4 The following is a consolidated list of the kernel parameters as
5 implemented by the __setup(), core_param() and module_param() macros
6 and sorted into English Dictionary order (defined as ignoring all
7 punctuation and sorting digits before letters in a case insensitive
8 manner), and with descriptions where known.
10 The kernel parses parameters from the kernel command line up to "--";
11 if it doesn't recognize a parameter and it doesn't contain a '.', the
12 parameter gets passed to init: parameters with '=' go into init's
13 environment, others are passed as command line arguments to init.
14 Everything after "--" is passed as an argument to init.
16 Module parameters can be specified in two ways: via the kernel command
17 line with a module name prefix, or via modprobe, e.g.:
19 (kernel command line) usbcore.blinkenlights=1
20 (modprobe command line) modprobe usbcore blinkenlights=1
22 Parameters for modules which are built into the kernel need to be
23 specified on the kernel command line. modprobe looks through the
24 kernel command line (/proc/cmdline) and collects module parameters
25 when it loads a module, so the kernel command line can be used for
28 Hyphens (dashes) and underscores are equivalent in parameter names, so
29 log_buf_len=1M print-fatal-signals=1
30 can also be entered as
31 log-buf-len=1M print_fatal_signals=1
33 Double-quotes can be used to protect spaces in values, e.g.:
34 param="spaces in here"
36 This document may not be entirely up to date and comprehensive. The command
37 "modinfo -p ${modulename}" shows a current list of all parameters of a loadable
38 module. Loadable modules, after being loaded into the running kernel, also
39 reveal their parameters in /sys/module/${modulename}/parameters/. Some of these
40 parameters may be changed at runtime by the command
41 "echo -n ${value} > /sys/module/${modulename}/parameters/${parm}".
43 The parameters listed below are only valid if certain kernel build options were
44 enabled and if respective hardware is present. The text in square brackets at
45 the beginning of each description states the restrictions within which a
46 parameter is applicable:
48 ACPI ACPI support is enabled.
49 AGP AGP (Accelerated Graphics Port) is enabled.
50 ALSA ALSA sound support is enabled.
51 APIC APIC support is enabled.
52 APM Advanced Power Management support is enabled.
53 ARM ARM architecture is enabled.
54 AVR32 AVR32 architecture is enabled.
55 AX25 Appropriate AX.25 support is enabled.
56 BLACKFIN Blackfin architecture is enabled.
57 CLK Common clock infrastructure is enabled.
58 CMA Contiguous Memory Area support is enabled.
59 DM Device mapper support is enabled.
60 DRM Direct Rendering Management support is enabled.
61 DYNAMIC_DEBUG Build in debug messages and enable them at runtime
62 EDD BIOS Enhanced Disk Drive Services (EDD) is enabled
63 EFI EFI Partitioning (GPT) is enabled
64 EIDE EIDE/ATAPI support is enabled.
65 EVM Extended Verification Module
66 FB The frame buffer device is enabled.
67 FTRACE Function tracing enabled.
68 GCOV GCOV profiling is enabled.
69 HW Appropriate hardware is enabled.
70 IA-64 IA-64 architecture is enabled.
71 IMA Integrity measurement architecture is enabled.
72 IOSCHED More than one I/O scheduler is enabled.
73 IP_PNP IP DHCP, BOOTP, or RARP is enabled.
74 IPV6 IPv6 support is enabled.
75 ISAPNP ISA PnP code is enabled.
76 ISDN Appropriate ISDN support is enabled.
77 JOY Appropriate joystick support is enabled.
78 KGDB Kernel debugger support is enabled.
79 KVM Kernel Virtual Machine support is enabled.
80 LIBATA Libata driver is enabled
81 LP Printer support is enabled.
82 LOOP Loopback device support is enabled.
83 M68k M68k architecture is enabled.
84 These options have more detailed description inside of
85 Documentation/m68k/kernel-options.txt.
86 MDA MDA console support is enabled.
87 MIPS MIPS architecture is enabled.
88 MOUSE Appropriate mouse support is enabled.
89 MSI Message Signaled Interrupts (PCI).
90 MTD MTD (Memory Technology Device) support is enabled.
91 NET Appropriate network support is enabled.
92 NUMA NUMA support is enabled.
93 NFS Appropriate NFS support is enabled.
94 OSS OSS sound support is enabled.
95 PV_OPS A paravirtualized kernel is enabled.
96 PARIDE The ParIDE (parallel port IDE) subsystem is enabled.
97 PARISC The PA-RISC architecture is enabled.
98 PCI PCI bus support is enabled.
99 PCIE PCI Express support is enabled.
100 PCMCIA The PCMCIA subsystem is enabled.
101 PNP Plug & Play support is enabled.
102 PPC PowerPC architecture is enabled.
103 PPT Parallel port support is enabled.
104 PS2 Appropriate PS/2 support is enabled.
105 RAM RAM disk support is enabled.
106 S390 S390 architecture is enabled.
107 SCSI Appropriate SCSI support is enabled.
108 A lot of drivers have their options described inside
109 the Documentation/scsi/ sub-directory.
110 SECURITY Different security models are enabled.
111 SELINUX SELinux support is enabled.
112 APPARMOR AppArmor support is enabled.
113 SERIAL Serial support is enabled.
114 SH SuperH architecture is enabled.
115 SMP The kernel is an SMP kernel.
116 SPARC Sparc architecture is enabled.
117 SWSUSP Software suspend (hibernation) is enabled.
118 SUSPEND System suspend states are enabled.
119 TPM TPM drivers are enabled.
120 TS Appropriate touchscreen support is enabled.
121 UMS USB Mass Storage support is enabled.
122 USB USB support is enabled.
123 USBHID USB Human Interface Device support is enabled.
124 V4L Video For Linux support is enabled.
125 VMMIO Driver for memory mapped virtio devices is enabled.
126 VGA The VGA console has been enabled.
127 VT Virtual terminal support is enabled.
128 WDT Watchdog support is enabled.
129 XT IBM PC/XT MFM hard disk support is enabled.
130 X86-32 X86-32, aka i386 architecture is enabled.
131 X86-64 X86-64 architecture is enabled.
132 More X86-64 boot options can be found in
133 Documentation/x86/x86_64/boot-options.txt .
134 X86 Either 32-bit or 64-bit x86 (same as X86-32+X86-64)
135 XEN Xen support is enabled
137 In addition, the following text indicates that the option:
139 BUGS= Relates to possible processor bugs on the said processor.
140 KNL Is a kernel start-up parameter.
141 BOOT Is a boot loader parameter.
143 Parameters denoted with BOOT are actually interpreted by the boot
144 loader, and have no meaning to the kernel directly.
145 Do not modify the syntax of boot loader parameters without extreme
146 need or coordination with <Documentation/x86/boot.txt>.
148 There are also arch-specific kernel-parameters not documented here.
149 See for example <Documentation/x86/x86_64/boot-options.txt>.
151 Note that ALL kernel parameters listed below are CASE SENSITIVE, and that
152 a trailing = on the name of any parameter states that that parameter will
153 be entered as an environment variable, whereas its absence indicates that
154 it will appear as a kernel argument readable via /proc/cmdline by programs
155 running once the system is up.
157 The number of kernel parameters is not limited, but the length of the
158 complete command line (parameters including spaces etc.) is limited to
159 a fixed number of characters. This limit depends on the architecture
160 and is between 256 and 4096 characters. It is defined in the file
161 ./include/asm/setup.h as COMMAND_LINE_SIZE.
163 Finally, the [KMG] suffix is commonly described after a number of kernel
164 parameter values. These 'K', 'M', and 'G' letters represent the _binary_
165 multipliers 'Kilo', 'Mega', and 'Giga', equalling 2^10, 2^20, and 2^30
166 bytes respectively. Such letter suffixes can also be entirely omitted.
169 acpi= [HW,ACPI,X86,ARM64]
170 Advanced Configuration and Power Interface
171 Format: { force | off | strict | noirq | rsdt |
173 force -- enable ACPI if default was off
174 off -- disable ACPI if default was on
175 noirq -- do not use ACPI for IRQ routing
176 strict -- Be less tolerant of platforms that are not
177 strictly ACPI specification compliant.
178 rsdt -- prefer RSDT over (default) XSDT
179 copy_dsdt -- copy DSDT to memory
180 For ARM64, ONLY "acpi=off" or "acpi=force" are available
182 See also Documentation/power/runtime_pm.txt, pci=noacpi
184 acpi_apic_instance= [ACPI, IOAPIC]
186 2: use 2nd APIC table, if available
187 1,0: use 1st APIC table
190 acpi_backlight= [HW,ACPI]
191 acpi_backlight=vendor
193 If set to vendor, prefer vendor specific driver
194 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
195 of the ACPI video.ko driver.
197 acpica_no_return_repair [HW, ACPI]
198 Disable AML predefined validation mechanism
199 This mechanism can repair the evaluation result to make
200 the return objects more ACPI specification compliant.
201 This option is useful for developers to identify the
202 root cause of an AML interpreter issue when the issue
203 has something to do with the repair mechanism.
205 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
206 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
208 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
209 debug output. Bits in debug_layer correspond to a
210 _COMPONENT in an ACPI source file, e.g.,
211 #define _COMPONENT ACPI_PCI_COMPONENT
212 Bits in debug_level correspond to a level in
213 ACPI_DEBUG_PRINT statements, e.g.,
214 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
215 The debug_level mask defaults to "info". See
216 Documentation/acpi/debug.txt for more information about
217 debug layers and levels.
219 Enable processor driver info messages:
220 acpi.debug_layer=0x20000000
221 Enable PCI/PCI interrupt routing info messages:
222 acpi.debug_layer=0x400000
223 Enable AML "Debug" output, i.e., stores to the Debug
224 object while interpreting AML:
225 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
226 Enable all messages related to ACPI hardware:
227 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
229 Some values produce so much output that the system is
230 unusable. The "log_buf_len" parameter may be useful
231 if you need to capture more output.
233 acpi_enforce_resources= [ACPI]
234 { strict | lax | no }
235 Check for resource conflicts between native drivers
236 and ACPI OperationRegions (SystemIO and SystemMemory
237 only). IO ports and memory declared in ACPI might be
238 used by the ACPI subsystem in arbitrary AML code and
239 can interfere with legacy drivers.
240 strict (default): access to resources claimed by ACPI
241 is denied; legacy drivers trying to access reserved
242 resources will fail to bind to device using them.
243 lax: access to resources claimed by ACPI is allowed;
244 legacy drivers trying to access reserved resources
245 will bind successfully but a warning message is logged.
246 no: ACPI OperationRegions are not marked as reserved,
247 no further checks are performed.
249 acpi_force_table_verification [HW,ACPI]
250 Enable table checksum verification during early stage.
251 By default, this is disabled due to x86 early mapping
254 acpi_irq_balance [HW,ACPI]
255 ACPI will balance active IRQs
258 acpi_irq_nobalance [HW,ACPI]
259 ACPI will not move active IRQs (default)
262 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
263 Format: <irq>,<irq>...
265 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
267 Format: <irq>,<irq>...
269 acpi_no_auto_serialize [HW,ACPI]
270 Disable auto-serialization of AML methods
271 AML control methods that contain the opcodes to create
272 named objects will be marked as "Serialized" by the
273 auto-serialization feature.
274 This feature is enabled by default.
275 This option allows to turn off the feature.
277 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
280 acpi_no_static_ssdt [HW,ACPI]
281 Disable installation of static SSDTs at early boot time
282 By default, SSDTs contained in the RSDT/XSDT will be
283 installed automatically and they will appear under
284 /sys/firmware/acpi/tables.
285 This option turns off this feature.
286 Note that specifying this option does not affect
287 dynamic table installation which will install SSDT
288 tables to /sys/firmware/acpi/tables/dynamic.
290 acpi_rsdp= [ACPI,EFI,KEXEC]
291 Pass the RSDP address to the kernel, mostly used
292 on machines running EFI runtime service to boot the
293 second kernel for kdump.
295 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
296 Format: To spoof as Windows 98: ="Microsoft Windows"
298 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
299 of 2 which is mandated by ACPI 6) as the supported ACPI
300 specification revision (when using this switch, it may
301 be necessary to carry out a cold reboot _twice_ in a
302 row to make it take effect on the platform firmware).
304 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
305 acpi_osi="string1" # add string1
306 acpi_osi="!string2" # remove string2
307 acpi_osi=!* # remove all strings
308 acpi_osi=! # disable all built-in OS vendor
310 acpi_osi= # disable all strings
312 'acpi_osi=!' can be used in combination with single or
313 multiple 'acpi_osi="string1"' to support specific OS
314 vendor string(s). Note that such command can only
315 affect the default state of the OS vendor strings, thus
316 it cannot affect the default state of the feature group
317 strings and the current state of the OS vendor strings,
318 specifying it multiple times through kernel command line
319 is meaningless. This command is useful when one do not
320 care about the state of the feature group strings which
321 should be controlled by the OSPM.
323 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
324 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
325 can make '_OSI("Windows 2000")' TRUE.
327 'acpi_osi=' cannot be used in combination with other
328 'acpi_osi=' command lines, the _OSI method will not
329 exist in the ACPI namespace. NOTE that such command can
330 only affect the _OSI support state, thus specifying it
331 multiple times through kernel command line is also
334 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
337 'acpi_osi=!*' can be used in combination with single or
338 multiple 'acpi_osi="string1"' to support specific
339 string(s). Note that such command can affect the
340 current state of both the OS vendor strings and the
341 feature group strings, thus specifying it multiple times
342 through kernel command line is meaningful. But it may
343 still not able to affect the final state of a string if
344 there are quirks related to this string. This command
345 is useful when one want to control the state of the
346 feature group strings to debug BIOS issues related to
349 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
350 '_OSI("Module Device")' FALSE.
351 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
352 '_OSI("Module Device")' TRUE.
353 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
355 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
357 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
358 they all will make '_OSI("Windows 2000")' TRUE.
361 Override the pmtimer bug detection: force the kernel
362 to assume that this machine's pmtimer latches its value
363 and always returns good values.
365 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
366 Format: { level | edge | high | low }
368 acpi_skip_timer_override [HW,ACPI]
369 Recognize and ignore IRQ0/pin2 Interrupt Override.
370 For broken nForce2 BIOS resulting in XT-PIC timer.
372 acpi_sleep= [HW,ACPI] Sleep options
373 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
374 old_ordering, nonvs, sci_force_enable }
375 See Documentation/power/video.txt for information on
377 s3_beep is for debugging; it makes the PC's speaker beep
378 as soon as the kernel's real-mode entry point is called.
379 s4_nohwsig prevents ACPI hardware signature from being
380 used during resume from hibernation.
381 old_ordering causes the ACPI 1.0 ordering of the _PTS
382 control method, with respect to putting devices into
383 low power states, to be enforced (the ACPI 2.0 ordering
384 of _PTS is used by default).
385 nonvs prevents the kernel from saving/restoring the
386 ACPI NVS memory during suspend/hibernation and resume.
387 sci_force_enable causes the kernel to set SCI_EN directly
388 on resume from S1/S3 (which is against the ACPI spec,
389 but some broken systems don't work without it).
391 acpi_use_timer_override [HW,ACPI]
392 Use timer override. For some broken Nvidia NF5 boards
393 that require a timer override, but don't have HPET
395 add_efi_memmap [EFI; X86] Include EFI memory map in
396 kernel's map of available physical RAM.
399 { off | try_unsupported }
400 off: disable AGP support
401 try_unsupported: try to drive unsupported chipsets
402 (may crash computer or cause data corruption)
405 See Documentation/sound/alsa/alsa-parameters.txt
408 Allow the default userspace alignment fault handler
409 behaviour to be specified. Bit 0 enables warnings,
410 bit 1 enables fixups, and bit 2 sends a segfault.
412 align_va_addr= [X86-64]
413 Align virtual addresses by clearing slice [14:12] when
414 allocating a VMA at process creation time. This option
415 gives you up to 3% performance improvement on AMD F15h
416 machines (where it is enabled by default) for a
417 CPU-intensive style benchmark, and it can vary highly in
418 a microbenchmark depending on workload and compiler.
420 32: only for 32-bit processes
421 64: only for 64-bit processes
422 on: enable for both 32- and 64-bit processes
423 off: disable for both 32- and 64-bit processes
425 alloc_snapshot [FTRACE]
426 Allocate the ftrace snapshot buffer on boot up when the
427 main buffer is allocated. This is handy if debugging
428 and you need to use tracing_snapshot() on boot up, and
429 do not want to use tracing_snapshot_alloc() as it needs
430 to be done where GFP_KERNEL allocations are allowed.
432 amd_iommu= [HW,X86-64]
433 Pass parameters to the AMD IOMMU driver in the system.
435 fullflush - enable flushing of IO/TLB entries when
436 they are unmapped. Otherwise they are
437 flushed before they will be reused, which
439 off - do not initialize any AMD IOMMU found in
441 force_isolation - Force device isolation for all
442 devices. The IOMMU driver is not
443 allowed anymore to lift isolation
444 requirements as needed. This option
445 does not override iommu=pt
447 amd_iommu_dump= [HW,X86-64]
448 Enable AMD IOMMU driver option to dump the ACPI table
449 for AMD IOMMU. With this option enabled, AMD IOMMU
450 driver will print ACPI tables for AMD IOMMU during
451 IOMMU initialization.
453 amijoy.map= [HW,JOY] Amiga joystick support
454 Map of devices attached to JOY0DAT and JOY1DAT
456 See also Documentation/input/joystick.txt
458 analog.map= [HW,JOY] Analog joystick and gamepad support
459 Specifies type or capabilities of an analog joystick
460 connected to one of 16 gameports
461 Format: <type1>,<type2>,..<type16>
464 Power management functions (SPARCstation-4/5 + deriv.)
466 Disable APC CPU standby support. SPARCstation-Fox does
467 not play well with APC CPU idle - disable it if you have
468 APC and your system crashes randomly.
470 apic= [APIC,X86-32] Advanced Programmable Interrupt Controller
471 Change the output verbosity whilst booting
472 Format: { quiet (default) | verbose | debug }
473 Change the amount of debugging information output
474 when initialising the APIC and IO-APIC components.
477 See Documentation/networking/ipv6.txt.
479 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
480 Limit apic dumping. The parameter defines the maximal
481 number of local apics being dumped. Also it is possible
482 to set it to "all" by meaning -- no limit here.
483 Format: { 1 (default) | 2 | ... | all }.
484 The parameter valid if only apic=debug or
485 apic=verbose is specified.
486 Example: apic=debug show_lapic=all
488 apm= [APM] Advanced Power Management
489 See header of arch/x86/kernel/apm_32.c.
491 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
492 Format: <io>,<irq>,<nodeID>
496 atarimouse= [HW,MOUSE] Atari Mouse
498 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
499 EzKey and similar keyboards
501 atkbd.reset= [HW] Reset keyboard during initialization
503 atkbd.set= [HW] Select keyboard code set
504 Format: <int> (2 = AT (default), 3 = PS/2)
506 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
509 atkbd.softraw= [HW] Choose between synthetic and real raw mode
510 Format: <bool> (0 = real, 1 = synthetic (default))
512 atkbd.softrepeat= [HW]
513 Use software keyboard repeat
515 audit= [KNL] Enable the audit sub-system
516 Format: { "0" | "1" } (0 = disabled, 1 = enabled)
517 0 - kernel audit is disabled and can not be enabled
518 until the next reboot
519 unset - kernel audit is initialized but disabled and
520 will be fully enabled by the userspace auditd.
521 1 - kernel audit is initialized and partially enabled,
522 storing at most audit_backlog_limit messages in
523 RAM until it is fully enabled by the userspace
527 audit_backlog_limit= [KNL] Set the audit queue size limit.
528 Format: <int> (must be >=0)
531 baycom_epp= [HW,AX25]
534 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
536 See header of drivers/net/hamradio/baycom_par.c.
538 baycom_ser_fdx= [HW,AX25]
539 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
540 Format: <io>,<irq>,<mode>[,<baud>]
541 See header of drivers/net/hamradio/baycom_ser_fdx.c.
543 baycom_ser_hdx= [HW,AX25]
544 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
545 Format: <io>,<irq>,<mode>
546 See header of drivers/net/hamradio/baycom_ser_hdx.c.
548 blkdevparts= Manual partition parsing of block device(s) for
549 embedded devices based on command line input.
550 See Documentation/block/cmdline-partition.txt
553 Rather than attempting to online all possible CPUs at
554 boot time, only online the specified set of CPUs.
556 boot_delay= Milliseconds to delay each printk during boot.
557 Values larger than 10 seconds (10000) are changed to
561 bootmem_debug [KNL] Enable bootmem allocator debug messages.
563 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
564 bttv.radio= Most important insmod options are available as
566 bttv.pll= See Documentation/video4linux/bttv/Insmod-options
569 bulk_remove=off [PPC] This parameter disables the use of the pSeries
570 firmware feature for flushing multiple hpte entries
573 c101= [NET] Moxa C101 synchronous serial card
575 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
576 Sometimes CPU hardware bugs make them report the cache
577 size incorrectly. The kernel will attempt work arounds
578 to fix known problems, but for some CPUs it is not
579 possible to determine what the correct size should be.
580 This option provides an override for these situations.
582 ca_keys= [KEYS] This parameter identifies a specific key(s) on
583 the system trusted keyring to be used for certificate
585 format: { id:<keyid> | builtin }
587 cca= [MIPS] Override the kernel pages' cache coherency
588 algorithm. Accepted values range from 0 to 7
589 inclusive. See arch/mips/include/asm/pgtable-bits.h
590 for platform specific values (SB1, Loongson3 and
593 ccw_timeout_log [S390]
594 See Documentation/s390/CommonIO for details.
596 cgroup_disable= [KNL] Disable a particular controller
597 Format: {name of the controller(s) to disable}
598 The effects of cgroup_disable=foo are:
599 - foo isn't auto-mounted if you mount all cgroups in
601 - foo isn't visible as an individually mountable
603 {Currently only "memory" controller deal with this and
604 cut the overhead, others just disable the usage. So
605 only cgroup_disable=memory is actually worthy}
607 checkreqprot [SELINUX] Set initial checkreqprot flag value.
608 Format: { "0" | "1" }
609 See security/selinux/Kconfig help text.
610 0 -- check protection applied by kernel (includes
611 any implied execute protection).
612 1 -- check protection requested by application.
613 Default value is set via a kernel config option.
614 Value can be changed at runtime via
615 /selinux/checkreqprot.
618 See Documentation/s390/CommonIO for details.
621 Prevents the clock framework from automatically gating
622 clocks that have not been explicitly enabled by a Linux
623 device driver but are enabled in hardware at reset or
624 by the bootloader/firmware. Note that this does not
625 force such clocks to be always-on nor does it reserve
626 those clocks in any way. This parameter is useful for
627 debug and development, but should not be needed on a
628 platform with proper driver support. For more
629 information, see Documentation/clk.txt.
631 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
633 Forces specified clocksource (if available) to be used
634 when calculating gettimeofday(). If specified
635 clocksource is not available, it defaults to PIT.
636 Format: { pit | tsc | cyclone | pmtmr }
638 clocksource= Override the default clocksource
640 Override the default clocksource and use the clocksource
641 with the name specified.
642 Some clocksource names to choose from, depending on
644 [all] jiffies (this is the base, fallback clocksource)
646 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
647 pxa_timer,timer3,32k_counter,timer0_1
649 [X86-32] pit,hpet,tsc;
650 scx200_hrt on Geode; cyclone on IBM x440
658 clearcpuid=BITNUM [X86]
659 Disable CPUID feature X for the kernel. See
660 arch/x86/include/asm/cpufeatures.h for the valid bit
661 numbers. Note the Linux specific bits are not necessarily
662 stable over kernel options, but the vendor specific
664 Also note that user programs calling CPUID directly
665 or using the feature without checking anything
666 will still see it. This just prevents it from
667 being used by the kernel or shown in /proc/cpuinfo.
668 Also note the kernel might malfunction if you disable
671 cma=nn[MG]@[start[MG][-end[MG]]]
673 Sets the size of kernel global memory area for
674 contiguous memory allocations and optionally the
675 placement constraint by the physical address range of
676 memory allocations. A value of 0 disables CMA
677 altogether. For more information, see
678 include/linux/dma-contiguous.h
680 cmo_free_hint= [PPC] Format: { yes | no }
681 Specify whether pages are marked as being inactive
682 when they are freed. This is used in CMO environments
683 to determine OS memory pressure for page stealing by
687 coherent_pool=nn[KMG] [ARM,KNL]
688 Sets the size of memory pool for coherent, atomic dma
689 allocations, by default set to 256K.
691 code_bytes [X86] How many bytes of object code to print
696 com20020= [HW,NET] ARCnet - COM20020 chipset
698 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
700 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
704 ARCnet - COM90xx chipset (memory-mapped buffers)
705 Format: <io>[,<irq>[,<memstart>]]
707 condev= [HW,S390] console device
710 console= [KNL] Output console device and options.
712 tty<n> Use the virtual console device <n>.
716 Use the specified serial port. The options are of
717 the form "bbbbpnf", where "bbbb" is the baud rate,
718 "p" is parity ("n", "o", or "e"), "n" is number of
719 bits, and "f" is flow control ("r" for RTS or
720 omit it). Default is "9600n8".
722 See Documentation/serial-console.txt for more
724 Documentation/networking/netconsole.txt for an
727 uart[8250],io,<addr>[,options]
728 uart[8250],mmio,<addr>[,options]
729 uart[8250],mmio32,<addr>[,options]
730 uart[8250],0x<addr>[,options]
731 Start an early, polled-mode console on the 8250/16550
732 UART at the specified I/O port or MMIO address,
733 switching to the matching ttyS device later.
734 MMIO inter-register address stride is either 8-bit
735 (mmio) or 32-bit (mmio32).
736 If none of [io|mmio|mmio32], <addr> is assumed to be
737 equivalent to 'mmio'. 'options' are specified in the
738 same format described for ttyS above; if unspecified,
739 the h/w is not re-initialized.
741 hvc<n> Use the hypervisor console device <n>. This is for
742 both Xen and PowerPC hypervisors.
744 If the device connected to the port is not a TTY but a braille
745 device, prepend "brl," before the device type, for instance
747 For now, only VisioBraille is supported.
749 consoleblank= [KNL] The console blank (screen saver) timeout in
750 seconds. Defaults to 10*60 = 10mins. A value of 0
751 disables the blank timer.
753 core_ctl_disable_cpumask= [SMP]
754 Exempt the CPUs from being managed by core_ctl.
755 core_ctl operates on a cluster basis. So all the
756 CPUs in a given cluster must be specified to disable
757 core_ctl for that cluster.
760 [KNL] Change the default value for
761 /proc/<pid>/coredump_filter.
762 See also Documentation/filesystems/proc.txt.
764 cpuidle.off=1 [CPU_IDLE]
765 disable the cpuidle sub-system
768 [X86] Delay for N microsec between assert and de-assert
769 of APIC INIT to start processors. This delay occurs
770 on every CPU online, such as boot, and resume from suspend.
773 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
775 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
777 crashkernel=size[KMG][@offset[KMG]]
778 [KNL] Using kexec, Linux can switch to a 'crash kernel'
779 upon panic. This parameter reserves the physical
780 memory region [offset, offset + size] for that kernel
781 image. If '@offset' is omitted, then a suitable offset
782 is selected automatically. Check
783 Documentation/kdump/kdump.txt for further details.
785 crashkernel=range1:size1[,range2:size2,...][@offset]
786 [KNL] Same as above, but depends on the memory
787 in the running system. The syntax of range is
788 start-[end] where start and end are both
789 a memory unit (amount[KMG]). See also
790 Documentation/kdump/kdump.txt for an example.
792 crashkernel=size[KMG],high
793 [KNL, x86_64] range could be above 4G. Allow kernel
794 to allocate physical memory region from top, so could
795 be above 4G if system have more than 4G ram installed.
796 Otherwise memory region will be allocated below 4G, if
798 It will be ignored if crashkernel=X is specified.
799 crashkernel=size[KMG],low
800 [KNL, x86_64] range under 4G. When crashkernel=X,high
801 is passed, kernel could allocate physical memory region
802 above 4G, that cause second kernel crash on system
803 that require some amount of low memory, e.g. swiotlb
804 requires at least 64M+32K low memory, also enough extra
805 low memory is needed to make sure DMA buffers for 32-bit
806 devices won't run out. Kernel would try to allocate at
807 at least 256M below 4G automatically.
808 This one let user to specify own low range under 4G
809 for second kernel instead.
810 0: to disable low allocation.
811 It will be ignored when crashkernel=X,high is not used
812 or memory reserved is below 4G.
817 cs89x0_media= [HW,NET]
818 Format: { rj45 | aui | bnc }
821 See header of drivers/s390/block/dasd_devmap.c.
823 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
824 (one device per port)
825 Format: <port#>,<type>
826 See also Documentation/input/joystick-parport.txt
828 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
829 time. See Documentation/dynamic-debug-howto.txt for
830 details. Deprecated, see dyndbg.
832 debug [KNL] Enable kernel debugging (events log level).
835 [KNL] verbose self-tests
837 Print debugging info while doing the locking API
839 We default to 0 (no extra messages), setting it to
840 1 will print _a lot_ more information - normally
841 only useful to kernel developers.
843 debug_objects [KNL] Enable object debugging
846 [KNL] Disable object debugging
848 debug_guardpage_minorder=
849 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
850 parameter allows control of the order of pages that will
851 be intentionally kept free (and hence protected) by the
852 buddy allocator. Bigger value increase the probability
853 of catching random memory corruption, but reduce the
854 amount of memory for normal system use. The maximum
855 possible value is MAX_ORDER/2. Setting this parameter
856 to 1 or 2 should be enough to identify most random
857 memory corruption problems caused by bugs in kernel or
858 driver code when a CPU writes to (or reads from) a
859 random memory location. Note that there exists a class
860 of memory corruptions problems caused by buggy H/W or
861 F/W or by drivers badly programing DMA (basically when
862 memory is written at bus level and the CPU MMU is
863 bypassed) which are not detectable by
864 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
865 tracking down these problems.
868 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
869 parameter enables the feature at boot time. In
870 default, it is disabled. We can avoid allocating huge
871 chunk of memory for debug pagealloc if we don't enable
872 it at boot time and the system will work mostly same
873 with the kernel built without CONFIG_DEBUG_PAGEALLOC.
874 on: enable the feature
876 debugpat [X86] Enable PAT debugging
878 decnet.addr= [HW,NET]
879 Format: <area>[,<node>]
880 See also Documentation/networking/decnet.txt.
883 [same as hugepagesz=] The size of the default
884 HugeTLB page size. This is the size represented by
885 the legacy /proc/ hugepages APIs, used for SHM, and
886 default size when mounting hugetlbfs filesystems.
887 Defaults to the default architecture's huge page size
891 Set number of hash buckets for dentry cache.
894 See Documentation/networking/ipv6.txt.
896 disable_cpu_apicid= [X86,APIC,SMP]
898 The number of initial APIC ID for the
899 corresponding CPU to be disabled at boot,
900 mostly used for the kdump 2nd kernel to
901 disable BSP to wake up multiple CPUs without
902 causing system reset or hang due to sending
905 disable_ddw [PPC/PSERIES]
906 Disable Dynamic DMA Window support. Use this if
907 to workaround buggy firmware.
910 See Documentation/networking/ipv6.txt.
912 disable_mtrr_cleanup [X86]
913 The kernel tries to adjust MTRR layout from continuous
914 to discrete, to make X server driver able to add WB
915 entry later. This parameter disables that.
917 disable_mtrr_trim [X86, Intel and AMD only]
918 By default the kernel will trim any uncacheable
919 memory out of your available memory pool based on
920 MTRR settings. This parameter disables that behavior,
921 possibly causing your machine to run very slowly.
923 disable_timer_pin_1 [X86]
924 Disable PIN 1 of APIC timer
925 Can be useful to work around chipset bugs.
927 dis_ucode_ldr [X86] Disable the microcode loader.
929 dm= [DM] Allows early creation of a device-mapper device.
930 See Documentation/device-mapper/boot.txt.
932 dmasound= [HW,OSS] Sound subsystem buff
934 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
935 this option disables the debugging code at boot.
937 dma_debug_entries=<number>
938 This option allows to tune the number of preallocated
939 entries for DMA-API debugging code. One entry is
940 required per DMA-API allocation. Use this if the
941 DMA-API debugging code disables itself because the
942 architectural default is too low.
944 dma_debug_driver=<driver_name>
945 With this option the DMA-API debugging driver
946 filter feature can be enabled at boot time. Just
947 pass the driver to filter for as the parameter.
948 The filter can be disabled or changed to another
949 driver later using sysfs.
951 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
952 Broken monitors, graphic adapters, KVMs and EDIDless
953 panels may send no or incorrect EDID data sets.
954 This parameter allows to specify an EDID data sets
955 in the /lib/firmware directory that are used instead.
956 Generic built-in EDID data sets are used, if one of
957 edid/1024x768.bin, edid/1280x1024.bin,
958 edid/1680x1050.bin, or edid/1920x1080.bin is given
959 and no file with the same name exists. Details and
960 instructions how to build your own EDID data are
961 available in Documentation/EDID/HOWTO.txt. An EDID
962 data set will only be used for a particular connector,
963 if its name and a colon are prepended to the EDID
964 name. Each connector may use a unique EDID data
965 set by separating the files with a comma. An EDID
966 data set with no connector name will be used for
967 any connectors not explicitly specified.
971 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
972 module.dyndbg[="val"]
973 Enable debug messages at boot time. See
974 Documentation/dynamic-debug-howto.txt for details.
976 nompx [X86] Disables Intel Memory Protection Extensions.
977 See Documentation/x86/intel_mpx.txt for more
978 information about the feature.
981 module.async_probe [KNL]
982 Enable asynchronous probe on this module.
984 early_ioremap_debug [KNL]
985 Enable debug messages in early_ioremap support. This
986 is useful for tracking down temporary early mappings
987 which are not unmapped.
989 earlycon= [KNL] Output early console device and options.
991 When used with no options, the early console is
992 determined by the stdout-path property in device
996 Start an early, polled-mode console on a cadence serial
997 port at the specified address. The cadence serial port
998 must already be setup and configured. Options are not
1001 uart[8250],io,<addr>[,options]
1002 uart[8250],mmio,<addr>[,options]
1003 uart[8250],mmio32,<addr>[,options]
1004 uart[8250],mmio32be,<addr>[,options]
1005 uart[8250],0x<addr>[,options]
1006 Start an early, polled-mode console on the 8250/16550
1007 UART at the specified I/O port or MMIO address.
1008 MMIO inter-register address stride is either 8-bit
1009 (mmio) or 32-bit (mmio32 or mmio32be).
1010 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1011 to be equivalent to 'mmio'. 'options' are specified
1012 in the same format described for "console=ttyS<n>"; if
1013 unspecified, the h/w is not initialized.
1016 Start an early, polled-mode console on a pl011 serial
1017 port at the specified address. The pl011 serial port
1018 must already be setup and configured. Options are not
1022 Start an early, polled-mode console on an msm serial
1023 port at the specified address. The serial port
1024 must already be setup and configured. Options are not
1027 msm_serial_dm,<addr>
1028 Start an early, polled-mode console on an msm serial
1029 dm port at the specified address. The serial port
1030 must already be setup and configured. Options are not
1033 smh Use ARM semihosting calls for early console.
1041 Use early console provided by serial driver available
1042 on Samsung SoCs, requires selecting proper type and
1043 a correct base address of the selected UART port. The
1044 serial port must already be setup and configured.
1045 Options are not yet supported.
1049 Use early console provided by Freescale LP UART driver
1050 found on Freescale Vybrid and QorIQ LS1021A processors.
1051 A valid base address must be provided, and the serial
1052 port must already be setup and configured.
1054 earlyprintk= [X86,SH,BLACKFIN,ARM,M68k]
1058 earlyprintk=serial[,ttySn[,baudrate]]
1059 earlyprintk=serial[,0x...[,baudrate]]
1060 earlyprintk=ttySn[,baudrate]
1061 earlyprintk=dbgp[debugController#]
1062 earlyprintk=pciserial,bus:device.function[,baudrate]
1064 earlyprintk is useful when the kernel crashes before
1065 the normal console is initialized. It is not enabled by
1066 default because it has some cosmetic problems.
1068 Append ",keep" to not disable it when the real console
1071 Only one of vga, efi, serial, or usb debug port can
1074 Currently only ttyS0 and ttyS1 may be specified by
1075 name. Other I/O ports may be explicitly specified
1076 on some architectures (x86 and arm at least) by
1077 replacing ttySn with an I/O port address, like this:
1078 earlyprintk=serial,0x1008,115200
1079 You can find the port for a given device in
1080 /proc/tty/driver/serial:
1081 2: uart:ST16650V2 port:00001008 irq:18 ...
1083 Interaction with the standard serial driver is not
1086 The VGA and EFI output is eventually overwritten by
1089 The xen output can only be used by Xen PV guests.
1091 edac_report= [HW,EDAC] Control how to report EDAC event
1092 Format: {"on" | "off" | "force"}
1093 on: enable EDAC to report H/W event. May be overridden
1094 by other higher priority error reporting module.
1095 off: disable H/W event reporting through EDAC.
1096 force: enforce the use of EDAC to report H/W event.
1099 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1102 This is designed to be used in conjunction with
1103 the boot argument: earlyprintk=vga
1106 Format: {"off" | "on" | "skip[mbr]"}
1109 Format: { "old_map", "nochunk", "noruntime", "debug" }
1110 old_map [X86-64]: switch to the old ioremap-based EFI
1111 runtime services mapping. 32-bit still uses this one by
1113 nochunk: disable reading files in "chunks" in the EFI
1114 boot stub, as chunking can cause problems with some
1115 firmware implementations.
1116 noruntime : disable EFI runtime services support
1117 debug: enable misc debug output
1119 efi_no_storage_paranoia [EFI; X86]
1120 Using this parameter you can use more than 50% of
1121 your efi variable storage. Use this parameter only if
1122 you are really sure that your UEFI does sane gc and
1123 fulfills the spec otherwise your board may brick.
1125 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1126 Add arbitrary attribute to specific memory range by
1127 updating original EFI memory map.
1128 Region of memory which aa attribute is added to is
1130 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1131 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1132 attribute is added to range 0x100000000-0x180000000 and
1133 0x10a0000000-0x1120000000.
1135 Using this parameter you can do debugging of EFI memmap
1136 related feature. For example, you can do debugging of
1137 Address Range Mirroring feature even if your box
1140 eisa_irq_edge= [PARISC,HW]
1141 See header of drivers/parisc/eisa.c.
1144 See comment before function elanfreq_setup() in
1145 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1148 Format: {"cfq" | "deadline" | "noop"}
1149 See Documentation/block/cfq-iosched.txt and
1150 Documentation/block/deadline-iosched.txt for details.
1152 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1153 Specifies physical address of start of kernel core
1154 image elf header and optionally the size. Generally
1155 kexec loader will pass this option to capture kernel.
1156 See Documentation/kdump/kdump.txt for details.
1158 enable_mtrr_cleanup [X86]
1159 The kernel tries to adjust MTRR layout from continuous
1160 to discrete, to make X server driver able to add WB
1161 entry later. This parameter enables that.
1163 enable_timer_pin_1 [X86]
1164 Enable PIN 1 of APIC timer
1165 Can be useful to work around chipset bugs
1166 (in particular on some ATI chipsets).
1167 The kernel tries to set a reasonable default.
1169 enforcing [SELINUX] Set initial enforcing status.
1171 See security/selinux/Kconfig help text.
1172 0 -- permissive (log only, no denials).
1173 1 -- enforcing (deny and log).
1175 Value can be changed at runtime via /selinux/enforce.
1178 Disable Error Record Serialization Table (ERST)
1181 ether= [HW,NET] Ethernet cards parameters
1182 This option is obsoleted by the "netdev=" option, which
1183 has equivalent usage. See its documentation for details.
1187 Permit 'security.evm' to be updated regardless of
1188 current integrity status.
1192 fail_make_request=[KNL]
1193 General fault injection mechanism.
1194 Format: <interval>,<probability>,<space>,<times>
1195 See also Documentation/fault-injection/.
1198 See Documentation/blockdev/floppy.txt.
1200 force_pal_cache_flush
1201 [IA-64] Avoid check_sal_cache_flush which may hang on
1202 buggy SAL_CACHE_FLUSH implementations. Using this
1203 parameter will force ia64_sal_cache_flush to call
1204 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1207 Forcefully enable Physical Address Extension (PAE).
1208 Many Pentium M systems disable PAE but may have a
1209 functionally usable PAE implementation.
1210 Warning: use of this parameter will taint the kernel
1211 and may cause unknown problems.
1214 [FTRACE] will set and start the specified tracer
1215 as early as possible in order to facilitate early
1218 ftrace_dump_on_oops[=orig_cpu]
1219 [FTRACE] will dump the trace buffers on oops.
1220 If no parameter is passed, ftrace will dump
1221 buffers of all CPUs, but if you pass orig_cpu, it will
1222 dump only the buffer of the CPU that triggered the
1225 ftrace_filter=[function-list]
1226 [FTRACE] Limit the functions traced by the function
1227 tracer at boot up. function-list is a comma separated
1228 list of functions. This list can be changed at run
1229 time by the set_ftrace_filter file in the debugfs
1232 ftrace_notrace=[function-list]
1233 [FTRACE] Do not trace the functions specified in
1234 function-list. This list can be changed at run time
1235 by the set_ftrace_notrace file in the debugfs
1238 ftrace_graph_filter=[function-list]
1239 [FTRACE] Limit the top level callers functions traced
1240 by the function graph tracer at boot up.
1241 function-list is a comma separated list of functions
1242 that can be changed at run time by the
1243 set_graph_function file in the debugfs tracing directory.
1245 ftrace_graph_notrace=[function-list]
1246 [FTRACE] Do not trace from the functions specified in
1247 function-list. This list is a comma separated list of
1248 functions that can be changed at run time by the
1249 set_graph_notrace file in the debugfs tracing directory.
1252 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1253 support via parallel port (up to 5 devices per port)
1254 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1255 See also Documentation/input/joystick-parport.txt
1259 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1263 gcov_persist= [GCOV] When non-zero (default), profiling data for
1264 kernel modules is saved and remains accessible via
1265 debugfs, even when the module is unloaded/reloaded.
1266 When zero, profiling data is discarded and associated
1267 debugfs files are removed at module unload time.
1269 goldfish [X86] Enable the goldfish android emulator platform.
1270 Don't use this when you are not running on the
1273 gpt [EFI] Forces disk with valid GPT signature but
1274 invalid Protective MBR to be treated as GPT. If the
1275 primary GPT is corrupted, it enables the backup/alternate
1276 GPT to be used instead.
1278 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1279 the "Enable 0" bit of the configuration register.
1282 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1283 the "Enable 0" bit of the configuration register.
1286 grcan.select= [HW] Select which physical interface to use.
1289 grcan.txsize= [HW] Sets the size of the tx buffer.
1290 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1292 grcan.rxsize= [HW] Sets the size of the rx buffer.
1293 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1296 hardlockup_all_cpu_backtrace=
1297 [KNL] Should the hard-lockup detector generate
1298 backtraces on all cpus.
1301 hashdist= [KNL,NUMA] Large hashes allocated during boot
1302 are distributed across NUMA nodes. Defaults on
1303 for 64-bit NUMA, off otherwise.
1304 Format: 0 | 1 (for off | on)
1306 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1308 hd= [EIDE] (E)IDE hard drive subsystem geometry
1309 Format: <cyl>,<head>,<sect>
1312 Disable Hardware Error Source Table (HEST) support;
1313 corresponding firmware-first mode error processing
1314 logic will be disabled.
1316 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1317 size of <nn>. This works even on boxes that have no
1318 highmem otherwise. This also works to reduce highmem
1319 size on bigger boxes.
1321 highres= [KNL] Enable/disable high resolution timer mode.
1322 Valid parameters: "on", "off"
1326 See Documentation/isdn/README.HiSax.
1330 hpet= [X86-32,HPET] option to control HPET usage
1331 Format: { enable (default) | disable | force |
1333 disable: disable HPET and use PIT instead
1334 force: allow force enabled of undocumented chips (ICH4,
1336 verbose: show contents of HPET registers during setup
1338 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1339 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1341 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1342 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
1343 On x86-64 and powerpc, this option can be specified
1344 multiple times interleaved with hugepages= to reserve
1345 huge pages of different sizes. Valid pages sizes on
1346 x86-64 are 2M (when the CPU supports "pse") and 1G
1347 (when the CPU supports the "pdpe1gb" cpuinfo flag).
1349 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1350 terminal devices. Valid values: 0..8
1351 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1352 If specified, z/VM IUCV HVC accepts connections
1353 from listed z/VM user IDs only.
1355 hwthread_map= [METAG] Comma-separated list of Linux cpu id to
1356 hardware thread id mappings.
1357 Format: <cpu>:<hwthread>
1360 Do not unregister boot console at start. This is only
1361 useful for debugging when something happens in the window
1362 between unregistering the boot console and initializing
1365 i2c_bus= [HW] Override the default board specific I2C bus speed
1366 or register an additional I2C bus that is not
1367 registered from board initialization code.
1371 i8042.debug [HW] Toggle i8042 debug mode
1372 i8042.unmask_kbd_data
1373 [HW] Enable printing of interrupt data from the KBD port
1374 (disabled by default, and as a pre-condition
1375 requires that i8042.debug=1 be enabled)
1376 i8042.direct [HW] Put keyboard port into non-translated mode
1377 i8042.dumbkbd [HW] Pretend that controller can only read data from
1378 keyboard and cannot control its state
1379 (Don't attempt to blink the leds)
1380 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1381 i8042.nokbd [HW] Don't check/create keyboard port
1382 i8042.noloop [HW] Disable the AUX Loopback command while probing
1384 i8042.nomux [HW] Don't check presence of an active multiplexing
1386 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1388 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1389 i8042.reset [HW] Reset the controller during init, cleanup and
1390 suspend-to-ram transitions, only during s2r
1391 transitions, or never reset
1392 Format: { 1 | Y | y | 0 | N | n }
1393 1, Y, y: always reset controller
1394 0, N, n: don't ever reset controller
1395 Default: only on s2r transitions on x86; most other
1396 architectures force reset to be always executed
1397 i8042.unlock [HW] Unlock (ignore) the keylock
1398 i8042.kbdreset [HW] Reset device connected to KBD port
1402 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1403 indicates that the driver is running on unsupported
1405 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1406 does not match list of supported models.
1408 [HW] Report power status in /proc/i8k
1409 (disabled by default)
1410 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1413 i915.invert_brightness=
1414 [DRM] Invert the sense of the variable that is used to
1415 set the brightness of the panel backlight. Normally a
1416 brightness value of 0 indicates backlight switched off,
1417 and the maximum of the brightness value sets the backlight
1418 to maximum brightness. If this parameter is set to 0
1419 (default) and the machine requires it, or this parameter
1420 is set to 1, a brightness value of 0 sets the backlight
1421 to maximum brightness, and the maximum of the brightness
1422 value switches the backlight off.
1423 -1 -- never invert brightness
1424 0 -- machine default
1425 1 -- force brightness inversion
1428 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1430 ide-core.nodma= [HW] (E)IDE subsystem
1431 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1432 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1433 .cdrom .chs .ignore_cable are additional options
1434 See Documentation/ide/ide.txt.
1436 ide-generic.probe-mask= [HW] (E)IDE subsystem
1438 Probe mask for legacy ISA IDE ports. Depending on
1439 platform up to 6 ports are supported, enabled by
1440 setting corresponding bits in the mask to 1. The
1441 default value is 0x0, which has a special meaning.
1442 On systems that have PCI, it triggers scanning the
1443 PCI bus for the first and the second port, which
1444 are then probed. On systems without PCI the value
1445 of 0x0 enables probing the two first ports as if it
1448 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1449 Claim all unknown PCI IDE storage controllers.
1452 Format: idle=poll, idle=halt, idle=nomwait
1453 Poll forces a polling idle loop that can slightly
1454 improve the performance of waking up a idle CPU, but
1455 will use a lot of power and make the system run hot.
1457 idle=halt: Halt is forced to be used for CPU idle.
1458 In such case C2/C3 won't be used again.
1459 idle=nomwait: Disable mwait for CPU C-states
1461 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1462 Format: { strict | legacy | 2008 | relaxed }
1465 Choose which programs will be accepted for execution
1466 based on the IEEE 754 NaN encoding(s) supported by
1467 the FPU and the NaN encoding requested with the value
1468 of an ELF file header flag individually set by each
1469 binary. Hardware implementations are permitted to
1470 support either or both of the legacy and the 2008 NaN
1473 Available settings are as follows:
1474 strict accept binaries that request a NaN encoding
1475 supported by the FPU
1476 legacy only accept legacy-NaN binaries, if supported
1478 2008 only accept 2008-NaN binaries, if supported
1480 relaxed accept any binaries regardless of whether
1481 supported by the FPU
1483 The FPU emulator is always able to support both NaN
1484 encodings, so if no FPU hardware is present or it has
1485 been disabled with 'nofpu', then the settings of
1486 'legacy' and '2008' strap the emulator accordingly,
1487 'relaxed' straps the emulator for both legacy-NaN and
1488 2008-NaN, whereas 'strict' enables legacy-NaN only on
1489 legacy processors and both NaN encodings on MIPS32 or
1492 The setting for ABS.fmt/NEG.fmt instruction execution
1493 mode generally follows that for the NaN encoding,
1494 except where unsupported by hardware.
1496 ignore_loglevel [KNL]
1497 Ignore loglevel setting - this will print /all/
1498 kernel messages to the console. Useful for debugging.
1499 We also add it as printk module parameter, so users
1500 could change it dynamically, usually by
1501 /sys/module/printk/parameters/ignore_loglevel.
1503 ihash_entries= [KNL]
1504 Set number of hash buckets for inode cache.
1506 ima_appraise= [IMA] appraise integrity measurements
1507 Format: { "off" | "enforce" | "fix" | "log" }
1510 ima_appraise_tcb [IMA]
1511 The builtin appraise policy appraises all files
1515 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1519 The list of supported hash algorithms is defined
1520 in crypto/hash_info.h.
1523 The builtin measurement policy to load during IMA
1524 setup. Specyfing "tcb" as the value, measures all
1525 programs exec'd, files mmap'd for exec, and all files
1526 opened with the read mode bit set by either the
1527 effective uid (euid=0) or uid=0.
1530 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1531 Load a policy which meets the needs of the Trusted
1532 Computing Base. This means IMA will measure all
1533 programs exec'd, files mmap'd for exec, and all files
1534 opened for read by uid=0.
1537 Select one of defined IMA measurements template formats.
1538 Formats: { "ima" | "ima-ng" | "ima-sig" }
1542 [IMA] Define a custom template format.
1543 Format: { "field1|...|fieldN" }
1545 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1546 Format: <min_file_size>
1547 Set the minimal file size for using asynchronous hash.
1548 If left unspecified, ahash usage is disabled.
1550 ahash performance varies for different data sizes on
1551 different crypto accelerators. This option can be used
1552 to achieve the best performance for a particular HW.
1554 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1556 Set hashing buffer size. Default: 4k.
1558 ahash performance varies for different chunk sizes on
1559 different crypto accelerators. This option can be used
1560 to achieve best performance for particular HW.
1564 Run specified binary instead of /sbin/init as init
1567 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1568 for working out where the kernel is dying during
1571 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1572 initcall functions. Useful for debugging built-in
1573 modules and initcalls.
1575 initrd= [BOOT] Specify the location of the initial ramdisk
1577 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1580 int_pln_enable [x86] Enable power limit notification interrupt
1582 integrity_audit=[IMA]
1583 Format: { "0" | "1" }
1584 0 -- basic integrity auditing messages. (Default)
1585 1 -- additional integrity auditing messages.
1587 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1589 Enable intel iommu driver.
1591 Disable intel iommu driver.
1592 igfx_off [Default Off]
1593 By default, gfx is mapped as normal device. If a gfx
1594 device has a dedicated DMAR unit, the DMAR unit is
1595 bypassed by not enabling DMAR with this option. In
1596 this case, gfx device will use physical address for
1599 With this option iommu will not optimize to look
1600 for io virtual address below 32-bit forcing dual
1601 address cycle on pci bus for cards supporting greater
1602 than 32-bit addressing. The default is to look
1603 for translation below 32-bit and if not available
1604 then look in the higher range.
1605 strict [Default Off]
1606 With this option on every unmap_single operation will
1607 result in a hardware IOTLB flush operation as opposed
1608 to batching them for performance.
1609 sp_off [Default Off]
1610 By default, super page will be supported if Intel IOMMU
1611 has the capability. With this option, super page will
1613 ecs_off [Default Off]
1614 By default, extended context tables will be supported if
1615 the hardware advertises that it has support both for the
1616 extended tables themselves, and also PASID support. With
1617 this option set, extended tables will not be used even
1618 on hardware which claims to support them.
1620 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1621 0 disables intel_idle and fall back on acpi_idle.
1622 1 to 6 specify maximum depth of C-state.
1626 Do not enable intel_pstate as the default
1627 scaling driver for the supported processors
1629 Enable intel_pstate on systems that prohibit it by default
1630 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1631 instead of acpi-cpufreq may disable platform features, such
1632 as thermal controls and power capping, that rely on ACPI
1633 P-States information being indicated to OSPM and therefore
1634 should be used with caution. This option does not work with
1635 processors that aren't supported by the intel_pstate driver
1636 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1638 Do not enable hardware P state control (HWP)
1641 Only load intel_pstate on systems which support
1642 hardware P state control (HWP) if available.
1644 intremap= [X86-64, Intel-IOMMU]
1645 on enable Interrupt Remapping (default)
1646 off disable Interrupt Remapping
1647 nosid disable Source ID checking
1649 BIOS x2APIC opt-out request will be ignored
1650 nopost disable Interrupt Posting
1652 iomem= Disable strict checking of access to MMIO memory
1653 strict regions from userspace.
1668 nobypass [PPC/POWERNV]
1669 Disable IOMMU bypass, using IOMMU for PCI devices.
1672 io7= [HW] IO7 for Marvel based alpha systems
1673 See comment before marvel_specify_io7 in
1674 arch/alpha/kernel/core_marvel.c.
1676 io_delay= [X86] I/O delay method
1678 Standard port 0x80 based delay
1680 Alternate port 0xed based delay (needed on some systems)
1682 Simple two microseconds delay
1687 See Documentation/filesystems/nfs/nfsroot.txt.
1690 When an interrupt is not handled search all handlers
1691 for it. Intended to get systems with badly broken
1695 When an interrupt is not handled search all handlers
1696 for it. Also check all handlers each timer
1697 interrupt. Intended to get systems with badly broken
1701 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1703 isolcpus= [KNL,SMP] Isolate CPUs from the general scheduler.
1705 <cpu number>,...,<cpu number>
1707 <cpu number>-<cpu number>
1708 (must be a positive range in ascending order)
1710 <cpu number>,...,<cpu number>-<cpu number>
1712 This option can be used to specify one or more CPUs
1713 to isolate from the general SMP balancing and scheduling
1714 algorithms. You can move a process onto or off an
1715 "isolated" CPU via the CPU affinity syscalls or cpuset.
1716 <cpu number> begins at 0 and the maximum value is
1717 "number of CPUs in system - 1".
1719 This option is the preferred way to isolate CPUs. The
1720 alternative -- manually setting the CPU mask of all
1721 tasks in the system -- can cause problems and
1722 suboptimal load balancer performance.
1726 ivrs_ioapic [HW,X86_64]
1727 Provide an override to the IOAPIC-ID<->DEVICE-ID
1728 mapping provided in the IVRS ACPI table. For
1729 example, to map IOAPIC-ID decimal 10 to
1730 PCI device 00:14.0 write the parameter as:
1731 ivrs_ioapic[10]=00:14.0
1733 ivrs_hpet [HW,X86_64]
1734 Provide an override to the HPET-ID<->DEVICE-ID
1735 mapping provided in the IVRS ACPI table. For
1736 example, to map HPET-ID decimal 0 to
1737 PCI device 00:14.0 write the parameter as:
1738 ivrs_hpet[0]=00:14.0
1740 js= [HW,JOY] Analog joystick
1741 See Documentation/input/joystick.txt.
1744 Enable/disable kernel and module base offset ASLR
1745 (Address Space Layout Randomization) if built into
1746 the kernel. When CONFIG_HIBERNATION is selected,
1747 kASLR is disabled by default. When kASLR is enabled,
1748 hibernation will be disabled.
1752 kernelcore=nn[KMG] [KNL,X86,IA-64,PPC] This parameter
1753 specifies the amount of memory usable by the kernel
1754 for non-movable allocations. The requested amount is
1755 spread evenly throughout all nodes in the system. The
1756 remaining memory in each node is used for Movable
1757 pages. In the event, a node is too small to have both
1758 kernelcore and Movable pages, kernelcore pages will
1759 take priority and other nodes will have a larger number
1760 of Movable pages. The Movable zone is used for the
1761 allocation of pages that may be reclaimed or moved
1762 by the page migration subsystem. This means that
1763 HugeTLB pages may not be allocated from this zone.
1764 Note that allocations like PTEs-from-HighMem still
1765 use the HighMem zone if it exists, and the Normal
1766 zone if it does not.
1768 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
1769 Format: <Controller#>[,poll interval]
1770 The controller # is the number of the ehci usb debug
1771 port as it is probed via PCI. The poll interval is
1772 optional and is the number seconds in between
1773 each poll cycle to the debug port in case you need
1774 the functionality for interrupting the kernel with
1775 gdb or control-c on the dbgp connection. When
1776 not using this parameter you use sysrq-g to break into
1777 the kernel debugger.
1779 kgdboc= [KGDB,HW] kgdb over consoles.
1780 Requires a tty driver that supports console polling,
1781 or a supported polling keyboard driver (non-usb).
1782 Serial only format: <serial_device>[,baud]
1783 keyboard only format: kbd
1784 keyboard and serial format: kbd,<serial_device>[,baud]
1785 Optional Kernel mode setting:
1786 kms, kbd format: kms,kbd
1787 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
1789 kgdbwait [KGDB] Stop kernel execution and enter the
1790 kernel debugger at the earliest opportunity.
1792 kmac= [MIPS] korina ethernet MAC address.
1793 Configure the RouterBoard 532 series on-chip
1794 Ethernet adapter MAC address.
1796 kmemleak= [KNL] Boot-time kmemleak enable/disable
1797 Valid arguments: on, off
1799 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
1802 kmemcheck= [X86] Boot-time kmemcheck enable/disable/one-shot mode
1803 Valid arguments: 0, 1, 2
1804 kmemcheck=0 (disabled)
1805 kmemcheck=1 (enabled)
1806 kmemcheck=2 (one-shot mode)
1807 Default: 2 (one-shot mode)
1809 kstack=N [X86] Print N words from the kernel stack
1812 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
1813 Default is 0 (don't ignore, but inject #GP)
1815 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
1819 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
1820 Default is 1 (enabled)
1822 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
1824 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
1826 kvm-intel.ept= [KVM,Intel] Disable extended page tables
1827 (virtualized MMU) support on capable Intel chips.
1828 Default is 1 (enabled)
1830 kvm-intel.emulate_invalid_guest_state=
1831 [KVM,Intel] Enable emulation of invalid guest states
1832 Default is 0 (disabled)
1834 kvm-intel.flexpriority=
1835 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
1836 Default is 1 (enabled)
1839 [KVM,Intel] Enable VMX nesting (nVMX).
1840 Default is 0 (disabled)
1842 kvm-intel.unrestricted_guest=
1843 [KVM,Intel] Disable unrestricted guest feature
1844 (virtualized real and unpaged mode) on capable
1845 Intel chips. Default is 1 (enabled)
1847 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
1848 feature (tagged TLBs) on capable Intel chips.
1849 Default is 1 (enabled)
1855 lapic [X86-32,APIC] Enable the local APIC even if BIOS
1858 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
1859 value for LAPIC timer one-shot implementation. Default
1860 back to the programmable timer unit in the LAPIC.
1862 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
1865 libata.dma= [LIBATA] DMA control
1866 libata.dma=0 Disable all PATA and SATA DMA
1867 libata.dma=1 PATA and SATA Disk DMA only
1868 libata.dma=2 ATAPI (CDROM) DMA only
1869 libata.dma=4 Compact Flash DMA only
1870 Combinations also work, so libata.dma=3 enables DMA
1871 for disks and CDROMs, but not CFs.
1873 libata.ignore_hpa= [LIBATA] Ignore HPA limit
1874 libata.ignore_hpa=0 keep BIOS limits (default)
1875 libata.ignore_hpa=1 ignore limits, using full disk
1877 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
1881 libata.force= [LIBATA] Force configurations. The format is comma
1882 separated list of "[ID:]VAL" where ID is
1883 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
1884 matching port, link or device. Basically, it matches
1885 the ATA ID string printed on console by libata. If
1886 the whole ID part is omitted, the last PORT and DEVICE
1887 values are used. If ID hasn't been specified yet, the
1888 configuration applies to all ports, links and devices.
1890 If only DEVICE is omitted, the parameter applies to
1891 the port and all links and devices behind it. DEVICE
1892 number of 0 either selects the first device or the
1893 first fan-out link behind PMP device. It does not
1894 select the host link. DEVICE number of 15 selects the
1895 host link and device attached to it.
1897 The VAL specifies the configuration to force. As long
1898 as there's no ambiguity shortcut notation is allowed.
1899 For example, both 1.5 and 1.5G would work for 1.5Gbps.
1900 The following configurations can be forced.
1902 * Cable type: 40c, 80c, short40c, unk, ign or sata.
1903 Any ID with matching PORT is used.
1905 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
1907 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
1908 udma[/][16,25,33,44,66,100,133] notation is also
1911 * [no]ncq: Turn on or off NCQ.
1913 * [no]ncqtrim: Turn off queued DSM TRIM.
1915 * nohrst, nosrst, norst: suppress hard, soft
1918 * rstonce: only attempt one reset during
1919 hot-unplug link recovery
1921 * dump_id: dump IDENTIFY data.
1923 * atapi_dmadir: Enable ATAPI DMADIR bridge support
1925 * disable: Disable this device.
1927 If there are multiple matching configurations changing
1928 the same attribute, the last one is used.
1930 memblock=debug [KNL] Enable memblock debug messages.
1932 load_ramdisk= [RAM] List of ramdisks to load from floppy
1933 See Documentation/blockdev/ramdisk.txt.
1935 lockd.nlm_grace_period=P [NFS] Assign grace period.
1938 lockd.nlm_tcpport=N [NFS] Assign TCP port.
1941 lockd.nlm_timeout=T [NFS] Assign timeout value.
1944 lockd.nlm_udpport=M [NFS] Assign UDP port.
1947 locktorture.nreaders_stress= [KNL]
1948 Set the number of locking read-acquisition kthreads.
1949 Defaults to being automatically set based on the
1950 number of online CPUs.
1952 locktorture.nwriters_stress= [KNL]
1953 Set the number of locking write-acquisition kthreads.
1955 locktorture.onoff_holdoff= [KNL]
1956 Set time (s) after boot for CPU-hotplug testing.
1958 locktorture.onoff_interval= [KNL]
1959 Set time (s) between CPU-hotplug operations, or
1960 zero to disable CPU-hotplug testing.
1962 locktorture.shuffle_interval= [KNL]
1963 Set task-shuffle interval (jiffies). Shuffling
1964 tasks allows some CPUs to go into dyntick-idle
1965 mode during the locktorture test.
1967 locktorture.shutdown_secs= [KNL]
1968 Set time (s) after boot system shutdown. This
1969 is useful for hands-off automated testing.
1971 locktorture.stat_interval= [KNL]
1972 Time (s) between statistics printk()s.
1974 locktorture.stutter= [KNL]
1975 Time (s) to stutter testing, for example,
1976 specifying five seconds causes the test to run for
1977 five seconds, wait for five seconds, and so on.
1978 This tests the locking primitive's ability to
1979 transition abruptly to and from idle.
1981 locktorture.torture_runnable= [BOOT]
1982 Start locktorture running at boot time.
1984 locktorture.torture_type= [KNL]
1985 Specify the locking implementation to test.
1987 locktorture.verbose= [KNL]
1988 Enable additional printk() statements.
1990 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
1993 loglevel= All Kernel Messages with a loglevel smaller than the
1994 console loglevel will be printed to the console. It can
1995 also be changed with klogd or other programs. The
1996 loglevels are defined as follows:
1998 0 (KERN_EMERG) system is unusable
1999 1 (KERN_ALERT) action must be taken immediately
2000 2 (KERN_CRIT) critical conditions
2001 3 (KERN_ERR) error conditions
2002 4 (KERN_WARNING) warning conditions
2003 5 (KERN_NOTICE) normal but significant condition
2004 6 (KERN_INFO) informational
2005 7 (KERN_DEBUG) debug-level messages
2007 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2008 in bytes. n must be a power of two and greater
2009 than the minimal size. The minimal size is defined
2010 by LOG_BUF_SHIFT kernel config parameter. There is
2011 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2012 that allows to increase the default size depending on
2013 the number of CPUs. See init/Kconfig for more details.
2015 logo.nologo [FB] Disables display of the built-in Linux logo.
2016 This may be used to provide more screen space for
2017 kernel log messages and is useful when debugging
2018 kernel boot problems.
2020 lp=0 [LP] Specify parallel ports to use, e.g,
2021 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2022 lp=reset first parallel port). 'lp=0' disables the
2023 lp=auto printer driver. 'lp=reset' (which can be
2024 specified in addition to the ports) causes
2025 attached printers to be reset. Using
2026 lp=port1,port2,... specifies the parallel ports
2027 to associate lp devices with, starting with
2028 lp0. A port specification may be 'none' to skip
2029 that lp device, or a parport name such as
2030 'parport0'. Specifying 'lp=auto' instead of a
2031 port specification list means that device IDs
2032 from each port should be examined, to see if
2033 an IEEE 1284-compliant printer is attached; if
2034 so, the driver will manage that printer.
2035 See also header of drivers/char/lp.c.
2038 Sets loops_per_jiffy to given constant, thus avoiding
2039 time-consuming boot-time autodetection (up to 250 ms per
2040 CPU). 0 enables autodetection (default). To determine
2041 the correct value for your kernel, boot with normal
2042 autodetection and see what value is printed. Note that
2043 on SMP systems the preset will be applied to all CPUs,
2044 which is likely to cause problems if your CPUs need
2045 significantly divergent settings. An incorrect value
2046 will cause delays in the kernel to be wrong, leading to
2047 unpredictable I/O errors and other breakage. Although
2048 unlikely, in the extreme case this might damage your
2052 Format: <io>,<irq>,<dma>
2054 machvec= [IA-64] Force the use of a particular machine-vector
2055 (machvec) in a generic kernel.
2056 Example: machvec=hpzx1_swiotlb
2058 machtype= [Loongson] Share the same kernel image file between different
2060 Example: machtype=lemote-yeeloong-2f-7inch
2062 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2063 than or equal to this physical address is ignored.
2065 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2066 should make use of. maxcpus=n : n >= 0 limits the
2067 kernel to using 'n' processors. n=0 is a special case,
2068 it is equivalent to "nosmp", which also disables
2071 max_loop= [LOOP] The number of loop block devices that get
2072 (loop.max_loop) unconditionally pre-created at init time. The default
2073 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2074 of statically allocating a predefined number, loop
2075 devices can be requested on-demand with the
2076 /dev/loop-control interface.
2078 mce [X86-32] Machine Check Exception
2080 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.txt
2082 md= [HW] RAID subsystems devices and level
2083 See Documentation/md.txt.
2086 Format: <first>,<last>
2087 Specifies range of consoles to be captured by the MDA.
2090 Control mitigation for the Micro-architectural Data
2091 Sampling (MDS) vulnerability.
2093 Certain CPUs are vulnerable to an exploit against CPU
2094 internal buffers which can forward information to a
2095 disclosure gadget under certain conditions.
2097 In vulnerable processors, the speculatively
2098 forwarded data can be used in a cache side channel
2099 attack, to access data to which the attacker does
2100 not have direct access.
2102 This parameter controls the MDS mitigation. The
2105 full - Enable MDS mitigation on vulnerable CPUs
2106 off - Unconditionally disable MDS mitigation
2108 Not specifying this option is equivalent to
2111 For details see: Documentation/hw-vuln/mds.rst
2113 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2114 Amount of memory to be used when the kernel is not able
2115 to see the whole system memory or for test.
2116 [X86] Work as limiting max address. Use together
2117 with memmap= to avoid physical address space collisions.
2118 Without memmap= PCI devices could be placed at addresses
2119 belonging to unused RAM.
2121 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2125 [KNL,SH] Allow user to override the default size for
2126 per-device physically contiguous DMA buffers.
2128 memmap=exactmap [KNL,X86] Enable setting of an exact
2129 E820 memory map, as specified by the user.
2130 Such memmap=exactmap lines can be constructed based on
2131 BIOS output or other requirements. See the memmap=nn@ss
2134 memmap=nn[KMG]@ss[KMG]
2135 [KNL] Force usage of a specific region of memory.
2136 Region of memory to be used is from ss to ss+nn.
2138 memmap=nn[KMG]#ss[KMG]
2139 [KNL,ACPI] Mark specific memory as ACPI data.
2140 Region of memory to be marked is from ss to ss+nn.
2142 memmap=nn[KMG]$ss[KMG]
2143 [KNL,ACPI] Mark specific memory as reserved.
2144 Region of memory to be reserved is from ss to ss+nn.
2145 Example: Exclude memory from 0x18690000-0x1869ffff
2146 memmap=64K$0x18690000
2148 memmap=0x10000$0x18690000
2150 memmap=nn[KMG]!ss[KMG]
2151 [KNL,X86] Mark specific memory as protected.
2152 Region of memory to be used, from ss to ss+nn.
2153 The memory region may be marked as e820 type 12 (0xc)
2154 and is NVDIMM or ADR memory.
2156 memory_corruption_check=0/1 [X86]
2157 Some BIOSes seem to corrupt the first 64k of
2158 memory when doing things like suspend/resume.
2159 Setting this option will scan the memory
2160 looking for corruption. Enabling this will
2161 both detect corruption and prevent the kernel
2162 from using the memory being corrupted.
2163 However, its intended as a diagnostic tool; if
2164 repeatable BIOS-originated corruption always
2165 affects the same memory, you can use memmap=
2166 to prevent the kernel from using that memory.
2168 memory_corruption_check_size=size [X86]
2169 By default it checks for corruption in the low
2170 64k, making this memory unavailable for normal
2171 use. Use this parameter to scan for
2172 corruption in more or less memory.
2174 memory_corruption_check_period=seconds [X86]
2175 By default it checks for corruption every 60
2176 seconds. Use this parameter to check at some
2177 other rate. 0 disables periodic checking.
2179 memtest= [KNL,X86,ARM] Enable memtest
2181 default : 0 <disable>
2182 Specifies the number of memtest passes to be
2183 performed. Each pass selects another test
2184 pattern from a given set of patterns. Memtest
2185 fills the memory with this pattern, validates
2186 memory contents and reserves bad memory
2187 regions that are detected.
2189 meye.*= [HW] Set MotionEye Camera parameters
2190 See Documentation/video4linux/meye.txt.
2192 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2193 Multi-Function General Purpose Timers on AMD Geode
2196 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2197 the BIOS has incorrectly applied a workaround. TinyBIOS
2198 version 0.98 is known to be affected, 0.99 fixes the
2199 problem by letting the user disable the workaround.
2203 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2204 physical address is ignored.
2206 mini2440= [ARM,HW,KNL]
2207 Format:[0..2][b][c][t]
2209 MINI2440 configuration specification:
2210 0 - The attached screen is the 3.5" TFT
2211 1 - The attached screen is the 7" TFT
2212 2 - The VGA Shield is attached (1024x768)
2213 Leaving out the screen size parameter will not load
2214 the TFT driver, and the framebuffer will be left
2216 b - Enable backlight. The TFT backlight pin will be
2217 linked to the kernel VESA blanking code and a GPIO
2218 LED. This parameter is not necessary when using the
2220 c - Enable the s3c camera interface.
2221 t - Reserved for enabling touchscreen support. The
2222 touchscreen support is not enabled in the mainstream
2223 kernel as of 2.6.30, a preliminary port can be found
2224 in the "bleeding edge" mini2440 support kernel at
2225 http://repo.or.cz/w/linux-2.6/mini2440.git
2228 [X86] Control optional mitigations for CPU
2229 vulnerabilities. This is a set of curated,
2230 arch-independent options, each of which is an
2231 aggregation of existing arch-specific options.
2234 Disable all optional CPU mitigations. This
2235 improves system performance, but it may also
2236 expose users to several CPU vulnerabilities.
2237 Equivalent to: nopti [X86]
2240 spectre_v2_user=off [X86]
2241 spec_store_bypass_disable=off [X86]
2245 Mitigate all CPU vulnerabilities, but leave SMT
2246 enabled, even if it's vulnerable. This is for
2247 users who don't want to be surprised by SMT
2248 getting disabled across kernel upgrades, or who
2249 have other ways of avoiding SMT-based attacks.
2250 Equivalent to: (default behavior)
2253 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2254 parameter allows control of the logging verbosity for
2255 the additional memory initialisation checks. A value
2256 of 0 disables mminit logging and a level of 4 will
2257 log everything. Information is printed at KERN_DEBUG
2258 so loglevel=8 may also need to be specified.
2261 [KNL] When CONFIG_MODULE_SIG is set, this means that
2262 modules without (valid) signatures will fail to load.
2263 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2264 is always true, so this option does nothing.
2267 [MOUSE] Maximum time between finger touching and
2268 leaving touchpad surface for touch to be considered
2269 a tap and be reported as a left button click (for
2270 touchpads working in absolute mode only).
2272 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2273 reporting absolute coordinates, such as tablets
2274 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2275 reporting absolute coordinates, such as tablets
2277 movablecore=nn[KMG] [KNL,X86,IA-64,PPC] This parameter
2278 is similar to kernelcore except it specifies the
2279 amount of memory used for migratable allocations.
2280 If both kernelcore and movablecore is specified,
2281 then kernelcore will be at *least* the specified
2282 value but may be more. If movablecore on its own
2283 is specified, the administrator must be careful
2284 that the amount of memory usable for all allocations
2287 movable_node [KNL,X86] Boot-time switch to enable the effects
2288 of CONFIG_MOVABLE_NODE=y. See mm/Kconfig for details.
2290 MTD_Partition= [MTD]
2291 Format: <name>,<region-number>,<size>,<offset>
2293 MTD_Region= [MTD] Format:
2294 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2297 See drivers/mtd/cmdlinepart.c.
2299 multitce=off [PPC] This parameter disables the use of the pSeries
2300 firmware feature for updating multiple TCE entries
2303 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2305 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2307 boundary - index of last SLC block on Flex-OneNAND.
2308 The remaining blocks are configured as MLC blocks.
2309 lock - Configure if Flex-OneNAND boundary should be locked.
2310 Once locked, the boundary cannot be changed.
2311 1 indicates lock status, 0 indicates unlock status.
2314 ARM/S3C2412 JIVE boot control
2316 See arch/arm/mach-s3c2412/mach-jive.c
2318 mtouchusb.raw_coordinates=
2319 [HW] Make the MicroTouch USB driver use raw coordinates
2320 ('y', default) or cooked coordinates ('n')
2322 mtrr_chunk_size=nn[KMG] [X86]
2323 used for mtrr cleanup. It is largest continuous chunk
2324 that could hold holes aka. UC entries.
2326 mtrr_gran_size=nn[KMG] [X86]
2327 Used for mtrr cleanup. It is granularity of mtrr block.
2329 Large value could prevent small alignment from
2332 mtrr_spare_reg_nr=n [X86]
2334 Range: 0,7 : spare reg number
2336 Used for mtrr cleanup. It is spare mtrr entries number.
2337 Set to 2 or more if your graphical card needs more.
2339 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2341 netdev= [NET] Network devices parameters
2342 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2343 Note that mem_start is often overloaded to mean
2344 something different and driver-specific.
2345 This usage is only documented in each driver source
2349 [NETFILTER] Enable connection tracking flow accounting
2350 0 to disable accounting
2351 1 to enable accounting
2354 nfsaddrs= [NFS] Deprecated. Use ip= instead.
2355 See Documentation/filesystems/nfs/nfsroot.txt.
2357 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
2358 See Documentation/filesystems/nfs/nfsroot.txt.
2360 nfsrootdebug [NFS] enable nfsroot debugging messages.
2361 See Documentation/filesystems/nfs/nfsroot.txt.
2363 nfs.callback_tcpport=
2364 [NFS] set the TCP port on which the NFSv4 callback
2365 channel should listen.
2368 [NFS] sets the pathname to the program which is used
2369 to update the NFS client cache entries.
2371 nfs.cache_getent_timeout=
2372 [NFS] sets the timeout after which an attempt to
2373 update a cache entry is deemed to have failed.
2375 nfs.idmap_cache_timeout=
2376 [NFS] set the maximum lifetime for idmapper cache
2380 [NFS] enable 64-bit inode numbers.
2381 If zero, the NFS client will fake up a 32-bit inode
2382 number for the readdir() and stat() syscalls instead
2383 of returning the full 64-bit number.
2384 The default is to return 64-bit inode numbers.
2386 nfs.max_session_slots=
2387 [NFSv4.1] Sets the maximum number of session slots
2388 the client will attempt to negotiate with the server.
2389 This limits the number of simultaneous RPC requests
2390 that the client can send to the NFSv4.1 server.
2391 Note that there is little point in setting this
2392 value higher than the max_tcp_slot_table_limit.
2394 nfs.nfs4_disable_idmapping=
2395 [NFSv4] When set to the default of '1', this option
2396 ensures that both the RPC level authentication
2397 scheme and the NFS level operations agree to use
2398 numeric uids/gids if the mount is using the
2399 'sec=sys' security flavour. In effect it is
2400 disabling idmapping, which can make migration from
2401 legacy NFSv2/v3 systems to NFSv4 easier.
2402 Servers that do not support this mode of operation
2403 will be autodetected by the client, and it will fall
2404 back to using the idmapper.
2405 To turn off this behaviour, set the value to '0'.
2407 [NFS4] Specify an additional fixed unique ident-
2408 ification string that NFSv4 clients can insert into
2409 their nfs_client_id4 string. This is typically a
2410 UUID that is generated at system install time.
2412 nfs.send_implementation_id =
2413 [NFSv4.1] Send client implementation identification
2414 information in exchange_id requests.
2415 If zero, no implementation identification information
2417 The default is to send the implementation identification
2420 nfs.recover_lost_locks =
2421 [NFSv4] Attempt to recover locks that were lost due
2422 to a lease timeout on the server. Please note that
2423 doing this risks data corruption, since there are
2424 no guarantees that the file will remain unchanged
2425 after the locks are lost.
2426 If you want to enable the kernel legacy behaviour of
2427 attempting to recover these locks, then set this
2429 The default parameter value of '0' causes the kernel
2430 not to attempt recovery of lost locks.
2432 nfs4.layoutstats_timer =
2433 [NFSv4.2] Change the rate at which the kernel sends
2434 layoutstats to the pNFS metadata server.
2436 Setting this to value to 0 causes the kernel to use
2437 whatever value is the default set by the layout
2438 driver. A non-zero value sets the minimum interval
2439 in seconds between layoutstats transmissions.
2441 nfsd.nfs4_disable_idmapping=
2442 [NFSv4] When set to the default of '1', the NFSv4
2443 server will return only numeric uids and gids to
2444 clients using auth_sys, and will accept numeric uids
2445 and gids from such clients. This is intended to ease
2446 migration from NFSv2/v3.
2448 objlayoutdriver.osd_login_prog=
2449 [NFS] [OBJLAYOUT] sets the pathname to the program which
2450 is used to automatically discover and login into new
2451 osd-targets. Please see:
2452 Documentation/filesystems/pnfs.txt for more explanations
2454 nmi_debug= [KNL,AVR32,SH] Specify one or more actions to take
2455 when a NMI is triggered.
2456 Format: [state][,regs][,debounce][,die]
2458 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
2459 Format: [panic,][nopanic,][num]
2461 0 - turn hardlockup detector in nmi_watchdog off
2462 1 - turn hardlockup detector in nmi_watchdog on
2463 When panic is specified, panic when an NMI watchdog
2464 timeout occurs (or 'nopanic' to override the opposite
2465 default). To disable both hard and soft lockup detectors,
2466 please see 'nowatchdog'.
2467 This is useful when you use a panic=... timeout and
2468 need the box quickly up again.
2470 netpoll.carrier_timeout=
2471 [NET] Specifies amount of time (in seconds) that
2472 netpoll should wait for a carrier. By default netpoll
2475 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
2476 emulation library even if a 387 maths coprocessor
2480 [HW] Never suspend the console
2481 Disable suspending of consoles during suspend and
2482 hibernate operations. Once disabled, debugging
2483 messages can reach various consoles while the rest
2484 of the system is being put to sleep (ie, while
2485 debugging driver suspend/resume hooks). This may
2486 not work reliably with all consoles, but is known
2487 to work with serial and VGA consoles.
2488 To facilitate more flexible debugging, we also add
2489 console_suspend, a printk module parameter to control
2490 it. Users could use console_suspend (usually
2491 /sys/module/printk/parameters/console_suspend) to
2492 turn on/off it dynamically.
2494 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
2495 caches in the slab allocator. Saves per-node memory,
2496 but will impact performance.
2500 noaltinstr [S390] Disables alternative instructions patching
2501 (CPU alternatives feature).
2503 noapic [SMP,APIC] Tells the kernel to not make use of any
2504 IOAPICs that may be present in the system.
2506 noautogroup Disable scheduler automatic task group creation.
2508 nobats [PPC] Do not use BATs for mapping kernel lowmem
2509 on "Classic" PPC cores.
2513 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
2515 nodelayacct [KNL] Disable per-task delay accounting
2517 nodisconnect [HW,SCSI,M68K] Disables SCSI disconnects.
2519 nodsp [SH] Disable hardware DSP at boot time.
2521 noefi Disable EFI runtime services support.
2526 On X86-32 available only on PAE configured kernels.
2527 noexec=on: enable non-executable mappings (default)
2528 noexec=off: disable non-executable mappings
2531 Force indicating stack and heap as non-executable or
2532 executable regardless of PT_GNU_STACK entry or CPU XI
2533 (execute inhibit) support. Valid valuess are: on, off.
2534 noexec=on: force indicating non-executable
2536 noexec=off: force indicating executable
2538 If this parameter is omitted, stack and heap will be
2539 indicated non-executable or executable as they are
2540 actually set up, which depends on PT_GNU_STACK entry
2541 and possibly other factors (for instance, CPU XI
2543 NOTE: Using noexec=on on a system without CPU XI
2544 support is not recommended since there is no actual
2545 HW support that provide non-executable stack/heap.
2546 Use only for debugging purposes and not in a
2547 production environment.
2550 Disable SMAP (Supervisor Mode Access Prevention)
2551 even if it is supported by processor.
2554 Disable SMEP (Supervisor Mode Execution Prevention)
2555 even if it is supported by processor.
2558 This affects only 32-bit executables.
2559 noexec32=on: enable non-executable mappings (default)
2560 read doesn't imply executable mappings
2561 noexec32=off: disable non-executable mappings
2562 read implies executable mappings
2564 nofpu [MIPS,SH] Disable hardware FPU at boot time.
2566 nofxsr [BUGS=X86-32] Disables x86 floating point extended
2567 register save and restore. The kernel will only save
2568 legacy floating-point registers on task switch.
2570 nohugeiomap [KNL,x86] Disable kernel huge I/O mappings.
2572 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
2573 (bounds check bypass). With this option data leaks are
2574 possible in the system.
2576 nospectre_v2 [X86,PPC_FSL_BOOK3E] Disable all mitigations for the Spectre variant 2
2577 (indirect branch prediction) vulnerability. System may
2578 allow data leaks with this option, which is equivalent
2581 nospec_store_bypass_disable
2582 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
2584 noxsave [BUGS=X86] Disables x86 extended register state save
2585 and restore using xsave. The kernel will fallback to
2586 enabling legacy floating-point and sse state.
2588 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
2589 register states. The kernel will fall back to use
2590 xsave to save the states. By using this parameter,
2591 performance of saving the states is degraded because
2592 xsave doesn't support modified optimization while
2593 xsaveopt supports it on xsaveopt enabled systems.
2595 noxsaves [X86] Disables xsaves and xrstors used in saving and
2596 restoring x86 extended register state in compacted
2597 form of xsave area. The kernel will fall back to use
2598 xsaveopt and xrstor to save and restore the states
2599 in standard form of xsave area. By using this
2600 parameter, xsave area per process might occupy more
2601 memory on xsaves enabled systems.
2603 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
2604 wfi(ARM) instruction doesn't work correctly and not to
2605 use it. This is also useful when using JTAG debugger.
2607 no_file_caps Tells the kernel not to honor file capabilities. The
2608 only way then for a file to be executed with privilege
2609 is to be setuid root or executed by root.
2611 nohalt [IA-64] Tells the kernel not to use the power saving
2612 function PAL_HALT_LIGHT when idle. This increases
2613 power-consumption. On the positive side, it reduces
2614 interrupt wake-up latency, which may improve performance
2615 in certain environments such as networked servers or
2618 nohibernate [HIBERNATION] Disable hibernation and resume.
2620 nohz= [KNL] Boottime enable/disable dynamic ticks
2621 Valid arguments: on, off
2624 nohz_full= [KNL,BOOT]
2625 In kernels built with CONFIG_NO_HZ_FULL=y, set
2626 the specified list of CPUs whose tick will be stopped
2627 whenever possible. The boot CPU will be forced outside
2628 the range to maintain the timekeeping.
2629 The CPUs in this range must also be included in the
2632 noiotrap [SH] Disables trapped I/O port accesses.
2634 noirqdebug [X86-32] Disables the code which attempts to detect and
2635 disable unhandled interrupt sources.
2637 no_timer_check [X86,APIC] Disables the code which tests for
2638 broken timer IRQ sources.
2640 noisapnp [ISAPNP] Disables ISA PnP code.
2642 noinitrd [RAM] Tells the kernel not to load any configured
2645 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
2647 [Deprecated - use intremap=off]
2651 noinvpcid [X86] Disable the INVPCID cpu feature.
2653 nojitter [IA-64] Disables jitter checking for ITC timers.
2655 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
2657 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
2660 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting.
2661 steal time is computed, but won't influence scheduler
2664 nolapic [X86-32,APIC] Do not enable or use the local APIC.
2666 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
2668 noltlbs [PPC] Do not use large page/tlb entries for kernel
2669 lowmem mapping on PPC40x.
2671 nomca [IA-64] Disable machine check abort handling
2673 nomce [X86-32] Disable Machine Check Exception
2675 nomfgpt [X86-32] Disable Multi-Function General Purpose
2676 Timer usage (for AMD Geode machines).
2678 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
2679 shutdown the other cpus. Instead use the REBOOT_VECTOR
2682 nomodule Disable module load
2684 nopat [X86] Disable PAT (page attribute table extension of
2685 pagetables) support.
2687 nopcid [X86-64] Disable the PCID cpu feature.
2689 norandmaps Don't use address space randomization. Equivalent to
2690 echo 0 > /proc/sys/kernel/randomize_va_space
2692 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
2693 with UP alternatives
2695 nordrand [X86] Disable kernel use of the RDRAND and
2696 RDSEED instructions even if they are supported
2697 by the processor. RDRAND and RDSEED are still
2698 available to user space applications.
2700 noresume [SWSUSP] Disables resume and restores original swap
2703 no-scroll [VGA] Disables scrollback.
2704 This is required for the Braillex ib80-piezo Braille
2705 reader made by F.H. Papenmeier (Germany).
2709 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
2711 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
2712 and disable the IO APIC. legacy for "maxcpus=0".
2714 nosoftlockup [KNL] Disable the soft-lockup detector.
2716 nosync [HW,M68K] Disables sync negotiation for all devices.
2718 notsc [BUGS=X86-32] Disable Time Stamp Counter
2720 nousb [USB] Disable the USB subsystem
2722 nowatchdog [KNL] Disable both lockup detectors, i.e.
2723 soft-lockup and NMI watchdog (hard-lockup).
2727 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
2729 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
2730 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
2731 Some features depend on CPU0. Known dependencies are:
2732 1. Resume from suspend/hibernate depends on CPU0.
2733 Suspend/hibernate will fail if CPU0 is offline and you
2734 need to online CPU0 before suspend/hibernate.
2735 2. PIC interrupts also depend on CPU0. CPU0 can't be
2736 removed if a PIC interrupt is detected.
2737 It's said poweroff/reboot may depend on CPU0 on some
2738 machines although I haven't seen such issues so far
2739 after CPU0 is offline on a few tested machines.
2740 If the dependencies are under your control, you can
2741 turn on cpu0_hotplug.
2743 nptcg= [IA-64] Override max number of concurrent global TLB
2744 purges which is reported from either PAL_VM_SUMMARY or
2747 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
2748 could support. nr_cpus=n : n >= 1 limits the kernel to
2749 supporting 'n' processors. Later in runtime you can not
2750 use hotplug cpu feature to put more cpu back to online.
2751 just like you compile the kernel NR_CPUS=n
2753 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
2755 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
2756 Allowed values are enable and disable
2758 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
2759 one of ['zone', 'node', 'default'] can be specified
2760 This can be set from sysctl after boot.
2761 See Documentation/sysctl/vm.txt for details.
2763 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
2764 See Documentation/debugging-via-ohci1394.txt for more
2767 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
2768 Rather than timing out after 20 ms if an EC
2769 command is not properly ACKed, override the length
2770 of the timeout. We have interrupts disabled while
2771 waiting for the ACK, so if this is set too high
2772 interrupts *may* be lost!
2774 omap_mux= [OMAP] Override bootloader pin multiplexing.
2775 Format: <mux_mode0.mode_name=value>...
2776 For example, to override I2C bus2:
2777 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
2779 oprofile.timer= [HW]
2780 Use timer interrupt instead of performance counters
2782 oprofile.cpu_type= Force an oprofile cpu type
2783 This might be useful if you have an older oprofile
2784 userland or if you want common events.
2785 Format: { arch_perfmon }
2786 arch_perfmon: [X86] Force use of architectural
2787 perfmon on Intel CPUs instead of the
2788 CPU specific event set.
2789 timer: [X86] Force use of architectural NMI
2790 timer mode (see also oprofile.timer
2791 for generic hr timer mode)
2792 [s390] Force legacy basic mode sampling
2793 (report cpu_type "timer")
2795 oops=panic Always panic on oopses. Default is to just kill the
2796 process, but there is a small probability of
2797 deadlocking the machine.
2798 This will also cause panics on machine check exceptions.
2799 Useful together with panic=30 to trigger a reboot.
2802 See Documentation/sound/oss/oss-parameters.txt
2804 page_owner= [KNL] Boot-time page_owner enabling option.
2805 Storage of the information about who allocated
2806 each page is disabled in default. With this switch,
2808 on: enable the feature
2810 page_poison= [KNL] Boot-time parameter changing the state of
2811 poisoning on the buddy allocator.
2812 off: turn off poisoning
2813 on: turn on poisoning
2815 panic= [KNL] Kernel behaviour on panic: delay <timeout>
2816 timeout > 0: seconds before rebooting
2817 timeout = 0: wait forever
2818 timeout < 0: reboot immediately
2821 panic_on_warn panic() instead of WARN(). Useful to cause kdump
2824 crash_kexec_post_notifiers
2825 Run kdump after running panic-notifiers and dumping
2826 kmsg. This only for the users who doubt kdump always
2827 succeeds in any situation.
2828 Note that this also increases risks of kdump failure,
2829 because some panic notifiers can make the crashed
2830 kernel more unstable.
2832 parkbd.port= [HW] Parallel port number the keyboard adapter is
2833 connected to, default is 0.
2835 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
2836 0 for XT, 1 for AT (default is AT).
2839 parport= [HW,PPT] Specify parallel ports. 0 disables.
2840 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
2841 Use 'auto' to force the driver to use any
2842 IRQ/DMA settings detected (the default is to
2843 ignore detected IRQ/DMA settings because of
2844 possible conflicts). You can specify the base
2845 address, IRQ, and DMA settings; IRQ and DMA
2846 should be numbers, or 'auto' (for using detected
2847 settings on that particular port), or 'nofifo'
2848 (to avoid using a FIFO even if it is detected).
2849 Parallel ports are assigned in the order they
2850 are specified on the command line, starting
2853 parport_init_mode= [HW,PPT]
2854 Configure VIA parallel port to operate in
2855 a specific mode. This is necessary on Pegasos
2856 computer where firmware has no options for setting
2857 up parallel port mode and sets it to spp.
2858 Currently this function knows 686a and 8231 chips.
2859 Format: [spp|ps2|epp|ecp|ecpepp]
2862 Halt all CPUs after the first oops has been printed for
2863 the specified number of seconds. This is to be used if
2864 your oopses keep scrolling off the screen.
2869 See header of drivers/block/paride/pcd.c.
2870 See also Documentation/blockdev/paride.txt.
2872 pci=option[,option...] [PCI] various PCI subsystem options:
2873 earlydump [X86] dump PCI config space before the kernel
2875 off [X86] don't probe for the PCI bus
2876 bios [X86-32] force use of PCI BIOS, don't access
2877 the hardware directly. Use this if your machine
2878 has a non-standard PCI host bridge.
2879 nobios [X86-32] disallow use of PCI BIOS, only direct
2880 hardware access methods are allowed. Use this
2881 if you experience crashes upon bootup and you
2882 suspect they are caused by the BIOS.
2883 conf1 [X86] Force use of PCI Configuration
2885 conf2 [X86] Force use of PCI Configuration
2887 noaer [PCIE] If the PCIEAER kernel config parameter is
2888 enabled, this kernel boot option can be used to
2889 disable the use of PCIE advanced error reporting.
2890 nodomains [PCI] Disable support for multiple PCI
2891 root domains (aka PCI segments, in ACPI-speak).
2892 nommconf [X86] Disable use of MMCONFIG for PCI
2894 check_enable_amd_mmconf [X86] check for and enable
2895 properly configured MMIO access to PCI
2896 config space on AMD family 10h CPU
2897 nomsi [MSI] If the PCI_MSI kernel config parameter is
2898 enabled, this kernel boot option can be used to
2899 disable the use of MSI interrupts system-wide.
2900 noioapicquirk [APIC] Disable all boot interrupt quirks.
2901 Safety option to keep boot IRQs enabled. This
2902 should never be necessary.
2903 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
2904 primary IO-APIC for bridges that cannot disable
2905 boot IRQs. This fixes a source of spurious IRQs
2906 when the system masks IRQs.
2907 noioapicreroute [APIC] Disable workaround that uses the
2908 boot IRQ equivalent of an IRQ that connects to
2909 a chipset where boot IRQs cannot be disabled.
2910 The opposite of ioapicreroute.
2911 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
2912 routing table. These calls are known to be buggy
2913 on several machines and they hang the machine
2914 when used, but on other computers it's the only
2915 way to get the interrupt routing table. Try
2916 this option if the kernel is unable to allocate
2917 IRQs or discover secondary PCI buses on your
2919 rom [X86] Assign address space to expansion ROMs.
2920 Use with caution as certain devices share
2921 address decoders between ROMs and other
2923 norom [X86] Do not assign address space to
2924 expansion ROMs that do not already have
2925 BIOS assigned address ranges.
2926 nobar [X86] Do not assign address space to the
2927 BARs that weren't assigned by the BIOS.
2928 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
2929 assigned automatically to PCI devices. You can
2930 make the kernel exclude IRQs of your ISA cards
2932 pirqaddr=0xAAAAA [X86] Specify the physical address
2933 of the PIRQ table (normally generated
2934 by the BIOS) if it is outside the
2935 F0000h-100000h range.
2936 lastbus=N [X86] Scan all buses thru bus #N. Can be
2937 useful if the kernel is unable to find your
2938 secondary buses and you want to tell it
2939 explicitly which ones they are.
2940 assign-busses [X86] Always assign all PCI bus
2941 numbers ourselves, overriding
2942 whatever the firmware may have done.
2943 usepirqmask [X86] Honor the possible IRQ mask stored
2944 in the BIOS $PIR table. This is needed on
2945 some systems with broken BIOSes, notably
2946 some HP Pavilion N5400 and Omnibook XE3
2947 notebooks. This will have no effect if ACPI
2948 IRQ routing is enabled.
2949 noacpi [X86] Do not use ACPI for IRQ routing
2950 or for PCI scanning.
2951 use_crs [X86] Use PCI host bridge window information
2952 from ACPI. On BIOSes from 2008 or later, this
2953 is enabled by default. If you need to use this,
2954 please report a bug.
2955 nocrs [X86] Ignore PCI host bridge windows from ACPI.
2956 If you need to use this, please report a bug.
2957 routeirq Do IRQ routing for all PCI devices.
2958 This is normally done in pci_enable_device(),
2959 so this option is a temporary workaround
2960 for broken drivers that don't call it.
2961 skip_isa_align [X86] do not align io start addr, so can
2962 handle more pci cards
2963 firmware [ARM] Do not re-enumerate the bus but instead
2964 just use the configuration from the
2965 bootloader. This is currently used on
2966 IXP2000 systems where the bus has to be
2967 configured a certain way for adjunct CPUs.
2968 noearly [X86] Don't do any early type 1 scanning.
2969 This might help on some broken boards which
2970 machine check when some devices' config space
2971 is read. But various workarounds are disabled
2972 and some IOMMU drivers will not work.
2973 bfsort Sort PCI devices into breadth-first order.
2974 This sorting is done to get a device
2975 order compatible with older (<= 2.4) kernels.
2976 nobfsort Don't sort PCI devices into breadth-first order.
2977 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
2978 tuning and use the BIOS-configured MPS defaults.
2979 pcie_bus_safe Set every device's MPS to the largest value
2980 supported by all devices below the root complex.
2981 pcie_bus_perf Set device MPS to the largest allowable MPS
2982 based on its parent bus. Also set MRRS (Max
2983 Read Request Size) to the largest supported
2984 value (no larger than the MPS that the device
2985 or bus can support) for best performance.
2986 pcie_bus_peer2peer Set every device's MPS to 128B, which
2987 every device is guaranteed to support. This
2988 configuration allows peer-to-peer DMA between
2989 any pair of devices, possibly at the cost of
2990 reduced performance. This also guarantees
2991 that hot-added devices will work.
2992 cbiosize=nn[KMG] The fixed amount of bus space which is
2993 reserved for the CardBus bridge's IO window.
2994 The default value is 256 bytes.
2995 cbmemsize=nn[KMG] The fixed amount of bus space which is
2996 reserved for the CardBus bridge's memory
2997 window. The default value is 64 megabytes.
3000 [<order of align>@][<domain>:]<bus>:<slot>.<func>[; ...]
3001 Specifies alignment and device to reassign
3002 aligned memory resources.
3003 If <order of align> is not specified,
3004 PAGE_SIZE is used as alignment.
3005 PCI-PCI bridge can be specified, if resource
3006 windows need to be expanded.
3007 ecrc= Enable/disable PCIe ECRC (transaction layer
3008 end-to-end CRC checking).
3009 bios: Use BIOS/firmware settings. This is the
3013 hpiosize=nn[KMG] The fixed amount of bus space which is
3014 reserved for hotplug bridge's IO window.
3015 Default size is 256 bytes.
3016 hpmemsize=nn[KMG] The fixed amount of bus space which is
3017 reserved for hotplug bridge's memory window.
3018 Default size is 2 megabytes.
3019 realloc= Enable/disable reallocating PCI bridge resources
3020 if allocations done by BIOS are too small to
3021 accommodate resources required by all child
3023 off: Turn realloc off
3025 realloc same as realloc=on
3026 noari do not use PCIe ARI.
3027 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3028 only look for one device below a PCIe downstream
3031 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3034 force Enable ASPM even on devices that claim not to support it.
3035 WARNING: Forcing ASPM on may cause system lockups.
3037 pcie_hp= [PCIE] PCI Express Hotplug driver options:
3038 nomsi Do not use MSI for PCI Express Native Hotplug (this
3039 makes all PCIe ports use INTx for hotplug services).
3041 pcie_ports= [PCIE] PCIe ports handling:
3042 auto Ask the BIOS whether or not to use native PCIe services
3043 associated with PCIe ports (PME, hot-plug, AER). Use
3044 them only if that is allowed by the BIOS.
3045 native Use native PCIe services associated with PCIe ports
3047 compat Treat PCIe ports as PCI-to-PCI bridges, disable the PCIe
3050 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3051 nomsi Do not use MSI for native PCIe PME signaling (this makes
3052 all PCIe root ports use INTx for all services).
3054 pcmv= [HW,PCMCIA] BadgePAD 4
3058 Keep all power-domains already enabled by bootloader on,
3059 even if no driver has claimed them. This is useful
3060 for debug and development, but should not be
3061 needed on a platform with proper driver support.
3064 See Documentation/blockdev/paride.txt.
3066 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3069 See arch/parisc/kernel/pdc_chassis.c
3071 percpu_alloc= Select which percpu first chunk allocator to use.
3072 Currently supported values are "embed" and "page".
3073 Archs may support subset or none of the selections.
3074 See comments in mm/percpu.c for details on each
3075 allocator. This parameter is primarily for debugging
3076 and performance comparison.
3079 See Documentation/blockdev/paride.txt.
3082 See Documentation/blockdev/paride.txt.
3084 pirq= [SMP,APIC] Manual mp-table setup
3085 See Documentation/x86/i386/IO-APIC.txt.
3087 plip= [PPT,NET] Parallel port network link
3088 Format: { parport<nr> | timid | 0 }
3089 See also Documentation/parport.txt.
3091 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3092 Override pmtimer IOPort with a hex value.
3096 Enable PNP debug messages (depends on the
3097 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3098 via /sys/module/pnp/parameters/debug. We always show
3099 current resource usage; turning this on also shows
3100 possible settings and some assignment information.
3106 { on | off | curr | res | no-curr | no-res }
3109 [ISAPNP] Exclude IRQs for the autoconfiguration
3112 [ISAPNP] Exclude DMAs for the autoconfiguration
3114 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3115 Ranges are in pairs (I/O port base and size).
3118 [ISAPNP] Exclude memory regions for the
3120 Ranges are in pairs (memory base and size).
3122 ports= [IP_VS_FTP] IPVS ftp helper module
3124 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3126 Format: <port>,<port>....
3128 print-fatal-signals=
3129 [KNL] debug: print fatal signals
3131 If enabled, warn about various signal handling
3132 related application anomalies: too many signals,
3133 too many POSIX.1 timers, fatal signals causing a
3136 If you hit the warning due to signal overflow,
3137 you might want to try "ulimit -i unlimited".
3141 printk.always_kmsg_dump=
3142 Trigger kmsg_dump for cases other than kernel oops or
3144 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3147 printk.time= Show timing data prefixed to each printk message line
3148 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3150 processor.max_cstate= [HW,ACPI]
3151 Limit processor to maximum C-state
3152 max_cstate=9 overrides any DMI blacklist limit.
3154 processor.nocst [HW,ACPI]
3155 Ignore the _CST method to determine C-states,
3156 instead using the legacy FADT method
3158 profile= [KNL] Enable kernel profiling via /proc/profile
3159 Format: [schedule,]<number>
3160 Param: "schedule" - profile schedule points.
3161 Param: <number> - step/bucket size as a power of 2 for
3162 statistical time based profiling.
3163 Param: "sleep" - profile D-state sleeping (millisecs).
3164 Requires CONFIG_SCHEDSTATS
3165 Param: "kvm" - profile VM exits.
3167 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3169 See Documentation/blockdev/ramdisk.txt.
3171 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3172 probe for; one of (bare|imps|exps|lifebook|any).
3173 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3175 psmouse.resetafter= [HW,MOUSE]
3176 Try to reset the device after so many bad packets
3179 [HW,MOUSE] Set desired mouse resolution, in dpi.
3180 psmouse.smartscroll=
3181 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
3182 0 = disabled, 1 = enabled (default).
3184 pstore.backend= Specify the name of the pstore backend to use
3187 See Documentation/blockdev/paride.txt.
3189 pti= [X86_64] Control Page Table Isolation of user and
3190 kernel address spaces. Disabling this feature
3191 removes hardening, but improves performance of
3192 system calls and interrupts.
3194 on - unconditionally enable
3195 off - unconditionally disable
3196 auto - kernel detects whether your CPU model is
3197 vulnerable to issues that PTI mitigates
3199 Not specifying this option is equivalent to pti=auto.
3202 Equivalent to pti=off
3205 [KNL] Number of legacy pty's. Overwrites compiled-in
3208 quiet [KNL] Disable most log messages
3213 See Documentation/md.txt.
3215 ramdisk_blocksize= [RAM]
3216 See Documentation/blockdev/ramdisk.txt.
3218 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
3219 See Documentation/blockdev/ramdisk.txt.
3222 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
3223 the specified list of CPUs to be no-callback CPUs.
3224 Invocation of these CPUs' RCU callbacks will
3225 be offloaded to "rcuox/N" kthreads created for
3226 that purpose, where "x" is "b" for RCU-bh, "p"
3227 for RCU-preempt, and "s" for RCU-sched, and "N"
3228 is the CPU number. This reduces OS jitter on the
3229 offloaded CPUs, which can be useful for HPC and
3230 real-time workloads. It can also improve energy
3231 efficiency for asymmetric multiprocessors.
3234 Rather than requiring that offloaded CPUs
3235 (specified by rcu_nocbs= above) explicitly
3236 awaken the corresponding "rcuoN" kthreads,
3237 make these kthreads poll for callbacks.
3238 This improves the real-time response for the
3239 offloaded CPUs by relieving them of the need to
3240 wake up the corresponding kthread, but degrades
3241 energy efficiency by requiring that the kthreads
3242 periodically wake up to do the polling.
3244 rcutree.blimit= [KNL]
3245 Set maximum number of finished RCU callbacks to
3246 process in one batch.
3248 rcutree.dump_tree= [KNL]
3249 Dump the structure of the rcu_node combining tree
3250 out at early boot. This is used for diagnostic
3251 purposes, to verify correct tree setup.
3253 rcutree.gp_cleanup_delay= [KNL]
3254 Set the number of jiffies to delay each step of
3255 RCU grace-period cleanup. This only has effect
3256 when CONFIG_RCU_TORTURE_TEST_SLOW_CLEANUP is set.
3258 rcutree.gp_init_delay= [KNL]
3259 Set the number of jiffies to delay each step of
3260 RCU grace-period initialization. This only has
3261 effect when CONFIG_RCU_TORTURE_TEST_SLOW_INIT
3264 rcutree.gp_preinit_delay= [KNL]
3265 Set the number of jiffies to delay each step of
3266 RCU grace-period pre-initialization, that is,
3267 the propagation of recent CPU-hotplug changes up
3268 the rcu_node combining tree. This only has effect
3269 when CONFIG_RCU_TORTURE_TEST_SLOW_PREINIT is set.
3271 rcutree.rcu_fanout_exact= [KNL]
3272 Disable autobalancing of the rcu_node combining
3273 tree. This is used by rcutorture, and might
3274 possibly be useful for architectures having high
3275 cache-to-cache transfer latencies.
3277 rcutree.rcu_fanout_leaf= [KNL]
3278 Change the number of CPUs assigned to each
3279 leaf rcu_node structure. Useful for very
3280 large systems, which will choose the value 64,
3281 and for NUMA systems with large remote-access
3282 latencies, which will choose a value aligned
3283 with the appropriate hardware boundaries.
3285 rcutree.jiffies_till_sched_qs= [KNL]
3286 Set required age in jiffies for a
3287 given grace period before RCU starts
3288 soliciting quiescent-state help from
3289 rcu_note_context_switch().
3291 rcutree.jiffies_till_first_fqs= [KNL]
3292 Set delay from grace-period initialization to
3293 first attempt to force quiescent states.
3294 Units are jiffies, minimum value is zero,
3295 and maximum value is HZ.
3297 rcutree.jiffies_till_next_fqs= [KNL]
3298 Set delay between subsequent attempts to force
3299 quiescent states. Units are jiffies, minimum
3300 value is one, and maximum value is HZ.
3302 rcutree.kthread_prio= [KNL,BOOT]
3303 Set the SCHED_FIFO priority of the RCU per-CPU
3304 kthreads (rcuc/N). This value is also used for
3305 the priority of the RCU boost threads (rcub/N)
3306 and for the RCU grace-period kthreads (rcu_bh,
3307 rcu_preempt, and rcu_sched). If RCU_BOOST is
3308 set, valid values are 1-99 and the default is 1
3309 (the least-favored priority). Otherwise, when
3310 RCU_BOOST is not set, valid values are 0-99 and
3311 the default is zero (non-realtime operation).
3313 rcutree.rcu_nocb_leader_stride= [KNL]
3314 Set the number of NOCB kthread groups, which
3315 defaults to the square root of the number of
3316 CPUs. Larger numbers reduces the wakeup overhead
3317 on the per-CPU grace-period kthreads, but increases
3318 that same overhead on each group's leader.
3320 rcutree.qhimark= [KNL]
3321 Set threshold of queued RCU callbacks beyond which
3322 batch limiting is disabled.
3324 rcutree.qlowmark= [KNL]
3325 Set threshold of queued RCU callbacks below which
3326 batch limiting is re-enabled.
3328 rcutree.rcu_idle_gp_delay= [KNL]
3329 Set wakeup interval for idle CPUs that have
3330 RCU callbacks (RCU_FAST_NO_HZ=y).
3332 rcutree.rcu_idle_lazy_gp_delay= [KNL]
3333 Set wakeup interval for idle CPUs that have
3334 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
3335 Lazy RCU callbacks are those which RCU can
3336 prove do nothing more than free memory.
3338 rcutorture.cbflood_inter_holdoff= [KNL]
3339 Set holdoff time (jiffies) between successive
3340 callback-flood tests.
3342 rcutorture.cbflood_intra_holdoff= [KNL]
3343 Set holdoff time (jiffies) between successive
3344 bursts of callbacks within a given callback-flood
3347 rcutorture.cbflood_n_burst= [KNL]
3348 Set the number of bursts making up a given
3349 callback-flood test. Set this to zero to
3350 disable callback-flood testing.
3352 rcutorture.cbflood_n_per_burst= [KNL]
3353 Set the number of callbacks to be registered
3354 in a given burst of a callback-flood test.
3356 rcutorture.fqs_duration= [KNL]
3357 Set duration of force_quiescent_state bursts
3360 rcutorture.fqs_holdoff= [KNL]
3361 Set holdoff time within force_quiescent_state bursts
3364 rcutorture.fqs_stutter= [KNL]
3365 Set wait time between force_quiescent_state bursts
3368 rcutorture.gp_cond= [KNL]
3369 Use conditional/asynchronous update-side
3370 primitives, if available.
3372 rcutorture.gp_exp= [KNL]
3373 Use expedited update-side primitives, if available.
3375 rcutorture.gp_normal= [KNL]
3376 Use normal (non-expedited) asynchronous
3377 update-side primitives, if available.
3379 rcutorture.gp_sync= [KNL]
3380 Use normal (non-expedited) synchronous
3381 update-side primitives, if available. If all
3382 of rcutorture.gp_cond=, rcutorture.gp_exp=,
3383 rcutorture.gp_normal=, and rcutorture.gp_sync=
3384 are zero, rcutorture acts as if is interpreted
3385 they are all non-zero.
3387 rcutorture.n_barrier_cbs= [KNL]
3388 Set callbacks/threads for rcu_barrier() testing.
3390 rcutorture.nfakewriters= [KNL]
3391 Set number of concurrent RCU writers. These just
3392 stress RCU, they don't participate in the actual
3393 test, hence the "fake".
3395 rcutorture.nreaders= [KNL]
3396 Set number of RCU readers. The value -1 selects
3397 N-1, where N is the number of CPUs. A value
3398 "n" less than -1 selects N-n-2, where N is again
3399 the number of CPUs. For example, -2 selects N
3400 (the number of CPUs), -3 selects N+1, and so on.
3402 rcutorture.object_debug= [KNL]
3403 Enable debug-object double-call_rcu() testing.
3405 rcutorture.onoff_holdoff= [KNL]
3406 Set time (s) after boot for CPU-hotplug testing.
3408 rcutorture.onoff_interval= [KNL]
3409 Set time (s) between CPU-hotplug operations, or
3410 zero to disable CPU-hotplug testing.
3412 rcutorture.shuffle_interval= [KNL]
3413 Set task-shuffle interval (s). Shuffling tasks
3414 allows some CPUs to go into dyntick-idle mode
3415 during the rcutorture test.
3417 rcutorture.shutdown_secs= [KNL]
3418 Set time (s) after boot system shutdown. This
3419 is useful for hands-off automated testing.
3421 rcutorture.stall_cpu= [KNL]
3422 Duration of CPU stall (s) to test RCU CPU stall
3423 warnings, zero to disable.
3425 rcutorture.stall_cpu_holdoff= [KNL]
3426 Time to wait (s) after boot before inducing stall.
3428 rcutorture.stat_interval= [KNL]
3429 Time (s) between statistics printk()s.
3431 rcutorture.stutter= [KNL]
3432 Time (s) to stutter testing, for example, specifying
3433 five seconds causes the test to run for five seconds,
3434 wait for five seconds, and so on. This tests RCU's
3435 ability to transition abruptly to and from idle.
3437 rcutorture.test_boost= [KNL]
3438 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
3439 "Maybe" means test if the RCU implementation
3440 under test support RCU priority boosting.
3442 rcutorture.test_boost_duration= [KNL]
3443 Duration (s) of each individual boost test.
3445 rcutorture.test_boost_interval= [KNL]
3446 Interval (s) between each boost test.
3448 rcutorture.test_no_idle_hz= [KNL]
3449 Test RCU's dyntick-idle handling. See also the
3450 rcutorture.shuffle_interval parameter.
3452 rcutorture.torture_runnable= [BOOT]
3453 Start rcutorture running at boot time.
3455 rcutorture.torture_type= [KNL]
3456 Specify the RCU implementation to test.
3458 rcutorture.verbose= [KNL]
3459 Enable additional printk() statements.
3461 rcupdate.rcu_expedited= [KNL]
3462 Use expedited grace-period primitives, for
3463 example, synchronize_rcu_expedited() instead
3464 of synchronize_rcu(). This reduces latency,
3465 but can increase CPU utilization, degrade
3466 real-time latency, and degrade energy efficiency.
3468 rcupdate.rcu_cpu_stall_suppress= [KNL]
3469 Suppress RCU CPU stall warning messages.
3471 rcupdate.rcu_cpu_stall_timeout= [KNL]
3472 Set timeout for RCU CPU stall warning messages.
3474 rcupdate.rcu_task_stall_timeout= [KNL]
3475 Set timeout in jiffies for RCU task stall warning
3476 messages. Disable with a value less than or equal
3479 rcupdate.rcu_self_test= [KNL]
3480 Run the RCU early boot self tests
3482 rcupdate.rcu_self_test_bh= [KNL]
3483 Run the RCU bh early boot self tests
3485 rcupdate.rcu_self_test_sched= [KNL]
3486 Run the RCU sched early boot self tests
3490 Run specified binary instead of /init from the ramdisk,
3491 used for early userspace startup. See initrd.
3494 force - Override the decision by the kernel to hide the
3495 advertisement of RDRAND support (this affects
3496 certain AMD processors because of buggy BIOS
3497 support, specifically around the suspend/resume
3501 Format (x86 or x86_64):
3502 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
3504 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
3506 Where reboot_mode is one of warm (soft) or cold (hard) or gpio,
3507 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
3508 reboot_force is either force or not specified,
3509 reboot_cpu is s[mp]#### with #### being the processor
3510 to be used for rebooting.
3513 [KNL, SMP] Set scheduler's default relax_domain_level.
3514 See Documentation/cgroups/cpusets.txt.
3516 relative_sleep_states=
3517 [SUSPEND] Use sleep state labeling where the deepest
3518 state available other than hibernation is always "mem".
3519 Format: { "0" | "1" }
3520 0 -- Traditional sleep state labels.
3521 1 -- Relative sleep state labels.
3523 reserve= [KNL,BUGS] Force the kernel to ignore some iomem area
3525 reservetop= [X86-32]
3527 Reserves a hole at the top of the kernel virtual
3532 Set the amount of memory to reserve for BIOS at
3533 the bottom of the address space.
3535 reset_devices [KNL] Force drivers to reset the underlying device
3536 during initialization.
3539 Specify the partition device for software suspend
3541 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
3543 resume_offset= [SWSUSP]
3544 Specify the offset from the beginning of the partition
3545 given by "resume=" at which the swap header is located,
3546 in <PAGE_SIZE> units (needed only for swap files).
3547 See Documentation/power/swsusp-and-swap-files.txt
3549 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
3550 read the resume files
3552 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
3553 Useful for devices that are detected asynchronously
3554 (e.g. USB and MMC devices).
3556 hibernate= [HIBERNATION]
3557 noresume Don't check if there's a hibernation image
3558 present during boot.
3559 nocompress Don't compress/decompress hibernation images.
3560 no Disable hibernation and resume.
3562 retain_initrd [RAM] Keep initrd memory after extraction
3564 rfkill.default_state=
3565 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
3566 etc. communication is blocked by default.
3569 rfkill.master_switch_mode=
3570 0 The "airplane mode" button does nothing.
3571 1 The "airplane mode" button toggles between everything
3572 blocked and the previous configuration.
3573 2 The "airplane mode" button toggles between everything
3574 blocked and everything unblocked.
3576 rhash_entries= [KNL,NET]
3577 Set number of hash buckets for route cache
3579 ro [KNL] Mount root device read-only on boot
3582 on Mark read-only kernel memory as read-only (default).
3583 off Leave read-only kernel memory writable for debugging.
3585 root= [KNL] Root filesystem
3586 See name_to_dev_t comment in init/do_mounts.c.
3588 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
3589 mount the root filesystem
3591 rootflags= [KNL] Set root filesystem mount option string
3593 rootfstype= [KNL] Set root filesystem type
3595 rootwait [KNL] Wait (indefinitely) for root device to show up.
3596 Useful for devices that are detected asynchronously
3597 (e.g. USB and MMC devices).
3599 rproc_mem=nn[KMG][@address]
3600 [KNL,ARM,CMA] Remoteproc physical memory block.
3601 Memory area to be used by remote processor image,
3604 rw [KNL] Mount root device read-write on boot
3606 S [KNL] Run init in single mode
3608 s390_iommu= [HW,S390]
3609 Set s390 IOTLB flushing mode
3611 With strict flushing every unmap operation will result in
3612 an IOTLB flush. Default is lazy flushing before reuse,
3616 See drivers/net/irda/sa1100_ir.c.
3618 sbni= [NET] Granch SBNI12 leased line adapter
3620 sched_debug [KNL] Enables verbose scheduler debug messages.
3622 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
3623 xtime_lock contention on larger systems, and/or RCU lock
3624 contention on all systems with CONFIG_MAXSMP set.
3625 Format: { "0" | "1" }
3626 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
3628 Note: increases power consumption, thus should only be
3629 enabled if running jitter sensitive (HPC/RT) workloads.
3631 security= [SECURITY] Choose a security module to enable at boot.
3632 If this boot parameter is not specified, only the first
3633 security module asking for security registration will be
3634 loaded. An invalid security module name will be treated
3635 as if no module has been chosen.
3637 selinux= [SELINUX] Disable or enable SELinux at boot time.
3638 Format: { "0" | "1" }
3639 See security/selinux/Kconfig help text.
3642 Default value is set via kernel config option.
3643 If enabled at boot time, /selinux/disable can be used
3644 later to disable prior to initial policy load.
3646 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
3647 Format: { "0" | "1" }
3648 See security/apparmor/Kconfig help text
3651 Default value is set via kernel config option.
3653 serialnumber [BUGS=X86-32]
3656 Maximal number of shapers.
3658 show_msr= [x86] show boot-time MSR settings
3659 Format: { <integer> }
3660 Show boot-time (BIOS-initialized) MSR settings.
3661 The parameter means the number of CPUs to show,
3662 for example 1 means boot CPU only.
3670 Disable merging of slabs with similar size. May be
3671 necessary if there is some reason to distinguish
3672 allocs to different slabs. Debug options disable
3673 merging on their own.
3674 For more information see Documentation/vm/slub.txt.
3676 slab_max_order= [MM, SLAB]
3677 Determines the maximum allowed order for slabs.
3678 A high setting may cause OOMs due to memory
3679 fragmentation. Defaults to 1 for systems with
3680 more than 32MB of RAM, 0 otherwise.
3682 slub_debug[=options[,slabs]] [MM, SLUB]
3683 Enabling slub_debug allows one to determine the
3684 culprit if slab objects become corrupted. Enabling
3685 slub_debug can create guard zones around objects and
3686 may poison objects when not in use. Also tracks the
3687 last alloc / free. For more information see
3688 Documentation/vm/slub.txt.
3690 slub_max_order= [MM, SLUB]
3691 Determines the maximum allowed order for slabs.
3692 A high setting may cause OOMs due to memory
3693 fragmentation. For more information see
3694 Documentation/vm/slub.txt.
3696 slub_min_objects= [MM, SLUB]
3697 The minimum number of objects per slab. SLUB will
3698 increase the slab order up to slub_max_order to
3699 generate a sufficiently large slab able to contain
3700 the number of objects indicated. The higher the number
3701 of objects the smaller the overhead of tracking slabs
3702 and the less frequently locks need to be acquired.
3703 For more information see Documentation/vm/slub.txt.
3705 slub_min_order= [MM, SLUB]
3706 Determines the minimum page order for slabs. Must be
3707 lower than slub_max_order.
3708 For more information see Documentation/vm/slub.txt.
3710 slub_nomerge [MM, SLUB]
3711 Same with slab_nomerge. This is supported for legacy.
3712 See slab_nomerge for more information.
3715 Format: <io1>[,<io2>[,...,<io8>]]
3717 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
3718 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
3719 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
3720 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
3721 smsc-ircc2.ircc_irq= [HW] IRQ line
3722 smsc-ircc2.ircc_dma= [HW] DMA channel
3723 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
3724 0: Toshiba Satellite 1800 (GP data pin select)
3725 1: Fast pin select (default)
3729 [KNL] Should the soft-lockup detector generate panics.
3732 softlockup_all_cpu_backtrace=
3733 [KNL] Should the soft-lockup detector generate
3734 backtraces on all cpus.
3737 sonypi.*= [HW] Sony Programmable I/O Control Device driver
3738 See Documentation/laptops/sonypi.txt
3740 spectre_v2= [X86] Control mitigation of Spectre variant 2
3741 (indirect branch speculation) vulnerability.
3742 The default operation protects the kernel from
3745 on - unconditionally enable, implies
3747 off - unconditionally disable, implies
3749 auto - kernel detects whether your CPU model is
3752 Selecting 'on' will, and 'auto' may, choose a
3753 mitigation method at run time according to the
3754 CPU, the available microcode, the setting of the
3755 CONFIG_RETPOLINE configuration option, and the
3756 compiler with which the kernel was built.
3758 Selecting 'on' will also enable the mitigation
3759 against user space to user space task attacks.
3761 Selecting 'off' will disable both the kernel and
3762 the user space protections.
3764 Specific mitigations can also be selected manually:
3766 retpoline - replace indirect branches
3767 retpoline,generic - google's original retpoline
3768 retpoline,amd - AMD-specific minimal thunk
3770 Not specifying this option is equivalent to
3774 [X86] Control mitigation of Spectre variant 2
3775 (indirect branch speculation) vulnerability between
3778 on - Unconditionally enable mitigations. Is
3779 enforced by spectre_v2=on
3781 off - Unconditionally disable mitigations. Is
3782 enforced by spectre_v2=off
3784 prctl - Indirect branch speculation is enabled,
3785 but mitigation can be enabled via prctl
3786 per thread. The mitigation control state
3787 is inherited on fork.
3790 - Like "prctl" above, but only STIBP is
3791 controlled per thread. IBPB is issued
3792 always when switching between different user
3796 - Same as "prctl" above, but all seccomp
3797 threads will enable the mitigation unless
3798 they explicitly opt out.
3801 - Like "seccomp" above, but only STIBP is
3802 controlled per thread. IBPB is issued
3803 always when switching between different
3804 user space processes.
3806 auto - Kernel selects the mitigation depending on
3807 the available CPU features and vulnerability.
3810 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl"
3812 Not specifying this option is equivalent to
3813 spectre_v2_user=auto.
3815 spec_store_bypass_disable=
3816 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
3817 (Speculative Store Bypass vulnerability)
3819 Certain CPUs are vulnerable to an exploit against a
3820 a common industry wide performance optimization known
3821 as "Speculative Store Bypass" in which recent stores
3822 to the same memory location may not be observed by
3823 later loads during speculative execution. The idea
3824 is that such stores are unlikely and that they can
3825 be detected prior to instruction retirement at the
3826 end of a particular speculation execution window.
3828 In vulnerable processors, the speculatively forwarded
3829 store can be used in a cache side channel attack, for
3830 example to read memory to which the attacker does not
3831 directly have access (e.g. inside sandboxed code).
3833 This parameter controls whether the Speculative Store
3834 Bypass optimization is used.
3836 on - Unconditionally disable Speculative Store Bypass
3837 off - Unconditionally enable Speculative Store Bypass
3838 auto - Kernel detects whether the CPU model contains an
3839 implementation of Speculative Store Bypass and
3840 picks the most appropriate mitigation. If the
3841 CPU is not vulnerable, "off" is selected. If the
3842 CPU is vulnerable the default mitigation is
3843 architecture and Kconfig dependent. See below.
3844 prctl - Control Speculative Store Bypass per thread
3845 via prctl. Speculative Store Bypass is enabled
3846 for a process by default. The state of the control
3847 is inherited on fork.
3848 seccomp - Same as "prctl" above, but all seccomp threads
3849 will disable SSB unless they explicitly opt out.
3851 Not specifying this option is equivalent to
3852 spec_store_bypass_disable=auto.
3854 Default mitigations:
3855 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
3857 spia_io_base= [HW,MTD]
3862 stack_guard_gap= [MM]
3863 override the default stack gap protection. The value
3864 is in page units and it defines how many pages prior
3865 to (for stacks growing down) resp. after (for stacks
3866 growing up) the main stack are reserved for no other
3867 mapping. Default value is 256 pages.
3870 Enabled the stack tracer on boot up.
3872 stacktrace_filter=[function-list]
3873 [FTRACE] Limit the functions that the stack tracer
3874 will trace at boot up. function-list is a comma separated
3875 list of functions. This list can be changed at run
3876 time by the stack_trace_filter file in the debugfs
3877 tracing directory. Note, this enables stack tracing
3878 and the stacktrace above is not needed.
3882 Set the STI (builtin display/keyboard on the HP-PARISC
3883 machines) console (graphic card) which should be used
3884 as the initial boot-console.
3885 See also comment in drivers/video/console/sticore.c.
3888 See comment in drivers/video/console/sticore.c.
3891 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
3893 sunrpc.min_resvport=
3894 sunrpc.max_resvport=
3896 SunRPC servers often require that client requests
3897 originate from a privileged port (i.e. a port in the
3898 range 0 < portnr < 1024).
3899 An administrator who wishes to reserve some of these
3900 ports for other uses may adjust the range that the
3901 kernel's sunrpc client considers to be privileged
3902 using these two parameters to set the minimum and
3903 maximum port values.
3907 Control how the NFS server code allocates CPUs to
3908 service thread pools. Depending on how many NICs
3909 you have and where their interrupts are bound, this
3910 option will affect which CPUs will do NFS serving.
3911 Note: this parameter cannot be changed while the
3912 NFS server is running.
3914 auto the server chooses an appropriate mode
3915 automatically using heuristics
3916 global a single global pool contains all CPUs
3917 percpu one pool for each CPU
3918 pernode one pool for each NUMA node (equivalent
3919 to global on non-NUMA machines)
3921 sunrpc.tcp_slot_table_entries=
3922 sunrpc.udp_slot_table_entries=
3924 Sets the upper limit on the number of simultaneous
3925 RPC calls that can be sent from the client to a
3926 server. Increasing these values may allow you to
3927 improve throughput, but will also increase the
3928 amount of memory reserved for use by the client.
3930 suspend.pm_test_delay=
3932 Sets the number of seconds to remain in a suspend test
3933 mode before resuming the system (see
3934 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
3935 is set. Default value is 5.
3938 [KNL] Enable accounting of swap in memory resource
3939 controller if no parameter or 1 is given or disable
3940 it if 0 is given (See Documentation/cgroups/memory.txt)
3942 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
3943 Format: { <int> | force }
3944 <int> -- Number of I/O TLB slabs
3945 force -- force using of bounce buffers even if they
3946 wouldn't be automatically used by the kernel
3950 sysfs.deprecated=0|1 [KNL]
3951 Enable/disable old style sysfs layout for old udev
3952 on older distributions. When this option is enabled
3953 very new udev will not work anymore. When this option
3954 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
3955 in older udev will not work anymore.
3956 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
3957 the kernel configuration.
3959 sysrq_always_enabled
3961 Ignore sysrq setting - this boot parameter will
3962 neutralize any effect of /proc/sys/kernel/sysrq.
3963 Useful for debugging.
3965 tcpmhash_entries= [KNL,NET]
3966 Set the number of tcp_metrics_hash slots.
3967 Default value is 8192 or 16384 depending on total
3968 ram pages. This is used to specify the TCP metrics
3969 cache size. See Documentation/networking/ip-sysctl.txt
3970 "tcp_no_metrics_save" section for more details.
3974 test_suspend= [SUSPEND][,N]
3975 Specify "mem" (for Suspend-to-RAM) or "standby" (for
3976 standby suspend) or "freeze" (for suspend type freeze)
3977 as the system sleep state during system startup with
3978 the optional capability to repeat N number of times.
3979 The system is woken from this state using a
3980 wakeup-capable RTC alarm.
3982 thash_entries= [KNL,NET]
3983 Set number of hash buckets for TCP connection
3985 thermal.act= [HW,ACPI]
3986 -1: disable all active trip points in all thermal zones
3987 <degrees C>: override all lowest active trip points
3989 thermal.crt= [HW,ACPI]
3990 -1: disable all critical trip points in all thermal zones
3991 <degrees C>: override all critical trip points
3993 thermal.nocrt= [HW,ACPI]
3994 Set to disable actions on ACPI thermal zone
3995 critical and hot trip points.
3997 thermal.off= [HW,ACPI]
3998 1: disable ACPI thermal control
4000 thermal.psv= [HW,ACPI]
4001 -1: disable all passive trip points
4002 <degrees C>: override all passive trip points to this
4005 thermal.tzp= [HW,ACPI]
4006 Specify global default ACPI thermal zone polling rate
4007 <deci-seconds>: poll all this frequency
4008 0: no polling (default)
4011 Force threading of all interrupt handlers except those
4012 marked explicitly IRQF_NO_THREAD.
4015 Enable the Transcendent memory driver if built-in.
4017 tmem.cleancache=0|1 [KNL, XEN]
4018 Default is on (1). Disable the usage of the cleancache
4019 API to send anonymous pages to the hypervisor.
4021 tmem.frontswap=0|1 [KNL, XEN]
4022 Default is on (1). Disable the usage of the frontswap
4023 API to send swap pages to the hypervisor. If disabled
4024 the selfballooning and selfshrinking are force disabled.
4026 tmem.selfballooning=0|1 [KNL, XEN]
4027 Default is on (1). Disable the driving of swap pages
4030 tmem.selfshrinking=0|1 [KNL, XEN]
4031 Default is on (1). Partial swapoff that immediately
4032 transfers pages from Xen hypervisor back to the
4033 kernel based on different criteria.
4037 Specify if the kernel should make use of the cpu
4038 topology information if the hardware supports this.
4039 The scheduler will make use of this information and
4040 e.g. base its process migration decisions on it.
4043 topology_updates= [KNL, PPC, NUMA]
4045 Specify if the kernel should ignore (off)
4046 topology updates sent by the hypervisor to this
4051 tpm_suspend_pcr=[HW,TPM]
4052 Format: integer pcr id
4053 Specify that at suspend time, the tpm driver
4054 should extend the specified pcr with zeros,
4055 as a workaround for some chips which fail to
4056 flush the last written pcr on TPM_SaveState.
4057 This will guarantee that all the other pcrs
4060 trace_buf_size=nn[KMG]
4061 [FTRACE] will set tracing buffer size on each cpu.
4063 trace_event=[event-list]
4064 [FTRACE] Set and start specified trace events in order
4065 to facilitate early boot debugging.
4066 See also Documentation/trace/events.txt
4068 trace_options=[option-list]
4069 [FTRACE] Enable or disable tracer options at boot.
4070 The option-list is a comma delimited list of options
4071 that can be enabled or disabled just as if you were
4072 to echo the option name into
4074 /sys/kernel/debug/tracing/trace_options
4076 For example, to enable stacktrace option (to dump the
4077 stack trace of each event), add to the command line:
4079 trace_options=stacktrace
4081 See also Documentation/trace/ftrace.txt "trace options"
4085 Have the tracepoints sent to printk as well as the
4086 tracing ring buffer. This is useful for early boot up
4087 where the system hangs or reboots and does not give the
4088 option for reading the tracing buffer or performing a
4089 ftrace_dump_on_oops.
4091 To turn off having tracepoints sent to printk,
4092 echo 0 > /proc/sys/kernel/tracepoint_printk
4093 Note, echoing 1 into this file without the
4094 tracepoint_printk kernel cmdline option has no effect.
4098 Having tracepoints sent to printk() and activating high
4099 frequency tracepoints such as irq or sched, can cause
4100 the system to live lock.
4103 [FTRACE] enable this option to disable tracing when a
4104 warning is hit. This turns off "tracing_on". Tracing can
4105 be enabled again by echoing '1' into the "tracing_on"
4106 file located in /sys/kernel/debug/tracing/
4108 This option is useful, as it disables the trace before
4109 the WARNING dump is called, which prevents the trace to
4110 be filled with content caused by the warning output.
4112 This option can also be set at run time via the sysctl
4113 option: kernel/traceoff_on_warning
4115 transparent_hugepage=
4117 Format: [always|madvise|never]
4118 Can be used to control the default behavior of the system
4119 with respect to transparent hugepages.
4120 See Documentation/vm/transhuge.txt for more details.
4122 tsc= Disable clocksource stability checks for TSC.
4124 [x86] reliable: mark tsc clocksource as reliable, this
4125 disables clocksource verification at runtime, as well
4126 as the stability checks done at bootup. Used to enable
4127 high-resolution timer mode on older hardware, and in
4128 virtualized environment.
4129 [x86] noirqtime: Do not use TSC to do irq accounting.
4130 Used to run time disable IRQ_TIME_ACCOUNTING on any
4131 platforms where RDTSC is slow and this accounting
4134 turbografx.map[2|3]= [HW,JOY]
4135 TurboGraFX parallel port interface
4137 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
4138 See also Documentation/input/joystick-parport.txt
4140 udbg-immortal [PPC] When debugging early kernel crashes that
4141 happen after console_init() and before a proper
4142 console driver takes over, this boot options might
4143 help "seeing" what's going on.
4145 uhash_entries= [KNL,NET]
4146 Set number of hash buckets for UDP/UDP-Lite connections
4149 [USB] Ignore overcurrent events (default N).
4150 Some badly-designed motherboards generate lots of
4151 bogus events, for ports that aren't wired to
4152 anything. Set this parameter to avoid log spamming.
4153 Note that genuine overcurrent events won't be
4157 [X86] Cause panic on unknown NMI.
4159 usbcore.authorized_default=
4160 [USB] Default USB device authorization:
4161 (default -1 = authorized except for wireless USB,
4162 0 = not authorized, 1 = authorized)
4164 usbcore.autosuspend=
4165 [USB] The autosuspend time delay (in seconds) used
4166 for newly-detected USB devices (default 2). This
4167 is the time required before an idle device will be
4168 autosuspended. Devices for which the delay is set
4169 to a negative value won't be autosuspended at all.
4171 usbcore.usbfs_snoop=
4172 [USB] Set to log all usbfs traffic (default 0 = off).
4174 usbcore.blinkenlights=
4175 [USB] Set to cycle leds on hubs (default 0 = off).
4177 usbcore.old_scheme_first=
4178 [USB] Start with the old device initialization
4179 scheme (default 0 = off).
4181 usbcore.usbfs_memory_mb=
4182 [USB] Memory limit (in MB) for buffers allocated by
4183 usbfs (default = 16, 0 = max = 2047).
4185 usbcore.use_both_schemes=
4186 [USB] Try the other device initialization scheme
4187 if the first one fails (default 1 = enabled).
4189 usbcore.initial_descriptor_timeout=
4190 [USB] Specifies timeout for the initial 64-byte
4191 USB_REQ_GET_DESCRIPTOR request in milliseconds
4192 (default 5000 = 5.0 seconds).
4195 [USBHID] The interval which mice are to be polled at.
4197 usb-storage.delay_use=
4198 [UMS] The delay in seconds before a new device is
4199 scanned for Logical Units (default 1).
4202 [UMS] A list of quirks entries to supplement or
4203 override the built-in unusual_devs list. List
4204 entries are separated by commas. Each entry has
4205 the form VID:PID:Flags where VID and PID are Vendor
4206 and Product ID values (4-digit hex numbers) and
4207 Flags is a set of characters, each corresponding
4208 to a common usb-storage quirk flag as follows:
4209 a = SANE_SENSE (collect more than 18 bytes
4211 b = BAD_SENSE (don't collect more than 18
4212 bytes of sense data);
4213 c = FIX_CAPACITY (decrease the reported
4214 device capacity by one sector);
4215 d = NO_READ_DISC_INFO (don't use
4216 READ_DISC_INFO command);
4217 e = NO_READ_CAPACITY_16 (don't use
4218 READ_CAPACITY_16 command);
4219 f = NO_REPORT_OPCODES (don't use report opcodes
4221 g = MAX_SECTORS_240 (don't transfer more than
4222 240 sectors at a time, uas only);
4223 h = CAPACITY_HEURISTICS (decrease the
4224 reported device capacity by one
4225 sector if the number is odd);
4226 i = IGNORE_DEVICE (don't bind to this
4228 j = NO_REPORT_LUNS (don't use report luns
4230 l = NOT_LOCKABLE (don't try to lock and
4231 unlock ejectable media);
4232 m = MAX_SECTORS_64 (don't transfer more
4233 than 64 sectors = 32 KB at a time);
4234 n = INITIAL_READ10 (force a retry of the
4235 initial READ(10) command);
4236 o = CAPACITY_OK (accept the capacity
4237 reported by the device);
4238 p = WRITE_CACHE (the device cache is ON
4240 r = IGNORE_RESIDUE (the device reports
4241 bogus residue values);
4242 s = SINGLE_LUN (the device has only one
4244 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
4245 commands, uas only);
4246 u = IGNORE_UAS (don't bind to the uas driver);
4247 w = NO_WP_DETECT (don't test whether the
4248 medium is write-protected).
4249 Example: quirks=0419:aaf5:rl,0421:0433:rc
4251 user_debug= [KNL,ARM]
4253 See arch/arm/Kconfig.debug help text.
4254 1 - undefined instruction events
4256 4 - invalid data aborts
4259 Example: user_debug=31
4262 [X86] Flags controlling user PTE allocations.
4264 nohigh = do not allocate PTE pages in
4265 HIGHMEM regardless of setting
4269 On X86_32, this is an alias for vdso32=. Otherwise:
4271 vdso=1: enable VDSO (the default)
4272 vdso=0: disable VDSO mapping
4274 vdso32= [X86] Control the 32-bit vDSO
4275 vdso32=1: enable 32-bit VDSO
4276 vdso32=0 or vdso32=2: disable 32-bit VDSO
4278 See the help text for CONFIG_COMPAT_VDSO for more
4279 details. If CONFIG_COMPAT_VDSO is set, the default is
4280 vdso32=0; otherwise, the default is vdso32=1.
4282 For compatibility with older kernels, vdso32=2 is an
4285 Try vdso32=0 if you encounter an error that says:
4286 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
4289 vector=percpu: enable percpu vector domain
4291 video= [FB] Frame buffer configuration
4292 See Documentation/fb/modedb.txt.
4294 video.brightness_switch_enabled= [0,1]
4295 If set to 1, on receiving an ACPI notify event
4296 generated by hotkey, video driver will adjust brightness
4297 level and then send out the event to user space through
4298 the allocated input device; If set to 0, video driver
4299 will only send out the event without touching backlight
4304 [VMMIO] Memory mapped virtio (platform) device.
4306 <size>@<baseaddr>:<irq>[:<id>]
4308 <size> := size (can use standard suffixes
4310 <baseaddr> := physical base address
4311 <irq> := interrupt number (as passed to
4313 <id> := (optional) platform device id
4315 virtio_mmio.device=1K@0x100b0000:48:7
4317 Can be used multiple times for multiple devices.
4319 vga= [BOOT,X86-32] Select a particular video mode
4320 See Documentation/x86/boot.txt and
4321 Documentation/svga.txt.
4322 Use vga=ask for menu.
4323 This is actually a boot loader parameter; the value is
4324 passed to the kernel using a special protocol.
4326 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
4327 size of <nn>. This can be used to increase the
4328 minimum size (128MB on x86). It can also be used to
4329 decrease the size and leave more room for directly
4332 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
4335 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
4338 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
4342 Controls the behavior of vsyscalls (i.e. calls to
4343 fixed addresses of 0xffffffffff600x00 from legacy
4344 code). Most statically-linked binaries and older
4345 versions of glibc use these calls. Because these
4346 functions are at fixed addresses, they make nice
4347 targets for exploits that can control RIP.
4349 emulate [default] Vsyscalls turn into traps and are
4350 emulated reasonably safely.
4352 native Vsyscalls are native syscall instructions.
4353 This is a little bit faster than trapping
4354 and makes a few dynamic recompilers work
4355 better than they would in emulation mode.
4356 It also makes exploits much easier to write.
4358 none Vsyscalls don't work at all. This makes
4359 them quite hard to use for exploits but
4360 might break your system.
4362 vt.color= [VT] Default text color.
4363 Format: 0xYX, X = foreground, Y = background.
4364 Default: 0x07 = light gray on black.
4366 vt.cur_default= [VT] Default cursor shape.
4367 Format: 0xCCBBAA, where AA, BB, and CC are the same as
4368 the parameters of the <Esc>[?A;B;Cc escape sequence;
4369 see VGA-softcursor.txt. Default: 2 = underline.
4371 vt.default_blu= [VT]
4372 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
4373 Change the default blue palette of the console.
4374 This is a 16-member array composed of values
4377 vt.default_grn= [VT]
4378 Format: <green0>,<green1>,<green2>,...,<green15>
4379 Change the default green palette of the console.
4380 This is a 16-member array composed of values
4383 vt.default_red= [VT]
4384 Format: <red0>,<red1>,<red2>,...,<red15>
4385 Change the default red palette of the console.
4386 This is a 16-member array composed of values
4392 Set system-wide default UTF-8 mode for all tty's.
4393 Default is 1, i.e. UTF-8 mode is enabled for all
4394 newly opened terminals.
4396 vt.global_cursor_default=
4399 Set system-wide default for whether a cursor
4400 is shown on new VTs. Default is -1,
4401 i.e. cursors will be created by default unless
4402 overridden by individual drivers. 0 will hide
4403 cursors, 1 will display them.
4405 vt.italic= [VT] Default color for italic text; 0-15.
4408 vt.underline= [VT] Default color for underlined text; 0-15.
4411 watchdog timers [HW,WDT] For information on watchdog timers,
4412 see Documentation/watchdog/watchdog-parameters.txt
4413 or other driver-specific files in the
4414 Documentation/watchdog/ directory.
4416 workqueue.watchdog_thresh=
4417 If CONFIG_WQ_WATCHDOG is configured, workqueue can
4418 warn stall conditions and dump internal state to
4419 help debugging. 0 disables workqueue stall
4420 detection; otherwise, it's the stall threshold
4421 duration in seconds. The default value is 30 and
4422 it can be updated at runtime by writing to the
4423 corresponding sysfs file.
4425 workqueue.disable_numa
4426 By default, all work items queued to unbound
4427 workqueues are affine to the NUMA nodes they're
4428 issued on, which results in better behavior in
4429 general. If NUMA affinity needs to be disabled for
4430 whatever reason, this option can be used. Note
4431 that this also can be controlled per-workqueue for
4432 workqueues visible under /sys/bus/workqueue/.
4434 workqueue.power_efficient
4435 Per-cpu workqueues are generally preferred because
4436 they show better performance thanks to cache
4437 locality; unfortunately, per-cpu workqueues tend to
4438 be more power hungry than unbound workqueues.
4440 Enabling this makes the per-cpu workqueues which
4441 were observed to contribute significantly to power
4442 consumption unbound, leading to measurably lower
4443 power usage at the cost of small performance
4446 The default value of this parameter is determined by
4447 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
4449 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
4450 default x2apic cluster mode on platforms
4453 x86_intel_mid_timer= [X86-32,APBT]
4454 Choose timer option for x86 Intel MID platform.
4455 Two valid options are apbt timer only and lapic timer
4456 plus one apbt timer for broadcast timer.
4457 x86_intel_mid_timer=apbt_only | lapic_and_apbt
4459 xen_512gb_limit [KNL,X86-64,XEN]
4460 Restricts the kernel running paravirtualized under Xen
4461 to use only up to 512 GB of RAM. The reason to do so is
4462 crash analysis tools and Xen tools for doing domain
4463 save/restore/migration must be enabled to handle larger
4466 xen_emul_unplug= [HW,X86,XEN]
4467 Unplug Xen emulated devices
4468 Format: [unplug0,][unplug1]
4469 ide-disks -- unplug primary master IDE devices
4470 aux-ide-disks -- unplug non-primary-master IDE devices
4471 nics -- unplug network devices
4472 all -- unplug all emulated devices (NICs and IDE disks)
4473 unnecessary -- unplugging emulated devices is
4474 unnecessary even if the host did not respond to
4476 never -- do not unplug even if version check succeeds
4478 xen_nopvspin [X86,XEN]
4479 Disables the ticketlock slowpath using Xen PV
4483 Disables the PV optimizations forcing the HVM guest to
4484 run as generic HVM guest with no PV drivers.
4486 xirc2ps_cs= [NET,PCMCIA]
4488 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
4490 ______________________________________________________________________
4494 Add more DRM drivers.