Linux Audio

Check our new training course

Loading...
v3.15
   1                          Kernel Parameters
   2                          ~~~~~~~~~~~~~~~~~
   3
   4The following is a consolidated list of the kernel parameters as implemented
   5(mostly) by the __setup() macro and sorted into English Dictionary order
   6(defined as ignoring all punctuation and sorting digits before letters in a
   7case insensitive manner), and with descriptions where known.
   8
   9Module parameters for loadable modules are specified only as the
  10parameter name with optional '=' and value as appropriate, such as:
  11
  12	modprobe usbcore blinkenlights=1
  13
  14Module parameters for modules that are built into the kernel image
  15are specified on the kernel command line with the module name plus
  16'.' plus parameter name, with '=' and value if appropriate, such as:
  17
  18	usbcore.blinkenlights=1
  19
  20Hyphens (dashes) and underscores are equivalent in parameter names, so
  21	log_buf_len=1M print-fatal-signals=1
  22can also be entered as
  23	log-buf-len=1M print_fatal_signals=1
  24
  25
  26This document may not be entirely up to date and comprehensive. The command
  27"modinfo -p ${modulename}" shows a current list of all parameters of a loadable
  28module. Loadable modules, after being loaded into the running kernel, also
  29reveal their parameters in /sys/module/${modulename}/parameters/. Some of these
  30parameters may be changed at runtime by the command
  31"echo -n ${value} > /sys/module/${modulename}/parameters/${parm}".
  32
  33The parameters listed below are only valid if certain kernel build options were
  34enabled and if respective hardware is present. The text in square brackets at
  35the beginning of each description states the restrictions within which a
  36parameter is applicable:
  37
  38	ACPI	ACPI support is enabled.
  39	AGP	AGP (Accelerated Graphics Port) is enabled.
  40	ALSA	ALSA sound support is enabled.
  41	APIC	APIC support is enabled.
  42	APM	Advanced Power Management support is enabled.
  43	ARM	ARM architecture is enabled.
  44	AVR32	AVR32 architecture is enabled.
  45	AX25	Appropriate AX.25 support is enabled.
  46	BLACKFIN Blackfin architecture is enabled.
  47	CLK	Common clock infrastructure is enabled.
  48	CMA	Contiguous Memory Area support is enabled.
  49	DRM	Direct Rendering Management support is enabled.
  50	DYNAMIC_DEBUG Build in debug messages and enable them at runtime
  51	EDD	BIOS Enhanced Disk Drive Services (EDD) is enabled
  52	EFI	EFI Partitioning (GPT) is enabled
  53	EIDE	EIDE/ATAPI support is enabled.
  54	EVM	Extended Verification Module
  55	FB	The frame buffer device is enabled.
  56	FTRACE	Function tracing enabled.
  57	GCOV	GCOV profiling is enabled.
  58	HW	Appropriate hardware is enabled.
  59	IA-64	IA-64 architecture is enabled.
  60	IMA     Integrity measurement architecture is enabled.
  61	IOSCHED	More than one I/O scheduler is enabled.
  62	IP_PNP	IP DHCP, BOOTP, or RARP is enabled.
  63	IPV6	IPv6 support is enabled.
  64	ISAPNP	ISA PnP code is enabled.
  65	ISDN	Appropriate ISDN support is enabled.
  66	JOY	Appropriate joystick support is enabled.
  67	KGDB	Kernel debugger support is enabled.
  68	KVM	Kernel Virtual Machine support is enabled.
  69	LIBATA  Libata driver is enabled
  70	LP	Printer support is enabled.
  71	LOOP	Loopback device support is enabled.
  72	M68k	M68k architecture is enabled.
  73			These options have more detailed description inside of
  74			Documentation/m68k/kernel-options.txt.
 
  75	MDA	MDA console support is enabled.
  76	MIPS	MIPS architecture is enabled.
  77	MOUSE	Appropriate mouse support is enabled.
  78	MSI	Message Signaled Interrupts (PCI).
  79	MTD	MTD (Memory Technology Device) support is enabled.
  80	NET	Appropriate network support is enabled.
  81	NUMA	NUMA support is enabled.
  82	NFS	Appropriate NFS support is enabled.
  83	OSS	OSS sound support is enabled.
  84	PV_OPS	A paravirtualized kernel is enabled.
  85	PARIDE	The ParIDE (parallel port IDE) subsystem is enabled.
  86	PARISC	The PA-RISC architecture is enabled.
  87	PCI	PCI bus support is enabled.
  88	PCIE	PCI Express support is enabled.
  89	PCMCIA	The PCMCIA subsystem is enabled.
  90	PNP	Plug & Play support is enabled.
  91	PPC	PowerPC architecture is enabled.
  92	PPT	Parallel port support is enabled.
  93	PS2	Appropriate PS/2 support is enabled.
  94	RAM	RAM disk support is enabled.
  95	S390	S390 architecture is enabled.
  96	SCSI	Appropriate SCSI support is enabled.
  97			A lot of drivers have their options described inside
  98			the Documentation/scsi/ sub-directory.
  99	SECURITY Different security models are enabled.
 100	SELINUX SELinux support is enabled.
 101	APPARMOR AppArmor support is enabled.
 102	SERIAL	Serial support is enabled.
 103	SH	SuperH architecture is enabled.
 104	SMP	The kernel is an SMP kernel.
 105	SPARC	Sparc architecture is enabled.
 106	SWSUSP	Software suspend (hibernation) is enabled.
 107	SUSPEND	System suspend states are enabled.
 108	TPM	TPM drivers are enabled.
 109	TS	Appropriate touchscreen support is enabled.
 110	UMS	USB Mass Storage support is enabled.
 111	USB	USB support is enabled.
 112	USBHID	USB Human Interface Device support is enabled.
 113	V4L	Video For Linux support is enabled.
 114	VMMIO   Driver for memory mapped virtio devices is enabled.
 115	VGA	The VGA console has been enabled.
 116	VT	Virtual terminal support is enabled.
 117	WDT	Watchdog support is enabled.
 118	XT	IBM PC/XT MFM hard disk support is enabled.
 119	X86-32	X86-32, aka i386 architecture is enabled.
 120	X86-64	X86-64 architecture is enabled.
 121			More X86-64 boot options can be found in
 122			Documentation/x86/x86_64/boot-options.txt .
 123	X86	Either 32-bit or 64-bit x86 (same as X86-32+X86-64)
 124	XEN	Xen support is enabled
 125
 126In addition, the following text indicates that the option:
 127
 128	BUGS=	Relates to possible processor bugs on the said processor.
 129	KNL	Is a kernel start-up parameter.
 130	BOOT	Is a boot loader parameter.
 131
 132Parameters denoted with BOOT are actually interpreted by the boot
 133loader, and have no meaning to the kernel directly.
 134Do not modify the syntax of boot loader parameters without extreme
 135need or coordination with <Documentation/x86/boot.txt>.
 136
 137There are also arch-specific kernel-parameters not documented here.
 138See for example <Documentation/x86/x86_64/boot-options.txt>.
 139
 140Note that ALL kernel parameters listed below are CASE SENSITIVE, and that
 141a trailing = on the name of any parameter states that that parameter will
 142be entered as an environment variable, whereas its absence indicates that
 143it will appear as a kernel argument readable via /proc/cmdline by programs
 144running once the system is up.
 145
 146The number of kernel parameters is not limited, but the length of the
 147complete command line (parameters including spaces etc.) is limited to
 148a fixed number of characters. This limit depends on the architecture
 149and is between 256 and 4096 characters. It is defined in the file
 150./include/asm/setup.h as COMMAND_LINE_SIZE.
 151
 152Finally, the [KMG] suffix is commonly described after a number of kernel
 153parameter values. These 'K', 'M', and 'G' letters represent the _binary_
 154multipliers 'Kilo', 'Mega', and 'Giga', equalling 2^10, 2^20, and 2^30
 155bytes respectively. Such letter suffixes can also be entirely omitted.
 156
 157
 158	acpi=		[HW,ACPI,X86]
 159			Advanced Configuration and Power Interface
 160			Format: { force | off | strict | noirq | rsdt }
 161			force -- enable ACPI if default was off
 162			off -- disable ACPI if default was on
 163			noirq -- do not use ACPI for IRQ routing
 164			strict -- Be less tolerant of platforms that are not
 165				strictly ACPI specification compliant.
 166			rsdt -- prefer RSDT over (default) XSDT
 167			copy_dsdt -- copy DSDT to memory
 168
 169			See also Documentation/power/runtime_pm.txt, pci=noacpi
 170
 171	acpi_rsdp=	[ACPI,EFI,KEXEC]
 172			Pass the RSDP address to the kernel, mostly used
 173			on machines running EFI runtime service to boot the
 174			second kernel for kdump.
 175
 176	acpi_apic_instance=	[ACPI, IOAPIC]
 177			Format: <int>
 178			2: use 2nd APIC table, if available
 179			1,0: use 1st APIC table
 180			default: 0
 181
 182	acpi_backlight=	[HW,ACPI]
 183			acpi_backlight=vendor
 184			acpi_backlight=video
 185			If set to vendor, prefer vendor specific driver
 186			(e.g. thinkpad_acpi, sony_acpi, etc.) instead
 187			of the ACPI video.ko driver.
 188
 189	acpi.debug_layer=	[HW,ACPI,ACPI_DEBUG]
 190	acpi.debug_level=	[HW,ACPI,ACPI_DEBUG]
 191			Format: <int>
 192			CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
 193			debug output.  Bits in debug_layer correspond to a
 194			_COMPONENT in an ACPI source file, e.g.,
 195			    #define _COMPONENT ACPI_PCI_COMPONENT
 196			Bits in debug_level correspond to a level in
 197			ACPI_DEBUG_PRINT statements, e.g.,
 198			    ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
 199			The debug_level mask defaults to "info".  See
 200			Documentation/acpi/debug.txt for more information about
 201			debug layers and levels.
 202
 203			Enable processor driver info messages:
 204			    acpi.debug_layer=0x20000000
 205			Enable PCI/PCI interrupt routing info messages:
 206			    acpi.debug_layer=0x400000
 207			Enable AML "Debug" output, i.e., stores to the Debug
 208			object while interpreting AML:
 209			    acpi.debug_layer=0xffffffff acpi.debug_level=0x2
 210			Enable all messages related to ACPI hardware:
 211			    acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
 212
 213			Some values produce so much output that the system is
 214			unusable.  The "log_buf_len" parameter may be useful
 215			if you need to capture more output.
 216
 217	acpi_irq_balance [HW,ACPI]
 218			ACPI will balance active IRQs
 219			default in APIC mode
 220
 221	acpi_irq_nobalance [HW,ACPI]
 222			ACPI will not move active IRQs (default)
 223			default in PIC mode
 224
 225	acpi_irq_isa=	[HW,ACPI] If irq_balance, mark listed IRQs used by ISA
 226			Format: <irq>,<irq>...
 227
 228	acpi_irq_pci=	[HW,ACPI] If irq_balance, clear listed IRQs for
 229			use by PCI
 230			Format: <irq>,<irq>...
 231
 232	acpi_no_auto_serialize	[HW,ACPI]
 233			Disable auto-serialization of AML methods
 234			AML control methods that contain the opcodes to create
 235			named objects will be marked as "Serialized" by the
 236			auto-serialization feature.
 237			This feature is enabled by default.
 238			This option allows to turn off the feature.
 239
 240	acpi_no_auto_ssdt	[HW,ACPI] Disable automatic loading of SSDT
 241
 242	acpica_no_return_repair [HW, ACPI]
 243			Disable AML predefined validation mechanism
 244			This mechanism can repair the evaluation result to make
 245			the return objects more ACPI specification compliant.
 246			This option is useful for developers to identify the
 247			root cause of an AML interpreter issue when the issue
 248			has something to do with the repair mechanism.
 249
 250	acpi_os_name=	[HW,ACPI] Tell ACPI BIOS the name of the OS
 251			Format: To spoof as Windows 98: ="Microsoft Windows"
 252
 253	acpi_osi=	[HW,ACPI] Modify list of supported OS interface strings
 254			acpi_osi="string1"	# add string1
 255			acpi_osi="!string2"	# remove string2
 256			acpi_osi=!*		# remove all strings
 257			acpi_osi=!		# disable all built-in OS vendor
 258						  strings
 259			acpi_osi=		# disable all strings
 260
 261			'acpi_osi=!' can be used in combination with single or
 262			multiple 'acpi_osi="string1"' to support specific OS
 263			vendor string(s).  Note that such command can only
 264			affect the default state of the OS vendor strings, thus
 265			it cannot affect the default state of the feature group
 266			strings and the current state of the OS vendor strings,
 267			specifying it multiple times through kernel command line
 268			is meaningless.  This command is useful when one do not
 269			care about the state of the feature group strings which
 270			should be controlled by the OSPM.
 271			Examples:
 272			  1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
 273			     to 'acpi_osi="Windows 2000" acpi_osi=!', they all
 274			     can make '_OSI("Windows 2000")' TRUE.
 275
 276			'acpi_osi=' cannot be used in combination with other
 277			'acpi_osi=' command lines, the _OSI method will not
 278			exist in the ACPI namespace.  NOTE that such command can
 279			only affect the _OSI support state, thus specifying it
 280			multiple times through kernel command line is also
 281			meaningless.
 282			Examples:
 283			  1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
 284			     FALSE.
 285
 286			'acpi_osi=!*' can be used in combination with single or
 287			multiple 'acpi_osi="string1"' to support specific
 288			string(s).  Note that such command can affect the
 289			current state of both the OS vendor strings and the
 290			feature group strings, thus specifying it multiple times
 291			through kernel command line is meaningful.  But it may
 292			still not able to affect the final state of a string if
 293			there are quirks related to this string.  This command
 294			is useful when one want to control the state of the
 295			feature group strings to debug BIOS issues related to
 296			the OSPM features.
 297			Examples:
 298			  1. 'acpi_osi="Module Device" acpi_osi=!*' can make
 299			     '_OSI("Module Device")' FALSE.
 300			  2. 'acpi_osi=!* acpi_osi="Module Device"' can make
 301			     '_OSI("Module Device")' TRUE.
 302			  3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
 303			     equivalent to
 304			     'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
 305			     and
 306			     'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
 307			     they all will make '_OSI("Windows 2000")' TRUE.
 308
 309	acpi_pm_good	[X86]
 310			Override the pmtimer bug detection: force the kernel
 311			to assume that this machine's pmtimer latches its value
 312			and always returns good values.
 313
 314	acpi_sci=	[HW,ACPI] ACPI System Control Interrupt trigger mode
 315			Format: { level | edge | high | low }
 316
 
 
 317	acpi_skip_timer_override [HW,ACPI]
 318			Recognize and ignore IRQ0/pin2 Interrupt Override.
 319			For broken nForce2 BIOS resulting in XT-PIC timer.
 320
 321	acpi_sleep=	[HW,ACPI] Sleep options
 322			Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
 323				  old_ordering, nonvs, sci_force_enable }
 324			See Documentation/power/video.txt for information on
 325			s3_bios and s3_mode.
 326			s3_beep is for debugging; it makes the PC's speaker beep
 327			as soon as the kernel's real-mode entry point is called.
 328			s4_nohwsig prevents ACPI hardware signature from being
 329			used during resume from hibernation.
 330			old_ordering causes the ACPI 1.0 ordering of the _PTS
 331			control method, with respect to putting devices into
 332			low power states, to be enforced (the ACPI 2.0 ordering
 333			of _PTS is used by default).
 334			nonvs prevents the kernel from saving/restoring the
 335			ACPI NVS memory during suspend/hibernation and resume.
 336			sci_force_enable causes the kernel to set SCI_EN directly
 337			on resume from S1/S3 (which is against the ACPI spec,
 338			but some broken systems don't work without it).
 339
 340	acpi_use_timer_override [HW,ACPI]
 341			Use timer override. For some broken Nvidia NF5 boards
 342			that require a timer override, but don't have HPET
 343
 344	acpi_enforce_resources=	[ACPI]
 345			{ strict | lax | no }
 346			Check for resource conflicts between native drivers
 347			and ACPI OperationRegions (SystemIO and SystemMemory
 348			only). IO ports and memory declared in ACPI might be
 349			used by the ACPI subsystem in arbitrary AML code and
 350			can interfere with legacy drivers.
 351			strict (default): access to resources claimed by ACPI
 352			is denied; legacy drivers trying to access reserved
 353			resources will fail to bind to device using them.
 354			lax: access to resources claimed by ACPI is allowed;
 355			legacy drivers trying to access reserved resources
 356			will bind successfully but a warning message is logged.
 357			no: ACPI OperationRegions are not marked as reserved,
 358			no further checks are performed.
 359
 360	acpi_no_memhotplug [ACPI] Disable memory hotplug.  Useful for kdump
 361			   kernels.
 362
 363	add_efi_memmap	[EFI; X86] Include EFI memory map in
 364			kernel's map of available physical RAM.
 365
 366	agp=		[AGP]
 367			{ off | try_unsupported }
 368			off: disable AGP support
 369			try_unsupported: try to drive unsupported chipsets
 370				(may crash computer or cause data corruption)
 371
 372	ALSA		[HW,ALSA]
 373			See Documentation/sound/alsa/alsa-parameters.txt
 374
 375	alignment=	[KNL,ARM]
 376			Allow the default userspace alignment fault handler
 377			behaviour to be specified.  Bit 0 enables warnings,
 378			bit 1 enables fixups, and bit 2 sends a segfault.
 379
 380	align_va_addr=	[X86-64]
 381			Align virtual addresses by clearing slice [14:12] when
 382			allocating a VMA at process creation time. This option
 383			gives you up to 3% performance improvement on AMD F15h
 384			machines (where it is enabled by default) for a
 385			CPU-intensive style benchmark, and it can vary highly in
 386			a microbenchmark depending on workload and compiler.
 387
 388			32: only for 32-bit processes
 389			64: only for 64-bit processes
 390			on: enable for both 32- and 64-bit processes
 391			off: disable for both 32- and 64-bit processes
 392
 393	alloc_snapshot	[FTRACE]
 394			Allocate the ftrace snapshot buffer on boot up when the
 395			main buffer is allocated. This is handy if debugging
 396			and you need to use tracing_snapshot() on boot up, and
 397			do not want to use tracing_snapshot_alloc() as it needs
 398			to be done where GFP_KERNEL allocations are allowed.
 399
 400	amd_iommu=	[HW,X86-64]
 401			Pass parameters to the AMD IOMMU driver in the system.
 402			Possible values are:
 403			fullflush - enable flushing of IO/TLB entries when
 404				    they are unmapped. Otherwise they are
 405				    flushed before they will be reused, which
 406				    is a lot of faster
 407			off	  - do not initialize any AMD IOMMU found in
 408				    the system
 409			force_isolation - Force device isolation for all
 410					  devices. The IOMMU driver is not
 411					  allowed anymore to lift isolation
 412					  requirements as needed. This option
 413					  does not override iommu=pt
 414
 415	amd_iommu_dump=	[HW,X86-64]
 416			Enable AMD IOMMU driver option to dump the ACPI table
 417			for AMD IOMMU. With this option enabled, AMD IOMMU
 418			driver will print ACPI tables for AMD IOMMU during
 419			IOMMU initialization.
 420
 421	amijoy.map=	[HW,JOY] Amiga joystick support
 422			Map of devices attached to JOY0DAT and JOY1DAT
 423			Format: <a>,<b>
 424			See also Documentation/input/joystick.txt
 425
 426	analog.map=	[HW,JOY] Analog joystick and gamepad support
 427			Specifies type or capabilities of an analog joystick
 428			connected to one of 16 gameports
 429			Format: <type1>,<type2>,..<type16>
 430
 431	apc=		[HW,SPARC]
 432			Power management functions (SPARCstation-4/5 + deriv.)
 433			Format: noidle
 434			Disable APC CPU standby support. SPARCstation-Fox does
 435			not play well with APC CPU idle - disable it if you have
 436			APC and your system crashes randomly.
 437
 438	apic=		[APIC,X86-32] Advanced Programmable Interrupt Controller
 439			Change the output verbosity whilst booting
 440			Format: { quiet (default) | verbose | debug }
 441			Change the amount of debugging information output
 442			when initialising the APIC and IO-APIC components.
 443
 444	autoconf=	[IPV6]
 445			See Documentation/networking/ipv6.txt.
 446
 447	show_lapic=	[APIC,X86] Advanced Programmable Interrupt Controller
 448			Limit apic dumping. The parameter defines the maximal
 449			number of local apics being dumped. Also it is possible
 450			to set it to "all" by meaning -- no limit here.
 451			Format: { 1 (default) | 2 | ... | all }.
 452			The parameter valid if only apic=debug or
 453			apic=verbose is specified.
 454			Example: apic=debug show_lapic=all
 455
 456	apm=		[APM] Advanced Power Management
 457			See header of arch/x86/kernel/apm_32.c.
 458
 459	arcrimi=	[HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
 460			Format: <io>,<irq>,<nodeID>
 461
 462	ataflop=	[HW,M68k]
 463
 464	atarimouse=	[HW,MOUSE] Atari Mouse
 465
 466	atkbd.extra=	[HW] Enable extra LEDs and keys on IBM RapidAccess,
 467			EzKey and similar keyboards
 468
 469	atkbd.reset=	[HW] Reset keyboard during initialization
 470
 471	atkbd.set=	[HW] Select keyboard code set
 472			Format: <int> (2 = AT (default), 3 = PS/2)
 473
 474	atkbd.scroll=	[HW] Enable scroll wheel on MS Office and similar
 475			keyboards
 476
 477	atkbd.softraw=	[HW] Choose between synthetic and real raw mode
 478			Format: <bool> (0 = real, 1 = synthetic (default))
 479
 480	atkbd.softrepeat= [HW]
 481			Use software keyboard repeat
 482
 483	audit=		[KNL] Enable the audit sub-system
 484			Format: { "0" | "1" } (0 = disabled, 1 = enabled)
 485			0 - kernel audit is disabled and can not be enabled
 486			    until the next reboot
 487			unset - kernel audit is initialized but disabled and
 488			    will be fully enabled by the userspace auditd.
 489			1 - kernel audit is initialized and partially enabled,
 490			    storing at most audit_backlog_limit messages in
 491			    RAM until it is fully enabled by the userspace
 492			    auditd.
 493			Default: unset
 494
 495	audit_backlog_limit= [KNL] Set the audit queue size limit.
 496			Format: <int> (must be >=0)
 497			Default: 64
 498
 499	baycom_epp=	[HW,AX25]
 500			Format: <io>,<mode>
 501
 502	baycom_par=	[HW,AX25] BayCom Parallel Port AX.25 Modem
 503			Format: <io>,<mode>
 504			See header of drivers/net/hamradio/baycom_par.c.
 505
 506	baycom_ser_fdx=	[HW,AX25]
 507			BayCom Serial Port AX.25 Modem (Full Duplex Mode)
 508			Format: <io>,<irq>,<mode>[,<baud>]
 509			See header of drivers/net/hamradio/baycom_ser_fdx.c.
 510
 511	baycom_ser_hdx=	[HW,AX25]
 512			BayCom Serial Port AX.25 Modem (Half Duplex Mode)
 513			Format: <io>,<irq>,<mode>
 514			See header of drivers/net/hamradio/baycom_ser_hdx.c.
 515
 516	blkdevparts=	Manual partition parsing of block device(s) for
 517			embedded devices based on command line input.
 518			See Documentation/block/cmdline-partition.txt
 519
 520	boot_delay=	Milliseconds to delay each printk during boot.
 521			Values larger than 10 seconds (10000) are changed to
 522			no delay (0).
 523			Format: integer
 524
 525	bootmem_debug	[KNL] Enable bootmem allocator debug messages.
 526
 527	bttv.card=	[HW,V4L] bttv (bt848 + bt878 based grabber cards)
 528	bttv.radio=	Most important insmod options are available as
 529			kernel args too.
 530	bttv.pll=	See Documentation/video4linux/bttv/Insmod-options
 531	bttv.tuner=
 532
 533	bulk_remove=off	[PPC]  This parameter disables the use of the pSeries
 534			firmware feature for flushing multiple hpte entries
 535			at a time.
 536
 537	c101=		[NET] Moxa C101 synchronous serial card
 538
 539	cachesize=	[BUGS=X86-32] Override level 2 CPU cache size detection.
 540			Sometimes CPU hardware bugs make them report the cache
 541			size incorrectly. The kernel will attempt work arounds
 542			to fix known problems, but for some CPUs it is not
 543			possible to determine what the correct size should be.
 544			This option provides an override for these situations.
 545
 
 
 
 
 
 
 546	ccw_timeout_log [S390]
 547			See Documentation/s390/CommonIO for details.
 548
 549	cgroup_disable= [KNL] Disable a particular controller
 550			Format: {name of the controller(s) to disable}
 551			The effects of cgroup_disable=foo are:
 552			- foo isn't auto-mounted if you mount all cgroups in
 553			  a single hierarchy
 554			- foo isn't visible as an individually mountable
 555			  subsystem
 556			{Currently only "memory" controller deal with this and
 557			cut the overhead, others just disable the usage. So
 558			only cgroup_disable=memory is actually worthy}
 559
 560	checkreqprot	[SELINUX] Set initial checkreqprot flag value.
 561			Format: { "0" | "1" }
 562			See security/selinux/Kconfig help text.
 563			0 -- check protection applied by kernel (includes
 564				any implied execute protection).
 565			1 -- check protection requested by application.
 566			Default value is set via a kernel config option.
 567			Value can be changed at runtime via
 568				/selinux/checkreqprot.
 569
 570	cio_ignore=	[S390]
 571			See Documentation/s390/CommonIO for details.
 572	clk_ignore_unused
 573			[CLK]
 574			Keep all clocks already enabled by bootloader on,
 575			even if no driver has claimed them. This is useful
 576			for debug and development, but should not be
 577			needed on a platform with proper driver support.
 578			For more information, see Documentation/clk.txt.
 579
 580	clock=		[BUGS=X86-32, HW] gettimeofday clocksource override.
 581			[Deprecated]
 582			Forces specified clocksource (if available) to be used
 583			when calculating gettimeofday(). If specified
 584			clocksource is not available, it defaults to PIT.
 585			Format: { pit | tsc | cyclone | pmtmr }
 586
 587	clocksource=	Override the default clocksource
 588			Format: <string>
 589			Override the default clocksource and use the clocksource
 590			with the name specified.
 591			Some clocksource names to choose from, depending on
 592			the platform:
 593			[all] jiffies (this is the base, fallback clocksource)
 594			[ACPI] acpi_pm
 595			[ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
 596				pxa_timer,timer3,32k_counter,timer0_1
 597			[AVR32] avr32
 598			[X86-32] pit,hpet,tsc;
 599				scx200_hrt on Geode; cyclone on IBM x440
 600			[MIPS] MIPS
 601			[PARISC] cr16
 602			[S390] tod
 603			[SH] SuperH
 604			[SPARC64] tick
 605			[X86-64] hpet,tsc
 606
 607	clearcpuid=BITNUM [X86]
 608			Disable CPUID feature X for the kernel. See
 609			arch/x86/include/asm/cpufeature.h for the valid bit
 610			numbers. Note the Linux specific bits are not necessarily
 611			stable over kernel options, but the vendor specific
 612			ones should be.
 613			Also note that user programs calling CPUID directly
 614			or using the feature without checking anything
 615			will still see it. This just prevents it from
 616			being used by the kernel or shown in /proc/cpuinfo.
 617			Also note the kernel might malfunction if you disable
 618			some critical bits.
 619
 620	cma=nn[MG]	[ARM,KNL]
 621			Sets the size of kernel global memory area for contiguous
 622			memory allocations. For more information, see
 623			include/linux/dma-contiguous.h
 624
 625	cmo_free_hint=	[PPC] Format: { yes | no }
 626			Specify whether pages are marked as being inactive
 627			when they are freed.  This is used in CMO environments
 628			to determine OS memory pressure for page stealing by
 629			a hypervisor.
 630			Default: yes
 631
 632	coherent_pool=nn[KMG]	[ARM,KNL]
 633			Sets the size of memory pool for coherent, atomic dma
 634			allocations, by default set to 256K.
 635
 636	code_bytes	[X86] How many bytes of object code to print
 637			in an oops report.
 638			Range: 0 - 8192
 639			Default: 64
 640
 641	com20020=	[HW,NET] ARCnet - COM20020 chipset
 642			Format:
 643			<io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
 644
 645	com90io=	[HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
 646			Format: <io>[,<irq>]
 647
 648	com90xx=	[HW,NET]
 649			ARCnet - COM90xx chipset (memory-mapped buffers)
 650			Format: <io>[,<irq>[,<memstart>]]
 651
 652	condev=		[HW,S390] console device
 653	conmode=
 654
 655	console=	[KNL] Output console device and options.
 656
 657		tty<n>	Use the virtual console device <n>.
 658
 659		ttyS<n>[,options]
 660		ttyUSB0[,options]
 661			Use the specified serial port.  The options are of
 662			the form "bbbbpnf", where "bbbb" is the baud rate,
 663			"p" is parity ("n", "o", or "e"), "n" is number of
 664			bits, and "f" is flow control ("r" for RTS or
 665			omit it).  Default is "9600n8".
 666
 667			See Documentation/serial-console.txt for more
 668			information.  See
 669			Documentation/networking/netconsole.txt for an
 670			alternative.
 671
 672		uart[8250],io,<addr>[,options]
 673		uart[8250],mmio,<addr>[,options]
 674			Start an early, polled-mode console on the 8250/16550
 675			UART at the specified I/O port or MMIO address,
 676			switching to the matching ttyS device later.  The
 677			options are the same as for ttyS, above.
 678		hvc<n>	Use the hypervisor console device <n>. This is for
 679			both Xen and PowerPC hypervisors.
 680
 681                If the device connected to the port is not a TTY but a braille
 682                device, prepend "brl," before the device type, for instance
 683			console=brl,ttyS0
 684		For now, only VisioBraille is supported.
 685
 686	consoleblank=	[KNL] The console blank (screen saver) timeout in
 687			seconds. Defaults to 10*60 = 10mins. A value of 0
 688			disables the blank timer.
 689
 690	coredump_filter=
 691			[KNL] Change the default value for
 692			/proc/<pid>/coredump_filter.
 693			See also Documentation/filesystems/proc.txt.
 694
 695	cpuidle.off=1	[CPU_IDLE]
 696			disable the cpuidle sub-system
 697
 698	cpcihp_generic=	[HW,PCI] Generic port I/O CompactPCI driver
 699			Format:
 700			<first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
 701
 702	crashkernel=size[KMG][@offset[KMG]]
 703			[KNL] Using kexec, Linux can switch to a 'crash kernel'
 704			upon panic. This parameter reserves the physical
 705			memory region [offset, offset + size] for that kernel
 706			image. If '@offset' is omitted, then a suitable offset
 707			is selected automatically. Check
 708			Documentation/kdump/kdump.txt for further details.
 709
 710	crashkernel=range1:size1[,range2:size2,...][@offset]
 711			[KNL] Same as above, but depends on the memory
 712			in the running system. The syntax of range is
 713			start-[end] where start and end are both
 714			a memory unit (amount[KMG]). See also
 715			Documentation/kdump/kdump.txt for an example.
 716
 717	crashkernel=size[KMG],high
 718			[KNL, x86_64] range could be above 4G. Allow kernel
 719			to allocate physical memory region from top, so could
 720			be above 4G if system have more than 4G ram installed.
 721			Otherwise memory region will be allocated below 4G, if
 722			available.
 723			It will be ignored if crashkernel=X is specified.
 724	crashkernel=size[KMG],low
 725			[KNL, x86_64] range under 4G. When crashkernel=X,high
 726			is passed, kernel could allocate physical memory region
 727			above 4G, that cause second kernel crash on system
 728			that require some amount of low memory, e.g. swiotlb
 729			requires at least 64M+32K low memory.  Kernel would
 730			try to allocate 72M below 4G automatically.
 731			This one let user to specify own low range under 4G
 732			for second kernel instead.
 733			0: to disable low allocation.
 734			It will be ignored when crashkernel=X,high is not used
 735			or memory reserved is below 4G.
 736
 737	cs89x0_dma=	[HW,NET]
 738			Format: <dma>
 739
 740	cs89x0_media=	[HW,NET]
 741			Format: { rj45 | aui | bnc }
 742
 743	dasd=		[HW,NET]
 744			See header of drivers/s390/block/dasd_devmap.c.
 745
 746	db9.dev[2|3]=	[HW,JOY] Multisystem joystick support via parallel port
 747			(one device per port)
 748			Format: <port#>,<type>
 749			See also Documentation/input/joystick-parport.txt
 750
 751	ddebug_query=   [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
 752			time. See Documentation/dynamic-debug-howto.txt for
 753			details.  Deprecated, see dyndbg.
 754
 755	debug		[KNL] Enable kernel debugging (events log level).
 756
 757	debug_locks_verbose=
 758			[KNL] verbose self-tests
 759			Format=<0|1>
 760			Print debugging info while doing the locking API
 761			self-tests.
 762			We default to 0 (no extra messages), setting it to
 763			1 will print _a lot_ more information - normally
 764			only useful to kernel developers.
 765
 766	debug_objects	[KNL] Enable object debugging
 767
 768	no_debug_objects
 769			[KNL] Disable object debugging
 770
 771	debug_guardpage_minorder=
 772			[KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
 773			parameter allows control of the order of pages that will
 774			be intentionally kept free (and hence protected) by the
 775			buddy allocator. Bigger value increase the probability
 776			of catching random memory corruption, but reduce the
 777			amount of memory for normal system use. The maximum
 778			possible value is MAX_ORDER/2.  Setting this parameter
 779			to 1 or 2 should be enough to identify most random
 780			memory corruption problems caused by bugs in kernel or
 781			driver code when a CPU writes to (or reads from) a
 782			random memory location. Note that there exists a class
 783			of memory corruptions problems caused by buggy H/W or
 784			F/W or by drivers badly programing DMA (basically when
 785			memory is written at bus level and the CPU MMU is
 786			bypassed) which are not detectable by
 787			CONFIG_DEBUG_PAGEALLOC, hence this option will not help
 788			tracking down these problems.
 789
 790	debugpat	[X86] Enable PAT debugging
 791
 792	decnet.addr=	[HW,NET]
 793			Format: <area>[,<node>]
 794			See also Documentation/networking/decnet.txt.
 795
 796	default_hugepagesz=
 797			[same as hugepagesz=] The size of the default
 798			HugeTLB page size. This is the size represented by
 799			the legacy /proc/ hugepages APIs, used for SHM, and
 800			default size when mounting hugetlbfs filesystems.
 801			Defaults to the default architecture's huge page size
 802			if not specified.
 803
 804	dhash_entries=	[KNL]
 805			Set number of hash buckets for dentry cache.
 806
 
 
 
 
 
 
 
 807	disable=	[IPV6]
 808			See Documentation/networking/ipv6.txt.
 809
 810	disable_cpu_apicid= [X86,APIC,SMP]
 811			Format: <int>
 812			The number of initial APIC ID for the
 813			corresponding CPU to be disabled at boot,
 814			mostly used for the kdump 2nd kernel to
 815			disable BSP to wake up multiple CPUs without
 816			causing system reset or hang due to sending
 817			INIT from AP to BSP.
 818
 819	disable_ddw     [PPC/PSERIES]
 820			Disable Dynamic DMA Window support. Use this if
 821			to workaround buggy firmware.
 822
 823	disable_ipv6=	[IPV6]
 824			See Documentation/networking/ipv6.txt.
 825
 826	disable_mtrr_cleanup [X86]
 827			The kernel tries to adjust MTRR layout from continuous
 828			to discrete, to make X server driver able to add WB
 829			entry later. This parameter disables that.
 830
 831	disable_mtrr_trim [X86, Intel and AMD only]
 832			By default the kernel will trim any uncacheable
 833			memory out of your available memory pool based on
 834			MTRR settings.  This parameter disables that behavior,
 835			possibly causing your machine to run very slowly.
 836
 837	disable_timer_pin_1 [X86]
 838			Disable PIN 1 of APIC timer
 839			Can be useful to work around chipset bugs.
 840
 841	dma_debug=off	If the kernel is compiled with DMA_API_DEBUG support,
 842			this option disables the debugging code at boot.
 843
 844	dma_debug_entries=<number>
 845			This option allows to tune the number of preallocated
 846			entries for DMA-API debugging code. One entry is
 847			required per DMA-API allocation. Use this if the
 848			DMA-API debugging code disables itself because the
 849			architectural default is too low.
 850
 851	dma_debug_driver=<driver_name>
 852			With this option the DMA-API debugging driver
 853			filter feature can be enabled at boot time. Just
 854			pass the driver to filter for as the parameter.
 855			The filter can be disabled or changed to another
 856			driver later using sysfs.
 857
 858	drm_kms_helper.edid_firmware=[<connector>:]<file>
 859			Broken monitors, graphic adapters and KVMs may
 860			send no or incorrect EDID data sets. This parameter
 861			allows to specify an EDID data set in the
 862			/lib/firmware directory that is used instead.
 863			Generic built-in EDID data sets are used, if one of
 864			edid/1024x768.bin, edid/1280x1024.bin,
 865			edid/1680x1050.bin, or edid/1920x1080.bin is given
 866			and no file with the same name exists. Details and
 867			instructions how to build your own EDID data are
 868			available in Documentation/EDID/HOWTO.txt. An EDID
 869			data set will only be used for a particular connector,
 870			if its name and a colon are prepended to the EDID
 871			name.
 872
 873	dscc4.setup=	[NET]
 874
 875	dyndbg[="val"]		[KNL,DYNAMIC_DEBUG]
 876	module.dyndbg[="val"]
 877			Enable debug messages at boot time.  See
 878			Documentation/dynamic-debug-howto.txt for details.
 879
 880	early_ioremap_debug [KNL]
 881			Enable debug messages in early_ioremap support. This
 882			is useful for tracking down temporary early mappings
 883			which are not unmapped.
 884
 885	earlycon=	[KNL] Output early console device and options.
 886		uart[8250],io,<addr>[,options]
 887		uart[8250],mmio,<addr>[,options]
 888		uart[8250],mmio32,<addr>[,options]
 889			Start an early, polled-mode console on the 8250/16550
 890			UART at the specified I/O port or MMIO address.
 891			MMIO inter-register address stride is either 8-bit
 892			(mmio) or 32-bit (mmio32).
 893			The options are the same as for ttyS, above.
 894
 895	earlyprintk=	[X86,SH,BLACKFIN,ARM]
 896			earlyprintk=vga
 897			earlyprintk=efi
 898			earlyprintk=xen
 899			earlyprintk=serial[,ttySn[,baudrate]]
 900			earlyprintk=serial[,0x...[,baudrate]]
 901			earlyprintk=ttySn[,baudrate]
 902			earlyprintk=dbgp[debugController#]
 903
 904			earlyprintk is useful when the kernel crashes before
 905			the normal console is initialized. It is not enabled by
 906			default because it has some cosmetic problems.
 907
 908			Append ",keep" to not disable it when the real console
 909			takes over.
 910
 911			Only one of vga, efi, serial, or usb debug port can
 912			be used at a time.
 913
 914			Currently only ttyS0 and ttyS1 may be specified by
 915			name.  Other I/O ports may be explicitly specified
 916			on some architectures (x86 and arm at least) by
 917			replacing ttySn with an I/O port address, like this:
 918				earlyprintk=serial,0x1008,115200
 919			You can find the port for a given device in
 920			/proc/tty/driver/serial:
 921				2: uart:ST16650V2 port:00001008 irq:18 ...
 922
 923			Interaction with the standard serial driver is not
 924			very good.
 925
 926			The VGA and EFI output is eventually overwritten by
 927			the real console.
 928
 929			The xen output can only be used by Xen PV guests.
 930
 931	edac_report=	[HW,EDAC] Control how to report EDAC event
 932			Format: {"on" | "off" | "force"}
 933			on: enable EDAC to report H/W event. May be overridden
 934			by other higher priority error reporting module.
 935			off: disable H/W event reporting through EDAC.
 936			force: enforce the use of EDAC to report H/W event.
 937			default: on.
 938
 939	ekgdboc=	[X86,KGDB] Allow early kernel console debugging
 940			ekgdboc=kbd
 941
 942			This is designed to be used in conjunction with
 943			the boot argument: earlyprintk=vga
 944
 945	edd=		[EDD]
 946			Format: {"off" | "on" | "skip[mbr]"}
 947
 948	efi=		[EFI]
 949			Format: { "old_map" }
 950			old_map [X86-64]: switch to the old ioremap-based EFI
 951			runtime services mapping. 32-bit still uses this one by
 952			default.
 953
 954	efi_no_storage_paranoia [EFI; X86]
 955			Using this parameter you can use more than 50% of
 956			your efi variable storage. Use this parameter only if
 957			you are really sure that your UEFI does sane gc and
 958			fulfills the spec otherwise your board may brick.
 959
 960	eisa_irq_edge=	[PARISC,HW]
 961			See header of drivers/parisc/eisa.c.
 962
 963	elanfreq=	[X86-32]
 964			See comment before function elanfreq_setup() in
 965			arch/x86/kernel/cpu/cpufreq/elanfreq.c.
 966
 967	elevator=	[IOSCHED]
 968			Format: {"cfq" | "deadline" | "noop"}
 969			See Documentation/block/cfq-iosched.txt and
 970			Documentation/block/deadline-iosched.txt for details.
 971
 972	elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
 973			Specifies physical address of start of kernel core
 974			image elf header and optionally the size. Generally
 975			kexec loader will pass this option to capture kernel.
 976			See Documentation/kdump/kdump.txt for details.
 977
 978	enable_mtrr_cleanup [X86]
 979			The kernel tries to adjust MTRR layout from continuous
 980			to discrete, to make X server driver able to add WB
 981			entry later. This parameter enables that.
 982
 983	enable_timer_pin_1 [X86]
 984			Enable PIN 1 of APIC timer
 985			Can be useful to work around chipset bugs
 986			(in particular on some ATI chipsets).
 987			The kernel tries to set a reasonable default.
 988
 989	enforcing	[SELINUX] Set initial enforcing status.
 990			Format: {"0" | "1"}
 991			See security/selinux/Kconfig help text.
 992			0 -- permissive (log only, no denials).
 993			1 -- enforcing (deny and log).
 994			Default value is 0.
 995			Value can be changed at runtime via /selinux/enforce.
 996
 997	erst_disable	[ACPI]
 998			Disable Error Record Serialization Table (ERST)
 999			support.
1000
1001	ether=		[HW,NET] Ethernet cards parameters
1002			This option is obsoleted by the "netdev=" option, which
1003			has equivalent usage. See its documentation for details.
1004
1005	evm=		[EVM]
1006			Format: { "fix" }
1007			Permit 'security.evm' to be updated regardless of
1008			current integrity status.
1009
1010	failslab=
1011	fail_page_alloc=
1012	fail_make_request=[KNL]
1013			General fault injection mechanism.
1014			Format: <interval>,<probability>,<space>,<times>
1015			See also Documentation/fault-injection/.
1016
1017	floppy=		[HW]
1018			See Documentation/blockdev/floppy.txt.
1019
1020	force_pal_cache_flush
1021			[IA-64] Avoid check_sal_cache_flush which may hang on
1022			buggy SAL_CACHE_FLUSH implementations. Using this
1023			parameter will force ia64_sal_cache_flush to call
1024			ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1025
1026	forcepae [X86-32]
1027			Forcefully enable Physical Address Extension (PAE).
1028			Many Pentium M systems disable PAE but may have a
1029			functionally usable PAE implementation.
1030			Warning: use of this parameter will taint the kernel
1031			and may cause unknown problems.
1032
1033	ftrace=[tracer]
1034			[FTRACE] will set and start the specified tracer
1035			as early as possible in order to facilitate early
1036			boot debugging.
1037
1038	ftrace_dump_on_oops[=orig_cpu]
1039			[FTRACE] will dump the trace buffers on oops.
1040			If no parameter is passed, ftrace will dump
1041			buffers of all CPUs, but if you pass orig_cpu, it will
1042			dump only the buffer of the CPU that triggered the
1043			oops.
1044
1045	ftrace_filter=[function-list]
1046			[FTRACE] Limit the functions traced by the function
1047			tracer at boot up. function-list is a comma separated
1048			list of functions. This list can be changed at run
1049			time by the set_ftrace_filter file in the debugfs
1050			tracing directory.
1051
1052	ftrace_notrace=[function-list]
1053			[FTRACE] Do not trace the functions specified in
1054			function-list. This list can be changed at run time
1055			by the set_ftrace_notrace file in the debugfs
1056			tracing directory.
1057
1058	ftrace_graph_filter=[function-list]
1059			[FTRACE] Limit the top level callers functions traced
1060			by the function graph tracer at boot up.
1061			function-list is a comma separated list of functions
1062			that can be changed at run time by the
1063			set_graph_function file in the debugfs tracing directory.
1064
1065	gamecon.map[2|3]=
1066			[HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1067			support via parallel port (up to 5 devices per port)
1068			Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1069			See also Documentation/input/joystick-parport.txt
1070
1071	gamma=		[HW,DRM]
1072
1073	gart_fix_e820=  [X86_64] disable the fix e820 for K8 GART
1074			Format: off | on
1075			default: on
1076
1077	gcov_persist=	[GCOV] When non-zero (default), profiling data for
1078			kernel modules is saved and remains accessible via
1079			debugfs, even when the module is unloaded/reloaded.
1080			When zero, profiling data is discarded and associated
1081			debugfs files are removed at module unload time.
1082
1083	gpt		[EFI] Forces disk with valid GPT signature but
1084			invalid Protective MBR to be treated as GPT. If the
1085			primary GPT is corrupted, it enables the backup/alternate
1086			GPT to be used instead.
1087
1088	grcan.enable0=	[HW] Configuration of physical interface 0. Determines
1089			the "Enable 0" bit of the configuration register.
1090			Format: 0 | 1
1091			Default: 0
1092	grcan.enable1=	[HW] Configuration of physical interface 1. Determines
1093			the "Enable 0" bit of the configuration register.
1094			Format: 0 | 1
1095			Default: 0
1096	grcan.select=	[HW] Select which physical interface to use.
1097			Format: 0 | 1
1098			Default: 0
1099	grcan.txsize=	[HW] Sets the size of the tx buffer.
1100			Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1101			Default: 1024
1102	grcan.rxsize=	[HW] Sets the size of the rx buffer.
1103			Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1104			Default: 1024
1105
1106	hashdist=	[KNL,NUMA] Large hashes allocated during boot
1107			are distributed across NUMA nodes.  Defaults on
1108			for 64-bit NUMA, off otherwise.
1109			Format: 0 | 1 (for off | on)
1110
1111	hcl=		[IA-64] SGI's Hardware Graph compatibility layer
1112
1113	hd=		[EIDE] (E)IDE hard drive subsystem geometry
1114			Format: <cyl>,<head>,<sect>
1115
1116	hest_disable	[ACPI]
1117			Disable Hardware Error Source Table (HEST) support;
1118			corresponding firmware-first mode error processing
1119			logic will be disabled.
1120
1121	highmem=nn[KMG]	[KNL,BOOT] forces the highmem zone to have an exact
1122			size of <nn>. This works even on boxes that have no
1123			highmem otherwise. This also works to reduce highmem
1124			size on bigger boxes.
1125
1126	highres=	[KNL] Enable/disable high resolution timer mode.
1127			Valid parameters: "on", "off"
1128			Default: "on"
1129
1130	hisax=		[HW,ISDN]
1131			See Documentation/isdn/README.HiSax.
1132
1133	hlt		[BUGS=ARM,SH]
1134
1135	hpet=		[X86-32,HPET] option to control HPET usage
1136			Format: { enable (default) | disable | force |
1137				verbose }
1138			disable: disable HPET and use PIT instead
1139			force: allow force enabled of undocumented chips (ICH4,
1140				VIA, nVidia)
1141			verbose: show contents of HPET registers during setup
1142
1143	hpet_mmap=	[X86, HPET_MMAP] Allow userspace to mmap HPET
1144			registers.  Default set by CONFIG_HPET_MMAP_DEFAULT.
1145
1146	hugepages=	[HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1147	hugepagesz=	[HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
1148			On x86-64 and powerpc, this option can be specified
1149			multiple times interleaved with hugepages= to reserve
1150			huge pages of different sizes. Valid pages sizes on
1151			x86-64 are 2M (when the CPU supports "pse") and 1G
1152			(when the CPU supports the "pdpe1gb" cpuinfo flag)
1153			Note that 1GB pages can only be allocated at boot time
1154			using hugepages= and not freed afterwards.
1155
1156	hvc_iucv=	[S390] Number of z/VM IUCV hypervisor console (HVC)
1157			       terminal devices. Valid values: 0..8
1158	hvc_iucv_allow=	[S390] Comma-separated list of z/VM user IDs.
1159			       If specified, z/VM IUCV HVC accepts connections
1160			       from listed z/VM user IDs only.
1161
1162	hwthread_map=	[METAG] Comma-separated list of Linux cpu id to
1163			        hardware thread id mappings.
1164				Format: <cpu>:<hwthread>
1165
1166	keep_bootcon	[KNL]
1167			Do not unregister boot console at start. This is only
1168			useful for debugging when something happens in the window
1169			between unregistering the boot console and initializing
1170			the real console.
1171
1172	i2c_bus=	[HW] Override the default board specific I2C bus speed
1173			     or register an additional I2C bus that is not
1174			     registered from board initialization code.
1175			     Format:
1176			     <bus_id>,<clkrate>
1177
1178	i8042.debug	[HW] Toggle i8042 debug mode
1179	i8042.direct	[HW] Put keyboard port into non-translated mode
1180	i8042.dumbkbd	[HW] Pretend that controller can only read data from
1181			     keyboard and cannot control its state
1182			     (Don't attempt to blink the leds)
1183	i8042.noaux	[HW] Don't check for auxiliary (== mouse) port
1184	i8042.nokbd	[HW] Don't check/create keyboard port
1185	i8042.noloop	[HW] Disable the AUX Loopback command while probing
1186			     for the AUX port
1187	i8042.nomux	[HW] Don't check presence of an active multiplexing
1188			     controller
1189	i8042.nopnp	[HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1190			     controllers
1191	i8042.notimeout	[HW] Ignore timeout condition signalled by controller
1192	i8042.reset	[HW] Reset the controller during init and cleanup
1193	i8042.unlock	[HW] Unlock (ignore) the keylock
1194
1195	i810=		[HW,DRM]
1196
1197	i8k.ignore_dmi	[HW] Continue probing hardware even if DMI data
1198			indicates that the driver is running on unsupported
1199			hardware.
1200	i8k.force	[HW] Activate i8k driver even if SMM BIOS signature
1201			does not match list of supported models.
1202	i8k.power_status
1203			[HW] Report power status in /proc/i8k
1204			(disabled by default)
1205	i8k.restricted	[HW] Allow controlling fans only if SYS_ADMIN
1206			capability is set.
1207
1208	i915.invert_brightness=
1209			[DRM] Invert the sense of the variable that is used to
1210			set the brightness of the panel backlight. Normally a
1211			brightness value of 0 indicates backlight switched off,
1212			and the maximum of the brightness value sets the backlight
1213			to maximum brightness. If this parameter is set to 0
1214			(default) and the machine requires it, or this parameter
1215			is set to 1, a brightness value of 0 sets the backlight
1216			to maximum brightness, and the maximum of the brightness
1217			value switches the backlight off.
1218			-1 -- never invert brightness
1219			 0 -- machine default
1220			 1 -- force brightness inversion
1221
1222	icn=		[HW,ISDN]
1223			Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1224
1225	ide-core.nodma=	[HW] (E)IDE subsystem
1226			Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1227			.vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1228			.cdrom .chs .ignore_cable are additional options
1229			See Documentation/ide/ide.txt.
1230
1231	ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1232			Claim all unknown PCI IDE storage controllers.
1233
1234	idle=		[X86]
1235			Format: idle=poll, idle=halt, idle=nomwait
1236			Poll forces a polling idle loop that can slightly
1237			improve the performance of waking up a idle CPU, but
1238			will use a lot of power and make the system run hot.
1239			Not recommended.
 
 
 
 
 
1240			idle=halt: Halt is forced to be used for CPU idle.
1241			In such case C2/C3 won't be used again.
1242			idle=nomwait: Disable mwait for CPU C-states
1243
1244	ignore_loglevel	[KNL]
1245			Ignore loglevel setting - this will print /all/
1246			kernel messages to the console. Useful for debugging.
1247			We also add it as printk module parameter, so users
1248			could change it dynamically, usually by
1249			/sys/module/printk/parameters/ignore_loglevel.
1250
1251	ihash_entries=	[KNL]
1252			Set number of hash buckets for inode cache.
1253
1254	ima_appraise=	[IMA] appraise integrity measurements
1255			Format: { "off" | "enforce" | "fix" }
1256			default: "enforce"
1257
1258	ima_appraise_tcb [IMA]
1259			The builtin appraise policy appraises all files
1260			owned by uid=0.
1261
1262	ima_hash=	[IMA]
1263			Format: { md5 | sha1 | rmd160 | sha256 | sha384
1264				   | sha512 | ... }
1265			default: "sha1"
1266
1267			The list of supported hash algorithms is defined
1268			in crypto/hash_info.h.
1269
1270	ima_tcb		[IMA]
1271			Load a policy which meets the needs of the Trusted
1272			Computing Base.  This means IMA will measure all
1273			programs exec'd, files mmap'd for exec, and all files
1274			opened for read by uid=0.
1275
1276	ima_template=   [IMA]
1277			Select one of defined IMA measurements template formats.
1278			Formats: { "ima" | "ima-ng" }
1279			Default: "ima-ng"
1280
1281	init=		[KNL]
1282			Format: <full_path>
1283			Run specified binary instead of /sbin/init as init
1284			process.
1285
1286	initcall_debug	[KNL] Trace initcalls as they are executed.  Useful
1287			for working out where the kernel is dying during
1288			startup.
1289
1290	initrd=		[BOOT] Specify the location of the initial ramdisk
1291
1292	inport.irq=	[HW] Inport (ATI XL and Microsoft) busmouse driver
1293			Format: <irq>
1294
1295	int_pln_enable  [x86] Enable power limit notification interrupt
1296
1297	integrity_audit=[IMA]
1298			Format: { "0" | "1" }
1299			0 -- basic integrity auditing messages. (Default)
1300			1 -- additional integrity auditing messages.
1301
1302	intel_iommu=	[DMAR] Intel IOMMU driver (DMAR) option
1303		on
1304			Enable intel iommu driver.
1305		off
1306			Disable intel iommu driver.
1307		igfx_off [Default Off]
1308			By default, gfx is mapped as normal device. If a gfx
1309			device has a dedicated DMAR unit, the DMAR unit is
1310			bypassed by not enabling DMAR with this option. In
1311			this case, gfx device will use physical address for
1312			DMA.
1313		forcedac [x86_64]
1314			With this option iommu will not optimize to look
1315			for io virtual address below 32-bit forcing dual
1316			address cycle on pci bus for cards supporting greater
1317			than 32-bit addressing. The default is to look
1318			for translation below 32-bit and if not available
1319			then look in the higher range.
1320		strict [Default Off]
1321			With this option on every unmap_single operation will
1322			result in a hardware IOTLB flush operation as opposed
1323			to batching them for performance.
1324		sp_off [Default Off]
1325			By default, super page will be supported if Intel IOMMU
1326			has the capability. With this option, super page will
1327			not be supported.
1328
1329	intel_idle.max_cstate=	[KNL,HW,ACPI,X86]
1330			0	disables intel_idle and fall back on acpi_idle.
1331			1 to 6	specify maximum depth of C-state.
1332
1333	intel_pstate=  [X86]
1334		       disable
1335		         Do not enable intel_pstate as the default
1336		         scaling driver for the supported processors
1337
1338	intremap=	[X86-64, Intel-IOMMU]
 
1339			on	enable Interrupt Remapping (default)
1340			off	disable Interrupt Remapping
1341			nosid	disable Source ID checking
1342			no_x2apic_optout
1343				BIOS x2APIC opt-out request will be ignored
1344
1345	iomem=		Disable strict checking of access to MMIO memory
1346		strict	regions from userspace.
1347		relaxed
1348
1349	iommu=		[x86]
1350		off
1351		force
1352		noforce
1353		biomerge
1354		panic
1355		nopanic
1356		merge
1357		nomerge
1358		forcesac
1359		soft
1360		pt		[x86, IA-64]
1361
1362
1363	io7=		[HW] IO7 for Marvel based alpha systems
1364			See comment before marvel_specify_io7 in
1365			arch/alpha/kernel/core_marvel.c.
1366
1367	io_delay=	[X86] I/O delay method
1368		0x80
1369			Standard port 0x80 based delay
1370		0xed
1371			Alternate port 0xed based delay (needed on some systems)
1372		udelay
1373			Simple two microseconds delay
1374		none
1375			No delay
1376
1377	ip=		[IP_PNP]
1378			See Documentation/filesystems/nfs/nfsroot.txt.
1379
1380	ip2=		[HW] Set IO/IRQ pairs for up to 4 IntelliPort boards
1381			See comment before ip2_setup() in
1382			drivers/char/ip2/ip2base.c.
1383
1384	irqfixup	[HW]
1385			When an interrupt is not handled search all handlers
1386			for it. Intended to get systems with badly broken
1387			firmware running.
1388
1389	irqpoll		[HW]
1390			When an interrupt is not handled search all handlers
1391			for it. Also check all handlers each timer
1392			interrupt. Intended to get systems with badly broken
1393			firmware running.
1394
1395	isapnp=		[ISAPNP]
1396			Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1397
1398	isolcpus=	[KNL,SMP] Isolate CPUs from the general scheduler.
1399			Format:
1400			<cpu number>,...,<cpu number>
1401			or
1402			<cpu number>-<cpu number>
1403			(must be a positive range in ascending order)
1404			or a mixture
1405			<cpu number>,...,<cpu number>-<cpu number>
1406
1407			This option can be used to specify one or more CPUs
1408			to isolate from the general SMP balancing and scheduling
1409			algorithms. You can move a process onto or off an
1410			"isolated" CPU via the CPU affinity syscalls or cpuset.
1411			<cpu number> begins at 0 and the maximum value is
1412			"number of CPUs in system - 1".
1413
1414			This option is the preferred way to isolate CPUs. The
1415			alternative -- manually setting the CPU mask of all
1416			tasks in the system -- can cause problems and
1417			suboptimal load balancer performance.
1418
1419	iucv=		[HW,NET]
1420
1421	ivrs_ioapic	[HW,X86_64]
1422			Provide an override to the IOAPIC-ID<->DEVICE-ID
1423			mapping provided in the IVRS ACPI table. For
1424			example, to map IOAPIC-ID decimal 10 to
1425			PCI device 00:14.0 write the parameter as:
1426				ivrs_ioapic[10]=00:14.0
1427
1428	ivrs_hpet	[HW,X86_64]
1429			Provide an override to the HPET-ID<->DEVICE-ID
1430			mapping provided in the IVRS ACPI table. For
1431			example, to map HPET-ID decimal 0 to
1432			PCI device 00:14.0 write the parameter as:
1433				ivrs_hpet[0]=00:14.0
1434
1435	js=		[HW,JOY] Analog joystick
1436			See Documentation/input/joystick.txt.
1437
1438	keepinitrd	[HW,ARM]
1439
1440	kernelcore=nn[KMG]	[KNL,X86,IA-64,PPC] This parameter
1441			specifies the amount of memory usable by the kernel
1442			for non-movable allocations.  The requested amount is
1443			spread evenly throughout all nodes in the system. The
1444			remaining memory in each node is used for Movable
1445			pages. In the event, a node is too small to have both
1446			kernelcore and Movable pages, kernelcore pages will
1447			take priority and other nodes will have a larger number
1448			of Movable pages.  The Movable zone is used for the
1449			allocation of pages that may be reclaimed or moved
1450			by the page migration subsystem.  This means that
1451			HugeTLB pages may not be allocated from this zone.
1452			Note that allocations like PTEs-from-HighMem still
1453			use the HighMem zone if it exists, and the Normal
1454			zone if it does not.
1455
1456	kgdbdbgp=	[KGDB,HW] kgdb over EHCI usb debug port.
1457			Format: <Controller#>[,poll interval]
1458			The controller # is the number of the ehci usb debug
1459			port as it is probed via PCI.  The poll interval is
1460			optional and is the number seconds in between
1461			each poll cycle to the debug port in case you need
1462			the functionality for interrupting the kernel with
1463			gdb or control-c on the dbgp connection.  When
1464			not using this parameter you use sysrq-g to break into
1465			the kernel debugger.
1466
1467	kgdboc=		[KGDB,HW] kgdb over consoles.
1468			Requires a tty driver that supports console polling,
1469			or a supported polling keyboard driver (non-usb).
1470			 Serial only format: <serial_device>[,baud]
1471			 keyboard only format: kbd
1472			 keyboard and serial format: kbd,<serial_device>[,baud]
1473			Optional Kernel mode setting:
1474			 kms, kbd format: kms,kbd
1475			 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
1476
1477	kgdbwait	[KGDB] Stop kernel execution and enter the
1478			kernel debugger at the earliest opportunity.
1479
1480	kmac=		[MIPS] korina ethernet MAC address.
1481			Configure the RouterBoard 532 series on-chip
1482			Ethernet adapter MAC address.
1483
1484	kmemleak=	[KNL] Boot-time kmemleak enable/disable
1485			Valid arguments: on, off
1486			Default: on
1487
1488	kmemcheck=	[X86] Boot-time kmemcheck enable/disable/one-shot mode
1489			Valid arguments: 0, 1, 2
1490			kmemcheck=0 (disabled)
1491			kmemcheck=1 (enabled)
1492			kmemcheck=2 (one-shot mode)
1493			Default: 2 (one-shot mode)
1494
1495	kstack=N	[X86] Print N words from the kernel stack
1496			in oops dumps.
1497
1498	kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
1499			Default is 0 (don't ignore, but inject #GP)
1500
 
 
 
1501	kvm.mmu_audit=	[KVM] This is a R/W parameter which allows audit
1502			KVM MMU at runtime.
1503			Default is 0 (off)
1504
1505	kvm-amd.nested=	[KVM,AMD] Allow nested virtualization in KVM/SVM.
1506			Default is 1 (enabled)
1507
1508	kvm-amd.npt=	[KVM,AMD] Disable nested paging (virtualized MMU)
1509			for all guests.
1510			Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
1511
1512	kvm-intel.ept=	[KVM,Intel] Disable extended page tables
1513			(virtualized MMU) support on capable Intel chips.
1514			Default is 1 (enabled)
1515
1516	kvm-intel.emulate_invalid_guest_state=
1517			[KVM,Intel] Enable emulation of invalid guest states
1518			Default is 0 (disabled)
1519
1520	kvm-intel.flexpriority=
1521			[KVM,Intel] Disable FlexPriority feature (TPR shadow).
1522			Default is 1 (enabled)
1523
1524	kvm-intel.nested=
1525			[KVM,Intel] Enable VMX nesting (nVMX).
1526			Default is 0 (disabled)
1527
1528	kvm-intel.unrestricted_guest=
1529			[KVM,Intel] Disable unrestricted guest feature
1530			(virtualized real and unpaged mode) on capable
1531			Intel chips. Default is 1 (enabled)
1532
1533	kvm-intel.vpid=	[KVM,Intel] Disable Virtual Processor Identification
1534			feature (tagged TLBs) on capable Intel chips.
1535			Default is 1 (enabled)
1536
1537	l2cr=		[PPC]
1538
1539	l3cr=		[PPC]
1540
1541	lapic		[X86-32,APIC] Enable the local APIC even if BIOS
1542			disabled it.
1543
1544	lapic=		[x86,APIC] "notscdeadline" Do not use TSC deadline
1545			value for LAPIC timer one-shot implementation. Default
1546			back to the programmable timer unit in the LAPIC.
1547
1548	lapic_timer_c2_ok	[X86,APIC] trust the local apic timer
1549			in C2 power state.
1550
1551	libata.dma=	[LIBATA] DMA control
1552			libata.dma=0	  Disable all PATA and SATA DMA
1553			libata.dma=1	  PATA and SATA Disk DMA only
1554			libata.dma=2	  ATAPI (CDROM) DMA only
1555			libata.dma=4	  Compact Flash DMA only
1556			Combinations also work, so libata.dma=3 enables DMA
1557			for disks and CDROMs, but not CFs.
1558
1559	libata.ignore_hpa=	[LIBATA] Ignore HPA limit
1560			libata.ignore_hpa=0	  keep BIOS limits (default)
1561			libata.ignore_hpa=1	  ignore limits, using full disk
1562
1563	libata.noacpi	[LIBATA] Disables use of ACPI in libata suspend/resume
1564			when set.
1565			Format: <int>
1566
1567	libata.force=	[LIBATA] Force configurations.  The format is comma
1568			separated list of "[ID:]VAL" where ID is
1569			PORT[.DEVICE].  PORT and DEVICE are decimal numbers
1570			matching port, link or device.  Basically, it matches
1571			the ATA ID string printed on console by libata.  If
1572			the whole ID part is omitted, the last PORT and DEVICE
1573			values are used.  If ID hasn't been specified yet, the
1574			configuration applies to all ports, links and devices.
1575
1576			If only DEVICE is omitted, the parameter applies to
1577			the port and all links and devices behind it.  DEVICE
1578			number of 0 either selects the first device or the
1579			first fan-out link behind PMP device.  It does not
1580			select the host link.  DEVICE number of 15 selects the
1581			host link and device attached to it.
1582
1583			The VAL specifies the configuration to force.  As long
1584			as there's no ambiguity shortcut notation is allowed.
1585			For example, both 1.5 and 1.5G would work for 1.5Gbps.
1586			The following configurations can be forced.
1587
1588			* Cable type: 40c, 80c, short40c, unk, ign or sata.
1589			  Any ID with matching PORT is used.
1590
1591			* SATA link speed limit: 1.5Gbps or 3.0Gbps.
1592
1593			* Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
1594			  udma[/][16,25,33,44,66,100,133] notation is also
1595			  allowed.
1596
1597			* [no]ncq: Turn on or off NCQ.
1598
1599			* nohrst, nosrst, norst: suppress hard, soft
1600                          and both resets.
1601
1602			* rstonce: only attempt one reset during
1603			  hot-unplug link recovery
1604
1605			* dump_id: dump IDENTIFY data.
1606
1607			* atapi_dmadir: Enable ATAPI DMADIR bridge support
1608
1609			* disable: Disable this device.
1610
1611			If there are multiple matching configurations changing
1612			the same attribute, the last one is used.
1613
1614	memblock=debug	[KNL] Enable memblock debug messages.
1615
1616	load_ramdisk=	[RAM] List of ramdisks to load from floppy
1617			See Documentation/blockdev/ramdisk.txt.
1618
1619	lockd.nlm_grace_period=P  [NFS] Assign grace period.
1620			Format: <integer>
1621
1622	lockd.nlm_tcpport=N	[NFS] Assign TCP port.
1623			Format: <integer>
1624
1625	lockd.nlm_timeout=T	[NFS] Assign timeout value.
1626			Format: <integer>
1627
1628	lockd.nlm_udpport=M	[NFS] Assign UDP port.
1629			Format: <integer>
1630
1631	logibm.irq=	[HW,MOUSE] Logitech Bus Mouse Driver
1632			Format: <irq>
1633
1634	loglevel=	All Kernel Messages with a loglevel smaller than the
1635			console loglevel will be printed to the console. It can
1636			also be changed with klogd or other programs. The
1637			loglevels are defined as follows:
1638
1639			0 (KERN_EMERG)		system is unusable
1640			1 (KERN_ALERT)		action must be taken immediately
1641			2 (KERN_CRIT)		critical conditions
1642			3 (KERN_ERR)		error conditions
1643			4 (KERN_WARNING)	warning conditions
1644			5 (KERN_NOTICE)		normal but significant condition
1645			6 (KERN_INFO)		informational
1646			7 (KERN_DEBUG)		debug-level messages
1647
1648	log_buf_len=n[KMG]	Sets the size of the printk ring buffer,
1649			in bytes.  n must be a power of two.  The default
1650			size is set in the kernel config file.
1651
1652	logo.nologo	[FB] Disables display of the built-in Linux logo.
1653			This may be used to provide more screen space for
1654			kernel log messages and is useful when debugging
1655			kernel boot problems.
1656
1657	lp=0		[LP]	Specify parallel ports to use, e.g,
1658	lp=port[,port...]	lp=none,parport0 (lp0 not configured, lp1 uses
1659	lp=reset		first parallel port). 'lp=0' disables the
1660	lp=auto			printer driver. 'lp=reset' (which can be
1661				specified in addition to the ports) causes
1662				attached printers to be reset. Using
1663				lp=port1,port2,... specifies the parallel ports
1664				to associate lp devices with, starting with
1665				lp0. A port specification may be 'none' to skip
1666				that lp device, or a parport name such as
1667				'parport0'. Specifying 'lp=auto' instead of a
1668				port specification list means that device IDs
1669				from each port should be examined, to see if
1670				an IEEE 1284-compliant printer is attached; if
1671				so, the driver will manage that printer.
1672				See also header of drivers/char/lp.c.
1673
1674	lpj=n		[KNL]
1675			Sets loops_per_jiffy to given constant, thus avoiding
1676			time-consuming boot-time autodetection (up to 250 ms per
1677			CPU). 0 enables autodetection (default). To determine
1678			the correct value for your kernel, boot with normal
1679			autodetection and see what value is printed. Note that
1680			on SMP systems the preset will be applied to all CPUs,
1681			which is likely to cause problems if your CPUs need
1682			significantly divergent settings. An incorrect value
1683			will cause delays in the kernel to be wrong, leading to
1684			unpredictable I/O errors and other breakage. Although
1685			unlikely, in the extreme case this might damage your
1686			hardware.
1687
1688	ltpc=		[NET]
1689			Format: <io>,<irq>,<dma>
1690
1691	machvec=	[IA-64] Force the use of a particular machine-vector
1692			(machvec) in a generic kernel.
1693			Example: machvec=hpzx1_swiotlb
1694
1695	machtype=	[Loongson] Share the same kernel image file between different
1696			 yeeloong laptop.
1697			Example: machtype=lemote-yeeloong-2f-7inch
1698
1699	max_addr=nn[KMG]	[KNL,BOOT,ia64] All physical memory greater
1700			than or equal to this physical address is ignored.
1701
1702	maxcpus=	[SMP] Maximum number of processors that	an SMP kernel
1703			should make use of.  maxcpus=n : n >= 0 limits the
1704			kernel to using 'n' processors.  n=0 is a special case,
1705			it is equivalent to "nosmp", which also disables
1706			the IO APIC.
1707
1708	max_loop=	[LOOP] The number of loop block devices that get
1709	(loop.max_loop)	unconditionally pre-created at init time. The default
1710			number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
1711			of statically allocating a predefined number, loop
1712			devices can be requested on-demand with the
1713			/dev/loop-control interface.
1714
 
 
1715	mce		[X86-32] Machine Check Exception
1716
1717	mce=option	[X86-64] See Documentation/x86/x86_64/boot-options.txt
1718
1719	md=		[HW] RAID subsystems devices and level
1720			See Documentation/md.txt.
1721
1722	mdacon=		[MDA]
1723			Format: <first>,<last>
1724			Specifies range of consoles to be captured by the MDA.
1725
1726	mem=nn[KMG]	[KNL,BOOT] Force usage of a specific amount of memory
1727			Amount of memory to be used when the kernel is not able
1728			to see the whole system memory or for test.
1729			[X86] Work as limiting max address. Use together
1730			with memmap= to avoid physical address space collisions.
1731			Without memmap= PCI devices could be placed at addresses
1732			belonging to unused RAM.
1733
1734	mem=nopentium	[BUGS=X86-32] Disable usage of 4MB pages for kernel
1735			memory.
1736
1737	memchunk=nn[KMG]
1738			[KNL,SH] Allow user to override the default size for
1739			per-device physically contiguous DMA buffers.
1740
1741	memmap=exactmap	[KNL,X86] Enable setting of an exact
1742			E820 memory map, as specified by the user.
1743			Such memmap=exactmap lines can be constructed based on
1744			BIOS output or other requirements. See the memmap=nn@ss
1745			option description.
1746
1747	memmap=nn[KMG]@ss[KMG]
1748			[KNL] Force usage of a specific region of memory.
1749			Region of memory to be used is from ss to ss+nn.
1750
1751	memmap=nn[KMG]#ss[KMG]
1752			[KNL,ACPI] Mark specific memory as ACPI data.
1753			Region of memory to be marked is from ss to ss+nn.
1754
1755	memmap=nn[KMG]$ss[KMG]
1756			[KNL,ACPI] Mark specific memory as reserved.
1757			Region of memory to be reserved is from ss to ss+nn.
1758			Example: Exclude memory from 0x18690000-0x1869ffff
1759			         memmap=64K$0x18690000
1760			         or
1761			         memmap=0x10000$0x18690000
1762
1763	memory_corruption_check=0/1 [X86]
1764			Some BIOSes seem to corrupt the first 64k of
1765			memory when doing things like suspend/resume.
1766			Setting this option will scan the memory
1767			looking for corruption.  Enabling this will
1768			both detect corruption and prevent the kernel
1769			from using the memory being corrupted.
1770			However, its intended as a diagnostic tool; if
1771			repeatable BIOS-originated corruption always
1772			affects the same memory, you can use memmap=
1773			to prevent the kernel from using that memory.
1774
1775	memory_corruption_check_size=size [X86]
1776			By default it checks for corruption in the low
1777			64k, making this memory unavailable for normal
1778			use.  Use this parameter to scan for
1779			corruption in more or less memory.
1780
1781	memory_corruption_check_period=seconds [X86]
1782			By default it checks for corruption every 60
1783			seconds.  Use this parameter to check at some
1784			other rate.  0 disables periodic checking.
1785
1786	memtest=	[KNL,X86] Enable memtest
1787			Format: <integer>
1788			default : 0 <disable>
1789			Specifies the number of memtest passes to be
1790			performed. Each pass selects another test
1791			pattern from a given set of patterns. Memtest
1792			fills the memory with this pattern, validates
1793			memory contents and reserves bad memory
1794			regions that are detected.
1795
1796	meye.*=		[HW] Set MotionEye Camera parameters
1797			See Documentation/video4linux/meye.txt.
1798
1799	mfgpt_irq=	[IA-32] Specify the IRQ to use for the
1800			Multi-Function General Purpose Timers on AMD Geode
1801			platforms.
1802
1803	mfgptfix	[X86-32] Fix MFGPT timers on AMD Geode platforms when
1804			the BIOS has incorrectly applied a workaround. TinyBIOS
1805			version 0.98 is known to be affected, 0.99 fixes the
1806			problem by letting the user disable the workaround.
1807
1808	mga=		[HW,DRM]
1809
1810	min_addr=nn[KMG]	[KNL,BOOT,ia64] All physical memory below this
1811			physical address is ignored.
1812
1813	mini2440=	[ARM,HW,KNL]
1814			Format:[0..2][b][c][t]
1815			Default: "0tb"
1816			MINI2440 configuration specification:
1817			0 - The attached screen is the 3.5" TFT
1818			1 - The attached screen is the 7" TFT
1819			2 - The VGA Shield is attached (1024x768)
1820			Leaving out the screen size parameter will not load
1821			the TFT driver, and the framebuffer will be left
1822			unconfigured.
1823			b - Enable backlight. The TFT backlight pin will be
1824			linked to the kernel VESA blanking code and a GPIO
1825			LED. This parameter is not necessary when using the
1826			VGA shield.
1827			c - Enable the s3c camera interface.
1828			t - Reserved for enabling touchscreen support. The
1829			touchscreen support is not enabled in the mainstream
1830			kernel as of 2.6.30, a preliminary port can be found
1831			in the "bleeding edge" mini2440 support kernel at
1832			http://repo.or.cz/w/linux-2.6/mini2440.git
1833
1834	mminit_loglevel=
1835			[KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
1836			parameter allows control of the logging verbosity for
1837			the additional memory initialisation checks. A value
1838			of 0 disables mminit logging and a level of 4 will
1839			log everything. Information is printed at KERN_DEBUG
1840			so loglevel=8 may also need to be specified.
1841
1842	module.sig_enforce
1843			[KNL] When CONFIG_MODULE_SIG is set, this means that
1844			modules without (valid) signatures will fail to load.
1845			Note that if CONFIG_MODULE_SIG_FORCE is set, that
1846			is always true, so this option does nothing.
1847
1848	mousedev.tap_time=
1849			[MOUSE] Maximum time between finger touching and
1850			leaving touchpad surface for touch to be considered
1851			a tap and be reported as a left button click (for
1852			touchpads working in absolute mode only).
1853			Format: <msecs>
1854	mousedev.xres=	[MOUSE] Horizontal screen resolution, used for devices
1855			reporting absolute coordinates, such as tablets
1856	mousedev.yres=	[MOUSE] Vertical screen resolution, used for devices
1857			reporting absolute coordinates, such as tablets
1858
1859	movablecore=nn[KMG]	[KNL,X86,IA-64,PPC] This parameter
1860			is similar to kernelcore except it specifies the
1861			amount of memory used for migratable allocations.
1862			If both kernelcore and movablecore is specified,
1863			then kernelcore will be at *least* the specified
1864			value but may be more. If movablecore on its own
1865			is specified, the administrator must be careful
1866			that the amount of memory usable for all allocations
1867			is not too small.
1868
1869	movable_node	[KNL,X86] Boot-time switch to enable the effects
1870			of CONFIG_MOVABLE_NODE=y. See mm/Kconfig for details.
1871
1872	MTD_Partition=	[MTD]
1873			Format: <name>,<region-number>,<size>,<offset>
1874
1875	MTD_Region=	[MTD] Format:
1876			<name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
1877
1878	mtdparts=	[MTD]
1879			See drivers/mtd/cmdlinepart.c.
1880
1881	multitce=off	[PPC]  This parameter disables the use of the pSeries
1882			firmware feature for updating multiple TCE entries
1883			at a time.
1884
1885	onenand.bdry=	[HW,MTD] Flex-OneNAND Boundary Configuration
1886
1887			Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
1888
1889			boundary - index of last SLC block on Flex-OneNAND.
1890				   The remaining blocks are configured as MLC blocks.
1891			lock	 - Configure if Flex-OneNAND boundary should be locked.
1892				   Once locked, the boundary cannot be changed.
1893				   1 indicates lock status, 0 indicates unlock status.
1894
1895	mtdset=		[ARM]
1896			ARM/S3C2412 JIVE boot control
1897
1898			See arch/arm/mach-s3c2412/mach-jive.c
1899
1900	mtouchusb.raw_coordinates=
1901			[HW] Make the MicroTouch USB driver use raw coordinates
1902			('y', default) or cooked coordinates ('n')
1903
1904	mtrr_chunk_size=nn[KMG] [X86]
1905			used for mtrr cleanup. It is largest continuous chunk
1906			that could hold holes aka. UC entries.
1907
1908	mtrr_gran_size=nn[KMG] [X86]
1909			Used for mtrr cleanup. It is granularity of mtrr block.
1910			Default is 1.
1911			Large value could prevent small alignment from
1912			using up MTRRs.
1913
1914	mtrr_spare_reg_nr=n [X86]
1915			Format: <integer>
1916			Range: 0,7 : spare reg number
1917			Default : 1
1918			Used for mtrr cleanup. It is spare mtrr entries number.
1919			Set to 2 or more if your graphical card needs more.
1920
1921	n2=		[NET] SDL Inc. RISCom/N2 synchronous serial card
1922
1923	netdev=		[NET] Network devices parameters
1924			Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
1925			Note that mem_start is often overloaded to mean
1926			something different and driver-specific.
1927			This usage is only documented in each driver source
1928			file if at all.
1929
1930	nf_conntrack.acct=
1931			[NETFILTER] Enable connection tracking flow accounting
1932			0 to disable accounting
1933			1 to enable accounting
1934			Default value is 0.
1935
1936	nfsaddrs=	[NFS] Deprecated.  Use ip= instead.
1937			See Documentation/filesystems/nfs/nfsroot.txt.
1938
1939	nfsroot=	[NFS] nfs root filesystem for disk-less boxes.
1940			See Documentation/filesystems/nfs/nfsroot.txt.
1941
1942	nfsrootdebug	[NFS] enable nfsroot debugging messages.
1943			See Documentation/filesystems/nfs/nfsroot.txt.
1944
1945	nfs.callback_tcpport=
1946			[NFS] set the TCP port on which the NFSv4 callback
1947			channel should listen.
1948
1949	nfs.cache_getent=
1950			[NFS] sets the pathname to the program which is used
1951			to update the NFS client cache entries.
1952
1953	nfs.cache_getent_timeout=
1954			[NFS] sets the timeout after which an attempt to
1955			update a cache entry is deemed to have failed.
1956
1957	nfs.idmap_cache_timeout=
1958			[NFS] set the maximum lifetime for idmapper cache
1959			entries.
1960
1961	nfs.enable_ino64=
1962			[NFS] enable 64-bit inode numbers.
1963			If zero, the NFS client will fake up a 32-bit inode
1964			number for the readdir() and stat() syscalls instead
1965			of returning the full 64-bit number.
1966			The default is to return 64-bit inode numbers.
1967
1968	nfs.max_session_slots=
1969			[NFSv4.1] Sets the maximum number of session slots
1970			the client will attempt to negotiate with the server.
1971			This limits the number of simultaneous RPC requests
1972			that the client can send to the NFSv4.1 server.
1973			Note that there is little point in setting this
1974			value higher than the max_tcp_slot_table_limit.
1975
1976	nfs.nfs4_disable_idmapping=
1977			[NFSv4] When set to the default of '1', this option
1978			ensures that both the RPC level authentication
1979			scheme and the NFS level operations agree to use
1980			numeric uids/gids if the mount is using the
1981			'sec=sys' security flavour. In effect it is
1982			disabling idmapping, which can make migration from
1983			legacy NFSv2/v3 systems to NFSv4 easier.
1984			Servers that do not support this mode of operation
1985			will be autodetected by the client, and it will fall
1986			back to using the idmapper.
1987			To turn off this behaviour, set the value to '0'.
1988	nfs.nfs4_unique_id=
1989			[NFS4] Specify an additional fixed unique ident-
1990			ification string that NFSv4 clients can insert into
1991			their nfs_client_id4 string.  This is typically a
1992			UUID that is generated at system install time.
1993
1994	nfs.send_implementation_id =
1995			[NFSv4.1] Send client implementation identification
1996			information in exchange_id requests.
1997			If zero, no implementation identification information
1998			will be sent.
1999			The default is to send the implementation identification
2000			information.
2001	
2002	nfs.recover_lost_locks =
2003			[NFSv4] Attempt to recover locks that were lost due
2004			to a lease timeout on the server. Please note that
2005			doing this risks data corruption, since there are
2006			no guarantees that the file will remain unchanged
2007			after the locks are lost.
2008			If you want to enable the kernel legacy behaviour of
2009			attempting to recover these locks, then set this
2010			parameter to '1'.
2011			The default parameter value of '0' causes the kernel
2012			not to attempt recovery of lost locks.
2013
2014	nfsd.nfs4_disable_idmapping=
2015			[NFSv4] When set to the default of '1', the NFSv4
2016			server will return only numeric uids and gids to
2017			clients using auth_sys, and will accept numeric uids
2018			and gids from such clients.  This is intended to ease
2019			migration from NFSv2/v3.
2020
2021	objlayoutdriver.osd_login_prog=
2022			[NFS] [OBJLAYOUT] sets the pathname to the program which
2023			is used to automatically discover and login into new
2024			osd-targets. Please see:
2025			Documentation/filesystems/pnfs.txt for more explanations
2026
2027	nmi_debug=	[KNL,AVR32,SH] Specify one or more actions to take
2028			when a NMI is triggered.
2029			Format: [state][,regs][,debounce][,die]
2030
2031	nmi_watchdog=	[KNL,BUGS=X86] Debugging features for SMP kernels
2032			Format: [panic,][nopanic,][num]
2033			Valid num: 0
2034			0 - turn nmi_watchdog off
2035			When panic is specified, panic when an NMI watchdog
2036			timeout occurs (or 'nopanic' to override the opposite
2037			default).
2038			This is useful when you use a panic=... timeout and
2039			need the box quickly up again.
2040
2041	netpoll.carrier_timeout=
2042			[NET] Specifies amount of time (in seconds) that
2043			netpoll should wait for a carrier. By default netpoll
2044			waits 4 seconds.
2045
2046	no387		[BUGS=X86-32] Tells the kernel to use the 387 maths
2047			emulation library even if a 387 maths coprocessor
2048			is present.
2049
2050	no_console_suspend
2051			[HW] Never suspend the console
2052			Disable suspending of consoles during suspend and
2053			hibernate operations.  Once disabled, debugging
2054			messages can reach various consoles while the rest
2055			of the system is being put to sleep (ie, while
2056			debugging driver suspend/resume hooks).  This may
2057			not work reliably with all consoles, but is known
2058			to work with serial and VGA consoles.
2059			To facilitate more flexible debugging, we also add
2060			console_suspend, a printk module parameter to control
2061			it. Users could use console_suspend (usually
2062			/sys/module/printk/parameters/console_suspend) to
2063			turn on/off it dynamically.
2064
2065	noaliencache	[MM, NUMA, SLAB] Disables the allocation of alien
2066			caches in the slab allocator.  Saves per-node memory,
2067			but will impact performance.
2068
2069	noalign		[KNL,ARM]
2070
2071	noapic		[SMP,APIC] Tells the kernel to not make use of any
2072			IOAPICs that may be present in the system.
2073
2074	nokaslr		[X86]
2075			Disable kernel and module base offset ASLR (Address
2076			Space Layout Randomization) if built into the kernel.
2077
2078	noautogroup	Disable scheduler automatic task group creation.
2079
2080	nobats		[PPC] Do not use BATs for mapping kernel lowmem
2081			on "Classic" PPC cores.
2082
2083	nocache		[ARM]
2084
2085	noclflush	[BUGS=X86] Don't use the CLFLUSH instruction
2086
2087	nodelayacct	[KNL] Disable per-task delay accounting
2088
2089	nodisconnect	[HW,SCSI,M68K] Disables SCSI disconnects.
2090
2091	nodsp		[SH] Disable hardware DSP at boot time.
2092
2093	noefi		[X86] Disable EFI runtime services support.
2094
2095	noexec		[IA-64]
2096
2097	noexec		[X86]
2098			On X86-32 available only on PAE configured kernels.
2099			noexec=on: enable non-executable mappings (default)
2100			noexec=off: disable non-executable mappings
2101
2102	nosmap		[X86]
2103			Disable SMAP (Supervisor Mode Access Prevention)
2104			even if it is supported by processor.
2105
2106	nosmep		[X86]
2107			Disable SMEP (Supervisor Mode Execution Prevention)
2108			even if it is supported by processor.
2109
2110	noexec32	[X86-64]
2111			This affects only 32-bit executables.
2112			noexec32=on: enable non-executable mappings (default)
2113				read doesn't imply executable mappings
2114			noexec32=off: disable non-executable mappings
2115				read implies executable mappings
2116
2117	nofpu		[SH] Disable hardware FPU at boot time.
2118
2119	nofxsr		[BUGS=X86-32] Disables x86 floating point extended
2120			register save and restore. The kernel will only save
2121			legacy floating-point registers on task switch.
2122
2123	noxsave		[BUGS=X86] Disables x86 extended register state save
2124			and restore using xsave. The kernel will fallback to
2125			enabling legacy floating-point and sse state.
2126
2127	eagerfpu=	[X86]
2128			on	enable eager fpu restore
2129			off	disable eager fpu restore
2130			auto	selects the default scheme, which automatically
2131				enables eagerfpu restore for xsaveopt.
2132
2133	nohlt		[BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
2134			wfi(ARM) instruction doesn't work correctly and not to
2135			use it. This is also useful when using JTAG debugger.
2136
 
 
 
 
2137	no_file_caps	Tells the kernel not to honor file capabilities.  The
2138			only way then for a file to be executed with privilege
2139			is to be setuid root or executed by root.
2140
2141	nohalt		[IA-64] Tells the kernel not to use the power saving
2142			function PAL_HALT_LIGHT when idle. This increases
2143			power-consumption. On the positive side, it reduces
2144			interrupt wake-up latency, which may improve performance
2145			in certain environments such as networked servers or
2146			real-time systems.
2147
2148	nohz=		[KNL] Boottime enable/disable dynamic ticks
2149			Valid arguments: on, off
2150			Default: on
2151
2152	nohz_full=	[KNL,BOOT]
2153			In kernels built with CONFIG_NO_HZ_FULL=y, set
2154			the specified list of CPUs whose tick will be stopped
2155			whenever possible. The boot CPU will be forced outside
2156			the range to maintain the timekeeping.
2157			The CPUs in this range must also be included in the
2158			rcu_nocbs= set.
2159
2160	noiotrap	[SH] Disables trapped I/O port accesses.
2161
2162	noirqdebug	[X86-32] Disables the code which attempts to detect and
2163			disable unhandled interrupt sources.
2164
2165	no_timer_check	[X86,APIC] Disables the code which tests for
2166			broken timer IRQ sources.
2167
2168	noisapnp	[ISAPNP] Disables ISA PnP code.
2169
2170	noinitrd	[RAM] Tells the kernel not to load any configured
2171			initial RAM disk.
2172
2173	nointremap	[X86-64, Intel-IOMMU] Do not enable interrupt
2174			remapping.
2175			[Deprecated - use intremap=off]
2176
2177	nointroute	[IA-64]
2178
2179	nojitter	[IA-64] Disables jitter checking for ITC timers.
2180
2181	no-kvmclock	[X86,KVM] Disable paravirtualized KVM clock driver
2182
2183	no-kvmapf	[X86,KVM] Disable paravirtualized asynchronous page
2184			fault handling.
2185
2186	no-steal-acc    [X86,KVM] Disable paravirtualized steal time accounting.
2187			steal time is computed, but won't influence scheduler
2188			behaviour
2189
2190	nolapic		[X86-32,APIC] Do not enable or use the local APIC.
2191
2192	nolapic_timer	[X86-32,APIC] Do not use the local APIC timer.
2193
2194	noltlbs		[PPC] Do not use large page/tlb entries for kernel
2195			lowmem mapping on PPC40x.
2196
2197	nomca		[IA-64] Disable machine check abort handling
2198
2199	nomce		[X86-32] Machine Check Exception
2200
2201	nomfgpt		[X86-32] Disable Multi-Function General Purpose
2202			Timer usage (for AMD Geode machines).
2203
2204	nonmi_ipi	[X86] Disable using NMI IPIs during panic/reboot to
2205			shutdown the other cpus.  Instead use the REBOOT_VECTOR
2206			irq.
2207
2208	nomodule	Disable module load
2209
2210	nopat		[X86] Disable PAT (page attribute table extension of
2211			pagetables) support.
2212
2213	norandmaps	Don't use address space randomization.  Equivalent to
2214			echo 0 > /proc/sys/kernel/randomize_va_space
2215
2216	noreplace-paravirt	[X86,IA-64,PV_OPS] Don't patch paravirt_ops
2217
2218	noreplace-smp	[X86-32,SMP] Don't replace SMP instructions
2219			with UP alternatives
2220
2221	nordrand	[X86] Disable kernel use of the RDRAND and
2222			RDSEED instructions even if they are supported
2223			by the processor.  RDRAND and RDSEED are still
2224			available to user space applications.
2225
2226	noresume	[SWSUSP] Disables resume and restores original swap
2227			space.
2228
2229	no-scroll	[VGA] Disables scrollback.
2230			This is required for the Braillex ib80-piezo Braille
2231			reader made by F.H. Papenmeier (Germany).
2232
2233	nosbagart	[IA-64]
2234
2235	nosep		[BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
2236
2237	nosmp		[SMP] Tells an SMP kernel to act as a UP kernel,
2238			and disable the IO APIC.  legacy for "maxcpus=0".
2239
2240	nosoftlockup	[KNL] Disable the soft-lockup detector.
2241
2242	nosync		[HW,M68K] Disables sync negotiation for all devices.
2243
2244	notsc		[BUGS=X86-32] Disable Time Stamp Counter
2245
2246	nousb		[USB] Disable the USB subsystem
2247
2248	nowatchdog	[KNL] Disable the lockup detector (NMI watchdog).
2249
2250	nowb		[ARM]
2251
2252	nox2apic	[X86-64,APIC] Do not enable x2APIC mode.
2253
2254	cpu0_hotplug	[X86] Turn on CPU0 hotplug feature when
2255			CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
2256			Some features depend on CPU0. Known dependencies are:
2257			1. Resume from suspend/hibernate depends on CPU0.
2258			Suspend/hibernate will fail if CPU0 is offline and you
2259			need to online CPU0 before suspend/hibernate.
2260			2. PIC interrupts also depend on CPU0. CPU0 can't be
2261			removed if a PIC interrupt is detected.
2262			It's said poweroff/reboot may depend on CPU0 on some
2263			machines although I haven't seen such issues so far
2264			after CPU0 is offline on a few tested machines.
2265			If the dependencies are under your control, you can
2266			turn on cpu0_hotplug.
2267
2268	nptcg=		[IA-64] Override max number of concurrent global TLB
2269			purges which is reported from either PAL_VM_SUMMARY or
2270			SAL PALO.
2271
2272	nr_cpus=	[SMP] Maximum number of processors that	an SMP kernel
2273			could support.  nr_cpus=n : n >= 1 limits the kernel to
2274			supporting 'n' processors. Later in runtime you can not
2275			use hotplug cpu feature to put more cpu back to online.
2276			just like you compile the kernel NR_CPUS=n
2277
2278	nr_uarts=	[SERIAL] maximum number of UARTs to be registered.
2279
2280	numa_balancing=	[KNL,X86] Enable or disable automatic NUMA balancing.
2281			Allowed values are enable and disable
2282
2283	numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
2284			one of ['zone', 'node', 'default'] can be specified
2285			This can be set from sysctl after boot.
2286			See Documentation/sysctl/vm.txt for details.
2287
2288	ohci1394_dma=early	[HW] enable debugging via the ohci1394 driver.
2289			See Documentation/debugging-via-ohci1394.txt for more
2290			info.
2291
2292	olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
2293			Rather than timing out after 20 ms if an EC
2294			command is not properly ACKed, override the length
2295			of the timeout.  We have interrupts disabled while
2296			waiting for the ACK, so if this is set too high
2297			interrupts *may* be lost!
2298
2299	omap_mux=	[OMAP] Override bootloader pin multiplexing.
2300			Format: <mux_mode0.mode_name=value>...
2301			For example, to override I2C bus2:
2302			omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
2303
2304	oprofile.timer=	[HW]
2305			Use timer interrupt instead of performance counters
2306
2307	oprofile.cpu_type=	Force an oprofile cpu type
2308			This might be useful if you have an older oprofile
2309			userland or if you want common events.
2310			Format: { arch_perfmon }
2311			arch_perfmon: [X86] Force use of architectural
2312				perfmon on Intel CPUs instead of the
2313				CPU specific event set.
2314			timer: [X86] Force use of architectural NMI
2315				timer mode (see also oprofile.timer
2316				for generic hr timer mode)
2317				[s390] Force legacy basic mode sampling
2318                                (report cpu_type "timer")
2319
2320	oops=panic	Always panic on oopses. Default is to just kill the
2321			process, but there is a small probability of
2322			deadlocking the machine.
2323			This will also cause panics on machine check exceptions.
2324			Useful together with panic=30 to trigger a reboot.
2325
2326	OSS		[HW,OSS]
2327			See Documentation/sound/oss/oss-parameters.txt
2328
2329	panic=		[KNL] Kernel behaviour on panic: delay <timeout>
2330			timeout > 0: seconds before rebooting
2331			timeout = 0: wait forever
2332			timeout < 0: reboot immediately
2333			Format: <timeout>
2334
2335	parkbd.port=	[HW] Parallel port number the keyboard adapter is
2336			connected to, default is 0.
2337			Format: <parport#>
2338	parkbd.mode=	[HW] Parallel port keyboard adapter mode of operation,
2339			0 for XT, 1 for AT (default is AT).
2340			Format: <mode>
2341
2342	parport=	[HW,PPT] Specify parallel ports. 0 disables.
2343			Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
2344			Use 'auto' to force the driver to use any
2345			IRQ/DMA settings detected (the default is to
2346			ignore detected IRQ/DMA settings because of
2347			possible conflicts). You can specify the base
2348			address, IRQ, and DMA settings; IRQ and DMA
2349			should be numbers, or 'auto' (for using detected
2350			settings on that particular port), or 'nofifo'
2351			(to avoid using a FIFO even if it is detected).
2352			Parallel ports are assigned in the order they
2353			are specified on the command line, starting
2354			with parport0.
2355
2356	parport_init_mode=	[HW,PPT]
2357			Configure VIA parallel port to operate in
2358			a specific mode. This is necessary on Pegasos
2359			computer where firmware has no options for setting
2360			up parallel port mode and sets it to spp.
2361			Currently this function knows 686a and 8231 chips.
2362			Format: [spp|ps2|epp|ecp|ecpepp]
2363
2364	pause_on_oops=
2365			Halt all CPUs after the first oops has been printed for
2366			the specified number of seconds.  This is to be used if
2367			your oopses keep scrolling off the screen.
2368
2369	pcbit=		[HW,ISDN]
2370
2371	pcd.		[PARIDE]
2372			See header of drivers/block/paride/pcd.c.
2373			See also Documentation/blockdev/paride.txt.
2374
2375	pci=option[,option...]	[PCI] various PCI subsystem options:
2376		earlydump	[X86] dump PCI config space before the kernel
2377			        changes anything
2378		off		[X86] don't probe for the PCI bus
2379		bios		[X86-32] force use of PCI BIOS, don't access
2380				the hardware directly. Use this if your machine
2381				has a non-standard PCI host bridge.
2382		nobios		[X86-32] disallow use of PCI BIOS, only direct
2383				hardware access methods are allowed. Use this
2384				if you experience crashes upon bootup and you
2385				suspect they are caused by the BIOS.
2386		conf1		[X86] Force use of PCI Configuration
2387				Mechanism 1.
2388		conf2		[X86] Force use of PCI Configuration
2389				Mechanism 2.
2390		noaer		[PCIE] If the PCIEAER kernel config parameter is
2391				enabled, this kernel boot option can be used to
2392				disable the use of PCIE advanced error reporting.
2393		nodomains	[PCI] Disable support for multiple PCI
2394				root domains (aka PCI segments, in ACPI-speak).
2395		nommconf	[X86] Disable use of MMCONFIG for PCI
2396				Configuration
2397		check_enable_amd_mmconf [X86] check for and enable
2398				properly configured MMIO access to PCI
2399				config space on AMD family 10h CPU
2400		nomsi		[MSI] If the PCI_MSI kernel config parameter is
2401				enabled, this kernel boot option can be used to
2402				disable the use of MSI interrupts system-wide.
2403		noioapicquirk	[APIC] Disable all boot interrupt quirks.
2404				Safety option to keep boot IRQs enabled. This
2405				should never be necessary.
2406		ioapicreroute	[APIC] Enable rerouting of boot IRQs to the
2407				primary IO-APIC for bridges that cannot disable
2408				boot IRQs. This fixes a source of spurious IRQs
2409				when the system masks IRQs.
2410		noioapicreroute	[APIC] Disable workaround that uses the
2411				boot IRQ equivalent of an IRQ that connects to
2412				a chipset where boot IRQs cannot be disabled.
2413				The opposite of ioapicreroute.
2414		biosirq		[X86-32] Use PCI BIOS calls to get the interrupt
2415				routing table. These calls are known to be buggy
2416				on several machines and they hang the machine
2417				when used, but on other computers it's the only
2418				way to get the interrupt routing table. Try
2419				this option if the kernel is unable to allocate
2420				IRQs or discover secondary PCI buses on your
2421				motherboard.
2422		rom		[X86] Assign address space to expansion ROMs.
2423				Use with caution as certain devices share
2424				address decoders between ROMs and other
2425				resources.
2426		norom		[X86] Do not assign address space to
2427				expansion ROMs that do not already have
2428				BIOS assigned address ranges.
2429		nobar		[X86] Do not assign address space to the
2430				BARs that weren't assigned by the BIOS.
2431		irqmask=0xMMMM	[X86] Set a bit mask of IRQs allowed to be
2432				assigned automatically to PCI devices. You can
2433				make the kernel exclude IRQs of your ISA cards
2434				this way.
2435		pirqaddr=0xAAAAA	[X86] Specify the physical address
2436				of the PIRQ table (normally generated
2437				by the BIOS) if it is outside the
2438				F0000h-100000h range.
2439		lastbus=N	[X86] Scan all buses thru bus #N. Can be
2440				useful if the kernel is unable to find your
2441				secondary buses and you want to tell it
2442				explicitly which ones they are.
2443		assign-busses	[X86] Always assign all PCI bus
2444				numbers ourselves, overriding
2445				whatever the firmware may have done.
2446		usepirqmask	[X86] Honor the possible IRQ mask stored
2447				in the BIOS $PIR table. This is needed on
2448				some systems with broken BIOSes, notably
2449				some HP Pavilion N5400 and Omnibook XE3
2450				notebooks. This will have no effect if ACPI
2451				IRQ routing is enabled.
2452		noacpi		[X86] Do not use ACPI for IRQ routing
2453				or for PCI scanning.
2454		use_crs		[X86] Use PCI host bridge window information
2455				from ACPI.  On BIOSes from 2008 or later, this
2456				is enabled by default.  If you need to use this,
2457				please report a bug.
2458		nocrs		[X86] Ignore PCI host bridge windows from ACPI.
2459			        If you need to use this, please report a bug.
2460		routeirq	Do IRQ routing for all PCI devices.
2461				This is normally done in pci_enable_device(),
2462				so this option is a temporary workaround
2463				for broken drivers that don't call it.
2464		skip_isa_align	[X86] do not align io start addr, so can
2465				handle more pci cards
2466		firmware	[ARM] Do not re-enumerate the bus but instead
2467				just use the configuration from the
2468				bootloader. This is currently used on
2469				IXP2000 systems where the bus has to be
2470				configured a certain way for adjunct CPUs.
2471		noearly		[X86] Don't do any early type 1 scanning.
2472				This might help on some broken boards which
2473				machine check when some devices' config space
2474				is read. But various workarounds are disabled
2475				and some IOMMU drivers will not work.
2476		bfsort		Sort PCI devices into breadth-first order.
2477				This sorting is done to get a device
2478				order compatible with older (<= 2.4) kernels.
2479		nobfsort	Don't sort PCI devices into breadth-first order.
2480		pcie_bus_tune_off	Disable PCIe MPS (Max Payload Size)
2481				tuning and use the BIOS-configured MPS defaults.
2482		pcie_bus_safe	Set every device's MPS to the largest value
2483				supported by all devices below the root complex.
2484		pcie_bus_perf	Set device MPS to the largest allowable MPS
2485				based on its parent bus. Also set MRRS (Max
2486				Read Request Size) to the largest supported
2487				value (no larger than the MPS that the device
2488				or bus can support) for best performance.
2489		pcie_bus_peer2peer	Set every device's MPS to 128B, which
2490				every device is guaranteed to support. This
2491				configuration allows peer-to-peer DMA between
2492				any pair of devices, possibly at the cost of
2493				reduced performance.  This also guarantees
2494				that hot-added devices will work.
2495		cbiosize=nn[KMG]	The fixed amount of bus space which is
2496				reserved for the CardBus bridge's IO window.
2497				The default value is 256 bytes.
2498		cbmemsize=nn[KMG]	The fixed amount of bus space which is
2499				reserved for the CardBus bridge's memory
2500				window. The default value is 64 megabytes.
2501		resource_alignment=
2502				Format:
2503				[<order of align>@][<domain>:]<bus>:<slot>.<func>[; ...]
2504				Specifies alignment and device to reassign
2505				aligned memory resources.
2506				If <order of align> is not specified,
2507				PAGE_SIZE is used as alignment.
2508				PCI-PCI bridge can be specified, if resource
2509				windows need to be expanded.
2510		ecrc=		Enable/disable PCIe ECRC (transaction layer
2511				end-to-end CRC checking).
2512				bios: Use BIOS/firmware settings. This is the
2513				the default.
2514				off: Turn ECRC off
2515				on: Turn ECRC on.
2516		hpiosize=nn[KMG]	The fixed amount of bus space which is
2517				reserved for hotplug bridge's IO window.
2518				Default size is 256 bytes.
2519		hpmemsize=nn[KMG]	The fixed amount of bus space which is
2520				reserved for hotplug bridge's memory window.
2521				Default size is 2 megabytes.
2522		realloc=	Enable/disable reallocating PCI bridge resources
2523				if allocations done by BIOS are too small to
2524				accommodate resources required by all child
2525				devices.
2526				off: Turn realloc off
2527				on: Turn realloc on
2528		realloc		same as realloc=on
2529		noari		do not use PCIe ARI.
2530		pcie_scan_all	Scan all possible PCIe devices.  Otherwise we
2531				only look for one device below a PCIe downstream
2532				port.
2533
2534	pcie_aspm=	[PCIE] Forcibly enable or disable PCIe Active State Power
2535			Management.
2536		off	Disable ASPM.
2537		force	Enable ASPM even on devices that claim not to support it.
2538			WARNING: Forcing ASPM on may cause system lockups.
2539
2540	pcie_hp=	[PCIE] PCI Express Hotplug driver options:
2541		nomsi	Do not use MSI for PCI Express Native Hotplug (this
2542			makes all PCIe ports use INTx for hotplug services).
2543
2544	pcie_ports=	[PCIE] PCIe ports handling:
2545		auto	Ask the BIOS whether or not to use native PCIe services
2546			associated with PCIe ports (PME, hot-plug, AER).  Use
2547			them only if that is allowed by the BIOS.
2548		native	Use native PCIe services associated with PCIe ports
2549			unconditionally.
2550		compat	Treat PCIe ports as PCI-to-PCI bridges, disable the PCIe
2551			ports driver.
2552
2553	pcie_pme=	[PCIE,PM] Native PCIe PME signaling options:
2554		nomsi	Do not use MSI for native PCIe PME signaling (this makes
2555			all PCIe root ports use INTx for all services).
2556
2557	pcmv=		[HW,PCMCIA] BadgePAD 4
2558
2559	pd_ignore_unused
2560			[PM]
2561			Keep all power-domains already enabled by bootloader on,
2562			even if no driver has claimed them. This is useful
2563			for debug and development, but should not be
2564			needed on a platform with proper driver support.
2565
2566	pd.		[PARIDE]
2567			See Documentation/blockdev/paride.txt.
2568
2569	pdcchassis=	[PARISC,HW] Disable/Enable PDC Chassis Status codes at
2570			boot time.
2571			Format: { 0 | 1 }
2572			See arch/parisc/kernel/pdc_chassis.c
2573
2574	percpu_alloc=	Select which percpu first chunk allocator to use.
2575			Currently supported values are "embed" and "page".
2576			Archs may support subset or none of the	selections.
2577			See comments in mm/percpu.c for details on each
2578			allocator.  This parameter is primarily	for debugging
2579			and performance comparison.
2580
2581	pf.		[PARIDE]
2582			See Documentation/blockdev/paride.txt.
2583
2584	pg.		[PARIDE]
2585			See Documentation/blockdev/paride.txt.
2586
2587	pirq=		[SMP,APIC] Manual mp-table setup
2588			See Documentation/x86/i386/IO-APIC.txt.
2589
2590	plip=		[PPT,NET] Parallel port network link
2591			Format: { parport<nr> | timid | 0 }
2592			See also Documentation/parport.txt.
2593
2594	pmtmr=		[X86] Manual setup of pmtmr I/O Port.
2595			Override pmtimer IOPort with a hex value.
2596			e.g. pmtmr=0x508
2597
2598	pnp.debug=1	[PNP]
2599			Enable PNP debug messages (depends on the
2600			CONFIG_PNP_DEBUG_MESSAGES option).  Change at run-time
2601			via /sys/module/pnp/parameters/debug.  We always show
2602			current resource usage; turning this on also shows
2603			possible settings and some assignment information.
2604
2605	pnpacpi=	[ACPI]
2606			{ off }
2607
2608	pnpbios=	[ISAPNP]
2609			{ on | off | curr | res | no-curr | no-res }
2610
2611	pnp_reserve_irq=
2612			[ISAPNP] Exclude IRQs for the autoconfiguration
2613
2614	pnp_reserve_dma=
2615			[ISAPNP] Exclude DMAs for the autoconfiguration
2616
2617	pnp_reserve_io=	[ISAPNP] Exclude I/O ports for the autoconfiguration
2618			Ranges are in pairs (I/O port base and size).
2619
2620	pnp_reserve_mem=
2621			[ISAPNP] Exclude memory regions for the
2622			autoconfiguration.
2623			Ranges are in pairs (memory base and size).
2624
2625	ports=		[IP_VS_FTP] IPVS ftp helper module
2626			Default is 21.
2627			Up to 8 (IP_VS_APP_MAX_PORTS) ports
2628			may be specified.
2629			Format: <port>,<port>....
2630
2631	print-fatal-signals=
2632			[KNL] debug: print fatal signals
2633
2634			If enabled, warn about various signal handling
2635			related application anomalies: too many signals,
2636			too many POSIX.1 timers, fatal signals causing a
2637			coredump - etc.
2638
2639			If you hit the warning due to signal overflow,
2640			you might want to try "ulimit -i unlimited".
2641
2642			default: off.
2643
2644	printk.always_kmsg_dump=
2645			Trigger kmsg_dump for cases other than kernel oops or
2646			panics
2647			Format: <bool>  (1/Y/y=enable, 0/N/n=disable)
2648			default: disabled
2649
2650	printk.time=	Show timing data prefixed to each printk message line
2651			Format: <bool>  (1/Y/y=enable, 0/N/n=disable)
2652
2653	processor.max_cstate=	[HW,ACPI]
2654			Limit processor to maximum C-state
2655			max_cstate=9 overrides any DMI blacklist limit.
2656
2657	processor.nocst	[HW,ACPI]
2658			Ignore the _CST method to determine C-states,
2659			instead using the legacy FADT method
2660
2661	profile=	[KNL] Enable kernel profiling via /proc/profile
2662			Format: [schedule,]<number>
2663			Param: "schedule" - profile schedule points.
2664			Param: <number> - step/bucket size as a power of 2 for
2665				statistical time based profiling.
2666			Param: "sleep" - profile D-state sleeping (millisecs).
2667				Requires CONFIG_SCHEDSTATS
2668			Param: "kvm" - profile VM exits.
2669
2670	prompt_ramdisk=	[RAM] List of RAM disks to prompt for floppy disk
2671			before loading.
2672			See Documentation/blockdev/ramdisk.txt.
2673
2674	psmouse.proto=	[HW,MOUSE] Highest PS2 mouse protocol extension to
2675			probe for; one of (bare|imps|exps|lifebook|any).
2676	psmouse.rate=	[HW,MOUSE] Set desired mouse report rate, in reports
2677			per second.
2678	psmouse.resetafter=	[HW,MOUSE]
2679			Try to reset the device after so many bad packets
2680			(0 = never).
2681	psmouse.resolution=
2682			[HW,MOUSE] Set desired mouse resolution, in dpi.
2683	psmouse.smartscroll=
2684			[HW,MOUSE] Controls Logitech smartscroll autorepeat.
2685			0 = disabled, 1 = enabled (default).
2686
2687	pstore.backend=	Specify the name of the pstore backend to use
2688
2689	pt.		[PARIDE]
2690			See Documentation/blockdev/paride.txt.
2691
2692	pty.legacy_count=
2693			[KNL] Number of legacy pty's. Overwrites compiled-in
2694			default number.
2695
2696	quiet		[KNL] Disable most log messages
2697
2698	r128=		[HW,DRM]
2699
2700	raid=		[HW,RAID]
2701			See Documentation/md.txt.
2702
2703	ramdisk_blocksize=	[RAM]
2704			See Documentation/blockdev/ramdisk.txt.
2705
2706	ramdisk_size=	[RAM] Sizes of RAM disks in kilobytes
2707			See Documentation/blockdev/ramdisk.txt.
2708
2709	rcu_nocbs=	[KNL]
2710			In kernels built with CONFIG_RCU_NOCB_CPU=y, set
2711			the specified list of CPUs to be no-callback CPUs.
2712			Invocation of these CPUs' RCU callbacks will
2713			be offloaded to "rcuox/N" kthreads created for
2714			that purpose, where "x" is "b" for RCU-bh, "p"
2715			for RCU-preempt, and "s" for RCU-sched, and "N"
2716			is the CPU number.  This reduces OS jitter on the
2717			offloaded CPUs, which can be useful for HPC and
2718			real-time workloads.  It can also improve energy
2719			efficiency for asymmetric multiprocessors.
2720
2721	rcu_nocb_poll	[KNL]
2722			Rather than requiring that offloaded CPUs
2723			(specified by rcu_nocbs= above) explicitly
2724			awaken the corresponding "rcuoN" kthreads,
2725			make these kthreads poll for callbacks.
2726			This improves the real-time response for the
2727			offloaded CPUs by relieving them of the need to
2728			wake up the corresponding kthread, but degrades
2729			energy efficiency by requiring that the kthreads
2730			periodically wake up to do the polling.
2731
2732	rcutree.blimit=	[KNL]
2733			Set maximum number of finished RCU callbacks to
2734			process in one batch.
2735
2736	rcutree.rcu_fanout_leaf= [KNL]
2737			Increase the number of CPUs assigned to each
2738			leaf rcu_node structure.  Useful for very large
2739			systems.
2740
2741	rcutree.jiffies_till_first_fqs= [KNL]
2742			Set delay from grace-period initialization to
2743			first attempt to force quiescent states.
2744			Units are jiffies, minimum value is zero,
2745			and maximum value is HZ.
2746
2747	rcutree.jiffies_till_next_fqs= [KNL]
2748			Set delay between subsequent attempts to force
2749			quiescent states.  Units are jiffies, minimum
2750			value is one, and maximum value is HZ.
2751
2752	rcutree.qhimark= [KNL]
2753			Set threshold of queued RCU callbacks beyond which
2754			batch limiting is disabled.
2755
2756	rcutree.qlowmark= [KNL]
2757			Set threshold of queued RCU callbacks below which
2758			batch limiting is re-enabled.
2759
2760	rcutree.rcu_idle_gp_delay= [KNL]
2761			Set wakeup interval for idle CPUs that have
2762			RCU callbacks (RCU_FAST_NO_HZ=y).
2763
2764	rcutree.rcu_idle_lazy_gp_delay= [KNL]
2765			Set wakeup interval for idle CPUs that have
2766			only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
2767			Lazy RCU callbacks are those which RCU can
2768			prove do nothing more than free memory.
2769
2770	rcutorture.fqs_duration= [KNL]
2771			Set duration of force_quiescent_state bursts.
2772
2773	rcutorture.fqs_holdoff= [KNL]
2774			Set holdoff time within force_quiescent_state bursts.
2775
2776	rcutorture.fqs_stutter= [KNL]
2777			Set wait time between force_quiescent_state bursts.
2778
2779	rcutorture.gp_exp= [KNL]
2780			Use expedited update-side primitives.
2781
2782	rcutorture.gp_normal= [KNL]
2783			Use normal (non-expedited) update-side primitives.
2784			If both gp_exp and gp_normal are set, do both.
2785			If neither gp_exp nor gp_normal are set, still
2786			do both.
2787
2788	rcutorture.n_barrier_cbs= [KNL]
2789			Set callbacks/threads for rcu_barrier() testing.
2790
2791	rcutorture.nfakewriters= [KNL]
2792			Set number of concurrent RCU writers.  These just
2793			stress RCU, they don't participate in the actual
2794			test, hence the "fake".
2795
2796	rcutorture.nreaders= [KNL]
2797			Set number of RCU readers.
2798
2799	rcutorture.object_debug= [KNL]
2800			Enable debug-object double-call_rcu() testing.
2801
2802	rcutorture.onoff_holdoff= [KNL]
2803			Set time (s) after boot for CPU-hotplug testing.
2804
2805	rcutorture.onoff_interval= [KNL]
2806			Set time (s) between CPU-hotplug operations, or
2807			zero to disable CPU-hotplug testing.
2808
2809	rcutorture.rcutorture_runnable= [BOOT]
2810			Start rcutorture running at boot time.
2811
2812	rcutorture.shuffle_interval= [KNL]
2813			Set task-shuffle interval (s).  Shuffling tasks
2814			allows some CPUs to go into dyntick-idle mode
2815			during the rcutorture test.
2816
2817	rcutorture.shutdown_secs= [KNL]
2818			Set time (s) after boot system shutdown.  This
2819			is useful for hands-off automated testing.
2820
2821	rcutorture.stall_cpu= [KNL]
2822			Duration of CPU stall (s) to test RCU CPU stall
2823			warnings, zero to disable.
2824
2825	rcutorture.stall_cpu_holdoff= [KNL]
2826			Time to wait (s) after boot before inducing stall.
2827
2828	rcutorture.stat_interval= [KNL]
2829			Time (s) between statistics printk()s.
2830
2831	rcutorture.stutter= [KNL]
2832			Time (s) to stutter testing, for example, specifying
2833			five seconds causes the test to run for five seconds,
2834			wait for five seconds, and so on.  This tests RCU's
2835			ability to transition abruptly to and from idle.
2836
2837	rcutorture.test_boost= [KNL]
2838			Test RCU priority boosting?  0=no, 1=maybe, 2=yes.
2839			"Maybe" means test if the RCU implementation
2840			under test support RCU priority boosting.
2841
2842	rcutorture.test_boost_duration= [KNL]
2843			Duration (s) of each individual boost test.
2844
2845	rcutorture.test_boost_interval= [KNL]
2846			Interval (s) between each boost test.
2847
2848	rcutorture.test_no_idle_hz= [KNL]
2849			Test RCU's dyntick-idle handling.  See also the
2850			rcutorture.shuffle_interval parameter.
2851
2852	rcutorture.torture_type= [KNL]
2853			Specify the RCU implementation to test.
2854
2855	rcutorture.verbose= [KNL]
2856			Enable additional printk() statements.
2857
2858	rcupdate.rcu_expedited= [KNL]
2859			Use expedited grace-period primitives, for
2860			example, synchronize_rcu_expedited() instead
2861			of synchronize_rcu().  This reduces latency,
2862			but can increase CPU utilization, degrade
2863			real-time latency, and degrade energy efficiency.
2864
2865	rcupdate.rcu_cpu_stall_suppress= [KNL]
2866			Suppress RCU CPU stall warning messages.
2867
2868	rcupdate.rcu_cpu_stall_timeout= [KNL]
2869			Set timeout for RCU CPU stall warning messages.
2870
2871	rdinit=		[KNL]
2872			Format: <full_path>
2873			Run specified binary instead of /init from the ramdisk,
2874			used for early userspace startup. See initrd.
2875
2876	reboot=		[KNL]
2877			Format (x86 or x86_64):
2878				[w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
2879				[[,]s[mp]#### \
2880				[[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
2881				[[,]f[orce]
2882			Where reboot_mode is one of warm (soft) or cold (hard) or gpio,
2883			      reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
2884			      reboot_force is either force or not specified,
2885			      reboot_cpu is s[mp]#### with #### being the processor
2886					to be used for rebooting.
2887
2888	relax_domain_level=
2889			[KNL, SMP] Set scheduler's default relax_domain_level.
2890			See Documentation/cgroups/cpusets.txt.
2891
2892	reserve=	[KNL,BUGS] Force the kernel to ignore some iomem area
2893
2894	reservetop=	[X86-32]
2895			Format: nn[KMG]
2896			Reserves a hole at the top of the kernel virtual
2897			address space.
2898
2899	reservelow=	[X86]
2900			Format: nn[K]
2901			Set the amount of memory to reserve for BIOS at
2902			the bottom of the address space.
2903
2904	reset_devices	[KNL] Force drivers to reset the underlying device
2905			during initialization.
2906
2907	resume=		[SWSUSP]
2908			Specify the partition device for software suspend
2909			Format:
2910			{/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
2911
2912	resume_offset=	[SWSUSP]
2913			Specify the offset from the beginning of the partition
2914			given by "resume=" at which the swap header is located,
2915			in <PAGE_SIZE> units (needed only for swap files).
2916			See  Documentation/power/swsusp-and-swap-files.txt
2917
2918	resumedelay=	[HIBERNATION] Delay (in seconds) to pause before attempting to
2919			read the resume files
2920
2921	resumewait	[HIBERNATION] Wait (indefinitely) for resume device to show up.
2922			Useful for devices that are detected asynchronously
2923			(e.g. USB and MMC devices).
2924
2925	hibernate=	[HIBERNATION]
2926		noresume	Don't check if there's a hibernation image
2927				present during boot.
2928		nocompress	Don't compress/decompress hibernation images.
2929
2930	retain_initrd	[RAM] Keep initrd memory after extraction
2931
2932	rhash_entries=	[KNL,NET]
2933			Set number of hash buckets for route cache
2934
 
 
 
2935	ro		[KNL] Mount root device read-only on boot
2936
2937	root=		[KNL] Root filesystem
2938			See name_to_dev_t comment in init/do_mounts.c.
2939
2940	rootdelay=	[KNL] Delay (in seconds) to pause before attempting to
2941			mount the root filesystem
2942
2943	rootflags=	[KNL] Set root filesystem mount option string
2944
2945	rootfstype=	[KNL] Set root filesystem type
2946
2947	rootwait	[KNL] Wait (indefinitely) for root device to show up.
2948			Useful for devices that are detected asynchronously
2949			(e.g. USB and MMC devices).
2950
2951	rproc_mem=nn[KMG][@address]
2952			[KNL,ARM,CMA] Remoteproc physical memory block.
2953			Memory area to be used by remote processor image,
2954			managed by CMA.
2955
2956	rw		[KNL] Mount root device read-write on boot
2957
2958	S		[KNL] Run init in single mode
2959
2960	sa1100ir	[NET]
2961			See drivers/net/irda/sa1100_ir.c.
2962
2963	sbni=		[NET] Granch SBNI12 leased line adapter
2964
2965	sched_debug	[KNL] Enables verbose scheduler debug messages.
2966
2967	skew_tick=	[KNL] Offset the periodic timer tick per cpu to mitigate
2968			xtime_lock contention on larger systems, and/or RCU lock
2969			contention on all systems with CONFIG_MAXSMP set.
2970			Format: { "0" | "1" }
2971			0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
2972			1 -- enable.
2973			Note: increases power consumption, thus should only be
2974			enabled if running jitter sensitive (HPC/RT) workloads.
2975
2976	security=	[SECURITY] Choose a security module to enable at boot.
2977			If this boot parameter is not specified, only the first
2978			security module asking for security registration will be
2979			loaded. An invalid security module name will be treated
2980			as if no module has been chosen.
2981
2982	selinux=	[SELINUX] Disable or enable SELinux at boot time.
2983			Format: { "0" | "1" }
2984			See security/selinux/Kconfig help text.
2985			0 -- disable.
2986			1 -- enable.
2987			Default value is set via kernel config option.
2988			If enabled at boot time, /selinux/disable can be used
2989			later to disable prior to initial policy load.
2990
2991	apparmor=	[APPARMOR] Disable or enable AppArmor at boot time
2992			Format: { "0" | "1" }
2993			See security/apparmor/Kconfig help text
2994			0 -- disable.
2995			1 -- enable.
2996			Default value is set via kernel config option.
2997
2998	serialnumber	[BUGS=X86-32]
2999
3000	shapers=	[NET]
3001			Maximal number of shapers.
3002
3003	show_msr=	[x86] show boot-time MSR settings
3004			Format: { <integer> }
3005			Show boot-time (BIOS-initialized) MSR settings.
3006			The parameter means the number of CPUs to show,
3007			for example 1 means boot CPU only.
3008
3009	simeth=		[IA-64]
3010	simscsi=
3011
3012	slram=		[HW,MTD]
3013
3014	slab_max_order=	[MM, SLAB]
3015			Determines the maximum allowed order for slabs.
3016			A high setting may cause OOMs due to memory
3017			fragmentation.  Defaults to 1 for systems with
3018			more than 32MB of RAM, 0 otherwise.
3019
3020	slub_debug[=options[,slabs]]	[MM, SLUB]
3021			Enabling slub_debug allows one to determine the
3022			culprit if slab objects become corrupted. Enabling
3023			slub_debug can create guard zones around objects and
3024			may poison objects when not in use. Also tracks the
3025			last alloc / free. For more information see
3026			Documentation/vm/slub.txt.
3027
3028	slub_max_order= [MM, SLUB]
3029			Determines the maximum allowed order for slabs.
3030			A high setting may cause OOMs due to memory
3031			fragmentation. For more information see
3032			Documentation/vm/slub.txt.
3033
3034	slub_min_objects=	[MM, SLUB]
3035			The minimum number of objects per slab. SLUB will
3036			increase the slab order up to slub_max_order to
3037			generate a sufficiently large slab able to contain
3038			the number of objects indicated. The higher the number
3039			of objects the smaller the overhead of tracking slabs
3040			and the less frequently locks need to be acquired.
3041			For more information see Documentation/vm/slub.txt.
3042
3043	slub_min_order=	[MM, SLUB]
3044			Determines the minimum page order for slabs. Must be
3045			lower than slub_max_order.
3046			For more information see Documentation/vm/slub.txt.
3047
3048	slub_nomerge	[MM, SLUB]
3049			Disable merging of slabs with similar size. May be
3050			necessary if there is some reason to distinguish
3051			allocs to different slabs. Debug options disable
3052			merging on their own.
3053			For more information see Documentation/vm/slub.txt.
3054
3055	smart2=		[HW]
3056			Format: <io1>[,<io2>[,...,<io8>]]
3057
 
 
 
3058	smsc-ircc2.nopnp	[HW] Don't use PNP to discover SMC devices
3059	smsc-ircc2.ircc_cfg=	[HW] Device configuration I/O port
3060	smsc-ircc2.ircc_sir=	[HW] SIR base I/O port
3061	smsc-ircc2.ircc_fir=	[HW] FIR base I/O port
3062	smsc-ircc2.ircc_irq=	[HW] IRQ line
3063	smsc-ircc2.ircc_dma=	[HW] DMA channel
3064	smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
3065				0: Toshiba Satellite 1800 (GP data pin select)
3066				1: Fast pin select (default)
3067				2: ATC IRMode
3068
3069	softlockup_panic=
3070			[KNL] Should the soft-lockup detector generate panics.
3071			Format: <integer>
3072
3073	sonypi.*=	[HW] Sony Programmable I/O Control Device driver
3074			See Documentation/laptops/sonypi.txt
 
 
 
3075
3076	spia_io_base=	[HW,MTD]
3077	spia_fio_base=
3078	spia_pedr=
3079	spia_peddr=
3080
3081	stacktrace	[FTRACE]
3082			Enabled the stack tracer on boot up.
3083
3084	stacktrace_filter=[function-list]
3085			[FTRACE] Limit the functions that the stack tracer
3086			will trace at boot up. function-list is a comma separated
3087			list of functions. This list can be changed at run
3088			time by the stack_trace_filter file in the debugfs
3089			tracing directory. Note, this enables stack tracing
3090			and the stacktrace above is not needed.
3091
3092	sti=		[PARISC,HW]
3093			Format: <num>
3094			Set the STI (builtin display/keyboard on the HP-PARISC
3095			machines) console (graphic card) which should be used
3096			as the initial boot-console.
3097			See also comment in drivers/video/console/sticore.c.
3098
3099	sti_font=	[HW]
3100			See comment in drivers/video/console/sticore.c.
3101
3102	stifb=		[HW]
3103			Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
3104
3105	sunrpc.min_resvport=
3106	sunrpc.max_resvport=
3107			[NFS,SUNRPC]
3108			SunRPC servers often require that client requests
3109			originate from a privileged port (i.e. a port in the
3110			range 0 < portnr < 1024).
3111			An administrator who wishes to reserve some of these
3112			ports for other uses may adjust the range that the
3113			kernel's sunrpc client considers to be privileged
3114			using these two parameters to set the minimum and
3115			maximum port values.
3116
3117	sunrpc.pool_mode=
3118			[NFS]
3119			Control how the NFS server code allocates CPUs to
3120			service thread pools.  Depending on how many NICs
3121			you have and where their interrupts are bound, this
3122			option will affect which CPUs will do NFS serving.
3123			Note: this parameter cannot be changed while the
3124			NFS server is running.
3125
3126			auto	    the server chooses an appropriate mode
3127				    automatically using heuristics
3128			global	    a single global pool contains all CPUs
3129			percpu	    one pool for each CPU
3130			pernode	    one pool for each NUMA node (equivalent
3131				    to global on non-NUMA machines)
3132
3133	sunrpc.tcp_slot_table_entries=
3134	sunrpc.udp_slot_table_entries=
3135			[NFS,SUNRPC]
3136			Sets the upper limit on the number of simultaneous
3137			RPC calls that can be sent from the client to a
3138			server. Increasing these values may allow you to
3139			improve throughput, but will also increase the
3140			amount of memory reserved for use by the client.
3141
3142	swapaccount=[0|1]
3143			[KNL] Enable accounting of swap in memory resource
3144			controller if no parameter or 1 is given or disable
3145			it if 0 is given (See Documentation/cgroups/memory.txt)
3146
3147	swiotlb=	[ARM,IA-64,PPC,MIPS,X86]
3148			Format: { <int> | force }
3149			<int> -- Number of I/O TLB slabs
3150			force -- force using of bounce buffers even if they
3151			         wouldn't be automatically used by the kernel
3152
3153	switches=	[HW,M68k]
3154
3155	sysfs.deprecated=0|1 [KNL]
3156			Enable/disable old style sysfs layout for old udev
3157			on older distributions. When this option is enabled
3158			very new udev will not work anymore. When this option
3159			is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
3160			in older udev will not work anymore.
3161			Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
3162			the kernel configuration.
3163
3164	sysrq_always_enabled
3165			[KNL]
3166			Ignore sysrq setting - this boot parameter will
3167			neutralize any effect of /proc/sys/kernel/sysrq.
3168			Useful for debugging.
3169
3170	tdfx=		[HW,DRM]
3171
3172	test_suspend=	[SUSPEND]
3173			Specify "mem" (for Suspend-to-RAM) or "standby" (for
3174			standby suspend) as the system sleep state to briefly
3175			enter during system startup.  The system is woken from
3176			this state using a wakeup-capable RTC alarm.
3177
3178	thash_entries=	[KNL,NET]
3179			Set number of hash buckets for TCP connection
3180
3181	thermal.act=	[HW,ACPI]
3182			-1: disable all active trip points in all thermal zones
3183			<degrees C>: override all lowest active trip points
3184
3185	thermal.crt=	[HW,ACPI]
3186			-1: disable all critical trip points in all thermal zones
3187			<degrees C>: override all critical trip points
3188
3189	thermal.nocrt=	[HW,ACPI]
3190			Set to disable actions on ACPI thermal zone
3191			critical and hot trip points.
3192
3193	thermal.off=	[HW,ACPI]
3194			1: disable ACPI thermal control
3195
3196	thermal.psv=	[HW,ACPI]
3197			-1: disable all passive trip points
3198			<degrees C>: override all passive trip points to this
3199			value
3200
3201	thermal.tzp=	[HW,ACPI]
3202			Specify global default ACPI thermal zone polling rate
3203			<deci-seconds>: poll all this frequency
3204			0: no polling (default)
3205
3206	threadirqs	[KNL]
3207			Force threading of all interrupt handlers except those
3208			marked explicitly IRQF_NO_THREAD.
3209
3210	tmem		[KNL,XEN]
3211			Enable the Transcendent memory driver if built-in.
3212
3213	tmem.cleancache=0|1 [KNL, XEN]
3214			Default is on (1). Disable the usage of the cleancache
3215			API to send anonymous pages to the hypervisor.
3216
3217	tmem.frontswap=0|1 [KNL, XEN]
3218			Default is on (1). Disable the usage of the frontswap
3219			API to send swap pages to the hypervisor. If disabled
3220			the selfballooning and selfshrinking are force disabled.
3221
3222	tmem.selfballooning=0|1 [KNL, XEN]
3223			Default is on (1). Disable the driving of swap pages
3224			to the hypervisor.
3225
3226	tmem.selfshrinking=0|1 [KNL, XEN]
3227			Default is on (1). Partial swapoff that immediately
3228			transfers pages from Xen hypervisor back to the
3229			kernel based on different criteria.
3230
3231	topology=	[S390]
3232			Format: {off | on}
3233			Specify if the kernel should make use of the cpu
3234			topology information if the hardware supports this.
3235			The scheduler will make use of this information and
3236			e.g. base its process migration decisions on it.
3237			Default is on.
3238
3239	tp720=		[HW,PS2]
3240
3241	tpm_suspend_pcr=[HW,TPM]
3242			Format: integer pcr id
3243			Specify that at suspend time, the tpm driver
3244			should extend the specified pcr with zeros,
3245			as a workaround for some chips which fail to
3246			flush the last written pcr on TPM_SaveState.
3247			This will guarantee that all the other pcrs
3248			are saved.
3249
3250	trace_buf_size=nn[KMG]
3251			[FTRACE] will set tracing buffer size.
3252
3253	trace_event=[event-list]
3254			[FTRACE] Set and start specified trace events in order
3255			to facilitate early boot debugging.
3256			See also Documentation/trace/events.txt
3257
3258	trace_options=[option-list]
3259			[FTRACE] Enable or disable tracer options at boot.
3260			The option-list is a comma delimited list of options
3261			that can be enabled or disabled just as if you were
3262			to echo the option name into
3263
3264			    /sys/kernel/debug/tracing/trace_options
3265
3266			For example, to enable stacktrace option (to dump the
3267			stack trace of each event), add to the command line:
3268
3269			      trace_options=stacktrace
3270
3271			See also Documentation/trace/ftrace.txt "trace options"
3272			section.
3273
3274	traceoff_on_warning
3275			[FTRACE] enable this option to disable tracing when a
3276			warning is hit. This turns off "tracing_on". Tracing can
3277			be enabled again by echoing '1' into the "tracing_on"
3278			file located in /sys/kernel/debug/tracing/
3279
3280			This option is useful, as it disables the trace before
3281			the WARNING dump is called, which prevents the trace to
3282			be filled with content caused by the warning output.
3283
3284			This option can also be set at run time via the sysctl
3285			option:  kernel/traceoff_on_warning
3286
3287	transparent_hugepage=
3288			[KNL]
3289			Format: [always|madvise|never]
3290			Can be used to control the default behavior of the system
3291			with respect to transparent hugepages.
3292			See Documentation/vm/transhuge.txt for more details.
3293
3294	tsc=		Disable clocksource stability checks for TSC.
3295			Format: <string>
3296			[x86] reliable: mark tsc clocksource as reliable, this
3297			disables clocksource verification at runtime, as well
3298			as the stability checks done at bootup.	Used to enable
3299			high-resolution timer mode on older hardware, and in
3300			virtualized environment.
3301			[x86] noirqtime: Do not use TSC to do irq accounting.
3302			Used to run time disable IRQ_TIME_ACCOUNTING on any
3303			platforms where RDTSC is slow and this accounting
3304			can add overhead.
3305
3306	turbografx.map[2|3]=	[HW,JOY]
3307			TurboGraFX parallel port interface
3308			Format:
3309			<port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
3310			See also Documentation/input/joystick-parport.txt
3311
3312	udbg-immortal	[PPC] When debugging early kernel crashes that
3313			happen after console_init() and before a proper 
3314			console driver takes over, this boot options might
3315			help "seeing" what's going on.
3316
3317	uhash_entries=	[KNL,NET]
3318			Set number of hash buckets for UDP/UDP-Lite connections
3319
3320	uhci-hcd.ignore_oc=
3321			[USB] Ignore overcurrent events (default N).
3322			Some badly-designed motherboards generate lots of
3323			bogus events, for ports that aren't wired to
3324			anything.  Set this parameter to avoid log spamming.
3325			Note that genuine overcurrent events won't be
3326			reported either.
3327
3328	unknown_nmi_panic
3329			[X86] Cause panic on unknown NMI.
3330
3331	usbcore.authorized_default=
3332			[USB] Default USB device authorization:
3333			(default -1 = authorized except for wireless USB,
3334			0 = not authorized, 1 = authorized)
3335
3336	usbcore.autosuspend=
3337			[USB] The autosuspend time delay (in seconds) used
3338			for newly-detected USB devices (default 2).  This
3339			is the time required before an idle device will be
3340			autosuspended.  Devices for which the delay is set
3341			to a negative value won't be autosuspended at all.
3342
3343	usbcore.usbfs_snoop=
3344			[USB] Set to log all usbfs traffic (default 0 = off).
3345
3346	usbcore.blinkenlights=
3347			[USB] Set to cycle leds on hubs (default 0 = off).
3348
3349	usbcore.old_scheme_first=
3350			[USB] Start with the old device initialization
3351			scheme (default 0 = off).
3352
3353	usbcore.usbfs_memory_mb=
3354			[USB] Memory limit (in MB) for buffers allocated by
3355			usbfs (default = 16, 0 = max = 2047).
3356
3357	usbcore.use_both_schemes=
3358			[USB] Try the other device initialization scheme
3359			if the first one fails (default 1 = enabled).
3360
3361	usbcore.initial_descriptor_timeout=
3362			[USB] Specifies timeout for the initial 64-byte
3363                        USB_REQ_GET_DESCRIPTOR request in milliseconds
3364			(default 5000 = 5.0 seconds).
3365
3366	usbhid.mousepoll=
3367			[USBHID] The interval which mice are to be polled at.
3368
3369	usb-storage.delay_use=
3370			[UMS] The delay in seconds before a new device is
3371			scanned for Logical Units (default 5).
3372
3373	usb-storage.quirks=
3374			[UMS] A list of quirks entries to supplement or
3375			override the built-in unusual_devs list.  List
3376			entries are separated by commas.  Each entry has
3377			the form VID:PID:Flags where VID and PID are Vendor
3378			and Product ID values (4-digit hex numbers) and
3379			Flags is a set of characters, each corresponding
3380			to a common usb-storage quirk flag as follows:
3381				a = SANE_SENSE (collect more than 18 bytes
3382					of sense data);
3383				b = BAD_SENSE (don't collect more than 18
3384					bytes of sense data);
3385				c = FIX_CAPACITY (decrease the reported
3386					device capacity by one sector);
3387				d = NO_READ_DISC_INFO (don't use
3388					READ_DISC_INFO command);
3389				e = NO_READ_CAPACITY_16 (don't use
3390					READ_CAPACITY_16 command);
3391				h = CAPACITY_HEURISTICS (decrease the
3392					reported device capacity by one
3393					sector if the number is odd);
3394				i = IGNORE_DEVICE (don't bind to this
3395					device);
3396				l = NOT_LOCKABLE (don't try to lock and
3397					unlock ejectable media);
3398				m = MAX_SECTORS_64 (don't transfer more
3399					than 64 sectors = 32 KB at a time);
3400				n = INITIAL_READ10 (force a retry of the
3401					initial READ(10) command);
3402				o = CAPACITY_OK (accept the capacity
3403					reported by the device);
3404				p = WRITE_CACHE (the device cache is ON
3405					by default);
3406				r = IGNORE_RESIDUE (the device reports
3407					bogus residue values);
3408				s = SINGLE_LUN (the device has only one
3409					Logical Unit);
3410				w = NO_WP_DETECT (don't test whether the
3411					medium is write-protected).
3412			Example: quirks=0419:aaf5:rl,0421:0433:rc
3413
3414	user_debug=	[KNL,ARM]
3415			Format: <int>
3416			See arch/arm/Kconfig.debug help text.
3417				 1 - undefined instruction events
3418				 2 - system calls
3419				 4 - invalid data aborts
3420				 8 - SIGSEGV faults
3421				16 - SIGBUS faults
3422			Example: user_debug=31
3423
3424	userpte=
3425			[X86] Flags controlling user PTE allocations.
3426
3427				nohigh = do not allocate PTE pages in
3428					HIGHMEM regardless of setting
3429					of CONFIG_HIGHPTE.
3430
3431	vdso=		[X86,SH]
3432			On X86_32, this is an alias for vdso32=.  Otherwise:
3433
3434			vdso=1: enable VDSO (the default)
3435			vdso=0: disable VDSO mapping
3436
3437	vdso32=		[X86] Control the 32-bit vDSO
3438			vdso32=1: enable 32-bit VDSO
3439			vdso32=0 or vdso32=2: disable 32-bit VDSO
3440
3441			See the help text for CONFIG_COMPAT_VDSO for more
3442			details.  If CONFIG_COMPAT_VDSO is set, the default is
3443			vdso32=0; otherwise, the default is vdso32=1.
3444
3445			For compatibility with older kernels, vdso32=2 is an
3446			alias for vdso32=0.
3447
3448			Try vdso32=0 if you encounter an error that says:
3449			dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
3450
3451	vector=		[IA-64,SMP]
3452			vector=percpu: enable percpu vector domain
3453
3454	video=		[FB] Frame buffer configuration
3455			See Documentation/fb/modedb.txt.
3456
3457	video.brightness_switch_enabled= [0,1]
3458			If set to 1, on receiving an ACPI notify event
3459			generated by hotkey, video driver will adjust brightness
3460			level and then send out the event to user space through
3461			the allocated input device; If set to 0, video driver
3462			will only send out the event without touching backlight
3463			brightness level.
3464			default: 1
3465
3466	virtio_mmio.device=
3467			[VMMIO] Memory mapped virtio (platform) device.
3468
3469				<size>@<baseaddr>:<irq>[:<id>]
3470			where:
3471				<size>     := size (can use standard suffixes
3472						like K, M and G)
3473				<baseaddr> := physical base address
3474				<irq>      := interrupt number (as passed to
3475						request_irq())
3476				<id>       := (optional) platform device id
3477			example:
3478				virtio_mmio.device=1K@0x100b0000:48:7
3479
3480			Can be used multiple times for multiple devices.
3481
3482	vga=		[BOOT,X86-32] Select a particular video mode
3483			See Documentation/x86/boot.txt and
3484			Documentation/svga.txt.
3485			Use vga=ask for menu.
3486			This is actually a boot loader parameter; the value is
3487			passed to the kernel using a special protocol.
3488
3489	vmalloc=nn[KMG]	[KNL,BOOT] Forces the vmalloc area to have an exact
3490			size of <nn>. This can be used to increase the
3491			minimum size (128MB on x86). It can also be used to
3492			decrease the size and leave more room for directly
3493			mapped kernel RAM.
3494
3495	vmhalt=		[KNL,S390] Perform z/VM CP command after system halt.
3496			Format: <command>
3497
3498	vmpanic=	[KNL,S390] Perform z/VM CP command after kernel panic.
3499			Format: <command>
3500
3501	vmpoff=		[KNL,S390] Perform z/VM CP command after power off.
3502			Format: <command>
3503
3504	vsyscall=	[X86-64]
3505			Controls the behavior of vsyscalls (i.e. calls to
3506			fixed addresses of 0xffffffffff600x00 from legacy
3507			code).  Most statically-linked binaries and older
3508			versions of glibc use these calls.  Because these
3509			functions are at fixed addresses, they make nice
3510			targets for exploits that can control RIP.
3511
3512			emulate     [default] Vsyscalls turn into traps and are
3513			            emulated reasonably safely.
3514
3515			native      Vsyscalls are native syscall instructions.
 
3516			            This is a little bit faster than trapping
3517			            and makes a few dynamic recompilers work
3518			            better than they would in emulation mode.
3519			            It also makes exploits much easier to write.
3520
3521			none        Vsyscalls don't work at all.  This makes
3522			            them quite hard to use for exploits but
3523			            might break your system.
3524
3525	vt.color=	[VT] Default text color.
3526			Format: 0xYX, X = foreground, Y = background.
3527			Default: 0x07 = light gray on black.
3528
3529	vt.cur_default=	[VT] Default cursor shape.
3530			Format: 0xCCBBAA, where AA, BB, and CC are the same as
3531			the parameters of the <Esc>[?A;B;Cc escape sequence;
3532			see VGA-softcursor.txt. Default: 2 = underline.
3533
3534	vt.default_blu=	[VT]
3535			Format: <blue0>,<blue1>,<blue2>,...,<blue15>
3536			Change the default blue palette of the console.
3537			This is a 16-member array composed of values
3538			ranging from 0-255.
3539
3540	vt.default_grn=	[VT]
3541			Format: <green0>,<green1>,<green2>,...,<green15>
3542			Change the default green palette of the console.
3543			This is a 16-member array composed of values
3544			ranging from 0-255.
3545
3546	vt.default_red=	[VT]
3547			Format: <red0>,<red1>,<red2>,...,<red15>
3548			Change the default red palette of the console.
3549			This is a 16-member array composed of values
3550			ranging from 0-255.
3551
3552	vt.default_utf8=
3553			[VT]
3554			Format=<0|1>
3555			Set system-wide default UTF-8 mode for all tty's.
3556			Default is 1, i.e. UTF-8 mode is enabled for all
3557			newly opened terminals.
3558
3559	vt.global_cursor_default=
3560			[VT]
3561			Format=<-1|0|1>
3562			Set system-wide default for whether a cursor
3563			is shown on new VTs. Default is -1,
3564			i.e. cursors will be created by default unless
3565			overridden by individual drivers. 0 will hide
3566			cursors, 1 will display them.
3567
3568	vt.italic=	[VT] Default color for italic text; 0-15.
3569			Default: 2 = green.
3570
3571	vt.underline=	[VT] Default color for underlined text; 0-15.
3572			Default: 3 = cyan.
3573
3574	watchdog timers	[HW,WDT] For information on watchdog timers,
3575			see Documentation/watchdog/watchdog-parameters.txt
3576			or other driver-specific files in the
3577			Documentation/watchdog/ directory.
3578
3579	workqueue.disable_numa
3580			By default, all work items queued to unbound
3581			workqueues are affine to the NUMA nodes they're
3582			issued on, which results in better behavior in
3583			general.  If NUMA affinity needs to be disabled for
3584			whatever reason, this option can be used.  Note
3585			that this also can be controlled per-workqueue for
3586			workqueues visible under /sys/bus/workqueue/.
3587
3588	workqueue.power_efficient
3589			Per-cpu workqueues are generally preferred because
3590			they show better performance thanks to cache
3591			locality; unfortunately, per-cpu workqueues tend to
3592			be more power hungry than unbound workqueues.
3593
3594			Enabling this makes the per-cpu workqueues which
3595			were observed to contribute significantly to power
3596			consumption unbound, leading to measurably lower
3597			power usage at the cost of small performance
3598			overhead.
3599
3600			The default value of this parameter is determined by
3601			the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
3602
3603	x2apic_phys	[X86-64,APIC] Use x2apic physical mode instead of
3604			default x2apic cluster mode on platforms
3605			supporting x2apic.
3606
3607	x86_intel_mid_timer= [X86-32,APBT]
3608			Choose timer option for x86 Intel MID platform.
3609			Two valid options are apbt timer only and lapic timer
3610			plus one apbt timer for broadcast timer.
3611			x86_intel_mid_timer=apbt_only | lapic_and_apbt
 
 
 
3612
3613	xen_emul_unplug=		[HW,X86,XEN]
3614			Unplug Xen emulated devices
3615			Format: [unplug0,][unplug1]
3616			ide-disks -- unplug primary master IDE devices
3617			aux-ide-disks -- unplug non-primary-master IDE devices
3618			nics -- unplug network devices
3619			all -- unplug all emulated devices (NICs and IDE disks)
3620			unnecessary -- unplugging emulated devices is
3621				unnecessary even if the host did not respond to
3622				the unplug protocol
3623			never -- do not unplug even if version check succeeds
3624
3625	xen_nopvspin	[X86,XEN]
3626			Disables the ticketlock slowpath using Xen PV
3627			optimizations.
3628
3629	xirc2ps_cs=	[NET,PCMCIA]
3630			Format:
3631			<irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
3632
3633______________________________________________________________________
3634
3635TODO:
3636
3637	Add more DRM drivers.
v3.1
   1                          Kernel Parameters
   2                          ~~~~~~~~~~~~~~~~~
   3
   4The following is a consolidated list of the kernel parameters as implemented
   5(mostly) by the __setup() macro and sorted into English Dictionary order
   6(defined as ignoring all punctuation and sorting digits before letters in a
   7case insensitive manner), and with descriptions where known.
   8
   9Module parameters for loadable modules are specified only as the
  10parameter name with optional '=' and value as appropriate, such as:
  11
  12	modprobe usbcore blinkenlights=1
  13
  14Module parameters for modules that are built into the kernel image
  15are specified on the kernel command line with the module name plus
  16'.' plus parameter name, with '=' and value if appropriate, such as:
  17
  18	usbcore.blinkenlights=1
  19
  20Hyphens (dashes) and underscores are equivalent in parameter names, so
  21	log_buf_len=1M print-fatal-signals=1
  22can also be entered as
  23	log-buf-len=1M print_fatal_signals=1
  24
  25
  26This document may not be entirely up to date and comprehensive. The command
  27"modinfo -p ${modulename}" shows a current list of all parameters of a loadable
  28module. Loadable modules, after being loaded into the running kernel, also
  29reveal their parameters in /sys/module/${modulename}/parameters/. Some of these
  30parameters may be changed at runtime by the command
  31"echo -n ${value} > /sys/module/${modulename}/parameters/${parm}".
  32
  33The parameters listed below are only valid if certain kernel build options were
  34enabled and if respective hardware is present. The text in square brackets at
  35the beginning of each description states the restrictions within which a
  36parameter is applicable:
  37
  38	ACPI	ACPI support is enabled.
  39	AGP	AGP (Accelerated Graphics Port) is enabled.
  40	ALSA	ALSA sound support is enabled.
  41	APIC	APIC support is enabled.
  42	APM	Advanced Power Management support is enabled.
  43	ARM	ARM architecture is enabled.
  44	AVR32	AVR32 architecture is enabled.
  45	AX25	Appropriate AX.25 support is enabled.
  46	BLACKFIN Blackfin architecture is enabled.
 
 
  47	DRM	Direct Rendering Management support is enabled.
  48	DYNAMIC_DEBUG Build in debug messages and enable them at runtime
  49	EDD	BIOS Enhanced Disk Drive Services (EDD) is enabled
  50	EFI	EFI Partitioning (GPT) is enabled
  51	EIDE	EIDE/ATAPI support is enabled.
 
  52	FB	The frame buffer device is enabled.
  53	FTRACE	Function tracing enabled.
  54	GCOV	GCOV profiling is enabled.
  55	HW	Appropriate hardware is enabled.
  56	IA-64	IA-64 architecture is enabled.
  57	IMA     Integrity measurement architecture is enabled.
  58	IOSCHED	More than one I/O scheduler is enabled.
  59	IP_PNP	IP DHCP, BOOTP, or RARP is enabled.
  60	IPV6	IPv6 support is enabled.
  61	ISAPNP	ISA PnP code is enabled.
  62	ISDN	Appropriate ISDN support is enabled.
  63	JOY	Appropriate joystick support is enabled.
  64	KGDB	Kernel debugger support is enabled.
  65	KVM	Kernel Virtual Machine support is enabled.
  66	LIBATA  Libata driver is enabled
  67	LP	Printer support is enabled.
  68	LOOP	Loopback device support is enabled.
  69	M68k	M68k architecture is enabled.
  70			These options have more detailed description inside of
  71			Documentation/m68k/kernel-options.txt.
  72	MCA	MCA bus support is enabled.
  73	MDA	MDA console support is enabled.
  74	MIPS	MIPS architecture is enabled.
  75	MOUSE	Appropriate mouse support is enabled.
  76	MSI	Message Signaled Interrupts (PCI).
  77	MTD	MTD (Memory Technology Device) support is enabled.
  78	NET	Appropriate network support is enabled.
  79	NUMA	NUMA support is enabled.
  80	NFS	Appropriate NFS support is enabled.
  81	OSS	OSS sound support is enabled.
  82	PV_OPS	A paravirtualized kernel is enabled.
  83	PARIDE	The ParIDE (parallel port IDE) subsystem is enabled.
  84	PARISC	The PA-RISC architecture is enabled.
  85	PCI	PCI bus support is enabled.
  86	PCIE	PCI Express support is enabled.
  87	PCMCIA	The PCMCIA subsystem is enabled.
  88	PNP	Plug & Play support is enabled.
  89	PPC	PowerPC architecture is enabled.
  90	PPT	Parallel port support is enabled.
  91	PS2	Appropriate PS/2 support is enabled.
  92	RAM	RAM disk support is enabled.
  93	S390	S390 architecture is enabled.
  94	SCSI	Appropriate SCSI support is enabled.
  95			A lot of drivers have their options described inside
  96			the Documentation/scsi/ sub-directory.
  97	SECURITY Different security models are enabled.
  98	SELINUX SELinux support is enabled.
  99	APPARMOR AppArmor support is enabled.
 100	SERIAL	Serial support is enabled.
 101	SH	SuperH architecture is enabled.
 102	SMP	The kernel is an SMP kernel.
 103	SPARC	Sparc architecture is enabled.
 104	SWSUSP	Software suspend (hibernation) is enabled.
 105	SUSPEND	System suspend states are enabled.
 106	TPM	TPM drivers are enabled.
 107	TS	Appropriate touchscreen support is enabled.
 108	UMS	USB Mass Storage support is enabled.
 109	USB	USB support is enabled.
 110	USBHID	USB Human Interface Device support is enabled.
 111	V4L	Video For Linux support is enabled.
 
 112	VGA	The VGA console has been enabled.
 113	VT	Virtual terminal support is enabled.
 114	WDT	Watchdog support is enabled.
 115	XT	IBM PC/XT MFM hard disk support is enabled.
 116	X86-32	X86-32, aka i386 architecture is enabled.
 117	X86-64	X86-64 architecture is enabled.
 118			More X86-64 boot options can be found in
 119			Documentation/x86/x86_64/boot-options.txt .
 120	X86	Either 32-bit or 64-bit x86 (same as X86-32+X86-64)
 121	XEN	Xen support is enabled
 122
 123In addition, the following text indicates that the option:
 124
 125	BUGS=	Relates to possible processor bugs on the said processor.
 126	KNL	Is a kernel start-up parameter.
 127	BOOT	Is a boot loader parameter.
 128
 129Parameters denoted with BOOT are actually interpreted by the boot
 130loader, and have no meaning to the kernel directly.
 131Do not modify the syntax of boot loader parameters without extreme
 132need or coordination with <Documentation/x86/boot.txt>.
 133
 134There are also arch-specific kernel-parameters not documented here.
 135See for example <Documentation/x86/x86_64/boot-options.txt>.
 136
 137Note that ALL kernel parameters listed below are CASE SENSITIVE, and that
 138a trailing = on the name of any parameter states that that parameter will
 139be entered as an environment variable, whereas its absence indicates that
 140it will appear as a kernel argument readable via /proc/cmdline by programs
 141running once the system is up.
 142
 143The number of kernel parameters is not limited, but the length of the
 144complete command line (parameters including spaces etc.) is limited to
 145a fixed number of characters. This limit depends on the architecture
 146and is between 256 and 4096 characters. It is defined in the file
 147./include/asm/setup.h as COMMAND_LINE_SIZE.
 148
 149Finally, the [KMG] suffix is commonly described after a number of kernel
 150parameter values. These 'K', 'M', and 'G' letters represent the _binary_
 151multipliers 'Kilo', 'Mega', and 'Giga', equalling 2^10, 2^20, and 2^30
 152bytes respectively. Such letter suffixes can also be entirely omitted.
 153
 154
 155	acpi=		[HW,ACPI,X86]
 156			Advanced Configuration and Power Interface
 157			Format: { force | off | strict | noirq | rsdt }
 158			force -- enable ACPI if default was off
 159			off -- disable ACPI if default was on
 160			noirq -- do not use ACPI for IRQ routing
 161			strict -- Be less tolerant of platforms that are not
 162				strictly ACPI specification compliant.
 163			rsdt -- prefer RSDT over (default) XSDT
 164			copy_dsdt -- copy DSDT to memory
 165
 166			See also Documentation/power/pm.txt, pci=noacpi
 167
 168	acpi_rsdp=	[ACPI,EFI,KEXEC]
 169			Pass the RSDP address to the kernel, mostly used
 170			on machines running EFI runtime service to boot the
 171			second kernel for kdump.
 172
 173	acpi_apic_instance=	[ACPI, IOAPIC]
 174			Format: <int>
 175			2: use 2nd APIC table, if available
 176			1,0: use 1st APIC table
 177			default: 0
 178
 179	acpi_backlight=	[HW,ACPI]
 180			acpi_backlight=vendor
 181			acpi_backlight=video
 182			If set to vendor, prefer vendor specific driver
 183			(e.g. thinkpad_acpi, sony_acpi, etc.) instead
 184			of the ACPI video.ko driver.
 185
 186	acpi.debug_layer=	[HW,ACPI,ACPI_DEBUG]
 187	acpi.debug_level=	[HW,ACPI,ACPI_DEBUG]
 188			Format: <int>
 189			CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
 190			debug output.  Bits in debug_layer correspond to a
 191			_COMPONENT in an ACPI source file, e.g.,
 192			    #define _COMPONENT ACPI_PCI_COMPONENT
 193			Bits in debug_level correspond to a level in
 194			ACPI_DEBUG_PRINT statements, e.g.,
 195			    ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
 196			The debug_level mask defaults to "info".  See
 197			Documentation/acpi/debug.txt for more information about
 198			debug layers and levels.
 199
 200			Enable processor driver info messages:
 201			    acpi.debug_layer=0x20000000
 202			Enable PCI/PCI interrupt routing info messages:
 203			    acpi.debug_layer=0x400000
 204			Enable AML "Debug" output, i.e., stores to the Debug
 205			object while interpreting AML:
 206			    acpi.debug_layer=0xffffffff acpi.debug_level=0x2
 207			Enable all messages related to ACPI hardware:
 208			    acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
 209
 210			Some values produce so much output that the system is
 211			unusable.  The "log_buf_len" parameter may be useful
 212			if you need to capture more output.
 213
 214	acpi_irq_balance [HW,ACPI]
 215			ACPI will balance active IRQs
 216			default in APIC mode
 217
 218	acpi_irq_nobalance [HW,ACPI]
 219			ACPI will not move active IRQs (default)
 220			default in PIC mode
 221
 222	acpi_irq_isa=	[HW,ACPI] If irq_balance, mark listed IRQs used by ISA
 223			Format: <irq>,<irq>...
 224
 225	acpi_irq_pci=	[HW,ACPI] If irq_balance, clear listed IRQs for
 226			use by PCI
 227			Format: <irq>,<irq>...
 228
 
 
 
 
 
 
 
 
 229	acpi_no_auto_ssdt	[HW,ACPI] Disable automatic loading of SSDT
 230
 
 
 
 
 
 
 
 
 231	acpi_os_name=	[HW,ACPI] Tell ACPI BIOS the name of the OS
 232			Format: To spoof as Windows 98: ="Microsoft Windows"
 233
 234	acpi_osi=	[HW,ACPI] Modify list of supported OS interface strings
 235			acpi_osi="string1"	# add string1 -- only one string
 236			acpi_osi="!string2"	# remove built-in string2
 
 
 
 237			acpi_osi=		# disable all strings
 238
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 239	acpi_pm_good	[X86]
 240			Override the pmtimer bug detection: force the kernel
 241			to assume that this machine's pmtimer latches its value
 242			and always returns good values.
 243
 244	acpi_sci=	[HW,ACPI] ACPI System Control Interrupt trigger mode
 245			Format: { level | edge | high | low }
 246
 247	acpi_serialize	[HW,ACPI] force serialization of AML methods
 248
 249	acpi_skip_timer_override [HW,ACPI]
 250			Recognize and ignore IRQ0/pin2 Interrupt Override.
 251			For broken nForce2 BIOS resulting in XT-PIC timer.
 252
 253	acpi_sleep=	[HW,ACPI] Sleep options
 254			Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
 255				  old_ordering, nonvs, sci_force_enable }
 256			See Documentation/power/video.txt for information on
 257			s3_bios and s3_mode.
 258			s3_beep is for debugging; it makes the PC's speaker beep
 259			as soon as the kernel's real-mode entry point is called.
 260			s4_nohwsig prevents ACPI hardware signature from being
 261			used during resume from hibernation.
 262			old_ordering causes the ACPI 1.0 ordering of the _PTS
 263			control method, with respect to putting devices into
 264			low power states, to be enforced (the ACPI 2.0 ordering
 265			of _PTS is used by default).
 266			nonvs prevents the kernel from saving/restoring the
 267			ACPI NVS memory during suspend/hibernation and resume.
 268			sci_force_enable causes the kernel to set SCI_EN directly
 269			on resume from S1/S3 (which is against the ACPI spec,
 270			but some broken systems don't work without it).
 271
 272	acpi_use_timer_override [HW,ACPI]
 273			Use timer override. For some broken Nvidia NF5 boards
 274			that require a timer override, but don't have HPET
 275
 276	acpi_enforce_resources=	[ACPI]
 277			{ strict | lax | no }
 278			Check for resource conflicts between native drivers
 279			and ACPI OperationRegions (SystemIO and SystemMemory
 280			only). IO ports and memory declared in ACPI might be
 281			used by the ACPI subsystem in arbitrary AML code and
 282			can interfere with legacy drivers.
 283			strict (default): access to resources claimed by ACPI
 284			is denied; legacy drivers trying to access reserved
 285			resources will fail to bind to device using them.
 286			lax: access to resources claimed by ACPI is allowed;
 287			legacy drivers trying to access reserved resources
 288			will bind successfully but a warning message is logged.
 289			no: ACPI OperationRegions are not marked as reserved,
 290			no further checks are performed.
 291
 
 
 
 292	add_efi_memmap	[EFI; X86] Include EFI memory map in
 293			kernel's map of available physical RAM.
 294
 295	agp=		[AGP]
 296			{ off | try_unsupported }
 297			off: disable AGP support
 298			try_unsupported: try to drive unsupported chipsets
 299				(may crash computer or cause data corruption)
 300
 301	ALSA		[HW,ALSA]
 302			See Documentation/sound/alsa/alsa-parameters.txt
 303
 304	alignment=	[KNL,ARM]
 305			Allow the default userspace alignment fault handler
 306			behaviour to be specified.  Bit 0 enables warnings,
 307			bit 1 enables fixups, and bit 2 sends a segfault.
 308
 309	amd_iommu=	[HW,X86-84]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 310			Pass parameters to the AMD IOMMU driver in the system.
 311			Possible values are:
 312			fullflush - enable flushing of IO/TLB entries when
 313				    they are unmapped. Otherwise they are
 314				    flushed before they will be reused, which
 315				    is a lot of faster
 316			off	  - do not initialize any AMD IOMMU found in
 317				    the system
 
 
 
 
 
 
 
 
 
 
 
 318
 319	amijoy.map=	[HW,JOY] Amiga joystick support
 320			Map of devices attached to JOY0DAT and JOY1DAT
 321			Format: <a>,<b>
 322			See also Documentation/kernel/input/joystick.txt
 323
 324	analog.map=	[HW,JOY] Analog joystick and gamepad support
 325			Specifies type or capabilities of an analog joystick
 326			connected to one of 16 gameports
 327			Format: <type1>,<type2>,..<type16>
 328
 329	apc=		[HW,SPARC]
 330			Power management functions (SPARCstation-4/5 + deriv.)
 331			Format: noidle
 332			Disable APC CPU standby support. SPARCstation-Fox does
 333			not play well with APC CPU idle - disable it if you have
 334			APC and your system crashes randomly.
 335
 336	apic=		[APIC,X86-32] Advanced Programmable Interrupt Controller
 337			Change the output verbosity whilst booting
 338			Format: { quiet (default) | verbose | debug }
 339			Change the amount of debugging information output
 340			when initialising the APIC and IO-APIC components.
 341
 342	autoconf=	[IPV6]
 343			See Documentation/networking/ipv6.txt.
 344
 345	show_lapic=	[APIC,X86] Advanced Programmable Interrupt Controller
 346			Limit apic dumping. The parameter defines the maximal
 347			number of local apics being dumped. Also it is possible
 348			to set it to "all" by meaning -- no limit here.
 349			Format: { 1 (default) | 2 | ... | all }.
 350			The parameter valid if only apic=debug or
 351			apic=verbose is specified.
 352			Example: apic=debug show_lapic=all
 353
 354	apm=		[APM] Advanced Power Management
 355			See header of arch/x86/kernel/apm_32.c.
 356
 357	arcrimi=	[HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
 358			Format: <io>,<irq>,<nodeID>
 359
 360	ataflop=	[HW,M68k]
 361
 362	atarimouse=	[HW,MOUSE] Atari Mouse
 363
 364	atkbd.extra=	[HW] Enable extra LEDs and keys on IBM RapidAccess,
 365			EzKey and similar keyboards
 366
 367	atkbd.reset=	[HW] Reset keyboard during initialization
 368
 369	atkbd.set=	[HW] Select keyboard code set
 370			Format: <int> (2 = AT (default), 3 = PS/2)
 371
 372	atkbd.scroll=	[HW] Enable scroll wheel on MS Office and similar
 373			keyboards
 374
 375	atkbd.softraw=	[HW] Choose between synthetic and real raw mode
 376			Format: <bool> (0 = real, 1 = synthetic (default))
 377
 378	atkbd.softrepeat= [HW]
 379			Use software keyboard repeat
 380
 381	autotest	[IA-64]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 382
 383	baycom_epp=	[HW,AX25]
 384			Format: <io>,<mode>
 385
 386	baycom_par=	[HW,AX25] BayCom Parallel Port AX.25 Modem
 387			Format: <io>,<mode>
 388			See header of drivers/net/hamradio/baycom_par.c.
 389
 390	baycom_ser_fdx=	[HW,AX25]
 391			BayCom Serial Port AX.25 Modem (Full Duplex Mode)
 392			Format: <io>,<irq>,<mode>[,<baud>]
 393			See header of drivers/net/hamradio/baycom_ser_fdx.c.
 394
 395	baycom_ser_hdx=	[HW,AX25]
 396			BayCom Serial Port AX.25 Modem (Half Duplex Mode)
 397			Format: <io>,<irq>,<mode>
 398			See header of drivers/net/hamradio/baycom_ser_hdx.c.
 399
 
 
 
 
 400	boot_delay=	Milliseconds to delay each printk during boot.
 401			Values larger than 10 seconds (10000) are changed to
 402			no delay (0).
 403			Format: integer
 404
 405	bootmem_debug	[KNL] Enable bootmem allocator debug messages.
 406
 407	bttv.card=	[HW,V4L] bttv (bt848 + bt878 based grabber cards)
 408	bttv.radio=	Most important insmod options are available as
 409			kernel args too.
 410	bttv.pll=	See Documentation/video4linux/bttv/Insmod-options
 411	bttv.tuner=	and Documentation/video4linux/bttv/CARDLIST
 412
 413	bulk_remove=off	[PPC]  This parameter disables the use of the pSeries
 414			firmware feature for flushing multiple hpte entries
 415			at a time.
 416
 417	c101=		[NET] Moxa C101 synchronous serial card
 418
 419	cachesize=	[BUGS=X86-32] Override level 2 CPU cache size detection.
 420			Sometimes CPU hardware bugs make them report the cache
 421			size incorrectly. The kernel will attempt work arounds
 422			to fix known problems, but for some CPUs it is not
 423			possible to determine what the correct size should be.
 424			This option provides an override for these situations.
 425
 426	capability.disable=
 427			[SECURITY] Disable capabilities.  This would normally
 428			be used only if an alternative security model is to be
 429			configured.  Potentially dangerous and should only be
 430			used if you are entirely sure of the consequences.
 431
 432	ccw_timeout_log [S390]
 433			See Documentation/s390/CommonIO for details.
 434
 435	cgroup_disable= [KNL] Disable a particular controller
 436			Format: {name of the controller(s) to disable}
 437				{Currently supported controllers - "memory"}
 
 
 
 
 
 
 
 438
 439	checkreqprot	[SELINUX] Set initial checkreqprot flag value.
 440			Format: { "0" | "1" }
 441			See security/selinux/Kconfig help text.
 442			0 -- check protection applied by kernel (includes
 443				any implied execute protection).
 444			1 -- check protection requested by application.
 445			Default value is set via a kernel config option.
 446			Value can be changed at runtime via
 447				/selinux/checkreqprot.
 448
 449	cio_ignore=	[S390]
 450			See Documentation/s390/CommonIO for details.
 
 
 
 
 
 
 
 451
 452	clock=		[BUGS=X86-32, HW] gettimeofday clocksource override.
 453			[Deprecated]
 454			Forces specified clocksource (if available) to be used
 455			when calculating gettimeofday(). If specified
 456			clocksource is not available, it defaults to PIT.
 457			Format: { pit | tsc | cyclone | pmtmr }
 458
 459	clocksource=	Override the default clocksource
 460			Format: <string>
 461			Override the default clocksource and use the clocksource
 462			with the name specified.
 463			Some clocksource names to choose from, depending on
 464			the platform:
 465			[all] jiffies (this is the base, fallback clocksource)
 466			[ACPI] acpi_pm
 467			[ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
 468				pxa_timer,timer3,32k_counter,timer0_1
 469			[AVR32] avr32
 470			[X86-32] pit,hpet,tsc;
 471				scx200_hrt on Geode; cyclone on IBM x440
 472			[MIPS] MIPS
 473			[PARISC] cr16
 474			[S390] tod
 475			[SH] SuperH
 476			[SPARC64] tick
 477			[X86-64] hpet,tsc
 478
 479	clearcpuid=BITNUM [X86]
 480			Disable CPUID feature X for the kernel. See
 481			arch/x86/include/asm/cpufeature.h for the valid bit
 482			numbers. Note the Linux specific bits are not necessarily
 483			stable over kernel options, but the vendor specific
 484			ones should be.
 485			Also note that user programs calling CPUID directly
 486			or using the feature without checking anything
 487			will still see it. This just prevents it from
 488			being used by the kernel or shown in /proc/cpuinfo.
 489			Also note the kernel might malfunction if you disable
 490			some critical bits.
 491
 
 
 
 
 
 492	cmo_free_hint=	[PPC] Format: { yes | no }
 493			Specify whether pages are marked as being inactive
 494			when they are freed.  This is used in CMO environments
 495			to determine OS memory pressure for page stealing by
 496			a hypervisor.
 497			Default: yes
 498
 
 
 
 
 499	code_bytes	[X86] How many bytes of object code to print
 500			in an oops report.
 501			Range: 0 - 8192
 502			Default: 64
 503
 504	com20020=	[HW,NET] ARCnet - COM20020 chipset
 505			Format:
 506			<io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
 507
 508	com90io=	[HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
 509			Format: <io>[,<irq>]
 510
 511	com90xx=	[HW,NET]
 512			ARCnet - COM90xx chipset (memory-mapped buffers)
 513			Format: <io>[,<irq>[,<memstart>]]
 514
 515	condev=		[HW,S390] console device
 516	conmode=
 517
 518	console=	[KNL] Output console device and options.
 519
 520		tty<n>	Use the virtual console device <n>.
 521
 522		ttyS<n>[,options]
 523		ttyUSB0[,options]
 524			Use the specified serial port.  The options are of
 525			the form "bbbbpnf", where "bbbb" is the baud rate,
 526			"p" is parity ("n", "o", or "e"), "n" is number of
 527			bits, and "f" is flow control ("r" for RTS or
 528			omit it).  Default is "9600n8".
 529
 530			See Documentation/serial-console.txt for more
 531			information.  See
 532			Documentation/networking/netconsole.txt for an
 533			alternative.
 534
 535		uart[8250],io,<addr>[,options]
 536		uart[8250],mmio,<addr>[,options]
 537			Start an early, polled-mode console on the 8250/16550
 538			UART at the specified I/O port or MMIO address,
 539			switching to the matching ttyS device later.  The
 540			options are the same as for ttyS, above.
 
 
 541
 542                If the device connected to the port is not a TTY but a braille
 543                device, prepend "brl," before the device type, for instance
 544			console=brl,ttyS0
 545		For now, only VisioBraille is supported.
 546
 547	consoleblank=	[KNL] The console blank (screen saver) timeout in
 548			seconds. Defaults to 10*60 = 10mins. A value of 0
 549			disables the blank timer.
 550
 551	coredump_filter=
 552			[KNL] Change the default value for
 553			/proc/<pid>/coredump_filter.
 554			See also Documentation/filesystems/proc.txt.
 555
 556	cpuidle.off=1	[CPU_IDLE]
 557			disable the cpuidle sub-system
 558
 559	cpcihp_generic=	[HW,PCI] Generic port I/O CompactPCI driver
 560			Format:
 561			<first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
 562
 563	crashkernel=size[KMG][@offset[KMG]]
 564			[KNL] Using kexec, Linux can switch to a 'crash kernel'
 565			upon panic. This parameter reserves the physical
 566			memory region [offset, offset + size] for that kernel
 567			image. If '@offset' is omitted, then a suitable offset
 568			is selected automatically. Check
 569			Documentation/kdump/kdump.txt for further details.
 570
 571	crashkernel=range1:size1[,range2:size2,...][@offset]
 572			[KNL] Same as above, but depends on the memory
 573			in the running system. The syntax of range is
 574			start-[end] where start and end are both
 575			a memory unit (amount[KMG]). See also
 576			Documentation/kdump/kdump.txt for an example.
 577
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 578	cs89x0_dma=	[HW,NET]
 579			Format: <dma>
 580
 581	cs89x0_media=	[HW,NET]
 582			Format: { rj45 | aui | bnc }
 583
 584	dasd=		[HW,NET]
 585			See header of drivers/s390/block/dasd_devmap.c.
 586
 587	db9.dev[2|3]=	[HW,JOY] Multisystem joystick support via parallel port
 588			(one device per port)
 589			Format: <port#>,<type>
 590			See also Documentation/input/joystick-parport.txt
 591
 592	ddebug_query=   [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
 593			time. See Documentation/dynamic-debug-howto.txt for
 594			details.
 595
 596	debug		[KNL] Enable kernel debugging (events log level).
 597
 598	debug_locks_verbose=
 599			[KNL] verbose self-tests
 600			Format=<0|1>
 601			Print debugging info while doing the locking API
 602			self-tests.
 603			We default to 0 (no extra messages), setting it to
 604			1 will print _a lot_ more information - normally
 605			only useful to kernel developers.
 606
 607	debug_objects	[KNL] Enable object debugging
 608
 609	no_debug_objects
 610			[KNL] Disable object debugging
 611
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 612	debugpat	[X86] Enable PAT debugging
 613
 614	decnet.addr=	[HW,NET]
 615			Format: <area>[,<node>]
 616			See also Documentation/networking/decnet.txt.
 617
 618	default_hugepagesz=
 619			[same as hugepagesz=] The size of the default
 620			HugeTLB page size. This is the size represented by
 621			the legacy /proc/ hugepages APIs, used for SHM, and
 622			default size when mounting hugetlbfs filesystems.
 623			Defaults to the default architecture's huge page size
 624			if not specified.
 625
 626	dhash_entries=	[KNL]
 627			Set number of hash buckets for dentry cache.
 628
 629	digi=		[HW,SERIAL]
 630			IO parameters + enable/disable command.
 631
 632	digiepca=	[HW,SERIAL]
 633			See drivers/char/README.epca and
 634			Documentation/serial/digiepca.txt.
 635
 636	disable=	[IPV6]
 637			See Documentation/networking/ipv6.txt.
 638
 
 
 
 
 
 
 
 
 
 639	disable_ddw     [PPC/PSERIES]
 640			Disable Dynamic DMA Window support. Use this if
 641			to workaround buggy firmware.
 642
 643	disable_ipv6=	[IPV6]
 644			See Documentation/networking/ipv6.txt.
 645
 646	disable_mtrr_cleanup [X86]
 647			The kernel tries to adjust MTRR layout from continuous
 648			to discrete, to make X server driver able to add WB
 649			entry later. This parameter disables that.
 650
 651	disable_mtrr_trim [X86, Intel and AMD only]
 652			By default the kernel will trim any uncacheable
 653			memory out of your available memory pool based on
 654			MTRR settings.  This parameter disables that behavior,
 655			possibly causing your machine to run very slowly.
 656
 657	disable_timer_pin_1 [X86]
 658			Disable PIN 1 of APIC timer
 659			Can be useful to work around chipset bugs.
 660
 661	dma_debug=off	If the kernel is compiled with DMA_API_DEBUG support,
 662			this option disables the debugging code at boot.
 663
 664	dma_debug_entries=<number>
 665			This option allows to tune the number of preallocated
 666			entries for DMA-API debugging code. One entry is
 667			required per DMA-API allocation. Use this if the
 668			DMA-API debugging code disables itself because the
 669			architectural default is too low.
 670
 671	dma_debug_driver=<driver_name>
 672			With this option the DMA-API debugging driver
 673			filter feature can be enabled at boot time. Just
 674			pass the driver to filter for as the parameter.
 675			The filter can be disabled or changed to another
 676			driver later using sysfs.
 677
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 678	dscc4.setup=	[NET]
 679
 
 
 
 
 
 
 
 
 
 
 680	earlycon=	[KNL] Output early console device and options.
 681		uart[8250],io,<addr>[,options]
 682		uart[8250],mmio,<addr>[,options]
 683		uart[8250],mmio32,<addr>[,options]
 684			Start an early, polled-mode console on the 8250/16550
 685			UART at the specified I/O port or MMIO address.
 686			MMIO inter-register address stride is either 8-bit
 687			(mmio) or 32-bit (mmio32).
 688			The options are the same as for ttyS, above.
 689
 690	earlyprintk=	[X86,SH,BLACKFIN]
 691			earlyprintk=vga
 
 
 692			earlyprintk=serial[,ttySn[,baudrate]]
 
 693			earlyprintk=ttySn[,baudrate]
 694			earlyprintk=dbgp[debugController#]
 695
 
 
 
 
 696			Append ",keep" to not disable it when the real console
 697			takes over.
 698
 699			Only vga or serial or usb debug port at a time.
 
 700
 701			Currently only ttyS0 and ttyS1 are supported.
 
 
 
 
 
 
 
 702
 703			Interaction with the standard serial driver is not
 704			very good.
 705
 706			The VGA output is eventually overwritten by the real
 707			console.
 
 
 
 
 
 
 
 
 
 
 708
 709	ekgdboc=	[X86,KGDB] Allow early kernel console debugging
 710			ekgdboc=kbd
 711
 712			This is designed to be used in conjunction with
 713			the boot argument: earlyprintk=vga
 714
 715	edd=		[EDD]
 716			Format: {"off" | "on" | "skip[mbr]"}
 717
 
 
 
 
 
 
 
 
 
 
 
 
 718	eisa_irq_edge=	[PARISC,HW]
 719			See header of drivers/parisc/eisa.c.
 720
 721	elanfreq=	[X86-32]
 722			See comment before function elanfreq_setup() in
 723			arch/x86/kernel/cpu/cpufreq/elanfreq.c.
 724
 725	elevator=	[IOSCHED]
 726			Format: {"cfq" | "deadline" | "noop"}
 727			See Documentation/block/as-iosched.txt and
 728			Documentation/block/deadline-iosched.txt for details.
 729
 730	elfcorehdr=	[IA-64,PPC,SH,X86]
 731			Specifies physical address of start of kernel core
 732			image elf header. Generally kexec loader will
 733			pass this option to capture kernel.
 734			See Documentation/kdump/kdump.txt for details.
 735
 736	enable_mtrr_cleanup [X86]
 737			The kernel tries to adjust MTRR layout from continuous
 738			to discrete, to make X server driver able to add WB
 739			entry later. This parameter enables that.
 740
 741	enable_timer_pin_1 [X86]
 742			Enable PIN 1 of APIC timer
 743			Can be useful to work around chipset bugs
 744			(in particular on some ATI chipsets).
 745			The kernel tries to set a reasonable default.
 746
 747	enforcing	[SELINUX] Set initial enforcing status.
 748			Format: {"0" | "1"}
 749			See security/selinux/Kconfig help text.
 750			0 -- permissive (log only, no denials).
 751			1 -- enforcing (deny and log).
 752			Default value is 0.
 753			Value can be changed at runtime via /selinux/enforce.
 754
 755	erst_disable	[ACPI]
 756			Disable Error Record Serialization Table (ERST)
 757			support.
 758
 759	ether=		[HW,NET] Ethernet cards parameters
 760			This option is obsoleted by the "netdev=" option, which
 761			has equivalent usage. See its documentation for details.
 762
 
 
 
 
 
 763	failslab=
 764	fail_page_alloc=
 765	fail_make_request=[KNL]
 766			General fault injection mechanism.
 767			Format: <interval>,<probability>,<space>,<times>
 768			See also /Documentation/fault-injection/.
 769
 770	floppy=		[HW]
 771			See Documentation/blockdev/floppy.txt.
 772
 773	force_pal_cache_flush
 774			[IA-64] Avoid check_sal_cache_flush which may hang on
 775			buggy SAL_CACHE_FLUSH implementations. Using this
 776			parameter will force ia64_sal_cache_flush to call
 777			ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
 778
 
 
 
 
 
 
 
 779	ftrace=[tracer]
 780			[FTRACE] will set and start the specified tracer
 781			as early as possible in order to facilitate early
 782			boot debugging.
 783
 784	ftrace_dump_on_oops[=orig_cpu]
 785			[FTRACE] will dump the trace buffers on oops.
 786			If no parameter is passed, ftrace will dump
 787			buffers of all CPUs, but if you pass orig_cpu, it will
 788			dump only the buffer of the CPU that triggered the
 789			oops.
 790
 791	ftrace_filter=[function-list]
 792			[FTRACE] Limit the functions traced by the function
 793			tracer at boot up. function-list is a comma separated
 794			list of functions. This list can be changed at run
 795			time by the set_ftrace_filter file in the debugfs
 796			tracing directory.
 797
 798	ftrace_notrace=[function-list]
 799			[FTRACE] Do not trace the functions specified in
 800			function-list. This list can be changed at run time
 801			by the set_ftrace_notrace file in the debugfs
 802			tracing directory.
 803
 804	ftrace_graph_filter=[function-list]
 805			[FTRACE] Limit the top level callers functions traced
 806			by the function graph tracer at boot up.
 807			function-list is a comma separated list of functions
 808			that can be changed at run time by the
 809			set_graph_function file in the debugfs tracing directory.
 810
 811	gamecon.map[2|3]=
 812			[HW,JOY] Multisystem joystick and NES/SNES/PSX pad
 813			support via parallel port (up to 5 devices per port)
 814			Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
 815			See also Documentation/input/joystick-parport.txt
 816
 817	gamma=		[HW,DRM]
 818
 819	gart_fix_e820=  [X86_64] disable the fix e820 for K8 GART
 820			Format: off | on
 821			default: on
 822
 823	gcov_persist=	[GCOV] When non-zero (default), profiling data for
 824			kernel modules is saved and remains accessible via
 825			debugfs, even when the module is unloaded/reloaded.
 826			When zero, profiling data is discarded and associated
 827			debugfs files are removed at module unload time.
 828
 829	gpt		[EFI] Forces disk with valid GPT signature but
 830			invalid Protective MBR to be treated as GPT.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 831
 832	hashdist=	[KNL,NUMA] Large hashes allocated during boot
 833			are distributed across NUMA nodes.  Defaults on
 834			for 64-bit NUMA, off otherwise.
 835			Format: 0 | 1 (for off | on)
 836
 837	hcl=		[IA-64] SGI's Hardware Graph compatibility layer
 838
 839	hd=		[EIDE] (E)IDE hard drive subsystem geometry
 840			Format: <cyl>,<head>,<sect>
 841
 842	hest_disable	[ACPI]
 843			Disable Hardware Error Source Table (HEST) support;
 844			corresponding firmware-first mode error processing
 845			logic will be disabled.
 846
 847	highmem=nn[KMG]	[KNL,BOOT] forces the highmem zone to have an exact
 848			size of <nn>. This works even on boxes that have no
 849			highmem otherwise. This also works to reduce highmem
 850			size on bigger boxes.
 851
 852	highres=	[KNL] Enable/disable high resolution timer mode.
 853			Valid parameters: "on", "off"
 854			Default: "on"
 855
 856	hisax=		[HW,ISDN]
 857			See Documentation/isdn/README.HiSax.
 858
 859	hlt		[BUGS=ARM,SH]
 860
 861	hpet=		[X86-32,HPET] option to control HPET usage
 862			Format: { enable (default) | disable | force |
 863				verbose }
 864			disable: disable HPET and use PIT instead
 865			force: allow force enabled of undocumented chips (ICH4,
 866				VIA, nVidia)
 867			verbose: show contents of HPET registers during setup
 868
 
 
 
 869	hugepages=	[HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
 870	hugepagesz=	[HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
 871			On x86-64 and powerpc, this option can be specified
 872			multiple times interleaved with hugepages= to reserve
 873			huge pages of different sizes. Valid pages sizes on
 874			x86-64 are 2M (when the CPU supports "pse") and 1G
 875			(when the CPU supports the "pdpe1gb" cpuinfo flag)
 876			Note that 1GB pages can only be allocated at boot time
 877			using hugepages= and not freed afterwards.
 878
 879	hvc_iucv=	[S390] Number of z/VM IUCV hypervisor console (HVC)
 880			       terminal devices. Valid values: 0..8
 881	hvc_iucv_allow=	[S390] Comma-separated list of z/VM user IDs.
 882			       If specified, z/VM IUCV HVC accepts connections
 883			       from listed z/VM user IDs only.
 884
 
 
 
 
 885	keep_bootcon	[KNL]
 886			Do not unregister boot console at start. This is only
 887			useful for debugging when something happens in the window
 888			between unregistering the boot console and initializing
 889			the real console.
 890
 891	i2c_bus=	[HW] Override the default board specific I2C bus speed
 892			     or register an additional I2C bus that is not
 893			     registered from board initialization code.
 894			     Format:
 895			     <bus_id>,<clkrate>
 896
 897	i8042.debug	[HW] Toggle i8042 debug mode
 898	i8042.direct	[HW] Put keyboard port into non-translated mode
 899	i8042.dumbkbd	[HW] Pretend that controller can only read data from
 900			     keyboard and cannot control its state
 901			     (Don't attempt to blink the leds)
 902	i8042.noaux	[HW] Don't check for auxiliary (== mouse) port
 903	i8042.nokbd	[HW] Don't check/create keyboard port
 904	i8042.noloop	[HW] Disable the AUX Loopback command while probing
 905			     for the AUX port
 906	i8042.nomux	[HW] Don't check presence of an active multiplexing
 907			     controller
 908	i8042.nopnp	[HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
 909			     controllers
 910	i8042.notimeout	[HW] Ignore timeout condition signalled by conroller
 911	i8042.reset	[HW] Reset the controller during init and cleanup
 912	i8042.unlock	[HW] Unlock (ignore) the keylock
 913
 914	i810=		[HW,DRM]
 915
 916	i8k.ignore_dmi	[HW] Continue probing hardware even if DMI data
 917			indicates that the driver is running on unsupported
 918			hardware.
 919	i8k.force	[HW] Activate i8k driver even if SMM BIOS signature
 920			does not match list of supported models.
 921	i8k.power_status
 922			[HW] Report power status in /proc/i8k
 923			(disabled by default)
 924	i8k.restricted	[HW] Allow controlling fans only if SYS_ADMIN
 925			capability is set.
 926
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 927	icn=		[HW,ISDN]
 928			Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
 929
 930	ide-core.nodma=	[HW] (E)IDE subsystem
 931			Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
 932			.vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
 933			.cdrom .chs .ignore_cable are additional options
 934			See Documentation/ide/ide.txt.
 935
 936	ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
 937			Claim all unknown PCI IDE storage controllers.
 938
 939	idle=		[X86]
 940			Format: idle=poll, idle=mwait, idle=halt, idle=nomwait
 941			Poll forces a polling idle loop that can slightly
 942			improve the performance of waking up a idle CPU, but
 943			will use a lot of power and make the system run hot.
 944			Not recommended.
 945			idle=mwait: On systems which support MONITOR/MWAIT but
 946			the kernel chose to not use it because it doesn't save
 947			as much power as a normal idle loop, use the
 948			MONITOR/MWAIT idle loop anyways. Performance should be
 949			the same as idle=poll.
 950			idle=halt: Halt is forced to be used for CPU idle.
 951			In such case C2/C3 won't be used again.
 952			idle=nomwait: Disable mwait for CPU C-states
 953
 954	ignore_loglevel	[KNL]
 955			Ignore loglevel setting - this will print /all/
 956			kernel messages to the console. Useful for debugging.
 
 
 
 957
 958	ihash_entries=	[KNL]
 959			Set number of hash buckets for inode cache.
 960
 961	ima_audit=	[IMA]
 962			Format: { "0" | "1" }
 963			0 -- integrity auditing messages. (Default)
 964			1 -- enable informational integrity auditing messages.
 
 
 
 965
 966	ima_hash=	[IMA]
 967			Format: { "sha1" | "md5" }
 
 968			default: "sha1"
 969
 
 
 
 970	ima_tcb		[IMA]
 971			Load a policy which meets the needs of the Trusted
 972			Computing Base.  This means IMA will measure all
 973			programs exec'd, files mmap'd for exec, and all files
 974			opened for read by uid=0.
 975
 
 
 
 
 
 976	init=		[KNL]
 977			Format: <full_path>
 978			Run specified binary instead of /sbin/init as init
 979			process.
 980
 981	initcall_debug	[KNL] Trace initcalls as they are executed.  Useful
 982			for working out where the kernel is dying during
 983			startup.
 984
 985	initrd=		[BOOT] Specify the location of the initial ramdisk
 986
 987	inport.irq=	[HW] Inport (ATI XL and Microsoft) busmouse driver
 988			Format: <irq>
 989
 
 
 
 
 
 
 
 990	intel_iommu=	[DMAR] Intel IOMMU driver (DMAR) option
 991		on
 992			Enable intel iommu driver.
 993		off
 994			Disable intel iommu driver.
 995		igfx_off [Default Off]
 996			By default, gfx is mapped as normal device. If a gfx
 997			device has a dedicated DMAR unit, the DMAR unit is
 998			bypassed by not enabling DMAR with this option. In
 999			this case, gfx device will use physical address for
1000			DMA.
1001		forcedac [x86_64]
1002			With this option iommu will not optimize to look
1003			for io virtual address below 32-bit forcing dual
1004			address cycle on pci bus for cards supporting greater
1005			than 32-bit addressing. The default is to look
1006			for translation below 32-bit and if not available
1007			then look in the higher range.
1008		strict [Default Off]
1009			With this option on every unmap_single operation will
1010			result in a hardware IOTLB flush operation as opposed
1011			to batching them for performance.
1012		sp_off [Default Off]
1013			By default, super page will be supported if Intel IOMMU
1014			has the capability. With this option, super page will
1015			not be supported.
 
 
 
 
 
 
 
 
 
 
1016	intremap=	[X86-64, Intel-IOMMU]
1017			Format: { on (default) | off | nosid }
1018			on	enable Interrupt Remapping (default)
1019			off	disable Interrupt Remapping
1020			nosid	disable Source ID checking
1021
1022	inttest=	[IA-64]
1023
1024	iomem=		Disable strict checking of access to MMIO memory
1025		strict	regions from userspace.
1026		relaxed
1027
1028	iommu=		[x86]
1029		off
1030		force
1031		noforce
1032		biomerge
1033		panic
1034		nopanic
1035		merge
1036		nomerge
1037		forcesac
1038		soft
1039		pt	[x86, IA-64]
 
1040
1041	io7=		[HW] IO7 for Marvel based alpha systems
1042			See comment before marvel_specify_io7 in
1043			arch/alpha/kernel/core_marvel.c.
1044
1045	io_delay=	[X86] I/O delay method
1046		0x80
1047			Standard port 0x80 based delay
1048		0xed
1049			Alternate port 0xed based delay (needed on some systems)
1050		udelay
1051			Simple two microseconds delay
1052		none
1053			No delay
1054
1055	ip=		[IP_PNP]
1056			See Documentation/filesystems/nfs/nfsroot.txt.
1057
1058	ip2=		[HW] Set IO/IRQ pairs for up to 4 IntelliPort boards
1059			See comment before ip2_setup() in
1060			drivers/char/ip2/ip2base.c.
1061
1062	irqfixup	[HW]
1063			When an interrupt is not handled search all handlers
1064			for it. Intended to get systems with badly broken
1065			firmware running.
1066
1067	irqpoll		[HW]
1068			When an interrupt is not handled search all handlers
1069			for it. Also check all handlers each timer
1070			interrupt. Intended to get systems with badly broken
1071			firmware running.
1072
1073	isapnp=		[ISAPNP]
1074			Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1075
1076	isolcpus=	[KNL,SMP] Isolate CPUs from the general scheduler.
1077			Format:
1078			<cpu number>,...,<cpu number>
1079			or
1080			<cpu number>-<cpu number>
1081			(must be a positive range in ascending order)
1082			or a mixture
1083			<cpu number>,...,<cpu number>-<cpu number>
1084
1085			This option can be used to specify one or more CPUs
1086			to isolate from the general SMP balancing and scheduling
1087			algorithms. You can move a process onto or off an
1088			"isolated" CPU via the CPU affinity syscalls or cpuset.
1089			<cpu number> begins at 0 and the maximum value is
1090			"number of CPUs in system - 1".
1091
1092			This option is the preferred way to isolate CPUs. The
1093			alternative -- manually setting the CPU mask of all
1094			tasks in the system -- can cause problems and
1095			suboptimal load balancer performance.
1096
1097	iucv=		[HW,NET]
1098
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1099	js=		[HW,JOY] Analog joystick
1100			See Documentation/input/joystick.txt.
1101
1102	keepinitrd	[HW,ARM]
1103
1104	kernelcore=nn[KMG]	[KNL,X86,IA-64,PPC] This parameter
1105			specifies the amount of memory usable by the kernel
1106			for non-movable allocations.  The requested amount is
1107			spread evenly throughout all nodes in the system. The
1108			remaining memory in each node is used for Movable
1109			pages. In the event, a node is too small to have both
1110			kernelcore and Movable pages, kernelcore pages will
1111			take priority and other nodes will have a larger number
1112			of kernelcore pages.  The Movable zone is used for the
1113			allocation of pages that may be reclaimed or moved
1114			by the page migration subsystem.  This means that
1115			HugeTLB pages may not be allocated from this zone.
1116			Note that allocations like PTEs-from-HighMem still
1117			use the HighMem zone if it exists, and the Normal
1118			zone if it does not.
1119
1120	kgdbdbgp=	[KGDB,HW] kgdb over EHCI usb debug port.
1121			Format: <Controller#>[,poll interval]
1122			The controller # is the number of the ehci usb debug
1123			port as it is probed via PCI.  The poll interval is
1124			optional and is the number seconds in between
1125			each poll cycle to the debug port in case you need
1126			the functionality for interrupting the kernel with
1127			gdb or control-c on the dbgp connection.  When
1128			not using this parameter you use sysrq-g to break into
1129			the kernel debugger.
1130
1131	kgdboc=		[KGDB,HW] kgdb over consoles.
1132			Requires a tty driver that supports console polling,
1133			or a supported polling keyboard driver (non-usb).
1134			 Serial only format: <serial_device>[,baud]
1135			 keyboard only format: kbd
1136			 keyboard and serial format: kbd,<serial_device>[,baud]
1137			Optional Kernel mode setting:
1138			 kms, kbd format: kms,kbd
1139			 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
1140
1141	kgdbwait	[KGDB] Stop kernel execution and enter the
1142			kernel debugger at the earliest opportunity.
1143
1144	kmac=		[MIPS] korina ethernet MAC address.
1145			Configure the RouterBoard 532 series on-chip
1146			Ethernet adapter MAC address.
1147
1148	kmemleak=	[KNL] Boot-time kmemleak enable/disable
1149			Valid arguments: on, off
1150			Default: on
1151
 
 
 
 
 
 
 
1152	kstack=N	[X86] Print N words from the kernel stack
1153			in oops dumps.
1154
1155	kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
1156			Default is 0 (don't ignore, but inject #GP)
1157
1158	kvm.oos_shadow=	[KVM] Disable out-of-sync shadow paging.
1159			Default is 1 (enabled)
1160
1161	kvm.mmu_audit=	[KVM] This is a R/W parameter which allows audit
1162			KVM MMU at runtime.
1163			Default is 0 (off)
1164
1165	kvm-amd.nested=	[KVM,AMD] Allow nested virtualization in KVM/SVM.
1166			Default is 1 (enabled)
1167
1168	kvm-amd.npt=	[KVM,AMD] Disable nested paging (virtualized MMU)
1169			for all guests.
1170			Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
1171
1172	kvm-intel.ept=	[KVM,Intel] Disable extended page tables
1173			(virtualized MMU) support on capable Intel chips.
1174			Default is 1 (enabled)
1175
1176	kvm-intel.emulate_invalid_guest_state=
1177			[KVM,Intel] Enable emulation of invalid guest states
1178			Default is 0 (disabled)
1179
1180	kvm-intel.flexpriority=
1181			[KVM,Intel] Disable FlexPriority feature (TPR shadow).
1182			Default is 1 (enabled)
1183
 
 
 
 
1184	kvm-intel.unrestricted_guest=
1185			[KVM,Intel] Disable unrestricted guest feature
1186			(virtualized real and unpaged mode) on capable
1187			Intel chips. Default is 1 (enabled)
1188
1189	kvm-intel.vpid=	[KVM,Intel] Disable Virtual Processor Identification
1190			feature (tagged TLBs) on capable Intel chips.
1191			Default is 1 (enabled)
1192
1193	l2cr=		[PPC]
1194
1195	l3cr=		[PPC]
1196
1197	lapic		[X86-32,APIC] Enable the local APIC even if BIOS
1198			disabled it.
1199
 
 
 
 
1200	lapic_timer_c2_ok	[X86,APIC] trust the local apic timer
1201			in C2 power state.
1202
1203	libata.dma=	[LIBATA] DMA control
1204			libata.dma=0	  Disable all PATA and SATA DMA
1205			libata.dma=1	  PATA and SATA Disk DMA only
1206			libata.dma=2	  ATAPI (CDROM) DMA only
1207			libata.dma=4	  Compact Flash DMA only
1208			Combinations also work, so libata.dma=3 enables DMA
1209			for disks and CDROMs, but not CFs.
1210
1211	libata.ignore_hpa=	[LIBATA] Ignore HPA limit
1212			libata.ignore_hpa=0	  keep BIOS limits (default)
1213			libata.ignore_hpa=1	  ignore limits, using full disk
1214
1215	libata.noacpi	[LIBATA] Disables use of ACPI in libata suspend/resume
1216			when set.
1217			Format: <int>
1218
1219	libata.force=	[LIBATA] Force configurations.  The format is comma
1220			separated list of "[ID:]VAL" where ID is
1221			PORT[.DEVICE].  PORT and DEVICE are decimal numbers
1222			matching port, link or device.  Basically, it matches
1223			the ATA ID string printed on console by libata.  If
1224			the whole ID part is omitted, the last PORT and DEVICE
1225			values are used.  If ID hasn't been specified yet, the
1226			configuration applies to all ports, links and devices.
1227
1228			If only DEVICE is omitted, the parameter applies to
1229			the port and all links and devices behind it.  DEVICE
1230			number of 0 either selects the first device or the
1231			first fan-out link behind PMP device.  It does not
1232			select the host link.  DEVICE number of 15 selects the
1233			host link and device attached to it.
1234
1235			The VAL specifies the configuration to force.  As long
1236			as there's no ambiguity shortcut notation is allowed.
1237			For example, both 1.5 and 1.5G would work for 1.5Gbps.
1238			The following configurations can be forced.
1239
1240			* Cable type: 40c, 80c, short40c, unk, ign or sata.
1241			  Any ID with matching PORT is used.
1242
1243			* SATA link speed limit: 1.5Gbps or 3.0Gbps.
1244
1245			* Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
1246			  udma[/][16,25,33,44,66,100,133] notation is also
1247			  allowed.
1248
1249			* [no]ncq: Turn on or off NCQ.
1250
1251			* nohrst, nosrst, norst: suppress hard, soft
1252                          and both resets.
1253
 
 
 
1254			* dump_id: dump IDENTIFY data.
1255
 
 
 
 
1256			If there are multiple matching configurations changing
1257			the same attribute, the last one is used.
1258
1259	memblock=debug	[KNL] Enable memblock debug messages.
1260
1261	load_ramdisk=	[RAM] List of ramdisks to load from floppy
1262			See Documentation/blockdev/ramdisk.txt.
1263
1264	lockd.nlm_grace_period=P  [NFS] Assign grace period.
1265			Format: <integer>
1266
1267	lockd.nlm_tcpport=N	[NFS] Assign TCP port.
1268			Format: <integer>
1269
1270	lockd.nlm_timeout=T	[NFS] Assign timeout value.
1271			Format: <integer>
1272
1273	lockd.nlm_udpport=M	[NFS] Assign UDP port.
1274			Format: <integer>
1275
1276	logibm.irq=	[HW,MOUSE] Logitech Bus Mouse Driver
1277			Format: <irq>
1278
1279	loglevel=	All Kernel Messages with a loglevel smaller than the
1280			console loglevel will be printed to the console. It can
1281			also be changed with klogd or other programs. The
1282			loglevels are defined as follows:
1283
1284			0 (KERN_EMERG)		system is unusable
1285			1 (KERN_ALERT)		action must be taken immediately
1286			2 (KERN_CRIT)		critical conditions
1287			3 (KERN_ERR)		error conditions
1288			4 (KERN_WARNING)	warning conditions
1289			5 (KERN_NOTICE)		normal but significant condition
1290			6 (KERN_INFO)		informational
1291			7 (KERN_DEBUG)		debug-level messages
1292
1293	log_buf_len=n[KMG]	Sets the size of the printk ring buffer,
1294			in bytes.  n must be a power of two.  The default
1295			size is set in the kernel config file.
1296
1297	logo.nologo	[FB] Disables display of the built-in Linux logo.
1298			This may be used to provide more screen space for
1299			kernel log messages and is useful when debugging
1300			kernel boot problems.
1301
1302	lp=0		[LP]	Specify parallel ports to use, e.g,
1303	lp=port[,port...]	lp=none,parport0 (lp0 not configured, lp1 uses
1304	lp=reset		first parallel port). 'lp=0' disables the
1305	lp=auto			printer driver. 'lp=reset' (which can be
1306				specified in addition to the ports) causes
1307				attached printers to be reset. Using
1308				lp=port1,port2,... specifies the parallel ports
1309				to associate lp devices with, starting with
1310				lp0. A port specification may be 'none' to skip
1311				that lp device, or a parport name such as
1312				'parport0'. Specifying 'lp=auto' instead of a
1313				port specification list means that device IDs
1314				from each port should be examined, to see if
1315				an IEEE 1284-compliant printer is attached; if
1316				so, the driver will manage that printer.
1317				See also header of drivers/char/lp.c.
1318
1319	lpj=n		[KNL]
1320			Sets loops_per_jiffy to given constant, thus avoiding
1321			time-consuming boot-time autodetection (up to 250 ms per
1322			CPU). 0 enables autodetection (default). To determine
1323			the correct value for your kernel, boot with normal
1324			autodetection and see what value is printed. Note that
1325			on SMP systems the preset will be applied to all CPUs,
1326			which is likely to cause problems if your CPUs need
1327			significantly divergent settings. An incorrect value
1328			will cause delays in the kernel to be wrong, leading to
1329			unpredictable I/O errors and other breakage. Although
1330			unlikely, in the extreme case this might damage your
1331			hardware.
1332
1333	ltpc=		[NET]
1334			Format: <io>,<irq>,<dma>
1335
1336	machvec=	[IA-64] Force the use of a particular machine-vector
1337			(machvec) in a generic kernel.
1338			Example: machvec=hpzx1_swiotlb
1339
1340	machtype=	[Loongson] Share the same kernel image file between different
1341			 yeeloong laptop.
1342			Example: machtype=lemote-yeeloong-2f-7inch
1343
1344	max_addr=nn[KMG]	[KNL,BOOT,ia64] All physical memory greater
1345			than or equal to this physical address is ignored.
1346
1347	maxcpus=	[SMP] Maximum number of processors that	an SMP kernel
1348			should make use of.  maxcpus=n : n >= 0 limits the
1349			kernel to using 'n' processors.  n=0 is a special case,
1350			it is equivalent to "nosmp", which also disables
1351			the IO APIC.
1352
1353	max_loop=	[LOOP] The number of loop block devices that get
1354	(loop.max_loop)	unconditionally pre-created at init time. The default
1355			number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
1356			of statically allocating a predefined number, loop
1357			devices can be requested on-demand with the
1358			/dev/loop-control interface.
1359
1360	mcatest=	[IA-64]
1361
1362	mce		[X86-32] Machine Check Exception
1363
1364	mce=option	[X86-64] See Documentation/x86/x86_64/boot-options.txt
1365
1366	md=		[HW] RAID subsystems devices and level
1367			See Documentation/md.txt.
1368
1369	mdacon=		[MDA]
1370			Format: <first>,<last>
1371			Specifies range of consoles to be captured by the MDA.
1372
1373	mem=nn[KMG]	[KNL,BOOT] Force usage of a specific amount of memory
1374			Amount of memory to be used when the kernel is not able
1375			to see the whole system memory or for test.
1376			[X86-32] Use together with memmap= to avoid physical
1377			address space collisions. Without memmap= PCI devices
1378			could be placed at addresses belonging to unused RAM.
 
1379
1380	mem=nopentium	[BUGS=X86-32] Disable usage of 4MB pages for kernel
1381			memory.
1382
1383	memchunk=nn[KMG]
1384			[KNL,SH] Allow user to override the default size for
1385			per-device physically contiguous DMA buffers.
1386
1387	memmap=exactmap	[KNL,X86] Enable setting of an exact
1388			E820 memory map, as specified by the user.
1389			Such memmap=exactmap lines can be constructed based on
1390			BIOS output or other requirements. See the memmap=nn@ss
1391			option description.
1392
1393	memmap=nn[KMG]@ss[KMG]
1394			[KNL] Force usage of a specific region of memory
1395			Region of memory to be used, from ss to ss+nn.
1396
1397	memmap=nn[KMG]#ss[KMG]
1398			[KNL,ACPI] Mark specific memory as ACPI data.
1399			Region of memory to be used, from ss to ss+nn.
1400
1401	memmap=nn[KMG]$ss[KMG]
1402			[KNL,ACPI] Mark specific memory as reserved.
1403			Region of memory to be used, from ss to ss+nn.
1404			Example: Exclude memory from 0x18690000-0x1869ffff
1405			         memmap=64K$0x18690000
1406			         or
1407			         memmap=0x10000$0x18690000
1408
1409	memory_corruption_check=0/1 [X86]
1410			Some BIOSes seem to corrupt the first 64k of
1411			memory when doing things like suspend/resume.
1412			Setting this option will scan the memory
1413			looking for corruption.  Enabling this will
1414			both detect corruption and prevent the kernel
1415			from using the memory being corrupted.
1416			However, its intended as a diagnostic tool; if
1417			repeatable BIOS-originated corruption always
1418			affects the same memory, you can use memmap=
1419			to prevent the kernel from using that memory.
1420
1421	memory_corruption_check_size=size [X86]
1422			By default it checks for corruption in the low
1423			64k, making this memory unavailable for normal
1424			use.  Use this parameter to scan for
1425			corruption in more or less memory.
1426
1427	memory_corruption_check_period=seconds [X86]
1428			By default it checks for corruption every 60
1429			seconds.  Use this parameter to check at some
1430			other rate.  0 disables periodic checking.
1431
1432	memtest=	[KNL,X86] Enable memtest
1433			Format: <integer>
1434			default : 0 <disable>
1435			Specifies the number of memtest passes to be
1436			performed. Each pass selects another test
1437			pattern from a given set of patterns. Memtest
1438			fills the memory with this pattern, validates
1439			memory contents and reserves bad memory
1440			regions that are detected.
1441
1442	meye.*=		[HW] Set MotionEye Camera parameters
1443			See Documentation/video4linux/meye.txt.
1444
1445	mfgpt_irq=	[IA-32] Specify the IRQ to use for the
1446			Multi-Function General Purpose Timers on AMD Geode
1447			platforms.
1448
1449	mfgptfix	[X86-32] Fix MFGPT timers on AMD Geode platforms when
1450			the BIOS has incorrectly applied a workaround. TinyBIOS
1451			version 0.98 is known to be affected, 0.99 fixes the
1452			problem by letting the user disable the workaround.
1453
1454	mga=		[HW,DRM]
1455
1456	min_addr=nn[KMG]	[KNL,BOOT,ia64] All physical memory below this
1457			physical address is ignored.
1458
1459	mini2440=	[ARM,HW,KNL]
1460			Format:[0..2][b][c][t]
1461			Default: "0tb"
1462			MINI2440 configuration specification:
1463			0 - The attached screen is the 3.5" TFT
1464			1 - The attached screen is the 7" TFT
1465			2 - The VGA Shield is attached (1024x768)
1466			Leaving out the screen size parameter will not load
1467			the TFT driver, and the framebuffer will be left
1468			unconfigured.
1469			b - Enable backlight. The TFT backlight pin will be
1470			linked to the kernel VESA blanking code and a GPIO
1471			LED. This parameter is not necessary when using the
1472			VGA shield.
1473			c - Enable the s3c camera interface.
1474			t - Reserved for enabling touchscreen support. The
1475			touchscreen support is not enabled in the mainstream
1476			kernel as of 2.6.30, a preliminary port can be found
1477			in the "bleeding edge" mini2440 support kernel at
1478			http://repo.or.cz/w/linux-2.6/mini2440.git
1479
1480	mminit_loglevel=
1481			[KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
1482			parameter allows control of the logging verbosity for
1483			the additional memory initialisation checks. A value
1484			of 0 disables mminit logging and a level of 4 will
1485			log everything. Information is printed at KERN_DEBUG
1486			so loglevel=8 may also need to be specified.
1487
 
 
 
 
 
 
1488	mousedev.tap_time=
1489			[MOUSE] Maximum time between finger touching and
1490			leaving touchpad surface for touch to be considered
1491			a tap and be reported as a left button click (for
1492			touchpads working in absolute mode only).
1493			Format: <msecs>
1494	mousedev.xres=	[MOUSE] Horizontal screen resolution, used for devices
1495			reporting absolute coordinates, such as tablets
1496	mousedev.yres=	[MOUSE] Vertical screen resolution, used for devices
1497			reporting absolute coordinates, such as tablets
1498
1499	movablecore=nn[KMG]	[KNL,X86,IA-64,PPC] This parameter
1500			is similar to kernelcore except it specifies the
1501			amount of memory used for migratable allocations.
1502			If both kernelcore and movablecore is specified,
1503			then kernelcore will be at *least* the specified
1504			value but may be more. If movablecore on its own
1505			is specified, the administrator must be careful
1506			that the amount of memory usable for all allocations
1507			is not too small.
1508
 
 
 
1509	MTD_Partition=	[MTD]
1510			Format: <name>,<region-number>,<size>,<offset>
1511
1512	MTD_Region=	[MTD] Format:
1513			<name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
1514
1515	mtdparts=	[MTD]
1516			See drivers/mtd/cmdlinepart.c.
1517
1518	multitce=off	[PPC]  This parameter disables the use of the pSeries
1519			firmware feature for updating multiple TCE entries
1520			at a time.
1521
1522	onenand.bdry=	[HW,MTD] Flex-OneNAND Boundary Configuration
1523
1524			Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
1525
1526			boundary - index of last SLC block on Flex-OneNAND.
1527				   The remaining blocks are configured as MLC blocks.
1528			lock	 - Configure if Flex-OneNAND boundary should be locked.
1529				   Once locked, the boundary cannot be changed.
1530				   1 indicates lock status, 0 indicates unlock status.
1531
1532	mtdset=		[ARM]
1533			ARM/S3C2412 JIVE boot control
1534
1535			See arch/arm/mach-s3c2412/mach-jive.c
1536
1537	mtouchusb.raw_coordinates=
1538			[HW] Make the MicroTouch USB driver use raw coordinates
1539			('y', default) or cooked coordinates ('n')
1540
1541	mtrr_chunk_size=nn[KMG] [X86]
1542			used for mtrr cleanup. It is largest continuous chunk
1543			that could hold holes aka. UC entries.
1544
1545	mtrr_gran_size=nn[KMG] [X86]
1546			Used for mtrr cleanup. It is granularity of mtrr block.
1547			Default is 1.
1548			Large value could prevent small alignment from
1549			using up MTRRs.
1550
1551	mtrr_spare_reg_nr=n [X86]
1552			Format: <integer>
1553			Range: 0,7 : spare reg number
1554			Default : 1
1555			Used for mtrr cleanup. It is spare mtrr entries number.
1556			Set to 2 or more if your graphical card needs more.
1557
1558	n2=		[NET] SDL Inc. RISCom/N2 synchronous serial card
1559
1560	netdev=		[NET] Network devices parameters
1561			Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
1562			Note that mem_start is often overloaded to mean
1563			something different and driver-specific.
1564			This usage is only documented in each driver source
1565			file if at all.
1566
1567	nf_conntrack.acct=
1568			[NETFILTER] Enable connection tracking flow accounting
1569			0 to disable accounting
1570			1 to enable accounting
1571			Default value is 0.
1572
1573	nfsaddrs=	[NFS] Deprecated.  Use ip= instead.
1574			See Documentation/filesystems/nfs/nfsroot.txt.
1575
1576	nfsroot=	[NFS] nfs root filesystem for disk-less boxes.
1577			See Documentation/filesystems/nfs/nfsroot.txt.
1578
1579	nfsrootdebug	[NFS] enable nfsroot debugging messages.
1580			See Documentation/filesystems/nfs/nfsroot.txt.
1581
1582	nfs.callback_tcpport=
1583			[NFS] set the TCP port on which the NFSv4 callback
1584			channel should listen.
1585
1586	nfs.cache_getent=
1587			[NFS] sets the pathname to the program which is used
1588			to update the NFS client cache entries.
1589
1590	nfs.cache_getent_timeout=
1591			[NFS] sets the timeout after which an attempt to
1592			update a cache entry is deemed to have failed.
1593
1594	nfs.idmap_cache_timeout=
1595			[NFS] set the maximum lifetime for idmapper cache
1596			entries.
1597
1598	nfs.enable_ino64=
1599			[NFS] enable 64-bit inode numbers.
1600			If zero, the NFS client will fake up a 32-bit inode
1601			number for the readdir() and stat() syscalls instead
1602			of returning the full 64-bit number.
1603			The default is to return 64-bit inode numbers.
1604
 
 
 
 
 
 
 
 
1605	nfs.nfs4_disable_idmapping=
1606			[NFSv4] When set, this option disables the NFSv4
1607			idmapper on the client, but only if the mount
1608			is using the 'sec=sys' security flavour. This may
1609			make migration from legacy NFSv2/v3 systems easier
1610			provided that the server has the appropriate support.
1611			The default is to always enable NFSv4 idmapping.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1612
1613	nmi_debug=	[KNL,AVR32,SH] Specify one or more actions to take
1614			when a NMI is triggered.
1615			Format: [state][,regs][,debounce][,die]
1616
1617	nmi_watchdog=	[KNL,BUGS=X86] Debugging features for SMP kernels
1618			Format: [panic,][nopanic,][num]
1619			Valid num: 0
1620			0 - turn nmi_watchdog off
1621			When panic is specified, panic when an NMI watchdog
1622			timeout occurs (or 'nopanic' to override the opposite
1623			default).
1624			This is useful when you use a panic=... timeout and
1625			need the box quickly up again.
1626
1627	netpoll.carrier_timeout=
1628			[NET] Specifies amount of time (in seconds) that
1629			netpoll should wait for a carrier. By default netpoll
1630			waits 4 seconds.
1631
1632	no387		[BUGS=X86-32] Tells the kernel to use the 387 maths
1633			emulation library even if a 387 maths coprocessor
1634			is present.
1635
1636	no_console_suspend
1637			[HW] Never suspend the console
1638			Disable suspending of consoles during suspend and
1639			hibernate operations.  Once disabled, debugging
1640			messages can reach various consoles while the rest
1641			of the system is being put to sleep (ie, while
1642			debugging driver suspend/resume hooks).  This may
1643			not work reliably with all consoles, but is known
1644			to work with serial and VGA consoles.
 
 
 
 
 
1645
1646	noaliencache	[MM, NUMA, SLAB] Disables the allocation of alien
1647			caches in the slab allocator.  Saves per-node memory,
1648			but will impact performance.
1649
1650	noalign		[KNL,ARM]
1651
1652	noapic		[SMP,APIC] Tells the kernel to not make use of any
1653			IOAPICs that may be present in the system.
1654
 
 
 
 
1655	noautogroup	Disable scheduler automatic task group creation.
1656
1657	nobats		[PPC] Do not use BATs for mapping kernel lowmem
1658			on "Classic" PPC cores.
1659
1660	nocache		[ARM]
1661
1662	noclflush	[BUGS=X86] Don't use the CLFLUSH instruction
1663
1664	nodelayacct	[KNL] Disable per-task delay accounting
1665
1666	nodisconnect	[HW,SCSI,M68K] Disables SCSI disconnects.
1667
1668	nodsp		[SH] Disable hardware DSP at boot time.
1669
1670	noefi		[X86] Disable EFI runtime services support.
1671
1672	noexec		[IA-64]
1673
1674	noexec		[X86]
1675			On X86-32 available only on PAE configured kernels.
1676			noexec=on: enable non-executable mappings (default)
1677			noexec=off: disable non-executable mappings
1678
 
 
 
 
1679	nosmep		[X86]
1680			Disable SMEP (Supervisor Mode Execution Protection)
1681			even if it is supported by processor.
1682
1683	noexec32	[X86-64]
1684			This affects only 32-bit executables.
1685			noexec32=on: enable non-executable mappings (default)
1686				read doesn't imply executable mappings
1687			noexec32=off: disable non-executable mappings
1688				read implies executable mappings
1689
1690	nofpu		[SH] Disable hardware FPU at boot time.
1691
1692	nofxsr		[BUGS=X86-32] Disables x86 floating point extended
1693			register save and restore. The kernel will only save
1694			legacy floating-point registers on task switch.
1695
1696	noxsave		[BUGS=X86] Disables x86 extended register state save
1697			and restore using xsave. The kernel will fallback to
1698			enabling legacy floating-point and sse state.
1699
 
 
 
 
 
 
1700	nohlt		[BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
1701			wfi(ARM) instruction doesn't work correctly and not to
1702			use it. This is also useful when using JTAG debugger.
1703
1704	no-hlt		[BUGS=X86-32] Tells the kernel that the hlt
1705			instruction doesn't work correctly and not to
1706			use it.
1707
1708	no_file_caps	Tells the kernel not to honor file capabilities.  The
1709			only way then for a file to be executed with privilege
1710			is to be setuid root or executed by root.
1711
1712	nohalt		[IA-64] Tells the kernel not to use the power saving
1713			function PAL_HALT_LIGHT when idle. This increases
1714			power-consumption. On the positive side, it reduces
1715			interrupt wake-up latency, which may improve performance
1716			in certain environments such as networked servers or
1717			real-time systems.
1718
1719	nohz=		[KNL] Boottime enable/disable dynamic ticks
1720			Valid arguments: on, off
1721			Default: on
1722
 
 
 
 
 
 
 
 
1723	noiotrap	[SH] Disables trapped I/O port accesses.
1724
1725	noirqdebug	[X86-32] Disables the code which attempts to detect and
1726			disable unhandled interrupt sources.
1727
1728	no_timer_check	[X86,APIC] Disables the code which tests for
1729			broken timer IRQ sources.
1730
1731	noisapnp	[ISAPNP] Disables ISA PnP code.
1732
1733	noinitrd	[RAM] Tells the kernel not to load any configured
1734			initial RAM disk.
1735
1736	nointremap	[X86-64, Intel-IOMMU] Do not enable interrupt
1737			remapping.
1738			[Deprecated - use intremap=off]
1739
1740	nointroute	[IA-64]
1741
1742	nojitter	[IA-64] Disables jitter checking for ITC timers.
1743
1744	no-kvmclock	[X86,KVM] Disable paravirtualized KVM clock driver
1745
1746	no-kvmapf	[X86,KVM] Disable paravirtualized asynchronous page
1747			fault handling.
1748
1749	no-steal-acc    [X86,KVM] Disable paravirtualized steal time accounting.
1750			steal time is computed, but won't influence scheduler
1751			behaviour
1752
1753	nolapic		[X86-32,APIC] Do not enable or use the local APIC.
1754
1755	nolapic_timer	[X86-32,APIC] Do not use the local APIC timer.
1756
1757	noltlbs		[PPC] Do not use large page/tlb entries for kernel
1758			lowmem mapping on PPC40x.
1759
1760	nomca		[IA-64] Disable machine check abort handling
1761
1762	nomce		[X86-32] Machine Check Exception
1763
1764	nomfgpt		[X86-32] Disable Multi-Function General Purpose
1765			Timer usage (for AMD Geode machines).
1766
 
 
 
 
 
 
1767	nopat		[X86] Disable PAT (page attribute table extension of
1768			pagetables) support.
1769
1770	norandmaps	Don't use address space randomization.  Equivalent to
1771			echo 0 > /proc/sys/kernel/randomize_va_space
1772
1773	noreplace-paravirt	[X86,IA-64,PV_OPS] Don't patch paravirt_ops
1774
1775	noreplace-smp	[X86-32,SMP] Don't replace SMP instructions
1776			with UP alternatives
1777
1778	noresidual	[PPC] Don't use residual data on PReP machines.
 
 
 
1779
1780	noresume	[SWSUSP] Disables resume and restores original swap
1781			space.
1782
1783	no-scroll	[VGA] Disables scrollback.
1784			This is required for the Braillex ib80-piezo Braille
1785			reader made by F.H. Papenmeier (Germany).
1786
1787	nosbagart	[IA-64]
1788
1789	nosep		[BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
1790
1791	nosmp		[SMP] Tells an SMP kernel to act as a UP kernel,
1792			and disable the IO APIC.  legacy for "maxcpus=0".
1793
1794	nosoftlockup	[KNL] Disable the soft-lockup detector.
1795
1796	nosync		[HW,M68K] Disables sync negotiation for all devices.
1797
1798	notsc		[BUGS=X86-32] Disable Time Stamp Counter
1799
1800	nousb		[USB] Disable the USB subsystem
1801
1802	nowatchdog	[KNL] Disable the lockup detector (NMI watchdog).
1803
1804	nowb		[ARM]
1805
1806	nox2apic	[X86-64,APIC] Do not enable x2APIC mode.
1807
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1808	nptcg=		[IA-64] Override max number of concurrent global TLB
1809			purges which is reported from either PAL_VM_SUMMARY or
1810			SAL PALO.
1811
1812	nr_cpus=	[SMP] Maximum number of processors that	an SMP kernel
1813			could support.  nr_cpus=n : n >= 1 limits the kernel to
1814			supporting 'n' processors. Later in runtime you can not
1815			use hotplug cpu feature to put more cpu back to online.
1816			just like you compile the kernel NR_CPUS=n
1817
1818	nr_uarts=	[SERIAL] maximum number of UARTs to be registered.
1819
 
 
 
1820	numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
1821			one of ['zone', 'node', 'default'] can be specified
1822			This can be set from sysctl after boot.
1823			See Documentation/sysctl/vm.txt for details.
1824
1825	ohci1394_dma=early	[HW] enable debugging via the ohci1394 driver.
1826			See Documentation/debugging-via-ohci1394.txt for more
1827			info.
1828
1829	olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
1830			Rather than timing out after 20 ms if an EC
1831			command is not properly ACKed, override the length
1832			of the timeout.  We have interrupts disabled while
1833			waiting for the ACK, so if this is set too high
1834			interrupts *may* be lost!
1835
1836	omap_mux=	[OMAP] Override bootloader pin multiplexing.
1837			Format: <mux_mode0.mode_name=value>...
1838			For example, to override I2C bus2:
1839			omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
1840
1841	oprofile.timer=	[HW]
1842			Use timer interrupt instead of performance counters
1843
1844	oprofile.cpu_type=	Force an oprofile cpu type
1845			This might be useful if you have an older oprofile
1846			userland or if you want common events.
1847			Format: { arch_perfmon }
1848			arch_perfmon: [X86] Force use of architectural
1849				perfmon on Intel CPUs instead of the
1850				CPU specific event set.
 
 
 
 
 
1851
1852	oops=panic	Always panic on oopses. Default is to just kill the
1853			process, but there is a small probability of
1854			deadlocking the machine.
1855			This will also cause panics on machine check exceptions.
1856			Useful together with panic=30 to trigger a reboot.
1857
1858	OSS		[HW,OSS]
1859			See Documentation/sound/oss/oss-parameters.txt
1860
1861	panic=		[KNL] Kernel behaviour on panic: delay <timeout>
1862			timeout > 0: seconds before rebooting
1863			timeout = 0: wait forever
1864			timeout < 0: reboot immediately
1865			Format: <timeout>
1866
1867	parkbd.port=	[HW] Parallel port number the keyboard adapter is
1868			connected to, default is 0.
1869			Format: <parport#>
1870	parkbd.mode=	[HW] Parallel port keyboard adapter mode of operation,
1871			0 for XT, 1 for AT (default is AT).
1872			Format: <mode>
1873
1874	parport=	[HW,PPT] Specify parallel ports. 0 disables.
1875			Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
1876			Use 'auto' to force the driver to use any
1877			IRQ/DMA settings detected (the default is to
1878			ignore detected IRQ/DMA settings because of
1879			possible conflicts). You can specify the base
1880			address, IRQ, and DMA settings; IRQ and DMA
1881			should be numbers, or 'auto' (for using detected
1882			settings on that particular port), or 'nofifo'
1883			(to avoid using a FIFO even if it is detected).
1884			Parallel ports are assigned in the order they
1885			are specified on the command line, starting
1886			with parport0.
1887
1888	parport_init_mode=	[HW,PPT]
1889			Configure VIA parallel port to operate in
1890			a specific mode. This is necessary on Pegasos
1891			computer where firmware has no options for setting
1892			up parallel port mode and sets it to spp.
1893			Currently this function knows 686a and 8231 chips.
1894			Format: [spp|ps2|epp|ecp|ecpepp]
1895
1896	pause_on_oops=
1897			Halt all CPUs after the first oops has been printed for
1898			the specified number of seconds.  This is to be used if
1899			your oopses keep scrolling off the screen.
1900
1901	pcbit=		[HW,ISDN]
1902
1903	pcd.		[PARIDE]
1904			See header of drivers/block/paride/pcd.c.
1905			See also Documentation/blockdev/paride.txt.
1906
1907	pci=option[,option...]	[PCI] various PCI subsystem options:
1908		earlydump	[X86] dump PCI config space before the kernel
1909			        changes anything
1910		off		[X86] don't probe for the PCI bus
1911		bios		[X86-32] force use of PCI BIOS, don't access
1912				the hardware directly. Use this if your machine
1913				has a non-standard PCI host bridge.
1914		nobios		[X86-32] disallow use of PCI BIOS, only direct
1915				hardware access methods are allowed. Use this
1916				if you experience crashes upon bootup and you
1917				suspect they are caused by the BIOS.
1918		conf1		[X86] Force use of PCI Configuration
1919				Mechanism 1.
1920		conf2		[X86] Force use of PCI Configuration
1921				Mechanism 2.
1922		noaer		[PCIE] If the PCIEAER kernel config parameter is
1923				enabled, this kernel boot option can be used to
1924				disable the use of PCIE advanced error reporting.
1925		nodomains	[PCI] Disable support for multiple PCI
1926				root domains (aka PCI segments, in ACPI-speak).
1927		nommconf	[X86] Disable use of MMCONFIG for PCI
1928				Configuration
1929		check_enable_amd_mmconf [X86] check for and enable
1930				properly configured MMIO access to PCI
1931				config space on AMD family 10h CPU
1932		nomsi		[MSI] If the PCI_MSI kernel config parameter is
1933				enabled, this kernel boot option can be used to
1934				disable the use of MSI interrupts system-wide.
1935		noioapicquirk	[APIC] Disable all boot interrupt quirks.
1936				Safety option to keep boot IRQs enabled. This
1937				should never be necessary.
1938		ioapicreroute	[APIC] Enable rerouting of boot IRQs to the
1939				primary IO-APIC for bridges that cannot disable
1940				boot IRQs. This fixes a source of spurious IRQs
1941				when the system masks IRQs.
1942		noioapicreroute	[APIC] Disable workaround that uses the
1943				boot IRQ equivalent of an IRQ that connects to
1944				a chipset where boot IRQs cannot be disabled.
1945				The opposite of ioapicreroute.
1946		biosirq		[X86-32] Use PCI BIOS calls to get the interrupt
1947				routing table. These calls are known to be buggy
1948				on several machines and they hang the machine
1949				when used, but on other computers it's the only
1950				way to get the interrupt routing table. Try
1951				this option if the kernel is unable to allocate
1952				IRQs or discover secondary PCI buses on your
1953				motherboard.
1954		rom		[X86] Assign address space to expansion ROMs.
1955				Use with caution as certain devices share
1956				address decoders between ROMs and other
1957				resources.
1958		norom		[X86] Do not assign address space to
1959				expansion ROMs that do not already have
1960				BIOS assigned address ranges.
1961		nobar		[X86] Do not assign address space to the
1962				BARs that weren't assigned by the BIOS.
1963		irqmask=0xMMMM	[X86] Set a bit mask of IRQs allowed to be
1964				assigned automatically to PCI devices. You can
1965				make the kernel exclude IRQs of your ISA cards
1966				this way.
1967		pirqaddr=0xAAAAA	[X86] Specify the physical address
1968				of the PIRQ table (normally generated
1969				by the BIOS) if it is outside the
1970				F0000h-100000h range.
1971		lastbus=N	[X86] Scan all buses thru bus #N. Can be
1972				useful if the kernel is unable to find your
1973				secondary buses and you want to tell it
1974				explicitly which ones they are.
1975		assign-busses	[X86] Always assign all PCI bus
1976				numbers ourselves, overriding
1977				whatever the firmware may have done.
1978		usepirqmask	[X86] Honor the possible IRQ mask stored
1979				in the BIOS $PIR table. This is needed on
1980				some systems with broken BIOSes, notably
1981				some HP Pavilion N5400 and Omnibook XE3
1982				notebooks. This will have no effect if ACPI
1983				IRQ routing is enabled.
1984		noacpi		[X86] Do not use ACPI for IRQ routing
1985				or for PCI scanning.
1986		use_crs		[X86] Use PCI host bridge window information
1987				from ACPI.  On BIOSes from 2008 or later, this
1988				is enabled by default.  If you need to use this,
1989				please report a bug.
1990		nocrs		[X86] Ignore PCI host bridge windows from ACPI.
1991			        If you need to use this, please report a bug.
1992		routeirq	Do IRQ routing for all PCI devices.
1993				This is normally done in pci_enable_device(),
1994				so this option is a temporary workaround
1995				for broken drivers that don't call it.
1996		skip_isa_align	[X86] do not align io start addr, so can
1997				handle more pci cards
1998		firmware	[ARM] Do not re-enumerate the bus but instead
1999				just use the configuration from the
2000				bootloader. This is currently used on
2001				IXP2000 systems where the bus has to be
2002				configured a certain way for adjunct CPUs.
2003		noearly		[X86] Don't do any early type 1 scanning.
2004				This might help on some broken boards which
2005				machine check when some devices' config space
2006				is read. But various workarounds are disabled
2007				and some IOMMU drivers will not work.
2008		bfsort		Sort PCI devices into breadth-first order.
2009				This sorting is done to get a device
2010				order compatible with older (<= 2.4) kernels.
2011		nobfsort	Don't sort PCI devices into breadth-first order.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2012		cbiosize=nn[KMG]	The fixed amount of bus space which is
2013				reserved for the CardBus bridge's IO window.
2014				The default value is 256 bytes.
2015		cbmemsize=nn[KMG]	The fixed amount of bus space which is
2016				reserved for the CardBus bridge's memory
2017				window. The default value is 64 megabytes.
2018		resource_alignment=
2019				Format:
2020				[<order of align>@][<domain>:]<bus>:<slot>.<func>[; ...]
2021				Specifies alignment and device to reassign
2022				aligned memory resources.
2023				If <order of align> is not specified,
2024				PAGE_SIZE is used as alignment.
2025				PCI-PCI bridge can be specified, if resource
2026				windows need to be expanded.
2027		ecrc=		Enable/disable PCIe ECRC (transaction layer
2028				end-to-end CRC checking).
2029				bios: Use BIOS/firmware settings. This is the
2030				the default.
2031				off: Turn ECRC off
2032				on: Turn ECRC on.
2033		realloc		reallocate PCI resources if allocations done by BIOS
2034				are erroneous.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2035
2036	pcie_aspm=	[PCIE] Forcibly enable or disable PCIe Active State Power
2037			Management.
2038		off	Disable ASPM.
2039		force	Enable ASPM even on devices that claim not to support it.
2040			WARNING: Forcing ASPM on may cause system lockups.
2041
 
 
 
 
2042	pcie_ports=	[PCIE] PCIe ports handling:
2043		auto	Ask the BIOS whether or not to use native PCIe services
2044			associated with PCIe ports (PME, hot-plug, AER).  Use
2045			them only if that is allowed by the BIOS.
2046		native	Use native PCIe services associated with PCIe ports
2047			unconditionally.
2048		compat	Treat PCIe ports as PCI-to-PCI bridges, disable the PCIe
2049			ports driver.
2050
2051	pcie_pme=	[PCIE,PM] Native PCIe PME signaling options:
2052		nomsi	Do not use MSI for native PCIe PME signaling (this makes
2053			all PCIe root ports use INTx for all services).
2054
2055	pcmv=		[HW,PCMCIA] BadgePAD 4
2056
 
 
 
 
 
 
 
2057	pd.		[PARIDE]
2058			See Documentation/blockdev/paride.txt.
2059
2060	pdcchassis=	[PARISC,HW] Disable/Enable PDC Chassis Status codes at
2061			boot time.
2062			Format: { 0 | 1 }
2063			See arch/parisc/kernel/pdc_chassis.c
2064
2065	percpu_alloc=	Select which percpu first chunk allocator to use.
2066			Currently supported values are "embed" and "page".
2067			Archs may support subset or none of the	selections.
2068			See comments in mm/percpu.c for details on each
2069			allocator.  This parameter is primarily	for debugging
2070			and performance comparison.
2071
2072	pf.		[PARIDE]
2073			See Documentation/blockdev/paride.txt.
2074
2075	pg.		[PARIDE]
2076			See Documentation/blockdev/paride.txt.
2077
2078	pirq=		[SMP,APIC] Manual mp-table setup
2079			See Documentation/x86/i386/IO-APIC.txt.
2080
2081	plip=		[PPT,NET] Parallel port network link
2082			Format: { parport<nr> | timid | 0 }
2083			See also Documentation/parport.txt.
2084
2085	pmtmr=		[X86] Manual setup of pmtmr I/O Port.
2086			Override pmtimer IOPort with a hex value.
2087			e.g. pmtmr=0x508
2088
2089	pnp.debug=1	[PNP]
2090			Enable PNP debug messages (depends on the
2091			CONFIG_PNP_DEBUG_MESSAGES option).  Change at run-time
2092			via /sys/module/pnp/parameters/debug.  We always show
2093			current resource usage; turning this on also shows
2094			possible settings and some assignment information.
2095
2096	pnpacpi=	[ACPI]
2097			{ off }
2098
2099	pnpbios=	[ISAPNP]
2100			{ on | off | curr | res | no-curr | no-res }
2101
2102	pnp_reserve_irq=
2103			[ISAPNP] Exclude IRQs for the autoconfiguration
2104
2105	pnp_reserve_dma=
2106			[ISAPNP] Exclude DMAs for the autoconfiguration
2107
2108	pnp_reserve_io=	[ISAPNP] Exclude I/O ports for the autoconfiguration
2109			Ranges are in pairs (I/O port base and size).
2110
2111	pnp_reserve_mem=
2112			[ISAPNP] Exclude memory regions for the
2113			autoconfiguration.
2114			Ranges are in pairs (memory base and size).
2115
2116	ports=		[IP_VS_FTP] IPVS ftp helper module
2117			Default is 21.
2118			Up to 8 (IP_VS_APP_MAX_PORTS) ports
2119			may be specified.
2120			Format: <port>,<port>....
2121
2122	print-fatal-signals=
2123			[KNL] debug: print fatal signals
2124
2125			If enabled, warn about various signal handling
2126			related application anomalies: too many signals,
2127			too many POSIX.1 timers, fatal signals causing a
2128			coredump - etc.
2129
2130			If you hit the warning due to signal overflow,
2131			you might want to try "ulimit -i unlimited".
2132
2133			default: off.
2134
 
 
 
 
 
 
2135	printk.time=	Show timing data prefixed to each printk message line
2136			Format: <bool>  (1/Y/y=enable, 0/N/n=disable)
2137
2138	processor.max_cstate=	[HW,ACPI]
2139			Limit processor to maximum C-state
2140			max_cstate=9 overrides any DMI blacklist limit.
2141
2142	processor.nocst	[HW,ACPI]
2143			Ignore the _CST method to determine C-states,
2144			instead using the legacy FADT method
2145
2146	profile=	[KNL] Enable kernel profiling via /proc/profile
2147			Format: [schedule,]<number>
2148			Param: "schedule" - profile schedule points.
2149			Param: <number> - step/bucket size as a power of 2 for
2150				statistical time based profiling.
2151			Param: "sleep" - profile D-state sleeping (millisecs).
2152				Requires CONFIG_SCHEDSTATS
2153			Param: "kvm" - profile VM exits.
2154
2155	prompt_ramdisk=	[RAM] List of RAM disks to prompt for floppy disk
2156			before loading.
2157			See Documentation/blockdev/ramdisk.txt.
2158
2159	psmouse.proto=	[HW,MOUSE] Highest PS2 mouse protocol extension to
2160			probe for; one of (bare|imps|exps|lifebook|any).
2161	psmouse.rate=	[HW,MOUSE] Set desired mouse report rate, in reports
2162			per second.
2163	psmouse.resetafter=	[HW,MOUSE]
2164			Try to reset the device after so many bad packets
2165			(0 = never).
2166	psmouse.resolution=
2167			[HW,MOUSE] Set desired mouse resolution, in dpi.
2168	psmouse.smartscroll=
2169			[HW,MOUSE] Controls Logitech smartscroll autorepeat.
2170			0 = disabled, 1 = enabled (default).
2171
2172	pstore.backend=	Specify the name of the pstore backend to use
2173
2174	pt.		[PARIDE]
2175			See Documentation/blockdev/paride.txt.
2176
2177	pty.legacy_count=
2178			[KNL] Number of legacy pty's. Overwrites compiled-in
2179			default number.
2180
2181	quiet		[KNL] Disable most log messages
2182
2183	r128=		[HW,DRM]
2184
2185	raid=		[HW,RAID]
2186			See Documentation/md.txt.
2187
2188	ramdisk_blocksize=	[RAM]
2189			See Documentation/blockdev/ramdisk.txt.
2190
2191	ramdisk_size=	[RAM] Sizes of RAM disks in kilobytes
2192			See Documentation/blockdev/ramdisk.txt.
2193
2194	rcupdate.blimit=	[KNL,BOOT]
2195			Set maximum number of finished RCU callbacks to process
2196			in one batch.
2197
2198	rcupdate.qhimark=	[KNL,BOOT]
2199			Set threshold of queued
2200			RCU callbacks over which batch limiting is disabled.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2201
2202	rcupdate.qlowmark=	[KNL,BOOT]
2203			Set threshold of queued RCU callbacks below which
2204			batch limiting is re-enabled.
2205
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2206	rdinit=		[KNL]
2207			Format: <full_path>
2208			Run specified binary instead of /init from the ramdisk,
2209			used for early userspace startup. See initrd.
2210
2211	reboot=		[BUGS=X86-32,BUGS=ARM,BUGS=IA-64] Rebooting mode
2212			Format: <reboot_mode>[,<reboot_mode2>[,...]]
2213			See arch/*/kernel/reboot.c or arch/*/kernel/process.c
 
 
 
 
 
 
 
 
2214
2215	relax_domain_level=
2216			[KNL, SMP] Set scheduler's default relax_domain_level.
2217			See Documentation/cgroups/cpusets.txt.
2218
2219	reserve=	[KNL,BUGS] Force the kernel to ignore some iomem area
2220
2221	reservetop=	[X86-32]
2222			Format: nn[KMG]
2223			Reserves a hole at the top of the kernel virtual
2224			address space.
2225
2226	reservelow=	[X86]
2227			Format: nn[K]
2228			Set the amount of memory to reserve for BIOS at
2229			the bottom of the address space.
2230
2231	reset_devices	[KNL] Force drivers to reset the underlying device
2232			during initialization.
2233
2234	resume=		[SWSUSP]
2235			Specify the partition device for software suspend
 
 
2236
2237	resume_offset=	[SWSUSP]
2238			Specify the offset from the beginning of the partition
2239			given by "resume=" at which the swap header is located,
2240			in <PAGE_SIZE> units (needed only for swap files).
2241			See  Documentation/power/swsusp-and-swap-files.txt
2242
 
 
 
 
 
 
 
2243	hibernate=	[HIBERNATION]
2244		noresume	Don't check if there's a hibernation image
2245				present during boot.
2246		nocompress	Don't compress/decompress hibernation images.
2247
2248	retain_initrd	[RAM] Keep initrd memory after extraction
2249
2250	rhash_entries=	[KNL,NET]
2251			Set number of hash buckets for route cache
2252
2253	riscom8=	[HW,SERIAL]
2254			Format: <io_board1>[,<io_board2>[,...<io_boardN>]]
2255
2256	ro		[KNL] Mount root device read-only on boot
2257
2258	root=		[KNL] Root filesystem
2259			See name_to_dev_t comment in init/do_mounts.c.
2260
2261	rootdelay=	[KNL] Delay (in seconds) to pause before attempting to
2262			mount the root filesystem
2263
2264	rootflags=	[KNL] Set root filesystem mount option string
2265
2266	rootfstype=	[KNL] Set root filesystem type
2267
2268	rootwait	[KNL] Wait (indefinitely) for root device to show up.
2269			Useful for devices that are detected asynchronously
2270			(e.g. USB and MMC devices).
2271
 
 
 
 
 
2272	rw		[KNL] Mount root device read-write on boot
2273
2274	S		[KNL] Run init in single mode
2275
2276	sa1100ir	[NET]
2277			See drivers/net/irda/sa1100_ir.c.
2278
2279	sbni=		[NET] Granch SBNI12 leased line adapter
2280
2281	sched_debug	[KNL] Enables verbose scheduler debug messages.
2282
 
 
 
 
 
 
 
 
 
2283	security=	[SECURITY] Choose a security module to enable at boot.
2284			If this boot parameter is not specified, only the first
2285			security module asking for security registration will be
2286			loaded. An invalid security module name will be treated
2287			as if no module has been chosen.
2288
2289	selinux=	[SELINUX] Disable or enable SELinux at boot time.
2290			Format: { "0" | "1" }
2291			See security/selinux/Kconfig help text.
2292			0 -- disable.
2293			1 -- enable.
2294			Default value is set via kernel config option.
2295			If enabled at boot time, /selinux/disable can be used
2296			later to disable prior to initial policy load.
2297
2298	apparmor=	[APPARMOR] Disable or enable AppArmor at boot time
2299			Format: { "0" | "1" }
2300			See security/apparmor/Kconfig help text
2301			0 -- disable.
2302			1 -- enable.
2303			Default value is set via kernel config option.
2304
2305	serialnumber	[BUGS=X86-32]
2306
2307	shapers=	[NET]
2308			Maximal number of shapers.
2309
2310	show_msr=	[x86] show boot-time MSR settings
2311			Format: { <integer> }
2312			Show boot-time (BIOS-initialized) MSR settings.
2313			The parameter means the number of CPUs to show,
2314			for example 1 means boot CPU only.
2315
2316	simeth=		[IA-64]
2317	simscsi=
2318
2319	slram=		[HW,MTD]
2320
 
 
 
 
 
 
2321	slub_debug[=options[,slabs]]	[MM, SLUB]
2322			Enabling slub_debug allows one to determine the
2323			culprit if slab objects become corrupted. Enabling
2324			slub_debug can create guard zones around objects and
2325			may poison objects when not in use. Also tracks the
2326			last alloc / free. For more information see
2327			Documentation/vm/slub.txt.
2328
2329	slub_max_order= [MM, SLUB]
2330			Determines the maximum allowed order for slabs.
2331			A high setting may cause OOMs due to memory
2332			fragmentation. For more information see
2333			Documentation/vm/slub.txt.
2334
2335	slub_min_objects=	[MM, SLUB]
2336			The minimum number of objects per slab. SLUB will
2337			increase the slab order up to slub_max_order to
2338			generate a sufficiently large slab able to contain
2339			the number of objects indicated. The higher the number
2340			of objects the smaller the overhead of tracking slabs
2341			and the less frequently locks need to be acquired.
2342			For more information see Documentation/vm/slub.txt.
2343
2344	slub_min_order=	[MM, SLUB]
2345			Determines the mininum page order for slabs. Must be
2346			lower than slub_max_order.
2347			For more information see Documentation/vm/slub.txt.
2348
2349	slub_nomerge	[MM, SLUB]
2350			Disable merging of slabs with similar size. May be
2351			necessary if there is some reason to distinguish
2352			allocs to different slabs. Debug options disable
2353			merging on their own.
2354			For more information see Documentation/vm/slub.txt.
2355
2356	smart2=		[HW]
2357			Format: <io1>[,<io2>[,...,<io8>]]
2358
2359	smp-alt-once	[X86-32,SMP] On a hotplug CPU system, only
2360			attempt to substitute SMP alternatives once at boot.
2361
2362	smsc-ircc2.nopnp	[HW] Don't use PNP to discover SMC devices
2363	smsc-ircc2.ircc_cfg=	[HW] Device configuration I/O port
2364	smsc-ircc2.ircc_sir=	[HW] SIR base I/O port
2365	smsc-ircc2.ircc_fir=	[HW] FIR base I/O port
2366	smsc-ircc2.ircc_irq=	[HW] IRQ line
2367	smsc-ircc2.ircc_dma=	[HW] DMA channel
2368	smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
2369				0: Toshiba Satellite 1800 (GP data pin select)
2370				1: Fast pin select (default)
2371				2: ATC IRMode
2372
2373	softlockup_panic=
2374			[KNL] Should the soft-lockup detector generate panics.
2375			Format: <integer>
2376
2377	sonypi.*=	[HW] Sony Programmable I/O Control Device driver
2378			See Documentation/sonypi.txt
2379
2380	specialix=	[HW,SERIAL] Specialix multi-serial port adapter
2381			See Documentation/serial/specialix.txt.
2382
2383	spia_io_base=	[HW,MTD]
2384	spia_fio_base=
2385	spia_pedr=
2386	spia_peddr=
2387
2388	stacktrace	[FTRACE]
2389			Enabled the stack tracer on boot up.
2390
 
 
 
 
 
 
 
 
2391	sti=		[PARISC,HW]
2392			Format: <num>
2393			Set the STI (builtin display/keyboard on the HP-PARISC
2394			machines) console (graphic card) which should be used
2395			as the initial boot-console.
2396			See also comment in drivers/video/console/sticore.c.
2397
2398	sti_font=	[HW]
2399			See comment in drivers/video/console/sticore.c.
2400
2401	stifb=		[HW]
2402			Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
2403
2404	sunrpc.min_resvport=
2405	sunrpc.max_resvport=
2406			[NFS,SUNRPC]
2407			SunRPC servers often require that client requests
2408			originate from a privileged port (i.e. a port in the
2409			range 0 < portnr < 1024).
2410			An administrator who wishes to reserve some of these
2411			ports for other uses may adjust the range that the
2412			kernel's sunrpc client considers to be privileged
2413			using these two parameters to set the minimum and
2414			maximum port values.
2415
2416	sunrpc.pool_mode=
2417			[NFS]
2418			Control how the NFS server code allocates CPUs to
2419			service thread pools.  Depending on how many NICs
2420			you have and where their interrupts are bound, this
2421			option will affect which CPUs will do NFS serving.
2422			Note: this parameter cannot be changed while the
2423			NFS server is running.
2424
2425			auto	    the server chooses an appropriate mode
2426				    automatically using heuristics
2427			global	    a single global pool contains all CPUs
2428			percpu	    one pool for each CPU
2429			pernode	    one pool for each NUMA node (equivalent
2430				    to global on non-NUMA machines)
2431
2432	sunrpc.tcp_slot_table_entries=
2433	sunrpc.udp_slot_table_entries=
2434			[NFS,SUNRPC]
2435			Sets the upper limit on the number of simultaneous
2436			RPC calls that can be sent from the client to a
2437			server. Increasing these values may allow you to
2438			improve throughput, but will also increase the
2439			amount of memory reserved for use by the client.
2440
2441	swapaccount[=0|1]
2442			[KNL] Enable accounting of swap in memory resource
2443			controller if no parameter or 1 is given or disable
2444			it if 0 is given (See Documentation/cgroups/memory.txt)
2445
2446	swiotlb=	[IA-64] Number of I/O TLB slabs
 
 
 
 
2447
2448	switches=	[HW,M68k]
2449
2450	sysfs.deprecated=0|1 [KNL]
2451			Enable/disable old style sysfs layout for old udev
2452			on older distributions. When this option is enabled
2453			very new udev will not work anymore. When this option
2454			is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
2455			in older udev will not work anymore.
2456			Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
2457			the kernel configuration.
2458
2459	sysrq_always_enabled
2460			[KNL]
2461			Ignore sysrq setting - this boot parameter will
2462			neutralize any effect of /proc/sys/kernel/sysrq.
2463			Useful for debugging.
2464
2465	tdfx=		[HW,DRM]
2466
2467	test_suspend=	[SUSPEND]
2468			Specify "mem" (for Suspend-to-RAM) or "standby" (for
2469			standby suspend) as the system sleep state to briefly
2470			enter during system startup.  The system is woken from
2471			this state using a wakeup-capable RTC alarm.
2472
2473	thash_entries=	[KNL,NET]
2474			Set number of hash buckets for TCP connection
2475
2476	thermal.act=	[HW,ACPI]
2477			-1: disable all active trip points in all thermal zones
2478			<degrees C>: override all lowest active trip points
2479
2480	thermal.crt=	[HW,ACPI]
2481			-1: disable all critical trip points in all thermal zones
2482			<degrees C>: override all critical trip points
2483
2484	thermal.nocrt=	[HW,ACPI]
2485			Set to disable actions on ACPI thermal zone
2486			critical and hot trip points.
2487
2488	thermal.off=	[HW,ACPI]
2489			1: disable ACPI thermal control
2490
2491	thermal.psv=	[HW,ACPI]
2492			-1: disable all passive trip points
2493			<degrees C>: override all passive trip points to this
2494			value
2495
2496	thermal.tzp=	[HW,ACPI]
2497			Specify global default ACPI thermal zone polling rate
2498			<deci-seconds>: poll all this frequency
2499			0: no polling (default)
2500
2501	threadirqs	[KNL]
2502			Force threading of all interrupt handlers except those
2503			marked explicitely IRQF_NO_THREAD.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2504
2505	topology=	[S390]
2506			Format: {off | on}
2507			Specify if the kernel should make use of the cpu
2508			topology information if the hardware supports this.
2509			The scheduler will make use of this information and
2510			e.g. base its process migration decisions on it.
2511			Default is on.
2512
2513	tp720=		[HW,PS2]
2514
2515	tpm_suspend_pcr=[HW,TPM]
2516			Format: integer pcr id
2517			Specify that at suspend time, the tpm driver
2518			should extend the specified pcr with zeros,
2519			as a workaround for some chips which fail to
2520			flush the last written pcr on TPM_SaveState.
2521			This will guarantee that all the other pcrs
2522			are saved.
2523
2524	trace_buf_size=nn[KMG]
2525			[FTRACE] will set tracing buffer size.
2526
2527	trace_event=[event-list]
2528			[FTRACE] Set and start specified trace events in order
2529			to facilitate early boot debugging.
2530			See also Documentation/trace/events.txt
2531
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2532	tsc=		Disable clocksource stability checks for TSC.
2533			Format: <string>
2534			[x86] reliable: mark tsc clocksource as reliable, this
2535			disables clocksource verification at runtime, as well
2536			as the stability checks done at bootup.	Used to enable
2537			high-resolution timer mode on older hardware, and in
2538			virtualized environment.
2539			[x86] noirqtime: Do not use TSC to do irq accounting.
2540			Used to run time disable IRQ_TIME_ACCOUNTING on any
2541			platforms where RDTSC is slow and this accounting
2542			can add overhead.
2543
2544	turbografx.map[2|3]=	[HW,JOY]
2545			TurboGraFX parallel port interface
2546			Format:
2547			<port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
2548			See also Documentation/input/joystick-parport.txt
2549
2550	udbg-immortal	[PPC] When debugging early kernel crashes that
2551			happen after console_init() and before a proper 
2552			console driver takes over, this boot options might
2553			help "seeing" what's going on.
2554
2555	uhash_entries=	[KNL,NET]
2556			Set number of hash buckets for UDP/UDP-Lite connections
2557
2558	uhci-hcd.ignore_oc=
2559			[USB] Ignore overcurrent events (default N).
2560			Some badly-designed motherboards generate lots of
2561			bogus events, for ports that aren't wired to
2562			anything.  Set this parameter to avoid log spamming.
2563			Note that genuine overcurrent events won't be
2564			reported either.
2565
2566	unknown_nmi_panic
2567			[X86] Cause panic on unknown NMI.
2568
2569	usbcore.authorized_default=
2570			[USB] Default USB device authorization:
2571			(default -1 = authorized except for wireless USB,
2572			0 = not authorized, 1 = authorized)
2573
2574	usbcore.autosuspend=
2575			[USB] The autosuspend time delay (in seconds) used
2576			for newly-detected USB devices (default 2).  This
2577			is the time required before an idle device will be
2578			autosuspended.  Devices for which the delay is set
2579			to a negative value won't be autosuspended at all.
2580
2581	usbcore.usbfs_snoop=
2582			[USB] Set to log all usbfs traffic (default 0 = off).
2583
2584	usbcore.blinkenlights=
2585			[USB] Set to cycle leds on hubs (default 0 = off).
2586
2587	usbcore.old_scheme_first=
2588			[USB] Start with the old device initialization
2589			scheme (default 0 = off).
2590
 
 
 
 
2591	usbcore.use_both_schemes=
2592			[USB] Try the other device initialization scheme
2593			if the first one fails (default 1 = enabled).
2594
2595	usbcore.initial_descriptor_timeout=
2596			[USB] Specifies timeout for the initial 64-byte
2597                        USB_REQ_GET_DESCRIPTOR request in milliseconds
2598			(default 5000 = 5.0 seconds).
2599
2600	usbhid.mousepoll=
2601			[USBHID] The interval which mice are to be polled at.
2602
2603	usb-storage.delay_use=
2604			[UMS] The delay in seconds before a new device is
2605			scanned for Logical Units (default 5).
2606
2607	usb-storage.quirks=
2608			[UMS] A list of quirks entries to supplement or
2609			override the built-in unusual_devs list.  List
2610			entries are separated by commas.  Each entry has
2611			the form VID:PID:Flags where VID and PID are Vendor
2612			and Product ID values (4-digit hex numbers) and
2613			Flags is a set of characters, each corresponding
2614			to a common usb-storage quirk flag as follows:
2615				a = SANE_SENSE (collect more than 18 bytes
2616					of sense data);
2617				b = BAD_SENSE (don't collect more than 18
2618					bytes of sense data);
2619				c = FIX_CAPACITY (decrease the reported
2620					device capacity by one sector);
2621				d = NO_READ_DISC_INFO (don't use
2622					READ_DISC_INFO command);
2623				e = NO_READ_CAPACITY_16 (don't use
2624					READ_CAPACITY_16 command);
2625				h = CAPACITY_HEURISTICS (decrease the
2626					reported device capacity by one
2627					sector if the number is odd);
2628				i = IGNORE_DEVICE (don't bind to this
2629					device);
2630				l = NOT_LOCKABLE (don't try to lock and
2631					unlock ejectable media);
2632				m = MAX_SECTORS_64 (don't transfer more
2633					than 64 sectors = 32 KB at a time);
2634				n = INITIAL_READ10 (force a retry of the
2635					initial READ(10) command);
2636				o = CAPACITY_OK (accept the capacity
2637					reported by the device);
 
 
2638				r = IGNORE_RESIDUE (the device reports
2639					bogus residue values);
2640				s = SINGLE_LUN (the device has only one
2641					Logical Unit);
2642				w = NO_WP_DETECT (don't test whether the
2643					medium is write-protected).
2644			Example: quirks=0419:aaf5:rl,0421:0433:rc
2645
2646	user_debug=	[KNL,ARM]
2647			Format: <int>
2648			See arch/arm/Kconfig.debug help text.
2649				 1 - undefined instruction events
2650				 2 - system calls
2651				 4 - invalid data aborts
2652				 8 - SIGSEGV faults
2653				16 - SIGBUS faults
2654			Example: user_debug=31
2655
2656	userpte=
2657			[X86] Flags controlling user PTE allocations.
2658
2659				nohigh = do not allocate PTE pages in
2660					HIGHMEM regardless of setting
2661					of CONFIG_HIGHPTE.
2662
2663	vdso=		[X86,SH]
2664			vdso=2: enable compat VDSO (default with COMPAT_VDSO)
2665			vdso=1: enable VDSO (default)
 
2666			vdso=0: disable VDSO mapping
2667
2668	vdso32=		[X86]
2669			vdso32=2: enable compat VDSO (default with COMPAT_VDSO)
2670			vdso32=1: enable 32-bit VDSO (default)
2671			vdso32=0: disable 32-bit VDSO mapping
 
 
 
 
 
 
 
 
 
2672
2673	vector=		[IA-64,SMP]
2674			vector=percpu: enable percpu vector domain
2675
2676	video=		[FB] Frame buffer configuration
2677			See Documentation/fb/modedb.txt.
2678
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2679	vga=		[BOOT,X86-32] Select a particular video mode
2680			See Documentation/x86/boot.txt and
2681			Documentation/svga.txt.
2682			Use vga=ask for menu.
2683			This is actually a boot loader parameter; the value is
2684			passed to the kernel using a special protocol.
2685
2686	vmalloc=nn[KMG]	[KNL,BOOT] Forces the vmalloc area to have an exact
2687			size of <nn>. This can be used to increase the
2688			minimum size (128MB on x86). It can also be used to
2689			decrease the size and leave more room for directly
2690			mapped kernel RAM.
2691
2692	vmhalt=		[KNL,S390] Perform z/VM CP command after system halt.
2693			Format: <command>
2694
2695	vmpanic=	[KNL,S390] Perform z/VM CP command after kernel panic.
2696			Format: <command>
2697
2698	vmpoff=		[KNL,S390] Perform z/VM CP command after power off.
2699			Format: <command>
2700
2701	vsyscall=	[X86-64]
2702			Controls the behavior of vsyscalls (i.e. calls to
2703			fixed addresses of 0xffffffffff600x00 from legacy
2704			code).  Most statically-linked binaries and older
2705			versions of glibc use these calls.  Because these
2706			functions are at fixed addresses, they make nice
2707			targets for exploits that can control RIP.
2708
2709			emulate     Vsyscalls turn into traps and are emulated
2710			            reasonably safely.
2711
2712			native      [default] Vsyscalls are native syscall
2713			            instructions.
2714			            This is a little bit faster than trapping
2715			            and makes a few dynamic recompilers work
2716			            better than they would in emulation mode.
2717			            It also makes exploits much easier to write.
2718
2719			none        Vsyscalls don't work at all.  This makes
2720			            them quite hard to use for exploits but
2721			            might break your system.
2722
 
 
 
 
2723	vt.cur_default=	[VT] Default cursor shape.
2724			Format: 0xCCBBAA, where AA, BB, and CC are the same as
2725			the parameters of the <Esc>[?A;B;Cc escape sequence;
2726			see VGA-softcursor.txt. Default: 2 = underline.
2727
2728	vt.default_blu=	[VT]
2729			Format: <blue0>,<blue1>,<blue2>,...,<blue15>
2730			Change the default blue palette of the console.
2731			This is a 16-member array composed of values
2732			ranging from 0-255.
2733
2734	vt.default_grn=	[VT]
2735			Format: <green0>,<green1>,<green2>,...,<green15>
2736			Change the default green palette of the console.
2737			This is a 16-member array composed of values
2738			ranging from 0-255.
2739
2740	vt.default_red=	[VT]
2741			Format: <red0>,<red1>,<red2>,...,<red15>
2742			Change the default red palette of the console.
2743			This is a 16-member array composed of values
2744			ranging from 0-255.
2745
2746	vt.default_utf8=
2747			[VT]
2748			Format=<0|1>
2749			Set system-wide default UTF-8 mode for all tty's.
2750			Default is 1, i.e. UTF-8 mode is enabled for all
2751			newly opened terminals.
2752
2753	vt.global_cursor_default=
2754			[VT]
2755			Format=<-1|0|1>
2756			Set system-wide default for whether a cursor
2757			is shown on new VTs. Default is -1,
2758			i.e. cursors will be created by default unless
2759			overridden by individual drivers. 0 will hide
2760			cursors, 1 will display them.
2761
 
 
 
 
 
 
2762	watchdog timers	[HW,WDT] For information on watchdog timers,
2763			see Documentation/watchdog/watchdog-parameters.txt
2764			or other driver-specific files in the
2765			Documentation/watchdog/ directory.
2766
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2767	x2apic_phys	[X86-64,APIC] Use x2apic physical mode instead of
2768			default x2apic cluster mode on platforms
2769			supporting x2apic.
2770
2771	x86_mrst_timer= [X86-32,APBT]
2772			Choose timer option for x86 Moorestown MID platform.
2773			Two valid options are apbt timer only and lapic timer
2774			plus one apbt timer for broadcast timer.
2775			x86_mrst_timer=apbt_only | lapic_and_apbt
2776
2777	xd=		[HW,XT] Original XT pre-IDE (RLL encoded) disks.
2778	xd_geo=		See header of drivers/block/xd.c.
2779
2780	xen_emul_unplug=		[HW,X86,XEN]
2781			Unplug Xen emulated devices
2782			Format: [unplug0,][unplug1]
2783			ide-disks -- unplug primary master IDE devices
2784			aux-ide-disks -- unplug non-primary-master IDE devices
2785			nics -- unplug network devices
2786			all -- unplug all emulated devices (NICs and IDE disks)
2787			unnecessary -- unplugging emulated devices is
2788				unnecessary even if the host did not respond to
2789				the unplug protocol
2790			never -- do not unplug even if version check succeeds
 
 
 
 
2791
2792	xirc2ps_cs=	[NET,PCMCIA]
2793			Format:
2794			<irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
2795
2796______________________________________________________________________
2797
2798TODO:
2799
2800	Add more DRM drivers.