4 The 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
7 case insensitive manner), and with descriptions where known.
9 Module parameters for loadable modules are specified only as the
10 parameter name with optional '=' and value as appropriate, such as:
12 modprobe usbcore blinkenlights=1
14 Module parameters for modules that are built into the kernel image
15 are specified on the kernel command line with the module name plus
16 '.' plus parameter name, with '=' and value if appropriate, such as:
18 usbcore.blinkenlights=1
20 Hyphens (dashes) and underscores are equivalent in parameter names, so
21 log_buf_len=1M print-fatal-signals=1
22 can also be entered as
23 log-buf-len=1M print_fatal_signals=1
26 This document may not be entirely up to date and comprehensive. The command
27 "modinfo -p ${modulename}" shows a current list of all parameters of a loadable
28 module. Loadable modules, after being loaded into the running kernel, also
29 reveal their parameters in /sys/module/${modulename}/parameters/. Some of these
30 parameters may be changed at runtime by the command
31 "echo -n ${value} > /sys/module/${modulename}/parameters/${parm}".
33 The parameters listed below are only valid if certain kernel build options were
34 enabled and if respective hardware is present. The text in square brackets at
35 the beginning of each description states the restrictions within which a
36 parameter is applicable:
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
126 In addition, the following text indicates that the option:
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.
132 Parameters denoted with BOOT are actually interpreted by the boot
133 loader, and have no meaning to the kernel directly.
134 Do not modify the syntax of boot loader parameters without extreme
135 need or coordination with <Documentation/x86/boot.txt>.
137 There are also arch-specific kernel-parameters not documented here.
138 See for example <Documentation/x86/x86_64/boot-options.txt>.
140 Note that ALL kernel parameters listed below are CASE SENSITIVE, and that
141 a trailing = on the name of any parameter states that that parameter will
142 be entered as an environment variable, whereas its absence indicates that
143 it will appear as a kernel argument readable via /proc/cmdline by programs
144 running once the system is up.
146 The number of kernel parameters is not limited, but the length of the
147 complete command line (parameters including spaces etc.) is limited to
148 a fixed number of characters. This limit depends on the architecture
149 and is between 256 and 4096 characters. It is defined in the file
150 ./include/asm/setup.h as COMMAND_LINE_SIZE.
152 Finally, the [KMG] suffix is commonly described after a number of kernel
153 parameter values. These 'K', 'M', and 'G' letters represent the _binary_
154 multipliers 'Kilo', 'Mega', and 'Giga', equalling 2^10, 2^20, and 2^30
155 bytes respectively. Such letter suffixes can also be entirely omitted.
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
169 See also Documentation/power/runtime_pm.txt, pci=noacpi
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.
176 acpi_apic_instance= [ACPI, IOAPIC]
178 2: use 2nd APIC table, if available
179 1,0: use 1st APIC table
182 acpi_backlight= [HW,ACPI]
183 acpi_backlight=vendor
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.
189 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
190 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
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.
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
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.
217 acpi_irq_balance [HW,ACPI]
218 ACPI will balance active IRQs
221 acpi_irq_nobalance [HW,ACPI]
222 ACPI will not move active IRQs (default)
225 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
226 Format: <irq>,<irq>...
228 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
230 Format: <irq>,<irq>...
232 acpi_no_auto_ssdt [HW,ACPI] Disable automatic loading of SSDT
234 acpica_no_return_repair [HW, ACPI]
235 Disable AML predefined validation mechanism
236 This mechanism can repair the evaluation result to make
237 the return objects more ACPI specification compliant.
238 This option is useful for developers to identify the
239 root cause of an AML interpreter issue when the issue
240 has something to do with the repair mechanism.
242 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
243 Format: To spoof as Windows 98: ="Microsoft Windows"
245 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
246 acpi_osi="string1" # add string1
247 acpi_osi="!string2" # remove string2
248 acpi_osi=!* # remove all strings
249 acpi_osi=! # disable all built-in OS vendor
251 acpi_osi= # disable all strings
253 'acpi_osi=!' can be used in combination with single or
254 multiple 'acpi_osi="string1"' to support specific OS
255 vendor string(s). Note that such command can only
256 affect the default state of the OS vendor strings, thus
257 it cannot affect the default state of the feature group
258 strings and the current state of the OS vendor strings,
259 specifying it multiple times through kernel command line
260 is meaningless. This command is useful when one do not
261 care about the state of the feature group strings which
262 should be controlled by the OSPM.
264 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
265 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
266 can make '_OSI("Windows 2000")' TRUE.
268 'acpi_osi=' cannot be used in combination with other
269 'acpi_osi=' command lines, the _OSI method will not
270 exist in the ACPI namespace. NOTE that such command can
271 only affect the _OSI support state, thus specifying it
272 multiple times through kernel command line is also
275 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
278 'acpi_osi=!*' can be used in combination with single or
279 multiple 'acpi_osi="string1"' to support specific
280 string(s). Note that such command can affect the
281 current state of both the OS vendor strings and the
282 feature group strings, thus specifying it multiple times
283 through kernel command line is meaningful. But it may
284 still not able to affect the final state of a string if
285 there are quirks related to this string. This command
286 is useful when one want to control the state of the
287 feature group strings to debug BIOS issues related to
290 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
291 '_OSI("Module Device")' FALSE.
292 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
293 '_OSI("Module Device")' TRUE.
294 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
296 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
298 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
299 they all will make '_OSI("Windows 2000")' TRUE.
302 Override the pmtimer bug detection: force the kernel
303 to assume that this machine's pmtimer latches its value
304 and always returns good values.
306 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
307 Format: { level | edge | high | low }
309 acpi_serialize [HW,ACPI] force serialization of AML methods
311 acpi_skip_timer_override [HW,ACPI]
312 Recognize and ignore IRQ0/pin2 Interrupt Override.
313 For broken nForce2 BIOS resulting in XT-PIC timer.
315 acpi_sleep= [HW,ACPI] Sleep options
316 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
317 old_ordering, nonvs, sci_force_enable }
318 See Documentation/power/video.txt for information on
320 s3_beep is for debugging; it makes the PC's speaker beep
321 as soon as the kernel's real-mode entry point is called.
322 s4_nohwsig prevents ACPI hardware signature from being
323 used during resume from hibernation.
324 old_ordering causes the ACPI 1.0 ordering of the _PTS
325 control method, with respect to putting devices into
326 low power states, to be enforced (the ACPI 2.0 ordering
327 of _PTS is used by default).
328 nonvs prevents the kernel from saving/restoring the
329 ACPI NVS memory during suspend/hibernation and resume.
330 sci_force_enable causes the kernel to set SCI_EN directly
331 on resume from S1/S3 (which is against the ACPI spec,
332 but some broken systems don't work without it).
334 acpi_use_timer_override [HW,ACPI]
335 Use timer override. For some broken Nvidia NF5 boards
336 that require a timer override, but don't have HPET
338 acpi_enforce_resources= [ACPI]
339 { strict | lax | no }
340 Check for resource conflicts between native drivers
341 and ACPI OperationRegions (SystemIO and SystemMemory
342 only). IO ports and memory declared in ACPI might be
343 used by the ACPI subsystem in arbitrary AML code and
344 can interfere with legacy drivers.
345 strict (default): access to resources claimed by ACPI
346 is denied; legacy drivers trying to access reserved
347 resources will fail to bind to device using them.
348 lax: access to resources claimed by ACPI is allowed;
349 legacy drivers trying to access reserved resources
350 will bind successfully but a warning message is logged.
351 no: ACPI OperationRegions are not marked as reserved,
352 no further checks are performed.
354 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
357 add_efi_memmap [EFI; X86] Include EFI memory map in
358 kernel's map of available physical RAM.
361 { off | try_unsupported }
362 off: disable AGP support
363 try_unsupported: try to drive unsupported chipsets
364 (may crash computer or cause data corruption)
367 See Documentation/sound/alsa/alsa-parameters.txt
370 Allow the default userspace alignment fault handler
371 behaviour to be specified. Bit 0 enables warnings,
372 bit 1 enables fixups, and bit 2 sends a segfault.
374 align_va_addr= [X86-64]
375 Align virtual addresses by clearing slice [14:12] when
376 allocating a VMA at process creation time. This option
377 gives you up to 3% performance improvement on AMD F15h
378 machines (where it is enabled by default) for a
379 CPU-intensive style benchmark, and it can vary highly in
380 a microbenchmark depending on workload and compiler.
382 32: only for 32-bit processes
383 64: only for 64-bit processes
384 on: enable for both 32- and 64-bit processes
385 off: disable for both 32- and 64-bit processes
387 alloc_snapshot [FTRACE]
388 Allocate the ftrace snapshot buffer on boot up when the
389 main buffer is allocated. This is handy if debugging
390 and you need to use tracing_snapshot() on boot up, and
391 do not want to use tracing_snapshot_alloc() as it needs
392 to be done where GFP_KERNEL allocations are allowed.
394 amd_iommu= [HW,X86-64]
395 Pass parameters to the AMD IOMMU driver in the system.
397 fullflush - enable flushing of IO/TLB entries when
398 they are unmapped. Otherwise they are
399 flushed before they will be reused, which
401 off - do not initialize any AMD IOMMU found in
403 force_isolation - Force device isolation for all
404 devices. The IOMMU driver is not
405 allowed anymore to lift isolation
406 requirements as needed. This option
407 does not override iommu=pt
409 amd_iommu_dump= [HW,X86-64]
410 Enable AMD IOMMU driver option to dump the ACPI table
411 for AMD IOMMU. With this option enabled, AMD IOMMU
412 driver will print ACPI tables for AMD IOMMU during
413 IOMMU initialization.
415 amijoy.map= [HW,JOY] Amiga joystick support
416 Map of devices attached to JOY0DAT and JOY1DAT
418 See also Documentation/input/joystick.txt
420 analog.map= [HW,JOY] Analog joystick and gamepad support
421 Specifies type or capabilities of an analog joystick
422 connected to one of 16 gameports
423 Format: <type1>,<type2>,..<type16>
426 Power management functions (SPARCstation-4/5 + deriv.)
428 Disable APC CPU standby support. SPARCstation-Fox does
429 not play well with APC CPU idle - disable it if you have
430 APC and your system crashes randomly.
432 apic= [APIC,X86-32] Advanced Programmable Interrupt Controller
433 Change the output verbosity whilst booting
434 Format: { quiet (default) | verbose | debug }
435 Change the amount of debugging information output
436 when initialising the APIC and IO-APIC components.
439 See Documentation/networking/ipv6.txt.
441 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
442 Limit apic dumping. The parameter defines the maximal
443 number of local apics being dumped. Also it is possible
444 to set it to "all" by meaning -- no limit here.
445 Format: { 1 (default) | 2 | ... | all }.
446 The parameter valid if only apic=debug or
447 apic=verbose is specified.
448 Example: apic=debug show_lapic=all
450 apm= [APM] Advanced Power Management
451 See header of arch/x86/kernel/apm_32.c.
453 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
454 Format: <io>,<irq>,<nodeID>
458 atarimouse= [HW,MOUSE] Atari Mouse
460 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
461 EzKey and similar keyboards
463 atkbd.reset= [HW] Reset keyboard during initialization
465 atkbd.set= [HW] Select keyboard code set
466 Format: <int> (2 = AT (default), 3 = PS/2)
468 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
471 atkbd.softraw= [HW] Choose between synthetic and real raw mode
472 Format: <bool> (0 = real, 1 = synthetic (default))
474 atkbd.softrepeat= [HW]
475 Use software keyboard repeat
477 audit= [KNL] Enable the audit sub-system
478 Format: { "0" | "1" } (0 = disabled, 1 = enabled)
479 0 - kernel audit is disabled and can not be enabled
480 until the next reboot
481 unset - kernel audit is initialized but disabled and
482 will be fully enabled by the userspace auditd.
483 1 - kernel audit is initialized and partially enabled,
484 storing at most audit_backlog_limit messages in
485 RAM until it is fully enabled by the userspace
489 audit_backlog_limit= [KNL] Set the audit queue size limit.
490 Format: <int> (must be >=0)
493 baycom_epp= [HW,AX25]
496 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
498 See header of drivers/net/hamradio/baycom_par.c.
500 baycom_ser_fdx= [HW,AX25]
501 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
502 Format: <io>,<irq>,<mode>[,<baud>]
503 See header of drivers/net/hamradio/baycom_ser_fdx.c.
505 baycom_ser_hdx= [HW,AX25]
506 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
507 Format: <io>,<irq>,<mode>
508 See header of drivers/net/hamradio/baycom_ser_hdx.c.
510 blkdevparts= Manual partition parsing of block device(s) for
511 embedded devices based on command line input.
512 See Documentation/block/cmdline-partition.txt
514 boot_delay= Milliseconds to delay each printk during boot.
515 Values larger than 10 seconds (10000) are changed to
519 bootmem_debug [KNL] Enable bootmem allocator debug messages.
521 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
522 bttv.radio= Most important insmod options are available as
524 bttv.pll= See Documentation/video4linux/bttv/Insmod-options
527 bulk_remove=off [PPC] This parameter disables the use of the pSeries
528 firmware feature for flushing multiple hpte entries
531 c101= [NET] Moxa C101 synchronous serial card
533 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
534 Sometimes CPU hardware bugs make them report the cache
535 size incorrectly. The kernel will attempt work arounds
536 to fix known problems, but for some CPUs it is not
537 possible to determine what the correct size should be.
538 This option provides an override for these situations.
540 ccw_timeout_log [S390]
541 See Documentation/s390/CommonIO for details.
543 cgroup_disable= [KNL] Disable a particular controller
544 Format: {name of the controller(s) to disable}
545 The effects of cgroup_disable=foo are:
546 - foo isn't auto-mounted if you mount all cgroups in
548 - foo isn't visible as an individually mountable
550 {Currently only "memory" controller deal with this and
551 cut the overhead, others just disable the usage. So
552 only cgroup_disable=memory is actually worthy}
554 checkreqprot [SELINUX] Set initial checkreqprot flag value.
555 Format: { "0" | "1" }
556 See security/selinux/Kconfig help text.
557 0 -- check protection applied by kernel (includes
558 any implied execute protection).
559 1 -- check protection requested by application.
560 Default value is set via a kernel config option.
561 Value can be changed at runtime via
562 /selinux/checkreqprot.
565 See Documentation/s390/CommonIO for details.
568 Keep all clocks already enabled by bootloader on,
569 even if no driver has claimed them. This is useful
570 for debug and development, but should not be
571 needed on a platform with proper driver support.
572 For more information, see Documentation/clk.txt.
574 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
576 Forces specified clocksource (if available) to be used
577 when calculating gettimeofday(). If specified
578 clocksource is not available, it defaults to PIT.
579 Format: { pit | tsc | cyclone | pmtmr }
581 clocksource= Override the default clocksource
583 Override the default clocksource and use the clocksource
584 with the name specified.
585 Some clocksource names to choose from, depending on
587 [all] jiffies (this is the base, fallback clocksource)
589 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
590 pxa_timer,timer3,32k_counter,timer0_1
592 [X86-32] pit,hpet,tsc;
593 scx200_hrt on Geode; cyclone on IBM x440
601 clearcpuid=BITNUM [X86]
602 Disable CPUID feature X for the kernel. See
603 arch/x86/include/asm/cpufeature.h for the valid bit
604 numbers. Note the Linux specific bits are not necessarily
605 stable over kernel options, but the vendor specific
607 Also note that user programs calling CPUID directly
608 or using the feature without checking anything
609 will still see it. This just prevents it from
610 being used by the kernel or shown in /proc/cpuinfo.
611 Also note the kernel might malfunction if you disable
615 Sets the size of kernel global memory area for contiguous
616 memory allocations. For more information, see
617 include/linux/dma-contiguous.h
619 cmo_free_hint= [PPC] Format: { yes | no }
620 Specify whether pages are marked as being inactive
621 when they are freed. This is used in CMO environments
622 to determine OS memory pressure for page stealing by
626 coherent_pool=nn[KMG] [ARM,KNL]
627 Sets the size of memory pool for coherent, atomic dma
628 allocations, by default set to 256K.
630 code_bytes [X86] How many bytes of object code to print
635 com20020= [HW,NET] ARCnet - COM20020 chipset
637 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
639 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
643 ARCnet - COM90xx chipset (memory-mapped buffers)
644 Format: <io>[,<irq>[,<memstart>]]
646 condev= [HW,S390] console device
649 console= [KNL] Output console device and options.
651 tty<n> Use the virtual console device <n>.
655 Use the specified serial port. The options are of
656 the form "bbbbpnf", where "bbbb" is the baud rate,
657 "p" is parity ("n", "o", or "e"), "n" is number of
658 bits, and "f" is flow control ("r" for RTS or
659 omit it). Default is "9600n8".
661 See Documentation/serial-console.txt for more
663 Documentation/networking/netconsole.txt for an
666 uart[8250],io,<addr>[,options]
667 uart[8250],mmio,<addr>[,options]
668 Start an early, polled-mode console on the 8250/16550
669 UART at the specified I/O port or MMIO address,
670 switching to the matching ttyS device later. The
671 options are the same as for ttyS, above.
672 hvc<n> Use the hypervisor console device <n>. This is for
673 both Xen and PowerPC hypervisors.
675 If the device connected to the port is not a TTY but a braille
676 device, prepend "brl," before the device type, for instance
678 For now, only VisioBraille is supported.
680 consoleblank= [KNL] The console blank (screen saver) timeout in
681 seconds. Defaults to 10*60 = 10mins. A value of 0
682 disables the blank timer.
685 [KNL] Change the default value for
686 /proc/<pid>/coredump_filter.
687 See also Documentation/filesystems/proc.txt.
689 cpuidle.off=1 [CPU_IDLE]
690 disable the cpuidle sub-system
692 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
694 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
696 crashkernel=size[KMG][@offset[KMG]]
697 [KNL] Using kexec, Linux can switch to a 'crash kernel'
698 upon panic. This parameter reserves the physical
699 memory region [offset, offset + size] for that kernel
700 image. If '@offset' is omitted, then a suitable offset
701 is selected automatically. Check
702 Documentation/kdump/kdump.txt for further details.
704 crashkernel=range1:size1[,range2:size2,...][@offset]
705 [KNL] Same as above, but depends on the memory
706 in the running system. The syntax of range is
707 start-[end] where start and end are both
708 a memory unit (amount[KMG]). See also
709 Documentation/kdump/kdump.txt for an example.
711 crashkernel=size[KMG],high
712 [KNL, x86_64] range could be above 4G. Allow kernel
713 to allocate physical memory region from top, so could
714 be above 4G if system have more than 4G ram installed.
715 Otherwise memory region will be allocated below 4G, if
717 It will be ignored if crashkernel=X is specified.
718 crashkernel=size[KMG],low
719 [KNL, x86_64] range under 4G. When crashkernel=X,high
720 is passed, kernel could allocate physical memory region
721 above 4G, that cause second kernel crash on system
722 that require some amount of low memory, e.g. swiotlb
723 requires at least 64M+32K low memory. Kernel would
724 try to allocate 72M below 4G automatically.
725 This one let user to specify own low range under 4G
726 for second kernel instead.
727 0: to disable low allocation.
728 It will be ignored when crashkernel=X,high is not used
729 or memory reserved is below 4G.
734 cs89x0_media= [HW,NET]
735 Format: { rj45 | aui | bnc }
738 See header of drivers/s390/block/dasd_devmap.c.
740 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
741 (one device per port)
742 Format: <port#>,<type>
743 See also Documentation/input/joystick-parport.txt
745 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
746 time. See Documentation/dynamic-debug-howto.txt for
747 details. Deprecated, see dyndbg.
749 debug [KNL] Enable kernel debugging (events log level).
752 [KNL] verbose self-tests
754 Print debugging info while doing the locking API
756 We default to 0 (no extra messages), setting it to
757 1 will print _a lot_ more information - normally
758 only useful to kernel developers.
760 debug_objects [KNL] Enable object debugging
763 [KNL] Disable object debugging
765 debug_guardpage_minorder=
766 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
767 parameter allows control of the order of pages that will
768 be intentionally kept free (and hence protected) by the
769 buddy allocator. Bigger value increase the probability
770 of catching random memory corruption, but reduce the
771 amount of memory for normal system use. The maximum
772 possible value is MAX_ORDER/2. Setting this parameter
773 to 1 or 2 should be enough to identify most random
774 memory corruption problems caused by bugs in kernel or
775 driver code when a CPU writes to (or reads from) a
776 random memory location. Note that there exists a class
777 of memory corruptions problems caused by buggy H/W or
778 F/W or by drivers badly programing DMA (basically when
779 memory is written at bus level and the CPU MMU is
780 bypassed) which are not detectable by
781 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
782 tracking down these problems.
784 debugpat [X86] Enable PAT debugging
786 decnet.addr= [HW,NET]
787 Format: <area>[,<node>]
788 See also Documentation/networking/decnet.txt.
791 [same as hugepagesz=] The size of the default
792 HugeTLB page size. This is the size represented by
793 the legacy /proc/ hugepages APIs, used for SHM, and
794 default size when mounting hugetlbfs filesystems.
795 Defaults to the default architecture's huge page size
799 Set number of hash buckets for dentry cache.
802 IO parameters + enable/disable command.
804 digiepca= [HW,SERIAL]
805 See drivers/char/README.epca and
806 Documentation/serial/digiepca.txt.
809 See Documentation/networking/ipv6.txt.
811 disable_cpu_apicid= [X86,APIC,SMP]
813 The number of initial APIC ID for the
814 corresponding CPU to be disabled at boot,
815 mostly used for the kdump 2nd kernel to
816 disable BSP to wake up multiple CPUs without
817 causing system reset or hang due to sending
820 disable_ddw [PPC/PSERIES]
821 Disable Dynamic DMA Window support. Use this if
822 to workaround buggy firmware.
825 See Documentation/networking/ipv6.txt.
827 disable_mtrr_cleanup [X86]
828 The kernel tries to adjust MTRR layout from continuous
829 to discrete, to make X server driver able to add WB
830 entry later. This parameter disables that.
832 disable_mtrr_trim [X86, Intel and AMD only]
833 By default the kernel will trim any uncacheable
834 memory out of your available memory pool based on
835 MTRR settings. This parameter disables that behavior,
836 possibly causing your machine to run very slowly.
838 disable_timer_pin_1 [X86]
839 Disable PIN 1 of APIC timer
840 Can be useful to work around chipset bugs.
842 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
843 this option disables the debugging code at boot.
845 dma_debug_entries=<number>
846 This option allows to tune the number of preallocated
847 entries for DMA-API debugging code. One entry is
848 required per DMA-API allocation. Use this if the
849 DMA-API debugging code disables itself because the
850 architectural default is too low.
852 dma_debug_driver=<driver_name>
853 With this option the DMA-API debugging driver
854 filter feature can be enabled at boot time. Just
855 pass the driver to filter for as the parameter.
856 The filter can be disabled or changed to another
857 driver later using sysfs.
859 drm_kms_helper.edid_firmware=[<connector>:]<file>
860 Broken monitors, graphic adapters and KVMs may
861 send no or incorrect EDID data sets. This parameter
862 allows to specify an EDID data set in the
863 /lib/firmware directory that is used instead.
864 Generic built-in EDID data sets are used, if one of
865 edid/1024x768.bin, edid/1280x1024.bin,
866 edid/1680x1050.bin, or edid/1920x1080.bin is given
867 and no file with the same name exists. Details and
868 instructions how to build your own EDID data are
869 available in Documentation/EDID/HOWTO.txt. An EDID
870 data set will only be used for a particular connector,
871 if its name and a colon are prepended to the EDID
876 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
877 module.dyndbg[="val"]
878 Enable debug messages at boot time. See
879 Documentation/dynamic-debug-howto.txt for details.
881 earlycon= [KNL] Output early console device and options.
882 uart[8250],io,<addr>[,options]
883 uart[8250],mmio,<addr>[,options]
884 uart[8250],mmio32,<addr>[,options]
885 Start an early, polled-mode console on the 8250/16550
886 UART at the specified I/O port or MMIO address.
887 MMIO inter-register address stride is either 8-bit
888 (mmio) or 32-bit (mmio32).
889 The options are the same as for ttyS, above.
891 earlyprintk= [X86,SH,BLACKFIN,ARM]
895 earlyprintk=serial[,ttySn[,baudrate]]
896 earlyprintk=serial[,0x...[,baudrate]]
897 earlyprintk=ttySn[,baudrate]
898 earlyprintk=dbgp[debugController#]
900 earlyprintk is useful when the kernel crashes before
901 the normal console is initialized. It is not enabled by
902 default because it has some cosmetic problems.
904 Append ",keep" to not disable it when the real console
907 Only one of vga, efi, serial, or usb debug port can
910 Currently only ttyS0 and ttyS1 may be specified by
911 name. Other I/O ports may be explicitly specified
912 on some architectures (x86 and arm at least) by
913 replacing ttySn with an I/O port address, like this:
914 earlyprintk=serial,0x1008,115200
915 You can find the port for a given device in
916 /proc/tty/driver/serial:
917 2: uart:ST16650V2 port:00001008 irq:18 ...
919 Interaction with the standard serial driver is not
922 The VGA and EFI output is eventually overwritten by
925 The xen output can only be used by Xen PV guests.
927 edac_report= [HW,EDAC] Control how to report EDAC event
928 Format: {"on" | "off" | "force"}
929 on: enable EDAC to report H/W event. May be overridden
930 by other higher priority error reporting module.
931 off: disable H/W event reporting through EDAC.
932 force: enforce the use of EDAC to report H/W event.
935 ekgdboc= [X86,KGDB] Allow early kernel console debugging
938 This is designed to be used in conjunction with
939 the boot argument: earlyprintk=vga
942 Format: {"off" | "on" | "skip[mbr]"}
945 Format: { "old_map" }
946 old_map [X86-64]: switch to the old ioremap-based EFI
947 runtime services mapping. 32-bit still uses this one by
950 efi_no_storage_paranoia [EFI; X86]
951 Using this parameter you can use more than 50% of
952 your efi variable storage. Use this parameter only if
953 you are really sure that your UEFI does sane gc and
954 fulfills the spec otherwise your board may brick.
956 eisa_irq_edge= [PARISC,HW]
957 See header of drivers/parisc/eisa.c.
960 See comment before function elanfreq_setup() in
961 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
964 Format: {"cfq" | "deadline" | "noop"}
965 See Documentation/block/cfq-iosched.txt and
966 Documentation/block/deadline-iosched.txt for details.
968 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
969 Specifies physical address of start of kernel core
970 image elf header and optionally the size. Generally
971 kexec loader will pass this option to capture kernel.
972 See Documentation/kdump/kdump.txt for details.
974 enable_mtrr_cleanup [X86]
975 The kernel tries to adjust MTRR layout from continuous
976 to discrete, to make X server driver able to add WB
977 entry later. This parameter enables that.
979 enable_timer_pin_1 [X86]
980 Enable PIN 1 of APIC timer
981 Can be useful to work around chipset bugs
982 (in particular on some ATI chipsets).
983 The kernel tries to set a reasonable default.
985 enforcing [SELINUX] Set initial enforcing status.
987 See security/selinux/Kconfig help text.
988 0 -- permissive (log only, no denials).
989 1 -- enforcing (deny and log).
991 Value can be changed at runtime via /selinux/enforce.
994 Disable Error Record Serialization Table (ERST)
997 ether= [HW,NET] Ethernet cards parameters
998 This option is obsoleted by the "netdev=" option, which
999 has equivalent usage. See its documentation for details.
1003 Permit 'security.evm' to be updated regardless of
1004 current integrity status.
1008 fail_make_request=[KNL]
1009 General fault injection mechanism.
1010 Format: <interval>,<probability>,<space>,<times>
1011 See also Documentation/fault-injection/.
1014 See Documentation/blockdev/floppy.txt.
1016 force_pal_cache_flush
1017 [IA-64] Avoid check_sal_cache_flush which may hang on
1018 buggy SAL_CACHE_FLUSH implementations. Using this
1019 parameter will force ia64_sal_cache_flush to call
1020 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1023 Forcefully enable Physical Address Extension (PAE).
1024 Many Pentium M systems disable PAE but may have a
1025 functionally usable PAE implementation.
1026 Warning: use of this parameter will taint the kernel
1027 and may cause unknown problems.
1030 [FTRACE] will set and start the specified tracer
1031 as early as possible in order to facilitate early
1034 ftrace_dump_on_oops[=orig_cpu]
1035 [FTRACE] will dump the trace buffers on oops.
1036 If no parameter is passed, ftrace will dump
1037 buffers of all CPUs, but if you pass orig_cpu, it will
1038 dump only the buffer of the CPU that triggered the
1041 ftrace_filter=[function-list]
1042 [FTRACE] Limit the functions traced by the function
1043 tracer at boot up. function-list is a comma separated
1044 list of functions. This list can be changed at run
1045 time by the set_ftrace_filter file in the debugfs
1048 ftrace_notrace=[function-list]
1049 [FTRACE] Do not trace the functions specified in
1050 function-list. This list can be changed at run time
1051 by the set_ftrace_notrace file in the debugfs
1054 ftrace_graph_filter=[function-list]
1055 [FTRACE] Limit the top level callers functions traced
1056 by the function graph tracer at boot up.
1057 function-list is a comma separated list of functions
1058 that can be changed at run time by the
1059 set_graph_function file in the debugfs tracing directory.
1062 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1063 support via parallel port (up to 5 devices per port)
1064 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1065 See also Documentation/input/joystick-parport.txt
1069 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1073 gcov_persist= [GCOV] When non-zero (default), profiling data for
1074 kernel modules is saved and remains accessible via
1075 debugfs, even when the module is unloaded/reloaded.
1076 When zero, profiling data is discarded and associated
1077 debugfs files are removed at module unload time.
1079 gpt [EFI] Forces disk with valid GPT signature but
1080 invalid Protective MBR to be treated as GPT. If the
1081 primary GPT is corrupted, it enables the backup/alternate
1082 GPT to be used instead.
1084 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1085 the "Enable 0" bit of the configuration register.
1088 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1089 the "Enable 0" bit of the configuration register.
1092 grcan.select= [HW] Select which physical interface to use.
1095 grcan.txsize= [HW] Sets the size of the tx buffer.
1096 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1098 grcan.rxsize= [HW] Sets the size of the rx buffer.
1099 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1102 hashdist= [KNL,NUMA] Large hashes allocated during boot
1103 are distributed across NUMA nodes. Defaults on
1104 for 64-bit NUMA, off otherwise.
1105 Format: 0 | 1 (for off | on)
1107 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1109 hd= [EIDE] (E)IDE hard drive subsystem geometry
1110 Format: <cyl>,<head>,<sect>
1113 Disable Hardware Error Source Table (HEST) support;
1114 corresponding firmware-first mode error processing
1115 logic will be disabled.
1117 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1118 size of <nn>. This works even on boxes that have no
1119 highmem otherwise. This also works to reduce highmem
1120 size on bigger boxes.
1122 highres= [KNL] Enable/disable high resolution timer mode.
1123 Valid parameters: "on", "off"
1127 See Documentation/isdn/README.HiSax.
1131 hpet= [X86-32,HPET] option to control HPET usage
1132 Format: { enable (default) | disable | force |
1134 disable: disable HPET and use PIT instead
1135 force: allow force enabled of undocumented chips (ICH4,
1137 verbose: show contents of HPET registers during setup
1139 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1140 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1142 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1143 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
1144 On x86-64 and powerpc, this option can be specified
1145 multiple times interleaved with hugepages= to reserve
1146 huge pages of different sizes. Valid pages sizes on
1147 x86-64 are 2M (when the CPU supports "pse") and 1G
1148 (when the CPU supports the "pdpe1gb" cpuinfo flag)
1149 Note that 1GB pages can only be allocated at boot time
1150 using hugepages= and not freed afterwards.
1152 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1153 terminal devices. Valid values: 0..8
1154 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1155 If specified, z/VM IUCV HVC accepts connections
1156 from listed z/VM user IDs only.
1158 hwthread_map= [METAG] Comma-separated list of Linux cpu id to
1159 hardware thread id mappings.
1160 Format: <cpu>:<hwthread>
1163 Do not unregister boot console at start. This is only
1164 useful for debugging when something happens in the window
1165 between unregistering the boot console and initializing
1168 i2c_bus= [HW] Override the default board specific I2C bus speed
1169 or register an additional I2C bus that is not
1170 registered from board initialization code.
1174 i8042.debug [HW] Toggle i8042 debug mode
1175 i8042.direct [HW] Put keyboard port into non-translated mode
1176 i8042.dumbkbd [HW] Pretend that controller can only read data from
1177 keyboard and cannot control its state
1178 (Don't attempt to blink the leds)
1179 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1180 i8042.nokbd [HW] Don't check/create keyboard port
1181 i8042.noloop [HW] Disable the AUX Loopback command while probing
1183 i8042.nomux [HW] Don't check presence of an active multiplexing
1185 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1187 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1188 i8042.reset [HW] Reset the controller during init and cleanup
1189 i8042.unlock [HW] Unlock (ignore) the keylock
1193 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1194 indicates that the driver is running on unsupported
1196 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1197 does not match list of supported models.
1199 [HW] Report power status in /proc/i8k
1200 (disabled by default)
1201 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1204 i915.invert_brightness=
1205 [DRM] Invert the sense of the variable that is used to
1206 set the brightness of the panel backlight. Normally a
1207 brightness value of 0 indicates backlight switched off,
1208 and the maximum of the brightness value sets the backlight
1209 to maximum brightness. If this parameter is set to 0
1210 (default) and the machine requires it, or this parameter
1211 is set to 1, a brightness value of 0 sets the backlight
1212 to maximum brightness, and the maximum of the brightness
1213 value switches the backlight off.
1214 -1 -- never invert brightness
1215 0 -- machine default
1216 1 -- force brightness inversion
1219 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1221 ide-core.nodma= [HW] (E)IDE subsystem
1222 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1223 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1224 .cdrom .chs .ignore_cable are additional options
1225 See Documentation/ide/ide.txt.
1227 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1228 Claim all unknown PCI IDE storage controllers.
1231 Format: idle=poll, idle=halt, idle=nomwait
1232 Poll forces a polling idle loop that can slightly
1233 improve the performance of waking up a idle CPU, but
1234 will use a lot of power and make the system run hot.
1236 idle=halt: Halt is forced to be used for CPU idle.
1237 In such case C2/C3 won't be used again.
1238 idle=nomwait: Disable mwait for CPU C-states
1240 ignore_loglevel [KNL]
1241 Ignore loglevel setting - this will print /all/
1242 kernel messages to the console. Useful for debugging.
1243 We also add it as printk module parameter, so users
1244 could change it dynamically, usually by
1245 /sys/module/printk/parameters/ignore_loglevel.
1247 ihash_entries= [KNL]
1248 Set number of hash buckets for inode cache.
1250 ima_appraise= [IMA] appraise integrity measurements
1251 Format: { "off" | "enforce" | "fix" }
1254 ima_appraise_tcb [IMA]
1255 The builtin appraise policy appraises all files
1259 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1263 The list of supported hash algorithms is defined
1264 in crypto/hash_info.h.
1267 Load a policy which meets the needs of the Trusted
1268 Computing Base. This means IMA will measure all
1269 programs exec'd, files mmap'd for exec, and all files
1270 opened for read by uid=0.
1273 Select one of defined IMA measurements template formats.
1274 Formats: { "ima" | "ima-ng" }
1279 Run specified binary instead of /sbin/init as init
1282 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1283 for working out where the kernel is dying during
1286 initrd= [BOOT] Specify the location of the initial ramdisk
1288 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1291 int_pln_enable [x86] Enable power limit notification interrupt
1293 integrity_audit=[IMA]
1294 Format: { "0" | "1" }
1295 0 -- basic integrity auditing messages. (Default)
1296 1 -- additional integrity auditing messages.
1298 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1300 Enable intel iommu driver.
1302 Disable intel iommu driver.
1303 igfx_off [Default Off]
1304 By default, gfx is mapped as normal device. If a gfx
1305 device has a dedicated DMAR unit, the DMAR unit is
1306 bypassed by not enabling DMAR with this option. In
1307 this case, gfx device will use physical address for
1310 With this option iommu will not optimize to look
1311 for io virtual address below 32-bit forcing dual
1312 address cycle on pci bus for cards supporting greater
1313 than 32-bit addressing. The default is to look
1314 for translation below 32-bit and if not available
1315 then look in the higher range.
1316 strict [Default Off]
1317 With this option on every unmap_single operation will
1318 result in a hardware IOTLB flush operation as opposed
1319 to batching them for performance.
1320 sp_off [Default Off]
1321 By default, super page will be supported if Intel IOMMU
1322 has the capability. With this option, super page will
1325 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1326 0 disables intel_idle and fall back on acpi_idle.
1327 1 to 6 specify maximum depth of C-state.
1331 Do not enable intel_pstate as the default
1332 scaling driver for the supported processors
1334 intremap= [X86-64, Intel-IOMMU]
1335 on enable Interrupt Remapping (default)
1336 off disable Interrupt Remapping
1337 nosid disable Source ID checking
1339 BIOS x2APIC opt-out request will be ignored
1341 iomem= Disable strict checking of access to MMIO memory
1342 strict regions from userspace.
1359 io7= [HW] IO7 for Marvel based alpha systems
1360 See comment before marvel_specify_io7 in
1361 arch/alpha/kernel/core_marvel.c.
1363 io_delay= [X86] I/O delay method
1365 Standard port 0x80 based delay
1367 Alternate port 0xed based delay (needed on some systems)
1369 Simple two microseconds delay
1374 See Documentation/filesystems/nfs/nfsroot.txt.
1376 ip2= [HW] Set IO/IRQ pairs for up to 4 IntelliPort boards
1377 See comment before ip2_setup() in
1378 drivers/char/ip2/ip2base.c.
1381 When an interrupt is not handled search all handlers
1382 for it. Intended to get systems with badly broken
1386 When an interrupt is not handled search all handlers
1387 for it. Also check all handlers each timer
1388 interrupt. Intended to get systems with badly broken
1392 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1394 isolcpus= [KNL,SMP] Isolate CPUs from the general scheduler.
1396 <cpu number>,...,<cpu number>
1398 <cpu number>-<cpu number>
1399 (must be a positive range in ascending order)
1401 <cpu number>,...,<cpu number>-<cpu number>
1403 This option can be used to specify one or more CPUs
1404 to isolate from the general SMP balancing and scheduling
1405 algorithms. You can move a process onto or off an
1406 "isolated" CPU via the CPU affinity syscalls or cpuset.
1407 <cpu number> begins at 0 and the maximum value is
1408 "number of CPUs in system - 1".
1410 This option is the preferred way to isolate CPUs. The
1411 alternative -- manually setting the CPU mask of all
1412 tasks in the system -- can cause problems and
1413 suboptimal load balancer performance.
1417 ivrs_ioapic [HW,X86_64]
1418 Provide an override to the IOAPIC-ID<->DEVICE-ID
1419 mapping provided in the IVRS ACPI table. For
1420 example, to map IOAPIC-ID decimal 10 to
1421 PCI device 00:14.0 write the parameter as:
1422 ivrs_ioapic[10]=00:14.0
1424 ivrs_hpet [HW,X86_64]
1425 Provide an override to the HPET-ID<->DEVICE-ID
1426 mapping provided in the IVRS ACPI table. For
1427 example, to map HPET-ID decimal 0 to
1428 PCI device 00:14.0 write the parameter as:
1429 ivrs_hpet[0]=00:14.0
1431 js= [HW,JOY] Analog joystick
1432 See Documentation/input/joystick.txt.
1436 kernelcore=nn[KMG] [KNL,X86,IA-64,PPC] This parameter
1437 specifies the amount of memory usable by the kernel
1438 for non-movable allocations. The requested amount is
1439 spread evenly throughout all nodes in the system. The
1440 remaining memory in each node is used for Movable
1441 pages. In the event, a node is too small to have both
1442 kernelcore and Movable pages, kernelcore pages will
1443 take priority and other nodes will have a larger number
1444 of Movable pages. The Movable zone is used for the
1445 allocation of pages that may be reclaimed or moved
1446 by the page migration subsystem. This means that
1447 HugeTLB pages may not be allocated from this zone.
1448 Note that allocations like PTEs-from-HighMem still
1449 use the HighMem zone if it exists, and the Normal
1450 zone if it does not.
1452 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
1453 Format: <Controller#>[,poll interval]
1454 The controller # is the number of the ehci usb debug
1455 port as it is probed via PCI. The poll interval is
1456 optional and is the number seconds in between
1457 each poll cycle to the debug port in case you need
1458 the functionality for interrupting the kernel with
1459 gdb or control-c on the dbgp connection. When
1460 not using this parameter you use sysrq-g to break into
1461 the kernel debugger.
1463 kgdboc= [KGDB,HW] kgdb over consoles.
1464 Requires a tty driver that supports console polling,
1465 or a supported polling keyboard driver (non-usb).
1466 Serial only format: <serial_device>[,baud]
1467 keyboard only format: kbd
1468 keyboard and serial format: kbd,<serial_device>[,baud]
1469 Optional Kernel mode setting:
1470 kms, kbd format: kms,kbd
1471 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
1473 kgdbwait [KGDB] Stop kernel execution and enter the
1474 kernel debugger at the earliest opportunity.
1476 kmac= [MIPS] korina ethernet MAC address.
1477 Configure the RouterBoard 532 series on-chip
1478 Ethernet adapter MAC address.
1480 kmemleak= [KNL] Boot-time kmemleak enable/disable
1481 Valid arguments: on, off
1484 kmemcheck= [X86] Boot-time kmemcheck enable/disable/one-shot mode
1485 Valid arguments: 0, 1, 2
1486 kmemcheck=0 (disabled)
1487 kmemcheck=1 (enabled)
1488 kmemcheck=2 (one-shot mode)
1489 Default: 2 (one-shot mode)
1491 kstack=N [X86] Print N words from the kernel stack
1494 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
1495 Default is 0 (don't ignore, but inject #GP)
1497 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
1501 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
1502 Default is 1 (enabled)
1504 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
1506 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
1508 kvm-intel.ept= [KVM,Intel] Disable extended page tables
1509 (virtualized MMU) support on capable Intel chips.
1510 Default is 1 (enabled)
1512 kvm-intel.emulate_invalid_guest_state=
1513 [KVM,Intel] Enable emulation of invalid guest states
1514 Default is 0 (disabled)
1516 kvm-intel.flexpriority=
1517 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
1518 Default is 1 (enabled)
1521 [KVM,Intel] Enable VMX nesting (nVMX).
1522 Default is 0 (disabled)
1524 kvm-intel.unrestricted_guest=
1525 [KVM,Intel] Disable unrestricted guest feature
1526 (virtualized real and unpaged mode) on capable
1527 Intel chips. Default is 1 (enabled)
1529 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
1530 feature (tagged TLBs) on capable Intel chips.
1531 Default is 1 (enabled)
1537 lapic [X86-32,APIC] Enable the local APIC even if BIOS
1540 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
1541 value for LAPIC timer one-shot implementation. Default
1542 back to the programmable timer unit in the LAPIC.
1544 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
1547 libata.dma= [LIBATA] DMA control
1548 libata.dma=0 Disable all PATA and SATA DMA
1549 libata.dma=1 PATA and SATA Disk DMA only
1550 libata.dma=2 ATAPI (CDROM) DMA only
1551 libata.dma=4 Compact Flash DMA only
1552 Combinations also work, so libata.dma=3 enables DMA
1553 for disks and CDROMs, but not CFs.
1555 libata.ignore_hpa= [LIBATA] Ignore HPA limit
1556 libata.ignore_hpa=0 keep BIOS limits (default)
1557 libata.ignore_hpa=1 ignore limits, using full disk
1559 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
1563 libata.force= [LIBATA] Force configurations. The format is comma
1564 separated list of "[ID:]VAL" where ID is
1565 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
1566 matching port, link or device. Basically, it matches
1567 the ATA ID string printed on console by libata. If
1568 the whole ID part is omitted, the last PORT and DEVICE
1569 values are used. If ID hasn't been specified yet, the
1570 configuration applies to all ports, links and devices.
1572 If only DEVICE is omitted, the parameter applies to
1573 the port and all links and devices behind it. DEVICE
1574 number of 0 either selects the first device or the
1575 first fan-out link behind PMP device. It does not
1576 select the host link. DEVICE number of 15 selects the
1577 host link and device attached to it.
1579 The VAL specifies the configuration to force. As long
1580 as there's no ambiguity shortcut notation is allowed.
1581 For example, both 1.5 and 1.5G would work for 1.5Gbps.
1582 The following configurations can be forced.
1584 * Cable type: 40c, 80c, short40c, unk, ign or sata.
1585 Any ID with matching PORT is used.
1587 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
1589 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
1590 udma[/][16,25,33,44,66,100,133] notation is also
1593 * [no]ncq: Turn on or off NCQ.
1595 * nohrst, nosrst, norst: suppress hard, soft
1598 * rstonce: only attempt one reset during
1599 hot-unplug link recovery
1601 * dump_id: dump IDENTIFY data.
1603 * atapi_dmadir: Enable ATAPI DMADIR bridge support
1605 * disable: Disable this device.
1607 If there are multiple matching configurations changing
1608 the same attribute, the last one is used.
1610 memblock=debug [KNL] Enable memblock debug messages.
1612 load_ramdisk= [RAM] List of ramdisks to load from floppy
1613 See Documentation/blockdev/ramdisk.txt.
1615 lockd.nlm_grace_period=P [NFS] Assign grace period.
1618 lockd.nlm_tcpport=N [NFS] Assign TCP port.
1621 lockd.nlm_timeout=T [NFS] Assign timeout value.
1624 lockd.nlm_udpport=M [NFS] Assign UDP port.
1627 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
1630 loglevel= All Kernel Messages with a loglevel smaller than the
1631 console loglevel will be printed to the console. It can
1632 also be changed with klogd or other programs. The
1633 loglevels are defined as follows:
1635 0 (KERN_EMERG) system is unusable
1636 1 (KERN_ALERT) action must be taken immediately
1637 2 (KERN_CRIT) critical conditions
1638 3 (KERN_ERR) error conditions
1639 4 (KERN_WARNING) warning conditions
1640 5 (KERN_NOTICE) normal but significant condition
1641 6 (KERN_INFO) informational
1642 7 (KERN_DEBUG) debug-level messages
1644 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
1645 in bytes. n must be a power of two. The default
1646 size is set in the kernel config file.
1648 logo.nologo [FB] Disables display of the built-in Linux logo.
1649 This may be used to provide more screen space for
1650 kernel log messages and is useful when debugging
1651 kernel boot problems.
1653 lp=0 [LP] Specify parallel ports to use, e.g,
1654 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
1655 lp=reset first parallel port). 'lp=0' disables the
1656 lp=auto printer driver. 'lp=reset' (which can be
1657 specified in addition to the ports) causes
1658 attached printers to be reset. Using
1659 lp=port1,port2,... specifies the parallel ports
1660 to associate lp devices with, starting with
1661 lp0. A port specification may be 'none' to skip
1662 that lp device, or a parport name such as
1663 'parport0'. Specifying 'lp=auto' instead of a
1664 port specification list means that device IDs
1665 from each port should be examined, to see if
1666 an IEEE 1284-compliant printer is attached; if
1667 so, the driver will manage that printer.
1668 See also header of drivers/char/lp.c.
1671 Sets loops_per_jiffy to given constant, thus avoiding
1672 time-consuming boot-time autodetection (up to 250 ms per
1673 CPU). 0 enables autodetection (default). To determine
1674 the correct value for your kernel, boot with normal
1675 autodetection and see what value is printed. Note that
1676 on SMP systems the preset will be applied to all CPUs,
1677 which is likely to cause problems if your CPUs need
1678 significantly divergent settings. An incorrect value
1679 will cause delays in the kernel to be wrong, leading to
1680 unpredictable I/O errors and other breakage. Although
1681 unlikely, in the extreme case this might damage your
1685 Format: <io>,<irq>,<dma>
1687 machvec= [IA-64] Force the use of a particular machine-vector
1688 (machvec) in a generic kernel.
1689 Example: machvec=hpzx1_swiotlb
1691 machtype= [Loongson] Share the same kernel image file between different
1693 Example: machtype=lemote-yeeloong-2f-7inch
1695 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
1696 than or equal to this physical address is ignored.
1698 maxcpus= [SMP] Maximum number of processors that an SMP kernel
1699 should make use of. maxcpus=n : n >= 0 limits the
1700 kernel to using 'n' processors. n=0 is a special case,
1701 it is equivalent to "nosmp", which also disables
1704 max_loop= [LOOP] The number of loop block devices that get
1705 (loop.max_loop) unconditionally pre-created at init time. The default
1706 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
1707 of statically allocating a predefined number, loop
1708 devices can be requested on-demand with the
1709 /dev/loop-control interface.
1711 mce [X86-32] Machine Check Exception
1713 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.txt
1715 md= [HW] RAID subsystems devices and level
1716 See Documentation/md.txt.
1719 Format: <first>,<last>
1720 Specifies range of consoles to be captured by the MDA.
1722 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
1723 Amount of memory to be used when the kernel is not able
1724 to see the whole system memory or for test.
1725 [X86] Work as limiting max address. Use together
1726 with memmap= to avoid physical address space collisions.
1727 Without memmap= PCI devices could be placed at addresses
1728 belonging to unused RAM.
1730 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
1734 [KNL,SH] Allow user to override the default size for
1735 per-device physically contiguous DMA buffers.
1737 memmap=exactmap [KNL,X86] Enable setting of an exact
1738 E820 memory map, as specified by the user.
1739 Such memmap=exactmap lines can be constructed based on
1740 BIOS output or other requirements. See the memmap=nn@ss
1743 memmap=nn[KMG]@ss[KMG]
1744 [KNL] Force usage of a specific region of memory.
1745 Region of memory to be used is from ss to ss+nn.
1747 memmap=nn[KMG]#ss[KMG]
1748 [KNL,ACPI] Mark specific memory as ACPI data.
1749 Region of memory to be marked is from ss to ss+nn.
1751 memmap=nn[KMG]$ss[KMG]
1752 [KNL,ACPI] Mark specific memory as reserved.
1753 Region of memory to be reserved is from ss to ss+nn.
1754 Example: Exclude memory from 0x18690000-0x1869ffff
1755 memmap=64K$0x18690000
1757 memmap=0x10000$0x18690000
1759 memory_corruption_check=0/1 [X86]
1760 Some BIOSes seem to corrupt the first 64k of
1761 memory when doing things like suspend/resume.
1762 Setting this option will scan the memory
1763 looking for corruption. Enabling this will
1764 both detect corruption and prevent the kernel
1765 from using the memory being corrupted.
1766 However, its intended as a diagnostic tool; if
1767 repeatable BIOS-originated corruption always
1768 affects the same memory, you can use memmap=
1769 to prevent the kernel from using that memory.
1771 memory_corruption_check_size=size [X86]
1772 By default it checks for corruption in the low
1773 64k, making this memory unavailable for normal
1774 use. Use this parameter to scan for
1775 corruption in more or less memory.
1777 memory_corruption_check_period=seconds [X86]
1778 By default it checks for corruption every 60
1779 seconds. Use this parameter to check at some
1780 other rate. 0 disables periodic checking.
1782 memtest= [KNL,X86] Enable memtest
1784 default : 0 <disable>
1785 Specifies the number of memtest passes to be
1786 performed. Each pass selects another test
1787 pattern from a given set of patterns. Memtest
1788 fills the memory with this pattern, validates
1789 memory contents and reserves bad memory
1790 regions that are detected.
1792 meye.*= [HW] Set MotionEye Camera parameters
1793 See Documentation/video4linux/meye.txt.
1795 mfgpt_irq= [IA-32] Specify the IRQ to use for the
1796 Multi-Function General Purpose Timers on AMD Geode
1799 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
1800 the BIOS has incorrectly applied a workaround. TinyBIOS
1801 version 0.98 is known to be affected, 0.99 fixes the
1802 problem by letting the user disable the workaround.
1806 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
1807 physical address is ignored.
1809 mini2440= [ARM,HW,KNL]
1810 Format:[0..2][b][c][t]
1812 MINI2440 configuration specification:
1813 0 - The attached screen is the 3.5" TFT
1814 1 - The attached screen is the 7" TFT
1815 2 - The VGA Shield is attached (1024x768)
1816 Leaving out the screen size parameter will not load
1817 the TFT driver, and the framebuffer will be left
1819 b - Enable backlight. The TFT backlight pin will be
1820 linked to the kernel VESA blanking code and a GPIO
1821 LED. This parameter is not necessary when using the
1823 c - Enable the s3c camera interface.
1824 t - Reserved for enabling touchscreen support. The
1825 touchscreen support is not enabled in the mainstream
1826 kernel as of 2.6.30, a preliminary port can be found
1827 in the "bleeding edge" mini2440 support kernel at
1828 http://repo.or.cz/w/linux-2.6/mini2440.git
1831 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
1832 parameter allows control of the logging verbosity for
1833 the additional memory initialisation checks. A value
1834 of 0 disables mminit logging and a level of 4 will
1835 log everything. Information is printed at KERN_DEBUG
1836 so loglevel=8 may also need to be specified.
1839 [KNL] When CONFIG_MODULE_SIG is set, this means that
1840 modules without (valid) signatures will fail to load.
1841 Note that if CONFIG_MODULE_SIG_FORCE is set, that
1842 is always true, so this option does nothing.
1845 [MOUSE] Maximum time between finger touching and
1846 leaving touchpad surface for touch to be considered
1847 a tap and be reported as a left button click (for
1848 touchpads working in absolute mode only).
1850 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
1851 reporting absolute coordinates, such as tablets
1852 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
1853 reporting absolute coordinates, such as tablets
1855 movablecore=nn[KMG] [KNL,X86,IA-64,PPC] This parameter
1856 is similar to kernelcore except it specifies the
1857 amount of memory used for migratable allocations.
1858 If both kernelcore and movablecore is specified,
1859 then kernelcore will be at *least* the specified
1860 value but may be more. If movablecore on its own
1861 is specified, the administrator must be careful
1862 that the amount of memory usable for all allocations
1865 movable_node [KNL,X86] Boot-time switch to enable the effects
1866 of CONFIG_MOVABLE_NODE=y. See mm/Kconfig for details.
1868 MTD_Partition= [MTD]
1869 Format: <name>,<region-number>,<size>,<offset>
1871 MTD_Region= [MTD] Format:
1872 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
1875 See drivers/mtd/cmdlinepart.c.
1877 multitce=off [PPC] This parameter disables the use of the pSeries
1878 firmware feature for updating multiple TCE entries
1881 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
1883 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
1885 boundary - index of last SLC block on Flex-OneNAND.
1886 The remaining blocks are configured as MLC blocks.
1887 lock - Configure if Flex-OneNAND boundary should be locked.
1888 Once locked, the boundary cannot be changed.
1889 1 indicates lock status, 0 indicates unlock status.
1892 ARM/S3C2412 JIVE boot control
1894 See arch/arm/mach-s3c2412/mach-jive.c
1896 mtouchusb.raw_coordinates=
1897 [HW] Make the MicroTouch USB driver use raw coordinates
1898 ('y', default) or cooked coordinates ('n')
1900 mtrr_chunk_size=nn[KMG] [X86]
1901 used for mtrr cleanup. It is largest continuous chunk
1902 that could hold holes aka. UC entries.
1904 mtrr_gran_size=nn[KMG] [X86]
1905 Used for mtrr cleanup. It is granularity of mtrr block.
1907 Large value could prevent small alignment from
1910 mtrr_spare_reg_nr=n [X86]
1912 Range: 0,7 : spare reg number
1914 Used for mtrr cleanup. It is spare mtrr entries number.
1915 Set to 2 or more if your graphical card needs more.
1917 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
1919 netdev= [NET] Network devices parameters
1920 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
1921 Note that mem_start is often overloaded to mean
1922 something different and driver-specific.
1923 This usage is only documented in each driver source
1927 [NETFILTER] Enable connection tracking flow accounting
1928 0 to disable accounting
1929 1 to enable accounting
1932 nfsaddrs= [NFS] Deprecated. Use ip= instead.
1933 See Documentation/filesystems/nfs/nfsroot.txt.
1935 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
1936 See Documentation/filesystems/nfs/nfsroot.txt.
1938 nfsrootdebug [NFS] enable nfsroot debugging messages.
1939 See Documentation/filesystems/nfs/nfsroot.txt.
1941 nfs.callback_tcpport=
1942 [NFS] set the TCP port on which the NFSv4 callback
1943 channel should listen.
1946 [NFS] sets the pathname to the program which is used
1947 to update the NFS client cache entries.
1949 nfs.cache_getent_timeout=
1950 [NFS] sets the timeout after which an attempt to
1951 update a cache entry is deemed to have failed.
1953 nfs.idmap_cache_timeout=
1954 [NFS] set the maximum lifetime for idmapper cache
1958 [NFS] enable 64-bit inode numbers.
1959 If zero, the NFS client will fake up a 32-bit inode
1960 number for the readdir() and stat() syscalls instead
1961 of returning the full 64-bit number.
1962 The default is to return 64-bit inode numbers.
1964 nfs.max_session_slots=
1965 [NFSv4.1] Sets the maximum number of session slots
1966 the client will attempt to negotiate with the server.
1967 This limits the number of simultaneous RPC requests
1968 that the client can send to the NFSv4.1 server.
1969 Note that there is little point in setting this
1970 value higher than the max_tcp_slot_table_limit.
1972 nfs.nfs4_disable_idmapping=
1973 [NFSv4] When set to the default of '1', this option
1974 ensures that both the RPC level authentication
1975 scheme and the NFS level operations agree to use
1976 numeric uids/gids if the mount is using the
1977 'sec=sys' security flavour. In effect it is
1978 disabling idmapping, which can make migration from
1979 legacy NFSv2/v3 systems to NFSv4 easier.
1980 Servers that do not support this mode of operation
1981 will be autodetected by the client, and it will fall
1982 back to using the idmapper.
1983 To turn off this behaviour, set the value to '0'.
1985 [NFS4] Specify an additional fixed unique ident-
1986 ification string that NFSv4 clients can insert into
1987 their nfs_client_id4 string. This is typically a
1988 UUID that is generated at system install time.
1990 nfs.send_implementation_id =
1991 [NFSv4.1] Send client implementation identification
1992 information in exchange_id requests.
1993 If zero, no implementation identification information
1995 The default is to send the implementation identification
1998 nfs.recover_lost_locks =
1999 [NFSv4] Attempt to recover locks that were lost due
2000 to a lease timeout on the server. Please note that
2001 doing this risks data corruption, since there are
2002 no guarantees that the file will remain unchanged
2003 after the locks are lost.
2004 If you want to enable the kernel legacy behaviour of
2005 attempting to recover these locks, then set this
2007 The default parameter value of '0' causes the kernel
2008 not to attempt recovery of lost locks.
2010 nfsd.nfs4_disable_idmapping=
2011 [NFSv4] When set to the default of '1', the NFSv4
2012 server will return only numeric uids and gids to
2013 clients using auth_sys, and will accept numeric uids
2014 and gids from such clients. This is intended to ease
2015 migration from NFSv2/v3.
2017 objlayoutdriver.osd_login_prog=
2018 [NFS] [OBJLAYOUT] sets the pathname to the program which
2019 is used to automatically discover and login into new
2020 osd-targets. Please see:
2021 Documentation/filesystems/pnfs.txt for more explanations
2023 nmi_debug= [KNL,AVR32,SH] Specify one or more actions to take
2024 when a NMI is triggered.
2025 Format: [state][,regs][,debounce][,die]
2027 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
2028 Format: [panic,][nopanic,][num]
2030 0 - turn nmi_watchdog off
2031 When panic is specified, panic when an NMI watchdog
2032 timeout occurs (or 'nopanic' to override the opposite
2034 This is useful when you use a panic=... timeout and
2035 need the box quickly up again.
2037 netpoll.carrier_timeout=
2038 [NET] Specifies amount of time (in seconds) that
2039 netpoll should wait for a carrier. By default netpoll
2042 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
2043 emulation library even if a 387 maths coprocessor
2047 [HW] Never suspend the console
2048 Disable suspending of consoles during suspend and
2049 hibernate operations. Once disabled, debugging
2050 messages can reach various consoles while the rest
2051 of the system is being put to sleep (ie, while
2052 debugging driver suspend/resume hooks). This may
2053 not work reliably with all consoles, but is known
2054 to work with serial and VGA consoles.
2055 To facilitate more flexible debugging, we also add
2056 console_suspend, a printk module parameter to control
2057 it. Users could use console_suspend (usually
2058 /sys/module/printk/parameters/console_suspend) to
2059 turn on/off it dynamically.
2061 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
2062 caches in the slab allocator. Saves per-node memory,
2063 but will impact performance.
2067 noapic [SMP,APIC] Tells the kernel to not make use of any
2068 IOAPICs that may be present in the system.
2071 Disable kernel and module base offset ASLR (Address
2072 Space Layout Randomization) if built into the kernel.
2074 noautogroup Disable scheduler automatic task group creation.
2076 nobats [PPC] Do not use BATs for mapping kernel lowmem
2077 on "Classic" PPC cores.
2081 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
2083 nodelayacct [KNL] Disable per-task delay accounting
2085 nodisconnect [HW,SCSI,M68K] Disables SCSI disconnects.
2087 nodsp [SH] Disable hardware DSP at boot time.
2089 noefi [X86] Disable EFI runtime services support.
2094 On X86-32 available only on PAE configured kernels.
2095 noexec=on: enable non-executable mappings (default)
2096 noexec=off: disable non-executable mappings
2099 Disable SMAP (Supervisor Mode Access Prevention)
2100 even if it is supported by processor.
2103 Disable SMEP (Supervisor Mode Execution Prevention)
2104 even if it is supported by processor.
2107 This affects only 32-bit executables.
2108 noexec32=on: enable non-executable mappings (default)
2109 read doesn't imply executable mappings
2110 noexec32=off: disable non-executable mappings
2111 read implies executable mappings
2113 nofpu [SH] Disable hardware FPU at boot time.
2115 nofxsr [BUGS=X86-32] Disables x86 floating point extended
2116 register save and restore. The kernel will only save
2117 legacy floating-point registers on task switch.
2119 noxsave [BUGS=X86] Disables x86 extended register state save
2120 and restore using xsave. The kernel will fallback to
2121 enabling legacy floating-point and sse state.
2124 on enable eager fpu restore
2125 off disable eager fpu restore
2126 auto selects the default scheme, which automatically
2127 enables eagerfpu restore for xsaveopt.
2129 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
2130 wfi(ARM) instruction doesn't work correctly and not to
2131 use it. This is also useful when using JTAG debugger.
2133 no_file_caps Tells the kernel not to honor file capabilities. The
2134 only way then for a file to be executed with privilege
2135 is to be setuid root or executed by root.
2137 nohalt [IA-64] Tells the kernel not to use the power saving
2138 function PAL_HALT_LIGHT when idle. This increases
2139 power-consumption. On the positive side, it reduces
2140 interrupt wake-up latency, which may improve performance
2141 in certain environments such as networked servers or
2144 nohz= [KNL] Boottime enable/disable dynamic ticks
2145 Valid arguments: on, off
2148 nohz_full= [KNL,BOOT]
2149 In kernels built with CONFIG_NO_HZ_FULL=y, set
2150 the specified list of CPUs whose tick will be stopped
2151 whenever possible. The boot CPU will be forced outside
2152 the range to maintain the timekeeping.
2153 The CPUs in this range must also be included in the
2156 noiotrap [SH] Disables trapped I/O port accesses.
2158 noirqdebug [X86-32] Disables the code which attempts to detect and
2159 disable unhandled interrupt sources.
2161 no_timer_check [X86,APIC] Disables the code which tests for
2162 broken timer IRQ sources.
2164 noisapnp [ISAPNP] Disables ISA PnP code.
2166 noinitrd [RAM] Tells the kernel not to load any configured
2169 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
2171 [Deprecated - use intremap=off]
2175 nojitter [IA-64] Disables jitter checking for ITC timers.
2177 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
2179 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
2182 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting.
2183 steal time is computed, but won't influence scheduler
2186 nolapic [X86-32,APIC] Do not enable or use the local APIC.
2188 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
2190 noltlbs [PPC] Do not use large page/tlb entries for kernel
2191 lowmem mapping on PPC40x.
2193 nomca [IA-64] Disable machine check abort handling
2195 nomce [X86-32] Machine Check Exception
2197 nomfgpt [X86-32] Disable Multi-Function General Purpose
2198 Timer usage (for AMD Geode machines).
2200 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
2201 shutdown the other cpus. Instead use the REBOOT_VECTOR
2204 nomodule Disable module load
2206 nopat [X86] Disable PAT (page attribute table extension of
2207 pagetables) support.
2209 norandmaps Don't use address space randomization. Equivalent to
2210 echo 0 > /proc/sys/kernel/randomize_va_space
2212 noreplace-paravirt [X86,IA-64,PV_OPS] Don't patch paravirt_ops
2214 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
2215 with UP alternatives
2217 nordrand [X86] Disable the direct use of the RDRAND
2218 instruction even if it is supported by the
2219 processor. RDRAND is still available to user
2222 noresume [SWSUSP] Disables resume and restores original swap
2225 no-scroll [VGA] Disables scrollback.
2226 This is required for the Braillex ib80-piezo Braille
2227 reader made by F.H. Papenmeier (Germany).
2231 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
2233 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
2234 and disable the IO APIC. legacy for "maxcpus=0".
2236 nosoftlockup [KNL] Disable the soft-lockup detector.
2238 nosync [HW,M68K] Disables sync negotiation for all devices.
2240 notsc [BUGS=X86-32] Disable Time Stamp Counter
2242 nousb [USB] Disable the USB subsystem
2244 nowatchdog [KNL] Disable the lockup detector (NMI watchdog).
2248 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
2250 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
2251 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
2252 Some features depend on CPU0. Known dependencies are:
2253 1. Resume from suspend/hibernate depends on CPU0.
2254 Suspend/hibernate will fail if CPU0 is offline and you
2255 need to online CPU0 before suspend/hibernate.
2256 2. PIC interrupts also depend on CPU0. CPU0 can't be
2257 removed if a PIC interrupt is detected.
2258 It's said poweroff/reboot may depend on CPU0 on some
2259 machines although I haven't seen such issues so far
2260 after CPU0 is offline on a few tested machines.
2261 If the dependencies are under your control, you can
2262 turn on cpu0_hotplug.
2264 nptcg= [IA-64] Override max number of concurrent global TLB
2265 purges which is reported from either PAL_VM_SUMMARY or
2268 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
2269 could support. nr_cpus=n : n >= 1 limits the kernel to
2270 supporting 'n' processors. Later in runtime you can not
2271 use hotplug cpu feature to put more cpu back to online.
2272 just like you compile the kernel NR_CPUS=n
2274 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
2276 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
2277 Allowed values are enable and disable
2279 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
2280 one of ['zone', 'node', 'default'] can be specified
2281 This can be set from sysctl after boot.
2282 See Documentation/sysctl/vm.txt for details.
2284 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
2285 See Documentation/debugging-via-ohci1394.txt for more
2288 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
2289 Rather than timing out after 20 ms if an EC
2290 command is not properly ACKed, override the length
2291 of the timeout. We have interrupts disabled while
2292 waiting for the ACK, so if this is set too high
2293 interrupts *may* be lost!
2295 omap_mux= [OMAP] Override bootloader pin multiplexing.
2296 Format: <mux_mode0.mode_name=value>...
2297 For example, to override I2C bus2:
2298 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
2300 oprofile.timer= [HW]
2301 Use timer interrupt instead of performance counters
2303 oprofile.cpu_type= Force an oprofile cpu type
2304 This might be useful if you have an older oprofile
2305 userland or if you want common events.
2306 Format: { arch_perfmon }
2307 arch_perfmon: [X86] Force use of architectural
2308 perfmon on Intel CPUs instead of the
2309 CPU specific event set.
2310 timer: [X86] Force use of architectural NMI
2311 timer mode (see also oprofile.timer
2312 for generic hr timer mode)
2313 [s390] Force legacy basic mode sampling
2314 (report cpu_type "timer")
2316 oops=panic Always panic on oopses. Default is to just kill the
2317 process, but there is a small probability of
2318 deadlocking the machine.
2319 This will also cause panics on machine check exceptions.
2320 Useful together with panic=30 to trigger a reboot.
2323 See Documentation/sound/oss/oss-parameters.txt
2325 panic= [KNL] Kernel behaviour on panic: delay <timeout>
2326 timeout > 0: seconds before rebooting
2327 timeout = 0: wait forever
2328 timeout < 0: reboot immediately
2331 parkbd.port= [HW] Parallel port number the keyboard adapter is
2332 connected to, default is 0.
2334 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
2335 0 for XT, 1 for AT (default is AT).
2338 parport= [HW,PPT] Specify parallel ports. 0 disables.
2339 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
2340 Use 'auto' to force the driver to use any
2341 IRQ/DMA settings detected (the default is to
2342 ignore detected IRQ/DMA settings because of
2343 possible conflicts). You can specify the base
2344 address, IRQ, and DMA settings; IRQ and DMA
2345 should be numbers, or 'auto' (for using detected
2346 settings on that particular port), or 'nofifo'
2347 (to avoid using a FIFO even if it is detected).
2348 Parallel ports are assigned in the order they
2349 are specified on the command line, starting
2352 parport_init_mode= [HW,PPT]
2353 Configure VIA parallel port to operate in
2354 a specific mode. This is necessary on Pegasos
2355 computer where firmware has no options for setting
2356 up parallel port mode and sets it to spp.
2357 Currently this function knows 686a and 8231 chips.
2358 Format: [spp|ps2|epp|ecp|ecpepp]
2361 Halt all CPUs after the first oops has been printed for
2362 the specified number of seconds. This is to be used if
2363 your oopses keep scrolling off the screen.
2368 See header of drivers/block/paride/pcd.c.
2369 See also Documentation/blockdev/paride.txt.
2371 pci=option[,option...] [PCI] various PCI subsystem options:
2372 earlydump [X86] dump PCI config space before the kernel
2374 off [X86] don't probe for the PCI bus
2375 bios [X86-32] force use of PCI BIOS, don't access
2376 the hardware directly. Use this if your machine
2377 has a non-standard PCI host bridge.
2378 nobios [X86-32] disallow use of PCI BIOS, only direct
2379 hardware access methods are allowed. Use this
2380 if you experience crashes upon bootup and you
2381 suspect they are caused by the BIOS.
2382 conf1 [X86] Force use of PCI Configuration
2384 conf2 [X86] Force use of PCI Configuration
2386 noaer [PCIE] If the PCIEAER kernel config parameter is
2387 enabled, this kernel boot option can be used to
2388 disable the use of PCIE advanced error reporting.
2389 nodomains [PCI] Disable support for multiple PCI
2390 root domains (aka PCI segments, in ACPI-speak).
2391 nommconf [X86] Disable use of MMCONFIG for PCI
2393 check_enable_amd_mmconf [X86] check for and enable
2394 properly configured MMIO access to PCI
2395 config space on AMD family 10h CPU
2396 nomsi [MSI] If the PCI_MSI kernel config parameter is
2397 enabled, this kernel boot option can be used to
2398 disable the use of MSI interrupts system-wide.
2399 noioapicquirk [APIC] Disable all boot interrupt quirks.
2400 Safety option to keep boot IRQs enabled. This
2401 should never be necessary.
2402 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
2403 primary IO-APIC for bridges that cannot disable
2404 boot IRQs. This fixes a source of spurious IRQs
2405 when the system masks IRQs.
2406 noioapicreroute [APIC] Disable workaround that uses the
2407 boot IRQ equivalent of an IRQ that connects to
2408 a chipset where boot IRQs cannot be disabled.
2409 The opposite of ioapicreroute.
2410 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
2411 routing table. These calls are known to be buggy
2412 on several machines and they hang the machine
2413 when used, but on other computers it's the only
2414 way to get the interrupt routing table. Try
2415 this option if the kernel is unable to allocate
2416 IRQs or discover secondary PCI buses on your
2418 rom [X86] Assign address space to expansion ROMs.
2419 Use with caution as certain devices share
2420 address decoders between ROMs and other
2422 norom [X86] Do not assign address space to
2423 expansion ROMs that do not already have
2424 BIOS assigned address ranges.
2425 nobar [X86] Do not assign address space to the
2426 BARs that weren't assigned by the BIOS.
2427 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
2428 assigned automatically to PCI devices. You can
2429 make the kernel exclude IRQs of your ISA cards
2431 pirqaddr=0xAAAAA [X86] Specify the physical address
2432 of the PIRQ table (normally generated
2433 by the BIOS) if it is outside the
2434 F0000h-100000h range.
2435 lastbus=N [X86] Scan all buses thru bus #N. Can be
2436 useful if the kernel is unable to find your
2437 secondary buses and you want to tell it
2438 explicitly which ones they are.
2439 assign-busses [X86] Always assign all PCI bus
2440 numbers ourselves, overriding
2441 whatever the firmware may have done.
2442 usepirqmask [X86] Honor the possible IRQ mask stored
2443 in the BIOS $PIR table. This is needed on
2444 some systems with broken BIOSes, notably
2445 some HP Pavilion N5400 and Omnibook XE3
2446 notebooks. This will have no effect if ACPI
2447 IRQ routing is enabled.
2448 noacpi [X86] Do not use ACPI for IRQ routing
2449 or for PCI scanning.
2450 use_crs [X86] Use PCI host bridge window information
2451 from ACPI. On BIOSes from 2008 or later, this
2452 is enabled by default. If you need to use this,
2453 please report a bug.
2454 nocrs [X86] Ignore PCI host bridge windows from ACPI.
2455 If you need to use this, please report a bug.
2456 routeirq Do IRQ routing for all PCI devices.
2457 This is normally done in pci_enable_device(),
2458 so this option is a temporary workaround
2459 for broken drivers that don't call it.
2460 skip_isa_align [X86] do not align io start addr, so can
2461 handle more pci cards
2462 firmware [ARM] Do not re-enumerate the bus but instead
2463 just use the configuration from the
2464 bootloader. This is currently used on
2465 IXP2000 systems where the bus has to be
2466 configured a certain way for adjunct CPUs.
2467 noearly [X86] Don't do any early type 1 scanning.
2468 This might help on some broken boards which
2469 machine check when some devices' config space
2470 is read. But various workarounds are disabled
2471 and some IOMMU drivers will not work.
2472 bfsort Sort PCI devices into breadth-first order.
2473 This sorting is done to get a device
2474 order compatible with older (<= 2.4) kernels.
2475 nobfsort Don't sort PCI devices into breadth-first order.
2476 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
2477 tuning and use the BIOS-configured MPS defaults.
2478 pcie_bus_safe Set every device's MPS to the largest value
2479 supported by all devices below the root complex.
2480 pcie_bus_perf Set device MPS to the largest allowable MPS
2481 based on its parent bus. Also set MRRS (Max
2482 Read Request Size) to the largest supported
2483 value (no larger than the MPS that the device
2484 or bus can support) for best performance.
2485 pcie_bus_peer2peer Set every device's MPS to 128B, which
2486 every device is guaranteed to support. This
2487 configuration allows peer-to-peer DMA between
2488 any pair of devices, possibly at the cost of
2489 reduced performance. This also guarantees
2490 that hot-added devices will work.
2491 cbiosize=nn[KMG] The fixed amount of bus space which is
2492 reserved for the CardBus bridge's IO window.
2493 The default value is 256 bytes.
2494 cbmemsize=nn[KMG] The fixed amount of bus space which is
2495 reserved for the CardBus bridge's memory
2496 window. The default value is 64 megabytes.
2499 [<order of align>@][<domain>:]<bus>:<slot>.<func>[; ...]
2500 Specifies alignment and device to reassign
2501 aligned memory resources.
2502 If <order of align> is not specified,
2503 PAGE_SIZE is used as alignment.
2504 PCI-PCI bridge can be specified, if resource
2505 windows need to be expanded.
2506 ecrc= Enable/disable PCIe ECRC (transaction layer
2507 end-to-end CRC checking).
2508 bios: Use BIOS/firmware settings. This is the
2512 hpiosize=nn[KMG] The fixed amount of bus space which is
2513 reserved for hotplug bridge's IO window.
2514 Default size is 256 bytes.
2515 hpmemsize=nn[KMG] The fixed amount of bus space which is
2516 reserved for hotplug bridge's memory window.
2517 Default size is 2 megabytes.
2518 realloc= Enable/disable reallocating PCI bridge resources
2519 if allocations done by BIOS are too small to
2520 accommodate resources required by all child
2522 off: Turn realloc off
2524 realloc same as realloc=on
2525 noari do not use PCIe ARI.
2526 pcie_scan_all Scan all possible PCIe devices. Otherwise we
2527 only look for one device below a PCIe downstream
2530 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
2533 force Enable ASPM even on devices that claim not to support it.
2534 WARNING: Forcing ASPM on may cause system lockups.
2536 pcie_hp= [PCIE] PCI Express Hotplug driver options:
2537 nomsi Do not use MSI for PCI Express Native Hotplug (this
2538 makes all PCIe ports use INTx for hotplug services).
2540 pcie_ports= [PCIE] PCIe ports handling:
2541 auto Ask the BIOS whether or not to use native PCIe services
2542 associated with PCIe ports (PME, hot-plug, AER). Use
2543 them only if that is allowed by the BIOS.
2544 native Use native PCIe services associated with PCIe ports
2546 compat Treat PCIe ports as PCI-to-PCI bridges, disable the PCIe
2549 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
2550 nomsi Do not use MSI for native PCIe PME signaling (this makes
2551 all PCIe root ports use INTx for all services).
2553 pcmv= [HW,PCMCIA] BadgePAD 4
2556 See Documentation/blockdev/paride.txt.
2558 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
2561 See arch/parisc/kernel/pdc_chassis.c
2563 percpu_alloc= Select which percpu first chunk allocator to use.
2564 Currently supported values are "embed" and "page".
2565 Archs may support subset or none of the selections.
2566 See comments in mm/percpu.c for details on each
2567 allocator. This parameter is primarily for debugging
2568 and performance comparison.
2571 See Documentation/blockdev/paride.txt.
2574 See Documentation/blockdev/paride.txt.
2576 pirq= [SMP,APIC] Manual mp-table setup
2577 See Documentation/x86/i386/IO-APIC.txt.
2579 plip= [PPT,NET] Parallel port network link
2580 Format: { parport<nr> | timid | 0 }
2581 See also Documentation/parport.txt.
2583 pmtmr= [X86] Manual setup of pmtmr I/O Port.
2584 Override pmtimer IOPort with a hex value.
2588 Enable PNP debug messages (depends on the
2589 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
2590 via /sys/module/pnp/parameters/debug. We always show
2591 current resource usage; turning this on also shows
2592 possible settings and some assignment information.
2598 { on | off | curr | res | no-curr | no-res }
2601 [ISAPNP] Exclude IRQs for the autoconfiguration
2604 [ISAPNP] Exclude DMAs for the autoconfiguration
2606 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
2607 Ranges are in pairs (I/O port base and size).
2610 [ISAPNP] Exclude memory regions for the
2612 Ranges are in pairs (memory base and size).
2614 ports= [IP_VS_FTP] IPVS ftp helper module
2616 Up to 8 (IP_VS_APP_MAX_PORTS) ports
2618 Format: <port>,<port>....
2620 print-fatal-signals=
2621 [KNL] debug: print fatal signals
2623 If enabled, warn about various signal handling
2624 related application anomalies: too many signals,
2625 too many POSIX.1 timers, fatal signals causing a
2628 If you hit the warning due to signal overflow,
2629 you might want to try "ulimit -i unlimited".
2633 printk.always_kmsg_dump=
2634 Trigger kmsg_dump for cases other than kernel oops or
2636 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
2639 printk.time= Show timing data prefixed to each printk message line
2640 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
2642 processor.max_cstate= [HW,ACPI]
2643 Limit processor to maximum C-state
2644 max_cstate=9 overrides any DMI blacklist limit.
2646 processor.nocst [HW,ACPI]
2647 Ignore the _CST method to determine C-states,
2648 instead using the legacy FADT method
2650 profile= [KNL] Enable kernel profiling via /proc/profile
2651 Format: [schedule,]<number>
2652 Param: "schedule" - profile schedule points.
2653 Param: <number> - step/bucket size as a power of 2 for
2654 statistical time based profiling.
2655 Param: "sleep" - profile D-state sleeping (millisecs).
2656 Requires CONFIG_SCHEDSTATS
2657 Param: "kvm" - profile VM exits.
2659 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
2661 See Documentation/blockdev/ramdisk.txt.
2663 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
2664 probe for; one of (bare|imps|exps|lifebook|any).
2665 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
2667 psmouse.resetafter= [HW,MOUSE]
2668 Try to reset the device after so many bad packets
2671 [HW,MOUSE] Set desired mouse resolution, in dpi.
2672 psmouse.smartscroll=
2673 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
2674 0 = disabled, 1 = enabled (default).
2676 pstore.backend= Specify the name of the pstore backend to use
2679 See Documentation/blockdev/paride.txt.
2682 [KNL] Number of legacy pty's. Overwrites compiled-in
2685 quiet [KNL] Disable most log messages
2690 See Documentation/md.txt.
2692 ramdisk_blocksize= [RAM]
2693 See Documentation/blockdev/ramdisk.txt.
2695 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
2696 See Documentation/blockdev/ramdisk.txt.
2699 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
2700 the specified list of CPUs to be no-callback CPUs.
2701 Invocation of these CPUs' RCU callbacks will
2702 be offloaded to "rcuox/N" kthreads created for
2703 that purpose, where "x" is "b" for RCU-bh, "p"
2704 for RCU-preempt, and "s" for RCU-sched, and "N"
2705 is the CPU number. This reduces OS jitter on the
2706 offloaded CPUs, which can be useful for HPC and
2707 real-time workloads. It can also improve energy
2708 efficiency for asymmetric multiprocessors.
2711 Rather than requiring that offloaded CPUs
2712 (specified by rcu_nocbs= above) explicitly
2713 awaken the corresponding "rcuoN" kthreads,
2714 make these kthreads poll for callbacks.
2715 This improves the real-time response for the
2716 offloaded CPUs by relieving them of the need to
2717 wake up the corresponding kthread, but degrades
2718 energy efficiency by requiring that the kthreads
2719 periodically wake up to do the polling.
2721 rcutree.blimit= [KNL]
2722 Set maximum number of finished RCU callbacks to
2723 process in one batch.
2725 rcutree.rcu_fanout_leaf= [KNL]
2726 Increase the number of CPUs assigned to each
2727 leaf rcu_node structure. Useful for very large
2730 rcutree.jiffies_till_first_fqs= [KNL]
2731 Set delay from grace-period initialization to
2732 first attempt to force quiescent states.
2733 Units are jiffies, minimum value is zero,
2734 and maximum value is HZ.
2736 rcutree.jiffies_till_next_fqs= [KNL]
2737 Set delay between subsequent attempts to force
2738 quiescent states. Units are jiffies, minimum
2739 value is one, and maximum value is HZ.
2741 rcutree.qhimark= [KNL]
2742 Set threshold of queued RCU callbacks beyond which
2743 batch limiting is disabled.
2745 rcutree.qlowmark= [KNL]
2746 Set threshold of queued RCU callbacks below which
2747 batch limiting is re-enabled.
2749 rcutree.rcu_idle_gp_delay= [KNL]
2750 Set wakeup interval for idle CPUs that have
2751 RCU callbacks (RCU_FAST_NO_HZ=y).
2753 rcutree.rcu_idle_lazy_gp_delay= [KNL]
2754 Set wakeup interval for idle CPUs that have
2755 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
2756 Lazy RCU callbacks are those which RCU can
2757 prove do nothing more than free memory.
2759 rcutorture.fqs_duration= [KNL]
2760 Set duration of force_quiescent_state bursts.
2762 rcutorture.fqs_holdoff= [KNL]
2763 Set holdoff time within force_quiescent_state bursts.
2765 rcutorture.fqs_stutter= [KNL]
2766 Set wait time between force_quiescent_state bursts.
2768 rcutorture.gp_exp= [KNL]
2769 Use expedited update-side primitives.
2771 rcutorture.gp_normal= [KNL]
2772 Use normal (non-expedited) update-side primitives.
2773 If both gp_exp and gp_normal are set, do both.
2774 If neither gp_exp nor gp_normal are set, still
2777 rcutorture.n_barrier_cbs= [KNL]
2778 Set callbacks/threads for rcu_barrier() testing.
2780 rcutorture.nfakewriters= [KNL]
2781 Set number of concurrent RCU writers. These just
2782 stress RCU, they don't participate in the actual
2783 test, hence the "fake".
2785 rcutorture.nreaders= [KNL]
2786 Set number of RCU readers.
2788 rcutorture.object_debug= [KNL]
2789 Enable debug-object double-call_rcu() testing.
2791 rcutorture.onoff_holdoff= [KNL]
2792 Set time (s) after boot for CPU-hotplug testing.
2794 rcutorture.onoff_interval= [KNL]
2795 Set time (s) between CPU-hotplug operations, or
2796 zero to disable CPU-hotplug testing.
2798 rcutorture.rcutorture_runnable= [BOOT]
2799 Start rcutorture running at boot time.
2801 rcutorture.shuffle_interval= [KNL]
2802 Set task-shuffle interval (s). Shuffling tasks
2803 allows some CPUs to go into dyntick-idle mode
2804 during the rcutorture test.
2806 rcutorture.shutdown_secs= [KNL]
2807 Set time (s) after boot system shutdown. This
2808 is useful for hands-off automated testing.
2810 rcutorture.stall_cpu= [KNL]
2811 Duration of CPU stall (s) to test RCU CPU stall
2812 warnings, zero to disable.
2814 rcutorture.stall_cpu_holdoff= [KNL]
2815 Time to wait (s) after boot before inducing stall.
2817 rcutorture.stat_interval= [KNL]
2818 Time (s) between statistics printk()s.
2820 rcutorture.stutter= [KNL]
2821 Time (s) to stutter testing, for example, specifying
2822 five seconds causes the test to run for five seconds,
2823 wait for five seconds, and so on. This tests RCU's
2824 ability to transition abruptly to and from idle.
2826 rcutorture.test_boost= [KNL]
2827 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
2828 "Maybe" means test if the RCU implementation
2829 under test support RCU priority boosting.
2831 rcutorture.test_boost_duration= [KNL]
2832 Duration (s) of each individual boost test.
2834 rcutorture.test_boost_interval= [KNL]
2835 Interval (s) between each boost test.
2837 rcutorture.test_no_idle_hz= [KNL]
2838 Test RCU's dyntick-idle handling. See also the
2839 rcutorture.shuffle_interval parameter.
2841 rcutorture.torture_type= [KNL]
2842 Specify the RCU implementation to test.
2844 rcutorture.verbose= [KNL]
2845 Enable additional printk() statements.
2847 rcupdate.rcu_expedited= [KNL]
2848 Use expedited grace-period primitives, for
2849 example, synchronize_rcu_expedited() instead
2850 of synchronize_rcu(). This reduces latency,
2851 but can increase CPU utilization, degrade
2852 real-time latency, and degrade energy efficiency.
2854 rcupdate.rcu_cpu_stall_suppress= [KNL]
2855 Suppress RCU CPU stall warning messages.
2857 rcupdate.rcu_cpu_stall_timeout= [KNL]
2858 Set timeout for RCU CPU stall warning messages.
2862 Run specified binary instead of /init from the ramdisk,
2863 used for early userspace startup. See initrd.
2866 Format (x86 or x86_64):
2867 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
2869 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
2871 Where reboot_mode is one of warm (soft) or cold (hard) or gpio,
2872 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
2873 reboot_force is either force or not specified,
2874 reboot_cpu is s[mp]#### with #### being the processor
2875 to be used for rebooting.
2878 [KNL, SMP] Set scheduler's default relax_domain_level.
2879 See Documentation/cgroups/cpusets.txt.
2881 reserve= [KNL,BUGS] Force the kernel to ignore some iomem area
2883 reservetop= [X86-32]
2885 Reserves a hole at the top of the kernel virtual
2890 Set the amount of memory to reserve for BIOS at
2891 the bottom of the address space.
2893 reset_devices [KNL] Force drivers to reset the underlying device
2894 during initialization.
2897 Specify the partition device for software suspend
2899 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
2901 resume_offset= [SWSUSP]
2902 Specify the offset from the beginning of the partition
2903 given by "resume=" at which the swap header is located,
2904 in <PAGE_SIZE> units (needed only for swap files).
2905 See Documentation/power/swsusp-and-swap-files.txt
2907 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
2908 read the resume files
2910 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
2911 Useful for devices that are detected asynchronously
2912 (e.g. USB and MMC devices).
2914 hibernate= [HIBERNATION]
2915 noresume Don't check if there's a hibernation image
2916 present during boot.
2917 nocompress Don't compress/decompress hibernation images.
2919 retain_initrd [RAM] Keep initrd memory after extraction
2921 rhash_entries= [KNL,NET]
2922 Set number of hash buckets for route cache
2924 riscom8= [HW,SERIAL]
2925 Format: <io_board1>[,<io_board2>[,...<io_boardN>]]
2927 ro [KNL] Mount root device read-only on boot
2929 root= [KNL] Root filesystem
2930 See name_to_dev_t comment in init/do_mounts.c.
2932 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
2933 mount the root filesystem
2935 rootflags= [KNL] Set root filesystem mount option string
2937 rootfstype= [KNL] Set root filesystem type
2939 rootwait [KNL] Wait (indefinitely) for root device to show up.
2940 Useful for devices that are detected asynchronously
2941 (e.g. USB and MMC devices).
2943 rproc_mem=nn[KMG][@address]
2944 [KNL,ARM,CMA] Remoteproc physical memory block.
2945 Memory area to be used by remote processor image,
2948 rw [KNL] Mount root device read-write on boot
2950 S [KNL] Run init in single mode
2953 See drivers/net/irda/sa1100_ir.c.
2955 sbni= [NET] Granch SBNI12 leased line adapter
2957 sched_debug [KNL] Enables verbose scheduler debug messages.
2959 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
2960 xtime_lock contention on larger systems, and/or RCU lock
2961 contention on all systems with CONFIG_MAXSMP set.
2962 Format: { "0" | "1" }
2963 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
2965 Note: increases power consumption, thus should only be
2966 enabled if running jitter sensitive (HPC/RT) workloads.
2968 security= [SECURITY] Choose a security module to enable at boot.
2969 If this boot parameter is not specified, only the first
2970 security module asking for security registration will be
2971 loaded. An invalid security module name will be treated
2972 as if no module has been chosen.
2974 selinux= [SELINUX] Disable or enable SELinux at boot time.
2975 Format: { "0" | "1" }
2976 See security/selinux/Kconfig help text.
2979 Default value is set via kernel config option.
2980 If enabled at boot time, /selinux/disable can be used
2981 later to disable prior to initial policy load.
2983 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
2984 Format: { "0" | "1" }
2985 See security/apparmor/Kconfig help text
2988 Default value is set via kernel config option.
2990 serialnumber [BUGS=X86-32]
2993 Maximal number of shapers.
2995 show_msr= [x86] show boot-time MSR settings
2996 Format: { <integer> }
2997 Show boot-time (BIOS-initialized) MSR settings.
2998 The parameter means the number of CPUs to show,
2999 for example 1 means boot CPU only.
3006 slab_max_order= [MM, SLAB]
3007 Determines the maximum allowed order for slabs.
3008 A high setting may cause OOMs due to memory
3009 fragmentation. Defaults to 1 for systems with
3010 more than 32MB of RAM, 0 otherwise.
3012 slub_debug[=options[,slabs]] [MM, SLUB]
3013 Enabling slub_debug allows one to determine the
3014 culprit if slab objects become corrupted. Enabling
3015 slub_debug can create guard zones around objects and
3016 may poison objects when not in use. Also tracks the
3017 last alloc / free. For more information see
3018 Documentation/vm/slub.txt.
3020 slub_max_order= [MM, SLUB]
3021 Determines the maximum allowed order for slabs.
3022 A high setting may cause OOMs due to memory
3023 fragmentation. For more information see
3024 Documentation/vm/slub.txt.
3026 slub_min_objects= [MM, SLUB]
3027 The minimum number of objects per slab. SLUB will
3028 increase the slab order up to slub_max_order to
3029 generate a sufficiently large slab able to contain
3030 the number of objects indicated. The higher the number
3031 of objects the smaller the overhead of tracking slabs
3032 and the less frequently locks need to be acquired.
3033 For more information see Documentation/vm/slub.txt.
3035 slub_min_order= [MM, SLUB]
3036 Determines the minimum page order for slabs. Must be
3037 lower than slub_max_order.
3038 For more information see Documentation/vm/slub.txt.
3040 slub_nomerge [MM, SLUB]
3041 Disable merging of slabs with similar size. May be
3042 necessary if there is some reason to distinguish
3043 allocs to different slabs. Debug options disable
3044 merging on their own.
3045 For more information see Documentation/vm/slub.txt.
3048 Format: <io1>[,<io2>[,...,<io8>]]
3050 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
3051 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
3052 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
3053 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
3054 smsc-ircc2.ircc_irq= [HW] IRQ line
3055 smsc-ircc2.ircc_dma= [HW] DMA channel
3056 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
3057 0: Toshiba Satellite 1800 (GP data pin select)
3058 1: Fast pin select (default)
3062 [KNL] Should the soft-lockup detector generate panics.
3065 sonypi.*= [HW] Sony Programmable I/O Control Device driver
3066 See Documentation/laptops/sonypi.txt
3068 specialix= [HW,SERIAL] Specialix multi-serial port adapter
3069 See Documentation/serial/specialix.txt.
3071 spia_io_base= [HW,MTD]
3077 Enabled the stack tracer on boot up.
3079 stacktrace_filter=[function-list]
3080 [FTRACE] Limit the functions that the stack tracer
3081 will trace at boot up. function-list is a comma separated
3082 list of functions. This list can be changed at run
3083 time by the stack_trace_filter file in the debugfs
3084 tracing directory. Note, this enables stack tracing
3085 and the stacktrace above is not needed.
3089 Set the STI (builtin display/keyboard on the HP-PARISC
3090 machines) console (graphic card) which should be used
3091 as the initial boot-console.
3092 See also comment in drivers/video/console/sticore.c.
3095 See comment in drivers/video/console/sticore.c.
3098 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
3100 sunrpc.min_resvport=
3101 sunrpc.max_resvport=
3103 SunRPC servers often require that client requests
3104 originate from a privileged port (i.e. a port in the
3105 range 0 < portnr < 1024).
3106 An administrator who wishes to reserve some of these
3107 ports for other uses may adjust the range that the
3108 kernel's sunrpc client considers to be privileged
3109 using these two parameters to set the minimum and
3110 maximum port values.
3114 Control how the NFS server code allocates CPUs to
3115 service thread pools. Depending on how many NICs
3116 you have and where their interrupts are bound, this
3117 option will affect which CPUs will do NFS serving.
3118 Note: this parameter cannot be changed while the
3119 NFS server is running.
3121 auto the server chooses an appropriate mode
3122 automatically using heuristics
3123 global a single global pool contains all CPUs
3124 percpu one pool for each CPU
3125 pernode one pool for each NUMA node (equivalent
3126 to global on non-NUMA machines)
3128 sunrpc.tcp_slot_table_entries=
3129 sunrpc.udp_slot_table_entries=
3131 Sets the upper limit on the number of simultaneous
3132 RPC calls that can be sent from the client to a
3133 server. Increasing these values may allow you to
3134 improve throughput, but will also increase the
3135 amount of memory reserved for use by the client.
3138 [KNL] Enable accounting of swap in memory resource
3139 controller if no parameter or 1 is given or disable
3140 it if 0 is given (See Documentation/cgroups/memory.txt)
3142 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
3143 Format: { <int> | force }
3144 <int> -- Number of I/O TLB slabs
3145 force -- force using of bounce buffers even if they
3146 wouldn't be automatically used by the kernel
3150 sysfs.deprecated=0|1 [KNL]
3151 Enable/disable old style sysfs layout for old udev
3152 on older distributions. When this option is enabled
3153 very new udev will not work anymore. When this option
3154 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
3155 in older udev will not work anymore.
3156 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
3157 the kernel configuration.
3159 sysrq_always_enabled
3161 Ignore sysrq setting - this boot parameter will
3162 neutralize any effect of /proc/sys/kernel/sysrq.
3163 Useful for debugging.
3167 test_suspend= [SUSPEND]
3168 Specify "mem" (for Suspend-to-RAM) or "standby" (for
3169 standby suspend) as the system sleep state to briefly
3170 enter during system startup. The system is woken from
3171 this state using a wakeup-capable RTC alarm.
3173 thash_entries= [KNL,NET]
3174 Set number of hash buckets for TCP connection
3176 thermal.act= [HW,ACPI]
3177 -1: disable all active trip points in all thermal zones
3178 <degrees C>: override all lowest active trip points
3180 thermal.crt= [HW,ACPI]
3181 -1: disable all critical trip points in all thermal zones
3182 <degrees C>: override all critical trip points
3184 thermal.nocrt= [HW,ACPI]
3185 Set to disable actions on ACPI thermal zone
3186 critical and hot trip points.
3188 thermal.off= [HW,ACPI]
3189 1: disable ACPI thermal control
3191 thermal.psv= [HW,ACPI]
3192 -1: disable all passive trip points
3193 <degrees C>: override all passive trip points to this
3196 thermal.tzp= [HW,ACPI]
3197 Specify global default ACPI thermal zone polling rate
3198 <deci-seconds>: poll all this frequency
3199 0: no polling (default)
3202 Force threading of all interrupt handlers except those
3203 marked explicitly IRQF_NO_THREAD.
3206 Enable the Transcendent memory driver if built-in.
3208 tmem.cleancache=0|1 [KNL, XEN]
3209 Default is on (1). Disable the usage of the cleancache
3210 API to send anonymous pages to the hypervisor.
3212 tmem.frontswap=0|1 [KNL, XEN]
3213 Default is on (1). Disable the usage of the frontswap
3214 API to send swap pages to the hypervisor. If disabled
3215 the selfballooning and selfshrinking are force disabled.
3217 tmem.selfballooning=0|1 [KNL, XEN]
3218 Default is on (1). Disable the driving of swap pages
3221 tmem.selfshrinking=0|1 [KNL, XEN]
3222 Default is on (1). Partial swapoff that immediately
3223 transfers pages from Xen hypervisor back to the
3224 kernel based on different criteria.
3228 Specify if the kernel should make use of the cpu
3229 topology information if the hardware supports this.
3230 The scheduler will make use of this information and
3231 e.g. base its process migration decisions on it.
3236 tpm_suspend_pcr=[HW,TPM]
3237 Format: integer pcr id
3238 Specify that at suspend time, the tpm driver
3239 should extend the specified pcr with zeros,
3240 as a workaround for some chips which fail to
3241 flush the last written pcr on TPM_SaveState.
3242 This will guarantee that all the other pcrs
3245 trace_buf_size=nn[KMG]
3246 [FTRACE] will set tracing buffer size.
3248 trace_event=[event-list]
3249 [FTRACE] Set and start specified trace events in order
3250 to facilitate early boot debugging.
3251 See also Documentation/trace/events.txt
3253 trace_options=[option-list]
3254 [FTRACE] Enable or disable tracer options at boot.
3255 The option-list is a comma delimited list of options
3256 that can be enabled or disabled just as if you were
3257 to echo the option name into
3259 /sys/kernel/debug/tracing/trace_options
3261 For example, to enable stacktrace option (to dump the
3262 stack trace of each event), add to the command line:
3264 trace_options=stacktrace
3266 See also Documentation/trace/ftrace.txt "trace options"
3270 [FTRACE] enable this option to disable tracing when a
3271 warning is hit. This turns off "tracing_on". Tracing can
3272 be enabled again by echoing '1' into the "tracing_on"
3273 file located in /sys/kernel/debug/tracing/
3275 This option is useful, as it disables the trace before
3276 the WARNING dump is called, which prevents the trace to
3277 be filled with content caused by the warning output.
3279 This option can also be set at run time via the sysctl
3280 option: kernel/traceoff_on_warning
3282 transparent_hugepage=
3284 Format: [always|madvise|never]
3285 Can be used to control the default behavior of the system
3286 with respect to transparent hugepages.
3287 See Documentation/vm/transhuge.txt for more details.
3289 tsc= Disable clocksource stability checks for TSC.
3291 [x86] reliable: mark tsc clocksource as reliable, this
3292 disables clocksource verification at runtime, as well
3293 as the stability checks done at bootup. Used to enable
3294 high-resolution timer mode on older hardware, and in
3295 virtualized environment.
3296 [x86] noirqtime: Do not use TSC to do irq accounting.
3297 Used to run time disable IRQ_TIME_ACCOUNTING on any
3298 platforms where RDTSC is slow and this accounting
3301 turbografx.map[2|3]= [HW,JOY]
3302 TurboGraFX parallel port interface
3304 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
3305 See also Documentation/input/joystick-parport.txt
3307 udbg-immortal [PPC] When debugging early kernel crashes that
3308 happen after console_init() and before a proper
3309 console driver takes over, this boot options might
3310 help "seeing" what's going on.
3312 uhash_entries= [KNL,NET]
3313 Set number of hash buckets for UDP/UDP-Lite connections
3316 [USB] Ignore overcurrent events (default N).
3317 Some badly-designed motherboards generate lots of
3318 bogus events, for ports that aren't wired to
3319 anything. Set this parameter to avoid log spamming.
3320 Note that genuine overcurrent events won't be
3324 [X86] Cause panic on unknown NMI.
3326 usbcore.authorized_default=
3327 [USB] Default USB device authorization:
3328 (default -1 = authorized except for wireless USB,
3329 0 = not authorized, 1 = authorized)
3331 usbcore.autosuspend=
3332 [USB] The autosuspend time delay (in seconds) used
3333 for newly-detected USB devices (default 2). This
3334 is the time required before an idle device will be
3335 autosuspended. Devices for which the delay is set
3336 to a negative value won't be autosuspended at all.
3338 usbcore.usbfs_snoop=
3339 [USB] Set to log all usbfs traffic (default 0 = off).
3341 usbcore.blinkenlights=
3342 [USB] Set to cycle leds on hubs (default 0 = off).
3344 usbcore.old_scheme_first=
3345 [USB] Start with the old device initialization
3346 scheme (default 0 = off).
3348 usbcore.usbfs_memory_mb=
3349 [USB] Memory limit (in MB) for buffers allocated by
3350 usbfs (default = 16, 0 = max = 2047).
3352 usbcore.use_both_schemes=
3353 [USB] Try the other device initialization scheme
3354 if the first one fails (default 1 = enabled).
3356 usbcore.initial_descriptor_timeout=
3357 [USB] Specifies timeout for the initial 64-byte
3358 USB_REQ_GET_DESCRIPTOR request in milliseconds
3359 (default 5000 = 5.0 seconds).
3362 [USBHID] The interval which mice are to be polled at.
3364 usb-storage.delay_use=
3365 [UMS] The delay in seconds before a new device is
3366 scanned for Logical Units (default 5).
3369 [UMS] A list of quirks entries to supplement or
3370 override the built-in unusual_devs list. List
3371 entries are separated by commas. Each entry has
3372 the form VID:PID:Flags where VID and PID are Vendor
3373 and Product ID values (4-digit hex numbers) and
3374 Flags is a set of characters, each corresponding
3375 to a common usb-storage quirk flag as follows:
3376 a = SANE_SENSE (collect more than 18 bytes
3378 b = BAD_SENSE (don't collect more than 18
3379 bytes of sense data);
3380 c = FIX_CAPACITY (decrease the reported
3381 device capacity by one sector);
3382 d = NO_READ_DISC_INFO (don't use
3383 READ_DISC_INFO command);
3384 e = NO_READ_CAPACITY_16 (don't use
3385 READ_CAPACITY_16 command);
3386 h = CAPACITY_HEURISTICS (decrease the
3387 reported device capacity by one
3388 sector if the number is odd);
3389 i = IGNORE_DEVICE (don't bind to this
3391 l = NOT_LOCKABLE (don't try to lock and
3392 unlock ejectable media);
3393 m = MAX_SECTORS_64 (don't transfer more
3394 than 64 sectors = 32 KB at a time);
3395 n = INITIAL_READ10 (force a retry of the
3396 initial READ(10) command);
3397 o = CAPACITY_OK (accept the capacity
3398 reported by the device);
3399 p = WRITE_CACHE (the device cache is ON
3401 r = IGNORE_RESIDUE (the device reports
3402 bogus residue values);
3403 s = SINGLE_LUN (the device has only one
3405 w = NO_WP_DETECT (don't test whether the
3406 medium is write-protected).
3407 Example: quirks=0419:aaf5:rl,0421:0433:rc
3409 user_debug= [KNL,ARM]
3411 See arch/arm/Kconfig.debug help text.
3412 1 - undefined instruction events
3414 4 - invalid data aborts
3417 Example: user_debug=31
3420 [X86] Flags controlling user PTE allocations.
3422 nohigh = do not allocate PTE pages in
3423 HIGHMEM regardless of setting
3427 vdso=2: enable compat VDSO (default with COMPAT_VDSO)
3428 vdso=1: enable VDSO (default)
3429 vdso=0: disable VDSO mapping
3432 vdso32=2: enable compat VDSO (default with COMPAT_VDSO)
3433 vdso32=1: enable 32-bit VDSO (default)
3434 vdso32=0: disable 32-bit VDSO mapping
3437 vector=percpu: enable percpu vector domain
3439 video= [FB] Frame buffer configuration
3440 See Documentation/fb/modedb.txt.
3442 video.brightness_switch_enabled= [0,1]
3443 If set to 1, on receiving an ACPI notify event
3444 generated by hotkey, video driver will adjust brightness
3445 level and then send out the event to user space through
3446 the allocated input device; If set to 0, video driver
3447 will only send out the event without touching backlight
3452 [VMMIO] Memory mapped virtio (platform) device.
3454 <size>@<baseaddr>:<irq>[:<id>]
3456 <size> := size (can use standard suffixes
3458 <baseaddr> := physical base address
3459 <irq> := interrupt number (as passed to
3461 <id> := (optional) platform device id
3463 virtio_mmio.device=1K@0x100b0000:48:7
3465 Can be used multiple times for multiple devices.
3467 vga= [BOOT,X86-32] Select a particular video mode
3468 See Documentation/x86/boot.txt and
3469 Documentation/svga.txt.
3470 Use vga=ask for menu.
3471 This is actually a boot loader parameter; the value is
3472 passed to the kernel using a special protocol.
3474 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
3475 size of <nn>. This can be used to increase the
3476 minimum size (128MB on x86). It can also be used to
3477 decrease the size and leave more room for directly
3480 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
3483 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
3486 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
3490 Controls the behavior of vsyscalls (i.e. calls to
3491 fixed addresses of 0xffffffffff600x00 from legacy
3492 code). Most statically-linked binaries and older
3493 versions of glibc use these calls. Because these
3494 functions are at fixed addresses, they make nice
3495 targets for exploits that can control RIP.
3497 emulate [default] Vsyscalls turn into traps and are
3498 emulated reasonably safely.
3500 native Vsyscalls are native syscall instructions.
3501 This is a little bit faster than trapping
3502 and makes a few dynamic recompilers work
3503 better than they would in emulation mode.
3504 It also makes exploits much easier to write.
3506 none Vsyscalls don't work at all. This makes
3507 them quite hard to use for exploits but
3508 might break your system.
3510 vt.color= [VT] Default text color.
3511 Format: 0xYX, X = foreground, Y = background.
3512 Default: 0x07 = light gray on black.
3514 vt.cur_default= [VT] Default cursor shape.
3515 Format: 0xCCBBAA, where AA, BB, and CC are the same as
3516 the parameters of the <Esc>[?A;B;Cc escape sequence;
3517 see VGA-softcursor.txt. Default: 2 = underline.
3519 vt.default_blu= [VT]
3520 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
3521 Change the default blue palette of the console.
3522 This is a 16-member array composed of values
3525 vt.default_grn= [VT]
3526 Format: <green0>,<green1>,<green2>,...,<green15>
3527 Change the default green palette of the console.
3528 This is a 16-member array composed of values
3531 vt.default_red= [VT]
3532 Format: <red0>,<red1>,<red2>,...,<red15>
3533 Change the default red palette of the console.
3534 This is a 16-member array composed of values
3540 Set system-wide default UTF-8 mode for all tty's.
3541 Default is 1, i.e. UTF-8 mode is enabled for all
3542 newly opened terminals.
3544 vt.global_cursor_default=
3547 Set system-wide default for whether a cursor
3548 is shown on new VTs. Default is -1,
3549 i.e. cursors will be created by default unless
3550 overridden by individual drivers. 0 will hide
3551 cursors, 1 will display them.
3553 vt.italic= [VT] Default color for italic text; 0-15.
3556 vt.underline= [VT] Default color for underlined text; 0-15.
3559 watchdog timers [HW,WDT] For information on watchdog timers,
3560 see Documentation/watchdog/watchdog-parameters.txt
3561 or other driver-specific files in the
3562 Documentation/watchdog/ directory.
3564 workqueue.disable_numa
3565 By default, all work items queued to unbound
3566 workqueues are affine to the NUMA nodes they're
3567 issued on, which results in better behavior in
3568 general. If NUMA affinity needs to be disabled for
3569 whatever reason, this option can be used. Note
3570 that this also can be controlled per-workqueue for
3571 workqueues visible under /sys/bus/workqueue/.
3573 workqueue.power_efficient
3574 Per-cpu workqueues are generally preferred because
3575 they show better performance thanks to cache
3576 locality; unfortunately, per-cpu workqueues tend to
3577 be more power hungry than unbound workqueues.
3579 Enabling this makes the per-cpu workqueues which
3580 were observed to contribute significantly to power
3581 consumption unbound, leading to measurably lower
3582 power usage at the cost of small performance
3585 The default value of this parameter is determined by
3586 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
3588 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
3589 default x2apic cluster mode on platforms
3592 x86_intel_mid_timer= [X86-32,APBT]
3593 Choose timer option for x86 Intel MID platform.
3594 Two valid options are apbt timer only and lapic timer
3595 plus one apbt timer for broadcast timer.
3596 x86_intel_mid_timer=apbt_only | lapic_and_apbt
3598 xen_emul_unplug= [HW,X86,XEN]
3599 Unplug Xen emulated devices
3600 Format: [unplug0,][unplug1]
3601 ide-disks -- unplug primary master IDE devices
3602 aux-ide-disks -- unplug non-primary-master IDE devices
3603 nics -- unplug network devices
3604 all -- unplug all emulated devices (NICs and IDE disks)
3605 unnecessary -- unplugging emulated devices is
3606 unnecessary even if the host did not respond to
3608 never -- do not unplug even if version check succeeds
3610 xen_nopvspin [X86,XEN]
3611 Disables the ticketlock slowpath using Xen PV
3614 xirc2ps_cs= [NET,PCMCIA]
3616 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
3618 ______________________________________________________________________
3622 Add more DRM drivers.