| blob | 87d40a72f6a1bec998be718138347a875c00117a |
1 Kernel Parameters
2 ~~~~~~~~~~~~~~~~~
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
26 loadable modules too.
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 DRM Direct Rendering Management support is enabled.
60 DYNAMIC_DEBUG Build in debug messages and enable them at runtime
61 EDD BIOS Enhanced Disk Drive Services (EDD) is enabled
62 EFI EFI Partitioning (GPT) is enabled
63 EIDE EIDE/ATAPI support is enabled.
64 EVM Extended Verification Module
65 FB The frame buffer device is enabled.
66 FTRACE Function tracing enabled.
67 GCOV GCOV profiling is enabled.
68 HW Appropriate hardware is enabled.
69 IA-64 IA-64 architecture is enabled.
70 IMA Integrity measurement architecture is enabled.
71 IOSCHED More than one I/O scheduler is enabled.
72 IP_PNP IP DHCP, BOOTP, or RARP is enabled.
73 IPV6 IPv6 support is enabled.
74 ISAPNP ISA PnP code is enabled.
75 ISDN Appropriate ISDN support is enabled.
76 JOY Appropriate joystick support is enabled.
77 KGDB Kernel debugger support is enabled.
78 KVM Kernel Virtual Machine support is enabled.
79 LIBATA Libata driver is enabled
80 LP Printer support is enabled.
81 LOOP Loopback device support is enabled.
82 M68k M68k architecture is enabled.
83 These options have more detailed description inside of
84 Documentation/m68k/kernel-options.txt.
85 MDA MDA console support is enabled.
86 MIPS MIPS architecture is enabled.
87 MOUSE Appropriate mouse support is enabled.
88 MSI Message Signaled Interrupts (PCI).
89 MTD MTD (Memory Technology Device) support is enabled.
90 NET Appropriate network support is enabled.
91 NUMA NUMA support is enabled.
92 NFS Appropriate NFS support is enabled.
93 OSS OSS sound support is enabled.
94 PV_OPS A paravirtualized kernel is enabled.
95 PARIDE The ParIDE (parallel port IDE) subsystem is enabled.
96 PARISC The PA-RISC architecture is enabled.
97 PCI PCI bus support is enabled.
98 PCIE PCI Express support is enabled.
99 PCMCIA The PCMCIA subsystem is enabled.
100 PNP Plug & Play support is enabled.
101 PPC PowerPC architecture is enabled.
102 PPT Parallel port support is enabled.
103 PS2 Appropriate PS/2 support is enabled.
104 RAM RAM disk support is enabled.
105 S390 S390 architecture is enabled.
106 SCSI Appropriate SCSI support is enabled.
107 A lot of drivers have their options described inside
108 the Documentation/scsi/ sub-directory.
109 SECURITY Different security models are enabled.
110 SELINUX SELinux support is enabled.
111 APPARMOR AppArmor support is enabled.
112 SERIAL Serial support is enabled.
113 SH SuperH architecture is enabled.
114 SMP The kernel is an SMP kernel.
115 SPARC Sparc architecture is enabled.
116 SWSUSP Software suspend (hibernation) is enabled.
117 SUSPEND System suspend states are enabled.
118 TPM TPM drivers are enabled.
119 TS Appropriate touchscreen support is enabled.
120 UMS USB Mass Storage support is enabled.
121 USB USB support is enabled.
122 USBHID USB Human Interface Device support is enabled.
123 V4L Video For Linux support is enabled.
124 VMMIO Driver for memory mapped virtio devices is enabled.
125 VGA The VGA console has been enabled.
126 VT Virtual terminal support is enabled.
127 WDT Watchdog support is enabled.
128 XT IBM PC/XT MFM hard disk support is enabled.
129 X86-32 X86-32, aka i386 architecture is enabled.
130 X86-64 X86-64 architecture is enabled.
131 More X86-64 boot options can be found in
132 Documentation/x86/x86_64/boot-options.txt .
133 X86 Either 32-bit or 64-bit x86 (same as X86-32+X86-64)
134 XEN Xen support is enabled
136 In addition, the following text indicates that the option:
138 BUGS= Relates to possible processor bugs on the said processor.
139 KNL Is a kernel start-up parameter.
140 BOOT Is a boot loader parameter.
142 Parameters denoted with BOOT are actually interpreted by the boot
143 loader, and have no meaning to the kernel directly.
144 Do not modify the syntax of boot loader parameters without extreme
145 need or coordination with <Documentation/x86/boot.txt>.
147 There are also arch-specific kernel-parameters not documented here.
148 See for example <Documentation/x86/x86_64/boot-options.txt>.
150 Note that ALL kernel parameters listed below are CASE SENSITIVE, and that
151 a trailing = on the name of any parameter states that that parameter will
152 be entered as an environment variable, whereas its absence indicates that
153 it will appear as a kernel argument readable via /proc/cmdline by programs
154 running once the system is up.
156 The number of kernel parameters is not limited, but the length of the
157 complete command line (parameters including spaces etc.) is limited to
158 a fixed number of characters. This limit depends on the architecture
159 and is between 256 and 4096 characters. It is defined in the file
160 ./include/asm/setup.h as COMMAND_LINE_SIZE.
162 Finally, the [KMG] suffix is commonly described after a number of kernel
163 parameter values. These 'K', 'M', and 'G' letters represent the _binary_
164 multipliers 'Kilo', 'Mega', and 'Giga', equalling 2^10, 2^20, and 2^30
165 bytes respectively. Such letter suffixes can also be entirely omitted.
168 acpi= [HW,ACPI,X86,ARM64]
169 Advanced Configuration and Power Interface
170 Format: { force | off | strict | noirq | rsdt |
171 copy_dsdt }
172 force -- enable ACPI if default was off
173 off -- disable ACPI if default was on
174 noirq -- do not use ACPI for IRQ routing
175 strict -- Be less tolerant of platforms that are not
176 strictly ACPI specification compliant.
177 rsdt -- prefer RSDT over (default) XSDT
178 copy_dsdt -- copy DSDT to memory
179 For ARM64, ONLY "acpi=off" or "acpi=force" are available
181 See also Documentation/power/runtime_pm.txt, pci=noacpi
183 acpi_apic_instance= [ACPI, IOAPIC]
184 Format: <int>
185 2: use 2nd APIC table, if available
186 1,0: use 1st APIC table
187 default: 0
189 acpi_backlight= [HW,ACPI]
190 acpi_backlight=vendor
191 acpi_backlight=video
192 If set to vendor, prefer vendor specific driver
193 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
194 of the ACPI video.ko driver.
196 acpica_no_return_repair [HW, ACPI]
197 Disable AML predefined validation mechanism
198 This mechanism can repair the evaluation result to make
199 the return objects more ACPI specification compliant.
200 This option is useful for developers to identify the
201 root cause of an AML interpreter issue when the issue
202 has something to do with the repair mechanism.
204 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
205 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
206 Format: <int>
207 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
208 debug output. Bits in debug_layer correspond to a
209 _COMPONENT in an ACPI source file, e.g.,
210 #define _COMPONENT ACPI_PCI_COMPONENT
211 Bits in debug_level correspond to a level in
212 ACPI_DEBUG_PRINT statements, e.g.,
213 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
214 The debug_level mask defaults to "info". See
215 Documentation/acpi/debug.txt for more information about
216 debug layers and levels.
218 Enable processor driver info messages:
219 acpi.debug_layer=0x20000000
220 Enable PCI/PCI interrupt routing info messages:
221 acpi.debug_layer=0x400000
222 Enable AML "Debug" output, i.e., stores to the Debug
223 object while interpreting AML:
224 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
225 Enable all messages related to ACPI hardware:
226 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
228 Some values produce so much output that the system is
229 unusable. The "log_buf_len" parameter may be useful
230 if you need to capture more output.
232 acpi_enforce_resources= [ACPI]
233 { strict | lax | no }
234 Check for resource conflicts between native drivers
235 and ACPI OperationRegions (SystemIO and SystemMemory
236 only). IO ports and memory declared in ACPI might be
237 used by the ACPI subsystem in arbitrary AML code and
238 can interfere with legacy drivers.
239 strict (default): access to resources claimed by ACPI
240 is denied; legacy drivers trying to access reserved
241 resources will fail to bind to device using them.
242 lax: access to resources claimed by ACPI is allowed;
243 legacy drivers trying to access reserved resources
244 will bind successfully but a warning message is logged.
245 no: ACPI OperationRegions are not marked as reserved,
246 no further checks are performed.
248 acpi_force_table_verification [HW,ACPI]
249 Enable table checksum verification during early stage.
250 By default, this is disabled due to x86 early mapping
251 size limitation.
253 acpi_irq_balance [HW,ACPI]
254 ACPI will balance active IRQs
255 default in APIC mode
257 acpi_irq_nobalance [HW,ACPI]
258 ACPI will not move active IRQs (default)
259 default in PIC mode
261 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
262 Format: <irq>,<irq>...
264 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
265 use by PCI
266 Format: <irq>,<irq>...
268 acpi_no_auto_serialize [HW,ACPI]
269 Disable auto-serialization of AML methods
270 AML control methods that contain the opcodes to create
271 named objects will be marked as "Serialized" by the
272 auto-serialization feature.
273 This feature is enabled by default.
274 This option allows to turn off the feature.
276 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
277 kernels.
279 acpi_no_static_ssdt [HW,ACPI]
280 Disable installation of static SSDTs at early boot time
281 By default, SSDTs contained in the RSDT/XSDT will be
282 installed automatically and they will appear under
283 /sys/firmware/acpi/tables.
284 This option turns off this feature.
285 Note that specifying this option does not affect
286 dynamic table installation which will install SSDT
287 tables to /sys/firmware/acpi/tables/dynamic.
289 acpi_rsdp= [ACPI,EFI,KEXEC]
290 Pass the RSDP address to the kernel, mostly used
291 on machines running EFI runtime service to boot the
292 second kernel for kdump.
294 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
295 Format: To spoof as Windows 98: ="Microsoft Windows"
297 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
298 of 2 which is mandated by ACPI 6) as the supported ACPI
299 specification revision (when using this switch, it may
300 be necessary to carry out a cold reboot _twice_ in a
301 row to make it take effect on the platform firmware).
303 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
304 acpi_osi="string1" # add string1
305 acpi_osi="!string2" # remove string2
306 acpi_osi=!* # remove all strings
307 acpi_osi=! # disable all built-in OS vendor
308 strings
309 acpi_osi= # disable all strings
311 'acpi_osi=!' can be used in combination with single or
312 multiple 'acpi_osi="string1"' to support specific OS
313 vendor string(s). Note that such command can only
314 affect the default state of the OS vendor strings, thus
315 it cannot affect the default state of the feature group
316 strings and the current state of the OS vendor strings,
317 specifying it multiple times through kernel command line
318 is meaningless. This command is useful when one do not
319 care about the state of the feature group strings which
320 should be controlled by the OSPM.
321 Examples:
322 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
323 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
324 can make '_OSI("Windows 2000")' TRUE.
326 'acpi_osi=' cannot be used in combination with other
327 'acpi_osi=' command lines, the _OSI method will not
328 exist in the ACPI namespace. NOTE that such command can
329 only affect the _OSI support state, thus specifying it
330 multiple times through kernel command line is also
331 meaningless.
332 Examples:
333 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
334 FALSE.
336 'acpi_osi=!*' can be used in combination with single or
337 multiple 'acpi_osi="string1"' to support specific
338 string(s). Note that such command can affect the
339 current state of both the OS vendor strings and the
340 feature group strings, thus specifying it multiple times
341 through kernel command line is meaningful. But it may
342 still not able to affect the final state of a string if
343 there are quirks related to this string. This command
344 is useful when one want to control the state of the
345 feature group strings to debug BIOS issues related to
346 the OSPM features.
347 Examples:
348 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
349 '_OSI("Module Device")' FALSE.
350 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
351 '_OSI("Module Device")' TRUE.
352 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
353 equivalent to
354 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
355 and
356 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
357 they all will make '_OSI("Windows 2000")' TRUE.
359 acpi_pm_good [X86]
360 Override the pmtimer bug detection: force the kernel
361 to assume that this machine's pmtimer latches its value
362 and always returns good values.
364 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
365 Format: { level | edge | high | low }
367 acpi_skip_timer_override [HW,ACPI]
368 Recognize and ignore IRQ0/pin2 Interrupt Override.
369 For broken nForce2 BIOS resulting in XT-PIC timer.
371 acpi_sleep= [HW,ACPI] Sleep options
372 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
373 old_ordering, nonvs, sci_force_enable }
374 See Documentation/power/video.txt for information on
375 s3_bios and s3_mode.
376 s3_beep is for debugging; it makes the PC's speaker beep
377 as soon as the kernel's real-mode entry point is called.
378 s4_nohwsig prevents ACPI hardware signature from being
379 used during resume from hibernation.
380 old_ordering causes the ACPI 1.0 ordering of the _PTS
381 control method, with respect to putting devices into
382 low power states, to be enforced (the ACPI 2.0 ordering
383 of _PTS is used by default).
384 nonvs prevents the kernel from saving/restoring the
385 ACPI NVS memory during suspend/hibernation and resume.
386 sci_force_enable causes the kernel to set SCI_EN directly
387 on resume from S1/S3 (which is against the ACPI spec,
388 but some broken systems don't work without it).
390 acpi_use_timer_override [HW,ACPI]
391 Use timer override. For some broken Nvidia NF5 boards
392 that require a timer override, but don't have HPET
394 add_efi_memmap [EFI; X86] Include EFI memory map in
395 kernel's map of available physical RAM.
397 agp= [AGP]
398 { off | try_unsupported }
399 off: disable AGP support
400 try_unsupported: try to drive unsupported chipsets
401 (may crash computer or cause data corruption)
403 ALSA [HW,ALSA]
404 See Documentation/sound/alsa/alsa-parameters.txt
406 alignment= [KNL,ARM]
407 Allow the default userspace alignment fault handler
408 behaviour to be specified. Bit 0 enables warnings,
409 bit 1 enables fixups, and bit 2 sends a segfault.
411 align_va_addr= [X86-64]
412 Align virtual addresses by clearing slice [14:12] when
413 allocating a VMA at process creation time. This option
414 gives you up to 3% performance improvement on AMD F15h
415 machines (where it is enabled by default) for a
416 CPU-intensive style benchmark, and it can vary highly in
417 a microbenchmark depending on workload and compiler.
419 32: only for 32-bit processes
420 64: only for 64-bit processes
421 on: enable for both 32- and 64-bit processes
422 off: disable for both 32- and 64-bit processes
424 alloc_snapshot [FTRACE]
425 Allocate the ftrace snapshot buffer on boot up when the
426 main buffer is allocated. This is handy if debugging
427 and you need to use tracing_snapshot() on boot up, and
428 do not want to use tracing_snapshot_alloc() as it needs
429 to be done where GFP_KERNEL allocations are allowed.
431 amd_iommu= [HW,X86-64]
432 Pass parameters to the AMD IOMMU driver in the system.
433 Possible values are:
434 fullflush - enable flushing of IO/TLB entries when
435 they are unmapped. Otherwise they are
436 flushed before they will be reused, which
437 is a lot of faster
438 off - do not initialize any AMD IOMMU found in
439 the system
440 force_isolation - Force device isolation for all
441 devices. The IOMMU driver is not
442 allowed anymore to lift isolation
443 requirements as needed. This option
444 does not override iommu=pt
446 amd_iommu_dump= [HW,X86-64]
447 Enable AMD IOMMU driver option to dump the ACPI table
448 for AMD IOMMU. With this option enabled, AMD IOMMU
449 driver will print ACPI tables for AMD IOMMU during
450 IOMMU initialization.
452 amijoy.map= [HW,JOY] Amiga joystick support
453 Map of devices attached to JOY0DAT and JOY1DAT
454 Format: <a>,<b>
455 See also Documentation/input/joystick.txt
457 analog.map= [HW,JOY] Analog joystick and gamepad support
458 Specifies type or capabilities of an analog joystick
459 connected to one of 16 gameports
460 Format: <type1>,<type2>,..<type16>
462 apc= [HW,SPARC]
463 Power management functions (SPARCstation-4/5 + deriv.)
464 Format: noidle
465 Disable APC CPU standby support. SPARCstation-Fox does
466 not play well with APC CPU idle - disable it if you have
467 APC and your system crashes randomly.
469 apic= [APIC,X86-32] Advanced Programmable Interrupt Controller
470 Change the output verbosity whilst booting
471 Format: { quiet (default) | verbose | debug }
472 Change the amount of debugging information output
473 when initialising the APIC and IO-APIC components.
475 apic_extnmi= [APIC,X86] External NMI delivery setting
476 Format: { bsp (default) | all | none }
477 bsp: External NMI is delivered only to CPU 0
478 all: External NMIs are broadcast to all CPUs as a
479 backup of CPU 0
480 none: External NMI is masked for all CPUs. This is
481 useful so that a dump capture kernel won't be
482 shot down by NMI
484 autoconf= [IPV6]
485 See Documentation/networking/ipv6.txt.
487 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
488 Limit apic dumping. The parameter defines the maximal
489 number of local apics being dumped. Also it is possible
490 to set it to "all" by meaning -- no limit here.
491 Format: { 1 (default) | 2 | ... | all }.
492 The parameter valid if only apic=debug or
493 apic=verbose is specified.
494 Example: apic=debug show_lapic=all
496 apm= [APM] Advanced Power Management
497 See header of arch/x86/kernel/apm_32.c.
499 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
500 Format: <io>,<irq>,<nodeID>
502 ataflop= [HW,M68k]
504 atarimouse= [HW,MOUSE] Atari Mouse
506 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
507 EzKey and similar keyboards
509 atkbd.reset= [HW] Reset keyboard during initialization
511 atkbd.set= [HW] Select keyboard code set
512 Format: <int> (2 = AT (default), 3 = PS/2)
514 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
515 keyboards
517 atkbd.softraw= [HW] Choose between synthetic and real raw mode
518 Format: <bool> (0 = real, 1 = synthetic (default))
520 atkbd.softrepeat= [HW]
521 Use software keyboard repeat
523 audit= [KNL] Enable the audit sub-system
524 Format: { "0" | "1" } (0 = disabled, 1 = enabled)
525 0 - kernel audit is disabled and can not be enabled
526 until the next reboot
527 unset - kernel audit is initialized but disabled and
528 will be fully enabled by the userspace auditd.
529 1 - kernel audit is initialized and partially enabled,
530 storing at most audit_backlog_limit messages in
531 RAM until it is fully enabled by the userspace
532 auditd.
533 Default: unset
535 audit_backlog_limit= [KNL] Set the audit queue size limit.
536 Format: <int> (must be >=0)
537 Default: 64
539 baycom_epp= [HW,AX25]
540 Format: <io>,<mode>
542 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
543 Format: <io>,<mode>
544 See header of drivers/net/hamradio/baycom_par.c.
546 baycom_ser_fdx= [HW,AX25]
547 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
548 Format: <io>,<irq>,<mode>[,<baud>]
549 See header of drivers/net/hamradio/baycom_ser_fdx.c.
551 baycom_ser_hdx= [HW,AX25]
552 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
553 Format: <io>,<irq>,<mode>
554 See header of drivers/net/hamradio/baycom_ser_hdx.c.
556 blkdevparts= Manual partition parsing of block device(s) for
557 embedded devices based on command line input.
558 See Documentation/block/cmdline-partition.txt
560 boot_delay= Milliseconds to delay each printk during boot.
561 Values larger than 10 seconds (10000) are changed to
562 no delay (0).
563 Format: integer
565 bootmem_debug [KNL] Enable bootmem allocator debug messages.
567 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
568 bttv.radio= Most important insmod options are available as
569 kernel args too.
570 bttv.pll= See Documentation/video4linux/bttv/Insmod-options
571 bttv.tuner=
573 bulk_remove=off [PPC] This parameter disables the use of the pSeries
574 firmware feature for flushing multiple hpte entries
575 at a time.
577 c101= [NET] Moxa C101 synchronous serial card
579 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
580 Sometimes CPU hardware bugs make them report the cache
581 size incorrectly. The kernel will attempt work arounds
582 to fix known problems, but for some CPUs it is not
583 possible to determine what the correct size should be.
584 This option provides an override for these situations.
586 ca_keys= [KEYS] This parameter identifies a specific key(s) on
587 the system trusted keyring to be used for certificate
588 trust validation.
589 format: { id:<keyid> | builtin }
591 cca= [MIPS] Override the kernel pages' cache coherency
592 algorithm. Accepted values range from 0 to 7
593 inclusive. See arch/mips/include/asm/pgtable-bits.h
594 for platform specific values (SB1, Loongson3 and
595 others).
597 ccw_timeout_log [S390]
598 See Documentation/s390/CommonIO for details.
600 cgroup_disable= [KNL] Disable a particular controller
601 Format: {name of the controller(s) to disable}
602 The effects of cgroup_disable=foo are:
603 - foo isn't auto-mounted if you mount all cgroups in
604 a single hierarchy
605 - foo isn't visible as an individually mountable
606 subsystem
607 {Currently only "memory" controller deal with this and
608 cut the overhead, others just disable the usage. So
609 only cgroup_disable=memory is actually worthy}
611 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
612 Format: <string>
613 nosocket -- Disable socket memory accounting.
614 nokmem -- Disable kernel memory accounting.
616 checkreqprot [SELINUX] Set initial checkreqprot flag value.
617 Format: { "0" | "1" }
618 See security/selinux/Kconfig help text.
619 0 -- check protection applied by kernel (includes
620 any implied execute protection).
621 1 -- check protection requested by application.
622 Default value is set via a kernel config option.
623 Value can be changed at runtime via
624 /selinux/checkreqprot.
626 cio_ignore= [S390]
627 See Documentation/s390/CommonIO for details.
628 clk_ignore_unused
629 [CLK]
630 Prevents the clock framework from automatically gating
631 clocks that have not been explicitly enabled by a Linux
632 device driver but are enabled in hardware at reset or
633 by the bootloader/firmware. Note that this does not
634 force such clocks to be always-on nor does it reserve
635 those clocks in any way. This parameter is useful for
636 debug and development, but should not be needed on a
637 platform with proper driver support. For more
638 information, see Documentation/clk.txt.
640 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
641 [Deprecated]
642 Forces specified clocksource (if available) to be used
643 when calculating gettimeofday(). If specified
644 clocksource is not available, it defaults to PIT.
645 Format: { pit | tsc | cyclone | pmtmr }
647 clocksource= Override the default clocksource
648 Format: <string>
649 Override the default clocksource and use the clocksource
650 with the name specified.
651 Some clocksource names to choose from, depending on
652 the platform:
653 [all] jiffies (this is the base, fallback clocksource)
654 [ACPI] acpi_pm
655 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
656 pxa_timer,timer3,32k_counter,timer0_1
657 [AVR32] avr32
658 [X86-32] pit,hpet,tsc;
659 scx200_hrt on Geode; cyclone on IBM x440
660 [MIPS] MIPS
661 [PARISC] cr16
662 [S390] tod
663 [SH] SuperH
664 [SPARC64] tick
665 [X86-64] hpet,tsc
667 clearcpuid=BITNUM [X86]
668 Disable CPUID feature X for the kernel. See
669 arch/x86/include/asm/cpufeature.h for the valid bit
670 numbers. Note the Linux specific bits are not necessarily
671 stable over kernel options, but the vendor specific
672 ones should be.
673 Also note that user programs calling CPUID directly
674 or using the feature without checking anything
675 will still see it. This just prevents it from
676 being used by the kernel or shown in /proc/cpuinfo.
677 Also note the kernel might malfunction if you disable
678 some critical bits.
680 cma=nn[MG]@[start[MG][-end[MG]]]
681 [ARM,X86,KNL]
682 Sets the size of kernel global memory area for
683 contiguous memory allocations and optionally the
684 placement constraint by the physical address range of
685 memory allocations. A value of 0 disables CMA
686 altogether. For more information, see
687 include/linux/dma-contiguous.h
689 cmo_free_hint= [PPC] Format: { yes | no }
690 Specify whether pages are marked as being inactive
691 when they are freed. This is used in CMO environments
692 to determine OS memory pressure for page stealing by
693 a hypervisor.
694 Default: yes
696 coherent_pool=nn[KMG] [ARM,KNL]
697 Sets the size of memory pool for coherent, atomic dma
698 allocations, by default set to 256K.
700 code_bytes [X86] How many bytes of object code to print
701 in an oops report.
702 Range: 0 - 8192
703 Default: 64
705 com20020= [HW,NET] ARCnet - COM20020 chipset
706 Format:
707 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
709 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
710 Format: <io>[,<irq>]
712 com90xx= [HW,NET]
713 ARCnet - COM90xx chipset (memory-mapped buffers)
714 Format: <io>[,<irq>[,<memstart>]]
716 condev= [HW,S390] console device
717 conmode=
719 console= [KNL] Output console device and options.
721 tty<n> Use the virtual console device <n>.
723 ttyS<n>[,options]
724 ttyUSB0[,options]
725 Use the specified serial port. The options are of
726 the form "bbbbpnf", where "bbbb" is the baud rate,
727 "p" is parity ("n", "o", or "e"), "n" is number of
728 bits, and "f" is flow control ("r" for RTS or
729 omit it). Default is "9600n8".
731 See Documentation/serial-console.txt for more
732 information. See
733 Documentation/networking/netconsole.txt for an
734 alternative.
736 uart[8250],io,<addr>[,options]
737 uart[8250],mmio,<addr>[,options]
738 uart[8250],mmio16,<addr>[,options]
739 uart[8250],mmio32,<addr>[,options]
740 uart[8250],0x<addr>[,options]
741 Start an early, polled-mode console on the 8250/16550
742 UART at the specified I/O port or MMIO address,
743 switching to the matching ttyS device later.
744 MMIO inter-register address stride is either 8-bit
745 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
746 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
747 to be equivalent to 'mmio'. 'options' are specified in
748 the same format described for ttyS above; if unspecified,
749 the h/w is not re-initialized.
751 hvc<n> Use the hypervisor console device <n>. This is for
752 both Xen and PowerPC hypervisors.
754 If the device connected to the port is not a TTY but a braille
755 device, prepend "brl," before the device type, for instance
756 console=brl,ttyS0
757 For now, only VisioBraille is supported.
759 consoleblank= [KNL] The console blank (screen saver) timeout in
760 seconds. Defaults to 10*60 = 10mins. A value of 0
761 disables the blank timer.
763 coredump_filter=
764 [KNL] Change the default value for
765 /proc/<pid>/coredump_filter.
766 See also Documentation/filesystems/proc.txt.
768 cpuidle.off=1 [CPU_IDLE]
769 disable the cpuidle sub-system
771 cpu_init_udelay=N
772 [X86] Delay for N microsec between assert and de-assert
773 of APIC INIT to start processors. This delay occurs
774 on every CPU online, such as boot, and resume from suspend.
775 Default: 10000
777 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
778 Format:
779 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
781 crashkernel=size[KMG][@offset[KMG]]
782 [KNL] Using kexec, Linux can switch to a 'crash kernel'
783 upon panic. This parameter reserves the physical
784 memory region [offset, offset + size] for that kernel
785 image. If '@offset' is omitted, then a suitable offset
786 is selected automatically. Check
787 Documentation/kdump/kdump.txt for further details.
789 crashkernel=range1:size1[,range2:size2,...][@offset]
790 [KNL] Same as above, but depends on the memory
791 in the running system. The syntax of range is
792 start-[end] where start and end are both
793 a memory unit (amount[KMG]). See also
794 Documentation/kdump/kdump.txt for an example.
796 crashkernel=size[KMG],high
797 [KNL, x86_64] range could be above 4G. Allow kernel
798 to allocate physical memory region from top, so could
799 be above 4G if system have more than 4G ram installed.
800 Otherwise memory region will be allocated below 4G, if
801 available.
802 It will be ignored if crashkernel=X is specified.
803 crashkernel=size[KMG],low
804 [KNL, x86_64] range under 4G. When crashkernel=X,high
805 is passed, kernel could allocate physical memory region
806 above 4G, that cause second kernel crash on system
807 that require some amount of low memory, e.g. swiotlb
808 requires at least 64M+32K low memory, also enough extra
809 low memory is needed to make sure DMA buffers for 32-bit
810 devices won't run out. Kernel would try to allocate at
811 at least 256M below 4G automatically.
812 This one let user to specify own low range under 4G
813 for second kernel instead.
814 0: to disable low allocation.
815 It will be ignored when crashkernel=X,high is not used
816 or memory reserved is below 4G.
818 cs89x0_dma= [HW,NET]
819 Format: <dma>
821 cs89x0_media= [HW,NET]
822 Format: { rj45 | aui | bnc }
824 dasd= [HW,NET]
825 See header of drivers/s390/block/dasd_devmap.c.
827 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
828 (one device per port)
829 Format: <port#>,<type>
830 See also Documentation/input/joystick-parport.txt
832 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
833 time. See Documentation/dynamic-debug-howto.txt for
834 details. Deprecated, see dyndbg.
836 debug [KNL] Enable kernel debugging (events log level).
838 debug_locks_verbose=
839 [KNL] verbose self-tests
840 Format=<0|1>
841 Print debugging info while doing the locking API
842 self-tests.
843 We default to 0 (no extra messages), setting it to
844 1 will print _a lot_ more information - normally
845 only useful to kernel developers.
847 debug_objects [KNL] Enable object debugging
849 no_debug_objects
850 [KNL] Disable object debugging
852 debug_guardpage_minorder=
853 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
854 parameter allows control of the order of pages that will
855 be intentionally kept free (and hence protected) by the
856 buddy allocator. Bigger value increase the probability
857 of catching random memory corruption, but reduce the
858 amount of memory for normal system use. The maximum
859 possible value is MAX_ORDER/2. Setting this parameter
860 to 1 or 2 should be enough to identify most random
861 memory corruption problems caused by bugs in kernel or
862 driver code when a CPU writes to (or reads from) a
863 random memory location. Note that there exists a class
864 of memory corruptions problems caused by buggy H/W or
865 F/W or by drivers badly programing DMA (basically when
866 memory is written at bus level and the CPU MMU is
867 bypassed) which are not detectable by
868 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
869 tracking down these problems.
871 debug_pagealloc=
872 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
873 parameter enables the feature at boot time. In
874 default, it is disabled. We can avoid allocating huge
875 chunk of memory for debug pagealloc if we don't enable
876 it at boot time and the system will work mostly same
877 with the kernel built without CONFIG_DEBUG_PAGEALLOC.
878 on: enable the feature
880 debugpat [X86] Enable PAT debugging
882 decnet.addr= [HW,NET]
883 Format: <area>[,<node>]
884 See also Documentation/networking/decnet.txt.
886 default_hugepagesz=
887 [same as hugepagesz=] The size of the default
888 HugeTLB page size. This is the size represented by
889 the legacy /proc/ hugepages APIs, used for SHM, and
890 default size when mounting hugetlbfs filesystems.
891 Defaults to the default architecture's huge page size
892 if not specified.
894 dhash_entries= [KNL]
895 Set number of hash buckets for dentry cache.
897 disable= [IPV6]
898 See Documentation/networking/ipv6.txt.
900 disable_cpu_apicid= [X86,APIC,SMP]
901 Format: <int>
902 The number of initial APIC ID for the
903 corresponding CPU to be disabled at boot,
904 mostly used for the kdump 2nd kernel to
905 disable BSP to wake up multiple CPUs without
906 causing system reset or hang due to sending
907 INIT from AP to BSP.
909 disable_ddw [PPC/PSERIES]
910 Disable Dynamic DMA Window support. Use this if
911 to workaround buggy firmware.
913 disable_ipv6= [IPV6]
914 See Documentation/networking/ipv6.txt.
916 disable_mtrr_cleanup [X86]
917 The kernel tries to adjust MTRR layout from continuous
918 to discrete, to make X server driver able to add WB
919 entry later. This parameter disables that.
921 disable_mtrr_trim [X86, Intel and AMD only]
922 By default the kernel will trim any uncacheable
923 memory out of your available memory pool based on
924 MTRR settings. This parameter disables that behavior,
925 possibly causing your machine to run very slowly.
927 disable_timer_pin_1 [X86]
928 Disable PIN 1 of APIC timer
929 Can be useful to work around chipset bugs.
931 dis_ucode_ldr [X86] Disable the microcode loader.
933 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
934 this option disables the debugging code at boot.
936 dma_debug_entries=<number>
937 This option allows to tune the number of preallocated
938 entries for DMA-API debugging code. One entry is
939 required per DMA-API allocation. Use this if the
940 DMA-API debugging code disables itself because the
941 architectural default is too low.
943 dma_debug_driver=<driver_name>
944 With this option the DMA-API debugging driver
945 filter feature can be enabled at boot time. Just
946 pass the driver to filter for as the parameter.
947 The filter can be disabled or changed to another
948 driver later using sysfs.
950 drm_kms_helper.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
951 Broken monitors, graphic adapters, KVMs and EDIDless
952 panels may send no or incorrect EDID data sets.
953 This parameter allows to specify an EDID data sets
954 in the /lib/firmware directory that are used instead.
955 Generic built-in EDID data sets are used, if one of
956 edid/1024x768.bin, edid/1280x1024.bin,
957 edid/1680x1050.bin, or edid/1920x1080.bin is given
958 and no file with the same name exists. Details and
959 instructions how to build your own EDID data are
960 available in Documentation/EDID/HOWTO.txt. An EDID
961 data set will only be used for a particular connector,
962 if its name and a colon are prepended to the EDID
963 name. Each connector may use a unique EDID data
964 set by separating the files with a comma. An EDID
965 data set with no connector name will be used for
966 any connectors not explicitly specified.
968 dscc4.setup= [NET]
970 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
971 module.dyndbg[="val"]
972 Enable debug messages at boot time. See
973 Documentation/dynamic-debug-howto.txt for details.
975 nompx [X86] Disables Intel Memory Protection Extensions.
976 See Documentation/x86/intel_mpx.txt for more
977 information about the feature.
979 eagerfpu= [X86]
980 on enable eager fpu restore
981 off disable eager fpu restore
982 auto selects the default scheme, which automatically
983 enables eagerfpu restore for xsaveopt.
985 module.async_probe [KNL]
986 Enable asynchronous probe on this module.
988 early_ioremap_debug [KNL]
989 Enable debug messages in early_ioremap support. This
990 is useful for tracking down temporary early mappings
991 which are not unmapped.
993 earlycon= [KNL] Output early console device and options.
995 When used with no options, the early console is
996 determined by the stdout-path property in device
997 tree's chosen node.
999 cdns,<addr>
1000 Start an early, polled-mode console on a cadence serial
1001 port at the specified address. The cadence serial port
1002 must already be setup and configured. Options are not
1003 yet supported.
1005 uart[8250],io,<addr>[,options]
1006 uart[8250],mmio,<addr>[,options]
1007 uart[8250],mmio32,<addr>[,options]
1008 uart[8250],mmio32be,<addr>[,options]
1009 uart[8250],0x<addr>[,options]
1010 Start an early, polled-mode console on the 8250/16550
1011 UART at the specified I/O port or MMIO address.
1012 MMIO inter-register address stride is either 8-bit
1013 (mmio) or 32-bit (mmio32 or mmio32be).
1014 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1015 to be equivalent to 'mmio'. 'options' are specified
1016 in the same format described for "console=ttyS<n>"; if
1017 unspecified, the h/w is not initialized.
1019 pl011,<addr>
1020 pl011,mmio32,<addr>
1021 Start an early, polled-mode console on a pl011 serial
1022 port at the specified address. The pl011 serial port
1023 must already be setup and configured. Options are not
1024 yet supported. If 'mmio32' is specified, then only
1025 the driver will use only 32-bit accessors to read/write
1026 the device registers.
1028 msm_serial,<addr>
1029 Start an early, polled-mode console on an msm serial
1030 port at the specified address. The serial port
1031 must already be setup and configured. Options are not
1032 yet supported.
1034 msm_serial_dm,<addr>
1035 Start an early, polled-mode console on an msm serial
1036 dm port at the specified address. The serial port
1037 must already be setup and configured. Options are not
1038 yet supported.
1040 smh Use ARM semihosting calls for early console.
1042 s3c2410,<addr>
1043 s3c2412,<addr>
1044 s3c2440,<addr>
1045 s3c6400,<addr>
1046 s5pv210,<addr>
1047 exynos4210,<addr>
1048 Use early console provided by serial driver available
1049 on Samsung SoCs, requires selecting proper type and
1050 a correct base address of the selected UART port. The
1051 serial port must already be setup and configured.
1052 Options are not yet supported.
1054 lpuart,<addr>
1055 lpuart32,<addr>
1056 Use early console provided by Freescale LP UART driver
1057 found on Freescale Vybrid and QorIQ LS1021A processors.
1058 A valid base address must be provided, and the serial
1059 port must already be setup and configured.
1061 earlyprintk= [X86,SH,BLACKFIN,ARM,M68k]
1062 earlyprintk=vga
1063 earlyprintk=efi
1064 earlyprintk=xen
1065 earlyprintk=serial[,ttySn[,baudrate]]
1066 earlyprintk=serial[,0x...[,baudrate]]
1067 earlyprintk=ttySn[,baudrate]
1068 earlyprintk=dbgp[debugController#]
1069 earlyprintk=pciserial,bus:device.function[,baudrate]
1071 earlyprintk is useful when the kernel crashes before
1072 the normal console is initialized. It is not enabled by
1073 default because it has some cosmetic problems.
1075 Append ",keep" to not disable it when the real console
1076 takes over.
1078 Only one of vga, efi, serial, or usb debug port can
1079 be used at a time.
1081 Currently only ttyS0 and ttyS1 may be specified by
1082 name. Other I/O ports may be explicitly specified
1083 on some architectures (x86 and arm at least) by
1084 replacing ttySn with an I/O port address, like this:
1085 earlyprintk=serial,0x1008,115200
1086 You can find the port for a given device in
1087 /proc/tty/driver/serial:
1088 2: uart:ST16650V2 port:00001008 irq:18 ...
1090 Interaction with the standard serial driver is not
1091 very good.
1093 The VGA and EFI output is eventually overwritten by
1094 the real console.
1096 The xen output can only be used by Xen PV guests.
1098 edac_report= [HW,EDAC] Control how to report EDAC event
1099 Format: {"on" | "off" | "force"}
1100 on: enable EDAC to report H/W event. May be overridden
1101 by other higher priority error reporting module.
1102 off: disable H/W event reporting through EDAC.
1103 force: enforce the use of EDAC to report H/W event.
1104 default: on.
1106 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1107 ekgdboc=kbd
1109 This is designed to be used in conjunction with
1110 the boot argument: earlyprintk=vga
1112 edd= [EDD]
1113 Format: {"off" | "on" | "skip[mbr]"}
1115 efi= [EFI]
1116 Format: { "old_map", "nochunk", "noruntime", "debug" }
1117 old_map [X86-64]: switch to the old ioremap-based EFI
1118 runtime services mapping. 32-bit still uses this one by
1119 default.
1120 nochunk: disable reading files in "chunks" in the EFI
1121 boot stub, as chunking can cause problems with some
1122 firmware implementations.
1123 noruntime : disable EFI runtime services support
1124 debug: enable misc debug output
1126 efi_no_storage_paranoia [EFI; X86]
1127 Using this parameter you can use more than 50% of
1128 your efi variable storage. Use this parameter only if
1129 you are really sure that your UEFI does sane gc and
1130 fulfills the spec otherwise your board may brick.
1132 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1133 Add arbitrary attribute to specific memory range by
1134 updating original EFI memory map.
1135 Region of memory which aa attribute is added to is
1136 from ss to ss+nn.
1137 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1138 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1139 attribute is added to range 0x100000000-0x180000000 and
1140 0x10a0000000-0x1120000000.
1142 Using this parameter you can do debugging of EFI memmap
1143 related feature. For example, you can do debugging of
1144 Address Range Mirroring feature even if your box
1145 doesn't support it.
1147 eisa_irq_edge= [PARISC,HW]
1148 See header of drivers/parisc/eisa.c.
1150 elanfreq= [X86-32]
1151 See comment before function elanfreq_setup() in
1152 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1154 elevator= [IOSCHED]
1155 Format: {"cfq" | "deadline" | "noop"}
1156 See Documentation/block/cfq-iosched.txt and
1157 Documentation/block/deadline-iosched.txt for details.
1159 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1160 Specifies physical address of start of kernel core
1161 image elf header and optionally the size. Generally
1162 kexec loader will pass this option to capture kernel.
1163 See Documentation/kdump/kdump.txt for details.
1165 enable_mtrr_cleanup [X86]
1166 The kernel tries to adjust MTRR layout from continuous
1167 to discrete, to make X server driver able to add WB
1168 entry later. This parameter enables that.
1170 enable_timer_pin_1 [X86]
1171 Enable PIN 1 of APIC timer
1172 Can be useful to work around chipset bugs
1173 (in particular on some ATI chipsets).
1174 The kernel tries to set a reasonable default.
1176 enforcing [SELINUX] Set initial enforcing status.
1177 Format: {"0" | "1"}
1178 See security/selinux/Kconfig help text.
1179 0 -- permissive (log only, no denials).
1180 1 -- enforcing (deny and log).
1181 Default value is 0.
1182 Value can be changed at runtime via /selinux/enforce.
1184 erst_disable [ACPI]
1185 Disable Error Record Serialization Table (ERST)
1186 support.
1188 ether= [HW,NET] Ethernet cards parameters
1189 This option is obsoleted by the "netdev=" option, which
1190 has equivalent usage. See its documentation for details.
1192 evm= [EVM]
1193 Format: { "fix" }
1194 Permit 'security.evm' to be updated regardless of
1195 current integrity status.
1197 failslab=
1198 fail_page_alloc=
1199 fail_make_request=[KNL]
1200 General fault injection mechanism.
1201 Format: <interval>,<probability>,<space>,<times>
1202 See also Documentation/fault-injection/.
1204 floppy= [HW]
1205 See Documentation/blockdev/floppy.txt.
1207 force_pal_cache_flush
1208 [IA-64] Avoid check_sal_cache_flush which may hang on
1209 buggy SAL_CACHE_FLUSH implementations. Using this
1210 parameter will force ia64_sal_cache_flush to call
1211 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1213 forcepae [X86-32]
1214 Forcefully enable Physical Address Extension (PAE).
1215 Many Pentium M systems disable PAE but may have a
1216 functionally usable PAE implementation.
1217 Warning: use of this parameter will taint the kernel
1218 and may cause unknown problems.
1220 ftrace=[tracer]
1221 [FTRACE] will set and start the specified tracer
1222 as early as possible in order to facilitate early
1223 boot debugging.
1225 ftrace_dump_on_oops[=orig_cpu]
1226 [FTRACE] will dump the trace buffers on oops.
1227 If no parameter is passed, ftrace will dump
1228 buffers of all CPUs, but if you pass orig_cpu, it will
1229 dump only the buffer of the CPU that triggered the
1230 oops.
1232 ftrace_filter=[function-list]
1233 [FTRACE] Limit the functions traced by the function
1234 tracer at boot up. function-list is a comma separated
1235 list of functions. This list can be changed at run
1236 time by the set_ftrace_filter file in the debugfs
1237 tracing directory.
1239 ftrace_notrace=[function-list]
1240 [FTRACE] Do not trace the functions specified in
1241 function-list. This list can be changed at run time
1242 by the set_ftrace_notrace file in the debugfs
1243 tracing directory.
1245 ftrace_graph_filter=[function-list]
1246 [FTRACE] Limit the top level callers functions traced
1247 by the function graph tracer at boot up.
1248 function-list is a comma separated list of functions
1249 that can be changed at run time by the
1250 set_graph_function file in the debugfs tracing directory.
1252 ftrace_graph_notrace=[function-list]
1253 [FTRACE] Do not trace from the functions specified in
1254 function-list. This list is a comma separated list of
1255 functions that can be changed at run time by the
1256 set_graph_notrace file in the debugfs tracing directory.
1258 gamecon.map[2|3]=
1259 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1260 support via parallel port (up to 5 devices per port)
1261 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1262 See also Documentation/input/joystick-parport.txt
1264 gamma= [HW,DRM]
1266 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1267 Format: off | on
1268 default: on
1270 gcov_persist= [GCOV] When non-zero (default), profiling data for
1271 kernel modules is saved and remains accessible via
1272 debugfs, even when the module is unloaded/reloaded.
1273 When zero, profiling data is discarded and associated
1274 debugfs files are removed at module unload time.
1276 gpt [EFI] Forces disk with valid GPT signature but
1277 invalid Protective MBR to be treated as GPT. If the
1278 primary GPT is corrupted, it enables the backup/alternate
1279 GPT to be used instead.
1281 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1282 the "Enable 0" bit of the configuration register.
1283 Format: 0 | 1
1284 Default: 0
1285 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1286 the "Enable 0" bit of the configuration register.
1287 Format: 0 | 1
1288 Default: 0
1289 grcan.select= [HW] Select which physical interface to use.
1290 Format: 0 | 1
1291 Default: 0
1292 grcan.txsize= [HW] Sets the size of the tx buffer.
1293 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1294 Default: 1024
1295 grcan.rxsize= [HW] Sets the size of the rx buffer.
1296 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1297 Default: 1024
1299 hardlockup_all_cpu_backtrace=
1300 [KNL] Should the hard-lockup detector generate
1301 backtraces on all cpus.
1302 Format: <integer>
1304 hashdist= [KNL,NUMA] Large hashes allocated during boot
1305 are distributed across NUMA nodes. Defaults on
1306 for 64-bit NUMA, off otherwise.
1307 Format: 0 | 1 (for off | on)
1309 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1311 hd= [EIDE] (E)IDE hard drive subsystem geometry
1312 Format: <cyl>,<head>,<sect>
1314 hest_disable [ACPI]
1315 Disable Hardware Error Source Table (HEST) support;
1316 corresponding firmware-first mode error processing
1317 logic will be disabled.
1319 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1320 size of <nn>. This works even on boxes that have no
1321 highmem otherwise. This also works to reduce highmem
1322 size on bigger boxes.
1324 highres= [KNL] Enable/disable high resolution timer mode.
1325 Valid parameters: "on", "off"
1326 Default: "on"
1328 hisax= [HW,ISDN]
1329 See Documentation/isdn/README.HiSax.
1331 hlt [BUGS=ARM,SH]
1333 hpet= [X86-32,HPET] option to control HPET usage
1334 Format: { enable (default) | disable | force |
1335 verbose }
1336 disable: disable HPET and use PIT instead
1337 force: allow force enabled of undocumented chips (ICH4,
1338 VIA, nVidia)
1339 verbose: show contents of HPET registers during setup
1341 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1342 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1344 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1345 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
1346 On x86-64 and powerpc, this option can be specified
1347 multiple times interleaved with hugepages= to reserve
1348 huge pages of different sizes. Valid pages sizes on
1349 x86-64 are 2M (when the CPU supports "pse") and 1G
1350 (when the CPU supports the "pdpe1gb" cpuinfo flag).
1352 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1353 terminal devices. Valid values: 0..8
1354 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1355 If specified, z/VM IUCV HVC accepts connections
1356 from listed z/VM user IDs only.
1358 hwthread_map= [METAG] Comma-separated list of Linux cpu id to
1359 hardware thread id mappings.
1360 Format: <cpu>:<hwthread>
1362 keep_bootcon [KNL]
1363 Do not unregister boot console at start. This is only
1364 useful for debugging when something happens in the window
1365 between unregistering the boot console and initializing
1366 the real console.
1368 i2c_bus= [HW] Override the default board specific I2C bus speed
1369 or register an additional I2C bus that is not
1370 registered from board initialization code.
1371 Format:
1372 <bus_id>,<clkrate>
1374 i8042.debug [HW] Toggle i8042 debug mode
1375 i8042.unmask_kbd_data
1376 [HW] Enable printing of interrupt data from the KBD port
1377 (disabled by default, and as a pre-condition
1378 requires that i8042.debug=1 be enabled)
1379 i8042.direct [HW] Put keyboard port into non-translated mode
1380 i8042.dumbkbd [HW] Pretend that controller can only read data from
1381 keyboard and cannot control its state
1382 (Don't attempt to blink the leds)
1383 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1384 i8042.nokbd [HW] Don't check/create keyboard port
1385 i8042.noloop [HW] Disable the AUX Loopback command while probing
1386 for the AUX port
1387 i8042.nomux [HW] Don't check presence of an active multiplexing
1388 controller
1389 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1390 controllers
1391 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1392 i8042.reset [HW] Reset the controller during init and cleanup
1393 i8042.unlock [HW] Unlock (ignore) the keylock
1394 i8042.kbdreset [HW] Reset device connected to KBD port
1396 i810= [HW,DRM]
1398 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1399 indicates that the driver is running on unsupported
1400 hardware.
1401 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1402 does not match list of supported models.
1403 i8k.power_status
1404 [HW] Report power status in /proc/i8k
1405 (disabled by default)
1406 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1407 capability is set.
1409 i915.invert_brightness=
1410 [DRM] Invert the sense of the variable that is used to
1411 set the brightness of the panel backlight. Normally a
1412 brightness value of 0 indicates backlight switched off,
1413 and the maximum of the brightness value sets the backlight
1414 to maximum brightness. If this parameter is set to 0
1415 (default) and the machine requires it, or this parameter
1416 is set to 1, a brightness value of 0 sets the backlight
1417 to maximum brightness, and the maximum of the brightness
1418 value switches the backlight off.
1419 -1 -- never invert brightness
1420 0 -- machine default
1421 1 -- force brightness inversion
1423 icn= [HW,ISDN]
1424 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1426 ide-core.nodma= [HW] (E)IDE subsystem
1427 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1428 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1429 .cdrom .chs .ignore_cable are additional options
1430 See Documentation/ide/ide.txt.
1432 ide-generic.probe-mask= [HW] (E)IDE subsystem
1433 Format: <int>
1434 Probe mask for legacy ISA IDE ports. Depending on
1435 platform up to 6 ports are supported, enabled by
1436 setting corresponding bits in the mask to 1. The
1437 default value is 0x0, which has a special meaning.
1438 On systems that have PCI, it triggers scanning the
1439 PCI bus for the first and the second port, which
1440 are then probed. On systems without PCI the value
1441 of 0x0 enables probing the two first ports as if it
1442 was 0x3.
1444 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1445 Claim all unknown PCI IDE storage controllers.
1447 idle= [X86]
1448 Format: idle=poll, idle=halt, idle=nomwait
1449 Poll forces a polling idle loop that can slightly
1450 improve the performance of waking up a idle CPU, but
1451 will use a lot of power and make the system run hot.
1452 Not recommended.
1453 idle=halt: Halt is forced to be used for CPU idle.
1454 In such case C2/C3 won't be used again.
1455 idle=nomwait: Disable mwait for CPU C-states
1457 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1458 Format: { strict | legacy | 2008 | relaxed }
1459 Default: strict
1461 Choose which programs will be accepted for execution
1462 based on the IEEE 754 NaN encoding(s) supported by
1463 the FPU and the NaN encoding requested with the value
1464 of an ELF file header flag individually set by each
1465 binary. Hardware implementations are permitted to
1466 support either or both of the legacy and the 2008 NaN
1467 encoding mode.
1469 Available settings are as follows:
1470 strict accept binaries that request a NaN encoding
1471 supported by the FPU
1472 legacy only accept legacy-NaN binaries, if supported
1473 by the FPU
1474 2008 only accept 2008-NaN binaries, if supported
1475 by the FPU
1476 relaxed accept any binaries regardless of whether
1477 supported by the FPU
1479 The FPU emulator is always able to support both NaN
1480 encodings, so if no FPU hardware is present or it has
1481 been disabled with 'nofpu', then the settings of
1482 'legacy' and '2008' strap the emulator accordingly,
1483 'relaxed' straps the emulator for both legacy-NaN and
1484 2008-NaN, whereas 'strict' enables legacy-NaN only on
1485 legacy processors and both NaN encodings on MIPS32 or
1486 MIPS64 CPUs.
1488 The setting for ABS.fmt/NEG.fmt instruction execution
1489 mode generally follows that for the NaN encoding,
1490 except where unsupported by hardware.
1492 ignore_loglevel [KNL]
1493 Ignore loglevel setting - this will print /all/
1494 kernel messages to the console. Useful for debugging.
1495 We also add it as printk module parameter, so users
1496 could change it dynamically, usually by
1497 /sys/module/printk/parameters/ignore_loglevel.
1499 ihash_entries= [KNL]
1500 Set number of hash buckets for inode cache.
1502 ima_appraise= [IMA] appraise integrity measurements
1503 Format: { "off" | "enforce" | "fix" | "log" }
1504 default: "enforce"
1506 ima_appraise_tcb [IMA]
1507 The builtin appraise policy appraises all files
1508 owned by uid=0.
1510 ima_hash= [IMA]
1511 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1512 | sha512 | ... }
1513 default: "sha1"
1515 The list of supported hash algorithms is defined
1516 in crypto/hash_info.h.
1518 ima_policy= [IMA]
1519 The builtin measurement policy to load during IMA
1520 setup. Specyfing "tcb" as the value, measures all
1521 programs exec'd, files mmap'd for exec, and all files
1522 opened with the read mode bit set by either the
1523 effective uid (euid=0) or uid=0.
1524 Format: "tcb"
1526 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1527 Load a policy which meets the needs of the Trusted
1528 Computing Base. This means IMA will measure all
1529 programs exec'd, files mmap'd for exec, and all files
1530 opened for read by uid=0.
1532 ima_template= [IMA]
1533 Select one of defined IMA measurements template formats.
1534 Formats: { "ima" | "ima-ng" | "ima-sig" }
1535 Default: "ima-ng"
1537 ima_template_fmt=
1538 [IMA] Define a custom template format.
1539 Format: { "field1|...|fieldN" }
1541 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1542 Format: <min_file_size>
1543 Set the minimal file size for using asynchronous hash.
1544 If left unspecified, ahash usage is disabled.
1546 ahash performance varies for different data sizes on
1547 different crypto accelerators. This option can be used
1548 to achieve the best performance for a particular HW.
1550 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1551 Format: <bufsize>
1552 Set hashing buffer size. Default: 4k.
1554 ahash performance varies for different chunk sizes on
1555 different crypto accelerators. This option can be used
1556 to achieve best performance for particular HW.
1558 init= [KNL]
1559 Format: <full_path>
1560 Run specified binary instead of /sbin/init as init
1561 process.
1563 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1564 for working out where the kernel is dying during
1565 startup.
1567 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1568 initcall functions. Useful for debugging built-in
1569 modules and initcalls.
1571 initrd= [BOOT] Specify the location of the initial ramdisk
1573 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1574 Format: <irq>
1576 int_pln_enable [x86] Enable power limit notification interrupt
1578 integrity_audit=[IMA]
1579 Format: { "0" | "1" }
1580 0 -- basic integrity auditing messages. (Default)
1581 1 -- additional integrity auditing messages.
1583 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1584 on
1585 Enable intel iommu driver.
1586 off
1587 Disable intel iommu driver.
1588 igfx_off [Default Off]
1589 By default, gfx is mapped as normal device. If a gfx
1590 device has a dedicated DMAR unit, the DMAR unit is
1591 bypassed by not enabling DMAR with this option. In
1592 this case, gfx device will use physical address for
1593 DMA.
1594 forcedac [x86_64]
1595 With this option iommu will not optimize to look
1596 for io virtual address below 32-bit forcing dual
1597 address cycle on pci bus for cards supporting greater
1598 than 32-bit addressing. The default is to look
1599 for translation below 32-bit and if not available
1600 then look in the higher range.
1601 strict [Default Off]
1602 With this option on every unmap_single operation will
1603 result in a hardware IOTLB flush operation as opposed
1604 to batching them for performance.
1605 sp_off [Default Off]
1606 By default, super page will be supported if Intel IOMMU
1607 has the capability. With this option, super page will
1608 not be supported.
1609 ecs_off [Default Off]
1610 By default, extended context tables will be supported if
1611 the hardware advertises that it has support both for the
1612 extended tables themselves, and also PASID support. With
1613 this option set, extended tables will not be used even
1614 on hardware which claims to support them.
1616 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1617 0 disables intel_idle and fall back on acpi_idle.
1618 1 to 6 specify maximum depth of C-state.
1620 intel_pstate= [X86]
1621 disable
1622 Do not enable intel_pstate as the default
1623 scaling driver for the supported processors
1624 force
1625 Enable intel_pstate on systems that prohibit it by default
1626 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1627 instead of acpi-cpufreq may disable platform features, such
1628 as thermal controls and power capping, that rely on ACPI
1629 P-States information being indicated to OSPM and therefore
1630 should be used with caution. This option does not work with
1631 processors that aren't supported by the intel_pstate driver
1632 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1633 no_hwp
1634 Do not enable hardware P state control (HWP)
1635 if available.
1636 hwp_only
1637 Only load intel_pstate on systems which support
1638 hardware P state control (HWP) if available.
1640 intremap= [X86-64, Intel-IOMMU]
1641 on enable Interrupt Remapping (default)
1642 off disable Interrupt Remapping
1643 nosid disable Source ID checking
1644 no_x2apic_optout
1645 BIOS x2APIC opt-out request will be ignored
1646 nopost disable Interrupt Posting
1648 iomem= Disable strict checking of access to MMIO memory
1649 strict regions from userspace.
1650 relaxed
1652 iommu= [x86]
1653 off
1654 force
1655 noforce
1656 biomerge
1657 panic
1658 nopanic
1659 merge
1660 nomerge
1661 forcesac
1662 soft
1663 pt [x86, IA-64]
1664 nobypass [PPC/POWERNV]
1665 Disable IOMMU bypass, using IOMMU for PCI devices.
1668 io7= [HW] IO7 for Marvel based alpha systems
1669 See comment before marvel_specify_io7 in
1670 arch/alpha/kernel/core_marvel.c.
1672 io_delay= [X86] I/O delay method
1673 0x80
1674 Standard port 0x80 based delay
1675 0xed
1676 Alternate port 0xed based delay (needed on some systems)
1677 udelay
1678 Simple two microseconds delay
1679 none
1680 No delay
1682 ip= [IP_PNP]
1683 See Documentation/filesystems/nfs/nfsroot.txt.
1685 irqfixup [HW]
1686 When an interrupt is not handled search all handlers
1687 for it. Intended to get systems with badly broken
1688 firmware running.
1690 irqpoll [HW]
1691 When an interrupt is not handled search all handlers
1692 for it. Also check all handlers each timer
1693 interrupt. Intended to get systems with badly broken
1694 firmware running.
1696 isapnp= [ISAPNP]
1697 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1699 isolcpus= [KNL,SMP] Isolate CPUs from the general scheduler.
1700 Format:
1701 <cpu number>,...,<cpu number>
1702 or
1703 <cpu number>-<cpu number>
1704 (must be a positive range in ascending order)
1705 or a mixture
1706 <cpu number>,...,<cpu number>-<cpu number>
1708 This option can be used to specify one or more CPUs
1709 to isolate from the general SMP balancing and scheduling
1710 algorithms. You can move a process onto or off an
1711 "isolated" CPU via the CPU affinity syscalls or cpuset.
1712 <cpu number> begins at 0 and the maximum value is
1713 "number of CPUs in system - 1".
1715 This option is the preferred way to isolate CPUs. The
1716 alternative -- manually setting the CPU mask of all
1717 tasks in the system -- can cause problems and
1718 suboptimal load balancer performance.
1720 iucv= [HW,NET]
1722 ivrs_ioapic [HW,X86_64]
1723 Provide an override to the IOAPIC-ID<->DEVICE-ID
1724 mapping provided in the IVRS ACPI table. For
1725 example, to map IOAPIC-ID decimal 10 to
1726 PCI device 00:14.0 write the parameter as:
1727 ivrs_ioapic[10]=00:14.0
1729 ivrs_hpet [HW,X86_64]
1730 Provide an override to the HPET-ID<->DEVICE-ID
1731 mapping provided in the IVRS ACPI table. For
1732 example, to map HPET-ID decimal 0 to
1733 PCI device 00:14.0 write the parameter as:
1734 ivrs_hpet[0]=00:14.0
1736 js= [HW,JOY] Analog joystick
1737 See Documentation/input/joystick.txt.
1739 kaslr/nokaslr [X86]
1740 Enable/disable kernel and module base offset ASLR
1741 (Address Space Layout Randomization) if built into
1742 the kernel. When CONFIG_HIBERNATION is selected,
1743 kASLR is disabled by default. When kASLR is enabled,
1744 hibernation will be disabled.
1746 keepinitrd [HW,ARM]
1748 kernelcore=nn[KMG] [KNL,X86,IA-64,PPC] This parameter
1749 specifies the amount of memory usable by the kernel
1750 for non-movable allocations. The requested amount is
1751 spread evenly throughout all nodes in the system. The
1752 remaining memory in each node is used for Movable
1753 pages. In the event, a node is too small to have both
1754 kernelcore and Movable pages, kernelcore pages will
1755 take priority and other nodes will have a larger number
1756 of Movable pages. The Movable zone is used for the
1757 allocation of pages that may be reclaimed or moved
1758 by the page migration subsystem. This means that
1759 HugeTLB pages may not be allocated from this zone.
1760 Note that allocations like PTEs-from-HighMem still
1761 use the HighMem zone if it exists, and the Normal
1762 zone if it does not.
1764 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
1765 Format: <Controller#>[,poll interval]
1766 The controller # is the number of the ehci usb debug
1767 port as it is probed via PCI. The poll interval is
1768 optional and is the number seconds in between
1769 each poll cycle to the debug port in case you need
1770 the functionality for interrupting the kernel with
1771 gdb or control-c on the dbgp connection. When
1772 not using this parameter you use sysrq-g to break into
1773 the kernel debugger.
1775 kgdboc= [KGDB,HW] kgdb over consoles.
1776 Requires a tty driver that supports console polling,
1777 or a supported polling keyboard driver (non-usb).
1778 Serial only format: <serial_device>[,baud]
1779 keyboard only format: kbd
1780 keyboard and serial format: kbd,<serial_device>[,baud]
1781 Optional Kernel mode setting:
1782 kms, kbd format: kms,kbd
1783 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
1785 kgdbwait [KGDB] Stop kernel execution and enter the
1786 kernel debugger at the earliest opportunity.
1788 kmac= [MIPS] korina ethernet MAC address.
1789 Configure the RouterBoard 532 series on-chip
1790 Ethernet adapter MAC address.
1792 kmemleak= [KNL] Boot-time kmemleak enable/disable
1793 Valid arguments: on, off
1794 Default: on
1795 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
1796 the default is off.
1798 kmemcheck= [X86] Boot-time kmemcheck enable/disable/one-shot mode
1799 Valid arguments: 0, 1, 2
1800 kmemcheck=0 (disabled)
1801 kmemcheck=1 (enabled)
1802 kmemcheck=2 (one-shot mode)
1803 Default: 2 (one-shot mode)
1805 kstack=N [X86] Print N words from the kernel stack
1806 in oops dumps.
1808 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
1809 Default is 0 (don't ignore, but inject #GP)
1811 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
1812 KVM MMU at runtime.
1813 Default is 0 (off)
1815 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
1816 Default is 1 (enabled)
1818 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
1819 for all guests.
1820 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
1822 kvm-intel.ept= [KVM,Intel] Disable extended page tables
1823 (virtualized MMU) support on capable Intel chips.
1824 Default is 1 (enabled)
1826 kvm-intel.emulate_invalid_guest_state=
1827 [KVM,Intel] Enable emulation of invalid guest states
1828 Default is 0 (disabled)
1830 kvm-intel.flexpriority=
1831 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
1832 Default is 1 (enabled)
1834 kvm-intel.nested=
1835 [KVM,Intel] Enable VMX nesting (nVMX).
1836 Default is 0 (disabled)
1838 kvm-intel.unrestricted_guest=
1839 [KVM,Intel] Disable unrestricted guest feature
1840 (virtualized real and unpaged mode) on capable
1841 Intel chips. Default is 1 (enabled)
1843 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
1844 feature (tagged TLBs) on capable Intel chips.
1845 Default is 1 (enabled)
1847 l2cr= [PPC]
1849 l3cr= [PPC]
1851 lapic [X86-32,APIC] Enable the local APIC even if BIOS
1852 disabled it.
1854 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
1855 value for LAPIC timer one-shot implementation. Default
1856 back to the programmable timer unit in the LAPIC.
1858 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
1859 in C2 power state.
1861 libata.dma= [LIBATA] DMA control
1862 libata.dma=0 Disable all PATA and SATA DMA
1863 libata.dma=1 PATA and SATA Disk DMA only
1864 libata.dma=2 ATAPI (CDROM) DMA only
1865 libata.dma=4 Compact Flash DMA only
1866 Combinations also work, so libata.dma=3 enables DMA
1867 for disks and CDROMs, but not CFs.
1869 libata.ignore_hpa= [LIBATA] Ignore HPA limit
1870 libata.ignore_hpa=0 keep BIOS limits (default)
1871 libata.ignore_hpa=1 ignore limits, using full disk
1873 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
1874 when set.
1875 Format: <int>
1877 libata.force= [LIBATA] Force configurations. The format is comma
1878 separated list of "[ID:]VAL" where ID is
1879 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
1880 matching port, link or device. Basically, it matches
1881 the ATA ID string printed on console by libata. If
1882 the whole ID part is omitted, the last PORT and DEVICE
1883 values are used. If ID hasn't been specified yet, the
1884 configuration applies to all ports, links and devices.
1886 If only DEVICE is omitted, the parameter applies to
1887 the port and all links and devices behind it. DEVICE
1888 number of 0 either selects the first device or the
1889 first fan-out link behind PMP device. It does not
1890 select the host link. DEVICE number of 15 selects the
1891 host link and device attached to it.
1893 The VAL specifies the configuration to force. As long
1894 as there's no ambiguity shortcut notation is allowed.
1895 For example, both 1.5 and 1.5G would work for 1.5Gbps.
1896 The following configurations can be forced.
1898 * Cable type: 40c, 80c, short40c, unk, ign or sata.
1899 Any ID with matching PORT is used.
1901 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
1903 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
1904 udma[/][16,25,33,44,66,100,133] notation is also
1905 allowed.
1907 * [no]ncq: Turn on or off NCQ.
1909 * [no]ncqtrim: Turn off queued DSM TRIM.
1911 * nohrst, nosrst, norst: suppress hard, soft
1912 and both resets.
1914 * rstonce: only attempt one reset during
1915 hot-unplug link recovery
1917 * dump_id: dump IDENTIFY data.
1919 * atapi_dmadir: Enable ATAPI DMADIR bridge support
1921 * disable: Disable this device.
1923 If there are multiple matching configurations changing
1924 the same attribute, the last one is used.
1926 memblock=debug [KNL] Enable memblock debug messages.
1928 load_ramdisk= [RAM] List of ramdisks to load from floppy
1929 See Documentation/blockdev/ramdisk.txt.
1931 lockd.nlm_grace_period=P [NFS] Assign grace period.
1932 Format: <integer>
1934 lockd.nlm_tcpport=N [NFS] Assign TCP port.
1935 Format: <integer>
1937 lockd.nlm_timeout=T [NFS] Assign timeout value.
1938 Format: <integer>
1940 lockd.nlm_udpport=M [NFS] Assign UDP port.
1941 Format: <integer>
1943 locktorture.nreaders_stress= [KNL]
1944 Set the number of locking read-acquisition kthreads.
1945 Defaults to being automatically set based on the
1946 number of online CPUs.
1948 locktorture.nwriters_stress= [KNL]
1949 Set the number of locking write-acquisition kthreads.
1951 locktorture.onoff_holdoff= [KNL]
1952 Set time (s) after boot for CPU-hotplug testing.
1954 locktorture.onoff_interval= [KNL]
1955 Set time (s) between CPU-hotplug operations, or
1956 zero to disable CPU-hotplug testing.
1958 locktorture.shuffle_interval= [KNL]
1959 Set task-shuffle interval (jiffies). Shuffling
1960 tasks allows some CPUs to go into dyntick-idle
1961 mode during the locktorture test.
1963 locktorture.shutdown_secs= [KNL]
1964 Set time (s) after boot system shutdown. This
1965 is useful for hands-off automated testing.
1967 locktorture.stat_interval= [KNL]
1968 Time (s) between statistics printk()s.
1970 locktorture.stutter= [KNL]
1971 Time (s) to stutter testing, for example,
1972 specifying five seconds causes the test to run for
1973 five seconds, wait for five seconds, and so on.
1974 This tests the locking primitive's ability to
1975 transition abruptly to and from idle.
1977 locktorture.torture_runnable= [BOOT]
1978 Start locktorture running at boot time.
1980 locktorture.torture_type= [KNL]
1981 Specify the locking implementation to test.
1983 locktorture.verbose= [KNL]
1984 Enable additional printk() statements.
1986 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
1987 Format: <irq>
1989 loglevel= All Kernel Messages with a loglevel smaller than the
1990 console loglevel will be printed to the console. It can
1991 also be changed with klogd or other programs. The
1992 loglevels are defined as follows:
1994 0 (KERN_EMERG) system is unusable
1995 1 (KERN_ALERT) action must be taken immediately
1996 2 (KERN_CRIT) critical conditions
1997 3 (KERN_ERR) error conditions
1998 4 (KERN_WARNING) warning conditions
1999 5 (KERN_NOTICE) normal but significant condition
2000 6 (KERN_INFO) informational
2001 7 (KERN_DEBUG) debug-level messages
2003 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2004 in bytes. n must be a power of two and greater
2005 than the minimal size. The minimal size is defined
2006 by LOG_BUF_SHIFT kernel config parameter. There is
2007 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2008 that allows to increase the default size depending on
2009 the number of CPUs. See init/Kconfig for more details.
2011 logo.nologo [FB] Disables display of the built-in Linux logo.
2012 This may be used to provide more screen space for
2013 kernel log messages and is useful when debugging
2014 kernel boot problems.
2016 lp=0 [LP] Specify parallel ports to use, e.g,
2017 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2018 lp=reset first parallel port). 'lp=0' disables the
2019 lp=auto printer driver. 'lp=reset' (which can be
2020 specified in addition to the ports) causes
2021 attached printers to be reset. Using
2022 lp=port1,port2,... specifies the parallel ports
2023 to associate lp devices with, starting with
2024 lp0. A port specification may be 'none' to skip
2025 that lp device, or a parport name such as
2026 'parport0'. Specifying 'lp=auto' instead of a
2027 port specification list means that device IDs
2028 from each port should be examined, to see if
2029 an IEEE 1284-compliant printer is attached; if
2030 so, the driver will manage that printer.
2031 See also header of drivers/char/lp.c.
2033 lpj=n [KNL]
2034 Sets loops_per_jiffy to given constant, thus avoiding
2035 time-consuming boot-time autodetection (up to 250 ms per
2036 CPU). 0 enables autodetection (default). To determine
2037 the correct value for your kernel, boot with normal
2038 autodetection and see what value is printed. Note that
2039 on SMP systems the preset will be applied to all CPUs,
2040 which is likely to cause problems if your CPUs need
2041 significantly divergent settings. An incorrect value
2042 will cause delays in the kernel to be wrong, leading to
2043 unpredictable I/O errors and other breakage. Although
2044 unlikely, in the extreme case this might damage your
2045 hardware.
2047 ltpc= [NET]
2048 Format: <io>,<irq>,<dma>
2050 machvec= [IA-64] Force the use of a particular machine-vector
2051 (machvec) in a generic kernel.
2052 Example: machvec=hpzx1_swiotlb
2054 machtype= [Loongson] Share the same kernel image file between different
2055 yeeloong laptop.
2056 Example: machtype=lemote-yeeloong-2f-7inch
2058 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2059 than or equal to this physical address is ignored.
2061 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2062 should make use of. maxcpus=n : n >= 0 limits the
2063 kernel to using 'n' processors. n=0 is a special case,
2064 it is equivalent to "nosmp", which also disables
2065 the IO APIC.
2067 max_loop= [LOOP] The number of loop block devices that get
2068 (loop.max_loop) unconditionally pre-created at init time. The default
2069 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2070 of statically allocating a predefined number, loop
2071 devices can be requested on-demand with the
2072 /dev/loop-control interface.
2074 mce [X86-32] Machine Check Exception
2076 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.txt
2078 md= [HW] RAID subsystems devices and level
2079 See Documentation/md.txt.
2081 mdacon= [MDA]
2082 Format: <first>,<last>
2083 Specifies range of consoles to be captured by the MDA.
2085 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2086 Amount of memory to be used when the kernel is not able
2087 to see the whole system memory or for test.
2088 [X86] Work as limiting max address. Use together
2089 with memmap= to avoid physical address space collisions.
2090 Without memmap= PCI devices could be placed at addresses
2091 belonging to unused RAM.
2093 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2094 memory.
2096 memchunk=nn[KMG]
2097 [KNL,SH] Allow user to override the default size for
2098 per-device physically contiguous DMA buffers.
2100 memmap=exactmap [KNL,X86] Enable setting of an exact
2101 E820 memory map, as specified by the user.
2102 Such memmap=exactmap lines can be constructed based on
2103 BIOS output or other requirements. See the memmap=nn@ss
2104 option description.
2106 memmap=nn[KMG]@ss[KMG]
2107 [KNL] Force usage of a specific region of memory.
2108 Region of memory to be used is from ss to ss+nn.
2110 memmap=nn[KMG]#ss[KMG]
2111 [KNL,ACPI] Mark specific memory as ACPI data.
2112 Region of memory to be marked is from ss to ss+nn.
2114 memmap=nn[KMG]$ss[KMG]
2115 [KNL,ACPI] Mark specific memory as reserved.
2116 Region of memory to be reserved is from ss to ss+nn.
2117 Example: Exclude memory from 0x18690000-0x1869ffff
2118 memmap=64K$0x18690000
2119 or
2120 memmap=0x10000$0x18690000
2122 memmap=nn[KMG]!ss[KMG]
2123 [KNL,X86] Mark specific memory as protected.
2124 Region of memory to be used, from ss to ss+nn.
2125 The memory region may be marked as e820 type 12 (0xc)
2126 and is NVDIMM or ADR memory.
2128 memory_corruption_check=0/1 [X86]
2129 Some BIOSes seem to corrupt the first 64k of
2130 memory when doing things like suspend/resume.
2131 Setting this option will scan the memory
2132 looking for corruption. Enabling this will
2133 both detect corruption and prevent the kernel
2134 from using the memory being corrupted.
2135 However, its intended as a diagnostic tool; if
2136 repeatable BIOS-originated corruption always
2137 affects the same memory, you can use memmap=
2138 to prevent the kernel from using that memory.
2140 memory_corruption_check_size=size [X86]
2141 By default it checks for corruption in the low
2142 64k, making this memory unavailable for normal
2143 use. Use this parameter to scan for
2144 corruption in more or less memory.
2146 memory_corruption_check_period=seconds [X86]
2147 By default it checks for corruption every 60
2148 seconds. Use this parameter to check at some
2149 other rate. 0 disables periodic checking.
2151 memtest= [KNL,X86,ARM] Enable memtest
2152 Format: <integer>
2153 default : 0 <disable>
2154 Specifies the number of memtest passes to be
2155 performed. Each pass selects another test
2156 pattern from a given set of patterns. Memtest
2157 fills the memory with this pattern, validates
2158 memory contents and reserves bad memory
2159 regions that are detected.
2161 meye.*= [HW] Set MotionEye Camera parameters
2162 See Documentation/video4linux/meye.txt.
2164 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2165 Multi-Function General Purpose Timers on AMD Geode
2166 platforms.
2168 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2169 the BIOS has incorrectly applied a workaround. TinyBIOS
2170 version 0.98 is known to be affected, 0.99 fixes the
2171 problem by letting the user disable the workaround.
2173 mga= [HW,DRM]
2175 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2176 physical address is ignored.
2178 mini2440= [ARM,HW,KNL]
2179 Format:[0..2][b][c][t]
2180 Default: "0tb"
2181 MINI2440 configuration specification:
2182 0 - The attached screen is the 3.5" TFT
2183 1 - The attached screen is the 7" TFT
2184 2 - The VGA Shield is attached (1024x768)
2185 Leaving out the screen size parameter will not load
2186 the TFT driver, and the framebuffer will be left
2187 unconfigured.
2188 b - Enable backlight. The TFT backlight pin will be
2189 linked to the kernel VESA blanking code and a GPIO
2190 LED. This parameter is not necessary when using the
2191 VGA shield.
2192 c - Enable the s3c camera interface.
2193 t - Reserved for enabling touchscreen support. The
2194 touchscreen support is not enabled in the mainstream
2195 kernel as of 2.6.30, a preliminary port can be found
2196 in the "bleeding edge" mini2440 support kernel at
2197 http://repo.or.cz/w/linux-2.6/mini2440.git
2199 mminit_loglevel=
2200 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2201 parameter allows control of the logging verbosity for
2202 the additional memory initialisation checks. A value
2203 of 0 disables mminit logging and a level of 4 will
2204 log everything. Information is printed at KERN_DEBUG
2205 so loglevel=8 may also need to be specified.
2207 module.sig_enforce
2208 [KNL] When CONFIG_MODULE_SIG is set, this means that
2209 modules without (valid) signatures will fail to load.
2210 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2211 is always true, so this option does nothing.
2213 mousedev.tap_time=
2214 [MOUSE] Maximum time between finger touching and
2215 leaving touchpad surface for touch to be considered
2216 a tap and be reported as a left button click (for
2217 touchpads working in absolute mode only).
2218 Format: <msecs>
2219 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2220 reporting absolute coordinates, such as tablets
2221 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2222 reporting absolute coordinates, such as tablets
2224 movablecore=nn[KMG] [KNL,X86,IA-64,PPC] This parameter
2225 is similar to kernelcore except it specifies the
2226 amount of memory used for migratable allocations.
2227 If both kernelcore and movablecore is specified,
2228 then kernelcore will be at *least* the specified
2229 value but may be more. If movablecore on its own
2230 is specified, the administrator must be careful
2231 that the amount of memory usable for all allocations
2232 is not too small.
2234 movable_node [KNL,X86] Boot-time switch to enable the effects
2235 of CONFIG_MOVABLE_NODE=y. See mm/Kconfig for details.
2237 MTD_Partition= [MTD]
2238 Format: <name>,<region-number>,<size>,<offset>
2240 MTD_Region= [MTD] Format:
2241 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2243 mtdparts= [MTD]
2244 See drivers/mtd/cmdlinepart.c.
2246 multitce=off [PPC] This parameter disables the use of the pSeries
2247 firmware feature for updating multiple TCE entries
2248 at a time.
2250 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2252 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2254 boundary - index of last SLC block on Flex-OneNAND.
2255 The remaining blocks are configured as MLC blocks.
2256 lock - Configure if Flex-OneNAND boundary should be locked.
2257 Once locked, the boundary cannot be changed.
2258 1 indicates lock status, 0 indicates unlock status.
2260 mtdset= [ARM]
2261 ARM/S3C2412 JIVE boot control
2263 See arch/arm/mach-s3c2412/mach-jive.c
2265 mtouchusb.raw_coordinates=
2266 [HW] Make the MicroTouch USB driver use raw coordinates
2267 ('y', default) or cooked coordinates ('n')
2269 mtrr_chunk_size=nn[KMG] [X86]
2270 used for mtrr cleanup. It is largest continuous chunk
2271 that could hold holes aka. UC entries.
2273 mtrr_gran_size=nn[KMG] [X86]
2274 Used for mtrr cleanup. It is granularity of mtrr block.
2275 Default is 1.
2276 Large value could prevent small alignment from
2277 using up MTRRs.
2279 mtrr_spare_reg_nr=n [X86]
2280 Format: <integer>
2281 Range: 0,7 : spare reg number
2282 Default : 1
2283 Used for mtrr cleanup. It is spare mtrr entries number.
2284 Set to 2 or more if your graphical card needs more.
2286 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2288 netdev= [NET] Network devices parameters
2289 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2290 Note that mem_start is often overloaded to mean
2291 something different and driver-specific.
2292 This usage is only documented in each driver source
2293 file if at all.
2295 nf_conntrack.acct=
2296 [NETFILTER] Enable connection tracking flow accounting
2297 0 to disable accounting
2298 1 to enable accounting
2299 Default value is 0.
2301 nfsaddrs= [NFS] Deprecated. Use ip= instead.
2302 See Documentation/filesystems/nfs/nfsroot.txt.
2304 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
2305 See Documentation/filesystems/nfs/nfsroot.txt.
2307 nfsrootdebug [NFS] enable nfsroot debugging messages.
2308 See Documentation/filesystems/nfs/nfsroot.txt.
2310 nfs.callback_tcpport=
2311 [NFS] set the TCP port on which the NFSv4 callback
2312 channel should listen.
2314 nfs.cache_getent=
2315 [NFS] sets the pathname to the program which is used
2316 to update the NFS client cache entries.
2318 nfs.cache_getent_timeout=
2319 [NFS] sets the timeout after which an attempt to
2320 update a cache entry is deemed to have failed.
2322 nfs.idmap_cache_timeout=
2323 [NFS] set the maximum lifetime for idmapper cache
2324 entries.
2326 nfs.enable_ino64=
2327 [NFS] enable 64-bit inode numbers.
2328 If zero, the NFS client will fake up a 32-bit inode
2329 number for the readdir() and stat() syscalls instead
2330 of returning the full 64-bit number.
2331 The default is to return 64-bit inode numbers.
2333 nfs.max_session_slots=
2334 [NFSv4.1] Sets the maximum number of session slots
2335 the client will attempt to negotiate with the server.
2336 This limits the number of simultaneous RPC requests
2337 that the client can send to the NFSv4.1 server.
2338 Note that there is little point in setting this
2339 value higher than the max_tcp_slot_table_limit.
2341 nfs.nfs4_disable_idmapping=
2342 [NFSv4] When set to the default of '1', this option
2343 ensures that both the RPC level authentication
2344 scheme and the NFS level operations agree to use
2345 numeric uids/gids if the mount is using the
2346 'sec=sys' security flavour. In effect it is
2347 disabling idmapping, which can make migration from
2348 legacy NFSv2/v3 systems to NFSv4 easier.
2349 Servers that do not support this mode of operation
2350 will be autodetected by the client, and it will fall
2351 back to using the idmapper.
2352 To turn off this behaviour, set the value to '0'.
2353 nfs.nfs4_unique_id=
2354 [NFS4] Specify an additional fixed unique ident-
2355 ification string that NFSv4 clients can insert into
2356 their nfs_client_id4 string. This is typically a
2357 UUID that is generated at system install time.
2359 nfs.send_implementation_id =
2360 [NFSv4.1] Send client implementation identification
2361 information in exchange_id requests.
2362 If zero, no implementation identification information
2363 will be sent.
2364 The default is to send the implementation identification
2365 information.
2367 nfs.recover_lost_locks =
2368 [NFSv4] Attempt to recover locks that were lost due
2369 to a lease timeout on the server. Please note that
2370 doing this risks data corruption, since there are
2371 no guarantees that the file will remain unchanged
2372 after the locks are lost.
2373 If you want to enable the kernel legacy behaviour of
2374 attempting to recover these locks, then set this
2375 parameter to '1'.
2376 The default parameter value of '0' causes the kernel
2377 not to attempt recovery of lost locks.
2379 nfs4.layoutstats_timer =
2380 [NFSv4.2] Change the rate at which the kernel sends
2381 layoutstats to the pNFS metadata server.
2383 Setting this to value to 0 causes the kernel to use
2384 whatever value is the default set by the layout
2385 driver. A non-zero value sets the minimum interval
2386 in seconds between layoutstats transmissions.
2388 nfsd.nfs4_disable_idmapping=
2389 [NFSv4] When set to the default of '1', the NFSv4
2390 server will return only numeric uids and gids to
2391 clients using auth_sys, and will accept numeric uids
2392 and gids from such clients. This is intended to ease
2393 migration from NFSv2/v3.
2395 objlayoutdriver.osd_login_prog=
2396 [NFS] [OBJLAYOUT] sets the pathname to the program which
2397 is used to automatically discover and login into new
2398 osd-targets. Please see:
2399 Documentation/filesystems/pnfs.txt for more explanations
2401 nmi_debug= [KNL,AVR32,SH] Specify one or more actions to take
2402 when a NMI is triggered.
2403 Format: [state][,regs][,debounce][,die]
2405 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
2406 Format: [panic,][nopanic,][num]
2407 Valid num: 0 or 1
2408 0 - turn hardlockup detector in nmi_watchdog off
2409 1 - turn hardlockup detector in nmi_watchdog on
2410 When panic is specified, panic when an NMI watchdog
2411 timeout occurs (or 'nopanic' to override the opposite
2412 default). To disable both hard and soft lockup detectors,
2413 please see 'nowatchdog'.
2414 This is useful when you use a panic=... timeout and
2415 need the box quickly up again.
2417 netpoll.carrier_timeout=
2418 [NET] Specifies amount of time (in seconds) that
2419 netpoll should wait for a carrier. By default netpoll
2420 waits 4 seconds.
2422 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
2423 emulation library even if a 387 maths coprocessor
2424 is present.
2426 no_console_suspend
2427 [HW] Never suspend the console
2428 Disable suspending of consoles during suspend and
2429 hibernate operations. Once disabled, debugging
2430 messages can reach various consoles while the rest
2431 of the system is being put to sleep (ie, while
2432 debugging driver suspend/resume hooks). This may
2433 not work reliably with all consoles, but is known
2434 to work with serial and VGA consoles.
2435 To facilitate more flexible debugging, we also add
2436 console_suspend, a printk module parameter to control
2437 it. Users could use console_suspend (usually
2438 /sys/module/printk/parameters/console_suspend) to
2439 turn on/off it dynamically.
2441 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
2442 caches in the slab allocator. Saves per-node memory,
2443 but will impact performance.
2445 noalign [KNL,ARM]
2447 noapic [SMP,APIC] Tells the kernel to not make use of any
2448 IOAPICs that may be present in the system.
2450 noautogroup Disable scheduler automatic task group creation.
2452 nobats [PPC] Do not use BATs for mapping kernel lowmem
2453 on "Classic" PPC cores.
2455 nocache [ARM]
2457 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
2459 nodelayacct [KNL] Disable per-task delay accounting
2461 nodisconnect [HW,SCSI,M68K] Disables SCSI disconnects.
2463 nodsp [SH] Disable hardware DSP at boot time.
2465 noefi Disable EFI runtime services support.
2467 noexec [IA-64]
2469 noexec [X86]
2470 On X86-32 available only on PAE configured kernels.
2471 noexec=on: enable non-executable mappings (default)
2472 noexec=off: disable non-executable mappings
2474 nosmap [X86]
2475 Disable SMAP (Supervisor Mode Access Prevention)
2476 even if it is supported by processor.
2478 nosmep [X86]
2479 Disable SMEP (Supervisor Mode Execution Prevention)
2480 even if it is supported by processor.
2482 noexec32 [X86-64]
2483 This affects only 32-bit executables.
2484 noexec32=on: enable non-executable mappings (default)
2485 read doesn't imply executable mappings
2486 noexec32=off: disable non-executable mappings
2487 read implies executable mappings
2489 nofpu [MIPS,SH] Disable hardware FPU at boot time.
2491 nofxsr [BUGS=X86-32] Disables x86 floating point extended
2492 register save and restore. The kernel will only save
2493 legacy floating-point registers on task switch.
2495 nohugeiomap [KNL,x86] Disable kernel huge I/O mappings.
2497 noxsave [BUGS=X86] Disables x86 extended register state save
2498 and restore using xsave. The kernel will fallback to
2499 enabling legacy floating-point and sse state.
2501 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
2502 register states. The kernel will fall back to use
2503 xsave to save the states. By using this parameter,
2504 performance of saving the states is degraded because
2505 xsave doesn't support modified optimization while
2506 xsaveopt supports it on xsaveopt enabled systems.
2508 noxsaves [X86] Disables xsaves and xrstors used in saving and
2509 restoring x86 extended register state in compacted
2510 form of xsave area. The kernel will fall back to use
2511 xsaveopt and xrstor to save and restore the states
2512 in standard form of xsave area. By using this
2513 parameter, xsave area per process might occupy more
2514 memory on xsaves enabled systems.
2516 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
2517 wfi(ARM) instruction doesn't work correctly and not to
2518 use it. This is also useful when using JTAG debugger.
2520 no_file_caps Tells the kernel not to honor file capabilities. The
2521 only way then for a file to be executed with privilege
2522 is to be setuid root or executed by root.
2524 nohalt [IA-64] Tells the kernel not to use the power saving
2525 function PAL_HALT_LIGHT when idle. This increases
2526 power-consumption. On the positive side, it reduces
2527 interrupt wake-up latency, which may improve performance
2528 in certain environments such as networked servers or
2529 real-time systems.
2531 nohibernate [HIBERNATION] Disable hibernation and resume.
2533 nohz= [KNL] Boottime enable/disable dynamic ticks
2534 Valid arguments: on, off
2535 Default: on
2537 nohz_full= [KNL,BOOT]
2538 In kernels built with CONFIG_NO_HZ_FULL=y, set
2539 the specified list of CPUs whose tick will be stopped
2540 whenever possible. The boot CPU will be forced outside
2541 the range to maintain the timekeeping.
2542 The CPUs in this range must also be included in the
2543 rcu_nocbs= set.
2545 noiotrap [SH] Disables trapped I/O port accesses.
2547 noirqdebug [X86-32] Disables the code which attempts to detect and
2548 disable unhandled interrupt sources.
2550 no_timer_check [X86,APIC] Disables the code which tests for
2551 broken timer IRQ sources.
2553 noisapnp [ISAPNP] Disables ISA PnP code.
2555 noinitrd [RAM] Tells the kernel not to load any configured
2556 initial RAM disk.
2558 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
2559 remapping.
2560 [Deprecated - use intremap=off]
2562 nointroute [IA-64]
2564 nojitter [IA-64] Disables jitter checking for ITC timers.
2566 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
2568 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
2569 fault handling.
2571 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting.
2572 steal time is computed, but won't influence scheduler
2573 behaviour
2575 nolapic [X86-32,APIC] Do not enable or use the local APIC.
2577 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
2579 noltlbs [PPC] Do not use large page/tlb entries for kernel
2580 lowmem mapping on PPC40x.
2582 nomca [IA-64] Disable machine check abort handling
2584 nomce [X86-32] Disable Machine Check Exception
2586 nomfgpt [X86-32] Disable Multi-Function General Purpose
2587 Timer usage (for AMD Geode machines).
2589 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
2590 shutdown the other cpus. Instead use the REBOOT_VECTOR
2591 irq.
2593 nomodule Disable module load
2595 nopat [X86] Disable PAT (page attribute table extension of
2596 pagetables) support.
2598 norandmaps Don't use address space randomization. Equivalent to
2599 echo 0 > /proc/sys/kernel/randomize_va_space
2601 noreplace-paravirt [X86,IA-64,PV_OPS] Don't patch paravirt_ops
2603 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
2604 with UP alternatives
2606 nordrand [X86] Disable kernel use of the RDRAND and
2607 RDSEED instructions even if they are supported
2608 by the processor. RDRAND and RDSEED are still
2609 available to user space applications.
2611 noresume [SWSUSP] Disables resume and restores original swap
2612 space.
2614 no-scroll [VGA] Disables scrollback.
2615 This is required for the Braillex ib80-piezo Braille
2616 reader made by F.H. Papenmeier (Germany).
2618 nosbagart [IA-64]
2620 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
2622 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
2623 and disable the IO APIC. legacy for "maxcpus=0".
2625 nosoftlockup [KNL] Disable the soft-lockup detector.
2627 nosync [HW,M68K] Disables sync negotiation for all devices.
2629 notsc [BUGS=X86-32] Disable Time Stamp Counter
2631 nowatchdog [KNL] Disable both lockup detectors, i.e.
2632 soft-lockup and NMI watchdog (hard-lockup).
2634 nowb [ARM]
2636 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
2638 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
2639 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
2640 Some features depend on CPU0. Known dependencies are:
2641 1. Resume from suspend/hibernate depends on CPU0.
2642 Suspend/hibernate will fail if CPU0 is offline and you
2643 need to online CPU0 before suspend/hibernate.
2644 2. PIC interrupts also depend on CPU0. CPU0 can't be
2645 removed if a PIC interrupt is detected.
2646 It's said poweroff/reboot may depend on CPU0 on some
2647 machines although I haven't seen such issues so far
2648 after CPU0 is offline on a few tested machines.
2649 If the dependencies are under your control, you can
2650 turn on cpu0_hotplug.
2652 nptcg= [IA-64] Override max number of concurrent global TLB
2653 purges which is reported from either PAL_VM_SUMMARY or
2654 SAL PALO.
2656 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
2657 could support. nr_cpus=n : n >= 1 limits the kernel to
2658 supporting 'n' processors. Later in runtime you can not
2659 use hotplug cpu feature to put more cpu back to online.
2660 just like you compile the kernel NR_CPUS=n
2662 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
2664 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
2665 Allowed values are enable and disable
2667 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
2668 one of ['zone', 'node', 'default'] can be specified
2669 This can be set from sysctl after boot.
2670 See Documentation/sysctl/vm.txt for details.
2672 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
2673 See Documentation/debugging-via-ohci1394.txt for more
2674 info.
2676 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
2677 Rather than timing out after 20 ms if an EC
2678 command is not properly ACKed, override the length
2679 of the timeout. We have interrupts disabled while
2680 waiting for the ACK, so if this is set too high
2681 interrupts *may* be lost!
2683 omap_mux= [OMAP] Override bootloader pin multiplexing.
2684 Format: <mux_mode0.mode_name=value>...
2685 For example, to override I2C bus2:
2686 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
2688 oprofile.timer= [HW]
2689 Use timer interrupt instead of performance counters
2691 oprofile.cpu_type= Force an oprofile cpu type
2692 This might be useful if you have an older oprofile
2693 userland or if you want common events.
2694 Format: { arch_perfmon }
2695 arch_perfmon: [X86] Force use of architectural
2696 perfmon on Intel CPUs instead of the
2697 CPU specific event set.
2698 timer: [X86] Force use of architectural NMI
2699 timer mode (see also oprofile.timer
2700 for generic hr timer mode)
2701 [s390] Force legacy basic mode sampling
2702 (report cpu_type "timer")
2704 oops=panic Always panic on oopses. Default is to just kill the
2705 process, but there is a small probability of
2706 deadlocking the machine.
2707 This will also cause panics on machine check exceptions.
2708 Useful together with panic=30 to trigger a reboot.
2710 OSS [HW,OSS]
2711 See Documentation/sound/oss/oss-parameters.txt
2713 page_owner= [KNL] Boot-time page_owner enabling option.
2714 Storage of the information about who allocated
2715 each page is disabled in default. With this switch,
2716 we can turn it on.
2717 on: enable the feature
2719 panic= [KNL] Kernel behaviour on panic: delay <timeout>
2720 timeout > 0: seconds before rebooting
2721 timeout = 0: wait forever
2722 timeout < 0: reboot immediately
2723 Format: <timeout>
2725 panic_on_warn panic() instead of WARN(). Useful to cause kdump
2726 on a WARN().
2728 crash_kexec_post_notifiers
2729 Run kdump after running panic-notifiers and dumping
2730 kmsg. This only for the users who doubt kdump always
2731 succeeds in any situation.
2732 Note that this also increases risks of kdump failure,
2733 because some panic notifiers can make the crashed
2734 kernel more unstable.
2736 parkbd.port= [HW] Parallel port number the keyboard adapter is
2737 connected to, default is 0.
2738 Format: <parport#>
2739 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
2740 0 for XT, 1 for AT (default is AT).
2741 Format: <mode>
2743 parport= [HW,PPT] Specify parallel ports. 0 disables.
2744 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
2745 Use 'auto' to force the driver to use any
2746 IRQ/DMA settings detected (the default is to
2747 ignore detected IRQ/DMA settings because of
2748 possible conflicts). You can specify the base
2749 address, IRQ, and DMA settings; IRQ and DMA
2750 should be numbers, or 'auto' (for using detected
2751 settings on that particular port), or 'nofifo'
2752 (to avoid using a FIFO even if it is detected).
2753 Parallel ports are assigned in the order they
2754 are specified on the command line, starting
2755 with parport0.
2757 parport_init_mode= [HW,PPT]
2758 Configure VIA parallel port to operate in
2759 a specific mode. This is necessary on Pegasos
2760 computer where firmware has no options for setting
2761 up parallel port mode and sets it to spp.
2762 Currently this function knows 686a and 8231 chips.
2763 Format: [spp|ps2|epp|ecp|ecpepp]
2765 pause_on_oops=
2766 Halt all CPUs after the first oops has been printed for
2767 the specified number of seconds. This is to be used if
2768 your oopses keep scrolling off the screen.
2770 pcbit= [HW,ISDN]
2772 pcd. [PARIDE]
2773 See header of drivers/block/paride/pcd.c.
2774 See also Documentation/blockdev/paride.txt.
2776 pci=option[,option...] [PCI] various PCI subsystem options:
2777 earlydump [X86] dump PCI config space before the kernel
2778 changes anything
2779 off [X86] don't probe for the PCI bus
2780 bios [X86-32] force use of PCI BIOS, don't access
2781 the hardware directly. Use this if your machine
2782 has a non-standard PCI host bridge.
2783 nobios [X86-32] disallow use of PCI BIOS, only direct
2784 hardware access methods are allowed. Use this
2785 if you experience crashes upon bootup and you
2786 suspect they are caused by the BIOS.
2787 conf1 [X86] Force use of PCI Configuration Access
2788 Mechanism 1 (config address in IO port 0xCF8,
2789 data in IO port 0xCFC, both 32-bit).
2790 conf2 [X86] Force use of PCI Configuration Access
2791 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
2792 the function, IO port 0xCFA, also 8-bit, sets
2793 bus number. The config space is then accessed
2794 through ports 0xC000-0xCFFF).
2795 See http://wiki.osdev.org/PCI for more info
2796 on the configuration access mechanisms.
2797 noaer [PCIE] If the PCIEAER kernel config parameter is
2798 enabled, this kernel boot option can be used to
2799 disable the use of PCIE advanced error reporting.
2800 nodomains [PCI] Disable support for multiple PCI
2801 root domains (aka PCI segments, in ACPI-speak).
2802 nommconf [X86] Disable use of MMCONFIG for PCI
2803 Configuration
2804 check_enable_amd_mmconf [X86] check for and enable
2805 properly configured MMIO access to PCI
2806 config space on AMD family 10h CPU
2807 nomsi [MSI] If the PCI_MSI kernel config parameter is
2808 enabled, this kernel boot option can be used to
2809 disable the use of MSI interrupts system-wide.
2810 noioapicquirk [APIC] Disable all boot interrupt quirks.
2811 Safety option to keep boot IRQs enabled. This
2812 should never be necessary.
2813 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
2814 primary IO-APIC for bridges that cannot disable
2815 boot IRQs. This fixes a source of spurious IRQs
2816 when the system masks IRQs.
2817 noioapicreroute [APIC] Disable workaround that uses the
2818 boot IRQ equivalent of an IRQ that connects to
2819 a chipset where boot IRQs cannot be disabled.
2820 The opposite of ioapicreroute.
2821 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
2822 routing table. These calls are known to be buggy
2823 on several machines and they hang the machine
2824 when used, but on other computers it's the only
2825 way to get the interrupt routing table. Try
2826 this option if the kernel is unable to allocate
2827 IRQs or discover secondary PCI buses on your
2828 motherboard.
2829 rom [X86] Assign address space to expansion ROMs.
2830 Use with caution as certain devices share
2831 address decoders between ROMs and other
2832 resources.
2833 norom [X86] Do not assign address space to
2834 expansion ROMs that do not already have
2835 BIOS assigned address ranges.
2836 nobar [X86] Do not assign address space to the
2837 BARs that weren't assigned by the BIOS.
2838 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
2839 assigned automatically to PCI devices. You can
2840 make the kernel exclude IRQs of your ISA cards
2841 this way.
2842 pirqaddr=0xAAAAA [X86] Specify the physical address
2843 of the PIRQ table (normally generated
2844 by the BIOS) if it is outside the
2845 F0000h-100000h range.
2846 lastbus=N [X86] Scan all buses thru bus #N. Can be
2847 useful if the kernel is unable to find your
2848 secondary buses and you want to tell it
2849 explicitly which ones they are.
2850 assign-busses [X86] Always assign all PCI bus
2851 numbers ourselves, overriding
2852 whatever the firmware may have done.
2853 usepirqmask [X86] Honor the possible IRQ mask stored
2854 in the BIOS $PIR table. This is needed on
2855 some systems with broken BIOSes, notably
2856 some HP Pavilion N5400 and Omnibook XE3
2857 notebooks. This will have no effect if ACPI
2858 IRQ routing is enabled.
2859 noacpi [X86] Do not use ACPI for IRQ routing
2860 or for PCI scanning.
2861 use_crs [X86] Use PCI host bridge window information
2862 from ACPI. On BIOSes from 2008 or later, this
2863 is enabled by default. If you need to use this,
2864 please report a bug.
2865 nocrs [X86] Ignore PCI host bridge windows from ACPI.
2866 If you need to use this, please report a bug.
2867 routeirq Do IRQ routing for all PCI devices.
2868 This is normally done in pci_enable_device(),
2869 so this option is a temporary workaround
2870 for broken drivers that don't call it.
2871 skip_isa_align [X86] do not align io start addr, so can
2872 handle more pci cards
2873 firmware [ARM] Do not re-enumerate the bus but instead
2874 just use the configuration from the
2875 bootloader. This is currently used on
2876 IXP2000 systems where the bus has to be
2877 configured a certain way for adjunct CPUs.
2878 noearly [X86] Don't do any early type 1 scanning.
2879 This might help on some broken boards which
2880 machine check when some devices' config space
2881 is read. But various workarounds are disabled
2882 and some IOMMU drivers will not work.
2883 bfsort Sort PCI devices into breadth-first order.
2884 This sorting is done to get a device
2885 order compatible with older (<= 2.4) kernels.
2886 nobfsort Don't sort PCI devices into breadth-first order.
2887 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
2888 tuning and use the BIOS-configured MPS defaults.
2889 pcie_bus_safe Set every device's MPS to the largest value
2890 supported by all devices below the root complex.
2891 pcie_bus_perf Set device MPS to the largest allowable MPS
2892 based on its parent bus. Also set MRRS (Max
2893 Read Request Size) to the largest supported
2894 value (no larger than the MPS that the device
2895 or bus can support) for best performance.
2896 pcie_bus_peer2peer Set every device's MPS to 128B, which
2897 every device is guaranteed to support. This
2898 configuration allows peer-to-peer DMA between
2899 any pair of devices, possibly at the cost of
2900 reduced performance. This also guarantees
2901 that hot-added devices will work.
2902 cbiosize=nn[KMG] The fixed amount of bus space which is
2903 reserved for the CardBus bridge's IO window.
2904 The default value is 256 bytes.
2905 cbmemsize=nn[KMG] The fixed amount of bus space which is
2906 reserved for the CardBus bridge's memory
2907 window. The default value is 64 megabytes.
2908 resource_alignment=
2909 Format:
2910 [<order of align>@][<domain>:]<bus>:<slot>.<func>[; ...]
2911 Specifies alignment and device to reassign
2912 aligned memory resources.
2913 If <order of align> is not specified,
2914 PAGE_SIZE is used as alignment.
2915 PCI-PCI bridge can be specified, if resource
2916 windows need to be expanded.
2917 ecrc= Enable/disable PCIe ECRC (transaction layer
2918 end-to-end CRC checking).
2919 bios: Use BIOS/firmware settings. This is the
2920 the default.
2921 off: Turn ECRC off
2922 on: Turn ECRC on.
2923 hpiosize=nn[KMG] The fixed amount of bus space which is
2924 reserved for hotplug bridge's IO window.
2925 Default size is 256 bytes.
2926 hpmemsize=nn[KMG] The fixed amount of bus space which is
2927 reserved for hotplug bridge's memory window.
2928 Default size is 2 megabytes.
2929 realloc= Enable/disable reallocating PCI bridge resources
2930 if allocations done by BIOS are too small to
2931 accommodate resources required by all child
2932 devices.
2933 off: Turn realloc off
2934 on: Turn realloc on
2935 realloc same as realloc=on
2936 noari do not use PCIe ARI.
2937 pcie_scan_all Scan all possible PCIe devices. Otherwise we
2938 only look for one device below a PCIe downstream
2939 port.
2941 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
2942 Management.
2943 off Disable ASPM.
2944 force Enable ASPM even on devices that claim not to support it.
2945 WARNING: Forcing ASPM on may cause system lockups.
2947 pcie_hp= [PCIE] PCI Express Hotplug driver options:
2948 nomsi Do not use MSI for PCI Express Native Hotplug (this
2949 makes all PCIe ports use INTx for hotplug services).
2951 pcie_ports= [PCIE] PCIe ports handling:
2952 auto Ask the BIOS whether or not to use native PCIe services
2953 associated with PCIe ports (PME, hot-plug, AER). Use
2954 them only if that is allowed by the BIOS.
2955 native Use native PCIe services associated with PCIe ports
2956 unconditionally.
2957 compat Treat PCIe ports as PCI-to-PCI bridges, disable the PCIe
2958 ports driver.
2960 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
2961 nomsi Do not use MSI for native PCIe PME signaling (this makes
2962 all PCIe root ports use INTx for all services).
2964 pcmv= [HW,PCMCIA] BadgePAD 4
2966 pd_ignore_unused
2967 [PM]
2968 Keep all power-domains already enabled by bootloader on,
2969 even if no driver has claimed them. This is useful
2970 for debug and development, but should not be
2971 needed on a platform with proper driver support.
2973 pd. [PARIDE]
2974 See Documentation/blockdev/paride.txt.
2976 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
2977 boot time.
2978 Format: { 0 | 1 }
2979 See arch/parisc/kernel/pdc_chassis.c
2981 percpu_alloc= Select which percpu first chunk allocator to use.
2982 Currently supported values are "embed" and "page".
2983 Archs may support subset or none of the selections.
2984 See comments in mm/percpu.c for details on each
2985 allocator. This parameter is primarily for debugging
2986 and performance comparison.
2988 pf. [PARIDE]
2989 See Documentation/blockdev/paride.txt.
2991 pg. [PARIDE]
2992 See Documentation/blockdev/paride.txt.
2994 pirq= [SMP,APIC] Manual mp-table setup
2995 See Documentation/x86/i386/IO-APIC.txt.
2997 plip= [PPT,NET] Parallel port network link
2998 Format: { parport<nr> | timid | 0 }
2999 See also Documentation/parport.txt.
3001 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3002 Override pmtimer IOPort with a hex value.
3003 e.g. pmtmr=0x508
3005 pnp.debug=1 [PNP]
3006 Enable PNP debug messages (depends on the
3007 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3008 via /sys/module/pnp/parameters/debug. We always show
3009 current resource usage; turning this on also shows
3010 possible settings and some assignment information.
3012 pnpacpi= [ACPI]
3013 { off }
3015 pnpbios= [ISAPNP]
3016 { on | off | curr | res | no-curr | no-res }
3018 pnp_reserve_irq=
3019 [ISAPNP] Exclude IRQs for the autoconfiguration
3021 pnp_reserve_dma=
3022 [ISAPNP] Exclude DMAs for the autoconfiguration
3024 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3025 Ranges are in pairs (I/O port base and size).
3027 pnp_reserve_mem=
3028 [ISAPNP] Exclude memory regions for the
3029 autoconfiguration.
3030 Ranges are in pairs (memory base and size).
3032 ports= [IP_VS_FTP] IPVS ftp helper module
3033 Default is 21.
3034 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3035 may be specified.
3036 Format: <port>,<port>....
3038 ppc_strict_facility_enable
3039 [PPC] This option catches any kernel floating point,
3040 Altivec, VSX and SPE outside of regions specifically
3041 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3042 There is some performance impact when enabling this.
3044 print-fatal-signals=
3045 [KNL] debug: print fatal signals
3047 If enabled, warn about various signal handling
3048 related application anomalies: too many signals,
3049 too many POSIX.1 timers, fatal signals causing a
3050 coredump - etc.
3052 If you hit the warning due to signal overflow,
3053 you might want to try "ulimit -i unlimited".
3055 default: off.
3057 printk.always_kmsg_dump=
3058 Trigger kmsg_dump for cases other than kernel oops or
3059 panics
3060 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3061 default: disabled
3063 printk.time= Show timing data prefixed to each printk message line
3064 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3066 processor.max_cstate= [HW,ACPI]
3067 Limit processor to maximum C-state
3068 max_cstate=9 overrides any DMI blacklist limit.
3070 processor.nocst [HW,ACPI]
3071 Ignore the _CST method to determine C-states,
3072 instead using the legacy FADT method
3074 profile= [KNL] Enable kernel profiling via /proc/profile
3075 Format: [schedule,]<number>
3076 Param: "schedule" - profile schedule points.
3077 Param: <number> - step/bucket size as a power of 2 for
3078 statistical time based profiling.
3079 Param: "sleep" - profile D-state sleeping (millisecs).
3080 Requires CONFIG_SCHEDSTATS
3081 Param: "kvm" - profile VM exits.
3083 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3084 before loading.
3085 See Documentation/blockdev/ramdisk.txt.
3087 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3088 probe for; one of (bare|imps|exps|lifebook|any).
3089 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3090 per second.
3091 psmouse.resetafter= [HW,MOUSE]
3092 Try to reset the device after so many bad packets
3093 (0 = never).
3094 psmouse.resolution=
3095 [HW,MOUSE] Set desired mouse resolution, in dpi.
3096 psmouse.smartscroll=
3097 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
3098 0 = disabled, 1 = enabled (default).
3100 pstore.backend= Specify the name of the pstore backend to use
3102 pt. [PARIDE]
3103 See Documentation/blockdev/paride.txt.
3105 pty.legacy_count=
3106 [KNL] Number of legacy pty's. Overwrites compiled-in
3107 default number.
3109 quiet [KNL] Disable most log messages
3111 r128= [HW,DRM]
3113 raid= [HW,RAID]
3114 See Documentation/md.txt.
3116 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
3117 See Documentation/blockdev/ramdisk.txt.
3119 rcu_nocbs= [KNL]
3120 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
3121 the specified list of CPUs to be no-callback CPUs.
3122 Invocation of these CPUs' RCU callbacks will
3123 be offloaded to "rcuox/N" kthreads created for
3124 that purpose, where "x" is "b" for RCU-bh, "p"
3125 for RCU-preempt, and "s" for RCU-sched, and "N"
3126 is the CPU number. This reduces OS jitter on the
3127 offloaded CPUs, which can be useful for HPC and
3128 real-time workloads. It can also improve energy
3129 efficiency for asymmetric multiprocessors.
3131 rcu_nocb_poll [KNL]
3132 Rather than requiring that offloaded CPUs
3133 (specified by rcu_nocbs= above) explicitly
3134 awaken the corresponding "rcuoN" kthreads,
3135 make these kthreads poll for callbacks.
3136 This improves the real-time response for the
3137 offloaded CPUs by relieving them of the need to
3138 wake up the corresponding kthread, but degrades
3139 energy efficiency by requiring that the kthreads
3140 periodically wake up to do the polling.
3142 rcutree.blimit= [KNL]
3143 Set maximum number of finished RCU callbacks to
3144 process in one batch.
3146 rcutree.dump_tree= [KNL]
3147 Dump the structure of the rcu_node combining tree
3148 out at early boot. This is used for diagnostic
3149 purposes, to verify correct tree setup.
3151 rcutree.gp_cleanup_delay= [KNL]
3152 Set the number of jiffies to delay each step of
3153 RCU grace-period cleanup. This only has effect
3154 when CONFIG_RCU_TORTURE_TEST_SLOW_CLEANUP is set.
3156 rcutree.gp_init_delay= [KNL]
3157 Set the number of jiffies to delay each step of
3158 RCU grace-period initialization. This only has
3159 effect when CONFIG_RCU_TORTURE_TEST_SLOW_INIT
3160 is set.
3162 rcutree.gp_preinit_delay= [KNL]
3163 Set the number of jiffies to delay each step of
3164 RCU grace-period pre-initialization, that is,
3165 the propagation of recent CPU-hotplug changes up
3166 the rcu_node combining tree. This only has effect
3167 when CONFIG_RCU_TORTURE_TEST_SLOW_PREINIT is set.
3169 rcutree.rcu_fanout_exact= [KNL]
3170 Disable autobalancing of the rcu_node combining
3171 tree. This is used by rcutorture, and might
3172 possibly be useful for architectures having high
3173 cache-to-cache transfer latencies.
3175 rcutree.rcu_fanout_leaf= [KNL]
3176 Change the number of CPUs assigned to each
3177 leaf rcu_node structure. Useful for very
3178 large systems, which will choose the value 64,
3179 and for NUMA systems with large remote-access
3180 latencies, which will choose a value aligned
3181 with the appropriate hardware boundaries.
3183 rcutree.jiffies_till_sched_qs= [KNL]
3184 Set required age in jiffies for a
3185 given grace period before RCU starts
3186 soliciting quiescent-state help from
3187 rcu_note_context_switch().
3189 rcutree.jiffies_till_first_fqs= [KNL]
3190 Set delay from grace-period initialization to
3191 first attempt to force quiescent states.
3192 Units are jiffies, minimum value is zero,
3193 and maximum value is HZ.
3195 rcutree.jiffies_till_next_fqs= [KNL]
3196 Set delay between subsequent attempts to force
3197 quiescent states. Units are jiffies, minimum
3198 value is one, and maximum value is HZ.
3200 rcutree.kthread_prio= [KNL,BOOT]
3201 Set the SCHED_FIFO priority of the RCU per-CPU
3202 kthreads (rcuc/N). This value is also used for
3203 the priority of the RCU boost threads (rcub/N)
3204 and for the RCU grace-period kthreads (rcu_bh,
3205 rcu_preempt, and rcu_sched). If RCU_BOOST is
3206 set, valid values are 1-99 and the default is 1
3207 (the least-favored priority). Otherwise, when
3208 RCU_BOOST is not set, valid values are 0-99 and
3209 the default is zero (non-realtime operation).
3211 rcutree.rcu_nocb_leader_stride= [KNL]
3212 Set the number of NOCB kthread groups, which
3213 defaults to the square root of the number of
3214 CPUs. Larger numbers reduces the wakeup overhead
3215 on the per-CPU grace-period kthreads, but increases
3216 that same overhead on each group's leader.
3218 rcutree.qhimark= [KNL]
3219 Set threshold of queued RCU callbacks beyond which
3220 batch limiting is disabled.
3222 rcutree.qlowmark= [KNL]
3223 Set threshold of queued RCU callbacks below which
3224 batch limiting is re-enabled.
3226 rcutree.rcu_idle_gp_delay= [KNL]
3227 Set wakeup interval for idle CPUs that have
3228 RCU callbacks (RCU_FAST_NO_HZ=y).
3230 rcutree.rcu_idle_lazy_gp_delay= [KNL]
3231 Set wakeup interval for idle CPUs that have
3232 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
3233 Lazy RCU callbacks are those which RCU can
3234 prove do nothing more than free memory.
3236 rcutorture.cbflood_inter_holdoff= [KNL]
3237 Set holdoff time (jiffies) between successive
3238 callback-flood tests.
3240 rcutorture.cbflood_intra_holdoff= [KNL]
3241 Set holdoff time (jiffies) between successive
3242 bursts of callbacks within a given callback-flood
3243 test.
3245 rcutorture.cbflood_n_burst= [KNL]
3246 Set the number of bursts making up a given
3247 callback-flood test. Set this to zero to
3248 disable callback-flood testing.
3250 rcutorture.cbflood_n_per_burst= [KNL]
3251 Set the number of callbacks to be registered
3252 in a given burst of a callback-flood test.
3254 rcutorture.fqs_duration= [KNL]
3255 Set duration of force_quiescent_state bursts
3256 in microseconds.
3258 rcutorture.fqs_holdoff= [KNL]
3259 Set holdoff time within force_quiescent_state bursts
3260 in microseconds.
3262 rcutorture.fqs_stutter= [KNL]
3263 Set wait time between force_quiescent_state bursts
3264 in seconds.
3266 rcutorture.gp_cond= [KNL]
3267 Use conditional/asynchronous update-side
3268 primitives, if available.
3270 rcutorture.gp_exp= [KNL]
3271 Use expedited update-side primitives, if available.
3273 rcutorture.gp_normal= [KNL]
3274 Use normal (non-expedited) asynchronous
3275 update-side primitives, if available.
3277 rcutorture.gp_sync= [KNL]
3278 Use normal (non-expedited) synchronous
3279 update-side primitives, if available. If all
3280 of rcutorture.gp_cond=, rcutorture.gp_exp=,
3281 rcutorture.gp_normal=, and rcutorture.gp_sync=
3282 are zero, rcutorture acts as if is interpreted
3283 they are all non-zero.
3285 rcutorture.n_barrier_cbs= [KNL]
3286 Set callbacks/threads for rcu_barrier() testing.
3288 rcutorture.nfakewriters= [KNL]
3289 Set number of concurrent RCU writers. These just
3290 stress RCU, they don't participate in the actual
3291 test, hence the "fake".
3293 rcutorture.nreaders= [KNL]
3294 Set number of RCU readers. The value -1 selects
3295 N-1, where N is the number of CPUs. A value
3296 "n" less than -1 selects N-n-2, where N is again
3297 the number of CPUs. For example, -2 selects N
3298 (the number of CPUs), -3 selects N+1, and so on.
3300 rcutorture.object_debug= [KNL]
3301 Enable debug-object double-call_rcu() testing.
3303 rcutorture.onoff_holdoff= [KNL]
3304 Set time (s) after boot for CPU-hotplug testing.
3306 rcutorture.onoff_interval= [KNL]
3307 Set time (s) between CPU-hotplug operations, or
3308 zero to disable CPU-hotplug testing.
3310 rcutorture.shuffle_interval= [KNL]
3311 Set task-shuffle interval (s). Shuffling tasks
3312 allows some CPUs to go into dyntick-idle mode
3313 during the rcutorture test.
3315 rcutorture.shutdown_secs= [KNL]
3316 Set time (s) after boot system shutdown. This
3317 is useful for hands-off automated testing.
3319 rcutorture.stall_cpu= [KNL]
3320 Duration of CPU stall (s) to test RCU CPU stall
3321 warnings, zero to disable.
3323 rcutorture.stall_cpu_holdoff= [KNL]
3324 Time to wait (s) after boot before inducing stall.
3326 rcutorture.stat_interval= [KNL]
3327 Time (s) between statistics printk()s.
3329 rcutorture.stutter= [KNL]
3330 Time (s) to stutter testing, for example, specifying
3331 five seconds causes the test to run for five seconds,
3332 wait for five seconds, and so on. This tests RCU's
3333 ability to transition abruptly to and from idle.
3335 rcutorture.test_boost= [KNL]
3336 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
3337 "Maybe" means test if the RCU implementation
3338 under test support RCU priority boosting.
3340 rcutorture.test_boost_duration= [KNL]
3341 Duration (s) of each individual boost test.
3343 rcutorture.test_boost_interval= [KNL]
3344 Interval (s) between each boost test.
3346 rcutorture.test_no_idle_hz= [KNL]
3347 Test RCU's dyntick-idle handling. See also the
3348 rcutorture.shuffle_interval parameter.
3350 rcutorture.torture_runnable= [BOOT]
3351 Start rcutorture running at boot time.
3353 rcutorture.torture_type= [KNL]
3354 Specify the RCU implementation to test.
3356 rcutorture.verbose= [KNL]
3357 Enable additional printk() statements.
3359 rcupdate.rcu_cpu_stall_suppress= [KNL]
3360 Suppress RCU CPU stall warning messages.
3362 rcupdate.rcu_cpu_stall_timeout= [KNL]
3363 Set timeout for RCU CPU stall warning messages.
3365 rcupdate.rcu_expedited= [KNL]
3366 Use expedited grace-period primitives, for
3367 example, synchronize_rcu_expedited() instead
3368 of synchronize_rcu(). This reduces latency,
3369 but can increase CPU utilization, degrade
3370 real-time latency, and degrade energy efficiency.
3371 No effect on CONFIG_TINY_RCU kernels.
3373 rcupdate.rcu_normal= [KNL]
3374 Use only normal grace-period primitives,
3375 for example, synchronize_rcu() instead of
3376 synchronize_rcu_expedited(). This improves
3377 real-time latency, CPU utilization, and
3378 energy efficiency, but can expose users to
3379 increased grace-period latency. This parameter
3380 overrides rcupdate.rcu_expedited. No effect on
3381 CONFIG_TINY_RCU kernels.
3383 rcupdate.rcu_normal_after_boot= [KNL]
3384 Once boot has completed (that is, after
3385 rcu_end_inkernel_boot() has been invoked), use
3386 only normal grace-period primitives. No effect
3387 on CONFIG_TINY_RCU kernels.
3389 rcupdate.rcu_task_stall_timeout= [KNL]
3390 Set timeout in jiffies for RCU task stall warning
3391 messages. Disable with a value less than or equal
3392 to zero.
3394 rcupdate.rcu_self_test= [KNL]
3395 Run the RCU early boot self tests
3397 rcupdate.rcu_self_test_bh= [KNL]
3398 Run the RCU bh early boot self tests
3400 rcupdate.rcu_self_test_sched= [KNL]
3401 Run the RCU sched early boot self tests
3403 rdinit= [KNL]
3404 Format: <full_path>
3405 Run specified binary instead of /init from the ramdisk,
3406 used for early userspace startup. See initrd.
3408 reboot= [KNL]
3409 Format (x86 or x86_64):
3410 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
3411 [[,]s[mp]#### \
3412 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
3413 [[,]f[orce]
3414 Where reboot_mode is one of warm (soft) or cold (hard) or gpio,
3415 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
3416 reboot_force is either force or not specified,
3417 reboot_cpu is s[mp]#### with #### being the processor
3418 to be used for rebooting.
3420 relax_domain_level=
3421 [KNL, SMP] Set scheduler's default relax_domain_level.
3422 See Documentation/cgroups/cpusets.txt.
3424 relative_sleep_states=
3425 [SUSPEND] Use sleep state labeling where the deepest
3426 state available other than hibernation is always "mem".
3427 Format: { "0" | "1" }
3428 0 -- Traditional sleep state labels.
3429 1 -- Relative sleep state labels.
3431 reserve= [KNL,BUGS] Force the kernel to ignore some iomem area
3433 reservetop= [X86-32]
3434 Format: nn[KMG]
3435 Reserves a hole at the top of the kernel virtual
3436 address space.
3438 reservelow= [X86]
3439 Format: nn[K]
3440 Set the amount of memory to reserve for BIOS at
3441 the bottom of the address space.
3443 reset_devices [KNL] Force drivers to reset the underlying device
3444 during initialization.
3446 resume= [SWSUSP]
3447 Specify the partition device for software suspend
3448 Format:
3449 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
3451 resume_offset= [SWSUSP]
3452 Specify the offset from the beginning of the partition
3453 given by "resume=" at which the swap header is located,
3454 in <PAGE_SIZE> units (needed only for swap files).
3455 See Documentation/power/swsusp-and-swap-files.txt
3457 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
3458 read the resume files
3460 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
3461 Useful for devices that are detected asynchronously
3462 (e.g. USB and MMC devices).
3464 hibernate= [HIBERNATION]
3465 noresume Don't check if there's a hibernation image
3466 present during boot.
3467 nocompress Don't compress/decompress hibernation images.
3468 no Disable hibernation and resume.
3470 retain_initrd [RAM] Keep initrd memory after extraction
3472 rfkill.default_state=
3473 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
3474 etc. communication is blocked by default.
3475 1 Unblocked.
3477 rfkill.master_switch_mode=
3478 0 The "airplane mode" button does nothing.
3479 1 The "airplane mode" button toggles between everything
3480 blocked and the previous configuration.
3481 2 The "airplane mode" button toggles between everything
3482 blocked and everything unblocked.
3484 rhash_entries= [KNL,NET]
3485 Set number of hash buckets for route cache
3487 ro [KNL] Mount root device read-only on boot
3489 root= [KNL] Root filesystem
3490 See name_to_dev_t comment in init/do_mounts.c.
3492 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
3493 mount the root filesystem
3495 rootflags= [KNL] Set root filesystem mount option string
3497 rootfstype= [KNL] Set root filesystem type
3499 rootwait [KNL] Wait (indefinitely) for root device to show up.
3500 Useful for devices that are detected asynchronously
3501 (e.g. USB and MMC devices).
3503 rproc_mem=nn[KMG][@address]
3504 [KNL,ARM,CMA] Remoteproc physical memory block.
3505 Memory area to be used by remote processor image,
3506 managed by CMA.
3508 rw [KNL] Mount root device read-write on boot
3510 S [KNL] Run init in single mode
3512 s390_iommu= [HW,S390]
3513 Set s390 IOTLB flushing mode
3514 strict
3515 With strict flushing every unmap operation will result in
3516 an IOTLB flush. Default is lazy flushing before reuse,
3517 which is faster.
3519 sa1100ir [NET]
3520 See drivers/net/irda/sa1100_ir.c.
3522 sbni= [NET] Granch SBNI12 leased line adapter
3524 sched_debug [KNL] Enables verbose scheduler debug messages.
3526 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
3527 xtime_lock contention on larger systems, and/or RCU lock
3528 contention on all systems with CONFIG_MAXSMP set.
3529 Format: { "0" | "1" }
3530 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
3531 1 -- enable.
3532 Note: increases power consumption, thus should only be
3533 enabled if running jitter sensitive (HPC/RT) workloads.
3535 security= [SECURITY] Choose a security module to enable at boot.
3536 If this boot parameter is not specified, only the first
3537 security module asking for security registration will be
3538 loaded. An invalid security module name will be treated
3539 as if no module has been chosen.
3541 selinux= [SELINUX] Disable or enable SELinux at boot time.
3542 Format: { "0" | "1" }
3543 See security/selinux/Kconfig help text.
3544 0 -- disable.
3545 1 -- enable.
3546 Default value is set via kernel config option.
3547 If enabled at boot time, /selinux/disable can be used
3548 later to disable prior to initial policy load.
3550 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
3551 Format: { "0" | "1" }
3552 See security/apparmor/Kconfig help text
3553 0 -- disable.
3554 1 -- enable.
3555 Default value is set via kernel config option.
3557 serialnumber [BUGS=X86-32]
3559 shapers= [NET]
3560 Maximal number of shapers.
3562 show_msr= [x86] show boot-time MSR settings
3563 Format: { <integer> }
3564 Show boot-time (BIOS-initialized) MSR settings.
3565 The parameter means the number of CPUs to show,
3566 for example 1 means boot CPU only.
3568 simeth= [IA-64]
3569 simscsi=
3571 slram= [HW,MTD]
3573 slab_nomerge [MM]
3574 Disable merging of slabs with similar size. May be
3575 necessary if there is some reason to distinguish
3576 allocs to different slabs. Debug options disable
3577 merging on their own.
3578 For more information see Documentation/vm/slub.txt.
3580 slab_max_order= [MM, SLAB]
3581 Determines the maximum allowed order for slabs.
3582 A high setting may cause OOMs due to memory
3583 fragmentation. Defaults to 1 for systems with
3584 more than 32MB of RAM, 0 otherwise.
3586 slub_debug[=options[,slabs]] [MM, SLUB]
3587 Enabling slub_debug allows one to determine the
3588 culprit if slab objects become corrupted. Enabling
3589 slub_debug can create guard zones around objects and
3590 may poison objects when not in use. Also tracks the
3591 last alloc / free. For more information see
3592 Documentation/vm/slub.txt.
3594 slub_max_order= [MM, SLUB]
3595 Determines the maximum allowed order for slabs.
3596 A high setting may cause OOMs due to memory
3597 fragmentation. For more information see
3598 Documentation/vm/slub.txt.
3600 slub_min_objects= [MM, SLUB]
3601 The minimum number of objects per slab. SLUB will
3602 increase the slab order up to slub_max_order to
3603 generate a sufficiently large slab able to contain
3604 the number of objects indicated. The higher the number
3605 of objects the smaller the overhead of tracking slabs
3606 and the less frequently locks need to be acquired.
3607 For more information see Documentation/vm/slub.txt.
3609 slub_min_order= [MM, SLUB]
3610 Determines the minimum page order for slabs. Must be
3611 lower than slub_max_order.
3612 For more information see Documentation/vm/slub.txt.
3614 slub_nomerge [MM, SLUB]
3615 Same with slab_nomerge. This is supported for legacy.
3616 See slab_nomerge for more information.
3618 smart2= [HW]
3619 Format: <io1>[,<io2>[,...,<io8>]]
3621 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
3622 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
3623 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
3624 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
3625 smsc-ircc2.ircc_irq= [HW] IRQ line
3626 smsc-ircc2.ircc_dma= [HW] DMA channel
3627 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
3628 0: Toshiba Satellite 1800 (GP data pin select)
3629 1: Fast pin select (default)
3630 2: ATC IRMode
3632 softlockup_panic=
3633 [KNL] Should the soft-lockup detector generate panics.
3634 Format: <integer>
3636 softlockup_all_cpu_backtrace=
3637 [KNL] Should the soft-lockup detector generate
3638 backtraces on all cpus.
3639 Format: <integer>
3641 sonypi.*= [HW] Sony Programmable I/O Control Device driver
3642 See Documentation/laptops/sonypi.txt
3644 spia_io_base= [HW,MTD]
3645 spia_fio_base=
3646 spia_pedr=
3647 spia_peddr=
3649 stacktrace [FTRACE]
3650 Enabled the stack tracer on boot up.
3652 stacktrace_filter=[function-list]
3653 [FTRACE] Limit the functions that the stack tracer
3654 will trace at boot up. function-list is a comma separated
3655 list of functions. This list can be changed at run
3656 time by the stack_trace_filter file in the debugfs
3657 tracing directory. Note, this enables stack tracing
3658 and the stacktrace above is not needed.
3660 sti= [PARISC,HW]
3661 Format: <num>
3662 Set the STI (builtin display/keyboard on the HP-PARISC
3663 machines) console (graphic card) which should be used
3664 as the initial boot-console.
3665 See also comment in drivers/video/console/sticore.c.
3667 sti_font= [HW]
3668 See comment in drivers/video/console/sticore.c.
3670 stifb= [HW]
3671 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
3673 sunrpc.min_resvport=
3674 sunrpc.max_resvport=
3675 [NFS,SUNRPC]
3676 SunRPC servers often require that client requests
3677 originate from a privileged port (i.e. a port in the
3678 range 0 < portnr < 1024).
3679 An administrator who wishes to reserve some of these
3680 ports for other uses may adjust the range that the
3681 kernel's sunrpc client considers to be privileged
3682 using these two parameters to set the minimum and
3683 maximum port values.
3685 sunrpc.pool_mode=
3686 [NFS]
3687 Control how the NFS server code allocates CPUs to
3688 service thread pools. Depending on how many NICs
3689 you have and where their interrupts are bound, this
3690 option will affect which CPUs will do NFS serving.
3691 Note: this parameter cannot be changed while the
3692 NFS server is running.
3694 auto the server chooses an appropriate mode
3695 automatically using heuristics
3696 global a single global pool contains all CPUs
3697 percpu one pool for each CPU
3698 pernode one pool for each NUMA node (equivalent
3699 to global on non-NUMA machines)
3701 sunrpc.tcp_slot_table_entries=
3702 sunrpc.udp_slot_table_entries=
3703 [NFS,SUNRPC]
3704 Sets the upper limit on the number of simultaneous
3705 RPC calls that can be sent from the client to a
3706 server. Increasing these values may allow you to
3707 improve throughput, but will also increase the
3708 amount of memory reserved for use by the client.
3710 suspend.pm_test_delay=
3711 [SUSPEND]
3712 Sets the number of seconds to remain in a suspend test
3713 mode before resuming the system (see
3714 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
3715 is set. Default value is 5.
3717 swapaccount=[0|1]
3718 [KNL] Enable accounting of swap in memory resource
3719 controller if no parameter or 1 is given or disable
3720 it if 0 is given (See Documentation/cgroups/memory.txt)
3722 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
3723 Format: { <int> | force }
3724 <int> -- Number of I/O TLB slabs
3725 force -- force using of bounce buffers even if they
3726 wouldn't be automatically used by the kernel
3728 switches= [HW,M68k]
3730 sysfs.deprecated=0|1 [KNL]
3731 Enable/disable old style sysfs layout for old udev
3732 on older distributions. When this option is enabled
3733 very new udev will not work anymore. When this option
3734 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
3735 in older udev will not work anymore.
3736 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
3737 the kernel configuration.
3739 sysrq_always_enabled
3740 [KNL]
3741 Ignore sysrq setting - this boot parameter will
3742 neutralize any effect of /proc/sys/kernel/sysrq.
3743 Useful for debugging.
3745 tcpmhash_entries= [KNL,NET]
3746 Set the number of tcp_metrics_hash slots.
3747 Default value is 8192 or 16384 depending on total
3748 ram pages. This is used to specify the TCP metrics
3749 cache size. See Documentation/networking/ip-sysctl.txt
3750 "tcp_no_metrics_save" section for more details.
3752 tdfx= [HW,DRM]
3754 test_suspend= [SUSPEND][,N]
3755 Specify "mem" (for Suspend-to-RAM) or "standby" (for
3756 standby suspend) or "freeze" (for suspend type freeze)
3757 as the system sleep state during system startup with
3758 the optional capability to repeat N number of times.
3759 The system is woken from this state using a
3760 wakeup-capable RTC alarm.
3762 thash_entries= [KNL,NET]
3763 Set number of hash buckets for TCP connection
3765 thermal.act= [HW,ACPI]
3766 -1: disable all active trip points in all thermal zones
3767 <degrees C>: override all lowest active trip points
3769 thermal.crt= [HW,ACPI]
3770 -1: disable all critical trip points in all thermal zones
3771 <degrees C>: override all critical trip points
3773 thermal.nocrt= [HW,ACPI]
3774 Set to disable actions on ACPI thermal zone
3775 critical and hot trip points.
3777 thermal.off= [HW,ACPI]
3778 1: disable ACPI thermal control
3780 thermal.psv= [HW,ACPI]
3781 -1: disable all passive trip points
3782 <degrees C>: override all passive trip points to this
3783 value
3785 thermal.tzp= [HW,ACPI]
3786 Specify global default ACPI thermal zone polling rate
3787 <deci-seconds>: poll all this frequency
3788 0: no polling (default)
3790 threadirqs [KNL]
3791 Force threading of all interrupt handlers except those
3792 marked explicitly IRQF_NO_THREAD.
3794 tmem [KNL,XEN]
3795 Enable the Transcendent memory driver if built-in.
3797 tmem.cleancache=0|1 [KNL, XEN]
3798 Default is on (1). Disable the usage of the cleancache
3799 API to send anonymous pages to the hypervisor.
3801 tmem.frontswap=0|1 [KNL, XEN]
3802 Default is on (1). Disable the usage of the frontswap
3803 API to send swap pages to the hypervisor. If disabled
3804 the selfballooning and selfshrinking are force disabled.
3806 tmem.selfballooning=0|1 [KNL, XEN]
3807 Default is on (1). Disable the driving of swap pages
3808 to the hypervisor.
3810 tmem.selfshrinking=0|1 [KNL, XEN]
3811 Default is on (1). Partial swapoff that immediately
3812 transfers pages from Xen hypervisor back to the
3813 kernel based on different criteria.
3815 topology= [S390]
3816 Format: {off | on}
3817 Specify if the kernel should make use of the cpu
3818 topology information if the hardware supports this.
3819 The scheduler will make use of this information and
3820 e.g. base its process migration decisions on it.
3821 Default is on.
3823 topology_updates= [KNL, PPC, NUMA]
3824 Format: {off}
3825 Specify if the kernel should ignore (off)
3826 topology updates sent by the hypervisor to this
3827 LPAR.
3829 tp720= [HW,PS2]
3831 tpm_suspend_pcr=[HW,TPM]
3832 Format: integer pcr id
3833 Specify that at suspend time, the tpm driver
3834 should extend the specified pcr with zeros,
3835 as a workaround for some chips which fail to
3836 flush the last written pcr on TPM_SaveState.
3837 This will guarantee that all the other pcrs
3838 are saved.
3840 trace_buf_size=nn[KMG]
3841 [FTRACE] will set tracing buffer size on each cpu.
3843 trace_event=[event-list]
3844 [FTRACE] Set and start specified trace events in order
3845 to facilitate early boot debugging.
3846 See also Documentation/trace/events.txt
3848 trace_options=[option-list]
3849 [FTRACE] Enable or disable tracer options at boot.
3850 The option-list is a comma delimited list of options
3851 that can be enabled or disabled just as if you were
3852 to echo the option name into
3854 /sys/kernel/debug/tracing/trace_options
3856 For example, to enable stacktrace option (to dump the
3857 stack trace of each event), add to the command line:
3859 trace_options=stacktrace
3861 See also Documentation/trace/ftrace.txt "trace options"
3862 section.
3864 tp_printk[FTRACE]
3865 Have the tracepoints sent to printk as well as the
3866 tracing ring buffer. This is useful for early boot up
3867 where the system hangs or reboots and does not give the
3868 option for reading the tracing buffer or performing a
3869 ftrace_dump_on_oops.
3871 To turn off having tracepoints sent to printk,
3872 echo 0 > /proc/sys/kernel/tracepoint_printk
3873 Note, echoing 1 into this file without the
3874 tracepoint_printk kernel cmdline option has no effect.
3876 ** CAUTION **
3878 Having tracepoints sent to printk() and activating high
3879 frequency tracepoints such as irq or sched, can cause
3880 the system to live lock.
3882 traceoff_on_warning
3883 [FTRACE] enable this option to disable tracing when a
3884 warning is hit. This turns off "tracing_on". Tracing can
3885 be enabled again by echoing '1' into the "tracing_on"
3886 file located in /sys/kernel/debug/tracing/
3888 This option is useful, as it disables the trace before
3889 the WARNING dump is called, which prevents the trace to
3890 be filled with content caused by the warning output.
3892 This option can also be set at run time via the sysctl
3893 option: kernel/traceoff_on_warning
3895 transparent_hugepage=
3896 [KNL]
3897 Format: [always|madvise|never]
3898 Can be used to control the default behavior of the system
3899 with respect to transparent hugepages.
3900 See Documentation/vm/transhuge.txt for more details.
3902 tsc= Disable clocksource stability checks for TSC.
3903 Format: <string>
3904 [x86] reliable: mark tsc clocksource as reliable, this
3905 disables clocksource verification at runtime, as well
3906 as the stability checks done at bootup. Used to enable
3907 high-resolution timer mode on older hardware, and in
3908 virtualized environment.
3909 [x86] noirqtime: Do not use TSC to do irq accounting.
3910 Used to run time disable IRQ_TIME_ACCOUNTING on any
3911 platforms where RDTSC is slow and this accounting
3912 can add overhead.
3914 turbografx.map[2|3]= [HW,JOY]
3915 TurboGraFX parallel port interface
3916 Format:
3917 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
3918 See also Documentation/input/joystick-parport.txt
3920 udbg-immortal [PPC] When debugging early kernel crashes that
3921 happen after console_init() and before a proper
3922 console driver takes over, this boot options might
3923 help "seeing" what's going on.
3925 uhash_entries= [KNL,NET]
3926 Set number of hash buckets for UDP/UDP-Lite connections
3928 uhci-hcd.ignore_oc=
3929 [USB] Ignore overcurrent events (default N).
3930 Some badly-designed motherboards generate lots of
3931 bogus events, for ports that aren't wired to
3932 anything. Set this parameter to avoid log spamming.
3933 Note that genuine overcurrent events won't be
3934 reported either.
3936 unknown_nmi_panic
3937 [X86] Cause panic on unknown NMI.
3939 usbcore.authorized_default=
3940 [USB] Default USB device authorization:
3941 (default -1 = authorized except for wireless USB,
3942 0 = not authorized, 1 = authorized)
3944 usbcore.autosuspend=
3945 [USB] The autosuspend time delay (in seconds) used
3946 for newly-detected USB devices (default 2). This
3947 is the time required before an idle device will be
3948 autosuspended. Devices for which the delay is set
3949 to a negative value won't be autosuspended at all.
3951 usbcore.usbfs_snoop=
3952 [USB] Set to log all usbfs traffic (default 0 = off).
3954 usbcore.usbfs_snoop_max=
3955 [USB] Maximum number of bytes to snoop in each URB
3956 (default = 65536).
3958 usbcore.blinkenlights=
3959 [USB] Set to cycle leds on hubs (default 0 = off).
3961 usbcore.old_scheme_first=
3962 [USB] Start with the old device initialization
3963 scheme (default 0 = off).
3965 usbcore.usbfs_memory_mb=
3966 [USB] Memory limit (in MB) for buffers allocated by
3967 usbfs (default = 16, 0 = max = 2047).
3969 usbcore.use_both_schemes=
3970 [USB] Try the other device initialization scheme
3971 if the first one fails (default 1 = enabled).
3973 usbcore.initial_descriptor_timeout=
3974 [USB] Specifies timeout for the initial 64-byte
3975 USB_REQ_GET_DESCRIPTOR request in milliseconds
3976 (default 5000 = 5.0 seconds).
3978 usbcore.nousb [USB] Disable the USB subsystem
3980 usbhid.mousepoll=
3981 [USBHID] The interval which mice are to be polled at.
3983 usb-storage.delay_use=
3984 [UMS] The delay in seconds before a new device is
3985 scanned for Logical Units (default 1).
3987 usb-storage.quirks=
3988 [UMS] A list of quirks entries to supplement or
3989 override the built-in unusual_devs list. List
3990 entries are separated by commas. Each entry has
3991 the form VID:PID:Flags where VID and PID are Vendor
3992 and Product ID values (4-digit hex numbers) and
3993 Flags is a set of characters, each corresponding
3994 to a common usb-storage quirk flag as follows:
3995 a = SANE_SENSE (collect more than 18 bytes
3996 of sense data);
3997 b = BAD_SENSE (don't collect more than 18
3998 bytes of sense data);
3999 c = FIX_CAPACITY (decrease the reported
4000 device capacity by one sector);
4001 d = NO_READ_DISC_INFO (don't use
4002 READ_DISC_INFO command);
4003 e = NO_READ_CAPACITY_16 (don't use
4004 READ_CAPACITY_16 command);
4005 f = NO_REPORT_OPCODES (don't use report opcodes
4006 command, uas only);
4007 g = MAX_SECTORS_240 (don't transfer more than
4008 240 sectors at a time, uas only);
4009 h = CAPACITY_HEURISTICS (decrease the
4010 reported device capacity by one
4011 sector if the number is odd);
4012 i = IGNORE_DEVICE (don't bind to this
4013 device);
4014 l = NOT_LOCKABLE (don't try to lock and
4015 unlock ejectable media);
4016 m = MAX_SECTORS_64 (don't transfer more
4017 than 64 sectors = 32 KB at a time);
4018 n = INITIAL_READ10 (force a retry of the
4019 initial READ(10) command);
4020 o = CAPACITY_OK (accept the capacity
4021 reported by the device);
4022 p = WRITE_CACHE (the device cache is ON
4023 by default);
4024 r = IGNORE_RESIDUE (the device reports
4025 bogus residue values);
4026 s = SINGLE_LUN (the device has only one
4027 Logical Unit);
4028 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
4029 commands, uas only);
4030 u = IGNORE_UAS (don't bind to the uas driver);
4031 w = NO_WP_DETECT (don't test whether the
4032 medium is write-protected).
4033 Example: quirks=0419:aaf5:rl,0421:0433:rc
4035 user_debug= [KNL,ARM]
4036 Format: <int>
4037 See arch/arm/Kconfig.debug help text.
4038 1 - undefined instruction events
4039 2 - system calls
4040 4 - invalid data aborts
4041 8 - SIGSEGV faults
4042 16 - SIGBUS faults
4043 Example: user_debug=31
4045 userpte=
4046 [X86] Flags controlling user PTE allocations.
4048 nohigh = do not allocate PTE pages in
4049 HIGHMEM regardless of setting
4050 of CONFIG_HIGHPTE.
4052 vdso= [X86,SH]
4053 On X86_32, this is an alias for vdso32=. Otherwise:
4055 vdso=1: enable VDSO (the default)
4056 vdso=0: disable VDSO mapping
4058 vdso32= [X86] Control the 32-bit vDSO
4059 vdso32=1: enable 32-bit VDSO
4060 vdso32=0 or vdso32=2: disable 32-bit VDSO
4062 See the help text for CONFIG_COMPAT_VDSO for more
4063 details. If CONFIG_COMPAT_VDSO is set, the default is
4064 vdso32=0; otherwise, the default is vdso32=1.
4066 For compatibility with older kernels, vdso32=2 is an
4067 alias for vdso32=0.
4069 Try vdso32=0 if you encounter an error that says:
4070 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
4072 vector= [IA-64,SMP]
4073 vector=percpu: enable percpu vector domain
4075 video= [FB] Frame buffer configuration
4076 See Documentation/fb/modedb.txt.
4078 video.brightness_switch_enabled= [0,1]
4079 If set to 1, on receiving an ACPI notify event
4080 generated by hotkey, video driver will adjust brightness
4081 level and then send out the event to user space through
4082 the allocated input device; If set to 0, video driver
4083 will only send out the event without touching backlight
4084 brightness level.
4085 default: 1
4087 virtio_mmio.device=
4088 [VMMIO] Memory mapped virtio (platform) device.
4090 <size>@<baseaddr>:<irq>[:<id>]
4091 where:
4092 <size> := size (can use standard suffixes
4093 like K, M and G)
4094 <baseaddr> := physical base address
4095 <irq> := interrupt number (as passed to
4096 request_irq())
4097 <id> := (optional) platform device id
4098 example:
4099 virtio_mmio.device=1K@0x100b0000:48:7
4101 Can be used multiple times for multiple devices.
4103 vga= [BOOT,X86-32] Select a particular video mode
4104 See Documentation/x86/boot.txt and
4105 Documentation/svga.txt.
4106 Use vga=ask for menu.
4107 This is actually a boot loader parameter; the value is
4108 passed to the kernel using a special protocol.
4110 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
4111 size of <nn>. This can be used to increase the
4112 minimum size (128MB on x86). It can also be used to
4113 decrease the size and leave more room for directly
4114 mapped kernel RAM.
4116 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
4117 Format: <command>
4119 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
4120 Format: <command>
4122 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
4123 Format: <command>
4125 vsyscall= [X86-64]
4126 Controls the behavior of vsyscalls (i.e. calls to
4127 fixed addresses of 0xffffffffff600x00 from legacy
4128 code). Most statically-linked binaries and older
4129 versions of glibc use these calls. Because these
4130 functions are at fixed addresses, they make nice
4131 targets for exploits that can control RIP.
4133 emulate [default] Vsyscalls turn into traps and are
4134 emulated reasonably safely.
4136 native Vsyscalls are native syscall instructions.
4137 This is a little bit faster than trapping
4138 and makes a few dynamic recompilers work
4139 better than they would in emulation mode.
4140 It also makes exploits much easier to write.
4142 none Vsyscalls don't work at all. This makes
4143 them quite hard to use for exploits but
4144 might break your system.
4146 vt.color= [VT] Default text color.
4147 Format: 0xYX, X = foreground, Y = background.
4148 Default: 0x07 = light gray on black.
4150 vt.cur_default= [VT] Default cursor shape.
4151 Format: 0xCCBBAA, where AA, BB, and CC are the same as
4152 the parameters of the <Esc>[?A;B;Cc escape sequence;
4153 see VGA-softcursor.txt. Default: 2 = underline.
4155 vt.default_blu= [VT]
4156 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
4157 Change the default blue palette of the console.
4158 This is a 16-member array composed of values
4159 ranging from 0-255.
4161 vt.default_grn= [VT]
4162 Format: <green0>,<green1>,<green2>,...,<green15>
4163 Change the default green palette of the console.
4164 This is a 16-member array composed of values
4165 ranging from 0-255.
4167 vt.default_red= [VT]
4168 Format: <red0>,<red1>,<red2>,...,<red15>
4169 Change the default red palette of the console.
4170 This is a 16-member array composed of values
4171 ranging from 0-255.
4173 vt.default_utf8=
4174 [VT]
4175 Format=<0|1>
4176 Set system-wide default UTF-8 mode for all tty's.
4177 Default is 1, i.e. UTF-8 mode is enabled for all
4178 newly opened terminals.
4180 vt.global_cursor_default=
4181 [VT]
4182 Format=<-1|0|1>
4183 Set system-wide default for whether a cursor
4184 is shown on new VTs. Default is -1,
4185 i.e. cursors will be created by default unless
4186 overridden by individual drivers. 0 will hide
4187 cursors, 1 will display them.
4189 vt.italic= [VT] Default color for italic text; 0-15.
4190 Default: 2 = green.
4192 vt.underline= [VT] Default color for underlined text; 0-15.
4193 Default: 3 = cyan.
4195 watchdog timers [HW,WDT] For information on watchdog timers,
4196 see Documentation/watchdog/watchdog-parameters.txt
4197 or other driver-specific files in the
4198 Documentation/watchdog/ directory.
4200 workqueue.watchdog_thresh=
4201 If CONFIG_WQ_WATCHDOG is configured, workqueue can
4202 warn stall conditions and dump internal state to
4203 help debugging. 0 disables workqueue stall
4204 detection; otherwise, it's the stall threshold
4205 duration in seconds. The default value is 30 and
4206 it can be updated at runtime by writing to the
4207 corresponding sysfs file.
4209 workqueue.disable_numa
4210 By default, all work items queued to unbound
4211 workqueues are affine to the NUMA nodes they're
4212 issued on, which results in better behavior in
4213 general. If NUMA affinity needs to be disabled for
4214 whatever reason, this option can be used. Note
4215 that this also can be controlled per-workqueue for
4216 workqueues visible under /sys/bus/workqueue/.
4218 workqueue.power_efficient
4219 Per-cpu workqueues are generally preferred because
4220 they show better performance thanks to cache
4221 locality; unfortunately, per-cpu workqueues tend to
4222 be more power hungry than unbound workqueues.
4224 Enabling this makes the per-cpu workqueues which
4225 were observed to contribute significantly to power
4226 consumption unbound, leading to measurably lower
4227 power usage at the cost of small performance
4228 overhead.
4230 The default value of this parameter is determined by
4231 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
4233 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
4234 default x2apic cluster mode on platforms
4235 supporting x2apic.
4237 x86_intel_mid_timer= [X86-32,APBT]
4238 Choose timer option for x86 Intel MID platform.
4239 Two valid options are apbt timer only and lapic timer
4240 plus one apbt timer for broadcast timer.
4241 x86_intel_mid_timer=apbt_only | lapic_and_apbt
4243 xen_512gb_limit [KNL,X86-64,XEN]
4244 Restricts the kernel running paravirtualized under Xen
4245 to use only up to 512 GB of RAM. The reason to do so is
4246 crash analysis tools and Xen tools for doing domain
4247 save/restore/migration must be enabled to handle larger
4248 domains.
4250 xen_emul_unplug= [HW,X86,XEN]
4251 Unplug Xen emulated devices
4252 Format: [unplug0,][unplug1]
4253 ide-disks -- unplug primary master IDE devices
4254 aux-ide-disks -- unplug non-primary-master IDE devices
4255 nics -- unplug network devices
4256 all -- unplug all emulated devices (NICs and IDE disks)
4257 unnecessary -- unplugging emulated devices is
4258 unnecessary even if the host did not respond to
4259 the unplug protocol
4260 never -- do not unplug even if version check succeeds
4262 xen_nopvspin [X86,XEN]
4263 Disables the ticketlock slowpath using Xen PV
4264 optimizations.
4266 xen_nopv [X86]
4267 Disables the PV optimizations forcing the HVM guest to
4268 run as generic HVM guest with no PV drivers.
4270 xirc2ps_cs= [NET,PCMCIA]
4271 Format:
4272 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
4274 ______________________________________________________________________
4276 TODO:
4278 Add more DRM drivers.