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 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
235 Format: To spoof as Windows 98: ="Microsoft Windows"
237 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
238 acpi_osi="string1" # add string1 -- only one string
239 acpi_osi="!string2" # remove built-in string2
240 acpi_osi= # disable all strings
243 Override the pmtimer bug detection: force the kernel
244 to assume that this machine's pmtimer latches its value
245 and always returns good values.
247 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
248 Format: { level | edge | high | low }
250 acpi_serialize [HW,ACPI] force serialization of AML methods
252 acpi_skip_timer_override [HW,ACPI]
253 Recognize and ignore IRQ0/pin2 Interrupt Override.
254 For broken nForce2 BIOS resulting in XT-PIC timer.
256 acpi_sleep= [HW,ACPI] Sleep options
257 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
258 old_ordering, nonvs, sci_force_enable }
259 See Documentation/power/video.txt for information on
261 s3_beep is for debugging; it makes the PC's speaker beep
262 as soon as the kernel's real-mode entry point is called.
263 s4_nohwsig prevents ACPI hardware signature from being
264 used during resume from hibernation.
265 old_ordering causes the ACPI 1.0 ordering of the _PTS
266 control method, with respect to putting devices into
267 low power states, to be enforced (the ACPI 2.0 ordering
268 of _PTS is used by default).
269 nonvs prevents the kernel from saving/restoring the
270 ACPI NVS memory during suspend/hibernation and resume.
271 sci_force_enable causes the kernel to set SCI_EN directly
272 on resume from S1/S3 (which is against the ACPI spec,
273 but some broken systems don't work without it).
275 acpi_use_timer_override [HW,ACPI]
276 Use timer override. For some broken Nvidia NF5 boards
277 that require a timer override, but don't have HPET
279 acpi_enforce_resources= [ACPI]
280 { strict | lax | no }
281 Check for resource conflicts between native drivers
282 and ACPI OperationRegions (SystemIO and SystemMemory
283 only). IO ports and memory declared in ACPI might be
284 used by the ACPI subsystem in arbitrary AML code and
285 can interfere with legacy drivers.
286 strict (default): access to resources claimed by ACPI
287 is denied; legacy drivers trying to access reserved
288 resources will fail to bind to device using them.
289 lax: access to resources claimed by ACPI is allowed;
290 legacy drivers trying to access reserved resources
291 will bind successfully but a warning message is logged.
292 no: ACPI OperationRegions are not marked as reserved,
293 no further checks are performed.
295 add_efi_memmap [EFI; X86] Include EFI memory map in
296 kernel's map of available physical RAM.
299 { off | try_unsupported }
300 off: disable AGP support
301 try_unsupported: try to drive unsupported chipsets
302 (may crash computer or cause data corruption)
305 See Documentation/sound/alsa/alsa-parameters.txt
308 Allow the default userspace alignment fault handler
309 behaviour to be specified. Bit 0 enables warnings,
310 bit 1 enables fixups, and bit 2 sends a segfault.
312 align_va_addr= [X86-64]
313 Align virtual addresses by clearing slice [14:12] when
314 allocating a VMA at process creation time. This option
315 gives you up to 3% performance improvement on AMD F15h
316 machines (where it is enabled by default) for a
317 CPU-intensive style benchmark, and it can vary highly in
318 a microbenchmark depending on workload and compiler.
320 32: only for 32-bit processes
321 64: only for 64-bit processes
322 on: enable for both 32- and 64-bit processes
323 off: disable for both 32- and 64-bit processes
325 alloc_snapshot [FTRACE]
326 Allocate the ftrace snapshot buffer on boot up when the
327 main buffer is allocated. This is handy if debugging
328 and you need to use tracing_snapshot() on boot up, and
329 do not want to use tracing_snapshot_alloc() as it needs
330 to be done where GFP_KERNEL allocations are allowed.
332 amd_iommu= [HW,X86-64]
333 Pass parameters to the AMD IOMMU driver in the system.
335 fullflush - enable flushing of IO/TLB entries when
336 they are unmapped. Otherwise they are
337 flushed before they will be reused, which
339 off - do not initialize any AMD IOMMU found in
341 force_isolation - Force device isolation for all
342 devices. The IOMMU driver is not
343 allowed anymore to lift isolation
344 requirements as needed. This option
345 does not override iommu=pt
347 amd_iommu_dump= [HW,X86-64]
348 Enable AMD IOMMU driver option to dump the ACPI table
349 for AMD IOMMU. With this option enabled, AMD IOMMU
350 driver will print ACPI tables for AMD IOMMU during
351 IOMMU initialization.
353 amijoy.map= [HW,JOY] Amiga joystick support
354 Map of devices attached to JOY0DAT and JOY1DAT
356 See also Documentation/input/joystick.txt
358 analog.map= [HW,JOY] Analog joystick and gamepad support
359 Specifies type or capabilities of an analog joystick
360 connected to one of 16 gameports
361 Format: <type1>,<type2>,..<type16>
364 Power management functions (SPARCstation-4/5 + deriv.)
366 Disable APC CPU standby support. SPARCstation-Fox does
367 not play well with APC CPU idle - disable it if you have
368 APC and your system crashes randomly.
370 apic= [APIC,X86-32] Advanced Programmable Interrupt Controller
371 Change the output verbosity whilst booting
372 Format: { quiet (default) | verbose | debug }
373 Change the amount of debugging information output
374 when initialising the APIC and IO-APIC components.
377 See Documentation/networking/ipv6.txt.
379 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
380 Limit apic dumping. The parameter defines the maximal
381 number of local apics being dumped. Also it is possible
382 to set it to "all" by meaning -- no limit here.
383 Format: { 1 (default) | 2 | ... | all }.
384 The parameter valid if only apic=debug or
385 apic=verbose is specified.
386 Example: apic=debug show_lapic=all
388 apm= [APM] Advanced Power Management
389 See header of arch/x86/kernel/apm_32.c.
391 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
392 Format: <io>,<irq>,<nodeID>
396 atarimouse= [HW,MOUSE] Atari Mouse
398 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
399 EzKey and similar keyboards
401 atkbd.reset= [HW] Reset keyboard during initialization
403 atkbd.set= [HW] Select keyboard code set
404 Format: <int> (2 = AT (default), 3 = PS/2)
406 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
409 atkbd.softraw= [HW] Choose between synthetic and real raw mode
410 Format: <bool> (0 = real, 1 = synthetic (default))
412 atkbd.softrepeat= [HW]
413 Use software keyboard repeat
415 baycom_epp= [HW,AX25]
418 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
420 See header of drivers/net/hamradio/baycom_par.c.
422 baycom_ser_fdx= [HW,AX25]
423 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
424 Format: <io>,<irq>,<mode>[,<baud>]
425 See header of drivers/net/hamradio/baycom_ser_fdx.c.
427 baycom_ser_hdx= [HW,AX25]
428 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
429 Format: <io>,<irq>,<mode>
430 See header of drivers/net/hamradio/baycom_ser_hdx.c.
432 boot_delay= Milliseconds to delay each printk during boot.
433 Values larger than 10 seconds (10000) are changed to
437 bootmem_debug [KNL] Enable bootmem allocator debug messages.
439 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
440 bttv.radio= Most important insmod options are available as
442 bttv.pll= See Documentation/video4linux/bttv/Insmod-options
445 bulk_remove=off [PPC] This parameter disables the use of the pSeries
446 firmware feature for flushing multiple hpte entries
449 c101= [NET] Moxa C101 synchronous serial card
451 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
452 Sometimes CPU hardware bugs make them report the cache
453 size incorrectly. The kernel will attempt work arounds
454 to fix known problems, but for some CPUs it is not
455 possible to determine what the correct size should be.
456 This option provides an override for these situations.
458 ccw_timeout_log [S390]
459 See Documentation/s390/CommonIO for details.
461 cgroup_disable= [KNL] Disable a particular controller
462 Format: {name of the controller(s) to disable}
463 {Currently supported controllers - "memory"}
465 checkreqprot [SELINUX] Set initial checkreqprot flag value.
466 Format: { "0" | "1" }
467 See security/selinux/Kconfig help text.
468 0 -- check protection applied by kernel (includes
469 any implied execute protection).
470 1 -- check protection requested by application.
471 Default value is set via a kernel config option.
472 Value can be changed at runtime via
473 /selinux/checkreqprot.
476 See Documentation/s390/CommonIO for details.
479 Keep all clocks already enabled by bootloader on,
480 even if no driver has claimed them. This is useful
481 for debug and development, but should not be
482 needed on a platform with proper driver support.
483 For more information, see Documentation/clk.txt.
485 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
487 Forces specified clocksource (if available) to be used
488 when calculating gettimeofday(). If specified
489 clocksource is not available, it defaults to PIT.
490 Format: { pit | tsc | cyclone | pmtmr }
492 clocksource= Override the default clocksource
494 Override the default clocksource and use the clocksource
495 with the name specified.
496 Some clocksource names to choose from, depending on
498 [all] jiffies (this is the base, fallback clocksource)
500 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
501 pxa_timer,timer3,32k_counter,timer0_1
503 [X86-32] pit,hpet,tsc;
504 scx200_hrt on Geode; cyclone on IBM x440
512 clearcpuid=BITNUM [X86]
513 Disable CPUID feature X for the kernel. See
514 arch/x86/include/asm/cpufeature.h for the valid bit
515 numbers. Note the Linux specific bits are not necessarily
516 stable over kernel options, but the vendor specific
518 Also note that user programs calling CPUID directly
519 or using the feature without checking anything
520 will still see it. This just prevents it from
521 being used by the kernel or shown in /proc/cpuinfo.
522 Also note the kernel might malfunction if you disable
526 Sets the size of kernel global memory area for contiguous
527 memory allocations. For more information, see
528 include/linux/dma-contiguous.h
530 cmo_free_hint= [PPC] Format: { yes | no }
531 Specify whether pages are marked as being inactive
532 when they are freed. This is used in CMO environments
533 to determine OS memory pressure for page stealing by
537 coherent_pool=nn[KMG] [ARM,KNL]
538 Sets the size of memory pool for coherent, atomic dma
539 allocations, by default set to 256K.
541 code_bytes [X86] How many bytes of object code to print
546 com20020= [HW,NET] ARCnet - COM20020 chipset
548 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
550 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
554 ARCnet - COM90xx chipset (memory-mapped buffers)
555 Format: <io>[,<irq>[,<memstart>]]
557 condev= [HW,S390] console device
560 console= [KNL] Output console device and options.
562 tty<n> Use the virtual console device <n>.
566 Use the specified serial port. The options are of
567 the form "bbbbpnf", where "bbbb" is the baud rate,
568 "p" is parity ("n", "o", or "e"), "n" is number of
569 bits, and "f" is flow control ("r" for RTS or
570 omit it). Default is "9600n8".
572 See Documentation/serial-console.txt for more
574 Documentation/networking/netconsole.txt for an
577 uart[8250],io,<addr>[,options]
578 uart[8250],mmio,<addr>[,options]
579 Start an early, polled-mode console on the 8250/16550
580 UART at the specified I/O port or MMIO address,
581 switching to the matching ttyS device later. The
582 options are the same as for ttyS, above.
583 hvc<n> Use the hypervisor console device <n>. This is for
584 both Xen and PowerPC hypervisors.
586 If the device connected to the port is not a TTY but a braille
587 device, prepend "brl," before the device type, for instance
589 For now, only VisioBraille is supported.
591 consoleblank= [KNL] The console blank (screen saver) timeout in
592 seconds. Defaults to 10*60 = 10mins. A value of 0
593 disables the blank timer.
596 [KNL] Change the default value for
597 /proc/<pid>/coredump_filter.
598 See also Documentation/filesystems/proc.txt.
600 cpuidle.off=1 [CPU_IDLE]
601 disable the cpuidle sub-system
603 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
605 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
607 crashkernel=size[KMG][@offset[KMG]]
608 [KNL] Using kexec, Linux can switch to a 'crash kernel'
609 upon panic. This parameter reserves the physical
610 memory region [offset, offset + size] for that kernel
611 image. If '@offset' is omitted, then a suitable offset
612 is selected automatically. Check
613 Documentation/kdump/kdump.txt for further details.
615 crashkernel=range1:size1[,range2:size2,...][@offset]
616 [KNL] Same as above, but depends on the memory
617 in the running system. The syntax of range is
618 start-[end] where start and end are both
619 a memory unit (amount[KMG]). See also
620 Documentation/kdump/kdump.txt for an example.
622 crashkernel=size[KMG],high
623 [KNL, x86_64] range could be above 4G. Allow kernel
624 to allocate physical memory region from top, so could
625 be above 4G if system have more than 4G ram installed.
626 Otherwise memory region will be allocated below 4G, if
628 It will be ignored if crashkernel=X is specified.
629 crashkernel=size[KMG],low
630 [KNL, x86_64] range under 4G. When crashkernel=X,high
631 is passed, kernel could allocate physical memory region
632 above 4G, that cause second kernel crash on system
633 that require some amount of low memory, e.g. swiotlb
634 requires at least 64M+32K low memory. Kernel would
635 try to allocate 72M below 4G automatically.
636 This one let user to specify own low range under 4G
637 for second kernel instead.
638 0: to disable low allocation.
639 It will be ignored when crashkernel=X,high is not used
640 or memory reserved is below 4G.
645 cs89x0_media= [HW,NET]
646 Format: { rj45 | aui | bnc }
649 See header of drivers/s390/block/dasd_devmap.c.
651 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
652 (one device per port)
653 Format: <port#>,<type>
654 See also Documentation/input/joystick-parport.txt
656 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
657 time. See Documentation/dynamic-debug-howto.txt for
658 details. Deprecated, see dyndbg.
660 debug [KNL] Enable kernel debugging (events log level).
663 [KNL] verbose self-tests
665 Print debugging info while doing the locking API
667 We default to 0 (no extra messages), setting it to
668 1 will print _a lot_ more information - normally
669 only useful to kernel developers.
671 debug_objects [KNL] Enable object debugging
674 [KNL] Disable object debugging
676 debug_guardpage_minorder=
677 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
678 parameter allows control of the order of pages that will
679 be intentionally kept free (and hence protected) by the
680 buddy allocator. Bigger value increase the probability
681 of catching random memory corruption, but reduce the
682 amount of memory for normal system use. The maximum
683 possible value is MAX_ORDER/2. Setting this parameter
684 to 1 or 2 should be enough to identify most random
685 memory corruption problems caused by bugs in kernel or
686 driver code when a CPU writes to (or reads from) a
687 random memory location. Note that there exists a class
688 of memory corruptions problems caused by buggy H/W or
689 F/W or by drivers badly programing DMA (basically when
690 memory is written at bus level and the CPU MMU is
691 bypassed) which are not detectable by
692 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
693 tracking down these problems.
695 debugpat [X86] Enable PAT debugging
697 decnet.addr= [HW,NET]
698 Format: <area>[,<node>]
699 See also Documentation/networking/decnet.txt.
702 [same as hugepagesz=] The size of the default
703 HugeTLB page size. This is the size represented by
704 the legacy /proc/ hugepages APIs, used for SHM, and
705 default size when mounting hugetlbfs filesystems.
706 Defaults to the default architecture's huge page size
710 Set number of hash buckets for dentry cache.
713 IO parameters + enable/disable command.
715 digiepca= [HW,SERIAL]
716 See drivers/char/README.epca and
717 Documentation/serial/digiepca.txt.
720 See Documentation/networking/ipv6.txt.
722 disable_ddw [PPC/PSERIES]
723 Disable Dynamic DMA Window support. Use this if
724 to workaround buggy firmware.
727 See Documentation/networking/ipv6.txt.
729 disable_mtrr_cleanup [X86]
730 The kernel tries to adjust MTRR layout from continuous
731 to discrete, to make X server driver able to add WB
732 entry later. This parameter disables that.
734 disable_mtrr_trim [X86, Intel and AMD only]
735 By default the kernel will trim any uncacheable
736 memory out of your available memory pool based on
737 MTRR settings. This parameter disables that behavior,
738 possibly causing your machine to run very slowly.
740 disable_timer_pin_1 [X86]
741 Disable PIN 1 of APIC timer
742 Can be useful to work around chipset bugs.
744 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
745 this option disables the debugging code at boot.
747 dma_debug_entries=<number>
748 This option allows to tune the number of preallocated
749 entries for DMA-API debugging code. One entry is
750 required per DMA-API allocation. Use this if the
751 DMA-API debugging code disables itself because the
752 architectural default is too low.
754 dma_debug_driver=<driver_name>
755 With this option the DMA-API debugging driver
756 filter feature can be enabled at boot time. Just
757 pass the driver to filter for as the parameter.
758 The filter can be disabled or changed to another
759 driver later using sysfs.
761 drm_kms_helper.edid_firmware=[<connector>:]<file>
762 Broken monitors, graphic adapters and KVMs may
763 send no or incorrect EDID data sets. This parameter
764 allows to specify an EDID data set in the
765 /lib/firmware directory that is used instead.
766 Generic built-in EDID data sets are used, if one of
767 edid/1024x768.bin, edid/1280x1024.bin,
768 edid/1680x1050.bin, or edid/1920x1080.bin is given
769 and no file with the same name exists. Details and
770 instructions how to build your own EDID data are
771 available in Documentation/EDID/HOWTO.txt. An EDID
772 data set will only be used for a particular connector,
773 if its name and a colon are prepended to the EDID
778 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
779 module.dyndbg[="val"]
780 Enable debug messages at boot time. See
781 Documentation/dynamic-debug-howto.txt for details.
783 earlycon= [KNL] Output early console device and options.
784 uart[8250],io,<addr>[,options]
785 uart[8250],mmio,<addr>[,options]
786 uart[8250],mmio32,<addr>[,options]
787 Start an early, polled-mode console on the 8250/16550
788 UART at the specified I/O port or MMIO address.
789 MMIO inter-register address stride is either 8-bit
790 (mmio) or 32-bit (mmio32).
791 The options are the same as for ttyS, above.
793 earlyprintk= [X86,SH,BLACKFIN,ARM]
796 earlyprintk=serial[,ttySn[,baudrate]]
797 earlyprintk=serial[,0x...[,baudrate]]
798 earlyprintk=ttySn[,baudrate]
799 earlyprintk=dbgp[debugController#]
801 earlyprintk is useful when the kernel crashes before
802 the normal console is initialized. It is not enabled by
803 default because it has some cosmetic problems.
805 Append ",keep" to not disable it when the real console
808 Only vga or serial or usb debug port at a time.
810 Currently only ttyS0 and ttyS1 may be specified by
811 name. Other I/O ports may be explicitly specified
812 on some architectures (x86 and arm at least) by
813 replacing ttySn with an I/O port address, like this:
814 earlyprintk=serial,0x1008,115200
815 You can find the port for a given device in
816 /proc/tty/driver/serial:
817 2: uart:ST16650V2 port:00001008 irq:18 ...
819 Interaction with the standard serial driver is not
822 The VGA output is eventually overwritten by the real
825 The xen output can only be used by Xen PV guests.
827 ekgdboc= [X86,KGDB] Allow early kernel console debugging
830 This is designed to be used in conjunction with
831 the boot argument: earlyprintk=vga
834 Format: {"off" | "on" | "skip[mbr]"}
836 efi_no_storage_paranoia [EFI; X86]
837 Using this parameter you can use more than 50% of
838 your efi variable storage. Use this parameter only if
839 you are really sure that your UEFI does sane gc and
840 fulfills the spec otherwise your board may brick.
842 eisa_irq_edge= [PARISC,HW]
843 See header of drivers/parisc/eisa.c.
846 See comment before function elanfreq_setup() in
847 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
850 Format: {"cfq" | "deadline" | "noop"}
851 See Documentation/block/cfq-iosched.txt and
852 Documentation/block/deadline-iosched.txt for details.
854 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
855 Specifies physical address of start of kernel core
856 image elf header and optionally the size. Generally
857 kexec loader will pass this option to capture kernel.
858 See Documentation/kdump/kdump.txt for details.
860 enable_mtrr_cleanup [X86]
861 The kernel tries to adjust MTRR layout from continuous
862 to discrete, to make X server driver able to add WB
863 entry later. This parameter enables that.
865 enable_timer_pin_1 [X86]
866 Enable PIN 1 of APIC timer
867 Can be useful to work around chipset bugs
868 (in particular on some ATI chipsets).
869 The kernel tries to set a reasonable default.
871 enforcing [SELINUX] Set initial enforcing status.
873 See security/selinux/Kconfig help text.
874 0 -- permissive (log only, no denials).
875 1 -- enforcing (deny and log).
877 Value can be changed at runtime via /selinux/enforce.
880 Disable Error Record Serialization Table (ERST)
883 ether= [HW,NET] Ethernet cards parameters
884 This option is obsoleted by the "netdev=" option, which
885 has equivalent usage. See its documentation for details.
889 Permit 'security.evm' to be updated regardless of
890 current integrity status.
894 fail_make_request=[KNL]
895 General fault injection mechanism.
896 Format: <interval>,<probability>,<space>,<times>
897 See also Documentation/fault-injection/.
900 See Documentation/blockdev/floppy.txt.
902 force_pal_cache_flush
903 [IA-64] Avoid check_sal_cache_flush which may hang on
904 buggy SAL_CACHE_FLUSH implementations. Using this
905 parameter will force ia64_sal_cache_flush to call
906 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
909 [FTRACE] will set and start the specified tracer
910 as early as possible in order to facilitate early
913 ftrace_dump_on_oops[=orig_cpu]
914 [FTRACE] will dump the trace buffers on oops.
915 If no parameter is passed, ftrace will dump
916 buffers of all CPUs, but if you pass orig_cpu, it will
917 dump only the buffer of the CPU that triggered the
920 ftrace_filter=[function-list]
921 [FTRACE] Limit the functions traced by the function
922 tracer at boot up. function-list is a comma separated
923 list of functions. This list can be changed at run
924 time by the set_ftrace_filter file in the debugfs
927 ftrace_notrace=[function-list]
928 [FTRACE] Do not trace the functions specified in
929 function-list. This list can be changed at run time
930 by the set_ftrace_notrace file in the debugfs
933 ftrace_graph_filter=[function-list]
934 [FTRACE] Limit the top level callers functions traced
935 by the function graph tracer at boot up.
936 function-list is a comma separated list of functions
937 that can be changed at run time by the
938 set_graph_function file in the debugfs tracing directory.
941 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
942 support via parallel port (up to 5 devices per port)
943 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
944 See also Documentation/input/joystick-parport.txt
948 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
952 gcov_persist= [GCOV] When non-zero (default), profiling data for
953 kernel modules is saved and remains accessible via
954 debugfs, even when the module is unloaded/reloaded.
955 When zero, profiling data is discarded and associated
956 debugfs files are removed at module unload time.
958 gpt [EFI] Forces disk with valid GPT signature but
959 invalid Protective MBR to be treated as GPT.
961 grcan.enable0= [HW] Configuration of physical interface 0. Determines
962 the "Enable 0" bit of the configuration register.
965 grcan.enable1= [HW] Configuration of physical interface 1. Determines
966 the "Enable 0" bit of the configuration register.
969 grcan.select= [HW] Select which physical interface to use.
972 grcan.txsize= [HW] Sets the size of the tx buffer.
973 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
975 grcan.rxsize= [HW] Sets the size of the rx buffer.
976 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
979 hashdist= [KNL,NUMA] Large hashes allocated during boot
980 are distributed across NUMA nodes. Defaults on
981 for 64-bit NUMA, off otherwise.
982 Format: 0 | 1 (for off | on)
984 hcl= [IA-64] SGI's Hardware Graph compatibility layer
986 hd= [EIDE] (E)IDE hard drive subsystem geometry
987 Format: <cyl>,<head>,<sect>
990 Disable Hardware Error Source Table (HEST) support;
991 corresponding firmware-first mode error processing
992 logic will be disabled.
994 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
995 size of <nn>. This works even on boxes that have no
996 highmem otherwise. This also works to reduce highmem
997 size on bigger boxes.
999 highres= [KNL] Enable/disable high resolution timer mode.
1000 Valid parameters: "on", "off"
1004 See Documentation/isdn/README.HiSax.
1008 hpet= [X86-32,HPET] option to control HPET usage
1009 Format: { enable (default) | disable | force |
1011 disable: disable HPET and use PIT instead
1012 force: allow force enabled of undocumented chips (ICH4,
1014 verbose: show contents of HPET registers during setup
1016 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1017 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
1018 On x86-64 and powerpc, this option can be specified
1019 multiple times interleaved with hugepages= to reserve
1020 huge pages of different sizes. Valid pages sizes on
1021 x86-64 are 2M (when the CPU supports "pse") and 1G
1022 (when the CPU supports the "pdpe1gb" cpuinfo flag)
1023 Note that 1GB pages can only be allocated at boot time
1024 using hugepages= and not freed afterwards.
1026 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1027 terminal devices. Valid values: 0..8
1028 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1029 If specified, z/VM IUCV HVC accepts connections
1030 from listed z/VM user IDs only.
1032 hwthread_map= [METAG] Comma-separated list of Linux cpu id to
1033 hardware thread id mappings.
1034 Format: <cpu>:<hwthread>
1037 Do not unregister boot console at start. This is only
1038 useful for debugging when something happens in the window
1039 between unregistering the boot console and initializing
1042 i2c_bus= [HW] Override the default board specific I2C bus speed
1043 or register an additional I2C bus that is not
1044 registered from board initialization code.
1048 i8042.debug [HW] Toggle i8042 debug mode
1049 i8042.direct [HW] Put keyboard port into non-translated mode
1050 i8042.dumbkbd [HW] Pretend that controller can only read data from
1051 keyboard and cannot control its state
1052 (Don't attempt to blink the leds)
1053 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1054 i8042.nokbd [HW] Don't check/create keyboard port
1055 i8042.noloop [HW] Disable the AUX Loopback command while probing
1057 i8042.nomux [HW] Don't check presence of an active multiplexing
1059 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1061 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1062 i8042.reset [HW] Reset the controller during init and cleanup
1063 i8042.unlock [HW] Unlock (ignore) the keylock
1067 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1068 indicates that the driver is running on unsupported
1070 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1071 does not match list of supported models.
1073 [HW] Report power status in /proc/i8k
1074 (disabled by default)
1075 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1078 i915.invert_brightness=
1079 [DRM] Invert the sense of the variable that is used to
1080 set the brightness of the panel backlight. Normally a
1081 brightness value of 0 indicates backlight switched off,
1082 and the maximum of the brightness value sets the backlight
1083 to maximum brightness. If this parameter is set to 0
1084 (default) and the machine requires it, or this parameter
1085 is set to 1, a brightness value of 0 sets the backlight
1086 to maximum brightness, and the maximum of the brightness
1087 value switches the backlight off.
1088 -1 -- never invert brightness
1089 0 -- machine default
1090 1 -- force brightness inversion
1093 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1095 ide-core.nodma= [HW] (E)IDE subsystem
1096 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1097 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1098 .cdrom .chs .ignore_cable are additional options
1099 See Documentation/ide/ide.txt.
1101 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1102 Claim all unknown PCI IDE storage controllers.
1105 Format: idle=poll, idle=halt, idle=nomwait
1106 Poll forces a polling idle loop that can slightly
1107 improve the performance of waking up a idle CPU, but
1108 will use a lot of power and make the system run hot.
1110 idle=halt: Halt is forced to be used for CPU idle.
1111 In such case C2/C3 won't be used again.
1112 idle=nomwait: Disable mwait for CPU C-states
1114 ignore_loglevel [KNL]
1115 Ignore loglevel setting - this will print /all/
1116 kernel messages to the console. Useful for debugging.
1117 We also add it as printk module parameter, so users
1118 could change it dynamically, usually by
1119 /sys/module/printk/parameters/ignore_loglevel.
1121 ihash_entries= [KNL]
1122 Set number of hash buckets for inode cache.
1124 ima_appraise= [IMA] appraise integrity measurements
1125 Format: { "off" | "enforce" | "fix" }
1128 ima_appraise_tcb [IMA]
1129 The builtin appraise policy appraises all files
1133 Format: { "0" | "1" }
1134 0 -- integrity auditing messages. (Default)
1135 1 -- enable informational integrity auditing messages.
1138 Format: { "sha1" | "md5" }
1142 Load a policy which meets the needs of the Trusted
1143 Computing Base. This means IMA will measure all
1144 programs exec'd, files mmap'd for exec, and all files
1145 opened for read by uid=0.
1149 Run specified binary instead of /sbin/init as init
1152 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1153 for working out where the kernel is dying during
1156 initrd= [BOOT] Specify the location of the initial ramdisk
1158 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1161 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1163 Enable intel iommu driver.
1165 Disable intel iommu driver.
1166 igfx_off [Default Off]
1167 By default, gfx is mapped as normal device. If a gfx
1168 device has a dedicated DMAR unit, the DMAR unit is
1169 bypassed by not enabling DMAR with this option. In
1170 this case, gfx device will use physical address for
1173 With this option iommu will not optimize to look
1174 for io virtual address below 32-bit forcing dual
1175 address cycle on pci bus for cards supporting greater
1176 than 32-bit addressing. The default is to look
1177 for translation below 32-bit and if not available
1178 then look in the higher range.
1179 strict [Default Off]
1180 With this option on every unmap_single operation will
1181 result in a hardware IOTLB flush operation as opposed
1182 to batching them for performance.
1183 sp_off [Default Off]
1184 By default, super page will be supported if Intel IOMMU
1185 has the capability. With this option, super page will
1188 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1189 0 disables intel_idle and fall back on acpi_idle.
1190 1 to 6 specify maximum depth of C-state.
1194 Do not enable intel_pstate as the default
1195 scaling driver for the supported processors
1197 intremap= [X86-64, Intel-IOMMU]
1198 on enable Interrupt Remapping (default)
1199 off disable Interrupt Remapping
1200 nosid disable Source ID checking
1202 BIOS x2APIC opt-out request will be ignored
1204 iomem= Disable strict checking of access to MMIO memory
1205 strict regions from userspace.
1222 io7= [HW] IO7 for Marvel based alpha systems
1223 See comment before marvel_specify_io7 in
1224 arch/alpha/kernel/core_marvel.c.
1226 io_delay= [X86] I/O delay method
1228 Standard port 0x80 based delay
1230 Alternate port 0xed based delay (needed on some systems)
1232 Simple two microseconds delay
1237 See Documentation/filesystems/nfs/nfsroot.txt.
1239 ip2= [HW] Set IO/IRQ pairs for up to 4 IntelliPort boards
1240 See comment before ip2_setup() in
1241 drivers/char/ip2/ip2base.c.
1244 When an interrupt is not handled search all handlers
1245 for it. Intended to get systems with badly broken
1249 When an interrupt is not handled search all handlers
1250 for it. Also check all handlers each timer
1251 interrupt. Intended to get systems with badly broken
1255 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1257 isolcpus= [KNL,SMP] Isolate CPUs from the general scheduler.
1259 <cpu number>,...,<cpu number>
1261 <cpu number>-<cpu number>
1262 (must be a positive range in ascending order)
1264 <cpu number>,...,<cpu number>-<cpu number>
1266 This option can be used to specify one or more CPUs
1267 to isolate from the general SMP balancing and scheduling
1268 algorithms. You can move a process onto or off an
1269 "isolated" CPU via the CPU affinity syscalls or cpuset.
1270 <cpu number> begins at 0 and the maximum value is
1271 "number of CPUs in system - 1".
1273 This option is the preferred way to isolate CPUs. The
1274 alternative -- manually setting the CPU mask of all
1275 tasks in the system -- can cause problems and
1276 suboptimal load balancer performance.
1280 js= [HW,JOY] Analog joystick
1281 See Documentation/input/joystick.txt.
1285 kernelcore=nn[KMG] [KNL,X86,IA-64,PPC] This parameter
1286 specifies the amount of memory usable by the kernel
1287 for non-movable allocations. The requested amount is
1288 spread evenly throughout all nodes in the system. The
1289 remaining memory in each node is used for Movable
1290 pages. In the event, a node is too small to have both
1291 kernelcore and Movable pages, kernelcore pages will
1292 take priority and other nodes will have a larger number
1293 of kernelcore pages. The Movable zone is used for the
1294 allocation of pages that may be reclaimed or moved
1295 by the page migration subsystem. This means that
1296 HugeTLB pages may not be allocated from this zone.
1297 Note that allocations like PTEs-from-HighMem still
1298 use the HighMem zone if it exists, and the Normal
1299 zone if it does not.
1301 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
1302 Format: <Controller#>[,poll interval]
1303 The controller # is the number of the ehci usb debug
1304 port as it is probed via PCI. The poll interval is
1305 optional and is the number seconds in between
1306 each poll cycle to the debug port in case you need
1307 the functionality for interrupting the kernel with
1308 gdb or control-c on the dbgp connection. When
1309 not using this parameter you use sysrq-g to break into
1310 the kernel debugger.
1312 kgdboc= [KGDB,HW] kgdb over consoles.
1313 Requires a tty driver that supports console polling,
1314 or a supported polling keyboard driver (non-usb).
1315 Serial only format: <serial_device>[,baud]
1316 keyboard only format: kbd
1317 keyboard and serial format: kbd,<serial_device>[,baud]
1318 Optional Kernel mode setting:
1319 kms, kbd format: kms,kbd
1320 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
1322 kgdbwait [KGDB] Stop kernel execution and enter the
1323 kernel debugger at the earliest opportunity.
1325 kmac= [MIPS] korina ethernet MAC address.
1326 Configure the RouterBoard 532 series on-chip
1327 Ethernet adapter MAC address.
1329 kmemleak= [KNL] Boot-time kmemleak enable/disable
1330 Valid arguments: on, off
1333 kstack=N [X86] Print N words from the kernel stack
1336 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
1337 Default is 0 (don't ignore, but inject #GP)
1339 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
1343 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
1344 Default is 1 (enabled)
1346 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
1348 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
1350 kvm-intel.ept= [KVM,Intel] Disable extended page tables
1351 (virtualized MMU) support on capable Intel chips.
1352 Default is 1 (enabled)
1354 kvm-intel.emulate_invalid_guest_state=
1355 [KVM,Intel] Enable emulation of invalid guest states
1356 Default is 0 (disabled)
1358 kvm-intel.flexpriority=
1359 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
1360 Default is 1 (enabled)
1363 [KVM,Intel] Enable VMX nesting (nVMX).
1364 Default is 0 (disabled)
1366 kvm-intel.unrestricted_guest=
1367 [KVM,Intel] Disable unrestricted guest feature
1368 (virtualized real and unpaged mode) on capable
1369 Intel chips. Default is 1 (enabled)
1371 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
1372 feature (tagged TLBs) on capable Intel chips.
1373 Default is 1 (enabled)
1379 lapic [X86-32,APIC] Enable the local APIC even if BIOS
1382 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
1383 value for LAPIC timer one-shot implementation. Default
1384 back to the programmable timer unit in the LAPIC.
1386 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
1389 libata.dma= [LIBATA] DMA control
1390 libata.dma=0 Disable all PATA and SATA DMA
1391 libata.dma=1 PATA and SATA Disk DMA only
1392 libata.dma=2 ATAPI (CDROM) DMA only
1393 libata.dma=4 Compact Flash DMA only
1394 Combinations also work, so libata.dma=3 enables DMA
1395 for disks and CDROMs, but not CFs.
1397 libata.ignore_hpa= [LIBATA] Ignore HPA limit
1398 libata.ignore_hpa=0 keep BIOS limits (default)
1399 libata.ignore_hpa=1 ignore limits, using full disk
1401 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
1405 libata.force= [LIBATA] Force configurations. The format is comma
1406 separated list of "[ID:]VAL" where ID is
1407 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
1408 matching port, link or device. Basically, it matches
1409 the ATA ID string printed on console by libata. If
1410 the whole ID part is omitted, the last PORT and DEVICE
1411 values are used. If ID hasn't been specified yet, the
1412 configuration applies to all ports, links and devices.
1414 If only DEVICE is omitted, the parameter applies to
1415 the port and all links and devices behind it. DEVICE
1416 number of 0 either selects the first device or the
1417 first fan-out link behind PMP device. It does not
1418 select the host link. DEVICE number of 15 selects the
1419 host link and device attached to it.
1421 The VAL specifies the configuration to force. As long
1422 as there's no ambiguity shortcut notation is allowed.
1423 For example, both 1.5 and 1.5G would work for 1.5Gbps.
1424 The following configurations can be forced.
1426 * Cable type: 40c, 80c, short40c, unk, ign or sata.
1427 Any ID with matching PORT is used.
1429 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
1431 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
1432 udma[/][16,25,33,44,66,100,133] notation is also
1435 * [no]ncq: Turn on or off NCQ.
1437 * nohrst, nosrst, norst: suppress hard, soft
1440 * rstonce: only attempt one reset during
1441 hot-unplug link recovery
1443 * dump_id: dump IDENTIFY data.
1445 If there are multiple matching configurations changing
1446 the same attribute, the last one is used.
1448 memblock=debug [KNL] Enable memblock debug messages.
1450 load_ramdisk= [RAM] List of ramdisks to load from floppy
1451 See Documentation/blockdev/ramdisk.txt.
1453 lockd.nlm_grace_period=P [NFS] Assign grace period.
1456 lockd.nlm_tcpport=N [NFS] Assign TCP port.
1459 lockd.nlm_timeout=T [NFS] Assign timeout value.
1462 lockd.nlm_udpport=M [NFS] Assign UDP port.
1465 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
1468 loglevel= All Kernel Messages with a loglevel smaller than the
1469 console loglevel will be printed to the console. It can
1470 also be changed with klogd or other programs. The
1471 loglevels are defined as follows:
1473 0 (KERN_EMERG) system is unusable
1474 1 (KERN_ALERT) action must be taken immediately
1475 2 (KERN_CRIT) critical conditions
1476 3 (KERN_ERR) error conditions
1477 4 (KERN_WARNING) warning conditions
1478 5 (KERN_NOTICE) normal but significant condition
1479 6 (KERN_INFO) informational
1480 7 (KERN_DEBUG) debug-level messages
1482 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
1483 in bytes. n must be a power of two. The default
1484 size is set in the kernel config file.
1486 logo.nologo [FB] Disables display of the built-in Linux logo.
1487 This may be used to provide more screen space for
1488 kernel log messages and is useful when debugging
1489 kernel boot problems.
1491 lp=0 [LP] Specify parallel ports to use, e.g,
1492 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
1493 lp=reset first parallel port). 'lp=0' disables the
1494 lp=auto printer driver. 'lp=reset' (which can be
1495 specified in addition to the ports) causes
1496 attached printers to be reset. Using
1497 lp=port1,port2,... specifies the parallel ports
1498 to associate lp devices with, starting with
1499 lp0. A port specification may be 'none' to skip
1500 that lp device, or a parport name such as
1501 'parport0'. Specifying 'lp=auto' instead of a
1502 port specification list means that device IDs
1503 from each port should be examined, to see if
1504 an IEEE 1284-compliant printer is attached; if
1505 so, the driver will manage that printer.
1506 See also header of drivers/char/lp.c.
1509 Sets loops_per_jiffy to given constant, thus avoiding
1510 time-consuming boot-time autodetection (up to 250 ms per
1511 CPU). 0 enables autodetection (default). To determine
1512 the correct value for your kernel, boot with normal
1513 autodetection and see what value is printed. Note that
1514 on SMP systems the preset will be applied to all CPUs,
1515 which is likely to cause problems if your CPUs need
1516 significantly divergent settings. An incorrect value
1517 will cause delays in the kernel to be wrong, leading to
1518 unpredictable I/O errors and other breakage. Although
1519 unlikely, in the extreme case this might damage your
1523 Format: <io>,<irq>,<dma>
1525 machvec= [IA-64] Force the use of a particular machine-vector
1526 (machvec) in a generic kernel.
1527 Example: machvec=hpzx1_swiotlb
1529 machtype= [Loongson] Share the same kernel image file between different
1531 Example: machtype=lemote-yeeloong-2f-7inch
1533 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
1534 than or equal to this physical address is ignored.
1536 maxcpus= [SMP] Maximum number of processors that an SMP kernel
1537 should make use of. maxcpus=n : n >= 0 limits the
1538 kernel to using 'n' processors. n=0 is a special case,
1539 it is equivalent to "nosmp", which also disables
1542 max_loop= [LOOP] The number of loop block devices that get
1543 (loop.max_loop) unconditionally pre-created at init time. The default
1544 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
1545 of statically allocating a predefined number, loop
1546 devices can be requested on-demand with the
1547 /dev/loop-control interface.
1549 mce [X86-32] Machine Check Exception
1551 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.txt
1553 md= [HW] RAID subsystems devices and level
1554 See Documentation/md.txt.
1557 Format: <first>,<last>
1558 Specifies range of consoles to be captured by the MDA.
1560 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
1561 Amount of memory to be used when the kernel is not able
1562 to see the whole system memory or for test.
1563 [X86] Work as limiting max address. Use together
1564 with memmap= to avoid physical address space collisions.
1565 Without memmap= PCI devices could be placed at addresses
1566 belonging to unused RAM.
1568 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
1572 [KNL,SH] Allow user to override the default size for
1573 per-device physically contiguous DMA buffers.
1575 memmap=exactmap [KNL,X86] Enable setting of an exact
1576 E820 memory map, as specified by the user.
1577 Such memmap=exactmap lines can be constructed based on
1578 BIOS output or other requirements. See the memmap=nn@ss
1581 memmap=nn[KMG]@ss[KMG]
1582 [KNL] Force usage of a specific region of memory
1583 Region of memory to be used, from ss to ss+nn.
1585 memmap=nn[KMG]#ss[KMG]
1586 [KNL,ACPI] Mark specific memory as ACPI data.
1587 Region of memory to be used, from ss to ss+nn.
1589 memmap=nn[KMG]$ss[KMG]
1590 [KNL,ACPI] Mark specific memory as reserved.
1591 Region of memory to be used, from ss to ss+nn.
1592 Example: Exclude memory from 0x18690000-0x1869ffff
1593 memmap=64K$0x18690000
1595 memmap=0x10000$0x18690000
1597 memory_corruption_check=0/1 [X86]
1598 Some BIOSes seem to corrupt the first 64k of
1599 memory when doing things like suspend/resume.
1600 Setting this option will scan the memory
1601 looking for corruption. Enabling this will
1602 both detect corruption and prevent the kernel
1603 from using the memory being corrupted.
1604 However, its intended as a diagnostic tool; if
1605 repeatable BIOS-originated corruption always
1606 affects the same memory, you can use memmap=
1607 to prevent the kernel from using that memory.
1609 memory_corruption_check_size=size [X86]
1610 By default it checks for corruption in the low
1611 64k, making this memory unavailable for normal
1612 use. Use this parameter to scan for
1613 corruption in more or less memory.
1615 memory_corruption_check_period=seconds [X86]
1616 By default it checks for corruption every 60
1617 seconds. Use this parameter to check at some
1618 other rate. 0 disables periodic checking.
1620 memtest= [KNL,X86] Enable memtest
1622 default : 0 <disable>
1623 Specifies the number of memtest passes to be
1624 performed. Each pass selects another test
1625 pattern from a given set of patterns. Memtest
1626 fills the memory with this pattern, validates
1627 memory contents and reserves bad memory
1628 regions that are detected.
1630 meye.*= [HW] Set MotionEye Camera parameters
1631 See Documentation/video4linux/meye.txt.
1633 mfgpt_irq= [IA-32] Specify the IRQ to use for the
1634 Multi-Function General Purpose Timers on AMD Geode
1637 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
1638 the BIOS has incorrectly applied a workaround. TinyBIOS
1639 version 0.98 is known to be affected, 0.99 fixes the
1640 problem by letting the user disable the workaround.
1644 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
1645 physical address is ignored.
1647 mini2440= [ARM,HW,KNL]
1648 Format:[0..2][b][c][t]
1650 MINI2440 configuration specification:
1651 0 - The attached screen is the 3.5" TFT
1652 1 - The attached screen is the 7" TFT
1653 2 - The VGA Shield is attached (1024x768)
1654 Leaving out the screen size parameter will not load
1655 the TFT driver, and the framebuffer will be left
1657 b - Enable backlight. The TFT backlight pin will be
1658 linked to the kernel VESA blanking code and a GPIO
1659 LED. This parameter is not necessary when using the
1661 c - Enable the s3c camera interface.
1662 t - Reserved for enabling touchscreen support. The
1663 touchscreen support is not enabled in the mainstream
1664 kernel as of 2.6.30, a preliminary port can be found
1665 in the "bleeding edge" mini2440 support kernel at
1666 http://repo.or.cz/w/linux-2.6/mini2440.git
1669 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
1670 parameter allows control of the logging verbosity for
1671 the additional memory initialisation checks. A value
1672 of 0 disables mminit logging and a level of 4 will
1673 log everything. Information is printed at KERN_DEBUG
1674 so loglevel=8 may also need to be specified.
1677 [KNL] When CONFIG_MODULE_SIG is set, this means that
1678 modules without (valid) signatures will fail to load.
1679 Note that if CONFIG_MODULE_SIG_FORCE is set, that
1680 is always true, so this option does nothing.
1683 [MOUSE] Maximum time between finger touching and
1684 leaving touchpad surface for touch to be considered
1685 a tap and be reported as a left button click (for
1686 touchpads working in absolute mode only).
1688 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
1689 reporting absolute coordinates, such as tablets
1690 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
1691 reporting absolute coordinates, such as tablets
1693 movablecore=nn[KMG] [KNL,X86,IA-64,PPC] This parameter
1694 is similar to kernelcore except it specifies the
1695 amount of memory used for migratable allocations.
1696 If both kernelcore and movablecore is specified,
1697 then kernelcore will be at *least* the specified
1698 value but may be more. If movablecore on its own
1699 is specified, the administrator must be careful
1700 that the amount of memory usable for all allocations
1703 MTD_Partition= [MTD]
1704 Format: <name>,<region-number>,<size>,<offset>
1706 MTD_Region= [MTD] Format:
1707 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
1710 See drivers/mtd/cmdlinepart.c.
1712 multitce=off [PPC] This parameter disables the use of the pSeries
1713 firmware feature for updating multiple TCE entries
1716 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
1718 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
1720 boundary - index of last SLC block on Flex-OneNAND.
1721 The remaining blocks are configured as MLC blocks.
1722 lock - Configure if Flex-OneNAND boundary should be locked.
1723 Once locked, the boundary cannot be changed.
1724 1 indicates lock status, 0 indicates unlock status.
1727 ARM/S3C2412 JIVE boot control
1729 See arch/arm/mach-s3c2412/mach-jive.c
1731 mtouchusb.raw_coordinates=
1732 [HW] Make the MicroTouch USB driver use raw coordinates
1733 ('y', default) or cooked coordinates ('n')
1735 mtrr_chunk_size=nn[KMG] [X86]
1736 used for mtrr cleanup. It is largest continuous chunk
1737 that could hold holes aka. UC entries.
1739 mtrr_gran_size=nn[KMG] [X86]
1740 Used for mtrr cleanup. It is granularity of mtrr block.
1742 Large value could prevent small alignment from
1745 mtrr_spare_reg_nr=n [X86]
1747 Range: 0,7 : spare reg number
1749 Used for mtrr cleanup. It is spare mtrr entries number.
1750 Set to 2 or more if your graphical card needs more.
1752 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
1754 netdev= [NET] Network devices parameters
1755 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
1756 Note that mem_start is often overloaded to mean
1757 something different and driver-specific.
1758 This usage is only documented in each driver source
1762 [NETFILTER] Enable connection tracking flow accounting
1763 0 to disable accounting
1764 1 to enable accounting
1767 nfsaddrs= [NFS] Deprecated. Use ip= instead.
1768 See Documentation/filesystems/nfs/nfsroot.txt.
1770 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
1771 See Documentation/filesystems/nfs/nfsroot.txt.
1773 nfsrootdebug [NFS] enable nfsroot debugging messages.
1774 See Documentation/filesystems/nfs/nfsroot.txt.
1776 nfs.callback_tcpport=
1777 [NFS] set the TCP port on which the NFSv4 callback
1778 channel should listen.
1781 [NFS] sets the pathname to the program which is used
1782 to update the NFS client cache entries.
1784 nfs.cache_getent_timeout=
1785 [NFS] sets the timeout after which an attempt to
1786 update a cache entry is deemed to have failed.
1788 nfs.idmap_cache_timeout=
1789 [NFS] set the maximum lifetime for idmapper cache
1793 [NFS] enable 64-bit inode numbers.
1794 If zero, the NFS client will fake up a 32-bit inode
1795 number for the readdir() and stat() syscalls instead
1796 of returning the full 64-bit number.
1797 The default is to return 64-bit inode numbers.
1799 nfs.max_session_slots=
1800 [NFSv4.1] Sets the maximum number of session slots
1801 the client will attempt to negotiate with the server.
1802 This limits the number of simultaneous RPC requests
1803 that the client can send to the NFSv4.1 server.
1804 Note that there is little point in setting this
1805 value higher than the max_tcp_slot_table_limit.
1807 nfs.nfs4_disable_idmapping=
1808 [NFSv4] When set to the default of '1', this option
1809 ensures that both the RPC level authentication
1810 scheme and the NFS level operations agree to use
1811 numeric uids/gids if the mount is using the
1812 'sec=sys' security flavour. In effect it is
1813 disabling idmapping, which can make migration from
1814 legacy NFSv2/v3 systems to NFSv4 easier.
1815 Servers that do not support this mode of operation
1816 will be autodetected by the client, and it will fall
1817 back to using the idmapper.
1818 To turn off this behaviour, set the value to '0'.
1820 [NFS4] Specify an additional fixed unique ident-
1821 ification string that NFSv4 clients can insert into
1822 their nfs_client_id4 string. This is typically a
1823 UUID that is generated at system install time.
1825 nfs.send_implementation_id =
1826 [NFSv4.1] Send client implementation identification
1827 information in exchange_id requests.
1828 If zero, no implementation identification information
1830 The default is to send the implementation identification
1833 nfsd.nfs4_disable_idmapping=
1834 [NFSv4] When set to the default of '1', the NFSv4
1835 server will return only numeric uids and gids to
1836 clients using auth_sys, and will accept numeric uids
1837 and gids from such clients. This is intended to ease
1838 migration from NFSv2/v3.
1840 objlayoutdriver.osd_login_prog=
1841 [NFS] [OBJLAYOUT] sets the pathname to the program which
1842 is used to automatically discover and login into new
1843 osd-targets. Please see:
1844 Documentation/filesystems/pnfs.txt for more explanations
1846 nmi_debug= [KNL,AVR32,SH] Specify one or more actions to take
1847 when a NMI is triggered.
1848 Format: [state][,regs][,debounce][,die]
1850 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
1851 Format: [panic,][nopanic,][num]
1853 0 - turn nmi_watchdog off
1854 When panic is specified, panic when an NMI watchdog
1855 timeout occurs (or 'nopanic' to override the opposite
1857 This is useful when you use a panic=... timeout and
1858 need the box quickly up again.
1860 netpoll.carrier_timeout=
1861 [NET] Specifies amount of time (in seconds) that
1862 netpoll should wait for a carrier. By default netpoll
1865 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
1866 emulation library even if a 387 maths coprocessor
1870 [HW] Never suspend the console
1871 Disable suspending of consoles during suspend and
1872 hibernate operations. Once disabled, debugging
1873 messages can reach various consoles while the rest
1874 of the system is being put to sleep (ie, while
1875 debugging driver suspend/resume hooks). This may
1876 not work reliably with all consoles, but is known
1877 to work with serial and VGA consoles.
1878 To facilitate more flexible debugging, we also add
1879 console_suspend, a printk module parameter to control
1880 it. Users could use console_suspend (usually
1881 /sys/module/printk/parameters/console_suspend) to
1882 turn on/off it dynamically.
1884 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
1885 caches in the slab allocator. Saves per-node memory,
1886 but will impact performance.
1890 noapic [SMP,APIC] Tells the kernel to not make use of any
1891 IOAPICs that may be present in the system.
1893 noautogroup Disable scheduler automatic task group creation.
1895 nobats [PPC] Do not use BATs for mapping kernel lowmem
1896 on "Classic" PPC cores.
1900 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
1902 nodelayacct [KNL] Disable per-task delay accounting
1904 nodisconnect [HW,SCSI,M68K] Disables SCSI disconnects.
1906 nodsp [SH] Disable hardware DSP at boot time.
1908 noefi [X86] Disable EFI runtime services support.
1913 On X86-32 available only on PAE configured kernels.
1914 noexec=on: enable non-executable mappings (default)
1915 noexec=off: disable non-executable mappings
1918 Disable SMAP (Supervisor Mode Access Prevention)
1919 even if it is supported by processor.
1922 Disable SMEP (Supervisor Mode Execution Prevention)
1923 even if it is supported by processor.
1926 This affects only 32-bit executables.
1927 noexec32=on: enable non-executable mappings (default)
1928 read doesn't imply executable mappings
1929 noexec32=off: disable non-executable mappings
1930 read implies executable mappings
1932 nofpu [SH] Disable hardware FPU at boot time.
1934 nofxsr [BUGS=X86-32] Disables x86 floating point extended
1935 register save and restore. The kernel will only save
1936 legacy floating-point registers on task switch.
1938 noxsave [BUGS=X86] Disables x86 extended register state save
1939 and restore using xsave. The kernel will fallback to
1940 enabling legacy floating-point and sse state.
1943 on enable eager fpu restore
1944 off disable eager fpu restore
1945 auto selects the default scheme, which automatically
1946 enables eagerfpu restore for xsaveopt.
1948 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
1949 wfi(ARM) instruction doesn't work correctly and not to
1950 use it. This is also useful when using JTAG debugger.
1952 no_file_caps Tells the kernel not to honor file capabilities. The
1953 only way then for a file to be executed with privilege
1954 is to be setuid root or executed by root.
1956 nohalt [IA-64] Tells the kernel not to use the power saving
1957 function PAL_HALT_LIGHT when idle. This increases
1958 power-consumption. On the positive side, it reduces
1959 interrupt wake-up latency, which may improve performance
1960 in certain environments such as networked servers or
1963 nohz= [KNL] Boottime enable/disable dynamic ticks
1964 Valid arguments: on, off
1967 noiotrap [SH] Disables trapped I/O port accesses.
1969 noirqdebug [X86-32] Disables the code which attempts to detect and
1970 disable unhandled interrupt sources.
1972 no_timer_check [X86,APIC] Disables the code which tests for
1973 broken timer IRQ sources.
1975 noisapnp [ISAPNP] Disables ISA PnP code.
1977 noinitrd [RAM] Tells the kernel not to load any configured
1980 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
1982 [Deprecated - use intremap=off]
1986 nojitter [IA-64] Disables jitter checking for ITC timers.
1988 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
1990 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
1993 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting.
1994 steal time is computed, but won't influence scheduler
1997 nolapic [X86-32,APIC] Do not enable or use the local APIC.
1999 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
2001 noltlbs [PPC] Do not use large page/tlb entries for kernel
2002 lowmem mapping on PPC40x.
2004 nomca [IA-64] Disable machine check abort handling
2006 nomce [X86-32] Machine Check Exception
2008 nomfgpt [X86-32] Disable Multi-Function General Purpose
2009 Timer usage (for AMD Geode machines).
2011 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
2012 shutdown the other cpus. Instead use the REBOOT_VECTOR
2015 nomodule Disable module load
2017 nopat [X86] Disable PAT (page attribute table extension of
2018 pagetables) support.
2020 norandmaps Don't use address space randomization. Equivalent to
2021 echo 0 > /proc/sys/kernel/randomize_va_space
2023 noreplace-paravirt [X86,IA-64,PV_OPS] Don't patch paravirt_ops
2025 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
2026 with UP alternatives
2028 nordrand [X86] Disable the direct use of the RDRAND
2029 instruction even if it is supported by the
2030 processor. RDRAND is still available to user
2033 noresume [SWSUSP] Disables resume and restores original swap
2036 no-scroll [VGA] Disables scrollback.
2037 This is required for the Braillex ib80-piezo Braille
2038 reader made by F.H. Papenmeier (Germany).
2042 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
2044 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
2045 and disable the IO APIC. legacy for "maxcpus=0".
2047 nosoftlockup [KNL] Disable the soft-lockup detector.
2049 nosync [HW,M68K] Disables sync negotiation for all devices.
2051 notsc [BUGS=X86-32] Disable Time Stamp Counter
2053 nousb [USB] Disable the USB subsystem
2055 nowatchdog [KNL] Disable the lockup detector (NMI watchdog).
2059 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
2061 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
2062 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
2063 Some features depend on CPU0. Known dependencies are:
2064 1. Resume from suspend/hibernate depends on CPU0.
2065 Suspend/hibernate will fail if CPU0 is offline and you
2066 need to online CPU0 before suspend/hibernate.
2067 2. PIC interrupts also depend on CPU0. CPU0 can't be
2068 removed if a PIC interrupt is detected.
2069 It's said poweroff/reboot may depend on CPU0 on some
2070 machines although I haven't seen such issues so far
2071 after CPU0 is offline on a few tested machines.
2072 If the dependencies are under your control, you can
2073 turn on cpu0_hotplug.
2075 nptcg= [IA-64] Override max number of concurrent global TLB
2076 purges which is reported from either PAL_VM_SUMMARY or
2079 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
2080 could support. nr_cpus=n : n >= 1 limits the kernel to
2081 supporting 'n' processors. Later in runtime you can not
2082 use hotplug cpu feature to put more cpu back to online.
2083 just like you compile the kernel NR_CPUS=n
2085 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
2087 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
2088 Allowed values are enable and disable
2090 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
2091 one of ['zone', 'node', 'default'] can be specified
2092 This can be set from sysctl after boot.
2093 See Documentation/sysctl/vm.txt for details.
2095 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
2096 See Documentation/debugging-via-ohci1394.txt for more
2099 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
2100 Rather than timing out after 20 ms if an EC
2101 command is not properly ACKed, override the length
2102 of the timeout. We have interrupts disabled while
2103 waiting for the ACK, so if this is set too high
2104 interrupts *may* be lost!
2106 omap_mux= [OMAP] Override bootloader pin multiplexing.
2107 Format: <mux_mode0.mode_name=value>...
2108 For example, to override I2C bus2:
2109 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
2111 oprofile.timer= [HW]
2112 Use timer interrupt instead of performance counters
2114 oprofile.cpu_type= Force an oprofile cpu type
2115 This might be useful if you have an older oprofile
2116 userland or if you want common events.
2117 Format: { arch_perfmon }
2118 arch_perfmon: [X86] Force use of architectural
2119 perfmon on Intel CPUs instead of the
2120 CPU specific event set.
2121 timer: [X86] Force use of architectural NMI
2122 timer mode (see also oprofile.timer
2123 for generic hr timer mode)
2124 [s390] Force legacy basic mode sampling
2125 (report cpu_type "timer")
2127 oops=panic Always panic on oopses. Default is to just kill the
2128 process, but there is a small probability of
2129 deadlocking the machine.
2130 This will also cause panics on machine check exceptions.
2131 Useful together with panic=30 to trigger a reboot.
2134 See Documentation/sound/oss/oss-parameters.txt
2136 panic= [KNL] Kernel behaviour on panic: delay <timeout>
2137 timeout > 0: seconds before rebooting
2138 timeout = 0: wait forever
2139 timeout < 0: reboot immediately
2142 parkbd.port= [HW] Parallel port number the keyboard adapter is
2143 connected to, default is 0.
2145 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
2146 0 for XT, 1 for AT (default is AT).
2149 parport= [HW,PPT] Specify parallel ports. 0 disables.
2150 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
2151 Use 'auto' to force the driver to use any
2152 IRQ/DMA settings detected (the default is to
2153 ignore detected IRQ/DMA settings because of
2154 possible conflicts). You can specify the base
2155 address, IRQ, and DMA settings; IRQ and DMA
2156 should be numbers, or 'auto' (for using detected
2157 settings on that particular port), or 'nofifo'
2158 (to avoid using a FIFO even if it is detected).
2159 Parallel ports are assigned in the order they
2160 are specified on the command line, starting
2163 parport_init_mode= [HW,PPT]
2164 Configure VIA parallel port to operate in
2165 a specific mode. This is necessary on Pegasos
2166 computer where firmware has no options for setting
2167 up parallel port mode and sets it to spp.
2168 Currently this function knows 686a and 8231 chips.
2169 Format: [spp|ps2|epp|ecp|ecpepp]
2172 Halt all CPUs after the first oops has been printed for
2173 the specified number of seconds. This is to be used if
2174 your oopses keep scrolling off the screen.
2179 See header of drivers/block/paride/pcd.c.
2180 See also Documentation/blockdev/paride.txt.
2182 pci=option[,option...] [PCI] various PCI subsystem options:
2183 earlydump [X86] dump PCI config space before the kernel
2185 off [X86] don't probe for the PCI bus
2186 bios [X86-32] force use of PCI BIOS, don't access
2187 the hardware directly. Use this if your machine
2188 has a non-standard PCI host bridge.
2189 nobios [X86-32] disallow use of PCI BIOS, only direct
2190 hardware access methods are allowed. Use this
2191 if you experience crashes upon bootup and you
2192 suspect they are caused by the BIOS.
2193 conf1 [X86] Force use of PCI Configuration
2195 conf2 [X86] Force use of PCI Configuration
2197 noaer [PCIE] If the PCIEAER kernel config parameter is
2198 enabled, this kernel boot option can be used to
2199 disable the use of PCIE advanced error reporting.
2200 nodomains [PCI] Disable support for multiple PCI
2201 root domains (aka PCI segments, in ACPI-speak).
2202 nommconf [X86] Disable use of MMCONFIG for PCI
2204 check_enable_amd_mmconf [X86] check for and enable
2205 properly configured MMIO access to PCI
2206 config space on AMD family 10h CPU
2207 nomsi [MSI] If the PCI_MSI kernel config parameter is
2208 enabled, this kernel boot option can be used to
2209 disable the use of MSI interrupts system-wide.
2210 noioapicquirk [APIC] Disable all boot interrupt quirks.
2211 Safety option to keep boot IRQs enabled. This
2212 should never be necessary.
2213 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
2214 primary IO-APIC for bridges that cannot disable
2215 boot IRQs. This fixes a source of spurious IRQs
2216 when the system masks IRQs.
2217 noioapicreroute [APIC] Disable workaround that uses the
2218 boot IRQ equivalent of an IRQ that connects to
2219 a chipset where boot IRQs cannot be disabled.
2220 The opposite of ioapicreroute.
2221 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
2222 routing table. These calls are known to be buggy
2223 on several machines and they hang the machine
2224 when used, but on other computers it's the only
2225 way to get the interrupt routing table. Try
2226 this option if the kernel is unable to allocate
2227 IRQs or discover secondary PCI buses on your
2229 rom [X86] Assign address space to expansion ROMs.
2230 Use with caution as certain devices share
2231 address decoders between ROMs and other
2233 norom [X86] Do not assign address space to
2234 expansion ROMs that do not already have
2235 BIOS assigned address ranges.
2236 nobar [X86] Do not assign address space to the
2237 BARs that weren't assigned by the BIOS.
2238 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
2239 assigned automatically to PCI devices. You can
2240 make the kernel exclude IRQs of your ISA cards
2242 pirqaddr=0xAAAAA [X86] Specify the physical address
2243 of the PIRQ table (normally generated
2244 by the BIOS) if it is outside the
2245 F0000h-100000h range.
2246 lastbus=N [X86] Scan all buses thru bus #N. Can be
2247 useful if the kernel is unable to find your
2248 secondary buses and you want to tell it
2249 explicitly which ones they are.
2250 assign-busses [X86] Always assign all PCI bus
2251 numbers ourselves, overriding
2252 whatever the firmware may have done.
2253 usepirqmask [X86] Honor the possible IRQ mask stored
2254 in the BIOS $PIR table. This is needed on
2255 some systems with broken BIOSes, notably
2256 some HP Pavilion N5400 and Omnibook XE3
2257 notebooks. This will have no effect if ACPI
2258 IRQ routing is enabled.
2259 noacpi [X86] Do not use ACPI for IRQ routing
2260 or for PCI scanning.
2261 use_crs [X86] Use PCI host bridge window information
2262 from ACPI. On BIOSes from 2008 or later, this
2263 is enabled by default. If you need to use this,
2264 please report a bug.
2265 nocrs [X86] Ignore PCI host bridge windows from ACPI.
2266 If you need to use this, please report a bug.
2267 routeirq Do IRQ routing for all PCI devices.
2268 This is normally done in pci_enable_device(),
2269 so this option is a temporary workaround
2270 for broken drivers that don't call it.
2271 skip_isa_align [X86] do not align io start addr, so can
2272 handle more pci cards
2273 firmware [ARM] Do not re-enumerate the bus but instead
2274 just use the configuration from the
2275 bootloader. This is currently used on
2276 IXP2000 systems where the bus has to be
2277 configured a certain way for adjunct CPUs.
2278 noearly [X86] Don't do any early type 1 scanning.
2279 This might help on some broken boards which
2280 machine check when some devices' config space
2281 is read. But various workarounds are disabled
2282 and some IOMMU drivers will not work.
2283 bfsort Sort PCI devices into breadth-first order.
2284 This sorting is done to get a device
2285 order compatible with older (<= 2.4) kernels.
2286 nobfsort Don't sort PCI devices into breadth-first order.
2287 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
2288 tuning and use the BIOS-configured MPS defaults.
2289 pcie_bus_safe Set every device's MPS to the largest value
2290 supported by all devices below the root complex.
2291 pcie_bus_perf Set device MPS to the largest allowable MPS
2292 based on its parent bus. Also set MRRS (Max
2293 Read Request Size) to the largest supported
2294 value (no larger than the MPS that the device
2295 or bus can support) for best performance.
2296 pcie_bus_peer2peer Set every device's MPS to 128B, which
2297 every device is guaranteed to support. This
2298 configuration allows peer-to-peer DMA between
2299 any pair of devices, possibly at the cost of
2300 reduced performance. This also guarantees
2301 that hot-added devices will work.
2302 cbiosize=nn[KMG] The fixed amount of bus space which is
2303 reserved for the CardBus bridge's IO window.
2304 The default value is 256 bytes.
2305 cbmemsize=nn[KMG] The fixed amount of bus space which is
2306 reserved for the CardBus bridge's memory
2307 window. The default value is 64 megabytes.
2310 [<order of align>@][<domain>:]<bus>:<slot>.<func>[; ...]
2311 Specifies alignment and device to reassign
2312 aligned memory resources.
2313 If <order of align> is not specified,
2314 PAGE_SIZE is used as alignment.
2315 PCI-PCI bridge can be specified, if resource
2316 windows need to be expanded.
2317 ecrc= Enable/disable PCIe ECRC (transaction layer
2318 end-to-end CRC checking).
2319 bios: Use BIOS/firmware settings. This is the
2323 hpiosize=nn[KMG] The fixed amount of bus space which is
2324 reserved for hotplug bridge's IO window.
2325 Default size is 256 bytes.
2326 hpmemsize=nn[KMG] The fixed amount of bus space which is
2327 reserved for hotplug bridge's memory window.
2328 Default size is 2 megabytes.
2329 realloc= Enable/disable reallocating PCI bridge resources
2330 if allocations done by BIOS are too small to
2331 accommodate resources required by all child
2333 off: Turn realloc off
2335 realloc same as realloc=on
2336 noari do not use PCIe ARI.
2337 pcie_scan_all Scan all possible PCIe devices. Otherwise we
2338 only look for one device below a PCIe downstream
2341 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
2344 force Enable ASPM even on devices that claim not to support it.
2345 WARNING: Forcing ASPM on may cause system lockups.
2347 pcie_hp= [PCIE] PCI Express Hotplug driver options:
2348 nomsi Do not use MSI for PCI Express Native Hotplug (this
2349 makes all PCIe ports use INTx for hotplug services).
2351 pcie_ports= [PCIE] PCIe ports handling:
2352 auto Ask the BIOS whether or not to use native PCIe services
2353 associated with PCIe ports (PME, hot-plug, AER). Use
2354 them only if that is allowed by the BIOS.
2355 native Use native PCIe services associated with PCIe ports
2357 compat Treat PCIe ports as PCI-to-PCI bridges, disable the PCIe
2360 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
2361 nomsi Do not use MSI for native PCIe PME signaling (this makes
2362 all PCIe root ports use INTx for all services).
2364 pcmv= [HW,PCMCIA] BadgePAD 4
2367 See Documentation/blockdev/paride.txt.
2369 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
2372 See arch/parisc/kernel/pdc_chassis.c
2374 percpu_alloc= Select which percpu first chunk allocator to use.
2375 Currently supported values are "embed" and "page".
2376 Archs may support subset or none of the selections.
2377 See comments in mm/percpu.c for details on each
2378 allocator. This parameter is primarily for debugging
2379 and performance comparison.
2382 See Documentation/blockdev/paride.txt.
2385 See Documentation/blockdev/paride.txt.
2387 pirq= [SMP,APIC] Manual mp-table setup
2388 See Documentation/x86/i386/IO-APIC.txt.
2390 plip= [PPT,NET] Parallel port network link
2391 Format: { parport<nr> | timid | 0 }
2392 See also Documentation/parport.txt.
2394 pmtmr= [X86] Manual setup of pmtmr I/O Port.
2395 Override pmtimer IOPort with a hex value.
2399 Enable PNP debug messages (depends on the
2400 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
2401 via /sys/module/pnp/parameters/debug. We always show
2402 current resource usage; turning this on also shows
2403 possible settings and some assignment information.
2409 { on | off | curr | res | no-curr | no-res }
2412 [ISAPNP] Exclude IRQs for the autoconfiguration
2415 [ISAPNP] Exclude DMAs for the autoconfiguration
2417 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
2418 Ranges are in pairs (I/O port base and size).
2421 [ISAPNP] Exclude memory regions for the
2423 Ranges are in pairs (memory base and size).
2425 ports= [IP_VS_FTP] IPVS ftp helper module
2427 Up to 8 (IP_VS_APP_MAX_PORTS) ports
2429 Format: <port>,<port>....
2431 print-fatal-signals=
2432 [KNL] debug: print fatal signals
2434 If enabled, warn about various signal handling
2435 related application anomalies: too many signals,
2436 too many POSIX.1 timers, fatal signals causing a
2439 If you hit the warning due to signal overflow,
2440 you might want to try "ulimit -i unlimited".
2444 printk.always_kmsg_dump=
2445 Trigger kmsg_dump for cases other than kernel oops or
2447 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
2450 printk.time= Show timing data prefixed to each printk message line
2451 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
2453 processor.max_cstate= [HW,ACPI]
2454 Limit processor to maximum C-state
2455 max_cstate=9 overrides any DMI blacklist limit.
2457 processor.nocst [HW,ACPI]
2458 Ignore the _CST method to determine C-states,
2459 instead using the legacy FADT method
2461 profile= [KNL] Enable kernel profiling via /proc/profile
2462 Format: [schedule,]<number>
2463 Param: "schedule" - profile schedule points.
2464 Param: <number> - step/bucket size as a power of 2 for
2465 statistical time based profiling.
2466 Param: "sleep" - profile D-state sleeping (millisecs).
2467 Requires CONFIG_SCHEDSTATS
2468 Param: "kvm" - profile VM exits.
2470 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
2472 See Documentation/blockdev/ramdisk.txt.
2474 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
2475 probe for; one of (bare|imps|exps|lifebook|any).
2476 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
2478 psmouse.resetafter= [HW,MOUSE]
2479 Try to reset the device after so many bad packets
2482 [HW,MOUSE] Set desired mouse resolution, in dpi.
2483 psmouse.smartscroll=
2484 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
2485 0 = disabled, 1 = enabled (default).
2487 pstore.backend= Specify the name of the pstore backend to use
2490 See Documentation/blockdev/paride.txt.
2493 [KNL] Number of legacy pty's. Overwrites compiled-in
2496 quiet [KNL] Disable most log messages
2501 See Documentation/md.txt.
2503 ramdisk_blocksize= [RAM]
2504 See Documentation/blockdev/ramdisk.txt.
2506 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
2507 See Documentation/blockdev/ramdisk.txt.
2509 rcu_nocbs= [KNL,BOOT]
2510 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
2511 the specified list of CPUs to be no-callback CPUs.
2512 Invocation of these CPUs' RCU callbacks will
2513 be offloaded to "rcuox/N" kthreads created for
2514 that purpose, where "x" is "b" for RCU-bh, "p"
2515 for RCU-preempt, and "s" for RCU-sched, and "N"
2516 is the CPU number. This reduces OS jitter on the
2517 offloaded CPUs, which can be useful for HPC and
2519 real-time workloads. It can also improve energy
2520 efficiency for asymmetric multiprocessors.
2522 rcu_nocb_poll [KNL,BOOT]
2523 Rather than requiring that offloaded CPUs
2524 (specified by rcu_nocbs= above) explicitly
2525 awaken the corresponding "rcuoN" kthreads,
2526 make these kthreads poll for callbacks.
2527 This improves the real-time response for the
2528 offloaded CPUs by relieving them of the need to
2529 wake up the corresponding kthread, but degrades
2530 energy efficiency by requiring that the kthreads
2531 periodically wake up to do the polling.
2533 rcutree.blimit= [KNL,BOOT]
2534 Set maximum number of finished RCU callbacks to process
2537 rcutree.fanout_leaf= [KNL,BOOT]
2538 Increase the number of CPUs assigned to each
2539 leaf rcu_node structure. Useful for very large
2542 rcutree.jiffies_till_first_fqs= [KNL,BOOT]
2543 Set delay from grace-period initialization to
2544 first attempt to force quiescent states.
2545 Units are jiffies, minimum value is zero,
2546 and maximum value is HZ.
2548 rcutree.jiffies_till_next_fqs= [KNL,BOOT]
2549 Set delay between subsequent attempts to force
2550 quiescent states. Units are jiffies, minimum
2551 value is one, and maximum value is HZ.
2553 rcutree.qhimark= [KNL,BOOT]
2554 Set threshold of queued
2555 RCU callbacks over which batch limiting is disabled.
2557 rcutree.qlowmark= [KNL,BOOT]
2558 Set threshold of queued RCU callbacks below which
2559 batch limiting is re-enabled.
2561 rcutree.rcu_cpu_stall_suppress= [KNL,BOOT]
2562 Suppress RCU CPU stall warning messages.
2564 rcutree.rcu_cpu_stall_timeout= [KNL,BOOT]
2565 Set timeout for RCU CPU stall warning messages.
2567 rcutree.rcu_idle_gp_delay= [KNL,BOOT]
2568 Set wakeup interval for idle CPUs that have
2569 RCU callbacks (RCU_FAST_NO_HZ=y).
2571 rcutree.rcu_idle_lazy_gp_delay= [KNL,BOOT]
2572 Set wakeup interval for idle CPUs that have
2573 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
2574 Lazy RCU callbacks are those which RCU can
2575 prove do nothing more than free memory.
2577 rcutorture.fqs_duration= [KNL,BOOT]
2578 Set duration of force_quiescent_state bursts.
2580 rcutorture.fqs_holdoff= [KNL,BOOT]
2581 Set holdoff time within force_quiescent_state bursts.
2583 rcutorture.fqs_stutter= [KNL,BOOT]
2584 Set wait time between force_quiescent_state bursts.
2586 rcutorture.irqreader= [KNL,BOOT]
2587 Test RCU readers from irq handlers.
2589 rcutorture.n_barrier_cbs= [KNL,BOOT]
2590 Set callbacks/threads for rcu_barrier() testing.
2592 rcutorture.nfakewriters= [KNL,BOOT]
2593 Set number of concurrent RCU writers. These just
2594 stress RCU, they don't participate in the actual
2595 test, hence the "fake".
2597 rcutorture.nreaders= [KNL,BOOT]
2598 Set number of RCU readers.
2600 rcutorture.onoff_holdoff= [KNL,BOOT]
2601 Set time (s) after boot for CPU-hotplug testing.
2603 rcutorture.onoff_interval= [KNL,BOOT]
2604 Set time (s) between CPU-hotplug operations, or
2605 zero to disable CPU-hotplug testing.
2607 rcutorture.shuffle_interval= [KNL,BOOT]
2608 Set task-shuffle interval (s). Shuffling tasks
2609 allows some CPUs to go into dyntick-idle mode
2610 during the rcutorture test.
2612 rcutorture.shutdown_secs= [KNL,BOOT]
2613 Set time (s) after boot system shutdown. This
2614 is useful for hands-off automated testing.
2616 rcutorture.stall_cpu= [KNL,BOOT]
2617 Duration of CPU stall (s) to test RCU CPU stall
2618 warnings, zero to disable.
2620 rcutorture.stall_cpu_holdoff= [KNL,BOOT]
2621 Time to wait (s) after boot before inducing stall.
2623 rcutorture.stat_interval= [KNL,BOOT]
2624 Time (s) between statistics printk()s.
2626 rcutorture.stutter= [KNL,BOOT]
2627 Time (s) to stutter testing, for example, specifying
2628 five seconds causes the test to run for five seconds,
2629 wait for five seconds, and so on. This tests RCU's
2630 ability to transition abruptly to and from idle.
2632 rcutorture.test_boost= [KNL,BOOT]
2633 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
2634 "Maybe" means test if the RCU implementation
2635 under test support RCU priority boosting.
2637 rcutorture.test_boost_duration= [KNL,BOOT]
2638 Duration (s) of each individual boost test.
2640 rcutorture.test_boost_interval= [KNL,BOOT]
2641 Interval (s) between each boost test.
2643 rcutorture.test_no_idle_hz= [KNL,BOOT]
2644 Test RCU's dyntick-idle handling. See also the
2645 rcutorture.shuffle_interval parameter.
2647 rcutorture.torture_type= [KNL,BOOT]
2648 Specify the RCU implementation to test.
2650 rcutorture.verbose= [KNL,BOOT]
2651 Enable additional printk() statements.
2655 Run specified binary instead of /init from the ramdisk,
2656 used for early userspace startup. See initrd.
2658 reboot= [BUGS=X86-32,BUGS=ARM,BUGS=IA-64] Rebooting mode
2659 Format: <reboot_mode>[,<reboot_mode2>[,...]]
2660 See arch/*/kernel/reboot.c or arch/*/kernel/process.c
2663 [KNL, SMP] Set scheduler's default relax_domain_level.
2664 See Documentation/cgroups/cpusets.txt.
2666 reserve= [KNL,BUGS] Force the kernel to ignore some iomem area
2668 reservetop= [X86-32]
2670 Reserves a hole at the top of the kernel virtual
2675 Set the amount of memory to reserve for BIOS at
2676 the bottom of the address space.
2678 reset_devices [KNL] Force drivers to reset the underlying device
2679 during initialization.
2682 Specify the partition device for software suspend
2684 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
2686 resume_offset= [SWSUSP]
2687 Specify the offset from the beginning of the partition
2688 given by "resume=" at which the swap header is located,
2689 in <PAGE_SIZE> units (needed only for swap files).
2690 See Documentation/power/swsusp-and-swap-files.txt
2692 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
2693 read the resume files
2695 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
2696 Useful for devices that are detected asynchronously
2697 (e.g. USB and MMC devices).
2699 hibernate= [HIBERNATION]
2700 noresume Don't check if there's a hibernation image
2701 present during boot.
2702 nocompress Don't compress/decompress hibernation images.
2704 retain_initrd [RAM] Keep initrd memory after extraction
2706 rhash_entries= [KNL,NET]
2707 Set number of hash buckets for route cache
2709 riscom8= [HW,SERIAL]
2710 Format: <io_board1>[,<io_board2>[,...<io_boardN>]]
2712 ro [KNL] Mount root device read-only on boot
2714 root= [KNL] Root filesystem
2715 See name_to_dev_t comment in init/do_mounts.c.
2717 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
2718 mount the root filesystem
2720 rootflags= [KNL] Set root filesystem mount option string
2722 rootfstype= [KNL] Set root filesystem type
2724 rootwait [KNL] Wait (indefinitely) for root device to show up.
2725 Useful for devices that are detected asynchronously
2726 (e.g. USB and MMC devices).
2728 rproc_mem=nn[KMG][@address]
2729 [KNL,ARM,CMA] Remoteproc physical memory block.
2730 Memory area to be used by remote processor image,
2733 rw [KNL] Mount root device read-write on boot
2735 S [KNL] Run init in single mode
2738 See drivers/net/irda/sa1100_ir.c.
2740 sbni= [NET] Granch SBNI12 leased line adapter
2742 sched_debug [KNL] Enables verbose scheduler debug messages.
2744 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
2745 xtime_lock contention on larger systems, and/or RCU lock
2746 contention on all systems with CONFIG_MAXSMP set.
2747 Format: { "0" | "1" }
2748 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
2750 Note: increases power consumption, thus should only be
2751 enabled if running jitter sensitive (HPC/RT) workloads.
2753 security= [SECURITY] Choose a security module to enable at boot.
2754 If this boot parameter is not specified, only the first
2755 security module asking for security registration will be
2756 loaded. An invalid security module name will be treated
2757 as if no module has been chosen.
2759 selinux= [SELINUX] Disable or enable SELinux at boot time.
2760 Format: { "0" | "1" }
2761 See security/selinux/Kconfig help text.
2764 Default value is set via kernel config option.
2765 If enabled at boot time, /selinux/disable can be used
2766 later to disable prior to initial policy load.
2768 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
2769 Format: { "0" | "1" }
2770 See security/apparmor/Kconfig help text
2773 Default value is set via kernel config option.
2775 serialnumber [BUGS=X86-32]
2778 Maximal number of shapers.
2780 show_msr= [x86] show boot-time MSR settings
2781 Format: { <integer> }
2782 Show boot-time (BIOS-initialized) MSR settings.
2783 The parameter means the number of CPUs to show,
2784 for example 1 means boot CPU only.
2791 slab_max_order= [MM, SLAB]
2792 Determines the maximum allowed order for slabs.
2793 A high setting may cause OOMs due to memory
2794 fragmentation. Defaults to 1 for systems with
2795 more than 32MB of RAM, 0 otherwise.
2797 slub_debug[=options[,slabs]] [MM, SLUB]
2798 Enabling slub_debug allows one to determine the
2799 culprit if slab objects become corrupted. Enabling
2800 slub_debug can create guard zones around objects and
2801 may poison objects when not in use. Also tracks the
2802 last alloc / free. For more information see
2803 Documentation/vm/slub.txt.
2805 slub_max_order= [MM, SLUB]
2806 Determines the maximum allowed order for slabs.
2807 A high setting may cause OOMs due to memory
2808 fragmentation. For more information see
2809 Documentation/vm/slub.txt.
2811 slub_min_objects= [MM, SLUB]
2812 The minimum number of objects per slab. SLUB will
2813 increase the slab order up to slub_max_order to
2814 generate a sufficiently large slab able to contain
2815 the number of objects indicated. The higher the number
2816 of objects the smaller the overhead of tracking slabs
2817 and the less frequently locks need to be acquired.
2818 For more information see Documentation/vm/slub.txt.
2820 slub_min_order= [MM, SLUB]
2821 Determines the minimum page order for slabs. Must be
2822 lower than slub_max_order.
2823 For more information see Documentation/vm/slub.txt.
2825 slub_nomerge [MM, SLUB]
2826 Disable merging of slabs with similar size. May be
2827 necessary if there is some reason to distinguish
2828 allocs to different slabs. Debug options disable
2829 merging on their own.
2830 For more information see Documentation/vm/slub.txt.
2833 Format: <io1>[,<io2>[,...,<io8>]]
2835 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
2836 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
2837 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
2838 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
2839 smsc-ircc2.ircc_irq= [HW] IRQ line
2840 smsc-ircc2.ircc_dma= [HW] DMA channel
2841 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
2842 0: Toshiba Satellite 1800 (GP data pin select)
2843 1: Fast pin select (default)
2847 [KNL] Should the soft-lockup detector generate panics.
2850 sonypi.*= [HW] Sony Programmable I/O Control Device driver
2851 See Documentation/laptops/sonypi.txt
2853 specialix= [HW,SERIAL] Specialix multi-serial port adapter
2854 See Documentation/serial/specialix.txt.
2856 spia_io_base= [HW,MTD]
2862 Enabled the stack tracer on boot up.
2864 stacktrace_filter=[function-list]
2865 [FTRACE] Limit the functions that the stack tracer
2866 will trace at boot up. function-list is a comma separated
2867 list of functions. This list can be changed at run
2868 time by the stack_trace_filter file in the debugfs
2869 tracing directory. Note, this enables stack tracing
2870 and the stacktrace above is not needed.
2874 Set the STI (builtin display/keyboard on the HP-PARISC
2875 machines) console (graphic card) which should be used
2876 as the initial boot-console.
2877 See also comment in drivers/video/console/sticore.c.
2880 See comment in drivers/video/console/sticore.c.
2883 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
2885 sunrpc.min_resvport=
2886 sunrpc.max_resvport=
2888 SunRPC servers often require that client requests
2889 originate from a privileged port (i.e. a port in the
2890 range 0 < portnr < 1024).
2891 An administrator who wishes to reserve some of these
2892 ports for other uses may adjust the range that the
2893 kernel's sunrpc client considers to be privileged
2894 using these two parameters to set the minimum and
2895 maximum port values.
2899 Control how the NFS server code allocates CPUs to
2900 service thread pools. Depending on how many NICs
2901 you have and where their interrupts are bound, this
2902 option will affect which CPUs will do NFS serving.
2903 Note: this parameter cannot be changed while the
2904 NFS server is running.
2906 auto the server chooses an appropriate mode
2907 automatically using heuristics
2908 global a single global pool contains all CPUs
2909 percpu one pool for each CPU
2910 pernode one pool for each NUMA node (equivalent
2911 to global on non-NUMA machines)
2913 sunrpc.tcp_slot_table_entries=
2914 sunrpc.udp_slot_table_entries=
2916 Sets the upper limit on the number of simultaneous
2917 RPC calls that can be sent from the client to a
2918 server. Increasing these values may allow you to
2919 improve throughput, but will also increase the
2920 amount of memory reserved for use by the client.
2923 [KNL] Enable accounting of swap in memory resource
2924 controller if no parameter or 1 is given or disable
2925 it if 0 is given (See Documentation/cgroups/memory.txt)
2927 swiotlb= [IA-64] Number of I/O TLB slabs
2931 sysfs.deprecated=0|1 [KNL]
2932 Enable/disable old style sysfs layout for old udev
2933 on older distributions. When this option is enabled
2934 very new udev will not work anymore. When this option
2935 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
2936 in older udev will not work anymore.
2937 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
2938 the kernel configuration.
2940 sysrq_always_enabled
2942 Ignore sysrq setting - this boot parameter will
2943 neutralize any effect of /proc/sys/kernel/sysrq.
2944 Useful for debugging.
2948 test_suspend= [SUSPEND]
2949 Specify "mem" (for Suspend-to-RAM) or "standby" (for
2950 standby suspend) as the system sleep state to briefly
2951 enter during system startup. The system is woken from
2952 this state using a wakeup-capable RTC alarm.
2954 thash_entries= [KNL,NET]
2955 Set number of hash buckets for TCP connection
2957 thermal.act= [HW,ACPI]
2958 -1: disable all active trip points in all thermal zones
2959 <degrees C>: override all lowest active trip points
2961 thermal.crt= [HW,ACPI]
2962 -1: disable all critical trip points in all thermal zones
2963 <degrees C>: override all critical trip points
2965 thermal.nocrt= [HW,ACPI]
2966 Set to disable actions on ACPI thermal zone
2967 critical and hot trip points.
2969 thermal.off= [HW,ACPI]
2970 1: disable ACPI thermal control
2972 thermal.psv= [HW,ACPI]
2973 -1: disable all passive trip points
2974 <degrees C>: override all passive trip points to this
2977 thermal.tzp= [HW,ACPI]
2978 Specify global default ACPI thermal zone polling rate
2979 <deci-seconds>: poll all this frequency
2980 0: no polling (default)
2983 Force threading of all interrupt handlers except those
2984 marked explicitly IRQF_NO_THREAD.
2988 Specify if the kernel should make use of the cpu
2989 topology information if the hardware supports this.
2990 The scheduler will make use of this information and
2991 e.g. base its process migration decisions on it.
2996 tpm_suspend_pcr=[HW,TPM]
2997 Format: integer pcr id
2998 Specify that at suspend time, the tpm driver
2999 should extend the specified pcr with zeros,
3000 as a workaround for some chips which fail to
3001 flush the last written pcr on TPM_SaveState.
3002 This will guarantee that all the other pcrs
3005 trace_buf_size=nn[KMG]
3006 [FTRACE] will set tracing buffer size.
3008 trace_event=[event-list]
3009 [FTRACE] Set and start specified trace events in order
3010 to facilitate early boot debugging.
3011 See also Documentation/trace/events.txt
3013 trace_options=[option-list]
3014 [FTRACE] Enable or disable tracer options at boot.
3015 The option-list is a comma delimited list of options
3016 that can be enabled or disabled just as if you were
3017 to echo the option name into
3019 /sys/kernel/debug/tracing/trace_options
3021 For example, to enable stacktrace option (to dump the
3022 stack trace of each event), add to the command line:
3024 trace_options=stacktrace
3026 See also Documentation/trace/ftrace.txt "trace options"
3029 transparent_hugepage=
3031 Format: [always|madvise|never]
3032 Can be used to control the default behavior of the system
3033 with respect to transparent hugepages.
3034 See Documentation/vm/transhuge.txt for more details.
3036 tsc= Disable clocksource stability checks for TSC.
3038 [x86] reliable: mark tsc clocksource as reliable, this
3039 disables clocksource verification at runtime, as well
3040 as the stability checks done at bootup. Used to enable
3041 high-resolution timer mode on older hardware, and in
3042 virtualized environment.
3043 [x86] noirqtime: Do not use TSC to do irq accounting.
3044 Used to run time disable IRQ_TIME_ACCOUNTING on any
3045 platforms where RDTSC is slow and this accounting
3048 turbografx.map[2|3]= [HW,JOY]
3049 TurboGraFX parallel port interface
3051 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
3052 See also Documentation/input/joystick-parport.txt
3054 udbg-immortal [PPC] When debugging early kernel crashes that
3055 happen after console_init() and before a proper
3056 console driver takes over, this boot options might
3057 help "seeing" what's going on.
3059 uhash_entries= [KNL,NET]
3060 Set number of hash buckets for UDP/UDP-Lite connections
3063 [USB] Ignore overcurrent events (default N).
3064 Some badly-designed motherboards generate lots of
3065 bogus events, for ports that aren't wired to
3066 anything. Set this parameter to avoid log spamming.
3067 Note that genuine overcurrent events won't be
3071 [X86] Cause panic on unknown NMI.
3073 usbcore.authorized_default=
3074 [USB] Default USB device authorization:
3075 (default -1 = authorized except for wireless USB,
3076 0 = not authorized, 1 = authorized)
3078 usbcore.autosuspend=
3079 [USB] The autosuspend time delay (in seconds) used
3080 for newly-detected USB devices (default 2). This
3081 is the time required before an idle device will be
3082 autosuspended. Devices for which the delay is set
3083 to a negative value won't be autosuspended at all.
3085 usbcore.usbfs_snoop=
3086 [USB] Set to log all usbfs traffic (default 0 = off).
3088 usbcore.blinkenlights=
3089 [USB] Set to cycle leds on hubs (default 0 = off).
3091 usbcore.old_scheme_first=
3092 [USB] Start with the old device initialization
3093 scheme (default 0 = off).
3095 usbcore.usbfs_memory_mb=
3096 [USB] Memory limit (in MB) for buffers allocated by
3097 usbfs (default = 16, 0 = max = 2047).
3099 usbcore.use_both_schemes=
3100 [USB] Try the other device initialization scheme
3101 if the first one fails (default 1 = enabled).
3103 usbcore.initial_descriptor_timeout=
3104 [USB] Specifies timeout for the initial 64-byte
3105 USB_REQ_GET_DESCRIPTOR request in milliseconds
3106 (default 5000 = 5.0 seconds).
3109 [USBHID] The interval which mice are to be polled at.
3111 usb-storage.delay_use=
3112 [UMS] The delay in seconds before a new device is
3113 scanned for Logical Units (default 5).
3116 [UMS] A list of quirks entries to supplement or
3117 override the built-in unusual_devs list. List
3118 entries are separated by commas. Each entry has
3119 the form VID:PID:Flags where VID and PID are Vendor
3120 and Product ID values (4-digit hex numbers) and
3121 Flags is a set of characters, each corresponding
3122 to a common usb-storage quirk flag as follows:
3123 a = SANE_SENSE (collect more than 18 bytes
3125 b = BAD_SENSE (don't collect more than 18
3126 bytes of sense data);
3127 c = FIX_CAPACITY (decrease the reported
3128 device capacity by one sector);
3129 d = NO_READ_DISC_INFO (don't use
3130 READ_DISC_INFO command);
3131 e = NO_READ_CAPACITY_16 (don't use
3132 READ_CAPACITY_16 command);
3133 h = CAPACITY_HEURISTICS (decrease the
3134 reported device capacity by one
3135 sector if the number is odd);
3136 i = IGNORE_DEVICE (don't bind to this
3138 l = NOT_LOCKABLE (don't try to lock and
3139 unlock ejectable media);
3140 m = MAX_SECTORS_64 (don't transfer more
3141 than 64 sectors = 32 KB at a time);
3142 n = INITIAL_READ10 (force a retry of the
3143 initial READ(10) command);
3144 o = CAPACITY_OK (accept the capacity
3145 reported by the device);
3146 p = WRITE_CACHE (the device cache is ON
3148 r = IGNORE_RESIDUE (the device reports
3149 bogus residue values);
3150 s = SINGLE_LUN (the device has only one
3152 w = NO_WP_DETECT (don't test whether the
3153 medium is write-protected).
3154 Example: quirks=0419:aaf5:rl,0421:0433:rc
3156 user_debug= [KNL,ARM]
3158 See arch/arm/Kconfig.debug help text.
3159 1 - undefined instruction events
3161 4 - invalid data aborts
3164 Example: user_debug=31
3167 [X86] Flags controlling user PTE allocations.
3169 nohigh = do not allocate PTE pages in
3170 HIGHMEM regardless of setting
3174 vdso=2: enable compat VDSO (default with COMPAT_VDSO)
3175 vdso=1: enable VDSO (default)
3176 vdso=0: disable VDSO mapping
3179 vdso32=2: enable compat VDSO (default with COMPAT_VDSO)
3180 vdso32=1: enable 32-bit VDSO (default)
3181 vdso32=0: disable 32-bit VDSO mapping
3184 vector=percpu: enable percpu vector domain
3186 video= [FB] Frame buffer configuration
3187 See Documentation/fb/modedb.txt.
3190 [VMMIO] Memory mapped virtio (platform) device.
3192 <size>@<baseaddr>:<irq>[:<id>]
3194 <size> := size (can use standard suffixes
3196 <baseaddr> := physical base address
3197 <irq> := interrupt number (as passed to
3199 <id> := (optional) platform device id
3201 virtio_mmio.device=1K@0x100b0000:48:7
3203 Can be used multiple times for multiple devices.
3205 vga= [BOOT,X86-32] Select a particular video mode
3206 See Documentation/x86/boot.txt and
3207 Documentation/svga.txt.
3208 Use vga=ask for menu.
3209 This is actually a boot loader parameter; the value is
3210 passed to the kernel using a special protocol.
3212 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
3213 size of <nn>. This can be used to increase the
3214 minimum size (128MB on x86). It can also be used to
3215 decrease the size and leave more room for directly
3218 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
3221 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
3224 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
3228 Controls the behavior of vsyscalls (i.e. calls to
3229 fixed addresses of 0xffffffffff600x00 from legacy
3230 code). Most statically-linked binaries and older
3231 versions of glibc use these calls. Because these
3232 functions are at fixed addresses, they make nice
3233 targets for exploits that can control RIP.
3235 emulate [default] Vsyscalls turn into traps and are
3236 emulated reasonably safely.
3238 native Vsyscalls are native syscall instructions.
3239 This is a little bit faster than trapping
3240 and makes a few dynamic recompilers work
3241 better than they would in emulation mode.
3242 It also makes exploits much easier to write.
3244 none Vsyscalls don't work at all. This makes
3245 them quite hard to use for exploits but
3246 might break your system.
3248 vt.cur_default= [VT] Default cursor shape.
3249 Format: 0xCCBBAA, where AA, BB, and CC are the same as
3250 the parameters of the <Esc>[?A;B;Cc escape sequence;
3251 see VGA-softcursor.txt. Default: 2 = underline.
3253 vt.default_blu= [VT]
3254 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
3255 Change the default blue palette of the console.
3256 This is a 16-member array composed of values
3259 vt.default_grn= [VT]
3260 Format: <green0>,<green1>,<green2>,...,<green15>
3261 Change the default green palette of the console.
3262 This is a 16-member array composed of values
3265 vt.default_red= [VT]
3266 Format: <red0>,<red1>,<red2>,...,<red15>
3267 Change the default red palette of the console.
3268 This is a 16-member array composed of values
3274 Set system-wide default UTF-8 mode for all tty's.
3275 Default is 1, i.e. UTF-8 mode is enabled for all
3276 newly opened terminals.
3278 vt.global_cursor_default=
3281 Set system-wide default for whether a cursor
3282 is shown on new VTs. Default is -1,
3283 i.e. cursors will be created by default unless
3284 overridden by individual drivers. 0 will hide
3285 cursors, 1 will display them.
3287 watchdog timers [HW,WDT] For information on watchdog timers,
3288 see Documentation/watchdog/watchdog-parameters.txt
3289 or other driver-specific files in the
3290 Documentation/watchdog/ directory.
3292 workqueue.disable_numa
3293 By default, all work items queued to unbound
3294 workqueues are affine to the NUMA nodes they're
3295 issued on, which results in better behavior in
3296 general. If NUMA affinity needs to be disabled for
3297 whatever reason, this option can be used. Note
3298 that this also can be controlled per-workqueue for
3299 workqueues visible under /sys/bus/workqueue/.
3301 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
3302 default x2apic cluster mode on platforms
3305 x86_mrst_timer= [X86-32,APBT]
3306 Choose timer option for x86 Moorestown MID platform.
3307 Two valid options are apbt timer only and lapic timer
3308 plus one apbt timer for broadcast timer.
3309 x86_mrst_timer=apbt_only | lapic_and_apbt
3311 xd= [HW,XT] Original XT pre-IDE (RLL encoded) disks.
3312 xd_geo= See header of drivers/block/xd.c.
3314 xen_emul_unplug= [HW,X86,XEN]
3315 Unplug Xen emulated devices
3316 Format: [unplug0,][unplug1]
3317 ide-disks -- unplug primary master IDE devices
3318 aux-ide-disks -- unplug non-primary-master IDE devices
3319 nics -- unplug network devices
3320 all -- unplug all emulated devices (NICs and IDE disks)
3321 unnecessary -- unplugging emulated devices is
3322 unnecessary even if the host did not respond to
3324 never -- do not unplug even if version check succeeds
3326 xirc2ps_cs= [NET,PCMCIA]
3328 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
3330 ______________________________________________________________________
3334 Add more DRM drivers.