]> git.proxmox.com Git - mirror_ubuntu-bionic-kernel.git/blame - Documentation/admin-guide/kernel-parameters.txt
arm64: Add 'ssbd' command-line option
[mirror_ubuntu-bionic-kernel.git] / Documentation / admin-guide / kernel-parameters.txt
CommitLineData
b10d79f7 1 acpi= [HW,ACPI,X86,ARM64]
03d926f8 2 Advanced Configuration and Power Interface
6a1f5471 3 Format: { force | on | off | strict | noirq | rsdt |
e58d154b 4 copy_dsdt }
1da177e4 5 force -- enable ACPI if default was off
6a1f5471 6 on -- enable ACPI but allow fallback to DT [arm64]
1da177e4
LT
7 off -- disable ACPI if default was on
8 noirq -- do not use ACPI for IRQ routing
a9913044 9 strict -- Be less tolerant of platforms that are not
1da177e4 10 strictly ACPI specification compliant.
237889bf 11 rsdt -- prefer RSDT over (default) XSDT
aa2110cb 12 copy_dsdt -- copy DSDT to memory
6a1f5471
AB
13 For ARM64, ONLY "acpi=off", "acpi=on" or "acpi=force"
14 are available
1da177e4 15
395cf969 16 See also Documentation/power/runtime_pm.txt, pci=noacpi
1da177e4 17
a1fdcc0d
LB
18 acpi_apic_instance= [ACPI, IOAPIC]
19 Format: <int>
20 2: use 2nd APIC table, if available
21 1,0: use 1st APIC table
4e381a4f 22 default: 0
a1fdcc0d 23
c3d6de69
TR
24 acpi_backlight= [HW,ACPI]
25 acpi_backlight=vendor
26 acpi_backlight=video
27 If set to vendor, prefer vendor specific driver
28 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
29 of the ACPI video.ko driver.
30
b2ca5dae
CIK
31 acpi_force_32bit_fadt_addr
32 force FADT to use 32 bit addresses rather than the
33 64 bit X_* addresses. Some firmware have broken 64
34 bit addresses for force ACPI ignore these and use
35 the older legacy 32 bit addresses.
36
ef69449b
DB
37 acpica_no_return_repair [HW, ACPI]
38 Disable AML predefined validation mechanism
39 This mechanism can repair the evaluation result to make
40 the return objects more ACPI specification compliant.
41 This option is useful for developers to identify the
42 root cause of an AML interpreter issue when the issue
43 has something to do with the repair mechanism.
44
a0d84a92
BH
45 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
46 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
1da177e4 47 Format: <int>
a0d84a92
BH
48 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
49 debug output. Bits in debug_layer correspond to a
50 _COMPONENT in an ACPI source file, e.g.,
51 #define _COMPONENT ACPI_PCI_COMPONENT
52 Bits in debug_level correspond to a level in
53 ACPI_DEBUG_PRINT statements, e.g.,
54 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
e76f4276
BH
55 The debug_level mask defaults to "info". See
56 Documentation/acpi/debug.txt for more information about
57 debug layers and levels.
a0d84a92 58
e76f4276
BH
59 Enable processor driver info messages:
60 acpi.debug_layer=0x20000000
61 Enable PCI/PCI interrupt routing info messages:
62 acpi.debug_layer=0x400000
a0d84a92
BH
63 Enable AML "Debug" output, i.e., stores to the Debug
64 object while interpreting AML:
65 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
a0d84a92
BH
66 Enable all messages related to ACPI hardware:
67 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
68
69 Some values produce so much output that the system is
70 unusable. The "log_buf_len" parameter may be useful
71 if you need to capture more output.
f989106c 72
ef69449b
DB
73 acpi_enforce_resources= [ACPI]
74 { strict | lax | no }
75 Check for resource conflicts between native drivers
76 and ACPI OperationRegions (SystemIO and SystemMemory
77 only). IO ports and memory declared in ACPI might be
78 used by the ACPI subsystem in arbitrary AML code and
79 can interfere with legacy drivers.
80 strict (default): access to resources claimed by ACPI
81 is denied; legacy drivers trying to access reserved
82 resources will fail to bind to device using them.
83 lax: access to resources claimed by ACPI is allowed;
84 legacy drivers trying to access reserved resources
85 will bind successfully but a warning message is logged.
86 no: ACPI OperationRegions are not marked as reserved,
87 no further checks are performed.
88
4fc0a7e8
LZ
89 acpi_force_table_verification [HW,ACPI]
90 Enable table checksum verification during early stage.
91 By default, this is disabled due to x86 early mapping
92 size limitation.
93
0cb55ad2
RD
94 acpi_irq_balance [HW,ACPI]
95 ACPI will balance active IRQs
96 default in APIC mode
97
98 acpi_irq_nobalance [HW,ACPI]
99 ACPI will not move active IRQs (default)
100 default in PIC mode
101
102 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
103 Format: <irq>,<irq>...
104
105 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
106 use by PCI
107 Format: <irq>,<irq>...
108
9c4aa1ee
LZ
109 acpi_mask_gpe= [HW,ACPI]
110 Due to the existence of _Lxx/_Exx, some GPEs triggered
111 by unsupported hardware/firmware features can result in
112 GPE floodings that cannot be automatically disabled by
113 the GPE dispatcher.
114 This facility can be used to prevent such uncontrolled
115 GPE floodings.
116 Format: <int>
117 Support masking of GPEs numbered from 0x00 to 0x7f.
118
08e1d7c0
LZ
119 acpi_no_auto_serialize [HW,ACPI]
120 Disable auto-serialization of AML methods
22b5afce
BM
121 AML control methods that contain the opcodes to create
122 named objects will be marked as "Serialized" by the
123 auto-serialization feature.
08e1d7c0
LZ
124 This feature is enabled by default.
125 This option allows to turn off the feature.
22b5afce 126
ef69449b
DB
127 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
128 kernels.
129
a94e88cd
LZ
130 acpi_no_static_ssdt [HW,ACPI]
131 Disable installation of static SSDTs at early boot time
132 By default, SSDTs contained in the RSDT/XSDT will be
133 installed automatically and they will appear under
134 /sys/firmware/acpi/tables.
135 This option turns off this feature.
136 Note that specifying this option does not affect
137 dynamic table installation which will install SSDT
138 tables to /sys/firmware/acpi/tables/dynamic.
0cb55ad2 139
ef69449b
DB
140 acpi_rsdp= [ACPI,EFI,KEXEC]
141 Pass the RSDP address to the kernel, mostly used
142 on machines running EFI runtime service to boot the
143 second kernel for kdump.
4dde507f 144
0cb55ad2
RD
145 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
146 Format: To spoof as Windows 98: ="Microsoft Windows"
147
18d78b64
RW
148 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
149 of 2 which is mandated by ACPI 6) as the supported ACPI
150 specification revision (when using this switch, it may
151 be necessary to carry out a cold reboot _twice_ in a
152 row to make it take effect on the platform firmware).
153
0cb55ad2 154 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
5dc17986
LZ
155 acpi_osi="string1" # add string1
156 acpi_osi="!string2" # remove string2
741d8128 157 acpi_osi=!* # remove all strings
5dc17986
LZ
158 acpi_osi=! # disable all built-in OS vendor
159 strings
a707edeb
LZ
160 acpi_osi=!! # enable all built-in OS vendor
161 strings
0cb55ad2
RD
162 acpi_osi= # disable all strings
163
5dc17986
LZ
164 'acpi_osi=!' can be used in combination with single or
165 multiple 'acpi_osi="string1"' to support specific OS
166 vendor string(s). Note that such command can only
167 affect the default state of the OS vendor strings, thus
168 it cannot affect the default state of the feature group
169 strings and the current state of the OS vendor strings,
170 specifying it multiple times through kernel command line
741d8128
LZ
171 is meaningless. This command is useful when one do not
172 care about the state of the feature group strings which
173 should be controlled by the OSPM.
5dc17986
LZ
174 Examples:
175 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
176 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
177 can make '_OSI("Windows 2000")' TRUE.
178
179 'acpi_osi=' cannot be used in combination with other
180 'acpi_osi=' command lines, the _OSI method will not
181 exist in the ACPI namespace. NOTE that such command can
182 only affect the _OSI support state, thus specifying it
183 multiple times through kernel command line is also
184 meaningless.
185 Examples:
186 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
187 FALSE.
188
741d8128
LZ
189 'acpi_osi=!*' can be used in combination with single or
190 multiple 'acpi_osi="string1"' to support specific
191 string(s). Note that such command can affect the
192 current state of both the OS vendor strings and the
193 feature group strings, thus specifying it multiple times
194 through kernel command line is meaningful. But it may
195 still not able to affect the final state of a string if
196 there are quirks related to this string. This command
197 is useful when one want to control the state of the
198 feature group strings to debug BIOS issues related to
199 the OSPM features.
200 Examples:
201 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
202 '_OSI("Module Device")' FALSE.
203 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
204 '_OSI("Module Device")' TRUE.
205 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
206 equivalent to
207 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
208 and
209 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
210 they all will make '_OSI("Windows 2000")' TRUE.
211
6cececfc 212 acpi_pm_good [X86]
0cb55ad2
RD
213 Override the pmtimer bug detection: force the kernel
214 to assume that this machine's pmtimer latches its value
215 and always returns good values.
216
4af94f39
RD
217 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
218 Format: { level | edge | high | low }
219
4af94f39
RD
220 acpi_skip_timer_override [HW,ACPI]
221 Recognize and ignore IRQ0/pin2 Interrupt Override.
222 For broken nForce2 BIOS resulting in XT-PIC timer.
223
224 acpi_sleep= [HW,ACPI] Sleep options
225 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
c3b0795c 226 old_ordering, nonvs, sci_force_enable }
4af94f39
RD
227 See Documentation/power/video.txt for information on
228 s3_bios and s3_mode.
229 s3_beep is for debugging; it makes the PC's speaker beep
230 as soon as the kernel's real-mode entry point is called.
231 s4_nohwsig prevents ACPI hardware signature from being
232 used during resume from hibernation.
233 old_ordering causes the ACPI 1.0 ordering of the _PTS
234 control method, with respect to putting devices into
235 low power states, to be enforced (the ACPI 2.0 ordering
236 of _PTS is used by default).
72ad5d77
RW
237 nonvs prevents the kernel from saving/restoring the
238 ACPI NVS memory during suspend/hibernation and resume.
d7f0eea9
ZR
239 sci_force_enable causes the kernel to set SCI_EN directly
240 on resume from S1/S3 (which is against the ACPI spec,
241 but some broken systems don't work without it).
4af94f39
RD
242
243 acpi_use_timer_override [HW,ACPI]
244 Use timer override. For some broken Nvidia NF5 boards
245 that require a timer override, but don't have HPET
246
4af94f39
RD
247 add_efi_memmap [EFI; X86] Include EFI memory map in
248 kernel's map of available physical RAM.
249
0cb55ad2
RD
250 agp= [AGP]
251 { off | try_unsupported }
252 off: disable AGP support
253 try_unsupported: try to drive unsupported chipsets
254 (may crash computer or cause data corruption)
255
bcfde334
RD
256 ALSA [HW,ALSA]
257 See Documentation/sound/alsa/alsa-parameters.txt
258
d944d549
RK
259 alignment= [KNL,ARM]
260 Allow the default userspace alignment fault handler
261 behaviour to be specified. Bit 0 enables warnings,
262 bit 1 enables fixups, and bit 2 sends a segfault.
263
dfb09f9b
BP
264 align_va_addr= [X86-64]
265 Align virtual addresses by clearing slice [14:12] when
266 allocating a VMA at process creation time. This option
267 gives you up to 3% performance improvement on AMD F15h
268 machines (where it is enabled by default) for a
269 CPU-intensive style benchmark, and it can vary highly in
270 a microbenchmark depending on workload and compiler.
271
8360ee2f
BP
272 32: only for 32-bit processes
273 64: only for 64-bit processes
dfb09f9b
BP
274 on: enable for both 32- and 64-bit processes
275 off: disable for both 32- and 64-bit processes
276
55034cd6
SRRH
277 alloc_snapshot [FTRACE]
278 Allocate the ftrace snapshot buffer on boot up when the
279 main buffer is allocated. This is handy if debugging
280 and you need to use tracing_snapshot() on boot up, and
281 do not want to use tracing_snapshot_alloc() as it needs
282 to be done where GFP_KERNEL allocations are allowed.
283
89e0b9a3 284 amd_iommu= [HW,X86-64]
54b4cbd2
JR
285 Pass parameters to the AMD IOMMU driver in the system.
286 Possible values are:
afa9fdc2
FT
287 fullflush - enable flushing of IO/TLB entries when
288 they are unmapped. Otherwise they are
289 flushed before they will be reused, which
290 is a lot of faster
a5235725
JR
291 off - do not initialize any AMD IOMMU found in
292 the system
5abcdba4
JR
293 force_isolation - Force device isolation for all
294 devices. The IOMMU driver is not
295 allowed anymore to lift isolation
296 requirements as needed. This option
297 does not override iommu=pt
afa9fdc2 298
c099cf17
SK
299 amd_iommu_dump= [HW,X86-64]
300 Enable AMD IOMMU driver option to dump the ACPI table
301 for AMD IOMMU. With this option enabled, AMD IOMMU
302 driver will print ACPI tables for AMD IOMMU during
303 IOMMU initialization.
304
3928aa3f
SS
305 amd_iommu_intr= [HW,X86-64]
306 Specifies one of the following AMD IOMMU interrupt
307 remapping modes:
308 legacy - Use legacy interrupt remapping mode.
309 vapic - Use virtual APIC mode, which allows IOMMU
310 to inject interrupts directly into guest.
311 This mode requires kvm-amd.avic=1.
312 (Default when IOMMU HW support is present.)
313
1da177e4
LT
314 amijoy.map= [HW,JOY] Amiga joystick support
315 Map of devices attached to JOY0DAT and JOY1DAT
316 Format: <a>,<b>
1752118d 317 See also Documentation/input/joydev/joystick.rst
1da177e4
LT
318
319 analog.map= [HW,JOY] Analog joystick and gamepad support
320 Specifies type or capabilities of an analog joystick
321 connected to one of 16 gameports
322 Format: <type1>,<type2>,..<type16>
323
a9913044
RD
324 apc= [HW,SPARC]
325 Power management functions (SPARCstation-4/5 + deriv.)
1da177e4
LT
326 Format: noidle
327 Disable APC CPU standby support. SPARCstation-Fox does
328 not play well with APC CPU idle - disable it if you have
329 APC and your system crashes randomly.
330
64e05d11 331 apic= [APIC,X86] Advanced Programmable Interrupt Controller
c8facbb6 332 Change the output verbosity whilst booting
1da177e4
LT
333 Format: { quiet (default) | verbose | debug }
334 Change the amount of debugging information output
335 when initialising the APIC and IO-APIC components.
64e05d11
DL
336 For X86-32, this can also be used to specify an APIC
337 driver name.
338 Format: apic=driver_name
339 Examples: apic=bigsmp
a9913044 340
b7c4948e
HK
341 apic_extnmi= [APIC,X86] External NMI delivery setting
342 Format: { bsp (default) | all | none }
343 bsp: External NMI is delivered only to CPU 0
344 all: External NMIs are broadcast to all CPUs as a
345 backup of CPU 0
346 none: External NMI is masked for all CPUs. This is
347 useful so that a dump capture kernel won't be
348 shot down by NMI
349
b0f83b28
BH
350 autoconf= [IPV6]
351 See Documentation/networking/ipv6.txt.
352
9636bc05
CG
353 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
354 Limit apic dumping. The parameter defines the maximal
355 number of local apics being dumped. Also it is possible
356 to set it to "all" by meaning -- no limit here.
357 Format: { 1 (default) | 2 | ... | all }.
358 The parameter valid if only apic=debug or
359 apic=verbose is specified.
360 Example: apic=debug show_lapic=all
361
1da177e4 362 apm= [APM] Advanced Power Management
71f77055 363 See header of arch/x86/kernel/apm_32.c.
1da177e4 364
1da177e4
LT
365 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
366 Format: <io>,<irq>,<nodeID>
367
368 ataflop= [HW,M68k]
369
370 atarimouse= [HW,MOUSE] Atari Mouse
371
1da177e4
LT
372 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
373 EzKey and similar keyboards
374
375 atkbd.reset= [HW] Reset keyboard during initialization
376
a9913044
RD
377 atkbd.set= [HW] Select keyboard code set
378 Format: <int> (2 = AT (default), 3 = PS/2)
1da177e4
LT
379
380 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
381 keyboards
382
383 atkbd.softraw= [HW] Choose between synthetic and real raw mode
384 Format: <bool> (0 = real, 1 = synthetic (default))
a9913044
RD
385
386 atkbd.softrepeat= [HW]
387 Use software keyboard repeat
1da177e4 388
a106fb0c
RGB
389 audit= [KNL] Enable the audit sub-system
390 Format: { "0" | "1" } (0 = disabled, 1 = enabled)
d7961148
EP
391 0 - kernel audit is disabled and can not be enabled
392 until the next reboot
393 unset - kernel audit is initialized but disabled and
394 will be fully enabled by the userspace auditd.
395 1 - kernel audit is initialized and partially enabled,
396 storing at most audit_backlog_limit messages in
397 RAM until it is fully enabled by the userspace
398 auditd.
a106fb0c 399 Default: unset
f3411cb2 400
f910fde7
RGB
401 audit_backlog_limit= [KNL] Set the audit queue size limit.
402 Format: <int> (must be >=0)
403 Default: 64
404
1c532e00
AT
405 bau= [X86_UV] Enable the BAU on SGI UV. The default
406 behavior is to disable the BAU (i.e. bau=0).
407 Format: { "0" | "1" }
408 0 - Disable the BAU.
409 1 - Enable the BAU.
410 unset - Disable the BAU.
411
1da177e4
LT
412 baycom_epp= [HW,AX25]
413 Format: <io>,<mode>
a9913044 414
1da177e4
LT
415 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
416 Format: <io>,<mode>
417 See header of drivers/net/hamradio/baycom_par.c.
418
a9913044
RD
419 baycom_ser_fdx= [HW,AX25]
420 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
1da177e4
LT
421 Format: <io>,<irq>,<mode>[,<baud>]
422 See header of drivers/net/hamradio/baycom_ser_fdx.c.
423
a9913044
RD
424 baycom_ser_hdx= [HW,AX25]
425 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
1da177e4
LT
426 Format: <io>,<irq>,<mode>
427 See header of drivers/net/hamradio/baycom_ser_hdx.c.
428
080506ad
PG
429 blkdevparts= Manual partition parsing of block device(s) for
430 embedded devices based on command line input.
431 See Documentation/block/cmdline-partition.txt
432
bfe8df3d
RD
433 boot_delay= Milliseconds to delay each printk during boot.
434 Values larger than 10 seconds (10000) are changed to
435 no delay (0).
436 Format: integer
437
35fc908d
AH
438 bootmem_debug [KNL] Enable bootmem allocator debug messages.
439
a3e2acc5
HY
440 bert_disable [ACPI]
441 Disable BERT OS support on buggy BIOSes.
442
1da177e4 443 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
a9913044
RD
444 bttv.radio= Most important insmod options are available as
445 kernel args too.
a405ed85 446 bttv.pll= See Documentation/media/v4l-drivers/bttv.rst
395cf969 447 bttv.tuner=
1da177e4 448
4e89a2d8
WS
449 bulk_remove=off [PPC] This parameter disables the use of the pSeries
450 firmware feature for flushing multiple hpte entries
451 at a time.
452
1da177e4
LT
453 c101= [NET] Moxa C101 synchronous serial card
454
cd4f0ef7 455 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
1da177e4
LT
456 Sometimes CPU hardware bugs make them report the cache
457 size incorrectly. The kernel will attempt work arounds
458 to fix known problems, but for some CPUs it is not
459 possible to determine what the correct size should be.
460 This option provides an override for these situations.
461
ffb70f61
DK
462 ca_keys= [KEYS] This parameter identifies a specific key(s) on
463 the system trusted keyring to be used for certificate
464 trust validation.
32c4741c 465 format: { id:<keyid> | builtin }
ffb70f61 466
fd1bb4c9
FF
467 cca= [MIPS] Override the kernel pages' cache coherency
468 algorithm. Accepted values range from 0 to 7
469 inclusive. See arch/mips/include/asm/pgtable-bits.h
470 for platform specific values (SB1, Loongson3 and
471 others).
472
14ff56bb
SO
473 ccw_timeout_log [S390]
474 See Documentation/s390/CommonIO for details.
1da177e4 475
8bab8dde
PM
476 cgroup_disable= [KNL] Disable a particular controller
477 Format: {name of the controller(s) to disable}
ca0bdbb5
QH
478 The effects of cgroup_disable=foo are:
479 - foo isn't auto-mounted if you mount all cgroups in
480 a single hierarchy
481 - foo isn't visible as an individually mountable
482 subsystem
483 {Currently only "memory" controller deal with this and
484 cut the overhead, others just disable the usage. So
485 only cgroup_disable=memory is actually worthy}
8bab8dde 486
1619b6d4
JW
487 cgroup_no_v1= [KNL] Disable one, multiple, all cgroup controllers in v1
488 Format: { controller[,controller...] | "all" }
489 Like cgroup_disable, but only applies to cgroup v1;
490 the blacklisted controllers remain available in cgroup2.
491
f7e1cb6e
JW
492 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
493 Format: <string>
494 nosocket -- Disable socket memory accounting.
04823c83 495 nokmem -- Disable kernel memory accounting.
f7e1cb6e 496
1da177e4
LT
497 checkreqprot [SELINUX] Set initial checkreqprot flag value.
498 Format: { "0" | "1" }
499 See security/selinux/Kconfig help text.
a9913044
RD
500 0 -- check protection applied by kernel (includes
501 any implied execute protection).
1da177e4
LT
502 1 -- check protection requested by application.
503 Default value is set via a kernel config option.
a9913044
RD
504 Value can be changed at runtime via
505 /selinux/checkreqprot.
506
661ca0da
SO
507 cio_ignore= [S390]
508 See Documentation/s390/CommonIO for details.
1e435256
OJ
509 clk_ignore_unused
510 [CLK]
e156ee56
MT
511 Prevents the clock framework from automatically gating
512 clocks that have not been explicitly enabled by a Linux
513 device driver but are enabled in hardware at reset or
514 by the bootloader/firmware. Note that this does not
515 force such clocks to be always-on nor does it reserve
516 those clocks in any way. This parameter is useful for
517 debug and development, but should not be needed on a
518 platform with proper driver support. For more
519 information, see Documentation/clk.txt.
661ca0da 520
cd4f0ef7 521 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
734efb46 522 [Deprecated]
3f6dee9b 523 Forces specified clocksource (if available) to be used
734efb46 524 when calculating gettimeofday(). If specified
3f6dee9b 525 clocksource is not available, it defaults to PIT.
1da177e4
LT
526 Format: { pit | tsc | cyclone | pmtmr }
527
592913ec 528 clocksource= Override the default clocksource
3d6ac984
RD
529 Format: <string>
530 Override the default clocksource and use the clocksource
531 with the name specified.
532 Some clocksource names to choose from, depending on
533 the platform:
534 [all] jiffies (this is the base, fallback clocksource)
535 [ACPI] acpi_pm
536 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
537 pxa_timer,timer3,32k_counter,timer0_1
9863c90f 538 [X86-32] pit,hpet,tsc;
3d6ac984
RD
539 scx200_hrt on Geode; cyclone on IBM x440
540 [MIPS] MIPS
541 [PARISC] cr16
542 [S390] tod
543 [SH] SuperH
544 [SPARC64] tick
545 [X86-64] hpet,tsc
546
46fd5c6b
WD
547 clocksource.arm_arch_timer.evtstrm=
548 [ARM,ARM64]
549 Format: <bool>
550 Enable/disable the eventstream feature of the ARM
551 architected timer so that code using WFE-based polling
552 loops can be debugged more effectively on production
553 systems.
554
ac72e788
AK
555 clearcpuid=BITNUM [X86]
556 Disable CPUID feature X for the kernel. See
cd4d09ec 557 arch/x86/include/asm/cpufeatures.h for the valid bit
07983f0e 558 numbers. Note the Linux specific bits are not necessarily
ac72e788
AK
559 stable over kernel options, but the vendor specific
560 ones should be.
561 Also note that user programs calling CPUID directly
562 or using the feature without checking anything
563 will still see it. This just prevents it from
564 being used by the kernel or shown in /proc/cpuinfo.
565 Also note the kernel might malfunction if you disable
566 some critical bits.
567
5ea3b1b2
AM
568 cma=nn[MG]@[start[MG][-end[MG]]]
569 [ARM,X86,KNL]
570 Sets the size of kernel global memory area for
571 contiguous memory allocations and optionally the
572 placement constraint by the physical address range of
f0d6d1f6
JD
573 memory allocations. A value of 0 disables CMA
574 altogether. For more information, see
c64be2bb
MS
575 include/linux/dma-contiguous.h
576
14f966e7
RJ
577 cmo_free_hint= [PPC] Format: { yes | no }
578 Specify whether pages are marked as being inactive
579 when they are freed. This is used in CMO environments
580 to determine OS memory pressure for page stealing by
581 a hypervisor.
582 Default: yes
583
c7909509
MS
584 coherent_pool=nn[KMG] [ARM,KNL]
585 Sets the size of memory pool for coherent, atomic dma
e9da6e99 586 allocations, by default set to 256K.
c7909509 587
6cececfc 588 code_bytes [X86] How many bytes of object code to print
a25bd949 589 in an oops report.
86c41837
CE
590 Range: 0 - 8192
591 Default: 64
592
1da177e4 593 com20020= [HW,NET] ARCnet - COM20020 chipset
a9913044
RD
594 Format:
595 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
1da177e4
LT
596
597 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
598 Format: <io>[,<irq>]
599
a9913044
RD
600 com90xx= [HW,NET]
601 ARCnet - COM90xx chipset (memory-mapped buffers)
1da177e4
LT
602 Format: <io>[,<irq>[,<memstart>]]
603
604 condev= [HW,S390] console device
605 conmode=
a9913044 606
1da177e4
LT
607 console= [KNL] Output console device and options.
608
609 tty<n> Use the virtual console device <n>.
610
611 ttyS<n>[,options]
f1a1c2dc 612 ttyUSB0[,options]
1da177e4 613 Use the specified serial port. The options are of
f1a1c2dc
RD
614 the form "bbbbpnf", where "bbbb" is the baud rate,
615 "p" is parity ("n", "o", or "e"), "n" is number of
616 bits, and "f" is flow control ("r" for RTS or
617 omit it). Default is "9600n8".
618
e52347bd 619 See Documentation/admin-guide/serial-console.rst for more
f1a1c2dc
RD
620 information. See
621 Documentation/networking/netconsole.txt for an
622 alternative.
1da177e4 623
18a8bd94
YL
624 uart[8250],io,<addr>[,options]
625 uart[8250],mmio,<addr>[,options]
bd94c407 626 uart[8250],mmio16,<addr>[,options]
ca782f16
PH
627 uart[8250],mmio32,<addr>[,options]
628 uart[8250],0x<addr>[,options]
1da177e4
LT
629 Start an early, polled-mode console on the 8250/16550
630 UART at the specified I/O port or MMIO address,
ca782f16
PH
631 switching to the matching ttyS device later.
632 MMIO inter-register address stride is either 8-bit
bd94c407
MY
633 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
634 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
635 to be equivalent to 'mmio'. 'options' are specified in
636 the same format described for ttyS above; if unspecified,
ca782f16
PH
637 the h/w is not re-initialized.
638
a2fd6419
KRW
639 hvc<n> Use the hypervisor console device <n>. This is for
640 both Xen and PowerPC hypervisors.
1da177e4 641
f7511d5f
ST
642 If the device connected to the port is not a TTY but a braille
643 device, prepend "brl," before the device type, for instance
644 console=brl,ttyS0
645 For now, only VisioBraille is supported.
646
f324edc8 647 consoleblank= [KNL] The console blank (screen saver) timeout in
ac0a314c
DX
648 seconds. A value of 0 disables the blank timer.
649 Defaults to 0.
f324edc8 650
4cb0e11b
HK
651 coredump_filter=
652 [KNL] Change the default value for
653 /proc/<pid>/coredump_filter.
654 See also Documentation/filesystems/proc.txt.
655
62a31ce1
LY
656 coresight_cpu_debug.enable
657 [ARM,ARM64]
658 Format: <bool>
659 Enable/disable the CPU sampling based debugging.
660 0: default value, disable debugging
661 1: enable debugging at boot time
662
8a77b5d5
TG
663 cpufreq_driver= [X86] Allow only the named cpu frequency scaling driver
664 to register. Example: cpufreq_driver=powernow-k8
665 Format: { none | STRING }
666
62027aea
LB
667 cpuidle.off=1 [CPU_IDLE]
668 disable the cpuidle sub-system
669
d82f2692
LB
670 cpufreq.off=1 [CPU_FREQ]
671 disable the cpufreq sub-system
672
d68921f9
LB
673 cpu_init_udelay=N
674 [X86] Delay for N microsec between assert and de-assert
675 of APIC INIT to start processors. This delay occurs
676 on every CPU online, such as boot, and resume from suspend.
677 Default: 10000
678
1da177e4 679 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
a9913044
RD
680 Format:
681 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
1da177e4 682
6f21e646
AD
683 crashkernel=size[KMG][@offset[KMG]]
684 [KNL] Using kexec, Linux can switch to a 'crash kernel'
685 upon panic. This parameter reserves the physical
686 memory region [offset, offset + size] for that kernel
687 image. If '@offset' is omitted, then a suitable offset
688 is selected automatically. Check
689 Documentation/kdump/kdump.txt for further details.
dc009d92 690
fb391599
BW
691 crashkernel=range1:size1[,range2:size2,...][@offset]
692 [KNL] Same as above, but depends on the memory
693 in the running system. The syntax of range is
694 start-[end] where start and end are both
695 a memory unit (amount[KMG]). See also
6f21e646 696 Documentation/kdump/kdump.txt for an example.
fb391599 697
adbc742b 698 crashkernel=size[KMG],high
55a20ee7
YL
699 [KNL, x86_64] range could be above 4G. Allow kernel
700 to allocate physical memory region from top, so could
701 be above 4G if system have more than 4G ram installed.
702 Otherwise memory region will be allocated below 4G, if
703 available.
704 It will be ignored if crashkernel=X is specified.
adbc742b
YL
705 crashkernel=size[KMG],low
706 [KNL, x86_64] range under 4G. When crashkernel=X,high
707 is passed, kernel could allocate physical memory region
c729de8f
YL
708 above 4G, that cause second kernel crash on system
709 that require some amount of low memory, e.g. swiotlb
c6045031
BH
710 requires at least 64M+32K low memory, also enough extra
711 low memory is needed to make sure DMA buffers for 32-bit
712 devices won't run out. Kernel would try to allocate at
713 at least 256M below 4G automatically.
c729de8f
YL
714 This one let user to specify own low range under 4G
715 for second kernel instead.
716 0: to disable low allocation.
adbc742b 717 It will be ignored when crashkernel=X,high is not used
55a20ee7 718 or memory reserved is below 4G.
c729de8f 719
9e5c9fe4
RJ
720 cryptomgr.notests
721 [KNL] Disable crypto self-tests
722
1da177e4
LT
723 cs89x0_dma= [HW,NET]
724 Format: <dma>
725
726 cs89x0_media= [HW,NET]
727 Format: { rj45 | aui | bnc }
a9913044 728
a9913044 729 dasd= [HW,NET]
1da177e4
LT
730 See header of drivers/s390/block/dasd_devmap.c.
731
732 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
733 (one device per port)
734 Format: <port#>,<type>
1752118d 735 See also Documentation/input/devices/joystick-parport.rst
1da177e4 736
a648ec05 737 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
787e3075
SM
738 time. See
739 Documentation/admin-guide/dynamic-debug-howto.rst for
29e36c9f 740 details. Deprecated, see dyndbg.
a648ec05 741
1da177e4
LT
742 debug [KNL] Enable kernel debugging (events log level).
743
cae2ed9a
IM
744 debug_locks_verbose=
745 [KNL] verbose self-tests
746 Format=<0|1>
747 Print debugging info while doing the locking API
748 self-tests.
749 We default to 0 (no extra messages), setting it to
750 1 will print _a lot_ more information - normally
751 only useful to kernel developers.
752
3ac7fe5a
TG
753 debug_objects [KNL] Enable object debugging
754
3e8ebb5c
KM
755 no_debug_objects
756 [KNL] Disable object debugging
757
c0a32fc5
SG
758 debug_guardpage_minorder=
759 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
760 parameter allows control of the order of pages that will
761 be intentionally kept free (and hence protected) by the
762 buddy allocator. Bigger value increase the probability
763 of catching random memory corruption, but reduce the
764 amount of memory for normal system use. The maximum
765 possible value is MAX_ORDER/2. Setting this parameter
766 to 1 or 2 should be enough to identify most random
767 memory corruption problems caused by bugs in kernel or
768 driver code when a CPU writes to (or reads from) a
769 random memory location. Note that there exists a class
770 of memory corruptions problems caused by buggy H/W or
771 F/W or by drivers badly programing DMA (basically when
772 memory is written at bus level and the CPU MMU is
773 bypassed) which are not detectable by
774 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
775 tracking down these problems.
776
031bc574
JK
777 debug_pagealloc=
778 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
779 parameter enables the feature at boot time. In
780 default, it is disabled. We can avoid allocating huge
781 chunk of memory for debug pagealloc if we don't enable
782 it at boot time and the system will work mostly same
783 with the kernel built without CONFIG_DEBUG_PAGEALLOC.
784 on: enable the feature
785
d3af01f1
TG
786 debugpat [X86] Enable PAT debugging
787
2d27a966 788 decnet.addr= [HW,NET]
1da177e4
LT
789 Format: <area>[,<node>]
790 See also Documentation/networking/decnet.txt.
791
0cb55ad2
RD
792 default_hugepagesz=
793 [same as hugepagesz=] The size of the default
794 HugeTLB page size. This is the size represented by
795 the legacy /proc/ hugepages APIs, used for SHM, and
796 default size when mounting hugetlbfs filesystems.
797 Defaults to the default architecture's huge page size
798 if not specified.
55ff9780 799
1da177e4
LT
800 dhash_entries= [KNL]
801 Set number of hash buckets for dentry cache.
a9913044 802
faf78829
OH
803 disable_1tb_segments [PPC]
804 Disables the use of 1TB hash page table segments. This
805 causes the kernel to fall back to 256MB segments which
806 can be useful when debugging issues that require an SLB
807 miss to occur.
808
b0f83b28
BH
809 disable= [IPV6]
810 See Documentation/networking/ipv6.txt.
811
b275bfb2
AK
812 disable_radix [PPC]
813 Disable RADIX MMU mode on POWER9
814
151e0c7d
HD
815 disable_cpu_apicid= [X86,APIC,SMP]
816 Format: <int>
817 The number of initial APIC ID for the
818 corresponding CPU to be disabled at boot,
819 mostly used for the kdump 2nd kernel to
820 disable BSP to wake up multiple CPUs without
821 causing system reset or hang due to sending
822 INIT from AP to BSP.
823
4e8b0cf4
NA
824 disable_ddw [PPC/PSERIES]
825 Disable Dynamic DMA Window support. Use this if
826 to workaround buggy firmware.
827
b0f83b28
BH
828 disable_ipv6= [IPV6]
829 See Documentation/networking/ipv6.txt.
830
95ffa243 831 disable_mtrr_cleanup [X86]
95ffa243
YL
832 The kernel tries to adjust MTRR layout from continuous
833 to discrete, to make X server driver able to add WB
0cb55ad2 834 entry later. This parameter disables that.
95ffa243 835
093af8d7 836 disable_mtrr_trim [X86, Intel and AMD only]
99fc8d42
JB
837 By default the kernel will trim any uncacheable
838 memory out of your available memory pool based on
839 MTRR settings. This parameter disables that behavior,
840 possibly causing your machine to run very slowly.
841
6cececfc 842 disable_timer_pin_1 [X86]
0cb55ad2
RD
843 Disable PIN 1 of APIC timer
844 Can be useful to work around chipset bugs.
845
ce14c583
PB
846 dis_ucode_ldr [X86] Disable the microcode loader.
847
0cb55ad2
RD
848 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
849 this option disables the debugging code at boot.
850
851 dma_debug_entries=<number>
852 This option allows to tune the number of preallocated
853 entries for DMA-API debugging code. One entry is
854 required per DMA-API allocation. Use this if the
855 DMA-API debugging code disables itself because the
856 architectural default is too low.
857
1745de5e
JR
858 dma_debug_driver=<driver_name>
859 With this option the DMA-API debugging driver
860 filter feature can be enabled at boot time. Just
861 pass the driver to filter for as the parameter.
862 The filter can be disabled or changed to another
863 driver later using sysfs.
864
53fd40a9 865 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
96206e29
BP
866 Broken monitors, graphic adapters, KVMs and EDIDless
867 panels may send no or incorrect EDID data sets.
868 This parameter allows to specify an EDID data sets
869 in the /lib/firmware directory that are used instead.
da0df92b
CE
870 Generic built-in EDID data sets are used, if one of
871 edid/1024x768.bin, edid/1280x1024.bin,
872 edid/1680x1050.bin, or edid/1920x1080.bin is given
873 and no file with the same name exists. Details and
874 instructions how to build your own EDID data are
875 available in Documentation/EDID/HOWTO.txt. An EDID
876 data set will only be used for a particular connector,
877 if its name and a colon are prepended to the EDID
96206e29
BP
878 name. Each connector may use a unique EDID data
879 set by separating the files with a comma. An EDID
880 data set with no connector name will be used for
881 any connectors not explicitly specified.
da0df92b 882
1da177e4
LT
883 dscc4.setup= [NET]
884
a2b05b7a
NP
885 dt_cpu_ftrs= [PPC]
886 Format: {"off" | "known"}
887 Control how the dt_cpu_ftrs device-tree binding is
888 used for CPU feature discovery and setup (if it
889 exists).
890 off: Do not use it, fall back to legacy cpu table.
891 known: Do not pass through unknown features to guests
892 or userspace, only those that the kernel is aware of.
893
58c5475a
LW
894 dump_apple_properties [X86]
895 Dump name and content of EFI device properties on
896 x86 Macs. Useful for driver authors to determine
897 what data is available or for reverse-engineering.
898
29e36c9f
JC
899 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
900 module.dyndbg[="val"]
901 Enable debug messages at boot time. See
787e3075
SM
902 Documentation/admin-guide/dynamic-debug-howto.rst
903 for details.
29e36c9f 904
8c3641e9
DH
905 nompx [X86] Disables Intel Memory Protection Extensions.
906 See Documentation/x86/intel_mpx.txt for more
907 information about the feature.
908
06976945
DH
909 nopku [X86] Disable Memory Protection Keys CPU feature found
910 in some Intel CPUs.
911
f2411da7
LR
912 module.async_probe [KNL]
913 Enable asynchronous probe on this module.
914
56aeeba8
MS
915 early_ioremap_debug [KNL]
916 Enable debug messages in early_ioremap support. This
917 is useful for tracking down temporary early mappings
918 which are not unmapped.
919
0cb55ad2 920 earlycon= [KNL] Output early console device and options.
0d3c673e 921
5664f764
SW
922 When used with no options, the early console is
923 determined by the stdout-path property in device
924 tree's chosen node.
925
c41251b1
ST
926 cdns,<addr>[,options]
927 Start an early, polled-mode console on a Cadence
928 (xuartps) serial port at the specified address. Only
929 supported option is baud rate. If baud rate is not
930 specified, the serial port must already be setup and
931 configured.
6fa62fc4 932
0cb55ad2
RD
933 uart[8250],io,<addr>[,options]
934 uart[8250],mmio,<addr>[,options]
1917ac76 935 uart[8250],mmio32,<addr>[,options]
6e63be3f 936 uart[8250],mmio32be,<addr>[,options]
ca782f16 937 uart[8250],0x<addr>[,options]
0cb55ad2
RD
938 Start an early, polled-mode console on the 8250/16550
939 UART at the specified I/O port or MMIO address.
16290246 940 MMIO inter-register address stride is either 8-bit
6e63be3f
NC
941 (mmio) or 32-bit (mmio32 or mmio32be).
942 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
943 to be equivalent to 'mmio'. 'options' are specified
944 in the same format described for "console=ttyS<n>"; if
ca782f16 945 unspecified, the h/w is not initialized.
0cb55ad2 946
0d3c673e 947 pl011,<addr>
3b78fae7 948 pl011,mmio32,<addr>
0d3c673e
RH
949 Start an early, polled-mode console on a pl011 serial
950 port at the specified address. The pl011 serial port
951 must already be setup and configured. Options are not
3b78fae7
TT
952 yet supported. If 'mmio32' is specified, then only
953 the driver will use only 32-bit accessors to read/write
954 the device registers.
0d3c673e 955
736d5538
AF
956 meson,<addr>
957 Start an early, polled-mode console on a meson serial
958 port at the specified address. The serial port must
959 already be setup and configured. Options are not yet
960 supported.
961
0efe7296
SB
962 msm_serial,<addr>
963 Start an early, polled-mode console on an msm serial
964 port at the specified address. The serial port
965 must already be setup and configured. Options are not
966 yet supported.
967
968 msm_serial_dm,<addr>
969 Start an early, polled-mode console on an msm serial
970 dm port at the specified address. The serial port
971 must already be setup and configured. Options are not
972 yet supported.
973
e36361d7
AF
974 owl,<addr>
975 Start an early, polled-mode console on a serial port
976 of an Actions Semi SoC, such as S500 or S900, at the
977 specified address. The serial port must already be
978 setup and configured. Options are not yet supported.
979
d50d7269
RH
980 smh Use ARM semihosting calls for early console.
981
b94ba032
TF
982 s3c2410,<addr>
983 s3c2412,<addr>
984 s3c2440,<addr>
985 s3c6400,<addr>
986 s5pv210,<addr>
987 exynos4210,<addr>
988 Use early console provided by serial driver available
989 on Samsung SoCs, requires selecting proper type and
990 a correct base address of the selected UART port. The
991 serial port must already be setup and configured.
992 Options are not yet supported.
993
ec84aa0a
MB
994 lantiq,<addr>
995 Start an early, polled-mode console on a lantiq serial
996 (lqasc) port at the specified address. The serial port
997 must already be setup and configured. Options are not
998 yet supported.
999
1d59b382
SA
1000 lpuart,<addr>
1001 lpuart32,<addr>
1002 Use early console provided by Freescale LP UART driver
1003 found on Freescale Vybrid and QorIQ LS1021A processors.
1004 A valid base address must be provided, and the serial
1005 port must already be setup and configured.
1006
f7c864e7 1007 ar3700_uart,<addr>
30530791
WD
1008 Start an early, polled-mode console on the
1009 Armada 3700 serial port at the specified
1010 address. The serial port must already be setup
1011 and configured. Options are not yet supported.
1012
89175cf7 1013 earlyprintk= [X86,SH,BLACKFIN,ARM,M68k,S390]
1da177e4 1014 earlyprintk=vga
72548e83 1015 earlyprintk=efi
89175cf7 1016 earlyprintk=sclp
2482a92e 1017 earlyprintk=xen
1da177e4 1018 earlyprintk=serial[,ttySn[,baudrate]]
147ea091 1019 earlyprintk=serial[,0x...[,baudrate]]
ea3acb19 1020 earlyprintk=ttySn[,baudrate]
9780bc41 1021 earlyprintk=dbgp[debugController#]
c43088e3 1022 earlyprintk=pciserial,bus:device.function[,baudrate]
1b5aeebf 1023 earlyprintk=xdbc[xhciController#]
1da177e4 1024
147ea091
DH
1025 earlyprintk is useful when the kernel crashes before
1026 the normal console is initialized. It is not enabled by
1027 default because it has some cosmetic problems.
1028
a9913044 1029 Append ",keep" to not disable it when the real console
1da177e4
LT
1030 takes over.
1031
72548e83
MF
1032 Only one of vga, efi, serial, or usb debug port can
1033 be used at a time.
1da177e4 1034
147ea091
DH
1035 Currently only ttyS0 and ttyS1 may be specified by
1036 name. Other I/O ports may be explicitly specified
1037 on some architectures (x86 and arm at least) by
1038 replacing ttySn with an I/O port address, like this:
1039 earlyprintk=serial,0x1008,115200
1040 You can find the port for a given device in
1041 /proc/tty/driver/serial:
1042 2: uart:ST16650V2 port:00001008 irq:18 ...
1da177e4
LT
1043
1044 Interaction with the standard serial driver is not
1045 very good.
1046
72548e83
MF
1047 The VGA and EFI output is eventually overwritten by
1048 the real console.
1da177e4 1049
2482a92e
KRW
1050 The xen output can only be used by Xen PV guests.
1051
89175cf7
HC
1052 The sclp output can only be used on s390.
1053
c700f013
CG
1054 edac_report= [HW,EDAC] Control how to report EDAC event
1055 Format: {"on" | "off" | "force"}
1056 on: enable EDAC to report H/W event. May be overridden
1057 by other higher priority error reporting module.
1058 off: disable H/W event reporting through EDAC.
1059 force: enforce the use of EDAC to report H/W event.
1060 default: on.
1061
9731191f
JW
1062 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1063 ekgdboc=kbd
1064
25985edc 1065 This is designed to be used in conjunction with
9731191f
JW
1066 the boot argument: earlyprintk=vga
1067
1da177e4 1068 edd= [EDD]
8c4dd606 1069 Format: {"off" | "on" | "skip[mbr]"}
1da177e4 1070
d2f7cbe7 1071 efi= [EFI]
fed6cefe 1072 Format: { "old_map", "nochunk", "noruntime", "debug" }
d2f7cbe7
BP
1073 old_map [X86-64]: switch to the old ioremap-based EFI
1074 runtime services mapping. 32-bit still uses this one by
1075 default.
5a17dae4
MF
1076 nochunk: disable reading files in "chunks" in the EFI
1077 boot stub, as chunking can cause problems with some
1078 firmware implementations.
5ae3683c 1079 noruntime : disable EFI runtime services support
fed6cefe 1080 debug: enable misc debug output
d2f7cbe7 1081
8c58bf3e
RW
1082 efi_no_storage_paranoia [EFI; X86]
1083 Using this parameter you can use more than 50% of
1084 your efi variable storage. Use this parameter only if
1085 you are really sure that your UEFI does sane gc and
1086 fulfills the spec otherwise your board may brick.
1087
0f96a99d
TI
1088 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1089 Add arbitrary attribute to specific memory range by
1090 updating original EFI memory map.
1091 Region of memory which aa attribute is added to is
1092 from ss to ss+nn.
1093 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1094 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1095 attribute is added to range 0x100000000-0x180000000 and
1096 0x10a0000000-0x1120000000.
1097
1098 Using this parameter you can do debugging of EFI memmap
1099 related feature. For example, you can do debugging of
1100 Address Range Mirroring feature even if your box
1101 doesn't support it.
1102
475fb4e8
OP
1103 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1104 that is to be dynamically loaded by Linux. If there are
1105 multiple variables with the same name but with different
1106 vendor GUIDs, all of them will be loaded. See
1107 Documentation/acpi/ssdt-overlays.txt for details.
1108
1109
1da177e4
LT
1110 eisa_irq_edge= [PARISC,HW]
1111 See header of drivers/parisc/eisa.c.
1112
cd4f0ef7 1113 elanfreq= [X86-32]
1da177e4 1114 See comment before function elanfreq_setup() in
71f77055 1115 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1da177e4
LT
1116
1117 elevator= [IOSCHED]
17a9e7bb 1118 Format: {"cfq" | "deadline" | "noop"}
395cf969 1119 See Documentation/block/cfq-iosched.txt and
a9913044
RD
1120 Documentation/block/deadline-iosched.txt for details.
1121
d3bf3795 1122 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
a9913044 1123 Specifies physical address of start of kernel core
d3bf3795
MH
1124 image elf header and optionally the size. Generally
1125 kexec loader will pass this option to capture kernel.
aac04b32 1126 See Documentation/kdump/kdump.txt for details.
1da177e4 1127
0cb55ad2
RD
1128 enable_mtrr_cleanup [X86]
1129 The kernel tries to adjust MTRR layout from continuous
1130 to discrete, to make X server driver able to add WB
1131 entry later. This parameter enables that.
1132
ca1eda2d 1133 enable_timer_pin_1 [X86]
0cb55ad2
RD
1134 Enable PIN 1 of APIC timer
1135 Can be useful to work around chipset bugs
1136 (in particular on some ATI chipsets).
1137 The kernel tries to set a reasonable default.
1138
1da177e4
LT
1139 enforcing [SELINUX] Set initial enforcing status.
1140 Format: {"0" | "1"}
1141 See security/selinux/Kconfig help text.
1142 0 -- permissive (log only, no denials).
1143 1 -- enforcing (deny and log).
1144 Default value is 0.
1145 Value can be changed at runtime via /selinux/enforce.
1146
a08f82d0
HY
1147 erst_disable [ACPI]
1148 Disable Error Record Serialization Table (ERST)
1149 support.
1150
1da177e4
LT
1151 ether= [HW,NET] Ethernet cards parameters
1152 This option is obsoleted by the "netdev=" option, which
1153 has equivalent usage. See its documentation for details.
1154
7102ebcd
MZ
1155 evm= [EVM]
1156 Format: { "fix" }
1157 Permit 'security.evm' to be updated regardless of
1158 current integrity status.
1159
de1ba09b
AM
1160 failslab=
1161 fail_page_alloc=
1162 fail_make_request=[KNL]
1163 General fault injection mechanism.
1164 Format: <interval>,<probability>,<space>,<times>
395cf969 1165 See also Documentation/fault-injection/.
de1ba09b 1166
1da177e4 1167 floppy= [HW]
31c00fc1 1168 See Documentation/blockdev/floppy.txt.
1da177e4 1169
f13ae30e
AC
1170 force_pal_cache_flush
1171 [IA-64] Avoid check_sal_cache_flush which may hang on
1172 buggy SAL_CACHE_FLUSH implementations. Using this
1173 parameter will force ia64_sal_cache_flush to call
1174 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1175
69f2366c
CB
1176 forcepae [X86-32]
1177 Forcefully enable Physical Address Extension (PAE).
1178 Many Pentium M systems disable PAE but may have a
1179 functionally usable PAE implementation.
1180 Warning: use of this parameter will taint the kernel
1181 and may cause unknown problems.
1182
d9e54076 1183 ftrace=[tracer]
2af15d6a 1184 [FTRACE] will set and start the specified tracer
d9e54076
PZ
1185 as early as possible in order to facilitate early
1186 boot debugging.
1187
cecbca96 1188 ftrace_dump_on_oops[=orig_cpu]
2af15d6a 1189 [FTRACE] will dump the trace buffers on oops.
cecbca96
FW
1190 If no parameter is passed, ftrace will dump
1191 buffers of all CPUs, but if you pass orig_cpu, it will
1192 dump only the buffer of the CPU that triggered the
1193 oops.
2af15d6a
SR
1194
1195 ftrace_filter=[function-list]
1196 [FTRACE] Limit the functions traced by the function
1197 tracer at boot up. function-list is a comma separated
1198 list of functions. This list can be changed at run
1199 time by the set_ftrace_filter file in the debugfs
16290246 1200 tracing directory.
2af15d6a
SR
1201
1202 ftrace_notrace=[function-list]
1203 [FTRACE] Do not trace the functions specified in
1204 function-list. This list can be changed at run time
1205 by the set_ftrace_notrace file in the debugfs
1206 tracing directory.
d9e54076 1207
369bc18f
SA
1208 ftrace_graph_filter=[function-list]
1209 [FTRACE] Limit the top level callers functions traced
1210 by the function graph tracer at boot up.
1211 function-list is a comma separated list of functions
1212 that can be changed at run time by the
1213 set_graph_function file in the debugfs tracing directory.
1214
0d7d9a16
NK
1215 ftrace_graph_notrace=[function-list]
1216 [FTRACE] Do not trace from the functions specified in
1217 function-list. This list is a comma separated list of
1218 functions that can be changed at run time by the
1219 set_graph_notrace file in the debugfs tracing directory.
1220
65a50c65
TB
1221 ftrace_graph_max_depth=<uint>
1222 [FTRACE] Used with the function graph tracer. This is
1223 the max depth it will trace into a function. This value
1224 can be changed at run time by the max_graph_depth file
1225 in the tracefs tracing directory. default: 0 (no limit)
1226
1da177e4
LT
1227 gamecon.map[2|3]=
1228 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1229 support via parallel port (up to 5 devices per port)
1230 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1752118d 1231 See also Documentation/input/devices/joystick-parport.rst
1da177e4
LT
1232
1233 gamma= [HW,DRM]
1234
aaf23042
YL
1235 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1236 Format: off | on
1237 default: on
1238
2521f2c2
PO
1239 gcov_persist= [GCOV] When non-zero (default), profiling data for
1240 kernel modules is saved and remains accessible via
1241 debugfs, even when the module is unloaded/reloaded.
1242 When zero, profiling data is discarded and associated
1243 debugfs files are removed at module unload time.
1244
47512cfd
TG
1245 goldfish [X86] Enable the goldfish android emulator platform.
1246 Don't use this when you are not running on the
1247 android emulator
1248
1da177e4 1249 gpt [EFI] Forces disk with valid GPT signature but
6c5de79b
DB
1250 invalid Protective MBR to be treated as GPT. If the
1251 primary GPT is corrupted, it enables the backup/alternate
1252 GPT to be used instead.
1da177e4 1253
6cec9b07
AL
1254 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1255 the "Enable 0" bit of the configuration register.
1256 Format: 0 | 1
1257 Default: 0
1258 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1259 the "Enable 0" bit of the configuration register.
1260 Format: 0 | 1
1261 Default: 0
1262 grcan.select= [HW] Select which physical interface to use.
1263 Format: 0 | 1
1264 Default: 0
1265 grcan.txsize= [HW] Sets the size of the tx buffer.
1266 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1267 Default: 1024
1268 grcan.rxsize= [HW] Sets the size of the rx buffer.
1269 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1270 Default: 1024
1271
0f98dd1b
BJZ
1272 gpio-mockup.gpio_mockup_ranges
1273 [HW] Sets the ranges of gpiochip of for this device.
1274 Format: <start1>,<end1>,<start2>,<end2>...
1275
55537871
JK
1276 hardlockup_all_cpu_backtrace=
1277 [KNL] Should the hard-lockup detector generate
1278 backtraces on all cpus.
1279 Format: <integer>
1280
1da177e4
LT
1281 hashdist= [KNL,NUMA] Large hashes allocated during boot
1282 are distributed across NUMA nodes. Defaults on
16290246 1283 for 64-bit NUMA, off otherwise.
a9913044 1284 Format: 0 | 1 (for off | on)
1da177e4
LT
1285
1286 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1287
1288 hd= [EIDE] (E)IDE hard drive subsystem geometry
1289 Format: <cyl>,<head>,<sect>
1290
ea8c071c
HY
1291 hest_disable [ACPI]
1292 Disable Hardware Error Source Table (HEST) support;
1293 corresponding firmware-first mode error processing
1294 logic will be disabled.
1295
1da177e4
LT
1296 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1297 size of <nn>. This works even on boxes that have no
1298 highmem otherwise. This also works to reduce highmem
1299 size on bigger boxes.
1300
54cdfdb4
TG
1301 highres= [KNL] Enable/disable high resolution timer mode.
1302 Valid parameters: "on", "off"
1303 Default: "on"
1304
1da177e4
LT
1305 hisax= [HW,ISDN]
1306 See Documentation/isdn/README.HiSax.
1307
0cb55ad2
RD
1308 hlt [BUGS=ARM,SH]
1309
1310 hpet= [X86-32,HPET] option to control HPET usage
1311 Format: { enable (default) | disable | force |
1312 verbose }
1313 disable: disable HPET and use PIT instead
1314 force: allow force enabled of undocumented chips (ICH4,
1315 VIA, nVidia)
1316 verbose: show contents of HPET registers during setup
1317
3d035f58
PB
1318 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1319 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1320
b4718e62
AK
1321 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1322 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
0d9ea754
JT
1323 On x86-64 and powerpc, this option can be specified
1324 multiple times interleaved with hugepages= to reserve
1325 huge pages of different sizes. Valid pages sizes on
1326 x86-64 are 2M (when the CPU supports "pse") and 1G
27ec26ec 1327 (when the CPU supports the "pdpe1gb" cpuinfo flag).
6902aa84 1328
555d61d6
HB
1329 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1330 terminal devices. Valid values: 0..8
431429ff
HB
1331 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1332 If specified, z/VM IUCV HVC accepts connections
1333 from listed z/VM user IDs only.
cef7125d 1334
fdabf525
JH
1335 hwthread_map= [METAG] Comma-separated list of Linux cpu id to
1336 hardware thread id mappings.
1337 Format: <cpu>:<hwthread>
1338
7bf69395
FDN
1339 keep_bootcon [KNL]
1340 Do not unregister boot console at start. This is only
1341 useful for debugging when something happens in the window
1342 between unregistering the boot console and initializing
1343 the real console.
1344
3a853fb9 1345 i2c_bus= [HW] Override the default board specific I2C bus speed
7954763b
JN
1346 or register an additional I2C bus that is not
1347 registered from board initialization code.
3a853fb9
JN
1348 Format:
1349 <bus_id>,<clkrate>
1350
36d95739 1351 i8042.debug [HW] Toggle i8042 debug mode
e1443d28
SCP
1352 i8042.unmask_kbd_data
1353 [HW] Enable printing of interrupt data from the KBD port
1354 (disabled by default, and as a pre-condition
1355 requires that i8042.debug=1 be enabled)
1da177e4 1356 i8042.direct [HW] Put keyboard port into non-translated mode
84eb8d06
ML
1357 i8042.dumbkbd [HW] Pretend that controller can only read data from
1358 keyboard and cannot control its state
1da177e4
LT
1359 (Don't attempt to blink the leds)
1360 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
945ef0d4 1361 i8042.nokbd [HW] Don't check/create keyboard port
75d08c78
JK
1362 i8042.noloop [HW] Disable the AUX Loopback command while probing
1363 for the AUX port
1da177e4 1364 i8042.nomux [HW] Don't check presence of an active multiplexing
e55a3366 1365 controller
1da177e4
LT
1366 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1367 controllers
24775d65 1368 i8042.notimeout [HW] Ignore timeout condition signalled by controller
930e1924
MPS
1369 i8042.reset [HW] Reset the controller during init, cleanup and
1370 suspend-to-ram transitions, only during s2r
1371 transitions, or never reset
1372 Format: { 1 | Y | y | 0 | N | n }
1373 1, Y, y: always reset controller
1374 0, N, n: don't ever reset controller
1375 Default: only on s2r transitions on x86; most other
1376 architectures force reset to be always executed
1da177e4 1377 i8042.unlock [HW] Unlock (ignore) the keylock
148e9a71 1378 i8042.kbdreset [HW] Reset device connected to KBD port
1da177e4
LT
1379
1380 i810= [HW,DRM]
1381
e70c9d5e
DT
1382 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1383 indicates that the driver is running on unsupported
1384 hardware.
1da177e4
LT
1385 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1386 does not match list of supported models.
1387 i8k.power_status
1388 [HW] Report power status in /proc/i8k
1389 (disabled by default)
1390 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1391 capability is set.
1392
4dca20ef 1393 i915.invert_brightness=
7bd90909
CE
1394 [DRM] Invert the sense of the variable that is used to
1395 set the brightness of the panel backlight. Normally a
4dca20ef
CE
1396 brightness value of 0 indicates backlight switched off,
1397 and the maximum of the brightness value sets the backlight
1398 to maximum brightness. If this parameter is set to 0
1399 (default) and the machine requires it, or this parameter
1400 is set to 1, a brightness value of 0 sets the backlight
1401 to maximum brightness, and the maximum of the brightness
1402 value switches the backlight off.
1403 -1 -- never invert brightness
1404 0 -- machine default
1405 1 -- force brightness inversion
7bd90909 1406
1da177e4
LT
1407 icn= [HW,ISDN]
1408 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1409
0af80c04
DF
1410 ide-core.nodma= [HW] (E)IDE subsystem
1411 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
075affcb
BZ
1412 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1413 .cdrom .chs .ignore_cable are additional options
1c10e938 1414 See Documentation/ide/ide.txt.
1da177e4 1415
0f8b7f5d
MR
1416 ide-generic.probe-mask= [HW] (E)IDE subsystem
1417 Format: <int>
1418 Probe mask for legacy ISA IDE ports. Depending on
1419 platform up to 6 ports are supported, enabled by
1420 setting corresponding bits in the mask to 1. The
1421 default value is 0x0, which has a special meaning.
1422 On systems that have PCI, it triggers scanning the
1423 PCI bus for the first and the second port, which
1424 are then probed. On systems without PCI the value
1425 of 0x0 enables probing the two first ports as if it
1426 was 0x3.
1427
0cb55ad2
RD
1428 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1429 Claim all unknown PCI IDE storage controllers.
1430
f039b754 1431 idle= [X86]
69fb3676 1432 Format: idle=poll, idle=halt, idle=nomwait
ada9cfdd
RD
1433 Poll forces a polling idle loop that can slightly
1434 improve the performance of waking up a idle CPU, but
1435 will use a lot of power and make the system run hot.
1436 Not recommended.
ada9cfdd 1437 idle=halt: Halt is forced to be used for CPU idle.
c1e3b377 1438 In such case C2/C3 won't be used again.
ada9cfdd 1439 idle=nomwait: Disable mwait for CPU C-states
a9913044 1440
503943e0
MR
1441 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1442 Format: { strict | legacy | 2008 | relaxed }
1443 Default: strict
1444
1445 Choose which programs will be accepted for execution
1446 based on the IEEE 754 NaN encoding(s) supported by
1447 the FPU and the NaN encoding requested with the value
1448 of an ELF file header flag individually set by each
1449 binary. Hardware implementations are permitted to
1450 support either or both of the legacy and the 2008 NaN
1451 encoding mode.
1452
1453 Available settings are as follows:
1454 strict accept binaries that request a NaN encoding
1455 supported by the FPU
1456 legacy only accept legacy-NaN binaries, if supported
1457 by the FPU
1458 2008 only accept 2008-NaN binaries, if supported
1459 by the FPU
1460 relaxed accept any binaries regardless of whether
1461 supported by the FPU
1462
1463 The FPU emulator is always able to support both NaN
1464 encodings, so if no FPU hardware is present or it has
1465 been disabled with 'nofpu', then the settings of
1466 'legacy' and '2008' strap the emulator accordingly,
1467 'relaxed' straps the emulator for both legacy-NaN and
1468 2008-NaN, whereas 'strict' enables legacy-NaN only on
1469 legacy processors and both NaN encodings on MIPS32 or
1470 MIPS64 CPUs.
1471
1472 The setting for ABS.fmt/NEG.fmt instruction execution
1473 mode generally follows that for the NaN encoding,
1474 except where unsupported by hardware.
1475
79290822
IM
1476 ignore_loglevel [KNL]
1477 Ignore loglevel setting - this will print /all/
1478 kernel messages to the console. Useful for debugging.
0eca6b7c
YZ
1479 We also add it as printk module parameter, so users
1480 could change it dynamically, usually by
1481 /sys/module/printk/parameters/ignore_loglevel.
79290822 1482
d977d56c
KK
1483 ignore_rlimit_data
1484 Ignore RLIMIT_DATA setting for data mappings,
1485 print warning at first misuse. Can be changed via
1486 /sys/module/kernel/parameters/ignore_rlimit_data.
1487
1da177e4
LT
1488 ihash_entries= [KNL]
1489 Set number of hash buckets for inode cache.
1490
2fe5d6de 1491 ima_appraise= [IMA] appraise integrity measurements
2faa6ef3 1492 Format: { "off" | "enforce" | "fix" | "log" }
2fe5d6de
MZ
1493 default: "enforce"
1494
07f6a794
MZ
1495 ima_appraise_tcb [IMA]
1496 The builtin appraise policy appraises all files
1497 owned by uid=0.
1498
d68a6fe9
MZ
1499 ima_canonical_fmt [IMA]
1500 Use the canonical format for the binary runtime
1501 measurements, instead of host native format.
1502
3323eec9 1503 ima_hash= [IMA]
e7a2ad7e
MZ
1504 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1505 | sha512 | ... }
3323eec9
MZ
1506 default: "sha1"
1507
e7a2ad7e
MZ
1508 The list of supported hash algorithms is defined
1509 in crypto/hash_info.h.
1510
24fd03c8 1511 ima_policy= [IMA]
33ce9549 1512 The builtin policies to load during IMA setup.
503ceaef 1513 Format: "tcb | appraise_tcb | secure_boot"
33ce9549
MZ
1514
1515 The "tcb" policy measures all programs exec'd, files
1516 mmap'd for exec, and all files opened with the read
1517 mode bit set by either the effective uid (euid=0) or
1518 uid=0.
1519
1520 The "appraise_tcb" policy appraises the integrity of
1521 all files owned by root. (This is the equivalent
1522 of ima_appraise_tcb.)
24fd03c8 1523
503ceaef
MZ
1524 The "secure_boot" policy appraises the integrity
1525 of files (eg. kexec kernel image, kernel modules,
1526 firmware, policy, etc) based on file signatures.
24fd03c8
MZ
1527
1528 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
5789ba3b
EP
1529 Load a policy which meets the needs of the Trusted
1530 Computing Base. This means IMA will measure all
1531 programs exec'd, files mmap'd for exec, and all files
1532 opened for read by uid=0.
1533
9b9d4ce5
RS
1534 ima_template= [IMA]
1535 Select one of defined IMA measurements template formats.
8265a2f8 1536 Formats: { "ima" | "ima-ng" | "ima-sig" }
9b9d4ce5
RS
1537 Default: "ima-ng"
1538
c2426d2a
RS
1539 ima_template_fmt=
1540 [IMA] Define a custom template format.
1541 Format: { "field1|...|fieldN" }
1542
3bcced39
DK
1543 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1544 Format: <min_file_size>
1545 Set the minimal file size for using asynchronous hash.
1546 If left unspecified, ahash usage is disabled.
1547
1548 ahash performance varies for different data sizes on
1549 different crypto accelerators. This option can be used
1550 to achieve the best performance for a particular HW.
1551
6edf7a89
DK
1552 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1553 Format: <bufsize>
1554 Set hashing buffer size. Default: 4k.
1555
1556 ahash performance varies for different chunk sizes on
1557 different crypto accelerators. This option can be used
1558 to achieve best performance for particular HW.
1559
1da177e4
LT
1560 init= [KNL]
1561 Format: <full_path>
1562 Run specified binary instead of /sbin/init as init
1563 process.
1564
1565 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1566 for working out where the kernel is dying during
1567 startup.
1568
7b0b73d7
PB
1569 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1570 initcall functions. Useful for debugging built-in
1571 modules and initcalls.
1572
1da177e4
LT
1573 initrd= [BOOT] Specify the location of the initial ramdisk
1574
acd547b2
DH
1575 init_pkru= [x86] Specify the default memory protection keys rights
1576 register contents for all processes. 0x55555554 by
1577 default (disallow access to all but pkey 0). Can
1578 override in debugfs after boot.
1579
1da177e4
LT
1580 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1581 Format: <irq>
1582
6bb2ff84
FY
1583 int_pln_enable [x86] Enable power limit notification interrupt
1584
d726d8d7
MZ
1585 integrity_audit=[IMA]
1586 Format: { "0" | "1" }
1587 0 -- basic integrity auditing messages. (Default)
1588 1 -- additional integrity auditing messages.
1589
ba395927 1590 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
0cd5c3c8
KM
1591 on
1592 Enable intel iommu driver.
ba395927
KA
1593 off
1594 Disable intel iommu driver.
1595 igfx_off [Default Off]
1596 By default, gfx is mapped as normal device. If a gfx
1597 device has a dedicated DMAR unit, the DMAR unit is
1598 bypassed by not enabling DMAR with this option. In
1599 this case, gfx device will use physical address for
1600 DMA.
7d3b03ce
KA
1601 forcedac [x86_64]
1602 With this option iommu will not optimize to look
16290246 1603 for io virtual address below 32-bit forcing dual
7d3b03ce 1604 address cycle on pci bus for cards supporting greater
16290246
RD
1605 than 32-bit addressing. The default is to look
1606 for translation below 32-bit and if not available
7d3b03ce 1607 then look in the higher range.
5e0d2a6f 1608 strict [Default Off]
1609 With this option on every unmap_single operation will
1610 result in a hardware IOTLB flush operation as opposed
1611 to batching them for performance.
6dd9a7c7
YS
1612 sp_off [Default Off]
1613 By default, super page will be supported if Intel IOMMU
1614 has the capability. With this option, super page will
1615 not be supported.
c83b2f20
DW
1616 ecs_off [Default Off]
1617 By default, extended context tables will be supported if
1618 the hardware advertises that it has support both for the
1619 extended tables themselves, and also PASID support. With
1620 this option set, extended tables will not be used even
1621 on hardware which claims to support them.
bfd20f1c
SL
1622 tboot_noforce [Default Off]
1623 Do not force the Intel IOMMU enabled under tboot.
1624 By default, tboot will force Intel IOMMU on, which
1625 could harm performance of some high-throughput
1626 devices like 40GBit network cards, even if identity
1627 mapping is enabled.
1628 Note that using this option lowers the security
1629 provided by tboot because it makes the system
1630 vulnerable to DMA attacks.
2e92c7ad
MI
1631
1632 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1633 0 disables intel_idle and fall back on acpi_idle.
22c6bbe4 1634 1 to 9 specify maximum depth of C-state.
2e92c7ad 1635
6be26498
DB
1636 intel_pstate= [X86]
1637 disable
1638 Do not enable intel_pstate as the default
1639 scaling driver for the supported processors
001c76f0
RW
1640 passive
1641 Use intel_pstate as a scaling driver, but configure it
1642 to work with generic cpufreq governors (instead of
1643 enabling its internal governor). This mode cannot be
1644 used along with the hardware-managed P-states (HWP)
1645 feature.
aa4ea34d
EZ
1646 force
1647 Enable intel_pstate on systems that prohibit it by default
1648 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1649 instead of acpi-cpufreq may disable platform features, such
1650 as thermal controls and power capping, that rely on ACPI
1651 P-States information being indicated to OSPM and therefore
1652 should be used with caution. This option does not work with
1653 processors that aren't supported by the intel_pstate driver
1654 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2f86dc4c
DB
1655 no_hwp
1656 Do not enable hardware P state control (HWP)
1657 if available.
d64c3b0b
KCA
1658 hwp_only
1659 Only load intel_pstate on systems which support
1660 hardware P state control (HWP) if available.
9522a2ff 1661 support_acpi_ppc
2b3ec765
SP
1662 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1663 Description Table, specifies preferred power management
1664 profile as "Enterprise Server" or "Performance Server",
1665 then this feature is turned on by default.
22449c0c
SP
1666 per_cpu_perf_limits
1667 Allow per-logical-CPU P-State performance control limits using
1668 cpufreq sysfs interface
6be26498 1669
d1423d56 1670 intremap= [X86-64, Intel-IOMMU]
d1423d56
CW
1671 on enable Interrupt Remapping (default)
1672 off disable Interrupt Remapping
1673 nosid disable Source ID checking
41750d31
SS
1674 no_x2apic_optout
1675 BIOS x2APIC opt-out request will be ignored
b7d20631 1676 nopost disable Interrupt Posting
d1423d56 1677
0cb55ad2
RD
1678 iomem= Disable strict checking of access to MMIO memory
1679 strict regions from userspace.
1680 relaxed
1681
1682 iommu= [x86]
1683 off
1684 force
1685 noforce
1686 biomerge
1687 panic
1688 nopanic
1689 merge
1690 nomerge
1691 forcesac
1692 soft
bcb71abe 1693 pt [x86, IA-64]
4e287840
TLSC
1694 nobypass [PPC/POWERNV]
1695 Disable IOMMU bypass, using IOMMU for PCI devices.
bcb71abe 1696
fccb4e3b
WD
1697 iommu.passthrough=
1698 [ARM64] Configure DMA to bypass the IOMMU by default.
1699 Format: { "0" | "1" }
1700 0 - Use IOMMU translation for DMA.
1701 1 - Bypass the IOMMU for DMA.
1702 unset - Use IOMMU translation for DMA.
0cb55ad2
RD
1703
1704 io7= [HW] IO7 for Marvel based alpha systems
1705 See comment before marvel_specify_io7 in
1706 arch/alpha/kernel/core_marvel.c.
1707
6cececfc 1708 io_delay= [X86] I/O delay method
6e7c4025
IM
1709 0x80
1710 Standard port 0x80 based delay
1711 0xed
1712 Alternate port 0xed based delay (needed on some systems)
b02aae9c 1713 udelay
6e7c4025
IM
1714 Simple two microseconds delay
1715 none
1716 No delay
b02aae9c 1717
1da177e4 1718 ip= [IP_PNP]
dc7a0816 1719 See Documentation/filesystems/nfs/nfsroot.txt.
1da177e4 1720
fbf19803 1721 irqaffinity= [SMP] Set the default irq affinity mask
2d13e6ca 1722 The argument is a cpu list, as described above.
fbf19803 1723
0962289b
MZ
1724 irqchip.gicv2_force_probe=
1725 [ARM, ARM64]
1726 Format: <bool>
1727 Force the kernel to look for the second 4kB page
1728 of a GICv2 controller even if the memory range
1729 exposed by the device tree is too small.
1730
200803df
AC
1731 irqfixup [HW]
1732 When an interrupt is not handled search all handlers
1733 for it. Intended to get systems with badly broken
1734 firmware running.
1735
1736 irqpoll [HW]
1737 When an interrupt is not handled search all handlers
1738 for it. Also check all handlers each timer
1739 interrupt. Intended to get systems with badly broken
1740 firmware running.
1741
1da177e4 1742 isapnp= [ISAPNP]
a9913044 1743 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1da177e4 1744
d94d1053 1745 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
b0d40d2b
FW
1746 [Deprecated - use cpusets instead]
1747 Format: [flag-list,]<cpu-list>
1748
1749 Specify one or more CPUs to isolate from disturbances
1750 specified in the flag list (default: domain):
1751
1752 nohz
1753 Disable the tick when a single task runs.
1754 domain
1755 Isolate from the general SMP balancing and scheduling
1756 algorithms. Note that performing domain isolation this way
1757 is irreversible: it's not possible to bring back a CPU to
1758 the domains once isolated through isolcpus. It's strongly
1759 advised to use cpusets instead to disable scheduler load
1760 balancing through the "cpuset.sched_load_balance" file.
1761 It offers a much more flexible interface where CPUs can
1762 move in and out of an isolated set anytime.
1763
1764 You can move a process onto or off an "isolated" CPU via
1765 the CPU affinity syscalls or cpuset.
1766 <cpu number> begins at 0 and the maximum value is
1767 "number of CPUs in system - 1".
1768
1769 The format of <cpu-list> is described above.
b225d44e 1770
1da177e4 1771
1da177e4 1772
a9913044 1773 iucv= [HW,NET]
1da177e4 1774
7d8bfa26
JR
1775 ivrs_ioapic [HW,X86_64]
1776 Provide an override to the IOAPIC-ID<->DEVICE-ID
1777 mapping provided in the IVRS ACPI table. For
1778 example, to map IOAPIC-ID decimal 10 to
1779 PCI device 00:14.0 write the parameter as:
1780 ivrs_ioapic[10]=00:14.0
1781
1782 ivrs_hpet [HW,X86_64]
1783 Provide an override to the HPET-ID<->DEVICE-ID
1784 mapping provided in the IVRS ACPI table. For
1785 example, to map HPET-ID decimal 0 to
1786 PCI device 00:14.0 write the parameter as:
1787 ivrs_hpet[0]=00:14.0
1788
ca3bf5d4
SS
1789 ivrs_acpihid [HW,X86_64]
1790 Provide an override to the ACPI-HID:UID<->DEVICE-ID
1791 mapping provided in the IVRS ACPI table. For
1792 example, to map UART-HID:UID AMD0020:0 to
1793 PCI device 00:14.5 write the parameter as:
1794 ivrs_acpihid[00:14.5]=AMD0020:0
1795
1da177e4 1796 js= [HW,JOY] Analog joystick
1752118d 1797 See Documentation/input/joydev/joystick.rst.
1da177e4 1798
65fe935d
KC
1799 nokaslr [KNL]
1800 When CONFIG_RANDOMIZE_BASE is set, this disables
1801 kernel and module base offset ASLR (Address Space
1802 Layout Randomization).
24f2e027 1803
b0845ce5
MR
1804 kasan_multi_shot
1805 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
1806 report on every invalid memory access. Without this
1807 parameter KASAN will print report only for the first
1808 invalid access.
1809
0cb55ad2
RD
1810 keepinitrd [HW,ARM]
1811
342332e6
TI
1812 kernelcore= [KNL,X86,IA-64,PPC]
1813 Format: nn[KMGTPE] | "mirror"
1814 This parameter
ed7ed365
MG
1815 specifies the amount of memory usable by the kernel
1816 for non-movable allocations. The requested amount is
1817 spread evenly throughout all nodes in the system. The
1818 remaining memory in each node is used for Movable
1819 pages. In the event, a node is too small to have both
1820 kernelcore and Movable pages, kernelcore pages will
1821 take priority and other nodes will have a larger number
675217fd 1822 of Movable pages. The Movable zone is used for the
ed7ed365
MG
1823 allocation of pages that may be reclaimed or moved
1824 by the page migration subsystem. This means that
1825 HugeTLB pages may not be allocated from this zone.
1826 Note that allocations like PTEs-from-HighMem still
1827 use the HighMem zone if it exists, and the Normal
1828 zone if it does not.
1829
342332e6
TI
1830 Instead of specifying the amount of memory (nn[KMGTPE]),
1831 you can specify "mirror" option. In case "mirror"
1832 option is specified, mirrored (reliable) memory is used
1833 for non-movable allocations and remaining memory is used
1834 for Movable pages. nn[KMGTPE] and "mirror" are exclusive,
1835 so you can NOT specify nn[KMGTPE] and "mirror" at the same
1836 time.
ed7ed365 1837
4fe1da4e
JW
1838 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
1839 Format: <Controller#>[,poll interval]
1840 The controller # is the number of the ehci usb debug
1841 port as it is probed via PCI. The poll interval is
1842 optional and is the number seconds in between
1843 each poll cycle to the debug port in case you need
1844 the functionality for interrupting the kernel with
1845 gdb or control-c on the dbgp connection. When
1846 not using this parameter you use sysrq-g to break into
1847 the kernel debugger.
1848
84c08fd6 1849 kgdboc= [KGDB,HW] kgdb over consoles.
ada64e4c
JW
1850 Requires a tty driver that supports console polling,
1851 or a supported polling keyboard driver (non-usb).
65b5ac14
JW
1852 Serial only format: <serial_device>[,baud]
1853 keyboard only format: kbd
1854 keyboard and serial format: kbd,<serial_device>[,baud]
1855 Optional Kernel mode setting:
1856 kms, kbd format: kms,kbd
1857 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
6cdf6e06 1858
84c08fd6
JW
1859 kgdbwait [KGDB] Stop kernel execution and enter the
1860 kernel debugger at the earliest opportunity.
1861
9bed90c6
FF
1862 kmac= [MIPS] korina ethernet MAC address.
1863 Configure the RouterBoard 532 series on-chip
1864 Ethernet adapter MAC address.
1865
04f70336
CM
1866 kmemleak= [KNL] Boot-time kmemleak enable/disable
1867 Valid arguments: on, off
1868 Default: on
47aeeddc
MI
1869 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
1870 the default is off.
04f70336 1871
fef07aae
AP
1872 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
1873 Default is 0 (don't ignore, but inject #GP)
1874
a182d873
XG
1875 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
1876 KVM MMU at runtime.
fef07aae
AP
1877 Default is 0 (off)
1878
fef07aae 1879 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
8475f94a 1880 Default is 1 (enabled)
fef07aae
AP
1881
1882 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
1883 for all guests.
16290246 1884 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
fef07aae 1885
e23f62f7
MZ
1886 kvm-arm.vgic_v3_group0_trap=
1887 [KVM,ARM] Trap guest accesses to GICv3 group-0
1888 system registers
1889
182936ee
MZ
1890 kvm-arm.vgic_v3_group1_trap=
1891 [KVM,ARM] Trap guest accesses to GICv3 group-1
1892 system registers
1893
ff89511e
MZ
1894 kvm-arm.vgic_v3_common_trap=
1895 [KVM,ARM] Trap guest accesses to GICv3 common
1896 system registers
1897
a7546054
MZ
1898 kvm-arm.vgic_v4_enable=
1899 [KVM,ARM] Allow use of GICv4 for direct injection of
1900 LPIs.
1901
fef07aae
AP
1902 kvm-intel.ept= [KVM,Intel] Disable extended page tables
1903 (virtualized MMU) support on capable Intel chips.
1904 Default is 1 (enabled)
1905
1906 kvm-intel.emulate_invalid_guest_state=
1907 [KVM,Intel] Enable emulation of invalid guest states
1908 Default is 0 (disabled)
1909
1910 kvm-intel.flexpriority=
1911 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
1912 Default is 1 (enabled)
1913
e1a72ae2
SL
1914 kvm-intel.nested=
1915 [KVM,Intel] Enable VMX nesting (nVMX).
1916 Default is 0 (disabled)
1917
fef07aae
AP
1918 kvm-intel.unrestricted_guest=
1919 [KVM,Intel] Disable unrestricted guest feature
1920 (virtualized real and unpaged mode) on capable
1921 Intel chips. Default is 1 (enabled)
1922
1749555e
KRW
1923 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
1924 CVE-2018-3620.
1925
1926 Valid arguments: never, cond, always
1927
1928 always: L1D cache flush on every VMENTER.
1929 cond: Flush L1D on VMENTER only when the code between
1930 VMEXIT and VMENTER can leak host memory.
1931 never: Disables the mitigation
1932
1933 Default is cond (do L1 cache flush in specific instances)
1934
fef07aae
AP
1935 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
1936 feature (tagged TLBs) on capable Intel chips.
1937 Default is 1 (enabled)
1938
24fcb53c
JK
1939 l1tf= [X86] Control mitigation of the L1TF vulnerability on
1940 affected CPUs
1941
1942 The kernel PTE inversion protection is unconditionally
1943 enabled and cannot be disabled.
1944
1945 full
1946 Provides all available mitigations for the
1947 L1TF vulnerability. Disables SMT and
1948 enables all mitigations in the
1949 hypervisors, i.e. unconditional L1D flush.
1950
1951 SMT control and L1D flush control via the
1952 sysfs interface is still possible after
1953 boot. Hypervisors will issue a warning
1954 when the first VM is started in a
1955 potentially insecure configuration,
1956 i.e. SMT enabled or L1D flush disabled.
1957
1958 full,force
1959 Same as 'full', but disables SMT and L1D
1960 flush runtime control. Implies the
1961 'nosmt=force' command line option.
1962 (i.e. sysfs control of SMT is disabled.)
1963
1964 flush
1965 Leaves SMT enabled and enables the default
1966 hypervisor mitigation, i.e. conditional
1967 L1D flush.
1968
1969 SMT control and L1D flush control via the
1970 sysfs interface is still possible after
1971 boot. Hypervisors will issue a warning
1972 when the first VM is started in a
1973 potentially insecure configuration,
1974 i.e. SMT enabled or L1D flush disabled.
1975
1976 flush,nosmt
1977
1978 Disables SMT and enables the default
1979 hypervisor mitigation.
1980
1981 SMT control and L1D flush control via the
1982 sysfs interface is still possible after
1983 boot. Hypervisors will issue a warning
1984 when the first VM is started in a
1985 potentially insecure configuration,
1986 i.e. SMT enabled or L1D flush disabled.
1987
1988 flush,nowarn
1989 Same as 'flush', but hypervisors will not
1990 warn when a VM is started in a potentially
1991 insecure configuration.
1992
1993 off
1994 Disables hypervisor mitigations and doesn't
1995 emit any warnings.
1996
1997 Default is 'flush'.
1998
1999 For details see: Documentation/admin-guide/l1tf.rst
2000
1da177e4
LT
2001 l2cr= [PPC]
2002
a78bfbfc
RB
2003 l3cr= [PPC]
2004
cd4f0ef7 2005 lapic [X86-32,APIC] Enable the local APIC even if BIOS
a9913044 2006 disabled it.
1da177e4 2007
279f1461
SS
2008 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
2009 value for LAPIC timer one-shot implementation. Default
2010 back to the programmable timer unit in the LAPIC.
2011
6cececfc 2012 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
ada9cfdd 2013 in C2 power state.
e585bef8 2014
fcb71f6f
FC
2015 libata.dma= [LIBATA] DMA control
2016 libata.dma=0 Disable all PATA and SATA DMA
2017 libata.dma=1 PATA and SATA Disk DMA only
2018 libata.dma=2 ATAPI (CDROM) DMA only
16290246 2019 libata.dma=4 Compact Flash DMA only
fcb71f6f
FC
2020 Combinations also work, so libata.dma=3 enables DMA
2021 for disks and CDROMs, but not CFs.
16290246 2022
20308871
MP
2023 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2024 libata.ignore_hpa=0 keep BIOS limits (default)
2025 libata.ignore_hpa=1 ignore limits, using full disk
fcb71f6f 2026
78e70c23
DJ
2027 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2028 when set.
2029 Format: <int>
2030
33267325
TH
2031 libata.force= [LIBATA] Force configurations. The format is comma
2032 separated list of "[ID:]VAL" where ID is
4c44f309 2033 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
33267325
TH
2034 matching port, link or device. Basically, it matches
2035 the ATA ID string printed on console by libata. If
2036 the whole ID part is omitted, the last PORT and DEVICE
2037 values are used. If ID hasn't been specified yet, the
2038 configuration applies to all ports, links and devices.
2039
2040 If only DEVICE is omitted, the parameter applies to
2041 the port and all links and devices behind it. DEVICE
2042 number of 0 either selects the first device or the
2043 first fan-out link behind PMP device. It does not
2044 select the host link. DEVICE number of 15 selects the
2045 host link and device attached to it.
2046
2047 The VAL specifies the configuration to force. As long
2048 as there's no ambiguity shortcut notation is allowed.
2049 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2050 The following configurations can be forced.
2051
2052 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2053 Any ID with matching PORT is used.
2054
2055 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2056
2057 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2058 udma[/][16,25,33,44,66,100,133] notation is also
2059 allowed.
2060
2061 * [no]ncq: Turn on or off NCQ.
2062
d7b16e4f
MP
2063 * [no]ncqtrim: Turn off queued DSM TRIM.
2064
05944bdf
TH
2065 * nohrst, nosrst, norst: suppress hard, soft
2066 and both resets.
2067
ca6d43b0
DW
2068 * rstonce: only attempt one reset during
2069 hot-unplug link recovery
2070
43c9c591
TH
2071 * dump_id: dump IDENTIFY data.
2072
966fbe19
VP
2073 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2074
b8bd6dc3
RJ
2075 * disable: Disable this device.
2076
33267325
TH
2077 If there are multiple matching configurations changing
2078 the same attribute, the last one is used.
2079
95f72d1e 2080 memblock=debug [KNL] Enable memblock debug messages.
7c4be253 2081
1da177e4 2082 load_ramdisk= [RAM] List of ramdisks to load from floppy
31c00fc1 2083 See Documentation/blockdev/ramdisk.txt.
1da177e4 2084
a6b25b67
RD
2085 lockd.nlm_grace_period=P [NFS] Assign grace period.
2086 Format: <integer>
1da177e4 2087
a6b25b67
RD
2088 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2089 Format: <integer>
2090
2091 lockd.nlm_timeout=T [NFS] Assign timeout value.
2092 Format: <integer>
2093
2094 lockd.nlm_udpport=M [NFS] Assign UDP port.
2095 Format: <integer>
1da177e4 2096
ec4518aa
PM
2097 locktorture.nreaders_stress= [KNL]
2098 Set the number of locking read-acquisition kthreads.
2099 Defaults to being automatically set based on the
2100 number of online CPUs.
2101
2102 locktorture.nwriters_stress= [KNL]
2103 Set the number of locking write-acquisition kthreads.
2104
2105 locktorture.onoff_holdoff= [KNL]
2106 Set time (s) after boot for CPU-hotplug testing.
2107
2108 locktorture.onoff_interval= [KNL]
2109 Set time (s) between CPU-hotplug operations, or
2110 zero to disable CPU-hotplug testing.
2111
2112 locktorture.shuffle_interval= [KNL]
2113 Set task-shuffle interval (jiffies). Shuffling
2114 tasks allows some CPUs to go into dyntick-idle
2115 mode during the locktorture test.
2116
2117 locktorture.shutdown_secs= [KNL]
2118 Set time (s) after boot system shutdown. This
2119 is useful for hands-off automated testing.
2120
2121 locktorture.stat_interval= [KNL]
2122 Time (s) between statistics printk()s.
2123
2124 locktorture.stutter= [KNL]
2125 Time (s) to stutter testing, for example,
2126 specifying five seconds causes the test to run for
2127 five seconds, wait for five seconds, and so on.
2128 This tests the locking primitive's ability to
2129 transition abruptly to and from idle.
2130
2131 locktorture.torture_runnable= [BOOT]
2132 Start locktorture running at boot time.
2133
2134 locktorture.torture_type= [KNL]
2135 Specify the locking implementation to test.
2136
2137 locktorture.verbose= [KNL]
2138 Enable additional printk() statements.
2139
1da177e4
LT
2140 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2141 Format: <irq>
2142
2143 loglevel= All Kernel Messages with a loglevel smaller than the
2144 console loglevel will be printed to the console. It can
2145 also be changed with klogd or other programs. The
2146 loglevels are defined as follows:
2147
2148 0 (KERN_EMERG) system is unusable
2149 1 (KERN_ALERT) action must be taken immediately
2150 2 (KERN_CRIT) critical conditions
2151 3 (KERN_ERR) error conditions
2152 4 (KERN_WARNING) warning conditions
2153 5 (KERN_NOTICE) normal but significant condition
2154 6 (KERN_INFO) informational
2155 7 (KERN_DEBUG) debug-level messages
2156
c756d08a 2157 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
23b2899f
LR
2158 in bytes. n must be a power of two and greater
2159 than the minimal size. The minimal size is defined
2160 by LOG_BUF_SHIFT kernel config parameter. There is
2161 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2162 that allows to increase the default size depending on
2163 the number of CPUs. See init/Kconfig for more details.
1da177e4 2164
accaa24c
RD
2165 logo.nologo [FB] Disables display of the built-in Linux logo.
2166 This may be used to provide more screen space for
2167 kernel log messages and is useful when debugging
2168 kernel boot problems.
2169
1da177e4
LT
2170 lp=0 [LP] Specify parallel ports to use, e.g,
2171 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2172 lp=reset first parallel port). 'lp=0' disables the
2173 lp=auto printer driver. 'lp=reset' (which can be
2174 specified in addition to the ports) causes
2175 attached printers to be reset. Using
2176 lp=port1,port2,... specifies the parallel ports
2177 to associate lp devices with, starting with
2178 lp0. A port specification may be 'none' to skip
2179 that lp device, or a parport name such as
2180 'parport0'. Specifying 'lp=auto' instead of a
2181 port specification list means that device IDs
2182 from each port should be examined, to see if
2183 an IEEE 1284-compliant printer is attached; if
2184 so, the driver will manage that printer.
2185 See also header of drivers/char/lp.c.
2186
2187 lpj=n [KNL]
2188 Sets loops_per_jiffy to given constant, thus avoiding
2189 time-consuming boot-time autodetection (up to 250 ms per
2190 CPU). 0 enables autodetection (default). To determine
2191 the correct value for your kernel, boot with normal
2192 autodetection and see what value is printed. Note that
2193 on SMP systems the preset will be applied to all CPUs,
2194 which is likely to cause problems if your CPUs need
2195 significantly divergent settings. An incorrect value
2196 will cause delays in the kernel to be wrong, leading to
2197 unpredictable I/O errors and other breakage. Although
2198 unlikely, in the extreme case this might damage your
2199 hardware.
2200
2201 ltpc= [NET]
2202 Format: <io>,<irq>,<dma>
2203
16290246 2204 machvec= [IA-64] Force the use of a particular machine-vector
a9913044
RD
2205 (machvec) in a generic kernel.
2206 Example: machvec=hpzx1_swiotlb
1da177e4 2207
3209e70e
WZ
2208 machtype= [Loongson] Share the same kernel image file between different
2209 yeeloong laptop.
2210 Example: machtype=lemote-yeeloong-2f-7inch
2211
0cb55ad2
RD
2212 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2213 than or equal to this physical address is ignored.
1da177e4
LT
2214
2215 maxcpus= [SMP] Maximum number of processors that an SMP kernel
7c142bfe
BH
2216 will bring up during bootup. maxcpus=n : n >= 0 limits
2217 the kernel to bring up 'n' processors. Surely after
2218 bootup you can bring up the other plugged cpu by executing
2219 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2220 only takes effect during system bootup.
2221 While n=0 is a special case, it is equivalent to "nosmp",
2222 which also disables the IO APIC.
1da177e4 2223
d134b00b
KS
2224 max_loop= [LOOP] The number of loop block devices that get
2225 (loop.max_loop) unconditionally pre-created at init time. The default
2226 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2227 of statically allocating a predefined number, loop
2228 devices can be requested on-demand with the
2229 /dev/loop-control interface.
2b2c3750 2230
cd4f0ef7 2231 mce [X86-32] Machine Check Exception
1da177e4 2232
71cced6e 2233 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.txt
909dd324 2234
1da177e4 2235 md= [HW] RAID subsystems devices and level
e52347bd 2236 See Documentation/admin-guide/md.rst.
a9913044 2237
1da177e4
LT
2238 mdacon= [MDA]
2239 Format: <first>,<last>
2240 Specifies range of consoles to be captured by the MDA.
a9913044 2241
1da177e4
LT
2242 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2243 Amount of memory to be used when the kernel is not able
2244 to see the whole system memory or for test.
fbb97d87
WC
2245 [X86] Work as limiting max address. Use together
2246 with memmap= to avoid physical address space collisions.
2247 Without memmap= PCI devices could be placed at addresses
2248 belonging to unused RAM.
1da177e4 2249
cd4f0ef7 2250 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
1da177e4
LT
2251 memory.
2252
6902aa84
PM
2253 memchunk=nn[KMG]
2254 [KNL,SH] Allow user to override the default size for
2255 per-device physically contiguous DMA buffers.
2256
86dd995d
VK
2257 memhp_default_state=online/offline
2258 [KNL] Set the initial state for the memory hotplug
2259 onlining policy. If not specified, the default value is
2260 set according to the
2261 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2262 option.
2263 See Documentation/memory-hotplug.txt.
2264
6cececfc 2265 memmap=exactmap [KNL,X86] Enable setting of an exact
1da177e4
LT
2266 E820 memory map, as specified by the user.
2267 Such memmap=exactmap lines can be constructed based on
2268 BIOS output or other requirements. See the memmap=nn@ss
2269 option description.
2270
2271 memmap=nn[KMG]@ss[KMG]
277cba1d
RD
2272 [KNL] Force usage of a specific region of memory.
2273 Region of memory to be used is from ss to ss+nn.
8fcc9bc3
BH
2274 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2275 which limits max address to nn[KMG].
2276 Multiple different regions can be specified,
2277 comma delimited.
2278 Example:
2279 memmap=100M@2G,100M#3G,1G!1024G
1da177e4
LT
2280
2281 memmap=nn[KMG]#ss[KMG]
2282 [KNL,ACPI] Mark specific memory as ACPI data.
277cba1d 2283 Region of memory to be marked is from ss to ss+nn.
1da177e4
LT
2284
2285 memmap=nn[KMG]$ss[KMG]
2286 [KNL,ACPI] Mark specific memory as reserved.
277cba1d 2287 Region of memory to be reserved is from ss to ss+nn.
1312848e
PM
2288 Example: Exclude memory from 0x18690000-0x1869ffff
2289 memmap=64K$0x18690000
2290 or
2291 memmap=0x10000$0x18690000
8fcc9bc3
BH
2292 Some bootloaders may need an escape character before '$',
2293 like Grub2, otherwise '$' and the following number
2294 will be eaten.
1da177e4 2295
ec776ef6
CH
2296 memmap=nn[KMG]!ss[KMG]
2297 [KNL,X86] Mark specific memory as protected.
2298 Region of memory to be used, from ss to ss+nn.
2299 The memory region may be marked as e820 type 12 (0xc)
2300 and is NVDIMM or ADR memory.
2301
9f077871
JF
2302 memory_corruption_check=0/1 [X86]
2303 Some BIOSes seem to corrupt the first 64k of
2304 memory when doing things like suspend/resume.
2305 Setting this option will scan the memory
2306 looking for corruption. Enabling this will
2307 both detect corruption and prevent the kernel
2308 from using the memory being corrupted.
2309 However, its intended as a diagnostic tool; if
2310 repeatable BIOS-originated corruption always
2311 affects the same memory, you can use memmap=
2312 to prevent the kernel from using that memory.
2313
2314 memory_corruption_check_size=size [X86]
2315 By default it checks for corruption in the low
2316 64k, making this memory unavailable for normal
2317 use. Use this parameter to scan for
2318 corruption in more or less memory.
2319
2320 memory_corruption_check_period=seconds [X86]
2321 By default it checks for corruption every 60
2322 seconds. Use this parameter to check at some
2323 other rate. 0 disables periodic checking.
2324
e4b0db72 2325 memtest= [KNL,X86,ARM] Enable memtest
c64df707 2326 Format: <integer>
c64df707 2327 default : 0 <disable>
9e5f6cf5
AH
2328 Specifies the number of memtest passes to be
2329 performed. Each pass selects another test
2330 pattern from a given set of patterns. Memtest
2331 fills the memory with this pattern, validates
2332 memory contents and reserves bad memory
2333 regions that are detected.
c64df707 2334
c262f3b9
TL
2335 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2336 Valid arguments: on, off
2337 Default (depends on kernel configuration option):
2338 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2339 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2340 mem_encrypt=on: Activate SME
2341 mem_encrypt=off: Do not activate SME
2342
2343 Refer to Documentation/x86/amd-memory-encryption.txt
2344 for details on when memory encryption can be activated.
2345
406e7938
RW
2346 mem_sleep_default= [SUSPEND] Default system suspend mode:
2347 s2idle - Suspend-To-Idle
2348 shallow - Power-On Suspend or equivalent (if supported)
2349 deep - Suspend-To-RAM or equivalent (if supported)
58e7cb9e 2350 See Documentation/admin-guide/pm/sleep-states.rst.
406e7938 2351
1da177e4 2352 meye.*= [HW] Set MotionEye Camera parameters
a405ed85 2353 See Documentation/media/v4l-drivers/meye.rst.
1da177e4 2354
8f36881b
AS
2355 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2356 Multi-Function General Purpose Timers on AMD Geode
2357 platforms.
2358
e6c4dc6c
WT
2359 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2360 the BIOS has incorrectly applied a workaround. TinyBIOS
2361 version 0.98 is known to be affected, 0.99 fixes the
2362 problem by letting the user disable the workaround.
2363
1da177e4
LT
2364 mga= [HW,DRM]
2365
1c207f95
RD
2366 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2367 physical address is ignored.
2368
39f45d7b
MP
2369 mini2440= [ARM,HW,KNL]
2370 Format:[0..2][b][c][t]
2371 Default: "0tb"
2372 MINI2440 configuration specification:
2373 0 - The attached screen is the 3.5" TFT
2374 1 - The attached screen is the 7" TFT
2375 2 - The VGA Shield is attached (1024x768)
2376 Leaving out the screen size parameter will not load
2377 the TFT driver, and the framebuffer will be left
2378 unconfigured.
2379 b - Enable backlight. The TFT backlight pin will be
2380 linked to the kernel VESA blanking code and a GPIO
2381 LED. This parameter is not necessary when using the
2382 VGA shield.
2383 c - Enable the s3c camera interface.
2384 t - Reserved for enabling touchscreen support. The
2385 touchscreen support is not enabled in the mainstream
2386 kernel as of 2.6.30, a preliminary port can be found
2387 in the "bleeding edge" mini2440 support kernel at
2388 http://repo.or.cz/w/linux-2.6/mini2440.git
2389
6b74ab97
MG
2390 mminit_loglevel=
2391 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2392 parameter allows control of the logging verbosity for
2393 the additional memory initialisation checks. A value
2394 of 0 disables mminit logging and a level of 4 will
2395 log everything. Information is printed at KERN_DEBUG
2396 so loglevel=8 may also need to be specified.
2397
106a4ee2
RR
2398 module.sig_enforce
2399 [KNL] When CONFIG_MODULE_SIG is set, this means that
2400 modules without (valid) signatures will fail to load.
2a039be7 2401 Note that if CONFIG_MODULE_SIG_FORCE is set, that
106a4ee2
RR
2402 is always true, so this option does nothing.
2403
be7de5f9
PB
2404 module_blacklist= [KNL] Do not load a comma-separated list of
2405 modules. Useful for debugging problem modules.
2406
1da177e4
LT
2407 mousedev.tap_time=
2408 [MOUSE] Maximum time between finger touching and
2409 leaving touchpad surface for touch to be considered
2410 a tap and be reported as a left button click (for
2411 touchpads working in absolute mode only).
2412 Format: <msecs>
2413 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2414 reporting absolute coordinates, such as tablets
2415 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2416 reporting absolute coordinates, such as tablets
2417
6cececfc 2418 movablecore=nn[KMG] [KNL,X86,IA-64,PPC] This parameter
0cb55ad2
RD
2419 is similar to kernelcore except it specifies the
2420 amount of memory used for migratable allocations.
2421 If both kernelcore and movablecore is specified,
2422 then kernelcore will be at *least* the specified
2423 value but may be more. If movablecore on its own
2424 is specified, the administrator must be careful
2425 that the amount of memory usable for all allocations
2426 is not too small.
2427
f70029bb
MH
2428 movable_node [KNL] Boot-time switch to make hotplugable memory
2429 NUMA nodes to be movable. This means that the memory
2430 of such nodes will be usable only for movable
2431 allocations which rules out almost all kernel
2432 allocations. Use with caution!
c5320926 2433
1da177e4
LT
2434 MTD_Partition= [MTD]
2435 Format: <name>,<region-number>,<size>,<offset>
2436
a9913044
RD
2437 MTD_Region= [MTD] Format:
2438 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
1da177e4
LT
2439
2440 mtdparts= [MTD]
c8facbb6 2441 See drivers/mtd/cmdlinepart.c.
1da177e4 2442
4e89a2d8
WS
2443 multitce=off [PPC] This parameter disables the use of the pSeries
2444 firmware feature for updating multiple TCE entries
2445 at a time.
2446
5988af23
RH
2447 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2448
2449 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2450
2451 boundary - index of last SLC block on Flex-OneNAND.
2452 The remaining blocks are configured as MLC blocks.
2453 lock - Configure if Flex-OneNAND boundary should be locked.
2454 Once locked, the boundary cannot be changed.
2455 1 indicates lock status, 0 indicates unlock status.
2456
9db829f4
BD
2457 mtdset= [ARM]
2458 ARM/S3C2412 JIVE boot control
2459
2460 See arch/arm/mach-s3c2412/mach-jive.c
2461
1da177e4 2462 mtouchusb.raw_coordinates=
a9913044
RD
2463 [HW] Make the MicroTouch USB driver use raw coordinates
2464 ('y', default) or cooked coordinates ('n')
1da177e4 2465
0cb55ad2 2466 mtrr_chunk_size=nn[KMG] [X86]
19f59460 2467 used for mtrr cleanup. It is largest continuous chunk
0cb55ad2
RD
2468 that could hold holes aka. UC entries.
2469
2470 mtrr_gran_size=nn[KMG] [X86]
2471 Used for mtrr cleanup. It is granularity of mtrr block.
2472 Default is 1.
2473 Large value could prevent small alignment from
2474 using up MTRRs.
2475
2476 mtrr_spare_reg_nr=n [X86]
2477 Format: <integer>
2478 Range: 0,7 : spare reg number
2479 Default : 1
2480 Used for mtrr cleanup. It is spare mtrr entries number.
2481 Set to 2 or more if your graphical card needs more.
2482
1da177e4
LT
2483 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2484
1da177e4
LT
2485 netdev= [NET] Network devices parameters
2486 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2487 Note that mem_start is often overloaded to mean
2488 something different and driver-specific.
a9913044
RD
2489 This usage is only documented in each driver source
2490 file if at all.
2491
58401572
KPO
2492 nf_conntrack.acct=
2493 [NETFILTER] Enable connection tracking flow accounting
2494 0 to disable accounting
2495 1 to enable accounting
d70a011d 2496 Default value is 0.
58401572 2497
306a0753 2498 nfsaddrs= [NFS] Deprecated. Use ip= instead.
dc7a0816 2499 See Documentation/filesystems/nfs/nfsroot.txt.
1da177e4
LT
2500
2501 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
dc7a0816 2502 See Documentation/filesystems/nfs/nfsroot.txt.
1da177e4 2503
306a0753
CL
2504 nfsrootdebug [NFS] enable nfsroot debugging messages.
2505 See Documentation/filesystems/nfs/nfsroot.txt.
2506
5405fc44
TM
2507 nfs.callback_nr_threads=
2508 [NFSv4] set the total number of threads that the
2509 NFS client will assign to service NFSv4 callback
2510 requests.
2511
a72b4422
TM
2512 nfs.callback_tcpport=
2513 [NFS] set the TCP port on which the NFSv4 callback
2514 channel should listen.
2515
e571cbf1
TM
2516 nfs.cache_getent=
2517 [NFS] sets the pathname to the program which is used
2518 to update the NFS client cache entries.
2519
2520 nfs.cache_getent_timeout=
2521 [NFS] sets the timeout after which an attempt to
2522 update a cache entry is deemed to have failed.
2523
58df095b
TM
2524 nfs.idmap_cache_timeout=
2525 [NFS] set the maximum lifetime for idmapper cache
2526 entries.
2527
f43bf0be
TM
2528 nfs.enable_ino64=
2529 [NFS] enable 64-bit inode numbers.
2530 If zero, the NFS client will fake up a 32-bit inode
2531 number for the readdir() and stat() syscalls instead
2532 of returning the full 64-bit number.
2533 The default is to return 64-bit inode numbers.
2534
5405fc44
TM
2535 nfs.max_session_cb_slots=
2536 [NFSv4.1] Sets the maximum number of session
2537 slots the client will assign to the callback
2538 channel. This determines the maximum number of
2539 callbacks the client will process in parallel for
2540 a particular server.
2541
ef159e91
TM
2542 nfs.max_session_slots=
2543 [NFSv4.1] Sets the maximum number of session slots
2544 the client will attempt to negotiate with the server.
2545 This limits the number of simultaneous RPC requests
2546 that the client can send to the NFSv4.1 server.
2547 Note that there is little point in setting this
2548 value higher than the max_tcp_slot_table_limit.
2549
b064eca2 2550 nfs.nfs4_disable_idmapping=
074b1d12
TM
2551 [NFSv4] When set to the default of '1', this option
2552 ensures that both the RPC level authentication
2553 scheme and the NFS level operations agree to use
2554 numeric uids/gids if the mount is using the
2555 'sec=sys' security flavour. In effect it is
2556 disabling idmapping, which can make migration from
2557 legacy NFSv2/v3 systems to NFSv4 easier.
2558 Servers that do not support this mode of operation
2559 will be autodetected by the client, and it will fall
2560 back to using the idmapper.
2561 To turn off this behaviour, set the value to '0'.
6f2ea7f2
CL
2562 nfs.nfs4_unique_id=
2563 [NFS4] Specify an additional fixed unique ident-
2564 ification string that NFSv4 clients can insert into
2565 their nfs_client_id4 string. This is typically a
2566 UUID that is generated at system install time.
b064eca2 2567
db8ac8ba
WAA
2568 nfs.send_implementation_id =
2569 [NFSv4.1] Send client implementation identification
2570 information in exchange_id requests.
2571 If zero, no implementation identification information
2572 will be sent.
2573 The default is to send the implementation identification
2574 information.
e52347bd 2575
f6de7a39
TM
2576 nfs.recover_lost_locks =
2577 [NFSv4] Attempt to recover locks that were lost due
2578 to a lease timeout on the server. Please note that
2579 doing this risks data corruption, since there are
2580 no guarantees that the file will remain unchanged
2581 after the locks are lost.
2582 If you want to enable the kernel legacy behaviour of
2583 attempting to recover these locks, then set this
2584 parameter to '1'.
2585 The default parameter value of '0' causes the kernel
2586 not to attempt recovery of lost locks.
db8ac8ba 2587
bbf58bf3
TM
2588 nfs4.layoutstats_timer =
2589 [NFSv4.2] Change the rate at which the kernel sends
2590 layoutstats to the pNFS metadata server.
2591
2592 Setting this to value to 0 causes the kernel to use
2593 whatever value is the default set by the layout
2594 driver. A non-zero value sets the minimum interval
2595 in seconds between layoutstats transmissions.
2596
e9541ce8
BF
2597 nfsd.nfs4_disable_idmapping=
2598 [NFSv4] When set to the default of '1', the NFSv4
2599 server will return only numeric uids and gids to
2600 clients using auth_sys, and will accept numeric uids
2601 and gids from such clients. This is intended to ease
2602 migration from NFSv2/v3.
db8ac8ba 2603
c0c74acb 2604 nmi_debug= [KNL,SH] Specify one or more actions to take
e7ba176b
HS
2605 when a NMI is triggered.
2606 Format: [state][,regs][,debounce][,die]
2607
6cececfc 2608 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
fef2c9bc 2609 Format: [panic,][nopanic,][num]
195daf66 2610 Valid num: 0 or 1
334bb79c
PK
2611 0 - turn hardlockup detector in nmi_watchdog off
2612 1 - turn hardlockup detector in nmi_watchdog on
0cb55ad2 2613 When panic is specified, panic when an NMI watchdog
fef2c9bc 2614 timeout occurs (or 'nopanic' to override the opposite
334bb79c
PK
2615 default). To disable both hard and soft lockup detectors,
2616 please see 'nowatchdog'.
0cb55ad2
RD
2617 This is useful when you use a panic=... timeout and
2618 need the box quickly up again.
1da177e4 2619
bff38771
AV
2620 netpoll.carrier_timeout=
2621 [NET] Specifies amount of time (in seconds) that
2622 netpoll should wait for a carrier. By default netpoll
2623 waits 4 seconds.
2624
cd4f0ef7 2625 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
1da177e4
LT
2626 emulation library even if a 387 maths coprocessor
2627 is present.
2628
0cb55ad2
RD
2629 no_console_suspend
2630 [HW] Never suspend the console
2631 Disable suspending of consoles during suspend and
2632 hibernate operations. Once disabled, debugging
2633 messages can reach various consoles while the rest
2634 of the system is being put to sleep (ie, while
2635 debugging driver suspend/resume hooks). This may
2636 not work reliably with all consoles, but is known
2637 to work with serial and VGA consoles.
134620f7
YZ
2638 To facilitate more flexible debugging, we also add
2639 console_suspend, a printk module parameter to control
2640 it. Users could use console_suspend (usually
2641 /sys/module/printk/parameters/console_suspend) to
2642 turn on/off it dynamically.
0cb55ad2 2643
c1aee215
CL
2644 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
2645 caches in the slab allocator. Saves per-node memory,
2646 but will impact performance.
3395ee05 2647
a9913044
RD
2648 noalign [KNL,ARM]
2649
686140a1
VG
2650 noaltinstr [S390] Disables alternative instructions patching
2651 (CPU alternatives feature).
2652
1da177e4
LT
2653 noapic [SMP,APIC] Tells the kernel to not make use of any
2654 IOAPICs that may be present in the system.
2655
5091faa4
MG
2656 noautogroup Disable scheduler automatic task group creation.
2657
1da177e4
LT
2658 nobats [PPC] Do not use BATs for mapping kernel lowmem
2659 on "Classic" PPC cores.
2660
2661 nocache [ARM]
a9913044 2662
0cb55ad2
RD
2663 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
2664
163ecdff
SN
2665 nodelayacct [KNL] Disable per-task delay accounting
2666
6902aa84
PM
2667 nodsp [SH] Disable hardware DSP at boot time.
2668
b2e0a54a 2669 noefi Disable EFI runtime services support.
8b2cb7a8 2670
1da177e4
LT
2671 noexec [IA-64]
2672
6cececfc 2673 noexec [X86]
f5a1b191 2674 On X86-32 available only on PAE configured kernels.
1da177e4 2675 noexec=on: enable non-executable mappings (default)
f5a1b191
JS
2676 noexec=off: disable non-executable mappings
2677
52b6179a
PA
2678 nosmap [X86]
2679 Disable SMAP (Supervisor Mode Access Prevention)
2680 even if it is supported by processor.
2681
de5397ad 2682 nosmep [X86]
52b6179a 2683 Disable SMEP (Supervisor Mode Execution Prevention)
de5397ad
FY
2684 even if it is supported by processor.
2685
f5a1b191
JS
2686 noexec32 [X86-64]
2687 This affects only 32-bit executables.
2688 noexec32=on: enable non-executable mappings (default)
2689 read doesn't imply executable mappings
2690 noexec32=off: disable non-executable mappings
2691 read implies executable mappings
1da177e4 2692
fab43ef4 2693 nofpu [MIPS,SH] Disable hardware FPU at boot time.
6902aa84 2694
cd4f0ef7 2695 nofxsr [BUGS=X86-32] Disables x86 floating point extended
4f886511
CE
2696 register save and restore. The kernel will only save
2697 legacy floating-point registers on task switch.
1da177e4 2698
0ddab1d2
TK
2699 nohugeiomap [KNL,x86] Disable kernel huge I/O mappings.
2700
52c48c51
SS
2701 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
2702 Equivalent to smt=1.
2703
47b4c679 2704 [KNL,x86] Disable symmetric multithreading (SMT).
e55ed090
TG
2705 nosmt=force: Force disable SMT, cannot be undone
2706 via the sysfs control file.
47b4c679 2707
da285121
DW
2708 nospectre_v2 [X86] Disable all mitigations for the Spectre variant 2
2709 (indirect branch prediction) vulnerability. System may
2710 allow data leaks with this option, which is equivalent
2711 to spectre_v2=off.
2712
e63490c8
KRW
2713 nospec_store_bypass_disable
2714 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
2715
0c752a93
SS
2716 noxsave [BUGS=X86] Disables x86 extended register state save
2717 and restore using xsave. The kernel will fallback to
2718 enabling legacy floating-point and sse state.
2719
b6f42a4a
FY
2720 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
2721 register states. The kernel will fall back to use
2722 xsave to save the states. By using this parameter,
2723 performance of saving the states is degraded because
2724 xsave doesn't support modified optimization while
2725 xsaveopt supports it on xsaveopt enabled systems.
2726
2727 noxsaves [X86] Disables xsaves and xrstors used in saving and
2728 restoring x86 extended register state in compacted
2729 form of xsave area. The kernel will fall back to use
2730 xsaveopt and xrstor to save and restore the states
2731 in standard form of xsave area. By using this
2732 parameter, xsave area per process might occupy more
2733 memory on xsaves enabled systems.
2734
01a24d2b
PZ
2735 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
2736 wfi(ARM) instruction doesn't work correctly and not to
2737 use it. This is also useful when using JTAG debugger.
a9913044 2738
1f29fae2
SH
2739 no_file_caps Tells the kernel not to honor file capabilities. The
2740 only way then for a file to be executed with privilege
2741 is to be setuid root or executed by root.
2742
1da177e4
LT
2743 nohalt [IA-64] Tells the kernel not to use the power saving
2744 function PAL_HALT_LIGHT when idle. This increases
2745 power-consumption. On the positive side, it reduces
2746 interrupt wake-up latency, which may improve performance
2747 in certain environments such as networked servers or
2748 real-time systems.
2749
a6e15a39
KC
2750 nohibernate [HIBERNATION] Disable hibernation and resume.
2751
79bf2bb3
TG
2752 nohz= [KNL] Boottime enable/disable dynamic ticks
2753 Valid arguments: on, off
2754 Default: on
2755
d94d1053 2756 nohz_full= [KNL,BOOT,SMP,ISOL]
2d13e6ca 2757 The argument is a cpu list, as described above.
c5bfece2 2758 In kernels built with CONFIG_NO_HZ_FULL=y, set
a831881b 2759 the specified list of CPUs whose tick will be stopped
0453b435 2760 whenever possible. The boot CPU will be forced outside
f99bcb2c
PM
2761 the range to maintain the timekeeping. Any CPUs
2762 in this list will have their RCU callbacks offloaded,
2763 just as if they had also been called out in the
2764 rcu_nocbs= boot parameter.
a831881b 2765
eeee7853
PM
2766 noiotrap [SH] Disables trapped I/O port accesses.
2767
cd4f0ef7 2768 noirqdebug [X86-32] Disables the code which attempts to detect and
1da177e4
LT
2769 disable unhandled interrupt sources.
2770
6cececfc 2771 no_timer_check [X86,APIC] Disables the code which tests for
8542b200
ZA
2772 broken timer IRQ sources.
2773
1da177e4
LT
2774 noisapnp [ISAPNP] Disables ISA PnP code.
2775
2776 noinitrd [RAM] Tells the kernel not to load any configured
2777 initial RAM disk.
2778
03ea8155
WH
2779 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
2780 remapping.
d1423d56 2781 [Deprecated - use intremap=off]
03ea8155 2782
1da177e4
LT
2783 nointroute [IA-64]
2784
d12a72b8
AL
2785 noinvpcid [X86] Disable the INVPCID cpu feature.
2786
16290246 2787 nojitter [IA-64] Disables jitter checking for ITC timers.
0aa366f3 2788
9cf4c4fc
JK
2789 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
2790
fd10cde9
GN
2791 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
2792 fault handling.
2793
80e9a4f2
AM
2794 no-vmw-sched-clock
2795 [X86,PV_OPS] Disable paravirtualized VMware scheduler
2796 clock and use the default one.
2797
d910f5c1
GC
2798 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting.
2799 steal time is computed, but won't influence scheduler
2800 behaviour
2801
cd4f0ef7 2802 nolapic [X86-32,APIC] Do not enable or use the local APIC.
1da177e4 2803
cd4f0ef7 2804 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
ad62ca2b 2805
1da177e4 2806 noltlbs [PPC] Do not use large page/tlb entries for kernel
f15eea66 2807 lowmem mapping on PPC40x and PPC8xx
1da177e4 2808
312f1f01
H
2809 nomca [IA-64] Disable machine check abort handling
2810
13696e0a 2811 nomce [X86-32] Disable Machine Check Exception
abe37e5a 2812
83d7384f
AS
2813 nomfgpt [X86-32] Disable Multi-Function General Purpose
2814 Timer usage (for AMD Geode machines).
2815
bda62633
DZ
2816 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
2817 shutdown the other cpus. Instead use the REBOOT_VECTOR
2818 irq.
2819
02608bef
DY
2820 nomodule Disable module load
2821
016ddd9b
JK
2822 nopat [X86] Disable PAT (page attribute table extension of
2823 pagetables) support.
2824
0790c9aa
AL
2825 nopcid [X86-64] Disable the PCID cpu feature.
2826
0cb55ad2
RD
2827 norandmaps Don't use address space randomization. Equivalent to
2828 echo 0 > /proc/sys/kernel/randomize_va_space
2829
cd4f0ef7 2830 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
b7fb4af0
JF
2831 with UP alternatives
2832
7a5091d5
PA
2833 nordrand [X86] Disable kernel use of the RDRAND and
2834 RDSEED instructions even if they are supported
2835 by the processor. RDRAND and RDSEED are still
2836 available to user space applications.
49d859d7 2837
a9913044
RD
2838 noresume [SWSUSP] Disables resume and restores original swap
2839 space.
2840
1da177e4
LT
2841 no-scroll [VGA] Disables scrollback.
2842 This is required for the Braillex ib80-piezo Braille
2843 reader made by F.H. Papenmeier (Germany).
2844
2845 nosbagart [IA-64]
2846
cd4f0ef7 2847 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
4f886511 2848
61ec7567
LB
2849 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
2850 and disable the IO APIC. legacy for "maxcpus=0".
1da177e4 2851
97842216
DJ
2852 nosoftlockup [KNL] Disable the soft-lockup detector.
2853
1da177e4
LT
2854 nosync [HW,M68K] Disables sync negotiation for all devices.
2855
cd4f0ef7 2856 notsc [BUGS=X86-32] Disable Time Stamp Counter
1da177e4 2857
195daf66
UO
2858 nowatchdog [KNL] Disable both lockup detectors, i.e.
2859 soft-lockup and NMI watchdog (hard-lockup).
58687acb 2860
1da177e4 2861 nowb [ARM]
a9913044 2862
2b2fd87a
WH
2863 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
2864
f78cff48
FY
2865 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
2866 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
2867 Some features depend on CPU0. Known dependencies are:
2868 1. Resume from suspend/hibernate depends on CPU0.
2869 Suspend/hibernate will fail if CPU0 is offline and you
2870 need to online CPU0 before suspend/hibernate.
2871 2. PIC interrupts also depend on CPU0. CPU0 can't be
2872 removed if a PIC interrupt is detected.
2873 It's said poweroff/reboot may depend on CPU0 on some
2874 machines although I haven't seen such issues so far
2875 after CPU0 is offline on a few tested machines.
2876 If the dependencies are under your control, you can
2877 turn on cpu0_hotplug.
2878
35b55ef2
NC
2879 nps_mtm_hs_ctr= [KNL,ARC]
2880 This parameter sets the maximum duration, in
2881 cycles, each HW thread of the CTOP can run
2882 without interruptions, before HW switches it.
2883 The actual maximum duration is 16 times this
2884 parameter's value.
2885 Format: integer between 1 and 255
2886 Default: 255
2887
16290246 2888 nptcg= [IA-64] Override max number of concurrent global TLB
a6c75b86
FY
2889 purges which is reported from either PAL_VM_SUMMARY or
2890 SAL PALO.
2891
2b633e3f
YL
2892 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
2893 could support. nr_cpus=n : n >= 1 limits the kernel to
7c142bfe
BH
2894 support 'n' processors. It could be larger than the
2895 number of already plugged CPU during bootup, later in
2896 runtime you can physically add extra cpu until it reaches
2897 n. So during boot up some boot time memory for per-cpu
2898 variables need be pre-allocated for later physical cpu
2899 hot plugging.
2b633e3f 2900
0cb55ad2
RD
2901 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
2902
1a687c2e
MG
2903 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
2904 Allowed values are enable and disable
2905
f0c0b2b8 2906 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
c9bff3ee 2907 'node', 'default' can be specified
f0c0b2b8
KH
2908 This can be set from sysctl after boot.
2909 See Documentation/sysctl/vm.txt for details.
2910
7c4be253
RD
2911 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
2912 See Documentation/debugging-via-ohci1394.txt for more
2913 info.
2914
3ef0e1f8
AS
2915 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
2916 Rather than timing out after 20 ms if an EC
2917 command is not properly ACKed, override the length
2918 of the timeout. We have interrupts disabled while
2919 waiting for the ACK, so if this is set too high
2920 interrupts *may* be lost!
2921
15ac7afe
TL
2922 omap_mux= [OMAP] Override bootloader pin multiplexing.
2923 Format: <mux_mode0.mode_name=value>...
2924 For example, to override I2C bus2:
2925 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
2926
1da177e4
LT
2927 oprofile.timer= [HW]
2928 Use timer interrupt instead of performance counters
2929
7e4e0bd5
RR
2930 oprofile.cpu_type= Force an oprofile cpu type
2931 This might be useful if you have an older oprofile
2932 userland or if you want common events.
8d7ff4f2
RR
2933 Format: { arch_perfmon }
2934 arch_perfmon: [X86] Force use of architectural
7e4e0bd5
RR
2935 perfmon on Intel CPUs instead of the
2936 CPU specific event set.
159a80b2
RR
2937 timer: [X86] Force use of architectural NMI
2938 timer mode (see also oprofile.timer
2939 for generic hr timer mode)
1dcdb5a9 2940
44a4dcf7
RD
2941 oops=panic Always panic on oopses. Default is to just kill the
2942 process, but there is a small probability of
2943 deadlocking the machine.
d404ab0a
OH
2944 This will also cause panics on machine check exceptions.
2945 Useful together with panic=30 to trigger a reboot.
2946
bcfde334
RD
2947 OSS [HW,OSS]
2948 See Documentation/sound/oss/oss-parameters.txt
2949
48c96a36
JK
2950 page_owner= [KNL] Boot-time page_owner enabling option.
2951 Storage of the information about who allocated
2952 each page is disabled in default. With this switch,
2953 we can turn it on.
2954 on: enable the feature
2955
8823b1db
LA
2956 page_poison= [KNL] Boot-time parameter changing the state of
2957 poisoning on the buddy allocator.
2958 off: turn off poisoning
2959 on: turn on poisoning
2960
44a4dcf7 2961 panic= [KNL] Kernel behaviour on panic: delay <timeout>
4302fbc8
HD
2962 timeout > 0: seconds before rebooting
2963 timeout = 0: wait forever
2964 timeout < 0: reboot immediately
1da177e4
LT
2965 Format: <timeout>
2966
9e3961a0
PB
2967 panic_on_warn panic() instead of WARN(). Useful to cause kdump
2968 on a WARN().
2969
f06e5153
MH
2970 crash_kexec_post_notifiers
2971 Run kdump after running panic-notifiers and dumping
2972 kmsg. This only for the users who doubt kdump always
2973 succeeds in any situation.
2974 Note that this also increases risks of kdump failure,
2975 because some panic notifiers can make the crashed
2976 kernel more unstable.
2977
1da177e4
LT
2978 parkbd.port= [HW] Parallel port number the keyboard adapter is
2979 connected to, default is 0.
2980 Format: <parport#>
2981 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
2982 0 for XT, 1 for AT (default is AT).
a9913044
RD
2983 Format: <mode>
2984
2985 parport= [HW,PPT] Specify parallel ports. 0 disables.
2986 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
2987 Use 'auto' to force the driver to use any
2988 IRQ/DMA settings detected (the default is to
2989 ignore detected IRQ/DMA settings because of
2990 possible conflicts). You can specify the base
2991 address, IRQ, and DMA settings; IRQ and DMA
2992 should be numbers, or 'auto' (for using detected
2993 settings on that particular port), or 'nofifo'
2994 (to avoid using a FIFO even if it is detected).
2995 Parallel ports are assigned in the order they
2996 are specified on the command line, starting
2997 with parport0.
2998
2999 parport_init_mode= [HW,PPT]
3000 Configure VIA parallel port to operate in
3001 a specific mode. This is necessary on Pegasos
3002 computer where firmware has no options for setting
3003 up parallel port mode and sets it to spp.
3004 Currently this function knows 686a and 8231 chips.
1da177e4
LT
3005 Format: [spp|ps2|epp|ecp|ecpepp]
3006
dd287796
AM
3007 pause_on_oops=
3008 Halt all CPUs after the first oops has been printed for
3009 the specified number of seconds. This is to be used if
3010 your oopses keep scrolling off the screen.
3011
1da177e4
LT
3012 pcbit= [HW,ISDN]
3013
3014 pcd. [PARIDE]
3015 See header of drivers/block/paride/pcd.c.
31c00fc1 3016 See also Documentation/blockdev/paride.txt.
1da177e4 3017
a9913044 3018 pci=option[,option...] [PCI] various PCI subsystem options:
1cc0ca26
BH
3019 earlydump [X86] dump PCI config space before the kernel
3020 changes anything
c0115606 3021 off [X86] don't probe for the PCI bus
cd4f0ef7 3022 bios [X86-32] force use of PCI BIOS, don't access
a9913044
RD
3023 the hardware directly. Use this if your machine
3024 has a non-standard PCI host bridge.
cd4f0ef7 3025 nobios [X86-32] disallow use of PCI BIOS, only direct
a9913044
RD
3026 hardware access methods are allowed. Use this
3027 if you experience crashes upon bootup and you
3028 suspect they are caused by the BIOS.
afd8c084
BP
3029 conf1 [X86] Force use of PCI Configuration Access
3030 Mechanism 1 (config address in IO port 0xCF8,
3031 data in IO port 0xCFC, both 32-bit).
3032 conf2 [X86] Force use of PCI Configuration Access
3033 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3034 the function, IO port 0xCFA, also 8-bit, sets
3035 bus number. The config space is then accessed
3036 through ports 0xC000-0xCFFF).
3037 See http://wiki.osdev.org/PCI for more info
3038 on the configuration access mechanisms.
7f785763
RD
3039 noaer [PCIE] If the PCIEAER kernel config parameter is
3040 enabled, this kernel boot option can be used to
3041 disable the use of PCIE advanced error reporting.
32a2eea7
JG
3042 nodomains [PCI] Disable support for multiple PCI
3043 root domains (aka PCI segments, in ACPI-speak).
6cececfc 3044 nommconf [X86] Disable use of MMCONFIG for PCI
61be6d66 3045 Configuration
12983077
AH
3046 check_enable_amd_mmconf [X86] check for and enable
3047 properly configured MMIO access to PCI
3048 config space on AMD family 10h CPU
309e57df
MW
3049 nomsi [MSI] If the PCI_MSI kernel config parameter is
3050 enabled, this kernel boot option can be used to
3051 disable the use of MSI interrupts system-wide.
a9322f64
SA
3052 noioapicquirk [APIC] Disable all boot interrupt quirks.
3053 Safety option to keep boot IRQs enabled. This
3054 should never be necessary.
9197979b
SA
3055 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3056 primary IO-APIC for bridges that cannot disable
3057 boot IRQs. This fixes a source of spurious IRQs
3058 when the system masks IRQs.
41b9eb26
SA
3059 noioapicreroute [APIC] Disable workaround that uses the
3060 boot IRQ equivalent of an IRQ that connects to
3061 a chipset where boot IRQs cannot be disabled.
3062 The opposite of ioapicreroute.
cd4f0ef7 3063 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
a9913044
RD
3064 routing table. These calls are known to be buggy
3065 on several machines and they hang the machine
3066 when used, but on other computers it's the only
3067 way to get the interrupt routing table. Try
3068 this option if the kernel is unable to allocate
3069 IRQs or discover secondary PCI buses on your
3070 motherboard.
c0115606 3071 rom [X86] Assign address space to expansion ROMs.
a9913044
RD
3072 Use with caution as certain devices share
3073 address decoders between ROMs and other
3074 resources.
c0115606 3075 norom [X86] Do not assign address space to
bb71ad88
GH
3076 expansion ROMs that do not already have
3077 BIOS assigned address ranges.
7bd1c365
MH
3078 nobar [X86] Do not assign address space to the
3079 BARs that weren't assigned by the BIOS.
c0115606 3080 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
a9913044
RD
3081 assigned automatically to PCI devices. You can
3082 make the kernel exclude IRQs of your ISA cards
3083 this way.
c0115606 3084 pirqaddr=0xAAAAA [X86] Specify the physical address
a9913044
RD
3085 of the PIRQ table (normally generated
3086 by the BIOS) if it is outside the
3087 F0000h-100000h range.
c0115606 3088 lastbus=N [X86] Scan all buses thru bus #N. Can be
a9913044
RD
3089 useful if the kernel is unable to find your
3090 secondary buses and you want to tell it
3091 explicitly which ones they are.
c0115606 3092 assign-busses [X86] Always assign all PCI bus
a9913044
RD
3093 numbers ourselves, overriding
3094 whatever the firmware may have done.
c0115606 3095 usepirqmask [X86] Honor the possible IRQ mask stored
a9913044
RD
3096 in the BIOS $PIR table. This is needed on
3097 some systems with broken BIOSes, notably
3098 some HP Pavilion N5400 and Omnibook XE3
3099 notebooks. This will have no effect if ACPI
3100 IRQ routing is enabled.
c0115606 3101 noacpi [X86] Do not use ACPI for IRQ routing
a9913044 3102 or for PCI scanning.
7bc5e3f2
BH
3103 use_crs [X86] Use PCI host bridge window information
3104 from ACPI. On BIOSes from 2008 or later, this
3105 is enabled by default. If you need to use this,
3106 please report a bug.
3107 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3108 If you need to use this, please report a bug.
a9913044
RD
3109 routeirq Do IRQ routing for all PCI devices.
3110 This is normally done in pci_enable_device(),
3111 so this option is a temporary workaround
3112 for broken drivers that don't call it.
13a6ddb0
YL
3113 skip_isa_align [X86] do not align io start addr, so can
3114 handle more pci cards
0637a70a
AK
3115 noearly [X86] Don't do any early type 1 scanning.
3116 This might help on some broken boards which
3117 machine check when some devices' config space
3118 is read. But various workarounds are disabled
3119 and some IOMMU drivers will not work.
6b4b78fe
MD
3120 bfsort Sort PCI devices into breadth-first order.
3121 This sorting is done to get a device
3122 order compatible with older (<= 2.4) kernels.
3123 nobfsort Don't sort PCI devices into breadth-first order.
fa238712
YW
3124 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3125 tuning and use the BIOS-configured MPS defaults.
3126 pcie_bus_safe Set every device's MPS to the largest value
3127 supported by all devices below the root complex.
3128 pcie_bus_perf Set device MPS to the largest allowable MPS
3129 based on its parent bus. Also set MRRS (Max
3130 Read Request Size) to the largest supported
3131 value (no larger than the MPS that the device
3132 or bus can support) for best performance.
3133 pcie_bus_peer2peer Set every device's MPS to 128B, which
3134 every device is guaranteed to support. This
3135 configuration allows peer-to-peer DMA between
3136 any pair of devices, possibly at the cost of
3137 reduced performance. This also guarantees
3138 that hot-added devices will work.
4516a618
AN
3139 cbiosize=nn[KMG] The fixed amount of bus space which is
3140 reserved for the CardBus bridge's IO window.
3141 The default value is 256 bytes.
3142 cbmemsize=nn[KMG] The fixed amount of bus space which is
3143 reserved for the CardBus bridge's memory
3144 window. The default value is 64 megabytes.
32a9a682
YS
3145 resource_alignment=
3146 Format:
3147 [<order of align>@][<domain>:]<bus>:<slot>.<func>[; ...]
644a544f
KMEE
3148 [<order of align>@]pci:<vendor>:<device>\
3149 [:<subvendor>:<subdevice>][; ...]
32a9a682
YS
3150 Specifies alignment and device to reassign
3151 aligned memory resources.
3152 If <order of align> is not specified,
3153 PAGE_SIZE is used as alignment.
3154 PCI-PCI bridge can be specified, if resource
3155 windows need to be expanded.
8b078c60
MK
3156 To specify the alignment for several
3157 instances of a device, the PCI vendor,
3158 device, subvendor, and subdevice may be
3159 specified, e.g., 4096@pci:8086:9c22:103c:198f
43c16408
AP
3160 ecrc= Enable/disable PCIe ECRC (transaction layer
3161 end-to-end CRC checking).
3162 bios: Use BIOS/firmware settings. This is the
3163 the default.
3164 off: Turn ECRC off
3165 on: Turn ECRC on.
8c8803c5
YW
3166 hpiosize=nn[KMG] The fixed amount of bus space which is
3167 reserved for hotplug bridge's IO window.
3168 Default size is 256 bytes.
3169 hpmemsize=nn[KMG] The fixed amount of bus space which is
3170 reserved for hotplug bridge's memory window.
3171 Default size is 2 megabytes.
e16b4660
KB
3172 hpbussize=nn The minimum amount of additional bus numbers
3173 reserved for buses below a hotplug bridge.
3174 Default is 1.
b55438fd
YL
3175 realloc= Enable/disable reallocating PCI bridge resources
3176 if allocations done by BIOS are too small to
3177 accommodate resources required by all child
3178 devices.
3179 off: Turn realloc off
3180 on: Turn realloc on
3181 realloc same as realloc=on
6748dcc2 3182 noari do not use PCIe ARI.
284f5f9d
BH
3183 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3184 only look for one device below a PCIe downstream
3185 port.
f32ab754
UCCB
3186 big_root_window Try to add a big 64bit memory window to the PCIe
3187 root complex on AMD CPUs. Some GFX hardware
3188 can resize a BAR to allow access to all VRAM.
3189 Adding the window is slightly risky (it may
3190 conflict with unreported devices), so this
3191 taints the kernel.
6b4b78fe 3192
e5665a45
CE
3193 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3194 Management.
3195 off Disable ASPM.
3196 force Enable ASPM even on devices that claim not to support it.
3197 WARNING: Forcing ASPM on may cause system lockups.
3198
7570a333
MT
3199 pcie_hp= [PCIE] PCI Express Hotplug driver options:
3200 nomsi Do not use MSI for PCI Express Native Hotplug (this
3201 makes all PCIe ports use INTx for hotplug services).
3202
79dd9182 3203 pcie_ports= [PCIE] PCIe ports handling:
28eb5f27
RW
3204 auto Ask the BIOS whether or not to use native PCIe services
3205 associated with PCIe ports (PME, hot-plug, AER). Use
3206 them only if that is allowed by the BIOS.
3207 native Use native PCIe services associated with PCIe ports
3208 unconditionally.
79dd9182
RW
3209 compat Treat PCIe ports as PCI-to-PCI bridges, disable the PCIe
3210 ports driver.
3211
9d26d3a8
MW
3212 pcie_port_pm= [PCIE] PCIe port power management handling:
3213 off Disable power management of all PCIe ports
3214 force Forcibly enable power management of all PCIe ports
3215
c7f48656 3216 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
c39fae14 3217 nomsi Do not use MSI for native PCIe PME signaling (this makes
28eb5f27 3218 all PCIe root ports use INTx for all services).
c7f48656 3219
1da177e4
LT
3220 pcmv= [HW,PCMCIA] BadgePAD 4
3221
39ac5ba5
TB
3222 pd_ignore_unused
3223 [PM]
3224 Keep all power-domains already enabled by bootloader on,
3225 even if no driver has claimed them. This is useful
3226 for debug and development, but should not be
3227 needed on a platform with proper driver support.
3228
1da177e4 3229 pd. [PARIDE]
31c00fc1 3230 See Documentation/blockdev/paride.txt.
1da177e4
LT
3231
3232 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3233 boot time.
3234 Format: { 0 | 1 }
3235 See arch/parisc/kernel/pdc_chassis.c
3236
f58dc01b 3237 percpu_alloc= Select which percpu first chunk allocator to use.
e933a73f
TH
3238 Currently supported values are "embed" and "page".
3239 Archs may support subset or none of the selections.
3240 See comments in mm/percpu.c for details on each
3241 allocator. This parameter is primarily for debugging
3242 and performance comparison.
fa8a7094 3243
1da177e4 3244 pf. [PARIDE]
31c00fc1 3245 See Documentation/blockdev/paride.txt.
1da177e4
LT
3246
3247 pg. [PARIDE]
31c00fc1 3248 See Documentation/blockdev/paride.txt.
1da177e4
LT
3249
3250 pirq= [SMP,APIC] Manual mp-table setup
71cced6e 3251 See Documentation/x86/i386/IO-APIC.txt.
1da177e4
LT
3252
3253 plip= [PPT,NET] Parallel port network link
3254 Format: { parport<nr> | timid | 0 }
3ba9b1b8 3255 See also Documentation/admin-guide/parport.rst.
1da177e4 3256
16290246 3257 pmtmr= [X86] Manual setup of pmtmr I/O Port.
de32a243
TG
3258 Override pmtimer IOPort with a hex value.
3259 e.g. pmtmr=0x508
3260
96242116
BH
3261 pnp.debug=1 [PNP]
3262 Enable PNP debug messages (depends on the
3263 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3264 via /sys/module/pnp/parameters/debug. We always show
3265 current resource usage; turning this on also shows
3266 possible settings and some assignment information.
97ef062b 3267
1da177e4
LT
3268 pnpacpi= [ACPI]
3269 { off }
3270
3271 pnpbios= [ISAPNP]
3272 { on | off | curr | res | no-curr | no-res }
3273
3274 pnp_reserve_irq=
3275 [ISAPNP] Exclude IRQs for the autoconfiguration
3276
3277 pnp_reserve_dma=
3278 [ISAPNP] Exclude DMAs for the autoconfiguration
3279
3280 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
a9913044 3281 Ranges are in pairs (I/O port base and size).
1da177e4
LT
3282
3283 pnp_reserve_mem=
a9913044
RD
3284 [ISAPNP] Exclude memory regions for the
3285 autoconfiguration.
1da177e4
LT
3286 Ranges are in pairs (memory base and size).
3287
4af94f39
RD
3288 ports= [IP_VS_FTP] IPVS ftp helper module
3289 Default is 21.
3290 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3291 may be specified.
3292 Format: <port>,<port>....
3293
c3cbd075
BS
3294 powersave=off [PPC] This option disables power saving features.
3295 It specifically disables cpuidle and sets the
3296 platform machine description specific power_save
3297 function to NULL. On Idle the CPU just reduces
3298 execution priority.
3299
3eb5d588
AB
3300 ppc_strict_facility_enable
3301 [PPC] This option catches any kernel floating point,
3302 Altivec, VSX and SPE outside of regions specifically
3303 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3304 There is some performance impact when enabling this.
3305
07fd1761
CB
3306 ppc_tm= [PPC]
3307 Format: {"off"}
3308 Disable Hardware Transactional Memory
3309
45807a1d
IM
3310 print-fatal-signals=
3311 [KNL] debug: print fatal signals
f84d49b2
NO
3312
3313 If enabled, warn about various signal handling
3314 related application anomalies: too many signals,
3315 too many POSIX.1 timers, fatal signals causing a
3316 coredump - etc.
3317
3318 If you hit the warning due to signal overflow,
3319 you might want to try "ulimit -i unlimited".
3320
45807a1d
IM
3321 default: off.
3322
c22ab332
MG
3323 printk.always_kmsg_dump=
3324 Trigger kmsg_dump for cases other than kernel oops or
3325 panics
3326 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3327 default: disabled
3328
750afe7b
BP
3329 printk.devkmsg={on,off,ratelimit}
3330 Control writing to /dev/kmsg.
3331 on - unlimited logging to /dev/kmsg from userspace
3332 off - logging to /dev/kmsg disabled
3333 ratelimit - ratelimit the logging
3334 Default: ratelimit
3335
e84845c4
RD
3336 printk.time= Show timing data prefixed to each printk message line
3337 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3338
0cb55ad2
RD
3339 processor.max_cstate= [HW,ACPI]
3340 Limit processor to maximum C-state
3341 max_cstate=9 overrides any DMI blacklist limit.
3342
3343 processor.nocst [HW,ACPI]
3344 Ignore the _CST method to determine C-states,
3345 instead using the legacy FADT method
3346
1da177e4 3347 profile= [KNL] Enable kernel profiling via /proc/profile
e7e61fc0
RD
3348 Format: [<profiletype>,]<number>
3349 Param: <profiletype>: "schedule", "sleep", or "kvm"
3350 [defaults to kernel profiling]
a9913044 3351 Param: "schedule" - profile schedule points.
b3da2a73
MG
3352 Param: "sleep" - profile D-state sleeping (millisecs).
3353 Requires CONFIG_SCHEDSTATS
c0fe2e69 3354 Param: "kvm" - profile VM exits.
e7e61fc0
RD
3355 Param: <number> - step/bucket size as a power of 2 for
3356 statistical time based profiling.
1da177e4 3357
1da177e4
LT
3358 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3359 before loading.
31c00fc1 3360 See Documentation/blockdev/ramdisk.txt.
1da177e4 3361
a9913044
RD
3362 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3363 probe for; one of (bare|imps|exps|lifebook|any).
1da177e4
LT
3364 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3365 per second.
a9913044
RD
3366 psmouse.resetafter= [HW,MOUSE]
3367 Try to reset the device after so many bad packets
1da177e4
LT
3368 (0 = never).
3369 psmouse.resolution=
3370 [HW,MOUSE] Set desired mouse resolution, in dpi.
3371 psmouse.smartscroll=
a9913044 3372 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
1da177e4
LT
3373 0 = disabled, 1 = enabled (default).
3374
dee28e72
MG
3375 pstore.backend= Specify the name of the pstore backend to use
3376
1da177e4 3377 pt. [PARIDE]
31c00fc1 3378 See Documentation/blockdev/paride.txt.
1da177e4 3379
01c9b17b
DH
3380 pti= [X86_64] Control Page Table Isolation of user and
3381 kernel address spaces. Disabling this feature
3382 removes hardening, but improves performance of
3383 system calls and interrupts.
3384
3385 on - unconditionally enable
3386 off - unconditionally disable
3387 auto - kernel detects whether your CPU model is
3388 vulnerable to issues that PTI mitigates
3389
3390 Not specifying this option is equivalent to pti=auto.
3391
3392 nopti [X86_64]
3393 Equivalent to pti=off
41f4c20b 3394
dc8c8587
KS
3395 pty.legacy_count=
3396 [KNL] Number of legacy pty's. Overwrites compiled-in
3397 default number.
3398
7d2c502f 3399 quiet [KNL] Disable most log messages
a9913044 3400
1da177e4
LT
3401 r128= [HW,DRM]
3402
3403 raid= [HW,RAID]
e52347bd 3404 See Documentation/admin-guide/md.rst.
1da177e4 3405
1da177e4 3406 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
31c00fc1 3407 See Documentation/blockdev/ramdisk.txt.
1da177e4 3408
011d8261
BP
3409 ras=option[,option,...] [KNL] RAS-specific options
3410
3411 cec_disable [X86]
3412 Disable the Correctable Errors Collector,
3413 see CONFIG_RAS_CEC help text.
3414
4102adab 3415 rcu_nocbs= [KNL]
2d13e6ca
NC
3416 The argument is a cpu list, as described above.
3417
3fbfbf7a
PM
3418 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
3419 the specified list of CPUs to be no-callback CPUs.
3420 Invocation of these CPUs' RCU callbacks will
a4889858
PM
3421 be offloaded to "rcuox/N" kthreads created for
3422 that purpose, where "x" is "b" for RCU-bh, "p"
3423 for RCU-preempt, and "s" for RCU-sched, and "N"
3424 is the CPU number. This reduces OS jitter on the
3fbfbf7a
PM
3425 offloaded CPUs, which can be useful for HPC and
3426 real-time workloads. It can also improve energy
3427 efficiency for asymmetric multiprocessors.
3428
4102adab 3429 rcu_nocb_poll [KNL]
3fbfbf7a
PM
3430 Rather than requiring that offloaded CPUs
3431 (specified by rcu_nocbs= above) explicitly
3432 awaken the corresponding "rcuoN" kthreads,
3433 make these kthreads poll for callbacks.
3434 This improves the real-time response for the
3435 offloaded CPUs by relieving them of the need to
3436 wake up the corresponding kthread, but degrades
3437 energy efficiency by requiring that the kthreads
3438 periodically wake up to do the polling.
3439
4102adab 3440 rcutree.blimit= [KNL]
97e63f0c
PM
3441 Set maximum number of finished RCU callbacks to
3442 process in one batch.
21a1ea9e 3443
a3dc2948
PM
3444 rcutree.dump_tree= [KNL]
3445 Dump the structure of the rcu_node combining tree
3446 out at early boot. This is used for diagnostic
3447 purposes, to verify correct tree setup.
3448
0f41c0dd
PM
3449 rcutree.gp_cleanup_delay= [KNL]
3450 Set the number of jiffies to delay each step of
90040c9e 3451 RCU grace-period cleanup.
0f41c0dd 3452
37745d28
PM
3453 rcutree.gp_init_delay= [KNL]
3454 Set the number of jiffies to delay each step of
90040c9e 3455 RCU grace-period initialization.
0f41c0dd
PM
3456
3457 rcutree.gp_preinit_delay= [KNL]
3458 Set the number of jiffies to delay each step of
3459 RCU grace-period pre-initialization, that is,
3460 the propagation of recent CPU-hotplug changes up
90040c9e 3461 the rcu_node combining tree.
37745d28 3462
7fa27001
PM
3463 rcutree.rcu_fanout_exact= [KNL]
3464 Disable autobalancing of the rcu_node combining
3465 tree. This is used by rcutorture, and might
3466 possibly be useful for architectures having high
3467 cache-to-cache transfer latencies.
37745d28 3468
4102adab 3469 rcutree.rcu_fanout_leaf= [KNL]
ee968ac6
PM
3470 Change the number of CPUs assigned to each
3471 leaf rcu_node structure. Useful for very
3472 large systems, which will choose the value 64,
3473 and for NUMA systems with large remote-access
3474 latencies, which will choose a value aligned
3475 with the appropriate hardware boundaries.
f885b7f2 3476
4a81e832
PM
3477 rcutree.jiffies_till_sched_qs= [KNL]
3478 Set required age in jiffies for a
3479 given grace period before RCU starts
3480 soliciting quiescent-state help from
3481 rcu_note_context_switch().
3482
4102adab 3483 rcutree.jiffies_till_first_fqs= [KNL]
c0f4dfd4
PM
3484 Set delay from grace-period initialization to
3485 first attempt to force quiescent states.
3486 Units are jiffies, minimum value is zero,
3487 and maximum value is HZ.
3488
4102adab 3489 rcutree.jiffies_till_next_fqs= [KNL]
c0f4dfd4
PM
3490 Set delay between subsequent attempts to force
3491 quiescent states. Units are jiffies, minimum
3492 value is one, and maximum value is HZ.
3493
21871d7e 3494 rcutree.kthread_prio= [KNL,BOOT]
d2af1ad7
PM
3495 Set the SCHED_FIFO priority of the RCU per-CPU
3496 kthreads (rcuc/N). This value is also used for
3497 the priority of the RCU boost threads (rcub/N)
3498 and for the RCU grace-period kthreads (rcu_bh,
3499 rcu_preempt, and rcu_sched). If RCU_BOOST is
3500 set, valid values are 1-99 and the default is 1
3501 (the least-favored priority). Otherwise, when
3502 RCU_BOOST is not set, valid values are 0-99 and
3503 the default is zero (non-realtime operation).
21871d7e 3504
fbce7497
PM
3505 rcutree.rcu_nocb_leader_stride= [KNL]
3506 Set the number of NOCB kthread groups, which
3507 defaults to the square root of the number of
3508 CPUs. Larger numbers reduces the wakeup overhead
3509 on the per-CPU grace-period kthreads, but increases
3510 that same overhead on each group's leader.
3511
4102adab 3512 rcutree.qhimark= [KNL]
97e63f0c
PM
3513 Set threshold of queued RCU callbacks beyond which
3514 batch limiting is disabled.
21a1ea9e 3515
4102adab 3516 rcutree.qlowmark= [KNL]
24aaef8d
RD
3517 Set threshold of queued RCU callbacks below which
3518 batch limiting is re-enabled.
21a1ea9e 3519
4102adab 3520 rcutree.rcu_idle_gp_delay= [KNL]
c0f4dfd4
PM
3521 Set wakeup interval for idle CPUs that have
3522 RCU callbacks (RCU_FAST_NO_HZ=y).
d40011f6 3523
4102adab 3524 rcutree.rcu_idle_lazy_gp_delay= [KNL]
c0f4dfd4
PM
3525 Set wakeup interval for idle CPUs that have
3526 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
3527 Lazy RCU callbacks are those which RCU can
3528 prove do nothing more than free memory.
d40011f6 3529
e3c50dfb
PM
3530 rcutree.rcu_kick_kthreads= [KNL]
3531 Cause the grace-period kthread to get an extra
3532 wake_up() if it sleeps three times longer than
3533 it should at force-quiescent-state time.
3534 This wake_up() will be accompanied by a
3535 WARN_ONCE() splat and an ftrace_dump().
3536
881ed593
PM
3537 rcuperf.gp_async= [KNL]
3538 Measure performance of asynchronous
3539 grace-period primitives such as call_rcu().
3540
3541 rcuperf.gp_async_max= [KNL]
3542 Specify the maximum number of outstanding
3543 callbacks per writer thread. When a writer
3544 thread exceeds this limit, it invokes the
3545 corresponding flavor of rcu_barrier() to allow
3546 previously posted callbacks to drain.
3547
bdea9e34
PM
3548 rcuperf.gp_exp= [KNL]
3549 Measure performance of expedited synchronous
3550 grace-period primitives.
3551
df37e66b
PM
3552 rcuperf.holdoff= [KNL]
3553 Set test-start holdoff period. The purpose of
3554 this parameter is to delay the start of the
3555 test until boot completes in order to avoid
3556 interference.
3557
bdea9e34
PM
3558 rcuperf.nreaders= [KNL]
3559 Set number of RCU readers. The value -1 selects
3560 N, where N is the number of CPUs. A value
3561 "n" less than -1 selects N-n+1, where N is again
3562 the number of CPUs. For example, -2 selects N
3563 (the number of CPUs), -3 selects N+1, and so on.
3564 A value of "n" less than or equal to -N selects
3565 a single reader.
3566
3567 rcuperf.nwriters= [KNL]
3568 Set number of RCU writers. The values operate
3569 the same as for rcuperf.nreaders.
3570 N, where N is the number of CPUs
3571
3572 rcuperf.perf_runnable= [BOOT]
3573 Start rcuperf running at boot time.
3574
820687a7
PM
3575 rcuperf.perf_type= [KNL]
3576 Specify the RCU implementation to test.
3577
bdea9e34
PM
3578 rcuperf.shutdown= [KNL]
3579 Shut the system down after performance tests
3580 complete. This is useful for hands-off automated
3581 testing.
3582
bdea9e34
PM
3583 rcuperf.verbose= [KNL]
3584 Enable additional printk() statements.
3585
820687a7
PM
3586 rcuperf.writer_holdoff= [KNL]
3587 Write-side holdoff between grace periods,
3588 in microseconds. The default of zero says
3589 no holdoff.
3590
38706bc5
PM
3591 rcutorture.cbflood_inter_holdoff= [KNL]
3592 Set holdoff time (jiffies) between successive
3593 callback-flood tests.
3594
3595 rcutorture.cbflood_intra_holdoff= [KNL]
3596 Set holdoff time (jiffies) between successive
3597 bursts of callbacks within a given callback-flood
3598 test.
3599
3600 rcutorture.cbflood_n_burst= [KNL]
3601 Set the number of bursts making up a given
3602 callback-flood test. Set this to zero to
3603 disable callback-flood testing.
3604
3605 rcutorture.cbflood_n_per_burst= [KNL]
3606 Set the number of callbacks to be registered
3607 in a given burst of a callback-flood test.
3608
4102adab 3609 rcutorture.fqs_duration= [KNL]
21b05de4
PM
3610 Set duration of force_quiescent_state bursts
3611 in microseconds.
dabb8aa9 3612
4102adab 3613 rcutorture.fqs_holdoff= [KNL]
21b05de4
PM
3614 Set holdoff time within force_quiescent_state bursts
3615 in microseconds.
dabb8aa9 3616
4102adab 3617 rcutorture.fqs_stutter= [KNL]
21b05de4
PM
3618 Set wait time between force_quiescent_state bursts
3619 in seconds.
3620
3621 rcutorture.gp_cond= [KNL]
3622 Use conditional/asynchronous update-side
3623 primitives, if available.
dabb8aa9 3624
4102adab 3625 rcutorture.gp_exp= [KNL]
21b05de4 3626 Use expedited update-side primitives, if available.
4102adab
PM
3627
3628 rcutorture.gp_normal= [KNL]
21b05de4
PM
3629 Use normal (non-expedited) asynchronous
3630 update-side primitives, if available.
3631
3632 rcutorture.gp_sync= [KNL]
3633 Use normal (non-expedited) synchronous
3634 update-side primitives, if available. If all
3635 of rcutorture.gp_cond=, rcutorture.gp_exp=,
3636 rcutorture.gp_normal=, and rcutorture.gp_sync=
3637 are zero, rcutorture acts as if is interpreted
3638 they are all non-zero.
dabb8aa9 3639
4102adab 3640 rcutorture.n_barrier_cbs= [KNL]
dabb8aa9
PM
3641 Set callbacks/threads for rcu_barrier() testing.
3642
4102adab 3643 rcutorture.nfakewriters= [KNL]
dabb8aa9
PM
3644 Set number of concurrent RCU writers. These just
3645 stress RCU, they don't participate in the actual
3646 test, hence the "fake".
3647
4102adab 3648 rcutorture.nreaders= [KNL]
3838cc18
PM
3649 Set number of RCU readers. The value -1 selects
3650 N-1, where N is the number of CPUs. A value
3651 "n" less than -1 selects N-n-2, where N is again
3652 the number of CPUs. For example, -2 selects N
3653 (the number of CPUs), -3 selects N+1, and so on.
dabb8aa9 3654
4102adab
PM
3655 rcutorture.object_debug= [KNL]
3656 Enable debug-object double-call_rcu() testing.
3657
3658 rcutorture.onoff_holdoff= [KNL]
dabb8aa9
PM
3659 Set time (s) after boot for CPU-hotplug testing.
3660
4102adab 3661 rcutorture.onoff_interval= [KNL]
dabb8aa9
PM
3662 Set time (s) between CPU-hotplug operations, or
3663 zero to disable CPU-hotplug testing.
3664
4102adab 3665 rcutorture.shuffle_interval= [KNL]
dabb8aa9
PM
3666 Set task-shuffle interval (s). Shuffling tasks
3667 allows some CPUs to go into dyntick-idle mode
3668 during the rcutorture test.
3669
4102adab 3670 rcutorture.shutdown_secs= [KNL]
dabb8aa9
PM
3671 Set time (s) after boot system shutdown. This
3672 is useful for hands-off automated testing.
3673
4102adab 3674 rcutorture.stall_cpu= [KNL]
dabb8aa9
PM
3675 Duration of CPU stall (s) to test RCU CPU stall
3676 warnings, zero to disable.
3677
4102adab 3678 rcutorture.stall_cpu_holdoff= [KNL]
dabb8aa9
PM
3679 Time to wait (s) after boot before inducing stall.
3680
2b1516e5
PM
3681 rcutorture.stall_cpu_irqsoff= [KNL]
3682 Disable interrupts while stalling if set.
3683
4102adab 3684 rcutorture.stat_interval= [KNL]
dabb8aa9
PM
3685 Time (s) between statistics printk()s.
3686
4102adab 3687 rcutorture.stutter= [KNL]
dabb8aa9
PM
3688 Time (s) to stutter testing, for example, specifying
3689 five seconds causes the test to run for five seconds,
3690 wait for five seconds, and so on. This tests RCU's
3691 ability to transition abruptly to and from idle.
3692
4102adab 3693 rcutorture.test_boost= [KNL]
dabb8aa9
PM
3694 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
3695 "Maybe" means test if the RCU implementation
3696 under test support RCU priority boosting.
3697
4102adab 3698 rcutorture.test_boost_duration= [KNL]
dabb8aa9
PM
3699 Duration (s) of each individual boost test.
3700
4102adab 3701 rcutorture.test_boost_interval= [KNL]
dabb8aa9
PM
3702 Interval (s) between each boost test.
3703
4102adab 3704 rcutorture.test_no_idle_hz= [KNL]
dabb8aa9
PM
3705 Test RCU's dyntick-idle handling. See also the
3706 rcutorture.shuffle_interval parameter.
3707
21b05de4
PM
3708 rcutorture.torture_runnable= [BOOT]
3709 Start rcutorture running at boot time.
3710
4102adab 3711 rcutorture.torture_type= [KNL]
dabb8aa9
PM
3712 Specify the RCU implementation to test.
3713
4102adab 3714 rcutorture.verbose= [KNL]
dabb8aa9
PM
3715 Enable additional printk() statements.
3716
5a9be7c6
PM
3717 rcupdate.rcu_cpu_stall_suppress= [KNL]
3718 Suppress RCU CPU stall warning messages.
3719
3720 rcupdate.rcu_cpu_stall_timeout= [KNL]
3721 Set timeout for RCU CPU stall warning messages.
3722
4102adab
PM
3723 rcupdate.rcu_expedited= [KNL]
3724 Use expedited grace-period primitives, for
3725 example, synchronize_rcu_expedited() instead
3726 of synchronize_rcu(). This reduces latency,
3727 but can increase CPU utilization, degrade
3728 real-time latency, and degrade energy efficiency.
79cfea02 3729 No effect on CONFIG_TINY_RCU kernels.
4102adab 3730
5a9be7c6
PM
3731 rcupdate.rcu_normal= [KNL]
3732 Use only normal grace-period primitives,
3733 for example, synchronize_rcu() instead of
3734 synchronize_rcu_expedited(). This improves
79cfea02
PM
3735 real-time latency, CPU utilization, and
3736 energy efficiency, but can expose users to
3737 increased grace-period latency. This parameter
3738 overrides rcupdate.rcu_expedited. No effect on
3739 CONFIG_TINY_RCU kernels.
4102adab 3740
3e42ec1a
PM
3741 rcupdate.rcu_normal_after_boot= [KNL]
3742 Once boot has completed (that is, after
3743 rcu_end_inkernel_boot() has been invoked), use
79cfea02
PM
3744 only normal grace-period primitives. No effect
3745 on CONFIG_TINY_RCU kernels.
3e42ec1a 3746
52db30ab
PM
3747 rcupdate.rcu_task_stall_timeout= [KNL]
3748 Set timeout in jiffies for RCU task stall warning
3749 messages. Disable with a value less than or equal
3750 to zero.
3751
74860fee
PK
3752 rcupdate.rcu_self_test= [KNL]
3753 Run the RCU early boot self tests
3754
3755 rcupdate.rcu_self_test_bh= [KNL]
3756 Run the RCU bh early boot self tests
3757
3758 rcupdate.rcu_self_test_sched= [KNL]
3759 Run the RCU sched early boot self tests
3760
ffdfc409
OJ
3761 rdinit= [KNL]
3762 Format: <full_path>
3763 Run specified binary instead of /init from the ramdisk,
3764 used for early userspace startup. See initrd.
3765
1d9807fc
TL
3766 rdt= [HW,X86,RDT]
3767 Turn on/off individual RDT features. List is:
86b59b6e
FY
3768 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
3769 mba.
1d9807fc
TL
3770 E.g. to turn on cmt and turn off mba use:
3771 rdt=cmt,!mba
3772
1b3a5d02
RH
3773 reboot= [KNL]
3774 Format (x86 or x86_64):
3775 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
3776 [[,]s[mp]#### \
3777 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
3778 [[,]f[orce]
3779 Where reboot_mode is one of warm (soft) or cold (hard) or gpio,
3780 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
3781 reboot_force is either force or not specified,
3782 reboot_cpu is s[mp]#### with #### being the processor
3783 to be used for rebooting.
1da177e4 3784
46b6d94e
PJ
3785 relax_domain_level=
3786 [KNL, SMP] Set scheduler's default relax_domain_level.
09c3bcce 3787 See Documentation/cgroup-v1/cpusets.txt.
46b6d94e 3788
1da177e4
LT
3789 reserve= [KNL,BUGS] Force the kernel to ignore some iomem area
3790
cd4f0ef7 3791 reservetop= [X86-32]
461a9aff
ZA
3792 Format: nn[KMG]
3793 Reserves a hole at the top of the kernel virtual
3794 address space.
3795
9ea77bdb
PA
3796 reservelow= [X86]
3797 Format: nn[K]
3798 Set the amount of memory to reserve for BIOS at
3799 the bottom of the address space.
3800
7e96287d
VG
3801 reset_devices [KNL] Force drivers to reset the underlying device
3802 during initialization.
3803
a9913044
RD
3804 resume= [SWSUSP]
3805 Specify the partition device for software suspend
2df83fa4
MB
3806 Format:
3807 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
1da177e4 3808
ecbd0da1
RW
3809 resume_offset= [SWSUSP]
3810 Specify the offset from the beginning of the partition
3811 given by "resume=" at which the swap header is located,
3812 in <PAGE_SIZE> units (needed only for swap files).
3813 See Documentation/power/swsusp-and-swap-files.txt
3814
f126f733
BS
3815 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
3816 read the resume files
3817
6f8d7022
BS
3818 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
3819 Useful for devices that are detected asynchronously
3820 (e.g. USB and MMC devices).
3821
f996fc96
BS
3822 hibernate= [HIBERNATION]
3823 noresume Don't check if there's a hibernation image
3824 present during boot.
3825 nocompress Don't compress/decompress hibernation images.
a6e15a39 3826 no Disable hibernation and resume.
4c0b6c10
RW
3827 protect_image Turn on image protection during restoration
3828 (that will set all pages holding image data
3829 during restoration read-only).
f996fc96 3830
0a7b35cb
MN
3831 retain_initrd [RAM] Keep initrd memory after extraction
3832
0efbb786
AC
3833 rfkill.default_state=
3834 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
3835 etc. communication is blocked by default.
3836 1 Unblocked.
3837
3838 rfkill.master_switch_mode=
3839 0 The "airplane mode" button does nothing.
3840 1 The "airplane mode" button toggles between everything
3841 blocked and the previous configuration.
3842 2 The "airplane mode" button toggles between everything
3843 blocked and everything unblocked.
3844
1da177e4
LT
3845 rhash_entries= [KNL,NET]
3846 Set number of hash buckets for route cache
3847
e16fd002
GA
3848 ring3mwait=disable
3849 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
3850 CPUs.
3851
1da177e4
LT
3852 ro [KNL] Mount root device read-only on boot
3853
d2aa1aca
KC
3854 rodata= [KNL]
3855 on Mark read-only kernel memory as read-only (default).
3856 off Leave read-only kernel memory writable for debugging.
3857
605df8af
HS
3858 rockchip.usb_uart
3859 Enable the uart passthrough on the designated usb port
3860 on Rockchip SoCs. When active, the signals of the
3861 debug-uart get routed to the D+ and D- pins of the usb
3862 port and the regular usb controller gets disabled.
3863
1da177e4 3864 root= [KNL] Root filesystem
f2d34fd9 3865 See name_to_dev_t comment in init/do_mounts.c.
1da177e4
LT
3866
3867 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
3868 mount the root filesystem
3869
3870 rootflags= [KNL] Set root filesystem mount option string
3871
3872 rootfstype= [KNL] Set root filesystem type
3873
cc1ed754
PO
3874 rootwait [KNL] Wait (indefinitely) for root device to show up.
3875 Useful for devices that are detected asynchronously
3876 (e.g. USB and MMC devices).
3877
5c71d618
RT
3878 rproc_mem=nn[KMG][@address]
3879 [KNL,ARM,CMA] Remoteproc physical memory block.
3880 Memory area to be used by remote processor image,
3881 managed by CMA.
3882
1da177e4
LT
3883 rw [KNL] Mount root device read-write on boot
3884
3885 S [KNL] Run init in single mode
3886
c60d1ae4
GS
3887 s390_iommu= [HW,S390]
3888 Set s390 IOTLB flushing mode
3889 strict
3890 With strict flushing every unmap operation will result in
3891 an IOTLB flush. Default is lazy flushing before reuse,
3892 which is faster.
3893
1da177e4
LT
3894 sa1100ir [NET]
3895 See drivers/net/irda/sa1100_ir.c.
3896
1da177e4 3897 sbni= [NET] Granch SBNI12 leased line adapter
a9913044 3898
f6630114
MT
3899 sched_debug [KNL] Enables verbose scheduler debug messages.
3900
cb251765
MG
3901 schedstats= [KNL,X86] Enable or disable scheduled statistics.
3902 Allowed values are enable and disable. This feature
3903 incurs a small amount of overhead in the scheduler
3904 but is useful for debugging and performance tuning.
f6630114 3905
5307c955
MG
3906 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
3907 xtime_lock contention on larger systems, and/or RCU lock
3908 contention on all systems with CONFIG_MAXSMP set.
3909 Format: { "0" | "1" }
3910 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
3911 1 -- enable.
3912 Note: increases power consumption, thus should only be
3913 enabled if running jitter sensitive (HPC/RT) workloads.
3914
0cb55ad2
RD
3915 security= [SECURITY] Choose a security module to enable at boot.
3916 If this boot parameter is not specified, only the first
3917 security module asking for security registration will be
3918 loaded. An invalid security module name will be treated
3919 as if no module has been chosen.
3920
3921 selinux= [SELINUX] Disable or enable SELinux at boot time.
1da177e4
LT
3922 Format: { "0" | "1" }
3923 See security/selinux/Kconfig help text.
3924 0 -- disable.
3925 1 -- enable.
3926 Default value is set via kernel config option.
3927 If enabled at boot time, /selinux/disable can be used
3928 later to disable prior to initial policy load.
3929
c1c124e9
JJ
3930 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
3931 Format: { "0" | "1" }
3932 See security/apparmor/Kconfig help text
3933 0 -- disable.
3934 1 -- enable.
3935 Default value is set via kernel config option.
3936
cd4f0ef7 3937 serialnumber [BUGS=X86-32]
1da177e4 3938
1da177e4
LT
3939 shapers= [NET]
3940 Maximal number of shapers.
a9913044 3941
1da177e4
LT
3942 simeth= [IA-64]
3943 simscsi=
a9913044 3944
1da177e4
LT
3945 slram= [HW,MTD]
3946
423c929c
JK
3947 slab_nomerge [MM]
3948 Disable merging of slabs with similar size. May be
3949 necessary if there is some reason to distinguish
7660a6fd
KC
3950 allocs to different slabs, especially in hardened
3951 environments where the risk of heap overflows and
3952 layout control by attackers can usually be
3953 frustrated by disabling merging. This will reduce
3954 most of the exposure of a heap attack to a single
3955 cache (risks via metadata attacks are mostly
3956 unchanged). Debug options disable merging on their
3957 own.
423c929c
JK
3958 For more information see Documentation/vm/slub.txt.
3959
3df1cccd
DR
3960 slab_max_order= [MM, SLAB]
3961 Determines the maximum allowed order for slabs.
3962 A high setting may cause OOMs due to memory
3963 fragmentation. Defaults to 1 for systems with
3964 more than 32MB of RAM, 0 otherwise.
3965
f0630fff
CL
3966 slub_debug[=options[,slabs]] [MM, SLUB]
3967 Enabling slub_debug allows one to determine the
3968 culprit if slab objects become corrupted. Enabling
3969 slub_debug can create guard zones around objects and
3970 may poison objects when not in use. Also tracks the
3971 last alloc / free. For more information see
3972 Documentation/vm/slub.txt.
c1aee215 3973
1663f26d
TH
3974 slub_memcg_sysfs= [MM, SLUB]
3975 Determines whether to enable sysfs directories for
3976 memory cgroup sub-caches. 1 to enable, 0 to disable.
3977 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON.
3978 Enabling this can lead to a very high number of debug
3979 directories and files being created under
3980 /sys/kernel/slub.
3981
c1aee215 3982 slub_max_order= [MM, SLUB]
f0630fff
CL
3983 Determines the maximum allowed order for slabs.
3984 A high setting may cause OOMs due to memory
3985 fragmentation. For more information see
3986 Documentation/vm/slub.txt.
c1aee215
CL
3987
3988 slub_min_objects= [MM, SLUB]
f0630fff
CL
3989 The minimum number of objects per slab. SLUB will
3990 increase the slab order up to slub_max_order to
3991 generate a sufficiently large slab able to contain
3992 the number of objects indicated. The higher the number
3993 of objects the smaller the overhead of tracking slabs
3994 and the less frequently locks need to be acquired.
c1aee215
CL
3995 For more information see Documentation/vm/slub.txt.
3996
3997 slub_min_order= [MM, SLUB]
24775d65 3998 Determines the minimum page order for slabs. Must be
f0630fff 3999 lower than slub_max_order.
c1aee215
CL
4000 For more information see Documentation/vm/slub.txt.
4001
4002 slub_nomerge [MM, SLUB]
423c929c
JK
4003 Same with slab_nomerge. This is supported for legacy.
4004 See slab_nomerge for more information.
c1aee215 4005
1da177e4
LT
4006 smart2= [HW]
4007 Format: <io1>[,<io2>[,...,<io8>]]
4008
d0d4f69b
BH
4009 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
4010 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
4011 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
4012 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
4013 smsc-ircc2.ircc_irq= [HW] IRQ line
4014 smsc-ircc2.ircc_dma= [HW] DMA channel
4015 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
4016 0: Toshiba Satellite 1800 (GP data pin select)
4017 1: Fast pin select (default)
4018 2: ATC IRMode
4019
52c48c51
SS
4020 smt [KNL,S390] Set the maximum number of threads (logical
4021 CPUs) to use per physical CPU on systems capable of
4022 symmetric multithreading (SMT). Will be capped to the
4023 actual hardware limit.
4024 Format: <integer>
4025 Default: -1 (no limit)
4026
9c44bc03
IM
4027 softlockup_panic=
4028 [KNL] Should the soft-lockup detector generate panics.
44a4dcf7 4029 Format: <integer>
9c44bc03 4030
3ce62385
BP
4031 A nonzero value instructs the soft-lockup detector
4032 to panic the machine when a soft-lockup occurs. This
4033 is also controlled by CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC
4034 which is the respective build-time switch to that
4035 functionality.
4036
ed235875
AT
4037 softlockup_all_cpu_backtrace=
4038 [KNL] Should the soft-lockup detector generate
4039 backtraces on all cpus.
4040 Format: <integer>
4041
1da177e4 4042 sonypi.*= [HW] Sony Programmable I/O Control Device driver
395cf969 4043 See Documentation/laptops/sonypi.txt
1da177e4 4044
da285121
DW
4045 spectre_v2= [X86] Control mitigation of Spectre variant 2
4046 (indirect branch speculation) vulnerability.
4047
4048 on - unconditionally enable
4049 off - unconditionally disable
4050 auto - kernel detects whether your CPU model is
4051 vulnerable
4052
4053 Selecting 'on' will, and 'auto' may, choose a
4054 mitigation method at run time according to the
4055 CPU, the available microcode, the setting of the
4056 CONFIG_RETPOLINE configuration option, and the
4057 compiler with which the kernel was built.
4058
4059 Specific mitigations can also be selected manually:
4060
4061 retpoline - replace indirect branches
4062 retpoline,generic - google's original retpoline
4063 retpoline,amd - AMD-specific minimal thunk
4064
4065 Not specifying this option is equivalent to
4066 spectre_v2=auto.
4067
e63490c8
KRW
4068 spec_store_bypass_disable=
4069 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
4070 (Speculative Store Bypass vulnerability)
4071
4072 Certain CPUs are vulnerable to an exploit against a
4073 a common industry wide performance optimization known
4074 as "Speculative Store Bypass" in which recent stores
4075 to the same memory location may not be observed by
4076 later loads during speculative execution. The idea
4077 is that such stores are unlikely and that they can
4078 be detected prior to instruction retirement at the
4079 end of a particular speculation execution window.
4080
4081 In vulnerable processors, the speculatively forwarded
4082 store can be used in a cache side channel attack, for
4083 example to read memory to which the attacker does not
4084 directly have access (e.g. inside sandboxed code).
4085
4086 This parameter controls whether the Speculative Store
4087 Bypass optimization is used.
4088
c7416003
KC
4089 on - Unconditionally disable Speculative Store Bypass
4090 off - Unconditionally enable Speculative Store Bypass
4091 auto - Kernel detects whether the CPU model contains an
4092 implementation of Speculative Store Bypass and
4093 picks the most appropriate mitigation. If the
4094 CPU is not vulnerable, "off" is selected. If the
4095 CPU is vulnerable the default mitigation is
4096 architecture and Kconfig dependent. See below.
4097 prctl - Control Speculative Store Bypass per thread
4098 via prctl. Speculative Store Bypass is enabled
4099 for a process by default. The state of the control
4100 is inherited on fork.
4101 seccomp - Same as "prctl" above, but all seccomp threads
4102 will disable SSB unless they explicitly opt out.
e63490c8
KRW
4103
4104 Not specifying this option is equivalent to
4105 spec_store_bypass_disable=auto.
4106
c7416003
KC
4107 Default mitigations:
4108 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
4109
1da177e4
LT
4110 spia_io_base= [HW,MTD]
4111 spia_fio_base=
4112 spia_pedr=
4113 spia_peddr=
4114
c350c008
PM
4115 srcutree.counter_wrap_check [KNL]
4116 Specifies how frequently to check for
4117 grace-period sequence counter wrap for the
4118 srcu_data structure's ->srcu_gp_seq_needed field.
4119 The greater the number of bits set in this kernel
4120 parameter, the less frequently counter wrap will
4121 be checked for. Note that the bottom two bits
4122 are ignored.
4123
22607d66
PM
4124 srcutree.exp_holdoff [KNL]
4125 Specifies how many nanoseconds must elapse
4126 since the end of the last SRCU grace period for
4127 a given srcu_struct until the next normal SRCU
4128 grace period will be considered for automatic
4129 expediting. Set to zero to disable automatic
4130 expediting.
4131
79db941c
MZ
4132 ssbd= [ARM64,HW]
4133 Speculative Store Bypass Disable control
4134
4135 On CPUs that are vulnerable to the Speculative
4136 Store Bypass vulnerability and offer a
4137 firmware based mitigation, this parameter
4138 indicates how the mitigation should be used:
4139
4140 force-on: Unconditionally enable mitigation for
4141 for both kernel and userspace
4142 force-off: Unconditionally disable mitigation for
4143 for both kernel and userspace
4144 kernel: Always enable mitigation in the
4145 kernel, and offer a prctl interface
4146 to allow userspace to register its
4147 interest in being mitigated too.
4148
1be7107f
HD
4149 stack_guard_gap= [MM]
4150 override the default stack gap protection. The value
4151 is in page units and it defines how many pages prior
4152 to (for stacks growing down) resp. after (for stacks
4153 growing up) the main stack are reserved for no other
4154 mapping. Default value is 256 pages.
4155
f38f1d2a
SR
4156 stacktrace [FTRACE]
4157 Enabled the stack tracer on boot up.
4158
762e1207
SR
4159 stacktrace_filter=[function-list]
4160 [FTRACE] Limit the functions that the stack tracer
4161 will trace at boot up. function-list is a comma separated
4162 list of functions. This list can be changed at run
4163 time by the stack_trace_filter file in the debugfs
4164 tracing directory. Note, this enables stack tracing
4165 and the stacktrace above is not needed.
4166
1da177e4
LT
4167 sti= [PARISC,HW]
4168 Format: <num>
4169 Set the STI (builtin display/keyboard on the HP-PARISC
4170 machines) console (graphic card) which should be used
4171 as the initial boot-console.
4172 See also comment in drivers/video/console/sticore.c.
4173
4174 sti_font= [HW]
4175 See comment in drivers/video/console/sticore.c.
4176
4177 stifb= [HW]
4178 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
4179
cbf11071
TM
4180 sunrpc.min_resvport=
4181 sunrpc.max_resvport=
4182 [NFS,SUNRPC]
4183 SunRPC servers often require that client requests
4184 originate from a privileged port (i.e. a port in the
4185 range 0 < portnr < 1024).
4186 An administrator who wishes to reserve some of these
4187 ports for other uses may adjust the range that the
4188 kernel's sunrpc client considers to be privileged
4189 using these two parameters to set the minimum and
4190 maximum port values.
4191
ff3ac5c3
TM
4192 sunrpc.svc_rpc_per_connection_limit=
4193 [NFS,SUNRPC]
4194 Limit the number of requests that the server will
4195 process in parallel from a single connection.
4196 The default value is 0 (no limit).
4197
42a7fc4a
GB
4198 sunrpc.pool_mode=
4199 [NFS]
4200 Control how the NFS server code allocates CPUs to
4201 service thread pools. Depending on how many NICs
4202 you have and where their interrupts are bound, this
4203 option will affect which CPUs will do NFS serving.
4204 Note: this parameter cannot be changed while the
4205 NFS server is running.
4206
4207 auto the server chooses an appropriate mode
4208 automatically using heuristics
4209 global a single global pool contains all CPUs
4210 percpu one pool for each CPU
4211 pernode one pool for each NUMA node (equivalent
4212 to global on non-NUMA machines)
4213
cbf11071
TM
4214 sunrpc.tcp_slot_table_entries=
4215 sunrpc.udp_slot_table_entries=
4216 [NFS,SUNRPC]
4217 Sets the upper limit on the number of simultaneous
4218 RPC calls that can be sent from the client to a
4219 server. Increasing these values may allow you to
4220 improve throughput, but will also increase the
4221 amount of memory reserved for use by the client.
4222
1d4a9c17
BN
4223 suspend.pm_test_delay=
4224 [SUSPEND]
4225 Sets the number of seconds to remain in a suspend test
4226 mode before resuming the system (see
4227 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
4228 is set. Default value is 5.
4229
07555ac1 4230 swapaccount=[0|1]
a42c390c
MH
4231 [KNL] Enable accounting of swap in memory resource
4232 controller if no parameter or 1 is given or disable
09c3bcce 4233 it if 0 is given (See Documentation/cgroup-v1/memory.txt)
a42c390c 4234
91fec0f5 4235 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
fff5d992 4236 Format: { <int> | force | noforce }
91fec0f5
JK
4237 <int> -- Number of I/O TLB slabs
4238 force -- force using of bounce buffers even if they
4239 wouldn't be automatically used by the kernel
fff5d992 4240 noforce -- Never use bounce buffers (for debugging)
a9913044 4241
1da177e4
LT
4242 switches= [HW,M68k]
4243
e52eec13
AK
4244 sysfs.deprecated=0|1 [KNL]
4245 Enable/disable old style sysfs layout for old udev
4246 on older distributions. When this option is enabled
4247 very new udev will not work anymore. When this option
4248 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
4249 in older udev will not work anymore.
4250 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
4251 the kernel configuration.
4252
5d6f647f
IM
4253 sysrq_always_enabled
4254 [KNL]
4255 Ignore sysrq setting - this boot parameter will
4256 neutralize any effect of /proc/sys/kernel/sysrq.
4257 Useful for debugging.
4258
747029a5
FF
4259 tcpmhash_entries= [KNL,NET]
4260 Set the number of tcp_metrics_hash slots.
4261 Default value is 8192 or 16384 depending on total
4262 ram pages. This is used to specify the TCP metrics
4263 cache size. See Documentation/networking/ip-sysctl.txt
4264 "tcp_no_metrics_save" section for more details.
4265
1da177e4
LT
4266 tdfx= [HW,DRM]
4267
acc82342 4268 test_suspend= [SUSPEND][,N]
77437fd4 4269 Specify "mem" (for Suspend-to-RAM) or "standby" (for
acc82342
SP
4270 standby suspend) or "freeze" (for suspend type freeze)
4271 as the system sleep state during system startup with
4272 the optional capability to repeat N number of times.
4273 The system is woken from this state using a
4274 wakeup-capable RTC alarm.
77437fd4 4275
1da177e4
LT
4276 thash_entries= [KNL,NET]
4277 Set number of hash buckets for TCP connection
4278
f8707ec9
LB
4279 thermal.act= [HW,ACPI]
4280 -1: disable all active trip points in all thermal zones
4281 <degrees C>: override all lowest active trip points
4282
c52a7419
LB
4283 thermal.crt= [HW,ACPI]
4284 -1: disable all critical trip points in all thermal zones
22a94d79 4285 <degrees C>: override all critical trip points
c52a7419 4286
f5487145
LB
4287 thermal.nocrt= [HW,ACPI]
4288 Set to disable actions on ACPI thermal zone
4289 critical and hot trip points.
4290
72b33ef8
LB
4291 thermal.off= [HW,ACPI]
4292 1: disable ACPI thermal control
4293
a70cdc52
LB
4294 thermal.psv= [HW,ACPI]
4295 -1: disable all passive trip points
ada9cfdd
RD
4296 <degrees C>: override all passive trip points to this
4297 value
a70cdc52 4298
730ff34d
LB
4299 thermal.tzp= [HW,ACPI]
4300 Specify global default ACPI thermal zone polling rate
4301 <deci-seconds>: poll all this frequency
4302 0: no polling (default)
4303
8d32a307
TG
4304 threadirqs [KNL]
4305 Force threading of all interrupt handlers except those
24775d65 4306 marked explicitly IRQF_NO_THREAD.
8d32a307 4307
2ca62b04
KRW
4308 tmem [KNL,XEN]
4309 Enable the Transcendent memory driver if built-in.
4310
4311 tmem.cleancache=0|1 [KNL, XEN]
4312 Default is on (1). Disable the usage of the cleancache
4313 API to send anonymous pages to the hypervisor.
4314
4315 tmem.frontswap=0|1 [KNL, XEN]
4316 Default is on (1). Disable the usage of the frontswap
37d46e15
KRW
4317 API to send swap pages to the hypervisor. If disabled
4318 the selfballooning and selfshrinking are force disabled.
2ca62b04
KRW
4319
4320 tmem.selfballooning=0|1 [KNL, XEN]
4321 Default is on (1). Disable the driving of swap pages
4322 to the hypervisor.
4323
4324 tmem.selfshrinking=0|1 [KNL, XEN]
4325 Default is on (1). Partial swapoff that immediately
4326 transfers pages from Xen hypervisor back to the
4327 kernel based on different criteria.
4328
2b1a61f0
HC
4329 topology= [S390]
4330 Format: {off | on}
4331 Specify if the kernel should make use of the cpu
f65e51d7
SL
4332 topology information if the hardware supports this.
4333 The scheduler will make use of this information and
2b1a61f0 4334 e.g. base its process migration decisions on it.
c9af3fa9 4335 Default is on.
2b1a61f0 4336
2d73bae1
NA
4337 topology_updates= [KNL, PPC, NUMA]
4338 Format: {off}
4339 Specify if the kernel should ignore (off)
4340 topology updates sent by the hypervisor to this
4341 LPAR.
4342
1da177e4
LT
4343 tp720= [HW,PS2]
4344
225a9be2
RA
4345 tpm_suspend_pcr=[HW,TPM]
4346 Format: integer pcr id
4347 Specify that at suspend time, the tpm driver
4348 should extend the specified pcr with zeros,
4349 as a workaround for some chips which fail to
4350 flush the last written pcr on TPM_SaveState.
4351 This will guarantee that all the other pcrs
4352 are saved.
4353
9d612bef 4354 trace_buf_size=nn[KMG]
3e6fb8e9 4355 [FTRACE] will set tracing buffer size on each cpu.
631595fb 4356
020e5f85
LZ
4357 trace_event=[event-list]
4358 [FTRACE] Set and start specified trace events in order
d81749ea
BN
4359 to facilitate early boot debugging. The event-list is a
4360 comma separated list of trace events to enable. See
4361 also Documentation/trace/events.txt
020e5f85 4362
7bcfaf54
SR
4363 trace_options=[option-list]
4364 [FTRACE] Enable or disable tracer options at boot.
4365 The option-list is a comma delimited list of options
4366 that can be enabled or disabled just as if you were
4367 to echo the option name into
4368
4369 /sys/kernel/debug/tracing/trace_options
4370
4371 For example, to enable stacktrace option (to dump the
4372 stack trace of each event), add to the command line:
4373
4374 trace_options=stacktrace
4375
4376 See also Documentation/trace/ftrace.txt "trace options"
4377 section.
4378
0daa2302
SRRH
4379 tp_printk[FTRACE]
4380 Have the tracepoints sent to printk as well as the
4381 tracing ring buffer. This is useful for early boot up
4382 where the system hangs or reboots and does not give the
4383 option for reading the tracing buffer or performing a
4384 ftrace_dump_on_oops.
4385
4386 To turn off having tracepoints sent to printk,
4387 echo 0 > /proc/sys/kernel/tracepoint_printk
4388 Note, echoing 1 into this file without the
4389 tracepoint_printk kernel cmdline option has no effect.
4390
4391 ** CAUTION **
4392
4393 Having tracepoints sent to printk() and activating high
4394 frequency tracepoints such as irq or sched, can cause
4395 the system to live lock.
4396
de7edd31
SRRH
4397 traceoff_on_warning
4398 [FTRACE] enable this option to disable tracing when a
4399 warning is hit. This turns off "tracing_on". Tracing can
4400 be enabled again by echoing '1' into the "tracing_on"
4401 file located in /sys/kernel/debug/tracing/
4402
4403 This option is useful, as it disables the trace before
4404 the WARNING dump is called, which prevents the trace to
4405 be filled with content caused by the warning output.
4406
4407 This option can also be set at run time via the sysctl
4408 option: kernel/traceoff_on_warning
4409
fcf4d821
JK
4410 transparent_hugepage=
4411 [KNL]
4412 Format: [always|madvise|never]
4413 Can be used to control the default behavior of the system
4414 with respect to transparent hugepages.
4415 See Documentation/vm/transhuge.txt for more details.
4416
d3b8f889 4417 tsc= Disable clocksource stability checks for TSC.
395628ef
AK
4418 Format: <string>
4419 [x86] reliable: mark tsc clocksource as reliable, this
d3b8f889
JS
4420 disables clocksource verification at runtime, as well
4421 as the stability checks done at bootup. Used to enable
4422 high-resolution timer mode on older hardware, and in
4423 virtualized environment.
e82b8e4e
VP
4424 [x86] noirqtime: Do not use TSC to do irq accounting.
4425 Used to run time disable IRQ_TIME_ACCOUNTING on any
4426 platforms where RDTSC is slow and this accounting
4427 can add overhead.
6be53520
DL
4428 [x86] unstable: mark the TSC clocksource as unstable, this
4429 marks the TSC unconditionally unstable at bootup and
4430 avoids any further wobbles once the TSC watchdog notices.
395628ef 4431
a9913044
RD
4432 turbografx.map[2|3]= [HW,JOY]
4433 TurboGraFX parallel port interface
4434 Format:
4435 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
1752118d 4436 See also Documentation/input/devices/joystick-parport.rst
1da177e4 4437
b6935f8c 4438 udbg-immortal [PPC] When debugging early kernel crashes that
e52347bd 4439 happen after console_init() and before a proper
b6935f8c
CK
4440 console driver takes over, this boot options might
4441 help "seeing" what's going on.
4442
f86dcc5a
ED
4443 uhash_entries= [KNL,NET]
4444 Set number of hash buckets for UDP/UDP-Lite connections
4445
5f8364b7
AS
4446 uhci-hcd.ignore_oc=
4447 [USB] Ignore overcurrent events (default N).
4448 Some badly-designed motherboards generate lots of
4449 bogus events, for ports that aren't wired to
4450 anything. Set this parameter to avoid log spamming.
4451 Note that genuine overcurrent events won't be
4452 reported either.
4453
e3a61b0a 4454 unknown_nmi_panic
44a4dcf7 4455 [X86] Cause panic on unknown NMI.
e3a61b0a 4456
c4fc2342
CDH
4457 usbcore.authorized_default=
4458 [USB] Default USB device authorization:
4459 (default -1 = authorized except for wireless USB,
4460 0 = not authorized, 1 = authorized)
4461
b5e795f8
AS
4462 usbcore.autosuspend=
4463 [USB] The autosuspend time delay (in seconds) used
4464 for newly-detected USB devices (default 2). This
4465 is the time required before an idle device will be
4466 autosuspended. Devices for which the delay is set
eaafbc3a 4467 to a negative value won't be autosuspended at all.
b5e795f8 4468
fd7c519d
JK
4469 usbcore.usbfs_snoop=
4470 [USB] Set to log all usbfs traffic (default 0 = off).
4471
0290cc9f
AS
4472 usbcore.usbfs_snoop_max=
4473 [USB] Maximum number of bytes to snoop in each URB
4474 (default = 65536).
4475
fd7c519d
JK
4476 usbcore.blinkenlights=
4477 [USB] Set to cycle leds on hubs (default 0 = off).
4478
4479 usbcore.old_scheme_first=
4480 [USB] Start with the old device initialization
4481 scheme (default 0 = off).
4482
3f5eb8d5
AS
4483 usbcore.usbfs_memory_mb=
4484 [USB] Memory limit (in MB) for buffers allocated by
4485 usbfs (default = 16, 0 = max = 2047).
4486
fd7c519d
JK
4487 usbcore.use_both_schemes=
4488 [USB] Try the other device initialization scheme
4489 if the first one fails (default 1 = enabled).
4490
4491 usbcore.initial_descriptor_timeout=
4492 [USB] Specifies timeout for the initial 64-byte
4493 USB_REQ_GET_DESCRIPTOR request in milliseconds
4494 (default 5000 = 5.0 seconds).
4495
40d58148
ON
4496 usbcore.nousb [USB] Disable the USB subsystem
4497
389cb8cf
KHF
4498 usbcore.quirks=
4499 [USB] A list of quirk entries to augment the built-in
4500 usb core quirk list. List entries are separated by
4501 commas. Each entry has the form
4502 VendorID:ProductID:Flags. The IDs are 4-digit hex
4503 numbers and Flags is a set of letters. Each letter
4504 will change the built-in quirk; setting it if it is
4505 clear and clearing it if it is set. The letters have
4506 the following meanings:
4507 a = USB_QUIRK_STRING_FETCH_255 (string
4508 descriptors must not be fetched using
4509 a 255-byte read);
4510 b = USB_QUIRK_RESET_RESUME (device can't resume
4511 correctly so reset it instead);
4512 c = USB_QUIRK_NO_SET_INTF (device can't handle
4513 Set-Interface requests);
4514 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
4515 handle its Configuration or Interface
4516 strings);
4517 e = USB_QUIRK_RESET (device can't be reset
4518 (e.g morph devices), don't use reset);
4519 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
4520 more interface descriptions than the
4521 bNumInterfaces count, and can't handle
4522 talking to these interfaces);
4523 g = USB_QUIRK_DELAY_INIT (device needs a pause
4524 during initialization, after we read
4525 the device descriptor);
4526 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
4527 high speed and super speed interrupt
4528 endpoints, the USB 2.0 and USB 3.0 spec
4529 require the interval in microframes (1
4530 microframe = 125 microseconds) to be
4531 calculated as interval = 2 ^
4532 (bInterval-1).
4533 Devices with this quirk report their
4534 bInterval as the result of this
4535 calculation instead of the exponent
4536 variable used in the calculation);
4537 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
4538 handle device_qualifier descriptor
4539 requests);
4540 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
4541 generates spurious wakeup, ignore
4542 remote wakeup capability);
4543 k = USB_QUIRK_NO_LPM (device can't handle Link
4544 Power Management);
4545 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
4546 (Device reports its bInterval as linear
4547 frames instead of the USB 2.0
4548 calculation);
4549 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
4550 to be disconnected before suspend to
68d8e350
KHF
4551 prevent spurious wakeup);
4552 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
4553 pause after every control message);
389cb8cf
KHF
4554 Example: quirks=0781:5580:bk,0a5c:5834:gij
4555
1da177e4
LT
4556 usbhid.mousepoll=
4557 [USBHID] The interval which mice are to be polled at.
a9913044 4558
933bfe4d
TJ
4559 usbhid.jspoll=
4560 [USBHID] The interval which joysticks are to be polled at.
4561
d4f373e5
AS
4562 usb-storage.delay_use=
4563 [UMS] The delay in seconds before a new device is
19101954 4564 scanned for Logical Units (default 1).
d4f373e5
AS
4565
4566 usb-storage.quirks=
4567 [UMS] A list of quirks entries to supplement or
4568 override the built-in unusual_devs list. List
4569 entries are separated by commas. Each entry has
4570 the form VID:PID:Flags where VID and PID are Vendor
4571 and Product ID values (4-digit hex numbers) and
4572 Flags is a set of characters, each corresponding
4573 to a common usb-storage quirk flag as follows:
c838ea46
AS
4574 a = SANE_SENSE (collect more than 18 bytes
4575 of sense data);
a0bb1081
AS
4576 b = BAD_SENSE (don't collect more than 18
4577 bytes of sense data);
d4f373e5
AS
4578 c = FIX_CAPACITY (decrease the reported
4579 device capacity by one sector);
5116901d
KR
4580 d = NO_READ_DISC_INFO (don't use
4581 READ_DISC_INFO command);
4582 e = NO_READ_CAPACITY_16 (don't use
4583 READ_CAPACITY_16 command);
734016b0
HG
4584 f = NO_REPORT_OPCODES (don't use report opcodes
4585 command, uas only);
ee136af4
HG
4586 g = MAX_SECTORS_240 (don't transfer more than
4587 240 sectors at a time, uas only);
c838ea46
AS
4588 h = CAPACITY_HEURISTICS (decrease the
4589 reported device capacity by one
4590 sector if the number is odd);
d4f373e5
AS
4591 i = IGNORE_DEVICE (don't bind to this
4592 device);
13630746
HG
4593 j = NO_REPORT_LUNS (don't use report luns
4594 command, uas only);
d4f373e5
AS
4595 l = NOT_LOCKABLE (don't try to lock and
4596 unlock ejectable media);
4597 m = MAX_SECTORS_64 (don't transfer more
4598 than 64 sectors = 32 KB at a time);
21c13a4f
AS
4599 n = INITIAL_READ10 (force a retry of the
4600 initial READ(10) command);
c838ea46
AS
4601 o = CAPACITY_OK (accept the capacity
4602 reported by the device);
eaa05dfc
NJ
4603 p = WRITE_CACHE (the device cache is ON
4604 by default);
d4f373e5
AS
4605 r = IGNORE_RESIDUE (the device reports
4606 bogus residue values);
4607 s = SINGLE_LUN (the device has only one
4608 Logical Unit);
59307852
HG
4609 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
4610 commands, uas only);
b6089f19 4611 u = IGNORE_UAS (don't bind to the uas driver);
d4f373e5
AS
4612 w = NO_WP_DETECT (don't test whether the
4613 medium is write-protected).
050bc4e8
ON
4614 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
4615 even if the device claims no cache)
d4f373e5
AS
4616 Example: quirks=0419:aaf5:rl,0421:0433:rc
4617
ac1667db
SB
4618 user_debug= [KNL,ARM]
4619 Format: <int>
4620 See arch/arm/Kconfig.debug help text.
4621 1 - undefined instruction events
4622 2 - system calls
4623 4 - invalid data aborts
4624 8 - SIGSEGV faults
4625 16 - SIGBUS faults
4626 Example: user_debug=31
4627
14315592
IC
4628 userpte=
4629 [X86] Flags controlling user PTE allocations.
4630
4631 nohigh = do not allocate PTE pages in
4632 HIGHMEM regardless of setting
4633 of CONFIG_HIGHPTE.
4634
6cececfc 4635 vdso= [X86,SH]
b0b49f26
AL
4636 On X86_32, this is an alias for vdso32=. Otherwise:
4637
4638 vdso=1: enable VDSO (the default)
e6e5494c
IM
4639 vdso=0: disable VDSO mapping
4640
b0b49f26
AL
4641 vdso32= [X86] Control the 32-bit vDSO
4642 vdso32=1: enable 32-bit VDSO
4643 vdso32=0 or vdso32=2: disable 32-bit VDSO
4644
4645 See the help text for CONFIG_COMPAT_VDSO for more
4646 details. If CONFIG_COMPAT_VDSO is set, the default is
4647 vdso32=0; otherwise, the default is vdso32=1.
4648
4649 For compatibility with older kernels, vdso32=2 is an
4650 alias for vdso32=0.
4651
4652 Try vdso32=0 if you encounter an error that says:
4653 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
af65d648 4654
d080d397
YI
4655 vector= [IA-64,SMP]
4656 vector=percpu: enable percpu vector domain
4657
1da177e4
LT
4658 video= [FB] Frame buffer configuration
4659 See Documentation/fb/modedb.txt.
4660
3afe6dab
AL
4661 video.brightness_switch_enabled= [0,1]
4662 If set to 1, on receiving an ACPI notify event
4663 generated by hotkey, video driver will adjust brightness
4664 level and then send out the event to user space through
4665 the allocated input device; If set to 0, video driver
4666 will only send out the event without touching backlight
4667 brightness level.
2843768b 4668 default: 1
3afe6dab 4669
81a054ce
PM
4670 virtio_mmio.device=
4671 [VMMIO] Memory mapped virtio (platform) device.
4672
4673 <size>@<baseaddr>:<irq>[:<id>]
4674 where:
4675 <size> := size (can use standard suffixes
4676 like K, M and G)
4677 <baseaddr> := physical base address
4678 <irq> := interrupt number (as passed to
4679 request_irq())
4680 <id> := (optional) platform device id
4681 example:
4682 virtio_mmio.device=1K@0x100b0000:48:7
4683
4684 Can be used multiple times for multiple devices.
4685
cd4f0ef7 4686 vga= [BOOT,X86-32] Select a particular video mode
954a8b81 4687 See Documentation/x86/boot.txt and
a9913044 4688 Documentation/svga.txt.
1da177e4
LT
4689 Use vga=ask for menu.
4690 This is actually a boot loader parameter; the value is
4691 passed to the kernel using a special protocol.
4692
a9913044 4693 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
1da177e4
LT
4694 size of <nn>. This can be used to increase the
4695 minimum size (128MB on x86). It can also be used to
4696 decrease the size and leave more room for directly
4697 mapped kernel RAM.
4698
3f429842
HC
4699 vmcp_cma=nn[MG] [KNL,S390]
4700 Sets the memory size reserved for contiguous memory
4701 allocations for the vmcp device driver.
4702
585c3047
PO
4703 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
4704 Format: <command>
1da177e4 4705
585c3047
PO
4706 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
4707 Format: <command>
4708
4709 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
4710 Format: <command>
a9913044 4711
3ae36655
AL
4712 vsyscall= [X86-64]
4713 Controls the behavior of vsyscalls (i.e. calls to
4714 fixed addresses of 0xffffffffff600x00 from legacy
4715 code). Most statically-linked binaries and older
4716 versions of glibc use these calls. Because these
4717 functions are at fixed addresses, they make nice
4718 targets for exploits that can control RIP.
4719
2e57ae05
AL
4720 emulate [default] Vsyscalls turn into traps and are
4721 emulated reasonably safely.
3ae36655 4722
2e57ae05 4723 native Vsyscalls are native syscall instructions.
3ae36655
AL
4724 This is a little bit faster than trapping
4725 and makes a few dynamic recompilers work
4726 better than they would in emulation mode.
4727 It also makes exploits much easier to write.
4728
4729 none Vsyscalls don't work at all. This makes
4730 them quite hard to use for exploits but
4731 might break your system.
4732
3855ae1c
CL
4733 vt.color= [VT] Default text color.
4734 Format: 0xYX, X = foreground, Y = background.
4735 Default: 0x07 = light gray on black.
4736
9ea9a886
CL
4737 vt.cur_default= [VT] Default cursor shape.
4738 Format: 0xCCBBAA, where AA, BB, and CC are the same as
4739 the parameters of the <Esc>[?A;B;Cc escape sequence;
4740 see VGA-softcursor.txt. Default: 2 = underline.
4741
0cb55ad2
RD
4742 vt.default_blu= [VT]
4743 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
4744 Change the default blue palette of the console.
4745 This is a 16-member array composed of values
4746 ranging from 0-255.
4747
4748 vt.default_grn= [VT]
4749 Format: <green0>,<green1>,<green2>,...,<green15>
4750 Change the default green palette of the console.
4751 This is a 16-member array composed of values
4752 ranging from 0-255.
4753
4754 vt.default_red= [VT]
4755 Format: <red0>,<red1>,<red2>,...,<red15>
4756 Change the default red palette of the console.
4757 This is a 16-member array composed of values
4758 ranging from 0-255.
4759
4760 vt.default_utf8=
4761 [VT]
4762 Format=<0|1>
4763 Set system-wide default UTF-8 mode for all tty's.
4764 Default is 1, i.e. UTF-8 mode is enabled for all
4765 newly opened terminals.
4766
f6c06b68
MG
4767 vt.global_cursor_default=
4768 [VT]
4769 Format=<-1|0|1>
4770 Set system-wide default for whether a cursor
4771 is shown on new VTs. Default is -1,
4772 i.e. cursors will be created by default unless
4773 overridden by individual drivers. 0 will hide
4774 cursors, 1 will display them.
4775
3855ae1c
CL
4776 vt.italic= [VT] Default color for italic text; 0-15.
4777 Default: 2 = green.
4778
4779 vt.underline= [VT] Default color for underlined text; 0-15.
4780 Default: 3 = cyan.
4781
4724ba57
RD
4782 watchdog timers [HW,WDT] For information on watchdog timers,
4783 see Documentation/watchdog/watchdog-parameters.txt
4784 or other driver-specific files in the
4785 Documentation/watchdog/ directory.
1da177e4 4786
82607adc
TH
4787 workqueue.watchdog_thresh=
4788 If CONFIG_WQ_WATCHDOG is configured, workqueue can
4789 warn stall conditions and dump internal state to
4790 help debugging. 0 disables workqueue stall
4791 detection; otherwise, it's the stall threshold
4792 duration in seconds. The default value is 30 and
4793 it can be updated at runtime by writing to the
4794 corresponding sysfs file.
4795
d55262c4
TH
4796 workqueue.disable_numa
4797 By default, all work items queued to unbound
4798 workqueues are affine to the NUMA nodes they're
4799 issued on, which results in better behavior in
4800 general. If NUMA affinity needs to be disabled for
4801 whatever reason, this option can be used. Note
4802 that this also can be controlled per-workqueue for
4803 workqueues visible under /sys/bus/workqueue/.
4804
cee22a15
VK
4805 workqueue.power_efficient
4806 Per-cpu workqueues are generally preferred because
4807 they show better performance thanks to cache
4808 locality; unfortunately, per-cpu workqueues tend to
4809 be more power hungry than unbound workqueues.
4810
4811 Enabling this makes the per-cpu workqueues which
4812 were observed to contribute significantly to power
4813 consumption unbound, leading to measurably lower
4814 power usage at the cost of small performance
4815 overhead.
4816
4817 The default value of this parameter is determined by
4818 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
4819
f303fccb
TH
4820 workqueue.debug_force_rr_cpu
4821 Workqueue used to implicitly guarantee that work
4822 items queued without explicit CPU specified are put
4823 on the local CPU. This guarantee is no longer true
4824 and while local CPU is still preferred work items
4825 may be put on foreign CPUs. This debug option
4826 forces round-robin CPU selection to flush out
4827 usages which depend on the now broken guarantee.
4828 When enabled, memory and cache locality will be
4829 impacted.
4830
0cb55ad2
RD
4831 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
4832 default x2apic cluster mode on platforms
4833 supporting x2apic.
4834
712b6aa8
KS
4835 x86_intel_mid_timer= [X86-32,APBT]
4836 Choose timer option for x86 Intel MID platform.
bb24c471
JP
4837 Two valid options are apbt timer only and lapic timer
4838 plus one apbt timer for broadcast timer.
712b6aa8 4839 x86_intel_mid_timer=apbt_only | lapic_and_apbt
bb24c471 4840
c70727a5
JG
4841 xen_512gb_limit [KNL,X86-64,XEN]
4842 Restricts the kernel running paravirtualized under Xen
4843 to use only up to 512 GB of RAM. The reason to do so is
4844 crash analysis tools and Xen tools for doing domain
4845 save/restore/migration must be enabled to handle larger
4846 domains.
4847
c1c5413a
SS
4848 xen_emul_unplug= [HW,X86,XEN]
4849 Unplug Xen emulated devices
4850 Format: [unplug0,][unplug1]
4851 ide-disks -- unplug primary master IDE devices
4852 aux-ide-disks -- unplug non-primary-master IDE devices
4853 nics -- unplug network devices
4854 all -- unplug all emulated devices (NICs and IDE disks)
1dc7ce99
IC
4855 unnecessary -- unplugging emulated devices is
4856 unnecessary even if the host did not respond to
4857 the unplug protocol
c93a4dfb 4858 never -- do not unplug even if version check succeeds
c1c5413a 4859
15a3eac0
KRW
4860 xen_nopvspin [X86,XEN]
4861 Disables the ticketlock slowpath using Xen PV
4862 optimizations.
4863
8d693b91
KRW
4864 xen_nopv [X86]
4865 Disables the PV optimizations forcing the HVM guest to
4866 run as generic HVM guest with no PV drivers.
4867
1da177e4 4868 xirc2ps_cs= [NET,PCMCIA]
a9913044
RD
4869 Format:
4870 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]