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
117 Support masking of GPEs numbered from 0x00 to 0x7f.
119 acpi_no_auto_serialize [HW,ACPI]
120 Disable auto-serialization of AML methods
121 AML control methods that contain the opcodes to create
122 named objects will be marked as "Serialized" by the
123 auto-serialization feature.
124 This feature is enabled by default.
125 This option allows to turn off the feature.
127 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
130 acpi_no_static_ssdt [HW,ACPI]
131 Disable installation of static SSDTs at early boot time
132 By default, SSDTs contained in the RSDT/XSDT will be
133 installed automatically and they will appear under
134 /sys/firmware/acpi/tables.
135 This option turns off this feature.
136 Note that specifying this option does not affect
137 dynamic table installation which will install SSDT
138 tables to /sys/firmware/acpi/tables/dynamic.
140 acpi_rsdp= [ACPI,EFI,KEXEC]
141 Pass the RSDP address to the kernel, mostly used
142 on machines running EFI runtime service to boot the
143 second kernel for kdump.
145 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
146 Format: To spoof as Windows 98: ="Microsoft Windows"
148 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
149 of 2 which is mandated by ACPI 6) as the supported ACPI
150 specification revision (when using this switch, it may
151 be necessary to carry out a cold reboot _twice_ in a
152 row to make it take effect on the platform firmware).
154 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
155 acpi_osi="string1" # add string1
156 acpi_osi="!string2" # remove string2
157 acpi_osi=!* # remove all strings
158 acpi_osi=! # disable all built-in OS vendor
160 acpi_osi=!! # enable all built-in OS vendor
162 acpi_osi= # disable all strings
164 'acpi_osi=!' can be used in combination with single or
165 multiple 'acpi_osi="string1"' to support specific OS
166 vendor string(s). Note that such command can only
167 affect the default state of the OS vendor strings, thus
168 it cannot affect the default state of the feature group
169 strings and the current state of the OS vendor strings,
170 specifying it multiple times through kernel command line
171 is meaningless. This command is useful when one do not
172 care about the state of the feature group strings which
173 should be controlled by the OSPM.
175 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
176 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
177 can make '_OSI("Windows 2000")' TRUE.
179 'acpi_osi=' cannot be used in combination with other
180 'acpi_osi=' command lines, the _OSI method will not
181 exist in the ACPI namespace. NOTE that such command can
182 only affect the _OSI support state, thus specifying it
183 multiple times through kernel command line is also
186 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
189 'acpi_osi=!*' can be used in combination with single or
190 multiple 'acpi_osi="string1"' to support specific
191 string(s). Note that such command can affect the
192 current state of both the OS vendor strings and the
193 feature group strings, thus specifying it multiple times
194 through kernel command line is meaningful. But it may
195 still not able to affect the final state of a string if
196 there are quirks related to this string. This command
197 is useful when one want to control the state of the
198 feature group strings to debug BIOS issues related to
201 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
202 '_OSI("Module Device")' FALSE.
203 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
204 '_OSI("Module Device")' TRUE.
205 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
207 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
209 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
210 they all will make '_OSI("Windows 2000")' TRUE.
213 Override the pmtimer bug detection: force the kernel
214 to assume that this machine's pmtimer latches its value
215 and always returns good values.
217 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
218 Format: { level | edge | high | low }
220 acpi_skip_timer_override [HW,ACPI]
221 Recognize and ignore IRQ0/pin2 Interrupt Override.
222 For broken nForce2 BIOS resulting in XT-PIC timer.
224 acpi_sleep= [HW,ACPI] Sleep options
225 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
226 old_ordering, nonvs, sci_force_enable }
227 See Documentation/power/video.txt for information on
229 s3_beep is for debugging; it makes the PC's speaker beep
230 as soon as the kernel's real-mode entry point is called.
231 s4_nohwsig prevents ACPI hardware signature from being
232 used during resume from hibernation.
233 old_ordering causes the ACPI 1.0 ordering of the _PTS
234 control method, with respect to putting devices into
235 low power states, to be enforced (the ACPI 2.0 ordering
236 of _PTS is used by default).
237 nonvs prevents the kernel from saving/restoring the
238 ACPI NVS memory during suspend/hibernation and resume.
239 sci_force_enable causes the kernel to set SCI_EN directly
240 on resume from S1/S3 (which is against the ACPI spec,
241 but some broken systems don't work without it).
243 acpi_use_timer_override [HW,ACPI]
244 Use timer override. For some broken Nvidia NF5 boards
245 that require a timer override, but don't have HPET
247 add_efi_memmap [EFI; X86] Include EFI memory map in
248 kernel's map of available physical RAM.
251 { off | try_unsupported }
252 off: disable AGP support
253 try_unsupported: try to drive unsupported chipsets
254 (may crash computer or cause data corruption)
257 See Documentation/sound/alsa/alsa-parameters.txt
260 Allow the default userspace alignment fault handler
261 behaviour to be specified. Bit 0 enables warnings,
262 bit 1 enables fixups, and bit 2 sends a segfault.
264 align_va_addr= [X86-64]
265 Align virtual addresses by clearing slice [14:12] when
266 allocating a VMA at process creation time. This option
267 gives you up to 3% performance improvement on AMD F15h
268 machines (where it is enabled by default) for a
269 CPU-intensive style benchmark, and it can vary highly in
270 a microbenchmark depending on workload and compiler.
272 32: only for 32-bit processes
273 64: only for 64-bit processes
274 on: enable for both 32- and 64-bit processes
275 off: disable for both 32- and 64-bit processes
277 alloc_snapshot [FTRACE]
278 Allocate the ftrace snapshot buffer on boot up when the
279 main buffer is allocated. This is handy if debugging
280 and you need to use tracing_snapshot() on boot up, and
281 do not want to use tracing_snapshot_alloc() as it needs
282 to be done where GFP_KERNEL allocations are allowed.
284 amd_iommu= [HW,X86-64]
285 Pass parameters to the AMD IOMMU driver in the system.
287 fullflush - enable flushing of IO/TLB entries when
288 they are unmapped. Otherwise they are
289 flushed before they will be reused, which
291 off - do not initialize any AMD IOMMU found in
293 force_isolation - Force device isolation for all
294 devices. The IOMMU driver is not
295 allowed anymore to lift isolation
296 requirements as needed. This option
297 does not override iommu=pt
299 amd_iommu_dump= [HW,X86-64]
300 Enable AMD IOMMU driver option to dump the ACPI table
301 for AMD IOMMU. With this option enabled, AMD IOMMU
302 driver will print ACPI tables for AMD IOMMU during
303 IOMMU initialization.
305 amd_iommu_intr= [HW,X86-64]
306 Specifies one of the following AMD IOMMU interrupt
308 legacy - Use legacy interrupt remapping mode.
309 vapic - Use virtual APIC mode, which allows IOMMU
310 to inject interrupts directly into guest.
311 This mode requires kvm-amd.avic=1.
312 (Default when IOMMU HW support is present.)
314 amijoy.map= [HW,JOY] Amiga joystick support
315 Map of devices attached to JOY0DAT and JOY1DAT
317 See also Documentation/input/joystick.txt
319 analog.map= [HW,JOY] Analog joystick and gamepad support
320 Specifies type or capabilities of an analog joystick
321 connected to one of 16 gameports
322 Format: <type1>,<type2>,..<type16>
325 Power management functions (SPARCstation-4/5 + deriv.)
327 Disable APC CPU standby support. SPARCstation-Fox does
328 not play well with APC CPU idle - disable it if you have
329 APC and your system crashes randomly.
331 apic= [APIC,X86-32] Advanced Programmable Interrupt Controller
332 Change the output verbosity whilst booting
333 Format: { quiet (default) | verbose | debug }
334 Change the amount of debugging information output
335 when initialising the APIC and IO-APIC components.
337 apic_extnmi= [APIC,X86] External NMI delivery setting
338 Format: { bsp (default) | all | none }
339 bsp: External NMI is delivered only to CPU 0
340 all: External NMIs are broadcast to all CPUs as a
342 none: External NMI is masked for all CPUs. This is
343 useful so that a dump capture kernel won't be
347 See Documentation/networking/ipv6.txt.
349 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
350 Limit apic dumping. The parameter defines the maximal
351 number of local apics being dumped. Also it is possible
352 to set it to "all" by meaning -- no limit here.
353 Format: { 1 (default) | 2 | ... | all }.
354 The parameter valid if only apic=debug or
355 apic=verbose is specified.
356 Example: apic=debug show_lapic=all
358 apm= [APM] Advanced Power Management
359 See header of arch/x86/kernel/apm_32.c.
361 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
362 Format: <io>,<irq>,<nodeID>
366 atarimouse= [HW,MOUSE] Atari Mouse
368 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
369 EzKey and similar keyboards
371 atkbd.reset= [HW] Reset keyboard during initialization
373 atkbd.set= [HW] Select keyboard code set
374 Format: <int> (2 = AT (default), 3 = PS/2)
376 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
379 atkbd.softraw= [HW] Choose between synthetic and real raw mode
380 Format: <bool> (0 = real, 1 = synthetic (default))
382 atkbd.softrepeat= [HW]
383 Use software keyboard repeat
385 audit= [KNL] Enable the audit sub-system
386 Format: { "0" | "1" } (0 = disabled, 1 = enabled)
387 0 - kernel audit is disabled and can not be enabled
388 until the next reboot
389 unset - kernel audit is initialized but disabled and
390 will be fully enabled by the userspace auditd.
391 1 - kernel audit is initialized and partially enabled,
392 storing at most audit_backlog_limit messages in
393 RAM until it is fully enabled by the userspace
397 audit_backlog_limit= [KNL] Set the audit queue size limit.
398 Format: <int> (must be >=0)
401 bau= [X86_UV] Enable the BAU on SGI UV. The default
402 behavior is to disable the BAU (i.e. bau=0).
403 Format: { "0" | "1" }
406 unset - Disable the BAU.
408 baycom_epp= [HW,AX25]
411 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
413 See header of drivers/net/hamradio/baycom_par.c.
415 baycom_ser_fdx= [HW,AX25]
416 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
417 Format: <io>,<irq>,<mode>[,<baud>]
418 See header of drivers/net/hamradio/baycom_ser_fdx.c.
420 baycom_ser_hdx= [HW,AX25]
421 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
422 Format: <io>,<irq>,<mode>
423 See header of drivers/net/hamradio/baycom_ser_hdx.c.
425 blkdevparts= Manual partition parsing of block device(s) for
426 embedded devices based on command line input.
427 See Documentation/block/cmdline-partition.txt
429 boot_delay= Milliseconds to delay each printk during boot.
430 Values larger than 10 seconds (10000) are changed to
434 bootmem_debug [KNL] Enable bootmem allocator debug messages.
437 Disable BERT OS support on buggy BIOSes.
439 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
440 bttv.radio= Most important insmod options are available as
442 bttv.pll= See Documentation/video4linux/bttv/Insmod-options
445 bulk_remove=off [PPC] This parameter disables the use of the pSeries
446 firmware feature for flushing multiple hpte entries
449 c101= [NET] Moxa C101 synchronous serial card
451 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
452 Sometimes CPU hardware bugs make them report the cache
453 size incorrectly. The kernel will attempt work arounds
454 to fix known problems, but for some CPUs it is not
455 possible to determine what the correct size should be.
456 This option provides an override for these situations.
458 ca_keys= [KEYS] This parameter identifies a specific key(s) on
459 the system trusted keyring to be used for certificate
461 format: { id:<keyid> | builtin }
463 cca= [MIPS] Override the kernel pages' cache coherency
464 algorithm. Accepted values range from 0 to 7
465 inclusive. See arch/mips/include/asm/pgtable-bits.h
466 for platform specific values (SB1, Loongson3 and
469 ccw_timeout_log [S390]
470 See Documentation/s390/CommonIO for details.
472 cgroup_disable= [KNL] Disable a particular controller
473 Format: {name of the controller(s) to disable}
474 The effects of cgroup_disable=foo are:
475 - foo isn't auto-mounted if you mount all cgroups in
477 - foo isn't visible as an individually mountable
479 {Currently only "memory" controller deal with this and
480 cut the overhead, others just disable the usage. So
481 only cgroup_disable=memory is actually worthy}
483 cgroup_no_v1= [KNL] Disable one, multiple, all cgroup controllers in v1
484 Format: { controller[,controller...] | "all" }
485 Like cgroup_disable, but only applies to cgroup v1;
486 the blacklisted controllers remain available in cgroup2.
488 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
490 nosocket -- Disable socket memory accounting.
491 nokmem -- Disable kernel memory accounting.
493 checkreqprot [SELINUX] Set initial checkreqprot flag value.
494 Format: { "0" | "1" }
495 See security/selinux/Kconfig help text.
496 0 -- check protection applied by kernel (includes
497 any implied execute protection).
498 1 -- check protection requested by application.
499 Default value is set via a kernel config option.
500 Value can be changed at runtime via
501 /selinux/checkreqprot.
504 See Documentation/s390/CommonIO for details.
507 Prevents the clock framework from automatically gating
508 clocks that have not been explicitly enabled by a Linux
509 device driver but are enabled in hardware at reset or
510 by the bootloader/firmware. Note that this does not
511 force such clocks to be always-on nor does it reserve
512 those clocks in any way. This parameter is useful for
513 debug and development, but should not be needed on a
514 platform with proper driver support. For more
515 information, see Documentation/clk.txt.
517 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
519 Forces specified clocksource (if available) to be used
520 when calculating gettimeofday(). If specified
521 clocksource is not available, it defaults to PIT.
522 Format: { pit | tsc | cyclone | pmtmr }
524 clocksource= Override the default clocksource
526 Override the default clocksource and use the clocksource
527 with the name specified.
528 Some clocksource names to choose from, depending on
530 [all] jiffies (this is the base, fallback clocksource)
532 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
533 pxa_timer,timer3,32k_counter,timer0_1
535 [X86-32] pit,hpet,tsc;
536 scx200_hrt on Geode; cyclone on IBM x440
544 clocksource.arm_arch_timer.evtstrm=
547 Enable/disable the eventstream feature of the ARM
548 architected timer so that code using WFE-based polling
549 loops can be debugged more effectively on production
552 clearcpuid=BITNUM [X86]
553 Disable CPUID feature X for the kernel. See
554 arch/x86/include/asm/cpufeatures.h for the valid bit
555 numbers. Note the Linux specific bits are not necessarily
556 stable over kernel options, but the vendor specific
558 Also note that user programs calling CPUID directly
559 or using the feature without checking anything
560 will still see it. This just prevents it from
561 being used by the kernel or shown in /proc/cpuinfo.
562 Also note the kernel might malfunction if you disable
565 cma=nn[MG]@[start[MG][-end[MG]]]
567 Sets the size of kernel global memory area for
568 contiguous memory allocations and optionally the
569 placement constraint by the physical address range of
570 memory allocations. A value of 0 disables CMA
571 altogether. For more information, see
572 include/linux/dma-contiguous.h
574 cmo_free_hint= [PPC] Format: { yes | no }
575 Specify whether pages are marked as being inactive
576 when they are freed. This is used in CMO environments
577 to determine OS memory pressure for page stealing by
581 coherent_pool=nn[KMG] [ARM,KNL]
582 Sets the size of memory pool for coherent, atomic dma
583 allocations, by default set to 256K.
585 code_bytes [X86] How many bytes of object code to print
590 com20020= [HW,NET] ARCnet - COM20020 chipset
592 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
594 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
598 ARCnet - COM90xx chipset (memory-mapped buffers)
599 Format: <io>[,<irq>[,<memstart>]]
601 condev= [HW,S390] console device
604 console= [KNL] Output console device and options.
606 tty<n> Use the virtual console device <n>.
610 Use the specified serial port. The options are of
611 the form "bbbbpnf", where "bbbb" is the baud rate,
612 "p" is parity ("n", "o", or "e"), "n" is number of
613 bits, and "f" is flow control ("r" for RTS or
614 omit it). Default is "9600n8".
616 See Documentation/admin-guide/serial-console.rst for more
618 Documentation/networking/netconsole.txt for an
621 uart[8250],io,<addr>[,options]
622 uart[8250],mmio,<addr>[,options]
623 uart[8250],mmio16,<addr>[,options]
624 uart[8250],mmio32,<addr>[,options]
625 uart[8250],0x<addr>[,options]
626 Start an early, polled-mode console on the 8250/16550
627 UART at the specified I/O port or MMIO address,
628 switching to the matching ttyS device later.
629 MMIO inter-register address stride is either 8-bit
630 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
631 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
632 to be equivalent to 'mmio'. 'options' are specified in
633 the same format described for ttyS above; if unspecified,
634 the h/w is not re-initialized.
636 hvc<n> Use the hypervisor console device <n>. This is for
637 both Xen and PowerPC hypervisors.
639 If the device connected to the port is not a TTY but a braille
640 device, prepend "brl," before the device type, for instance
642 For now, only VisioBraille is supported.
644 consoleblank= [KNL] The console blank (screen saver) timeout in
645 seconds. Defaults to 10*60 = 10mins. A value of 0
646 disables the blank timer.
649 [KNL] Change the default value for
650 /proc/<pid>/coredump_filter.
651 See also Documentation/filesystems/proc.txt.
653 cpuidle.off=1 [CPU_IDLE]
654 disable the cpuidle sub-system
656 cpufreq.off=1 [CPU_FREQ]
657 disable the cpufreq sub-system
660 [X86] Delay for N microsec between assert and de-assert
661 of APIC INIT to start processors. This delay occurs
662 on every CPU online, such as boot, and resume from suspend.
665 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
667 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
669 crashkernel=size[KMG][@offset[KMG]]
670 [KNL] Using kexec, Linux can switch to a 'crash kernel'
671 upon panic. This parameter reserves the physical
672 memory region [offset, offset + size] for that kernel
673 image. If '@offset' is omitted, then a suitable offset
674 is selected automatically. Check
675 Documentation/kdump/kdump.txt for further details.
677 crashkernel=range1:size1[,range2:size2,...][@offset]
678 [KNL] Same as above, but depends on the memory
679 in the running system. The syntax of range is
680 start-[end] where start and end are both
681 a memory unit (amount[KMG]). See also
682 Documentation/kdump/kdump.txt for an example.
684 crashkernel=size[KMG],high
685 [KNL, x86_64] range could be above 4G. Allow kernel
686 to allocate physical memory region from top, so could
687 be above 4G if system have more than 4G ram installed.
688 Otherwise memory region will be allocated below 4G, if
690 It will be ignored if crashkernel=X is specified.
691 crashkernel=size[KMG],low
692 [KNL, x86_64] range under 4G. When crashkernel=X,high
693 is passed, kernel could allocate physical memory region
694 above 4G, that cause second kernel crash on system
695 that require some amount of low memory, e.g. swiotlb
696 requires at least 64M+32K low memory, also enough extra
697 low memory is needed to make sure DMA buffers for 32-bit
698 devices won't run out. Kernel would try to allocate at
699 at least 256M below 4G automatically.
700 This one let user to specify own low range under 4G
701 for second kernel instead.
702 0: to disable low allocation.
703 It will be ignored when crashkernel=X,high is not used
704 or memory reserved is below 4G.
707 [KNL] Disable crypto self-tests
712 cs89x0_media= [HW,NET]
713 Format: { rj45 | aui | bnc }
716 See header of drivers/s390/block/dasd_devmap.c.
718 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
719 (one device per port)
720 Format: <port#>,<type>
721 See also Documentation/input/joystick-parport.txt
723 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
724 time. See Documentation/dynamic-debug-howto.txt for
725 details. Deprecated, see dyndbg.
727 debug [KNL] Enable kernel debugging (events log level).
730 [KNL] verbose self-tests
732 Print debugging info while doing the locking API
734 We default to 0 (no extra messages), setting it to
735 1 will print _a lot_ more information - normally
736 only useful to kernel developers.
738 debug_objects [KNL] Enable object debugging
741 [KNL] Disable object debugging
743 debug_guardpage_minorder=
744 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
745 parameter allows control of the order of pages that will
746 be intentionally kept free (and hence protected) by the
747 buddy allocator. Bigger value increase the probability
748 of catching random memory corruption, but reduce the
749 amount of memory for normal system use. The maximum
750 possible value is MAX_ORDER/2. Setting this parameter
751 to 1 or 2 should be enough to identify most random
752 memory corruption problems caused by bugs in kernel or
753 driver code when a CPU writes to (or reads from) a
754 random memory location. Note that there exists a class
755 of memory corruptions problems caused by buggy H/W or
756 F/W or by drivers badly programing DMA (basically when
757 memory is written at bus level and the CPU MMU is
758 bypassed) which are not detectable by
759 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
760 tracking down these problems.
763 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
764 parameter enables the feature at boot time. In
765 default, it is disabled. We can avoid allocating huge
766 chunk of memory for debug pagealloc if we don't enable
767 it at boot time and the system will work mostly same
768 with the kernel built without CONFIG_DEBUG_PAGEALLOC.
769 on: enable the feature
771 debugpat [X86] Enable PAT debugging
773 decnet.addr= [HW,NET]
774 Format: <area>[,<node>]
775 See also Documentation/networking/decnet.txt.
778 [same as hugepagesz=] The size of the default
779 HugeTLB page size. This is the size represented by
780 the legacy /proc/ hugepages APIs, used for SHM, and
781 default size when mounting hugetlbfs filesystems.
782 Defaults to the default architecture's huge page size
786 Set number of hash buckets for dentry cache.
788 disable_1tb_segments [PPC]
789 Disables the use of 1TB hash page table segments. This
790 causes the kernel to fall back to 256MB segments which
791 can be useful when debugging issues that require an SLB
795 See Documentation/networking/ipv6.txt.
798 Disable RADIX MMU mode on POWER9
800 disable_cpu_apicid= [X86,APIC,SMP]
802 The number of initial APIC ID for the
803 corresponding CPU to be disabled at boot,
804 mostly used for the kdump 2nd kernel to
805 disable BSP to wake up multiple CPUs without
806 causing system reset or hang due to sending
809 disable_ddw [PPC/PSERIES]
810 Disable Dynamic DMA Window support. Use this if
811 to workaround buggy firmware.
814 See Documentation/networking/ipv6.txt.
816 disable_mtrr_cleanup [X86]
817 The kernel tries to adjust MTRR layout from continuous
818 to discrete, to make X server driver able to add WB
819 entry later. This parameter disables that.
821 disable_mtrr_trim [X86, Intel and AMD only]
822 By default the kernel will trim any uncacheable
823 memory out of your available memory pool based on
824 MTRR settings. This parameter disables that behavior,
825 possibly causing your machine to run very slowly.
827 disable_timer_pin_1 [X86]
828 Disable PIN 1 of APIC timer
829 Can be useful to work around chipset bugs.
831 dis_ucode_ldr [X86] Disable the microcode loader.
833 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
834 this option disables the debugging code at boot.
836 dma_debug_entries=<number>
837 This option allows to tune the number of preallocated
838 entries for DMA-API debugging code. One entry is
839 required per DMA-API allocation. Use this if the
840 DMA-API debugging code disables itself because the
841 architectural default is too low.
843 dma_debug_driver=<driver_name>
844 With this option the DMA-API debugging driver
845 filter feature can be enabled at boot time. Just
846 pass the driver to filter for as the parameter.
847 The filter can be disabled or changed to another
848 driver later using sysfs.
850 drm_kms_helper.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
851 Broken monitors, graphic adapters, KVMs and EDIDless
852 panels may send no or incorrect EDID data sets.
853 This parameter allows to specify an EDID data sets
854 in the /lib/firmware directory that are used instead.
855 Generic built-in EDID data sets are used, if one of
856 edid/1024x768.bin, edid/1280x1024.bin,
857 edid/1680x1050.bin, or edid/1920x1080.bin is given
858 and no file with the same name exists. Details and
859 instructions how to build your own EDID data are
860 available in Documentation/EDID/HOWTO.txt. An EDID
861 data set will only be used for a particular connector,
862 if its name and a colon are prepended to the EDID
863 name. Each connector may use a unique EDID data
864 set by separating the files with a comma. An EDID
865 data set with no connector name will be used for
866 any connectors not explicitly specified.
870 dump_apple_properties [X86]
871 Dump name and content of EFI device properties on
872 x86 Macs. Useful for driver authors to determine
873 what data is available or for reverse-engineering.
875 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
876 module.dyndbg[="val"]
877 Enable debug messages at boot time. See
878 Documentation/dynamic-debug-howto.txt for details.
880 nompx [X86] Disables Intel Memory Protection Extensions.
881 See Documentation/x86/intel_mpx.txt for more
882 information about the feature.
884 nopku [X86] Disable Memory Protection Keys CPU feature found
887 module.async_probe [KNL]
888 Enable asynchronous probe on this module.
890 early_ioremap_debug [KNL]
891 Enable debug messages in early_ioremap support. This
892 is useful for tracking down temporary early mappings
893 which are not unmapped.
895 earlycon= [KNL] Output early console device and options.
897 When used with no options, the early console is
898 determined by the stdout-path property in device
901 cdns,<addr>[,options]
902 Start an early, polled-mode console on a Cadence
903 (xuartps) serial port at the specified address. Only
904 supported option is baud rate. If baud rate is not
905 specified, the serial port must already be setup and
908 uart[8250],io,<addr>[,options]
909 uart[8250],mmio,<addr>[,options]
910 uart[8250],mmio32,<addr>[,options]
911 uart[8250],mmio32be,<addr>[,options]
912 uart[8250],0x<addr>[,options]
913 Start an early, polled-mode console on the 8250/16550
914 UART at the specified I/O port or MMIO address.
915 MMIO inter-register address stride is either 8-bit
916 (mmio) or 32-bit (mmio32 or mmio32be).
917 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
918 to be equivalent to 'mmio'. 'options' are specified
919 in the same format described for "console=ttyS<n>"; if
920 unspecified, the h/w is not initialized.
924 Start an early, polled-mode console on a pl011 serial
925 port at the specified address. The pl011 serial port
926 must already be setup and configured. Options are not
927 yet supported. If 'mmio32' is specified, then only
928 the driver will use only 32-bit accessors to read/write
929 the device registers.
932 Start an early, polled-mode console on a meson serial
933 port at the specified address. The serial port must
934 already be setup and configured. Options are not yet
938 Start an early, polled-mode console on an msm serial
939 port at the specified address. The serial port
940 must already be setup and configured. Options are not
944 Start an early, polled-mode console on an msm serial
945 dm port at the specified address. The serial port
946 must already be setup and configured. Options are not
949 smh Use ARM semihosting calls for early console.
957 Use early console provided by serial driver available
958 on Samsung SoCs, requires selecting proper type and
959 a correct base address of the selected UART port. The
960 serial port must already be setup and configured.
961 Options are not yet supported.
964 Start an early, polled-mode console on a lantiq serial
965 (lqasc) port at the specified address. The serial port
966 must already be setup and configured. Options are not
971 Use early console provided by Freescale LP UART driver
972 found on Freescale Vybrid and QorIQ LS1021A processors.
973 A valid base address must be provided, and the serial
974 port must already be setup and configured.
976 armada3700_uart,<addr>
977 Start an early, polled-mode console on the
978 Armada 3700 serial port at the specified
979 address. The serial port must already be setup
980 and configured. Options are not yet supported.
982 earlyprintk= [X86,SH,BLACKFIN,ARM,M68k,S390]
987 earlyprintk=serial[,ttySn[,baudrate]]
988 earlyprintk=serial[,0x...[,baudrate]]
989 earlyprintk=ttySn[,baudrate]
990 earlyprintk=dbgp[debugController#]
991 earlyprintk=pciserial,bus:device.function[,baudrate]
993 earlyprintk is useful when the kernel crashes before
994 the normal console is initialized. It is not enabled by
995 default because it has some cosmetic problems.
997 Append ",keep" to not disable it when the real console
1000 Only one of vga, efi, serial, or usb debug port can
1003 Currently only ttyS0 and ttyS1 may be specified by
1004 name. Other I/O ports may be explicitly specified
1005 on some architectures (x86 and arm at least) by
1006 replacing ttySn with an I/O port address, like this:
1007 earlyprintk=serial,0x1008,115200
1008 You can find the port for a given device in
1009 /proc/tty/driver/serial:
1010 2: uart:ST16650V2 port:00001008 irq:18 ...
1012 Interaction with the standard serial driver is not
1015 The VGA and EFI output is eventually overwritten by
1018 The xen output can only be used by Xen PV guests.
1020 The sclp output can only be used on s390.
1022 edac_report= [HW,EDAC] Control how to report EDAC event
1023 Format: {"on" | "off" | "force"}
1024 on: enable EDAC to report H/W event. May be overridden
1025 by other higher priority error reporting module.
1026 off: disable H/W event reporting through EDAC.
1027 force: enforce the use of EDAC to report H/W event.
1030 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1033 This is designed to be used in conjunction with
1034 the boot argument: earlyprintk=vga
1037 Format: {"off" | "on" | "skip[mbr]"}
1040 Format: { "old_map", "nochunk", "noruntime", "debug" }
1041 old_map [X86-64]: switch to the old ioremap-based EFI
1042 runtime services mapping. 32-bit still uses this one by
1044 nochunk: disable reading files in "chunks" in the EFI
1045 boot stub, as chunking can cause problems with some
1046 firmware implementations.
1047 noruntime : disable EFI runtime services support
1048 debug: enable misc debug output
1050 efi_no_storage_paranoia [EFI; X86]
1051 Using this parameter you can use more than 50% of
1052 your efi variable storage. Use this parameter only if
1053 you are really sure that your UEFI does sane gc and
1054 fulfills the spec otherwise your board may brick.
1056 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1057 Add arbitrary attribute to specific memory range by
1058 updating original EFI memory map.
1059 Region of memory which aa attribute is added to is
1061 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1062 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1063 attribute is added to range 0x100000000-0x180000000 and
1064 0x10a0000000-0x1120000000.
1066 Using this parameter you can do debugging of EFI memmap
1067 related feature. For example, you can do debugging of
1068 Address Range Mirroring feature even if your box
1071 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1072 that is to be dynamically loaded by Linux. If there are
1073 multiple variables with the same name but with different
1074 vendor GUIDs, all of them will be loaded. See
1075 Documentation/acpi/ssdt-overlays.txt for details.
1078 eisa_irq_edge= [PARISC,HW]
1079 See header of drivers/parisc/eisa.c.
1082 See comment before function elanfreq_setup() in
1083 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1086 Format: {"cfq" | "deadline" | "noop"}
1087 See Documentation/block/cfq-iosched.txt and
1088 Documentation/block/deadline-iosched.txt for details.
1090 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1091 Specifies physical address of start of kernel core
1092 image elf header and optionally the size. Generally
1093 kexec loader will pass this option to capture kernel.
1094 See Documentation/kdump/kdump.txt for details.
1096 enable_mtrr_cleanup [X86]
1097 The kernel tries to adjust MTRR layout from continuous
1098 to discrete, to make X server driver able to add WB
1099 entry later. This parameter enables that.
1101 enable_timer_pin_1 [X86]
1102 Enable PIN 1 of APIC timer
1103 Can be useful to work around chipset bugs
1104 (in particular on some ATI chipsets).
1105 The kernel tries to set a reasonable default.
1107 enforcing [SELINUX] Set initial enforcing status.
1109 See security/selinux/Kconfig help text.
1110 0 -- permissive (log only, no denials).
1111 1 -- enforcing (deny and log).
1113 Value can be changed at runtime via /selinux/enforce.
1116 Disable Error Record Serialization Table (ERST)
1119 ether= [HW,NET] Ethernet cards parameters
1120 This option is obsoleted by the "netdev=" option, which
1121 has equivalent usage. See its documentation for details.
1125 Permit 'security.evm' to be updated regardless of
1126 current integrity status.
1130 fail_make_request=[KNL]
1131 General fault injection mechanism.
1132 Format: <interval>,<probability>,<space>,<times>
1133 See also Documentation/fault-injection/.
1136 See Documentation/blockdev/floppy.txt.
1138 force_pal_cache_flush
1139 [IA-64] Avoid check_sal_cache_flush which may hang on
1140 buggy SAL_CACHE_FLUSH implementations. Using this
1141 parameter will force ia64_sal_cache_flush to call
1142 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1145 Forcefully enable Physical Address Extension (PAE).
1146 Many Pentium M systems disable PAE but may have a
1147 functionally usable PAE implementation.
1148 Warning: use of this parameter will taint the kernel
1149 and may cause unknown problems.
1152 [FTRACE] will set and start the specified tracer
1153 as early as possible in order to facilitate early
1156 ftrace_dump_on_oops[=orig_cpu]
1157 [FTRACE] will dump the trace buffers on oops.
1158 If no parameter is passed, ftrace will dump
1159 buffers of all CPUs, but if you pass orig_cpu, it will
1160 dump only the buffer of the CPU that triggered the
1163 ftrace_filter=[function-list]
1164 [FTRACE] Limit the functions traced by the function
1165 tracer at boot up. function-list is a comma separated
1166 list of functions. This list can be changed at run
1167 time by the set_ftrace_filter file in the debugfs
1170 ftrace_notrace=[function-list]
1171 [FTRACE] Do not trace the functions specified in
1172 function-list. This list can be changed at run time
1173 by the set_ftrace_notrace file in the debugfs
1176 ftrace_graph_filter=[function-list]
1177 [FTRACE] Limit the top level callers functions traced
1178 by the function graph tracer at boot up.
1179 function-list is a comma separated list of functions
1180 that can be changed at run time by the
1181 set_graph_function file in the debugfs tracing directory.
1183 ftrace_graph_notrace=[function-list]
1184 [FTRACE] Do not trace from the functions specified in
1185 function-list. This list is a comma separated list of
1186 functions that can be changed at run time by the
1187 set_graph_notrace file in the debugfs tracing directory.
1189 ftrace_graph_max_depth=<uint>
1190 [FTRACE] Used with the function graph tracer. This is
1191 the max depth it will trace into a function. This value
1192 can be changed at run time by the max_graph_depth file
1193 in the tracefs tracing directory. default: 0 (no limit)
1196 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1197 support via parallel port (up to 5 devices per port)
1198 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1199 See also Documentation/input/joystick-parport.txt
1203 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1207 gcov_persist= [GCOV] When non-zero (default), profiling data for
1208 kernel modules is saved and remains accessible via
1209 debugfs, even when the module is unloaded/reloaded.
1210 When zero, profiling data is discarded and associated
1211 debugfs files are removed at module unload time.
1213 goldfish [X86] Enable the goldfish android emulator platform.
1214 Don't use this when you are not running on the
1217 gpt [EFI] Forces disk with valid GPT signature but
1218 invalid Protective MBR to be treated as GPT. If the
1219 primary GPT is corrupted, it enables the backup/alternate
1220 GPT to be used instead.
1222 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1223 the "Enable 0" bit of the configuration register.
1226 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1227 the "Enable 0" bit of the configuration register.
1230 grcan.select= [HW] Select which physical interface to use.
1233 grcan.txsize= [HW] Sets the size of the tx buffer.
1234 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1236 grcan.rxsize= [HW] Sets the size of the rx buffer.
1237 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1240 gpio-mockup.gpio_mockup_ranges
1241 [HW] Sets the ranges of gpiochip of for this device.
1242 Format: <start1>,<end1>,<start2>,<end2>...
1244 hardlockup_all_cpu_backtrace=
1245 [KNL] Should the hard-lockup detector generate
1246 backtraces on all cpus.
1249 hashdist= [KNL,NUMA] Large hashes allocated during boot
1250 are distributed across NUMA nodes. Defaults on
1251 for 64-bit NUMA, off otherwise.
1252 Format: 0 | 1 (for off | on)
1254 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1256 hd= [EIDE] (E)IDE hard drive subsystem geometry
1257 Format: <cyl>,<head>,<sect>
1260 Disable Hardware Error Source Table (HEST) support;
1261 corresponding firmware-first mode error processing
1262 logic will be disabled.
1264 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1265 size of <nn>. This works even on boxes that have no
1266 highmem otherwise. This also works to reduce highmem
1267 size on bigger boxes.
1269 highres= [KNL] Enable/disable high resolution timer mode.
1270 Valid parameters: "on", "off"
1274 See Documentation/isdn/README.HiSax.
1278 hpet= [X86-32,HPET] option to control HPET usage
1279 Format: { enable (default) | disable | force |
1281 disable: disable HPET and use PIT instead
1282 force: allow force enabled of undocumented chips (ICH4,
1284 verbose: show contents of HPET registers during setup
1286 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1287 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1289 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1290 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
1291 On x86-64 and powerpc, this option can be specified
1292 multiple times interleaved with hugepages= to reserve
1293 huge pages of different sizes. Valid pages sizes on
1294 x86-64 are 2M (when the CPU supports "pse") and 1G
1295 (when the CPU supports the "pdpe1gb" cpuinfo flag).
1297 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1298 terminal devices. Valid values: 0..8
1299 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1300 If specified, z/VM IUCV HVC accepts connections
1301 from listed z/VM user IDs only.
1303 hwthread_map= [METAG] Comma-separated list of Linux cpu id to
1304 hardware thread id mappings.
1305 Format: <cpu>:<hwthread>
1308 Do not unregister boot console at start. This is only
1309 useful for debugging when something happens in the window
1310 between unregistering the boot console and initializing
1313 i2c_bus= [HW] Override the default board specific I2C bus speed
1314 or register an additional I2C bus that is not
1315 registered from board initialization code.
1319 i8042.debug [HW] Toggle i8042 debug mode
1320 i8042.unmask_kbd_data
1321 [HW] Enable printing of interrupt data from the KBD port
1322 (disabled by default, and as a pre-condition
1323 requires that i8042.debug=1 be enabled)
1324 i8042.direct [HW] Put keyboard port into non-translated mode
1325 i8042.dumbkbd [HW] Pretend that controller can only read data from
1326 keyboard and cannot control its state
1327 (Don't attempt to blink the leds)
1328 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1329 i8042.nokbd [HW] Don't check/create keyboard port
1330 i8042.noloop [HW] Disable the AUX Loopback command while probing
1332 i8042.nomux [HW] Don't check presence of an active multiplexing
1334 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1336 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1337 i8042.reset [HW] Reset the controller during init, cleanup and
1338 suspend-to-ram transitions, only during s2r
1339 transitions, or never reset
1340 Format: { 1 | Y | y | 0 | N | n }
1341 1, Y, y: always reset controller
1342 0, N, n: don't ever reset controller
1343 Default: only on s2r transitions on x86; most other
1344 architectures force reset to be always executed
1345 i8042.unlock [HW] Unlock (ignore) the keylock
1346 i8042.kbdreset [HW] Reset device connected to KBD port
1350 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1351 indicates that the driver is running on unsupported
1353 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1354 does not match list of supported models.
1356 [HW] Report power status in /proc/i8k
1357 (disabled by default)
1358 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1361 i915.invert_brightness=
1362 [DRM] Invert the sense of the variable that is used to
1363 set the brightness of the panel backlight. Normally a
1364 brightness value of 0 indicates backlight switched off,
1365 and the maximum of the brightness value sets the backlight
1366 to maximum brightness. If this parameter is set to 0
1367 (default) and the machine requires it, or this parameter
1368 is set to 1, a brightness value of 0 sets the backlight
1369 to maximum brightness, and the maximum of the brightness
1370 value switches the backlight off.
1371 -1 -- never invert brightness
1372 0 -- machine default
1373 1 -- force brightness inversion
1376 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1378 ide-core.nodma= [HW] (E)IDE subsystem
1379 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1380 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1381 .cdrom .chs .ignore_cable are additional options
1382 See Documentation/ide/ide.txt.
1384 ide-generic.probe-mask= [HW] (E)IDE subsystem
1386 Probe mask for legacy ISA IDE ports. Depending on
1387 platform up to 6 ports are supported, enabled by
1388 setting corresponding bits in the mask to 1. The
1389 default value is 0x0, which has a special meaning.
1390 On systems that have PCI, it triggers scanning the
1391 PCI bus for the first and the second port, which
1392 are then probed. On systems without PCI the value
1393 of 0x0 enables probing the two first ports as if it
1396 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1397 Claim all unknown PCI IDE storage controllers.
1400 Format: idle=poll, idle=halt, idle=nomwait
1401 Poll forces a polling idle loop that can slightly
1402 improve the performance of waking up a idle CPU, but
1403 will use a lot of power and make the system run hot.
1405 idle=halt: Halt is forced to be used for CPU idle.
1406 In such case C2/C3 won't be used again.
1407 idle=nomwait: Disable mwait for CPU C-states
1409 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1410 Format: { strict | legacy | 2008 | relaxed }
1413 Choose which programs will be accepted for execution
1414 based on the IEEE 754 NaN encoding(s) supported by
1415 the FPU and the NaN encoding requested with the value
1416 of an ELF file header flag individually set by each
1417 binary. Hardware implementations are permitted to
1418 support either or both of the legacy and the 2008 NaN
1421 Available settings are as follows:
1422 strict accept binaries that request a NaN encoding
1423 supported by the FPU
1424 legacy only accept legacy-NaN binaries, if supported
1426 2008 only accept 2008-NaN binaries, if supported
1428 relaxed accept any binaries regardless of whether
1429 supported by the FPU
1431 The FPU emulator is always able to support both NaN
1432 encodings, so if no FPU hardware is present or it has
1433 been disabled with 'nofpu', then the settings of
1434 'legacy' and '2008' strap the emulator accordingly,
1435 'relaxed' straps the emulator for both legacy-NaN and
1436 2008-NaN, whereas 'strict' enables legacy-NaN only on
1437 legacy processors and both NaN encodings on MIPS32 or
1440 The setting for ABS.fmt/NEG.fmt instruction execution
1441 mode generally follows that for the NaN encoding,
1442 except where unsupported by hardware.
1444 ignore_loglevel [KNL]
1445 Ignore loglevel setting - this will print /all/
1446 kernel messages to the console. Useful for debugging.
1447 We also add it as printk module parameter, so users
1448 could change it dynamically, usually by
1449 /sys/module/printk/parameters/ignore_loglevel.
1452 Ignore RLIMIT_DATA setting for data mappings,
1453 print warning at first misuse. Can be changed via
1454 /sys/module/kernel/parameters/ignore_rlimit_data.
1456 ihash_entries= [KNL]
1457 Set number of hash buckets for inode cache.
1459 ima_appraise= [IMA] appraise integrity measurements
1460 Format: { "off" | "enforce" | "fix" | "log" }
1463 ima_appraise_tcb [IMA]
1464 The builtin appraise policy appraises all files
1467 ima_canonical_fmt [IMA]
1468 Use the canonical format for the binary runtime
1469 measurements, instead of host native format.
1472 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1476 The list of supported hash algorithms is defined
1477 in crypto/hash_info.h.
1480 The builtin measurement policy to load during IMA
1481 setup. Specyfing "tcb" as the value, measures all
1482 programs exec'd, files mmap'd for exec, and all files
1483 opened with the read mode bit set by either the
1484 effective uid (euid=0) or uid=0.
1487 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1488 Load a policy which meets the needs of the Trusted
1489 Computing Base. This means IMA will measure all
1490 programs exec'd, files mmap'd for exec, and all files
1491 opened for read by uid=0.
1494 Select one of defined IMA measurements template formats.
1495 Formats: { "ima" | "ima-ng" | "ima-sig" }
1499 [IMA] Define a custom template format.
1500 Format: { "field1|...|fieldN" }
1502 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1503 Format: <min_file_size>
1504 Set the minimal file size for using asynchronous hash.
1505 If left unspecified, ahash usage is disabled.
1507 ahash performance varies for different data sizes on
1508 different crypto accelerators. This option can be used
1509 to achieve the best performance for a particular HW.
1511 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1513 Set hashing buffer size. Default: 4k.
1515 ahash performance varies for different chunk sizes on
1516 different crypto accelerators. This option can be used
1517 to achieve best performance for particular HW.
1521 Run specified binary instead of /sbin/init as init
1524 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1525 for working out where the kernel is dying during
1528 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1529 initcall functions. Useful for debugging built-in
1530 modules and initcalls.
1532 initrd= [BOOT] Specify the location of the initial ramdisk
1534 init_pkru= [x86] Specify the default memory protection keys rights
1535 register contents for all processes. 0x55555554 by
1536 default (disallow access to all but pkey 0). Can
1537 override in debugfs after boot.
1539 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1542 int_pln_enable [x86] Enable power limit notification interrupt
1544 integrity_audit=[IMA]
1545 Format: { "0" | "1" }
1546 0 -- basic integrity auditing messages. (Default)
1547 1 -- additional integrity auditing messages.
1549 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1551 Enable intel iommu driver.
1553 Disable intel iommu driver.
1554 igfx_off [Default Off]
1555 By default, gfx is mapped as normal device. If a gfx
1556 device has a dedicated DMAR unit, the DMAR unit is
1557 bypassed by not enabling DMAR with this option. In
1558 this case, gfx device will use physical address for
1561 With this option iommu will not optimize to look
1562 for io virtual address below 32-bit forcing dual
1563 address cycle on pci bus for cards supporting greater
1564 than 32-bit addressing. The default is to look
1565 for translation below 32-bit and if not available
1566 then look in the higher range.
1567 strict [Default Off]
1568 With this option on every unmap_single operation will
1569 result in a hardware IOTLB flush operation as opposed
1570 to batching them for performance.
1571 sp_off [Default Off]
1572 By default, super page will be supported if Intel IOMMU
1573 has the capability. With this option, super page will
1575 ecs_off [Default Off]
1576 By default, extended context tables will be supported if
1577 the hardware advertises that it has support both for the
1578 extended tables themselves, and also PASID support. With
1579 this option set, extended tables will not be used even
1580 on hardware which claims to support them.
1582 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1583 0 disables intel_idle and fall back on acpi_idle.
1584 1 to 9 specify maximum depth of C-state.
1588 Do not enable intel_pstate as the default
1589 scaling driver for the supported processors
1591 Use intel_pstate as a scaling driver, but configure it
1592 to work with generic cpufreq governors (instead of
1593 enabling its internal governor). This mode cannot be
1594 used along with the hardware-managed P-states (HWP)
1597 Enable intel_pstate on systems that prohibit it by default
1598 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1599 instead of acpi-cpufreq may disable platform features, such
1600 as thermal controls and power capping, that rely on ACPI
1601 P-States information being indicated to OSPM and therefore
1602 should be used with caution. This option does not work with
1603 processors that aren't supported by the intel_pstate driver
1604 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1606 Do not enable hardware P state control (HWP)
1609 Only load intel_pstate on systems which support
1610 hardware P state control (HWP) if available.
1612 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1613 Description Table, specifies preferred power management
1614 profile as "Enterprise Server" or "Performance Server",
1615 then this feature is turned on by default.
1617 Allow per-logical-CPU P-State performance control limits using
1618 cpufreq sysfs interface
1620 intremap= [X86-64, Intel-IOMMU]
1621 on enable Interrupt Remapping (default)
1622 off disable Interrupt Remapping
1623 nosid disable Source ID checking
1625 BIOS x2APIC opt-out request will be ignored
1626 nopost disable Interrupt Posting
1628 iomem= Disable strict checking of access to MMIO memory
1629 strict regions from userspace.
1644 nobypass [PPC/POWERNV]
1645 Disable IOMMU bypass, using IOMMU for PCI devices.
1648 io7= [HW] IO7 for Marvel based alpha systems
1649 See comment before marvel_specify_io7 in
1650 arch/alpha/kernel/core_marvel.c.
1652 io_delay= [X86] I/O delay method
1654 Standard port 0x80 based delay
1656 Alternate port 0xed based delay (needed on some systems)
1658 Simple two microseconds delay
1663 See Documentation/filesystems/nfs/nfsroot.txt.
1665 irqaffinity= [SMP] Set the default irq affinity mask
1666 The argument is a cpu list, as described above.
1669 When an interrupt is not handled search all handlers
1670 for it. Intended to get systems with badly broken
1674 When an interrupt is not handled search all handlers
1675 for it. Also check all handlers each timer
1676 interrupt. Intended to get systems with badly broken
1680 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1682 isolcpus= [KNL,SMP] Isolate CPUs from the general scheduler.
1683 The argument is a cpu list, as described above.
1685 This option can be used to specify one or more CPUs
1686 to isolate from the general SMP balancing and scheduling
1687 algorithms. You can move a process onto or off an
1688 "isolated" CPU via the CPU affinity syscalls or cpuset.
1689 <cpu number> begins at 0 and the maximum value is
1690 "number of CPUs in system - 1".
1692 This option is the preferred way to isolate CPUs. The
1693 alternative -- manually setting the CPU mask of all
1694 tasks in the system -- can cause problems and
1695 suboptimal load balancer performance.
1699 ivrs_ioapic [HW,X86_64]
1700 Provide an override to the IOAPIC-ID<->DEVICE-ID
1701 mapping provided in the IVRS ACPI table. For
1702 example, to map IOAPIC-ID decimal 10 to
1703 PCI device 00:14.0 write the parameter as:
1704 ivrs_ioapic[10]=00:14.0
1706 ivrs_hpet [HW,X86_64]
1707 Provide an override to the HPET-ID<->DEVICE-ID
1708 mapping provided in the IVRS ACPI table. For
1709 example, to map HPET-ID decimal 0 to
1710 PCI device 00:14.0 write the parameter as:
1711 ivrs_hpet[0]=00:14.0
1713 ivrs_acpihid [HW,X86_64]
1714 Provide an override to the ACPI-HID:UID<->DEVICE-ID
1715 mapping provided in the IVRS ACPI table. For
1716 example, to map UART-HID:UID AMD0020:0 to
1717 PCI device 00:14.5 write the parameter as:
1718 ivrs_acpihid[00:14.5]=AMD0020:0
1720 js= [HW,JOY] Analog joystick
1721 See Documentation/input/joystick.txt.
1724 When CONFIG_RANDOMIZE_BASE is set, this disables
1725 kernel and module base offset ASLR (Address Space
1726 Layout Randomization).
1729 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
1730 report on every invalid memory access. Without this
1731 parameter KASAN will print report only for the first
1736 kernelcore= [KNL,X86,IA-64,PPC]
1737 Format: nn[KMGTPE] | "mirror"
1739 specifies the amount of memory usable by the kernel
1740 for non-movable allocations. The requested amount is
1741 spread evenly throughout all nodes in the system. The
1742 remaining memory in each node is used for Movable
1743 pages. In the event, a node is too small to have both
1744 kernelcore and Movable pages, kernelcore pages will
1745 take priority and other nodes will have a larger number
1746 of Movable pages. The Movable zone is used for the
1747 allocation of pages that may be reclaimed or moved
1748 by the page migration subsystem. This means that
1749 HugeTLB pages may not be allocated from this zone.
1750 Note that allocations like PTEs-from-HighMem still
1751 use the HighMem zone if it exists, and the Normal
1752 zone if it does not.
1754 Instead of specifying the amount of memory (nn[KMGTPE]),
1755 you can specify "mirror" option. In case "mirror"
1756 option is specified, mirrored (reliable) memory is used
1757 for non-movable allocations and remaining memory is used
1758 for Movable pages. nn[KMGTPE] and "mirror" are exclusive,
1759 so you can NOT specify nn[KMGTPE] and "mirror" at the same
1762 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
1763 Format: <Controller#>[,poll interval]
1764 The controller # is the number of the ehci usb debug
1765 port as it is probed via PCI. The poll interval is
1766 optional and is the number seconds in between
1767 each poll cycle to the debug port in case you need
1768 the functionality for interrupting the kernel with
1769 gdb or control-c on the dbgp connection. When
1770 not using this parameter you use sysrq-g to break into
1771 the kernel debugger.
1773 kgdboc= [KGDB,HW] kgdb over consoles.
1774 Requires a tty driver that supports console polling,
1775 or a supported polling keyboard driver (non-usb).
1776 Serial only format: <serial_device>[,baud]
1777 keyboard only format: kbd
1778 keyboard and serial format: kbd,<serial_device>[,baud]
1779 Optional Kernel mode setting:
1780 kms, kbd format: kms,kbd
1781 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
1783 kgdbwait [KGDB] Stop kernel execution and enter the
1784 kernel debugger at the earliest opportunity.
1786 kmac= [MIPS] korina ethernet MAC address.
1787 Configure the RouterBoard 532 series on-chip
1788 Ethernet adapter MAC address.
1790 kmemleak= [KNL] Boot-time kmemleak enable/disable
1791 Valid arguments: on, off
1793 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
1796 kmemcheck= [X86] Boot-time kmemcheck enable/disable/one-shot mode
1797 Valid arguments: 0, 1, 2
1798 kmemcheck=0 (disabled)
1799 kmemcheck=1 (enabled)
1800 kmemcheck=2 (one-shot mode)
1801 Default: 2 (one-shot mode)
1803 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
1804 Default is 0 (don't ignore, but inject #GP)
1806 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
1810 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
1811 Default is 1 (enabled)
1813 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
1815 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
1817 kvm-intel.ept= [KVM,Intel] Disable extended page tables
1818 (virtualized MMU) support on capable Intel chips.
1819 Default is 1 (enabled)
1821 kvm-intel.emulate_invalid_guest_state=
1822 [KVM,Intel] Enable emulation of invalid guest states
1823 Default is 0 (disabled)
1825 kvm-intel.flexpriority=
1826 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
1827 Default is 1 (enabled)
1830 [KVM,Intel] Enable VMX nesting (nVMX).
1831 Default is 0 (disabled)
1833 kvm-intel.unrestricted_guest=
1834 [KVM,Intel] Disable unrestricted guest feature
1835 (virtualized real and unpaged mode) on capable
1836 Intel chips. Default is 1 (enabled)
1838 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
1839 feature (tagged TLBs) on capable Intel chips.
1840 Default is 1 (enabled)
1846 lapic [X86-32,APIC] Enable the local APIC even if BIOS
1849 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
1850 value for LAPIC timer one-shot implementation. Default
1851 back to the programmable timer unit in the LAPIC.
1853 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
1856 libata.dma= [LIBATA] DMA control
1857 libata.dma=0 Disable all PATA and SATA DMA
1858 libata.dma=1 PATA and SATA Disk DMA only
1859 libata.dma=2 ATAPI (CDROM) DMA only
1860 libata.dma=4 Compact Flash DMA only
1861 Combinations also work, so libata.dma=3 enables DMA
1862 for disks and CDROMs, but not CFs.
1864 libata.ignore_hpa= [LIBATA] Ignore HPA limit
1865 libata.ignore_hpa=0 keep BIOS limits (default)
1866 libata.ignore_hpa=1 ignore limits, using full disk
1868 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
1872 libata.force= [LIBATA] Force configurations. The format is comma
1873 separated list of "[ID:]VAL" where ID is
1874 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
1875 matching port, link or device. Basically, it matches
1876 the ATA ID string printed on console by libata. If
1877 the whole ID part is omitted, the last PORT and DEVICE
1878 values are used. If ID hasn't been specified yet, the
1879 configuration applies to all ports, links and devices.
1881 If only DEVICE is omitted, the parameter applies to
1882 the port and all links and devices behind it. DEVICE
1883 number of 0 either selects the first device or the
1884 first fan-out link behind PMP device. It does not
1885 select the host link. DEVICE number of 15 selects the
1886 host link and device attached to it.
1888 The VAL specifies the configuration to force. As long
1889 as there's no ambiguity shortcut notation is allowed.
1890 For example, both 1.5 and 1.5G would work for 1.5Gbps.
1891 The following configurations can be forced.
1893 * Cable type: 40c, 80c, short40c, unk, ign or sata.
1894 Any ID with matching PORT is used.
1896 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
1898 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
1899 udma[/][16,25,33,44,66,100,133] notation is also
1902 * [no]ncq: Turn on or off NCQ.
1904 * [no]ncqtrim: Turn off queued DSM TRIM.
1906 * nohrst, nosrst, norst: suppress hard, soft
1909 * rstonce: only attempt one reset during
1910 hot-unplug link recovery
1912 * dump_id: dump IDENTIFY data.
1914 * atapi_dmadir: Enable ATAPI DMADIR bridge support
1916 * disable: Disable this device.
1918 If there are multiple matching configurations changing
1919 the same attribute, the last one is used.
1921 memblock=debug [KNL] Enable memblock debug messages.
1923 load_ramdisk= [RAM] List of ramdisks to load from floppy
1924 See Documentation/blockdev/ramdisk.txt.
1926 lockd.nlm_grace_period=P [NFS] Assign grace period.
1929 lockd.nlm_tcpport=N [NFS] Assign TCP port.
1932 lockd.nlm_timeout=T [NFS] Assign timeout value.
1935 lockd.nlm_udpport=M [NFS] Assign UDP port.
1938 locktorture.nreaders_stress= [KNL]
1939 Set the number of locking read-acquisition kthreads.
1940 Defaults to being automatically set based on the
1941 number of online CPUs.
1943 locktorture.nwriters_stress= [KNL]
1944 Set the number of locking write-acquisition kthreads.
1946 locktorture.onoff_holdoff= [KNL]
1947 Set time (s) after boot for CPU-hotplug testing.
1949 locktorture.onoff_interval= [KNL]
1950 Set time (s) between CPU-hotplug operations, or
1951 zero to disable CPU-hotplug testing.
1953 locktorture.shuffle_interval= [KNL]
1954 Set task-shuffle interval (jiffies). Shuffling
1955 tasks allows some CPUs to go into dyntick-idle
1956 mode during the locktorture test.
1958 locktorture.shutdown_secs= [KNL]
1959 Set time (s) after boot system shutdown. This
1960 is useful for hands-off automated testing.
1962 locktorture.stat_interval= [KNL]
1963 Time (s) between statistics printk()s.
1965 locktorture.stutter= [KNL]
1966 Time (s) to stutter testing, for example,
1967 specifying five seconds causes the test to run for
1968 five seconds, wait for five seconds, and so on.
1969 This tests the locking primitive's ability to
1970 transition abruptly to and from idle.
1972 locktorture.torture_runnable= [BOOT]
1973 Start locktorture running at boot time.
1975 locktorture.torture_type= [KNL]
1976 Specify the locking implementation to test.
1978 locktorture.verbose= [KNL]
1979 Enable additional printk() statements.
1981 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
1984 loglevel= All Kernel Messages with a loglevel smaller than the
1985 console loglevel will be printed to the console. It can
1986 also be changed with klogd or other programs. The
1987 loglevels are defined as follows:
1989 0 (KERN_EMERG) system is unusable
1990 1 (KERN_ALERT) action must be taken immediately
1991 2 (KERN_CRIT) critical conditions
1992 3 (KERN_ERR) error conditions
1993 4 (KERN_WARNING) warning conditions
1994 5 (KERN_NOTICE) normal but significant condition
1995 6 (KERN_INFO) informational
1996 7 (KERN_DEBUG) debug-level messages
1998 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
1999 in bytes. n must be a power of two and greater
2000 than the minimal size. The minimal size is defined
2001 by LOG_BUF_SHIFT kernel config parameter. There is
2002 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2003 that allows to increase the default size depending on
2004 the number of CPUs. See init/Kconfig for more details.
2006 logo.nologo [FB] Disables display of the built-in Linux logo.
2007 This may be used to provide more screen space for
2008 kernel log messages and is useful when debugging
2009 kernel boot problems.
2011 lp=0 [LP] Specify parallel ports to use, e.g,
2012 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2013 lp=reset first parallel port). 'lp=0' disables the
2014 lp=auto printer driver. 'lp=reset' (which can be
2015 specified in addition to the ports) causes
2016 attached printers to be reset. Using
2017 lp=port1,port2,... specifies the parallel ports
2018 to associate lp devices with, starting with
2019 lp0. A port specification may be 'none' to skip
2020 that lp device, or a parport name such as
2021 'parport0'. Specifying 'lp=auto' instead of a
2022 port specification list means that device IDs
2023 from each port should be examined, to see if
2024 an IEEE 1284-compliant printer is attached; if
2025 so, the driver will manage that printer.
2026 See also header of drivers/char/lp.c.
2029 Sets loops_per_jiffy to given constant, thus avoiding
2030 time-consuming boot-time autodetection (up to 250 ms per
2031 CPU). 0 enables autodetection (default). To determine
2032 the correct value for your kernel, boot with normal
2033 autodetection and see what value is printed. Note that
2034 on SMP systems the preset will be applied to all CPUs,
2035 which is likely to cause problems if your CPUs need
2036 significantly divergent settings. An incorrect value
2037 will cause delays in the kernel to be wrong, leading to
2038 unpredictable I/O errors and other breakage. Although
2039 unlikely, in the extreme case this might damage your
2043 Format: <io>,<irq>,<dma>
2045 machvec= [IA-64] Force the use of a particular machine-vector
2046 (machvec) in a generic kernel.
2047 Example: machvec=hpzx1_swiotlb
2049 machtype= [Loongson] Share the same kernel image file between different
2051 Example: machtype=lemote-yeeloong-2f-7inch
2053 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2054 than or equal to this physical address is ignored.
2056 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2057 will bring up during bootup. maxcpus=n : n >= 0 limits
2058 the kernel to bring up 'n' processors. Surely after
2059 bootup you can bring up the other plugged cpu by executing
2060 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2061 only takes effect during system bootup.
2062 While n=0 is a special case, it is equivalent to "nosmp",
2063 which also disables the IO APIC.
2065 max_loop= [LOOP] The number of loop block devices that get
2066 (loop.max_loop) unconditionally pre-created at init time. The default
2067 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2068 of statically allocating a predefined number, loop
2069 devices can be requested on-demand with the
2070 /dev/loop-control interface.
2072 mce [X86-32] Machine Check Exception
2074 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.txt
2076 md= [HW] RAID subsystems devices and level
2077 See Documentation/admin-guide/md.rst.
2080 Format: <first>,<last>
2081 Specifies range of consoles to be captured by the MDA.
2083 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2084 Amount of memory to be used when the kernel is not able
2085 to see the whole system memory or for test.
2086 [X86] Work as limiting max address. Use together
2087 with memmap= to avoid physical address space collisions.
2088 Without memmap= PCI devices could be placed at addresses
2089 belonging to unused RAM.
2091 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2095 [KNL,SH] Allow user to override the default size for
2096 per-device physically contiguous DMA buffers.
2098 memhp_default_state=online/offline
2099 [KNL] Set the initial state for the memory hotplug
2100 onlining policy. If not specified, the default value is
2101 set according to the
2102 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2104 See Documentation/memory-hotplug.txt.
2106 memmap=exactmap [KNL,X86] Enable setting of an exact
2107 E820 memory map, as specified by the user.
2108 Such memmap=exactmap lines can be constructed based on
2109 BIOS output or other requirements. See the memmap=nn@ss
2112 memmap=nn[KMG]@ss[KMG]
2113 [KNL] Force usage of a specific region of memory.
2114 Region of memory to be used is from ss to ss+nn.
2116 memmap=nn[KMG]#ss[KMG]
2117 [KNL,ACPI] Mark specific memory as ACPI data.
2118 Region of memory to be marked is from ss to ss+nn.
2120 memmap=nn[KMG]$ss[KMG]
2121 [KNL,ACPI] Mark specific memory as reserved.
2122 Region of memory to be reserved is from ss to ss+nn.
2123 Example: Exclude memory from 0x18690000-0x1869ffff
2124 memmap=64K$0x18690000
2126 memmap=0x10000$0x18690000
2128 memmap=nn[KMG]!ss[KMG]
2129 [KNL,X86] Mark specific memory as protected.
2130 Region of memory to be used, from ss to ss+nn.
2131 The memory region may be marked as e820 type 12 (0xc)
2132 and is NVDIMM or ADR memory.
2134 memory_corruption_check=0/1 [X86]
2135 Some BIOSes seem to corrupt the first 64k of
2136 memory when doing things like suspend/resume.
2137 Setting this option will scan the memory
2138 looking for corruption. Enabling this will
2139 both detect corruption and prevent the kernel
2140 from using the memory being corrupted.
2141 However, its intended as a diagnostic tool; if
2142 repeatable BIOS-originated corruption always
2143 affects the same memory, you can use memmap=
2144 to prevent the kernel from using that memory.
2146 memory_corruption_check_size=size [X86]
2147 By default it checks for corruption in the low
2148 64k, making this memory unavailable for normal
2149 use. Use this parameter to scan for
2150 corruption in more or less memory.
2152 memory_corruption_check_period=seconds [X86]
2153 By default it checks for corruption every 60
2154 seconds. Use this parameter to check at some
2155 other rate. 0 disables periodic checking.
2157 memtest= [KNL,X86,ARM] Enable memtest
2159 default : 0 <disable>
2160 Specifies the number of memtest passes to be
2161 performed. Each pass selects another test
2162 pattern from a given set of patterns. Memtest
2163 fills the memory with this pattern, validates
2164 memory contents and reserves bad memory
2165 regions that are detected.
2167 mem_sleep_default= [SUSPEND] Default system suspend mode:
2168 s2idle - Suspend-To-Idle
2169 shallow - Power-On Suspend or equivalent (if supported)
2170 deep - Suspend-To-RAM or equivalent (if supported)
2171 See Documentation/power/states.txt.
2173 meye.*= [HW] Set MotionEye Camera parameters
2174 See Documentation/video4linux/meye.txt.
2176 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2177 Multi-Function General Purpose Timers on AMD Geode
2180 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2181 the BIOS has incorrectly applied a workaround. TinyBIOS
2182 version 0.98 is known to be affected, 0.99 fixes the
2183 problem by letting the user disable the workaround.
2187 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2188 physical address is ignored.
2190 mini2440= [ARM,HW,KNL]
2191 Format:[0..2][b][c][t]
2193 MINI2440 configuration specification:
2194 0 - The attached screen is the 3.5" TFT
2195 1 - The attached screen is the 7" TFT
2196 2 - The VGA Shield is attached (1024x768)
2197 Leaving out the screen size parameter will not load
2198 the TFT driver, and the framebuffer will be left
2200 b - Enable backlight. The TFT backlight pin will be
2201 linked to the kernel VESA blanking code and a GPIO
2202 LED. This parameter is not necessary when using the
2204 c - Enable the s3c camera interface.
2205 t - Reserved for enabling touchscreen support. The
2206 touchscreen support is not enabled in the mainstream
2207 kernel as of 2.6.30, a preliminary port can be found
2208 in the "bleeding edge" mini2440 support kernel at
2209 http://repo.or.cz/w/linux-2.6/mini2440.git
2212 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2213 parameter allows control of the logging verbosity for
2214 the additional memory initialisation checks. A value
2215 of 0 disables mminit logging and a level of 4 will
2216 log everything. Information is printed at KERN_DEBUG
2217 so loglevel=8 may also need to be specified.
2220 [KNL] When CONFIG_MODULE_SIG is set, this means that
2221 modules without (valid) signatures will fail to load.
2222 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2223 is always true, so this option does nothing.
2225 module_blacklist= [KNL] Do not load a comma-separated list of
2226 modules. Useful for debugging problem modules.
2229 [MOUSE] Maximum time between finger touching and
2230 leaving touchpad surface for touch to be considered
2231 a tap and be reported as a left button click (for
2232 touchpads working in absolute mode only).
2234 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2235 reporting absolute coordinates, such as tablets
2236 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2237 reporting absolute coordinates, such as tablets
2239 movablecore=nn[KMG] [KNL,X86,IA-64,PPC] This parameter
2240 is similar to kernelcore except it specifies the
2241 amount of memory used for migratable allocations.
2242 If both kernelcore and movablecore is specified,
2243 then kernelcore will be at *least* the specified
2244 value but may be more. If movablecore on its own
2245 is specified, the administrator must be careful
2246 that the amount of memory usable for all allocations
2249 movable_node [KNL] Boot-time switch to enable the effects
2250 of CONFIG_MOVABLE_NODE=y. See mm/Kconfig for details.
2252 MTD_Partition= [MTD]
2253 Format: <name>,<region-number>,<size>,<offset>
2255 MTD_Region= [MTD] Format:
2256 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2259 See drivers/mtd/cmdlinepart.c.
2261 multitce=off [PPC] This parameter disables the use of the pSeries
2262 firmware feature for updating multiple TCE entries
2265 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2267 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2269 boundary - index of last SLC block on Flex-OneNAND.
2270 The remaining blocks are configured as MLC blocks.
2271 lock - Configure if Flex-OneNAND boundary should be locked.
2272 Once locked, the boundary cannot be changed.
2273 1 indicates lock status, 0 indicates unlock status.
2276 ARM/S3C2412 JIVE boot control
2278 See arch/arm/mach-s3c2412/mach-jive.c
2280 mtouchusb.raw_coordinates=
2281 [HW] Make the MicroTouch USB driver use raw coordinates
2282 ('y', default) or cooked coordinates ('n')
2284 mtrr_chunk_size=nn[KMG] [X86]
2285 used for mtrr cleanup. It is largest continuous chunk
2286 that could hold holes aka. UC entries.
2288 mtrr_gran_size=nn[KMG] [X86]
2289 Used for mtrr cleanup. It is granularity of mtrr block.
2291 Large value could prevent small alignment from
2294 mtrr_spare_reg_nr=n [X86]
2296 Range: 0,7 : spare reg number
2298 Used for mtrr cleanup. It is spare mtrr entries number.
2299 Set to 2 or more if your graphical card needs more.
2301 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2303 netdev= [NET] Network devices parameters
2304 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2305 Note that mem_start is often overloaded to mean
2306 something different and driver-specific.
2307 This usage is only documented in each driver source
2311 [NETFILTER] Enable connection tracking flow accounting
2312 0 to disable accounting
2313 1 to enable accounting
2316 nfsaddrs= [NFS] Deprecated. Use ip= instead.
2317 See Documentation/filesystems/nfs/nfsroot.txt.
2319 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
2320 See Documentation/filesystems/nfs/nfsroot.txt.
2322 nfsrootdebug [NFS] enable nfsroot debugging messages.
2323 See Documentation/filesystems/nfs/nfsroot.txt.
2325 nfs.callback_nr_threads=
2326 [NFSv4] set the total number of threads that the
2327 NFS client will assign to service NFSv4 callback
2330 nfs.callback_tcpport=
2331 [NFS] set the TCP port on which the NFSv4 callback
2332 channel should listen.
2335 [NFS] sets the pathname to the program which is used
2336 to update the NFS client cache entries.
2338 nfs.cache_getent_timeout=
2339 [NFS] sets the timeout after which an attempt to
2340 update a cache entry is deemed to have failed.
2342 nfs.idmap_cache_timeout=
2343 [NFS] set the maximum lifetime for idmapper cache
2347 [NFS] enable 64-bit inode numbers.
2348 If zero, the NFS client will fake up a 32-bit inode
2349 number for the readdir() and stat() syscalls instead
2350 of returning the full 64-bit number.
2351 The default is to return 64-bit inode numbers.
2353 nfs.max_session_cb_slots=
2354 [NFSv4.1] Sets the maximum number of session
2355 slots the client will assign to the callback
2356 channel. This determines the maximum number of
2357 callbacks the client will process in parallel for
2358 a particular server.
2360 nfs.max_session_slots=
2361 [NFSv4.1] Sets the maximum number of session slots
2362 the client will attempt to negotiate with the server.
2363 This limits the number of simultaneous RPC requests
2364 that the client can send to the NFSv4.1 server.
2365 Note that there is little point in setting this
2366 value higher than the max_tcp_slot_table_limit.
2368 nfs.nfs4_disable_idmapping=
2369 [NFSv4] When set to the default of '1', this option
2370 ensures that both the RPC level authentication
2371 scheme and the NFS level operations agree to use
2372 numeric uids/gids if the mount is using the
2373 'sec=sys' security flavour. In effect it is
2374 disabling idmapping, which can make migration from
2375 legacy NFSv2/v3 systems to NFSv4 easier.
2376 Servers that do not support this mode of operation
2377 will be autodetected by the client, and it will fall
2378 back to using the idmapper.
2379 To turn off this behaviour, set the value to '0'.
2381 [NFS4] Specify an additional fixed unique ident-
2382 ification string that NFSv4 clients can insert into
2383 their nfs_client_id4 string. This is typically a
2384 UUID that is generated at system install time.
2386 nfs.send_implementation_id =
2387 [NFSv4.1] Send client implementation identification
2388 information in exchange_id requests.
2389 If zero, no implementation identification information
2391 The default is to send the implementation identification
2394 nfs.recover_lost_locks =
2395 [NFSv4] Attempt to recover locks that were lost due
2396 to a lease timeout on the server. Please note that
2397 doing this risks data corruption, since there are
2398 no guarantees that the file will remain unchanged
2399 after the locks are lost.
2400 If you want to enable the kernel legacy behaviour of
2401 attempting to recover these locks, then set this
2403 The default parameter value of '0' causes the kernel
2404 not to attempt recovery of lost locks.
2406 nfs4.layoutstats_timer =
2407 [NFSv4.2] Change the rate at which the kernel sends
2408 layoutstats to the pNFS metadata server.
2410 Setting this to value to 0 causes the kernel to use
2411 whatever value is the default set by the layout
2412 driver. A non-zero value sets the minimum interval
2413 in seconds between layoutstats transmissions.
2415 nfsd.nfs4_disable_idmapping=
2416 [NFSv4] When set to the default of '1', the NFSv4
2417 server will return only numeric uids and gids to
2418 clients using auth_sys, and will accept numeric uids
2419 and gids from such clients. This is intended to ease
2420 migration from NFSv2/v3.
2422 nmi_debug= [KNL,AVR32,SH] Specify one or more actions to take
2423 when a NMI is triggered.
2424 Format: [state][,regs][,debounce][,die]
2426 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
2427 Format: [panic,][nopanic,][num]
2429 0 - turn hardlockup detector in nmi_watchdog off
2430 1 - turn hardlockup detector in nmi_watchdog on
2431 When panic is specified, panic when an NMI watchdog
2432 timeout occurs (or 'nopanic' to override the opposite
2433 default). To disable both hard and soft lockup detectors,
2434 please see 'nowatchdog'.
2435 This is useful when you use a panic=... timeout and
2436 need the box quickly up again.
2438 netpoll.carrier_timeout=
2439 [NET] Specifies amount of time (in seconds) that
2440 netpoll should wait for a carrier. By default netpoll
2443 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
2444 emulation library even if a 387 maths coprocessor
2448 [HW] Never suspend the console
2449 Disable suspending of consoles during suspend and
2450 hibernate operations. Once disabled, debugging
2451 messages can reach various consoles while the rest
2452 of the system is being put to sleep (ie, while
2453 debugging driver suspend/resume hooks). This may
2454 not work reliably with all consoles, but is known
2455 to work with serial and VGA consoles.
2456 To facilitate more flexible debugging, we also add
2457 console_suspend, a printk module parameter to control
2458 it. Users could use console_suspend (usually
2459 /sys/module/printk/parameters/console_suspend) to
2460 turn on/off it dynamically.
2462 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
2463 caches in the slab allocator. Saves per-node memory,
2464 but will impact performance.
2468 noapic [SMP,APIC] Tells the kernel to not make use of any
2469 IOAPICs that may be present in the system.
2471 noautogroup Disable scheduler automatic task group creation.
2473 nobats [PPC] Do not use BATs for mapping kernel lowmem
2474 on "Classic" PPC cores.
2478 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
2480 nodelayacct [KNL] Disable per-task delay accounting
2482 nodsp [SH] Disable hardware DSP at boot time.
2484 noefi Disable EFI runtime services support.
2489 On X86-32 available only on PAE configured kernels.
2490 noexec=on: enable non-executable mappings (default)
2491 noexec=off: disable non-executable mappings
2494 Disable SMAP (Supervisor Mode Access Prevention)
2495 even if it is supported by processor.
2498 Disable SMEP (Supervisor Mode Execution Prevention)
2499 even if it is supported by processor.
2502 This affects only 32-bit executables.
2503 noexec32=on: enable non-executable mappings (default)
2504 read doesn't imply executable mappings
2505 noexec32=off: disable non-executable mappings
2506 read implies executable mappings
2508 nofpu [MIPS,SH] Disable hardware FPU at boot time.
2510 nofxsr [BUGS=X86-32] Disables x86 floating point extended
2511 register save and restore. The kernel will only save
2512 legacy floating-point registers on task switch.
2514 nohugeiomap [KNL,x86] Disable kernel huge I/O mappings.
2516 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
2517 Equivalent to smt=1.
2519 noxsave [BUGS=X86] Disables x86 extended register state save
2520 and restore using xsave. The kernel will fallback to
2521 enabling legacy floating-point and sse state.
2523 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
2524 register states. The kernel will fall back to use
2525 xsave to save the states. By using this parameter,
2526 performance of saving the states is degraded because
2527 xsave doesn't support modified optimization while
2528 xsaveopt supports it on xsaveopt enabled systems.
2530 noxsaves [X86] Disables xsaves and xrstors used in saving and
2531 restoring x86 extended register state in compacted
2532 form of xsave area. The kernel will fall back to use
2533 xsaveopt and xrstor to save and restore the states
2534 in standard form of xsave area. By using this
2535 parameter, xsave area per process might occupy more
2536 memory on xsaves enabled systems.
2538 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
2539 wfi(ARM) instruction doesn't work correctly and not to
2540 use it. This is also useful when using JTAG debugger.
2542 no_file_caps Tells the kernel not to honor file capabilities. The
2543 only way then for a file to be executed with privilege
2544 is to be setuid root or executed by root.
2546 nohalt [IA-64] Tells the kernel not to use the power saving
2547 function PAL_HALT_LIGHT when idle. This increases
2548 power-consumption. On the positive side, it reduces
2549 interrupt wake-up latency, which may improve performance
2550 in certain environments such as networked servers or
2553 nohibernate [HIBERNATION] Disable hibernation and resume.
2555 nohz= [KNL] Boottime enable/disable dynamic ticks
2556 Valid arguments: on, off
2559 nohz_full= [KNL,BOOT]
2560 The argument is a cpu list, as described above.
2561 In kernels built with CONFIG_NO_HZ_FULL=y, set
2562 the specified list of CPUs whose tick will be stopped
2563 whenever possible. The boot CPU will be forced outside
2564 the range to maintain the timekeeping.
2565 The CPUs in this range must also be included in the
2568 noiotrap [SH] Disables trapped I/O port accesses.
2570 noirqdebug [X86-32] Disables the code which attempts to detect and
2571 disable unhandled interrupt sources.
2573 no_timer_check [X86,APIC] Disables the code which tests for
2574 broken timer IRQ sources.
2576 noisapnp [ISAPNP] Disables ISA PnP code.
2578 noinitrd [RAM] Tells the kernel not to load any configured
2581 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
2583 [Deprecated - use intremap=off]
2587 noinvpcid [X86] Disable the INVPCID cpu feature.
2589 nojitter [IA-64] Disables jitter checking for ITC timers.
2591 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
2593 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
2597 [X86,PV_OPS] Disable paravirtualized VMware scheduler
2598 clock and use the default one.
2600 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting.
2601 steal time is computed, but won't influence scheduler
2604 nolapic [X86-32,APIC] Do not enable or use the local APIC.
2606 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
2608 noltlbs [PPC] Do not use large page/tlb entries for kernel
2609 lowmem mapping on PPC40x and PPC8xx
2611 nomca [IA-64] Disable machine check abort handling
2613 nomce [X86-32] Disable Machine Check Exception
2615 nomfgpt [X86-32] Disable Multi-Function General Purpose
2616 Timer usage (for AMD Geode machines).
2618 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
2619 shutdown the other cpus. Instead use the REBOOT_VECTOR
2622 nomodule Disable module load
2624 nopat [X86] Disable PAT (page attribute table extension of
2625 pagetables) support.
2627 norandmaps Don't use address space randomization. Equivalent to
2628 echo 0 > /proc/sys/kernel/randomize_va_space
2630 noreplace-paravirt [X86,IA-64,PV_OPS] Don't patch paravirt_ops
2632 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
2633 with UP alternatives
2635 nordrand [X86] Disable kernel use of the RDRAND and
2636 RDSEED instructions even if they are supported
2637 by the processor. RDRAND and RDSEED are still
2638 available to user space applications.
2640 noresume [SWSUSP] Disables resume and restores original swap
2643 no-scroll [VGA] Disables scrollback.
2644 This is required for the Braillex ib80-piezo Braille
2645 reader made by F.H. Papenmeier (Germany).
2649 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
2651 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
2652 and disable the IO APIC. legacy for "maxcpus=0".
2654 nosoftlockup [KNL] Disable the soft-lockup detector.
2656 nosync [HW,M68K] Disables sync negotiation for all devices.
2658 notsc [BUGS=X86-32] Disable Time Stamp Counter
2660 nowatchdog [KNL] Disable both lockup detectors, i.e.
2661 soft-lockup and NMI watchdog (hard-lockup).
2665 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
2667 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
2668 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
2669 Some features depend on CPU0. Known dependencies are:
2670 1. Resume from suspend/hibernate depends on CPU0.
2671 Suspend/hibernate will fail if CPU0 is offline and you
2672 need to online CPU0 before suspend/hibernate.
2673 2. PIC interrupts also depend on CPU0. CPU0 can't be
2674 removed if a PIC interrupt is detected.
2675 It's said poweroff/reboot may depend on CPU0 on some
2676 machines although I haven't seen such issues so far
2677 after CPU0 is offline on a few tested machines.
2678 If the dependencies are under your control, you can
2679 turn on cpu0_hotplug.
2681 nptcg= [IA-64] Override max number of concurrent global TLB
2682 purges which is reported from either PAL_VM_SUMMARY or
2685 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
2686 could support. nr_cpus=n : n >= 1 limits the kernel to
2687 support 'n' processors. It could be larger than the
2688 number of already plugged CPU during bootup, later in
2689 runtime you can physically add extra cpu until it reaches
2690 n. So during boot up some boot time memory for per-cpu
2691 variables need be pre-allocated for later physical cpu
2694 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
2696 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
2697 Allowed values are enable and disable
2699 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
2700 one of ['zone', 'node', 'default'] can be specified
2701 This can be set from sysctl after boot.
2702 See Documentation/sysctl/vm.txt for details.
2704 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
2705 See Documentation/debugging-via-ohci1394.txt for more
2708 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
2709 Rather than timing out after 20 ms if an EC
2710 command is not properly ACKed, override the length
2711 of the timeout. We have interrupts disabled while
2712 waiting for the ACK, so if this is set too high
2713 interrupts *may* be lost!
2715 omap_mux= [OMAP] Override bootloader pin multiplexing.
2716 Format: <mux_mode0.mode_name=value>...
2717 For example, to override I2C bus2:
2718 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
2720 oprofile.timer= [HW]
2721 Use timer interrupt instead of performance counters
2723 oprofile.cpu_type= Force an oprofile cpu type
2724 This might be useful if you have an older oprofile
2725 userland or if you want common events.
2726 Format: { arch_perfmon }
2727 arch_perfmon: [X86] Force use of architectural
2728 perfmon on Intel CPUs instead of the
2729 CPU specific event set.
2730 timer: [X86] Force use of architectural NMI
2731 timer mode (see also oprofile.timer
2732 for generic hr timer mode)
2734 oops=panic Always panic on oopses. Default is to just kill the
2735 process, but there is a small probability of
2736 deadlocking the machine.
2737 This will also cause panics on machine check exceptions.
2738 Useful together with panic=30 to trigger a reboot.
2741 See Documentation/sound/oss/oss-parameters.txt
2743 page_owner= [KNL] Boot-time page_owner enabling option.
2744 Storage of the information about who allocated
2745 each page is disabled in default. With this switch,
2747 on: enable the feature
2749 page_poison= [KNL] Boot-time parameter changing the state of
2750 poisoning on the buddy allocator.
2751 off: turn off poisoning
2752 on: turn on poisoning
2754 panic= [KNL] Kernel behaviour on panic: delay <timeout>
2755 timeout > 0: seconds before rebooting
2756 timeout = 0: wait forever
2757 timeout < 0: reboot immediately
2760 panic_on_warn panic() instead of WARN(). Useful to cause kdump
2763 crash_kexec_post_notifiers
2764 Run kdump after running panic-notifiers and dumping
2765 kmsg. This only for the users who doubt kdump always
2766 succeeds in any situation.
2767 Note that this also increases risks of kdump failure,
2768 because some panic notifiers can make the crashed
2769 kernel more unstable.
2771 parkbd.port= [HW] Parallel port number the keyboard adapter is
2772 connected to, default is 0.
2774 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
2775 0 for XT, 1 for AT (default is AT).
2778 parport= [HW,PPT] Specify parallel ports. 0 disables.
2779 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
2780 Use 'auto' to force the driver to use any
2781 IRQ/DMA settings detected (the default is to
2782 ignore detected IRQ/DMA settings because of
2783 possible conflicts). You can specify the base
2784 address, IRQ, and DMA settings; IRQ and DMA
2785 should be numbers, or 'auto' (for using detected
2786 settings on that particular port), or 'nofifo'
2787 (to avoid using a FIFO even if it is detected).
2788 Parallel ports are assigned in the order they
2789 are specified on the command line, starting
2792 parport_init_mode= [HW,PPT]
2793 Configure VIA parallel port to operate in
2794 a specific mode. This is necessary on Pegasos
2795 computer where firmware has no options for setting
2796 up parallel port mode and sets it to spp.
2797 Currently this function knows 686a and 8231 chips.
2798 Format: [spp|ps2|epp|ecp|ecpepp]
2801 Halt all CPUs after the first oops has been printed for
2802 the specified number of seconds. This is to be used if
2803 your oopses keep scrolling off the screen.
2808 See header of drivers/block/paride/pcd.c.
2809 See also Documentation/blockdev/paride.txt.
2811 pci=option[,option...] [PCI] various PCI subsystem options:
2812 earlydump [X86] dump PCI config space before the kernel
2814 off [X86] don't probe for the PCI bus
2815 bios [X86-32] force use of PCI BIOS, don't access
2816 the hardware directly. Use this if your machine
2817 has a non-standard PCI host bridge.
2818 nobios [X86-32] disallow use of PCI BIOS, only direct
2819 hardware access methods are allowed. Use this
2820 if you experience crashes upon bootup and you
2821 suspect they are caused by the BIOS.
2822 conf1 [X86] Force use of PCI Configuration Access
2823 Mechanism 1 (config address in IO port 0xCF8,
2824 data in IO port 0xCFC, both 32-bit).
2825 conf2 [X86] Force use of PCI Configuration Access
2826 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
2827 the function, IO port 0xCFA, also 8-bit, sets
2828 bus number. The config space is then accessed
2829 through ports 0xC000-0xCFFF).
2830 See http://wiki.osdev.org/PCI for more info
2831 on the configuration access mechanisms.
2832 noaer [PCIE] If the PCIEAER kernel config parameter is
2833 enabled, this kernel boot option can be used to
2834 disable the use of PCIE advanced error reporting.
2835 nodomains [PCI] Disable support for multiple PCI
2836 root domains (aka PCI segments, in ACPI-speak).
2837 nommconf [X86] Disable use of MMCONFIG for PCI
2839 check_enable_amd_mmconf [X86] check for and enable
2840 properly configured MMIO access to PCI
2841 config space on AMD family 10h CPU
2842 nomsi [MSI] If the PCI_MSI kernel config parameter is
2843 enabled, this kernel boot option can be used to
2844 disable the use of MSI interrupts system-wide.
2845 noioapicquirk [APIC] Disable all boot interrupt quirks.
2846 Safety option to keep boot IRQs enabled. This
2847 should never be necessary.
2848 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
2849 primary IO-APIC for bridges that cannot disable
2850 boot IRQs. This fixes a source of spurious IRQs
2851 when the system masks IRQs.
2852 noioapicreroute [APIC] Disable workaround that uses the
2853 boot IRQ equivalent of an IRQ that connects to
2854 a chipset where boot IRQs cannot be disabled.
2855 The opposite of ioapicreroute.
2856 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
2857 routing table. These calls are known to be buggy
2858 on several machines and they hang the machine
2859 when used, but on other computers it's the only
2860 way to get the interrupt routing table. Try
2861 this option if the kernel is unable to allocate
2862 IRQs or discover secondary PCI buses on your
2864 rom [X86] Assign address space to expansion ROMs.
2865 Use with caution as certain devices share
2866 address decoders between ROMs and other
2868 norom [X86] Do not assign address space to
2869 expansion ROMs that do not already have
2870 BIOS assigned address ranges.
2871 nobar [X86] Do not assign address space to the
2872 BARs that weren't assigned by the BIOS.
2873 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
2874 assigned automatically to PCI devices. You can
2875 make the kernel exclude IRQs of your ISA cards
2877 pirqaddr=0xAAAAA [X86] Specify the physical address
2878 of the PIRQ table (normally generated
2879 by the BIOS) if it is outside the
2880 F0000h-100000h range.
2881 lastbus=N [X86] Scan all buses thru bus #N. Can be
2882 useful if the kernel is unable to find your
2883 secondary buses and you want to tell it
2884 explicitly which ones they are.
2885 assign-busses [X86] Always assign all PCI bus
2886 numbers ourselves, overriding
2887 whatever the firmware may have done.
2888 usepirqmask [X86] Honor the possible IRQ mask stored
2889 in the BIOS $PIR table. This is needed on
2890 some systems with broken BIOSes, notably
2891 some HP Pavilion N5400 and Omnibook XE3
2892 notebooks. This will have no effect if ACPI
2893 IRQ routing is enabled.
2894 noacpi [X86] Do not use ACPI for IRQ routing
2895 or for PCI scanning.
2896 use_crs [X86] Use PCI host bridge window information
2897 from ACPI. On BIOSes from 2008 or later, this
2898 is enabled by default. If you need to use this,
2899 please report a bug.
2900 nocrs [X86] Ignore PCI host bridge windows from ACPI.
2901 If you need to use this, please report a bug.
2902 routeirq Do IRQ routing for all PCI devices.
2903 This is normally done in pci_enable_device(),
2904 so this option is a temporary workaround
2905 for broken drivers that don't call it.
2906 skip_isa_align [X86] do not align io start addr, so can
2907 handle more pci cards
2908 noearly [X86] Don't do any early type 1 scanning.
2909 This might help on some broken boards which
2910 machine check when some devices' config space
2911 is read. But various workarounds are disabled
2912 and some IOMMU drivers will not work.
2913 bfsort Sort PCI devices into breadth-first order.
2914 This sorting is done to get a device
2915 order compatible with older (<= 2.4) kernels.
2916 nobfsort Don't sort PCI devices into breadth-first order.
2917 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
2918 tuning and use the BIOS-configured MPS defaults.
2919 pcie_bus_safe Set every device's MPS to the largest value
2920 supported by all devices below the root complex.
2921 pcie_bus_perf Set device MPS to the largest allowable MPS
2922 based on its parent bus. Also set MRRS (Max
2923 Read Request Size) to the largest supported
2924 value (no larger than the MPS that the device
2925 or bus can support) for best performance.
2926 pcie_bus_peer2peer Set every device's MPS to 128B, which
2927 every device is guaranteed to support. This
2928 configuration allows peer-to-peer DMA between
2929 any pair of devices, possibly at the cost of
2930 reduced performance. This also guarantees
2931 that hot-added devices will work.
2932 cbiosize=nn[KMG] The fixed amount of bus space which is
2933 reserved for the CardBus bridge's IO window.
2934 The default value is 256 bytes.
2935 cbmemsize=nn[KMG] The fixed amount of bus space which is
2936 reserved for the CardBus bridge's memory
2937 window. The default value is 64 megabytes.
2940 [<order of align>@][<domain>:]<bus>:<slot>.<func>[; ...]
2941 [<order of align>@]pci:<vendor>:<device>\
2942 [:<subvendor>:<subdevice>][; ...]
2943 Specifies alignment and device to reassign
2944 aligned memory resources.
2945 If <order of align> is not specified,
2946 PAGE_SIZE is used as alignment.
2947 PCI-PCI bridge can be specified, if resource
2948 windows need to be expanded.
2949 To specify the alignment for several
2950 instances of a device, the PCI vendor,
2951 device, subvendor, and subdevice may be
2952 specified, e.g., 4096@pci:8086:9c22:103c:198f
2953 ecrc= Enable/disable PCIe ECRC (transaction layer
2954 end-to-end CRC checking).
2955 bios: Use BIOS/firmware settings. This is the
2959 hpiosize=nn[KMG] The fixed amount of bus space which is
2960 reserved for hotplug bridge's IO window.
2961 Default size is 256 bytes.
2962 hpmemsize=nn[KMG] The fixed amount of bus space which is
2963 reserved for hotplug bridge's memory window.
2964 Default size is 2 megabytes.
2965 hpbussize=nn The minimum amount of additional bus numbers
2966 reserved for buses below a hotplug bridge.
2968 realloc= Enable/disable reallocating PCI bridge resources
2969 if allocations done by BIOS are too small to
2970 accommodate resources required by all child
2972 off: Turn realloc off
2974 realloc same as realloc=on
2975 noari do not use PCIe ARI.
2976 pcie_scan_all Scan all possible PCIe devices. Otherwise we
2977 only look for one device below a PCIe downstream
2980 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
2983 force Enable ASPM even on devices that claim not to support it.
2984 WARNING: Forcing ASPM on may cause system lockups.
2986 pcie_hp= [PCIE] PCI Express Hotplug driver options:
2987 nomsi Do not use MSI for PCI Express Native Hotplug (this
2988 makes all PCIe ports use INTx for hotplug services).
2990 pcie_ports= [PCIE] PCIe ports handling:
2991 auto Ask the BIOS whether or not to use native PCIe services
2992 associated with PCIe ports (PME, hot-plug, AER). Use
2993 them only if that is allowed by the BIOS.
2994 native Use native PCIe services associated with PCIe ports
2996 compat Treat PCIe ports as PCI-to-PCI bridges, disable the PCIe
2999 pcie_port_pm= [PCIE] PCIe port power management handling:
3000 off Disable power management of all PCIe ports
3001 force Forcibly enable power management of all PCIe ports
3003 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3004 nomsi Do not use MSI for native PCIe PME signaling (this makes
3005 all PCIe root ports use INTx for all services).
3007 pcmv= [HW,PCMCIA] BadgePAD 4
3011 Keep all power-domains already enabled by bootloader on,
3012 even if no driver has claimed them. This is useful
3013 for debug and development, but should not be
3014 needed on a platform with proper driver support.
3017 See Documentation/blockdev/paride.txt.
3019 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3022 See arch/parisc/kernel/pdc_chassis.c
3024 percpu_alloc= Select which percpu first chunk allocator to use.
3025 Currently supported values are "embed" and "page".
3026 Archs may support subset or none of the selections.
3027 See comments in mm/percpu.c for details on each
3028 allocator. This parameter is primarily for debugging
3029 and performance comparison.
3032 See Documentation/blockdev/paride.txt.
3035 See Documentation/blockdev/paride.txt.
3037 pirq= [SMP,APIC] Manual mp-table setup
3038 See Documentation/x86/i386/IO-APIC.txt.
3040 plip= [PPT,NET] Parallel port network link
3041 Format: { parport<nr> | timid | 0 }
3042 See also Documentation/parport.txt.
3044 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3045 Override pmtimer IOPort with a hex value.
3049 Enable PNP debug messages (depends on the
3050 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3051 via /sys/module/pnp/parameters/debug. We always show
3052 current resource usage; turning this on also shows
3053 possible settings and some assignment information.
3059 { on | off | curr | res | no-curr | no-res }
3062 [ISAPNP] Exclude IRQs for the autoconfiguration
3065 [ISAPNP] Exclude DMAs for the autoconfiguration
3067 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3068 Ranges are in pairs (I/O port base and size).
3071 [ISAPNP] Exclude memory regions for the
3073 Ranges are in pairs (memory base and size).
3075 ports= [IP_VS_FTP] IPVS ftp helper module
3077 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3079 Format: <port>,<port>....
3081 powersave=off [PPC] This option disables power saving features.
3082 It specifically disables cpuidle and sets the
3083 platform machine description specific power_save
3084 function to NULL. On Idle the CPU just reduces
3087 ppc_strict_facility_enable
3088 [PPC] This option catches any kernel floating point,
3089 Altivec, VSX and SPE outside of regions specifically
3090 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3091 There is some performance impact when enabling this.
3093 print-fatal-signals=
3094 [KNL] debug: print fatal signals
3096 If enabled, warn about various signal handling
3097 related application anomalies: too many signals,
3098 too many POSIX.1 timers, fatal signals causing a
3101 If you hit the warning due to signal overflow,
3102 you might want to try "ulimit -i unlimited".
3106 printk.always_kmsg_dump=
3107 Trigger kmsg_dump for cases other than kernel oops or
3109 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3112 printk.devkmsg={on,off,ratelimit}
3113 Control writing to /dev/kmsg.
3114 on - unlimited logging to /dev/kmsg from userspace
3115 off - logging to /dev/kmsg disabled
3116 ratelimit - ratelimit the logging
3119 printk.time= Show timing data prefixed to each printk message line
3120 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3122 processor.max_cstate= [HW,ACPI]
3123 Limit processor to maximum C-state
3124 max_cstate=9 overrides any DMI blacklist limit.
3126 processor.nocst [HW,ACPI]
3127 Ignore the _CST method to determine C-states,
3128 instead using the legacy FADT method
3130 profile= [KNL] Enable kernel profiling via /proc/profile
3131 Format: [schedule,]<number>
3132 Param: "schedule" - profile schedule points.
3133 Param: <number> - step/bucket size as a power of 2 for
3134 statistical time based profiling.
3135 Param: "sleep" - profile D-state sleeping (millisecs).
3136 Requires CONFIG_SCHEDSTATS
3137 Param: "kvm" - profile VM exits.
3139 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3141 See Documentation/blockdev/ramdisk.txt.
3143 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3144 probe for; one of (bare|imps|exps|lifebook|any).
3145 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3147 psmouse.resetafter= [HW,MOUSE]
3148 Try to reset the device after so many bad packets
3151 [HW,MOUSE] Set desired mouse resolution, in dpi.
3152 psmouse.smartscroll=
3153 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
3154 0 = disabled, 1 = enabled (default).
3156 pstore.backend= Specify the name of the pstore backend to use
3159 See Documentation/blockdev/paride.txt.
3162 [KNL] Number of legacy pty's. Overwrites compiled-in
3165 quiet [KNL] Disable most log messages
3170 See Documentation/admin-guide/md.rst.
3172 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
3173 See Documentation/blockdev/ramdisk.txt.
3176 The argument is a cpu list, as described above.
3178 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
3179 the specified list of CPUs to be no-callback CPUs.
3180 Invocation of these CPUs' RCU callbacks will
3181 be offloaded to "rcuox/N" kthreads created for
3182 that purpose, where "x" is "b" for RCU-bh, "p"
3183 for RCU-preempt, and "s" for RCU-sched, and "N"
3184 is the CPU number. This reduces OS jitter on the
3185 offloaded CPUs, which can be useful for HPC and
3186 real-time workloads. It can also improve energy
3187 efficiency for asymmetric multiprocessors.
3190 Rather than requiring that offloaded CPUs
3191 (specified by rcu_nocbs= above) explicitly
3192 awaken the corresponding "rcuoN" kthreads,
3193 make these kthreads poll for callbacks.
3194 This improves the real-time response for the
3195 offloaded CPUs by relieving them of the need to
3196 wake up the corresponding kthread, but degrades
3197 energy efficiency by requiring that the kthreads
3198 periodically wake up to do the polling.
3200 rcutree.blimit= [KNL]
3201 Set maximum number of finished RCU callbacks to
3202 process in one batch.
3204 rcutree.dump_tree= [KNL]
3205 Dump the structure of the rcu_node combining tree
3206 out at early boot. This is used for diagnostic
3207 purposes, to verify correct tree setup.
3209 rcutree.gp_cleanup_delay= [KNL]
3210 Set the number of jiffies to delay each step of
3211 RCU grace-period cleanup. This only has effect
3212 when CONFIG_RCU_TORTURE_TEST_SLOW_CLEANUP is set.
3214 rcutree.gp_init_delay= [KNL]
3215 Set the number of jiffies to delay each step of
3216 RCU grace-period initialization. This only has
3217 effect when CONFIG_RCU_TORTURE_TEST_SLOW_INIT
3220 rcutree.gp_preinit_delay= [KNL]
3221 Set the number of jiffies to delay each step of
3222 RCU grace-period pre-initialization, that is,
3223 the propagation of recent CPU-hotplug changes up
3224 the rcu_node combining tree. This only has effect
3225 when CONFIG_RCU_TORTURE_TEST_SLOW_PREINIT is set.
3227 rcutree.rcu_fanout_exact= [KNL]
3228 Disable autobalancing of the rcu_node combining
3229 tree. This is used by rcutorture, and might
3230 possibly be useful for architectures having high
3231 cache-to-cache transfer latencies.
3233 rcutree.rcu_fanout_leaf= [KNL]
3234 Change the number of CPUs assigned to each
3235 leaf rcu_node structure. Useful for very
3236 large systems, which will choose the value 64,
3237 and for NUMA systems with large remote-access
3238 latencies, which will choose a value aligned
3239 with the appropriate hardware boundaries.
3241 rcutree.jiffies_till_sched_qs= [KNL]
3242 Set required age in jiffies for a
3243 given grace period before RCU starts
3244 soliciting quiescent-state help from
3245 rcu_note_context_switch().
3247 rcutree.jiffies_till_first_fqs= [KNL]
3248 Set delay from grace-period initialization to
3249 first attempt to force quiescent states.
3250 Units are jiffies, minimum value is zero,
3251 and maximum value is HZ.
3253 rcutree.jiffies_till_next_fqs= [KNL]
3254 Set delay between subsequent attempts to force
3255 quiescent states. Units are jiffies, minimum
3256 value is one, and maximum value is HZ.
3258 rcutree.kthread_prio= [KNL,BOOT]
3259 Set the SCHED_FIFO priority of the RCU per-CPU
3260 kthreads (rcuc/N). This value is also used for
3261 the priority of the RCU boost threads (rcub/N)
3262 and for the RCU grace-period kthreads (rcu_bh,
3263 rcu_preempt, and rcu_sched). If RCU_BOOST is
3264 set, valid values are 1-99 and the default is 1
3265 (the least-favored priority). Otherwise, when
3266 RCU_BOOST is not set, valid values are 0-99 and
3267 the default is zero (non-realtime operation).
3269 rcutree.rcu_nocb_leader_stride= [KNL]
3270 Set the number of NOCB kthread groups, which
3271 defaults to the square root of the number of
3272 CPUs. Larger numbers reduces the wakeup overhead
3273 on the per-CPU grace-period kthreads, but increases
3274 that same overhead on each group's leader.
3276 rcutree.qhimark= [KNL]
3277 Set threshold of queued RCU callbacks beyond which
3278 batch limiting is disabled.
3280 rcutree.qlowmark= [KNL]
3281 Set threshold of queued RCU callbacks below which
3282 batch limiting is re-enabled.
3284 rcutree.rcu_idle_gp_delay= [KNL]
3285 Set wakeup interval for idle CPUs that have
3286 RCU callbacks (RCU_FAST_NO_HZ=y).
3288 rcutree.rcu_idle_lazy_gp_delay= [KNL]
3289 Set wakeup interval for idle CPUs that have
3290 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
3291 Lazy RCU callbacks are those which RCU can
3292 prove do nothing more than free memory.
3294 rcutree.rcu_kick_kthreads= [KNL]
3295 Cause the grace-period kthread to get an extra
3296 wake_up() if it sleeps three times longer than
3297 it should at force-quiescent-state time.
3298 This wake_up() will be accompanied by a
3299 WARN_ONCE() splat and an ftrace_dump().
3301 rcuperf.gp_exp= [KNL]
3302 Measure performance of expedited synchronous
3303 grace-period primitives.
3305 rcuperf.holdoff= [KNL]
3306 Set test-start holdoff period. The purpose of
3307 this parameter is to delay the start of the
3308 test until boot completes in order to avoid
3311 rcuperf.nreaders= [KNL]
3312 Set number of RCU readers. The value -1 selects
3313 N, where N is the number of CPUs. A value
3314 "n" less than -1 selects N-n+1, where N is again
3315 the number of CPUs. For example, -2 selects N
3316 (the number of CPUs), -3 selects N+1, and so on.
3317 A value of "n" less than or equal to -N selects
3320 rcuperf.nwriters= [KNL]
3321 Set number of RCU writers. The values operate
3322 the same as for rcuperf.nreaders.
3323 N, where N is the number of CPUs
3325 rcuperf.perf_runnable= [BOOT]
3326 Start rcuperf running at boot time.
3328 rcuperf.shutdown= [KNL]
3329 Shut the system down after performance tests
3330 complete. This is useful for hands-off automated
3333 rcuperf.perf_type= [KNL]
3334 Specify the RCU implementation to test.
3336 rcuperf.verbose= [KNL]
3337 Enable additional printk() statements.
3339 rcutorture.cbflood_inter_holdoff= [KNL]
3340 Set holdoff time (jiffies) between successive
3341 callback-flood tests.
3343 rcutorture.cbflood_intra_holdoff= [KNL]
3344 Set holdoff time (jiffies) between successive
3345 bursts of callbacks within a given callback-flood
3348 rcutorture.cbflood_n_burst= [KNL]
3349 Set the number of bursts making up a given
3350 callback-flood test. Set this to zero to
3351 disable callback-flood testing.
3353 rcutorture.cbflood_n_per_burst= [KNL]
3354 Set the number of callbacks to be registered
3355 in a given burst of a callback-flood test.
3357 rcutorture.fqs_duration= [KNL]
3358 Set duration of force_quiescent_state bursts
3361 rcutorture.fqs_holdoff= [KNL]
3362 Set holdoff time within force_quiescent_state bursts
3365 rcutorture.fqs_stutter= [KNL]
3366 Set wait time between force_quiescent_state bursts
3369 rcutorture.gp_cond= [KNL]
3370 Use conditional/asynchronous update-side
3371 primitives, if available.
3373 rcutorture.gp_exp= [KNL]
3374 Use expedited update-side primitives, if available.
3376 rcutorture.gp_normal= [KNL]
3377 Use normal (non-expedited) asynchronous
3378 update-side primitives, if available.
3380 rcutorture.gp_sync= [KNL]
3381 Use normal (non-expedited) synchronous
3382 update-side primitives, if available. If all
3383 of rcutorture.gp_cond=, rcutorture.gp_exp=,
3384 rcutorture.gp_normal=, and rcutorture.gp_sync=
3385 are zero, rcutorture acts as if is interpreted
3386 they are all non-zero.
3388 rcutorture.n_barrier_cbs= [KNL]
3389 Set callbacks/threads for rcu_barrier() testing.
3391 rcutorture.nfakewriters= [KNL]
3392 Set number of concurrent RCU writers. These just
3393 stress RCU, they don't participate in the actual
3394 test, hence the "fake".
3396 rcutorture.nreaders= [KNL]
3397 Set number of RCU readers. The value -1 selects
3398 N-1, where N is the number of CPUs. A value
3399 "n" less than -1 selects N-n-2, where N is again
3400 the number of CPUs. For example, -2 selects N
3401 (the number of CPUs), -3 selects N+1, and so on.
3403 rcutorture.object_debug= [KNL]
3404 Enable debug-object double-call_rcu() testing.
3406 rcutorture.onoff_holdoff= [KNL]
3407 Set time (s) after boot for CPU-hotplug testing.
3409 rcutorture.onoff_interval= [KNL]
3410 Set time (s) between CPU-hotplug operations, or
3411 zero to disable CPU-hotplug testing.
3413 rcutorture.shuffle_interval= [KNL]
3414 Set task-shuffle interval (s). Shuffling tasks
3415 allows some CPUs to go into dyntick-idle mode
3416 during the rcutorture test.
3418 rcutorture.shutdown_secs= [KNL]
3419 Set time (s) after boot system shutdown. This
3420 is useful for hands-off automated testing.
3422 rcutorture.stall_cpu= [KNL]
3423 Duration of CPU stall (s) to test RCU CPU stall
3424 warnings, zero to disable.
3426 rcutorture.stall_cpu_holdoff= [KNL]
3427 Time to wait (s) after boot before inducing stall.
3429 rcutorture.stat_interval= [KNL]
3430 Time (s) between statistics printk()s.
3432 rcutorture.stutter= [KNL]
3433 Time (s) to stutter testing, for example, specifying
3434 five seconds causes the test to run for five seconds,
3435 wait for five seconds, and so on. This tests RCU's
3436 ability to transition abruptly to and from idle.
3438 rcutorture.test_boost= [KNL]
3439 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
3440 "Maybe" means test if the RCU implementation
3441 under test support RCU priority boosting.
3443 rcutorture.test_boost_duration= [KNL]
3444 Duration (s) of each individual boost test.
3446 rcutorture.test_boost_interval= [KNL]
3447 Interval (s) between each boost test.
3449 rcutorture.test_no_idle_hz= [KNL]
3450 Test RCU's dyntick-idle handling. See also the
3451 rcutorture.shuffle_interval parameter.
3453 rcutorture.torture_runnable= [BOOT]
3454 Start rcutorture running at boot time.
3456 rcutorture.torture_type= [KNL]
3457 Specify the RCU implementation to test.
3459 rcutorture.verbose= [KNL]
3460 Enable additional printk() statements.
3462 rcupdate.rcu_cpu_stall_suppress= [KNL]
3463 Suppress RCU CPU stall warning messages.
3465 rcupdate.rcu_cpu_stall_timeout= [KNL]
3466 Set timeout for RCU CPU stall warning messages.
3468 rcupdate.rcu_expedited= [KNL]
3469 Use expedited grace-period primitives, for
3470 example, synchronize_rcu_expedited() instead
3471 of synchronize_rcu(). This reduces latency,
3472 but can increase CPU utilization, degrade
3473 real-time latency, and degrade energy efficiency.
3474 No effect on CONFIG_TINY_RCU kernels.
3476 rcupdate.rcu_normal= [KNL]
3477 Use only normal grace-period primitives,
3478 for example, synchronize_rcu() instead of
3479 synchronize_rcu_expedited(). This improves
3480 real-time latency, CPU utilization, and
3481 energy efficiency, but can expose users to
3482 increased grace-period latency. This parameter
3483 overrides rcupdate.rcu_expedited. No effect on
3484 CONFIG_TINY_RCU kernels.
3486 rcupdate.rcu_normal_after_boot= [KNL]
3487 Once boot has completed (that is, after
3488 rcu_end_inkernel_boot() has been invoked), use
3489 only normal grace-period primitives. No effect
3490 on CONFIG_TINY_RCU kernels.
3492 rcupdate.rcu_task_stall_timeout= [KNL]
3493 Set timeout in jiffies for RCU task stall warning
3494 messages. Disable with a value less than or equal
3497 rcupdate.rcu_self_test= [KNL]
3498 Run the RCU early boot self tests
3500 rcupdate.rcu_self_test_bh= [KNL]
3501 Run the RCU bh early boot self tests
3503 rcupdate.rcu_self_test_sched= [KNL]
3504 Run the RCU sched early boot self tests
3508 Run specified binary instead of /init from the ramdisk,
3509 used for early userspace startup. See initrd.
3512 Format (x86 or x86_64):
3513 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
3515 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
3517 Where reboot_mode is one of warm (soft) or cold (hard) or gpio,
3518 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
3519 reboot_force is either force or not specified,
3520 reboot_cpu is s[mp]#### with #### being the processor
3521 to be used for rebooting.
3524 [KNL, SMP] Set scheduler's default relax_domain_level.
3525 See Documentation/cgroup-v1/cpusets.txt.
3527 reserve= [KNL,BUGS] Force the kernel to ignore some iomem area
3529 reservetop= [X86-32]
3531 Reserves a hole at the top of the kernel virtual
3536 Set the amount of memory to reserve for BIOS at
3537 the bottom of the address space.
3539 reset_devices [KNL] Force drivers to reset the underlying device
3540 during initialization.
3543 Specify the partition device for software suspend
3545 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
3547 resume_offset= [SWSUSP]
3548 Specify the offset from the beginning of the partition
3549 given by "resume=" at which the swap header is located,
3550 in <PAGE_SIZE> units (needed only for swap files).
3551 See Documentation/power/swsusp-and-swap-files.txt
3553 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
3554 read the resume files
3556 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
3557 Useful for devices that are detected asynchronously
3558 (e.g. USB and MMC devices).
3560 hibernate= [HIBERNATION]
3561 noresume Don't check if there's a hibernation image
3562 present during boot.
3563 nocompress Don't compress/decompress hibernation images.
3564 no Disable hibernation and resume.
3565 protect_image Turn on image protection during restoration
3566 (that will set all pages holding image data
3567 during restoration read-only).
3569 retain_initrd [RAM] Keep initrd memory after extraction
3571 rfkill.default_state=
3572 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
3573 etc. communication is blocked by default.
3576 rfkill.master_switch_mode=
3577 0 The "airplane mode" button does nothing.
3578 1 The "airplane mode" button toggles between everything
3579 blocked and the previous configuration.
3580 2 The "airplane mode" button toggles between everything
3581 blocked and everything unblocked.
3583 rhash_entries= [KNL,NET]
3584 Set number of hash buckets for route cache
3587 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
3590 ro [KNL] Mount root device read-only on boot
3593 on Mark read-only kernel memory as read-only (default).
3594 off Leave read-only kernel memory writable for debugging.
3597 Enable the uart passthrough on the designated usb port
3598 on Rockchip SoCs. When active, the signals of the
3599 debug-uart get routed to the D+ and D- pins of the usb
3600 port and the regular usb controller gets disabled.
3602 root= [KNL] Root filesystem
3603 See name_to_dev_t comment in init/do_mounts.c.
3605 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
3606 mount the root filesystem
3608 rootflags= [KNL] Set root filesystem mount option string
3610 rootfstype= [KNL] Set root filesystem type
3612 rootwait [KNL] Wait (indefinitely) for root device to show up.
3613 Useful for devices that are detected asynchronously
3614 (e.g. USB and MMC devices).
3616 rproc_mem=nn[KMG][@address]
3617 [KNL,ARM,CMA] Remoteproc physical memory block.
3618 Memory area to be used by remote processor image,
3621 rw [KNL] Mount root device read-write on boot
3623 S [KNL] Run init in single mode
3625 s390_iommu= [HW,S390]
3626 Set s390 IOTLB flushing mode
3628 With strict flushing every unmap operation will result in
3629 an IOTLB flush. Default is lazy flushing before reuse,
3633 See drivers/net/irda/sa1100_ir.c.
3635 sbni= [NET] Granch SBNI12 leased line adapter
3637 sched_debug [KNL] Enables verbose scheduler debug messages.
3639 schedstats= [KNL,X86] Enable or disable scheduled statistics.
3640 Allowed values are enable and disable. This feature
3641 incurs a small amount of overhead in the scheduler
3642 but is useful for debugging and performance tuning.
3644 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
3645 xtime_lock contention on larger systems, and/or RCU lock
3646 contention on all systems with CONFIG_MAXSMP set.
3647 Format: { "0" | "1" }
3648 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
3650 Note: increases power consumption, thus should only be
3651 enabled if running jitter sensitive (HPC/RT) workloads.
3653 security= [SECURITY] Choose a security module to enable at boot.
3654 If this boot parameter is not specified, only the first
3655 security module asking for security registration will be
3656 loaded. An invalid security module name will be treated
3657 as if no module has been chosen.
3659 selinux= [SELINUX] Disable or enable SELinux at boot time.
3660 Format: { "0" | "1" }
3661 See security/selinux/Kconfig help text.
3664 Default value is set via kernel config option.
3665 If enabled at boot time, /selinux/disable can be used
3666 later to disable prior to initial policy load.
3668 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
3669 Format: { "0" | "1" }
3670 See security/apparmor/Kconfig help text
3673 Default value is set via kernel config option.
3675 serialnumber [BUGS=X86-32]
3678 Maximal number of shapers.
3686 Disable merging of slabs with similar size. May be
3687 necessary if there is some reason to distinguish
3688 allocs to different slabs. Debug options disable
3689 merging on their own.
3690 For more information see Documentation/vm/slub.txt.
3692 slab_max_order= [MM, SLAB]
3693 Determines the maximum allowed order for slabs.
3694 A high setting may cause OOMs due to memory
3695 fragmentation. Defaults to 1 for systems with
3696 more than 32MB of RAM, 0 otherwise.
3698 slub_debug[=options[,slabs]] [MM, SLUB]
3699 Enabling slub_debug allows one to determine the
3700 culprit if slab objects become corrupted. Enabling
3701 slub_debug can create guard zones around objects and
3702 may poison objects when not in use. Also tracks the
3703 last alloc / free. For more information see
3704 Documentation/vm/slub.txt.
3706 slub_memcg_sysfs= [MM, SLUB]
3707 Determines whether to enable sysfs directories for
3708 memory cgroup sub-caches. 1 to enable, 0 to disable.
3709 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON.
3710 Enabling this can lead to a very high number of debug
3711 directories and files being created under
3714 slub_max_order= [MM, SLUB]
3715 Determines the maximum allowed order for slabs.
3716 A high setting may cause OOMs due to memory
3717 fragmentation. For more information see
3718 Documentation/vm/slub.txt.
3720 slub_min_objects= [MM, SLUB]
3721 The minimum number of objects per slab. SLUB will
3722 increase the slab order up to slub_max_order to
3723 generate a sufficiently large slab able to contain
3724 the number of objects indicated. The higher the number
3725 of objects the smaller the overhead of tracking slabs
3726 and the less frequently locks need to be acquired.
3727 For more information see Documentation/vm/slub.txt.
3729 slub_min_order= [MM, SLUB]
3730 Determines the minimum page order for slabs. Must be
3731 lower than slub_max_order.
3732 For more information see Documentation/vm/slub.txt.
3734 slub_nomerge [MM, SLUB]
3735 Same with slab_nomerge. This is supported for legacy.
3736 See slab_nomerge for more information.
3739 Format: <io1>[,<io2>[,...,<io8>]]
3741 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
3742 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
3743 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
3744 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
3745 smsc-ircc2.ircc_irq= [HW] IRQ line
3746 smsc-ircc2.ircc_dma= [HW] DMA channel
3747 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
3748 0: Toshiba Satellite 1800 (GP data pin select)
3749 1: Fast pin select (default)
3752 smt [KNL,S390] Set the maximum number of threads (logical
3753 CPUs) to use per physical CPU on systems capable of
3754 symmetric multithreading (SMT). Will be capped to the
3755 actual hardware limit.
3757 Default: -1 (no limit)
3760 [KNL] Should the soft-lockup detector generate panics.
3763 softlockup_all_cpu_backtrace=
3764 [KNL] Should the soft-lockup detector generate
3765 backtraces on all cpus.
3768 sonypi.*= [HW] Sony Programmable I/O Control Device driver
3769 See Documentation/laptops/sonypi.txt
3771 spia_io_base= [HW,MTD]
3777 Enabled the stack tracer on boot up.
3779 stacktrace_filter=[function-list]
3780 [FTRACE] Limit the functions that the stack tracer
3781 will trace at boot up. function-list is a comma separated
3782 list of functions. This list can be changed at run
3783 time by the stack_trace_filter file in the debugfs
3784 tracing directory. Note, this enables stack tracing
3785 and the stacktrace above is not needed.
3789 Set the STI (builtin display/keyboard on the HP-PARISC
3790 machines) console (graphic card) which should be used
3791 as the initial boot-console.
3792 See also comment in drivers/video/console/sticore.c.
3795 See comment in drivers/video/console/sticore.c.
3798 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
3800 sunrpc.min_resvport=
3801 sunrpc.max_resvport=
3803 SunRPC servers often require that client requests
3804 originate from a privileged port (i.e. a port in the
3805 range 0 < portnr < 1024).
3806 An administrator who wishes to reserve some of these
3807 ports for other uses may adjust the range that the
3808 kernel's sunrpc client considers to be privileged
3809 using these two parameters to set the minimum and
3810 maximum port values.
3812 sunrpc.svc_rpc_per_connection_limit=
3814 Limit the number of requests that the server will
3815 process in parallel from a single connection.
3816 The default value is 0 (no limit).
3820 Control how the NFS server code allocates CPUs to
3821 service thread pools. Depending on how many NICs
3822 you have and where their interrupts are bound, this
3823 option will affect which CPUs will do NFS serving.
3824 Note: this parameter cannot be changed while the
3825 NFS server is running.
3827 auto the server chooses an appropriate mode
3828 automatically using heuristics
3829 global a single global pool contains all CPUs
3830 percpu one pool for each CPU
3831 pernode one pool for each NUMA node (equivalent
3832 to global on non-NUMA machines)
3834 sunrpc.tcp_slot_table_entries=
3835 sunrpc.udp_slot_table_entries=
3837 Sets the upper limit on the number of simultaneous
3838 RPC calls that can be sent from the client to a
3839 server. Increasing these values may allow you to
3840 improve throughput, but will also increase the
3841 amount of memory reserved for use by the client.
3843 suspend.pm_test_delay=
3845 Sets the number of seconds to remain in a suspend test
3846 mode before resuming the system (see
3847 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
3848 is set. Default value is 5.
3851 [KNL] Enable accounting of swap in memory resource
3852 controller if no parameter or 1 is given or disable
3853 it if 0 is given (See Documentation/cgroup-v1/memory.txt)
3855 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
3856 Format: { <int> | force | noforce }
3857 <int> -- Number of I/O TLB slabs
3858 force -- force using of bounce buffers even if they
3859 wouldn't be automatically used by the kernel
3860 noforce -- Never use bounce buffers (for debugging)
3864 sysfs.deprecated=0|1 [KNL]
3865 Enable/disable old style sysfs layout for old udev
3866 on older distributions. When this option is enabled
3867 very new udev will not work anymore. When this option
3868 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
3869 in older udev will not work anymore.
3870 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
3871 the kernel configuration.
3873 sysrq_always_enabled
3875 Ignore sysrq setting - this boot parameter will
3876 neutralize any effect of /proc/sys/kernel/sysrq.
3877 Useful for debugging.
3879 tcpmhash_entries= [KNL,NET]
3880 Set the number of tcp_metrics_hash slots.
3881 Default value is 8192 or 16384 depending on total
3882 ram pages. This is used to specify the TCP metrics
3883 cache size. See Documentation/networking/ip-sysctl.txt
3884 "tcp_no_metrics_save" section for more details.
3888 test_suspend= [SUSPEND][,N]
3889 Specify "mem" (for Suspend-to-RAM) or "standby" (for
3890 standby suspend) or "freeze" (for suspend type freeze)
3891 as the system sleep state during system startup with
3892 the optional capability to repeat N number of times.
3893 The system is woken from this state using a
3894 wakeup-capable RTC alarm.
3896 thash_entries= [KNL,NET]
3897 Set number of hash buckets for TCP connection
3899 thermal.act= [HW,ACPI]
3900 -1: disable all active trip points in all thermal zones
3901 <degrees C>: override all lowest active trip points
3903 thermal.crt= [HW,ACPI]
3904 -1: disable all critical trip points in all thermal zones
3905 <degrees C>: override all critical trip points
3907 thermal.nocrt= [HW,ACPI]
3908 Set to disable actions on ACPI thermal zone
3909 critical and hot trip points.
3911 thermal.off= [HW,ACPI]
3912 1: disable ACPI thermal control
3914 thermal.psv= [HW,ACPI]
3915 -1: disable all passive trip points
3916 <degrees C>: override all passive trip points to this
3919 thermal.tzp= [HW,ACPI]
3920 Specify global default ACPI thermal zone polling rate
3921 <deci-seconds>: poll all this frequency
3922 0: no polling (default)
3925 Force threading of all interrupt handlers except those
3926 marked explicitly IRQF_NO_THREAD.
3929 Enable the Transcendent memory driver if built-in.
3931 tmem.cleancache=0|1 [KNL, XEN]
3932 Default is on (1). Disable the usage of the cleancache
3933 API to send anonymous pages to the hypervisor.
3935 tmem.frontswap=0|1 [KNL, XEN]
3936 Default is on (1). Disable the usage of the frontswap
3937 API to send swap pages to the hypervisor. If disabled
3938 the selfballooning and selfshrinking are force disabled.
3940 tmem.selfballooning=0|1 [KNL, XEN]
3941 Default is on (1). Disable the driving of swap pages
3944 tmem.selfshrinking=0|1 [KNL, XEN]
3945 Default is on (1). Partial swapoff that immediately
3946 transfers pages from Xen hypervisor back to the
3947 kernel based on different criteria.
3951 Specify if the kernel should make use of the cpu
3952 topology information if the hardware supports this.
3953 The scheduler will make use of this information and
3954 e.g. base its process migration decisions on it.
3957 topology_updates= [KNL, PPC, NUMA]
3959 Specify if the kernel should ignore (off)
3960 topology updates sent by the hypervisor to this
3965 tpm_suspend_pcr=[HW,TPM]
3966 Format: integer pcr id
3967 Specify that at suspend time, the tpm driver
3968 should extend the specified pcr with zeros,
3969 as a workaround for some chips which fail to
3970 flush the last written pcr on TPM_SaveState.
3971 This will guarantee that all the other pcrs
3974 trace_buf_size=nn[KMG]
3975 [FTRACE] will set tracing buffer size on each cpu.
3977 trace_event=[event-list]
3978 [FTRACE] Set and start specified trace events in order
3979 to facilitate early boot debugging. The event-list is a
3980 comma separated list of trace events to enable. See
3981 also Documentation/trace/events.txt
3983 trace_options=[option-list]
3984 [FTRACE] Enable or disable tracer options at boot.
3985 The option-list is a comma delimited list of options
3986 that can be enabled or disabled just as if you were
3987 to echo the option name into
3989 /sys/kernel/debug/tracing/trace_options
3991 For example, to enable stacktrace option (to dump the
3992 stack trace of each event), add to the command line:
3994 trace_options=stacktrace
3996 See also Documentation/trace/ftrace.txt "trace options"
4000 Have the tracepoints sent to printk as well as the
4001 tracing ring buffer. This is useful for early boot up
4002 where the system hangs or reboots and does not give the
4003 option for reading the tracing buffer or performing a
4004 ftrace_dump_on_oops.
4006 To turn off having tracepoints sent to printk,
4007 echo 0 > /proc/sys/kernel/tracepoint_printk
4008 Note, echoing 1 into this file without the
4009 tracepoint_printk kernel cmdline option has no effect.
4013 Having tracepoints sent to printk() and activating high
4014 frequency tracepoints such as irq or sched, can cause
4015 the system to live lock.
4018 [FTRACE] enable this option to disable tracing when a
4019 warning is hit. This turns off "tracing_on". Tracing can
4020 be enabled again by echoing '1' into the "tracing_on"
4021 file located in /sys/kernel/debug/tracing/
4023 This option is useful, as it disables the trace before
4024 the WARNING dump is called, which prevents the trace to
4025 be filled with content caused by the warning output.
4027 This option can also be set at run time via the sysctl
4028 option: kernel/traceoff_on_warning
4030 transparent_hugepage=
4032 Format: [always|madvise|never]
4033 Can be used to control the default behavior of the system
4034 with respect to transparent hugepages.
4035 See Documentation/vm/transhuge.txt for more details.
4037 tsc= Disable clocksource stability checks for TSC.
4039 [x86] reliable: mark tsc clocksource as reliable, this
4040 disables clocksource verification at runtime, as well
4041 as the stability checks done at bootup. Used to enable
4042 high-resolution timer mode on older hardware, and in
4043 virtualized environment.
4044 [x86] noirqtime: Do not use TSC to do irq accounting.
4045 Used to run time disable IRQ_TIME_ACCOUNTING on any
4046 platforms where RDTSC is slow and this accounting
4049 turbografx.map[2|3]= [HW,JOY]
4050 TurboGraFX parallel port interface
4052 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
4053 See also Documentation/input/joystick-parport.txt
4055 udbg-immortal [PPC] When debugging early kernel crashes that
4056 happen after console_init() and before a proper
4057 console driver takes over, this boot options might
4058 help "seeing" what's going on.
4060 uhash_entries= [KNL,NET]
4061 Set number of hash buckets for UDP/UDP-Lite connections
4064 [USB] Ignore overcurrent events (default N).
4065 Some badly-designed motherboards generate lots of
4066 bogus events, for ports that aren't wired to
4067 anything. Set this parameter to avoid log spamming.
4068 Note that genuine overcurrent events won't be
4072 [X86] Cause panic on unknown NMI.
4074 usbcore.authorized_default=
4075 [USB] Default USB device authorization:
4076 (default -1 = authorized except for wireless USB,
4077 0 = not authorized, 1 = authorized)
4079 usbcore.autosuspend=
4080 [USB] The autosuspend time delay (in seconds) used
4081 for newly-detected USB devices (default 2). This
4082 is the time required before an idle device will be
4083 autosuspended. Devices for which the delay is set
4084 to a negative value won't be autosuspended at all.
4086 usbcore.usbfs_snoop=
4087 [USB] Set to log all usbfs traffic (default 0 = off).
4089 usbcore.usbfs_snoop_max=
4090 [USB] Maximum number of bytes to snoop in each URB
4093 usbcore.blinkenlights=
4094 [USB] Set to cycle leds on hubs (default 0 = off).
4096 usbcore.old_scheme_first=
4097 [USB] Start with the old device initialization
4098 scheme (default 0 = off).
4100 usbcore.usbfs_memory_mb=
4101 [USB] Memory limit (in MB) for buffers allocated by
4102 usbfs (default = 16, 0 = max = 2047).
4104 usbcore.use_both_schemes=
4105 [USB] Try the other device initialization scheme
4106 if the first one fails (default 1 = enabled).
4108 usbcore.initial_descriptor_timeout=
4109 [USB] Specifies timeout for the initial 64-byte
4110 USB_REQ_GET_DESCRIPTOR request in milliseconds
4111 (default 5000 = 5.0 seconds).
4113 usbcore.nousb [USB] Disable the USB subsystem
4116 [USBHID] The interval which mice are to be polled at.
4118 usb-storage.delay_use=
4119 [UMS] The delay in seconds before a new device is
4120 scanned for Logical Units (default 1).
4123 [UMS] A list of quirks entries to supplement or
4124 override the built-in unusual_devs list. List
4125 entries are separated by commas. Each entry has
4126 the form VID:PID:Flags where VID and PID are Vendor
4127 and Product ID values (4-digit hex numbers) and
4128 Flags is a set of characters, each corresponding
4129 to a common usb-storage quirk flag as follows:
4130 a = SANE_SENSE (collect more than 18 bytes
4132 b = BAD_SENSE (don't collect more than 18
4133 bytes of sense data);
4134 c = FIX_CAPACITY (decrease the reported
4135 device capacity by one sector);
4136 d = NO_READ_DISC_INFO (don't use
4137 READ_DISC_INFO command);
4138 e = NO_READ_CAPACITY_16 (don't use
4139 READ_CAPACITY_16 command);
4140 f = NO_REPORT_OPCODES (don't use report opcodes
4142 g = MAX_SECTORS_240 (don't transfer more than
4143 240 sectors at a time, uas only);
4144 h = CAPACITY_HEURISTICS (decrease the
4145 reported device capacity by one
4146 sector if the number is odd);
4147 i = IGNORE_DEVICE (don't bind to this
4149 j = NO_REPORT_LUNS (don't use report luns
4151 l = NOT_LOCKABLE (don't try to lock and
4152 unlock ejectable media);
4153 m = MAX_SECTORS_64 (don't transfer more
4154 than 64 sectors = 32 KB at a time);
4155 n = INITIAL_READ10 (force a retry of the
4156 initial READ(10) command);
4157 o = CAPACITY_OK (accept the capacity
4158 reported by the device);
4159 p = WRITE_CACHE (the device cache is ON
4161 r = IGNORE_RESIDUE (the device reports
4162 bogus residue values);
4163 s = SINGLE_LUN (the device has only one
4165 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
4166 commands, uas only);
4167 u = IGNORE_UAS (don't bind to the uas driver);
4168 w = NO_WP_DETECT (don't test whether the
4169 medium is write-protected).
4170 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
4171 even if the device claims no cache)
4172 Example: quirks=0419:aaf5:rl,0421:0433:rc
4174 user_debug= [KNL,ARM]
4176 See arch/arm/Kconfig.debug help text.
4177 1 - undefined instruction events
4179 4 - invalid data aborts
4182 Example: user_debug=31
4185 [X86] Flags controlling user PTE allocations.
4187 nohigh = do not allocate PTE pages in
4188 HIGHMEM regardless of setting
4192 On X86_32, this is an alias for vdso32=. Otherwise:
4194 vdso=1: enable VDSO (the default)
4195 vdso=0: disable VDSO mapping
4197 vdso32= [X86] Control the 32-bit vDSO
4198 vdso32=1: enable 32-bit VDSO
4199 vdso32=0 or vdso32=2: disable 32-bit VDSO
4201 See the help text for CONFIG_COMPAT_VDSO for more
4202 details. If CONFIG_COMPAT_VDSO is set, the default is
4203 vdso32=0; otherwise, the default is vdso32=1.
4205 For compatibility with older kernels, vdso32=2 is an
4208 Try vdso32=0 if you encounter an error that says:
4209 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
4212 vector=percpu: enable percpu vector domain
4214 video= [FB] Frame buffer configuration
4215 See Documentation/fb/modedb.txt.
4217 video.brightness_switch_enabled= [0,1]
4218 If set to 1, on receiving an ACPI notify event
4219 generated by hotkey, video driver will adjust brightness
4220 level and then send out the event to user space through
4221 the allocated input device; If set to 0, video driver
4222 will only send out the event without touching backlight
4227 [VMMIO] Memory mapped virtio (platform) device.
4229 <size>@<baseaddr>:<irq>[:<id>]
4231 <size> := size (can use standard suffixes
4233 <baseaddr> := physical base address
4234 <irq> := interrupt number (as passed to
4236 <id> := (optional) platform device id
4238 virtio_mmio.device=1K@0x100b0000:48:7
4240 Can be used multiple times for multiple devices.
4242 vga= [BOOT,X86-32] Select a particular video mode
4243 See Documentation/x86/boot.txt and
4244 Documentation/svga.txt.
4245 Use vga=ask for menu.
4246 This is actually a boot loader parameter; the value is
4247 passed to the kernel using a special protocol.
4249 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
4250 size of <nn>. This can be used to increase the
4251 minimum size (128MB on x86). It can also be used to
4252 decrease the size and leave more room for directly
4255 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
4258 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
4261 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
4265 Controls the behavior of vsyscalls (i.e. calls to
4266 fixed addresses of 0xffffffffff600x00 from legacy
4267 code). Most statically-linked binaries and older
4268 versions of glibc use these calls. Because these
4269 functions are at fixed addresses, they make nice
4270 targets for exploits that can control RIP.
4272 emulate [default] Vsyscalls turn into traps and are
4273 emulated reasonably safely.
4275 native Vsyscalls are native syscall instructions.
4276 This is a little bit faster than trapping
4277 and makes a few dynamic recompilers work
4278 better than they would in emulation mode.
4279 It also makes exploits much easier to write.
4281 none Vsyscalls don't work at all. This makes
4282 them quite hard to use for exploits but
4283 might break your system.
4285 vt.color= [VT] Default text color.
4286 Format: 0xYX, X = foreground, Y = background.
4287 Default: 0x07 = light gray on black.
4289 vt.cur_default= [VT] Default cursor shape.
4290 Format: 0xCCBBAA, where AA, BB, and CC are the same as
4291 the parameters of the <Esc>[?A;B;Cc escape sequence;
4292 see VGA-softcursor.txt. Default: 2 = underline.
4294 vt.default_blu= [VT]
4295 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
4296 Change the default blue palette of the console.
4297 This is a 16-member array composed of values
4300 vt.default_grn= [VT]
4301 Format: <green0>,<green1>,<green2>,...,<green15>
4302 Change the default green palette of the console.
4303 This is a 16-member array composed of values
4306 vt.default_red= [VT]
4307 Format: <red0>,<red1>,<red2>,...,<red15>
4308 Change the default red palette of the console.
4309 This is a 16-member array composed of values
4315 Set system-wide default UTF-8 mode for all tty's.
4316 Default is 1, i.e. UTF-8 mode is enabled for all
4317 newly opened terminals.
4319 vt.global_cursor_default=
4322 Set system-wide default for whether a cursor
4323 is shown on new VTs. Default is -1,
4324 i.e. cursors will be created by default unless
4325 overridden by individual drivers. 0 will hide
4326 cursors, 1 will display them.
4328 vt.italic= [VT] Default color for italic text; 0-15.
4331 vt.underline= [VT] Default color for underlined text; 0-15.
4334 watchdog timers [HW,WDT] For information on watchdog timers,
4335 see Documentation/watchdog/watchdog-parameters.txt
4336 or other driver-specific files in the
4337 Documentation/watchdog/ directory.
4339 workqueue.watchdog_thresh=
4340 If CONFIG_WQ_WATCHDOG is configured, workqueue can
4341 warn stall conditions and dump internal state to
4342 help debugging. 0 disables workqueue stall
4343 detection; otherwise, it's the stall threshold
4344 duration in seconds. The default value is 30 and
4345 it can be updated at runtime by writing to the
4346 corresponding sysfs file.
4348 workqueue.disable_numa
4349 By default, all work items queued to unbound
4350 workqueues are affine to the NUMA nodes they're
4351 issued on, which results in better behavior in
4352 general. If NUMA affinity needs to be disabled for
4353 whatever reason, this option can be used. Note
4354 that this also can be controlled per-workqueue for
4355 workqueues visible under /sys/bus/workqueue/.
4357 workqueue.power_efficient
4358 Per-cpu workqueues are generally preferred because
4359 they show better performance thanks to cache
4360 locality; unfortunately, per-cpu workqueues tend to
4361 be more power hungry than unbound workqueues.
4363 Enabling this makes the per-cpu workqueues which
4364 were observed to contribute significantly to power
4365 consumption unbound, leading to measurably lower
4366 power usage at the cost of small performance
4369 The default value of this parameter is determined by
4370 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
4372 workqueue.debug_force_rr_cpu
4373 Workqueue used to implicitly guarantee that work
4374 items queued without explicit CPU specified are put
4375 on the local CPU. This guarantee is no longer true
4376 and while local CPU is still preferred work items
4377 may be put on foreign CPUs. This debug option
4378 forces round-robin CPU selection to flush out
4379 usages which depend on the now broken guarantee.
4380 When enabled, memory and cache locality will be
4383 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
4384 default x2apic cluster mode on platforms
4387 x86_intel_mid_timer= [X86-32,APBT]
4388 Choose timer option for x86 Intel MID platform.
4389 Two valid options are apbt timer only and lapic timer
4390 plus one apbt timer for broadcast timer.
4391 x86_intel_mid_timer=apbt_only | lapic_and_apbt
4393 xen_512gb_limit [KNL,X86-64,XEN]
4394 Restricts the kernel running paravirtualized under Xen
4395 to use only up to 512 GB of RAM. The reason to do so is
4396 crash analysis tools and Xen tools for doing domain
4397 save/restore/migration must be enabled to handle larger
4400 xen_emul_unplug= [HW,X86,XEN]
4401 Unplug Xen emulated devices
4402 Format: [unplug0,][unplug1]
4403 ide-disks -- unplug primary master IDE devices
4404 aux-ide-disks -- unplug non-primary-master IDE devices
4405 nics -- unplug network devices
4406 all -- unplug all emulated devices (NICs and IDE disks)
4407 unnecessary -- unplugging emulated devices is
4408 unnecessary even if the host did not respond to
4410 never -- do not unplug even if version check succeeds
4412 xen_nopvspin [X86,XEN]
4413 Disables the ticketlock slowpath using Xen PV
4417 Disables the PV optimizations forcing the HVM guest to
4418 run as generic HVM guest with no PV drivers.
4420 xirc2ps_cs= [NET,PCMCIA]
4422 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]