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.
585 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
586 bttv.radio= Most important insmod options are available as
588 bttv.pll= See Documentation/video4linux/bttv/Insmod-options
591 bulk_remove=off [PPC] This parameter disables the use of the pSeries
592 firmware feature for flushing multiple hpte entries
595 c101= [NET] Moxa C101 synchronous serial card
597 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
598 Sometimes CPU hardware bugs make them report the cache
599 size incorrectly. The kernel will attempt work arounds
600 to fix known problems, but for some CPUs it is not
601 possible to determine what the correct size should be.
602 This option provides an override for these situations.
604 ca_keys= [KEYS] This parameter identifies a specific key(s) on
605 the system trusted keyring to be used for certificate
607 format: { id:<keyid> | builtin }
609 cca= [MIPS] Override the kernel pages' cache coherency
610 algorithm. Accepted values range from 0 to 7
611 inclusive. See arch/mips/include/asm/pgtable-bits.h
612 for platform specific values (SB1, Loongson3 and
615 ccw_timeout_log [S390]
616 See Documentation/s390/CommonIO for details.
618 cgroup_disable= [KNL] Disable a particular controller
619 Format: {name of the controller(s) to disable}
620 The effects of cgroup_disable=foo are:
621 - foo isn't auto-mounted if you mount all cgroups in
623 - foo isn't visible as an individually mountable
625 {Currently only "memory" controller deal with this and
626 cut the overhead, others just disable the usage. So
627 only cgroup_disable=memory is actually worthy}
629 cgroup_no_v1= [KNL] Disable one, multiple, all cgroup controllers in v1
630 Format: { controller[,controller...] | "all" }
631 Like cgroup_disable, but only applies to cgroup v1;
632 the blacklisted controllers remain available in cgroup2.
634 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
636 nosocket -- Disable socket memory accounting.
637 nokmem -- Disable kernel memory accounting.
639 checkreqprot [SELINUX] Set initial checkreqprot flag value.
640 Format: { "0" | "1" }
641 See security/selinux/Kconfig help text.
642 0 -- check protection applied by kernel (includes
643 any implied execute protection).
644 1 -- check protection requested by application.
645 Default value is set via a kernel config option.
646 Value can be changed at runtime via
647 /selinux/checkreqprot.
650 See Documentation/s390/CommonIO for details.
653 Prevents the clock framework from automatically gating
654 clocks that have not been explicitly enabled by a Linux
655 device driver but are enabled in hardware at reset or
656 by the bootloader/firmware. Note that this does not
657 force such clocks to be always-on nor does it reserve
658 those clocks in any way. This parameter is useful for
659 debug and development, but should not be needed on a
660 platform with proper driver support. For more
661 information, see Documentation/clk.txt.
663 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
665 Forces specified clocksource (if available) to be used
666 when calculating gettimeofday(). If specified
667 clocksource is not available, it defaults to PIT.
668 Format: { pit | tsc | cyclone | pmtmr }
670 clocksource= Override the default clocksource
672 Override the default clocksource and use the clocksource
673 with the name specified.
674 Some clocksource names to choose from, depending on
676 [all] jiffies (this is the base, fallback clocksource)
678 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
679 pxa_timer,timer3,32k_counter,timer0_1
681 [X86-32] pit,hpet,tsc;
682 scx200_hrt on Geode; cyclone on IBM x440
690 clocksource.arm_arch_timer.evtstrm=
693 Enable/disable the eventstream feature of the ARM
694 architected timer so that code using WFE-based polling
695 loops can be debugged more effectively on production
698 clearcpuid=BITNUM [X86]
699 Disable CPUID feature X for the kernel. See
700 arch/x86/include/asm/cpufeatures.h for the valid bit
701 numbers. Note the Linux specific bits are not necessarily
702 stable over kernel options, but the vendor specific
704 Also note that user programs calling CPUID directly
705 or using the feature without checking anything
706 will still see it. This just prevents it from
707 being used by the kernel or shown in /proc/cpuinfo.
708 Also note the kernel might malfunction if you disable
711 cma=nn[MG]@[start[MG][-end[MG]]]
713 Sets the size of kernel global memory area for
714 contiguous memory allocations and optionally the
715 placement constraint by the physical address range of
716 memory allocations. A value of 0 disables CMA
717 altogether. For more information, see
718 include/linux/dma-contiguous.h
720 cmo_free_hint= [PPC] Format: { yes | no }
721 Specify whether pages are marked as being inactive
722 when they are freed. This is used in CMO environments
723 to determine OS memory pressure for page stealing by
727 coherent_pool=nn[KMG] [ARM,KNL]
728 Sets the size of memory pool for coherent, atomic dma
729 allocations, by default set to 256K.
731 code_bytes [X86] How many bytes of object code to print
736 com20020= [HW,NET] ARCnet - COM20020 chipset
738 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
740 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
744 ARCnet - COM90xx chipset (memory-mapped buffers)
745 Format: <io>[,<irq>[,<memstart>]]
747 condev= [HW,S390] console device
750 console= [KNL] Output console device and options.
752 tty<n> Use the virtual console device <n>.
756 Use the specified serial port. The options are of
757 the form "bbbbpnf", where "bbbb" is the baud rate,
758 "p" is parity ("n", "o", or "e"), "n" is number of
759 bits, and "f" is flow control ("r" for RTS or
760 omit it). Default is "9600n8".
762 See Documentation/serial-console.txt for more
764 Documentation/networking/netconsole.txt for an
767 uart[8250],io,<addr>[,options]
768 uart[8250],mmio,<addr>[,options]
769 uart[8250],mmio16,<addr>[,options]
770 uart[8250],mmio32,<addr>[,options]
771 uart[8250],0x<addr>[,options]
772 Start an early, polled-mode console on the 8250/16550
773 UART at the specified I/O port or MMIO address,
774 switching to the matching ttyS device later.
775 MMIO inter-register address stride is either 8-bit
776 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
777 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
778 to be equivalent to 'mmio'. 'options' are specified in
779 the same format described for ttyS above; if unspecified,
780 the h/w is not re-initialized.
782 hvc<n> Use the hypervisor console device <n>. This is for
783 both Xen and PowerPC hypervisors.
785 If the device connected to the port is not a TTY but a braille
786 device, prepend "brl," before the device type, for instance
788 For now, only VisioBraille is supported.
790 consoleblank= [KNL] The console blank (screen saver) timeout in
791 seconds. Defaults to 10*60 = 10mins. A value of 0
792 disables the blank timer.
795 [KNL] Change the default value for
796 /proc/<pid>/coredump_filter.
797 See also Documentation/filesystems/proc.txt.
799 cpuidle.off=1 [CPU_IDLE]
800 disable the cpuidle sub-system
803 [X86] Delay for N microsec between assert and de-assert
804 of APIC INIT to start processors. This delay occurs
805 on every CPU online, such as boot, and resume from suspend.
808 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
810 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
812 crashkernel=size[KMG][@offset[KMG]]
813 [KNL] Using kexec, Linux can switch to a 'crash kernel'
814 upon panic. This parameter reserves the physical
815 memory region [offset, offset + size] for that kernel
816 image. If '@offset' is omitted, then a suitable offset
817 is selected automatically. Check
818 Documentation/kdump/kdump.txt for further details.
820 crashkernel=range1:size1[,range2:size2,...][@offset]
821 [KNL] Same as above, but depends on the memory
822 in the running system. The syntax of range is
823 start-[end] where start and end are both
824 a memory unit (amount[KMG]). See also
825 Documentation/kdump/kdump.txt for an example.
827 crashkernel=size[KMG],high
828 [KNL, x86_64] range could be above 4G. Allow kernel
829 to allocate physical memory region from top, so could
830 be above 4G if system have more than 4G ram installed.
831 Otherwise memory region will be allocated below 4G, if
833 It will be ignored if crashkernel=X is specified.
834 crashkernel=size[KMG],low
835 [KNL, x86_64] range under 4G. When crashkernel=X,high
836 is passed, kernel could allocate physical memory region
837 above 4G, that cause second kernel crash on system
838 that require some amount of low memory, e.g. swiotlb
839 requires at least 64M+32K low memory, also enough extra
840 low memory is needed to make sure DMA buffers for 32-bit
841 devices won't run out. Kernel would try to allocate at
842 at least 256M below 4G automatically.
843 This one let user to specify own low range under 4G
844 for second kernel instead.
845 0: to disable low allocation.
846 It will be ignored when crashkernel=X,high is not used
847 or memory reserved is below 4G.
850 [KNL] Disable crypto self-tests
855 cs89x0_media= [HW,NET]
856 Format: { rj45 | aui | bnc }
859 See header of drivers/s390/block/dasd_devmap.c.
861 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
862 (one device per port)
863 Format: <port#>,<type>
864 See also Documentation/input/joystick-parport.txt
866 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
867 time. See Documentation/dynamic-debug-howto.txt for
868 details. Deprecated, see dyndbg.
870 debug [KNL] Enable kernel debugging (events log level).
873 [KNL] verbose self-tests
875 Print debugging info while doing the locking API
877 We default to 0 (no extra messages), setting it to
878 1 will print _a lot_ more information - normally
879 only useful to kernel developers.
881 debug_objects [KNL] Enable object debugging
884 [KNL] Disable object debugging
886 debug_guardpage_minorder=
887 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
888 parameter allows control of the order of pages that will
889 be intentionally kept free (and hence protected) by the
890 buddy allocator. Bigger value increase the probability
891 of catching random memory corruption, but reduce the
892 amount of memory for normal system use. The maximum
893 possible value is MAX_ORDER/2. Setting this parameter
894 to 1 or 2 should be enough to identify most random
895 memory corruption problems caused by bugs in kernel or
896 driver code when a CPU writes to (or reads from) a
897 random memory location. Note that there exists a class
898 of memory corruptions problems caused by buggy H/W or
899 F/W or by drivers badly programing DMA (basically when
900 memory is written at bus level and the CPU MMU is
901 bypassed) which are not detectable by
902 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
903 tracking down these problems.
906 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
907 parameter enables the feature at boot time. In
908 default, it is disabled. We can avoid allocating huge
909 chunk of memory for debug pagealloc if we don't enable
910 it at boot time and the system will work mostly same
911 with the kernel built without CONFIG_DEBUG_PAGEALLOC.
912 on: enable the feature
914 debugpat [X86] Enable PAT debugging
916 decnet.addr= [HW,NET]
917 Format: <area>[,<node>]
918 See also Documentation/networking/decnet.txt.
921 [same as hugepagesz=] The size of the default
922 HugeTLB page size. This is the size represented by
923 the legacy /proc/ hugepages APIs, used for SHM, and
924 default size when mounting hugetlbfs filesystems.
925 Defaults to the default architecture's huge page size
929 Set number of hash buckets for dentry cache.
932 See Documentation/networking/ipv6.txt.
934 disable_cpu_apicid= [X86,APIC,SMP]
936 The number of initial APIC ID for the
937 corresponding CPU to be disabled at boot,
938 mostly used for the kdump 2nd kernel to
939 disable BSP to wake up multiple CPUs without
940 causing system reset or hang due to sending
943 disable_ddw [PPC/PSERIES]
944 Disable Dynamic DMA Window support. Use this if
945 to workaround buggy firmware.
948 See Documentation/networking/ipv6.txt.
950 disable_mtrr_cleanup [X86]
951 The kernel tries to adjust MTRR layout from continuous
952 to discrete, to make X server driver able to add WB
953 entry later. This parameter disables that.
955 disable_mtrr_trim [X86, Intel and AMD only]
956 By default the kernel will trim any uncacheable
957 memory out of your available memory pool based on
958 MTRR settings. This parameter disables that behavior,
959 possibly causing your machine to run very slowly.
961 disable_timer_pin_1 [X86]
962 Disable PIN 1 of APIC timer
963 Can be useful to work around chipset bugs.
965 dis_ucode_ldr [X86] Disable the microcode loader.
967 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
968 this option disables the debugging code at boot.
970 dma_debug_entries=<number>
971 This option allows to tune the number of preallocated
972 entries for DMA-API debugging code. One entry is
973 required per DMA-API allocation. Use this if the
974 DMA-API debugging code disables itself because the
975 architectural default is too low.
977 dma_debug_driver=<driver_name>
978 With this option the DMA-API debugging driver
979 filter feature can be enabled at boot time. Just
980 pass the driver to filter for as the parameter.
981 The filter can be disabled or changed to another
982 driver later using sysfs.
984 drm_kms_helper.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
985 Broken monitors, graphic adapters, KVMs and EDIDless
986 panels may send no or incorrect EDID data sets.
987 This parameter allows to specify an EDID data sets
988 in the /lib/firmware directory that are used instead.
989 Generic built-in EDID data sets are used, if one of
990 edid/1024x768.bin, edid/1280x1024.bin,
991 edid/1680x1050.bin, or edid/1920x1080.bin is given
992 and no file with the same name exists. Details and
993 instructions how to build your own EDID data are
994 available in Documentation/EDID/HOWTO.txt. An EDID
995 data set will only be used for a particular connector,
996 if its name and a colon are prepended to the EDID
997 name. Each connector may use a unique EDID data
998 set by separating the files with a comma. An EDID
999 data set with no connector name will be used for
1000 any connectors not explicitly specified.
1004 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1005 module.dyndbg[="val"]
1006 Enable debug messages at boot time. See
1007 Documentation/dynamic-debug-howto.txt for details.
1009 nompx [X86] Disables Intel Memory Protection Extensions.
1010 See Documentation/x86/intel_mpx.txt for more
1011 information about the feature.
1013 nopku [X86] Disable Memory Protection Keys CPU feature found
1017 on enable eager fpu restore
1018 off disable eager fpu restore
1019 auto selects the default scheme, which automatically
1020 enables eagerfpu restore for xsaveopt.
1022 module.async_probe [KNL]
1023 Enable asynchronous probe on this module.
1025 early_ioremap_debug [KNL]
1026 Enable debug messages in early_ioremap support. This
1027 is useful for tracking down temporary early mappings
1028 which are not unmapped.
1030 earlycon= [KNL] Output early console device and options.
1032 When used with no options, the early console is
1033 determined by the stdout-path property in device
1037 Start an early, polled-mode console on a cadence serial
1038 port at the specified address. The cadence serial port
1039 must already be setup and configured. Options are not
1042 uart[8250],io,<addr>[,options]
1043 uart[8250],mmio,<addr>[,options]
1044 uart[8250],mmio32,<addr>[,options]
1045 uart[8250],mmio32be,<addr>[,options]
1046 uart[8250],0x<addr>[,options]
1047 Start an early, polled-mode console on the 8250/16550
1048 UART at the specified I/O port or MMIO address.
1049 MMIO inter-register address stride is either 8-bit
1050 (mmio) or 32-bit (mmio32 or mmio32be).
1051 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1052 to be equivalent to 'mmio'. 'options' are specified
1053 in the same format described for "console=ttyS<n>"; if
1054 unspecified, the h/w is not initialized.
1058 Start an early, polled-mode console on a pl011 serial
1059 port at the specified address. The pl011 serial port
1060 must already be setup and configured. Options are not
1061 yet supported. If 'mmio32' is specified, then only
1062 the driver will use only 32-bit accessors to read/write
1063 the device registers.
1066 Start an early, polled-mode console on a meson serial
1067 port at the specified address. The serial port must
1068 already be setup and configured. Options are not yet
1072 Start an early, polled-mode console on an msm serial
1073 port at the specified address. The serial port
1074 must already be setup and configured. Options are not
1077 msm_serial_dm,<addr>
1078 Start an early, polled-mode console on an msm serial
1079 dm port at the specified address. The serial port
1080 must already be setup and configured. Options are not
1083 smh Use ARM semihosting calls for early console.
1091 Use early console provided by serial driver available
1092 on Samsung SoCs, requires selecting proper type and
1093 a correct base address of the selected UART port. The
1094 serial port must already be setup and configured.
1095 Options are not yet supported.
1099 Use early console provided by Freescale LP UART driver
1100 found on Freescale Vybrid and QorIQ LS1021A processors.
1101 A valid base address must be provided, and the serial
1102 port must already be setup and configured.
1104 armada3700_uart,<addr>
1105 Start an early, polled-mode console on the
1106 Armada 3700 serial port at the specified
1107 address. The serial port must already be setup
1108 and configured. Options are not yet supported.
1110 earlyprintk= [X86,SH,BLACKFIN,ARM,M68k]
1114 earlyprintk=serial[,ttySn[,baudrate]]
1115 earlyprintk=serial[,0x...[,baudrate]]
1116 earlyprintk=ttySn[,baudrate]
1117 earlyprintk=dbgp[debugController#]
1118 earlyprintk=pciserial,bus:device.function[,baudrate]
1120 earlyprintk is useful when the kernel crashes before
1121 the normal console is initialized. It is not enabled by
1122 default because it has some cosmetic problems.
1124 Append ",keep" to not disable it when the real console
1127 Only one of vga, efi, serial, or usb debug port can
1130 Currently only ttyS0 and ttyS1 may be specified by
1131 name. Other I/O ports may be explicitly specified
1132 on some architectures (x86 and arm at least) by
1133 replacing ttySn with an I/O port address, like this:
1134 earlyprintk=serial,0x1008,115200
1135 You can find the port for a given device in
1136 /proc/tty/driver/serial:
1137 2: uart:ST16650V2 port:00001008 irq:18 ...
1139 Interaction with the standard serial driver is not
1142 The VGA and EFI output is eventually overwritten by
1145 The xen output can only be used by Xen PV guests.
1147 edac_report= [HW,EDAC] Control how to report EDAC event
1148 Format: {"on" | "off" | "force"}
1149 on: enable EDAC to report H/W event. May be overridden
1150 by other higher priority error reporting module.
1151 off: disable H/W event reporting through EDAC.
1152 force: enforce the use of EDAC to report H/W event.
1155 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1158 This is designed to be used in conjunction with
1159 the boot argument: earlyprintk=vga
1162 Format: {"off" | "on" | "skip[mbr]"}
1165 Format: { "old_map", "nochunk", "noruntime", "debug" }
1166 old_map [X86-64]: switch to the old ioremap-based EFI
1167 runtime services mapping. 32-bit still uses this one by
1169 nochunk: disable reading files in "chunks" in the EFI
1170 boot stub, as chunking can cause problems with some
1171 firmware implementations.
1172 noruntime : disable EFI runtime services support
1173 debug: enable misc debug output
1175 efi_no_storage_paranoia [EFI; X86]
1176 Using this parameter you can use more than 50% of
1177 your efi variable storage. Use this parameter only if
1178 you are really sure that your UEFI does sane gc and
1179 fulfills the spec otherwise your board may brick.
1181 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1182 Add arbitrary attribute to specific memory range by
1183 updating original EFI memory map.
1184 Region of memory which aa attribute is added to is
1186 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1187 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1188 attribute is added to range 0x100000000-0x180000000 and
1189 0x10a0000000-0x1120000000.
1191 Using this parameter you can do debugging of EFI memmap
1192 related feature. For example, you can do debugging of
1193 Address Range Mirroring feature even if your box
1196 eisa_irq_edge= [PARISC,HW]
1197 See header of drivers/parisc/eisa.c.
1200 See comment before function elanfreq_setup() in
1201 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1204 Format: {"cfq" | "deadline" | "noop"}
1205 See Documentation/block/cfq-iosched.txt and
1206 Documentation/block/deadline-iosched.txt for details.
1208 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1209 Specifies physical address of start of kernel core
1210 image elf header and optionally the size. Generally
1211 kexec loader will pass this option to capture kernel.
1212 See Documentation/kdump/kdump.txt for details.
1214 enable_mtrr_cleanup [X86]
1215 The kernel tries to adjust MTRR layout from continuous
1216 to discrete, to make X server driver able to add WB
1217 entry later. This parameter enables that.
1219 enable_timer_pin_1 [X86]
1220 Enable PIN 1 of APIC timer
1221 Can be useful to work around chipset bugs
1222 (in particular on some ATI chipsets).
1223 The kernel tries to set a reasonable default.
1225 enforcing [SELINUX] Set initial enforcing status.
1227 See security/selinux/Kconfig help text.
1228 0 -- permissive (log only, no denials).
1229 1 -- enforcing (deny and log).
1231 Value can be changed at runtime via /selinux/enforce.
1234 Disable Error Record Serialization Table (ERST)
1237 ether= [HW,NET] Ethernet cards parameters
1238 This option is obsoleted by the "netdev=" option, which
1239 has equivalent usage. See its documentation for details.
1243 Permit 'security.evm' to be updated regardless of
1244 current integrity status.
1248 fail_make_request=[KNL]
1249 General fault injection mechanism.
1250 Format: <interval>,<probability>,<space>,<times>
1251 See also Documentation/fault-injection/.
1254 See Documentation/blockdev/floppy.txt.
1256 force_pal_cache_flush
1257 [IA-64] Avoid check_sal_cache_flush which may hang on
1258 buggy SAL_CACHE_FLUSH implementations. Using this
1259 parameter will force ia64_sal_cache_flush to call
1260 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1263 Forcefully enable Physical Address Extension (PAE).
1264 Many Pentium M systems disable PAE but may have a
1265 functionally usable PAE implementation.
1266 Warning: use of this parameter will taint the kernel
1267 and may cause unknown problems.
1270 [FTRACE] will set and start the specified tracer
1271 as early as possible in order to facilitate early
1274 ftrace_dump_on_oops[=orig_cpu]
1275 [FTRACE] will dump the trace buffers on oops.
1276 If no parameter is passed, ftrace will dump
1277 buffers of all CPUs, but if you pass orig_cpu, it will
1278 dump only the buffer of the CPU that triggered the
1281 ftrace_filter=[function-list]
1282 [FTRACE] Limit the functions traced by the function
1283 tracer at boot up. function-list is a comma separated
1284 list of functions. This list can be changed at run
1285 time by the set_ftrace_filter file in the debugfs
1288 ftrace_notrace=[function-list]
1289 [FTRACE] Do not trace the functions specified in
1290 function-list. This list can be changed at run time
1291 by the set_ftrace_notrace file in the debugfs
1294 ftrace_graph_filter=[function-list]
1295 [FTRACE] Limit the top level callers functions traced
1296 by the function graph tracer at boot up.
1297 function-list is a comma separated list of functions
1298 that can be changed at run time by the
1299 set_graph_function file in the debugfs tracing directory.
1301 ftrace_graph_notrace=[function-list]
1302 [FTRACE] Do not trace from the functions specified in
1303 function-list. This list is a comma separated list of
1304 functions that can be changed at run time by the
1305 set_graph_notrace file in the debugfs tracing directory.
1308 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1309 support via parallel port (up to 5 devices per port)
1310 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1311 See also Documentation/input/joystick-parport.txt
1315 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1319 gcov_persist= [GCOV] When non-zero (default), profiling data for
1320 kernel modules is saved and remains accessible via
1321 debugfs, even when the module is unloaded/reloaded.
1322 When zero, profiling data is discarded and associated
1323 debugfs files are removed at module unload time.
1325 gpt [EFI] Forces disk with valid GPT signature but
1326 invalid Protective MBR to be treated as GPT. If the
1327 primary GPT is corrupted, it enables the backup/alternate
1328 GPT to be used instead.
1330 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1331 the "Enable 0" bit of the configuration register.
1334 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1335 the "Enable 0" bit of the configuration register.
1338 grcan.select= [HW] Select which physical interface to use.
1341 grcan.txsize= [HW] Sets the size of the tx buffer.
1342 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1344 grcan.rxsize= [HW] Sets the size of the rx buffer.
1345 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1348 hardlockup_all_cpu_backtrace=
1349 [KNL] Should the hard-lockup detector generate
1350 backtraces on all cpus.
1353 hashdist= [KNL,NUMA] Large hashes allocated during boot
1354 are distributed across NUMA nodes. Defaults on
1355 for 64-bit NUMA, off otherwise.
1356 Format: 0 | 1 (for off | on)
1358 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1360 hd= [EIDE] (E)IDE hard drive subsystem geometry
1361 Format: <cyl>,<head>,<sect>
1364 Disable Hardware Error Source Table (HEST) support;
1365 corresponding firmware-first mode error processing
1366 logic will be disabled.
1368 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1369 size of <nn>. This works even on boxes that have no
1370 highmem otherwise. This also works to reduce highmem
1371 size on bigger boxes.
1373 highres= [KNL] Enable/disable high resolution timer mode.
1374 Valid parameters: "on", "off"
1378 See Documentation/isdn/README.HiSax.
1382 hpet= [X86-32,HPET] option to control HPET usage
1383 Format: { enable (default) | disable | force |
1385 disable: disable HPET and use PIT instead
1386 force: allow force enabled of undocumented chips (ICH4,
1388 verbose: show contents of HPET registers during setup
1390 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1391 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1393 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1394 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
1395 On x86-64 and powerpc, this option can be specified
1396 multiple times interleaved with hugepages= to reserve
1397 huge pages of different sizes. Valid pages sizes on
1398 x86-64 are 2M (when the CPU supports "pse") and 1G
1399 (when the CPU supports the "pdpe1gb" cpuinfo flag).
1401 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1402 terminal devices. Valid values: 0..8
1403 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1404 If specified, z/VM IUCV HVC accepts connections
1405 from listed z/VM user IDs only.
1407 hwthread_map= [METAG] Comma-separated list of Linux cpu id to
1408 hardware thread id mappings.
1409 Format: <cpu>:<hwthread>
1412 Do not unregister boot console at start. This is only
1413 useful for debugging when something happens in the window
1414 between unregistering the boot console and initializing
1417 i2c_bus= [HW] Override the default board specific I2C bus speed
1418 or register an additional I2C bus that is not
1419 registered from board initialization code.
1423 i8042.debug [HW] Toggle i8042 debug mode
1424 i8042.unmask_kbd_data
1425 [HW] Enable printing of interrupt data from the KBD port
1426 (disabled by default, and as a pre-condition
1427 requires that i8042.debug=1 be enabled)
1428 i8042.direct [HW] Put keyboard port into non-translated mode
1429 i8042.dumbkbd [HW] Pretend that controller can only read data from
1430 keyboard and cannot control its state
1431 (Don't attempt to blink the leds)
1432 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1433 i8042.nokbd [HW] Don't check/create keyboard port
1434 i8042.noloop [HW] Disable the AUX Loopback command while probing
1436 i8042.nomux [HW] Don't check presence of an active multiplexing
1438 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1440 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1441 i8042.reset [HW] Reset the controller during init and cleanup
1442 i8042.unlock [HW] Unlock (ignore) the keylock
1443 i8042.kbdreset [HW] Reset device connected to KBD port
1447 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1448 indicates that the driver is running on unsupported
1450 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1451 does not match list of supported models.
1453 [HW] Report power status in /proc/i8k
1454 (disabled by default)
1455 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1458 i915.invert_brightness=
1459 [DRM] Invert the sense of the variable that is used to
1460 set the brightness of the panel backlight. Normally a
1461 brightness value of 0 indicates backlight switched off,
1462 and the maximum of the brightness value sets the backlight
1463 to maximum brightness. If this parameter is set to 0
1464 (default) and the machine requires it, or this parameter
1465 is set to 1, a brightness value of 0 sets the backlight
1466 to maximum brightness, and the maximum of the brightness
1467 value switches the backlight off.
1468 -1 -- never invert brightness
1469 0 -- machine default
1470 1 -- force brightness inversion
1473 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1475 ide-core.nodma= [HW] (E)IDE subsystem
1476 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1477 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1478 .cdrom .chs .ignore_cable are additional options
1479 See Documentation/ide/ide.txt.
1481 ide-generic.probe-mask= [HW] (E)IDE subsystem
1483 Probe mask for legacy ISA IDE ports. Depending on
1484 platform up to 6 ports are supported, enabled by
1485 setting corresponding bits in the mask to 1. The
1486 default value is 0x0, which has a special meaning.
1487 On systems that have PCI, it triggers scanning the
1488 PCI bus for the first and the second port, which
1489 are then probed. On systems without PCI the value
1490 of 0x0 enables probing the two first ports as if it
1493 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1494 Claim all unknown PCI IDE storage controllers.
1497 Format: idle=poll, idle=halt, idle=nomwait
1498 Poll forces a polling idle loop that can slightly
1499 improve the performance of waking up a idle CPU, but
1500 will use a lot of power and make the system run hot.
1502 idle=halt: Halt is forced to be used for CPU idle.
1503 In such case C2/C3 won't be used again.
1504 idle=nomwait: Disable mwait for CPU C-states
1506 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1507 Format: { strict | legacy | 2008 | relaxed }
1510 Choose which programs will be accepted for execution
1511 based on the IEEE 754 NaN encoding(s) supported by
1512 the FPU and the NaN encoding requested with the value
1513 of an ELF file header flag individually set by each
1514 binary. Hardware implementations are permitted to
1515 support either or both of the legacy and the 2008 NaN
1518 Available settings are as follows:
1519 strict accept binaries that request a NaN encoding
1520 supported by the FPU
1521 legacy only accept legacy-NaN binaries, if supported
1523 2008 only accept 2008-NaN binaries, if supported
1525 relaxed accept any binaries regardless of whether
1526 supported by the FPU
1528 The FPU emulator is always able to support both NaN
1529 encodings, so if no FPU hardware is present or it has
1530 been disabled with 'nofpu', then the settings of
1531 'legacy' and '2008' strap the emulator accordingly,
1532 'relaxed' straps the emulator for both legacy-NaN and
1533 2008-NaN, whereas 'strict' enables legacy-NaN only on
1534 legacy processors and both NaN encodings on MIPS32 or
1537 The setting for ABS.fmt/NEG.fmt instruction execution
1538 mode generally follows that for the NaN encoding,
1539 except where unsupported by hardware.
1541 ignore_loglevel [KNL]
1542 Ignore loglevel setting - this will print /all/
1543 kernel messages to the console. Useful for debugging.
1544 We also add it as printk module parameter, so users
1545 could change it dynamically, usually by
1546 /sys/module/printk/parameters/ignore_loglevel.
1549 Ignore RLIMIT_DATA setting for data mappings,
1550 print warning at first misuse. Can be changed via
1551 /sys/module/kernel/parameters/ignore_rlimit_data.
1553 ihash_entries= [KNL]
1554 Set number of hash buckets for inode cache.
1556 ima_appraise= [IMA] appraise integrity measurements
1557 Format: { "off" | "enforce" | "fix" | "log" }
1560 ima_appraise_tcb [IMA]
1561 The builtin appraise policy appraises all files
1565 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1569 The list of supported hash algorithms is defined
1570 in crypto/hash_info.h.
1573 The builtin measurement policy to load during IMA
1574 setup. Specyfing "tcb" as the value, measures all
1575 programs exec'd, files mmap'd for exec, and all files
1576 opened with the read mode bit set by either the
1577 effective uid (euid=0) or uid=0.
1580 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1581 Load a policy which meets the needs of the Trusted
1582 Computing Base. This means IMA will measure all
1583 programs exec'd, files mmap'd for exec, and all files
1584 opened for read by uid=0.
1587 Select one of defined IMA measurements template formats.
1588 Formats: { "ima" | "ima-ng" | "ima-sig" }
1592 [IMA] Define a custom template format.
1593 Format: { "field1|...|fieldN" }
1595 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1596 Format: <min_file_size>
1597 Set the minimal file size for using asynchronous hash.
1598 If left unspecified, ahash usage is disabled.
1600 ahash performance varies for different data sizes on
1601 different crypto accelerators. This option can be used
1602 to achieve the best performance for a particular HW.
1604 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1606 Set hashing buffer size. Default: 4k.
1608 ahash performance varies for different chunk sizes on
1609 different crypto accelerators. This option can be used
1610 to achieve best performance for particular HW.
1614 Run specified binary instead of /sbin/init as init
1617 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1618 for working out where the kernel is dying during
1621 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1622 initcall functions. Useful for debugging built-in
1623 modules and initcalls.
1625 initrd= [BOOT] Specify the location of the initial ramdisk
1627 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1630 int_pln_enable [x86] Enable power limit notification interrupt
1632 integrity_audit=[IMA]
1633 Format: { "0" | "1" }
1634 0 -- basic integrity auditing messages. (Default)
1635 1 -- additional integrity auditing messages.
1637 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1639 Enable intel iommu driver.
1641 Disable intel iommu driver.
1642 igfx_off [Default Off]
1643 By default, gfx is mapped as normal device. If a gfx
1644 device has a dedicated DMAR unit, the DMAR unit is
1645 bypassed by not enabling DMAR with this option. In
1646 this case, gfx device will use physical address for
1649 With this option iommu will not optimize to look
1650 for io virtual address below 32-bit forcing dual
1651 address cycle on pci bus for cards supporting greater
1652 than 32-bit addressing. The default is to look
1653 for translation below 32-bit and if not available
1654 then look in the higher range.
1655 strict [Default Off]
1656 With this option on every unmap_single operation will
1657 result in a hardware IOTLB flush operation as opposed
1658 to batching them for performance.
1659 sp_off [Default Off]
1660 By default, super page will be supported if Intel IOMMU
1661 has the capability. With this option, super page will
1663 ecs_off [Default Off]
1664 By default, extended context tables will be supported if
1665 the hardware advertises that it has support both for the
1666 extended tables themselves, and also PASID support. With
1667 this option set, extended tables will not be used even
1668 on hardware which claims to support them.
1670 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1671 0 disables intel_idle and fall back on acpi_idle.
1672 1 to 6 specify maximum depth of C-state.
1676 Do not enable intel_pstate as the default
1677 scaling driver for the supported processors
1679 Enable intel_pstate on systems that prohibit it by default
1680 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1681 instead of acpi-cpufreq may disable platform features, such
1682 as thermal controls and power capping, that rely on ACPI
1683 P-States information being indicated to OSPM and therefore
1684 should be used with caution. This option does not work with
1685 processors that aren't supported by the intel_pstate driver
1686 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1688 Do not enable hardware P state control (HWP)
1691 Only load intel_pstate on systems which support
1692 hardware P state control (HWP) if available.
1694 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1695 Description Table, specifies preferred power management
1696 profile as "Enterprise Server" or "Performance Server",
1697 then this feature is turned on by default.
1699 intremap= [X86-64, Intel-IOMMU]
1700 on enable Interrupt Remapping (default)
1701 off disable Interrupt Remapping
1702 nosid disable Source ID checking
1704 BIOS x2APIC opt-out request will be ignored
1705 nopost disable Interrupt Posting
1707 iomem= Disable strict checking of access to MMIO memory
1708 strict regions from userspace.
1723 nobypass [PPC/POWERNV]
1724 Disable IOMMU bypass, using IOMMU for PCI devices.
1727 io7= [HW] IO7 for Marvel based alpha systems
1728 See comment before marvel_specify_io7 in
1729 arch/alpha/kernel/core_marvel.c.
1731 io_delay= [X86] I/O delay method
1733 Standard port 0x80 based delay
1735 Alternate port 0xed based delay (needed on some systems)
1737 Simple two microseconds delay
1742 See Documentation/filesystems/nfs/nfsroot.txt.
1744 irqaffinity= [SMP] Set the default irq affinity mask
1746 <cpu number>,...,<cpu number>
1748 <cpu number>-<cpu number>
1749 (must be a positive range in ascending order)
1751 <cpu number>,...,<cpu number>-<cpu number>
1754 When an interrupt is not handled search all handlers
1755 for it. Intended to get systems with badly broken
1759 When an interrupt is not handled search all handlers
1760 for it. Also check all handlers each timer
1761 interrupt. Intended to get systems with badly broken
1765 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1767 isolcpus= [KNL,SMP] Isolate CPUs from the general scheduler.
1769 <cpu number>,...,<cpu number>
1771 <cpu number>-<cpu number>
1772 (must be a positive range in ascending order)
1774 <cpu number>,...,<cpu number>-<cpu number>
1776 This option can be used to specify one or more CPUs
1777 to isolate from the general SMP balancing and scheduling
1778 algorithms. You can move a process onto or off an
1779 "isolated" CPU via the CPU affinity syscalls or cpuset.
1780 <cpu number> begins at 0 and the maximum value is
1781 "number of CPUs in system - 1".
1783 This option is the preferred way to isolate CPUs. The
1784 alternative -- manually setting the CPU mask of all
1785 tasks in the system -- can cause problems and
1786 suboptimal load balancer performance.
1790 ivrs_ioapic [HW,X86_64]
1791 Provide an override to the IOAPIC-ID<->DEVICE-ID
1792 mapping provided in the IVRS ACPI table. For
1793 example, to map IOAPIC-ID decimal 10 to
1794 PCI device 00:14.0 write the parameter as:
1795 ivrs_ioapic[10]=00:14.0
1797 ivrs_hpet [HW,X86_64]
1798 Provide an override to the HPET-ID<->DEVICE-ID
1799 mapping provided in the IVRS ACPI table. For
1800 example, to map HPET-ID decimal 0 to
1801 PCI device 00:14.0 write the parameter as:
1802 ivrs_hpet[0]=00:14.0
1804 ivrs_acpihid [HW,X86_64]
1805 Provide an override to the ACPI-HID:UID<->DEVICE-ID
1806 mapping provided in the IVRS ACPI table. For
1807 example, to map UART-HID:UID AMD0020:0 to
1808 PCI device 00:14.5 write the parameter as:
1809 ivrs_acpihid[00:14.5]=AMD0020:0
1811 js= [HW,JOY] Analog joystick
1812 See Documentation/input/joystick.txt.
1815 When CONFIG_RANDOMIZE_BASE is set, this disables
1816 kernel and module base offset ASLR (Address Space
1817 Layout Randomization).
1821 kernelcore= [KNL,X86,IA-64,PPC]
1822 Format: nn[KMGTPE] | "mirror"
1824 specifies the amount of memory usable by the kernel
1825 for non-movable allocations. The requested amount is
1826 spread evenly throughout all nodes in the system. The
1827 remaining memory in each node is used for Movable
1828 pages. In the event, a node is too small to have both
1829 kernelcore and Movable pages, kernelcore pages will
1830 take priority and other nodes will have a larger number
1831 of Movable pages. The Movable zone is used for the
1832 allocation of pages that may be reclaimed or moved
1833 by the page migration subsystem. This means that
1834 HugeTLB pages may not be allocated from this zone.
1835 Note that allocations like PTEs-from-HighMem still
1836 use the HighMem zone if it exists, and the Normal
1837 zone if it does not.
1839 Instead of specifying the amount of memory (nn[KMGTPE]),
1840 you can specify "mirror" option. In case "mirror"
1841 option is specified, mirrored (reliable) memory is used
1842 for non-movable allocations and remaining memory is used
1843 for Movable pages. nn[KMGTPE] and "mirror" are exclusive,
1844 so you can NOT specify nn[KMGTPE] and "mirror" at the same
1847 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
1848 Format: <Controller#>[,poll interval]
1849 The controller # is the number of the ehci usb debug
1850 port as it is probed via PCI. The poll interval is
1851 optional and is the number seconds in between
1852 each poll cycle to the debug port in case you need
1853 the functionality for interrupting the kernel with
1854 gdb or control-c on the dbgp connection. When
1855 not using this parameter you use sysrq-g to break into
1856 the kernel debugger.
1858 kgdboc= [KGDB,HW] kgdb over consoles.
1859 Requires a tty driver that supports console polling,
1860 or a supported polling keyboard driver (non-usb).
1861 Serial only format: <serial_device>[,baud]
1862 keyboard only format: kbd
1863 keyboard and serial format: kbd,<serial_device>[,baud]
1864 Optional Kernel mode setting:
1865 kms, kbd format: kms,kbd
1866 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
1868 kgdbwait [KGDB] Stop kernel execution and enter the
1869 kernel debugger at the earliest opportunity.
1871 kmac= [MIPS] korina ethernet MAC address.
1872 Configure the RouterBoard 532 series on-chip
1873 Ethernet adapter MAC address.
1875 kmemleak= [KNL] Boot-time kmemleak enable/disable
1876 Valid arguments: on, off
1878 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
1881 kmemcheck= [X86] Boot-time kmemcheck enable/disable/one-shot mode
1882 Valid arguments: 0, 1, 2
1883 kmemcheck=0 (disabled)
1884 kmemcheck=1 (enabled)
1885 kmemcheck=2 (one-shot mode)
1886 Default: 2 (one-shot mode)
1888 kstack=N [X86] Print N words from the kernel stack
1891 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
1892 Default is 0 (don't ignore, but inject #GP)
1894 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
1898 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
1899 Default is 1 (enabled)
1901 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
1903 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
1905 kvm-intel.ept= [KVM,Intel] Disable extended page tables
1906 (virtualized MMU) support on capable Intel chips.
1907 Default is 1 (enabled)
1909 kvm-intel.emulate_invalid_guest_state=
1910 [KVM,Intel] Enable emulation of invalid guest states
1911 Default is 0 (disabled)
1913 kvm-intel.flexpriority=
1914 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
1915 Default is 1 (enabled)
1918 [KVM,Intel] Enable VMX nesting (nVMX).
1919 Default is 0 (disabled)
1921 kvm-intel.unrestricted_guest=
1922 [KVM,Intel] Disable unrestricted guest feature
1923 (virtualized real and unpaged mode) on capable
1924 Intel chips. Default is 1 (enabled)
1926 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
1927 feature (tagged TLBs) on capable Intel chips.
1928 Default is 1 (enabled)
1934 lapic [X86-32,APIC] Enable the local APIC even if BIOS
1937 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
1938 value for LAPIC timer one-shot implementation. Default
1939 back to the programmable timer unit in the LAPIC.
1941 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
1944 libata.dma= [LIBATA] DMA control
1945 libata.dma=0 Disable all PATA and SATA DMA
1946 libata.dma=1 PATA and SATA Disk DMA only
1947 libata.dma=2 ATAPI (CDROM) DMA only
1948 libata.dma=4 Compact Flash DMA only
1949 Combinations also work, so libata.dma=3 enables DMA
1950 for disks and CDROMs, but not CFs.
1952 libata.ignore_hpa= [LIBATA] Ignore HPA limit
1953 libata.ignore_hpa=0 keep BIOS limits (default)
1954 libata.ignore_hpa=1 ignore limits, using full disk
1956 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
1960 libata.force= [LIBATA] Force configurations. The format is comma
1961 separated list of "[ID:]VAL" where ID is
1962 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
1963 matching port, link or device. Basically, it matches
1964 the ATA ID string printed on console by libata. If
1965 the whole ID part is omitted, the last PORT and DEVICE
1966 values are used. If ID hasn't been specified yet, the
1967 configuration applies to all ports, links and devices.
1969 If only DEVICE is omitted, the parameter applies to
1970 the port and all links and devices behind it. DEVICE
1971 number of 0 either selects the first device or the
1972 first fan-out link behind PMP device. It does not
1973 select the host link. DEVICE number of 15 selects the
1974 host link and device attached to it.
1976 The VAL specifies the configuration to force. As long
1977 as there's no ambiguity shortcut notation is allowed.
1978 For example, both 1.5 and 1.5G would work for 1.5Gbps.
1979 The following configurations can be forced.
1981 * Cable type: 40c, 80c, short40c, unk, ign or sata.
1982 Any ID with matching PORT is used.
1984 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
1986 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
1987 udma[/][16,25,33,44,66,100,133] notation is also
1990 * [no]ncq: Turn on or off NCQ.
1992 * [no]ncqtrim: Turn off queued DSM TRIM.
1994 * nohrst, nosrst, norst: suppress hard, soft
1997 * rstonce: only attempt one reset during
1998 hot-unplug link recovery
2000 * dump_id: dump IDENTIFY data.
2002 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2004 * disable: Disable this device.
2006 If there are multiple matching configurations changing
2007 the same attribute, the last one is used.
2009 memblock=debug [KNL] Enable memblock debug messages.
2011 load_ramdisk= [RAM] List of ramdisks to load from floppy
2012 See Documentation/blockdev/ramdisk.txt.
2014 lockd.nlm_grace_period=P [NFS] Assign grace period.
2017 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2020 lockd.nlm_timeout=T [NFS] Assign timeout value.
2023 lockd.nlm_udpport=M [NFS] Assign UDP port.
2026 locktorture.nreaders_stress= [KNL]
2027 Set the number of locking read-acquisition kthreads.
2028 Defaults to being automatically set based on the
2029 number of online CPUs.
2031 locktorture.nwriters_stress= [KNL]
2032 Set the number of locking write-acquisition kthreads.
2034 locktorture.onoff_holdoff= [KNL]
2035 Set time (s) after boot for CPU-hotplug testing.
2037 locktorture.onoff_interval= [KNL]
2038 Set time (s) between CPU-hotplug operations, or
2039 zero to disable CPU-hotplug testing.
2041 locktorture.shuffle_interval= [KNL]
2042 Set task-shuffle interval (jiffies). Shuffling
2043 tasks allows some CPUs to go into dyntick-idle
2044 mode during the locktorture test.
2046 locktorture.shutdown_secs= [KNL]
2047 Set time (s) after boot system shutdown. This
2048 is useful for hands-off automated testing.
2050 locktorture.stat_interval= [KNL]
2051 Time (s) between statistics printk()s.
2053 locktorture.stutter= [KNL]
2054 Time (s) to stutter testing, for example,
2055 specifying five seconds causes the test to run for
2056 five seconds, wait for five seconds, and so on.
2057 This tests the locking primitive's ability to
2058 transition abruptly to and from idle.
2060 locktorture.torture_runnable= [BOOT]
2061 Start locktorture running at boot time.
2063 locktorture.torture_type= [KNL]
2064 Specify the locking implementation to test.
2066 locktorture.verbose= [KNL]
2067 Enable additional printk() statements.
2069 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2072 loglevel= All Kernel Messages with a loglevel smaller than the
2073 console loglevel will be printed to the console. It can
2074 also be changed with klogd or other programs. The
2075 loglevels are defined as follows:
2077 0 (KERN_EMERG) system is unusable
2078 1 (KERN_ALERT) action must be taken immediately
2079 2 (KERN_CRIT) critical conditions
2080 3 (KERN_ERR) error conditions
2081 4 (KERN_WARNING) warning conditions
2082 5 (KERN_NOTICE) normal but significant condition
2083 6 (KERN_INFO) informational
2084 7 (KERN_DEBUG) debug-level messages
2086 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2087 in bytes. n must be a power of two and greater
2088 than the minimal size. The minimal size is defined
2089 by LOG_BUF_SHIFT kernel config parameter. There is
2090 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2091 that allows to increase the default size depending on
2092 the number of CPUs. See init/Kconfig for more details.
2094 logo.nologo [FB] Disables display of the built-in Linux logo.
2095 This may be used to provide more screen space for
2096 kernel log messages and is useful when debugging
2097 kernel boot problems.
2099 lp=0 [LP] Specify parallel ports to use, e.g,
2100 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2101 lp=reset first parallel port). 'lp=0' disables the
2102 lp=auto printer driver. 'lp=reset' (which can be
2103 specified in addition to the ports) causes
2104 attached printers to be reset. Using
2105 lp=port1,port2,... specifies the parallel ports
2106 to associate lp devices with, starting with
2107 lp0. A port specification may be 'none' to skip
2108 that lp device, or a parport name such as
2109 'parport0'. Specifying 'lp=auto' instead of a
2110 port specification list means that device IDs
2111 from each port should be examined, to see if
2112 an IEEE 1284-compliant printer is attached; if
2113 so, the driver will manage that printer.
2114 See also header of drivers/char/lp.c.
2117 Sets loops_per_jiffy to given constant, thus avoiding
2118 time-consuming boot-time autodetection (up to 250 ms per
2119 CPU). 0 enables autodetection (default). To determine
2120 the correct value for your kernel, boot with normal
2121 autodetection and see what value is printed. Note that
2122 on SMP systems the preset will be applied to all CPUs,
2123 which is likely to cause problems if your CPUs need
2124 significantly divergent settings. An incorrect value
2125 will cause delays in the kernel to be wrong, leading to
2126 unpredictable I/O errors and other breakage. Although
2127 unlikely, in the extreme case this might damage your
2131 Format: <io>,<irq>,<dma>
2133 machvec= [IA-64] Force the use of a particular machine-vector
2134 (machvec) in a generic kernel.
2135 Example: machvec=hpzx1_swiotlb
2137 machtype= [Loongson] Share the same kernel image file between different
2139 Example: machtype=lemote-yeeloong-2f-7inch
2141 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2142 than or equal to this physical address is ignored.
2144 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2145 should make use of. maxcpus=n : n >= 0 limits the
2146 kernel to using 'n' processors. n=0 is a special case,
2147 it is equivalent to "nosmp", which also disables
2150 max_loop= [LOOP] The number of loop block devices that get
2151 (loop.max_loop) unconditionally pre-created at init time. The default
2152 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2153 of statically allocating a predefined number, loop
2154 devices can be requested on-demand with the
2155 /dev/loop-control interface.
2157 mce [X86-32] Machine Check Exception
2159 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.txt
2161 md= [HW] RAID subsystems devices and level
2162 See Documentation/md.txt.
2165 Format: <first>,<last>
2166 Specifies range of consoles to be captured by the MDA.
2168 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2169 Amount of memory to be used when the kernel is not able
2170 to see the whole system memory or for test.
2171 [X86] Work as limiting max address. Use together
2172 with memmap= to avoid physical address space collisions.
2173 Without memmap= PCI devices could be placed at addresses
2174 belonging to unused RAM.
2176 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2180 [KNL,SH] Allow user to override the default size for
2181 per-device physically contiguous DMA buffers.
2183 memhp_default_state=online/offline
2184 [KNL] Set the initial state for the memory hotplug
2185 onlining policy. If not specified, the default value is
2186 set according to the
2187 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2189 See Documentation/memory-hotplug.txt.
2191 memmap=exactmap [KNL,X86] Enable setting of an exact
2192 E820 memory map, as specified by the user.
2193 Such memmap=exactmap lines can be constructed based on
2194 BIOS output or other requirements. See the memmap=nn@ss
2197 memmap=nn[KMG]@ss[KMG]
2198 [KNL] Force usage of a specific region of memory.
2199 Region of memory to be used is from ss to ss+nn.
2201 memmap=nn[KMG]#ss[KMG]
2202 [KNL,ACPI] Mark specific memory as ACPI data.
2203 Region of memory to be marked is from ss to ss+nn.
2205 memmap=nn[KMG]$ss[KMG]
2206 [KNL,ACPI] Mark specific memory as reserved.
2207 Region of memory to be reserved is from ss to ss+nn.
2208 Example: Exclude memory from 0x18690000-0x1869ffff
2209 memmap=64K$0x18690000
2211 memmap=0x10000$0x18690000
2213 memmap=nn[KMG]!ss[KMG]
2214 [KNL,X86] Mark specific memory as protected.
2215 Region of memory to be used, from ss to ss+nn.
2216 The memory region may be marked as e820 type 12 (0xc)
2217 and is NVDIMM or ADR memory.
2219 memory_corruption_check=0/1 [X86]
2220 Some BIOSes seem to corrupt the first 64k of
2221 memory when doing things like suspend/resume.
2222 Setting this option will scan the memory
2223 looking for corruption. Enabling this will
2224 both detect corruption and prevent the kernel
2225 from using the memory being corrupted.
2226 However, its intended as a diagnostic tool; if
2227 repeatable BIOS-originated corruption always
2228 affects the same memory, you can use memmap=
2229 to prevent the kernel from using that memory.
2231 memory_corruption_check_size=size [X86]
2232 By default it checks for corruption in the low
2233 64k, making this memory unavailable for normal
2234 use. Use this parameter to scan for
2235 corruption in more or less memory.
2237 memory_corruption_check_period=seconds [X86]
2238 By default it checks for corruption every 60
2239 seconds. Use this parameter to check at some
2240 other rate. 0 disables periodic checking.
2242 memtest= [KNL,X86,ARM] Enable memtest
2244 default : 0 <disable>
2245 Specifies the number of memtest passes to be
2246 performed. Each pass selects another test
2247 pattern from a given set of patterns. Memtest
2248 fills the memory with this pattern, validates
2249 memory contents and reserves bad memory
2250 regions that are detected.
2252 meye.*= [HW] Set MotionEye Camera parameters
2253 See Documentation/video4linux/meye.txt.
2255 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2256 Multi-Function General Purpose Timers on AMD Geode
2259 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2260 the BIOS has incorrectly applied a workaround. TinyBIOS
2261 version 0.98 is known to be affected, 0.99 fixes the
2262 problem by letting the user disable the workaround.
2266 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2267 physical address is ignored.
2269 mini2440= [ARM,HW,KNL]
2270 Format:[0..2][b][c][t]
2272 MINI2440 configuration specification:
2273 0 - The attached screen is the 3.5" TFT
2274 1 - The attached screen is the 7" TFT
2275 2 - The VGA Shield is attached (1024x768)
2276 Leaving out the screen size parameter will not load
2277 the TFT driver, and the framebuffer will be left
2279 b - Enable backlight. The TFT backlight pin will be
2280 linked to the kernel VESA blanking code and a GPIO
2281 LED. This parameter is not necessary when using the
2283 c - Enable the s3c camera interface.
2284 t - Reserved for enabling touchscreen support. The
2285 touchscreen support is not enabled in the mainstream
2286 kernel as of 2.6.30, a preliminary port can be found
2287 in the "bleeding edge" mini2440 support kernel at
2288 http://repo.or.cz/w/linux-2.6/mini2440.git
2291 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2292 parameter allows control of the logging verbosity for
2293 the additional memory initialisation checks. A value
2294 of 0 disables mminit logging and a level of 4 will
2295 log everything. Information is printed at KERN_DEBUG
2296 so loglevel=8 may also need to be specified.
2299 [KNL] When CONFIG_MODULE_SIG is set, this means that
2300 modules without (valid) signatures will fail to load.
2301 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2302 is always true, so this option does nothing.
2305 [MOUSE] Maximum time between finger touching and
2306 leaving touchpad surface for touch to be considered
2307 a tap and be reported as a left button click (for
2308 touchpads working in absolute mode only).
2310 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2311 reporting absolute coordinates, such as tablets
2312 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2313 reporting absolute coordinates, such as tablets
2315 movablecore=nn[KMG] [KNL,X86,IA-64,PPC] This parameter
2316 is similar to kernelcore except it specifies the
2317 amount of memory used for migratable allocations.
2318 If both kernelcore and movablecore is specified,
2319 then kernelcore will be at *least* the specified
2320 value but may be more. If movablecore on its own
2321 is specified, the administrator must be careful
2322 that the amount of memory usable for all allocations
2325 movable_node [KNL,X86] Boot-time switch to enable the effects
2326 of CONFIG_MOVABLE_NODE=y. See mm/Kconfig for details.
2328 MTD_Partition= [MTD]
2329 Format: <name>,<region-number>,<size>,<offset>
2331 MTD_Region= [MTD] Format:
2332 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2335 See drivers/mtd/cmdlinepart.c.
2337 multitce=off [PPC] This parameter disables the use of the pSeries
2338 firmware feature for updating multiple TCE entries
2341 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2343 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2345 boundary - index of last SLC block on Flex-OneNAND.
2346 The remaining blocks are configured as MLC blocks.
2347 lock - Configure if Flex-OneNAND boundary should be locked.
2348 Once locked, the boundary cannot be changed.
2349 1 indicates lock status, 0 indicates unlock status.
2352 ARM/S3C2412 JIVE boot control
2354 See arch/arm/mach-s3c2412/mach-jive.c
2356 mtouchusb.raw_coordinates=
2357 [HW] Make the MicroTouch USB driver use raw coordinates
2358 ('y', default) or cooked coordinates ('n')
2360 mtrr_chunk_size=nn[KMG] [X86]
2361 used for mtrr cleanup. It is largest continuous chunk
2362 that could hold holes aka. UC entries.
2364 mtrr_gran_size=nn[KMG] [X86]
2365 Used for mtrr cleanup. It is granularity of mtrr block.
2367 Large value could prevent small alignment from
2370 mtrr_spare_reg_nr=n [X86]
2372 Range: 0,7 : spare reg number
2374 Used for mtrr cleanup. It is spare mtrr entries number.
2375 Set to 2 or more if your graphical card needs more.
2377 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2379 netdev= [NET] Network devices parameters
2380 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2381 Note that mem_start is often overloaded to mean
2382 something different and driver-specific.
2383 This usage is only documented in each driver source
2387 [NETFILTER] Enable connection tracking flow accounting
2388 0 to disable accounting
2389 1 to enable accounting
2392 nfsaddrs= [NFS] Deprecated. Use ip= instead.
2393 See Documentation/filesystems/nfs/nfsroot.txt.
2395 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
2396 See Documentation/filesystems/nfs/nfsroot.txt.
2398 nfsrootdebug [NFS] enable nfsroot debugging messages.
2399 See Documentation/filesystems/nfs/nfsroot.txt.
2401 nfs.callback_tcpport=
2402 [NFS] set the TCP port on which the NFSv4 callback
2403 channel should listen.
2406 [NFS] sets the pathname to the program which is used
2407 to update the NFS client cache entries.
2409 nfs.cache_getent_timeout=
2410 [NFS] sets the timeout after which an attempt to
2411 update a cache entry is deemed to have failed.
2413 nfs.idmap_cache_timeout=
2414 [NFS] set the maximum lifetime for idmapper cache
2418 [NFS] enable 64-bit inode numbers.
2419 If zero, the NFS client will fake up a 32-bit inode
2420 number for the readdir() and stat() syscalls instead
2421 of returning the full 64-bit number.
2422 The default is to return 64-bit inode numbers.
2424 nfs.max_session_slots=
2425 [NFSv4.1] Sets the maximum number of session slots
2426 the client will attempt to negotiate with the server.
2427 This limits the number of simultaneous RPC requests
2428 that the client can send to the NFSv4.1 server.
2429 Note that there is little point in setting this
2430 value higher than the max_tcp_slot_table_limit.
2432 nfs.nfs4_disable_idmapping=
2433 [NFSv4] When set to the default of '1', this option
2434 ensures that both the RPC level authentication
2435 scheme and the NFS level operations agree to use
2436 numeric uids/gids if the mount is using the
2437 'sec=sys' security flavour. In effect it is
2438 disabling idmapping, which can make migration from
2439 legacy NFSv2/v3 systems to NFSv4 easier.
2440 Servers that do not support this mode of operation
2441 will be autodetected by the client, and it will fall
2442 back to using the idmapper.
2443 To turn off this behaviour, set the value to '0'.
2445 [NFS4] Specify an additional fixed unique ident-
2446 ification string that NFSv4 clients can insert into
2447 their nfs_client_id4 string. This is typically a
2448 UUID that is generated at system install time.
2450 nfs.send_implementation_id =
2451 [NFSv4.1] Send client implementation identification
2452 information in exchange_id requests.
2453 If zero, no implementation identification information
2455 The default is to send the implementation identification
2458 nfs.recover_lost_locks =
2459 [NFSv4] Attempt to recover locks that were lost due
2460 to a lease timeout on the server. Please note that
2461 doing this risks data corruption, since there are
2462 no guarantees that the file will remain unchanged
2463 after the locks are lost.
2464 If you want to enable the kernel legacy behaviour of
2465 attempting to recover these locks, then set this
2467 The default parameter value of '0' causes the kernel
2468 not to attempt recovery of lost locks.
2470 nfs4.layoutstats_timer =
2471 [NFSv4.2] Change the rate at which the kernel sends
2472 layoutstats to the pNFS metadata server.
2474 Setting this to value to 0 causes the kernel to use
2475 whatever value is the default set by the layout
2476 driver. A non-zero value sets the minimum interval
2477 in seconds between layoutstats transmissions.
2479 nfsd.nfs4_disable_idmapping=
2480 [NFSv4] When set to the default of '1', the NFSv4
2481 server will return only numeric uids and gids to
2482 clients using auth_sys, and will accept numeric uids
2483 and gids from such clients. This is intended to ease
2484 migration from NFSv2/v3.
2486 objlayoutdriver.osd_login_prog=
2487 [NFS] [OBJLAYOUT] sets the pathname to the program which
2488 is used to automatically discover and login into new
2489 osd-targets. Please see:
2490 Documentation/filesystems/pnfs.txt for more explanations
2492 nmi_debug= [KNL,AVR32,SH] Specify one or more actions to take
2493 when a NMI is triggered.
2494 Format: [state][,regs][,debounce][,die]
2496 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
2497 Format: [panic,][nopanic,][num]
2499 0 - turn hardlockup detector in nmi_watchdog off
2500 1 - turn hardlockup detector in nmi_watchdog on
2501 When panic is specified, panic when an NMI watchdog
2502 timeout occurs (or 'nopanic' to override the opposite
2503 default). To disable both hard and soft lockup detectors,
2504 please see 'nowatchdog'.
2505 This is useful when you use a panic=... timeout and
2506 need the box quickly up again.
2508 netpoll.carrier_timeout=
2509 [NET] Specifies amount of time (in seconds) that
2510 netpoll should wait for a carrier. By default netpoll
2513 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
2514 emulation library even if a 387 maths coprocessor
2518 [HW] Never suspend the console
2519 Disable suspending of consoles during suspend and
2520 hibernate operations. Once disabled, debugging
2521 messages can reach various consoles while the rest
2522 of the system is being put to sleep (ie, while
2523 debugging driver suspend/resume hooks). This may
2524 not work reliably with all consoles, but is known
2525 to work with serial and VGA consoles.
2526 To facilitate more flexible debugging, we also add
2527 console_suspend, a printk module parameter to control
2528 it. Users could use console_suspend (usually
2529 /sys/module/printk/parameters/console_suspend) to
2530 turn on/off it dynamically.
2532 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
2533 caches in the slab allocator. Saves per-node memory,
2534 but will impact performance.
2538 noapic [SMP,APIC] Tells the kernel to not make use of any
2539 IOAPICs that may be present in the system.
2541 noautogroup Disable scheduler automatic task group creation.
2543 nobats [PPC] Do not use BATs for mapping kernel lowmem
2544 on "Classic" PPC cores.
2548 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
2550 nodelayacct [KNL] Disable per-task delay accounting
2552 nodisconnect [HW,SCSI,M68K] Disables SCSI disconnects.
2554 nodsp [SH] Disable hardware DSP at boot time.
2556 noefi Disable EFI runtime services support.
2561 On X86-32 available only on PAE configured kernels.
2562 noexec=on: enable non-executable mappings (default)
2563 noexec=off: disable non-executable mappings
2566 Disable SMAP (Supervisor Mode Access Prevention)
2567 even if it is supported by processor.
2570 Disable SMEP (Supervisor Mode Execution Prevention)
2571 even if it is supported by processor.
2574 This affects only 32-bit executables.
2575 noexec32=on: enable non-executable mappings (default)
2576 read doesn't imply executable mappings
2577 noexec32=off: disable non-executable mappings
2578 read implies executable mappings
2580 nofpu [MIPS,SH] Disable hardware FPU at boot time.
2582 nofxsr [BUGS=X86-32] Disables x86 floating point extended
2583 register save and restore. The kernel will only save
2584 legacy floating-point registers on task switch.
2586 nohugeiomap [KNL,x86] Disable kernel huge I/O mappings.
2588 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
2589 Equivalent to smt=1.
2591 noxsave [BUGS=X86] Disables x86 extended register state save
2592 and restore using xsave. The kernel will fallback to
2593 enabling legacy floating-point and sse state.
2595 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
2596 register states. The kernel will fall back to use
2597 xsave to save the states. By using this parameter,
2598 performance of saving the states is degraded because
2599 xsave doesn't support modified optimization while
2600 xsaveopt supports it on xsaveopt enabled systems.
2602 noxsaves [X86] Disables xsaves and xrstors used in saving and
2603 restoring x86 extended register state in compacted
2604 form of xsave area. The kernel will fall back to use
2605 xsaveopt and xrstor to save and restore the states
2606 in standard form of xsave area. By using this
2607 parameter, xsave area per process might occupy more
2608 memory on xsaves enabled systems.
2610 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
2611 wfi(ARM) instruction doesn't work correctly and not to
2612 use it. This is also useful when using JTAG debugger.
2614 no_file_caps Tells the kernel not to honor file capabilities. The
2615 only way then for a file to be executed with privilege
2616 is to be setuid root or executed by root.
2618 nohalt [IA-64] Tells the kernel not to use the power saving
2619 function PAL_HALT_LIGHT when idle. This increases
2620 power-consumption. On the positive side, it reduces
2621 interrupt wake-up latency, which may improve performance
2622 in certain environments such as networked servers or
2625 nohibernate [HIBERNATION] Disable hibernation and resume.
2627 nohz= [KNL] Boottime enable/disable dynamic ticks
2628 Valid arguments: on, off
2631 nohz_full= [KNL,BOOT]
2632 In kernels built with CONFIG_NO_HZ_FULL=y, set
2633 the specified list of CPUs whose tick will be stopped
2634 whenever possible. The boot CPU will be forced outside
2635 the range to maintain the timekeeping.
2636 The CPUs in this range must also be included in the
2639 noiotrap [SH] Disables trapped I/O port accesses.
2641 noirqdebug [X86-32] Disables the code which attempts to detect and
2642 disable unhandled interrupt sources.
2644 no_timer_check [X86,APIC] Disables the code which tests for
2645 broken timer IRQ sources.
2647 noisapnp [ISAPNP] Disables ISA PnP code.
2649 noinitrd [RAM] Tells the kernel not to load any configured
2652 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
2654 [Deprecated - use intremap=off]
2658 noinvpcid [X86] Disable the INVPCID cpu feature.
2660 nojitter [IA-64] Disables jitter checking for ITC timers.
2662 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
2664 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
2667 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting.
2668 steal time is computed, but won't influence scheduler
2671 nolapic [X86-32,APIC] Do not enable or use the local APIC.
2673 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
2675 noltlbs [PPC] Do not use large page/tlb entries for kernel
2676 lowmem mapping on PPC40x and PPC8xx
2678 nomca [IA-64] Disable machine check abort handling
2680 nomce [X86-32] Disable Machine Check Exception
2682 nomfgpt [X86-32] Disable Multi-Function General Purpose
2683 Timer usage (for AMD Geode machines).
2685 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
2686 shutdown the other cpus. Instead use the REBOOT_VECTOR
2689 nomodule Disable module load
2691 nopat [X86] Disable PAT (page attribute table extension of
2692 pagetables) support.
2694 norandmaps Don't use address space randomization. Equivalent to
2695 echo 0 > /proc/sys/kernel/randomize_va_space
2697 noreplace-paravirt [X86,IA-64,PV_OPS] Don't patch paravirt_ops
2699 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
2700 with UP alternatives
2702 nordrand [X86] Disable kernel use of the RDRAND and
2703 RDSEED instructions even if they are supported
2704 by the processor. RDRAND and RDSEED are still
2705 available to user space applications.
2707 noresume [SWSUSP] Disables resume and restores original swap
2710 no-scroll [VGA] Disables scrollback.
2711 This is required for the Braillex ib80-piezo Braille
2712 reader made by F.H. Papenmeier (Germany).
2716 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
2718 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
2719 and disable the IO APIC. legacy for "maxcpus=0".
2721 nosoftlockup [KNL] Disable the soft-lockup detector.
2723 nosync [HW,M68K] Disables sync negotiation for all devices.
2725 notsc [BUGS=X86-32] Disable Time Stamp Counter
2727 nowatchdog [KNL] Disable both lockup detectors, i.e.
2728 soft-lockup and NMI watchdog (hard-lockup).
2732 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
2734 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
2735 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
2736 Some features depend on CPU0. Known dependencies are:
2737 1. Resume from suspend/hibernate depends on CPU0.
2738 Suspend/hibernate will fail if CPU0 is offline and you
2739 need to online CPU0 before suspend/hibernate.
2740 2. PIC interrupts also depend on CPU0. CPU0 can't be
2741 removed if a PIC interrupt is detected.
2742 It's said poweroff/reboot may depend on CPU0 on some
2743 machines although I haven't seen such issues so far
2744 after CPU0 is offline on a few tested machines.
2745 If the dependencies are under your control, you can
2746 turn on cpu0_hotplug.
2748 nptcg= [IA-64] Override max number of concurrent global TLB
2749 purges which is reported from either PAL_VM_SUMMARY or
2752 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
2753 could support. nr_cpus=n : n >= 1 limits the kernel to
2754 supporting 'n' processors. Later in runtime you can not
2755 use hotplug cpu feature to put more cpu back to online.
2756 just like you compile the kernel NR_CPUS=n
2758 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
2760 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
2761 Allowed values are enable and disable
2763 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
2764 one of ['zone', 'node', 'default'] can be specified
2765 This can be set from sysctl after boot.
2766 See Documentation/sysctl/vm.txt for details.
2768 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
2769 See Documentation/debugging-via-ohci1394.txt for more
2772 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
2773 Rather than timing out after 20 ms if an EC
2774 command is not properly ACKed, override the length
2775 of the timeout. We have interrupts disabled while
2776 waiting for the ACK, so if this is set too high
2777 interrupts *may* be lost!
2779 omap_mux= [OMAP] Override bootloader pin multiplexing.
2780 Format: <mux_mode0.mode_name=value>...
2781 For example, to override I2C bus2:
2782 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
2784 oprofile.timer= [HW]
2785 Use timer interrupt instead of performance counters
2787 oprofile.cpu_type= Force an oprofile cpu type
2788 This might be useful if you have an older oprofile
2789 userland or if you want common events.
2790 Format: { arch_perfmon }
2791 arch_perfmon: [X86] Force use of architectural
2792 perfmon on Intel CPUs instead of the
2793 CPU specific event set.
2794 timer: [X86] Force use of architectural NMI
2795 timer mode (see also oprofile.timer
2796 for generic hr timer mode)
2798 oops=panic Always panic on oopses. Default is to just kill the
2799 process, but there is a small probability of
2800 deadlocking the machine.
2801 This will also cause panics on machine check exceptions.
2802 Useful together with panic=30 to trigger a reboot.
2805 See Documentation/sound/oss/oss-parameters.txt
2807 page_owner= [KNL] Boot-time page_owner enabling option.
2808 Storage of the information about who allocated
2809 each page is disabled in default. With this switch,
2811 on: enable the feature
2813 page_poison= [KNL] Boot-time parameter changing the state of
2814 poisoning on the buddy allocator.
2815 off: turn off poisoning
2816 on: turn on poisoning
2818 panic= [KNL] Kernel behaviour on panic: delay <timeout>
2819 timeout > 0: seconds before rebooting
2820 timeout = 0: wait forever
2821 timeout < 0: reboot immediately
2824 panic_on_warn panic() instead of WARN(). Useful to cause kdump
2827 crash_kexec_post_notifiers
2828 Run kdump after running panic-notifiers and dumping
2829 kmsg. This only for the users who doubt kdump always
2830 succeeds in any situation.
2831 Note that this also increases risks of kdump failure,
2832 because some panic notifiers can make the crashed
2833 kernel more unstable.
2835 parkbd.port= [HW] Parallel port number the keyboard adapter is
2836 connected to, default is 0.
2838 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
2839 0 for XT, 1 for AT (default is AT).
2842 parport= [HW,PPT] Specify parallel ports. 0 disables.
2843 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
2844 Use 'auto' to force the driver to use any
2845 IRQ/DMA settings detected (the default is to
2846 ignore detected IRQ/DMA settings because of
2847 possible conflicts). You can specify the base
2848 address, IRQ, and DMA settings; IRQ and DMA
2849 should be numbers, or 'auto' (for using detected
2850 settings on that particular port), or 'nofifo'
2851 (to avoid using a FIFO even if it is detected).
2852 Parallel ports are assigned in the order they
2853 are specified on the command line, starting
2856 parport_init_mode= [HW,PPT]
2857 Configure VIA parallel port to operate in
2858 a specific mode. This is necessary on Pegasos
2859 computer where firmware has no options for setting
2860 up parallel port mode and sets it to spp.
2861 Currently this function knows 686a and 8231 chips.
2862 Format: [spp|ps2|epp|ecp|ecpepp]
2865 Halt all CPUs after the first oops has been printed for
2866 the specified number of seconds. This is to be used if
2867 your oopses keep scrolling off the screen.
2872 See header of drivers/block/paride/pcd.c.
2873 See also Documentation/blockdev/paride.txt.
2875 pci=option[,option...] [PCI] various PCI subsystem options:
2876 earlydump [X86] dump PCI config space before the kernel
2878 off [X86] don't probe for the PCI bus
2879 bios [X86-32] force use of PCI BIOS, don't access
2880 the hardware directly. Use this if your machine
2881 has a non-standard PCI host bridge.
2882 nobios [X86-32] disallow use of PCI BIOS, only direct
2883 hardware access methods are allowed. Use this
2884 if you experience crashes upon bootup and you
2885 suspect they are caused by the BIOS.
2886 conf1 [X86] Force use of PCI Configuration Access
2887 Mechanism 1 (config address in IO port 0xCF8,
2888 data in IO port 0xCFC, both 32-bit).
2889 conf2 [X86] Force use of PCI Configuration Access
2890 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
2891 the function, IO port 0xCFA, also 8-bit, sets
2892 bus number. The config space is then accessed
2893 through ports 0xC000-0xCFFF).
2894 See http://wiki.osdev.org/PCI for more info
2895 on the configuration access mechanisms.
2896 noaer [PCIE] If the PCIEAER kernel config parameter is
2897 enabled, this kernel boot option can be used to
2898 disable the use of PCIE advanced error reporting.
2899 nodomains [PCI] Disable support for multiple PCI
2900 root domains (aka PCI segments, in ACPI-speak).
2901 nommconf [X86] Disable use of MMCONFIG for PCI
2903 check_enable_amd_mmconf [X86] check for and enable
2904 properly configured MMIO access to PCI
2905 config space on AMD family 10h CPU
2906 nomsi [MSI] If the PCI_MSI kernel config parameter is
2907 enabled, this kernel boot option can be used to
2908 disable the use of MSI interrupts system-wide.
2909 noioapicquirk [APIC] Disable all boot interrupt quirks.
2910 Safety option to keep boot IRQs enabled. This
2911 should never be necessary.
2912 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
2913 primary IO-APIC for bridges that cannot disable
2914 boot IRQs. This fixes a source of spurious IRQs
2915 when the system masks IRQs.
2916 noioapicreroute [APIC] Disable workaround that uses the
2917 boot IRQ equivalent of an IRQ that connects to
2918 a chipset where boot IRQs cannot be disabled.
2919 The opposite of ioapicreroute.
2920 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
2921 routing table. These calls are known to be buggy
2922 on several machines and they hang the machine
2923 when used, but on other computers it's the only
2924 way to get the interrupt routing table. Try
2925 this option if the kernel is unable to allocate
2926 IRQs or discover secondary PCI buses on your
2928 rom [X86] Assign address space to expansion ROMs.
2929 Use with caution as certain devices share
2930 address decoders between ROMs and other
2932 norom [X86] Do not assign address space to
2933 expansion ROMs that do not already have
2934 BIOS assigned address ranges.
2935 nobar [X86] Do not assign address space to the
2936 BARs that weren't assigned by the BIOS.
2937 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
2938 assigned automatically to PCI devices. You can
2939 make the kernel exclude IRQs of your ISA cards
2941 pirqaddr=0xAAAAA [X86] Specify the physical address
2942 of the PIRQ table (normally generated
2943 by the BIOS) if it is outside the
2944 F0000h-100000h range.
2945 lastbus=N [X86] Scan all buses thru bus #N. Can be
2946 useful if the kernel is unable to find your
2947 secondary buses and you want to tell it
2948 explicitly which ones they are.
2949 assign-busses [X86] Always assign all PCI bus
2950 numbers ourselves, overriding
2951 whatever the firmware may have done.
2952 usepirqmask [X86] Honor the possible IRQ mask stored
2953 in the BIOS $PIR table. This is needed on
2954 some systems with broken BIOSes, notably
2955 some HP Pavilion N5400 and Omnibook XE3
2956 notebooks. This will have no effect if ACPI
2957 IRQ routing is enabled.
2958 noacpi [X86] Do not use ACPI for IRQ routing
2959 or for PCI scanning.
2960 use_crs [X86] Use PCI host bridge window information
2961 from ACPI. On BIOSes from 2008 or later, this
2962 is enabled by default. If you need to use this,
2963 please report a bug.
2964 nocrs [X86] Ignore PCI host bridge windows from ACPI.
2965 If you need to use this, please report a bug.
2966 routeirq Do IRQ routing for all PCI devices.
2967 This is normally done in pci_enable_device(),
2968 so this option is a temporary workaround
2969 for broken drivers that don't call it.
2970 skip_isa_align [X86] do not align io start addr, so can
2971 handle more pci cards
2972 noearly [X86] Don't do any early type 1 scanning.
2973 This might help on some broken boards which
2974 machine check when some devices' config space
2975 is read. But various workarounds are disabled
2976 and some IOMMU drivers will not work.
2977 bfsort Sort PCI devices into breadth-first order.
2978 This sorting is done to get a device
2979 order compatible with older (<= 2.4) kernels.
2980 nobfsort Don't sort PCI devices into breadth-first order.
2981 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
2982 tuning and use the BIOS-configured MPS defaults.
2983 pcie_bus_safe Set every device's MPS to the largest value
2984 supported by all devices below the root complex.
2985 pcie_bus_perf Set device MPS to the largest allowable MPS
2986 based on its parent bus. Also set MRRS (Max
2987 Read Request Size) to the largest supported
2988 value (no larger than the MPS that the device
2989 or bus can support) for best performance.
2990 pcie_bus_peer2peer Set every device's MPS to 128B, which
2991 every device is guaranteed to support. This
2992 configuration allows peer-to-peer DMA between
2993 any pair of devices, possibly at the cost of
2994 reduced performance. This also guarantees
2995 that hot-added devices will work.
2996 cbiosize=nn[KMG] The fixed amount of bus space which is
2997 reserved for the CardBus bridge's IO window.
2998 The default value is 256 bytes.
2999 cbmemsize=nn[KMG] The fixed amount of bus space which is
3000 reserved for the CardBus bridge's memory
3001 window. The default value is 64 megabytes.
3004 [<order of align>@][<domain>:]<bus>:<slot>.<func>[; ...]
3005 Specifies alignment and device to reassign
3006 aligned memory resources.
3007 If <order of align> is not specified,
3008 PAGE_SIZE is used as alignment.
3009 PCI-PCI bridge can be specified, if resource
3010 windows need to be expanded.
3011 ecrc= Enable/disable PCIe ECRC (transaction layer
3012 end-to-end CRC checking).
3013 bios: Use BIOS/firmware settings. This is the
3017 hpiosize=nn[KMG] The fixed amount of bus space which is
3018 reserved for hotplug bridge's IO window.
3019 Default size is 256 bytes.
3020 hpmemsize=nn[KMG] The fixed amount of bus space which is
3021 reserved for hotplug bridge's memory window.
3022 Default size is 2 megabytes.
3023 realloc= Enable/disable reallocating PCI bridge resources
3024 if allocations done by BIOS are too small to
3025 accommodate resources required by all child
3027 off: Turn realloc off
3029 realloc same as realloc=on
3030 noari do not use PCIe ARI.
3031 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3032 only look for one device below a PCIe downstream
3035 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3038 force Enable ASPM even on devices that claim not to support it.
3039 WARNING: Forcing ASPM on may cause system lockups.
3041 pcie_hp= [PCIE] PCI Express Hotplug driver options:
3042 nomsi Do not use MSI for PCI Express Native Hotplug (this
3043 makes all PCIe ports use INTx for hotplug services).
3045 pcie_ports= [PCIE] PCIe ports handling:
3046 auto Ask the BIOS whether or not to use native PCIe services
3047 associated with PCIe ports (PME, hot-plug, AER). Use
3048 them only if that is allowed by the BIOS.
3049 native Use native PCIe services associated with PCIe ports
3051 compat Treat PCIe ports as PCI-to-PCI bridges, disable the PCIe
3054 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3055 nomsi Do not use MSI for native PCIe PME signaling (this makes
3056 all PCIe root ports use INTx for all services).
3058 pcmv= [HW,PCMCIA] BadgePAD 4
3062 Keep all power-domains already enabled by bootloader on,
3063 even if no driver has claimed them. This is useful
3064 for debug and development, but should not be
3065 needed on a platform with proper driver support.
3068 See Documentation/blockdev/paride.txt.
3070 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3073 See arch/parisc/kernel/pdc_chassis.c
3075 percpu_alloc= Select which percpu first chunk allocator to use.
3076 Currently supported values are "embed" and "page".
3077 Archs may support subset or none of the selections.
3078 See comments in mm/percpu.c for details on each
3079 allocator. This parameter is primarily for debugging
3080 and performance comparison.
3083 See Documentation/blockdev/paride.txt.
3086 See Documentation/blockdev/paride.txt.
3088 pirq= [SMP,APIC] Manual mp-table setup
3089 See Documentation/x86/i386/IO-APIC.txt.
3091 plip= [PPT,NET] Parallel port network link
3092 Format: { parport<nr> | timid | 0 }
3093 See also Documentation/parport.txt.
3095 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3096 Override pmtimer IOPort with a hex value.
3100 Enable PNP debug messages (depends on the
3101 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3102 via /sys/module/pnp/parameters/debug. We always show
3103 current resource usage; turning this on also shows
3104 possible settings and some assignment information.
3110 { on | off | curr | res | no-curr | no-res }
3113 [ISAPNP] Exclude IRQs for the autoconfiguration
3116 [ISAPNP] Exclude DMAs for the autoconfiguration
3118 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3119 Ranges are in pairs (I/O port base and size).
3122 [ISAPNP] Exclude memory regions for the
3124 Ranges are in pairs (memory base and size).
3126 ports= [IP_VS_FTP] IPVS ftp helper module
3128 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3130 Format: <port>,<port>....
3132 ppc_strict_facility_enable
3133 [PPC] This option catches any kernel floating point,
3134 Altivec, VSX and SPE outside of regions specifically
3135 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3136 There is some performance impact when enabling this.
3138 print-fatal-signals=
3139 [KNL] debug: print fatal signals
3141 If enabled, warn about various signal handling
3142 related application anomalies: too many signals,
3143 too many POSIX.1 timers, fatal signals causing a
3146 If you hit the warning due to signal overflow,
3147 you might want to try "ulimit -i unlimited".
3151 printk.always_kmsg_dump=
3152 Trigger kmsg_dump for cases other than kernel oops or
3154 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3157 printk.time= Show timing data prefixed to each printk message line
3158 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3160 processor.max_cstate= [HW,ACPI]
3161 Limit processor to maximum C-state
3162 max_cstate=9 overrides any DMI blacklist limit.
3164 processor.nocst [HW,ACPI]
3165 Ignore the _CST method to determine C-states,
3166 instead using the legacy FADT method
3168 profile= [KNL] Enable kernel profiling via /proc/profile
3169 Format: [schedule,]<number>
3170 Param: "schedule" - profile schedule points.
3171 Param: <number> - step/bucket size as a power of 2 for
3172 statistical time based profiling.
3173 Param: "sleep" - profile D-state sleeping (millisecs).
3174 Requires CONFIG_SCHEDSTATS
3175 Param: "kvm" - profile VM exits.
3177 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3179 See Documentation/blockdev/ramdisk.txt.
3181 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3182 probe for; one of (bare|imps|exps|lifebook|any).
3183 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3185 psmouse.resetafter= [HW,MOUSE]
3186 Try to reset the device after so many bad packets
3189 [HW,MOUSE] Set desired mouse resolution, in dpi.
3190 psmouse.smartscroll=
3191 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
3192 0 = disabled, 1 = enabled (default).
3194 pstore.backend= Specify the name of the pstore backend to use
3197 See Documentation/blockdev/paride.txt.
3200 [KNL] Number of legacy pty's. Overwrites compiled-in
3203 quiet [KNL] Disable most log messages
3208 See Documentation/md.txt.
3210 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
3211 See Documentation/blockdev/ramdisk.txt.
3214 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
3215 the specified list of CPUs to be no-callback CPUs.
3216 Invocation of these CPUs' RCU callbacks will
3217 be offloaded to "rcuox/N" kthreads created for
3218 that purpose, where "x" is "b" for RCU-bh, "p"
3219 for RCU-preempt, and "s" for RCU-sched, and "N"
3220 is the CPU number. This reduces OS jitter on the
3221 offloaded CPUs, which can be useful for HPC and
3222 real-time workloads. It can also improve energy
3223 efficiency for asymmetric multiprocessors.
3226 Rather than requiring that offloaded CPUs
3227 (specified by rcu_nocbs= above) explicitly
3228 awaken the corresponding "rcuoN" kthreads,
3229 make these kthreads poll for callbacks.
3230 This improves the real-time response for the
3231 offloaded CPUs by relieving them of the need to
3232 wake up the corresponding kthread, but degrades
3233 energy efficiency by requiring that the kthreads
3234 periodically wake up to do the polling.
3236 rcutree.blimit= [KNL]
3237 Set maximum number of finished RCU callbacks to
3238 process in one batch.
3240 rcutree.dump_tree= [KNL]
3241 Dump the structure of the rcu_node combining tree
3242 out at early boot. This is used for diagnostic
3243 purposes, to verify correct tree setup.
3245 rcutree.gp_cleanup_delay= [KNL]
3246 Set the number of jiffies to delay each step of
3247 RCU grace-period cleanup. This only has effect
3248 when CONFIG_RCU_TORTURE_TEST_SLOW_CLEANUP is set.
3250 rcutree.gp_init_delay= [KNL]
3251 Set the number of jiffies to delay each step of
3252 RCU grace-period initialization. This only has
3253 effect when CONFIG_RCU_TORTURE_TEST_SLOW_INIT
3256 rcutree.gp_preinit_delay= [KNL]
3257 Set the number of jiffies to delay each step of
3258 RCU grace-period pre-initialization, that is,
3259 the propagation of recent CPU-hotplug changes up
3260 the rcu_node combining tree. This only has effect
3261 when CONFIG_RCU_TORTURE_TEST_SLOW_PREINIT is set.
3263 rcutree.rcu_fanout_exact= [KNL]
3264 Disable autobalancing of the rcu_node combining
3265 tree. This is used by rcutorture, and might
3266 possibly be useful for architectures having high
3267 cache-to-cache transfer latencies.
3269 rcutree.rcu_fanout_leaf= [KNL]
3270 Change the number of CPUs assigned to each
3271 leaf rcu_node structure. Useful for very
3272 large systems, which will choose the value 64,
3273 and for NUMA systems with large remote-access
3274 latencies, which will choose a value aligned
3275 with the appropriate hardware boundaries.
3277 rcutree.jiffies_till_sched_qs= [KNL]
3278 Set required age in jiffies for a
3279 given grace period before RCU starts
3280 soliciting quiescent-state help from
3281 rcu_note_context_switch().
3283 rcutree.jiffies_till_first_fqs= [KNL]
3284 Set delay from grace-period initialization to
3285 first attempt to force quiescent states.
3286 Units are jiffies, minimum value is zero,
3287 and maximum value is HZ.
3289 rcutree.jiffies_till_next_fqs= [KNL]
3290 Set delay between subsequent attempts to force
3291 quiescent states. Units are jiffies, minimum
3292 value is one, and maximum value is HZ.
3294 rcutree.kthread_prio= [KNL,BOOT]
3295 Set the SCHED_FIFO priority of the RCU per-CPU
3296 kthreads (rcuc/N). This value is also used for
3297 the priority of the RCU boost threads (rcub/N)
3298 and for the RCU grace-period kthreads (rcu_bh,
3299 rcu_preempt, and rcu_sched). If RCU_BOOST is
3300 set, valid values are 1-99 and the default is 1
3301 (the least-favored priority). Otherwise, when
3302 RCU_BOOST is not set, valid values are 0-99 and
3303 the default is zero (non-realtime operation).
3305 rcutree.rcu_nocb_leader_stride= [KNL]
3306 Set the number of NOCB kthread groups, which
3307 defaults to the square root of the number of
3308 CPUs. Larger numbers reduces the wakeup overhead
3309 on the per-CPU grace-period kthreads, but increases
3310 that same overhead on each group's leader.
3312 rcutree.qhimark= [KNL]
3313 Set threshold of queued RCU callbacks beyond which
3314 batch limiting is disabled.
3316 rcutree.qlowmark= [KNL]
3317 Set threshold of queued RCU callbacks below which
3318 batch limiting is re-enabled.
3320 rcutree.rcu_idle_gp_delay= [KNL]
3321 Set wakeup interval for idle CPUs that have
3322 RCU callbacks (RCU_FAST_NO_HZ=y).
3324 rcutree.rcu_idle_lazy_gp_delay= [KNL]
3325 Set wakeup interval for idle CPUs that have
3326 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
3327 Lazy RCU callbacks are those which RCU can
3328 prove do nothing more than free memory.
3330 rcuperf.gp_exp= [KNL]
3331 Measure performance of expedited synchronous
3332 grace-period primitives.
3334 rcuperf.holdoff= [KNL]
3335 Set test-start holdoff period. The purpose of
3336 this parameter is to delay the start of the
3337 test until boot completes in order to avoid
3340 rcuperf.nreaders= [KNL]
3341 Set number of RCU readers. The value -1 selects
3342 N, where N is the number of CPUs. A value
3343 "n" less than -1 selects N-n+1, where N is again
3344 the number of CPUs. For example, -2 selects N
3345 (the number of CPUs), -3 selects N+1, and so on.
3346 A value of "n" less than or equal to -N selects
3349 rcuperf.nwriters= [KNL]
3350 Set number of RCU writers. The values operate
3351 the same as for rcuperf.nreaders.
3352 N, where N is the number of CPUs
3354 rcuperf.perf_runnable= [BOOT]
3355 Start rcuperf running at boot time.
3357 rcuperf.shutdown= [KNL]
3358 Shut the system down after performance tests
3359 complete. This is useful for hands-off automated
3362 rcuperf.perf_type= [KNL]
3363 Specify the RCU implementation to test.
3365 rcuperf.verbose= [KNL]
3366 Enable additional printk() statements.
3368 rcutorture.cbflood_inter_holdoff= [KNL]
3369 Set holdoff time (jiffies) between successive
3370 callback-flood tests.
3372 rcutorture.cbflood_intra_holdoff= [KNL]
3373 Set holdoff time (jiffies) between successive
3374 bursts of callbacks within a given callback-flood
3377 rcutorture.cbflood_n_burst= [KNL]
3378 Set the number of bursts making up a given
3379 callback-flood test. Set this to zero to
3380 disable callback-flood testing.
3382 rcutorture.cbflood_n_per_burst= [KNL]
3383 Set the number of callbacks to be registered
3384 in a given burst of a callback-flood test.
3386 rcutorture.fqs_duration= [KNL]
3387 Set duration of force_quiescent_state bursts
3390 rcutorture.fqs_holdoff= [KNL]
3391 Set holdoff time within force_quiescent_state bursts
3394 rcutorture.fqs_stutter= [KNL]
3395 Set wait time between force_quiescent_state bursts
3398 rcutorture.gp_cond= [KNL]
3399 Use conditional/asynchronous update-side
3400 primitives, if available.
3402 rcutorture.gp_exp= [KNL]
3403 Use expedited update-side primitives, if available.
3405 rcutorture.gp_normal= [KNL]
3406 Use normal (non-expedited) asynchronous
3407 update-side primitives, if available.
3409 rcutorture.gp_sync= [KNL]
3410 Use normal (non-expedited) synchronous
3411 update-side primitives, if available. If all
3412 of rcutorture.gp_cond=, rcutorture.gp_exp=,
3413 rcutorture.gp_normal=, and rcutorture.gp_sync=
3414 are zero, rcutorture acts as if is interpreted
3415 they are all non-zero.
3417 rcutorture.n_barrier_cbs= [KNL]
3418 Set callbacks/threads for rcu_barrier() testing.
3420 rcutorture.nfakewriters= [KNL]
3421 Set number of concurrent RCU writers. These just
3422 stress RCU, they don't participate in the actual
3423 test, hence the "fake".
3425 rcutorture.nreaders= [KNL]
3426 Set number of RCU readers. The value -1 selects
3427 N-1, where N is the number of CPUs. A value
3428 "n" less than -1 selects N-n-2, where N is again
3429 the number of CPUs. For example, -2 selects N
3430 (the number of CPUs), -3 selects N+1, and so on.
3432 rcutorture.object_debug= [KNL]
3433 Enable debug-object double-call_rcu() testing.
3435 rcutorture.onoff_holdoff= [KNL]
3436 Set time (s) after boot for CPU-hotplug testing.
3438 rcutorture.onoff_interval= [KNL]
3439 Set time (s) between CPU-hotplug operations, or
3440 zero to disable CPU-hotplug testing.
3442 rcutorture.shuffle_interval= [KNL]
3443 Set task-shuffle interval (s). Shuffling tasks
3444 allows some CPUs to go into dyntick-idle mode
3445 during the rcutorture test.
3447 rcutorture.shutdown_secs= [KNL]
3448 Set time (s) after boot system shutdown. This
3449 is useful for hands-off automated testing.
3451 rcutorture.stall_cpu= [KNL]
3452 Duration of CPU stall (s) to test RCU CPU stall
3453 warnings, zero to disable.
3455 rcutorture.stall_cpu_holdoff= [KNL]
3456 Time to wait (s) after boot before inducing stall.
3458 rcutorture.stat_interval= [KNL]
3459 Time (s) between statistics printk()s.
3461 rcutorture.stutter= [KNL]
3462 Time (s) to stutter testing, for example, specifying
3463 five seconds causes the test to run for five seconds,
3464 wait for five seconds, and so on. This tests RCU's
3465 ability to transition abruptly to and from idle.
3467 rcutorture.test_boost= [KNL]
3468 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
3469 "Maybe" means test if the RCU implementation
3470 under test support RCU priority boosting.
3472 rcutorture.test_boost_duration= [KNL]
3473 Duration (s) of each individual boost test.
3475 rcutorture.test_boost_interval= [KNL]
3476 Interval (s) between each boost test.
3478 rcutorture.test_no_idle_hz= [KNL]
3479 Test RCU's dyntick-idle handling. See also the
3480 rcutorture.shuffle_interval parameter.
3482 rcutorture.torture_runnable= [BOOT]
3483 Start rcutorture running at boot time.
3485 rcutorture.torture_type= [KNL]
3486 Specify the RCU implementation to test.
3488 rcutorture.verbose= [KNL]
3489 Enable additional printk() statements.
3491 rcupdate.rcu_cpu_stall_suppress= [KNL]
3492 Suppress RCU CPU stall warning messages.
3494 rcupdate.rcu_cpu_stall_timeout= [KNL]
3495 Set timeout for RCU CPU stall warning messages.
3497 rcupdate.rcu_expedited= [KNL]
3498 Use expedited grace-period primitives, for
3499 example, synchronize_rcu_expedited() instead
3500 of synchronize_rcu(). This reduces latency,
3501 but can increase CPU utilization, degrade
3502 real-time latency, and degrade energy efficiency.
3503 No effect on CONFIG_TINY_RCU kernels.
3505 rcupdate.rcu_normal= [KNL]
3506 Use only normal grace-period primitives,
3507 for example, synchronize_rcu() instead of
3508 synchronize_rcu_expedited(). This improves
3509 real-time latency, CPU utilization, and
3510 energy efficiency, but can expose users to
3511 increased grace-period latency. This parameter
3512 overrides rcupdate.rcu_expedited. No effect on
3513 CONFIG_TINY_RCU kernels.
3515 rcupdate.rcu_normal_after_boot= [KNL]
3516 Once boot has completed (that is, after
3517 rcu_end_inkernel_boot() has been invoked), use
3518 only normal grace-period primitives. No effect
3519 on CONFIG_TINY_RCU kernels.
3521 rcupdate.rcu_task_stall_timeout= [KNL]
3522 Set timeout in jiffies for RCU task stall warning
3523 messages. Disable with a value less than or equal
3526 rcupdate.rcu_self_test= [KNL]
3527 Run the RCU early boot self tests
3529 rcupdate.rcu_self_test_bh= [KNL]
3530 Run the RCU bh early boot self tests
3532 rcupdate.rcu_self_test_sched= [KNL]
3533 Run the RCU sched early boot self tests
3537 Run specified binary instead of /init from the ramdisk,
3538 used for early userspace startup. See initrd.
3541 Format (x86 or x86_64):
3542 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
3544 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
3546 Where reboot_mode is one of warm (soft) or cold (hard) or gpio,
3547 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
3548 reboot_force is either force or not specified,
3549 reboot_cpu is s[mp]#### with #### being the processor
3550 to be used for rebooting.
3553 [KNL, SMP] Set scheduler's default relax_domain_level.
3554 See Documentation/cgroups/cpusets.txt.
3556 relative_sleep_states=
3557 [SUSPEND] Use sleep state labeling where the deepest
3558 state available other than hibernation is always "mem".
3559 Format: { "0" | "1" }
3560 0 -- Traditional sleep state labels.
3561 1 -- Relative sleep state labels.
3563 reserve= [KNL,BUGS] Force the kernel to ignore some iomem area
3565 reservetop= [X86-32]
3567 Reserves a hole at the top of the kernel virtual
3572 Set the amount of memory to reserve for BIOS at
3573 the bottom of the address space.
3575 reset_devices [KNL] Force drivers to reset the underlying device
3576 during initialization.
3579 Specify the partition device for software suspend
3581 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
3583 resume_offset= [SWSUSP]
3584 Specify the offset from the beginning of the partition
3585 given by "resume=" at which the swap header is located,
3586 in <PAGE_SIZE> units (needed only for swap files).
3587 See Documentation/power/swsusp-and-swap-files.txt
3589 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
3590 read the resume files
3592 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
3593 Useful for devices that are detected asynchronously
3594 (e.g. USB and MMC devices).
3596 hibernate= [HIBERNATION]
3597 noresume Don't check if there's a hibernation image
3598 present during boot.
3599 nocompress Don't compress/decompress hibernation images.
3600 no Disable hibernation and resume.
3602 retain_initrd [RAM] Keep initrd memory after extraction
3604 rfkill.default_state=
3605 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
3606 etc. communication is blocked by default.
3609 rfkill.master_switch_mode=
3610 0 The "airplane mode" button does nothing.
3611 1 The "airplane mode" button toggles between everything
3612 blocked and the previous configuration.
3613 2 The "airplane mode" button toggles between everything
3614 blocked and everything unblocked.
3616 rhash_entries= [KNL,NET]
3617 Set number of hash buckets for route cache
3619 ro [KNL] Mount root device read-only on boot
3622 on Mark read-only kernel memory as read-only (default).
3623 off Leave read-only kernel memory writable for debugging.
3626 Enable the uart passthrough on the designated usb port
3627 on Rockchip SoCs. When active, the signals of the
3628 debug-uart get routed to the D+ and D- pins of the usb
3629 port and the regular usb controller gets disabled.
3631 root= [KNL] Root filesystem
3632 See name_to_dev_t comment in init/do_mounts.c.
3634 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
3635 mount the root filesystem
3637 rootflags= [KNL] Set root filesystem mount option string
3639 rootfstype= [KNL] Set root filesystem type
3641 rootwait [KNL] Wait (indefinitely) for root device to show up.
3642 Useful for devices that are detected asynchronously
3643 (e.g. USB and MMC devices).
3645 rproc_mem=nn[KMG][@address]
3646 [KNL,ARM,CMA] Remoteproc physical memory block.
3647 Memory area to be used by remote processor image,
3650 rw [KNL] Mount root device read-write on boot
3652 S [KNL] Run init in single mode
3654 s390_iommu= [HW,S390]
3655 Set s390 IOTLB flushing mode
3657 With strict flushing every unmap operation will result in
3658 an IOTLB flush. Default is lazy flushing before reuse,
3662 See drivers/net/irda/sa1100_ir.c.
3664 sbni= [NET] Granch SBNI12 leased line adapter
3666 sched_debug [KNL] Enables verbose scheduler debug messages.
3668 schedstats= [KNL,X86] Enable or disable scheduled statistics.
3669 Allowed values are enable and disable. This feature
3670 incurs a small amount of overhead in the scheduler
3671 but is useful for debugging and performance tuning.
3673 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
3674 xtime_lock contention on larger systems, and/or RCU lock
3675 contention on all systems with CONFIG_MAXSMP set.
3676 Format: { "0" | "1" }
3677 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
3679 Note: increases power consumption, thus should only be
3680 enabled if running jitter sensitive (HPC/RT) workloads.
3682 security= [SECURITY] Choose a security module to enable at boot.
3683 If this boot parameter is not specified, only the first
3684 security module asking for security registration will be
3685 loaded. An invalid security module name will be treated
3686 as if no module has been chosen.
3688 selinux= [SELINUX] Disable or enable SELinux at boot time.
3689 Format: { "0" | "1" }
3690 See security/selinux/Kconfig help text.
3693 Default value is set via kernel config option.
3694 If enabled at boot time, /selinux/disable can be used
3695 later to disable prior to initial policy load.
3697 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
3698 Format: { "0" | "1" }
3699 See security/apparmor/Kconfig help text
3702 Default value is set via kernel config option.
3704 serialnumber [BUGS=X86-32]
3707 Maximal number of shapers.
3709 show_msr= [x86] show boot-time MSR settings
3710 Format: { <integer> }
3711 Show boot-time (BIOS-initialized) MSR settings.
3712 The parameter means the number of CPUs to show,
3713 for example 1 means boot CPU only.
3721 Disable merging of slabs with similar size. May be
3722 necessary if there is some reason to distinguish
3723 allocs to different slabs. Debug options disable
3724 merging on their own.
3725 For more information see Documentation/vm/slub.txt.
3727 slab_max_order= [MM, SLAB]
3728 Determines the maximum allowed order for slabs.
3729 A high setting may cause OOMs due to memory
3730 fragmentation. Defaults to 1 for systems with
3731 more than 32MB of RAM, 0 otherwise.
3733 slub_debug[=options[,slabs]] [MM, SLUB]
3734 Enabling slub_debug allows one to determine the
3735 culprit if slab objects become corrupted. Enabling
3736 slub_debug can create guard zones around objects and
3737 may poison objects when not in use. Also tracks the
3738 last alloc / free. For more information see
3739 Documentation/vm/slub.txt.
3741 slub_max_order= [MM, SLUB]
3742 Determines the maximum allowed order for slabs.
3743 A high setting may cause OOMs due to memory
3744 fragmentation. For more information see
3745 Documentation/vm/slub.txt.
3747 slub_min_objects= [MM, SLUB]
3748 The minimum number of objects per slab. SLUB will
3749 increase the slab order up to slub_max_order to
3750 generate a sufficiently large slab able to contain
3751 the number of objects indicated. The higher the number
3752 of objects the smaller the overhead of tracking slabs
3753 and the less frequently locks need to be acquired.
3754 For more information see Documentation/vm/slub.txt.
3756 slub_min_order= [MM, SLUB]
3757 Determines the minimum page order for slabs. Must be
3758 lower than slub_max_order.
3759 For more information see Documentation/vm/slub.txt.
3761 slub_nomerge [MM, SLUB]
3762 Same with slab_nomerge. This is supported for legacy.
3763 See slab_nomerge for more information.
3766 Format: <io1>[,<io2>[,...,<io8>]]
3768 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
3769 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
3770 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
3771 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
3772 smsc-ircc2.ircc_irq= [HW] IRQ line
3773 smsc-ircc2.ircc_dma= [HW] DMA channel
3774 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
3775 0: Toshiba Satellite 1800 (GP data pin select)
3776 1: Fast pin select (default)
3779 smt [KNL,S390] Set the maximum number of threads (logical
3780 CPUs) to use per physical CPU on systems capable of
3781 symmetric multithreading (SMT). Will be capped to the
3782 actual hardware limit.
3784 Default: -1 (no limit)
3787 [KNL] Should the soft-lockup detector generate panics.
3790 softlockup_all_cpu_backtrace=
3791 [KNL] Should the soft-lockup detector generate
3792 backtraces on all cpus.
3795 sonypi.*= [HW] Sony Programmable I/O Control Device driver
3796 See Documentation/laptops/sonypi.txt
3798 spia_io_base= [HW,MTD]
3804 Enabled the stack tracer on boot up.
3806 stacktrace_filter=[function-list]
3807 [FTRACE] Limit the functions that the stack tracer
3808 will trace at boot up. function-list is a comma separated
3809 list of functions. This list can be changed at run
3810 time by the stack_trace_filter file in the debugfs
3811 tracing directory. Note, this enables stack tracing
3812 and the stacktrace above is not needed.
3816 Set the STI (builtin display/keyboard on the HP-PARISC
3817 machines) console (graphic card) which should be used
3818 as the initial boot-console.
3819 See also comment in drivers/video/console/sticore.c.
3822 See comment in drivers/video/console/sticore.c.
3825 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
3827 sunrpc.min_resvport=
3828 sunrpc.max_resvport=
3830 SunRPC servers often require that client requests
3831 originate from a privileged port (i.e. a port in the
3832 range 0 < portnr < 1024).
3833 An administrator who wishes to reserve some of these
3834 ports for other uses may adjust the range that the
3835 kernel's sunrpc client considers to be privileged
3836 using these two parameters to set the minimum and
3837 maximum port values.
3841 Control how the NFS server code allocates CPUs to
3842 service thread pools. Depending on how many NICs
3843 you have and where their interrupts are bound, this
3844 option will affect which CPUs will do NFS serving.
3845 Note: this parameter cannot be changed while the
3846 NFS server is running.
3848 auto the server chooses an appropriate mode
3849 automatically using heuristics
3850 global a single global pool contains all CPUs
3851 percpu one pool for each CPU
3852 pernode one pool for each NUMA node (equivalent
3853 to global on non-NUMA machines)
3855 sunrpc.tcp_slot_table_entries=
3856 sunrpc.udp_slot_table_entries=
3858 Sets the upper limit on the number of simultaneous
3859 RPC calls that can be sent from the client to a
3860 server. Increasing these values may allow you to
3861 improve throughput, but will also increase the
3862 amount of memory reserved for use by the client.
3864 suspend.pm_test_delay=
3866 Sets the number of seconds to remain in a suspend test
3867 mode before resuming the system (see
3868 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
3869 is set. Default value is 5.
3872 [KNL] Enable accounting of swap in memory resource
3873 controller if no parameter or 1 is given or disable
3874 it if 0 is given (See Documentation/cgroups/memory.txt)
3876 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
3877 Format: { <int> | force }
3878 <int> -- Number of I/O TLB slabs
3879 force -- force using of bounce buffers even if they
3880 wouldn't be automatically used by the kernel
3884 sysfs.deprecated=0|1 [KNL]
3885 Enable/disable old style sysfs layout for old udev
3886 on older distributions. When this option is enabled
3887 very new udev will not work anymore. When this option
3888 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
3889 in older udev will not work anymore.
3890 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
3891 the kernel configuration.
3893 sysrq_always_enabled
3895 Ignore sysrq setting - this boot parameter will
3896 neutralize any effect of /proc/sys/kernel/sysrq.
3897 Useful for debugging.
3899 tcpmhash_entries= [KNL,NET]
3900 Set the number of tcp_metrics_hash slots.
3901 Default value is 8192 or 16384 depending on total
3902 ram pages. This is used to specify the TCP metrics
3903 cache size. See Documentation/networking/ip-sysctl.txt
3904 "tcp_no_metrics_save" section for more details.
3908 test_suspend= [SUSPEND][,N]
3909 Specify "mem" (for Suspend-to-RAM) or "standby" (for
3910 standby suspend) or "freeze" (for suspend type freeze)
3911 as the system sleep state during system startup with
3912 the optional capability to repeat N number of times.
3913 The system is woken from this state using a
3914 wakeup-capable RTC alarm.
3916 thash_entries= [KNL,NET]
3917 Set number of hash buckets for TCP connection
3919 thermal.act= [HW,ACPI]
3920 -1: disable all active trip points in all thermal zones
3921 <degrees C>: override all lowest active trip points
3923 thermal.crt= [HW,ACPI]
3924 -1: disable all critical trip points in all thermal zones
3925 <degrees C>: override all critical trip points
3927 thermal.nocrt= [HW,ACPI]
3928 Set to disable actions on ACPI thermal zone
3929 critical and hot trip points.
3931 thermal.off= [HW,ACPI]
3932 1: disable ACPI thermal control
3934 thermal.psv= [HW,ACPI]
3935 -1: disable all passive trip points
3936 <degrees C>: override all passive trip points to this
3939 thermal.tzp= [HW,ACPI]
3940 Specify global default ACPI thermal zone polling rate
3941 <deci-seconds>: poll all this frequency
3942 0: no polling (default)
3945 Force threading of all interrupt handlers except those
3946 marked explicitly IRQF_NO_THREAD.
3949 Enable the Transcendent memory driver if built-in.
3951 tmem.cleancache=0|1 [KNL, XEN]
3952 Default is on (1). Disable the usage of the cleancache
3953 API to send anonymous pages to the hypervisor.
3955 tmem.frontswap=0|1 [KNL, XEN]
3956 Default is on (1). Disable the usage of the frontswap
3957 API to send swap pages to the hypervisor. If disabled
3958 the selfballooning and selfshrinking are force disabled.
3960 tmem.selfballooning=0|1 [KNL, XEN]
3961 Default is on (1). Disable the driving of swap pages
3964 tmem.selfshrinking=0|1 [KNL, XEN]
3965 Default is on (1). Partial swapoff that immediately
3966 transfers pages from Xen hypervisor back to the
3967 kernel based on different criteria.
3971 Specify if the kernel should make use of the cpu
3972 topology information if the hardware supports this.
3973 The scheduler will make use of this information and
3974 e.g. base its process migration decisions on it.
3977 topology_updates= [KNL, PPC, NUMA]
3979 Specify if the kernel should ignore (off)
3980 topology updates sent by the hypervisor to this
3985 tpm_suspend_pcr=[HW,TPM]
3986 Format: integer pcr id
3987 Specify that at suspend time, the tpm driver
3988 should extend the specified pcr with zeros,
3989 as a workaround for some chips which fail to
3990 flush the last written pcr on TPM_SaveState.
3991 This will guarantee that all the other pcrs
3994 trace_buf_size=nn[KMG]
3995 [FTRACE] will set tracing buffer size on each cpu.
3997 trace_event=[event-list]
3998 [FTRACE] Set and start specified trace events in order
3999 to facilitate early boot debugging. The event-list is a
4000 comma separated list of trace events to enable. See
4001 also Documentation/trace/events.txt
4003 trace_options=[option-list]
4004 [FTRACE] Enable or disable tracer options at boot.
4005 The option-list is a comma delimited list of options
4006 that can be enabled or disabled just as if you were
4007 to echo the option name into
4009 /sys/kernel/debug/tracing/trace_options
4011 For example, to enable stacktrace option (to dump the
4012 stack trace of each event), add to the command line:
4014 trace_options=stacktrace
4016 See also Documentation/trace/ftrace.txt "trace options"
4020 Have the tracepoints sent to printk as well as the
4021 tracing ring buffer. This is useful for early boot up
4022 where the system hangs or reboots and does not give the
4023 option for reading the tracing buffer or performing a
4024 ftrace_dump_on_oops.
4026 To turn off having tracepoints sent to printk,
4027 echo 0 > /proc/sys/kernel/tracepoint_printk
4028 Note, echoing 1 into this file without the
4029 tracepoint_printk kernel cmdline option has no effect.
4033 Having tracepoints sent to printk() and activating high
4034 frequency tracepoints such as irq or sched, can cause
4035 the system to live lock.
4038 [FTRACE] enable this option to disable tracing when a
4039 warning is hit. This turns off "tracing_on". Tracing can
4040 be enabled again by echoing '1' into the "tracing_on"
4041 file located in /sys/kernel/debug/tracing/
4043 This option is useful, as it disables the trace before
4044 the WARNING dump is called, which prevents the trace to
4045 be filled with content caused by the warning output.
4047 This option can also be set at run time via the sysctl
4048 option: kernel/traceoff_on_warning
4050 transparent_hugepage=
4052 Format: [always|madvise|never]
4053 Can be used to control the default behavior of the system
4054 with respect to transparent hugepages.
4055 See Documentation/vm/transhuge.txt for more details.
4057 tsc= Disable clocksource stability checks for TSC.
4059 [x86] reliable: mark tsc clocksource as reliable, this
4060 disables clocksource verification at runtime, as well
4061 as the stability checks done at bootup. Used to enable
4062 high-resolution timer mode on older hardware, and in
4063 virtualized environment.
4064 [x86] noirqtime: Do not use TSC to do irq accounting.
4065 Used to run time disable IRQ_TIME_ACCOUNTING on any
4066 platforms where RDTSC is slow and this accounting
4069 turbografx.map[2|3]= [HW,JOY]
4070 TurboGraFX parallel port interface
4072 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
4073 See also Documentation/input/joystick-parport.txt
4075 udbg-immortal [PPC] When debugging early kernel crashes that
4076 happen after console_init() and before a proper
4077 console driver takes over, this boot options might
4078 help "seeing" what's going on.
4080 uhash_entries= [KNL,NET]
4081 Set number of hash buckets for UDP/UDP-Lite connections
4084 [USB] Ignore overcurrent events (default N).
4085 Some badly-designed motherboards generate lots of
4086 bogus events, for ports that aren't wired to
4087 anything. Set this parameter to avoid log spamming.
4088 Note that genuine overcurrent events won't be
4092 [X86] Cause panic on unknown NMI.
4094 usbcore.authorized_default=
4095 [USB] Default USB device authorization:
4096 (default -1 = authorized except for wireless USB,
4097 0 = not authorized, 1 = authorized)
4099 usbcore.autosuspend=
4100 [USB] The autosuspend time delay (in seconds) used
4101 for newly-detected USB devices (default 2). This
4102 is the time required before an idle device will be
4103 autosuspended. Devices for which the delay is set
4104 to a negative value won't be autosuspended at all.
4106 usbcore.usbfs_snoop=
4107 [USB] Set to log all usbfs traffic (default 0 = off).
4109 usbcore.usbfs_snoop_max=
4110 [USB] Maximum number of bytes to snoop in each URB
4113 usbcore.blinkenlights=
4114 [USB] Set to cycle leds on hubs (default 0 = off).
4116 usbcore.old_scheme_first=
4117 [USB] Start with the old device initialization
4118 scheme (default 0 = off).
4120 usbcore.usbfs_memory_mb=
4121 [USB] Memory limit (in MB) for buffers allocated by
4122 usbfs (default = 16, 0 = max = 2047).
4124 usbcore.use_both_schemes=
4125 [USB] Try the other device initialization scheme
4126 if the first one fails (default 1 = enabled).
4128 usbcore.initial_descriptor_timeout=
4129 [USB] Specifies timeout for the initial 64-byte
4130 USB_REQ_GET_DESCRIPTOR request in milliseconds
4131 (default 5000 = 5.0 seconds).
4133 usbcore.nousb [USB] Disable the USB subsystem
4136 [USBHID] The interval which mice are to be polled at.
4138 usb-storage.delay_use=
4139 [UMS] The delay in seconds before a new device is
4140 scanned for Logical Units (default 1).
4143 [UMS] A list of quirks entries to supplement or
4144 override the built-in unusual_devs list. List
4145 entries are separated by commas. Each entry has
4146 the form VID:PID:Flags where VID and PID are Vendor
4147 and Product ID values (4-digit hex numbers) and
4148 Flags is a set of characters, each corresponding
4149 to a common usb-storage quirk flag as follows:
4150 a = SANE_SENSE (collect more than 18 bytes
4152 b = BAD_SENSE (don't collect more than 18
4153 bytes of sense data);
4154 c = FIX_CAPACITY (decrease the reported
4155 device capacity by one sector);
4156 d = NO_READ_DISC_INFO (don't use
4157 READ_DISC_INFO command);
4158 e = NO_READ_CAPACITY_16 (don't use
4159 READ_CAPACITY_16 command);
4160 f = NO_REPORT_OPCODES (don't use report opcodes
4162 g = MAX_SECTORS_240 (don't transfer more than
4163 240 sectors at a time, uas only);
4164 h = CAPACITY_HEURISTICS (decrease the
4165 reported device capacity by one
4166 sector if the number is odd);
4167 i = IGNORE_DEVICE (don't bind to this
4169 j = NO_REPORT_LUNS (don't use report luns
4171 l = NOT_LOCKABLE (don't try to lock and
4172 unlock ejectable media);
4173 m = MAX_SECTORS_64 (don't transfer more
4174 than 64 sectors = 32 KB at a time);
4175 n = INITIAL_READ10 (force a retry of the
4176 initial READ(10) command);
4177 o = CAPACITY_OK (accept the capacity
4178 reported by the device);
4179 p = WRITE_CACHE (the device cache is ON
4181 r = IGNORE_RESIDUE (the device reports
4182 bogus residue values);
4183 s = SINGLE_LUN (the device has only one
4185 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
4186 commands, uas only);
4187 u = IGNORE_UAS (don't bind to the uas driver);
4188 w = NO_WP_DETECT (don't test whether the
4189 medium is write-protected).
4190 Example: quirks=0419:aaf5:rl,0421:0433:rc
4192 user_debug= [KNL,ARM]
4194 See arch/arm/Kconfig.debug help text.
4195 1 - undefined instruction events
4197 4 - invalid data aborts
4200 Example: user_debug=31
4203 [X86] Flags controlling user PTE allocations.
4205 nohigh = do not allocate PTE pages in
4206 HIGHMEM regardless of setting
4210 On X86_32, this is an alias for vdso32=. Otherwise:
4212 vdso=1: enable VDSO (the default)
4213 vdso=0: disable VDSO mapping
4215 vdso32= [X86] Control the 32-bit vDSO
4216 vdso32=1: enable 32-bit VDSO
4217 vdso32=0 or vdso32=2: disable 32-bit VDSO
4219 See the help text for CONFIG_COMPAT_VDSO for more
4220 details. If CONFIG_COMPAT_VDSO is set, the default is
4221 vdso32=0; otherwise, the default is vdso32=1.
4223 For compatibility with older kernels, vdso32=2 is an
4226 Try vdso32=0 if you encounter an error that says:
4227 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
4230 vector=percpu: enable percpu vector domain
4232 video= [FB] Frame buffer configuration
4233 See Documentation/fb/modedb.txt.
4235 video.brightness_switch_enabled= [0,1]
4236 If set to 1, on receiving an ACPI notify event
4237 generated by hotkey, video driver will adjust brightness
4238 level and then send out the event to user space through
4239 the allocated input device; If set to 0, video driver
4240 will only send out the event without touching backlight
4245 [VMMIO] Memory mapped virtio (platform) device.
4247 <size>@<baseaddr>:<irq>[:<id>]
4249 <size> := size (can use standard suffixes
4251 <baseaddr> := physical base address
4252 <irq> := interrupt number (as passed to
4254 <id> := (optional) platform device id
4256 virtio_mmio.device=1K@0x100b0000:48:7
4258 Can be used multiple times for multiple devices.
4260 vga= [BOOT,X86-32] Select a particular video mode
4261 See Documentation/x86/boot.txt and
4262 Documentation/svga.txt.
4263 Use vga=ask for menu.
4264 This is actually a boot loader parameter; the value is
4265 passed to the kernel using a special protocol.
4267 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
4268 size of <nn>. This can be used to increase the
4269 minimum size (128MB on x86). It can also be used to
4270 decrease the size and leave more room for directly
4273 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
4276 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
4279 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
4283 Controls the behavior of vsyscalls (i.e. calls to
4284 fixed addresses of 0xffffffffff600x00 from legacy
4285 code). Most statically-linked binaries and older
4286 versions of glibc use these calls. Because these
4287 functions are at fixed addresses, they make nice
4288 targets for exploits that can control RIP.
4290 emulate [default] Vsyscalls turn into traps and are
4291 emulated reasonably safely.
4293 native Vsyscalls are native syscall instructions.
4294 This is a little bit faster than trapping
4295 and makes a few dynamic recompilers work
4296 better than they would in emulation mode.
4297 It also makes exploits much easier to write.
4299 none Vsyscalls don't work at all. This makes
4300 them quite hard to use for exploits but
4301 might break your system.
4303 vt.color= [VT] Default text color.
4304 Format: 0xYX, X = foreground, Y = background.
4305 Default: 0x07 = light gray on black.
4307 vt.cur_default= [VT] Default cursor shape.
4308 Format: 0xCCBBAA, where AA, BB, and CC are the same as
4309 the parameters of the <Esc>[?A;B;Cc escape sequence;
4310 see VGA-softcursor.txt. Default: 2 = underline.
4312 vt.default_blu= [VT]
4313 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
4314 Change the default blue palette of the console.
4315 This is a 16-member array composed of values
4318 vt.default_grn= [VT]
4319 Format: <green0>,<green1>,<green2>,...,<green15>
4320 Change the default green palette of the console.
4321 This is a 16-member array composed of values
4324 vt.default_red= [VT]
4325 Format: <red0>,<red1>,<red2>,...,<red15>
4326 Change the default red palette of the console.
4327 This is a 16-member array composed of values
4333 Set system-wide default UTF-8 mode for all tty's.
4334 Default is 1, i.e. UTF-8 mode is enabled for all
4335 newly opened terminals.
4337 vt.global_cursor_default=
4340 Set system-wide default for whether a cursor
4341 is shown on new VTs. Default is -1,
4342 i.e. cursors will be created by default unless
4343 overridden by individual drivers. 0 will hide
4344 cursors, 1 will display them.
4346 vt.italic= [VT] Default color for italic text; 0-15.
4349 vt.underline= [VT] Default color for underlined text; 0-15.
4352 watchdog timers [HW,WDT] For information on watchdog timers,
4353 see Documentation/watchdog/watchdog-parameters.txt
4354 or other driver-specific files in the
4355 Documentation/watchdog/ directory.
4357 workqueue.watchdog_thresh=
4358 If CONFIG_WQ_WATCHDOG is configured, workqueue can
4359 warn stall conditions and dump internal state to
4360 help debugging. 0 disables workqueue stall
4361 detection; otherwise, it's the stall threshold
4362 duration in seconds. The default value is 30 and
4363 it can be updated at runtime by writing to the
4364 corresponding sysfs file.
4366 workqueue.disable_numa
4367 By default, all work items queued to unbound
4368 workqueues are affine to the NUMA nodes they're
4369 issued on, which results in better behavior in
4370 general. If NUMA affinity needs to be disabled for
4371 whatever reason, this option can be used. Note
4372 that this also can be controlled per-workqueue for
4373 workqueues visible under /sys/bus/workqueue/.
4375 workqueue.power_efficient
4376 Per-cpu workqueues are generally preferred because
4377 they show better performance thanks to cache
4378 locality; unfortunately, per-cpu workqueues tend to
4379 be more power hungry than unbound workqueues.
4381 Enabling this makes the per-cpu workqueues which
4382 were observed to contribute significantly to power
4383 consumption unbound, leading to measurably lower
4384 power usage at the cost of small performance
4387 The default value of this parameter is determined by
4388 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
4390 workqueue.debug_force_rr_cpu
4391 Workqueue used to implicitly guarantee that work
4392 items queued without explicit CPU specified are put
4393 on the local CPU. This guarantee is no longer true
4394 and while local CPU is still preferred work items
4395 may be put on foreign CPUs. This debug option
4396 forces round-robin CPU selection to flush out
4397 usages which depend on the now broken guarantee.
4398 When enabled, memory and cache locality will be
4401 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
4402 default x2apic cluster mode on platforms
4405 x86_intel_mid_timer= [X86-32,APBT]
4406 Choose timer option for x86 Intel MID platform.
4407 Two valid options are apbt timer only and lapic timer
4408 plus one apbt timer for broadcast timer.
4409 x86_intel_mid_timer=apbt_only | lapic_and_apbt
4411 xen_512gb_limit [KNL,X86-64,XEN]
4412 Restricts the kernel running paravirtualized under Xen
4413 to use only up to 512 GB of RAM. The reason to do so is
4414 crash analysis tools and Xen tools for doing domain
4415 save/restore/migration must be enabled to handle larger
4418 xen_emul_unplug= [HW,X86,XEN]
4419 Unplug Xen emulated devices
4420 Format: [unplug0,][unplug1]
4421 ide-disks -- unplug primary master IDE devices
4422 aux-ide-disks -- unplug non-primary-master IDE devices
4423 nics -- unplug network devices
4424 all -- unplug all emulated devices (NICs and IDE disks)
4425 unnecessary -- unplugging emulated devices is
4426 unnecessary even if the host did not respond to
4428 never -- do not unplug even if version check succeeds
4430 xen_nopvspin [X86,XEN]
4431 Disables the ticketlock slowpath using Xen PV
4435 Disables the PV optimizations forcing the HVM guest to
4436 run as generic HVM guest with no PV drivers.
4438 xirc2ps_cs= [NET,PCMCIA]
4440 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
4442 ______________________________________________________________________
4446 Add more DRM drivers.