4 The following is a consolidated list of the kernel parameters as
5 implemented by the __setup(), core_param() and module_param() macros
6 and sorted into English Dictionary order (defined as ignoring all
7 punctuation and sorting digits before letters in a case insensitive
8 manner), and with descriptions where known.
10 The kernel parses parameters from the kernel command line up to "--";
11 if it doesn't recognize a parameter and it doesn't contain a '.', the
12 parameter gets passed to init: parameters with '=' go into init's
13 environment, others are passed as command line arguments to init.
14 Everything after "--" is passed as an argument to init.
16 Module parameters can be specified in two ways: via the kernel command
17 line with a module name prefix, or via modprobe, e.g.:
19 (kernel command line) usbcore.blinkenlights=1
20 (modprobe command line) modprobe usbcore blinkenlights=1
22 Parameters for modules which are built into the kernel need to be
23 specified on the kernel command line. modprobe looks through the
24 kernel command line (/proc/cmdline) and collects module parameters
25 when it loads a module, so the kernel command line can be used for
28 Hyphens (dashes) and underscores are equivalent in parameter names, so
29 log_buf_len=1M print-fatal-signals=1
30 can also be entered as
31 log-buf-len=1M print_fatal_signals=1
33 Double-quotes can be used to protect spaces in values, e.g.:
34 param="spaces in here"
36 This document may not be entirely up to date and comprehensive. The command
37 "modinfo -p ${modulename}" shows a current list of all parameters of a loadable
38 module. Loadable modules, after being loaded into the running kernel, also
39 reveal their parameters in /sys/module/${modulename}/parameters/. Some of these
40 parameters may be changed at runtime by the command
41 "echo -n ${value} > /sys/module/${modulename}/parameters/${parm}".
43 The parameters listed below are only valid if certain kernel build options were
44 enabled and if respective hardware is present. The text in square brackets at
45 the beginning of each description states the restrictions within which a
46 parameter is applicable:
48 ACPI ACPI support is enabled.
49 AGP AGP (Accelerated Graphics Port) is enabled.
50 ALSA ALSA sound support is enabled.
51 APIC APIC support is enabled.
52 APM Advanced Power Management support is enabled.
53 ARM ARM architecture is enabled.
54 AVR32 AVR32 architecture is enabled.
55 AX25 Appropriate AX.25 support is enabled.
56 BLACKFIN Blackfin architecture is enabled.
57 CLK Common clock infrastructure is enabled.
58 CMA Contiguous Memory Area support is enabled.
59 DRM Direct Rendering Management support is enabled.
60 DYNAMIC_DEBUG Build in debug messages and enable them at runtime
61 EDD BIOS Enhanced Disk Drive Services (EDD) is enabled
62 EFI EFI Partitioning (GPT) is enabled
63 EIDE EIDE/ATAPI support is enabled.
64 EVM Extended Verification Module
65 FB The frame buffer device is enabled.
66 FTRACE Function tracing enabled.
67 GCOV GCOV profiling is enabled.
68 HW Appropriate hardware is enabled.
69 IA-64 IA-64 architecture is enabled.
70 IMA Integrity measurement architecture is enabled.
71 IOSCHED More than one I/O scheduler is enabled.
72 IP_PNP IP DHCP, BOOTP, or RARP is enabled.
73 IPV6 IPv6 support is enabled.
74 ISAPNP ISA PnP code is enabled.
75 ISDN Appropriate ISDN support is enabled.
76 JOY Appropriate joystick support is enabled.
77 KGDB Kernel debugger support is enabled.
78 KVM Kernel Virtual Machine support is enabled.
79 LIBATA Libata driver is enabled
80 LP Printer support is enabled.
81 LOOP Loopback device support is enabled.
82 M68k M68k architecture is enabled.
83 These options have more detailed description inside of
84 Documentation/m68k/kernel-options.txt.
85 MDA MDA console support is enabled.
86 MIPS MIPS architecture is enabled.
87 MOUSE Appropriate mouse support is enabled.
88 MSI Message Signaled Interrupts (PCI).
89 MTD MTD (Memory Technology Device) support is enabled.
90 NET Appropriate network support is enabled.
91 NUMA NUMA support is enabled.
92 NFS Appropriate NFS support is enabled.
93 OSS OSS sound support is enabled.
94 PV_OPS A paravirtualized kernel is enabled.
95 PARIDE The ParIDE (parallel port IDE) subsystem is enabled.
96 PARISC The PA-RISC architecture is enabled.
97 PCI PCI bus support is enabled.
98 PCIE PCI Express support is enabled.
99 PCMCIA The PCMCIA subsystem is enabled.
100 PNP Plug & Play support is enabled.
101 PPC PowerPC architecture is enabled.
102 PPT Parallel port support is enabled.
103 PS2 Appropriate PS/2 support is enabled.
104 RAM RAM disk support is enabled.
105 S390 S390 architecture is enabled.
106 SCSI Appropriate SCSI support is enabled.
107 A lot of drivers have their options described inside
108 the Documentation/scsi/ sub-directory.
109 SECURITY Different security models are enabled.
110 SELINUX SELinux support is enabled.
111 APPARMOR AppArmor support is enabled.
112 SERIAL Serial support is enabled.
113 SH SuperH architecture is enabled.
114 SMP The kernel is an SMP kernel.
115 SPARC Sparc architecture is enabled.
116 SWSUSP Software suspend (hibernation) is enabled.
117 SUSPEND System suspend states are enabled.
118 TPM TPM drivers are enabled.
119 TS Appropriate touchscreen support is enabled.
120 UMS USB Mass Storage support is enabled.
121 USB USB support is enabled.
122 USBHID USB Human Interface Device support is enabled.
123 V4L Video For Linux support is enabled.
124 VMMIO Driver for memory mapped virtio devices is enabled.
125 VGA The VGA console has been enabled.
126 VT Virtual terminal support is enabled.
127 WDT Watchdog support is enabled.
128 XT IBM PC/XT MFM hard disk support is enabled.
129 X86-32 X86-32, aka i386 architecture is enabled.
130 X86-64 X86-64 architecture is enabled.
131 More X86-64 boot options can be found in
132 Documentation/x86/x86_64/boot-options.txt .
133 X86 Either 32-bit or 64-bit x86 (same as X86-32+X86-64)
134 X86_UV SGI UV support is enabled.
135 XEN Xen support is enabled
137 In addition, the following text indicates that the option:
139 BUGS= Relates to possible processor bugs on the said processor.
140 KNL Is a kernel start-up parameter.
141 BOOT Is a boot loader parameter.
143 Parameters denoted with BOOT are actually interpreted by the boot
144 loader, and have no meaning to the kernel directly.
145 Do not modify the syntax of boot loader parameters without extreme
146 need or coordination with <Documentation/x86/boot.txt>.
148 There are also arch-specific kernel-parameters not documented here.
149 See for example <Documentation/x86/x86_64/boot-options.txt>.
151 Note that ALL kernel parameters listed below are CASE SENSITIVE, and that
152 a trailing = on the name of any parameter states that that parameter will
153 be entered as an environment variable, whereas its absence indicates that
154 it will appear as a kernel argument readable via /proc/cmdline by programs
155 running once the system is up.
157 The number of kernel parameters is not limited, but the length of the
158 complete command line (parameters including spaces etc.) is limited to
159 a fixed number of characters. This limit depends on the architecture
160 and is between 256 and 4096 characters. It is defined in the file
161 ./include/asm/setup.h as COMMAND_LINE_SIZE.
163 Finally, the [KMG] suffix is commonly described after a number of kernel
164 parameter values. These 'K', 'M', and 'G' letters represent the _binary_
165 multipliers 'Kilo', 'Mega', and 'Giga', equalling 2^10, 2^20, and 2^30
166 bytes respectively. Such letter suffixes can also be entirely omitted.
169 acpi= [HW,ACPI,X86,ARM64]
170 Advanced Configuration and Power Interface
171 Format: { force | on | off | strict | noirq | rsdt |
173 force -- enable ACPI if default was off
174 on -- enable ACPI but allow fallback to DT [arm64]
175 off -- disable ACPI if default was on
176 noirq -- do not use ACPI for IRQ routing
177 strict -- Be less tolerant of platforms that are not
178 strictly ACPI specification compliant.
179 rsdt -- prefer RSDT over (default) XSDT
180 copy_dsdt -- copy DSDT to memory
181 For ARM64, ONLY "acpi=off", "acpi=on" or "acpi=force"
184 See also Documentation/power/runtime_pm.txt, pci=noacpi
186 acpi_apic_instance= [ACPI, IOAPIC]
188 2: use 2nd APIC table, if available
189 1,0: use 1st APIC table
192 acpi_backlight= [HW,ACPI]
193 acpi_backlight=vendor
195 If set to vendor, prefer vendor specific driver
196 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
197 of the ACPI video.ko driver.
199 acpi_force_32bit_fadt_addr
200 force FADT to use 32 bit addresses rather than the
201 64 bit X_* addresses. Some firmware have broken 64
202 bit addresses for force ACPI ignore these and use
203 the older legacy 32 bit addresses.
205 acpica_no_return_repair [HW, ACPI]
206 Disable AML predefined validation mechanism
207 This mechanism can repair the evaluation result to make
208 the return objects more ACPI specification compliant.
209 This option is useful for developers to identify the
210 root cause of an AML interpreter issue when the issue
211 has something to do with the repair mechanism.
213 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
214 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
216 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
217 debug output. Bits in debug_layer correspond to a
218 _COMPONENT in an ACPI source file, e.g.,
219 #define _COMPONENT ACPI_PCI_COMPONENT
220 Bits in debug_level correspond to a level in
221 ACPI_DEBUG_PRINT statements, e.g.,
222 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
223 The debug_level mask defaults to "info". See
224 Documentation/acpi/debug.txt for more information about
225 debug layers and levels.
227 Enable processor driver info messages:
228 acpi.debug_layer=0x20000000
229 Enable PCI/PCI interrupt routing info messages:
230 acpi.debug_layer=0x400000
231 Enable AML "Debug" output, i.e., stores to the Debug
232 object while interpreting AML:
233 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
234 Enable all messages related to ACPI hardware:
235 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
237 Some values produce so much output that the system is
238 unusable. The "log_buf_len" parameter may be useful
239 if you need to capture more output.
241 acpi_enforce_resources= [ACPI]
242 { strict | lax | no }
243 Check for resource conflicts between native drivers
244 and ACPI OperationRegions (SystemIO and SystemMemory
245 only). IO ports and memory declared in ACPI might be
246 used by the ACPI subsystem in arbitrary AML code and
247 can interfere with legacy drivers.
248 strict (default): access to resources claimed by ACPI
249 is denied; legacy drivers trying to access reserved
250 resources will fail to bind to device using them.
251 lax: access to resources claimed by ACPI is allowed;
252 legacy drivers trying to access reserved resources
253 will bind successfully but a warning message is logged.
254 no: ACPI OperationRegions are not marked as reserved,
255 no further checks are performed.
257 acpi_force_table_verification [HW,ACPI]
258 Enable table checksum verification during early stage.
259 By default, this is disabled due to x86 early mapping
262 acpi_irq_balance [HW,ACPI]
263 ACPI will balance active IRQs
266 acpi_irq_nobalance [HW,ACPI]
267 ACPI will not move active IRQs (default)
270 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
271 Format: <irq>,<irq>...
273 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
275 Format: <irq>,<irq>...
277 acpi_no_auto_serialize [HW,ACPI]
278 Disable auto-serialization of AML methods
279 AML control methods that contain the opcodes to create
280 named objects will be marked as "Serialized" by the
281 auto-serialization feature.
282 This feature is enabled by default.
283 This option allows to turn off the feature.
285 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
288 acpi_no_static_ssdt [HW,ACPI]
289 Disable installation of static SSDTs at early boot time
290 By default, SSDTs contained in the RSDT/XSDT will be
291 installed automatically and they will appear under
292 /sys/firmware/acpi/tables.
293 This option turns off this feature.
294 Note that specifying this option does not affect
295 dynamic table installation which will install SSDT
296 tables to /sys/firmware/acpi/tables/dynamic.
298 acpi_rsdp= [ACPI,EFI,KEXEC]
299 Pass the RSDP address to the kernel, mostly used
300 on machines running EFI runtime service to boot the
301 second kernel for kdump.
303 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
304 Format: To spoof as Windows 98: ="Microsoft Windows"
306 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
307 of 2 which is mandated by ACPI 6) as the supported ACPI
308 specification revision (when using this switch, it may
309 be necessary to carry out a cold reboot _twice_ in a
310 row to make it take effect on the platform firmware).
312 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
313 acpi_osi="string1" # add string1
314 acpi_osi="!string2" # remove string2
315 acpi_osi=!* # remove all strings
316 acpi_osi=! # disable all built-in OS vendor
318 acpi_osi=!! # enable all built-in OS vendor
320 acpi_osi= # disable all strings
322 'acpi_osi=!' can be used in combination with single or
323 multiple 'acpi_osi="string1"' to support specific OS
324 vendor string(s). Note that such command can only
325 affect the default state of the OS vendor strings, thus
326 it cannot affect the default state of the feature group
327 strings and the current state of the OS vendor strings,
328 specifying it multiple times through kernel command line
329 is meaningless. This command is useful when one do not
330 care about the state of the feature group strings which
331 should be controlled by the OSPM.
333 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
334 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
335 can make '_OSI("Windows 2000")' TRUE.
337 'acpi_osi=' cannot be used in combination with other
338 'acpi_osi=' command lines, the _OSI method will not
339 exist in the ACPI namespace. NOTE that such command can
340 only affect the _OSI support state, thus specifying it
341 multiple times through kernel command line is also
344 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
347 'acpi_osi=!*' can be used in combination with single or
348 multiple 'acpi_osi="string1"' to support specific
349 string(s). Note that such command can affect the
350 current state of both the OS vendor strings and the
351 feature group strings, thus specifying it multiple times
352 through kernel command line is meaningful. But it may
353 still not able to affect the final state of a string if
354 there are quirks related to this string. This command
355 is useful when one want to control the state of the
356 feature group strings to debug BIOS issues related to
359 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
360 '_OSI("Module Device")' FALSE.
361 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
362 '_OSI("Module Device")' TRUE.
363 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
365 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
367 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
368 they all will make '_OSI("Windows 2000")' TRUE.
371 Override the pmtimer bug detection: force the kernel
372 to assume that this machine's pmtimer latches its value
373 and always returns good values.
375 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
376 Format: { level | edge | high | low }
378 acpi_skip_timer_override [HW,ACPI]
379 Recognize and ignore IRQ0/pin2 Interrupt Override.
380 For broken nForce2 BIOS resulting in XT-PIC timer.
382 acpi_sleep= [HW,ACPI] Sleep options
383 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
384 old_ordering, nonvs, sci_force_enable }
385 See Documentation/power/video.txt for information on
387 s3_beep is for debugging; it makes the PC's speaker beep
388 as soon as the kernel's real-mode entry point is called.
389 s4_nohwsig prevents ACPI hardware signature from being
390 used during resume from hibernation.
391 old_ordering causes the ACPI 1.0 ordering of the _PTS
392 control method, with respect to putting devices into
393 low power states, to be enforced (the ACPI 2.0 ordering
394 of _PTS is used by default).
395 nonvs prevents the kernel from saving/restoring the
396 ACPI NVS memory during suspend/hibernation and resume.
397 sci_force_enable causes the kernel to set SCI_EN directly
398 on resume from S1/S3 (which is against the ACPI spec,
399 but some broken systems don't work without it).
401 acpi_use_timer_override [HW,ACPI]
402 Use timer override. For some broken Nvidia NF5 boards
403 that require a timer override, but don't have HPET
405 add_efi_memmap [EFI; X86] Include EFI memory map in
406 kernel's map of available physical RAM.
409 { off | try_unsupported }
410 off: disable AGP support
411 try_unsupported: try to drive unsupported chipsets
412 (may crash computer or cause data corruption)
415 See Documentation/sound/alsa/alsa-parameters.txt
418 Allow the default userspace alignment fault handler
419 behaviour to be specified. Bit 0 enables warnings,
420 bit 1 enables fixups, and bit 2 sends a segfault.
422 align_va_addr= [X86-64]
423 Align virtual addresses by clearing slice [14:12] when
424 allocating a VMA at process creation time. This option
425 gives you up to 3% performance improvement on AMD F15h
426 machines (where it is enabled by default) for a
427 CPU-intensive style benchmark, and it can vary highly in
428 a microbenchmark depending on workload and compiler.
430 32: only for 32-bit processes
431 64: only for 64-bit processes
432 on: enable for both 32- and 64-bit processes
433 off: disable for both 32- and 64-bit processes
435 alloc_snapshot [FTRACE]
436 Allocate the ftrace snapshot buffer on boot up when the
437 main buffer is allocated. This is handy if debugging
438 and you need to use tracing_snapshot() on boot up, and
439 do not want to use tracing_snapshot_alloc() as it needs
440 to be done where GFP_KERNEL allocations are allowed.
442 amd_iommu= [HW,X86-64]
443 Pass parameters to the AMD IOMMU driver in the system.
445 fullflush - enable flushing of IO/TLB entries when
446 they are unmapped. Otherwise they are
447 flushed before they will be reused, which
449 off - do not initialize any AMD IOMMU found in
451 force_isolation - Force device isolation for all
452 devices. The IOMMU driver is not
453 allowed anymore to lift isolation
454 requirements as needed. This option
455 does not override iommu=pt
457 amd_iommu_dump= [HW,X86-64]
458 Enable AMD IOMMU driver option to dump the ACPI table
459 for AMD IOMMU. With this option enabled, AMD IOMMU
460 driver will print ACPI tables for AMD IOMMU during
461 IOMMU initialization.
463 amijoy.map= [HW,JOY] Amiga joystick support
464 Map of devices attached to JOY0DAT and JOY1DAT
466 See also Documentation/input/joystick.txt
468 analog.map= [HW,JOY] Analog joystick and gamepad support
469 Specifies type or capabilities of an analog joystick
470 connected to one of 16 gameports
471 Format: <type1>,<type2>,..<type16>
474 Power management functions (SPARCstation-4/5 + deriv.)
476 Disable APC CPU standby support. SPARCstation-Fox does
477 not play well with APC CPU idle - disable it if you have
478 APC and your system crashes randomly.
480 apic= [APIC,X86-32] Advanced Programmable Interrupt Controller
481 Change the output verbosity whilst booting
482 Format: { quiet (default) | verbose | debug }
483 Change the amount of debugging information output
484 when initialising the APIC and IO-APIC components.
486 apic_extnmi= [APIC,X86] External NMI delivery setting
487 Format: { bsp (default) | all | none }
488 bsp: External NMI is delivered only to CPU 0
489 all: External NMIs are broadcast to all CPUs as a
491 none: External NMI is masked for all CPUs. This is
492 useful so that a dump capture kernel won't be
496 See Documentation/networking/ipv6.txt.
498 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
499 Limit apic dumping. The parameter defines the maximal
500 number of local apics being dumped. Also it is possible
501 to set it to "all" by meaning -- no limit here.
502 Format: { 1 (default) | 2 | ... | all }.
503 The parameter valid if only apic=debug or
504 apic=verbose is specified.
505 Example: apic=debug show_lapic=all
507 apm= [APM] Advanced Power Management
508 See header of arch/x86/kernel/apm_32.c.
510 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
511 Format: <io>,<irq>,<nodeID>
515 atarimouse= [HW,MOUSE] Atari Mouse
517 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
518 EzKey and similar keyboards
520 atkbd.reset= [HW] Reset keyboard during initialization
522 atkbd.set= [HW] Select keyboard code set
523 Format: <int> (2 = AT (default), 3 = PS/2)
525 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
528 atkbd.softraw= [HW] Choose between synthetic and real raw mode
529 Format: <bool> (0 = real, 1 = synthetic (default))
531 atkbd.softrepeat= [HW]
532 Use software keyboard repeat
534 audit= [KNL] Enable the audit sub-system
535 Format: { "0" | "1" } (0 = disabled, 1 = enabled)
536 0 - kernel audit is disabled and can not be enabled
537 until the next reboot
538 unset - kernel audit is initialized but disabled and
539 will be fully enabled by the userspace auditd.
540 1 - kernel audit is initialized and partially enabled,
541 storing at most audit_backlog_limit messages in
542 RAM until it is fully enabled by the userspace
546 audit_backlog_limit= [KNL] Set the audit queue size limit.
547 Format: <int> (must be >=0)
550 bau= [X86_UV] Enable the BAU on SGI UV. The default
551 behavior is to disable the BAU (i.e. bau=0).
552 Format: { "0" | "1" }
555 unset - Disable the BAU.
557 baycom_epp= [HW,AX25]
560 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
562 See header of drivers/net/hamradio/baycom_par.c.
564 baycom_ser_fdx= [HW,AX25]
565 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
566 Format: <io>,<irq>,<mode>[,<baud>]
567 See header of drivers/net/hamradio/baycom_ser_fdx.c.
569 baycom_ser_hdx= [HW,AX25]
570 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
571 Format: <io>,<irq>,<mode>
572 See header of drivers/net/hamradio/baycom_ser_hdx.c.
574 blkdevparts= Manual partition parsing of block device(s) for
575 embedded devices based on command line input.
576 See Documentation/block/cmdline-partition.txt
578 boot_delay= Milliseconds to delay each printk during boot.
579 Values larger than 10 seconds (10000) are changed to
583 bootmem_debug [KNL] Enable bootmem allocator debug messages.
586 Disable BERT OS support on buggy BIOSes.
588 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
589 bttv.radio= Most important insmod options are available as
591 bttv.pll= See Documentation/video4linux/bttv/Insmod-options
594 bulk_remove=off [PPC] This parameter disables the use of the pSeries
595 firmware feature for flushing multiple hpte entries
598 c101= [NET] Moxa C101 synchronous serial card
600 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
601 Sometimes CPU hardware bugs make them report the cache
602 size incorrectly. The kernel will attempt work arounds
603 to fix known problems, but for some CPUs it is not
604 possible to determine what the correct size should be.
605 This option provides an override for these situations.
607 ca_keys= [KEYS] This parameter identifies a specific key(s) on
608 the system trusted keyring to be used for certificate
610 format: { id:<keyid> | builtin }
612 cca= [MIPS] Override the kernel pages' cache coherency
613 algorithm. Accepted values range from 0 to 7
614 inclusive. See arch/mips/include/asm/pgtable-bits.h
615 for platform specific values (SB1, Loongson3 and
618 ccw_timeout_log [S390]
619 See Documentation/s390/CommonIO for details.
621 cgroup_disable= [KNL] Disable a particular controller
622 Format: {name of the controller(s) to disable}
623 The effects of cgroup_disable=foo are:
624 - foo isn't auto-mounted if you mount all cgroups in
626 - foo isn't visible as an individually mountable
628 {Currently only "memory" controller deal with this and
629 cut the overhead, others just disable the usage. So
630 only cgroup_disable=memory is actually worthy}
632 cgroup_no_v1= [KNL] Disable one, multiple, all cgroup controllers in v1
633 Format: { controller[,controller...] | "all" }
634 Like cgroup_disable, but only applies to cgroup v1;
635 the blacklisted controllers remain available in cgroup2.
637 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
639 nosocket -- Disable socket memory accounting.
640 nokmem -- Disable kernel memory accounting.
642 checkreqprot [SELINUX] Set initial checkreqprot flag value.
643 Format: { "0" | "1" }
644 See security/selinux/Kconfig help text.
645 0 -- check protection applied by kernel (includes
646 any implied execute protection).
647 1 -- check protection requested by application.
648 Default value is set via a kernel config option.
649 Value can be changed at runtime via
650 /selinux/checkreqprot.
653 See Documentation/s390/CommonIO for details.
656 Prevents the clock framework from automatically gating
657 clocks that have not been explicitly enabled by a Linux
658 device driver but are enabled in hardware at reset or
659 by the bootloader/firmware. Note that this does not
660 force such clocks to be always-on nor does it reserve
661 those clocks in any way. This parameter is useful for
662 debug and development, but should not be needed on a
663 platform with proper driver support. For more
664 information, see Documentation/clk.txt.
666 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
668 Forces specified clocksource (if available) to be used
669 when calculating gettimeofday(). If specified
670 clocksource is not available, it defaults to PIT.
671 Format: { pit | tsc | cyclone | pmtmr }
673 clocksource= Override the default clocksource
675 Override the default clocksource and use the clocksource
676 with the name specified.
677 Some clocksource names to choose from, depending on
679 [all] jiffies (this is the base, fallback clocksource)
681 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
682 pxa_timer,timer3,32k_counter,timer0_1
684 [X86-32] pit,hpet,tsc;
685 scx200_hrt on Geode; cyclone on IBM x440
693 clocksource.arm_arch_timer.evtstrm=
696 Enable/disable the eventstream feature of the ARM
697 architected timer so that code using WFE-based polling
698 loops can be debugged more effectively on production
701 clearcpuid=BITNUM [X86]
702 Disable CPUID feature X for the kernel. See
703 arch/x86/include/asm/cpufeatures.h for the valid bit
704 numbers. Note the Linux specific bits are not necessarily
705 stable over kernel options, but the vendor specific
707 Also note that user programs calling CPUID directly
708 or using the feature without checking anything
709 will still see it. This just prevents it from
710 being used by the kernel or shown in /proc/cpuinfo.
711 Also note the kernel might malfunction if you disable
714 cma=nn[MG]@[start[MG][-end[MG]]]
716 Sets the size of kernel global memory area for
717 contiguous memory allocations and optionally the
718 placement constraint by the physical address range of
719 memory allocations. A value of 0 disables CMA
720 altogether. For more information, see
721 include/linux/dma-contiguous.h
723 cmo_free_hint= [PPC] Format: { yes | no }
724 Specify whether pages are marked as being inactive
725 when they are freed. This is used in CMO environments
726 to determine OS memory pressure for page stealing by
730 coherent_pool=nn[KMG] [ARM,KNL]
731 Sets the size of memory pool for coherent, atomic dma
732 allocations, by default set to 256K.
734 code_bytes [X86] How many bytes of object code to print
739 com20020= [HW,NET] ARCnet - COM20020 chipset
741 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
743 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
747 ARCnet - COM90xx chipset (memory-mapped buffers)
748 Format: <io>[,<irq>[,<memstart>]]
750 condev= [HW,S390] console device
753 console= [KNL] Output console device and options.
755 tty<n> Use the virtual console device <n>.
759 Use the specified serial port. The options are of
760 the form "bbbbpnf", where "bbbb" is the baud rate,
761 "p" is parity ("n", "o", or "e"), "n" is number of
762 bits, and "f" is flow control ("r" for RTS or
763 omit it). Default is "9600n8".
765 See Documentation/serial-console.txt for more
767 Documentation/networking/netconsole.txt for an
770 uart[8250],io,<addr>[,options]
771 uart[8250],mmio,<addr>[,options]
772 uart[8250],mmio16,<addr>[,options]
773 uart[8250],mmio32,<addr>[,options]
774 uart[8250],0x<addr>[,options]
775 Start an early, polled-mode console on the 8250/16550
776 UART at the specified I/O port or MMIO address,
777 switching to the matching ttyS device later.
778 MMIO inter-register address stride is either 8-bit
779 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
780 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
781 to be equivalent to 'mmio'. 'options' are specified in
782 the same format described for ttyS above; if unspecified,
783 the h/w is not re-initialized.
785 hvc<n> Use the hypervisor console device <n>. This is for
786 both Xen and PowerPC hypervisors.
788 If the device connected to the port is not a TTY but a braille
789 device, prepend "brl," before the device type, for instance
791 For now, only VisioBraille is supported.
793 consoleblank= [KNL] The console blank (screen saver) timeout in
794 seconds. Defaults to 10*60 = 10mins. A value of 0
795 disables the blank timer.
798 [KNL] Change the default value for
799 /proc/<pid>/coredump_filter.
800 See also Documentation/filesystems/proc.txt.
802 cpuidle.off=1 [CPU_IDLE]
803 disable the cpuidle sub-system
806 [X86] Delay for N microsec between assert and de-assert
807 of APIC INIT to start processors. This delay occurs
808 on every CPU online, such as boot, and resume from suspend.
811 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
813 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
815 crashkernel=size[KMG][@offset[KMG]]
816 [KNL] Using kexec, Linux can switch to a 'crash kernel'
817 upon panic. This parameter reserves the physical
818 memory region [offset, offset + size] for that kernel
819 image. If '@offset' is omitted, then a suitable offset
820 is selected automatically. Check
821 Documentation/kdump/kdump.txt for further details.
823 crashkernel=range1:size1[,range2:size2,...][@offset]
824 [KNL] Same as above, but depends on the memory
825 in the running system. The syntax of range is
826 start-[end] where start and end are both
827 a memory unit (amount[KMG]). See also
828 Documentation/kdump/kdump.txt for an example.
830 crashkernel=size[KMG],high
831 [KNL, x86_64] range could be above 4G. Allow kernel
832 to allocate physical memory region from top, so could
833 be above 4G if system have more than 4G ram installed.
834 Otherwise memory region will be allocated below 4G, if
836 It will be ignored if crashkernel=X is specified.
837 crashkernel=size[KMG],low
838 [KNL, x86_64] range under 4G. When crashkernel=X,high
839 is passed, kernel could allocate physical memory region
840 above 4G, that cause second kernel crash on system
841 that require some amount of low memory, e.g. swiotlb
842 requires at least 64M+32K low memory, also enough extra
843 low memory is needed to make sure DMA buffers for 32-bit
844 devices won't run out. Kernel would try to allocate at
845 at least 256M below 4G automatically.
846 This one let user to specify own low range under 4G
847 for second kernel instead.
848 0: to disable low allocation.
849 It will be ignored when crashkernel=X,high is not used
850 or memory reserved is below 4G.
853 [KNL] Disable crypto self-tests
858 cs89x0_media= [HW,NET]
859 Format: { rj45 | aui | bnc }
862 See header of drivers/s390/block/dasd_devmap.c.
864 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
865 (one device per port)
866 Format: <port#>,<type>
867 See also Documentation/input/joystick-parport.txt
869 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
870 time. See Documentation/dynamic-debug-howto.txt for
871 details. Deprecated, see dyndbg.
873 debug [KNL] Enable kernel debugging (events log level).
876 [KNL] verbose self-tests
878 Print debugging info while doing the locking API
880 We default to 0 (no extra messages), setting it to
881 1 will print _a lot_ more information - normally
882 only useful to kernel developers.
884 debug_objects [KNL] Enable object debugging
887 [KNL] Disable object debugging
889 debug_guardpage_minorder=
890 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
891 parameter allows control of the order of pages that will
892 be intentionally kept free (and hence protected) by the
893 buddy allocator. Bigger value increase the probability
894 of catching random memory corruption, but reduce the
895 amount of memory for normal system use. The maximum
896 possible value is MAX_ORDER/2. Setting this parameter
897 to 1 or 2 should be enough to identify most random
898 memory corruption problems caused by bugs in kernel or
899 driver code when a CPU writes to (or reads from) a
900 random memory location. Note that there exists a class
901 of memory corruptions problems caused by buggy H/W or
902 F/W or by drivers badly programing DMA (basically when
903 memory is written at bus level and the CPU MMU is
904 bypassed) which are not detectable by
905 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
906 tracking down these problems.
909 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
910 parameter enables the feature at boot time. In
911 default, it is disabled. We can avoid allocating huge
912 chunk of memory for debug pagealloc if we don't enable
913 it at boot time and the system will work mostly same
914 with the kernel built without CONFIG_DEBUG_PAGEALLOC.
915 on: enable the feature
917 debugpat [X86] Enable PAT debugging
919 decnet.addr= [HW,NET]
920 Format: <area>[,<node>]
921 See also Documentation/networking/decnet.txt.
924 [same as hugepagesz=] The size of the default
925 HugeTLB page size. This is the size represented by
926 the legacy /proc/ hugepages APIs, used for SHM, and
927 default size when mounting hugetlbfs filesystems.
928 Defaults to the default architecture's huge page size
932 Set number of hash buckets for dentry cache.
935 See Documentation/networking/ipv6.txt.
937 disable_cpu_apicid= [X86,APIC,SMP]
939 The number of initial APIC ID for the
940 corresponding CPU to be disabled at boot,
941 mostly used for the kdump 2nd kernel to
942 disable BSP to wake up multiple CPUs without
943 causing system reset or hang due to sending
946 disable_ddw [PPC/PSERIES]
947 Disable Dynamic DMA Window support. Use this if
948 to workaround buggy firmware.
951 See Documentation/networking/ipv6.txt.
953 disable_mtrr_cleanup [X86]
954 The kernel tries to adjust MTRR layout from continuous
955 to discrete, to make X server driver able to add WB
956 entry later. This parameter disables that.
958 disable_mtrr_trim [X86, Intel and AMD only]
959 By default the kernel will trim any uncacheable
960 memory out of your available memory pool based on
961 MTRR settings. This parameter disables that behavior,
962 possibly causing your machine to run very slowly.
964 disable_timer_pin_1 [X86]
965 Disable PIN 1 of APIC timer
966 Can be useful to work around chipset bugs.
968 dis_ucode_ldr [X86] Disable the microcode loader.
970 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
971 this option disables the debugging code at boot.
973 dma_debug_entries=<number>
974 This option allows to tune the number of preallocated
975 entries for DMA-API debugging code. One entry is
976 required per DMA-API allocation. Use this if the
977 DMA-API debugging code disables itself because the
978 architectural default is too low.
980 dma_debug_driver=<driver_name>
981 With this option the DMA-API debugging driver
982 filter feature can be enabled at boot time. Just
983 pass the driver to filter for as the parameter.
984 The filter can be disabled or changed to another
985 driver later using sysfs.
987 drm_kms_helper.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
988 Broken monitors, graphic adapters, KVMs and EDIDless
989 panels may send no or incorrect EDID data sets.
990 This parameter allows to specify an EDID data sets
991 in the /lib/firmware directory that are used instead.
992 Generic built-in EDID data sets are used, if one of
993 edid/1024x768.bin, edid/1280x1024.bin,
994 edid/1680x1050.bin, or edid/1920x1080.bin is given
995 and no file with the same name exists. Details and
996 instructions how to build your own EDID data are
997 available in Documentation/EDID/HOWTO.txt. An EDID
998 data set will only be used for a particular connector,
999 if its name and a colon are prepended to the EDID
1000 name. Each connector may use a unique EDID data
1001 set by separating the files with a comma. An EDID
1002 data set with no connector name will be used for
1003 any connectors not explicitly specified.
1007 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1008 module.dyndbg[="val"]
1009 Enable debug messages at boot time. See
1010 Documentation/dynamic-debug-howto.txt for details.
1012 nompx [X86] Disables Intel Memory Protection Extensions.
1013 See Documentation/x86/intel_mpx.txt for more
1014 information about the feature.
1016 nopku [X86] Disable Memory Protection Keys CPU feature found
1020 on enable eager fpu restore
1021 off disable eager fpu restore
1022 auto selects the default scheme, which automatically
1023 enables eagerfpu restore for xsaveopt.
1025 module.async_probe [KNL]
1026 Enable asynchronous probe on this module.
1028 early_ioremap_debug [KNL]
1029 Enable debug messages in early_ioremap support. This
1030 is useful for tracking down temporary early mappings
1031 which are not unmapped.
1033 earlycon= [KNL] Output early console device and options.
1035 When used with no options, the early console is
1036 determined by the stdout-path property in device
1040 Start an early, polled-mode console on a cadence serial
1041 port at the specified address. The cadence serial port
1042 must already be setup and configured. Options are not
1045 uart[8250],io,<addr>[,options]
1046 uart[8250],mmio,<addr>[,options]
1047 uart[8250],mmio32,<addr>[,options]
1048 uart[8250],mmio32be,<addr>[,options]
1049 uart[8250],0x<addr>[,options]
1050 Start an early, polled-mode console on the 8250/16550
1051 UART at the specified I/O port or MMIO address.
1052 MMIO inter-register address stride is either 8-bit
1053 (mmio) or 32-bit (mmio32 or mmio32be).
1054 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1055 to be equivalent to 'mmio'. 'options' are specified
1056 in the same format described for "console=ttyS<n>"; if
1057 unspecified, the h/w is not initialized.
1061 Start an early, polled-mode console on a pl011 serial
1062 port at the specified address. The pl011 serial port
1063 must already be setup and configured. Options are not
1064 yet supported. If 'mmio32' is specified, then only
1065 the driver will use only 32-bit accessors to read/write
1066 the device registers.
1069 Start an early, polled-mode console on a meson serial
1070 port at the specified address. The serial port must
1071 already be setup and configured. Options are not yet
1075 Start an early, polled-mode console on an msm serial
1076 port at the specified address. The serial port
1077 must already be setup and configured. Options are not
1080 msm_serial_dm,<addr>
1081 Start an early, polled-mode console on an msm serial
1082 dm port at the specified address. The serial port
1083 must already be setup and configured. Options are not
1086 smh Use ARM semihosting calls for early console.
1094 Use early console provided by serial driver available
1095 on Samsung SoCs, requires selecting proper type and
1096 a correct base address of the selected UART port. The
1097 serial port must already be setup and configured.
1098 Options are not yet supported.
1102 Use early console provided by Freescale LP UART driver
1103 found on Freescale Vybrid and QorIQ LS1021A processors.
1104 A valid base address must be provided, and the serial
1105 port must already be setup and configured.
1107 armada3700_uart,<addr>
1108 Start an early, polled-mode console on the
1109 Armada 3700 serial port at the specified
1110 address. The serial port must already be setup
1111 and configured. Options are not yet supported.
1113 earlyprintk= [X86,SH,BLACKFIN,ARM,M68k]
1117 earlyprintk=serial[,ttySn[,baudrate]]
1118 earlyprintk=serial[,0x...[,baudrate]]
1119 earlyprintk=ttySn[,baudrate]
1120 earlyprintk=dbgp[debugController#]
1121 earlyprintk=pciserial,bus:device.function[,baudrate]
1123 earlyprintk is useful when the kernel crashes before
1124 the normal console is initialized. It is not enabled by
1125 default because it has some cosmetic problems.
1127 Append ",keep" to not disable it when the real console
1130 Only one of vga, efi, serial, or usb debug port can
1133 Currently only ttyS0 and ttyS1 may be specified by
1134 name. Other I/O ports may be explicitly specified
1135 on some architectures (x86 and arm at least) by
1136 replacing ttySn with an I/O port address, like this:
1137 earlyprintk=serial,0x1008,115200
1138 You can find the port for a given device in
1139 /proc/tty/driver/serial:
1140 2: uart:ST16650V2 port:00001008 irq:18 ...
1142 Interaction with the standard serial driver is not
1145 The VGA and EFI output is eventually overwritten by
1148 The xen output can only be used by Xen PV guests.
1150 edac_report= [HW,EDAC] Control how to report EDAC event
1151 Format: {"on" | "off" | "force"}
1152 on: enable EDAC to report H/W event. May be overridden
1153 by other higher priority error reporting module.
1154 off: disable H/W event reporting through EDAC.
1155 force: enforce the use of EDAC to report H/W event.
1158 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1161 This is designed to be used in conjunction with
1162 the boot argument: earlyprintk=vga
1165 Format: {"off" | "on" | "skip[mbr]"}
1168 Format: { "old_map", "nochunk", "noruntime", "debug" }
1169 old_map [X86-64]: switch to the old ioremap-based EFI
1170 runtime services mapping. 32-bit still uses this one by
1172 nochunk: disable reading files in "chunks" in the EFI
1173 boot stub, as chunking can cause problems with some
1174 firmware implementations.
1175 noruntime : disable EFI runtime services support
1176 debug: enable misc debug output
1178 efi_no_storage_paranoia [EFI; X86]
1179 Using this parameter you can use more than 50% of
1180 your efi variable storage. Use this parameter only if
1181 you are really sure that your UEFI does sane gc and
1182 fulfills the spec otherwise your board may brick.
1184 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1185 Add arbitrary attribute to specific memory range by
1186 updating original EFI memory map.
1187 Region of memory which aa attribute is added to is
1189 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1190 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1191 attribute is added to range 0x100000000-0x180000000 and
1192 0x10a0000000-0x1120000000.
1194 Using this parameter you can do debugging of EFI memmap
1195 related feature. For example, you can do debugging of
1196 Address Range Mirroring feature even if your box
1199 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1200 that is to be dynamically loaded by Linux. If there are
1201 multiple variables with the same name but with different
1202 vendor GUIDs, all of them will be loaded. See
1203 Documentation/acpi/ssdt-overlays.txt for details.
1206 eisa_irq_edge= [PARISC,HW]
1207 See header of drivers/parisc/eisa.c.
1210 See comment before function elanfreq_setup() in
1211 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1214 Format: {"cfq" | "deadline" | "noop"}
1215 See Documentation/block/cfq-iosched.txt and
1216 Documentation/block/deadline-iosched.txt for details.
1218 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1219 Specifies physical address of start of kernel core
1220 image elf header and optionally the size. Generally
1221 kexec loader will pass this option to capture kernel.
1222 See Documentation/kdump/kdump.txt for details.
1224 enable_mtrr_cleanup [X86]
1225 The kernel tries to adjust MTRR layout from continuous
1226 to discrete, to make X server driver able to add WB
1227 entry later. This parameter enables that.
1229 enable_timer_pin_1 [X86]
1230 Enable PIN 1 of APIC timer
1231 Can be useful to work around chipset bugs
1232 (in particular on some ATI chipsets).
1233 The kernel tries to set a reasonable default.
1235 enforcing [SELINUX] Set initial enforcing status.
1237 See security/selinux/Kconfig help text.
1238 0 -- permissive (log only, no denials).
1239 1 -- enforcing (deny and log).
1241 Value can be changed at runtime via /selinux/enforce.
1244 Disable Error Record Serialization Table (ERST)
1247 ether= [HW,NET] Ethernet cards parameters
1248 This option is obsoleted by the "netdev=" option, which
1249 has equivalent usage. See its documentation for details.
1253 Permit 'security.evm' to be updated regardless of
1254 current integrity status.
1258 fail_make_request=[KNL]
1259 General fault injection mechanism.
1260 Format: <interval>,<probability>,<space>,<times>
1261 See also Documentation/fault-injection/.
1264 See Documentation/blockdev/floppy.txt.
1266 force_pal_cache_flush
1267 [IA-64] Avoid check_sal_cache_flush which may hang on
1268 buggy SAL_CACHE_FLUSH implementations. Using this
1269 parameter will force ia64_sal_cache_flush to call
1270 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1273 Forcefully enable Physical Address Extension (PAE).
1274 Many Pentium M systems disable PAE but may have a
1275 functionally usable PAE implementation.
1276 Warning: use of this parameter will taint the kernel
1277 and may cause unknown problems.
1280 [FTRACE] will set and start the specified tracer
1281 as early as possible in order to facilitate early
1284 ftrace_dump_on_oops[=orig_cpu]
1285 [FTRACE] will dump the trace buffers on oops.
1286 If no parameter is passed, ftrace will dump
1287 buffers of all CPUs, but if you pass orig_cpu, it will
1288 dump only the buffer of the CPU that triggered the
1291 ftrace_filter=[function-list]
1292 [FTRACE] Limit the functions traced by the function
1293 tracer at boot up. function-list is a comma separated
1294 list of functions. This list can be changed at run
1295 time by the set_ftrace_filter file in the debugfs
1298 ftrace_notrace=[function-list]
1299 [FTRACE] Do not trace the functions specified in
1300 function-list. This list can be changed at run time
1301 by the set_ftrace_notrace file in the debugfs
1304 ftrace_graph_filter=[function-list]
1305 [FTRACE] Limit the top level callers functions traced
1306 by the function graph tracer at boot up.
1307 function-list is a comma separated list of functions
1308 that can be changed at run time by the
1309 set_graph_function file in the debugfs tracing directory.
1311 ftrace_graph_notrace=[function-list]
1312 [FTRACE] Do not trace from the functions specified in
1313 function-list. This list is a comma separated list of
1314 functions that can be changed at run time by the
1315 set_graph_notrace file in the debugfs tracing directory.
1318 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1319 support via parallel port (up to 5 devices per port)
1320 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1321 See also Documentation/input/joystick-parport.txt
1325 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1329 gcov_persist= [GCOV] When non-zero (default), profiling data for
1330 kernel modules is saved and remains accessible via
1331 debugfs, even when the module is unloaded/reloaded.
1332 When zero, profiling data is discarded and associated
1333 debugfs files are removed at module unload time.
1335 gpt [EFI] Forces disk with valid GPT signature but
1336 invalid Protective MBR to be treated as GPT. If the
1337 primary GPT is corrupted, it enables the backup/alternate
1338 GPT to be used instead.
1340 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1341 the "Enable 0" bit of the configuration register.
1344 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1345 the "Enable 0" bit of the configuration register.
1348 grcan.select= [HW] Select which physical interface to use.
1351 grcan.txsize= [HW] Sets the size of the tx buffer.
1352 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1354 grcan.rxsize= [HW] Sets the size of the rx buffer.
1355 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1358 hardlockup_all_cpu_backtrace=
1359 [KNL] Should the hard-lockup detector generate
1360 backtraces on all cpus.
1363 hashdist= [KNL,NUMA] Large hashes allocated during boot
1364 are distributed across NUMA nodes. Defaults on
1365 for 64-bit NUMA, off otherwise.
1366 Format: 0 | 1 (for off | on)
1368 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1370 hd= [EIDE] (E)IDE hard drive subsystem geometry
1371 Format: <cyl>,<head>,<sect>
1374 Disable Hardware Error Source Table (HEST) support;
1375 corresponding firmware-first mode error processing
1376 logic will be disabled.
1378 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1379 size of <nn>. This works even on boxes that have no
1380 highmem otherwise. This also works to reduce highmem
1381 size on bigger boxes.
1383 highres= [KNL] Enable/disable high resolution timer mode.
1384 Valid parameters: "on", "off"
1388 See Documentation/isdn/README.HiSax.
1392 hpet= [X86-32,HPET] option to control HPET usage
1393 Format: { enable (default) | disable | force |
1395 disable: disable HPET and use PIT instead
1396 force: allow force enabled of undocumented chips (ICH4,
1398 verbose: show contents of HPET registers during setup
1400 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1401 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1403 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1404 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
1405 On x86-64 and powerpc, this option can be specified
1406 multiple times interleaved with hugepages= to reserve
1407 huge pages of different sizes. Valid pages sizes on
1408 x86-64 are 2M (when the CPU supports "pse") and 1G
1409 (when the CPU supports the "pdpe1gb" cpuinfo flag).
1411 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1412 terminal devices. Valid values: 0..8
1413 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1414 If specified, z/VM IUCV HVC accepts connections
1415 from listed z/VM user IDs only.
1417 hwthread_map= [METAG] Comma-separated list of Linux cpu id to
1418 hardware thread id mappings.
1419 Format: <cpu>:<hwthread>
1422 Do not unregister boot console at start. This is only
1423 useful for debugging when something happens in the window
1424 between unregistering the boot console and initializing
1427 i2c_bus= [HW] Override the default board specific I2C bus speed
1428 or register an additional I2C bus that is not
1429 registered from board initialization code.
1433 i8042.debug [HW] Toggle i8042 debug mode
1434 i8042.unmask_kbd_data
1435 [HW] Enable printing of interrupt data from the KBD port
1436 (disabled by default, and as a pre-condition
1437 requires that i8042.debug=1 be enabled)
1438 i8042.direct [HW] Put keyboard port into non-translated mode
1439 i8042.dumbkbd [HW] Pretend that controller can only read data from
1440 keyboard and cannot control its state
1441 (Don't attempt to blink the leds)
1442 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1443 i8042.nokbd [HW] Don't check/create keyboard port
1444 i8042.noloop [HW] Disable the AUX Loopback command while probing
1446 i8042.nomux [HW] Don't check presence of an active multiplexing
1448 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1450 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1451 i8042.reset [HW] Reset the controller during init and cleanup
1452 i8042.unlock [HW] Unlock (ignore) the keylock
1453 i8042.kbdreset [HW] Reset device connected to KBD port
1457 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1458 indicates that the driver is running on unsupported
1460 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1461 does not match list of supported models.
1463 [HW] Report power status in /proc/i8k
1464 (disabled by default)
1465 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1468 i915.invert_brightness=
1469 [DRM] Invert the sense of the variable that is used to
1470 set the brightness of the panel backlight. Normally a
1471 brightness value of 0 indicates backlight switched off,
1472 and the maximum of the brightness value sets the backlight
1473 to maximum brightness. If this parameter is set to 0
1474 (default) and the machine requires it, or this parameter
1475 is set to 1, a brightness value of 0 sets the backlight
1476 to maximum brightness, and the maximum of the brightness
1477 value switches the backlight off.
1478 -1 -- never invert brightness
1479 0 -- machine default
1480 1 -- force brightness inversion
1483 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1485 ide-core.nodma= [HW] (E)IDE subsystem
1486 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1487 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1488 .cdrom .chs .ignore_cable are additional options
1489 See Documentation/ide/ide.txt.
1491 ide-generic.probe-mask= [HW] (E)IDE subsystem
1493 Probe mask for legacy ISA IDE ports. Depending on
1494 platform up to 6 ports are supported, enabled by
1495 setting corresponding bits in the mask to 1. The
1496 default value is 0x0, which has a special meaning.
1497 On systems that have PCI, it triggers scanning the
1498 PCI bus for the first and the second port, which
1499 are then probed. On systems without PCI the value
1500 of 0x0 enables probing the two first ports as if it
1503 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1504 Claim all unknown PCI IDE storage controllers.
1507 Format: idle=poll, idle=halt, idle=nomwait
1508 Poll forces a polling idle loop that can slightly
1509 improve the performance of waking up a idle CPU, but
1510 will use a lot of power and make the system run hot.
1512 idle=halt: Halt is forced to be used for CPU idle.
1513 In such case C2/C3 won't be used again.
1514 idle=nomwait: Disable mwait for CPU C-states
1516 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1517 Format: { strict | legacy | 2008 | relaxed }
1520 Choose which programs will be accepted for execution
1521 based on the IEEE 754 NaN encoding(s) supported by
1522 the FPU and the NaN encoding requested with the value
1523 of an ELF file header flag individually set by each
1524 binary. Hardware implementations are permitted to
1525 support either or both of the legacy and the 2008 NaN
1528 Available settings are as follows:
1529 strict accept binaries that request a NaN encoding
1530 supported by the FPU
1531 legacy only accept legacy-NaN binaries, if supported
1533 2008 only accept 2008-NaN binaries, if supported
1535 relaxed accept any binaries regardless of whether
1536 supported by the FPU
1538 The FPU emulator is always able to support both NaN
1539 encodings, so if no FPU hardware is present or it has
1540 been disabled with 'nofpu', then the settings of
1541 'legacy' and '2008' strap the emulator accordingly,
1542 'relaxed' straps the emulator for both legacy-NaN and
1543 2008-NaN, whereas 'strict' enables legacy-NaN only on
1544 legacy processors and both NaN encodings on MIPS32 or
1547 The setting for ABS.fmt/NEG.fmt instruction execution
1548 mode generally follows that for the NaN encoding,
1549 except where unsupported by hardware.
1551 ignore_loglevel [KNL]
1552 Ignore loglevel setting - this will print /all/
1553 kernel messages to the console. Useful for debugging.
1554 We also add it as printk module parameter, so users
1555 could change it dynamically, usually by
1556 /sys/module/printk/parameters/ignore_loglevel.
1559 Ignore RLIMIT_DATA setting for data mappings,
1560 print warning at first misuse. Can be changed via
1561 /sys/module/kernel/parameters/ignore_rlimit_data.
1563 ihash_entries= [KNL]
1564 Set number of hash buckets for inode cache.
1566 ima_appraise= [IMA] appraise integrity measurements
1567 Format: { "off" | "enforce" | "fix" | "log" }
1570 ima_appraise_tcb [IMA]
1571 The builtin appraise policy appraises all files
1575 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1579 The list of supported hash algorithms is defined
1580 in crypto/hash_info.h.
1583 The builtin measurement policy to load during IMA
1584 setup. Specyfing "tcb" as the value, measures all
1585 programs exec'd, files mmap'd for exec, and all files
1586 opened with the read mode bit set by either the
1587 effective uid (euid=0) or uid=0.
1590 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1591 Load a policy which meets the needs of the Trusted
1592 Computing Base. This means IMA will measure all
1593 programs exec'd, files mmap'd for exec, and all files
1594 opened for read by uid=0.
1597 Select one of defined IMA measurements template formats.
1598 Formats: { "ima" | "ima-ng" | "ima-sig" }
1602 [IMA] Define a custom template format.
1603 Format: { "field1|...|fieldN" }
1605 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1606 Format: <min_file_size>
1607 Set the minimal file size for using asynchronous hash.
1608 If left unspecified, ahash usage is disabled.
1610 ahash performance varies for different data sizes on
1611 different crypto accelerators. This option can be used
1612 to achieve the best performance for a particular HW.
1614 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1616 Set hashing buffer size. Default: 4k.
1618 ahash performance varies for different chunk sizes on
1619 different crypto accelerators. This option can be used
1620 to achieve best performance for particular HW.
1624 Run specified binary instead of /sbin/init as init
1627 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1628 for working out where the kernel is dying during
1631 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1632 initcall functions. Useful for debugging built-in
1633 modules and initcalls.
1635 initrd= [BOOT] Specify the location of the initial ramdisk
1637 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1640 int_pln_enable [x86] Enable power limit notification interrupt
1642 integrity_audit=[IMA]
1643 Format: { "0" | "1" }
1644 0 -- basic integrity auditing messages. (Default)
1645 1 -- additional integrity auditing messages.
1647 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1649 Enable intel iommu driver.
1651 Disable intel iommu driver.
1652 igfx_off [Default Off]
1653 By default, gfx is mapped as normal device. If a gfx
1654 device has a dedicated DMAR unit, the DMAR unit is
1655 bypassed by not enabling DMAR with this option. In
1656 this case, gfx device will use physical address for
1659 With this option iommu will not optimize to look
1660 for io virtual address below 32-bit forcing dual
1661 address cycle on pci bus for cards supporting greater
1662 than 32-bit addressing. The default is to look
1663 for translation below 32-bit and if not available
1664 then look in the higher range.
1665 strict [Default Off]
1666 With this option on every unmap_single operation will
1667 result in a hardware IOTLB flush operation as opposed
1668 to batching them for performance.
1669 sp_off [Default Off]
1670 By default, super page will be supported if Intel IOMMU
1671 has the capability. With this option, super page will
1673 ecs_off [Default Off]
1674 By default, extended context tables will be supported if
1675 the hardware advertises that it has support both for the
1676 extended tables themselves, and also PASID support. With
1677 this option set, extended tables will not be used even
1678 on hardware which claims to support them.
1680 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1681 0 disables intel_idle and fall back on acpi_idle.
1682 1 to 6 specify maximum depth of C-state.
1686 Do not enable intel_pstate as the default
1687 scaling driver for the supported processors
1689 Enable intel_pstate on systems that prohibit it by default
1690 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1691 instead of acpi-cpufreq may disable platform features, such
1692 as thermal controls and power capping, that rely on ACPI
1693 P-States information being indicated to OSPM and therefore
1694 should be used with caution. This option does not work with
1695 processors that aren't supported by the intel_pstate driver
1696 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1698 Do not enable hardware P state control (HWP)
1701 Only load intel_pstate on systems which support
1702 hardware P state control (HWP) if available.
1704 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1705 Description Table, specifies preferred power management
1706 profile as "Enterprise Server" or "Performance Server",
1707 then this feature is turned on by default.
1709 intremap= [X86-64, Intel-IOMMU]
1710 on enable Interrupt Remapping (default)
1711 off disable Interrupt Remapping
1712 nosid disable Source ID checking
1714 BIOS x2APIC opt-out request will be ignored
1715 nopost disable Interrupt Posting
1717 iomem= Disable strict checking of access to MMIO memory
1718 strict regions from userspace.
1733 nobypass [PPC/POWERNV]
1734 Disable IOMMU bypass, using IOMMU for PCI devices.
1737 io7= [HW] IO7 for Marvel based alpha systems
1738 See comment before marvel_specify_io7 in
1739 arch/alpha/kernel/core_marvel.c.
1741 io_delay= [X86] I/O delay method
1743 Standard port 0x80 based delay
1745 Alternate port 0xed based delay (needed on some systems)
1747 Simple two microseconds delay
1752 See Documentation/filesystems/nfs/nfsroot.txt.
1754 irqaffinity= [SMP] Set the default irq affinity mask
1756 <cpu number>,...,<cpu number>
1758 <cpu number>-<cpu number>
1759 (must be a positive range in ascending order)
1761 <cpu number>,...,<cpu number>-<cpu number>
1764 When an interrupt is not handled search all handlers
1765 for it. Intended to get systems with badly broken
1769 When an interrupt is not handled search all handlers
1770 for it. Also check all handlers each timer
1771 interrupt. Intended to get systems with badly broken
1775 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1777 isolcpus= [KNL,SMP] Isolate CPUs from the general scheduler.
1779 <cpu number>,...,<cpu number>
1781 <cpu number>-<cpu number>
1782 (must be a positive range in ascending order)
1784 <cpu number>,...,<cpu number>-<cpu number>
1786 This option can be used to specify one or more CPUs
1787 to isolate from the general SMP balancing and scheduling
1788 algorithms. You can move a process onto or off an
1789 "isolated" CPU via the CPU affinity syscalls or cpuset.
1790 <cpu number> begins at 0 and the maximum value is
1791 "number of CPUs in system - 1".
1793 This option is the preferred way to isolate CPUs. The
1794 alternative -- manually setting the CPU mask of all
1795 tasks in the system -- can cause problems and
1796 suboptimal load balancer performance.
1800 ivrs_ioapic [HW,X86_64]
1801 Provide an override to the IOAPIC-ID<->DEVICE-ID
1802 mapping provided in the IVRS ACPI table. For
1803 example, to map IOAPIC-ID decimal 10 to
1804 PCI device 00:14.0 write the parameter as:
1805 ivrs_ioapic[10]=00:14.0
1807 ivrs_hpet [HW,X86_64]
1808 Provide an override to the HPET-ID<->DEVICE-ID
1809 mapping provided in the IVRS ACPI table. For
1810 example, to map HPET-ID decimal 0 to
1811 PCI device 00:14.0 write the parameter as:
1812 ivrs_hpet[0]=00:14.0
1814 ivrs_acpihid [HW,X86_64]
1815 Provide an override to the ACPI-HID:UID<->DEVICE-ID
1816 mapping provided in the IVRS ACPI table. For
1817 example, to map UART-HID:UID AMD0020:0 to
1818 PCI device 00:14.5 write the parameter as:
1819 ivrs_acpihid[00:14.5]=AMD0020:0
1821 js= [HW,JOY] Analog joystick
1822 See Documentation/input/joystick.txt.
1825 When CONFIG_RANDOMIZE_BASE is set, this disables
1826 kernel and module base offset ASLR (Address Space
1827 Layout Randomization).
1831 kernelcore= [KNL,X86,IA-64,PPC]
1832 Format: nn[KMGTPE] | "mirror"
1834 specifies the amount of memory usable by the kernel
1835 for non-movable allocations. The requested amount is
1836 spread evenly throughout all nodes in the system. The
1837 remaining memory in each node is used for Movable
1838 pages. In the event, a node is too small to have both
1839 kernelcore and Movable pages, kernelcore pages will
1840 take priority and other nodes will have a larger number
1841 of Movable pages. The Movable zone is used for the
1842 allocation of pages that may be reclaimed or moved
1843 by the page migration subsystem. This means that
1844 HugeTLB pages may not be allocated from this zone.
1845 Note that allocations like PTEs-from-HighMem still
1846 use the HighMem zone if it exists, and the Normal
1847 zone if it does not.
1849 Instead of specifying the amount of memory (nn[KMGTPE]),
1850 you can specify "mirror" option. In case "mirror"
1851 option is specified, mirrored (reliable) memory is used
1852 for non-movable allocations and remaining memory is used
1853 for Movable pages. nn[KMGTPE] and "mirror" are exclusive,
1854 so you can NOT specify nn[KMGTPE] and "mirror" at the same
1857 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
1858 Format: <Controller#>[,poll interval]
1859 The controller # is the number of the ehci usb debug
1860 port as it is probed via PCI. The poll interval is
1861 optional and is the number seconds in between
1862 each poll cycle to the debug port in case you need
1863 the functionality for interrupting the kernel with
1864 gdb or control-c on the dbgp connection. When
1865 not using this parameter you use sysrq-g to break into
1866 the kernel debugger.
1868 kgdboc= [KGDB,HW] kgdb over consoles.
1869 Requires a tty driver that supports console polling,
1870 or a supported polling keyboard driver (non-usb).
1871 Serial only format: <serial_device>[,baud]
1872 keyboard only format: kbd
1873 keyboard and serial format: kbd,<serial_device>[,baud]
1874 Optional Kernel mode setting:
1875 kms, kbd format: kms,kbd
1876 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
1878 kgdbwait [KGDB] Stop kernel execution and enter the
1879 kernel debugger at the earliest opportunity.
1881 kmac= [MIPS] korina ethernet MAC address.
1882 Configure the RouterBoard 532 series on-chip
1883 Ethernet adapter MAC address.
1885 kmemleak= [KNL] Boot-time kmemleak enable/disable
1886 Valid arguments: on, off
1888 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
1891 kmemcheck= [X86] Boot-time kmemcheck enable/disable/one-shot mode
1892 Valid arguments: 0, 1, 2
1893 kmemcheck=0 (disabled)
1894 kmemcheck=1 (enabled)
1895 kmemcheck=2 (one-shot mode)
1896 Default: 2 (one-shot mode)
1898 kstack=N [X86] Print N words from the kernel stack
1901 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
1902 Default is 0 (don't ignore, but inject #GP)
1904 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
1908 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
1909 Default is 1 (enabled)
1911 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
1913 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
1915 kvm-intel.ept= [KVM,Intel] Disable extended page tables
1916 (virtualized MMU) support on capable Intel chips.
1917 Default is 1 (enabled)
1919 kvm-intel.emulate_invalid_guest_state=
1920 [KVM,Intel] Enable emulation of invalid guest states
1921 Default is 0 (disabled)
1923 kvm-intel.flexpriority=
1924 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
1925 Default is 1 (enabled)
1928 [KVM,Intel] Enable VMX nesting (nVMX).
1929 Default is 0 (disabled)
1931 kvm-intel.unrestricted_guest=
1932 [KVM,Intel] Disable unrestricted guest feature
1933 (virtualized real and unpaged mode) on capable
1934 Intel chips. Default is 1 (enabled)
1936 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
1937 feature (tagged TLBs) on capable Intel chips.
1938 Default is 1 (enabled)
1944 lapic [X86-32,APIC] Enable the local APIC even if BIOS
1947 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
1948 value for LAPIC timer one-shot implementation. Default
1949 back to the programmable timer unit in the LAPIC.
1951 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
1954 libata.dma= [LIBATA] DMA control
1955 libata.dma=0 Disable all PATA and SATA DMA
1956 libata.dma=1 PATA and SATA Disk DMA only
1957 libata.dma=2 ATAPI (CDROM) DMA only
1958 libata.dma=4 Compact Flash DMA only
1959 Combinations also work, so libata.dma=3 enables DMA
1960 for disks and CDROMs, but not CFs.
1962 libata.ignore_hpa= [LIBATA] Ignore HPA limit
1963 libata.ignore_hpa=0 keep BIOS limits (default)
1964 libata.ignore_hpa=1 ignore limits, using full disk
1966 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
1970 libata.force= [LIBATA] Force configurations. The format is comma
1971 separated list of "[ID:]VAL" where ID is
1972 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
1973 matching port, link or device. Basically, it matches
1974 the ATA ID string printed on console by libata. If
1975 the whole ID part is omitted, the last PORT and DEVICE
1976 values are used. If ID hasn't been specified yet, the
1977 configuration applies to all ports, links and devices.
1979 If only DEVICE is omitted, the parameter applies to
1980 the port and all links and devices behind it. DEVICE
1981 number of 0 either selects the first device or the
1982 first fan-out link behind PMP device. It does not
1983 select the host link. DEVICE number of 15 selects the
1984 host link and device attached to it.
1986 The VAL specifies the configuration to force. As long
1987 as there's no ambiguity shortcut notation is allowed.
1988 For example, both 1.5 and 1.5G would work for 1.5Gbps.
1989 The following configurations can be forced.
1991 * Cable type: 40c, 80c, short40c, unk, ign or sata.
1992 Any ID with matching PORT is used.
1994 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
1996 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
1997 udma[/][16,25,33,44,66,100,133] notation is also
2000 * [no]ncq: Turn on or off NCQ.
2002 * [no]ncqtrim: Turn off queued DSM TRIM.
2004 * nohrst, nosrst, norst: suppress hard, soft
2007 * rstonce: only attempt one reset during
2008 hot-unplug link recovery
2010 * dump_id: dump IDENTIFY data.
2012 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2014 * disable: Disable this device.
2016 If there are multiple matching configurations changing
2017 the same attribute, the last one is used.
2019 memblock=debug [KNL] Enable memblock debug messages.
2021 load_ramdisk= [RAM] List of ramdisks to load from floppy
2022 See Documentation/blockdev/ramdisk.txt.
2024 lockd.nlm_grace_period=P [NFS] Assign grace period.
2027 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2030 lockd.nlm_timeout=T [NFS] Assign timeout value.
2033 lockd.nlm_udpport=M [NFS] Assign UDP port.
2036 locktorture.nreaders_stress= [KNL]
2037 Set the number of locking read-acquisition kthreads.
2038 Defaults to being automatically set based on the
2039 number of online CPUs.
2041 locktorture.nwriters_stress= [KNL]
2042 Set the number of locking write-acquisition kthreads.
2044 locktorture.onoff_holdoff= [KNL]
2045 Set time (s) after boot for CPU-hotplug testing.
2047 locktorture.onoff_interval= [KNL]
2048 Set time (s) between CPU-hotplug operations, or
2049 zero to disable CPU-hotplug testing.
2051 locktorture.shuffle_interval= [KNL]
2052 Set task-shuffle interval (jiffies). Shuffling
2053 tasks allows some CPUs to go into dyntick-idle
2054 mode during the locktorture test.
2056 locktorture.shutdown_secs= [KNL]
2057 Set time (s) after boot system shutdown. This
2058 is useful for hands-off automated testing.
2060 locktorture.stat_interval= [KNL]
2061 Time (s) between statistics printk()s.
2063 locktorture.stutter= [KNL]
2064 Time (s) to stutter testing, for example,
2065 specifying five seconds causes the test to run for
2066 five seconds, wait for five seconds, and so on.
2067 This tests the locking primitive's ability to
2068 transition abruptly to and from idle.
2070 locktorture.torture_runnable= [BOOT]
2071 Start locktorture running at boot time.
2073 locktorture.torture_type= [KNL]
2074 Specify the locking implementation to test.
2076 locktorture.verbose= [KNL]
2077 Enable additional printk() statements.
2079 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2082 loglevel= All Kernel Messages with a loglevel smaller than the
2083 console loglevel will be printed to the console. It can
2084 also be changed with klogd or other programs. The
2085 loglevels are defined as follows:
2087 0 (KERN_EMERG) system is unusable
2088 1 (KERN_ALERT) action must be taken immediately
2089 2 (KERN_CRIT) critical conditions
2090 3 (KERN_ERR) error conditions
2091 4 (KERN_WARNING) warning conditions
2092 5 (KERN_NOTICE) normal but significant condition
2093 6 (KERN_INFO) informational
2094 7 (KERN_DEBUG) debug-level messages
2096 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2097 in bytes. n must be a power of two and greater
2098 than the minimal size. The minimal size is defined
2099 by LOG_BUF_SHIFT kernel config parameter. There is
2100 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2101 that allows to increase the default size depending on
2102 the number of CPUs. See init/Kconfig for more details.
2104 logo.nologo [FB] Disables display of the built-in Linux logo.
2105 This may be used to provide more screen space for
2106 kernel log messages and is useful when debugging
2107 kernel boot problems.
2109 lp=0 [LP] Specify parallel ports to use, e.g,
2110 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2111 lp=reset first parallel port). 'lp=0' disables the
2112 lp=auto printer driver. 'lp=reset' (which can be
2113 specified in addition to the ports) causes
2114 attached printers to be reset. Using
2115 lp=port1,port2,... specifies the parallel ports
2116 to associate lp devices with, starting with
2117 lp0. A port specification may be 'none' to skip
2118 that lp device, or a parport name such as
2119 'parport0'. Specifying 'lp=auto' instead of a
2120 port specification list means that device IDs
2121 from each port should be examined, to see if
2122 an IEEE 1284-compliant printer is attached; if
2123 so, the driver will manage that printer.
2124 See also header of drivers/char/lp.c.
2127 Sets loops_per_jiffy to given constant, thus avoiding
2128 time-consuming boot-time autodetection (up to 250 ms per
2129 CPU). 0 enables autodetection (default). To determine
2130 the correct value for your kernel, boot with normal
2131 autodetection and see what value is printed. Note that
2132 on SMP systems the preset will be applied to all CPUs,
2133 which is likely to cause problems if your CPUs need
2134 significantly divergent settings. An incorrect value
2135 will cause delays in the kernel to be wrong, leading to
2136 unpredictable I/O errors and other breakage. Although
2137 unlikely, in the extreme case this might damage your
2141 Format: <io>,<irq>,<dma>
2143 machvec= [IA-64] Force the use of a particular machine-vector
2144 (machvec) in a generic kernel.
2145 Example: machvec=hpzx1_swiotlb
2147 machtype= [Loongson] Share the same kernel image file between different
2149 Example: machtype=lemote-yeeloong-2f-7inch
2151 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2152 than or equal to this physical address is ignored.
2154 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2155 should make use of. maxcpus=n : n >= 0 limits the
2156 kernel to using 'n' processors. n=0 is a special case,
2157 it is equivalent to "nosmp", which also disables
2160 max_loop= [LOOP] The number of loop block devices that get
2161 (loop.max_loop) unconditionally pre-created at init time. The default
2162 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2163 of statically allocating a predefined number, loop
2164 devices can be requested on-demand with the
2165 /dev/loop-control interface.
2167 mce [X86-32] Machine Check Exception
2169 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.txt
2171 md= [HW] RAID subsystems devices and level
2172 See Documentation/md.txt.
2175 Format: <first>,<last>
2176 Specifies range of consoles to be captured by the MDA.
2178 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2179 Amount of memory to be used when the kernel is not able
2180 to see the whole system memory or for test.
2181 [X86] Work as limiting max address. Use together
2182 with memmap= to avoid physical address space collisions.
2183 Without memmap= PCI devices could be placed at addresses
2184 belonging to unused RAM.
2186 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2190 [KNL,SH] Allow user to override the default size for
2191 per-device physically contiguous DMA buffers.
2193 memhp_default_state=online/offline
2194 [KNL] Set the initial state for the memory hotplug
2195 onlining policy. If not specified, the default value is
2196 set according to the
2197 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2199 See Documentation/memory-hotplug.txt.
2201 memmap=exactmap [KNL,X86] Enable setting of an exact
2202 E820 memory map, as specified by the user.
2203 Such memmap=exactmap lines can be constructed based on
2204 BIOS output or other requirements. See the memmap=nn@ss
2207 memmap=nn[KMG]@ss[KMG]
2208 [KNL] Force usage of a specific region of memory.
2209 Region of memory to be used is from ss to ss+nn.
2211 memmap=nn[KMG]#ss[KMG]
2212 [KNL,ACPI] Mark specific memory as ACPI data.
2213 Region of memory to be marked is from ss to ss+nn.
2215 memmap=nn[KMG]$ss[KMG]
2216 [KNL,ACPI] Mark specific memory as reserved.
2217 Region of memory to be reserved is from ss to ss+nn.
2218 Example: Exclude memory from 0x18690000-0x1869ffff
2219 memmap=64K$0x18690000
2221 memmap=0x10000$0x18690000
2223 memmap=nn[KMG]!ss[KMG]
2224 [KNL,X86] Mark specific memory as protected.
2225 Region of memory to be used, from ss to ss+nn.
2226 The memory region may be marked as e820 type 12 (0xc)
2227 and is NVDIMM or ADR memory.
2229 memory_corruption_check=0/1 [X86]
2230 Some BIOSes seem to corrupt the first 64k of
2231 memory when doing things like suspend/resume.
2232 Setting this option will scan the memory
2233 looking for corruption. Enabling this will
2234 both detect corruption and prevent the kernel
2235 from using the memory being corrupted.
2236 However, its intended as a diagnostic tool; if
2237 repeatable BIOS-originated corruption always
2238 affects the same memory, you can use memmap=
2239 to prevent the kernel from using that memory.
2241 memory_corruption_check_size=size [X86]
2242 By default it checks for corruption in the low
2243 64k, making this memory unavailable for normal
2244 use. Use this parameter to scan for
2245 corruption in more or less memory.
2247 memory_corruption_check_period=seconds [X86]
2248 By default it checks for corruption every 60
2249 seconds. Use this parameter to check at some
2250 other rate. 0 disables periodic checking.
2252 memtest= [KNL,X86,ARM] Enable memtest
2254 default : 0 <disable>
2255 Specifies the number of memtest passes to be
2256 performed. Each pass selects another test
2257 pattern from a given set of patterns. Memtest
2258 fills the memory with this pattern, validates
2259 memory contents and reserves bad memory
2260 regions that are detected.
2262 meye.*= [HW] Set MotionEye Camera parameters
2263 See Documentation/video4linux/meye.txt.
2265 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2266 Multi-Function General Purpose Timers on AMD Geode
2269 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2270 the BIOS has incorrectly applied a workaround. TinyBIOS
2271 version 0.98 is known to be affected, 0.99 fixes the
2272 problem by letting the user disable the workaround.
2276 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2277 physical address is ignored.
2279 mini2440= [ARM,HW,KNL]
2280 Format:[0..2][b][c][t]
2282 MINI2440 configuration specification:
2283 0 - The attached screen is the 3.5" TFT
2284 1 - The attached screen is the 7" TFT
2285 2 - The VGA Shield is attached (1024x768)
2286 Leaving out the screen size parameter will not load
2287 the TFT driver, and the framebuffer will be left
2289 b - Enable backlight. The TFT backlight pin will be
2290 linked to the kernel VESA blanking code and a GPIO
2291 LED. This parameter is not necessary when using the
2293 c - Enable the s3c camera interface.
2294 t - Reserved for enabling touchscreen support. The
2295 touchscreen support is not enabled in the mainstream
2296 kernel as of 2.6.30, a preliminary port can be found
2297 in the "bleeding edge" mini2440 support kernel at
2298 http://repo.or.cz/w/linux-2.6/mini2440.git
2301 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2302 parameter allows control of the logging verbosity for
2303 the additional memory initialisation checks. A value
2304 of 0 disables mminit logging and a level of 4 will
2305 log everything. Information is printed at KERN_DEBUG
2306 so loglevel=8 may also need to be specified.
2309 [KNL] When CONFIG_MODULE_SIG is set, this means that
2310 modules without (valid) signatures will fail to load.
2311 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2312 is always true, so this option does nothing.
2315 [MOUSE] Maximum time between finger touching and
2316 leaving touchpad surface for touch to be considered
2317 a tap and be reported as a left button click (for
2318 touchpads working in absolute mode only).
2320 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2321 reporting absolute coordinates, such as tablets
2322 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2323 reporting absolute coordinates, such as tablets
2325 movablecore=nn[KMG] [KNL,X86,IA-64,PPC] This parameter
2326 is similar to kernelcore except it specifies the
2327 amount of memory used for migratable allocations.
2328 If both kernelcore and movablecore is specified,
2329 then kernelcore will be at *least* the specified
2330 value but may be more. If movablecore on its own
2331 is specified, the administrator must be careful
2332 that the amount of memory usable for all allocations
2335 movable_node [KNL,X86] Boot-time switch to enable the effects
2336 of CONFIG_MOVABLE_NODE=y. See mm/Kconfig for details.
2338 MTD_Partition= [MTD]
2339 Format: <name>,<region-number>,<size>,<offset>
2341 MTD_Region= [MTD] Format:
2342 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2345 See drivers/mtd/cmdlinepart.c.
2347 multitce=off [PPC] This parameter disables the use of the pSeries
2348 firmware feature for updating multiple TCE entries
2351 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2353 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2355 boundary - index of last SLC block on Flex-OneNAND.
2356 The remaining blocks are configured as MLC blocks.
2357 lock - Configure if Flex-OneNAND boundary should be locked.
2358 Once locked, the boundary cannot be changed.
2359 1 indicates lock status, 0 indicates unlock status.
2362 ARM/S3C2412 JIVE boot control
2364 See arch/arm/mach-s3c2412/mach-jive.c
2366 mtouchusb.raw_coordinates=
2367 [HW] Make the MicroTouch USB driver use raw coordinates
2368 ('y', default) or cooked coordinates ('n')
2370 mtrr_chunk_size=nn[KMG] [X86]
2371 used for mtrr cleanup. It is largest continuous chunk
2372 that could hold holes aka. UC entries.
2374 mtrr_gran_size=nn[KMG] [X86]
2375 Used for mtrr cleanup. It is granularity of mtrr block.
2377 Large value could prevent small alignment from
2380 mtrr_spare_reg_nr=n [X86]
2382 Range: 0,7 : spare reg number
2384 Used for mtrr cleanup. It is spare mtrr entries number.
2385 Set to 2 or more if your graphical card needs more.
2387 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2389 netdev= [NET] Network devices parameters
2390 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2391 Note that mem_start is often overloaded to mean
2392 something different and driver-specific.
2393 This usage is only documented in each driver source
2397 [NETFILTER] Enable connection tracking flow accounting
2398 0 to disable accounting
2399 1 to enable accounting
2402 nfsaddrs= [NFS] Deprecated. Use ip= instead.
2403 See Documentation/filesystems/nfs/nfsroot.txt.
2405 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
2406 See Documentation/filesystems/nfs/nfsroot.txt.
2408 nfsrootdebug [NFS] enable nfsroot debugging messages.
2409 See Documentation/filesystems/nfs/nfsroot.txt.
2411 nfs.callback_tcpport=
2412 [NFS] set the TCP port on which the NFSv4 callback
2413 channel should listen.
2416 [NFS] sets the pathname to the program which is used
2417 to update the NFS client cache entries.
2419 nfs.cache_getent_timeout=
2420 [NFS] sets the timeout after which an attempt to
2421 update a cache entry is deemed to have failed.
2423 nfs.idmap_cache_timeout=
2424 [NFS] set the maximum lifetime for idmapper cache
2428 [NFS] enable 64-bit inode numbers.
2429 If zero, the NFS client will fake up a 32-bit inode
2430 number for the readdir() and stat() syscalls instead
2431 of returning the full 64-bit number.
2432 The default is to return 64-bit inode numbers.
2434 nfs.max_session_slots=
2435 [NFSv4.1] Sets the maximum number of session slots
2436 the client will attempt to negotiate with the server.
2437 This limits the number of simultaneous RPC requests
2438 that the client can send to the NFSv4.1 server.
2439 Note that there is little point in setting this
2440 value higher than the max_tcp_slot_table_limit.
2442 nfs.nfs4_disable_idmapping=
2443 [NFSv4] When set to the default of '1', this option
2444 ensures that both the RPC level authentication
2445 scheme and the NFS level operations agree to use
2446 numeric uids/gids if the mount is using the
2447 'sec=sys' security flavour. In effect it is
2448 disabling idmapping, which can make migration from
2449 legacy NFSv2/v3 systems to NFSv4 easier.
2450 Servers that do not support this mode of operation
2451 will be autodetected by the client, and it will fall
2452 back to using the idmapper.
2453 To turn off this behaviour, set the value to '0'.
2455 [NFS4] Specify an additional fixed unique ident-
2456 ification string that NFSv4 clients can insert into
2457 their nfs_client_id4 string. This is typically a
2458 UUID that is generated at system install time.
2460 nfs.send_implementation_id =
2461 [NFSv4.1] Send client implementation identification
2462 information in exchange_id requests.
2463 If zero, no implementation identification information
2465 The default is to send the implementation identification
2468 nfs.recover_lost_locks =
2469 [NFSv4] Attempt to recover locks that were lost due
2470 to a lease timeout on the server. Please note that
2471 doing this risks data corruption, since there are
2472 no guarantees that the file will remain unchanged
2473 after the locks are lost.
2474 If you want to enable the kernel legacy behaviour of
2475 attempting to recover these locks, then set this
2477 The default parameter value of '0' causes the kernel
2478 not to attempt recovery of lost locks.
2480 nfs4.layoutstats_timer =
2481 [NFSv4.2] Change the rate at which the kernel sends
2482 layoutstats to the pNFS metadata server.
2484 Setting this to value to 0 causes the kernel to use
2485 whatever value is the default set by the layout
2486 driver. A non-zero value sets the minimum interval
2487 in seconds between layoutstats transmissions.
2489 nfsd.nfs4_disable_idmapping=
2490 [NFSv4] When set to the default of '1', the NFSv4
2491 server will return only numeric uids and gids to
2492 clients using auth_sys, and will accept numeric uids
2493 and gids from such clients. This is intended to ease
2494 migration from NFSv2/v3.
2496 objlayoutdriver.osd_login_prog=
2497 [NFS] [OBJLAYOUT] sets the pathname to the program which
2498 is used to automatically discover and login into new
2499 osd-targets. Please see:
2500 Documentation/filesystems/pnfs.txt for more explanations
2502 nmi_debug= [KNL,AVR32,SH] Specify one or more actions to take
2503 when a NMI is triggered.
2504 Format: [state][,regs][,debounce][,die]
2506 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
2507 Format: [panic,][nopanic,][num]
2509 0 - turn hardlockup detector in nmi_watchdog off
2510 1 - turn hardlockup detector in nmi_watchdog on
2511 When panic is specified, panic when an NMI watchdog
2512 timeout occurs (or 'nopanic' to override the opposite
2513 default). To disable both hard and soft lockup detectors,
2514 please see 'nowatchdog'.
2515 This is useful when you use a panic=... timeout and
2516 need the box quickly up again.
2518 netpoll.carrier_timeout=
2519 [NET] Specifies amount of time (in seconds) that
2520 netpoll should wait for a carrier. By default netpoll
2523 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
2524 emulation library even if a 387 maths coprocessor
2528 [HW] Never suspend the console
2529 Disable suspending of consoles during suspend and
2530 hibernate operations. Once disabled, debugging
2531 messages can reach various consoles while the rest
2532 of the system is being put to sleep (ie, while
2533 debugging driver suspend/resume hooks). This may
2534 not work reliably with all consoles, but is known
2535 to work with serial and VGA consoles.
2536 To facilitate more flexible debugging, we also add
2537 console_suspend, a printk module parameter to control
2538 it. Users could use console_suspend (usually
2539 /sys/module/printk/parameters/console_suspend) to
2540 turn on/off it dynamically.
2542 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
2543 caches in the slab allocator. Saves per-node memory,
2544 but will impact performance.
2548 noapic [SMP,APIC] Tells the kernel to not make use of any
2549 IOAPICs that may be present in the system.
2551 noautogroup Disable scheduler automatic task group creation.
2553 nobats [PPC] Do not use BATs for mapping kernel lowmem
2554 on "Classic" PPC cores.
2558 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
2560 nodelayacct [KNL] Disable per-task delay accounting
2562 nodisconnect [HW,SCSI,M68K] Disables SCSI disconnects.
2564 nodsp [SH] Disable hardware DSP at boot time.
2566 noefi Disable EFI runtime services support.
2571 On X86-32 available only on PAE configured kernels.
2572 noexec=on: enable non-executable mappings (default)
2573 noexec=off: disable non-executable mappings
2576 Disable SMAP (Supervisor Mode Access Prevention)
2577 even if it is supported by processor.
2580 Disable SMEP (Supervisor Mode Execution Prevention)
2581 even if it is supported by processor.
2584 This affects only 32-bit executables.
2585 noexec32=on: enable non-executable mappings (default)
2586 read doesn't imply executable mappings
2587 noexec32=off: disable non-executable mappings
2588 read implies executable mappings
2590 nofpu [MIPS,SH] Disable hardware FPU at boot time.
2592 nofxsr [BUGS=X86-32] Disables x86 floating point extended
2593 register save and restore. The kernel will only save
2594 legacy floating-point registers on task switch.
2596 nohugeiomap [KNL,x86] Disable kernel huge I/O mappings.
2598 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
2599 Equivalent to smt=1.
2601 noxsave [BUGS=X86] Disables x86 extended register state save
2602 and restore using xsave. The kernel will fallback to
2603 enabling legacy floating-point and sse state.
2605 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
2606 register states. The kernel will fall back to use
2607 xsave to save the states. By using this parameter,
2608 performance of saving the states is degraded because
2609 xsave doesn't support modified optimization while
2610 xsaveopt supports it on xsaveopt enabled systems.
2612 noxsaves [X86] Disables xsaves and xrstors used in saving and
2613 restoring x86 extended register state in compacted
2614 form of xsave area. The kernel will fall back to use
2615 xsaveopt and xrstor to save and restore the states
2616 in standard form of xsave area. By using this
2617 parameter, xsave area per process might occupy more
2618 memory on xsaves enabled systems.
2620 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
2621 wfi(ARM) instruction doesn't work correctly and not to
2622 use it. This is also useful when using JTAG debugger.
2624 no_file_caps Tells the kernel not to honor file capabilities. The
2625 only way then for a file to be executed with privilege
2626 is to be setuid root or executed by root.
2628 nohalt [IA-64] Tells the kernel not to use the power saving
2629 function PAL_HALT_LIGHT when idle. This increases
2630 power-consumption. On the positive side, it reduces
2631 interrupt wake-up latency, which may improve performance
2632 in certain environments such as networked servers or
2635 nohibernate [HIBERNATION] Disable hibernation and resume.
2637 nohz= [KNL] Boottime enable/disable dynamic ticks
2638 Valid arguments: on, off
2641 nohz_full= [KNL,BOOT]
2642 In kernels built with CONFIG_NO_HZ_FULL=y, set
2643 the specified list of CPUs whose tick will be stopped
2644 whenever possible. The boot CPU will be forced outside
2645 the range to maintain the timekeeping.
2646 The CPUs in this range must also be included in the
2649 noiotrap [SH] Disables trapped I/O port accesses.
2651 noirqdebug [X86-32] Disables the code which attempts to detect and
2652 disable unhandled interrupt sources.
2654 no_timer_check [X86,APIC] Disables the code which tests for
2655 broken timer IRQ sources.
2657 noisapnp [ISAPNP] Disables ISA PnP code.
2659 noinitrd [RAM] Tells the kernel not to load any configured
2662 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
2664 [Deprecated - use intremap=off]
2668 noinvpcid [X86] Disable the INVPCID cpu feature.
2670 nojitter [IA-64] Disables jitter checking for ITC timers.
2672 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
2674 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
2677 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting.
2678 steal time is computed, but won't influence scheduler
2681 nolapic [X86-32,APIC] Do not enable or use the local APIC.
2683 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
2685 noltlbs [PPC] Do not use large page/tlb entries for kernel
2686 lowmem mapping on PPC40x and PPC8xx
2688 nomca [IA-64] Disable machine check abort handling
2690 nomce [X86-32] Disable Machine Check Exception
2692 nomfgpt [X86-32] Disable Multi-Function General Purpose
2693 Timer usage (for AMD Geode machines).
2695 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
2696 shutdown the other cpus. Instead use the REBOOT_VECTOR
2699 nomodule Disable module load
2701 nopat [X86] Disable PAT (page attribute table extension of
2702 pagetables) support.
2704 norandmaps Don't use address space randomization. Equivalent to
2705 echo 0 > /proc/sys/kernel/randomize_va_space
2707 noreplace-paravirt [X86,IA-64,PV_OPS] Don't patch paravirt_ops
2709 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
2710 with UP alternatives
2712 nordrand [X86] Disable kernel use of the RDRAND and
2713 RDSEED instructions even if they are supported
2714 by the processor. RDRAND and RDSEED are still
2715 available to user space applications.
2717 noresume [SWSUSP] Disables resume and restores original swap
2720 no-scroll [VGA] Disables scrollback.
2721 This is required for the Braillex ib80-piezo Braille
2722 reader made by F.H. Papenmeier (Germany).
2726 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
2728 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
2729 and disable the IO APIC. legacy for "maxcpus=0".
2731 nosoftlockup [KNL] Disable the soft-lockup detector.
2733 nosync [HW,M68K] Disables sync negotiation for all devices.
2735 notsc [BUGS=X86-32] Disable Time Stamp Counter
2737 nowatchdog [KNL] Disable both lockup detectors, i.e.
2738 soft-lockup and NMI watchdog (hard-lockup).
2742 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
2744 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
2745 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
2746 Some features depend on CPU0. Known dependencies are:
2747 1. Resume from suspend/hibernate depends on CPU0.
2748 Suspend/hibernate will fail if CPU0 is offline and you
2749 need to online CPU0 before suspend/hibernate.
2750 2. PIC interrupts also depend on CPU0. CPU0 can't be
2751 removed if a PIC interrupt is detected.
2752 It's said poweroff/reboot may depend on CPU0 on some
2753 machines although I haven't seen such issues so far
2754 after CPU0 is offline on a few tested machines.
2755 If the dependencies are under your control, you can
2756 turn on cpu0_hotplug.
2758 nptcg= [IA-64] Override max number of concurrent global TLB
2759 purges which is reported from either PAL_VM_SUMMARY or
2762 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
2763 could support. nr_cpus=n : n >= 1 limits the kernel to
2764 supporting 'n' processors. Later in runtime you can not
2765 use hotplug cpu feature to put more cpu back to online.
2766 just like you compile the kernel NR_CPUS=n
2768 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
2770 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
2771 Allowed values are enable and disable
2773 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
2774 one of ['zone', 'node', 'default'] can be specified
2775 This can be set from sysctl after boot.
2776 See Documentation/sysctl/vm.txt for details.
2778 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
2779 See Documentation/debugging-via-ohci1394.txt for more
2782 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
2783 Rather than timing out after 20 ms if an EC
2784 command is not properly ACKed, override the length
2785 of the timeout. We have interrupts disabled while
2786 waiting for the ACK, so if this is set too high
2787 interrupts *may* be lost!
2789 omap_mux= [OMAP] Override bootloader pin multiplexing.
2790 Format: <mux_mode0.mode_name=value>...
2791 For example, to override I2C bus2:
2792 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
2794 oprofile.timer= [HW]
2795 Use timer interrupt instead of performance counters
2797 oprofile.cpu_type= Force an oprofile cpu type
2798 This might be useful if you have an older oprofile
2799 userland or if you want common events.
2800 Format: { arch_perfmon }
2801 arch_perfmon: [X86] Force use of architectural
2802 perfmon on Intel CPUs instead of the
2803 CPU specific event set.
2804 timer: [X86] Force use of architectural NMI
2805 timer mode (see also oprofile.timer
2806 for generic hr timer mode)
2808 oops=panic Always panic on oopses. Default is to just kill the
2809 process, but there is a small probability of
2810 deadlocking the machine.
2811 This will also cause panics on machine check exceptions.
2812 Useful together with panic=30 to trigger a reboot.
2815 See Documentation/sound/oss/oss-parameters.txt
2817 page_owner= [KNL] Boot-time page_owner enabling option.
2818 Storage of the information about who allocated
2819 each page is disabled in default. With this switch,
2821 on: enable the feature
2823 page_poison= [KNL] Boot-time parameter changing the state of
2824 poisoning on the buddy allocator.
2825 off: turn off poisoning
2826 on: turn on poisoning
2828 panic= [KNL] Kernel behaviour on panic: delay <timeout>
2829 timeout > 0: seconds before rebooting
2830 timeout = 0: wait forever
2831 timeout < 0: reboot immediately
2834 panic_on_warn panic() instead of WARN(). Useful to cause kdump
2837 crash_kexec_post_notifiers
2838 Run kdump after running panic-notifiers and dumping
2839 kmsg. This only for the users who doubt kdump always
2840 succeeds in any situation.
2841 Note that this also increases risks of kdump failure,
2842 because some panic notifiers can make the crashed
2843 kernel more unstable.
2845 parkbd.port= [HW] Parallel port number the keyboard adapter is
2846 connected to, default is 0.
2848 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
2849 0 for XT, 1 for AT (default is AT).
2852 parport= [HW,PPT] Specify parallel ports. 0 disables.
2853 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
2854 Use 'auto' to force the driver to use any
2855 IRQ/DMA settings detected (the default is to
2856 ignore detected IRQ/DMA settings because of
2857 possible conflicts). You can specify the base
2858 address, IRQ, and DMA settings; IRQ and DMA
2859 should be numbers, or 'auto' (for using detected
2860 settings on that particular port), or 'nofifo'
2861 (to avoid using a FIFO even if it is detected).
2862 Parallel ports are assigned in the order they
2863 are specified on the command line, starting
2866 parport_init_mode= [HW,PPT]
2867 Configure VIA parallel port to operate in
2868 a specific mode. This is necessary on Pegasos
2869 computer where firmware has no options for setting
2870 up parallel port mode and sets it to spp.
2871 Currently this function knows 686a and 8231 chips.
2872 Format: [spp|ps2|epp|ecp|ecpepp]
2875 Halt all CPUs after the first oops has been printed for
2876 the specified number of seconds. This is to be used if
2877 your oopses keep scrolling off the screen.
2882 See header of drivers/block/paride/pcd.c.
2883 See also Documentation/blockdev/paride.txt.
2885 pci=option[,option...] [PCI] various PCI subsystem options:
2886 earlydump [X86] dump PCI config space before the kernel
2888 off [X86] don't probe for the PCI bus
2889 bios [X86-32] force use of PCI BIOS, don't access
2890 the hardware directly. Use this if your machine
2891 has a non-standard PCI host bridge.
2892 nobios [X86-32] disallow use of PCI BIOS, only direct
2893 hardware access methods are allowed. Use this
2894 if you experience crashes upon bootup and you
2895 suspect they are caused by the BIOS.
2896 conf1 [X86] Force use of PCI Configuration Access
2897 Mechanism 1 (config address in IO port 0xCF8,
2898 data in IO port 0xCFC, both 32-bit).
2899 conf2 [X86] Force use of PCI Configuration Access
2900 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
2901 the function, IO port 0xCFA, also 8-bit, sets
2902 bus number. The config space is then accessed
2903 through ports 0xC000-0xCFFF).
2904 See http://wiki.osdev.org/PCI for more info
2905 on the configuration access mechanisms.
2906 noaer [PCIE] If the PCIEAER kernel config parameter is
2907 enabled, this kernel boot option can be used to
2908 disable the use of PCIE advanced error reporting.
2909 nodomains [PCI] Disable support for multiple PCI
2910 root domains (aka PCI segments, in ACPI-speak).
2911 nommconf [X86] Disable use of MMCONFIG for PCI
2913 check_enable_amd_mmconf [X86] check for and enable
2914 properly configured MMIO access to PCI
2915 config space on AMD family 10h CPU
2916 nomsi [MSI] If the PCI_MSI kernel config parameter is
2917 enabled, this kernel boot option can be used to
2918 disable the use of MSI interrupts system-wide.
2919 noioapicquirk [APIC] Disable all boot interrupt quirks.
2920 Safety option to keep boot IRQs enabled. This
2921 should never be necessary.
2922 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
2923 primary IO-APIC for bridges that cannot disable
2924 boot IRQs. This fixes a source of spurious IRQs
2925 when the system masks IRQs.
2926 noioapicreroute [APIC] Disable workaround that uses the
2927 boot IRQ equivalent of an IRQ that connects to
2928 a chipset where boot IRQs cannot be disabled.
2929 The opposite of ioapicreroute.
2930 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
2931 routing table. These calls are known to be buggy
2932 on several machines and they hang the machine
2933 when used, but on other computers it's the only
2934 way to get the interrupt routing table. Try
2935 this option if the kernel is unable to allocate
2936 IRQs or discover secondary PCI buses on your
2938 rom [X86] Assign address space to expansion ROMs.
2939 Use with caution as certain devices share
2940 address decoders between ROMs and other
2942 norom [X86] Do not assign address space to
2943 expansion ROMs that do not already have
2944 BIOS assigned address ranges.
2945 nobar [X86] Do not assign address space to the
2946 BARs that weren't assigned by the BIOS.
2947 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
2948 assigned automatically to PCI devices. You can
2949 make the kernel exclude IRQs of your ISA cards
2951 pirqaddr=0xAAAAA [X86] Specify the physical address
2952 of the PIRQ table (normally generated
2953 by the BIOS) if it is outside the
2954 F0000h-100000h range.
2955 lastbus=N [X86] Scan all buses thru bus #N. Can be
2956 useful if the kernel is unable to find your
2957 secondary buses and you want to tell it
2958 explicitly which ones they are.
2959 assign-busses [X86] Always assign all PCI bus
2960 numbers ourselves, overriding
2961 whatever the firmware may have done.
2962 usepirqmask [X86] Honor the possible IRQ mask stored
2963 in the BIOS $PIR table. This is needed on
2964 some systems with broken BIOSes, notably
2965 some HP Pavilion N5400 and Omnibook XE3
2966 notebooks. This will have no effect if ACPI
2967 IRQ routing is enabled.
2968 noacpi [X86] Do not use ACPI for IRQ routing
2969 or for PCI scanning.
2970 use_crs [X86] Use PCI host bridge window information
2971 from ACPI. On BIOSes from 2008 or later, this
2972 is enabled by default. If you need to use this,
2973 please report a bug.
2974 nocrs [X86] Ignore PCI host bridge windows from ACPI.
2975 If you need to use this, please report a bug.
2976 routeirq Do IRQ routing for all PCI devices.
2977 This is normally done in pci_enable_device(),
2978 so this option is a temporary workaround
2979 for broken drivers that don't call it.
2980 skip_isa_align [X86] do not align io start addr, so can
2981 handle more pci cards
2982 noearly [X86] Don't do any early type 1 scanning.
2983 This might help on some broken boards which
2984 machine check when some devices' config space
2985 is read. But various workarounds are disabled
2986 and some IOMMU drivers will not work.
2987 bfsort Sort PCI devices into breadth-first order.
2988 This sorting is done to get a device
2989 order compatible with older (<= 2.4) kernels.
2990 nobfsort Don't sort PCI devices into breadth-first order.
2991 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
2992 tuning and use the BIOS-configured MPS defaults.
2993 pcie_bus_safe Set every device's MPS to the largest value
2994 supported by all devices below the root complex.
2995 pcie_bus_perf Set device MPS to the largest allowable MPS
2996 based on its parent bus. Also set MRRS (Max
2997 Read Request Size) to the largest supported
2998 value (no larger than the MPS that the device
2999 or bus can support) for best performance.
3000 pcie_bus_peer2peer Set every device's MPS to 128B, which
3001 every device is guaranteed to support. This
3002 configuration allows peer-to-peer DMA between
3003 any pair of devices, possibly at the cost of
3004 reduced performance. This also guarantees
3005 that hot-added devices will work.
3006 cbiosize=nn[KMG] The fixed amount of bus space which is
3007 reserved for the CardBus bridge's IO window.
3008 The default value is 256 bytes.
3009 cbmemsize=nn[KMG] The fixed amount of bus space which is
3010 reserved for the CardBus bridge's memory
3011 window. The default value is 64 megabytes.
3014 [<order of align>@][<domain>:]<bus>:<slot>.<func>[; ...]
3015 Specifies alignment and device to reassign
3016 aligned memory resources.
3017 If <order of align> is not specified,
3018 PAGE_SIZE is used as alignment.
3019 PCI-PCI bridge can be specified, if resource
3020 windows need to be expanded.
3021 ecrc= Enable/disable PCIe ECRC (transaction layer
3022 end-to-end CRC checking).
3023 bios: Use BIOS/firmware settings. This is the
3027 hpiosize=nn[KMG] The fixed amount of bus space which is
3028 reserved for hotplug bridge's IO window.
3029 Default size is 256 bytes.
3030 hpmemsize=nn[KMG] The fixed amount of bus space which is
3031 reserved for hotplug bridge's memory window.
3032 Default size is 2 megabytes.
3033 realloc= Enable/disable reallocating PCI bridge resources
3034 if allocations done by BIOS are too small to
3035 accommodate resources required by all child
3037 off: Turn realloc off
3039 realloc same as realloc=on
3040 noari do not use PCIe ARI.
3041 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3042 only look for one device below a PCIe downstream
3045 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3048 force Enable ASPM even on devices that claim not to support it.
3049 WARNING: Forcing ASPM on may cause system lockups.
3051 pcie_hp= [PCIE] PCI Express Hotplug driver options:
3052 nomsi Do not use MSI for PCI Express Native Hotplug (this
3053 makes all PCIe ports use INTx for hotplug services).
3055 pcie_ports= [PCIE] PCIe ports handling:
3056 auto Ask the BIOS whether or not to use native PCIe services
3057 associated with PCIe ports (PME, hot-plug, AER). Use
3058 them only if that is allowed by the BIOS.
3059 native Use native PCIe services associated with PCIe ports
3061 compat Treat PCIe ports as PCI-to-PCI bridges, disable the PCIe
3064 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3065 nomsi Do not use MSI for native PCIe PME signaling (this makes
3066 all PCIe root ports use INTx for all services).
3068 pcmv= [HW,PCMCIA] BadgePAD 4
3072 Keep all power-domains already enabled by bootloader on,
3073 even if no driver has claimed them. This is useful
3074 for debug and development, but should not be
3075 needed on a platform with proper driver support.
3078 See Documentation/blockdev/paride.txt.
3080 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3083 See arch/parisc/kernel/pdc_chassis.c
3085 percpu_alloc= Select which percpu first chunk allocator to use.
3086 Currently supported values are "embed" and "page".
3087 Archs may support subset or none of the selections.
3088 See comments in mm/percpu.c for details on each
3089 allocator. This parameter is primarily for debugging
3090 and performance comparison.
3093 See Documentation/blockdev/paride.txt.
3096 See Documentation/blockdev/paride.txt.
3098 pirq= [SMP,APIC] Manual mp-table setup
3099 See Documentation/x86/i386/IO-APIC.txt.
3101 plip= [PPT,NET] Parallel port network link
3102 Format: { parport<nr> | timid | 0 }
3103 See also Documentation/parport.txt.
3105 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3106 Override pmtimer IOPort with a hex value.
3110 Enable PNP debug messages (depends on the
3111 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3112 via /sys/module/pnp/parameters/debug. We always show
3113 current resource usage; turning this on also shows
3114 possible settings and some assignment information.
3120 { on | off | curr | res | no-curr | no-res }
3123 [ISAPNP] Exclude IRQs for the autoconfiguration
3126 [ISAPNP] Exclude DMAs for the autoconfiguration
3128 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3129 Ranges are in pairs (I/O port base and size).
3132 [ISAPNP] Exclude memory regions for the
3134 Ranges are in pairs (memory base and size).
3136 ports= [IP_VS_FTP] IPVS ftp helper module
3138 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3140 Format: <port>,<port>....
3142 ppc_strict_facility_enable
3143 [PPC] This option catches any kernel floating point,
3144 Altivec, VSX and SPE outside of regions specifically
3145 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3146 There is some performance impact when enabling this.
3148 print-fatal-signals=
3149 [KNL] debug: print fatal signals
3151 If enabled, warn about various signal handling
3152 related application anomalies: too many signals,
3153 too many POSIX.1 timers, fatal signals causing a
3156 If you hit the warning due to signal overflow,
3157 you might want to try "ulimit -i unlimited".
3161 printk.always_kmsg_dump=
3162 Trigger kmsg_dump for cases other than kernel oops or
3164 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3167 printk.time= Show timing data prefixed to each printk message line
3168 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3170 processor.max_cstate= [HW,ACPI]
3171 Limit processor to maximum C-state
3172 max_cstate=9 overrides any DMI blacklist limit.
3174 processor.nocst [HW,ACPI]
3175 Ignore the _CST method to determine C-states,
3176 instead using the legacy FADT method
3178 profile= [KNL] Enable kernel profiling via /proc/profile
3179 Format: [schedule,]<number>
3180 Param: "schedule" - profile schedule points.
3181 Param: <number> - step/bucket size as a power of 2 for
3182 statistical time based profiling.
3183 Param: "sleep" - profile D-state sleeping (millisecs).
3184 Requires CONFIG_SCHEDSTATS
3185 Param: "kvm" - profile VM exits.
3187 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3189 See Documentation/blockdev/ramdisk.txt.
3191 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3192 probe for; one of (bare|imps|exps|lifebook|any).
3193 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3195 psmouse.resetafter= [HW,MOUSE]
3196 Try to reset the device after so many bad packets
3199 [HW,MOUSE] Set desired mouse resolution, in dpi.
3200 psmouse.smartscroll=
3201 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
3202 0 = disabled, 1 = enabled (default).
3204 pstore.backend= Specify the name of the pstore backend to use
3207 See Documentation/blockdev/paride.txt.
3210 [KNL] Number of legacy pty's. Overwrites compiled-in
3213 quiet [KNL] Disable most log messages
3218 See Documentation/md.txt.
3220 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
3221 See Documentation/blockdev/ramdisk.txt.
3224 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
3225 the specified list of CPUs to be no-callback CPUs.
3226 Invocation of these CPUs' RCU callbacks will
3227 be offloaded to "rcuox/N" kthreads created for
3228 that purpose, where "x" is "b" for RCU-bh, "p"
3229 for RCU-preempt, and "s" for RCU-sched, and "N"
3230 is the CPU number. This reduces OS jitter on the
3231 offloaded CPUs, which can be useful for HPC and
3232 real-time workloads. It can also improve energy
3233 efficiency for asymmetric multiprocessors.
3236 Rather than requiring that offloaded CPUs
3237 (specified by rcu_nocbs= above) explicitly
3238 awaken the corresponding "rcuoN" kthreads,
3239 make these kthreads poll for callbacks.
3240 This improves the real-time response for the
3241 offloaded CPUs by relieving them of the need to
3242 wake up the corresponding kthread, but degrades
3243 energy efficiency by requiring that the kthreads
3244 periodically wake up to do the polling.
3246 rcutree.blimit= [KNL]
3247 Set maximum number of finished RCU callbacks to
3248 process in one batch.
3250 rcutree.dump_tree= [KNL]
3251 Dump the structure of the rcu_node combining tree
3252 out at early boot. This is used for diagnostic
3253 purposes, to verify correct tree setup.
3255 rcutree.gp_cleanup_delay= [KNL]
3256 Set the number of jiffies to delay each step of
3257 RCU grace-period cleanup. This only has effect
3258 when CONFIG_RCU_TORTURE_TEST_SLOW_CLEANUP is set.
3260 rcutree.gp_init_delay= [KNL]
3261 Set the number of jiffies to delay each step of
3262 RCU grace-period initialization. This only has
3263 effect when CONFIG_RCU_TORTURE_TEST_SLOW_INIT
3266 rcutree.gp_preinit_delay= [KNL]
3267 Set the number of jiffies to delay each step of
3268 RCU grace-period pre-initialization, that is,
3269 the propagation of recent CPU-hotplug changes up
3270 the rcu_node combining tree. This only has effect
3271 when CONFIG_RCU_TORTURE_TEST_SLOW_PREINIT is set.
3273 rcutree.rcu_fanout_exact= [KNL]
3274 Disable autobalancing of the rcu_node combining
3275 tree. This is used by rcutorture, and might
3276 possibly be useful for architectures having high
3277 cache-to-cache transfer latencies.
3279 rcutree.rcu_fanout_leaf= [KNL]
3280 Change the number of CPUs assigned to each
3281 leaf rcu_node structure. Useful for very
3282 large systems, which will choose the value 64,
3283 and for NUMA systems with large remote-access
3284 latencies, which will choose a value aligned
3285 with the appropriate hardware boundaries.
3287 rcutree.jiffies_till_sched_qs= [KNL]
3288 Set required age in jiffies for a
3289 given grace period before RCU starts
3290 soliciting quiescent-state help from
3291 rcu_note_context_switch().
3293 rcutree.jiffies_till_first_fqs= [KNL]
3294 Set delay from grace-period initialization to
3295 first attempt to force quiescent states.
3296 Units are jiffies, minimum value is zero,
3297 and maximum value is HZ.
3299 rcutree.jiffies_till_next_fqs= [KNL]
3300 Set delay between subsequent attempts to force
3301 quiescent states. Units are jiffies, minimum
3302 value is one, and maximum value is HZ.
3304 rcutree.kthread_prio= [KNL,BOOT]
3305 Set the SCHED_FIFO priority of the RCU per-CPU
3306 kthreads (rcuc/N). This value is also used for
3307 the priority of the RCU boost threads (rcub/N)
3308 and for the RCU grace-period kthreads (rcu_bh,
3309 rcu_preempt, and rcu_sched). If RCU_BOOST is
3310 set, valid values are 1-99 and the default is 1
3311 (the least-favored priority). Otherwise, when
3312 RCU_BOOST is not set, valid values are 0-99 and
3313 the default is zero (non-realtime operation).
3315 rcutree.rcu_nocb_leader_stride= [KNL]
3316 Set the number of NOCB kthread groups, which
3317 defaults to the square root of the number of
3318 CPUs. Larger numbers reduces the wakeup overhead
3319 on the per-CPU grace-period kthreads, but increases
3320 that same overhead on each group's leader.
3322 rcutree.qhimark= [KNL]
3323 Set threshold of queued RCU callbacks beyond which
3324 batch limiting is disabled.
3326 rcutree.qlowmark= [KNL]
3327 Set threshold of queued RCU callbacks below which
3328 batch limiting is re-enabled.
3330 rcutree.rcu_idle_gp_delay= [KNL]
3331 Set wakeup interval for idle CPUs that have
3332 RCU callbacks (RCU_FAST_NO_HZ=y).
3334 rcutree.rcu_idle_lazy_gp_delay= [KNL]
3335 Set wakeup interval for idle CPUs that have
3336 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
3337 Lazy RCU callbacks are those which RCU can
3338 prove do nothing more than free memory.
3340 rcuperf.gp_exp= [KNL]
3341 Measure performance of expedited synchronous
3342 grace-period primitives.
3344 rcuperf.holdoff= [KNL]
3345 Set test-start holdoff period. The purpose of
3346 this parameter is to delay the start of the
3347 test until boot completes in order to avoid
3350 rcuperf.nreaders= [KNL]
3351 Set number of RCU readers. The value -1 selects
3352 N, where N is the number of CPUs. A value
3353 "n" less than -1 selects N-n+1, where N is again
3354 the number of CPUs. For example, -2 selects N
3355 (the number of CPUs), -3 selects N+1, and so on.
3356 A value of "n" less than or equal to -N selects
3359 rcuperf.nwriters= [KNL]
3360 Set number of RCU writers. The values operate
3361 the same as for rcuperf.nreaders.
3362 N, where N is the number of CPUs
3364 rcuperf.perf_runnable= [BOOT]
3365 Start rcuperf running at boot time.
3367 rcuperf.shutdown= [KNL]
3368 Shut the system down after performance tests
3369 complete. This is useful for hands-off automated
3372 rcuperf.perf_type= [KNL]
3373 Specify the RCU implementation to test.
3375 rcuperf.verbose= [KNL]
3376 Enable additional printk() statements.
3378 rcutorture.cbflood_inter_holdoff= [KNL]
3379 Set holdoff time (jiffies) between successive
3380 callback-flood tests.
3382 rcutorture.cbflood_intra_holdoff= [KNL]
3383 Set holdoff time (jiffies) between successive
3384 bursts of callbacks within a given callback-flood
3387 rcutorture.cbflood_n_burst= [KNL]
3388 Set the number of bursts making up a given
3389 callback-flood test. Set this to zero to
3390 disable callback-flood testing.
3392 rcutorture.cbflood_n_per_burst= [KNL]
3393 Set the number of callbacks to be registered
3394 in a given burst of a callback-flood test.
3396 rcutorture.fqs_duration= [KNL]
3397 Set duration of force_quiescent_state bursts
3400 rcutorture.fqs_holdoff= [KNL]
3401 Set holdoff time within force_quiescent_state bursts
3404 rcutorture.fqs_stutter= [KNL]
3405 Set wait time between force_quiescent_state bursts
3408 rcutorture.gp_cond= [KNL]
3409 Use conditional/asynchronous update-side
3410 primitives, if available.
3412 rcutorture.gp_exp= [KNL]
3413 Use expedited update-side primitives, if available.
3415 rcutorture.gp_normal= [KNL]
3416 Use normal (non-expedited) asynchronous
3417 update-side primitives, if available.
3419 rcutorture.gp_sync= [KNL]
3420 Use normal (non-expedited) synchronous
3421 update-side primitives, if available. If all
3422 of rcutorture.gp_cond=, rcutorture.gp_exp=,
3423 rcutorture.gp_normal=, and rcutorture.gp_sync=
3424 are zero, rcutorture acts as if is interpreted
3425 they are all non-zero.
3427 rcutorture.n_barrier_cbs= [KNL]
3428 Set callbacks/threads for rcu_barrier() testing.
3430 rcutorture.nfakewriters= [KNL]
3431 Set number of concurrent RCU writers. These just
3432 stress RCU, they don't participate in the actual
3433 test, hence the "fake".
3435 rcutorture.nreaders= [KNL]
3436 Set number of RCU readers. The value -1 selects
3437 N-1, where N is the number of CPUs. A value
3438 "n" less than -1 selects N-n-2, where N is again
3439 the number of CPUs. For example, -2 selects N
3440 (the number of CPUs), -3 selects N+1, and so on.
3442 rcutorture.object_debug= [KNL]
3443 Enable debug-object double-call_rcu() testing.
3445 rcutorture.onoff_holdoff= [KNL]
3446 Set time (s) after boot for CPU-hotplug testing.
3448 rcutorture.onoff_interval= [KNL]
3449 Set time (s) between CPU-hotplug operations, or
3450 zero to disable CPU-hotplug testing.
3452 rcutorture.shuffle_interval= [KNL]
3453 Set task-shuffle interval (s). Shuffling tasks
3454 allows some CPUs to go into dyntick-idle mode
3455 during the rcutorture test.
3457 rcutorture.shutdown_secs= [KNL]
3458 Set time (s) after boot system shutdown. This
3459 is useful for hands-off automated testing.
3461 rcutorture.stall_cpu= [KNL]
3462 Duration of CPU stall (s) to test RCU CPU stall
3463 warnings, zero to disable.
3465 rcutorture.stall_cpu_holdoff= [KNL]
3466 Time to wait (s) after boot before inducing stall.
3468 rcutorture.stat_interval= [KNL]
3469 Time (s) between statistics printk()s.
3471 rcutorture.stutter= [KNL]
3472 Time (s) to stutter testing, for example, specifying
3473 five seconds causes the test to run for five seconds,
3474 wait for five seconds, and so on. This tests RCU's
3475 ability to transition abruptly to and from idle.
3477 rcutorture.test_boost= [KNL]
3478 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
3479 "Maybe" means test if the RCU implementation
3480 under test support RCU priority boosting.
3482 rcutorture.test_boost_duration= [KNL]
3483 Duration (s) of each individual boost test.
3485 rcutorture.test_boost_interval= [KNL]
3486 Interval (s) between each boost test.
3488 rcutorture.test_no_idle_hz= [KNL]
3489 Test RCU's dyntick-idle handling. See also the
3490 rcutorture.shuffle_interval parameter.
3492 rcutorture.torture_runnable= [BOOT]
3493 Start rcutorture running at boot time.
3495 rcutorture.torture_type= [KNL]
3496 Specify the RCU implementation to test.
3498 rcutorture.verbose= [KNL]
3499 Enable additional printk() statements.
3501 rcupdate.rcu_cpu_stall_suppress= [KNL]
3502 Suppress RCU CPU stall warning messages.
3504 rcupdate.rcu_cpu_stall_timeout= [KNL]
3505 Set timeout for RCU CPU stall warning messages.
3507 rcupdate.rcu_expedited= [KNL]
3508 Use expedited grace-period primitives, for
3509 example, synchronize_rcu_expedited() instead
3510 of synchronize_rcu(). This reduces latency,
3511 but can increase CPU utilization, degrade
3512 real-time latency, and degrade energy efficiency.
3513 No effect on CONFIG_TINY_RCU kernels.
3515 rcupdate.rcu_normal= [KNL]
3516 Use only normal grace-period primitives,
3517 for example, synchronize_rcu() instead of
3518 synchronize_rcu_expedited(). This improves
3519 real-time latency, CPU utilization, and
3520 energy efficiency, but can expose users to
3521 increased grace-period latency. This parameter
3522 overrides rcupdate.rcu_expedited. No effect on
3523 CONFIG_TINY_RCU kernels.
3525 rcupdate.rcu_normal_after_boot= [KNL]
3526 Once boot has completed (that is, after
3527 rcu_end_inkernel_boot() has been invoked), use
3528 only normal grace-period primitives. No effect
3529 on CONFIG_TINY_RCU kernels.
3531 rcupdate.rcu_task_stall_timeout= [KNL]
3532 Set timeout in jiffies for RCU task stall warning
3533 messages. Disable with a value less than or equal
3536 rcupdate.rcu_self_test= [KNL]
3537 Run the RCU early boot self tests
3539 rcupdate.rcu_self_test_bh= [KNL]
3540 Run the RCU bh early boot self tests
3542 rcupdate.rcu_self_test_sched= [KNL]
3543 Run the RCU sched early boot self tests
3547 Run specified binary instead of /init from the ramdisk,
3548 used for early userspace startup. See initrd.
3551 Format (x86 or x86_64):
3552 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
3554 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
3556 Where reboot_mode is one of warm (soft) or cold (hard) or gpio,
3557 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
3558 reboot_force is either force or not specified,
3559 reboot_cpu is s[mp]#### with #### being the processor
3560 to be used for rebooting.
3563 [KNL, SMP] Set scheduler's default relax_domain_level.
3564 See Documentation/cgroups/cpusets.txt.
3566 relative_sleep_states=
3567 [SUSPEND] Use sleep state labeling where the deepest
3568 state available other than hibernation is always "mem".
3569 Format: { "0" | "1" }
3570 0 -- Traditional sleep state labels.
3571 1 -- Relative sleep state labels.
3573 reserve= [KNL,BUGS] Force the kernel to ignore some iomem area
3575 reservetop= [X86-32]
3577 Reserves a hole at the top of the kernel virtual
3582 Set the amount of memory to reserve for BIOS at
3583 the bottom of the address space.
3585 reset_devices [KNL] Force drivers to reset the underlying device
3586 during initialization.
3589 Specify the partition device for software suspend
3591 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
3593 resume_offset= [SWSUSP]
3594 Specify the offset from the beginning of the partition
3595 given by "resume=" at which the swap header is located,
3596 in <PAGE_SIZE> units (needed only for swap files).
3597 See Documentation/power/swsusp-and-swap-files.txt
3599 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
3600 read the resume files
3602 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
3603 Useful for devices that are detected asynchronously
3604 (e.g. USB and MMC devices).
3606 hibernate= [HIBERNATION]
3607 noresume Don't check if there's a hibernation image
3608 present during boot.
3609 nocompress Don't compress/decompress hibernation images.
3610 no Disable hibernation and resume.
3611 protect_image Turn on image protection during restoration
3612 (that will set all pages holding image data
3613 during restoration read-only).
3615 retain_initrd [RAM] Keep initrd memory after extraction
3617 rfkill.default_state=
3618 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
3619 etc. communication is blocked by default.
3622 rfkill.master_switch_mode=
3623 0 The "airplane mode" button does nothing.
3624 1 The "airplane mode" button toggles between everything
3625 blocked and the previous configuration.
3626 2 The "airplane mode" button toggles between everything
3627 blocked and everything unblocked.
3629 rhash_entries= [KNL,NET]
3630 Set number of hash buckets for route cache
3632 ro [KNL] Mount root device read-only on boot
3635 on Mark read-only kernel memory as read-only (default).
3636 off Leave read-only kernel memory writable for debugging.
3639 Enable the uart passthrough on the designated usb port
3640 on Rockchip SoCs. When active, the signals of the
3641 debug-uart get routed to the D+ and D- pins of the usb
3642 port and the regular usb controller gets disabled.
3644 root= [KNL] Root filesystem
3645 See name_to_dev_t comment in init/do_mounts.c.
3647 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
3648 mount the root filesystem
3650 rootflags= [KNL] Set root filesystem mount option string
3652 rootfstype= [KNL] Set root filesystem type
3654 rootwait [KNL] Wait (indefinitely) for root device to show up.
3655 Useful for devices that are detected asynchronously
3656 (e.g. USB and MMC devices).
3658 rproc_mem=nn[KMG][@address]
3659 [KNL,ARM,CMA] Remoteproc physical memory block.
3660 Memory area to be used by remote processor image,
3663 rw [KNL] Mount root device read-write on boot
3665 S [KNL] Run init in single mode
3667 s390_iommu= [HW,S390]
3668 Set s390 IOTLB flushing mode
3670 With strict flushing every unmap operation will result in
3671 an IOTLB flush. Default is lazy flushing before reuse,
3675 See drivers/net/irda/sa1100_ir.c.
3677 sbni= [NET] Granch SBNI12 leased line adapter
3679 sched_debug [KNL] Enables verbose scheduler debug messages.
3681 schedstats= [KNL,X86] Enable or disable scheduled statistics.
3682 Allowed values are enable and disable. This feature
3683 incurs a small amount of overhead in the scheduler
3684 but is useful for debugging and performance tuning.
3686 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
3687 xtime_lock contention on larger systems, and/or RCU lock
3688 contention on all systems with CONFIG_MAXSMP set.
3689 Format: { "0" | "1" }
3690 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
3692 Note: increases power consumption, thus should only be
3693 enabled if running jitter sensitive (HPC/RT) workloads.
3695 security= [SECURITY] Choose a security module to enable at boot.
3696 If this boot parameter is not specified, only the first
3697 security module asking for security registration will be
3698 loaded. An invalid security module name will be treated
3699 as if no module has been chosen.
3701 selinux= [SELINUX] Disable or enable SELinux at boot time.
3702 Format: { "0" | "1" }
3703 See security/selinux/Kconfig help text.
3706 Default value is set via kernel config option.
3707 If enabled at boot time, /selinux/disable can be used
3708 later to disable prior to initial policy load.
3710 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
3711 Format: { "0" | "1" }
3712 See security/apparmor/Kconfig help text
3715 Default value is set via kernel config option.
3717 serialnumber [BUGS=X86-32]
3720 Maximal number of shapers.
3722 show_msr= [x86] show boot-time MSR settings
3723 Format: { <integer> }
3724 Show boot-time (BIOS-initialized) MSR settings.
3725 The parameter means the number of CPUs to show,
3726 for example 1 means boot CPU only.
3734 Disable merging of slabs with similar size. May be
3735 necessary if there is some reason to distinguish
3736 allocs to different slabs. Debug options disable
3737 merging on their own.
3738 For more information see Documentation/vm/slub.txt.
3740 slab_max_order= [MM, SLAB]
3741 Determines the maximum allowed order for slabs.
3742 A high setting may cause OOMs due to memory
3743 fragmentation. Defaults to 1 for systems with
3744 more than 32MB of RAM, 0 otherwise.
3746 slub_debug[=options[,slabs]] [MM, SLUB]
3747 Enabling slub_debug allows one to determine the
3748 culprit if slab objects become corrupted. Enabling
3749 slub_debug can create guard zones around objects and
3750 may poison objects when not in use. Also tracks the
3751 last alloc / free. For more information see
3752 Documentation/vm/slub.txt.
3754 slub_max_order= [MM, SLUB]
3755 Determines the maximum allowed order for slabs.
3756 A high setting may cause OOMs due to memory
3757 fragmentation. For more information see
3758 Documentation/vm/slub.txt.
3760 slub_min_objects= [MM, SLUB]
3761 The minimum number of objects per slab. SLUB will
3762 increase the slab order up to slub_max_order to
3763 generate a sufficiently large slab able to contain
3764 the number of objects indicated. The higher the number
3765 of objects the smaller the overhead of tracking slabs
3766 and the less frequently locks need to be acquired.
3767 For more information see Documentation/vm/slub.txt.
3769 slub_min_order= [MM, SLUB]
3770 Determines the minimum page order for slabs. Must be
3771 lower than slub_max_order.
3772 For more information see Documentation/vm/slub.txt.
3774 slub_nomerge [MM, SLUB]
3775 Same with slab_nomerge. This is supported for legacy.
3776 See slab_nomerge for more information.
3779 Format: <io1>[,<io2>[,...,<io8>]]
3781 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
3782 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
3783 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
3784 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
3785 smsc-ircc2.ircc_irq= [HW] IRQ line
3786 smsc-ircc2.ircc_dma= [HW] DMA channel
3787 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
3788 0: Toshiba Satellite 1800 (GP data pin select)
3789 1: Fast pin select (default)
3792 smt [KNL,S390] Set the maximum number of threads (logical
3793 CPUs) to use per physical CPU on systems capable of
3794 symmetric multithreading (SMT). Will be capped to the
3795 actual hardware limit.
3797 Default: -1 (no limit)
3800 [KNL] Should the soft-lockup detector generate panics.
3803 softlockup_all_cpu_backtrace=
3804 [KNL] Should the soft-lockup detector generate
3805 backtraces on all cpus.
3808 sonypi.*= [HW] Sony Programmable I/O Control Device driver
3809 See Documentation/laptops/sonypi.txt
3811 spia_io_base= [HW,MTD]
3817 Enabled the stack tracer on boot up.
3819 stacktrace_filter=[function-list]
3820 [FTRACE] Limit the functions that the stack tracer
3821 will trace at boot up. function-list is a comma separated
3822 list of functions. This list can be changed at run
3823 time by the stack_trace_filter file in the debugfs
3824 tracing directory. Note, this enables stack tracing
3825 and the stacktrace above is not needed.
3829 Set the STI (builtin display/keyboard on the HP-PARISC
3830 machines) console (graphic card) which should be used
3831 as the initial boot-console.
3832 See also comment in drivers/video/console/sticore.c.
3835 See comment in drivers/video/console/sticore.c.
3838 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
3840 sunrpc.min_resvport=
3841 sunrpc.max_resvport=
3843 SunRPC servers often require that client requests
3844 originate from a privileged port (i.e. a port in the
3845 range 0 < portnr < 1024).
3846 An administrator who wishes to reserve some of these
3847 ports for other uses may adjust the range that the
3848 kernel's sunrpc client considers to be privileged
3849 using these two parameters to set the minimum and
3850 maximum port values.
3854 Control how the NFS server code allocates CPUs to
3855 service thread pools. Depending on how many NICs
3856 you have and where their interrupts are bound, this
3857 option will affect which CPUs will do NFS serving.
3858 Note: this parameter cannot be changed while the
3859 NFS server is running.
3861 auto the server chooses an appropriate mode
3862 automatically using heuristics
3863 global a single global pool contains all CPUs
3864 percpu one pool for each CPU
3865 pernode one pool for each NUMA node (equivalent
3866 to global on non-NUMA machines)
3868 sunrpc.tcp_slot_table_entries=
3869 sunrpc.udp_slot_table_entries=
3871 Sets the upper limit on the number of simultaneous
3872 RPC calls that can be sent from the client to a
3873 server. Increasing these values may allow you to
3874 improve throughput, but will also increase the
3875 amount of memory reserved for use by the client.
3877 suspend.pm_test_delay=
3879 Sets the number of seconds to remain in a suspend test
3880 mode before resuming the system (see
3881 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
3882 is set. Default value is 5.
3885 [KNL] Enable accounting of swap in memory resource
3886 controller if no parameter or 1 is given or disable
3887 it if 0 is given (See Documentation/cgroups/memory.txt)
3889 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
3890 Format: { <int> | force }
3891 <int> -- Number of I/O TLB slabs
3892 force -- force using of bounce buffers even if they
3893 wouldn't be automatically used by the kernel
3897 sysfs.deprecated=0|1 [KNL]
3898 Enable/disable old style sysfs layout for old udev
3899 on older distributions. When this option is enabled
3900 very new udev will not work anymore. When this option
3901 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
3902 in older udev will not work anymore.
3903 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
3904 the kernel configuration.
3906 sysrq_always_enabled
3908 Ignore sysrq setting - this boot parameter will
3909 neutralize any effect of /proc/sys/kernel/sysrq.
3910 Useful for debugging.
3912 tcpmhash_entries= [KNL,NET]
3913 Set the number of tcp_metrics_hash slots.
3914 Default value is 8192 or 16384 depending on total
3915 ram pages. This is used to specify the TCP metrics
3916 cache size. See Documentation/networking/ip-sysctl.txt
3917 "tcp_no_metrics_save" section for more details.
3921 test_suspend= [SUSPEND][,N]
3922 Specify "mem" (for Suspend-to-RAM) or "standby" (for
3923 standby suspend) or "freeze" (for suspend type freeze)
3924 as the system sleep state during system startup with
3925 the optional capability to repeat N number of times.
3926 The system is woken from this state using a
3927 wakeup-capable RTC alarm.
3929 thash_entries= [KNL,NET]
3930 Set number of hash buckets for TCP connection
3932 thermal.act= [HW,ACPI]
3933 -1: disable all active trip points in all thermal zones
3934 <degrees C>: override all lowest active trip points
3936 thermal.crt= [HW,ACPI]
3937 -1: disable all critical trip points in all thermal zones
3938 <degrees C>: override all critical trip points
3940 thermal.nocrt= [HW,ACPI]
3941 Set to disable actions on ACPI thermal zone
3942 critical and hot trip points.
3944 thermal.off= [HW,ACPI]
3945 1: disable ACPI thermal control
3947 thermal.psv= [HW,ACPI]
3948 -1: disable all passive trip points
3949 <degrees C>: override all passive trip points to this
3952 thermal.tzp= [HW,ACPI]
3953 Specify global default ACPI thermal zone polling rate
3954 <deci-seconds>: poll all this frequency
3955 0: no polling (default)
3958 Force threading of all interrupt handlers except those
3959 marked explicitly IRQF_NO_THREAD.
3962 Enable the Transcendent memory driver if built-in.
3964 tmem.cleancache=0|1 [KNL, XEN]
3965 Default is on (1). Disable the usage of the cleancache
3966 API to send anonymous pages to the hypervisor.
3968 tmem.frontswap=0|1 [KNL, XEN]
3969 Default is on (1). Disable the usage of the frontswap
3970 API to send swap pages to the hypervisor. If disabled
3971 the selfballooning and selfshrinking are force disabled.
3973 tmem.selfballooning=0|1 [KNL, XEN]
3974 Default is on (1). Disable the driving of swap pages
3977 tmem.selfshrinking=0|1 [KNL, XEN]
3978 Default is on (1). Partial swapoff that immediately
3979 transfers pages from Xen hypervisor back to the
3980 kernel based on different criteria.
3984 Specify if the kernel should make use of the cpu
3985 topology information if the hardware supports this.
3986 The scheduler will make use of this information and
3987 e.g. base its process migration decisions on it.
3990 topology_updates= [KNL, PPC, NUMA]
3992 Specify if the kernel should ignore (off)
3993 topology updates sent by the hypervisor to this
3998 tpm_suspend_pcr=[HW,TPM]
3999 Format: integer pcr id
4000 Specify that at suspend time, the tpm driver
4001 should extend the specified pcr with zeros,
4002 as a workaround for some chips which fail to
4003 flush the last written pcr on TPM_SaveState.
4004 This will guarantee that all the other pcrs
4007 trace_buf_size=nn[KMG]
4008 [FTRACE] will set tracing buffer size on each cpu.
4010 trace_event=[event-list]
4011 [FTRACE] Set and start specified trace events in order
4012 to facilitate early boot debugging. The event-list is a
4013 comma separated list of trace events to enable. See
4014 also Documentation/trace/events.txt
4016 trace_options=[option-list]
4017 [FTRACE] Enable or disable tracer options at boot.
4018 The option-list is a comma delimited list of options
4019 that can be enabled or disabled just as if you were
4020 to echo the option name into
4022 /sys/kernel/debug/tracing/trace_options
4024 For example, to enable stacktrace option (to dump the
4025 stack trace of each event), add to the command line:
4027 trace_options=stacktrace
4029 See also Documentation/trace/ftrace.txt "trace options"
4033 Have the tracepoints sent to printk as well as the
4034 tracing ring buffer. This is useful for early boot up
4035 where the system hangs or reboots and does not give the
4036 option for reading the tracing buffer or performing a
4037 ftrace_dump_on_oops.
4039 To turn off having tracepoints sent to printk,
4040 echo 0 > /proc/sys/kernel/tracepoint_printk
4041 Note, echoing 1 into this file without the
4042 tracepoint_printk kernel cmdline option has no effect.
4046 Having tracepoints sent to printk() and activating high
4047 frequency tracepoints such as irq or sched, can cause
4048 the system to live lock.
4051 [FTRACE] enable this option to disable tracing when a
4052 warning is hit. This turns off "tracing_on". Tracing can
4053 be enabled again by echoing '1' into the "tracing_on"
4054 file located in /sys/kernel/debug/tracing/
4056 This option is useful, as it disables the trace before
4057 the WARNING dump is called, which prevents the trace to
4058 be filled with content caused by the warning output.
4060 This option can also be set at run time via the sysctl
4061 option: kernel/traceoff_on_warning
4063 transparent_hugepage=
4065 Format: [always|madvise|never]
4066 Can be used to control the default behavior of the system
4067 with respect to transparent hugepages.
4068 See Documentation/vm/transhuge.txt for more details.
4070 tsc= Disable clocksource stability checks for TSC.
4072 [x86] reliable: mark tsc clocksource as reliable, this
4073 disables clocksource verification at runtime, as well
4074 as the stability checks done at bootup. Used to enable
4075 high-resolution timer mode on older hardware, and in
4076 virtualized environment.
4077 [x86] noirqtime: Do not use TSC to do irq accounting.
4078 Used to run time disable IRQ_TIME_ACCOUNTING on any
4079 platforms where RDTSC is slow and this accounting
4082 turbografx.map[2|3]= [HW,JOY]
4083 TurboGraFX parallel port interface
4085 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
4086 See also Documentation/input/joystick-parport.txt
4088 udbg-immortal [PPC] When debugging early kernel crashes that
4089 happen after console_init() and before a proper
4090 console driver takes over, this boot options might
4091 help "seeing" what's going on.
4093 uhash_entries= [KNL,NET]
4094 Set number of hash buckets for UDP/UDP-Lite connections
4097 [USB] Ignore overcurrent events (default N).
4098 Some badly-designed motherboards generate lots of
4099 bogus events, for ports that aren't wired to
4100 anything. Set this parameter to avoid log spamming.
4101 Note that genuine overcurrent events won't be
4105 [X86] Cause panic on unknown NMI.
4107 usbcore.authorized_default=
4108 [USB] Default USB device authorization:
4109 (default -1 = authorized except for wireless USB,
4110 0 = not authorized, 1 = authorized)
4112 usbcore.autosuspend=
4113 [USB] The autosuspend time delay (in seconds) used
4114 for newly-detected USB devices (default 2). This
4115 is the time required before an idle device will be
4116 autosuspended. Devices for which the delay is set
4117 to a negative value won't be autosuspended at all.
4119 usbcore.usbfs_snoop=
4120 [USB] Set to log all usbfs traffic (default 0 = off).
4122 usbcore.usbfs_snoop_max=
4123 [USB] Maximum number of bytes to snoop in each URB
4126 usbcore.blinkenlights=
4127 [USB] Set to cycle leds on hubs (default 0 = off).
4129 usbcore.old_scheme_first=
4130 [USB] Start with the old device initialization
4131 scheme (default 0 = off).
4133 usbcore.usbfs_memory_mb=
4134 [USB] Memory limit (in MB) for buffers allocated by
4135 usbfs (default = 16, 0 = max = 2047).
4137 usbcore.use_both_schemes=
4138 [USB] Try the other device initialization scheme
4139 if the first one fails (default 1 = enabled).
4141 usbcore.initial_descriptor_timeout=
4142 [USB] Specifies timeout for the initial 64-byte
4143 USB_REQ_GET_DESCRIPTOR request in milliseconds
4144 (default 5000 = 5.0 seconds).
4146 usbcore.nousb [USB] Disable the USB subsystem
4149 [USBHID] The interval which mice are to be polled at.
4151 usb-storage.delay_use=
4152 [UMS] The delay in seconds before a new device is
4153 scanned for Logical Units (default 1).
4156 [UMS] A list of quirks entries to supplement or
4157 override the built-in unusual_devs list. List
4158 entries are separated by commas. Each entry has
4159 the form VID:PID:Flags where VID and PID are Vendor
4160 and Product ID values (4-digit hex numbers) and
4161 Flags is a set of characters, each corresponding
4162 to a common usb-storage quirk flag as follows:
4163 a = SANE_SENSE (collect more than 18 bytes
4165 b = BAD_SENSE (don't collect more than 18
4166 bytes of sense data);
4167 c = FIX_CAPACITY (decrease the reported
4168 device capacity by one sector);
4169 d = NO_READ_DISC_INFO (don't use
4170 READ_DISC_INFO command);
4171 e = NO_READ_CAPACITY_16 (don't use
4172 READ_CAPACITY_16 command);
4173 f = NO_REPORT_OPCODES (don't use report opcodes
4175 g = MAX_SECTORS_240 (don't transfer more than
4176 240 sectors at a time, uas only);
4177 h = CAPACITY_HEURISTICS (decrease the
4178 reported device capacity by one
4179 sector if the number is odd);
4180 i = IGNORE_DEVICE (don't bind to this
4182 j = NO_REPORT_LUNS (don't use report luns
4184 l = NOT_LOCKABLE (don't try to lock and
4185 unlock ejectable media);
4186 m = MAX_SECTORS_64 (don't transfer more
4187 than 64 sectors = 32 KB at a time);
4188 n = INITIAL_READ10 (force a retry of the
4189 initial READ(10) command);
4190 o = CAPACITY_OK (accept the capacity
4191 reported by the device);
4192 p = WRITE_CACHE (the device cache is ON
4194 r = IGNORE_RESIDUE (the device reports
4195 bogus residue values);
4196 s = SINGLE_LUN (the device has only one
4198 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
4199 commands, uas only);
4200 u = IGNORE_UAS (don't bind to the uas driver);
4201 w = NO_WP_DETECT (don't test whether the
4202 medium is write-protected).
4203 Example: quirks=0419:aaf5:rl,0421:0433:rc
4205 user_debug= [KNL,ARM]
4207 See arch/arm/Kconfig.debug help text.
4208 1 - undefined instruction events
4210 4 - invalid data aborts
4213 Example: user_debug=31
4216 [X86] Flags controlling user PTE allocations.
4218 nohigh = do not allocate PTE pages in
4219 HIGHMEM regardless of setting
4223 On X86_32, this is an alias for vdso32=. Otherwise:
4225 vdso=1: enable VDSO (the default)
4226 vdso=0: disable VDSO mapping
4228 vdso32= [X86] Control the 32-bit vDSO
4229 vdso32=1: enable 32-bit VDSO
4230 vdso32=0 or vdso32=2: disable 32-bit VDSO
4232 See the help text for CONFIG_COMPAT_VDSO for more
4233 details. If CONFIG_COMPAT_VDSO is set, the default is
4234 vdso32=0; otherwise, the default is vdso32=1.
4236 For compatibility with older kernels, vdso32=2 is an
4239 Try vdso32=0 if you encounter an error that says:
4240 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
4243 vector=percpu: enable percpu vector domain
4245 video= [FB] Frame buffer configuration
4246 See Documentation/fb/modedb.txt.
4248 video.brightness_switch_enabled= [0,1]
4249 If set to 1, on receiving an ACPI notify event
4250 generated by hotkey, video driver will adjust brightness
4251 level and then send out the event to user space through
4252 the allocated input device; If set to 0, video driver
4253 will only send out the event without touching backlight
4258 [VMMIO] Memory mapped virtio (platform) device.
4260 <size>@<baseaddr>:<irq>[:<id>]
4262 <size> := size (can use standard suffixes
4264 <baseaddr> := physical base address
4265 <irq> := interrupt number (as passed to
4267 <id> := (optional) platform device id
4269 virtio_mmio.device=1K@0x100b0000:48:7
4271 Can be used multiple times for multiple devices.
4273 vga= [BOOT,X86-32] Select a particular video mode
4274 See Documentation/x86/boot.txt and
4275 Documentation/svga.txt.
4276 Use vga=ask for menu.
4277 This is actually a boot loader parameter; the value is
4278 passed to the kernel using a special protocol.
4280 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
4281 size of <nn>. This can be used to increase the
4282 minimum size (128MB on x86). It can also be used to
4283 decrease the size and leave more room for directly
4286 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
4289 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
4292 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
4296 Controls the behavior of vsyscalls (i.e. calls to
4297 fixed addresses of 0xffffffffff600x00 from legacy
4298 code). Most statically-linked binaries and older
4299 versions of glibc use these calls. Because these
4300 functions are at fixed addresses, they make nice
4301 targets for exploits that can control RIP.
4303 emulate [default] Vsyscalls turn into traps and are
4304 emulated reasonably safely.
4306 native Vsyscalls are native syscall instructions.
4307 This is a little bit faster than trapping
4308 and makes a few dynamic recompilers work
4309 better than they would in emulation mode.
4310 It also makes exploits much easier to write.
4312 none Vsyscalls don't work at all. This makes
4313 them quite hard to use for exploits but
4314 might break your system.
4316 vt.color= [VT] Default text color.
4317 Format: 0xYX, X = foreground, Y = background.
4318 Default: 0x07 = light gray on black.
4320 vt.cur_default= [VT] Default cursor shape.
4321 Format: 0xCCBBAA, where AA, BB, and CC are the same as
4322 the parameters of the <Esc>[?A;B;Cc escape sequence;
4323 see VGA-softcursor.txt. Default: 2 = underline.
4325 vt.default_blu= [VT]
4326 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
4327 Change the default blue palette of the console.
4328 This is a 16-member array composed of values
4331 vt.default_grn= [VT]
4332 Format: <green0>,<green1>,<green2>,...,<green15>
4333 Change the default green palette of the console.
4334 This is a 16-member array composed of values
4337 vt.default_red= [VT]
4338 Format: <red0>,<red1>,<red2>,...,<red15>
4339 Change the default red palette of the console.
4340 This is a 16-member array composed of values
4346 Set system-wide default UTF-8 mode for all tty's.
4347 Default is 1, i.e. UTF-8 mode is enabled for all
4348 newly opened terminals.
4350 vt.global_cursor_default=
4353 Set system-wide default for whether a cursor
4354 is shown on new VTs. Default is -1,
4355 i.e. cursors will be created by default unless
4356 overridden by individual drivers. 0 will hide
4357 cursors, 1 will display them.
4359 vt.italic= [VT] Default color for italic text; 0-15.
4362 vt.underline= [VT] Default color for underlined text; 0-15.
4365 watchdog timers [HW,WDT] For information on watchdog timers,
4366 see Documentation/watchdog/watchdog-parameters.txt
4367 or other driver-specific files in the
4368 Documentation/watchdog/ directory.
4370 workqueue.watchdog_thresh=
4371 If CONFIG_WQ_WATCHDOG is configured, workqueue can
4372 warn stall conditions and dump internal state to
4373 help debugging. 0 disables workqueue stall
4374 detection; otherwise, it's the stall threshold
4375 duration in seconds. The default value is 30 and
4376 it can be updated at runtime by writing to the
4377 corresponding sysfs file.
4379 workqueue.disable_numa
4380 By default, all work items queued to unbound
4381 workqueues are affine to the NUMA nodes they're
4382 issued on, which results in better behavior in
4383 general. If NUMA affinity needs to be disabled for
4384 whatever reason, this option can be used. Note
4385 that this also can be controlled per-workqueue for
4386 workqueues visible under /sys/bus/workqueue/.
4388 workqueue.power_efficient
4389 Per-cpu workqueues are generally preferred because
4390 they show better performance thanks to cache
4391 locality; unfortunately, per-cpu workqueues tend to
4392 be more power hungry than unbound workqueues.
4394 Enabling this makes the per-cpu workqueues which
4395 were observed to contribute significantly to power
4396 consumption unbound, leading to measurably lower
4397 power usage at the cost of small performance
4400 The default value of this parameter is determined by
4401 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
4403 workqueue.debug_force_rr_cpu
4404 Workqueue used to implicitly guarantee that work
4405 items queued without explicit CPU specified are put
4406 on the local CPU. This guarantee is no longer true
4407 and while local CPU is still preferred work items
4408 may be put on foreign CPUs. This debug option
4409 forces round-robin CPU selection to flush out
4410 usages which depend on the now broken guarantee.
4411 When enabled, memory and cache locality will be
4414 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
4415 default x2apic cluster mode on platforms
4418 x86_intel_mid_timer= [X86-32,APBT]
4419 Choose timer option for x86 Intel MID platform.
4420 Two valid options are apbt timer only and lapic timer
4421 plus one apbt timer for broadcast timer.
4422 x86_intel_mid_timer=apbt_only | lapic_and_apbt
4424 xen_512gb_limit [KNL,X86-64,XEN]
4425 Restricts the kernel running paravirtualized under Xen
4426 to use only up to 512 GB of RAM. The reason to do so is
4427 crash analysis tools and Xen tools for doing domain
4428 save/restore/migration must be enabled to handle larger
4431 xen_emul_unplug= [HW,X86,XEN]
4432 Unplug Xen emulated devices
4433 Format: [unplug0,][unplug1]
4434 ide-disks -- unplug primary master IDE devices
4435 aux-ide-disks -- unplug non-primary-master IDE devices
4436 nics -- unplug network devices
4437 all -- unplug all emulated devices (NICs and IDE disks)
4438 unnecessary -- unplugging emulated devices is
4439 unnecessary even if the host did not respond to
4441 never -- do not unplug even if version check succeeds
4443 xen_nopvspin [X86,XEN]
4444 Disables the ticketlock slowpath using Xen PV
4448 Disables the PV optimizations forcing the HVM guest to
4449 run as generic HVM guest with no PV drivers.
4451 xirc2ps_cs= [NET,PCMCIA]
4453 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
4455 ______________________________________________________________________
4459 Add more DRM drivers.