1>,<start2>,<end2>...
1288 hardlockup_all_cpu_backtrace=
1289 [KNL] Should the hard-lockup detector generate
1290 backtraces on all cpus.
1293 hashdist= [KNL,NUMA] Large hashes allocated during boot
1294 are distributed across NUMA nodes. Defaults on
1295 for 64-bit NUMA, off otherwise.
1296 Format: 0 | 1 (for off | on)
1298 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1300 hd= [EIDE] (E)IDE hard drive subsystem geometry
1301 Format: <cyl>,<head>,<sect>
1304 Disable Hardware Error Source Table (HEST) support;
1305 corresponding firmware-first mode error processing
1306 logic will be disabled.
1308 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1309 size of <nn>. This works even on boxes that have no
1310 highmem otherwise. This also works to reduce highmem
1311 size on bigger boxes.
1313 highres= [KNL] Enable/disable high resolution timer mode.
1314 Valid parameters: "on", "off"
1318 See Documentation/isdn/README.HiSax.
1322 hpet= [X86-32,HPET] option to control HPET usage
1323 Format: { enable (default) | disable | force |
1325 disable: disable HPET and use PIT instead
1326 force: allow force enabled of undocumented chips (ICH4,
1328 verbose: show contents of HPET registers during setup
1330 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1331 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1333 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1334 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
1335 On x86-64 and powerpc, this option can be specified
1336 multiple times interleaved with hugepages= to reserve
1337 huge pages of different sizes. Valid pages sizes on
1338 x86-64 are 2M (when the CPU supports "pse") and 1G
1339 (when the CPU supports the "pdpe1gb" cpuinfo flag).
1341 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1342 terminal devices. Valid values: 0..8
1343 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1344 If specified, z/VM IUCV HVC accepts connections
1345 from listed z/VM user IDs only.
1347 hwthread_map= [METAG] Comma-separated list of Linux cpu id to
1348 hardware thread id mappings.
1349 Format: <cpu>:<hwthread>
1352 Do not unregister boot console at start. This is only
1353 useful for debugging when something happens in the window
1354 between unregistering the boot console and initializing
1357 i2c_bus= [HW] Override the default board specific I2C bus speed
1358 or register an additional I2C bus that is not
1359 registered from board initialization code.
1363 i8042.debug [HW] Toggle i8042 debug mode
1364 i8042.unmask_kbd_data
1365 [HW] Enable printing of interrupt data from the KBD port
1366 (disabled by default, and as a pre-condition
1367 requires that i8042.debug=1 be enabled)
1368 i8042.direct [HW] Put keyboard port into non-translated mode
1369 i8042.dumbkbd [HW] Pretend that controller can only read data from
1370 keyboard and cannot control its state
1371 (Don't attempt to blink the leds)
1372 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1373 i8042.nokbd [HW] Don't check/create keyboard port
1374 i8042.noloop [HW] Disable the AUX Loopback command while probing
1376 i8042.nomux [HW] Don't check presence of an active multiplexing
1378 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1380 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1381 i8042.reset [HW] Reset the controller during init, cleanup and
1382 suspend-to-ram transitions, only during s2r
1383 transitions, or never reset
1384 Format: { 1 | Y | y | 0 | N | n }
1385 1, Y, y: always reset controller
1386 0, N, n: don't ever reset controller
1387 Default: only on s2r transitions on x86; most other
1388 architectures force reset to be always executed
1389 i8042.unlock [HW] Unlock (ignore) the keylock
1390 i8042.kbdreset [HW] Reset device connected to KBD port
1394 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1395 indicates that the driver is running on unsupported
1397 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1398 does not match list of supported models.
1400 [HW] Report power status in /proc/i8k
1401 (disabled by default)
1402 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1405 i915.invert_brightness=
1406 [DRM] Invert the sense of the variable that is used to
1407 set the brightness of the panel backlight. Normally a
1408 brightness value of 0 indicates backlight switched off,
1409 and the maximum of the brightness value sets the backlight
1410 to maximum brightness. If this parameter is set to 0
1411 (default) and the machine requires it, or this parameter
1412 is set to 1, a brightness value of 0 sets the backlight
1413 to maximum brightness, and the maximum of the brightness
1414 value switches the backlight off.
1415 -1 -- never invert brightness
1416 0 -- machine default
1417 1 -- force brightness inversion
1420 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1422 ide-core.nodma= [HW] (E)IDE subsystem
1423 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1424 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1425 .cdrom .chs .ignore_cable are additional options
1426 See Documentation/ide/ide.txt.
1428 ide-generic.probe-mask= [HW] (E)IDE subsystem
1430 Probe mask for legacy ISA IDE ports. Depending on
1431 platform up to 6 ports are supported, enabled by
1432 setting corresponding bits in the mask to 1. The
1433 default value is 0x0, which has a special meaning.
1434 On systems that have PCI, it triggers scanning the
1435 PCI bus for the first and the second port, which
1436 are then probed. On systems without PCI the value
1437 of 0x0 enables probing the two first ports as if it
1440 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1441 Claim all unknown PCI IDE storage controllers.
1444 Format: idle=poll, idle=halt, idle=nomwait
1445 Poll forces a polling idle loop that can slightly
1446 improve the performance of waking up a idle CPU, but
1447 will use a lot of power and make the system run hot.
1449 idle=halt: Halt is forced to be used for CPU idle.
1450 In such case C2/C3 won't be used again.
1451 idle=nomwait: Disable mwait for CPU C-states
1453 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1454 Format: { strict | legacy | 2008 | relaxed }
1457 Choose which programs will be accepted for execution
1458 based on the IEEE 754 NaN encoding(s) supported by
1459 the FPU and the NaN encoding requested with the value
1460 of an ELF file header flag individually set by each
1461 binary. Hardware implementations are permitted to
1462 support either or both of the legacy and the 2008 NaN
1465 Available settings are as follows:
1466 strict accept binaries that request a NaN encoding
1467 supported by the FPU
1468 legacy only accept legacy-NaN binaries, if supported
1470 2008 only accept 2008-NaN binaries, if supported
1472 relaxed accept any binaries regardless of whether
1473 supported by the FPU
1475 The FPU emulator is always able to support both NaN
1476 encodings, so if no FPU hardware is present or it has
1477 been disabled with 'nofpu', then the settings of
1478 'legacy' and '2008' strap the emulator accordingly,
1479 'relaxed' straps the emulator for both legacy-NaN and
1480 2008-NaN, whereas 'strict' enables legacy-NaN only on
1481 legacy processors and both NaN encodings on MIPS32 or
1484 The setting for ABS.fmt/NEG.fmt instruction execution
1485 mode generally follows that for the NaN encoding,
1486 except where unsupported by hardware.
1488 ignore_loglevel [KNL]
1489 Ignore loglevel setting - this will print /all/
1490 kernel messages to the console. Useful for debugging.
1491 We also add it as printk module parameter, so users
1492 could change it dynamically, usually by
1493 /sys/module/printk/parameters/ignore_loglevel.
1496 Ignore RLIMIT_DATA setting for data mappings,
1497 print warning at first misuse. Can be changed via
1498 /sys/module/kernel/parameters/ignore_rlimit_data.
1500 ihash_entries= [KNL]
1501 Set number of hash buckets for inode cache.
1503 ima_appraise= [IMA] appraise integrity measurements
1504 Format: { "off" | "enforce" | "fix" | "log" }
1507 ima_appraise_tcb [IMA]
1508 The builtin appraise policy appraises all files
1511 ima_canonical_fmt [IMA]
1512 Use the canonical format for the binary runtime
1513 measurements, instead of host native format.
1516 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1520 The list of supported hash algorithms is defined
1521 in crypto/hash_info.h.
1524 The builtin policies to load during IMA setup.
1525 Format: "tcb | appraise_tcb | secure_boot"
1527 The "tcb" policy measures all programs exec'd, files
1528 mmap'd for exec, and all files opened with the read
1529 mode bit set by either the effective uid (euid=0) or
1532 The "appraise_tcb" policy appraises the integrity of
1533 all files owned by root. (This is the equivalent
1534 of ima_appraise_tcb.)
1536 The "secure_boot" policy appraises the integrity
1537 of files (eg. kexec kernel image, kernel modules,
1538 firmware, policy, etc) based on file signatures.
1540 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1541 Load a policy which meets the needs of the Trusted
1542 Computing Base. This means IMA will measure all
1543 programs exec'd, files mmap'd for exec, and all files
1544 opened for read by uid=0.
1547 Select one of defined IMA measurements template formats.
1548 Formats: { "ima" | "ima-ng" | "ima-sig" }
1552 [IMA] Define a custom template format.
1553 Format: { "field1|...|fieldN" }
1555 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1556 Format: <min_file_size>
1557 Set the minimal file size for using asynchronous hash.
1558 If left unspecified, ahash usage is disabled.
1560 ahash performance varies for different data sizes on
1561 different crypto accelerators. This option can be used
1562 to achieve the best performance for a particular HW.
1564 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1566 Set hashing buffer size. Default: 4k.
1568 ahash performance varies for different chunk sizes on
1569 different crypto accelerators. This option can be used
1570 to achieve best performance for particular HW.
1574 Run specified binary instead of /sbin/init as init
1577 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1578 for working out where the kernel is dying during
1581 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1582 initcall functions. Useful for debugging built-in
1583 modules and initcalls.
1585 initrd= [BOOT] Specify the location of the initial ramdisk
1587 init_pkru= [x86] Specify the default memory protection keys rights
1588 register contents for all processes. 0x55555554 by
1589 default (disallow access to all but pkey 0). Can
1590 override in debugfs after boot.
1592 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1595 int_pln_enable [x86] Enable power limit notification interrupt
1597 integrity_audit=[IMA]
1598 Format: { "0" | "1" }
1599 0 -- basic integrity auditing messages. (Default)
1600 1 -- additional integrity auditing messages.
1602 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1604 Enable intel iommu driver.
1606 Disable intel iommu driver.
1607 igfx_off [Default Off]
1608 By default, gfx is mapped as normal device. If a gfx
1609 device has a dedicated DMAR unit, the DMAR unit is
1610 bypassed by not enabling DMAR with this option. In
1611 this case, gfx device will use physical address for
1614 With this option iommu will not optimize to look
1615 for io virtual address below 32-bit forcing dual
1616 address cycle on pci bus for cards supporting greater
1617 than 32-bit addressing. The default is to look
1618 for translation below 32-bit and if not available
1619 then look in the higher range.
1620 strict [Default Off]
1621 With this option on every unmap_single operation will
1622 result in a hardware IOTLB flush operation as opposed
1623 to batching them for performance.
1624 sp_off [Default Off]
1625 By default, super page will be supported if Intel IOMMU
1626 has the capability. With this option, super page will
1628 ecs_off [Default Off]
1629 By default, extended context tables will be supported if
1630 the hardware advertises that it has support both for the
1631 extended tables themselves, and also PASID support. With
1632 this option set, extended tables will not be used even
1633 on hardware which claims to support them.
1634 tboot_noforce [Default Off]
1635 Do not force the Intel IOMMU enabled under tboot.
1636 By default, tboot will force Intel IOMMU on, which
1637 could harm performance of some high-throughput
1638 devices like 40GBit network cards, even if identity
1640 Note that using this option lowers the security
1641 provided by tboot because it makes the system
1642 vulnerable to DMA attacks.
1644 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1645 0 disables intel_idle and fall back on acpi_idle.
1646 1 to 9 specify maximum depth of C-state.
1650 Do not enable intel_pstate as the default
1651 scaling driver for the supported processors
1653 Use intel_pstate as a scaling driver, but configure it
1654 to work with generic cpufreq governors (instead of
1655 enabling its internal governor). This mode cannot be
1656 used along with the hardware-managed P-states (HWP)
1659 Enable intel_pstate on systems that prohibit it by default
1660 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1661 instead of acpi-cpufreq may disable platform features, such
1662 as thermal controls and power capping, that rely on ACPI
1663 P-States information being indicated to OSPM and therefore
1664 should be used with caution. This option does not work with
1665 processors that aren't supported by the intel_pstate driver
1666 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1668 Do not enable hardware P state control (HWP)
1671 Only load intel_pstate on systems which support
1672 hardware P state control (HWP) if available.
1674 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1675 Description Table, specifies preferred power management
1676 profile as "Enterprise Server" or "Performance Server",
1677 then this feature is turned on by default.
1679 Allow per-logical-CPU P-State performance control limits using
1680 cpufreq sysfs interface
1682 intremap= [X86-64, Intel-IOMMU]
1683 on enable Interrupt Remapping (default)
1684 off disable Interrupt Remapping
1685 nosid disable Source ID checking
1687 BIOS x2APIC opt-out request will be ignored
1688 nopost disable Interrupt Posting
1690 iomem= Disable strict checking of access to MMIO memory
1691 strict regions from userspace.
1706 nobypass [PPC/POWERNV]
1707 Disable IOMMU bypass, using IOMMU for PCI devices.
1710 [ARM64] Configure DMA to bypass the IOMMU by default.
1711 Format: { "0" | "1" }
1712 0 - Use IOMMU translation for DMA.
1713 1 - Bypass the IOMMU for DMA.
1714 unset - Use IOMMU translation for DMA.
1716 io7= [HW] IO7 for Marvel based alpha systems
1717 See comment before marvel_specify_io7 in
1718 arch/alpha/kernel/core_marvel.c.
1720 io_delay= [X86] I/O delay method
1722 Standard port 0x80 based delay
1724 Alternate port 0xed based delay (needed on some systems)
1726 Simple two microseconds delay
1731 See Documentation/filesystems/nfs/nfsroot.txt.
1733 irqaffinity= [SMP] Set the default irq affinity mask
1734 The argument is a cpu list, as described above.
1736 irqchip.gicv2_force_probe=
1739 Force the kernel to look for the second 4kB page
1740 of a GICv2 controller even if the memory range
1741 exposed by the device tree is too small.
1744 When an interrupt is not handled search all handlers
1745 for it. Intended to get systems with badly broken
1749 When an interrupt is not handled search all handlers
1750 for it. Also check all handlers each timer
1751 interrupt. Intended to get systems with badly broken
1755 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1757 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
1758 [Deprecated - use cpusets instead]
1759 Format: [flag-list,]<cpu-list>
1761 Specify one or more CPUs to isolate from disturbances
1762 specified in the flag list (default: domain):
1765 Disable the tick when a single task runs.
1767 Isolate from the general SMP balancing and scheduling
1768 algorithms. Note that performing domain isolation this way
1769 is irreversible: it's not possible to bring back a CPU to
1770 the domains once isolated through isolcpus. It's strongly
1771 advised to use cpusets instead to disable scheduler load
1772 balancing through the "cpuset.sched_load_balance" file.
1773 It offers a much more flexible interface where CPUs can
1774 move in and out of an isolated set anytime.
1776 You can move a process onto or off an "isolated" CPU via
1777 the CPU affinity syscalls or cpuset.
1778 <cpu number> begins at 0 and the maximum value is
1779 "number of CPUs in system - 1".
1781 The format of <cpu-list> is described above.
1787 ivrs_ioapic [HW,X86_64]
1788 Provide an override to the IOAPIC-ID<->DEVICE-ID
1789 mapping provided in the IVRS ACPI table. For
1790 example, to map IOAPIC-ID decimal 10 to
1791 PCI device 00:14.0 write the parameter as:
1792 ivrs_ioapic[10]=00:14.0
1794 ivrs_hpet [HW,X86_64]
1795 Provide an override to the HPET-ID<->DEVICE-ID
1796 mapping provided in the IVRS ACPI table. For
1797 example, to map HPET-ID decimal 0 to
1798 PCI device 00:14.0 write the parameter as:
1799 ivrs_hpet[0]=00:14.0
1801 ivrs_acpihid [HW,X86_64]
1802 Provide an override to the ACPI-HID:UID<->DEVICE-ID
1803 mapping provided in the IVRS ACPI table. For
1804 example, to map UART-HID:UID AMD0020:0 to
1805 PCI device 00:14.5 write the parameter as:
1806 ivrs_acpihid[00:14.5]=AMD0020:0
1808 js= [HW,JOY] Analog joystick
1809 See Documentation/input/joydev/joystick.rst.
1812 When CONFIG_RANDOMIZE_BASE is set, this disables
1813 kernel and module base offset ASLR (Address Space
1814 Layout Randomization).
1817 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
1818 report on every invalid memory access. Without this
1819 parameter KASAN will print report only for the first
1824 kernelcore= [KNL,X86,IA-64,PPC]
1825 Format: nn[KMGTPE] | "mirror"
1827 specifies the amount of memory usable by the kernel
1828 for non-movable allocations. The requested amount is
1829 spread evenly throughout all nodes in the system. The
1830 remaining memory in each node is used for Movable
1831 pages. In the event, a node is too small to have both
1832 kernelcore and Movable pages, kernelcore pages will
1833 take priority and other nodes will have a larger number
1834 of Movable pages. The Movable zone is used for the
1835 allocation of pages that may be reclaimed or moved
1836 by the page migration subsystem. This means that
1837 HugeTLB pages may not be allocated from this zone.
1838 Note that allocations like PTEs-from-HighMem still
1839 use the HighMem zone if it exists, and the Normal
1840 zone if it does not.
1842 Instead of specifying the amount of memory (nn[KMGTPE]),
1843 you can specify "mirror" option. In case "mirror"
1844 option is specified, mirrored (reliable) memory is used
1845 for non-movable allocations and remaining memory is used
1846 for Movable pages. nn[KMGTPE] and "mirror" are exclusive,
1847 so you can NOT specify nn[KMGTPE] and "mirror" at the same
1850 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
1851 Format: <Controller#>[,poll interval]
1852 The controller # is the number of the ehci usb debug
1853 port as it is probed via PCI. The poll interval is
1854 optional and is the number seconds in between
1855 each poll cycle to the debug port in case you need
1856 the functionality for interrupting the kernel with
1857 gdb or control-c on the dbgp connection. When
1858 not using this parameter you use sysrq-g to break into
1859 the kernel debugger.
1861 kgdboc= [KGDB,HW] kgdb over consoles.
1862 Requires a tty driver that supports console polling,
1863 or a supported polling keyboard driver (non-usb).
1864 Serial only format: <serial_device>[,baud]
1865 keyboard only format: kbd
1866 keyboard and serial format: kbd,<serial_device>[,baud]
1867 Optional Kernel mode setting:
1868 kms, kbd format: kms,kbd
1869 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
1871 kgdbwait [KGDB] Stop kernel execution and enter the
1872 kernel debugger at the earliest opportunity.
1874 kmac= [MIPS] korina ethernet MAC address.
1875 Configure the RouterBoard 532 series on-chip
1876 Ethernet adapter MAC address.
1878 kmemleak= [KNL] Boot-time kmemleak enable/disable
1879 Valid arguments: on, off
1881 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
1884 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
1885 Default is 0 (don't ignore, but inject #GP)
1887 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
1891 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
1892 Default is 1 (enabled)
1894 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
1896 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
1898 kvm-arm.vgic_v3_group0_trap=
1899 [KVM,ARM] Trap guest accesses to GICv3 group-0
1902 kvm-arm.vgic_v3_group1_trap=
1903 [KVM,ARM] Trap guest accesses to GICv3 group-1
1906 kvm-arm.vgic_v3_common_trap=
1907 [KVM,ARM] Trap guest accesses to GICv3 common
1910 kvm-arm.vgic_v4_enable=
1911 [KVM,ARM] Allow use of GICv4 for direct injection of
1914 kvm-intel.ept= [KVM,Intel] Disable extended page tables
1915 (virtualized MMU) support on capable Intel chips.
1916 Default is 1 (enabled)
1918 kvm-intel.emulate_invalid_guest_state=
1919 [KVM,Intel] Enable emulation of invalid guest states
1920 Default is 0 (disabled)
1922 kvm-intel.flexpriority=
1923 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
1924 Default is 1 (enabled)
1927 [KVM,Intel] Enable VMX nesting (nVMX).
1928 Default is 0 (disabled)
1930 kvm-intel.unrestricted_guest=
1931 [KVM,Intel] Disable unrestricted guest feature
1932 (virtualized real and unpaged mode) on capable
1933 Intel chips. Default is 1 (enabled)
1935 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
1936 feature (tagged TLBs) on capable Intel chips.
1937 Default is 1 (enabled)
1943 lapic [X86-32,APIC] Enable the local APIC even if BIOS
1946 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
1947 value for LAPIC timer one-shot implementation. Default
1948 back to the programmable timer unit in the LAPIC.
1950 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
1953 libata.dma= [LIBATA] DMA control
1954 libata.dma=0 Disable all PATA and SATA DMA
1955 libata.dma=1 PATA and SATA Disk DMA only
1956 libata.dma=2 ATAPI (CDROM) DMA only
1957 libata.dma=4 Compact Flash DMA only
1958 Combinations also work, so libata.dma=3 enables DMA
1959 for disks and CDROMs, but not CFs.
1961 libata.ignore_hpa= [LIBATA] Ignore HPA limit
1962 libata.ignore_hpa=0 keep BIOS limits (default)
1963 libata.ignore_hpa=1 ignore limits, using full disk
1965 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
1969 libata.force= [LIBATA] Force configurations. The format is comma
1970 separated list of "[ID:]VAL" where ID is
1971 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
1972 matching port, link or device. Basically, it matches
1973 the ATA ID string printed on console by libata. If
1974 the whole ID part is omitted, the last PORT and DEVICE
1975 values are used. If ID hasn't been specified yet, the
1976 configuration applies to all ports, links and devices.
1978 If only DEVICE is omitted, the parameter applies to
1979 the port and all links and devices behind it. DEVICE
1980 number of 0 either selects the first device or the
1981 first fan-out link behind PMP device. It does not
1982 select the host link. DEVICE number of 15 selects the
1983 host link and device attached to it.
1985 The VAL specifies the configuration to force. As long
1986 as there's no ambiguity shortcut notation is allowed.
1987 For example, both 1.5 and 1.5G would work for 1.5Gbps.
1988 The following configurations can be forced.
1990 * Cable type: 40c, 80c, short40c, unk, ign or sata.
1991 Any ID with matching PORT is used.
1993 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
1995 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
1996 udma[/][16,25,33,44,66,100,133] notation is also
1999 * [no]ncq: Turn on or off NCQ.
2001 * [no]ncqtrim: Turn off queued DSM TRIM.
2003 * nohrst, nosrst, norst: suppress hard, soft
2006 * rstonce: only attempt one reset during
2007 hot-unplug link recovery
2009 * dump_id: dump IDENTIFY data.
2011 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2013 * disable: Disable this device.
2015 If there are multiple matching configurations changing
2016 the same attribute, the last one is used.
2018 memblock=debug [KNL] Enable memblock debug messages.
2020 load_ramdisk= [RAM] List of ramdisks to load from floppy
2021 See Documentation/blockdev/ramdisk.txt.
2023 lockd.nlm_grace_period=P [NFS] Assign grace period.
2026 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2029 lockd.nlm_timeout=T [NFS] Assign timeout value.
2032 lockd.nlm_udpport=M [NFS] Assign UDP port.
2035 locktorture.nreaders_stress= [KNL]
2036 Set the number of locking read-acquisition kthreads.
2037 Defaults to being automatically set based on the
2038 number of online CPUs.
2040 locktorture.nwriters_stress= [KNL]
2041 Set the number of locking write-acquisition kthreads.
2043 locktorture.onoff_holdoff= [KNL]
2044 Set time (s) after boot for CPU-hotplug testing.
2046 locktorture.onoff_interval= [KNL]
2047 Set time (s) between CPU-hotplug operations, or
2048 zero to disable CPU-hotplug testing.
2050 locktorture.shuffle_interval= [KNL]
2051 Set task-shuffle interval (jiffies). Shuffling
2052 tasks allows some CPUs to go into dyntick-idle
2053 mode during the locktorture test.
2055 locktorture.shutdown_secs= [KNL]
2056 Set time (s) after boot system shutdown. This
2057 is useful for hands-off automated testing.
2059 locktorture.stat_interval= [KNL]
2060 Time (s) between statistics printk()s.
2062 locktorture.stutter= [KNL]
2063 Time (s) to stutter testing, for example,
2064 specifying five seconds causes the test to run for
2065 five seconds, wait for five seconds, and so on.
2066 This tests the locking primitive's ability to
2067 transition abruptly to and from idle.
2069 locktorture.torture_type= [KNL]
2070 Specify the locking implementation to test.
2072 locktorture.verbose= [KNL]
2073 Enable additional printk() statements.
2075 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2078 loglevel= All Kernel Messages with a loglevel smaller than the
2079 console loglevel will be printed to the console. It can
2080 also be changed with klogd or other programs. The
2081 loglevels are defined as follows:
2083 0 (KERN_EMERG) system is unusable
2084 1 (KERN_ALERT) action must be taken immediately
2085 2 (KERN_CRIT) critical conditions
2086 3 (KERN_ERR) error conditions
2087 4 (KERN_WARNING) warning conditions
2088 5 (KERN_NOTICE) normal but significant condition
2089 6 (KERN_INFO) informational
2090 7 (KERN_DEBUG) debug-level messages
2092 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2093 in bytes. n must be a power of two and greater
2094 than the minimal size. The minimal size is defined
2095 by LOG_BUF_SHIFT kernel config parameter. There is
2096 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2097 that allows to increase the default size depending on
2098 the number of CPUs. See init/Kconfig for more details.
2100 logo.nologo [FB] Disables display of the built-in Linux logo.
2101 This may be used to provide more screen space for
2102 kernel log messages and is useful when debugging
2103 kernel boot problems.
2105 lp=0 [LP] Specify parallel ports to use, e.g,
2106 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2107 lp=reset first parallel port). 'lp=0' disables the
2108 lp=auto printer driver. 'lp=reset' (which can be
2109 specified in addition to the ports) causes
2110 attached printers to be reset. Using
2111 lp=port1,port2,... specifies the parallel ports
2112 to associate lp devices with, starting with
2113 lp0. A port specification may be 'none' to skip
2114 that lp device, or a parport name such as
2115 'parport0'. Specifying 'lp=auto' instead of a
2116 port specification list means that device IDs
2117 from each port should be examined, to see if
2118 an IEEE 1284-compliant printer is attached; if
2119 so, the driver will manage that printer.
2120 See also header of drivers/char/lp.c.
2123 Sets loops_per_jiffy to given constant, thus avoiding
2124 time-consuming boot-time autodetection (up to 250 ms per
2125 CPU). 0 enables autodetection (default). To determine
2126 the correct value for your kernel, boot with normal
2127 autodetection and see what value is printed. Note that
2128 on SMP systems the preset will be applied to all CPUs,
2129 which is likely to cause problems if your CPUs need
2130 significantly divergent settings. An incorrect value
2131 will cause delays in the kernel to be wrong, leading to
2132 unpredictable I/O errors and other breakage. Although
2133 unlikely, in the extreme case this might damage your
2137 Format: <io>,<irq>,<dma>
2139 machvec= [IA-64] Force the use of a particular machine-vector
2140 (machvec) in a generic kernel.
2141 Example: machvec=hpzx1_swiotlb
2143 machtype= [Loongson] Share the same kernel image file between different
2145 Example: machtype=lemote-yeeloong-2f-7inch
2147 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2148 than or equal to this physical address is ignored.
2150 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2151 will bring up during bootup. maxcpus=n : n >= 0 limits
2152 the kernel to bring up 'n' processors. Surely after
2153 bootup you can bring up the other plugged cpu by executing
2154 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2155 only takes effect during system bootup.
2156 While n=0 is a special case, it is equivalent to "nosmp",
2157 which also disables the IO APIC.
2159 max_loop= [LOOP] The number of loop block devices that get
2160 (loop.max_loop) unconditionally pre-created at init time. The default
2161 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2162 of statically allocating a predefined number, loop
2163 devices can be requested on-demand with the
2164 /dev/loop-control interface.
2166 mce [X86-32] Machine Check Exception
2168 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.txt
2170 md= [HW] RAID subsystems devices and level
2171 See Documentation/admin-guide/md.rst.
2174 Format: <first>,<last>
2175 Specifies range of consoles to be captured by the MDA.
2177 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2178 Amount of memory to be used when the kernel is not able
2179 to see the whole system memory or for test.
2180 [X86] Work as limiting max address. Use together
2181 with memmap= to avoid physical address space collisions.
2182 Without memmap= PCI devices could be placed at addresses
2183 belonging to unused RAM.
2185 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2189 [KNL,SH] Allow user to override the default size for
2190 per-device physically contiguous DMA buffers.
2192 memhp_default_state=online/offline
2193 [KNL] Set the initial state for the memory hotplug
2194 onlining policy. If not specified, the default value is
2195 set according to the
2196 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2198 See Documentation/memory-hotplug.txt.
2200 memmap=exactmap [KNL,X86] Enable setting of an exact
2201 E820 memory map, as specified by the user.
2202 Such memmap=exactmap lines can be constructed based on
2203 BIOS output or other requirements. See the memmap=nn@ss
2206 memmap=nn[KMG]@ss[KMG]
2207 [KNL] Force usage of a specific region of memory.
2208 Region of memory to be used is from ss to ss+nn.
2209 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2210 which limits max address to nn[KMG].
2211 Multiple different regions can be specified,
2214 memmap=100M@2G,100M#3G,1G!1024G
2216 memmap=nn[KMG]#ss[KMG]
2217 [KNL,ACPI] Mark specific memory as ACPI data.
2218 Region of memory to be marked is from ss to ss+nn.
2220 memmap=nn[KMG]$ss[KMG]
2221 [KNL,ACPI] Mark specific memory as reserved.
2222 Region of memory to be reserved is from ss to ss+nn.
2223 Example: Exclude memory from 0x18690000-0x1869ffff
2224 memmap=64K$0x18690000
2226 memmap=0x10000$0x18690000
2227 Some bootloaders may need an escape character before '$',
2228 like Grub2, otherwise '$' and the following number
2231 memmap=nn[KMG]!ss[KMG]
2232 [KNL,X86] Mark specific memory as protected.
2233 Region of memory to be used, from ss to ss+nn.
2234 The memory region may be marked as e820 type 12 (0xc)
2235 and is NVDIMM or ADR memory.
2237 memory_corruption_check=0/1 [X86]
2238 Some BIOSes seem to corrupt the first 64k of
2239 memory when doing things like suspend/resume.
2240 Setting this option will scan the memory
2241 looking for corruption. Enabling this will
2242 both detect corruption and prevent the kernel
2243 from using the memory being corrupted.
2244 However, its intended as a diagnostic tool; if
2245 repeatable BIOS-originated corruption always
2246 affects the same memory, you can use memmap=
2247 to prevent the kernel from using that memory.
2249 memory_corruption_check_size=size [X86]
2250 By default it checks for corruption in the low
2251 64k, making this memory unavailable for normal
2252 use. Use this parameter to scan for
2253 corruption in more or less memory.
2255 memory_corruption_check_period=seconds [X86]
2256 By default it checks for corruption every 60
2257 seconds. Use this parameter to check at some
2258 other rate. 0 disables periodic checking.
2260 memtest= [KNL,X86,ARM] Enable memtest
2262 default : 0 <disable>
2263 Specifies the number of memtest passes to be
2264 performed. Each pass selects another test
2265 pattern from a given set of patterns. Memtest
2266 fills the memory with this pattern, validates
2267 memory contents and reserves bad memory
2268 regions that are detected.
2270 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2271 Valid arguments: on, off
2272 Default (depends on kernel configuration option):
2273 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2274 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2275 mem_encrypt=on: Activate SME
2276 mem_encrypt=off: Do not activate SME
2278 Refer to Documentation/x86/amd-memory-encryption.txt
2279 for details on when memory encryption can be activated.
2281 mem_sleep_default= [SUSPEND] Default system suspend mode:
2282 s2idle - Suspend-To-Idle
2283 shallow - Power-On Suspend or equivalent (if supported)
2284 deep - Suspend-To-RAM or equivalent (if supported)
2285 See Documentation/admin-guide/pm/sleep-states.rst.
2287 meye.*= [HW] Set MotionEye Camera parameters
2288 See Documentation/media/v4l-drivers/meye.rst.
2290 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2291 Multi-Function General Purpose Timers on AMD Geode
2294 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2295 the BIOS has incorrectly applied a workaround. TinyBIOS
2296 version 0.98 is known to be affected, 0.99 fixes the
2297 problem by letting the user disable the workaround.
2301 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2302 physical address is ignored.
2304 mini2440= [ARM,HW,KNL]
2305 Format:[0..2][b][c][t]
2307 MINI2440 configuration specification:
2308 0 - The attached screen is the 3.5" TFT
2309 1 - The attached screen is the 7" TFT
2310 2 - The VGA Shield is attached (1024x768)
2311 Leaving out the screen size parameter will not load
2312 the TFT driver, and the framebuffer will be left
2314 b - Enable backlight. The TFT backlight pin will be
2315 linked to the kernel VESA blanking code and a GPIO
2316 LED. This parameter is not necessary when using the
2318 c - Enable the s3c camera interface.
2319 t - Reserved for enabling touchscreen support. The
2320 touchscreen support is not enabled in the mainstream
2321 kernel as of 2.6.30, a preliminary port can be found
2322 in the "bleeding edge" mini2440 support kernel at
2323 http://repo.or.cz/w/linux-2.6/mini2440.git
2326 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2327 parameter allows control of the logging verbosity for
2328 the additional memory initialisation checks. A value
2329 of 0 disables mminit logging and a level of 4 will
2330 log everything. Information is printed at KERN_DEBUG
2331 so loglevel=8 may also need to be specified.
2334 [KNL] When CONFIG_MODULE_SIG is set, this means that
2335 modules without (valid) signatures will fail to load.
2336 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2337 is always true, so this option does nothing.
2339 module_blacklist= [KNL] Do not load a comma-separated list of
2340 modules. Useful for debugging problem modules.
2343 [MOUSE] Maximum time between finger touching and
2344 leaving touchpad surface for touch to be considered
2345 a tap and be reported as a left button click (for
2346 touchpads working in absolute mode only).
2348 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2349 reporting absolute coordinates, such as tablets
2350 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2351 reporting absolute coordinates, such as tablets
2353 movablecore=nn[KMG] [KNL,X86,IA-64,PPC] This parameter
2354 is similar to kernelcore except it specifies the
2355 amount of memory used for migratable allocations.
2356 If both kernelcore and movablecore is specified,
2357 then kernelcore will be at *least* the specified
2358 value but may be more. If movablecore on its own
2359 is specified, the administrator must be careful
2360 that the amount of memory usable for all allocations
2363 movable_node [KNL] Boot-time switch to make hotplugable memory
2364 NUMA nodes to be movable. This means that the memory
2365 of such nodes will be usable only for movable
2366 allocations which rules out almost all kernel
2367 allocations. Use with caution!
2369 MTD_Partition= [MTD]
2370 Format: <name>,<region-number>,<size>,<offset>
2372 MTD_Region= [MTD] Format:
2373 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2376 See drivers/mtd/cmdlinepart.c.
2378 multitce=off [PPC] This parameter disables the use of the pSeries
2379 firmware feature for updating multiple TCE entries
2382 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2384 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2386 boundary - index of last SLC block on Flex-OneNAND.
2387 The remaining blocks are configured as MLC blocks.
2388 lock - Configure if Flex-OneNAND boundary should be locked.
2389 Once locked, the boundary cannot be changed.
2390 1 indicates lock status, 0 indicates unlock status.
2393 ARM/S3C2412 JIVE boot control
2395 See arch/arm/mach-s3c2412/mach-jive.c
2397 mtouchusb.raw_coordinates=
2398 [HW] Make the MicroTouch USB driver use raw coordinates
2399 ('y', default) or cooked coordinates ('n')
2401 mtrr_chunk_size=nn[KMG] [X86]
2402 used for mtrr cleanup. It is largest continuous chunk
2403 that could hold holes aka. UC entries.
2405 mtrr_gran_size=nn[KMG] [X86]
2406 Used for mtrr cleanup. It is granularity of mtrr block.
2408 Large value could prevent small alignment from
2411 mtrr_spare_reg_nr=n [X86]
2413 Range: 0,7 : spare reg number
2415 Used for mtrr cleanup. It is spare mtrr entries number.
2416 Set to 2 or more if your graphical card needs more.
2418 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2420 netdev= [NET] Network devices parameters
2421 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2422 Note that mem_start is often overloaded to mean
2423 something different and driver-specific.
2424 This usage is only documented in each driver source
2428 [NETFILTER] Enable connection tracking flow accounting
2429 0 to disable accounting
2430 1 to enable accounting
2433 nfsaddrs= [NFS] Deprecated. Use ip= instead.
2434 See Documentation/filesystems/nfs/nfsroot.txt.
2436 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
2437 See Documentation/filesystems/nfs/nfsroot.txt.
2439 nfsrootdebug [NFS] enable nfsroot debugging messages.
2440 See Documentation/filesystems/nfs/nfsroot.txt.
2442 nfs.callback_nr_threads=
2443 [NFSv4] set the total number of threads that the
2444 NFS client will assign to service NFSv4 callback
2447 nfs.callback_tcpport=
2448 [NFS] set the TCP port on which the NFSv4 callback
2449 channel should listen.
2452 [NFS] sets the pathname to the program which is used
2453 to update the NFS client cache entries.
2455 nfs.cache_getent_timeout=
2456 [NFS] sets the timeout after which an attempt to
2457 update a cache entry is deemed to have failed.
2459 nfs.idmap_cache_timeout=
2460 [NFS] set the maximum lifetime for idmapper cache
2464 [NFS] enable 64-bit inode numbers.
2465 If zero, the NFS client will fake up a 32-bit inode
2466 number for the readdir() and stat() syscalls instead
2467 of returning the full 64-bit number.
2468 The default is to return 64-bit inode numbers.
2470 nfs.max_session_cb_slots=
2471 [NFSv4.1] Sets the maximum number of session
2472 slots the client will assign to the callback
2473 channel. This determines the maximum number of
2474 callbacks the client will process in parallel for
2475 a particular server.
2477 nfs.max_session_slots=
2478 [NFSv4.1] Sets the maximum number of session slots
2479 the client will attempt to negotiate with the server.
2480 This limits the number of simultaneous RPC requests
2481 that the client can send to the NFSv4.1 server.
2482 Note that there is little point in setting this
2483 value higher than the max_tcp_slot_table_limit.
2485 nfs.nfs4_disable_idmapping=
2486 [NFSv4] When set to the default of '1', this option
2487 ensures that both the RPC level authentication
2488 scheme and the NFS level operations agree to use
2489 numeric uids/gids if the mount is using the
2490 'sec=sys' security flavour. In effect it is
2491 disabling idmapping, which can make migration from
2492 legacy NFSv2/v3 systems to NFSv4 easier.
2493 Servers that do not support this mode of operation
2494 will be autodetected by the client, and it will fall
2495 back to using the idmapper.
2496 To turn off this behaviour, set the value to '0'.
2498 [NFS4] Specify an additional fixed unique ident-
2499 ification string that NFSv4 clients can insert into
2500 their nfs_client_id4 string. This is typically a
2501 UUID that is generated at system install time.
2503 nfs.send_implementation_id =
2504 [NFSv4.1] Send client implementation identification
2505 information in exchange_id requests.
2506 If zero, no implementation identification information
2508 The default is to send the implementation identification
2511 nfs.recover_lost_locks =
2512 [NFSv4] Attempt to recover locks that were lost due
2513 to a lease timeout on the server. Please note that
2514 doing this risks data corruption, since there are
2515 no guarantees that the file will remain unchanged
2516 after the locks are lost.
2517 If you want to enable the kernel legacy behaviour of
2518 attempting to recover these locks, then set this
2520 The default parameter value of '0' causes the kernel
2521 not to attempt recovery of lost locks.
2523 nfs4.layoutstats_timer =
2524 [NFSv4.2] Change the rate at which the kernel sends
2525 layoutstats to the pNFS metadata server.
2527 Setting this to value to 0 causes the kernel to use
2528 whatever value is the default set by the layout
2529 driver. A non-zero value sets the minimum interval
2530 in seconds between layoutstats transmissions.
2532 nfsd.nfs4_disable_idmapping=
2533 [NFSv4] When set to the default of '1', the NFSv4
2534 server will return only numeric uids and gids to
2535 clients using auth_sys, and will accept numeric uids
2536 and gids from such clients. This is intended to ease
2537 migration from NFSv2/v3.
2539 nmi_debug= [KNL,SH] Specify one or more actions to take
2540 when a NMI is triggered.
2541 Format: [state][,regs][,debounce][,die]
2543 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
2544 Format: [panic,][nopanic,][num]
2546 0 - turn hardlockup detector in nmi_watchdog off
2547 1 - turn hardlockup detector in nmi_watchdog on
2548 When panic is specified, panic when an NMI watchdog
2549 timeout occurs (or 'nopanic' to override the opposite
2550 default). To disable both hard and soft lockup detectors,
2551 please see 'nowatchdog'.
2552 This is useful when you use a panic=... timeout and
2553 need the box quickly up again.
2555 These settings can be accessed at runtime via
2556 the nmi_watchdog and hardlockup_panic sysctls.
2558 netpoll.carrier_timeout=
2559 [NET] Specifies amount of time (in seconds) that
2560 netpoll should wait for a carrier. By default netpoll
2563 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
2564 emulation library even if a 387 maths coprocessor
2568 [HW] Never suspend the console
2569 Disable suspending of consoles during suspend and
2570 hibernate operations. Once disabled, debugging
2571 messages can reach various consoles while the rest
2572 of the system is being put to sleep (ie, while
2573 debugging driver suspend/resume hooks). This may
2574 not work reliably with all consoles, but is known
2575 to work with serial and VGA consoles.
2576 To facilitate more flexible debugging, we also add
2577 console_suspend, a printk module parameter to control
2578 it. Users could use console_suspend (usually
2579 /sys/module/printk/parameters/console_suspend) to
2580 turn on/off it dynamically.
2582 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
2583 caches in the slab allocator. Saves per-node memory,
2584 but will impact performance.
2588 noaltinstr [S390] Disables alternative instructions patching
2589 (CPU alternatives feature).
2591 noapic [SMP,APIC] Tells the kernel to not make use of any
2592 IOAPICs that may be present in the system.
2594 noautogroup Disable scheduler automatic task group creation.
2596 nobats [PPC] Do not use BATs for mapping kernel lowmem
2597 on "Classic" PPC cores.
2601 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
2603 nodelayacct [KNL] Disable per-task delay accounting
2605 nodsp [SH] Disable hardware DSP at boot time.
2607 noefi Disable EFI runtime services support.
2612 On X86-32 available only on PAE configured kernels.
2613 noexec=on: enable non-executable mappings (default)
2614 noexec=off: disable non-executable mappings
2617 Disable SMAP (Supervisor Mode Access Prevention)
2618 even if it is supported by processor.
2621 Disable SMEP (Supervisor Mode Execution Prevention)
2622 even if it is supported by processor.
2625 This affects only 32-bit executables.
2626 noexec32=on: enable non-executable mappings (default)
2627 read doesn't imply executable mappings
2628 noexec32=off: disable non-executable mappings
2629 read implies executable mappings
2631 nofpu [MIPS,SH] Disable hardware FPU at boot time.
2633 nofxsr [BUGS=X86-32] Disables x86 floating point extended
2634 register save and restore. The kernel will only save
2635 legacy floating-point registers on task switch.
2637 nohugeiomap [KNL,x86] Disable kernel huge I/O mappings.
2639 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
2640 Equivalent to smt=1.
2642 nospectre_v2 [X86] Disable all mitigations for the Spectre variant 2
2643 (indirect branch prediction) vulnerability. System may
2644 allow data leaks with this option, which is equivalent
2647 noxsave [BUGS=X86] Disables x86 extended register state save
2648 and restore using xsave. The kernel will fallback to
2649 enabling legacy floating-point and sse state.
2651 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
2652 register states. The kernel will fall back to use
2653 xsave to save the states. By using this parameter,
2654 performance of saving the states is degraded because
2655 xsave doesn't support modified optimization while
2656 xsaveopt supports it on xsaveopt enabled systems.
2658 noxsaves [X86] Disables xsaves and xrstors used in saving and
2659 restoring x86 extended register state in compacted
2660 form of xsave area. The kernel will fall back to use
2661 xsaveopt and xrstor to save and restore the states
2662 in standard form of xsave area. By using this
2663 parameter, xsave area per process might occupy more
2664 memory on xsaves enabled systems.
2666 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
2667 wfi(ARM) instruction doesn't work correctly and not to
2668 use it. This is also useful when using JTAG debugger.
2670 no_file_caps Tells the kernel not to honor file capabilities. The
2671 only way then for a file to be executed with privilege
2672 is to be setuid root or executed by root.
2674 nohalt [IA-64] Tells the kernel not to use the power saving
2675 function PAL_HALT_LIGHT when idle. This increases
2676 power-consumption. On the positive side, it reduces
2677 interrupt wake-up latency, which may improve performance
2678 in certain environments such as networked servers or
2681 nohibernate [HIBERNATION] Disable hibernation and resume.
2683 nohz= [KNL] Boottime enable/disable dynamic ticks
2684 Valid arguments: on, off
2687 nohz_full= [KNL,BOOT,SMP,ISOL]
2688 The argument is a cpu list, as described above.
2689 In kernels built with CONFIG_NO_HZ_FULL=y, set
2690 the specified list of CPUs whose tick will be stopped
2691 whenever possible. The boot CPU will be forced outside
2692 the range to maintain the timekeeping. Any CPUs
2693 in this list will have their RCU callbacks offloaded,
2694 just as if they had also been called out in the
2695 rcu_nocbs= boot parameter.
2697 noiotrap [SH] Disables trapped I/O port accesses.
2699 noirqdebug [X86-32] Disables the code which attempts to detect and
2700 disable unhandled interrupt sources.
2702 no_timer_check [X86,APIC] Disables the code which tests for
2703 broken timer IRQ sources.
2705 noisapnp [ISAPNP] Disables ISA PnP code.
2707 noinitrd [RAM] Tells the kernel not to load any configured
2710 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
2712 [Deprecated - use intremap=off]
2716 noinvpcid [X86] Disable the INVPCID cpu feature.
2718 nojitter [IA-64] Disables jitter checking for ITC timers.
2720 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
2722 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
2726 [X86,PV_OPS] Disable paravirtualized VMware scheduler
2727 clock and use the default one.
2729 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting.
2730 steal time is computed, but won't influence scheduler
2733 nolapic [X86-32,APIC] Do not enable or use the local APIC.
2735 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
2737 noltlbs [PPC] Do not use large page/tlb entries for kernel
2738 lowmem mapping on PPC40x and PPC8xx
2740 nomca [IA-64] Disable machine check abort handling
2742 nomce [X86-32] Disable Machine Check Exception
2744 nomfgpt [X86-32] Disable Multi-Function General Purpose
2745 Timer usage (for AMD Geode machines).
2747 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
2748 shutdown the other cpus. Instead use the REBOOT_VECTOR
2751 nomodule Disable module load
2753 nopat [X86] Disable PAT (page attribute table extension of
2754 pagetables) support.
2756 nopcid [X86-64] Disable the PCID cpu feature.
2758 norandmaps Don't use address space randomization. Equivalent to
2759 echo 0 > /proc/sys/kernel/randomize_va_space
2761 noreplace-paravirt [X86,IA-64,PV_OPS] Don't patch paravirt_ops
2763 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
2764 with UP alternatives
2766 nordrand [X86] Disable kernel use of the RDRAND and
2767 RDSEED instructions even if they are supported
2768 by the processor. RDRAND and RDSEED are still
2769 available to user space applications.
2771 noresume [SWSUSP] Disables resume and restores original swap
2774 no-scroll [VGA] Disables scrollback.
2775 This is required for the Braillex ib80-piezo Braille
2776 reader made by F.H. Papenmeier (Germany).
2780 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
2782 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
2783 and disable the IO APIC. legacy for "maxcpus=0".
2785 nosoftlockup [KNL] Disable the soft-lockup detector.
2787 nosync [HW,M68K] Disables sync negotiation for all devices.
2789 notsc [BUGS=X86-32] Disable Time Stamp Counter
2791 nowatchdog [KNL] Disable both lockup detectors, i.e.
2792 soft-lockup and NMI watchdog (hard-lockup).
2796 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
2798 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
2799 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
2800 Some features depend on CPU0. Known dependencies are:
2801 1. Resume from suspend/hibernate depends on CPU0.
2802 Suspend/hibernate will fail if CPU0 is offline and you
2803 need to online CPU0 before suspend/hibernate.
2804 2. PIC interrupts also depend on CPU0. CPU0 can't be
2805 removed if a PIC interrupt is detected.
2806 It's said poweroff/reboot may depend on CPU0 on some
2807 machines although I haven't seen such issues so far
2808 after CPU0 is offline on a few tested machines.
2809 If the dependencies are under your control, you can
2810 turn on cpu0_hotplug.
2812 nps_mtm_hs_ctr= [KNL,ARC]
2813 This parameter sets the maximum duration, in
2814 cycles, each HW thread of the CTOP can run
2815 without interruptions, before HW switches it.
2816 The actual maximum duration is 16 times this
2818 Format: integer between 1 and 255
2821 nptcg= [IA-64] Override max number of concurrent global TLB
2822 purges which is reported from either PAL_VM_SUMMARY or
2825 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
2826 could support. nr_cpus=n : n >= 1 limits the kernel to
2827 support 'n' processors. It could be larger than the
2828 number of already plugged CPU during bootup, later in
2829 runtime you can physically add extra cpu until it reaches
2830 n. So during boot up some boot time memory for per-cpu
2831 variables need be pre-allocated for later physical cpu
2834 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
2836 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
2837 Allowed values are enable and disable
2839 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
2840 'node', 'default' can be specified
2841 This can be set from sysctl after boot.
2842 See Documentation/sysctl/vm.txt for details.
2844 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
2845 See Documentation/debugging-via-ohci1394.txt for more
2848 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
2849 Rather than timing out after 20 ms if an EC
2850 command is not properly ACKed, override the length
2851 of the timeout. We have interrupts disabled while
2852 waiting for the ACK, so if this is set too high
2853 interrupts *may* be lost!
2855 omap_mux= [OMAP] Override bootloader pin multiplexing.
2856 Format: <mux_mode0.mode_name=value>...
2857 For example, to override I2C bus2:
2858 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
2860 oprofile.timer= [HW]
2861 Use timer interrupt instead of performance counters
2863 oprofile.cpu_type= Force an oprofile cpu type
2864 This might be useful if you have an older oprofile
2865 userland or if you want common events.
2866 Format: { arch_perfmon }
2867 arch_perfmon: [X86] Force use of architectural
2868 perfmon on Intel CPUs instead of the
2869 CPU specific event set.
2870 timer: [X86] Force use of architectural NMI
2871 timer mode (see also oprofile.timer
2872 for generic hr timer mode)
2874 oops=panic Always panic on oopses. Default is to just kill the
2875 process, but there is a small probability of
2876 deadlocking the machine.
2877 This will also cause panics on machine check exceptions.
2878 Useful together with panic=30 to trigger a reboot.
2881 See Documentation/sound/oss/oss-parameters.txt
2883 page_owner= [KNL] Boot-time page_owner enabling option.
2884 Storage of the information about who allocated
2885 each page is disabled in default. With this switch,
2887 on: enable the feature
2889 page_poison= [KNL] Boot-time parameter changing the state of
2890 poisoning on the buddy allocator.
2891 off: turn off poisoning
2892 on: turn on poisoning
2894 panic= [KNL] Kernel behaviour on panic: delay <timeout>
2895 timeout > 0: seconds before rebooting
2896 timeout = 0: wait forever
2897 timeout < 0: reboot immediately
2900 panic_on_warn panic() instead of WARN(). Useful to cause kdump
2903 crash_kexec_post_notifiers
2904 Run kdump after running panic-notifiers and dumping
2905 kmsg. This only for the users who doubt kdump always
2906 succeeds in any situation.
2907 Note that this also increases risks of kdump failure,
2908 because some panic notifiers can make the crashed
2909 kernel more unstable.
2911 parkbd.port= [HW] Parallel port number the keyboard adapter is
2912 connected to, default is 0.
2914 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
2915 0 for XT, 1 for AT (default is AT).
2918 parport= [HW,PPT] Specify parallel ports. 0 disables.
2919 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
2920 Use 'auto' to force the driver to use any
2921 IRQ/DMA settings detected (the default is to
2922 ignore detected IRQ/DMA settings because of
2923 possible conflicts). You can specify the base
2924 address, IRQ, and DMA settings; IRQ and DMA
2925 should be numbers, or 'auto' (for using detected
2926 settings on that particular port), or 'nofifo'
2927 (to avoid using a FIFO even if it is detected).
2928 Parallel ports are assigned in the order they
2929 are specified on the command line, starting
2932 parport_init_mode= [HW,PPT]
2933 Configure VIA parallel port to operate in
2934 a specific mode. This is necessary on Pegasos
2935 computer where firmware has no options for setting
2936 up parallel port mode and sets it to spp.
2937 Currently this function knows 686a and 8231 chips.
2938 Format: [spp|ps2|epp|ecp|ecpepp]
2941 Halt all CPUs after the first oops has been printed for
2942 the specified number of seconds. This is to be used if
2943 your oopses keep scrolling off the screen.
2948 See header of drivers/block/paride/pcd.c.
2949 See also Documentation/blockdev/paride.txt.
2951 pci=option[,option...] [PCI] various PCI subsystem options:
2952 earlydump [X86] dump PCI config space before the kernel
2954 off [X86] don't probe for the PCI bus
2955 bios [X86-32] force use of PCI BIOS, don't access
2956 the hardware directly. Use this if your machine
2957 has a non-standard PCI host bridge.
2958 nobios [X86-32] disallow use of PCI BIOS, only direct
2959 hardware access methods are allowed. Use this
2960 if you experience crashes upon bootup and you
2961 suspect they are caused by the BIOS.
2962 conf1 [X86] Force use of PCI Configuration Access
2963 Mechanism 1 (config address in IO port 0xCF8,
2964 data in IO port 0xCFC, both 32-bit).
2965 conf2 [X86] Force use of PCI Configuration Access
2966 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
2967 the function, IO port 0xCFA, also 8-bit, sets
2968 bus number. The config space is then accessed
2969 through ports 0xC000-0xCFFF).
2970 See http://wiki.osdev.org/PCI for more info
2971 on the configuration access mechanisms.
2972 noaer [PCIE] If the PCIEAER kernel config parameter is
2973 enabled, this kernel boot option can be used to
2974 disable the use of PCIE advanced error reporting.
2975 nodomains [PCI] Disable support for multiple PCI
2976 root domains (aka PCI segments, in ACPI-speak).
2977 nommconf [X86] Disable use of MMCONFIG for PCI
2979 check_enable_amd_mmconf [X86] check for and enable
2980 properly configured MMIO access to PCI
2981 config space on AMD family 10h CPU
2982 nomsi [MSI] If the PCI_MSI kernel config parameter is
2983 enabled, this kernel boot option can be used to
2984 disable the use of MSI interrupts system-wide.
2985 noioapicquirk [APIC] Disable all boot interrupt quirks.
2986 Safety option to keep boot IRQs enabled. This
2987 should never be necessary.
2988 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
2989 primary IO-APIC for bridges that cannot disable
2990 boot IRQs. This fixes a source of spurious IRQs
2991 when the system masks IRQs.
2992 noioapicreroute [APIC] Disable workaround that uses the
2993 boot IRQ equivalent of an IRQ that connects to
2994 a chipset where boot IRQs cannot be disabled.
2995 The opposite of ioapicreroute.
2996 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
2997 routing table. These calls are known to be buggy
2998 on several machines and they hang the machine
2999 when used, but on other computers it's the only
3000 way to get the interrupt routing table. Try
3001 this option if the kernel is unable to allocate
3002 IRQs or discover secondary PCI buses on your
3004 rom [X86] Assign address space to expansion ROMs.
3005 Use with caution as certain devices share
3006 address decoders between ROMs and other
3008 norom [X86] Do not assign address space to
3009 expansion ROMs that do not already have
3010 BIOS assigned address ranges.
3011 nobar [X86] Do not assign address space to the
3012 BARs that weren't assigned by the BIOS.
3013 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3014 assigned automatically to PCI devices. You can
3015 make the kernel exclude IRQs of your ISA cards
3017 pirqaddr=0xAAAAA [X86] Specify the physical address
3018 of the PIRQ table (normally generated
3019 by the BIOS) if it is outside the
3020 F0000h-100000h range.
3021 lastbus=N [X86] Scan all buses thru bus #N. Can be
3022 useful if the kernel is unable to find your
3023 secondary buses and you want to tell it
3024 explicitly which ones they are.
3025 assign-busses [X86] Always assign all PCI bus
3026 numbers ourselves, overriding
3027 whatever the firmware may have done.
3028 usepirqmask [X86] Honor the possible IRQ mask stored
3029 in the BIOS $PIR table. This is needed on
3030 some systems with broken BIOSes, notably
3031 some HP Pavilion N5400 and Omnibook XE3
3032 notebooks. This will have no effect if ACPI
3033 IRQ routing is enabled.
3034 noacpi [X86] Do not use ACPI for IRQ routing
3035 or for PCI scanning.
3036 use_crs [X86] Use PCI host bridge window information
3037 from ACPI. On BIOSes from 2008 or later, this
3038 is enabled by default. If you need to use this,
3039 please report a bug.
3040 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3041 If you need to use this, please report a bug.
3042 routeirq Do IRQ routing for all PCI devices.
3043 This is normally done in pci_enable_device(),
3044 so this option is a temporary workaround
3045 for broken drivers that don't call it.
3046 skip_isa_align [X86] do not align io start addr, so can
3047 handle more pci cards
3048 noearly [X86] Don't do any early type 1 scanning.
3049 This might help on some broken boards which
3050 machine check when some devices' config space
3051 is read. But various workarounds are disabled
3052 and some IOMMU drivers will not work.
3053 bfsort Sort PCI devices into breadth-first order.
3054 This sorting is done to get a device
3055 order compatible with older (<= 2.4) kernels.
3056 nobfsort Don't sort PCI devices into breadth-first order.
3057 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3058 tuning and use the BIOS-configured MPS defaults.
3059 pcie_bus_safe Set every device's MPS to the largest value
3060 supported by all devices below the root complex.
3061 pcie_bus_perf Set device MPS to the largest allowable MPS
3062 based on its parent bus. Also set MRRS (Max
3063 Read Request Size) to the largest supported
3064 value (no larger than the MPS that the device
3065 or bus can support) for best performance.
3066 pcie_bus_peer2peer Set every device's MPS to 128B, which
3067 every device is guaranteed to support. This
3068 configuration allows peer-to-peer DMA between
3069 any pair of devices, possibly at the cost of
3070 reduced performance. This also guarantees
3071 that hot-added devices will work.
3072 cbiosize=nn[KMG] The fixed amount of bus space which is
3073 reserved for the CardBus bridge's IO window.
3074 The default value is 256 bytes.
3075 cbmemsize=nn[KMG] The fixed amount of bus space which is
3076 reserved for the CardBus bridge's memory
3077 window. The default value is 64 megabytes.
3080 [<order of align>@][<domain>:]<bus>:<slot>.<func>[; ...]
3081 [<order of align>@]pci:<vendor>:<device>\
3082 [:<subvendor>:<subdevice>][; ...]
3083 Specifies alignment and device to reassign
3084 aligned memory resources.
3085 If <order of align> is not specified,
3086 PAGE_SIZE is used as alignment.
3087 PCI-PCI bridge can be specified, if resource
3088 windows need to be expanded.
3089 To specify the alignment for several
3090 instances of a device, the PCI vendor,
3091 device, subvendor, and subdevice may be
3092 specified, e.g., 4096@pci:8086:9c22:103c:198f
3093 ecrc= Enable/disable PCIe ECRC (transaction layer
3094 end-to-end CRC checking).
3095 bios: Use BIOS/firmware settings. This is the
3099 hpiosize=nn[KMG] The fixed amount of bus space which is
3100 reserved for hotplug bridge's IO window.
3101 Default size is 256 bytes.
3102 hpmemsize=nn[KMG] The fixed amount of bus space which is
3103 reserved for hotplug bridge's memory window.
3104 Default size is 2 megabytes.
3105 hpbussize=nn The minimum amount of additional bus numbers
3106 reserved for buses below a hotplug bridge.
3108 realloc= Enable/disable reallocating PCI bridge resources
3109 if allocations done by BIOS are too small to
3110 accommodate resources required by all child
3112 off: Turn realloc off
3114 realloc same as realloc=on
3115 noari do not use PCIe ARI.
3116 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3117 only look for one device below a PCIe downstream
3119 big_root_window Try to add a big 64bit memory window to the PCIe
3120 root complex on AMD CPUs. Some GFX hardware
3121 can resize a BAR to allow access to all VRAM.
3122 Adding the window is slightly risky (it may
3123 conflict with unreported devices), so this
3126 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3129 force Enable ASPM even on devices that claim not to support it.
3130 WARNING: Forcing ASPM on may cause system lockups.
3132 pcie_hp= [PCIE] PCI Express Hotplug driver options:
3133 nomsi Do not use MSI for PCI Express Native Hotplug (this
3134 makes all PCIe ports use INTx for hotplug services).
3136 pcie_ports= [PCIE] PCIe ports handling:
3137 auto Ask the BIOS whether or not to use native PCIe services
3138 associated with PCIe ports (PME, hot-plug, AER). Use
3139 them only if that is allowed by the BIOS.
3140 native Use native PCIe services associated with PCIe ports
3142 compat Treat PCIe ports as PCI-to-PCI bridges, disable the PCIe
3145 pcie_port_pm= [PCIE] PCIe port power management handling:
3146 off Disable power management of all PCIe ports
3147 force Forcibly enable power management of all PCIe ports
3149 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3150 nomsi Do not use MSI for native PCIe PME signaling (this makes
3151 all PCIe root ports use INTx for all services).
3153 pcmv= [HW,PCMCIA] BadgePAD 4
3157 Keep all power-domains already enabled by bootloader on,
3158 even if no driver has claimed them. This is useful
3159 for debug and development, but should not be
3160 needed on a platform with proper driver support.
3163 See Documentation/blockdev/paride.txt.
3165 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3168 See arch/parisc/kernel/pdc_chassis.c
3170 percpu_alloc= Select which percpu first chunk allocator to use.
3171 Currently supported values are "embed" and "page".
3172 Archs may support subset or none of the selections.
3173 See comments in mm/percpu.c for details on each
3174 allocator. This parameter is primarily for debugging
3175 and performance comparison.
3178 See Documentation/blockdev/paride.txt.
3181 See Documentation/blockdev/paride.txt.
3183 pirq= [SMP,APIC] Manual mp-table setup
3184 See Documentation/x86/i386/IO-APIC.txt.
3186 plip= [PPT,NET] Parallel port network link
3187 Format: { parport<nr> | timid | 0 }
3188 See also Documentation/admin-guide/parport.rst.
3190 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3191 Override pmtimer IOPort with a hex value.
3195 Enable PNP debug messages (depends on the
3196 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3197 via /sys/module/pnp/parameters/debug. We always show
3198 current resource usage; turning this on also shows
3199 possible settings and some assignment information.
3205 { on | off | curr | res | no-curr | no-res }
3208 [ISAPNP] Exclude IRQs for the autoconfiguration
3211 [ISAPNP] Exclude DMAs for the autoconfiguration
3213 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3214 Ranges are in pairs (I/O port base and size).
3217 [ISAPNP] Exclude memory regions for the
3219 Ranges are in pairs (memory base and size).
3221 ports= [IP_VS_FTP] IPVS ftp helper module
3223 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3225 Format: <port>,<port>....
3227 powersave=off [PPC] This option disables power saving features.
3228 It specifically disables cpuidle and sets the
3229 platform machine description specific power_save
3230 function to NULL. On Idle the CPU just reduces
3233 ppc_strict_facility_enable
3234 [PPC] This option catches any kernel floating point,
3235 Altivec, VSX and SPE outside of regions specifically
3236 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3237 There is some performance impact when enabling this.
3241 Disable Hardware Transactional Memory
3243 print-fatal-signals=
3244 [KNL] debug: print fatal signals
3246 If enabled, warn about various signal handling
3247 related application anomalies: too many signals,
3248 too many POSIX.1 timers, fatal signals causing a
3251 If you hit the warning due to signal overflow,
3252 you might want to try "ulimit -i unlimited".
3256 printk.always_kmsg_dump=
3257 Trigger kmsg_dump for cases other than kernel oops or
3259 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3262 printk.devkmsg={on,off,ratelimit}
3263 Control writing to /dev/kmsg.
3264 on - unlimited logging to /dev/kmsg from userspace
3265 off - logging to /dev/kmsg disabled
3266 ratelimit - ratelimit the logging
3269 printk.time= Show timing data prefixed to each printk message line
3270 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3272 processor.max_cstate= [HW,ACPI]
3273 Limit processor to maximum C-state
3274 max_cstate=9 overrides any DMI blacklist limit.
3276 processor.nocst [HW,ACPI]
3277 Ignore the _CST method to determine C-states,
3278 instead using the legacy FADT method
3280 profile= [KNL] Enable kernel profiling via /proc/profile
3281 Format: [<profiletype>,]<number>
3282 Param: <profiletype>: "schedule", "sleep", or "kvm"
3283 [defaults to kernel profiling]
3284 Param: "schedule" - profile schedule points.
3285 Param: "sleep" - profile D-state sleeping (millisecs).
3286 Requires CONFIG_SCHEDSTATS
3287 Param: "kvm" - profile VM exits.
3288 Param: <number> - step/bucket size as a power of 2 for
3289 statistical time based profiling.
3291 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3293 See Documentation/blockdev/ramdisk.txt.
3295 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3296 probe for; one of (bare|imps|exps|lifebook|any).
3297 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3299 psmouse.resetafter= [HW,MOUSE]
3300 Try to reset the device after so many bad packets
3303 [HW,MOUSE] Set desired mouse resolution, in dpi.
3304 psmouse.smartscroll=
3305 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
3306 0 = disabled, 1 = enabled (default).
3308 pstore.backend= Specify the name of the pstore backend to use
3311 See Documentation/blockdev/paride.txt.
3313 pti= [X86_64] Control Page Table Isolation of user and
3314 kernel address spaces. Disabling this feature
3315 removes hardening, but improves performance of
3316 system calls and interrupts.
3318 on - unconditionally enable
3319 off - unconditionally disable
3320 auto - kernel detects whether your CPU model is
3321 vulnerable to issues that PTI mitigates
3323 Not specifying this option is equivalent to pti=auto.
3326 Equivalent to pti=off
3329 [KNL] Number of legacy pty's. Overwrites compiled-in
3332 quiet [KNL] Disable most log messages
3337 See Documentation/admin-guide/md.rst.
3339 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
3340 See Documentation/blockdev/ramdisk.txt.
3342 ras=option[,option,...] [KNL] RAS-specific options
3345 Disable the Correctable Errors Collector,
3346 see CONFIG_RAS_CEC help text.
3349 The argument is a cpu list, as described above.
3351 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
3352 the specified list of CPUs to be no-callback CPUs.
3353 Invocation of these CPUs' RCU callbacks will
3354 be offloaded to "rcuox/N" kthreads created for
3355 that purpose, where "x" is "b" for RCU-bh, "p"
3356 for RCU-preempt, and "s" for RCU-sched, and "N"
3357 is the CPU number. This reduces OS jitter on the
3358 offloaded CPUs, which can be useful for HPC and
3359 real-time workloads. It can also improve energy
3360 efficiency for asymmetric multiprocessors.
3363 Rather than requiring that offloaded CPUs
3364 (specified by rcu_nocbs= above) explicitly
3365 awaken the corresponding "rcuoN" kthreads,
3366 make these kthreads poll for callbacks.
3367 This improves the real-time response for the
3368 offloaded CPUs by relieving them of the need to
3369 wake up the corresponding kthread, but degrades
3370 energy efficiency by requiring that the kthreads
3371 periodically wake up to do the polling.
3373 rcutree.blimit= [KNL]
3374 Set maximum number of finished RCU callbacks to
3375 process in one batch.
3377 rcutree.dump_tree= [KNL]
3378 Dump the structure of the rcu_node combining tree
3379 out at early boot. This is used for diagnostic
3380 purposes, to verify correct tree setup.
3382 rcutree.gp_cleanup_delay= [KNL]
3383 Set the number of jiffies to delay each step of
3384 RCU grace-period cleanup.
3386 rcutree.gp_init_delay= [KNL]
3387 Set the number of jiffies to delay each step of
3388 RCU grace-period initialization.
3390 rcutree.gp_preinit_delay= [KNL]
3391 Set the number of jiffies to delay each step of
3392 RCU grace-period pre-initialization, that is,
3393 the propagation of recent CPU-hotplug changes up
3394 the rcu_node combining tree.
3396 rcutree.rcu_fanout_exact= [KNL]
3397 Disable autobalancing of the rcu_node combining
3398 tree. This is used by rcutorture, and might
3399 possibly be useful for architectures having high
3400 cache-to-cache transfer latencies.
3402 rcutree.rcu_fanout_leaf= [KNL]
3403 Change the number of CPUs assigned to each
3404 leaf rcu_node structure. Useful for very
3405 large systems, which will choose the value 64,
3406 and for NUMA systems with large remote-access
3407 latencies, which will choose a value aligned
3408 with the appropriate hardware boundaries.
3410 rcutree.jiffies_till_sched_qs= [KNL]
3411 Set required age in jiffies for a
3412 given grace period before RCU starts
3413 soliciting quiescent-state help from
3414 rcu_note_context_switch().
3416 rcutree.jiffies_till_first_fqs= [KNL]
3417 Set delay from grace-period initialization to
3418 first attempt to force quiescent states.
3419 Units are jiffies, minimum value is zero,
3420 and maximum value is HZ.
3422 rcutree.jiffies_till_next_fqs= [KNL]
3423 Set delay between subsequent attempts to force
3424 quiescent states. Units are jiffies, minimum
3425 value is one, and maximum value is HZ.
3427 rcutree.kthread_prio= [KNL,BOOT]
3428 Set the SCHED_FIFO priority of the RCU per-CPU
3429 kthreads (rcuc/N). This value is also used for
3430 the priority of the RCU boost threads (rcub/N)
3431 and for the RCU grace-period kthreads (rcu_bh,
3432 rcu_preempt, and rcu_sched). If RCU_BOOST is
3433 set, valid values are 1-99 and the default is 1
3434 (the least-favored priority). Otherwise, when
3435 RCU_BOOST is not set, valid values are 0-99 and
3436 the default is zero (non-realtime operation).
3438 rcutree.rcu_nocb_leader_stride= [KNL]
3439 Set the number of NOCB kthread groups, which
3440 defaults to the square root of the number of
3441 CPUs. Larger numbers reduces the wakeup overhead
3442 on the per-CPU grace-period kthreads, but increases
3443 that same overhead on each group's leader.
3445 rcutree.qhimark= [KNL]
3446 Set threshold of queued RCU callbacks beyond which
3447 batch limiting is disabled.
3449 rcutree.qlowmark= [KNL]
3450 Set threshold of queued RCU callbacks below which
3451 batch limiting is re-enabled.
3453 rcutree.rcu_idle_gp_delay= [KNL]
3454 Set wakeup interval for idle CPUs that have
3455 RCU callbacks (RCU_FAST_NO_HZ=y).
3457 rcutree.rcu_idle_lazy_gp_delay= [KNL]
3458 Set wakeup interval for idle CPUs that have
3459 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
3460 Lazy RCU callbacks are those which RCU can
3461 prove do nothing more than free memory.
3463 rcutree.rcu_kick_kthreads= [KNL]
3464 Cause the grace-period kthread to get an extra
3465 wake_up() if it sleeps three times longer than
3466 it should at force-quiescent-state time.
3467 This wake_up() will be accompanied by a
3468 WARN_ONCE() splat and an ftrace_dump().
3470 rcuperf.gp_async= [KNL]
3471 Measure performance of asynchronous
3472 grace-period primitives such as call_rcu().
3474 rcuperf.gp_async_max= [KNL]
3475 Specify the maximum number of outstanding
3476 callbacks per writer thread. When a writer
3477 thread exceeds this limit, it invokes the
3478 corresponding flavor of rcu_barrier() to allow
3479 previously posted callbacks to drain.
3481 rcuperf.gp_exp= [KNL]
3482 Measure performance of expedited synchronous
3483 grace-period primitives.
3485 rcuperf.holdoff= [KNL]
3486 Set test-start holdoff period. The purpose of
3487 this parameter is to delay the start of the
3488 test until boot completes in order to avoid
3491 rcuperf.nreaders= [KNL]
3492 Set number of RCU readers. The value -1 selects
3493 N, where N is the number of CPUs. A value
3494 "n" less than -1 selects N-n+1, where N is again
3495 the number of CPUs. For example, -2 selects N
3496 (the number of CPUs), -3 selects N+1, and so on.
3497 A value of "n" less than or equal to -N selects
3500 rcuperf.nwriters= [KNL]
3501 Set number of RCU writers. The values operate
3502 the same as for rcuperf.nreaders.
3503 N, where N is the number of CPUs
3505 rcuperf.perf_type= [KNL]
3506 Specify the RCU implementation to test.
3508 rcuperf.shutdown= [KNL]
3509 Shut the system down after performance tests
3510 complete. This is useful for hands-off automated
3513 rcuperf.verbose= [KNL]
3514 Enable additional printk() statements.
3516 rcuperf.writer_holdoff= [KNL]
3517 Write-side holdoff between grace periods,
3518 in microseconds. The default of zero says
3521 rcutorture.cbflood_inter_holdoff= [KNL]
3522 Set holdoff time (jiffies) between successive
3523 callback-flood tests.
3525 rcutorture.cbflood_intra_holdoff= [KNL]
3526 Set holdoff time (jiffies) between successive
3527 bursts of callbacks within a given callback-flood
3530 rcutorture.cbflood_n_burst= [KNL]
3531 Set the number of bursts making up a given
3532 callback-flood test. Set this to zero to
3533 disable callback-flood testing.
3535 rcutorture.cbflood_n_per_burst= [KNL]
3536 Set the number of callbacks to be registered
3537 in a given burst of a callback-flood test.
3539 rcutorture.fqs_duration= [KNL]
3540 Set duration of force_quiescent_state bursts
3543 rcutorture.fqs_holdoff= [KNL]
3544 Set holdoff time within force_quiescent_state bursts
3547 rcutorture.fqs_stutter= [KNL]
3548 Set wait time between force_quiescent_state bursts
3551 rcutorture.gp_cond= [KNL]
3552 Use conditional/asynchronous update-side
3553 primitives, if available.
3555 rcutorture.gp_exp= [KNL]
3556 Use expedited update-side primitives, if available.
3558 rcutorture.gp_normal= [KNL]
3559 Use normal (non-expedited) asynchronous
3560 update-side primitives, if available.
3562 rcutorture.gp_sync= [KNL]
3563 Use normal (non-expedited) synchronous
3564 update-side primitives, if available. If all
3565 of rcutorture.gp_cond=, rcutorture.gp_exp=,
3566 rcutorture.gp_normal=, and rcutorture.gp_sync=
3567 are zero, rcutorture acts as if is interpreted
3568 they are all non-zero.
3570 rcutorture.n_barrier_cbs= [KNL]
3571 Set callbacks/threads for rcu_barrier() testing.
3573 rcutorture.nfakewriters= [KNL]
3574 Set number of concurrent RCU writers. These just
3575 stress RCU, they don't participate in the actual
3576 test, hence the "fake".
3578 rcutorture.nreaders= [KNL]
3579 Set number of RCU readers. The value -1 selects
3580 N-1, where N is the number of CPUs. A value
3581 "n" less than -1 selects N-n-2, where N is again
3582 the number of CPUs. For example, -2 selects N
3583 (the number of CPUs), -3 selects N+1, and so on.
3585 rcutorture.object_debug= [KNL]
3586 Enable debug-object double-call_rcu() testing.
3588 rcutorture.onoff_holdoff= [KNL]
3589 Set time (s) after boot for CPU-hotplug testing.
3591 rcutorture.onoff_interval= [KNL]
3592 Set time (s) between CPU-hotplug operations, or
3593 zero to disable CPU-hotplug testing.
3595 rcutorture.shuffle_interval= [KNL]
3596 Set task-shuffle interval (s). Shuffling tasks
3597 allows some CPUs to go into dyntick-idle mode
3598 during the rcutorture test.
3600 rcutorture.shutdown_secs= [KNL]
3601 Set time (s) after boot system shutdown. This
3602 is useful for hands-off automated testing.
3604 rcutorture.stall_cpu= [KNL]
3605 Duration of CPU stall (s) to test RCU CPU stall
3606 warnings, zero to disable.
3608 rcutorture.stall_cpu_holdoff= [KNL]
3609 Time to wait (s) after boot before inducing stall.
3611 rcutorture.stall_cpu_irqsoff= [KNL]
3612 Disable interrupts while stalling if set.
3614 rcutorture.stat_interval= [KNL]
3615 Time (s) between statistics printk()s.
3617 rcutorture.stutter= [KNL]
3618 Time (s) to stutter testing, for example, specifying
3619 five seconds causes the test to run for five seconds,
3620 wait for five seconds, and so on. This tests RCU's
3621 ability to transition abruptly to and from idle.
3623 rcutorture.test_boost= [KNL]
3624 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
3625 "Maybe" means test if the RCU implementation
3626 under test support RCU priority boosting.
3628 rcutorture.test_boost_duration= [KNL]
3629 Duration (s) of each individual boost test.
3631 rcutorture.test_boost_interval= [KNL]
3632 Interval (s) between each boost test.
3634 rcutorture.test_no_idle_hz= [KNL]
3635 Test RCU's dyntick-idle handling. See also the
3636 rcutorture.shuffle_interval parameter.
3638 rcutorture.torture_type= [KNL]
3639 Specify the RCU implementation to test.
3641 rcutorture.verbose= [KNL]
3642 Enable additional printk() statements.
3644 rcupdate.rcu_cpu_stall_suppress= [KNL]
3645 Suppress RCU CPU stall warning messages.
3647 rcupdate.rcu_cpu_stall_timeout= [KNL]
3648 Set timeout for RCU CPU stall warning messages.
3650 rcupdate.rcu_expedited= [KNL]
3651 Use expedited grace-period primitives, for
3652 example, synchronize_rcu_expedited() instead
3653 of synchronize_rcu(). This reduces latency,
3654 but can increase CPU utilization, degrade
3655 real-time latency, and degrade energy efficiency.
3656 No effect on CONFIG_TINY_RCU kernels.
3658 rcupdate.rcu_normal= [KNL]
3659 Use only normal grace-period primitives,
3660 for example, synchronize_rcu() instead of
3661 synchronize_rcu_expedited(). This improves
3662 real-time latency, CPU utilization, and
3663 energy efficiency, but can expose users to
3664 increased grace-period latency. This parameter
3665 overrides rcupdate.rcu_expedited. No effect on
3666 CONFIG_TINY_RCU kernels.
3668 rcupdate.rcu_normal_after_boot= [KNL]
3669 Once boot has completed (that is, after
3670 rcu_end_inkernel_boot() has been invoked), use
3671 only normal grace-period primitives. No effect
3672 on CONFIG_TINY_RCU kernels.
3674 rcupdate.rcu_task_stall_timeout= [KNL]
3675 Set timeout in jiffies for RCU task stall warning
3676 messages. Disable with a value less than or equal
3679 rcupdate.rcu_self_test= [KNL]
3680 Run the RCU early boot self tests
3682 rcupdate.rcu_self_test_bh= [KNL]
3683 Run the RCU bh early boot self tests
3685 rcupdate.rcu_self_test_sched= [KNL]
3686 Run the RCU sched early boot self tests
3690 Run specified binary instead of /init from the ramdisk,
3691 used for early userspace startup. See initrd.
3694 Turn on/off individual RDT features. List is:
3695 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
3697 E.g. to turn on cmt and turn off mba use:
3701 Format (x86 or x86_64):
3702 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
3704 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
3706 Where reboot_mode is one of warm (soft) or cold (hard) or gpio,
3707 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
3708 reboot_force is either force or not specified,
3709 reboot_cpu is s[mp]#### with #### being the processor
3710 to be used for rebooting.
3713 [KNL, SMP] Set scheduler's default relax_domain_level.
3714 See Documentation/cgroup-v1/cpusets.txt.
3716 reserve= [KNL,BUGS] Force the kernel to ignore some iomem area
3718 reservetop= [X86-32]
3720 Reserves a hole at the top of the kernel virtual
3725 Set the amount of memory to reserve for BIOS at
3726 the bottom of the address space.
3728 reset_devices [KNL] Force drivers to reset the underlying device
3729 during initialization.
3732 Specify the partition device for software suspend
3734 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
3736 resume_offset= [SWSUSP]
3737 Specify the offset from the beginning of the partition
3738 given by "resume=" at which the swap header is located,
3739 in <PAGE_SIZE> units (needed only for swap files).
3740 See Documentation/power/swsusp-and-swap-files.txt
3742 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
3743 read the resume files
3745 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
3746 Useful for devices that are detected asynchronously
3747 (e.g. USB and MMC devices).
3749 hibernate= [HIBERNATION]
3750 noresume Don't check if there's a hibernation image
3751 present during boot.
3752 nocompress Don't compress/decompress hibernation images.
3753 no Disable hibernation and resume.
3754 protect_image Turn on image protection during restoration
3755 (that will set all pages holding image data
3756 during restoration read-only).
3758 retain_initrd [RAM] Keep initrd memory after extraction
3760 rfkill.default_state=
3761 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
3762 etc. communication is blocked by default.
3765 rfkill.master_switch_mode=
3766 0 The "airplane mode" button does nothing.
3767 1 The "airplane mode" button toggles between everything
3768 blocked and the previous configuration.
3769 2 The "airplane mode" button toggles between everything
3770 blocked and everything unblocked.
3772 rhash_entries= [KNL,NET]
3773 Set number of hash buckets for route cache
3776 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
3779 ro [KNL] Mount root device read-only on boot
3782 on Mark read-only kernel memory as read-only (default).
3783 off Leave read-only kernel memory writable for debugging.
3786 Enable the uart passthrough on the designated usb port
3787 on Rockchip SoCs. When active, the signals of the
3788 debug-uart get routed to the D+ and D- pins of the usb
3789 port and the regular usb controller gets disabled.
3791 root= [KNL] Root filesystem
3792 See name_to_dev_t comment in init/do_mounts.c.
3794 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
3795 mount the root filesystem
3797 rootflags= [KNL] Set root filesystem mount option string
3799 rootfstype= [KNL] Set root filesystem type
3801 rootwait [KNL] Wait (indefinitely) for root device to show up.
3802 Useful for devices that are detected asynchronously
3803 (e.g. USB and MMC devices).
3805 rproc_mem=nn[KMG][@address]
3806 [KNL,ARM,CMA] Remoteproc physical memory block.
3807 Memory area to be used by remote processor image,
3810 rw [KNL] Mount root device read-write on boot
3812 S [KNL] Run init in single mode
3814 s390_iommu= [HW,S390]
3815 Set s390 IOTLB flushing mode
3817 With strict flushing every unmap operation will result in
3818 an IOTLB flush. Default is lazy flushing before reuse,
3822 See drivers/net/irda/sa1100_ir.c.
3824 sbni= [NET] Granch SBNI12 leased line adapter
3826 sched_debug [KNL] Enables verbose scheduler debug messages.
3828 schedstats= [KNL,X86] Enable or disable scheduled statistics.
3829 Allowed values are enable and disable. This feature
3830 incurs a small amount of overhead in the scheduler
3831 but is useful for debugging and performance tuning.
3833 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
3834 xtime_lock contention on larger systems, and/or RCU lock
3835 contention on all systems with CONFIG_MAXSMP set.
3836 Format: { "0" | "1" }
3837 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
3839 Note: increases power consumption, thus should only be
3840 enabled if running jitter sensitive (HPC/RT) workloads.
3842 security= [SECURITY] Choose a security module to enable at boot.
3843 If this boot parameter is not specified, only the first
3844 security module asking for security registration will be
3845 loaded. An invalid security module name will be treated
3846 as if no module has been chosen.
3848 selinux= [SELINUX] Disable or enable SELinux at boot time.
3849 Format: { "0" | "1" }
3850 See security/selinux/Kconfig help text.
3853 Default value is set via kernel config option.
3854 If enabled at boot time, /selinux/disable can be used
3855 later to disable prior to initial policy load.
3857 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
3858 Format: { "0" | "1" }
3859 See security/apparmor/Kconfig help text
3862 Default value is set via kernel config option.
3864 serialnumber [BUGS=X86-32]
3867 Maximal number of shapers.
3875 Disable merging of slabs with similar size. May be
3876 necessary if there is some reason to distinguish
3877 allocs to different slabs, especially in hardened
3878 environments where the risk of heap overflows and
3879 layout control by attackers can usually be
3880 frustrated by disabling merging. This will reduce
3881 most of the exposure of a heap attack to a single
3882 cache (risks via metadata attacks are mostly
3883 unchanged). Debug options disable merging on their
3885 For more information see Documentation/vm/slub.txt.
3887 slab_max_order= [MM, SLAB]
3888 Determines the maximum allowed order for slabs.
3889 A high setting may cause OOMs due to memory
3890 fragmentation. Defaults to 1 for systems with
3891 more than 32MB of RAM, 0 otherwise.
3893 slub_debug[=options[,slabs]] [MM, SLUB]
3894 Enabling slub_debug allows one to determine the
3895 culprit if slab objects become corrupted. Enabling
3896 slub_debug can create guard zones around objects and
3897 may poison objects when not in use. Also tracks the
3898 last alloc / free. For more information see
3899 Documentation/vm/slub.txt.
3901 slub_memcg_sysfs= [MM, SLUB]
3902 Determines whether to enable sysfs directories for
3903 memory cgroup sub-caches. 1 to enable, 0 to disable.
3904 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON.
3905 Enabling this can lead to a very high number of debug
3906 directories and files being created under
3909 slub_max_order= [MM, SLUB]
3910 Determines the maximum allowed order for slabs.
3911 A high setting may cause OOMs due to memory
3912 fragmentation. For more information see
3913 Documentation/vm/slub.txt.
3915 slub_min_objects= [MM, SLUB]
3916 The minimum number of objects per slab. SLUB will
3917 increase the slab order up to slub_max_order to
3918 generate a sufficiently large slab able to contain
3919 the number of objects indicated. The higher the number
3920 of objects the smaller the overhead of tracking slabs
3921 and the less frequently locks need to be acquired.
3922 For more information see Documentation/vm/slub.txt.
3924 slub_min_order= [MM, SLUB]
3925 Determines the minimum page order for slabs. Must be
3926 lower than slub_max_order.
3927 For more information see Documentation/vm/slub.txt.
3929 slub_nomerge [MM, SLUB]
3930 Same with slab_nomerge. This is supported for legacy.
3931 See slab_nomerge for more information.
3934 Format: <io1>[,<io2>[,...,<io8>]]
3936 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
3937 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
3938 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
3939 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
3940 smsc-ircc2.ircc_irq= [HW] IRQ line
3941 smsc-ircc2.ircc_dma= [HW] DMA channel
3942 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
3943 0: Toshiba Satellite 1800 (GP data pin select)
3944 1: Fast pin select (default)
3947 smt [KNL,S390] Set the maximum number of threads (logical
3948 CPUs) to use per physical CPU on systems capable of
3949 symmetric multithreading (SMT). Will be capped to the
3950 actual hardware limit.
3952 Default: -1 (no limit)
3955 [KNL] Should the soft-lockup detector generate panics.
3958 A nonzero value instructs the soft-lockup detector
3959 to panic the machine when a soft-lockup occurs. This
3960 is also controlled by CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC
3961 which is the respective build-time switch to that
3964 softlockup_all_cpu_backtrace=
3965 [KNL] Should the soft-lockup detector generate
3966 backtraces on all cpus.
3969 sonypi.*= [HW] Sony Programmable I/O Control Device driver
3970 See Documentation/laptops/sonypi.txt
3972 spectre_v2= [X86] Control mitigation of Spectre variant 2
3973 (indirect branch speculation) vulnerability.
3975 on - unconditionally enable
3976 off - unconditionally disable
3977 auto - kernel detects whether your CPU model is
3980 Selecting 'on' will, and 'auto' may, choose a
3981 mitigation method at run time according to the
3982 CPU, the available microcode, the setting of the
3983 CONFIG_RETPOLINE configuration option, and the
3984 compiler with which the kernel was built.
3986 Specific mitigations can also be selected manually:
3988 retpoline - replace indirect branches
3989 retpoline,generic - google's original retpoline
3990 retpoline,amd - AMD-specific minimal thunk
3992 Not specifying this option is equivalent to
3995 spia_io_base= [HW,MTD]
4000 srcutree.counter_wrap_check [KNL]
4001 Specifies how frequently to check for
4002 grace-period sequence counter wrap for the
4003 srcu_data structure's ->srcu_gp_seq_needed field.
4004 The greater the number of bits set in this kernel
4005 parameter, the less frequently counter wrap will
4006 be checked for. Note that the bottom two bits
4009 srcutree.exp_holdoff [KNL]
4010 Specifies how many nanoseconds must elapse
4011 since the end of the last SRCU grace period for
4012 a given srcu_struct until the next normal SRCU
4013 grace period will be considered for automatic
4014 expediting. Set to zero to disable automatic
4017 stack_guard_gap= [MM]
4018 override the default stack gap protection. The value
4019 is in page units and it defines how many pages prior
4020 to (for stacks growing down) resp. after (for stacks
4021 growing up) the main stack are reserved for no other
4022 mapping. Default value is 256 pages.
4025 Enabled the stack tracer on boot up.
4027 stacktrace_filter=[function-list]
4028 [FTRACE] Limit the functions that the stack tracer
4029 will trace at boot up. function-list is a comma separated
4030 list of functions. This list can be changed at run
4031 time by the stack_trace_filter file in the debugfs
4032 tracing directory. Note, this enables stack tracing
4033 and the stacktrace above is not needed.
4037 Set the STI (builtin display/keyboard on the HP-PARISC
4038 machines) console (graphic card) which should be used
4039 as the initial boot-console.
4040 See also comment in drivers/video/console/sticore.c.
4043 See comment in drivers/video/console/sticore.c.
4046 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
4048 sunrpc.min_resvport=
4049 sunrpc.max_resvport=
4051 SunRPC servers often require that client requests
4052 originate from a privileged port (i.e. a port in the
4053 range 0 < portnr < 1024).
4054 An administrator who wishes to reserve some of these
4055 ports for other uses may adjust the range that the
4056 kernel's sunrpc client considers to be privileged
4057 using these two parameters to set the minimum and
4058 maximum port values.
4060 sunrpc.svc_rpc_per_connection_limit=
4062 Limit the number of requests that the server will
4063 process in parallel from a single connection.
4064 The default value is 0 (no limit).
4068 Control how the NFS server code allocates CPUs to
4069 service thread pools. Depending on how many NICs
4070 you have and where their interrupts are bound, this
4071 option will affect which CPUs will do NFS serving.
4072 Note: this parameter cannot be changed while the
4073 NFS server is running.
4075 auto the server chooses an appropriate mode
4076 automatically using heuristics
4077 global a single global pool contains all CPUs
4078 percpu one pool for each CPU
4079 pernode one pool for each NUMA node (equivalent
4080 to global on non-NUMA machines)
4082 sunrpc.tcp_slot_table_entries=
4083 sunrpc.udp_slot_table_entries=
4085 Sets the upper limit on the number of simultaneous
4086 RPC calls that can be sent from the client to a
4087 server. Increasing these values may allow you to
4088 improve throughput, but will also increase the
4089 amount of memory reserved for use by the client.
4091 suspend.pm_test_delay=
4093 Sets the number of seconds to remain in a suspend test
4094 mode before resuming the system (see
4095 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
4096 is set. Default value is 5.
4099 [KNL] Enable accounting of swap in memory resource
4100 controller if no parameter or 1 is given or disable
4101 it if 0 is given (See Documentation/cgroup-v1/memory.txt)
4103 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
4104 Format: { <int> | force | noforce }
4105 <int> -- Number of I/O TLB slabs
4106 force -- force using of bounce buffers even if they
4107 wouldn't be automatically used by the kernel
4108 noforce -- Never use bounce buffers (for debugging)
4112 sysfs.deprecated=0|1 [KNL]
4113 Enable/disable old style sysfs layout for old udev
4114 on older distributions. When this option is enabled
4115 very new udev will not work anymore. When this option
4116 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
4117 in older udev will not work anymore.
4118 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
4119 the kernel configuration.
4121 sysrq_always_enabled
4123 Ignore sysrq setting - this boot parameter will
4124 neutralize any effect of /proc/sys/kernel/sysrq.
4125 Useful for debugging.
4127 tcpmhash_entries= [KNL,NET]
4128 Set the number of tcp_metrics_hash slots.
4129 Default value is 8192 or 16384 depending on total
4130 ram pages. This is used to specify the TCP metrics
4131 cache size. See Documentation/networking/ip-sysctl.txt
4132 "tcp_no_metrics_save" section for more details.
4136 test_suspend= [SUSPEND][,N]
4137 Specify "mem" (for Suspend-to-RAM) or "standby" (for
4138 standby suspend) or "freeze" (for suspend type freeze)
4139 as the system sleep state during system startup with
4140 the optional capability to repeat N number of times.
4141 The system is woken from this state using a
4142 wakeup-capable RTC alarm.
4144 thash_entries= [KNL,NET]
4145 Set number of hash buckets for TCP connection
4147 thermal.act= [HW,ACPI]
4148 -1: disable all active trip points in all thermal zones
4149 <degrees C>: override all lowest active trip points
4151 thermal.crt= [HW,ACPI]
4152 -1: disable all critical trip points in all thermal zones
4153 <degrees C>: override all critical trip points
4155 thermal.nocrt= [HW,ACPI]
4156 Set to disable actions on ACPI thermal zone
4157 critical and hot trip points.
4159 thermal.off= [HW,ACPI]
4160 1: disable ACPI thermal control
4162 thermal.psv= [HW,ACPI]
4163 -1: disable all passive trip points
4164 <degrees C>: override all passive trip points to this
4167 thermal.tzp= [HW,ACPI]
4168 Specify global default ACPI thermal zone polling rate
4169 <deci-seconds>: poll all this frequency
4170 0: no polling (default)
4173 Force threading of all interrupt handlers except those
4174 marked explicitly IRQF_NO_THREAD.
4177 Enable the Transcendent memory driver if built-in.
4179 tmem.cleancache=0|1 [KNL, XEN]
4180 Default is on (1). Disable the usage of the cleancache
4181 API to send anonymous pages to the hypervisor.
4183 tmem.frontswap=0|1 [KNL, XEN]
4184 Default is on (1). Disable the usage of the frontswap
4185 API to send swap pages to the hypervisor. If disabled
4186 the selfballooning and selfshrinking are force disabled.
4188 tmem.selfballooning=0|1 [KNL, XEN]
4189 Default is on (1). Disable the driving of swap pages
4192 tmem.selfshrinking=0|1 [KNL, XEN]
4193 Default is on (1). Partial swapoff that immediately
4194 transfers pages from Xen hypervisor back to the
4195 kernel based on different criteria.
4199 Specify if the kernel should make use of the cpu
4200 topology information if the hardware supports this.
4201 The scheduler will make use of this information and
4202 e.g. base its process migration decisions on it.
4205 topology_updates= [KNL, PPC, NUMA]
4207 Specify if the kernel should ignore (off)
4208 topology updates sent by the hypervisor to this
4213 tpm_suspend_pcr=[HW,TPM]
4214 Format: integer pcr id
4215 Specify that at suspend time, the tpm driver
4216 should extend the specified pcr with zeros,
4217 as a workaround for some chips which fail to
4218 flush the last written pcr on TPM_SaveState.
4219 This will guarantee that all the other pcrs
4222 trace_buf_size=nn[KMG]
4223 [FTRACE] will set tracing buffer size on each cpu.
4225 trace_event=[event-list]
4226 [FTRACE] Set and start specified trace events in order
4227 to facilitate early boot debugging. The event-list is a
4228 comma separated list of trace events to enable. See
4229 also Documentation/trace/events.txt
4231 trace_options=[option-list]
4232 [FTRACE] Enable or disable tracer options at boot.
4233 The option-list is a comma delimited list of options
4234 that can be enabled or disabled just as if you were
4235 to echo the option name into
4237 /sys/kernel/debug/tracing/trace_options
4239 For example, to enable stacktrace option (to dump the
4240 stack trace of each event), add to the command line:
4242 trace_options=stacktrace
4244 See also Documentation/trace/ftrace.txt "trace options"
4248 Have the tracepoints sent to printk as well as the
4249 tracing ring buffer. This is useful for early boot up
4250 where the system hangs or reboots and does not give the
4251 option for reading the tracing buffer or performing a
4252 ftrace_dump_on_oops.
4254 To turn off having tracepoints sent to printk,
4255 echo 0 > /proc/sys/kernel/tracepoint_printk
4256 Note, echoing 1 into this file without the
4257 tracepoint_printk kernel cmdline option has no effect.
4261 Having tracepoints sent to printk() and activating high
4262 frequency tracepoints such as irq or sched, can cause
4263 the system to live lock.
4266 [FTRACE] enable this option to disable tracing when a
4267 warning is hit. This turns off "tracing_on". Tracing can
4268 be enabled again by echoing '1' into the "tracing_on"
4269 file located in /sys/kernel/debug/tracing/
4271 This option is useful, as it disables the trace before
4272 the WARNING dump is called, which prevents the trace to
4273 be filled with content caused by the warning output.
4275 This option can also be set at run time via the sysctl
4276 option: kernel/traceoff_on_warning
4278 transparent_hugepage=
4280 Format: [always|madvise|never]
4281 Can be used to control the default behavior of the system
4282 with respect to transparent hugepages.
4283 See Documentation/vm/transhuge.txt for more details.
4285 tsc= Disable clocksource stability checks for TSC.
4287 [x86] reliable: mark tsc clocksource as reliable, this
4288 disables clocksource verification at runtime, as well
4289 as the stability checks done at bootup. Used to enable
4290 high-resolution timer mode on older hardware, and in
4291 virtualized environment.
4292 [x86] noirqtime: Do not use TSC to do irq accounting.
4293 Used to run time disable IRQ_TIME_ACCOUNTING on any
4294 platforms where RDTSC is slow and this accounting
4296 [x86] unstable: mark the TSC clocksource as unstable, this
4297 marks the TSC unconditionally unstable at bootup and
4298 avoids any further wobbles once the TSC watchdog notices.
4300 turbografx.map[2|3]= [HW,JOY]
4301 TurboGraFX parallel port interface
4303 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
4304 See also Documentation/input/devices/joystick-parport.rst
4306 udbg-immortal [PPC] When debugging early kernel crashes that
4307 happen after console_init() and before a proper
4308 console driver takes over, this boot options might
4309 help "seeing" what's going on.
4311 uhash_entries= [KNL,NET]
4312 Set number of hash buckets for UDP/UDP-Lite connections
4315 [USB] Ignore overcurrent events (default N).
4316 Some badly-designed motherboards generate lots of
4317 bogus events, for ports that aren't wired to
4318 anything. Set this parameter to avoid log spamming.
4319 Note that genuine overcurrent events won't be
4323 [X86] Cause panic on unknown NMI.
4325 usbcore.authorized_default=
4326 [USB] Default USB device authorization:
4327 (default -1 = authorized except for wireless USB,
4328 0 = not authorized, 1 = authorized)
4330 usbcore.autosuspend=
4331 [USB] The autosuspend time delay (in seconds) used
4332 for newly-detected USB devices (default 2). This
4333 is the time required before an idle device will be
4334 autosuspended. Devices for which the delay is set
4335 to a negative value won't be autosuspended at all.
4337 usbcore.usbfs_snoop=
4338 [USB] Set to log all usbfs traffic (default 0 = off).
4340 usbcore.usbfs_snoop_max=
4341 [USB] Maximum number of bytes to snoop in each URB
4344 usbcore.blinkenlights=
4345 [USB] Set to cycle leds on hubs (default 0 = off).
4347 usbcore.old_scheme_first=
4348 [USB] Start with the old device initialization
4349 scheme (default 0 = off).
4351 usbcore.usbfs_memory_mb=
4352 [USB] Memory limit (in MB) for buffers allocated by
4353 usbfs (default = 16, 0 = max = 2047).
4355 usbcore.use_both_schemes=
4356 [USB] Try the other device initialization scheme
4357 if the first one fails (default 1 = enabled).
4359 usbcore.initial_descriptor_timeout=
4360 [USB] Specifies timeout for the initial 64-byte
4361 USB_REQ_GET_DESCRIPTOR request in milliseconds
4362 (default 5000 = 5.0 seconds).
4364 usbcore.nousb [USB] Disable the USB subsystem
4367 [USBHID] The interval which mice are to be polled at.
4370 [USBHID] The interval which joysticks are to be polled at.
4372 usb-storage.delay_use=
4373 [UMS] The delay in seconds before a new device is
4374 scanned for Logical Units (default 1).
4377 [UMS] A list of quirks entries to supplement or
4378 override the built-in unusual_devs list. List
4379 entries are separated by commas. Each entry has
4380 the form VID:PID:Flags where VID and PID are Vendor
4381 and Product ID values (4-digit hex numbers) and
4382 Flags is a set of characters, each corresponding
4383 to a common usb-storage quirk flag as follows:
4384 a = SANE_SENSE (collect more than 18 bytes
4386 b = BAD_SENSE (don't collect more than 18
4387 bytes of sense data);
4388 c = FIX_CAPACITY (decrease the reported
4389 device capacity by one sector);
4390 d = NO_READ_DISC_INFO (don't use
4391 READ_DISC_INFO command);
4392 e = NO_READ_CAPACITY_16 (don't use
4393 READ_CAPACITY_16 command);
4394 f = NO_REPORT_OPCODES (don't use report opcodes
4396 g = MAX_SECTORS_240 (don't transfer more than
4397 240 sectors at a time, uas only);
4398 h = CAPACITY_HEURISTICS (decrease the
4399 reported device capacity by one
4400 sector if the number is odd);
4401 i = IGNORE_DEVICE (don't bind to this
4403 j = NO_REPORT_LUNS (don't use report luns
4405 l = NOT_LOCKABLE (don't try to lock and
4406 unlock ejectable media);
4407 m = MAX_SECTORS_64 (don't transfer more
4408 than 64 sectors = 32 KB at a time);
4409 n = INITIAL_READ10 (force a retry of the
4410 initial READ(10) command);
4411 o = CAPACITY_OK (accept the capacity
4412 reported by the device);
4413 p = WRITE_CACHE (the device cache is ON
4415 r = IGNORE_RESIDUE (the device reports
4416 bogus residue values);
4417 s = SINGLE_LUN (the device has only one
4419 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
4420 commands, uas only);
4421 u = IGNORE_UAS (don't bind to the uas driver);
4422 w = NO_WP_DETECT (don't test whether the
4423 medium is write-protected).
4424 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
4425 even if the device claims no cache)
4426 Example: quirks=0419:aaf5:rl,0421:0433:rc
4428 user_debug= [KNL,ARM]
4430 See arch/arm/Kconfig.debug help text.
4431 1 - undefined instruction events
4433 4 - invalid data aborts
4436 Example: user_debug=31
4439 [X86] Flags controlling user PTE allocations.
4441 nohigh = do not allocate PTE pages in
4442 HIGHMEM regardless of setting
4446 On X86_32, this is an alias for vdso32=. Otherwise:
4448 vdso=1: enable VDSO (the default)
4449 vdso=0: disable VDSO mapping
4451 vdso32= [X86] Control the 32-bit vDSO
4452 vdso32=1: enable 32-bit VDSO
4453 vdso32=0 or vdso32=2: disable 32-bit VDSO
4455 See the help text for CONFIG_COMPAT_VDSO for more
4456 details. If CONFIG_COMPAT_VDSO is set, the default is
4457 vdso32=0; otherwise, the default is vdso32=1.
4459 For compatibility with older kernels, vdso32=2 is an
4462 Try vdso32=0 if you encounter an error that says:
4463 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
4466 vector=percpu: enable percpu vector domain
4468 video= [FB] Frame buffer configuration
4469 See Documentation/fb/modedb.txt.
4471 video.brightness_switch_enabled= [0,1]
4472 If set to 1, on receiving an ACPI notify event
4473 generated by hotkey, video driver will adjust brightness
4474 level and then send out the event to user space through
4475 the allocated input device; If set to 0, video driver
4476 will only send out the event without touching backlight
4481 [VMMIO] Memory mapped virtio (platform) device.
4483 <size>@<baseaddr>:<irq>[:<id>]
4485 <size> := size (can use standard suffixes
4487 <baseaddr> := physical base address
4488 <irq> := interrupt number (as passed to
4490 <id> := (optional) platform device id
4492 virtio_mmio.device=1K@0x100b0000:48:7
4494 Can be used multiple times for multiple devices.
4496 vga= [BOOT,X86-32] Select a particular video mode
4497 See Documentation/x86/boot.txt and
4498 Documentation/svga.txt.
4499 Use vga=ask for menu.
4500 This is actually a boot loader parameter; the value is
4501 passed to the kernel using a special protocol.
4503 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
4504 size of <nn>. This can be used to increase the
4505 minimum size (128MB on x86). It can also be used to
4506 decrease the size and leave more room for directly
4509 vmcp_cma=nn[MG] [KNL,S390]
4510 Sets the memory size reserved for contiguous memory
4511 allocations for the vmcp device driver.
4513 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
4516 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
4519 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
4523 Controls the behavior of vsyscalls (i.e. calls to
4524 fixed addresses of 0xffffffffff600x00 from legacy
4525 code). Most statically-linked binaries and older
4526 versions of glibc use these calls. Because these
4527 functions are at fixed addresses, they make nice
4528 targets for exploits that can control RIP.
4530 emulate [default] Vsyscalls turn into traps and are
4531 emulated reasonably safely.
4533 native Vsyscalls are native syscall instructions.
4534 This is a little bit faster than trapping
4535 and makes a few dynamic recompilers work
4536 better than they would in emulation mode.
4537 It also makes exploits much easier to write.
4539 none Vsyscalls don't work at all. This makes
4540 them quite hard to use for exploits but
4541 might break your system.
4543 vt.color= [VT] Default text color.
4544 Format: 0xYX, X = foreground, Y = background.
4545 Default: 0x07 = light gray on black.
4547 vt.cur_default= [VT] Default cursor shape.
4548 Format: 0xCCBBAA, where AA, BB, and CC are the same as
4549 the parameters of the <Esc>[?A;B;Cc escape sequence;
4550 see VGA-softcursor.txt. Default: 2 = underline.
4552 vt.default_blu= [VT]
4553 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
4554 Change the default blue palette of the console.
4555 This is a 16-member array composed of values
4558 vt.default_grn= [VT]
4559 Format: <green0>,<green1>,<green2>,...,<green15>
4560 Change the default green palette of the console.
4561 This is a 16-member array composed of values
4564 vt.default_red= [VT]
4565 Format: <red0>,<red1>,<red2>,...,<red15>
4566 Change the default red palette of the console.
4567 This is a 16-member array composed of values
4573 Set system-wide default UTF-8 mode for all tty's.
4574 Default is 1, i.e. UTF-8 mode is enabled for all
4575 newly opened terminals.
4577 vt.global_cursor_default=
4580 Set system-wide default for whether a cursor
4581 is shown on new VTs. Default is -1,
4582 i.e. cursors will be created by default unless
4583 overridden by individual drivers. 0 will hide
4584 cursors, 1 will display them.
4586 vt.italic= [VT] Default color for italic text; 0-15.
4589 vt.underline= [VT] Default color for underlined text; 0-15.
4592 watchdog timers [HW,WDT] For information on watchdog timers,
4593 see Documentation/watchdog/watchdog-parameters.txt
4594 or other driver-specific files in the
4595 Documentation/watchdog/ directory.
4597 workqueue.watchdog_thresh=
4598 If CONFIG_WQ_WATCHDOG is configured, workqueue can
4599 warn stall conditions and dump internal state to
4600 help debugging. 0 disables workqueue stall
4601 detection; otherwise, it's the stall threshold
4602 duration in seconds. The default value is 30 and
4603 it can be updated at runtime by writing to the
4604 corresponding sysfs file.
4606 workqueue.disable_numa
4607 By default, all work items queued to unbound
4608 workqueues are affine to the NUMA nodes they're
4609 issued on, which results in better behavior in
4610 general. If NUMA affinity needs to be disabled for
4611 whatever reason, this option can be used. Note
4612 that this also can be controlled per-workqueue for
4613 workqueues visible under /sys/bus/workqueue/.
4615 workqueue.power_efficient
4616 Per-cpu workqueues are generally preferred because
4617 they show better performance thanks to cache
4618 locality; unfortunately, per-cpu workqueues tend to
4619 be more power hungry than unbound workqueues.
4621 Enabling this makes the per-cpu workqueues which
4622 were observed to contribute significantly to power
4623 consumption unbound, leading to measurably lower
4624 power usage at the cost of small performance
4627 The default value of this parameter is determined by
4628 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
4630 workqueue.debug_force_rr_cpu
4631 Workqueue used to implicitly guarantee that work
4632 items queued without explicit CPU specified are put
4633 on the local CPU. This guarantee is no longer true
4634 and while local CPU is still preferred work items
4635 may be put on foreign CPUs. This debug option
4636 forces round-robin CPU selection to flush out
4637 usages which depend on the now broken guarantee.
4638 When enabled, memory and cache locality will be
4641 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
4642 default x2apic cluster mode on platforms
4645 x86_intel_mid_timer= [X86-32,APBT]
4646 Choose timer option for x86 Intel MID platform.
4647 Two valid options are apbt timer only and lapic timer
4648 plus one apbt timer for broadcast timer.
4649 x86_intel_mid_timer=apbt_only | lapic_and_apbt
4651 xen_512gb_limit [KNL,X86-64,XEN]
4652 Restricts the kernel running paravirtualized under Xen
4653 to use only up to 512 GB of RAM. The reason to do so is
4654 crash analysis tools and Xen tools for doing domain
4655 save/restore/migration must be enabled to handle larger
4658 xen_emul_unplug= [HW,X86,XEN]
4659 Unplug Xen emulated devices
4660 Format: [unplug0,][unplug1]
4661 ide-disks -- unplug primary master IDE devices
4662 aux-ide-disks -- unplug non-primary-master IDE devices
4663 nics -- unplug network devices
4664 all -- unplug all emulated devices (NICs and IDE disks)
4665 unnecessary -- unplugging emulated devices is
4666 unnecessary even if the host did not respond to
4668 never -- do not unplug even if version check succeeds
4670 xen_nopvspin [X86,XEN]
4671 Disables the ticketlock slowpath using Xen PV
4675 Disables the PV optimizations forcing the HVM guest to
4676 run as generic HVM guest with no PV drivers.
4678 xirc2ps_cs= [NET,PCMCIA]
4680 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]