Loading...
1 acpi= [HW,ACPI,X86,ARM64]
2 Advanced Configuration and Power Interface
3 Format: { force | on | off | strict | noirq | rsdt |
4 copy_dsdt }
5 force -- enable ACPI if default was off
6 on -- enable ACPI but allow fallback to DT [arm64]
7 off -- disable ACPI if default was on
8 noirq -- do not use ACPI for IRQ routing
9 strict -- Be less tolerant of platforms that are not
10 strictly ACPI specification compliant.
11 rsdt -- prefer RSDT over (default) XSDT
12 copy_dsdt -- copy DSDT to memory
13 For ARM64, ONLY "acpi=off", "acpi=on" or "acpi=force"
14 are available
15
16 See also Documentation/power/runtime_pm.txt, pci=noacpi
17
18 acpi_apic_instance= [ACPI, IOAPIC]
19 Format: <int>
20 2: use 2nd APIC table, if available
21 1,0: use 1st APIC table
22 default: 0
23
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
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
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
45 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
46 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
47 Format: <int>
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, ...
55 The debug_level mask defaults to "info". See
56 Documentation/acpi/debug.txt for more information about
57 debug layers and levels.
58
59 Enable processor driver info messages:
60 acpi.debug_layer=0x20000000
61 Enable PCI/PCI interrupt routing info messages:
62 acpi.debug_layer=0x400000
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
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.
72
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
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
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
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
118 acpi_no_auto_serialize [HW,ACPI]
119 Disable auto-serialization of AML methods
120 AML control methods that contain the opcodes to create
121 named objects will be marked as "Serialized" by the
122 auto-serialization feature.
123 This feature is enabled by default.
124 This option allows to turn off the feature.
125
126 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
127 kernels.
128
129 acpi_no_static_ssdt [HW,ACPI]
130 Disable installation of static SSDTs at early boot time
131 By default, SSDTs contained in the RSDT/XSDT will be
132 installed automatically and they will appear under
133 /sys/firmware/acpi/tables.
134 This option turns off this feature.
135 Note that specifying this option does not affect
136 dynamic table installation which will install SSDT
137 tables to /sys/firmware/acpi/tables/dynamic.
138
139 acpi_rsdp= [ACPI,EFI,KEXEC]
140 Pass the RSDP address to the kernel, mostly used
141 on machines running EFI runtime service to boot the
142 second kernel for kdump.
143
144 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
145 Format: To spoof as Windows 98: ="Microsoft Windows"
146
147 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
148 of 2 which is mandated by ACPI 6) as the supported ACPI
149 specification revision (when using this switch, it may
150 be necessary to carry out a cold reboot _twice_ in a
151 row to make it take effect on the platform firmware).
152
153 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
154 acpi_osi="string1" # add string1
155 acpi_osi="!string2" # remove string2
156 acpi_osi=!* # remove all strings
157 acpi_osi=! # disable all built-in OS vendor
158 strings
159 acpi_osi=!! # enable all built-in OS vendor
160 strings
161 acpi_osi= # disable all strings
162
163 'acpi_osi=!' can be used in combination with single or
164 multiple 'acpi_osi="string1"' to support specific OS
165 vendor string(s). Note that such command can only
166 affect the default state of the OS vendor strings, thus
167 it cannot affect the default state of the feature group
168 strings and the current state of the OS vendor strings,
169 specifying it multiple times through kernel command line
170 is meaningless. This command is useful when one do not
171 care about the state of the feature group strings which
172 should be controlled by the OSPM.
173 Examples:
174 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
175 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
176 can make '_OSI("Windows 2000")' TRUE.
177
178 'acpi_osi=' cannot be used in combination with other
179 'acpi_osi=' command lines, the _OSI method will not
180 exist in the ACPI namespace. NOTE that such command can
181 only affect the _OSI support state, thus specifying it
182 multiple times through kernel command line is also
183 meaningless.
184 Examples:
185 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
186 FALSE.
187
188 'acpi_osi=!*' can be used in combination with single or
189 multiple 'acpi_osi="string1"' to support specific
190 string(s). Note that such command can affect the
191 current state of both the OS vendor strings and the
192 feature group strings, thus specifying it multiple times
193 through kernel command line is meaningful. But it may
194 still not able to affect the final state of a string if
195 there are quirks related to this string. This command
196 is useful when one want to control the state of the
197 feature group strings to debug BIOS issues related to
198 the OSPM features.
199 Examples:
200 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
201 '_OSI("Module Device")' FALSE.
202 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
203 '_OSI("Module Device")' TRUE.
204 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
205 equivalent to
206 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
207 and
208 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
209 they all will make '_OSI("Windows 2000")' TRUE.
210
211 acpi_pm_good [X86]
212 Override the pmtimer bug detection: force the kernel
213 to assume that this machine's pmtimer latches its value
214 and always returns good values.
215
216 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
217 Format: { level | edge | high | low }
218
219 acpi_skip_timer_override [HW,ACPI]
220 Recognize and ignore IRQ0/pin2 Interrupt Override.
221 For broken nForce2 BIOS resulting in XT-PIC timer.
222
223 acpi_sleep= [HW,ACPI] Sleep options
224 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
225 old_ordering, nonvs, sci_force_enable, nobl }
226 See Documentation/power/video.txt for information on
227 s3_bios and s3_mode.
228 s3_beep is for debugging; it makes the PC's speaker beep
229 as soon as the kernel's real-mode entry point is called.
230 s4_nohwsig prevents ACPI hardware signature from being
231 used during resume from hibernation.
232 old_ordering causes the ACPI 1.0 ordering of the _PTS
233 control method, with respect to putting devices into
234 low power states, to be enforced (the ACPI 2.0 ordering
235 of _PTS is used by default).
236 nonvs prevents the kernel from saving/restoring the
237 ACPI NVS memory during suspend/hibernation and resume.
238 sci_force_enable causes the kernel to set SCI_EN directly
239 on resume from S1/S3 (which is against the ACPI spec,
240 but some broken systems don't work without it).
241 nobl causes the internal blacklist of systems known to
242 behave incorrectly in some ways with respect to system
243 suspend and resume to be ignored (use wisely).
244
245 acpi_use_timer_override [HW,ACPI]
246 Use timer override. For some broken Nvidia NF5 boards
247 that require a timer override, but don't have HPET
248
249 add_efi_memmap [EFI; X86] Include EFI memory map in
250 kernel's map of available physical RAM.
251
252 agp= [AGP]
253 { off | try_unsupported }
254 off: disable AGP support
255 try_unsupported: try to drive unsupported chipsets
256 (may crash computer or cause data corruption)
257
258 ALSA [HW,ALSA]
259 See Documentation/sound/alsa/alsa-parameters.txt
260
261 alignment= [KNL,ARM]
262 Allow the default userspace alignment fault handler
263 behaviour to be specified. Bit 0 enables warnings,
264 bit 1 enables fixups, and bit 2 sends a segfault.
265
266 align_va_addr= [X86-64]
267 Align virtual addresses by clearing slice [14:12] when
268 allocating a VMA at process creation time. This option
269 gives you up to 3% performance improvement on AMD F15h
270 machines (where it is enabled by default) for a
271 CPU-intensive style benchmark, and it can vary highly in
272 a microbenchmark depending on workload and compiler.
273
274 32: only for 32-bit processes
275 64: only for 64-bit processes
276 on: enable for both 32- and 64-bit processes
277 off: disable for both 32- and 64-bit processes
278
279 alloc_snapshot [FTRACE]
280 Allocate the ftrace snapshot buffer on boot up when the
281 main buffer is allocated. This is handy if debugging
282 and you need to use tracing_snapshot() on boot up, and
283 do not want to use tracing_snapshot_alloc() as it needs
284 to be done where GFP_KERNEL allocations are allowed.
285
286 amd_iommu= [HW,X86-64]
287 Pass parameters to the AMD IOMMU driver in the system.
288 Possible values are:
289 fullflush - enable flushing of IO/TLB entries when
290 they are unmapped. Otherwise they are
291 flushed before they will be reused, which
292 is a lot of faster
293 off - do not initialize any AMD IOMMU found in
294 the system
295 force_isolation - Force device isolation for all
296 devices. The IOMMU driver is not
297 allowed anymore to lift isolation
298 requirements as needed. This option
299 does not override iommu=pt
300
301 amd_iommu_dump= [HW,X86-64]
302 Enable AMD IOMMU driver option to dump the ACPI table
303 for AMD IOMMU. With this option enabled, AMD IOMMU
304 driver will print ACPI tables for AMD IOMMU during
305 IOMMU initialization.
306
307 amd_iommu_intr= [HW,X86-64]
308 Specifies one of the following AMD IOMMU interrupt
309 remapping modes:
310 legacy - Use legacy interrupt remapping mode.
311 vapic - Use virtual APIC mode, which allows IOMMU
312 to inject interrupts directly into guest.
313 This mode requires kvm-amd.avic=1.
314 (Default when IOMMU HW support is present.)
315
316 amijoy.map= [HW,JOY] Amiga joystick support
317 Map of devices attached to JOY0DAT and JOY1DAT
318 Format: <a>,<b>
319 See also Documentation/input/joydev/joystick.rst
320
321 analog.map= [HW,JOY] Analog joystick and gamepad support
322 Specifies type or capabilities of an analog joystick
323 connected to one of 16 gameports
324 Format: <type1>,<type2>,..<type16>
325
326 apc= [HW,SPARC]
327 Power management functions (SPARCstation-4/5 + deriv.)
328 Format: noidle
329 Disable APC CPU standby support. SPARCstation-Fox does
330 not play well with APC CPU idle - disable it if you have
331 APC and your system crashes randomly.
332
333 apic= [APIC,X86] Advanced Programmable Interrupt Controller
334 Change the output verbosity whilst booting
335 Format: { quiet (default) | verbose | debug }
336 Change the amount of debugging information output
337 when initialising the APIC and IO-APIC components.
338 For X86-32, this can also be used to specify an APIC
339 driver name.
340 Format: apic=driver_name
341 Examples: apic=bigsmp
342
343 apic_extnmi= [APIC,X86] External NMI delivery setting
344 Format: { bsp (default) | all | none }
345 bsp: External NMI is delivered only to CPU 0
346 all: External NMIs are broadcast to all CPUs as a
347 backup of CPU 0
348 none: External NMI is masked for all CPUs. This is
349 useful so that a dump capture kernel won't be
350 shot down by NMI
351
352 autoconf= [IPV6]
353 See Documentation/networking/ipv6.txt.
354
355 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
356 Limit apic dumping. The parameter defines the maximal
357 number of local apics being dumped. Also it is possible
358 to set it to "all" by meaning -- no limit here.
359 Format: { 1 (default) | 2 | ... | all }.
360 The parameter valid if only apic=debug or
361 apic=verbose is specified.
362 Example: apic=debug show_lapic=all
363
364 apm= [APM] Advanced Power Management
365 See header of arch/x86/kernel/apm_32.c.
366
367 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
368 Format: <io>,<irq>,<nodeID>
369
370 ataflop= [HW,M68k]
371
372 atarimouse= [HW,MOUSE] Atari Mouse
373
374 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
375 EzKey and similar keyboards
376
377 atkbd.reset= [HW] Reset keyboard during initialization
378
379 atkbd.set= [HW] Select keyboard code set
380 Format: <int> (2 = AT (default), 3 = PS/2)
381
382 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
383 keyboards
384
385 atkbd.softraw= [HW] Choose between synthetic and real raw mode
386 Format: <bool> (0 = real, 1 = synthetic (default))
387
388 atkbd.softrepeat= [HW]
389 Use software keyboard repeat
390
391 audit= [KNL] Enable the audit sub-system
392 Format: { "0" | "1" | "off" | "on" }
393 0 | off - kernel audit is disabled and can not be
394 enabled until the next reboot
395 unset - kernel audit is initialized but disabled and
396 will be fully enabled by the userspace auditd.
397 1 | on - kernel audit is initialized and partially
398 enabled, storing at most audit_backlog_limit
399 messages in RAM until it is fully enabled by the
400 userspace auditd.
401 Default: unset
402
403 audit_backlog_limit= [KNL] Set the audit queue size limit.
404 Format: <int> (must be >=0)
405 Default: 64
406
407 bau= [X86_UV] Enable the BAU on SGI UV. The default
408 behavior is to disable the BAU (i.e. bau=0).
409 Format: { "0" | "1" }
410 0 - Disable the BAU.
411 1 - Enable the BAU.
412 unset - Disable the BAU.
413
414 baycom_epp= [HW,AX25]
415 Format: <io>,<mode>
416
417 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
418 Format: <io>,<mode>
419 See header of drivers/net/hamradio/baycom_par.c.
420
421 baycom_ser_fdx= [HW,AX25]
422 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
423 Format: <io>,<irq>,<mode>[,<baud>]
424 See header of drivers/net/hamradio/baycom_ser_fdx.c.
425
426 baycom_ser_hdx= [HW,AX25]
427 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
428 Format: <io>,<irq>,<mode>
429 See header of drivers/net/hamradio/baycom_ser_hdx.c.
430
431 blkdevparts= Manual partition parsing of block device(s) for
432 embedded devices based on command line input.
433 See Documentation/block/cmdline-partition.txt
434
435 boot_delay= Milliseconds to delay each printk during boot.
436 Values larger than 10 seconds (10000) are changed to
437 no delay (0).
438 Format: integer
439
440 bootmem_debug [KNL] Enable bootmem allocator debug messages.
441
442 bert_disable [ACPI]
443 Disable BERT OS support on buggy BIOSes.
444
445 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
446 bttv.radio= Most important insmod options are available as
447 kernel args too.
448 bttv.pll= See Documentation/media/v4l-drivers/bttv.rst
449 bttv.tuner=
450
451 bulk_remove=off [PPC] This parameter disables the use of the pSeries
452 firmware feature for flushing multiple hpte entries
453 at a time.
454
455 c101= [NET] Moxa C101 synchronous serial card
456
457 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
458 Sometimes CPU hardware bugs make them report the cache
459 size incorrectly. The kernel will attempt work arounds
460 to fix known problems, but for some CPUs it is not
461 possible to determine what the correct size should be.
462 This option provides an override for these situations.
463
464 ca_keys= [KEYS] This parameter identifies a specific key(s) on
465 the system trusted keyring to be used for certificate
466 trust validation.
467 format: { id:<keyid> | builtin }
468
469 cca= [MIPS] Override the kernel pages' cache coherency
470 algorithm. Accepted values range from 0 to 7
471 inclusive. See arch/mips/include/asm/pgtable-bits.h
472 for platform specific values (SB1, Loongson3 and
473 others).
474
475 ccw_timeout_log [S390]
476 See Documentation/s390/CommonIO for details.
477
478 cgroup_disable= [KNL] Disable a particular controller
479 Format: {name of the controller(s) to disable}
480 The effects of cgroup_disable=foo are:
481 - foo isn't auto-mounted if you mount all cgroups in
482 a single hierarchy
483 - foo isn't visible as an individually mountable
484 subsystem
485 {Currently only "memory" controller deal with this and
486 cut the overhead, others just disable the usage. So
487 only cgroup_disable=memory is actually worthy}
488
489 cgroup_no_v1= [KNL] Disable one, multiple, all cgroup controllers in v1
490 Format: { controller[,controller...] | "all" }
491 Like cgroup_disable, but only applies to cgroup v1;
492 the blacklisted controllers remain available in cgroup2.
493
494 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
495 Format: <string>
496 nosocket -- Disable socket memory accounting.
497 nokmem -- Disable kernel memory accounting.
498
499 checkreqprot [SELINUX] Set initial checkreqprot flag value.
500 Format: { "0" | "1" }
501 See security/selinux/Kconfig help text.
502 0 -- check protection applied by kernel (includes
503 any implied execute protection).
504 1 -- check protection requested by application.
505 Default value is set via a kernel config option.
506 Value can be changed at runtime via
507 /selinux/checkreqprot.
508
509 cio_ignore= [S390]
510 See Documentation/s390/CommonIO for details.
511 clk_ignore_unused
512 [CLK]
513 Prevents the clock framework from automatically gating
514 clocks that have not been explicitly enabled by a Linux
515 device driver but are enabled in hardware at reset or
516 by the bootloader/firmware. Note that this does not
517 force such clocks to be always-on nor does it reserve
518 those clocks in any way. This parameter is useful for
519 debug and development, but should not be needed on a
520 platform with proper driver support. For more
521 information, see Documentation/clk.txt.
522
523 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
524 [Deprecated]
525 Forces specified clocksource (if available) to be used
526 when calculating gettimeofday(). If specified
527 clocksource is not available, it defaults to PIT.
528 Format: { pit | tsc | cyclone | pmtmr }
529
530 clocksource= Override the default clocksource
531 Format: <string>
532 Override the default clocksource and use the clocksource
533 with the name specified.
534 Some clocksource names to choose from, depending on
535 the platform:
536 [all] jiffies (this is the base, fallback clocksource)
537 [ACPI] acpi_pm
538 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
539 pxa_timer,timer3,32k_counter,timer0_1
540 [X86-32] pit,hpet,tsc;
541 scx200_hrt on Geode; cyclone on IBM x440
542 [MIPS] MIPS
543 [PARISC] cr16
544 [S390] tod
545 [SH] SuperH
546 [SPARC64] tick
547 [X86-64] hpet,tsc
548
549 clocksource.arm_arch_timer.evtstrm=
550 [ARM,ARM64]
551 Format: <bool>
552 Enable/disable the eventstream feature of the ARM
553 architected timer so that code using WFE-based polling
554 loops can be debugged more effectively on production
555 systems.
556
557 clearcpuid=BITNUM [X86]
558 Disable CPUID feature X for the kernel. See
559 arch/x86/include/asm/cpufeatures.h for the valid bit
560 numbers. Note the Linux specific bits are not necessarily
561 stable over kernel options, but the vendor specific
562 ones should be.
563 Also note that user programs calling CPUID directly
564 or using the feature without checking anything
565 will still see it. This just prevents it from
566 being used by the kernel or shown in /proc/cpuinfo.
567 Also note the kernel might malfunction if you disable
568 some critical bits.
569
570 cma=nn[MG]@[start[MG][-end[MG]]]
571 [ARM,X86,KNL]
572 Sets the size of kernel global memory area for
573 contiguous memory allocations and optionally the
574 placement constraint by the physical address range of
575 memory allocations. A value of 0 disables CMA
576 altogether. For more information, see
577 include/linux/dma-contiguous.h
578
579 cmo_free_hint= [PPC] Format: { yes | no }
580 Specify whether pages are marked as being inactive
581 when they are freed. This is used in CMO environments
582 to determine OS memory pressure for page stealing by
583 a hypervisor.
584 Default: yes
585
586 coherent_pool=nn[KMG] [ARM,KNL]
587 Sets the size of memory pool for coherent, atomic dma
588 allocations, by default set to 256K.
589
590 code_bytes [X86] How many bytes of object code to print
591 in an oops report.
592 Range: 0 - 8192
593 Default: 64
594
595 com20020= [HW,NET] ARCnet - COM20020 chipset
596 Format:
597 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
598
599 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
600 Format: <io>[,<irq>]
601
602 com90xx= [HW,NET]
603 ARCnet - COM90xx chipset (memory-mapped buffers)
604 Format: <io>[,<irq>[,<memstart>]]
605
606 condev= [HW,S390] console device
607 conmode=
608
609 console= [KNL] Output console device and options.
610
611 tty<n> Use the virtual console device <n>.
612
613 ttyS<n>[,options]
614 ttyUSB0[,options]
615 Use the specified serial port. The options are of
616 the form "bbbbpnf", where "bbbb" is the baud rate,
617 "p" is parity ("n", "o", or "e"), "n" is number of
618 bits, and "f" is flow control ("r" for RTS or
619 omit it). Default is "9600n8".
620
621 See Documentation/admin-guide/serial-console.rst for more
622 information. See
623 Documentation/networking/netconsole.txt for an
624 alternative.
625
626 uart[8250],io,<addr>[,options]
627 uart[8250],mmio,<addr>[,options]
628 uart[8250],mmio16,<addr>[,options]
629 uart[8250],mmio32,<addr>[,options]
630 uart[8250],0x<addr>[,options]
631 Start an early, polled-mode console on the 8250/16550
632 UART at the specified I/O port or MMIO address,
633 switching to the matching ttyS device later.
634 MMIO inter-register address stride is either 8-bit
635 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
636 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
637 to be equivalent to 'mmio'. 'options' are specified in
638 the same format described for ttyS above; if unspecified,
639 the h/w is not re-initialized.
640
641 hvc<n> Use the hypervisor console device <n>. This is for
642 both Xen and PowerPC hypervisors.
643
644 If the device connected to the port is not a TTY but a braille
645 device, prepend "brl," before the device type, for instance
646 console=brl,ttyS0
647 For now, only VisioBraille is supported.
648
649 console_msg_format=
650 [KNL] Change console messages format
651 default
652 By default we print messages on consoles in
653 "[time stamp] text\n" format (time stamp may not be
654 printed, depending on CONFIG_PRINTK_TIME or
655 `printk_time' param).
656 syslog
657 Switch to syslog format: "<%u>[time stamp] text\n"
658 IOW, each message will have a facility and loglevel
659 prefix. The format is similar to one used by syslog()
660 syscall, or to executing "dmesg -S --raw" or to reading
661 from /proc/kmsg.
662
663 consoleblank= [KNL] The console blank (screen saver) timeout in
664 seconds. A value of 0 disables the blank timer.
665 Defaults to 0.
666
667 coredump_filter=
668 [KNL] Change the default value for
669 /proc/<pid>/coredump_filter.
670 See also Documentation/filesystems/proc.txt.
671
672 coresight_cpu_debug.enable
673 [ARM,ARM64]
674 Format: <bool>
675 Enable/disable the CPU sampling based debugging.
676 0: default value, disable debugging
677 1: enable debugging at boot time
678
679 cpuidle.off=1 [CPU_IDLE]
680 disable the cpuidle sub-system
681
682 cpufreq.off=1 [CPU_FREQ]
683 disable the cpufreq sub-system
684
685 cpu_init_udelay=N
686 [X86] Delay for N microsec between assert and de-assert
687 of APIC INIT to start processors. This delay occurs
688 on every CPU online, such as boot, and resume from suspend.
689 Default: 10000
690
691 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
692 Format:
693 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
694
695 crashkernel=size[KMG][@offset[KMG]]
696 [KNL] Using kexec, Linux can switch to a 'crash kernel'
697 upon panic. This parameter reserves the physical
698 memory region [offset, offset + size] for that kernel
699 image. If '@offset' is omitted, then a suitable offset
700 is selected automatically. Check
701 Documentation/kdump/kdump.txt for further details.
702
703 crashkernel=range1:size1[,range2:size2,...][@offset]
704 [KNL] Same as above, but depends on the memory
705 in the running system. The syntax of range is
706 start-[end] where start and end are both
707 a memory unit (amount[KMG]). See also
708 Documentation/kdump/kdump.txt for an example.
709
710 crashkernel=size[KMG],high
711 [KNL, x86_64] range could be above 4G. Allow kernel
712 to allocate physical memory region from top, so could
713 be above 4G if system have more than 4G ram installed.
714 Otherwise memory region will be allocated below 4G, if
715 available.
716 It will be ignored if crashkernel=X is specified.
717 crashkernel=size[KMG],low
718 [KNL, x86_64] range under 4G. When crashkernel=X,high
719 is passed, kernel could allocate physical memory region
720 above 4G, that cause second kernel crash on system
721 that require some amount of low memory, e.g. swiotlb
722 requires at least 64M+32K low memory, also enough extra
723 low memory is needed to make sure DMA buffers for 32-bit
724 devices won't run out. Kernel would try to allocate at
725 at least 256M below 4G automatically.
726 This one let user to specify own low range under 4G
727 for second kernel instead.
728 0: to disable low allocation.
729 It will be ignored when crashkernel=X,high is not used
730 or memory reserved is below 4G.
731
732 cryptomgr.notests
733 [KNL] Disable crypto self-tests
734
735 cs89x0_dma= [HW,NET]
736 Format: <dma>
737
738 cs89x0_media= [HW,NET]
739 Format: { rj45 | aui | bnc }
740
741 dasd= [HW,NET]
742 See header of drivers/s390/block/dasd_devmap.c.
743
744 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
745 (one device per port)
746 Format: <port#>,<type>
747 See also Documentation/input/devices/joystick-parport.rst
748
749 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
750 time. See
751 Documentation/admin-guide/dynamic-debug-howto.rst for
752 details. Deprecated, see dyndbg.
753
754 debug [KNL] Enable kernel debugging (events log level).
755
756 debug_locks_verbose=
757 [KNL] verbose self-tests
758 Format=<0|1>
759 Print debugging info while doing the locking API
760 self-tests.
761 We default to 0 (no extra messages), setting it to
762 1 will print _a lot_ more information - normally
763 only useful to kernel developers.
764
765 debug_objects [KNL] Enable object debugging
766
767 no_debug_objects
768 [KNL] Disable object debugging
769
770 debug_guardpage_minorder=
771 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
772 parameter allows control of the order of pages that will
773 be intentionally kept free (and hence protected) by the
774 buddy allocator. Bigger value increase the probability
775 of catching random memory corruption, but reduce the
776 amount of memory for normal system use. The maximum
777 possible value is MAX_ORDER/2. Setting this parameter
778 to 1 or 2 should be enough to identify most random
779 memory corruption problems caused by bugs in kernel or
780 driver code when a CPU writes to (or reads from) a
781 random memory location. Note that there exists a class
782 of memory corruptions problems caused by buggy H/W or
783 F/W or by drivers badly programing DMA (basically when
784 memory is written at bus level and the CPU MMU is
785 bypassed) which are not detectable by
786 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
787 tracking down these problems.
788
789 debug_pagealloc=
790 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
791 parameter enables the feature at boot time. In
792 default, it is disabled. We can avoid allocating huge
793 chunk of memory for debug pagealloc if we don't enable
794 it at boot time and the system will work mostly same
795 with the kernel built without CONFIG_DEBUG_PAGEALLOC.
796 on: enable the feature
797
798 debugpat [X86] Enable PAT debugging
799
800 decnet.addr= [HW,NET]
801 Format: <area>[,<node>]
802 See also Documentation/networking/decnet.txt.
803
804 default_hugepagesz=
805 [same as hugepagesz=] The size of the default
806 HugeTLB page size. This is the size represented by
807 the legacy /proc/ hugepages APIs, used for SHM, and
808 default size when mounting hugetlbfs filesystems.
809 Defaults to the default architecture's huge page size
810 if not specified.
811
812 dhash_entries= [KNL]
813 Set number of hash buckets for dentry cache.
814
815 disable_1tb_segments [PPC]
816 Disables the use of 1TB hash page table segments. This
817 causes the kernel to fall back to 256MB segments which
818 can be useful when debugging issues that require an SLB
819 miss to occur.
820
821 disable= [IPV6]
822 See Documentation/networking/ipv6.txt.
823
824 disable_radix [PPC]
825 Disable RADIX MMU mode on POWER9
826
827 disable_cpu_apicid= [X86,APIC,SMP]
828 Format: <int>
829 The number of initial APIC ID for the
830 corresponding CPU to be disabled at boot,
831 mostly used for the kdump 2nd kernel to
832 disable BSP to wake up multiple CPUs without
833 causing system reset or hang due to sending
834 INIT from AP to BSP.
835
836 disable_ddw [PPC/PSERIES]
837 Disable Dynamic DMA Window support. Use this if
838 to workaround buggy firmware.
839
840 disable_ipv6= [IPV6]
841 See Documentation/networking/ipv6.txt.
842
843 disable_mtrr_cleanup [X86]
844 The kernel tries to adjust MTRR layout from continuous
845 to discrete, to make X server driver able to add WB
846 entry later. This parameter disables that.
847
848 disable_mtrr_trim [X86, Intel and AMD only]
849 By default the kernel will trim any uncacheable
850 memory out of your available memory pool based on
851 MTRR settings. This parameter disables that behavior,
852 possibly causing your machine to run very slowly.
853
854 disable_timer_pin_1 [X86]
855 Disable PIN 1 of APIC timer
856 Can be useful to work around chipset bugs.
857
858 dis_ucode_ldr [X86] Disable the microcode loader.
859
860 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
861 this option disables the debugging code at boot.
862
863 dma_debug_entries=<number>
864 This option allows to tune the number of preallocated
865 entries for DMA-API debugging code. One entry is
866 required per DMA-API allocation. Use this if the
867 DMA-API debugging code disables itself because the
868 architectural default is too low.
869
870 dma_debug_driver=<driver_name>
871 With this option the DMA-API debugging driver
872 filter feature can be enabled at boot time. Just
873 pass the driver to filter for as the parameter.
874 The filter can be disabled or changed to another
875 driver later using sysfs.
876
877 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
878 Broken monitors, graphic adapters, KVMs and EDIDless
879 panels may send no or incorrect EDID data sets.
880 This parameter allows to specify an EDID data sets
881 in the /lib/firmware directory that are used instead.
882 Generic built-in EDID data sets are used, if one of
883 edid/1024x768.bin, edid/1280x1024.bin,
884 edid/1680x1050.bin, or edid/1920x1080.bin is given
885 and no file with the same name exists. Details and
886 instructions how to build your own EDID data are
887 available in Documentation/EDID/HOWTO.txt. An EDID
888 data set will only be used for a particular connector,
889 if its name and a colon are prepended to the EDID
890 name. Each connector may use a unique EDID data
891 set by separating the files with a comma. An EDID
892 data set with no connector name will be used for
893 any connectors not explicitly specified.
894
895 dscc4.setup= [NET]
896
897 dt_cpu_ftrs= [PPC]
898 Format: {"off" | "known"}
899 Control how the dt_cpu_ftrs device-tree binding is
900 used for CPU feature discovery and setup (if it
901 exists).
902 off: Do not use it, fall back to legacy cpu table.
903 known: Do not pass through unknown features to guests
904 or userspace, only those that the kernel is aware of.
905
906 dump_apple_properties [X86]
907 Dump name and content of EFI device properties on
908 x86 Macs. Useful for driver authors to determine
909 what data is available or for reverse-engineering.
910
911 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
912 module.dyndbg[="val"]
913 Enable debug messages at boot time. See
914 Documentation/admin-guide/dynamic-debug-howto.rst
915 for details.
916
917 nompx [X86] Disables Intel Memory Protection Extensions.
918 See Documentation/x86/intel_mpx.txt for more
919 information about the feature.
920
921 nopku [X86] Disable Memory Protection Keys CPU feature found
922 in some Intel CPUs.
923
924 module.async_probe [KNL]
925 Enable asynchronous probe on this module.
926
927 early_ioremap_debug [KNL]
928 Enable debug messages in early_ioremap support. This
929 is useful for tracking down temporary early mappings
930 which are not unmapped.
931
932 earlycon= [KNL] Output early console device and options.
933
934 [ARM64] The early console is determined by the
935 stdout-path property in device tree's chosen node,
936 or determined by the ACPI SPCR table.
937
938 [X86] When used with no options the early console is
939 determined by the ACPI SPCR table.
940
941 cdns,<addr>[,options]
942 Start an early, polled-mode console on a Cadence
943 (xuartps) serial port at the specified address. Only
944 supported option is baud rate. If baud rate is not
945 specified, the serial port must already be setup and
946 configured.
947
948 uart[8250],io,<addr>[,options]
949 uart[8250],mmio,<addr>[,options]
950 uart[8250],mmio32,<addr>[,options]
951 uart[8250],mmio32be,<addr>[,options]
952 uart[8250],0x<addr>[,options]
953 Start an early, polled-mode console on the 8250/16550
954 UART at the specified I/O port or MMIO address.
955 MMIO inter-register address stride is either 8-bit
956 (mmio) or 32-bit (mmio32 or mmio32be).
957 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
958 to be equivalent to 'mmio'. 'options' are specified
959 in the same format described for "console=ttyS<n>"; if
960 unspecified, the h/w is not initialized.
961
962 pl011,<addr>
963 pl011,mmio32,<addr>
964 Start an early, polled-mode console on a pl011 serial
965 port at the specified address. The pl011 serial port
966 must already be setup and configured. Options are not
967 yet supported. If 'mmio32' is specified, then only
968 the driver will use only 32-bit accessors to read/write
969 the device registers.
970
971 meson,<addr>
972 Start an early, polled-mode console on a meson serial
973 port at the specified address. The serial port must
974 already be setup and configured. Options are not yet
975 supported.
976
977 msm_serial,<addr>
978 Start an early, polled-mode console on an msm serial
979 port at the specified address. The serial port
980 must already be setup and configured. Options are not
981 yet supported.
982
983 msm_serial_dm,<addr>
984 Start an early, polled-mode console on an msm serial
985 dm port at the specified address. The serial port
986 must already be setup and configured. Options are not
987 yet supported.
988
989 owl,<addr>
990 Start an early, polled-mode console on a serial port
991 of an Actions Semi SoC, such as S500 or S900, at the
992 specified address. The serial port must already be
993 setup and configured. Options are not yet supported.
994
995 smh Use ARM semihosting calls for early console.
996
997 s3c2410,<addr>
998 s3c2412,<addr>
999 s3c2440,<addr>
1000 s3c6400,<addr>
1001 s5pv210,<addr>
1002 exynos4210,<addr>
1003 Use early console provided by serial driver available
1004 on Samsung SoCs, requires selecting proper type and
1005 a correct base address of the selected UART port. The
1006 serial port must already be setup and configured.
1007 Options are not yet supported.
1008
1009 lantiq,<addr>
1010 Start an early, polled-mode console on a lantiq serial
1011 (lqasc) port at the specified address. The serial port
1012 must already be setup and configured. Options are not
1013 yet supported.
1014
1015 lpuart,<addr>
1016 lpuart32,<addr>
1017 Use early console provided by Freescale LP UART driver
1018 found on Freescale Vybrid and QorIQ LS1021A processors.
1019 A valid base address must be provided, and the serial
1020 port must already be setup and configured.
1021
1022 ar3700_uart,<addr>
1023 Start an early, polled-mode console on the
1024 Armada 3700 serial port at the specified
1025 address. The serial port must already be setup
1026 and configured. Options are not yet supported.
1027
1028 earlyprintk= [X86,SH,ARM,M68k,S390]
1029 earlyprintk=vga
1030 earlyprintk=efi
1031 earlyprintk=sclp
1032 earlyprintk=xen
1033 earlyprintk=serial[,ttySn[,baudrate]]
1034 earlyprintk=serial[,0x...[,baudrate]]
1035 earlyprintk=ttySn[,baudrate]
1036 earlyprintk=dbgp[debugController#]
1037 earlyprintk=pciserial,bus:device.function[,baudrate]
1038 earlyprintk=xdbc[xhciController#]
1039
1040 earlyprintk is useful when the kernel crashes before
1041 the normal console is initialized. It is not enabled by
1042 default because it has some cosmetic problems.
1043
1044 Append ",keep" to not disable it when the real console
1045 takes over.
1046
1047 Only one of vga, efi, serial, or usb debug port can
1048 be used at a time.
1049
1050 Currently only ttyS0 and ttyS1 may be specified by
1051 name. Other I/O ports may be explicitly specified
1052 on some architectures (x86 and arm at least) by
1053 replacing ttySn with an I/O port address, like this:
1054 earlyprintk=serial,0x1008,115200
1055 You can find the port for a given device in
1056 /proc/tty/driver/serial:
1057 2: uart:ST16650V2 port:00001008 irq:18 ...
1058
1059 Interaction with the standard serial driver is not
1060 very good.
1061
1062 The VGA and EFI output is eventually overwritten by
1063 the real console.
1064
1065 The xen output can only be used by Xen PV guests.
1066
1067 The sclp output can only be used on s390.
1068
1069 edac_report= [HW,EDAC] Control how to report EDAC event
1070 Format: {"on" | "off" | "force"}
1071 on: enable EDAC to report H/W event. May be overridden
1072 by other higher priority error reporting module.
1073 off: disable H/W event reporting through EDAC.
1074 force: enforce the use of EDAC to report H/W event.
1075 default: on.
1076
1077 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1078 ekgdboc=kbd
1079
1080 This is designed to be used in conjunction with
1081 the boot argument: earlyprintk=vga
1082
1083 edd= [EDD]
1084 Format: {"off" | "on" | "skip[mbr]"}
1085
1086 efi= [EFI]
1087 Format: { "old_map", "nochunk", "noruntime", "debug" }
1088 old_map [X86-64]: switch to the old ioremap-based EFI
1089 runtime services mapping. 32-bit still uses this one by
1090 default.
1091 nochunk: disable reading files in "chunks" in the EFI
1092 boot stub, as chunking can cause problems with some
1093 firmware implementations.
1094 noruntime : disable EFI runtime services support
1095 debug: enable misc debug output
1096
1097 efi_no_storage_paranoia [EFI; X86]
1098 Using this parameter you can use more than 50% of
1099 your efi variable storage. Use this parameter only if
1100 you are really sure that your UEFI does sane gc and
1101 fulfills the spec otherwise your board may brick.
1102
1103 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1104 Add arbitrary attribute to specific memory range by
1105 updating original EFI memory map.
1106 Region of memory which aa attribute is added to is
1107 from ss to ss+nn.
1108 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1109 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1110 attribute is added to range 0x100000000-0x180000000 and
1111 0x10a0000000-0x1120000000.
1112
1113 Using this parameter you can do debugging of EFI memmap
1114 related feature. For example, you can do debugging of
1115 Address Range Mirroring feature even if your box
1116 doesn't support it.
1117
1118 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1119 that is to be dynamically loaded by Linux. If there are
1120 multiple variables with the same name but with different
1121 vendor GUIDs, all of them will be loaded. See
1122 Documentation/acpi/ssdt-overlays.txt for details.
1123
1124
1125 eisa_irq_edge= [PARISC,HW]
1126 See header of drivers/parisc/eisa.c.
1127
1128 elanfreq= [X86-32]
1129 See comment before function elanfreq_setup() in
1130 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1131
1132 elevator= [IOSCHED]
1133 Format: {"cfq" | "deadline" | "noop"}
1134 See Documentation/block/cfq-iosched.txt and
1135 Documentation/block/deadline-iosched.txt for details.
1136
1137 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1138 Specifies physical address of start of kernel core
1139 image elf header and optionally the size. Generally
1140 kexec loader will pass this option to capture kernel.
1141 See Documentation/kdump/kdump.txt for details.
1142
1143 enable_mtrr_cleanup [X86]
1144 The kernel tries to adjust MTRR layout from continuous
1145 to discrete, to make X server driver able to add WB
1146 entry later. This parameter enables that.
1147
1148 enable_timer_pin_1 [X86]
1149 Enable PIN 1 of APIC timer
1150 Can be useful to work around chipset bugs
1151 (in particular on some ATI chipsets).
1152 The kernel tries to set a reasonable default.
1153
1154 enforcing [SELINUX] Set initial enforcing status.
1155 Format: {"0" | "1"}
1156 See security/selinux/Kconfig help text.
1157 0 -- permissive (log only, no denials).
1158 1 -- enforcing (deny and log).
1159 Default value is 0.
1160 Value can be changed at runtime via /selinux/enforce.
1161
1162 erst_disable [ACPI]
1163 Disable Error Record Serialization Table (ERST)
1164 support.
1165
1166 ether= [HW,NET] Ethernet cards parameters
1167 This option is obsoleted by the "netdev=" option, which
1168 has equivalent usage. See its documentation for details.
1169
1170 evm= [EVM]
1171 Format: { "fix" }
1172 Permit 'security.evm' to be updated regardless of
1173 current integrity status.
1174
1175 failslab=
1176 fail_page_alloc=
1177 fail_make_request=[KNL]
1178 General fault injection mechanism.
1179 Format: <interval>,<probability>,<space>,<times>
1180 See also Documentation/fault-injection/.
1181
1182 floppy= [HW]
1183 See Documentation/blockdev/floppy.txt.
1184
1185 force_pal_cache_flush
1186 [IA-64] Avoid check_sal_cache_flush which may hang on
1187 buggy SAL_CACHE_FLUSH implementations. Using this
1188 parameter will force ia64_sal_cache_flush to call
1189 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1190
1191 forcepae [X86-32]
1192 Forcefully enable Physical Address Extension (PAE).
1193 Many Pentium M systems disable PAE but may have a
1194 functionally usable PAE implementation.
1195 Warning: use of this parameter will taint the kernel
1196 and may cause unknown problems.
1197
1198 ftrace=[tracer]
1199 [FTRACE] will set and start the specified tracer
1200 as early as possible in order to facilitate early
1201 boot debugging.
1202
1203 ftrace_dump_on_oops[=orig_cpu]
1204 [FTRACE] will dump the trace buffers on oops.
1205 If no parameter is passed, ftrace will dump
1206 buffers of all CPUs, but if you pass orig_cpu, it will
1207 dump only the buffer of the CPU that triggered the
1208 oops.
1209
1210 ftrace_filter=[function-list]
1211 [FTRACE] Limit the functions traced by the function
1212 tracer at boot up. function-list is a comma separated
1213 list of functions. This list can be changed at run
1214 time by the set_ftrace_filter file in the debugfs
1215 tracing directory.
1216
1217 ftrace_notrace=[function-list]
1218 [FTRACE] Do not trace the functions specified in
1219 function-list. This list can be changed at run time
1220 by the set_ftrace_notrace file in the debugfs
1221 tracing directory.
1222
1223 ftrace_graph_filter=[function-list]
1224 [FTRACE] Limit the top level callers functions traced
1225 by the function graph tracer at boot up.
1226 function-list is a comma separated list of functions
1227 that can be changed at run time by the
1228 set_graph_function file in the debugfs tracing directory.
1229
1230 ftrace_graph_notrace=[function-list]
1231 [FTRACE] Do not trace from the functions specified in
1232 function-list. This list is a comma separated list of
1233 functions that can be changed at run time by the
1234 set_graph_notrace file in the debugfs tracing directory.
1235
1236 ftrace_graph_max_depth=<uint>
1237 [FTRACE] Used with the function graph tracer. This is
1238 the max depth it will trace into a function. This value
1239 can be changed at run time by the max_graph_depth file
1240 in the tracefs tracing directory. default: 0 (no limit)
1241
1242 gamecon.map[2|3]=
1243 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1244 support via parallel port (up to 5 devices per port)
1245 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1246 See also Documentation/input/devices/joystick-parport.rst
1247
1248 gamma= [HW,DRM]
1249
1250 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1251 Format: off | on
1252 default: on
1253
1254 gcov_persist= [GCOV] When non-zero (default), profiling data for
1255 kernel modules is saved and remains accessible via
1256 debugfs, even when the module is unloaded/reloaded.
1257 When zero, profiling data is discarded and associated
1258 debugfs files are removed at module unload time.
1259
1260 goldfish [X86] Enable the goldfish android emulator platform.
1261 Don't use this when you are not running on the
1262 android emulator
1263
1264 gpt [EFI] Forces disk with valid GPT signature but
1265 invalid Protective MBR to be treated as GPT. If the
1266 primary GPT is corrupted, it enables the backup/alternate
1267 GPT to be used instead.
1268
1269 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1270 the "Enable 0" bit of the configuration register.
1271 Format: 0 | 1
1272 Default: 0
1273 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1274 the "Enable 0" bit of the configuration register.
1275 Format: 0 | 1
1276 Default: 0
1277 grcan.select= [HW] Select which physical interface to use.
1278 Format: 0 | 1
1279 Default: 0
1280 grcan.txsize= [HW] Sets the size of the tx buffer.
1281 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1282 Default: 1024
1283 grcan.rxsize= [HW] Sets the size of the rx buffer.
1284 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1285 Default: 1024
1286
1287 gpio-mockup.gpio_mockup_ranges
1288 [HW] Sets the ranges of gpiochip of for this device.
1289 Format: <start1>,<end1>,<start2>,<end2>...
1290
1291 hardlockup_all_cpu_backtrace=
1292 [KNL] Should the hard-lockup detector generate
1293 backtraces on all cpus.
1294 Format: <integer>
1295
1296 hashdist= [KNL,NUMA] Large hashes allocated during boot
1297 are distributed across NUMA nodes. Defaults on
1298 for 64-bit NUMA, off otherwise.
1299 Format: 0 | 1 (for off | on)
1300
1301 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1302
1303 hd= [EIDE] (E)IDE hard drive subsystem geometry
1304 Format: <cyl>,<head>,<sect>
1305
1306 hest_disable [ACPI]
1307 Disable Hardware Error Source Table (HEST) support;
1308 corresponding firmware-first mode error processing
1309 logic will be disabled.
1310
1311 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1312 size of <nn>. This works even on boxes that have no
1313 highmem otherwise. This also works to reduce highmem
1314 size on bigger boxes.
1315
1316 highres= [KNL] Enable/disable high resolution timer mode.
1317 Valid parameters: "on", "off"
1318 Default: "on"
1319
1320 hisax= [HW,ISDN]
1321 See Documentation/isdn/README.HiSax.
1322
1323 hlt [BUGS=ARM,SH]
1324
1325 hpet= [X86-32,HPET] option to control HPET usage
1326 Format: { enable (default) | disable | force |
1327 verbose }
1328 disable: disable HPET and use PIT instead
1329 force: allow force enabled of undocumented chips (ICH4,
1330 VIA, nVidia)
1331 verbose: show contents of HPET registers during setup
1332
1333 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1334 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1335
1336 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1337 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
1338 On x86-64 and powerpc, this option can be specified
1339 multiple times interleaved with hugepages= to reserve
1340 huge pages of different sizes. Valid pages sizes on
1341 x86-64 are 2M (when the CPU supports "pse") and 1G
1342 (when the CPU supports the "pdpe1gb" cpuinfo flag).
1343
1344 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1345 terminal devices. Valid values: 0..8
1346 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1347 If specified, z/VM IUCV HVC accepts connections
1348 from listed z/VM user IDs only.
1349
1350 keep_bootcon [KNL]
1351 Do not unregister boot console at start. This is only
1352 useful for debugging when something happens in the window
1353 between unregistering the boot console and initializing
1354 the real console.
1355
1356 i2c_bus= [HW] Override the default board specific I2C bus speed
1357 or register an additional I2C bus that is not
1358 registered from board initialization code.
1359 Format:
1360 <bus_id>,<clkrate>
1361
1362 i8042.debug [HW] Toggle i8042 debug mode
1363 i8042.unmask_kbd_data
1364 [HW] Enable printing of interrupt data from the KBD port
1365 (disabled by default, and as a pre-condition
1366 requires that i8042.debug=1 be enabled)
1367 i8042.direct [HW] Put keyboard port into non-translated mode
1368 i8042.dumbkbd [HW] Pretend that controller can only read data from
1369 keyboard and cannot control its state
1370 (Don't attempt to blink the leds)
1371 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1372 i8042.nokbd [HW] Don't check/create keyboard port
1373 i8042.noloop [HW] Disable the AUX Loopback command while probing
1374 for the AUX port
1375 i8042.nomux [HW] Don't check presence of an active multiplexing
1376 controller
1377 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1378 controllers
1379 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1380 i8042.reset [HW] Reset the controller during init, cleanup and
1381 suspend-to-ram transitions, only during s2r
1382 transitions, or never reset
1383 Format: { 1 | Y | y | 0 | N | n }
1384 1, Y, y: always reset controller
1385 0, N, n: don't ever reset controller
1386 Default: only on s2r transitions on x86; most other
1387 architectures force reset to be always executed
1388 i8042.unlock [HW] Unlock (ignore) the keylock
1389 i8042.kbdreset [HW] Reset device connected to KBD port
1390
1391 i810= [HW,DRM]
1392
1393 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1394 indicates that the driver is running on unsupported
1395 hardware.
1396 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1397 does not match list of supported models.
1398 i8k.power_status
1399 [HW] Report power status in /proc/i8k
1400 (disabled by default)
1401 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1402 capability is set.
1403
1404 i915.invert_brightness=
1405 [DRM] Invert the sense of the variable that is used to
1406 set the brightness of the panel backlight. Normally a
1407 brightness value of 0 indicates backlight switched off,
1408 and the maximum of the brightness value sets the backlight
1409 to maximum brightness. If this parameter is set to 0
1410 (default) and the machine requires it, or this parameter
1411 is set to 1, a brightness value of 0 sets the backlight
1412 to maximum brightness, and the maximum of the brightness
1413 value switches the backlight off.
1414 -1 -- never invert brightness
1415 0 -- machine default
1416 1 -- force brightness inversion
1417
1418 icn= [HW,ISDN]
1419 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1420
1421 ide-core.nodma= [HW] (E)IDE subsystem
1422 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1423 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1424 .cdrom .chs .ignore_cable are additional options
1425 See Documentation/ide/ide.txt.
1426
1427 ide-generic.probe-mask= [HW] (E)IDE subsystem
1428 Format: <int>
1429 Probe mask for legacy ISA IDE ports. Depending on
1430 platform up to 6 ports are supported, enabled by
1431 setting corresponding bits in the mask to 1. The
1432 default value is 0x0, which has a special meaning.
1433 On systems that have PCI, it triggers scanning the
1434 PCI bus for the first and the second port, which
1435 are then probed. On systems without PCI the value
1436 of 0x0 enables probing the two first ports as if it
1437 was 0x3.
1438
1439 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1440 Claim all unknown PCI IDE storage controllers.
1441
1442 idle= [X86]
1443 Format: idle=poll, idle=halt, idle=nomwait
1444 Poll forces a polling idle loop that can slightly
1445 improve the performance of waking up a idle CPU, but
1446 will use a lot of power and make the system run hot.
1447 Not recommended.
1448 idle=halt: Halt is forced to be used for CPU idle.
1449 In such case C2/C3 won't be used again.
1450 idle=nomwait: Disable mwait for CPU C-states
1451
1452 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1453 Format: { strict | legacy | 2008 | relaxed }
1454 Default: strict
1455
1456 Choose which programs will be accepted for execution
1457 based on the IEEE 754 NaN encoding(s) supported by
1458 the FPU and the NaN encoding requested with the value
1459 of an ELF file header flag individually set by each
1460 binary. Hardware implementations are permitted to
1461 support either or both of the legacy and the 2008 NaN
1462 encoding mode.
1463
1464 Available settings are as follows:
1465 strict accept binaries that request a NaN encoding
1466 supported by the FPU
1467 legacy only accept legacy-NaN binaries, if supported
1468 by the FPU
1469 2008 only accept 2008-NaN binaries, if supported
1470 by the FPU
1471 relaxed accept any binaries regardless of whether
1472 supported by the FPU
1473
1474 The FPU emulator is always able to support both NaN
1475 encodings, so if no FPU hardware is present or it has
1476 been disabled with 'nofpu', then the settings of
1477 'legacy' and '2008' strap the emulator accordingly,
1478 'relaxed' straps the emulator for both legacy-NaN and
1479 2008-NaN, whereas 'strict' enables legacy-NaN only on
1480 legacy processors and both NaN encodings on MIPS32 or
1481 MIPS64 CPUs.
1482
1483 The setting for ABS.fmt/NEG.fmt instruction execution
1484 mode generally follows that for the NaN encoding,
1485 except where unsupported by hardware.
1486
1487 ignore_loglevel [KNL]
1488 Ignore loglevel setting - this will print /all/
1489 kernel messages to the console. Useful for debugging.
1490 We also add it as printk module parameter, so users
1491 could change it dynamically, usually by
1492 /sys/module/printk/parameters/ignore_loglevel.
1493
1494 ignore_rlimit_data
1495 Ignore RLIMIT_DATA setting for data mappings,
1496 print warning at first misuse. Can be changed via
1497 /sys/module/kernel/parameters/ignore_rlimit_data.
1498
1499 ihash_entries= [KNL]
1500 Set number of hash buckets for inode cache.
1501
1502 ima_appraise= [IMA] appraise integrity measurements
1503 Format: { "off" | "enforce" | "fix" | "log" }
1504 default: "enforce"
1505
1506 ima_appraise_tcb [IMA]
1507 The builtin appraise policy appraises all files
1508 owned by uid=0.
1509
1510 ima_canonical_fmt [IMA]
1511 Use the canonical format for the binary runtime
1512 measurements, instead of host native format.
1513
1514 ima_hash= [IMA]
1515 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1516 | sha512 | ... }
1517 default: "sha1"
1518
1519 The list of supported hash algorithms is defined
1520 in crypto/hash_info.h.
1521
1522 ima_policy= [IMA]
1523 The builtin policies to load during IMA setup.
1524 Format: "tcb | appraise_tcb | secure_boot |
1525 fail_securely"
1526
1527 The "tcb" policy measures all programs exec'd, files
1528 mmap'd for exec, and all files opened with the read
1529 mode bit set by either the effective uid (euid=0) or
1530 uid=0.
1531
1532 The "appraise_tcb" policy appraises the integrity of
1533 all files owned by root. (This is the equivalent
1534 of ima_appraise_tcb.)
1535
1536 The "secure_boot" policy appraises the integrity
1537 of files (eg. kexec kernel image, kernel modules,
1538 firmware, policy, etc) based on file signatures.
1539
1540 The "fail_securely" policy forces file signature
1541 verification failure also on privileged mounted
1542 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1543 flag.
1544
1545 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1546 Load a policy which meets the needs of the Trusted
1547 Computing Base. This means IMA will measure all
1548 programs exec'd, files mmap'd for exec, and all files
1549 opened for read by uid=0.
1550
1551 ima_template= [IMA]
1552 Select one of defined IMA measurements template formats.
1553 Formats: { "ima" | "ima-ng" | "ima-sig" }
1554 Default: "ima-ng"
1555
1556 ima_template_fmt=
1557 [IMA] Define a custom template format.
1558 Format: { "field1|...|fieldN" }
1559
1560 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1561 Format: <min_file_size>
1562 Set the minimal file size for using asynchronous hash.
1563 If left unspecified, ahash usage is disabled.
1564
1565 ahash performance varies for different data sizes on
1566 different crypto accelerators. This option can be used
1567 to achieve the best performance for a particular HW.
1568
1569 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1570 Format: <bufsize>
1571 Set hashing buffer size. Default: 4k.
1572
1573 ahash performance varies for different chunk sizes on
1574 different crypto accelerators. This option can be used
1575 to achieve best performance for particular HW.
1576
1577 init= [KNL]
1578 Format: <full_path>
1579 Run specified binary instead of /sbin/init as init
1580 process.
1581
1582 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1583 for working out where the kernel is dying during
1584 startup.
1585
1586 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1587 initcall functions. Useful for debugging built-in
1588 modules and initcalls.
1589
1590 initrd= [BOOT] Specify the location of the initial ramdisk
1591
1592 init_pkru= [x86] Specify the default memory protection keys rights
1593 register contents for all processes. 0x55555554 by
1594 default (disallow access to all but pkey 0). Can
1595 override in debugfs after boot.
1596
1597 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1598 Format: <irq>
1599
1600 int_pln_enable [x86] Enable power limit notification interrupt
1601
1602 integrity_audit=[IMA]
1603 Format: { "0" | "1" }
1604 0 -- basic integrity auditing messages. (Default)
1605 1 -- additional integrity auditing messages.
1606
1607 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1608 on
1609 Enable intel iommu driver.
1610 off
1611 Disable intel iommu driver.
1612 igfx_off [Default Off]
1613 By default, gfx is mapped as normal device. If a gfx
1614 device has a dedicated DMAR unit, the DMAR unit is
1615 bypassed by not enabling DMAR with this option. In
1616 this case, gfx device will use physical address for
1617 DMA.
1618 forcedac [x86_64]
1619 With this option iommu will not optimize to look
1620 for io virtual address below 32-bit forcing dual
1621 address cycle on pci bus for cards supporting greater
1622 than 32-bit addressing. The default is to look
1623 for translation below 32-bit and if not available
1624 then look in the higher range.
1625 strict [Default Off]
1626 With this option on every unmap_single operation will
1627 result in a hardware IOTLB flush operation as opposed
1628 to batching them for performance.
1629 sp_off [Default Off]
1630 By default, super page will be supported if Intel IOMMU
1631 has the capability. With this option, super page will
1632 not be supported.
1633 ecs_off [Default Off]
1634 By default, extended context tables will be supported if
1635 the hardware advertises that it has support both for the
1636 extended tables themselves, and also PASID support. With
1637 this option set, extended tables will not be used even
1638 on hardware which claims to support them.
1639 tboot_noforce [Default Off]
1640 Do not force the Intel IOMMU enabled under tboot.
1641 By default, tboot will force Intel IOMMU on, which
1642 could harm performance of some high-throughput
1643 devices like 40GBit network cards, even if identity
1644 mapping is enabled.
1645 Note that using this option lowers the security
1646 provided by tboot because it makes the system
1647 vulnerable to DMA attacks.
1648
1649 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1650 0 disables intel_idle and fall back on acpi_idle.
1651 1 to 9 specify maximum depth of C-state.
1652
1653 intel_pstate= [X86]
1654 disable
1655 Do not enable intel_pstate as the default
1656 scaling driver for the supported processors
1657 passive
1658 Use intel_pstate as a scaling driver, but configure it
1659 to work with generic cpufreq governors (instead of
1660 enabling its internal governor). This mode cannot be
1661 used along with the hardware-managed P-states (HWP)
1662 feature.
1663 force
1664 Enable intel_pstate on systems that prohibit it by default
1665 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1666 instead of acpi-cpufreq may disable platform features, such
1667 as thermal controls and power capping, that rely on ACPI
1668 P-States information being indicated to OSPM and therefore
1669 should be used with caution. This option does not work with
1670 processors that aren't supported by the intel_pstate driver
1671 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1672 no_hwp
1673 Do not enable hardware P state control (HWP)
1674 if available.
1675 hwp_only
1676 Only load intel_pstate on systems which support
1677 hardware P state control (HWP) if available.
1678 support_acpi_ppc
1679 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1680 Description Table, specifies preferred power management
1681 profile as "Enterprise Server" or "Performance Server",
1682 then this feature is turned on by default.
1683 per_cpu_perf_limits
1684 Allow per-logical-CPU P-State performance control limits using
1685 cpufreq sysfs interface
1686
1687 intremap= [X86-64, Intel-IOMMU]
1688 on enable Interrupt Remapping (default)
1689 off disable Interrupt Remapping
1690 nosid disable Source ID checking
1691 no_x2apic_optout
1692 BIOS x2APIC opt-out request will be ignored
1693 nopost disable Interrupt Posting
1694
1695 iomem= Disable strict checking of access to MMIO memory
1696 strict regions from userspace.
1697 relaxed
1698
1699 iommu= [x86]
1700 off
1701 force
1702 noforce
1703 biomerge
1704 panic
1705 nopanic
1706 merge
1707 nomerge
1708 forcesac
1709 soft
1710 pt [x86, IA-64]
1711 nobypass [PPC/POWERNV]
1712 Disable IOMMU bypass, using IOMMU for PCI devices.
1713
1714 iommu.passthrough=
1715 [ARM64] Configure DMA to bypass the IOMMU by default.
1716 Format: { "0" | "1" }
1717 0 - Use IOMMU translation for DMA.
1718 1 - Bypass the IOMMU for DMA.
1719 unset - Use IOMMU translation for DMA.
1720
1721 io7= [HW] IO7 for Marvel based alpha systems
1722 See comment before marvel_specify_io7 in
1723 arch/alpha/kernel/core_marvel.c.
1724
1725 io_delay= [X86] I/O delay method
1726 0x80
1727 Standard port 0x80 based delay
1728 0xed
1729 Alternate port 0xed based delay (needed on some systems)
1730 udelay
1731 Simple two microseconds delay
1732 none
1733 No delay
1734
1735 ip= [IP_PNP]
1736 See Documentation/filesystems/nfs/nfsroot.txt.
1737
1738 irqaffinity= [SMP] Set the default irq affinity mask
1739 The argument is a cpu list, as described above.
1740
1741 irqchip.gicv2_force_probe=
1742 [ARM, ARM64]
1743 Format: <bool>
1744 Force the kernel to look for the second 4kB page
1745 of a GICv2 controller even if the memory range
1746 exposed by the device tree is too small.
1747
1748 irqchip.gicv3_nolpi=
1749 [ARM, ARM64]
1750 Force the kernel to ignore the availability of
1751 LPIs (and by consequence ITSs). Intended for system
1752 that use the kernel as a bootloader, and thus want
1753 to let secondary kernels in charge of setting up
1754 LPIs.
1755
1756 irqfixup [HW]
1757 When an interrupt is not handled search all handlers
1758 for it. Intended to get systems with badly broken
1759 firmware running.
1760
1761 irqpoll [HW]
1762 When an interrupt is not handled search all handlers
1763 for it. Also check all handlers each timer
1764 interrupt. Intended to get systems with badly broken
1765 firmware running.
1766
1767 isapnp= [ISAPNP]
1768 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1769
1770 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
1771 [Deprecated - use cpusets instead]
1772 Format: [flag-list,]<cpu-list>
1773
1774 Specify one or more CPUs to isolate from disturbances
1775 specified in the flag list (default: domain):
1776
1777 nohz
1778 Disable the tick when a single task runs.
1779
1780 A residual 1Hz tick is offloaded to workqueues, which you
1781 need to affine to housekeeping through the global
1782 workqueue's affinity configured via the
1783 /sys/devices/virtual/workqueue/cpumask sysfs file, or
1784 by using the 'domain' flag described below.
1785
1786 NOTE: by default the global workqueue runs on all CPUs,
1787 so to protect individual CPUs the 'cpumask' file has to
1788 be configured manually after bootup.
1789
1790 domain
1791 Isolate from the general SMP balancing and scheduling
1792 algorithms. Note that performing domain isolation this way
1793 is irreversible: it's not possible to bring back a CPU to
1794 the domains once isolated through isolcpus. It's strongly
1795 advised to use cpusets instead to disable scheduler load
1796 balancing through the "cpuset.sched_load_balance" file.
1797 It offers a much more flexible interface where CPUs can
1798 move in and out of an isolated set anytime.
1799
1800 You can move a process onto or off an "isolated" CPU via
1801 the CPU affinity syscalls or cpuset.
1802 <cpu number> begins at 0 and the maximum value is
1803 "number of CPUs in system - 1".
1804
1805 The format of <cpu-list> is described above.
1806
1807
1808
1809 iucv= [HW,NET]
1810
1811 ivrs_ioapic [HW,X86_64]
1812 Provide an override to the IOAPIC-ID<->DEVICE-ID
1813 mapping provided in the IVRS ACPI table. For
1814 example, to map IOAPIC-ID decimal 10 to
1815 PCI device 00:14.0 write the parameter as:
1816 ivrs_ioapic[10]=00:14.0
1817
1818 ivrs_hpet [HW,X86_64]
1819 Provide an override to the HPET-ID<->DEVICE-ID
1820 mapping provided in the IVRS ACPI table. For
1821 example, to map HPET-ID decimal 0 to
1822 PCI device 00:14.0 write the parameter as:
1823 ivrs_hpet[0]=00:14.0
1824
1825 ivrs_acpihid [HW,X86_64]
1826 Provide an override to the ACPI-HID:UID<->DEVICE-ID
1827 mapping provided in the IVRS ACPI table. For
1828 example, to map UART-HID:UID AMD0020:0 to
1829 PCI device 00:14.5 write the parameter as:
1830 ivrs_acpihid[00:14.5]=AMD0020:0
1831
1832 js= [HW,JOY] Analog joystick
1833 See Documentation/input/joydev/joystick.rst.
1834
1835 nokaslr [KNL]
1836 When CONFIG_RANDOMIZE_BASE is set, this disables
1837 kernel and module base offset ASLR (Address Space
1838 Layout Randomization).
1839
1840 kasan_multi_shot
1841 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
1842 report on every invalid memory access. Without this
1843 parameter KASAN will print report only for the first
1844 invalid access.
1845
1846 keepinitrd [HW,ARM]
1847
1848 kernelcore= [KNL,X86,IA-64,PPC]
1849 Format: nn[KMGTPE] | nn% | "mirror"
1850 This parameter specifies the amount of memory usable by
1851 the kernel for non-movable allocations. The requested
1852 amount is spread evenly throughout all nodes in the
1853 system as ZONE_NORMAL. The remaining memory is used for
1854 movable memory in its own zone, ZONE_MOVABLE. In the
1855 event, a node is too small to have both ZONE_NORMAL and
1856 ZONE_MOVABLE, kernelcore memory will take priority and
1857 other nodes will have a larger ZONE_MOVABLE.
1858
1859 ZONE_MOVABLE is used for the allocation of pages that
1860 may be reclaimed or moved by the page migration
1861 subsystem. Note that allocations like PTEs-from-HighMem
1862 still use the HighMem zone if it exists, and the Normal
1863 zone if it does not.
1864
1865 It is possible to specify the exact amount of memory in
1866 the form of "nn[KMGTPE]", a percentage of total system
1867 memory in the form of "nn%", or "mirror". If "mirror"
1868 option is specified, mirrored (reliable) memory is used
1869 for non-movable allocations and remaining memory is used
1870 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
1871 are exclusive, so you cannot specify multiple forms.
1872
1873 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
1874 Format: <Controller#>[,poll interval]
1875 The controller # is the number of the ehci usb debug
1876 port as it is probed via PCI. The poll interval is
1877 optional and is the number seconds in between
1878 each poll cycle to the debug port in case you need
1879 the functionality for interrupting the kernel with
1880 gdb or control-c on the dbgp connection. When
1881 not using this parameter you use sysrq-g to break into
1882 the kernel debugger.
1883
1884 kgdboc= [KGDB,HW] kgdb over consoles.
1885 Requires a tty driver that supports console polling,
1886 or a supported polling keyboard driver (non-usb).
1887 Serial only format: <serial_device>[,baud]
1888 keyboard only format: kbd
1889 keyboard and serial format: kbd,<serial_device>[,baud]
1890 Optional Kernel mode setting:
1891 kms, kbd format: kms,kbd
1892 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
1893
1894 kgdbwait [KGDB] Stop kernel execution and enter the
1895 kernel debugger at the earliest opportunity.
1896
1897 kmac= [MIPS] korina ethernet MAC address.
1898 Configure the RouterBoard 532 series on-chip
1899 Ethernet adapter MAC address.
1900
1901 kmemleak= [KNL] Boot-time kmemleak enable/disable
1902 Valid arguments: on, off
1903 Default: on
1904 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
1905 the default is off.
1906
1907 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
1908 Default is 0 (don't ignore, but inject #GP)
1909
1910 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
1911 Default is false (don't support).
1912
1913 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
1914 KVM MMU at runtime.
1915 Default is 0 (off)
1916
1917 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
1918 Default is 1 (enabled)
1919
1920 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
1921 for all guests.
1922 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
1923
1924 kvm-arm.vgic_v3_group0_trap=
1925 [KVM,ARM] Trap guest accesses to GICv3 group-0
1926 system registers
1927
1928 kvm-arm.vgic_v3_group1_trap=
1929 [KVM,ARM] Trap guest accesses to GICv3 group-1
1930 system registers
1931
1932 kvm-arm.vgic_v3_common_trap=
1933 [KVM,ARM] Trap guest accesses to GICv3 common
1934 system registers
1935
1936 kvm-arm.vgic_v4_enable=
1937 [KVM,ARM] Allow use of GICv4 for direct injection of
1938 LPIs.
1939
1940 kvm-intel.ept= [KVM,Intel] Disable extended page tables
1941 (virtualized MMU) support on capable Intel chips.
1942 Default is 1 (enabled)
1943
1944 kvm-intel.emulate_invalid_guest_state=
1945 [KVM,Intel] Enable emulation of invalid guest states
1946 Default is 0 (disabled)
1947
1948 kvm-intel.flexpriority=
1949 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
1950 Default is 1 (enabled)
1951
1952 kvm-intel.nested=
1953 [KVM,Intel] Enable VMX nesting (nVMX).
1954 Default is 0 (disabled)
1955
1956 kvm-intel.unrestricted_guest=
1957 [KVM,Intel] Disable unrestricted guest feature
1958 (virtualized real and unpaged mode) on capable
1959 Intel chips. Default is 1 (enabled)
1960
1961 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
1962 feature (tagged TLBs) on capable Intel chips.
1963 Default is 1 (enabled)
1964
1965 l2cr= [PPC]
1966
1967 l3cr= [PPC]
1968
1969 lapic [X86-32,APIC] Enable the local APIC even if BIOS
1970 disabled it.
1971
1972 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
1973 value for LAPIC timer one-shot implementation. Default
1974 back to the programmable timer unit in the LAPIC.
1975
1976 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
1977 in C2 power state.
1978
1979 libata.dma= [LIBATA] DMA control
1980 libata.dma=0 Disable all PATA and SATA DMA
1981 libata.dma=1 PATA and SATA Disk DMA only
1982 libata.dma=2 ATAPI (CDROM) DMA only
1983 libata.dma=4 Compact Flash DMA only
1984 Combinations also work, so libata.dma=3 enables DMA
1985 for disks and CDROMs, but not CFs.
1986
1987 libata.ignore_hpa= [LIBATA] Ignore HPA limit
1988 libata.ignore_hpa=0 keep BIOS limits (default)
1989 libata.ignore_hpa=1 ignore limits, using full disk
1990
1991 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
1992 when set.
1993 Format: <int>
1994
1995 libata.force= [LIBATA] Force configurations. The format is comma
1996 separated list of "[ID:]VAL" where ID is
1997 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
1998 matching port, link or device. Basically, it matches
1999 the ATA ID string printed on console by libata. If
2000 the whole ID part is omitted, the last PORT and DEVICE
2001 values are used. If ID hasn't been specified yet, the
2002 configuration applies to all ports, links and devices.
2003
2004 If only DEVICE is omitted, the parameter applies to
2005 the port and all links and devices behind it. DEVICE
2006 number of 0 either selects the first device or the
2007 first fan-out link behind PMP device. It does not
2008 select the host link. DEVICE number of 15 selects the
2009 host link and device attached to it.
2010
2011 The VAL specifies the configuration to force. As long
2012 as there's no ambiguity shortcut notation is allowed.
2013 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2014 The following configurations can be forced.
2015
2016 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2017 Any ID with matching PORT is used.
2018
2019 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2020
2021 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2022 udma[/][16,25,33,44,66,100,133] notation is also
2023 allowed.
2024
2025 * [no]ncq: Turn on or off NCQ.
2026
2027 * [no]ncqtrim: Turn off queued DSM TRIM.
2028
2029 * nohrst, nosrst, norst: suppress hard, soft
2030 and both resets.
2031
2032 * rstonce: only attempt one reset during
2033 hot-unplug link recovery
2034
2035 * dump_id: dump IDENTIFY data.
2036
2037 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2038
2039 * disable: Disable this device.
2040
2041 If there are multiple matching configurations changing
2042 the same attribute, the last one is used.
2043
2044 memblock=debug [KNL] Enable memblock debug messages.
2045
2046 load_ramdisk= [RAM] List of ramdisks to load from floppy
2047 See Documentation/blockdev/ramdisk.txt.
2048
2049 lockd.nlm_grace_period=P [NFS] Assign grace period.
2050 Format: <integer>
2051
2052 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2053 Format: <integer>
2054
2055 lockd.nlm_timeout=T [NFS] Assign timeout value.
2056 Format: <integer>
2057
2058 lockd.nlm_udpport=M [NFS] Assign UDP port.
2059 Format: <integer>
2060
2061 locktorture.nreaders_stress= [KNL]
2062 Set the number of locking read-acquisition kthreads.
2063 Defaults to being automatically set based on the
2064 number of online CPUs.
2065
2066 locktorture.nwriters_stress= [KNL]
2067 Set the number of locking write-acquisition kthreads.
2068
2069 locktorture.onoff_holdoff= [KNL]
2070 Set time (s) after boot for CPU-hotplug testing.
2071
2072 locktorture.onoff_interval= [KNL]
2073 Set time (s) between CPU-hotplug operations, or
2074 zero to disable CPU-hotplug testing.
2075
2076 locktorture.shuffle_interval= [KNL]
2077 Set task-shuffle interval (jiffies). Shuffling
2078 tasks allows some CPUs to go into dyntick-idle
2079 mode during the locktorture test.
2080
2081 locktorture.shutdown_secs= [KNL]
2082 Set time (s) after boot system shutdown. This
2083 is useful for hands-off automated testing.
2084
2085 locktorture.stat_interval= [KNL]
2086 Time (s) between statistics printk()s.
2087
2088 locktorture.stutter= [KNL]
2089 Time (s) to stutter testing, for example,
2090 specifying five seconds causes the test to run for
2091 five seconds, wait for five seconds, and so on.
2092 This tests the locking primitive's ability to
2093 transition abruptly to and from idle.
2094
2095 locktorture.torture_type= [KNL]
2096 Specify the locking implementation to test.
2097
2098 locktorture.verbose= [KNL]
2099 Enable additional printk() statements.
2100
2101 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2102 Format: <irq>
2103
2104 loglevel= All Kernel Messages with a loglevel smaller than the
2105 console loglevel will be printed to the console. It can
2106 also be changed with klogd or other programs. The
2107 loglevels are defined as follows:
2108
2109 0 (KERN_EMERG) system is unusable
2110 1 (KERN_ALERT) action must be taken immediately
2111 2 (KERN_CRIT) critical conditions
2112 3 (KERN_ERR) error conditions
2113 4 (KERN_WARNING) warning conditions
2114 5 (KERN_NOTICE) normal but significant condition
2115 6 (KERN_INFO) informational
2116 7 (KERN_DEBUG) debug-level messages
2117
2118 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2119 in bytes. n must be a power of two and greater
2120 than the minimal size. The minimal size is defined
2121 by LOG_BUF_SHIFT kernel config parameter. There is
2122 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2123 that allows to increase the default size depending on
2124 the number of CPUs. See init/Kconfig for more details.
2125
2126 logo.nologo [FB] Disables display of the built-in Linux logo.
2127 This may be used to provide more screen space for
2128 kernel log messages and is useful when debugging
2129 kernel boot problems.
2130
2131 lp=0 [LP] Specify parallel ports to use, e.g,
2132 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2133 lp=reset first parallel port). 'lp=0' disables the
2134 lp=auto printer driver. 'lp=reset' (which can be
2135 specified in addition to the ports) causes
2136 attached printers to be reset. Using
2137 lp=port1,port2,... specifies the parallel ports
2138 to associate lp devices with, starting with
2139 lp0. A port specification may be 'none' to skip
2140 that lp device, or a parport name such as
2141 'parport0'. Specifying 'lp=auto' instead of a
2142 port specification list means that device IDs
2143 from each port should be examined, to see if
2144 an IEEE 1284-compliant printer is attached; if
2145 so, the driver will manage that printer.
2146 See also header of drivers/char/lp.c.
2147
2148 lpj=n [KNL]
2149 Sets loops_per_jiffy to given constant, thus avoiding
2150 time-consuming boot-time autodetection (up to 250 ms per
2151 CPU). 0 enables autodetection (default). To determine
2152 the correct value for your kernel, boot with normal
2153 autodetection and see what value is printed. Note that
2154 on SMP systems the preset will be applied to all CPUs,
2155 which is likely to cause problems if your CPUs need
2156 significantly divergent settings. An incorrect value
2157 will cause delays in the kernel to be wrong, leading to
2158 unpredictable I/O errors and other breakage. Although
2159 unlikely, in the extreme case this might damage your
2160 hardware.
2161
2162 ltpc= [NET]
2163 Format: <io>,<irq>,<dma>
2164
2165 machvec= [IA-64] Force the use of a particular machine-vector
2166 (machvec) in a generic kernel.
2167 Example: machvec=hpzx1_swiotlb
2168
2169 machtype= [Loongson] Share the same kernel image file between different
2170 yeeloong laptop.
2171 Example: machtype=lemote-yeeloong-2f-7inch
2172
2173 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2174 than or equal to this physical address is ignored.
2175
2176 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2177 will bring up during bootup. maxcpus=n : n >= 0 limits
2178 the kernel to bring up 'n' processors. Surely after
2179 bootup you can bring up the other plugged cpu by executing
2180 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2181 only takes effect during system bootup.
2182 While n=0 is a special case, it is equivalent to "nosmp",
2183 which also disables the IO APIC.
2184
2185 max_loop= [LOOP] The number of loop block devices that get
2186 (loop.max_loop) unconditionally pre-created at init time. The default
2187 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2188 of statically allocating a predefined number, loop
2189 devices can be requested on-demand with the
2190 /dev/loop-control interface.
2191
2192 mce [X86-32] Machine Check Exception
2193
2194 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.txt
2195
2196 md= [HW] RAID subsystems devices and level
2197 See Documentation/admin-guide/md.rst.
2198
2199 mdacon= [MDA]
2200 Format: <first>,<last>
2201 Specifies range of consoles to be captured by the MDA.
2202
2203 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2204 Amount of memory to be used when the kernel is not able
2205 to see the whole system memory or for test.
2206 [X86] Work as limiting max address. Use together
2207 with memmap= to avoid physical address space collisions.
2208 Without memmap= PCI devices could be placed at addresses
2209 belonging to unused RAM.
2210
2211 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2212 memory.
2213
2214 memchunk=nn[KMG]
2215 [KNL,SH] Allow user to override the default size for
2216 per-device physically contiguous DMA buffers.
2217
2218 memhp_default_state=online/offline
2219 [KNL] Set the initial state for the memory hotplug
2220 onlining policy. If not specified, the default value is
2221 set according to the
2222 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2223 option.
2224 See Documentation/memory-hotplug.txt.
2225
2226 memmap=exactmap [KNL,X86] Enable setting of an exact
2227 E820 memory map, as specified by the user.
2228 Such memmap=exactmap lines can be constructed based on
2229 BIOS output or other requirements. See the memmap=nn@ss
2230 option description.
2231
2232 memmap=nn[KMG]@ss[KMG]
2233 [KNL] Force usage of a specific region of memory.
2234 Region of memory to be used is from ss to ss+nn.
2235 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2236 which limits max address to nn[KMG].
2237 Multiple different regions can be specified,
2238 comma delimited.
2239 Example:
2240 memmap=100M@2G,100M#3G,1G!1024G
2241
2242 memmap=nn[KMG]#ss[KMG]
2243 [KNL,ACPI] Mark specific memory as ACPI data.
2244 Region of memory to be marked is from ss to ss+nn.
2245
2246 memmap=nn[KMG]$ss[KMG]
2247 [KNL,ACPI] Mark specific memory as reserved.
2248 Region of memory to be reserved is from ss to ss+nn.
2249 Example: Exclude memory from 0x18690000-0x1869ffff
2250 memmap=64K$0x18690000
2251 or
2252 memmap=0x10000$0x18690000
2253 Some bootloaders may need an escape character before '$',
2254 like Grub2, otherwise '$' and the following number
2255 will be eaten.
2256
2257 memmap=nn[KMG]!ss[KMG]
2258 [KNL,X86] Mark specific memory as protected.
2259 Region of memory to be used, from ss to ss+nn.
2260 The memory region may be marked as e820 type 12 (0xc)
2261 and is NVDIMM or ADR memory.
2262
2263 memmap=<size>%<offset>-<oldtype>+<newtype>
2264 [KNL,ACPI] Convert memory within the specified region
2265 from <oldtype> to <newtype>. If "-<oldtype>" is left
2266 out, the whole region will be marked as <newtype>,
2267 even if previously unavailable. If "+<newtype>" is left
2268 out, matching memory will be removed. Types are
2269 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2270 3 = ACPI, 12 = PRAM.
2271
2272 memory_corruption_check=0/1 [X86]
2273 Some BIOSes seem to corrupt the first 64k of
2274 memory when doing things like suspend/resume.
2275 Setting this option will scan the memory
2276 looking for corruption. Enabling this will
2277 both detect corruption and prevent the kernel
2278 from using the memory being corrupted.
2279 However, its intended as a diagnostic tool; if
2280 repeatable BIOS-originated corruption always
2281 affects the same memory, you can use memmap=
2282 to prevent the kernel from using that memory.
2283
2284 memory_corruption_check_size=size [X86]
2285 By default it checks for corruption in the low
2286 64k, making this memory unavailable for normal
2287 use. Use this parameter to scan for
2288 corruption in more or less memory.
2289
2290 memory_corruption_check_period=seconds [X86]
2291 By default it checks for corruption every 60
2292 seconds. Use this parameter to check at some
2293 other rate. 0 disables periodic checking.
2294
2295 memtest= [KNL,X86,ARM] Enable memtest
2296 Format: <integer>
2297 default : 0 <disable>
2298 Specifies the number of memtest passes to be
2299 performed. Each pass selects another test
2300 pattern from a given set of patterns. Memtest
2301 fills the memory with this pattern, validates
2302 memory contents and reserves bad memory
2303 regions that are detected.
2304
2305 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2306 Valid arguments: on, off
2307 Default (depends on kernel configuration option):
2308 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2309 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2310 mem_encrypt=on: Activate SME
2311 mem_encrypt=off: Do not activate SME
2312
2313 Refer to Documentation/x86/amd-memory-encryption.txt
2314 for details on when memory encryption can be activated.
2315
2316 mem_sleep_default= [SUSPEND] Default system suspend mode:
2317 s2idle - Suspend-To-Idle
2318 shallow - Power-On Suspend or equivalent (if supported)
2319 deep - Suspend-To-RAM or equivalent (if supported)
2320 See Documentation/admin-guide/pm/sleep-states.rst.
2321
2322 meye.*= [HW] Set MotionEye Camera parameters
2323 See Documentation/media/v4l-drivers/meye.rst.
2324
2325 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2326 Multi-Function General Purpose Timers on AMD Geode
2327 platforms.
2328
2329 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2330 the BIOS has incorrectly applied a workaround. TinyBIOS
2331 version 0.98 is known to be affected, 0.99 fixes the
2332 problem by letting the user disable the workaround.
2333
2334 mga= [HW,DRM]
2335
2336 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2337 physical address is ignored.
2338
2339 mini2440= [ARM,HW,KNL]
2340 Format:[0..2][b][c][t]
2341 Default: "0tb"
2342 MINI2440 configuration specification:
2343 0 - The attached screen is the 3.5" TFT
2344 1 - The attached screen is the 7" TFT
2345 2 - The VGA Shield is attached (1024x768)
2346 Leaving out the screen size parameter will not load
2347 the TFT driver, and the framebuffer will be left
2348 unconfigured.
2349 b - Enable backlight. The TFT backlight pin will be
2350 linked to the kernel VESA blanking code and a GPIO
2351 LED. This parameter is not necessary when using the
2352 VGA shield.
2353 c - Enable the s3c camera interface.
2354 t - Reserved for enabling touchscreen support. The
2355 touchscreen support is not enabled in the mainstream
2356 kernel as of 2.6.30, a preliminary port can be found
2357 in the "bleeding edge" mini2440 support kernel at
2358 http://repo.or.cz/w/linux-2.6/mini2440.git
2359
2360 mminit_loglevel=
2361 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2362 parameter allows control of the logging verbosity for
2363 the additional memory initialisation checks. A value
2364 of 0 disables mminit logging and a level of 4 will
2365 log everything. Information is printed at KERN_DEBUG
2366 so loglevel=8 may also need to be specified.
2367
2368 module.sig_enforce
2369 [KNL] When CONFIG_MODULE_SIG is set, this means that
2370 modules without (valid) signatures will fail to load.
2371 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2372 is always true, so this option does nothing.
2373
2374 module_blacklist= [KNL] Do not load a comma-separated list of
2375 modules. Useful for debugging problem modules.
2376
2377 mousedev.tap_time=
2378 [MOUSE] Maximum time between finger touching and
2379 leaving touchpad surface for touch to be considered
2380 a tap and be reported as a left button click (for
2381 touchpads working in absolute mode only).
2382 Format: <msecs>
2383 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2384 reporting absolute coordinates, such as tablets
2385 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2386 reporting absolute coordinates, such as tablets
2387
2388 movablecore= [KNL,X86,IA-64,PPC]
2389 Format: nn[KMGTPE] | nn%
2390 This parameter is the complement to kernelcore=, it
2391 specifies the amount of memory used for migratable
2392 allocations. If both kernelcore and movablecore is
2393 specified, then kernelcore will be at *least* the
2394 specified value but may be more. If movablecore on its
2395 own is specified, the administrator must be careful
2396 that the amount of memory usable for all allocations
2397 is not too small.
2398
2399 movable_node [KNL] Boot-time switch to make hotplugable memory
2400 NUMA nodes to be movable. This means that the memory
2401 of such nodes will be usable only for movable
2402 allocations which rules out almost all kernel
2403 allocations. Use with caution!
2404
2405 MTD_Partition= [MTD]
2406 Format: <name>,<region-number>,<size>,<offset>
2407
2408 MTD_Region= [MTD] Format:
2409 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2410
2411 mtdparts= [MTD]
2412 See drivers/mtd/cmdlinepart.c.
2413
2414 multitce=off [PPC] This parameter disables the use of the pSeries
2415 firmware feature for updating multiple TCE entries
2416 at a time.
2417
2418 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2419
2420 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2421
2422 boundary - index of last SLC block on Flex-OneNAND.
2423 The remaining blocks are configured as MLC blocks.
2424 lock - Configure if Flex-OneNAND boundary should be locked.
2425 Once locked, the boundary cannot be changed.
2426 1 indicates lock status, 0 indicates unlock status.
2427
2428 mtdset= [ARM]
2429 ARM/S3C2412 JIVE boot control
2430
2431 See arch/arm/mach-s3c2412/mach-jive.c
2432
2433 mtouchusb.raw_coordinates=
2434 [HW] Make the MicroTouch USB driver use raw coordinates
2435 ('y', default) or cooked coordinates ('n')
2436
2437 mtrr_chunk_size=nn[KMG] [X86]
2438 used for mtrr cleanup. It is largest continuous chunk
2439 that could hold holes aka. UC entries.
2440
2441 mtrr_gran_size=nn[KMG] [X86]
2442 Used for mtrr cleanup. It is granularity of mtrr block.
2443 Default is 1.
2444 Large value could prevent small alignment from
2445 using up MTRRs.
2446
2447 mtrr_spare_reg_nr=n [X86]
2448 Format: <integer>
2449 Range: 0,7 : spare reg number
2450 Default : 1
2451 Used for mtrr cleanup. It is spare mtrr entries number.
2452 Set to 2 or more if your graphical card needs more.
2453
2454 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2455
2456 netdev= [NET] Network devices parameters
2457 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2458 Note that mem_start is often overloaded to mean
2459 something different and driver-specific.
2460 This usage is only documented in each driver source
2461 file if at all.
2462
2463 nf_conntrack.acct=
2464 [NETFILTER] Enable connection tracking flow accounting
2465 0 to disable accounting
2466 1 to enable accounting
2467 Default value is 0.
2468
2469 nfsaddrs= [NFS] Deprecated. Use ip= instead.
2470 See Documentation/filesystems/nfs/nfsroot.txt.
2471
2472 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
2473 See Documentation/filesystems/nfs/nfsroot.txt.
2474
2475 nfsrootdebug [NFS] enable nfsroot debugging messages.
2476 See Documentation/filesystems/nfs/nfsroot.txt.
2477
2478 nfs.callback_nr_threads=
2479 [NFSv4] set the total number of threads that the
2480 NFS client will assign to service NFSv4 callback
2481 requests.
2482
2483 nfs.callback_tcpport=
2484 [NFS] set the TCP port on which the NFSv4 callback
2485 channel should listen.
2486
2487 nfs.cache_getent=
2488 [NFS] sets the pathname to the program which is used
2489 to update the NFS client cache entries.
2490
2491 nfs.cache_getent_timeout=
2492 [NFS] sets the timeout after which an attempt to
2493 update a cache entry is deemed to have failed.
2494
2495 nfs.idmap_cache_timeout=
2496 [NFS] set the maximum lifetime for idmapper cache
2497 entries.
2498
2499 nfs.enable_ino64=
2500 [NFS] enable 64-bit inode numbers.
2501 If zero, the NFS client will fake up a 32-bit inode
2502 number for the readdir() and stat() syscalls instead
2503 of returning the full 64-bit number.
2504 The default is to return 64-bit inode numbers.
2505
2506 nfs.max_session_cb_slots=
2507 [NFSv4.1] Sets the maximum number of session
2508 slots the client will assign to the callback
2509 channel. This determines the maximum number of
2510 callbacks the client will process in parallel for
2511 a particular server.
2512
2513 nfs.max_session_slots=
2514 [NFSv4.1] Sets the maximum number of session slots
2515 the client will attempt to negotiate with the server.
2516 This limits the number of simultaneous RPC requests
2517 that the client can send to the NFSv4.1 server.
2518 Note that there is little point in setting this
2519 value higher than the max_tcp_slot_table_limit.
2520
2521 nfs.nfs4_disable_idmapping=
2522 [NFSv4] When set to the default of '1', this option
2523 ensures that both the RPC level authentication
2524 scheme and the NFS level operations agree to use
2525 numeric uids/gids if the mount is using the
2526 'sec=sys' security flavour. In effect it is
2527 disabling idmapping, which can make migration from
2528 legacy NFSv2/v3 systems to NFSv4 easier.
2529 Servers that do not support this mode of operation
2530 will be autodetected by the client, and it will fall
2531 back to using the idmapper.
2532 To turn off this behaviour, set the value to '0'.
2533 nfs.nfs4_unique_id=
2534 [NFS4] Specify an additional fixed unique ident-
2535 ification string that NFSv4 clients can insert into
2536 their nfs_client_id4 string. This is typically a
2537 UUID that is generated at system install time.
2538
2539 nfs.send_implementation_id =
2540 [NFSv4.1] Send client implementation identification
2541 information in exchange_id requests.
2542 If zero, no implementation identification information
2543 will be sent.
2544 The default is to send the implementation identification
2545 information.
2546
2547 nfs.recover_lost_locks =
2548 [NFSv4] Attempt to recover locks that were lost due
2549 to a lease timeout on the server. Please note that
2550 doing this risks data corruption, since there are
2551 no guarantees that the file will remain unchanged
2552 after the locks are lost.
2553 If you want to enable the kernel legacy behaviour of
2554 attempting to recover these locks, then set this
2555 parameter to '1'.
2556 The default parameter value of '0' causes the kernel
2557 not to attempt recovery of lost locks.
2558
2559 nfs4.layoutstats_timer =
2560 [NFSv4.2] Change the rate at which the kernel sends
2561 layoutstats to the pNFS metadata server.
2562
2563 Setting this to value to 0 causes the kernel to use
2564 whatever value is the default set by the layout
2565 driver. A non-zero value sets the minimum interval
2566 in seconds between layoutstats transmissions.
2567
2568 nfsd.nfs4_disable_idmapping=
2569 [NFSv4] When set to the default of '1', the NFSv4
2570 server will return only numeric uids and gids to
2571 clients using auth_sys, and will accept numeric uids
2572 and gids from such clients. This is intended to ease
2573 migration from NFSv2/v3.
2574
2575 nmi_debug= [KNL,SH] Specify one or more actions to take
2576 when a NMI is triggered.
2577 Format: [state][,regs][,debounce][,die]
2578
2579 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
2580 Format: [panic,][nopanic,][num]
2581 Valid num: 0 or 1
2582 0 - turn hardlockup detector in nmi_watchdog off
2583 1 - turn hardlockup detector in nmi_watchdog on
2584 When panic is specified, panic when an NMI watchdog
2585 timeout occurs (or 'nopanic' to override the opposite
2586 default). To disable both hard and soft lockup detectors,
2587 please see 'nowatchdog'.
2588 This is useful when you use a panic=... timeout and
2589 need the box quickly up again.
2590
2591 These settings can be accessed at runtime via
2592 the nmi_watchdog and hardlockup_panic sysctls.
2593
2594 netpoll.carrier_timeout=
2595 [NET] Specifies amount of time (in seconds) that
2596 netpoll should wait for a carrier. By default netpoll
2597 waits 4 seconds.
2598
2599 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
2600 emulation library even if a 387 maths coprocessor
2601 is present.
2602
2603 no_console_suspend
2604 [HW] Never suspend the console
2605 Disable suspending of consoles during suspend and
2606 hibernate operations. Once disabled, debugging
2607 messages can reach various consoles while the rest
2608 of the system is being put to sleep (ie, while
2609 debugging driver suspend/resume hooks). This may
2610 not work reliably with all consoles, but is known
2611 to work with serial and VGA consoles.
2612 To facilitate more flexible debugging, we also add
2613 console_suspend, a printk module parameter to control
2614 it. Users could use console_suspend (usually
2615 /sys/module/printk/parameters/console_suspend) to
2616 turn on/off it dynamically.
2617
2618 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
2619 caches in the slab allocator. Saves per-node memory,
2620 but will impact performance.
2621
2622 noalign [KNL,ARM]
2623
2624 noaltinstr [S390] Disables alternative instructions patching
2625 (CPU alternatives feature).
2626
2627 noapic [SMP,APIC] Tells the kernel to not make use of any
2628 IOAPICs that may be present in the system.
2629
2630 noautogroup Disable scheduler automatic task group creation.
2631
2632 nobats [PPC] Do not use BATs for mapping kernel lowmem
2633 on "Classic" PPC cores.
2634
2635 nocache [ARM]
2636
2637 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
2638
2639 nodelayacct [KNL] Disable per-task delay accounting
2640
2641 nodsp [SH] Disable hardware DSP at boot time.
2642
2643 noefi Disable EFI runtime services support.
2644
2645 noexec [IA-64]
2646
2647 noexec [X86]
2648 On X86-32 available only on PAE configured kernels.
2649 noexec=on: enable non-executable mappings (default)
2650 noexec=off: disable non-executable mappings
2651
2652 nosmap [X86]
2653 Disable SMAP (Supervisor Mode Access Prevention)
2654 even if it is supported by processor.
2655
2656 nosmep [X86]
2657 Disable SMEP (Supervisor Mode Execution Prevention)
2658 even if it is supported by processor.
2659
2660 noexec32 [X86-64]
2661 This affects only 32-bit executables.
2662 noexec32=on: enable non-executable mappings (default)
2663 read doesn't imply executable mappings
2664 noexec32=off: disable non-executable mappings
2665 read implies executable mappings
2666
2667 nofpu [MIPS,SH] Disable hardware FPU at boot time.
2668
2669 nofxsr [BUGS=X86-32] Disables x86 floating point extended
2670 register save and restore. The kernel will only save
2671 legacy floating-point registers on task switch.
2672
2673 nohugeiomap [KNL,x86] Disable kernel huge I/O mappings.
2674
2675 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
2676 Equivalent to smt=1.
2677
2678 nospectre_v2 [X86] Disable all mitigations for the Spectre variant 2
2679 (indirect branch prediction) vulnerability. System may
2680 allow data leaks with this option, which is equivalent
2681 to spectre_v2=off.
2682
2683 nospec_store_bypass_disable
2684 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
2685
2686 noxsave [BUGS=X86] Disables x86 extended register state save
2687 and restore using xsave. The kernel will fallback to
2688 enabling legacy floating-point and sse state.
2689
2690 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
2691 register states. The kernel will fall back to use
2692 xsave to save the states. By using this parameter,
2693 performance of saving the states is degraded because
2694 xsave doesn't support modified optimization while
2695 xsaveopt supports it on xsaveopt enabled systems.
2696
2697 noxsaves [X86] Disables xsaves and xrstors used in saving and
2698 restoring x86 extended register state in compacted
2699 form of xsave area. The kernel will fall back to use
2700 xsaveopt and xrstor to save and restore the states
2701 in standard form of xsave area. By using this
2702 parameter, xsave area per process might occupy more
2703 memory on xsaves enabled systems.
2704
2705 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
2706 wfi(ARM) instruction doesn't work correctly and not to
2707 use it. This is also useful when using JTAG debugger.
2708
2709 no_file_caps Tells the kernel not to honor file capabilities. The
2710 only way then for a file to be executed with privilege
2711 is to be setuid root or executed by root.
2712
2713 nohalt [IA-64] Tells the kernel not to use the power saving
2714 function PAL_HALT_LIGHT when idle. This increases
2715 power-consumption. On the positive side, it reduces
2716 interrupt wake-up latency, which may improve performance
2717 in certain environments such as networked servers or
2718 real-time systems.
2719
2720 nohibernate [HIBERNATION] Disable hibernation and resume.
2721
2722 nohz= [KNL] Boottime enable/disable dynamic ticks
2723 Valid arguments: on, off
2724 Default: on
2725
2726 nohz_full= [KNL,BOOT,SMP,ISOL]
2727 The argument is a cpu list, as described above.
2728 In kernels built with CONFIG_NO_HZ_FULL=y, set
2729 the specified list of CPUs whose tick will be stopped
2730 whenever possible. The boot CPU will be forced outside
2731 the range to maintain the timekeeping. Any CPUs
2732 in this list will have their RCU callbacks offloaded,
2733 just as if they had also been called out in the
2734 rcu_nocbs= boot parameter.
2735
2736 noiotrap [SH] Disables trapped I/O port accesses.
2737
2738 noirqdebug [X86-32] Disables the code which attempts to detect and
2739 disable unhandled interrupt sources.
2740
2741 no_timer_check [X86,APIC] Disables the code which tests for
2742 broken timer IRQ sources.
2743
2744 noisapnp [ISAPNP] Disables ISA PnP code.
2745
2746 noinitrd [RAM] Tells the kernel not to load any configured
2747 initial RAM disk.
2748
2749 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
2750 remapping.
2751 [Deprecated - use intremap=off]
2752
2753 nointroute [IA-64]
2754
2755 noinvpcid [X86] Disable the INVPCID cpu feature.
2756
2757 nojitter [IA-64] Disables jitter checking for ITC timers.
2758
2759 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
2760
2761 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
2762 fault handling.
2763
2764 no-vmw-sched-clock
2765 [X86,PV_OPS] Disable paravirtualized VMware scheduler
2766 clock and use the default one.
2767
2768 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting.
2769 steal time is computed, but won't influence scheduler
2770 behaviour
2771
2772 nolapic [X86-32,APIC] Do not enable or use the local APIC.
2773
2774 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
2775
2776 noltlbs [PPC] Do not use large page/tlb entries for kernel
2777 lowmem mapping on PPC40x and PPC8xx
2778
2779 nomca [IA-64] Disable machine check abort handling
2780
2781 nomce [X86-32] Disable Machine Check Exception
2782
2783 nomfgpt [X86-32] Disable Multi-Function General Purpose
2784 Timer usage (for AMD Geode machines).
2785
2786 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
2787 shutdown the other cpus. Instead use the REBOOT_VECTOR
2788 irq.
2789
2790 nomodule Disable module load
2791
2792 nopat [X86] Disable PAT (page attribute table extension of
2793 pagetables) support.
2794
2795 nopcid [X86-64] Disable the PCID cpu feature.
2796
2797 norandmaps Don't use address space randomization. Equivalent to
2798 echo 0 > /proc/sys/kernel/randomize_va_space
2799
2800 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
2801 with UP alternatives
2802
2803 nordrand [X86] Disable kernel use of the RDRAND and
2804 RDSEED instructions even if they are supported
2805 by the processor. RDRAND and RDSEED are still
2806 available to user space applications.
2807
2808 noresume [SWSUSP] Disables resume and restores original swap
2809 space.
2810
2811 no-scroll [VGA] Disables scrollback.
2812 This is required for the Braillex ib80-piezo Braille
2813 reader made by F.H. Papenmeier (Germany).
2814
2815 nosbagart [IA-64]
2816
2817 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
2818
2819 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
2820 and disable the IO APIC. legacy for "maxcpus=0".
2821
2822 nosoftlockup [KNL] Disable the soft-lockup detector.
2823
2824 nosync [HW,M68K] Disables sync negotiation for all devices.
2825
2826 notsc [BUGS=X86-32] Disable Time Stamp Counter
2827
2828 nowatchdog [KNL] Disable both lockup detectors, i.e.
2829 soft-lockup and NMI watchdog (hard-lockup).
2830
2831 nowb [ARM]
2832
2833 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
2834
2835 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
2836 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
2837 Some features depend on CPU0. Known dependencies are:
2838 1. Resume from suspend/hibernate depends on CPU0.
2839 Suspend/hibernate will fail if CPU0 is offline and you
2840 need to online CPU0 before suspend/hibernate.
2841 2. PIC interrupts also depend on CPU0. CPU0 can't be
2842 removed if a PIC interrupt is detected.
2843 It's said poweroff/reboot may depend on CPU0 on some
2844 machines although I haven't seen such issues so far
2845 after CPU0 is offline on a few tested machines.
2846 If the dependencies are under your control, you can
2847 turn on cpu0_hotplug.
2848
2849 nps_mtm_hs_ctr= [KNL,ARC]
2850 This parameter sets the maximum duration, in
2851 cycles, each HW thread of the CTOP can run
2852 without interruptions, before HW switches it.
2853 The actual maximum duration is 16 times this
2854 parameter's value.
2855 Format: integer between 1 and 255
2856 Default: 255
2857
2858 nptcg= [IA-64] Override max number of concurrent global TLB
2859 purges which is reported from either PAL_VM_SUMMARY or
2860 SAL PALO.
2861
2862 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
2863 could support. nr_cpus=n : n >= 1 limits the kernel to
2864 support 'n' processors. It could be larger than the
2865 number of already plugged CPU during bootup, later in
2866 runtime you can physically add extra cpu until it reaches
2867 n. So during boot up some boot time memory for per-cpu
2868 variables need be pre-allocated for later physical cpu
2869 hot plugging.
2870
2871 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
2872
2873 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
2874 Allowed values are enable and disable
2875
2876 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
2877 'node', 'default' can be specified
2878 This can be set from sysctl after boot.
2879 See Documentation/sysctl/vm.txt for details.
2880
2881 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
2882 See Documentation/debugging-via-ohci1394.txt for more
2883 info.
2884
2885 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
2886 Rather than timing out after 20 ms if an EC
2887 command is not properly ACKed, override the length
2888 of the timeout. We have interrupts disabled while
2889 waiting for the ACK, so if this is set too high
2890 interrupts *may* be lost!
2891
2892 omap_mux= [OMAP] Override bootloader pin multiplexing.
2893 Format: <mux_mode0.mode_name=value>...
2894 For example, to override I2C bus2:
2895 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
2896
2897 oprofile.timer= [HW]
2898 Use timer interrupt instead of performance counters
2899
2900 oprofile.cpu_type= Force an oprofile cpu type
2901 This might be useful if you have an older oprofile
2902 userland or if you want common events.
2903 Format: { arch_perfmon }
2904 arch_perfmon: [X86] Force use of architectural
2905 perfmon on Intel CPUs instead of the
2906 CPU specific event set.
2907 timer: [X86] Force use of architectural NMI
2908 timer mode (see also oprofile.timer
2909 for generic hr timer mode)
2910
2911 oops=panic Always panic on oopses. Default is to just kill the
2912 process, but there is a small probability of
2913 deadlocking the machine.
2914 This will also cause panics on machine check exceptions.
2915 Useful together with panic=30 to trigger a reboot.
2916
2917 OSS [HW,OSS]
2918 See Documentation/sound/oss/oss-parameters.txt
2919
2920 page_owner= [KNL] Boot-time page_owner enabling option.
2921 Storage of the information about who allocated
2922 each page is disabled in default. With this switch,
2923 we can turn it on.
2924 on: enable the feature
2925
2926 page_poison= [KNL] Boot-time parameter changing the state of
2927 poisoning on the buddy allocator.
2928 off: turn off poisoning
2929 on: turn on poisoning
2930
2931 panic= [KNL] Kernel behaviour on panic: delay <timeout>
2932 timeout > 0: seconds before rebooting
2933 timeout = 0: wait forever
2934 timeout < 0: reboot immediately
2935 Format: <timeout>
2936
2937 panic_on_warn panic() instead of WARN(). Useful to cause kdump
2938 on a WARN().
2939
2940 crash_kexec_post_notifiers
2941 Run kdump after running panic-notifiers and dumping
2942 kmsg. This only for the users who doubt kdump always
2943 succeeds in any situation.
2944 Note that this also increases risks of kdump failure,
2945 because some panic notifiers can make the crashed
2946 kernel more unstable.
2947
2948 parkbd.port= [HW] Parallel port number the keyboard adapter is
2949 connected to, default is 0.
2950 Format: <parport#>
2951 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
2952 0 for XT, 1 for AT (default is AT).
2953 Format: <mode>
2954
2955 parport= [HW,PPT] Specify parallel ports. 0 disables.
2956 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
2957 Use 'auto' to force the driver to use any
2958 IRQ/DMA settings detected (the default is to
2959 ignore detected IRQ/DMA settings because of
2960 possible conflicts). You can specify the base
2961 address, IRQ, and DMA settings; IRQ and DMA
2962 should be numbers, or 'auto' (for using detected
2963 settings on that particular port), or 'nofifo'
2964 (to avoid using a FIFO even if it is detected).
2965 Parallel ports are assigned in the order they
2966 are specified on the command line, starting
2967 with parport0.
2968
2969 parport_init_mode= [HW,PPT]
2970 Configure VIA parallel port to operate in
2971 a specific mode. This is necessary on Pegasos
2972 computer where firmware has no options for setting
2973 up parallel port mode and sets it to spp.
2974 Currently this function knows 686a and 8231 chips.
2975 Format: [spp|ps2|epp|ecp|ecpepp]
2976
2977 pause_on_oops=
2978 Halt all CPUs after the first oops has been printed for
2979 the specified number of seconds. This is to be used if
2980 your oopses keep scrolling off the screen.
2981
2982 pcbit= [HW,ISDN]
2983
2984 pcd. [PARIDE]
2985 See header of drivers/block/paride/pcd.c.
2986 See also Documentation/blockdev/paride.txt.
2987
2988 pci=option[,option...] [PCI] various PCI subsystem options:
2989 earlydump [X86] dump PCI config space before the kernel
2990 changes anything
2991 off [X86] don't probe for the PCI bus
2992 bios [X86-32] force use of PCI BIOS, don't access
2993 the hardware directly. Use this if your machine
2994 has a non-standard PCI host bridge.
2995 nobios [X86-32] disallow use of PCI BIOS, only direct
2996 hardware access methods are allowed. Use this
2997 if you experience crashes upon bootup and you
2998 suspect they are caused by the BIOS.
2999 conf1 [X86] Force use of PCI Configuration Access
3000 Mechanism 1 (config address in IO port 0xCF8,
3001 data in IO port 0xCFC, both 32-bit).
3002 conf2 [X86] Force use of PCI Configuration Access
3003 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3004 the function, IO port 0xCFA, also 8-bit, sets
3005 bus number. The config space is then accessed
3006 through ports 0xC000-0xCFFF).
3007 See http://wiki.osdev.org/PCI for more info
3008 on the configuration access mechanisms.
3009 noaer [PCIE] If the PCIEAER kernel config parameter is
3010 enabled, this kernel boot option can be used to
3011 disable the use of PCIE advanced error reporting.
3012 nodomains [PCI] Disable support for multiple PCI
3013 root domains (aka PCI segments, in ACPI-speak).
3014 nommconf [X86] Disable use of MMCONFIG for PCI
3015 Configuration
3016 check_enable_amd_mmconf [X86] check for and enable
3017 properly configured MMIO access to PCI
3018 config space on AMD family 10h CPU
3019 nomsi [MSI] If the PCI_MSI kernel config parameter is
3020 enabled, this kernel boot option can be used to
3021 disable the use of MSI interrupts system-wide.
3022 noioapicquirk [APIC] Disable all boot interrupt quirks.
3023 Safety option to keep boot IRQs enabled. This
3024 should never be necessary.
3025 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3026 primary IO-APIC for bridges that cannot disable
3027 boot IRQs. This fixes a source of spurious IRQs
3028 when the system masks IRQs.
3029 noioapicreroute [APIC] Disable workaround that uses the
3030 boot IRQ equivalent of an IRQ that connects to
3031 a chipset where boot IRQs cannot be disabled.
3032 The opposite of ioapicreroute.
3033 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3034 routing table. These calls are known to be buggy
3035 on several machines and they hang the machine
3036 when used, but on other computers it's the only
3037 way to get the interrupt routing table. Try
3038 this option if the kernel is unable to allocate
3039 IRQs or discover secondary PCI buses on your
3040 motherboard.
3041 rom [X86] Assign address space to expansion ROMs.
3042 Use with caution as certain devices share
3043 address decoders between ROMs and other
3044 resources.
3045 norom [X86] Do not assign address space to
3046 expansion ROMs that do not already have
3047 BIOS assigned address ranges.
3048 nobar [X86] Do not assign address space to the
3049 BARs that weren't assigned by the BIOS.
3050 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3051 assigned automatically to PCI devices. You can
3052 make the kernel exclude IRQs of your ISA cards
3053 this way.
3054 pirqaddr=0xAAAAA [X86] Specify the physical address
3055 of the PIRQ table (normally generated
3056 by the BIOS) if it is outside the
3057 F0000h-100000h range.
3058 lastbus=N [X86] Scan all buses thru bus #N. Can be
3059 useful if the kernel is unable to find your
3060 secondary buses and you want to tell it
3061 explicitly which ones they are.
3062 assign-busses [X86] Always assign all PCI bus
3063 numbers ourselves, overriding
3064 whatever the firmware may have done.
3065 usepirqmask [X86] Honor the possible IRQ mask stored
3066 in the BIOS $PIR table. This is needed on
3067 some systems with broken BIOSes, notably
3068 some HP Pavilion N5400 and Omnibook XE3
3069 notebooks. This will have no effect if ACPI
3070 IRQ routing is enabled.
3071 noacpi [X86] Do not use ACPI for IRQ routing
3072 or for PCI scanning.
3073 use_crs [X86] Use PCI host bridge window information
3074 from ACPI. On BIOSes from 2008 or later, this
3075 is enabled by default. If you need to use this,
3076 please report a bug.
3077 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3078 If you need to use this, please report a bug.
3079 routeirq Do IRQ routing for all PCI devices.
3080 This is normally done in pci_enable_device(),
3081 so this option is a temporary workaround
3082 for broken drivers that don't call it.
3083 skip_isa_align [X86] do not align io start addr, so can
3084 handle more pci cards
3085 noearly [X86] Don't do any early type 1 scanning.
3086 This might help on some broken boards which
3087 machine check when some devices' config space
3088 is read. But various workarounds are disabled
3089 and some IOMMU drivers will not work.
3090 bfsort Sort PCI devices into breadth-first order.
3091 This sorting is done to get a device
3092 order compatible with older (<= 2.4) kernels.
3093 nobfsort Don't sort PCI devices into breadth-first order.
3094 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3095 tuning and use the BIOS-configured MPS defaults.
3096 pcie_bus_safe Set every device's MPS to the largest value
3097 supported by all devices below the root complex.
3098 pcie_bus_perf Set device MPS to the largest allowable MPS
3099 based on its parent bus. Also set MRRS (Max
3100 Read Request Size) to the largest supported
3101 value (no larger than the MPS that the device
3102 or bus can support) for best performance.
3103 pcie_bus_peer2peer Set every device's MPS to 128B, which
3104 every device is guaranteed to support. This
3105 configuration allows peer-to-peer DMA between
3106 any pair of devices, possibly at the cost of
3107 reduced performance. This also guarantees
3108 that hot-added devices will work.
3109 cbiosize=nn[KMG] The fixed amount of bus space which is
3110 reserved for the CardBus bridge's IO window.
3111 The default value is 256 bytes.
3112 cbmemsize=nn[KMG] The fixed amount of bus space which is
3113 reserved for the CardBus bridge's memory
3114 window. The default value is 64 megabytes.
3115 resource_alignment=
3116 Format:
3117 [<order of align>@][<domain>:]<bus>:<slot>.<func>[; ...]
3118 [<order of align>@]pci:<vendor>:<device>\
3119 [:<subvendor>:<subdevice>][; ...]
3120 Specifies alignment and device to reassign
3121 aligned memory resources.
3122 If <order of align> is not specified,
3123 PAGE_SIZE is used as alignment.
3124 PCI-PCI bridge can be specified, if resource
3125 windows need to be expanded.
3126 To specify the alignment for several
3127 instances of a device, the PCI vendor,
3128 device, subvendor, and subdevice may be
3129 specified, e.g., 4096@pci:8086:9c22:103c:198f
3130 ecrc= Enable/disable PCIe ECRC (transaction layer
3131 end-to-end CRC checking).
3132 bios: Use BIOS/firmware settings. This is the
3133 the default.
3134 off: Turn ECRC off
3135 on: Turn ECRC on.
3136 hpiosize=nn[KMG] The fixed amount of bus space which is
3137 reserved for hotplug bridge's IO window.
3138 Default size is 256 bytes.
3139 hpmemsize=nn[KMG] The fixed amount of bus space which is
3140 reserved for hotplug bridge's memory window.
3141 Default size is 2 megabytes.
3142 hpbussize=nn The minimum amount of additional bus numbers
3143 reserved for buses below a hotplug bridge.
3144 Default is 1.
3145 realloc= Enable/disable reallocating PCI bridge resources
3146 if allocations done by BIOS are too small to
3147 accommodate resources required by all child
3148 devices.
3149 off: Turn realloc off
3150 on: Turn realloc on
3151 realloc same as realloc=on
3152 noari do not use PCIe ARI.
3153 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3154 only look for one device below a PCIe downstream
3155 port.
3156 big_root_window Try to add a big 64bit memory window to the PCIe
3157 root complex on AMD CPUs. Some GFX hardware
3158 can resize a BAR to allow access to all VRAM.
3159 Adding the window is slightly risky (it may
3160 conflict with unreported devices), so this
3161 taints the kernel.
3162
3163 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3164 Management.
3165 off Disable ASPM.
3166 force Enable ASPM even on devices that claim not to support it.
3167 WARNING: Forcing ASPM on may cause system lockups.
3168
3169 pcie_ports= [PCIE] PCIe port services handling:
3170 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
3171 even if the platform doesn't give the OS permission to
3172 use them. This may cause conflicts if the platform
3173 also tries to use these services.
3174 compat Disable native PCIe services (PME, AER, DPC, PCIe
3175 hotplug).
3176
3177 pcie_port_pm= [PCIE] PCIe port power management handling:
3178 off Disable power management of all PCIe ports
3179 force Forcibly enable power management of all PCIe ports
3180
3181 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3182 nomsi Do not use MSI for native PCIe PME signaling (this makes
3183 all PCIe root ports use INTx for all services).
3184
3185 pcmv= [HW,PCMCIA] BadgePAD 4
3186
3187 pd_ignore_unused
3188 [PM]
3189 Keep all power-domains already enabled by bootloader on,
3190 even if no driver has claimed them. This is useful
3191 for debug and development, but should not be
3192 needed on a platform with proper driver support.
3193
3194 pd. [PARIDE]
3195 See Documentation/blockdev/paride.txt.
3196
3197 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3198 boot time.
3199 Format: { 0 | 1 }
3200 See arch/parisc/kernel/pdc_chassis.c
3201
3202 percpu_alloc= Select which percpu first chunk allocator to use.
3203 Currently supported values are "embed" and "page".
3204 Archs may support subset or none of the selections.
3205 See comments in mm/percpu.c for details on each
3206 allocator. This parameter is primarily for debugging
3207 and performance comparison.
3208
3209 pf. [PARIDE]
3210 See Documentation/blockdev/paride.txt.
3211
3212 pg. [PARIDE]
3213 See Documentation/blockdev/paride.txt.
3214
3215 pirq= [SMP,APIC] Manual mp-table setup
3216 See Documentation/x86/i386/IO-APIC.txt.
3217
3218 plip= [PPT,NET] Parallel port network link
3219 Format: { parport<nr> | timid | 0 }
3220 See also Documentation/admin-guide/parport.rst.
3221
3222 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3223 Override pmtimer IOPort with a hex value.
3224 e.g. pmtmr=0x508
3225
3226 pnp.debug=1 [PNP]
3227 Enable PNP debug messages (depends on the
3228 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3229 via /sys/module/pnp/parameters/debug. We always show
3230 current resource usage; turning this on also shows
3231 possible settings and some assignment information.
3232
3233 pnpacpi= [ACPI]
3234 { off }
3235
3236 pnpbios= [ISAPNP]
3237 { on | off | curr | res | no-curr | no-res }
3238
3239 pnp_reserve_irq=
3240 [ISAPNP] Exclude IRQs for the autoconfiguration
3241
3242 pnp_reserve_dma=
3243 [ISAPNP] Exclude DMAs for the autoconfiguration
3244
3245 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3246 Ranges are in pairs (I/O port base and size).
3247
3248 pnp_reserve_mem=
3249 [ISAPNP] Exclude memory regions for the
3250 autoconfiguration.
3251 Ranges are in pairs (memory base and size).
3252
3253 ports= [IP_VS_FTP] IPVS ftp helper module
3254 Default is 21.
3255 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3256 may be specified.
3257 Format: <port>,<port>....
3258
3259 powersave=off [PPC] This option disables power saving features.
3260 It specifically disables cpuidle and sets the
3261 platform machine description specific power_save
3262 function to NULL. On Idle the CPU just reduces
3263 execution priority.
3264
3265 ppc_strict_facility_enable
3266 [PPC] This option catches any kernel floating point,
3267 Altivec, VSX and SPE outside of regions specifically
3268 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3269 There is some performance impact when enabling this.
3270
3271 ppc_tm= [PPC]
3272 Format: {"off"}
3273 Disable Hardware Transactional Memory
3274
3275 print-fatal-signals=
3276 [KNL] debug: print fatal signals
3277
3278 If enabled, warn about various signal handling
3279 related application anomalies: too many signals,
3280 too many POSIX.1 timers, fatal signals causing a
3281 coredump - etc.
3282
3283 If you hit the warning due to signal overflow,
3284 you might want to try "ulimit -i unlimited".
3285
3286 default: off.
3287
3288 printk.always_kmsg_dump=
3289 Trigger kmsg_dump for cases other than kernel oops or
3290 panics
3291 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3292 default: disabled
3293
3294 printk.devkmsg={on,off,ratelimit}
3295 Control writing to /dev/kmsg.
3296 on - unlimited logging to /dev/kmsg from userspace
3297 off - logging to /dev/kmsg disabled
3298 ratelimit - ratelimit the logging
3299 Default: ratelimit
3300
3301 printk.time= Show timing data prefixed to each printk message line
3302 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3303
3304 processor.max_cstate= [HW,ACPI]
3305 Limit processor to maximum C-state
3306 max_cstate=9 overrides any DMI blacklist limit.
3307
3308 processor.nocst [HW,ACPI]
3309 Ignore the _CST method to determine C-states,
3310 instead using the legacy FADT method
3311
3312 profile= [KNL] Enable kernel profiling via /proc/profile
3313 Format: [<profiletype>,]<number>
3314 Param: <profiletype>: "schedule", "sleep", or "kvm"
3315 [defaults to kernel profiling]
3316 Param: "schedule" - profile schedule points.
3317 Param: "sleep" - profile D-state sleeping (millisecs).
3318 Requires CONFIG_SCHEDSTATS
3319 Param: "kvm" - profile VM exits.
3320 Param: <number> - step/bucket size as a power of 2 for
3321 statistical time based profiling.
3322
3323 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3324 before loading.
3325 See Documentation/blockdev/ramdisk.txt.
3326
3327 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3328 probe for; one of (bare|imps|exps|lifebook|any).
3329 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3330 per second.
3331 psmouse.resetafter= [HW,MOUSE]
3332 Try to reset the device after so many bad packets
3333 (0 = never).
3334 psmouse.resolution=
3335 [HW,MOUSE] Set desired mouse resolution, in dpi.
3336 psmouse.smartscroll=
3337 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
3338 0 = disabled, 1 = enabled (default).
3339
3340 pstore.backend= Specify the name of the pstore backend to use
3341
3342 pt. [PARIDE]
3343 See Documentation/blockdev/paride.txt.
3344
3345 pti= [X86_64] Control Page Table Isolation of user and
3346 kernel address spaces. Disabling this feature
3347 removes hardening, but improves performance of
3348 system calls and interrupts.
3349
3350 on - unconditionally enable
3351 off - unconditionally disable
3352 auto - kernel detects whether your CPU model is
3353 vulnerable to issues that PTI mitigates
3354
3355 Not specifying this option is equivalent to pti=auto.
3356
3357 nopti [X86_64]
3358 Equivalent to pti=off
3359
3360 pty.legacy_count=
3361 [KNL] Number of legacy pty's. Overwrites compiled-in
3362 default number.
3363
3364 quiet [KNL] Disable most log messages
3365
3366 r128= [HW,DRM]
3367
3368 raid= [HW,RAID]
3369 See Documentation/admin-guide/md.rst.
3370
3371 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
3372 See Documentation/blockdev/ramdisk.txt.
3373
3374 ras=option[,option,...] [KNL] RAS-specific options
3375
3376 cec_disable [X86]
3377 Disable the Correctable Errors Collector,
3378 see CONFIG_RAS_CEC help text.
3379
3380 rcu_nocbs= [KNL]
3381 The argument is a cpu list, as described above.
3382
3383 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
3384 the specified list of CPUs to be no-callback CPUs.
3385 Invocation of these CPUs' RCU callbacks will
3386 be offloaded to "rcuox/N" kthreads created for
3387 that purpose, where "x" is "b" for RCU-bh, "p"
3388 for RCU-preempt, and "s" for RCU-sched, and "N"
3389 is the CPU number. This reduces OS jitter on the
3390 offloaded CPUs, which can be useful for HPC and
3391 real-time workloads. It can also improve energy
3392 efficiency for asymmetric multiprocessors.
3393
3394 rcu_nocb_poll [KNL]
3395 Rather than requiring that offloaded CPUs
3396 (specified by rcu_nocbs= above) explicitly
3397 awaken the corresponding "rcuoN" kthreads,
3398 make these kthreads poll for callbacks.
3399 This improves the real-time response for the
3400 offloaded CPUs by relieving them of the need to
3401 wake up the corresponding kthread, but degrades
3402 energy efficiency by requiring that the kthreads
3403 periodically wake up to do the polling.
3404
3405 rcutree.blimit= [KNL]
3406 Set maximum number of finished RCU callbacks to
3407 process in one batch.
3408
3409 rcutree.dump_tree= [KNL]
3410 Dump the structure of the rcu_node combining tree
3411 out at early boot. This is used for diagnostic
3412 purposes, to verify correct tree setup.
3413
3414 rcutree.gp_cleanup_delay= [KNL]
3415 Set the number of jiffies to delay each step of
3416 RCU grace-period cleanup.
3417
3418 rcutree.gp_init_delay= [KNL]
3419 Set the number of jiffies to delay each step of
3420 RCU grace-period initialization.
3421
3422 rcutree.gp_preinit_delay= [KNL]
3423 Set the number of jiffies to delay each step of
3424 RCU grace-period pre-initialization, that is,
3425 the propagation of recent CPU-hotplug changes up
3426 the rcu_node combining tree.
3427
3428 rcutree.rcu_fanout_exact= [KNL]
3429 Disable autobalancing of the rcu_node combining
3430 tree. This is used by rcutorture, and might
3431 possibly be useful for architectures having high
3432 cache-to-cache transfer latencies.
3433
3434 rcutree.rcu_fanout_leaf= [KNL]
3435 Change the number of CPUs assigned to each
3436 leaf rcu_node structure. Useful for very
3437 large systems, which will choose the value 64,
3438 and for NUMA systems with large remote-access
3439 latencies, which will choose a value aligned
3440 with the appropriate hardware boundaries.
3441
3442 rcutree.jiffies_till_sched_qs= [KNL]
3443 Set required age in jiffies for a
3444 given grace period before RCU starts
3445 soliciting quiescent-state help from
3446 rcu_note_context_switch().
3447
3448 rcutree.jiffies_till_first_fqs= [KNL]
3449 Set delay from grace-period initialization to
3450 first attempt to force quiescent states.
3451 Units are jiffies, minimum value is zero,
3452 and maximum value is HZ.
3453
3454 rcutree.jiffies_till_next_fqs= [KNL]
3455 Set delay between subsequent attempts to force
3456 quiescent states. Units are jiffies, minimum
3457 value is one, and maximum value is HZ.
3458
3459 rcutree.kthread_prio= [KNL,BOOT]
3460 Set the SCHED_FIFO priority of the RCU per-CPU
3461 kthreads (rcuc/N). This value is also used for
3462 the priority of the RCU boost threads (rcub/N)
3463 and for the RCU grace-period kthreads (rcu_bh,
3464 rcu_preempt, and rcu_sched). If RCU_BOOST is
3465 set, valid values are 1-99 and the default is 1
3466 (the least-favored priority). Otherwise, when
3467 RCU_BOOST is not set, valid values are 0-99 and
3468 the default is zero (non-realtime operation).
3469
3470 rcutree.rcu_nocb_leader_stride= [KNL]
3471 Set the number of NOCB kthread groups, which
3472 defaults to the square root of the number of
3473 CPUs. Larger numbers reduces the wakeup overhead
3474 on the per-CPU grace-period kthreads, but increases
3475 that same overhead on each group's leader.
3476
3477 rcutree.qhimark= [KNL]
3478 Set threshold of queued RCU callbacks beyond which
3479 batch limiting is disabled.
3480
3481 rcutree.qlowmark= [KNL]
3482 Set threshold of queued RCU callbacks below which
3483 batch limiting is re-enabled.
3484
3485 rcutree.rcu_idle_gp_delay= [KNL]
3486 Set wakeup interval for idle CPUs that have
3487 RCU callbacks (RCU_FAST_NO_HZ=y).
3488
3489 rcutree.rcu_idle_lazy_gp_delay= [KNL]
3490 Set wakeup interval for idle CPUs that have
3491 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
3492 Lazy RCU callbacks are those which RCU can
3493 prove do nothing more than free memory.
3494
3495 rcutree.rcu_kick_kthreads= [KNL]
3496 Cause the grace-period kthread to get an extra
3497 wake_up() if it sleeps three times longer than
3498 it should at force-quiescent-state time.
3499 This wake_up() will be accompanied by a
3500 WARN_ONCE() splat and an ftrace_dump().
3501
3502 rcuperf.gp_async= [KNL]
3503 Measure performance of asynchronous
3504 grace-period primitives such as call_rcu().
3505
3506 rcuperf.gp_async_max= [KNL]
3507 Specify the maximum number of outstanding
3508 callbacks per writer thread. When a writer
3509 thread exceeds this limit, it invokes the
3510 corresponding flavor of rcu_barrier() to allow
3511 previously posted callbacks to drain.
3512
3513 rcuperf.gp_exp= [KNL]
3514 Measure performance of expedited synchronous
3515 grace-period primitives.
3516
3517 rcuperf.holdoff= [KNL]
3518 Set test-start holdoff period. The purpose of
3519 this parameter is to delay the start of the
3520 test until boot completes in order to avoid
3521 interference.
3522
3523 rcuperf.nreaders= [KNL]
3524 Set number of RCU readers. The value -1 selects
3525 N, where N is the number of CPUs. A value
3526 "n" less than -1 selects N-n+1, where N is again
3527 the number of CPUs. For example, -2 selects N
3528 (the number of CPUs), -3 selects N+1, and so on.
3529 A value of "n" less than or equal to -N selects
3530 a single reader.
3531
3532 rcuperf.nwriters= [KNL]
3533 Set number of RCU writers. The values operate
3534 the same as for rcuperf.nreaders.
3535 N, where N is the number of CPUs
3536
3537 rcuperf.perf_type= [KNL]
3538 Specify the RCU implementation to test.
3539
3540 rcuperf.shutdown= [KNL]
3541 Shut the system down after performance tests
3542 complete. This is useful for hands-off automated
3543 testing.
3544
3545 rcuperf.verbose= [KNL]
3546 Enable additional printk() statements.
3547
3548 rcuperf.writer_holdoff= [KNL]
3549 Write-side holdoff between grace periods,
3550 in microseconds. The default of zero says
3551 no holdoff.
3552
3553 rcutorture.cbflood_inter_holdoff= [KNL]
3554 Set holdoff time (jiffies) between successive
3555 callback-flood tests.
3556
3557 rcutorture.cbflood_intra_holdoff= [KNL]
3558 Set holdoff time (jiffies) between successive
3559 bursts of callbacks within a given callback-flood
3560 test.
3561
3562 rcutorture.cbflood_n_burst= [KNL]
3563 Set the number of bursts making up a given
3564 callback-flood test. Set this to zero to
3565 disable callback-flood testing.
3566
3567 rcutorture.cbflood_n_per_burst= [KNL]
3568 Set the number of callbacks to be registered
3569 in a given burst of a callback-flood test.
3570
3571 rcutorture.fqs_duration= [KNL]
3572 Set duration of force_quiescent_state bursts
3573 in microseconds.
3574
3575 rcutorture.fqs_holdoff= [KNL]
3576 Set holdoff time within force_quiescent_state bursts
3577 in microseconds.
3578
3579 rcutorture.fqs_stutter= [KNL]
3580 Set wait time between force_quiescent_state bursts
3581 in seconds.
3582
3583 rcutorture.gp_cond= [KNL]
3584 Use conditional/asynchronous update-side
3585 primitives, if available.
3586
3587 rcutorture.gp_exp= [KNL]
3588 Use expedited update-side primitives, if available.
3589
3590 rcutorture.gp_normal= [KNL]
3591 Use normal (non-expedited) asynchronous
3592 update-side primitives, if available.
3593
3594 rcutorture.gp_sync= [KNL]
3595 Use normal (non-expedited) synchronous
3596 update-side primitives, if available. If all
3597 of rcutorture.gp_cond=, rcutorture.gp_exp=,
3598 rcutorture.gp_normal=, and rcutorture.gp_sync=
3599 are zero, rcutorture acts as if is interpreted
3600 they are all non-zero.
3601
3602 rcutorture.n_barrier_cbs= [KNL]
3603 Set callbacks/threads for rcu_barrier() testing.
3604
3605 rcutorture.nfakewriters= [KNL]
3606 Set number of concurrent RCU writers. These just
3607 stress RCU, they don't participate in the actual
3608 test, hence the "fake".
3609
3610 rcutorture.nreaders= [KNL]
3611 Set number of RCU readers. The value -1 selects
3612 N-1, where N is the number of CPUs. A value
3613 "n" less than -1 selects N-n-2, where N is again
3614 the number of CPUs. For example, -2 selects N
3615 (the number of CPUs), -3 selects N+1, and so on.
3616
3617 rcutorture.object_debug= [KNL]
3618 Enable debug-object double-call_rcu() testing.
3619
3620 rcutorture.onoff_holdoff= [KNL]
3621 Set time (s) after boot for CPU-hotplug testing.
3622
3623 rcutorture.onoff_interval= [KNL]
3624 Set time (s) between CPU-hotplug operations, or
3625 zero to disable CPU-hotplug testing.
3626
3627 rcutorture.shuffle_interval= [KNL]
3628 Set task-shuffle interval (s). Shuffling tasks
3629 allows some CPUs to go into dyntick-idle mode
3630 during the rcutorture test.
3631
3632 rcutorture.shutdown_secs= [KNL]
3633 Set time (s) after boot system shutdown. This
3634 is useful for hands-off automated testing.
3635
3636 rcutorture.stall_cpu= [KNL]
3637 Duration of CPU stall (s) to test RCU CPU stall
3638 warnings, zero to disable.
3639
3640 rcutorture.stall_cpu_holdoff= [KNL]
3641 Time to wait (s) after boot before inducing stall.
3642
3643 rcutorture.stall_cpu_irqsoff= [KNL]
3644 Disable interrupts while stalling if set.
3645
3646 rcutorture.stat_interval= [KNL]
3647 Time (s) between statistics printk()s.
3648
3649 rcutorture.stutter= [KNL]
3650 Time (s) to stutter testing, for example, specifying
3651 five seconds causes the test to run for five seconds,
3652 wait for five seconds, and so on. This tests RCU's
3653 ability to transition abruptly to and from idle.
3654
3655 rcutorture.test_boost= [KNL]
3656 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
3657 "Maybe" means test if the RCU implementation
3658 under test support RCU priority boosting.
3659
3660 rcutorture.test_boost_duration= [KNL]
3661 Duration (s) of each individual boost test.
3662
3663 rcutorture.test_boost_interval= [KNL]
3664 Interval (s) between each boost test.
3665
3666 rcutorture.test_no_idle_hz= [KNL]
3667 Test RCU's dyntick-idle handling. See also the
3668 rcutorture.shuffle_interval parameter.
3669
3670 rcutorture.torture_type= [KNL]
3671 Specify the RCU implementation to test.
3672
3673 rcutorture.verbose= [KNL]
3674 Enable additional printk() statements.
3675
3676 rcupdate.rcu_cpu_stall_suppress= [KNL]
3677 Suppress RCU CPU stall warning messages.
3678
3679 rcupdate.rcu_cpu_stall_timeout= [KNL]
3680 Set timeout for RCU CPU stall warning messages.
3681
3682 rcupdate.rcu_expedited= [KNL]
3683 Use expedited grace-period primitives, for
3684 example, synchronize_rcu_expedited() instead
3685 of synchronize_rcu(). This reduces latency,
3686 but can increase CPU utilization, degrade
3687 real-time latency, and degrade energy efficiency.
3688 No effect on CONFIG_TINY_RCU kernels.
3689
3690 rcupdate.rcu_normal= [KNL]
3691 Use only normal grace-period primitives,
3692 for example, synchronize_rcu() instead of
3693 synchronize_rcu_expedited(). This improves
3694 real-time latency, CPU utilization, and
3695 energy efficiency, but can expose users to
3696 increased grace-period latency. This parameter
3697 overrides rcupdate.rcu_expedited. No effect on
3698 CONFIG_TINY_RCU kernels.
3699
3700 rcupdate.rcu_normal_after_boot= [KNL]
3701 Once boot has completed (that is, after
3702 rcu_end_inkernel_boot() has been invoked), use
3703 only normal grace-period primitives. No effect
3704 on CONFIG_TINY_RCU kernels.
3705
3706 rcupdate.rcu_task_stall_timeout= [KNL]
3707 Set timeout in jiffies for RCU task stall warning
3708 messages. Disable with a value less than or equal
3709 to zero.
3710
3711 rcupdate.rcu_self_test= [KNL]
3712 Run the RCU early boot self tests
3713
3714 rcupdate.rcu_self_test_bh= [KNL]
3715 Run the RCU bh early boot self tests
3716
3717 rcupdate.rcu_self_test_sched= [KNL]
3718 Run the RCU sched early boot self tests
3719
3720 rdinit= [KNL]
3721 Format: <full_path>
3722 Run specified binary instead of /init from the ramdisk,
3723 used for early userspace startup. See initrd.
3724
3725 rdt= [HW,X86,RDT]
3726 Turn on/off individual RDT features. List is:
3727 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
3728 mba.
3729 E.g. to turn on cmt and turn off mba use:
3730 rdt=cmt,!mba
3731
3732 reboot= [KNL]
3733 Format (x86 or x86_64):
3734 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
3735 [[,]s[mp]#### \
3736 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
3737 [[,]f[orce]
3738 Where reboot_mode is one of warm (soft) or cold (hard) or gpio,
3739 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
3740 reboot_force is either force or not specified,
3741 reboot_cpu is s[mp]#### with #### being the processor
3742 to be used for rebooting.
3743
3744 relax_domain_level=
3745 [KNL, SMP] Set scheduler's default relax_domain_level.
3746 See Documentation/cgroup-v1/cpusets.txt.
3747
3748 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
3749 Format: <base1>,<size1>[,<base2>,<size2>,...]
3750 Reserve I/O ports or memory so the kernel won't use
3751 them. If <base> is less than 0x10000, the region
3752 is assumed to be I/O ports; otherwise it is memory.
3753
3754 reservetop= [X86-32]
3755 Format: nn[KMG]
3756 Reserves a hole at the top of the kernel virtual
3757 address space.
3758
3759 reservelow= [X86]
3760 Format: nn[K]
3761 Set the amount of memory to reserve for BIOS at
3762 the bottom of the address space.
3763
3764 reset_devices [KNL] Force drivers to reset the underlying device
3765 during initialization.
3766
3767 resume= [SWSUSP]
3768 Specify the partition device for software suspend
3769 Format:
3770 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
3771
3772 resume_offset= [SWSUSP]
3773 Specify the offset from the beginning of the partition
3774 given by "resume=" at which the swap header is located,
3775 in <PAGE_SIZE> units (needed only for swap files).
3776 See Documentation/power/swsusp-and-swap-files.txt
3777
3778 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
3779 read the resume files
3780
3781 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
3782 Useful for devices that are detected asynchronously
3783 (e.g. USB and MMC devices).
3784
3785 hibernate= [HIBERNATION]
3786 noresume Don't check if there's a hibernation image
3787 present during boot.
3788 nocompress Don't compress/decompress hibernation images.
3789 no Disable hibernation and resume.
3790 protect_image Turn on image protection during restoration
3791 (that will set all pages holding image data
3792 during restoration read-only).
3793
3794 retain_initrd [RAM] Keep initrd memory after extraction
3795
3796 rfkill.default_state=
3797 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
3798 etc. communication is blocked by default.
3799 1 Unblocked.
3800
3801 rfkill.master_switch_mode=
3802 0 The "airplane mode" button does nothing.
3803 1 The "airplane mode" button toggles between everything
3804 blocked and the previous configuration.
3805 2 The "airplane mode" button toggles between everything
3806 blocked and everything unblocked.
3807
3808 rhash_entries= [KNL,NET]
3809 Set number of hash buckets for route cache
3810
3811 ring3mwait=disable
3812 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
3813 CPUs.
3814
3815 ro [KNL] Mount root device read-only on boot
3816
3817 rodata= [KNL]
3818 on Mark read-only kernel memory as read-only (default).
3819 off Leave read-only kernel memory writable for debugging.
3820
3821 rockchip.usb_uart
3822 Enable the uart passthrough on the designated usb port
3823 on Rockchip SoCs. When active, the signals of the
3824 debug-uart get routed to the D+ and D- pins of the usb
3825 port and the regular usb controller gets disabled.
3826
3827 root= [KNL] Root filesystem
3828 See name_to_dev_t comment in init/do_mounts.c.
3829
3830 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
3831 mount the root filesystem
3832
3833 rootflags= [KNL] Set root filesystem mount option string
3834
3835 rootfstype= [KNL] Set root filesystem type
3836
3837 rootwait [KNL] Wait (indefinitely) for root device to show up.
3838 Useful for devices that are detected asynchronously
3839 (e.g. USB and MMC devices).
3840
3841 rproc_mem=nn[KMG][@address]
3842 [KNL,ARM,CMA] Remoteproc physical memory block.
3843 Memory area to be used by remote processor image,
3844 managed by CMA.
3845
3846 rw [KNL] Mount root device read-write on boot
3847
3848 S [KNL] Run init in single mode
3849
3850 s390_iommu= [HW,S390]
3851 Set s390 IOTLB flushing mode
3852 strict
3853 With strict flushing every unmap operation will result in
3854 an IOTLB flush. Default is lazy flushing before reuse,
3855 which is faster.
3856
3857 sa1100ir [NET]
3858 See drivers/net/irda/sa1100_ir.c.
3859
3860 sbni= [NET] Granch SBNI12 leased line adapter
3861
3862 sched_debug [KNL] Enables verbose scheduler debug messages.
3863
3864 schedstats= [KNL,X86] Enable or disable scheduled statistics.
3865 Allowed values are enable and disable. This feature
3866 incurs a small amount of overhead in the scheduler
3867 but is useful for debugging and performance tuning.
3868
3869 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
3870 xtime_lock contention on larger systems, and/or RCU lock
3871 contention on all systems with CONFIG_MAXSMP set.
3872 Format: { "0" | "1" }
3873 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
3874 1 -- enable.
3875 Note: increases power consumption, thus should only be
3876 enabled if running jitter sensitive (HPC/RT) workloads.
3877
3878 security= [SECURITY] Choose a security module to enable at boot.
3879 If this boot parameter is not specified, only the first
3880 security module asking for security registration will be
3881 loaded. An invalid security module name will be treated
3882 as if no module has been chosen.
3883
3884 selinux= [SELINUX] Disable or enable SELinux at boot time.
3885 Format: { "0" | "1" }
3886 See security/selinux/Kconfig help text.
3887 0 -- disable.
3888 1 -- enable.
3889 Default value is set via kernel config option.
3890 If enabled at boot time, /selinux/disable can be used
3891 later to disable prior to initial policy load.
3892
3893 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
3894 Format: { "0" | "1" }
3895 See security/apparmor/Kconfig help text
3896 0 -- disable.
3897 1 -- enable.
3898 Default value is set via kernel config option.
3899
3900 serialnumber [BUGS=X86-32]
3901
3902 shapers= [NET]
3903 Maximal number of shapers.
3904
3905 simeth= [IA-64]
3906 simscsi=
3907
3908 slram= [HW,MTD]
3909
3910 slab_nomerge [MM]
3911 Disable merging of slabs with similar size. May be
3912 necessary if there is some reason to distinguish
3913 allocs to different slabs, especially in hardened
3914 environments where the risk of heap overflows and
3915 layout control by attackers can usually be
3916 frustrated by disabling merging. This will reduce
3917 most of the exposure of a heap attack to a single
3918 cache (risks via metadata attacks are mostly
3919 unchanged). Debug options disable merging on their
3920 own.
3921 For more information see Documentation/vm/slub.txt.
3922
3923 slab_max_order= [MM, SLAB]
3924 Determines the maximum allowed order for slabs.
3925 A high setting may cause OOMs due to memory
3926 fragmentation. Defaults to 1 for systems with
3927 more than 32MB of RAM, 0 otherwise.
3928
3929 slub_debug[=options[,slabs]] [MM, SLUB]
3930 Enabling slub_debug allows one to determine the
3931 culprit if slab objects become corrupted. Enabling
3932 slub_debug can create guard zones around objects and
3933 may poison objects when not in use. Also tracks the
3934 last alloc / free. For more information see
3935 Documentation/vm/slub.txt.
3936
3937 slub_memcg_sysfs= [MM, SLUB]
3938 Determines whether to enable sysfs directories for
3939 memory cgroup sub-caches. 1 to enable, 0 to disable.
3940 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON.
3941 Enabling this can lead to a very high number of debug
3942 directories and files being created under
3943 /sys/kernel/slub.
3944
3945 slub_max_order= [MM, SLUB]
3946 Determines the maximum allowed order for slabs.
3947 A high setting may cause OOMs due to memory
3948 fragmentation. For more information see
3949 Documentation/vm/slub.txt.
3950
3951 slub_min_objects= [MM, SLUB]
3952 The minimum number of objects per slab. SLUB will
3953 increase the slab order up to slub_max_order to
3954 generate a sufficiently large slab able to contain
3955 the number of objects indicated. The higher the number
3956 of objects the smaller the overhead of tracking slabs
3957 and the less frequently locks need to be acquired.
3958 For more information see Documentation/vm/slub.txt.
3959
3960 slub_min_order= [MM, SLUB]
3961 Determines the minimum page order for slabs. Must be
3962 lower than slub_max_order.
3963 For more information see Documentation/vm/slub.txt.
3964
3965 slub_nomerge [MM, SLUB]
3966 Same with slab_nomerge. This is supported for legacy.
3967 See slab_nomerge for more information.
3968
3969 smart2= [HW]
3970 Format: <io1>[,<io2>[,...,<io8>]]
3971
3972 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
3973 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
3974 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
3975 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
3976 smsc-ircc2.ircc_irq= [HW] IRQ line
3977 smsc-ircc2.ircc_dma= [HW] DMA channel
3978 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
3979 0: Toshiba Satellite 1800 (GP data pin select)
3980 1: Fast pin select (default)
3981 2: ATC IRMode
3982
3983 smt [KNL,S390] Set the maximum number of threads (logical
3984 CPUs) to use per physical CPU on systems capable of
3985 symmetric multithreading (SMT). Will be capped to the
3986 actual hardware limit.
3987 Format: <integer>
3988 Default: -1 (no limit)
3989
3990 softlockup_panic=
3991 [KNL] Should the soft-lockup detector generate panics.
3992 Format: <integer>
3993
3994 A nonzero value instructs the soft-lockup detector
3995 to panic the machine when a soft-lockup occurs. This
3996 is also controlled by CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC
3997 which is the respective build-time switch to that
3998 functionality.
3999
4000 softlockup_all_cpu_backtrace=
4001 [KNL] Should the soft-lockup detector generate
4002 backtraces on all cpus.
4003 Format: <integer>
4004
4005 sonypi.*= [HW] Sony Programmable I/O Control Device driver
4006 See Documentation/laptops/sonypi.txt
4007
4008 spectre_v2= [X86] Control mitigation of Spectre variant 2
4009 (indirect branch speculation) vulnerability.
4010
4011 on - unconditionally enable
4012 off - unconditionally disable
4013 auto - kernel detects whether your CPU model is
4014 vulnerable
4015
4016 Selecting 'on' will, and 'auto' may, choose a
4017 mitigation method at run time according to the
4018 CPU, the available microcode, the setting of the
4019 CONFIG_RETPOLINE configuration option, and the
4020 compiler with which the kernel was built.
4021
4022 Specific mitigations can also be selected manually:
4023
4024 retpoline - replace indirect branches
4025 retpoline,generic - google's original retpoline
4026 retpoline,amd - AMD-specific minimal thunk
4027
4028 Not specifying this option is equivalent to
4029 spectre_v2=auto.
4030
4031 spec_store_bypass_disable=
4032 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
4033 (Speculative Store Bypass vulnerability)
4034
4035 Certain CPUs are vulnerable to an exploit against a
4036 a common industry wide performance optimization known
4037 as "Speculative Store Bypass" in which recent stores
4038 to the same memory location may not be observed by
4039 later loads during speculative execution. The idea
4040 is that such stores are unlikely and that they can
4041 be detected prior to instruction retirement at the
4042 end of a particular speculation execution window.
4043
4044 In vulnerable processors, the speculatively forwarded
4045 store can be used in a cache side channel attack, for
4046 example to read memory to which the attacker does not
4047 directly have access (e.g. inside sandboxed code).
4048
4049 This parameter controls whether the Speculative Store
4050 Bypass optimization is used.
4051
4052 on - Unconditionally disable Speculative Store Bypass
4053 off - Unconditionally enable Speculative Store Bypass
4054 auto - Kernel detects whether the CPU model contains an
4055 implementation of Speculative Store Bypass and
4056 picks the most appropriate mitigation. If the
4057 CPU is not vulnerable, "off" is selected. If the
4058 CPU is vulnerable the default mitigation is
4059 architecture and Kconfig dependent. See below.
4060 prctl - Control Speculative Store Bypass per thread
4061 via prctl. Speculative Store Bypass is enabled
4062 for a process by default. The state of the control
4063 is inherited on fork.
4064 seccomp - Same as "prctl" above, but all seccomp threads
4065 will disable SSB unless they explicitly opt out.
4066
4067 Not specifying this option is equivalent to
4068 spec_store_bypass_disable=auto.
4069
4070 Default mitigations:
4071 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
4072
4073 spia_io_base= [HW,MTD]
4074 spia_fio_base=
4075 spia_pedr=
4076 spia_peddr=
4077
4078 srcutree.counter_wrap_check [KNL]
4079 Specifies how frequently to check for
4080 grace-period sequence counter wrap for the
4081 srcu_data structure's ->srcu_gp_seq_needed field.
4082 The greater the number of bits set in this kernel
4083 parameter, the less frequently counter wrap will
4084 be checked for. Note that the bottom two bits
4085 are ignored.
4086
4087 srcutree.exp_holdoff [KNL]
4088 Specifies how many nanoseconds must elapse
4089 since the end of the last SRCU grace period for
4090 a given srcu_struct until the next normal SRCU
4091 grace period will be considered for automatic
4092 expediting. Set to zero to disable automatic
4093 expediting.
4094
4095 stack_guard_gap= [MM]
4096 override the default stack gap protection. The value
4097 is in page units and it defines how many pages prior
4098 to (for stacks growing down) resp. after (for stacks
4099 growing up) the main stack are reserved for no other
4100 mapping. Default value is 256 pages.
4101
4102 stacktrace [FTRACE]
4103 Enabled the stack tracer on boot up.
4104
4105 stacktrace_filter=[function-list]
4106 [FTRACE] Limit the functions that the stack tracer
4107 will trace at boot up. function-list is a comma separated
4108 list of functions. This list can be changed at run
4109 time by the stack_trace_filter file in the debugfs
4110 tracing directory. Note, this enables stack tracing
4111 and the stacktrace above is not needed.
4112
4113 sti= [PARISC,HW]
4114 Format: <num>
4115 Set the STI (builtin display/keyboard on the HP-PARISC
4116 machines) console (graphic card) which should be used
4117 as the initial boot-console.
4118 See also comment in drivers/video/console/sticore.c.
4119
4120 sti_font= [HW]
4121 See comment in drivers/video/console/sticore.c.
4122
4123 stifb= [HW]
4124 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
4125
4126 sunrpc.min_resvport=
4127 sunrpc.max_resvport=
4128 [NFS,SUNRPC]
4129 SunRPC servers often require that client requests
4130 originate from a privileged port (i.e. a port in the
4131 range 0 < portnr < 1024).
4132 An administrator who wishes to reserve some of these
4133 ports for other uses may adjust the range that the
4134 kernel's sunrpc client considers to be privileged
4135 using these two parameters to set the minimum and
4136 maximum port values.
4137
4138 sunrpc.svc_rpc_per_connection_limit=
4139 [NFS,SUNRPC]
4140 Limit the number of requests that the server will
4141 process in parallel from a single connection.
4142 The default value is 0 (no limit).
4143
4144 sunrpc.pool_mode=
4145 [NFS]
4146 Control how the NFS server code allocates CPUs to
4147 service thread pools. Depending on how many NICs
4148 you have and where their interrupts are bound, this
4149 option will affect which CPUs will do NFS serving.
4150 Note: this parameter cannot be changed while the
4151 NFS server is running.
4152
4153 auto the server chooses an appropriate mode
4154 automatically using heuristics
4155 global a single global pool contains all CPUs
4156 percpu one pool for each CPU
4157 pernode one pool for each NUMA node (equivalent
4158 to global on non-NUMA machines)
4159
4160 sunrpc.tcp_slot_table_entries=
4161 sunrpc.udp_slot_table_entries=
4162 [NFS,SUNRPC]
4163 Sets the upper limit on the number of simultaneous
4164 RPC calls that can be sent from the client to a
4165 server. Increasing these values may allow you to
4166 improve throughput, but will also increase the
4167 amount of memory reserved for use by the client.
4168
4169 suspend.pm_test_delay=
4170 [SUSPEND]
4171 Sets the number of seconds to remain in a suspend test
4172 mode before resuming the system (see
4173 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
4174 is set. Default value is 5.
4175
4176 swapaccount=[0|1]
4177 [KNL] Enable accounting of swap in memory resource
4178 controller if no parameter or 1 is given or disable
4179 it if 0 is given (See Documentation/cgroup-v1/memory.txt)
4180
4181 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
4182 Format: { <int> | force | noforce }
4183 <int> -- Number of I/O TLB slabs
4184 force -- force using of bounce buffers even if they
4185 wouldn't be automatically used by the kernel
4186 noforce -- Never use bounce buffers (for debugging)
4187
4188 switches= [HW,M68k]
4189
4190 sysfs.deprecated=0|1 [KNL]
4191 Enable/disable old style sysfs layout for old udev
4192 on older distributions. When this option is enabled
4193 very new udev will not work anymore. When this option
4194 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
4195 in older udev will not work anymore.
4196 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
4197 the kernel configuration.
4198
4199 sysrq_always_enabled
4200 [KNL]
4201 Ignore sysrq setting - this boot parameter will
4202 neutralize any effect of /proc/sys/kernel/sysrq.
4203 Useful for debugging.
4204
4205 tcpmhash_entries= [KNL,NET]
4206 Set the number of tcp_metrics_hash slots.
4207 Default value is 8192 or 16384 depending on total
4208 ram pages. This is used to specify the TCP metrics
4209 cache size. See Documentation/networking/ip-sysctl.txt
4210 "tcp_no_metrics_save" section for more details.
4211
4212 tdfx= [HW,DRM]
4213
4214 test_suspend= [SUSPEND][,N]
4215 Specify "mem" (for Suspend-to-RAM) or "standby" (for
4216 standby suspend) or "freeze" (for suspend type freeze)
4217 as the system sleep state during system startup with
4218 the optional capability to repeat N number of times.
4219 The system is woken from this state using a
4220 wakeup-capable RTC alarm.
4221
4222 thash_entries= [KNL,NET]
4223 Set number of hash buckets for TCP connection
4224
4225 thermal.act= [HW,ACPI]
4226 -1: disable all active trip points in all thermal zones
4227 <degrees C>: override all lowest active trip points
4228
4229 thermal.crt= [HW,ACPI]
4230 -1: disable all critical trip points in all thermal zones
4231 <degrees C>: override all critical trip points
4232
4233 thermal.nocrt= [HW,ACPI]
4234 Set to disable actions on ACPI thermal zone
4235 critical and hot trip points.
4236
4237 thermal.off= [HW,ACPI]
4238 1: disable ACPI thermal control
4239
4240 thermal.psv= [HW,ACPI]
4241 -1: disable all passive trip points
4242 <degrees C>: override all passive trip points to this
4243 value
4244
4245 thermal.tzp= [HW,ACPI]
4246 Specify global default ACPI thermal zone polling rate
4247 <deci-seconds>: poll all this frequency
4248 0: no polling (default)
4249
4250 threadirqs [KNL]
4251 Force threading of all interrupt handlers except those
4252 marked explicitly IRQF_NO_THREAD.
4253
4254 tmem [KNL,XEN]
4255 Enable the Transcendent memory driver if built-in.
4256
4257 tmem.cleancache=0|1 [KNL, XEN]
4258 Default is on (1). Disable the usage of the cleancache
4259 API to send anonymous pages to the hypervisor.
4260
4261 tmem.frontswap=0|1 [KNL, XEN]
4262 Default is on (1). Disable the usage of the frontswap
4263 API to send swap pages to the hypervisor. If disabled
4264 the selfballooning and selfshrinking are force disabled.
4265
4266 tmem.selfballooning=0|1 [KNL, XEN]
4267 Default is on (1). Disable the driving of swap pages
4268 to the hypervisor.
4269
4270 tmem.selfshrinking=0|1 [KNL, XEN]
4271 Default is on (1). Partial swapoff that immediately
4272 transfers pages from Xen hypervisor back to the
4273 kernel based on different criteria.
4274
4275 topology= [S390]
4276 Format: {off | on}
4277 Specify if the kernel should make use of the cpu
4278 topology information if the hardware supports this.
4279 The scheduler will make use of this information and
4280 e.g. base its process migration decisions on it.
4281 Default is on.
4282
4283 topology_updates= [KNL, PPC, NUMA]
4284 Format: {off}
4285 Specify if the kernel should ignore (off)
4286 topology updates sent by the hypervisor to this
4287 LPAR.
4288
4289 tp720= [HW,PS2]
4290
4291 tpm_suspend_pcr=[HW,TPM]
4292 Format: integer pcr id
4293 Specify that at suspend time, the tpm driver
4294 should extend the specified pcr with zeros,
4295 as a workaround for some chips which fail to
4296 flush the last written pcr on TPM_SaveState.
4297 This will guarantee that all the other pcrs
4298 are saved.
4299
4300 trace_buf_size=nn[KMG]
4301 [FTRACE] will set tracing buffer size on each cpu.
4302
4303 trace_event=[event-list]
4304 [FTRACE] Set and start specified trace events in order
4305 to facilitate early boot debugging. The event-list is a
4306 comma separated list of trace events to enable. See
4307 also Documentation/trace/events.txt
4308
4309 trace_options=[option-list]
4310 [FTRACE] Enable or disable tracer options at boot.
4311 The option-list is a comma delimited list of options
4312 that can be enabled or disabled just as if you were
4313 to echo the option name into
4314
4315 /sys/kernel/debug/tracing/trace_options
4316
4317 For example, to enable stacktrace option (to dump the
4318 stack trace of each event), add to the command line:
4319
4320 trace_options=stacktrace
4321
4322 See also Documentation/trace/ftrace.txt "trace options"
4323 section.
4324
4325 tp_printk[FTRACE]
4326 Have the tracepoints sent to printk as well as the
4327 tracing ring buffer. This is useful for early boot up
4328 where the system hangs or reboots and does not give the
4329 option for reading the tracing buffer or performing a
4330 ftrace_dump_on_oops.
4331
4332 To turn off having tracepoints sent to printk,
4333 echo 0 > /proc/sys/kernel/tracepoint_printk
4334 Note, echoing 1 into this file without the
4335 tracepoint_printk kernel cmdline option has no effect.
4336
4337 ** CAUTION **
4338
4339 Having tracepoints sent to printk() and activating high
4340 frequency tracepoints such as irq or sched, can cause
4341 the system to live lock.
4342
4343 traceoff_on_warning
4344 [FTRACE] enable this option to disable tracing when a
4345 warning is hit. This turns off "tracing_on". Tracing can
4346 be enabled again by echoing '1' into the "tracing_on"
4347 file located in /sys/kernel/debug/tracing/
4348
4349 This option is useful, as it disables the trace before
4350 the WARNING dump is called, which prevents the trace to
4351 be filled with content caused by the warning output.
4352
4353 This option can also be set at run time via the sysctl
4354 option: kernel/traceoff_on_warning
4355
4356 transparent_hugepage=
4357 [KNL]
4358 Format: [always|madvise|never]
4359 Can be used to control the default behavior of the system
4360 with respect to transparent hugepages.
4361 See Documentation/vm/transhuge.txt for more details.
4362
4363 tsc= Disable clocksource stability checks for TSC.
4364 Format: <string>
4365 [x86] reliable: mark tsc clocksource as reliable, this
4366 disables clocksource verification at runtime, as well
4367 as the stability checks done at bootup. Used to enable
4368 high-resolution timer mode on older hardware, and in
4369 virtualized environment.
4370 [x86] noirqtime: Do not use TSC to do irq accounting.
4371 Used to run time disable IRQ_TIME_ACCOUNTING on any
4372 platforms where RDTSC is slow and this accounting
4373 can add overhead.
4374 [x86] unstable: mark the TSC clocksource as unstable, this
4375 marks the TSC unconditionally unstable at bootup and
4376 avoids any further wobbles once the TSC watchdog notices.
4377
4378 turbografx.map[2|3]= [HW,JOY]
4379 TurboGraFX parallel port interface
4380 Format:
4381 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
4382 See also Documentation/input/devices/joystick-parport.rst
4383
4384 udbg-immortal [PPC] When debugging early kernel crashes that
4385 happen after console_init() and before a proper
4386 console driver takes over, this boot options might
4387 help "seeing" what's going on.
4388
4389 uhash_entries= [KNL,NET]
4390 Set number of hash buckets for UDP/UDP-Lite connections
4391
4392 uhci-hcd.ignore_oc=
4393 [USB] Ignore overcurrent events (default N).
4394 Some badly-designed motherboards generate lots of
4395 bogus events, for ports that aren't wired to
4396 anything. Set this parameter to avoid log spamming.
4397 Note that genuine overcurrent events won't be
4398 reported either.
4399
4400 unknown_nmi_panic
4401 [X86] Cause panic on unknown NMI.
4402
4403 usbcore.authorized_default=
4404 [USB] Default USB device authorization:
4405 (default -1 = authorized except for wireless USB,
4406 0 = not authorized, 1 = authorized)
4407
4408 usbcore.autosuspend=
4409 [USB] The autosuspend time delay (in seconds) used
4410 for newly-detected USB devices (default 2). This
4411 is the time required before an idle device will be
4412 autosuspended. Devices for which the delay is set
4413 to a negative value won't be autosuspended at all.
4414
4415 usbcore.usbfs_snoop=
4416 [USB] Set to log all usbfs traffic (default 0 = off).
4417
4418 usbcore.usbfs_snoop_max=
4419 [USB] Maximum number of bytes to snoop in each URB
4420 (default = 65536).
4421
4422 usbcore.blinkenlights=
4423 [USB] Set to cycle leds on hubs (default 0 = off).
4424
4425 usbcore.old_scheme_first=
4426 [USB] Start with the old device initialization
4427 scheme (default 0 = off).
4428
4429 usbcore.usbfs_memory_mb=
4430 [USB] Memory limit (in MB) for buffers allocated by
4431 usbfs (default = 16, 0 = max = 2047).
4432
4433 usbcore.use_both_schemes=
4434 [USB] Try the other device initialization scheme
4435 if the first one fails (default 1 = enabled).
4436
4437 usbcore.initial_descriptor_timeout=
4438 [USB] Specifies timeout for the initial 64-byte
4439 USB_REQ_GET_DESCRIPTOR request in milliseconds
4440 (default 5000 = 5.0 seconds).
4441
4442 usbcore.nousb [USB] Disable the USB subsystem
4443
4444 usbcore.quirks=
4445 [USB] A list of quirk entries to augment the built-in
4446 usb core quirk list. List entries are separated by
4447 commas. Each entry has the form
4448 VendorID:ProductID:Flags. The IDs are 4-digit hex
4449 numbers and Flags is a set of letters. Each letter
4450 will change the built-in quirk; setting it if it is
4451 clear and clearing it if it is set. The letters have
4452 the following meanings:
4453 a = USB_QUIRK_STRING_FETCH_255 (string
4454 descriptors must not be fetched using
4455 a 255-byte read);
4456 b = USB_QUIRK_RESET_RESUME (device can't resume
4457 correctly so reset it instead);
4458 c = USB_QUIRK_NO_SET_INTF (device can't handle
4459 Set-Interface requests);
4460 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
4461 handle its Configuration or Interface
4462 strings);
4463 e = USB_QUIRK_RESET (device can't be reset
4464 (e.g morph devices), don't use reset);
4465 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
4466 more interface descriptions than the
4467 bNumInterfaces count, and can't handle
4468 talking to these interfaces);
4469 g = USB_QUIRK_DELAY_INIT (device needs a pause
4470 during initialization, after we read
4471 the device descriptor);
4472 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
4473 high speed and super speed interrupt
4474 endpoints, the USB 2.0 and USB 3.0 spec
4475 require the interval in microframes (1
4476 microframe = 125 microseconds) to be
4477 calculated as interval = 2 ^
4478 (bInterval-1).
4479 Devices with this quirk report their
4480 bInterval as the result of this
4481 calculation instead of the exponent
4482 variable used in the calculation);
4483 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
4484 handle device_qualifier descriptor
4485 requests);
4486 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
4487 generates spurious wakeup, ignore
4488 remote wakeup capability);
4489 k = USB_QUIRK_NO_LPM (device can't handle Link
4490 Power Management);
4491 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
4492 (Device reports its bInterval as linear
4493 frames instead of the USB 2.0
4494 calculation);
4495 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
4496 to be disconnected before suspend to
4497 prevent spurious wakeup);
4498 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
4499 pause after every control message);
4500 Example: quirks=0781:5580:bk,0a5c:5834:gij
4501
4502 usbhid.mousepoll=
4503 [USBHID] The interval which mice are to be polled at.
4504
4505 usbhid.jspoll=
4506 [USBHID] The interval which joysticks are to be polled at.
4507
4508 usbhid.kbpoll=
4509 [USBHID] The interval which keyboards are to be polled at.
4510
4511 usb-storage.delay_use=
4512 [UMS] The delay in seconds before a new device is
4513 scanned for Logical Units (default 1).
4514
4515 usb-storage.quirks=
4516 [UMS] A list of quirks entries to supplement or
4517 override the built-in unusual_devs list. List
4518 entries are separated by commas. Each entry has
4519 the form VID:PID:Flags where VID and PID are Vendor
4520 and Product ID values (4-digit hex numbers) and
4521 Flags is a set of characters, each corresponding
4522 to a common usb-storage quirk flag as follows:
4523 a = SANE_SENSE (collect more than 18 bytes
4524 of sense data);
4525 b = BAD_SENSE (don't collect more than 18
4526 bytes of sense data);
4527 c = FIX_CAPACITY (decrease the reported
4528 device capacity by one sector);
4529 d = NO_READ_DISC_INFO (don't use
4530 READ_DISC_INFO command);
4531 e = NO_READ_CAPACITY_16 (don't use
4532 READ_CAPACITY_16 command);
4533 f = NO_REPORT_OPCODES (don't use report opcodes
4534 command, uas only);
4535 g = MAX_SECTORS_240 (don't transfer more than
4536 240 sectors at a time, uas only);
4537 h = CAPACITY_HEURISTICS (decrease the
4538 reported device capacity by one
4539 sector if the number is odd);
4540 i = IGNORE_DEVICE (don't bind to this
4541 device);
4542 j = NO_REPORT_LUNS (don't use report luns
4543 command, uas only);
4544 l = NOT_LOCKABLE (don't try to lock and
4545 unlock ejectable media);
4546 m = MAX_SECTORS_64 (don't transfer more
4547 than 64 sectors = 32 KB at a time);
4548 n = INITIAL_READ10 (force a retry of the
4549 initial READ(10) command);
4550 o = CAPACITY_OK (accept the capacity
4551 reported by the device);
4552 p = WRITE_CACHE (the device cache is ON
4553 by default);
4554 r = IGNORE_RESIDUE (the device reports
4555 bogus residue values);
4556 s = SINGLE_LUN (the device has only one
4557 Logical Unit);
4558 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
4559 commands, uas only);
4560 u = IGNORE_UAS (don't bind to the uas driver);
4561 w = NO_WP_DETECT (don't test whether the
4562 medium is write-protected).
4563 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
4564 even if the device claims no cache)
4565 Example: quirks=0419:aaf5:rl,0421:0433:rc
4566
4567 user_debug= [KNL,ARM]
4568 Format: <int>
4569 See arch/arm/Kconfig.debug help text.
4570 1 - undefined instruction events
4571 2 - system calls
4572 4 - invalid data aborts
4573 8 - SIGSEGV faults
4574 16 - SIGBUS faults
4575 Example: user_debug=31
4576
4577 userpte=
4578 [X86] Flags controlling user PTE allocations.
4579
4580 nohigh = do not allocate PTE pages in
4581 HIGHMEM regardless of setting
4582 of CONFIG_HIGHPTE.
4583
4584 vdso= [X86,SH]
4585 On X86_32, this is an alias for vdso32=. Otherwise:
4586
4587 vdso=1: enable VDSO (the default)
4588 vdso=0: disable VDSO mapping
4589
4590 vdso32= [X86] Control the 32-bit vDSO
4591 vdso32=1: enable 32-bit VDSO
4592 vdso32=0 or vdso32=2: disable 32-bit VDSO
4593
4594 See the help text for CONFIG_COMPAT_VDSO for more
4595 details. If CONFIG_COMPAT_VDSO is set, the default is
4596 vdso32=0; otherwise, the default is vdso32=1.
4597
4598 For compatibility with older kernels, vdso32=2 is an
4599 alias for vdso32=0.
4600
4601 Try vdso32=0 if you encounter an error that says:
4602 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
4603
4604 vector= [IA-64,SMP]
4605 vector=percpu: enable percpu vector domain
4606
4607 video= [FB] Frame buffer configuration
4608 See Documentation/fb/modedb.txt.
4609
4610 video.brightness_switch_enabled= [0,1]
4611 If set to 1, on receiving an ACPI notify event
4612 generated by hotkey, video driver will adjust brightness
4613 level and then send out the event to user space through
4614 the allocated input device; If set to 0, video driver
4615 will only send out the event without touching backlight
4616 brightness level.
4617 default: 1
4618
4619 virtio_mmio.device=
4620 [VMMIO] Memory mapped virtio (platform) device.
4621
4622 <size>@<baseaddr>:<irq>[:<id>]
4623 where:
4624 <size> := size (can use standard suffixes
4625 like K, M and G)
4626 <baseaddr> := physical base address
4627 <irq> := interrupt number (as passed to
4628 request_irq())
4629 <id> := (optional) platform device id
4630 example:
4631 virtio_mmio.device=1K@0x100b0000:48:7
4632
4633 Can be used multiple times for multiple devices.
4634
4635 vga= [BOOT,X86-32] Select a particular video mode
4636 See Documentation/x86/boot.txt and
4637 Documentation/svga.txt.
4638 Use vga=ask for menu.
4639 This is actually a boot loader parameter; the value is
4640 passed to the kernel using a special protocol.
4641
4642 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
4643 size of <nn>. This can be used to increase the
4644 minimum size (128MB on x86). It can also be used to
4645 decrease the size and leave more room for directly
4646 mapped kernel RAM.
4647
4648 vmcp_cma=nn[MG] [KNL,S390]
4649 Sets the memory size reserved for contiguous memory
4650 allocations for the vmcp device driver.
4651
4652 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
4653 Format: <command>
4654
4655 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
4656 Format: <command>
4657
4658 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
4659 Format: <command>
4660
4661 vsyscall= [X86-64]
4662 Controls the behavior of vsyscalls (i.e. calls to
4663 fixed addresses of 0xffffffffff600x00 from legacy
4664 code). Most statically-linked binaries and older
4665 versions of glibc use these calls. Because these
4666 functions are at fixed addresses, they make nice
4667 targets for exploits that can control RIP.
4668
4669 emulate [default] Vsyscalls turn into traps and are
4670 emulated reasonably safely.
4671
4672 native Vsyscalls are native syscall instructions.
4673 This is a little bit faster than trapping
4674 and makes a few dynamic recompilers work
4675 better than they would in emulation mode.
4676 It also makes exploits much easier to write.
4677
4678 none Vsyscalls don't work at all. This makes
4679 them quite hard to use for exploits but
4680 might break your system.
4681
4682 vt.color= [VT] Default text color.
4683 Format: 0xYX, X = foreground, Y = background.
4684 Default: 0x07 = light gray on black.
4685
4686 vt.cur_default= [VT] Default cursor shape.
4687 Format: 0xCCBBAA, where AA, BB, and CC are the same as
4688 the parameters of the <Esc>[?A;B;Cc escape sequence;
4689 see VGA-softcursor.txt. Default: 2 = underline.
4690
4691 vt.default_blu= [VT]
4692 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
4693 Change the default blue palette of the console.
4694 This is a 16-member array composed of values
4695 ranging from 0-255.
4696
4697 vt.default_grn= [VT]
4698 Format: <green0>,<green1>,<green2>,...,<green15>
4699 Change the default green palette of the console.
4700 This is a 16-member array composed of values
4701 ranging from 0-255.
4702
4703 vt.default_red= [VT]
4704 Format: <red0>,<red1>,<red2>,...,<red15>
4705 Change the default red palette of the console.
4706 This is a 16-member array composed of values
4707 ranging from 0-255.
4708
4709 vt.default_utf8=
4710 [VT]
4711 Format=<0|1>
4712 Set system-wide default UTF-8 mode for all tty's.
4713 Default is 1, i.e. UTF-8 mode is enabled for all
4714 newly opened terminals.
4715
4716 vt.global_cursor_default=
4717 [VT]
4718 Format=<-1|0|1>
4719 Set system-wide default for whether a cursor
4720 is shown on new VTs. Default is -1,
4721 i.e. cursors will be created by default unless
4722 overridden by individual drivers. 0 will hide
4723 cursors, 1 will display them.
4724
4725 vt.italic= [VT] Default color for italic text; 0-15.
4726 Default: 2 = green.
4727
4728 vt.underline= [VT] Default color for underlined text; 0-15.
4729 Default: 3 = cyan.
4730
4731 watchdog timers [HW,WDT] For information on watchdog timers,
4732 see Documentation/watchdog/watchdog-parameters.txt
4733 or other driver-specific files in the
4734 Documentation/watchdog/ directory.
4735
4736 workqueue.watchdog_thresh=
4737 If CONFIG_WQ_WATCHDOG is configured, workqueue can
4738 warn stall conditions and dump internal state to
4739 help debugging. 0 disables workqueue stall
4740 detection; otherwise, it's the stall threshold
4741 duration in seconds. The default value is 30 and
4742 it can be updated at runtime by writing to the
4743 corresponding sysfs file.
4744
4745 workqueue.disable_numa
4746 By default, all work items queued to unbound
4747 workqueues are affine to the NUMA nodes they're
4748 issued on, which results in better behavior in
4749 general. If NUMA affinity needs to be disabled for
4750 whatever reason, this option can be used. Note
4751 that this also can be controlled per-workqueue for
4752 workqueues visible under /sys/bus/workqueue/.
4753
4754 workqueue.power_efficient
4755 Per-cpu workqueues are generally preferred because
4756 they show better performance thanks to cache
4757 locality; unfortunately, per-cpu workqueues tend to
4758 be more power hungry than unbound workqueues.
4759
4760 Enabling this makes the per-cpu workqueues which
4761 were observed to contribute significantly to power
4762 consumption unbound, leading to measurably lower
4763 power usage at the cost of small performance
4764 overhead.
4765
4766 The default value of this parameter is determined by
4767 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
4768
4769 workqueue.debug_force_rr_cpu
4770 Workqueue used to implicitly guarantee that work
4771 items queued without explicit CPU specified are put
4772 on the local CPU. This guarantee is no longer true
4773 and while local CPU is still preferred work items
4774 may be put on foreign CPUs. This debug option
4775 forces round-robin CPU selection to flush out
4776 usages which depend on the now broken guarantee.
4777 When enabled, memory and cache locality will be
4778 impacted.
4779
4780 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
4781 default x2apic cluster mode on platforms
4782 supporting x2apic.
4783
4784 x86_intel_mid_timer= [X86-32,APBT]
4785 Choose timer option for x86 Intel MID platform.
4786 Two valid options are apbt timer only and lapic timer
4787 plus one apbt timer for broadcast timer.
4788 x86_intel_mid_timer=apbt_only | lapic_and_apbt
4789
4790 xen_512gb_limit [KNL,X86-64,XEN]
4791 Restricts the kernel running paravirtualized under Xen
4792 to use only up to 512 GB of RAM. The reason to do so is
4793 crash analysis tools and Xen tools for doing domain
4794 save/restore/migration must be enabled to handle larger
4795 domains.
4796
4797 xen_emul_unplug= [HW,X86,XEN]
4798 Unplug Xen emulated devices
4799 Format: [unplug0,][unplug1]
4800 ide-disks -- unplug primary master IDE devices
4801 aux-ide-disks -- unplug non-primary-master IDE devices
4802 nics -- unplug network devices
4803 all -- unplug all emulated devices (NICs and IDE disks)
4804 unnecessary -- unplugging emulated devices is
4805 unnecessary even if the host did not respond to
4806 the unplug protocol
4807 never -- do not unplug even if version check succeeds
4808
4809 xen_nopvspin [X86,XEN]
4810 Disables the ticketlock slowpath using Xen PV
4811 optimizations.
4812
4813 xen_nopv [X86]
4814 Disables the PV optimizations forcing the HVM guest to
4815 run as generic HVM guest with no PV drivers.
4816
4817 xirc2ps_cs= [NET,PCMCIA]
4818 Format:
4819 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
1 acpi= [HW,ACPI,X86,ARM64]
2 Advanced Configuration and Power Interface
3 Format: { force | on | off | strict | noirq | rsdt |
4 copy_dsdt }
5 force -- enable ACPI if default was off
6 on -- enable ACPI but allow fallback to DT [arm64]
7 off -- disable ACPI if default was on
8 noirq -- do not use ACPI for IRQ routing
9 strict -- Be less tolerant of platforms that are not
10 strictly ACPI specification compliant.
11 rsdt -- prefer RSDT over (default) XSDT
12 copy_dsdt -- copy DSDT to memory
13 For ARM64, ONLY "acpi=off", "acpi=on" or "acpi=force"
14 are available
15
16 See also Documentation/power/runtime_pm.rst, pci=noacpi
17
18 acpi_apic_instance= [ACPI, IOAPIC]
19 Format: <int>
20 2: use 2nd APIC table, if available
21 1,0: use 1st APIC table
22 default: 0
23
24 acpi_backlight= [HW,ACPI]
25 { vendor | video | native | none }
26 If set to vendor, prefer vendor-specific driver
27 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
28 of the ACPI video.ko driver.
29 If set to video, use the ACPI video.ko driver.
30 If set to native, use the device's native backlight mode.
31 If set to none, disable the ACPI backlight interface.
32
33 acpi_force_32bit_fadt_addr
34 force FADT to use 32 bit addresses rather than the
35 64 bit X_* addresses. Some firmware have broken 64
36 bit addresses for force ACPI ignore these and use
37 the older legacy 32 bit addresses.
38
39 acpica_no_return_repair [HW, ACPI]
40 Disable AML predefined validation mechanism
41 This mechanism can repair the evaluation result to make
42 the return objects more ACPI specification compliant.
43 This option is useful for developers to identify the
44 root cause of an AML interpreter issue when the issue
45 has something to do with the repair mechanism.
46
47 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
48 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
49 Format: <int>
50 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
51 debug output. Bits in debug_layer correspond to a
52 _COMPONENT in an ACPI source file, e.g.,
53 #define _COMPONENT ACPI_EVENTS
54 Bits in debug_level correspond to a level in
55 ACPI_DEBUG_PRINT statements, e.g.,
56 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
57 The debug_level mask defaults to "info". See
58 Documentation/firmware-guide/acpi/debug.rst for more information about
59 debug layers and levels.
60
61 Enable processor driver info messages:
62 acpi.debug_layer=0x20000000
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
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.
72
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
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
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
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: <byte> or <bitmap-list>
117
118 acpi_no_auto_serialize [HW,ACPI]
119 Disable auto-serialization of AML methods
120 AML control methods that contain the opcodes to create
121 named objects will be marked as "Serialized" by the
122 auto-serialization feature.
123 This feature is enabled by default.
124 This option allows to turn off the feature.
125
126 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
127 kernels.
128
129 acpi_no_static_ssdt [HW,ACPI]
130 Disable installation of static SSDTs at early boot time
131 By default, SSDTs contained in the RSDT/XSDT will be
132 installed automatically and they will appear under
133 /sys/firmware/acpi/tables.
134 This option turns off this feature.
135 Note that specifying this option does not affect
136 dynamic table installation which will install SSDT
137 tables to /sys/firmware/acpi/tables/dynamic.
138
139 acpi_no_watchdog [HW,ACPI,WDT]
140 Ignore the ACPI-based watchdog interface (WDAT) and let
141 a native driver control the watchdog device instead.
142
143 acpi_rsdp= [ACPI,EFI,KEXEC]
144 Pass the RSDP address to the kernel, mostly used
145 on machines running EFI runtime service to boot the
146 second kernel for kdump.
147
148 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
149 Format: To spoof as Windows 98: ="Microsoft Windows"
150
151 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
152 of 2 which is mandated by ACPI 6) as the supported ACPI
153 specification revision (when using this switch, it may
154 be necessary to carry out a cold reboot _twice_ in a
155 row to make it take effect on the platform firmware).
156
157 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
158 acpi_osi="string1" # add string1
159 acpi_osi="!string2" # remove string2
160 acpi_osi=!* # remove all strings
161 acpi_osi=! # disable all built-in OS vendor
162 strings
163 acpi_osi=!! # enable all built-in OS vendor
164 strings
165 acpi_osi= # disable all strings
166
167 'acpi_osi=!' can be used in combination with single or
168 multiple 'acpi_osi="string1"' to support specific OS
169 vendor string(s). Note that such command can only
170 affect the default state of the OS vendor strings, thus
171 it cannot affect the default state of the feature group
172 strings and the current state of the OS vendor strings,
173 specifying it multiple times through kernel command line
174 is meaningless. This command is useful when one do not
175 care about the state of the feature group strings which
176 should be controlled by the OSPM.
177 Examples:
178 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
179 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
180 can make '_OSI("Windows 2000")' TRUE.
181
182 'acpi_osi=' cannot be used in combination with other
183 'acpi_osi=' command lines, the _OSI method will not
184 exist in the ACPI namespace. NOTE that such command can
185 only affect the _OSI support state, thus specifying it
186 multiple times through kernel command line is also
187 meaningless.
188 Examples:
189 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
190 FALSE.
191
192 'acpi_osi=!*' can be used in combination with single or
193 multiple 'acpi_osi="string1"' to support specific
194 string(s). Note that such command can affect the
195 current state of both the OS vendor strings and the
196 feature group strings, thus specifying it multiple times
197 through kernel command line is meaningful. But it may
198 still not able to affect the final state of a string if
199 there are quirks related to this string. This command
200 is useful when one want to control the state of the
201 feature group strings to debug BIOS issues related to
202 the OSPM features.
203 Examples:
204 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
205 '_OSI("Module Device")' FALSE.
206 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
207 '_OSI("Module Device")' TRUE.
208 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
209 equivalent to
210 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
211 and
212 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
213 they all will make '_OSI("Windows 2000")' TRUE.
214
215 acpi_pm_good [X86]
216 Override the pmtimer bug detection: force the kernel
217 to assume that this machine's pmtimer latches its value
218 and always returns good values.
219
220 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
221 Format: { level | edge | high | low }
222
223 acpi_skip_timer_override [HW,ACPI]
224 Recognize and ignore IRQ0/pin2 Interrupt Override.
225 For broken nForce2 BIOS resulting in XT-PIC timer.
226
227 acpi_sleep= [HW,ACPI] Sleep options
228 Format: { s3_bios, s3_mode, s3_beep, s4_hwsig,
229 s4_nohwsig, old_ordering, nonvs,
230 sci_force_enable, nobl }
231 See Documentation/power/video.rst for information on
232 s3_bios and s3_mode.
233 s3_beep is for debugging; it makes the PC's speaker beep
234 as soon as the kernel's real-mode entry point is called.
235 s4_hwsig causes the kernel to check the ACPI hardware
236 signature during resume from hibernation, and gracefully
237 refuse to resume if it has changed. This complies with
238 the ACPI specification but not with reality, since
239 Windows does not do this and many laptops do change it
240 on docking. So the default behaviour is to allow resume
241 and simply warn when the signature changes, unless the
242 s4_hwsig option is enabled.
243 s4_nohwsig prevents ACPI hardware signature from being
244 used (or even warned about) during resume.
245 old_ordering causes the ACPI 1.0 ordering of the _PTS
246 control method, with respect to putting devices into
247 low power states, to be enforced (the ACPI 2.0 ordering
248 of _PTS is used by default).
249 nonvs prevents the kernel from saving/restoring the
250 ACPI NVS memory during suspend/hibernation and resume.
251 sci_force_enable causes the kernel to set SCI_EN directly
252 on resume from S1/S3 (which is against the ACPI spec,
253 but some broken systems don't work without it).
254 nobl causes the internal blacklist of systems known to
255 behave incorrectly in some ways with respect to system
256 suspend and resume to be ignored (use wisely).
257
258 acpi_use_timer_override [HW,ACPI]
259 Use timer override. For some broken Nvidia NF5 boards
260 that require a timer override, but don't have HPET
261
262 add_efi_memmap [EFI; X86] Include EFI memory map in
263 kernel's map of available physical RAM.
264
265 agp= [AGP]
266 { off | try_unsupported }
267 off: disable AGP support
268 try_unsupported: try to drive unsupported chipsets
269 (may crash computer or cause data corruption)
270
271 ALSA [HW,ALSA]
272 See Documentation/sound/alsa-configuration.rst
273
274 alignment= [KNL,ARM]
275 Allow the default userspace alignment fault handler
276 behaviour to be specified. Bit 0 enables warnings,
277 bit 1 enables fixups, and bit 2 sends a segfault.
278
279 align_va_addr= [X86-64]
280 Align virtual addresses by clearing slice [14:12] when
281 allocating a VMA at process creation time. This option
282 gives you up to 3% performance improvement on AMD F15h
283 machines (where it is enabled by default) for a
284 CPU-intensive style benchmark, and it can vary highly in
285 a microbenchmark depending on workload and compiler.
286
287 32: only for 32-bit processes
288 64: only for 64-bit processes
289 on: enable for both 32- and 64-bit processes
290 off: disable for both 32- and 64-bit processes
291
292 alloc_snapshot [FTRACE]
293 Allocate the ftrace snapshot buffer on boot up when the
294 main buffer is allocated. This is handy if debugging
295 and you need to use tracing_snapshot() on boot up, and
296 do not want to use tracing_snapshot_alloc() as it needs
297 to be done where GFP_KERNEL allocations are allowed.
298
299 allow_mismatched_32bit_el0 [ARM64]
300 Allow execve() of 32-bit applications and setting of the
301 PER_LINUX32 personality on systems where only a strict
302 subset of the CPUs support 32-bit EL0. When this
303 parameter is present, the set of CPUs supporting 32-bit
304 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0
305 and hot-unplug operations may be restricted.
306
307 See Documentation/arm64/asymmetric-32bit.rst for more
308 information.
309
310 amd_iommu= [HW,X86-64]
311 Pass parameters to the AMD IOMMU driver in the system.
312 Possible values are:
313 fullflush - Deprecated, equivalent to iommu.strict=1
314 off - do not initialize any AMD IOMMU found in
315 the system
316 force_isolation - Force device isolation for all
317 devices. The IOMMU driver is not
318 allowed anymore to lift isolation
319 requirements as needed. This option
320 does not override iommu=pt
321 force_enable - Force enable the IOMMU on platforms known
322 to be buggy with IOMMU enabled. Use this
323 option with care.
324 pgtbl_v1 - Use v1 page table for DMA-API (Default).
325 pgtbl_v2 - Use v2 page table for DMA-API.
326
327 amd_iommu_dump= [HW,X86-64]
328 Enable AMD IOMMU driver option to dump the ACPI table
329 for AMD IOMMU. With this option enabled, AMD IOMMU
330 driver will print ACPI tables for AMD IOMMU during
331 IOMMU initialization.
332
333 amd_iommu_intr= [HW,X86-64]
334 Specifies one of the following AMD IOMMU interrupt
335 remapping modes:
336 legacy - Use legacy interrupt remapping mode.
337 vapic - Use virtual APIC mode, which allows IOMMU
338 to inject interrupts directly into guest.
339 This mode requires kvm-amd.avic=1.
340 (Default when IOMMU HW support is present.)
341
342 amijoy.map= [HW,JOY] Amiga joystick support
343 Map of devices attached to JOY0DAT and JOY1DAT
344 Format: <a>,<b>
345 See also Documentation/input/joydev/joystick.rst
346
347 analog.map= [HW,JOY] Analog joystick and gamepad support
348 Specifies type or capabilities of an analog joystick
349 connected to one of 16 gameports
350 Format: <type1>,<type2>,..<type16>
351
352 apc= [HW,SPARC]
353 Power management functions (SPARCstation-4/5 + deriv.)
354 Format: noidle
355 Disable APC CPU standby support. SPARCstation-Fox does
356 not play well with APC CPU idle - disable it if you have
357 APC and your system crashes randomly.
358
359 apic= [APIC,X86] Advanced Programmable Interrupt Controller
360 Change the output verbosity while booting
361 Format: { quiet (default) | verbose | debug }
362 Change the amount of debugging information output
363 when initialising the APIC and IO-APIC components.
364 For X86-32, this can also be used to specify an APIC
365 driver name.
366 Format: apic=driver_name
367 Examples: apic=bigsmp
368
369 apic_extnmi= [APIC,X86] External NMI delivery setting
370 Format: { bsp (default) | all | none }
371 bsp: External NMI is delivered only to CPU 0
372 all: External NMIs are broadcast to all CPUs as a
373 backup of CPU 0
374 none: External NMI is masked for all CPUs. This is
375 useful so that a dump capture kernel won't be
376 shot down by NMI
377
378 autoconf= [IPV6]
379 See Documentation/networking/ipv6.rst.
380
381 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
382 Limit apic dumping. The parameter defines the maximal
383 number of local apics being dumped. Also it is possible
384 to set it to "all" by meaning -- no limit here.
385 Format: { 1 (default) | 2 | ... | all }.
386 The parameter valid if only apic=debug or
387 apic=verbose is specified.
388 Example: apic=debug show_lapic=all
389
390 apm= [APM] Advanced Power Management
391 See header of arch/x86/kernel/apm_32.c.
392
393 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
394 Format: <io>,<irq>,<nodeID>
395
396 arm64.nobti [ARM64] Unconditionally disable Branch Target
397 Identification support
398
399 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication
400 support
401
402 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension
403 support
404
405 arm64.nosve [ARM64] Unconditionally disable Scalable Vector
406 Extension support
407
408 arm64.nosme [ARM64] Unconditionally disable Scalable Matrix
409 Extension support
410
411 ataflop= [HW,M68k]
412
413 atarimouse= [HW,MOUSE] Atari Mouse
414
415 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
416 EzKey and similar keyboards
417
418 atkbd.reset= [HW] Reset keyboard during initialization
419
420 atkbd.set= [HW] Select keyboard code set
421 Format: <int> (2 = AT (default), 3 = PS/2)
422
423 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
424 keyboards
425
426 atkbd.softraw= [HW] Choose between synthetic and real raw mode
427 Format: <bool> (0 = real, 1 = synthetic (default))
428
429 atkbd.softrepeat= [HW]
430 Use software keyboard repeat
431
432 audit= [KNL] Enable the audit sub-system
433 Format: { "0" | "1" | "off" | "on" }
434 0 | off - kernel audit is disabled and can not be
435 enabled until the next reboot
436 unset - kernel audit is initialized but disabled and
437 will be fully enabled by the userspace auditd.
438 1 | on - kernel audit is initialized and partially
439 enabled, storing at most audit_backlog_limit
440 messages in RAM until it is fully enabled by the
441 userspace auditd.
442 Default: unset
443
444 audit_backlog_limit= [KNL] Set the audit queue size limit.
445 Format: <int> (must be >=0)
446 Default: 64
447
448 bau= [X86_UV] Enable the BAU on SGI UV. The default
449 behavior is to disable the BAU (i.e. bau=0).
450 Format: { "0" | "1" }
451 0 - Disable the BAU.
452 1 - Enable the BAU.
453 unset - Disable the BAU.
454
455 baycom_epp= [HW,AX25]
456 Format: <io>,<mode>
457
458 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
459 Format: <io>,<mode>
460 See header of drivers/net/hamradio/baycom_par.c.
461
462 baycom_ser_fdx= [HW,AX25]
463 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
464 Format: <io>,<irq>,<mode>[,<baud>]
465 See header of drivers/net/hamradio/baycom_ser_fdx.c.
466
467 baycom_ser_hdx= [HW,AX25]
468 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
469 Format: <io>,<irq>,<mode>
470 See header of drivers/net/hamradio/baycom_ser_hdx.c.
471
472 bert_disable [ACPI]
473 Disable BERT OS support on buggy BIOSes.
474
475 bgrt_disable [ACPI][X86]
476 Disable BGRT to avoid flickering OEM logo.
477
478 blkdevparts= Manual partition parsing of block device(s) for
479 embedded devices based on command line input.
480 See Documentation/block/cmdline-partition.rst
481
482 boot_delay= Milliseconds to delay each printk during boot.
483 Values larger than 10 seconds (10000) are changed to
484 no delay (0).
485 Format: integer
486
487 bootconfig [KNL]
488 Extended command line options can be added to an initrd
489 and this will cause the kernel to look for it.
490
491 See Documentation/admin-guide/bootconfig.rst
492
493 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
494 bttv.radio= Most important insmod options are available as
495 kernel args too.
496 bttv.pll= See Documentation/admin-guide/media/bttv.rst
497 bttv.tuner=
498
499 bulk_remove=off [PPC] This parameter disables the use of the pSeries
500 firmware feature for flushing multiple hpte entries
501 at a time.
502
503 c101= [NET] Moxa C101 synchronous serial card
504
505 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
506 Sometimes CPU hardware bugs make them report the cache
507 size incorrectly. The kernel will attempt work arounds
508 to fix known problems, but for some CPUs it is not
509 possible to determine what the correct size should be.
510 This option provides an override for these situations.
511
512 carrier_timeout=
513 [NET] Specifies amount of time (in seconds) that
514 the kernel should wait for a network carrier. By default
515 it waits 120 seconds.
516
517 ca_keys= [KEYS] This parameter identifies a specific key(s) on
518 the system trusted keyring to be used for certificate
519 trust validation.
520 format: { id:<keyid> | builtin }
521
522 cca= [MIPS] Override the kernel pages' cache coherency
523 algorithm. Accepted values range from 0 to 7
524 inclusive. See arch/mips/include/asm/pgtable-bits.h
525 for platform specific values (SB1, Loongson3 and
526 others).
527
528 ccw_timeout_log [S390]
529 See Documentation/s390/common_io.rst for details.
530
531 cgroup_disable= [KNL] Disable a particular controller or optional feature
532 Format: {name of the controller(s) or feature(s) to disable}
533 The effects of cgroup_disable=foo are:
534 - foo isn't auto-mounted if you mount all cgroups in
535 a single hierarchy
536 - foo isn't visible as an individually mountable
537 subsystem
538 - if foo is an optional feature then the feature is
539 disabled and corresponding cgroup files are not
540 created
541 {Currently only "memory" controller deal with this and
542 cut the overhead, others just disable the usage. So
543 only cgroup_disable=memory is actually worthy}
544 Specifying "pressure" disables per-cgroup pressure
545 stall information accounting feature
546
547 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
548 Format: { { controller | "all" | "named" }
549 [,{ controller | "all" | "named" }...] }
550 Like cgroup_disable, but only applies to cgroup v1;
551 the blacklisted controllers remain available in cgroup2.
552 "all" blacklists all controllers and "named" disables
553 named mounts. Specifying both "all" and "named" disables
554 all v1 hierarchies.
555
556 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
557 Format: <string>
558 nosocket -- Disable socket memory accounting.
559 nokmem -- Disable kernel memory accounting.
560
561 checkreqprot= [SELINUX] Set initial checkreqprot flag value.
562 Format: { "0" | "1" }
563 See security/selinux/Kconfig help text.
564 0 -- check protection applied by kernel (includes
565 any implied execute protection).
566 1 -- check protection requested by application.
567 Default value is set via a kernel config option.
568 Value can be changed at runtime via
569 /sys/fs/selinux/checkreqprot.
570 Setting checkreqprot to 1 is deprecated.
571
572 cio_ignore= [S390]
573 See Documentation/s390/common_io.rst for details.
574
575 clearcpuid=X[,X...] [X86]
576 Disable CPUID feature X for the kernel. See
577 arch/x86/include/asm/cpufeatures.h for the valid bit
578 numbers X. Note the Linux-specific bits are not necessarily
579 stable over kernel options, but the vendor-specific
580 ones should be.
581 X can also be a string as appearing in the flags: line
582 in /proc/cpuinfo which does not have the above
583 instability issue. However, not all features have names
584 in /proc/cpuinfo.
585 Note that using this option will taint your kernel.
586 Also note that user programs calling CPUID directly
587 or using the feature without checking anything
588 will still see it. This just prevents it from
589 being used by the kernel or shown in /proc/cpuinfo.
590 Also note the kernel might malfunction if you disable
591 some critical bits.
592
593 clk_ignore_unused
594 [CLK]
595 Prevents the clock framework from automatically gating
596 clocks that have not been explicitly enabled by a Linux
597 device driver but are enabled in hardware at reset or
598 by the bootloader/firmware. Note that this does not
599 force such clocks to be always-on nor does it reserve
600 those clocks in any way. This parameter is useful for
601 debug and development, but should not be needed on a
602 platform with proper driver support. For more
603 information, see Documentation/driver-api/clk.rst.
604
605 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
606 [Deprecated]
607 Forces specified clocksource (if available) to be used
608 when calculating gettimeofday(). If specified
609 clocksource is not available, it defaults to PIT.
610 Format: { pit | tsc | cyclone | pmtmr }
611
612 clocksource= Override the default clocksource
613 Format: <string>
614 Override the default clocksource and use the clocksource
615 with the name specified.
616 Some clocksource names to choose from, depending on
617 the platform:
618 [all] jiffies (this is the base, fallback clocksource)
619 [ACPI] acpi_pm
620 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
621 pxa_timer,timer3,32k_counter,timer0_1
622 [X86-32] pit,hpet,tsc;
623 scx200_hrt on Geode; cyclone on IBM x440
624 [MIPS] MIPS
625 [PARISC] cr16
626 [S390] tod
627 [SH] SuperH
628 [SPARC64] tick
629 [X86-64] hpet,tsc
630
631 clocksource.arm_arch_timer.evtstrm=
632 [ARM,ARM64]
633 Format: <bool>
634 Enable/disable the eventstream feature of the ARM
635 architected timer so that code using WFE-based polling
636 loops can be debugged more effectively on production
637 systems.
638
639 clocksource.max_cswd_read_retries= [KNL]
640 Number of clocksource_watchdog() retries due to
641 external delays before the clock will be marked
642 unstable. Defaults to two retries, that is,
643 three attempts to read the clock under test.
644
645 clocksource.verify_n_cpus= [KNL]
646 Limit the number of CPUs checked for clocksources
647 marked with CLOCK_SOURCE_VERIFY_PERCPU that
648 are marked unstable due to excessive skew.
649 A negative value says to check all CPUs, while
650 zero says not to check any. Values larger than
651 nr_cpu_ids are silently truncated to nr_cpu_ids.
652 The actual CPUs are chosen randomly, with
653 no replacement if the same CPU is chosen twice.
654
655 clocksource-wdtest.holdoff= [KNL]
656 Set the time in seconds that the clocksource
657 watchdog test waits before commencing its tests.
658 Defaults to zero when built as a module and to
659 10 seconds when built into the kernel.
660
661 cma=nn[MG]@[start[MG][-end[MG]]]
662 [KNL,CMA]
663 Sets the size of kernel global memory area for
664 contiguous memory allocations and optionally the
665 placement constraint by the physical address range of
666 memory allocations. A value of 0 disables CMA
667 altogether. For more information, see
668 kernel/dma/contiguous.c
669
670 cma_pernuma=nn[MG]
671 [ARM64,KNL,CMA]
672 Sets the size of kernel per-numa memory area for
673 contiguous memory allocations. A value of 0 disables
674 per-numa CMA altogether. And If this option is not
675 specificed, the default value is 0.
676 With per-numa CMA enabled, DMA users on node nid will
677 first try to allocate buffer from the pernuma area
678 which is located in node nid, if the allocation fails,
679 they will fallback to the global default memory area.
680
681 cmo_free_hint= [PPC] Format: { yes | no }
682 Specify whether pages are marked as being inactive
683 when they are freed. This is used in CMO environments
684 to determine OS memory pressure for page stealing by
685 a hypervisor.
686 Default: yes
687
688 coherent_pool=nn[KMG] [ARM,KNL]
689 Sets the size of memory pool for coherent, atomic dma
690 allocations, by default set to 256K.
691
692 com20020= [HW,NET] ARCnet - COM20020 chipset
693 Format:
694 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
695
696 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
697 Format: <io>[,<irq>]
698
699 com90xx= [HW,NET]
700 ARCnet - COM90xx chipset (memory-mapped buffers)
701 Format: <io>[,<irq>[,<memstart>]]
702
703 condev= [HW,S390] console device
704 conmode=
705
706 con3215_drop= [S390] 3215 console drop mode.
707 Format: y|n|Y|N|1|0
708 When set to true, drop data on the 3215 console when
709 the console buffer is full. In this case the
710 operator using a 3270 terminal emulator (for example
711 x3270) does not have to enter the clear key for the
712 console output to advance and the kernel to continue.
713 This leads to a much faster boot time when a 3270
714 terminal emulator is active. If no 3270 terminal
715 emulator is used, this parameter has no effect.
716
717 console= [KNL] Output console device and options.
718
719 tty<n> Use the virtual console device <n>.
720
721 ttyS<n>[,options]
722 ttyUSB0[,options]
723 Use the specified serial port. The options are of
724 the form "bbbbpnf", where "bbbb" is the baud rate,
725 "p" is parity ("n", "o", or "e"), "n" is number of
726 bits, and "f" is flow control ("r" for RTS or
727 omit it). Default is "9600n8".
728
729 See Documentation/admin-guide/serial-console.rst for more
730 information. See
731 Documentation/networking/netconsole.rst for an
732 alternative.
733
734 uart[8250],io,<addr>[,options]
735 uart[8250],mmio,<addr>[,options]
736 uart[8250],mmio16,<addr>[,options]
737 uart[8250],mmio32,<addr>[,options]
738 uart[8250],0x<addr>[,options]
739 Start an early, polled-mode console on the 8250/16550
740 UART at the specified I/O port or MMIO address,
741 switching to the matching ttyS device later.
742 MMIO inter-register address stride is either 8-bit
743 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
744 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
745 to be equivalent to 'mmio'. 'options' are specified in
746 the same format described for ttyS above; if unspecified,
747 the h/w is not re-initialized.
748
749 hvc<n> Use the hypervisor console device <n>. This is for
750 both Xen and PowerPC hypervisors.
751
752 { null | "" }
753 Use to disable console output, i.e., to have kernel
754 console messages discarded.
755 This must be the only console= parameter used on the
756 kernel command line.
757
758 If the device connected to the port is not a TTY but a braille
759 device, prepend "brl," before the device type, for instance
760 console=brl,ttyS0
761 For now, only VisioBraille is supported.
762
763 console_msg_format=
764 [KNL] Change console messages format
765 default
766 By default we print messages on consoles in
767 "[time stamp] text\n" format (time stamp may not be
768 printed, depending on CONFIG_PRINTK_TIME or
769 `printk_time' param).
770 syslog
771 Switch to syslog format: "<%u>[time stamp] text\n"
772 IOW, each message will have a facility and loglevel
773 prefix. The format is similar to one used by syslog()
774 syscall, or to executing "dmesg -S --raw" or to reading
775 from /proc/kmsg.
776
777 consoleblank= [KNL] The console blank (screen saver) timeout in
778 seconds. A value of 0 disables the blank timer.
779 Defaults to 0.
780
781 coredump_filter=
782 [KNL] Change the default value for
783 /proc/<pid>/coredump_filter.
784 See also Documentation/filesystems/proc.rst.
785
786 coresight_cpu_debug.enable
787 [ARM,ARM64]
788 Format: <bool>
789 Enable/disable the CPU sampling based debugging.
790 0: default value, disable debugging
791 1: enable debugging at boot time
792
793 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
794 Format:
795 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
796
797 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
798 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
799 Some features depend on CPU0. Known dependencies are:
800 1. Resume from suspend/hibernate depends on CPU0.
801 Suspend/hibernate will fail if CPU0 is offline and you
802 need to online CPU0 before suspend/hibernate.
803 2. PIC interrupts also depend on CPU0. CPU0 can't be
804 removed if a PIC interrupt is detected.
805 It's said poweroff/reboot may depend on CPU0 on some
806 machines although I haven't seen such issues so far
807 after CPU0 is offline on a few tested machines.
808 If the dependencies are under your control, you can
809 turn on cpu0_hotplug.
810
811 cpuidle.off=1 [CPU_IDLE]
812 disable the cpuidle sub-system
813
814 cpuidle.governor=
815 [CPU_IDLE] Name of the cpuidle governor to use.
816
817 cpufreq.off=1 [CPU_FREQ]
818 disable the cpufreq sub-system
819
820 cpufreq.default_governor=
821 [CPU_FREQ] Name of the default cpufreq governor or
822 policy to use. This governor must be registered in the
823 kernel before the cpufreq driver probes.
824
825 cpu_init_udelay=N
826 [X86] Delay for N microsec between assert and de-assert
827 of APIC INIT to start processors. This delay occurs
828 on every CPU online, such as boot, and resume from suspend.
829 Default: 10000
830
831 crash_kexec_post_notifiers
832 Run kdump after running panic-notifiers and dumping
833 kmsg. This only for the users who doubt kdump always
834 succeeds in any situation.
835 Note that this also increases risks of kdump failure,
836 because some panic notifiers can make the crashed
837 kernel more unstable.
838
839 crashkernel=size[KMG][@offset[KMG]]
840 [KNL] Using kexec, Linux can switch to a 'crash kernel'
841 upon panic. This parameter reserves the physical
842 memory region [offset, offset + size] for that kernel
843 image. If '@offset' is omitted, then a suitable offset
844 is selected automatically.
845 [KNL, X86-64, ARM64] Select a region under 4G first, and
846 fall back to reserve region above 4G when '@offset'
847 hasn't been specified.
848 See Documentation/admin-guide/kdump/kdump.rst for further details.
849
850 crashkernel=range1:size1[,range2:size2,...][@offset]
851 [KNL] Same as above, but depends on the memory
852 in the running system. The syntax of range is
853 start-[end] where start and end are both
854 a memory unit (amount[KMG]). See also
855 Documentation/admin-guide/kdump/kdump.rst for an example.
856
857 crashkernel=size[KMG],high
858 [KNL, X86-64, ARM64] range could be above 4G. Allow kernel
859 to allocate physical memory region from top, so could
860 be above 4G if system have more than 4G ram installed.
861 Otherwise memory region will be allocated below 4G, if
862 available.
863 It will be ignored if crashkernel=X is specified.
864 crashkernel=size[KMG],low
865 [KNL, X86-64, ARM64] range under 4G. When crashkernel=X,high
866 is passed, kernel could allocate physical memory region
867 above 4G, that cause second kernel crash on system
868 that require some amount of low memory, e.g. swiotlb
869 requires at least 64M+32K low memory, also enough extra
870 low memory is needed to make sure DMA buffers for 32-bit
871 devices won't run out. Kernel would try to allocate
872 default size of memory below 4G automatically. The default
873 size is platform dependent.
874 --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB)
875 --> arm64: 128MiB
876 This one lets the user specify own low range under 4G
877 for second kernel instead.
878 0: to disable low allocation.
879 It will be ignored when crashkernel=X,high is not used
880 or memory reserved is below 4G.
881
882 cryptomgr.notests
883 [KNL] Disable crypto self-tests
884
885 cs89x0_dma= [HW,NET]
886 Format: <dma>
887
888 cs89x0_media= [HW,NET]
889 Format: { rj45 | aui | bnc }
890
891 csdlock_debug= [KNL] Enable debug add-ons of cross-CPU function call
892 handling. When switched on, additional debug data is
893 printed to the console in case a hanging CPU is
894 detected, and that CPU is pinged again in order to try
895 to resolve the hang situation.
896 0: disable csdlock debugging (default)
897 1: enable basic csdlock debugging (minor impact)
898 ext: enable extended csdlock debugging (more impact,
899 but more data)
900
901 dasd= [HW,NET]
902 See header of drivers/s390/block/dasd_devmap.c.
903
904 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
905 (one device per port)
906 Format: <port#>,<type>
907 See also Documentation/input/devices/joystick-parport.rst
908
909 debug [KNL] Enable kernel debugging (events log level).
910
911 debug_boot_weak_hash
912 [KNL] Enable printing [hashed] pointers early in the
913 boot sequence. If enabled, we use a weak hash instead
914 of siphash to hash pointers. Use this option if you are
915 seeing instances of '(___ptrval___)') and need to see a
916 value (hashed pointer) instead. Cryptographically
917 insecure, please do not use on production kernels.
918
919 debug_locks_verbose=
920 [KNL] verbose locking self-tests
921 Format: <int>
922 Print debugging info while doing the locking API
923 self-tests.
924 Bitmask for the various LOCKTYPE_ tests. Defaults to 0
925 (no extra messages), setting it to -1 (all bits set)
926 will print _a_lot_ more information - normally only
927 useful to lockdep developers.
928
929 debug_objects [KNL] Enable object debugging
930
931 no_debug_objects
932 [KNL] Disable object debugging
933
934 debug_guardpage_minorder=
935 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
936 parameter allows control of the order of pages that will
937 be intentionally kept free (and hence protected) by the
938 buddy allocator. Bigger value increase the probability
939 of catching random memory corruption, but reduce the
940 amount of memory for normal system use. The maximum
941 possible value is MAX_ORDER/2. Setting this parameter
942 to 1 or 2 should be enough to identify most random
943 memory corruption problems caused by bugs in kernel or
944 driver code when a CPU writes to (or reads from) a
945 random memory location. Note that there exists a class
946 of memory corruptions problems caused by buggy H/W or
947 F/W or by drivers badly programing DMA (basically when
948 memory is written at bus level and the CPU MMU is
949 bypassed) which are not detectable by
950 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
951 tracking down these problems.
952
953 debug_pagealloc=
954 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
955 enables the feature at boot time. By default, it is
956 disabled and the system will work mostly the same as a
957 kernel built without CONFIG_DEBUG_PAGEALLOC.
958 Note: to get most of debug_pagealloc error reports, it's
959 useful to also enable the page_owner functionality.
960 on: enable the feature
961
962 debugfs= [KNL] This parameter enables what is exposed to userspace
963 and debugfs internal clients.
964 Format: { on, no-mount, off }
965 on: All functions are enabled.
966 no-mount:
967 Filesystem is not registered but kernel clients can
968 access APIs and a crashkernel can be used to read
969 its content. There is nothing to mount.
970 off: Filesystem is not registered and clients
971 get a -EPERM as result when trying to register files
972 or directories within debugfs.
973 This is equivalent of the runtime functionality if
974 debugfs was not enabled in the kernel at all.
975 Default value is set in build-time with a kernel configuration.
976
977 debugpat [X86] Enable PAT debugging
978
979 default_hugepagesz=
980 [HW] The size of the default HugeTLB page. This is
981 the size represented by the legacy /proc/ hugepages
982 APIs. In addition, this is the default hugetlb size
983 used for shmget(), mmap() and mounting hugetlbfs
984 filesystems. If not specified, defaults to the
985 architecture's default huge page size. Huge page
986 sizes are architecture dependent. See also
987 Documentation/admin-guide/mm/hugetlbpage.rst.
988 Format: size[KMG]
989
990 deferred_probe_timeout=
991 [KNL] Debugging option to set a timeout in seconds for
992 deferred probe to give up waiting on dependencies to
993 probe. Only specific dependencies (subsystems or
994 drivers) that have opted in will be ignored. A timeout
995 of 0 will timeout at the end of initcalls. If the time
996 out hasn't expired, it'll be restarted by each
997 successful driver registration. This option will also
998 dump out devices still on the deferred probe list after
999 retrying.
1000
1001 delayacct [KNL] Enable per-task delay accounting
1002
1003 dell_smm_hwmon.ignore_dmi=
1004 [HW] Continue probing hardware even if DMI data
1005 indicates that the driver is running on unsupported
1006 hardware.
1007
1008 dell_smm_hwmon.force=
1009 [HW] Activate driver even if SMM BIOS signature does
1010 not match list of supported models and enable otherwise
1011 blacklisted features.
1012
1013 dell_smm_hwmon.power_status=
1014 [HW] Report power status in /proc/i8k
1015 (disabled by default).
1016
1017 dell_smm_hwmon.restricted=
1018 [HW] Allow controlling fans only if SYS_ADMIN
1019 capability is set.
1020
1021 dell_smm_hwmon.fan_mult=
1022 [HW] Factor to multiply fan speed with.
1023
1024 dell_smm_hwmon.fan_max=
1025 [HW] Maximum configurable fan speed.
1026
1027 dfltcc= [HW,S390]
1028 Format: { on | off | def_only | inf_only | always }
1029 on: s390 zlib hardware support for compression on
1030 level 1 and decompression (default)
1031 off: No s390 zlib hardware support
1032 def_only: s390 zlib hardware support for deflate
1033 only (compression on level 1)
1034 inf_only: s390 zlib hardware support for inflate
1035 only (decompression)
1036 always: Same as 'on' but ignores the selected compression
1037 level always using hardware support (used for debugging)
1038
1039 dhash_entries= [KNL]
1040 Set number of hash buckets for dentry cache.
1041
1042 disable_1tb_segments [PPC]
1043 Disables the use of 1TB hash page table segments. This
1044 causes the kernel to fall back to 256MB segments which
1045 can be useful when debugging issues that require an SLB
1046 miss to occur.
1047
1048 stress_slb [PPC]
1049 Limits the number of kernel SLB entries, and flushes
1050 them frequently to increase the rate of SLB faults
1051 on kernel addresses.
1052
1053 stress_hpt [PPC]
1054 Limits the number of kernel HPT entries in the hash
1055 page table to increase the rate of hash page table
1056 faults on kernel addresses.
1057
1058 disable= [IPV6]
1059 See Documentation/networking/ipv6.rst.
1060
1061 disable_radix [PPC]
1062 Disable RADIX MMU mode on POWER9
1063
1064 radix_hcall_invalidate=on [PPC/PSERIES]
1065 Disable RADIX GTSE feature and use hcall for TLB
1066 invalidate.
1067
1068 disable_tlbie [PPC]
1069 Disable TLBIE instruction. Currently does not work
1070 with KVM, with HASH MMU, or with coherent accelerators.
1071
1072 disable_cpu_apicid= [X86,APIC,SMP]
1073 Format: <int>
1074 The number of initial APIC ID for the
1075 corresponding CPU to be disabled at boot,
1076 mostly used for the kdump 2nd kernel to
1077 disable BSP to wake up multiple CPUs without
1078 causing system reset or hang due to sending
1079 INIT from AP to BSP.
1080
1081 disable_ddw [PPC/PSERIES]
1082 Disable Dynamic DMA Window support. Use this
1083 to workaround buggy firmware.
1084
1085 disable_ipv6= [IPV6]
1086 See Documentation/networking/ipv6.rst.
1087
1088 disable_mtrr_cleanup [X86]
1089 The kernel tries to adjust MTRR layout from continuous
1090 to discrete, to make X server driver able to add WB
1091 entry later. This parameter disables that.
1092
1093 disable_mtrr_trim [X86, Intel and AMD only]
1094 By default the kernel will trim any uncacheable
1095 memory out of your available memory pool based on
1096 MTRR settings. This parameter disables that behavior,
1097 possibly causing your machine to run very slowly.
1098
1099 disable_timer_pin_1 [X86]
1100 Disable PIN 1 of APIC timer
1101 Can be useful to work around chipset bugs.
1102
1103 dis_ucode_ldr [X86] Disable the microcode loader.
1104
1105 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
1106 this option disables the debugging code at boot.
1107
1108 dma_debug_entries=<number>
1109 This option allows to tune the number of preallocated
1110 entries for DMA-API debugging code. One entry is
1111 required per DMA-API allocation. Use this if the
1112 DMA-API debugging code disables itself because the
1113 architectural default is too low.
1114
1115 dma_debug_driver=<driver_name>
1116 With this option the DMA-API debugging driver
1117 filter feature can be enabled at boot time. Just
1118 pass the driver to filter for as the parameter.
1119 The filter can be disabled or changed to another
1120 driver later using sysfs.
1121
1122 driver_async_probe= [KNL]
1123 List of driver names to be probed asynchronously. *
1124 matches with all driver names. If * is specified, the
1125 rest of the listed driver names are those that will NOT
1126 match the *.
1127 Format: <driver_name1>,<driver_name2>...
1128
1129 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
1130 Broken monitors, graphic adapters, KVMs and EDIDless
1131 panels may send no or incorrect EDID data sets.
1132 This parameter allows to specify an EDID data sets
1133 in the /lib/firmware directory that are used instead.
1134 Generic built-in EDID data sets are used, if one of
1135 edid/1024x768.bin, edid/1280x1024.bin,
1136 edid/1680x1050.bin, or edid/1920x1080.bin is given
1137 and no file with the same name exists. Details and
1138 instructions how to build your own EDID data are
1139 available in Documentation/admin-guide/edid.rst. An EDID
1140 data set will only be used for a particular connector,
1141 if its name and a colon are prepended to the EDID
1142 name. Each connector may use a unique EDID data
1143 set by separating the files with a comma. An EDID
1144 data set with no connector name will be used for
1145 any connectors not explicitly specified.
1146
1147 dscc4.setup= [NET]
1148
1149 dt_cpu_ftrs= [PPC]
1150 Format: {"off" | "known"}
1151 Control how the dt_cpu_ftrs device-tree binding is
1152 used for CPU feature discovery and setup (if it
1153 exists).
1154 off: Do not use it, fall back to legacy cpu table.
1155 known: Do not pass through unknown features to guests
1156 or userspace, only those that the kernel is aware of.
1157
1158 dump_apple_properties [X86]
1159 Dump name and content of EFI device properties on
1160 x86 Macs. Useful for driver authors to determine
1161 what data is available or for reverse-engineering.
1162
1163 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1164 <module>.dyndbg[="val"]
1165 Enable debug messages at boot time. See
1166 Documentation/admin-guide/dynamic-debug-howto.rst
1167 for details.
1168
1169 nopku [X86] Disable Memory Protection Keys CPU feature found
1170 in some Intel CPUs.
1171
1172 <module>.async_probe[=<bool>] [KNL]
1173 If no <bool> value is specified or if the value
1174 specified is not a valid <bool>, enable asynchronous
1175 probe on this module. Otherwise, enable/disable
1176 asynchronous probe on this module as indicated by the
1177 <bool> value. See also: module.async_probe
1178
1179 early_ioremap_debug [KNL]
1180 Enable debug messages in early_ioremap support. This
1181 is useful for tracking down temporary early mappings
1182 which are not unmapped.
1183
1184 earlycon= [KNL] Output early console device and options.
1185
1186 When used with no options, the early console is
1187 determined by stdout-path property in device tree's
1188 chosen node or the ACPI SPCR table if supported by
1189 the platform.
1190
1191 cdns,<addr>[,options]
1192 Start an early, polled-mode console on a Cadence
1193 (xuartps) serial port at the specified address. Only
1194 supported option is baud rate. If baud rate is not
1195 specified, the serial port must already be setup and
1196 configured.
1197
1198 uart[8250],io,<addr>[,options]
1199 uart[8250],mmio,<addr>[,options]
1200 uart[8250],mmio32,<addr>[,options]
1201 uart[8250],mmio32be,<addr>[,options]
1202 uart[8250],0x<addr>[,options]
1203 Start an early, polled-mode console on the 8250/16550
1204 UART at the specified I/O port or MMIO address.
1205 MMIO inter-register address stride is either 8-bit
1206 (mmio) or 32-bit (mmio32 or mmio32be).
1207 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1208 to be equivalent to 'mmio'. 'options' are specified
1209 in the same format described for "console=ttyS<n>"; if
1210 unspecified, the h/w is not initialized.
1211
1212 pl011,<addr>
1213 pl011,mmio32,<addr>
1214 Start an early, polled-mode console on a pl011 serial
1215 port at the specified address. The pl011 serial port
1216 must already be setup and configured. Options are not
1217 yet supported. If 'mmio32' is specified, then only
1218 the driver will use only 32-bit accessors to read/write
1219 the device registers.
1220
1221 liteuart,<addr>
1222 Start an early console on a litex serial port at the
1223 specified address. The serial port must already be
1224 setup and configured. Options are not yet supported.
1225
1226 meson,<addr>
1227 Start an early, polled-mode console on a meson serial
1228 port at the specified address. The serial port must
1229 already be setup and configured. Options are not yet
1230 supported.
1231
1232 msm_serial,<addr>
1233 Start an early, polled-mode console on an msm serial
1234 port at the specified address. The serial port
1235 must already be setup and configured. Options are not
1236 yet supported.
1237
1238 msm_serial_dm,<addr>
1239 Start an early, polled-mode console on an msm serial
1240 dm port at the specified address. The serial port
1241 must already be setup and configured. Options are not
1242 yet supported.
1243
1244 owl,<addr>
1245 Start an early, polled-mode console on a serial port
1246 of an Actions Semi SoC, such as S500 or S900, at the
1247 specified address. The serial port must already be
1248 setup and configured. Options are not yet supported.
1249
1250 rda,<addr>
1251 Start an early, polled-mode console on a serial port
1252 of an RDA Micro SoC, such as RDA8810PL, at the
1253 specified address. The serial port must already be
1254 setup and configured. Options are not yet supported.
1255
1256 sbi
1257 Use RISC-V SBI (Supervisor Binary Interface) for early
1258 console.
1259
1260 smh Use ARM semihosting calls for early console.
1261
1262 s3c2410,<addr>
1263 s3c2412,<addr>
1264 s3c2440,<addr>
1265 s3c6400,<addr>
1266 s5pv210,<addr>
1267 exynos4210,<addr>
1268 Use early console provided by serial driver available
1269 on Samsung SoCs, requires selecting proper type and
1270 a correct base address of the selected UART port. The
1271 serial port must already be setup and configured.
1272 Options are not yet supported.
1273
1274 lantiq,<addr>
1275 Start an early, polled-mode console on a lantiq serial
1276 (lqasc) port at the specified address. The serial port
1277 must already be setup and configured. Options are not
1278 yet supported.
1279
1280 lpuart,<addr>
1281 lpuart32,<addr>
1282 Use early console provided by Freescale LP UART driver
1283 found on Freescale Vybrid and QorIQ LS1021A processors.
1284 A valid base address must be provided, and the serial
1285 port must already be setup and configured.
1286
1287 ec_imx21,<addr>
1288 ec_imx6q,<addr>
1289 Start an early, polled-mode, output-only console on the
1290 Freescale i.MX UART at the specified address. The UART
1291 must already be setup and configured.
1292
1293 ar3700_uart,<addr>
1294 Start an early, polled-mode console on the
1295 Armada 3700 serial port at the specified
1296 address. The serial port must already be setup
1297 and configured. Options are not yet supported.
1298
1299 qcom_geni,<addr>
1300 Start an early, polled-mode console on a Qualcomm
1301 Generic Interface (GENI) based serial port at the
1302 specified address. The serial port must already be
1303 setup and configured. Options are not yet supported.
1304
1305 efifb,[options]
1306 Start an early, unaccelerated console on the EFI
1307 memory mapped framebuffer (if available). On cache
1308 coherent non-x86 systems that use system memory for
1309 the framebuffer, pass the 'ram' option so that it is
1310 mapped with the correct attributes.
1311
1312 linflex,<addr>
1313 Use early console provided by Freescale LINFlexD UART
1314 serial driver for NXP S32V234 SoCs. A valid base
1315 address must be provided, and the serial port must
1316 already be setup and configured.
1317
1318 earlyprintk= [X86,SH,ARM,M68k,S390]
1319 earlyprintk=vga
1320 earlyprintk=sclp
1321 earlyprintk=xen
1322 earlyprintk=serial[,ttySn[,baudrate]]
1323 earlyprintk=serial[,0x...[,baudrate]]
1324 earlyprintk=ttySn[,baudrate]
1325 earlyprintk=dbgp[debugController#]
1326 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1327 earlyprintk=xdbc[xhciController#]
1328
1329 earlyprintk is useful when the kernel crashes before
1330 the normal console is initialized. It is not enabled by
1331 default because it has some cosmetic problems.
1332
1333 Append ",keep" to not disable it when the real console
1334 takes over.
1335
1336 Only one of vga, serial, or usb debug port can
1337 be used at a time.
1338
1339 Currently only ttyS0 and ttyS1 may be specified by
1340 name. Other I/O ports may be explicitly specified
1341 on some architectures (x86 and arm at least) by
1342 replacing ttySn with an I/O port address, like this:
1343 earlyprintk=serial,0x1008,115200
1344 You can find the port for a given device in
1345 /proc/tty/driver/serial:
1346 2: uart:ST16650V2 port:00001008 irq:18 ...
1347
1348 Interaction with the standard serial driver is not
1349 very good.
1350
1351 The VGA output is eventually overwritten by
1352 the real console.
1353
1354 The xen option can only be used in Xen domains.
1355
1356 The sclp output can only be used on s390.
1357
1358 The optional "force" to "pciserial" enables use of a
1359 PCI device even when its classcode is not of the
1360 UART class.
1361
1362 edac_report= [HW,EDAC] Control how to report EDAC event
1363 Format: {"on" | "off" | "force"}
1364 on: enable EDAC to report H/W event. May be overridden
1365 by other higher priority error reporting module.
1366 off: disable H/W event reporting through EDAC.
1367 force: enforce the use of EDAC to report H/W event.
1368 default: on.
1369
1370 edd= [EDD]
1371 Format: {"off" | "on" | "skip[mbr]"}
1372
1373 efi= [EFI]
1374 Format: { "debug", "disable_early_pci_dma",
1375 "nochunk", "noruntime", "nosoftreserve",
1376 "novamap", "no_disable_early_pci_dma" }
1377 debug: enable misc debug output.
1378 disable_early_pci_dma: disable the busmaster bit on all
1379 PCI bridges while in the EFI boot stub.
1380 nochunk: disable reading files in "chunks" in the EFI
1381 boot stub, as chunking can cause problems with some
1382 firmware implementations.
1383 noruntime : disable EFI runtime services support
1384 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1385 attribute may cause the kernel to reserve the
1386 memory range for a memory mapping driver to
1387 claim. Specify efi=nosoftreserve to disable this
1388 reservation and treat the memory by its base type
1389 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1390 novamap: do not call SetVirtualAddressMap().
1391 no_disable_early_pci_dma: Leave the busmaster bit set
1392 on all PCI bridges while in the EFI boot stub
1393
1394 efi_no_storage_paranoia [EFI; X86]
1395 Using this parameter you can use more than 50% of
1396 your efi variable storage. Use this parameter only if
1397 you are really sure that your UEFI does sane gc and
1398 fulfills the spec otherwise your board may brick.
1399
1400 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1401 Add arbitrary attribute to specific memory range by
1402 updating original EFI memory map.
1403 Region of memory which aa attribute is added to is
1404 from ss to ss+nn.
1405
1406 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1407 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1408 attribute is added to range 0x100000000-0x180000000 and
1409 0x10a0000000-0x1120000000.
1410
1411 If efi_fake_mem=8G@9G:0x40000 is specified, the
1412 EFI_MEMORY_SP(0x40000) attribute is added to
1413 range 0x240000000-0x43fffffff.
1414
1415 Using this parameter you can do debugging of EFI memmap
1416 related features. For example, you can do debugging of
1417 Address Range Mirroring feature even if your box
1418 doesn't support it, or mark specific memory as
1419 "soft reserved".
1420
1421 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1422 that is to be dynamically loaded by Linux. If there are
1423 multiple variables with the same name but with different
1424 vendor GUIDs, all of them will be loaded. See
1425 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1426
1427
1428 eisa_irq_edge= [PARISC,HW]
1429 See header of drivers/parisc/eisa.c.
1430
1431 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1432 Format: ekgdboc=kbd
1433
1434 This is designed to be used in conjunction with
1435 the boot argument: earlyprintk=vga
1436
1437 This parameter works in place of the kgdboc parameter
1438 but can only be used if the backing tty is available
1439 very early in the boot process. For early debugging
1440 via a serial port see kgdboc_earlycon instead.
1441
1442 elanfreq= [X86-32]
1443 See comment before function elanfreq_setup() in
1444 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1445
1446 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1447 Specifies physical address of start of kernel core
1448 image elf header and optionally the size. Generally
1449 kexec loader will pass this option to capture kernel.
1450 See Documentation/admin-guide/kdump/kdump.rst for details.
1451
1452 enable_mtrr_cleanup [X86]
1453 The kernel tries to adjust MTRR layout from continuous
1454 to discrete, to make X server driver able to add WB
1455 entry later. This parameter enables that.
1456
1457 enable_timer_pin_1 [X86]
1458 Enable PIN 1 of APIC timer
1459 Can be useful to work around chipset bugs
1460 (in particular on some ATI chipsets).
1461 The kernel tries to set a reasonable default.
1462
1463 enforcing= [SELINUX] Set initial enforcing status.
1464 Format: {"0" | "1"}
1465 See security/selinux/Kconfig help text.
1466 0 -- permissive (log only, no denials).
1467 1 -- enforcing (deny and log).
1468 Default value is 0.
1469 Value can be changed at runtime via
1470 /sys/fs/selinux/enforce.
1471
1472 erst_disable [ACPI]
1473 Disable Error Record Serialization Table (ERST)
1474 support.
1475
1476 ether= [HW,NET] Ethernet cards parameters
1477 This option is obsoleted by the "netdev=" option, which
1478 has equivalent usage. See its documentation for details.
1479
1480 evm= [EVM]
1481 Format: { "fix" }
1482 Permit 'security.evm' to be updated regardless of
1483 current integrity status.
1484
1485 early_page_ext [KNL] Enforces page_ext initialization to earlier
1486 stages so cover more early boot allocations.
1487 Please note that as side effect some optimizations
1488 might be disabled to achieve that (e.g. parallelized
1489 memory initialization is disabled) so the boot process
1490 might take longer, especially on systems with a lot of
1491 memory. Available with CONFIG_PAGE_EXTENSION=y.
1492
1493 failslab=
1494 fail_usercopy=
1495 fail_page_alloc=
1496 fail_make_request=[KNL]
1497 General fault injection mechanism.
1498 Format: <interval>,<probability>,<space>,<times>
1499 See also Documentation/fault-injection/.
1500
1501 fb_tunnels= [NET]
1502 Format: { initns | none }
1503 See Documentation/admin-guide/sysctl/net.rst for
1504 fb_tunnels_only_for_init_ns
1505
1506 floppy= [HW]
1507 See Documentation/admin-guide/blockdev/floppy.rst.
1508
1509 force_pal_cache_flush
1510 [IA-64] Avoid check_sal_cache_flush which may hang on
1511 buggy SAL_CACHE_FLUSH implementations. Using this
1512 parameter will force ia64_sal_cache_flush to call
1513 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1514
1515 forcepae [X86-32]
1516 Forcefully enable Physical Address Extension (PAE).
1517 Many Pentium M systems disable PAE but may have a
1518 functionally usable PAE implementation.
1519 Warning: use of this parameter will taint the kernel
1520 and may cause unknown problems.
1521
1522 ftrace=[tracer]
1523 [FTRACE] will set and start the specified tracer
1524 as early as possible in order to facilitate early
1525 boot debugging.
1526
1527 ftrace_boot_snapshot
1528 [FTRACE] On boot up, a snapshot will be taken of the
1529 ftrace ring buffer that can be read at:
1530 /sys/kernel/tracing/snapshot.
1531 This is useful if you need tracing information from kernel
1532 boot up that is likely to be overridden by user space
1533 start up functionality.
1534
1535 ftrace_dump_on_oops[=orig_cpu]
1536 [FTRACE] will dump the trace buffers on oops.
1537 If no parameter is passed, ftrace will dump
1538 buffers of all CPUs, but if you pass orig_cpu, it will
1539 dump only the buffer of the CPU that triggered the
1540 oops.
1541
1542 ftrace_filter=[function-list]
1543 [FTRACE] Limit the functions traced by the function
1544 tracer at boot up. function-list is a comma-separated
1545 list of functions. This list can be changed at run
1546 time by the set_ftrace_filter file in the debugfs
1547 tracing directory.
1548
1549 ftrace_notrace=[function-list]
1550 [FTRACE] Do not trace the functions specified in
1551 function-list. This list can be changed at run time
1552 by the set_ftrace_notrace file in the debugfs
1553 tracing directory.
1554
1555 ftrace_graph_filter=[function-list]
1556 [FTRACE] Limit the top level callers functions traced
1557 by the function graph tracer at boot up.
1558 function-list is a comma-separated list of functions
1559 that can be changed at run time by the
1560 set_graph_function file in the debugfs tracing directory.
1561
1562 ftrace_graph_notrace=[function-list]
1563 [FTRACE] Do not trace from the functions specified in
1564 function-list. This list is a comma-separated list of
1565 functions that can be changed at run time by the
1566 set_graph_notrace file in the debugfs tracing directory.
1567
1568 ftrace_graph_max_depth=<uint>
1569 [FTRACE] Used with the function graph tracer. This is
1570 the max depth it will trace into a function. This value
1571 can be changed at run time by the max_graph_depth file
1572 in the tracefs tracing directory. default: 0 (no limit)
1573
1574 fw_devlink= [KNL] Create device links between consumer and supplier
1575 devices by scanning the firmware to infer the
1576 consumer/supplier relationships. This feature is
1577 especially useful when drivers are loaded as modules as
1578 it ensures proper ordering of tasks like device probing
1579 (suppliers first, then consumers), supplier boot state
1580 clean up (only after all consumers have probed),
1581 suspend/resume & runtime PM (consumers first, then
1582 suppliers).
1583 Format: { off | permissive | on | rpm }
1584 off -- Don't create device links from firmware info.
1585 permissive -- Create device links from firmware info
1586 but use it only for ordering boot state clean
1587 up (sync_state() calls).
1588 on -- Create device links from firmware info and use it
1589 to enforce probe and suspend/resume ordering.
1590 rpm -- Like "on", but also use to order runtime PM.
1591
1592 fw_devlink.strict=<bool>
1593 [KNL] Treat all inferred dependencies as mandatory
1594 dependencies. This only applies for fw_devlink=on|rpm.
1595 Format: <bool>
1596
1597 gamecon.map[2|3]=
1598 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1599 support via parallel port (up to 5 devices per port)
1600 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1601 See also Documentation/input/devices/joystick-parport.rst
1602
1603 gamma= [HW,DRM]
1604
1605 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART
1606 Format: off | on
1607 default: on
1608
1609 gcov_persist= [GCOV] When non-zero (default), profiling data for
1610 kernel modules is saved and remains accessible via
1611 debugfs, even when the module is unloaded/reloaded.
1612 When zero, profiling data is discarded and associated
1613 debugfs files are removed at module unload time.
1614
1615 goldfish [X86] Enable the goldfish android emulator platform.
1616 Don't use this when you are not running on the
1617 android emulator
1618
1619 gpio-mockup.gpio_mockup_ranges
1620 [HW] Sets the ranges of gpiochip of for this device.
1621 Format: <start1>,<end1>,<start2>,<end2>...
1622 gpio-mockup.gpio_mockup_named_lines
1623 [HW] Let the driver know GPIO lines should be named.
1624
1625 gpt [EFI] Forces disk with valid GPT signature but
1626 invalid Protective MBR to be treated as GPT. If the
1627 primary GPT is corrupted, it enables the backup/alternate
1628 GPT to be used instead.
1629
1630 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1631 the "Enable 0" bit of the configuration register.
1632 Format: 0 | 1
1633 Default: 0
1634 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1635 the "Enable 0" bit of the configuration register.
1636 Format: 0 | 1
1637 Default: 0
1638 grcan.select= [HW] Select which physical interface to use.
1639 Format: 0 | 1
1640 Default: 0
1641 grcan.txsize= [HW] Sets the size of the tx buffer.
1642 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1643 Default: 1024
1644 grcan.rxsize= [HW] Sets the size of the rx buffer.
1645 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1646 Default: 1024
1647
1648 hardened_usercopy=
1649 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
1650 hardening is enabled for this boot. Hardened
1651 usercopy checking is used to protect the kernel
1652 from reading or writing beyond known memory
1653 allocation boundaries as a proactive defense
1654 against bounds-checking flaws in the kernel's
1655 copy_to_user()/copy_from_user() interface.
1656 on Perform hardened usercopy checks (default).
1657 off Disable hardened usercopy checks.
1658
1659 hardlockup_all_cpu_backtrace=
1660 [KNL] Should the hard-lockup detector generate
1661 backtraces on all cpus.
1662 Format: 0 | 1
1663
1664 hashdist= [KNL,NUMA] Large hashes allocated during boot
1665 are distributed across NUMA nodes. Defaults on
1666 for 64-bit NUMA, off otherwise.
1667 Format: 0 | 1 (for off | on)
1668
1669 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1670
1671 hd= [EIDE] (E)IDE hard drive subsystem geometry
1672 Format: <cyl>,<head>,<sect>
1673
1674 hest_disable [ACPI]
1675 Disable Hardware Error Source Table (HEST) support;
1676 corresponding firmware-first mode error processing
1677 logic will be disabled.
1678
1679 hibernate= [HIBERNATION]
1680 noresume Don't check if there's a hibernation image
1681 present during boot.
1682 nocompress Don't compress/decompress hibernation images.
1683 no Disable hibernation and resume.
1684 protect_image Turn on image protection during restoration
1685 (that will set all pages holding image data
1686 during restoration read-only).
1687
1688 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1689 size of <nn>. This works even on boxes that have no
1690 highmem otherwise. This also works to reduce highmem
1691 size on bigger boxes.
1692
1693 highres= [KNL] Enable/disable high resolution timer mode.
1694 Valid parameters: "on", "off"
1695 Default: "on"
1696
1697 hlt [BUGS=ARM,SH]
1698
1699 hostname= [KNL] Set the hostname (aka UTS nodename).
1700 Format: <string>
1701 This allows setting the system's hostname during early
1702 startup. This sets the name returned by gethostname.
1703 Using this parameter to set the hostname makes it
1704 possible to ensure the hostname is correctly set before
1705 any userspace processes run, avoiding the possibility
1706 that a process may call gethostname before the hostname
1707 has been explicitly set, resulting in the calling
1708 process getting an incorrect result. The string must
1709 not exceed the maximum allowed hostname length (usually
1710 64 characters) and will be truncated otherwise.
1711
1712 hpet= [X86-32,HPET] option to control HPET usage
1713 Format: { enable (default) | disable | force |
1714 verbose }
1715 disable: disable HPET and use PIT instead
1716 force: allow force enabled of undocumented chips (ICH4,
1717 VIA, nVidia)
1718 verbose: show contents of HPET registers during setup
1719
1720 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1721 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1722
1723 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1724 If this follows hugepagesz (below), it specifies
1725 the number of pages of hugepagesz to be allocated.
1726 If this is the first HugeTLB parameter on the command
1727 line, it specifies the number of pages to allocate for
1728 the default huge page size. If using node format, the
1729 number of pages to allocate per-node can be specified.
1730 See also Documentation/admin-guide/mm/hugetlbpage.rst.
1731 Format: <integer> or (node format)
1732 <node>:<integer>[,<node>:<integer>]
1733
1734 hugepagesz=
1735 [HW] The size of the HugeTLB pages. This is used in
1736 conjunction with hugepages (above) to allocate huge
1737 pages of a specific size at boot. The pair
1738 hugepagesz=X hugepages=Y can be specified once for
1739 each supported huge page size. Huge page sizes are
1740 architecture dependent. See also
1741 Documentation/admin-guide/mm/hugetlbpage.rst.
1742 Format: size[KMG]
1743
1744 hugetlb_cma= [HW,CMA] The size of a CMA area used for allocation
1745 of gigantic hugepages. Or using node format, the size
1746 of a CMA area per node can be specified.
1747 Format: nn[KMGTPE] or (node format)
1748 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]]
1749
1750 Reserve a CMA area of given size and allocate gigantic
1751 hugepages using the CMA allocator. If enabled, the
1752 boot-time allocation of gigantic hugepages is skipped.
1753
1754 hugetlb_free_vmemmap=
1755 [KNL] Reguires CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP
1756 enabled.
1757 Control if HugeTLB Vmemmap Optimization (HVO) is enabled.
1758 Allows heavy hugetlb users to free up some more
1759 memory (7 * PAGE_SIZE for each 2MB hugetlb page).
1760 Format: { on | off (default) }
1761
1762 on: enable HVO
1763 off: disable HVO
1764
1765 Built with CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y,
1766 the default is on.
1767
1768 Note that the vmemmap pages may be allocated from the added
1769 memory block itself when memory_hotplug.memmap_on_memory is
1770 enabled, those vmemmap pages cannot be optimized even if this
1771 feature is enabled. Other vmemmap pages not allocated from
1772 the added memory block itself do not be affected.
1773
1774 hung_task_panic=
1775 [KNL] Should the hung task detector generate panics.
1776 Format: 0 | 1
1777
1778 A value of 1 instructs the kernel to panic when a
1779 hung task is detected. The default value is controlled
1780 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1781 option. The value selected by this boot parameter can
1782 be changed later by the kernel.hung_task_panic sysctl.
1783
1784 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1785 terminal devices. Valid values: 0..8
1786 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1787 If specified, z/VM IUCV HVC accepts connections
1788 from listed z/VM user IDs only.
1789
1790 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1791 which allow the hypervisor to 'idle' the
1792 guest on lock contention.
1793
1794 keep_bootcon [KNL]
1795 Do not unregister boot console at start. This is only
1796 useful for debugging when something happens in the window
1797 between unregistering the boot console and initializing
1798 the real console.
1799
1800 i2c_bus= [HW] Override the default board specific I2C bus speed
1801 or register an additional I2C bus that is not
1802 registered from board initialization code.
1803 Format:
1804 <bus_id>,<clkrate>
1805
1806 i8042.debug [HW] Toggle i8042 debug mode
1807 i8042.unmask_kbd_data
1808 [HW] Enable printing of interrupt data from the KBD port
1809 (disabled by default, and as a pre-condition
1810 requires that i8042.debug=1 be enabled)
1811 i8042.direct [HW] Put keyboard port into non-translated mode
1812 i8042.dumbkbd [HW] Pretend that controller can only read data from
1813 keyboard and cannot control its state
1814 (Don't attempt to blink the leds)
1815 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1816 i8042.nokbd [HW] Don't check/create keyboard port
1817 i8042.noloop [HW] Disable the AUX Loopback command while probing
1818 for the AUX port
1819 i8042.nomux [HW] Don't check presence of an active multiplexing
1820 controller
1821 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1822 controllers
1823 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1824 i8042.reset [HW] Reset the controller during init, cleanup and
1825 suspend-to-ram transitions, only during s2r
1826 transitions, or never reset
1827 Format: { 1 | Y | y | 0 | N | n }
1828 1, Y, y: always reset controller
1829 0, N, n: don't ever reset controller
1830 Default: only on s2r transitions on x86; most other
1831 architectures force reset to be always executed
1832 i8042.unlock [HW] Unlock (ignore) the keylock
1833 i8042.kbdreset [HW] Reset device connected to KBD port
1834 i8042.probe_defer
1835 [HW] Allow deferred probing upon i8042 probe errors
1836
1837 i810= [HW,DRM]
1838
1839 i915.invert_brightness=
1840 [DRM] Invert the sense of the variable that is used to
1841 set the brightness of the panel backlight. Normally a
1842 brightness value of 0 indicates backlight switched off,
1843 and the maximum of the brightness value sets the backlight
1844 to maximum brightness. If this parameter is set to 0
1845 (default) and the machine requires it, or this parameter
1846 is set to 1, a brightness value of 0 sets the backlight
1847 to maximum brightness, and the maximum of the brightness
1848 value switches the backlight off.
1849 -1 -- never invert brightness
1850 0 -- machine default
1851 1 -- force brightness inversion
1852
1853 icn= [HW,ISDN]
1854 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1855
1856
1857 idle= [X86]
1858 Format: idle=poll, idle=halt, idle=nomwait
1859 Poll forces a polling idle loop that can slightly
1860 improve the performance of waking up a idle CPU, but
1861 will use a lot of power and make the system run hot.
1862 Not recommended.
1863 idle=halt: Halt is forced to be used for CPU idle.
1864 In such case C2/C3 won't be used again.
1865 idle=nomwait: Disable mwait for CPU C-states
1866
1867 idxd.sva= [HW]
1868 Format: <bool>
1869 Allow force disabling of Shared Virtual Memory (SVA)
1870 support for the idxd driver. By default it is set to
1871 true (1).
1872
1873 idxd.tc_override= [HW]
1874 Format: <bool>
1875 Allow override of default traffic class configuration
1876 for the device. By default it is set to false (0).
1877
1878 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1879 Format: { strict | legacy | 2008 | relaxed }
1880 Default: strict
1881
1882 Choose which programs will be accepted for execution
1883 based on the IEEE 754 NaN encoding(s) supported by
1884 the FPU and the NaN encoding requested with the value
1885 of an ELF file header flag individually set by each
1886 binary. Hardware implementations are permitted to
1887 support either or both of the legacy and the 2008 NaN
1888 encoding mode.
1889
1890 Available settings are as follows:
1891 strict accept binaries that request a NaN encoding
1892 supported by the FPU
1893 legacy only accept legacy-NaN binaries, if supported
1894 by the FPU
1895 2008 only accept 2008-NaN binaries, if supported
1896 by the FPU
1897 relaxed accept any binaries regardless of whether
1898 supported by the FPU
1899
1900 The FPU emulator is always able to support both NaN
1901 encodings, so if no FPU hardware is present or it has
1902 been disabled with 'nofpu', then the settings of
1903 'legacy' and '2008' strap the emulator accordingly,
1904 'relaxed' straps the emulator for both legacy-NaN and
1905 2008-NaN, whereas 'strict' enables legacy-NaN only on
1906 legacy processors and both NaN encodings on MIPS32 or
1907 MIPS64 CPUs.
1908
1909 The setting for ABS.fmt/NEG.fmt instruction execution
1910 mode generally follows that for the NaN encoding,
1911 except where unsupported by hardware.
1912
1913 ignore_loglevel [KNL]
1914 Ignore loglevel setting - this will print /all/
1915 kernel messages to the console. Useful for debugging.
1916 We also add it as printk module parameter, so users
1917 could change it dynamically, usually by
1918 /sys/module/printk/parameters/ignore_loglevel.
1919
1920 ignore_rlimit_data
1921 Ignore RLIMIT_DATA setting for data mappings,
1922 print warning at first misuse. Can be changed via
1923 /sys/module/kernel/parameters/ignore_rlimit_data.
1924
1925 ihash_entries= [KNL]
1926 Set number of hash buckets for inode cache.
1927
1928 ima_appraise= [IMA] appraise integrity measurements
1929 Format: { "off" | "enforce" | "fix" | "log" }
1930 default: "enforce"
1931
1932 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1933 The builtin appraise policy appraises all files
1934 owned by uid=0.
1935
1936 ima_canonical_fmt [IMA]
1937 Use the canonical format for the binary runtime
1938 measurements, instead of host native format.
1939
1940 ima_hash= [IMA]
1941 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1942 | sha512 | ... }
1943 default: "sha1"
1944
1945 The list of supported hash algorithms is defined
1946 in crypto/hash_info.h.
1947
1948 ima_policy= [IMA]
1949 The builtin policies to load during IMA setup.
1950 Format: "tcb | appraise_tcb | secure_boot |
1951 fail_securely | critical_data"
1952
1953 The "tcb" policy measures all programs exec'd, files
1954 mmap'd for exec, and all files opened with the read
1955 mode bit set by either the effective uid (euid=0) or
1956 uid=0.
1957
1958 The "appraise_tcb" policy appraises the integrity of
1959 all files owned by root.
1960
1961 The "secure_boot" policy appraises the integrity
1962 of files (eg. kexec kernel image, kernel modules,
1963 firmware, policy, etc) based on file signatures.
1964
1965 The "fail_securely" policy forces file signature
1966 verification failure also on privileged mounted
1967 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1968 flag.
1969
1970 The "critical_data" policy measures kernel integrity
1971 critical data.
1972
1973 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1974 Load a policy which meets the needs of the Trusted
1975 Computing Base. This means IMA will measure all
1976 programs exec'd, files mmap'd for exec, and all files
1977 opened for read by uid=0.
1978
1979 ima_template= [IMA]
1980 Select one of defined IMA measurements template formats.
1981 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" |
1982 "ima-sigv2" }
1983 Default: "ima-ng"
1984
1985 ima_template_fmt=
1986 [IMA] Define a custom template format.
1987 Format: { "field1|...|fieldN" }
1988
1989 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1990 Format: <min_file_size>
1991 Set the minimal file size for using asynchronous hash.
1992 If left unspecified, ahash usage is disabled.
1993
1994 ahash performance varies for different data sizes on
1995 different crypto accelerators. This option can be used
1996 to achieve the best performance for a particular HW.
1997
1998 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1999 Format: <bufsize>
2000 Set hashing buffer size. Default: 4k.
2001
2002 ahash performance varies for different chunk sizes on
2003 different crypto accelerators. This option can be used
2004 to achieve best performance for particular HW.
2005
2006 init= [KNL]
2007 Format: <full_path>
2008 Run specified binary instead of /sbin/init as init
2009 process.
2010
2011 initcall_debug [KNL] Trace initcalls as they are executed. Useful
2012 for working out where the kernel is dying during
2013 startup.
2014
2015 initcall_blacklist= [KNL] Do not execute a comma-separated list of
2016 initcall functions. Useful for debugging built-in
2017 modules and initcalls.
2018
2019 initramfs_async= [KNL]
2020 Format: <bool>
2021 Default: 1
2022 This parameter controls whether the initramfs
2023 image is unpacked asynchronously, concurrently
2024 with devices being probed and
2025 initialized. This should normally just work,
2026 but as a debugging aid, one can get the
2027 historical behaviour of the initramfs
2028 unpacking being completed before device_ and
2029 late_ initcalls.
2030
2031 initrd= [BOOT] Specify the location of the initial ramdisk
2032
2033 initrdmem= [KNL] Specify a physical address and size from which to
2034 load the initrd. If an initrd is compiled in or
2035 specified in the bootparams, it takes priority over this
2036 setting.
2037 Format: ss[KMG],nn[KMG]
2038 Default is 0, 0
2039
2040 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
2041 zeroes.
2042 Format: 0 | 1
2043 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
2044
2045 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
2046 Format: 0 | 1
2047 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
2048
2049 init_pkru= [X86] Specify the default memory protection keys rights
2050 register contents for all processes. 0x55555554 by
2051 default (disallow access to all but pkey 0). Can
2052 override in debugfs after boot.
2053
2054 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
2055 Format: <irq>
2056
2057 int_pln_enable [X86] Enable power limit notification interrupt
2058
2059 integrity_audit=[IMA]
2060 Format: { "0" | "1" }
2061 0 -- basic integrity auditing messages. (Default)
2062 1 -- additional integrity auditing messages.
2063
2064 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
2065 on
2066 Enable intel iommu driver.
2067 off
2068 Disable intel iommu driver.
2069 igfx_off [Default Off]
2070 By default, gfx is mapped as normal device. If a gfx
2071 device has a dedicated DMAR unit, the DMAR unit is
2072 bypassed by not enabling DMAR with this option. In
2073 this case, gfx device will use physical address for
2074 DMA.
2075 strict [Default Off]
2076 Deprecated, equivalent to iommu.strict=1.
2077 sp_off [Default Off]
2078 By default, super page will be supported if Intel IOMMU
2079 has the capability. With this option, super page will
2080 not be supported.
2081 sm_on
2082 Enable the Intel IOMMU scalable mode if the hardware
2083 advertises that it has support for the scalable mode
2084 translation.
2085 sm_off
2086 Disallow use of the Intel IOMMU scalable mode.
2087 tboot_noforce [Default Off]
2088 Do not force the Intel IOMMU enabled under tboot.
2089 By default, tboot will force Intel IOMMU on, which
2090 could harm performance of some high-throughput
2091 devices like 40GBit network cards, even if identity
2092 mapping is enabled.
2093 Note that using this option lowers the security
2094 provided by tboot because it makes the system
2095 vulnerable to DMA attacks.
2096
2097 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
2098 0 disables intel_idle and fall back on acpi_idle.
2099 1 to 9 specify maximum depth of C-state.
2100
2101 intel_pstate= [X86]
2102 disable
2103 Do not enable intel_pstate as the default
2104 scaling driver for the supported processors
2105 passive
2106 Use intel_pstate as a scaling driver, but configure it
2107 to work with generic cpufreq governors (instead of
2108 enabling its internal governor). This mode cannot be
2109 used along with the hardware-managed P-states (HWP)
2110 feature.
2111 force
2112 Enable intel_pstate on systems that prohibit it by default
2113 in favor of acpi-cpufreq. Forcing the intel_pstate driver
2114 instead of acpi-cpufreq may disable platform features, such
2115 as thermal controls and power capping, that rely on ACPI
2116 P-States information being indicated to OSPM and therefore
2117 should be used with caution. This option does not work with
2118 processors that aren't supported by the intel_pstate driver
2119 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2120 no_hwp
2121 Do not enable hardware P state control (HWP)
2122 if available.
2123 hwp_only
2124 Only load intel_pstate on systems which support
2125 hardware P state control (HWP) if available.
2126 support_acpi_ppc
2127 Enforce ACPI _PPC performance limits. If the Fixed ACPI
2128 Description Table, specifies preferred power management
2129 profile as "Enterprise Server" or "Performance Server",
2130 then this feature is turned on by default.
2131 per_cpu_perf_limits
2132 Allow per-logical-CPU P-State performance control limits using
2133 cpufreq sysfs interface
2134
2135 intremap= [X86-64, Intel-IOMMU]
2136 on enable Interrupt Remapping (default)
2137 off disable Interrupt Remapping
2138 nosid disable Source ID checking
2139 no_x2apic_optout
2140 BIOS x2APIC opt-out request will be ignored
2141 nopost disable Interrupt Posting
2142
2143 iomem= Disable strict checking of access to MMIO memory
2144 strict regions from userspace.
2145 relaxed
2146
2147 iommu= [X86]
2148 off
2149 force
2150 noforce
2151 biomerge
2152 panic
2153 nopanic
2154 merge
2155 nomerge
2156 soft
2157 pt [X86]
2158 nopt [X86]
2159 nobypass [PPC/POWERNV]
2160 Disable IOMMU bypass, using IOMMU for PCI devices.
2161
2162 iommu.forcedac= [ARM64, X86] Control IOVA allocation for PCI devices.
2163 Format: { "0" | "1" }
2164 0 - Try to allocate a 32-bit DMA address first, before
2165 falling back to the full range if needed.
2166 1 - Allocate directly from the full usable range,
2167 forcing Dual Address Cycle for PCI cards supporting
2168 greater than 32-bit addressing.
2169
2170 iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour
2171 Format: { "0" | "1" }
2172 0 - Lazy mode.
2173 Request that DMA unmap operations use deferred
2174 invalidation of hardware TLBs, for increased
2175 throughput at the cost of reduced device isolation.
2176 Will fall back to strict mode if not supported by
2177 the relevant IOMMU driver.
2178 1 - Strict mode.
2179 DMA unmap operations invalidate IOMMU hardware TLBs
2180 synchronously.
2181 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}.
2182 Note: on x86, strict mode specified via one of the
2183 legacy driver-specific options takes precedence.
2184
2185 iommu.passthrough=
2186 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
2187 Format: { "0" | "1" }
2188 0 - Use IOMMU translation for DMA.
2189 1 - Bypass the IOMMU for DMA.
2190 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2191
2192 io7= [HW] IO7 for Marvel-based Alpha systems
2193 See comment before marvel_specify_io7 in
2194 arch/alpha/kernel/core_marvel.c.
2195
2196 io_delay= [X86] I/O delay method
2197 0x80
2198 Standard port 0x80 based delay
2199 0xed
2200 Alternate port 0xed based delay (needed on some systems)
2201 udelay
2202 Simple two microseconds delay
2203 none
2204 No delay
2205
2206 ip= [IP_PNP]
2207 See Documentation/admin-guide/nfs/nfsroot.rst.
2208
2209 ipcmni_extend [KNL] Extend the maximum number of unique System V
2210 IPC identifiers from 32,768 to 16,777,216.
2211
2212 irqaffinity= [SMP] Set the default irq affinity mask
2213 The argument is a cpu list, as described above.
2214
2215 irqchip.gicv2_force_probe=
2216 [ARM, ARM64]
2217 Format: <bool>
2218 Force the kernel to look for the second 4kB page
2219 of a GICv2 controller even if the memory range
2220 exposed by the device tree is too small.
2221
2222 irqchip.gicv3_nolpi=
2223 [ARM, ARM64]
2224 Force the kernel to ignore the availability of
2225 LPIs (and by consequence ITSs). Intended for system
2226 that use the kernel as a bootloader, and thus want
2227 to let secondary kernels in charge of setting up
2228 LPIs.
2229
2230 irqchip.gicv3_pseudo_nmi= [ARM64]
2231 Enables support for pseudo-NMIs in the kernel. This
2232 requires the kernel to be built with
2233 CONFIG_ARM64_PSEUDO_NMI.
2234
2235 irqfixup [HW]
2236 When an interrupt is not handled search all handlers
2237 for it. Intended to get systems with badly broken
2238 firmware running.
2239
2240 irqpoll [HW]
2241 When an interrupt is not handled search all handlers
2242 for it. Also check all handlers each timer
2243 interrupt. Intended to get systems with badly broken
2244 firmware running.
2245
2246 isapnp= [ISAPNP]
2247 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2248
2249 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2250 [Deprecated - use cpusets instead]
2251 Format: [flag-list,]<cpu-list>
2252
2253 Specify one or more CPUs to isolate from disturbances
2254 specified in the flag list (default: domain):
2255
2256 nohz
2257 Disable the tick when a single task runs.
2258
2259 A residual 1Hz tick is offloaded to workqueues, which you
2260 need to affine to housekeeping through the global
2261 workqueue's affinity configured via the
2262 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2263 by using the 'domain' flag described below.
2264
2265 NOTE: by default the global workqueue runs on all CPUs,
2266 so to protect individual CPUs the 'cpumask' file has to
2267 be configured manually after bootup.
2268
2269 domain
2270 Isolate from the general SMP balancing and scheduling
2271 algorithms. Note that performing domain isolation this way
2272 is irreversible: it's not possible to bring back a CPU to
2273 the domains once isolated through isolcpus. It's strongly
2274 advised to use cpusets instead to disable scheduler load
2275 balancing through the "cpuset.sched_load_balance" file.
2276 It offers a much more flexible interface where CPUs can
2277 move in and out of an isolated set anytime.
2278
2279 You can move a process onto or off an "isolated" CPU via
2280 the CPU affinity syscalls or cpuset.
2281 <cpu number> begins at 0 and the maximum value is
2282 "number of CPUs in system - 1".
2283
2284 managed_irq
2285
2286 Isolate from being targeted by managed interrupts
2287 which have an interrupt mask containing isolated
2288 CPUs. The affinity of managed interrupts is
2289 handled by the kernel and cannot be changed via
2290 the /proc/irq/* interfaces.
2291
2292 This isolation is best effort and only effective
2293 if the automatically assigned interrupt mask of a
2294 device queue contains isolated and housekeeping
2295 CPUs. If housekeeping CPUs are online then such
2296 interrupts are directed to the housekeeping CPU
2297 so that IO submitted on the housekeeping CPU
2298 cannot disturb the isolated CPU.
2299
2300 If a queue's affinity mask contains only isolated
2301 CPUs then this parameter has no effect on the
2302 interrupt routing decision, though interrupts are
2303 only delivered when tasks running on those
2304 isolated CPUs submit IO. IO submitted on
2305 housekeeping CPUs has no influence on those
2306 queues.
2307
2308 The format of <cpu-list> is described above.
2309
2310 iucv= [HW,NET]
2311
2312 ivrs_ioapic [HW,X86-64]
2313 Provide an override to the IOAPIC-ID<->DEVICE-ID
2314 mapping provided in the IVRS ACPI table.
2315 By default, PCI segment is 0, and can be omitted.
2316
2317 For example, to map IOAPIC-ID decimal 10 to
2318 PCI segment 0x1 and PCI device 00:14.0,
2319 write the parameter as:
2320 ivrs_ioapic=10@0001:00:14.0
2321
2322 Deprecated formats:
2323 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0
2324 write the parameter as:
2325 ivrs_ioapic[10]=00:14.0
2326 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and
2327 PCI device 00:14.0 write the parameter as:
2328 ivrs_ioapic[10]=0001:00:14.0
2329
2330 ivrs_hpet [HW,X86-64]
2331 Provide an override to the HPET-ID<->DEVICE-ID
2332 mapping provided in the IVRS ACPI table.
2333 By default, PCI segment is 0, and can be omitted.
2334
2335 For example, to map HPET-ID decimal 10 to
2336 PCI segment 0x1 and PCI device 00:14.0,
2337 write the parameter as:
2338 ivrs_hpet=10@0001:00:14.0
2339
2340 Deprecated formats:
2341 * To map HPET-ID decimal 0 to PCI device 00:14.0
2342 write the parameter as:
2343 ivrs_hpet[0]=00:14.0
2344 * To map HPET-ID decimal 10 to PCI segment 0x1 and
2345 PCI device 00:14.0 write the parameter as:
2346 ivrs_ioapic[10]=0001:00:14.0
2347
2348 ivrs_acpihid [HW,X86-64]
2349 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2350 mapping provided in the IVRS ACPI table.
2351 By default, PCI segment is 0, and can be omitted.
2352
2353 For example, to map UART-HID:UID AMD0020:0 to
2354 PCI segment 0x1 and PCI device ID 00:14.5,
2355 write the parameter as:
2356 ivrs_acpihid=AMD0020:0@0001:00:14.5
2357
2358 Deprecated formats:
2359 * To map UART-HID:UID AMD0020:0 to PCI segment is 0,
2360 PCI device ID 00:14.5, write the parameter as:
2361 ivrs_acpihid[00:14.5]=AMD0020:0
2362 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and
2363 PCI device ID 00:14.5, write the parameter as:
2364 ivrs_acpihid[0001:00:14.5]=AMD0020:0
2365
2366 js= [HW,JOY] Analog joystick
2367 See Documentation/input/joydev/joystick.rst.
2368
2369 nokaslr [KNL]
2370 When CONFIG_RANDOMIZE_BASE is set, this disables
2371 kernel and module base offset ASLR (Address Space
2372 Layout Randomization).
2373
2374 kasan_multi_shot
2375 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2376 report on every invalid memory access. Without this
2377 parameter KASAN will print report only for the first
2378 invalid access.
2379
2380 keepinitrd [HW,ARM]
2381
2382 kernelcore= [KNL,X86,IA-64,PPC]
2383 Format: nn[KMGTPE] | nn% | "mirror"
2384 This parameter specifies the amount of memory usable by
2385 the kernel for non-movable allocations. The requested
2386 amount is spread evenly throughout all nodes in the
2387 system as ZONE_NORMAL. The remaining memory is used for
2388 movable memory in its own zone, ZONE_MOVABLE. In the
2389 event, a node is too small to have both ZONE_NORMAL and
2390 ZONE_MOVABLE, kernelcore memory will take priority and
2391 other nodes will have a larger ZONE_MOVABLE.
2392
2393 ZONE_MOVABLE is used for the allocation of pages that
2394 may be reclaimed or moved by the page migration
2395 subsystem. Note that allocations like PTEs-from-HighMem
2396 still use the HighMem zone if it exists, and the Normal
2397 zone if it does not.
2398
2399 It is possible to specify the exact amount of memory in
2400 the form of "nn[KMGTPE]", a percentage of total system
2401 memory in the form of "nn%", or "mirror". If "mirror"
2402 option is specified, mirrored (reliable) memory is used
2403 for non-movable allocations and remaining memory is used
2404 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2405 are exclusive, so you cannot specify multiple forms.
2406
2407 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2408 Format: <Controller#>[,poll interval]
2409 The controller # is the number of the ehci usb debug
2410 port as it is probed via PCI. The poll interval is
2411 optional and is the number seconds in between
2412 each poll cycle to the debug port in case you need
2413 the functionality for interrupting the kernel with
2414 gdb or control-c on the dbgp connection. When
2415 not using this parameter you use sysrq-g to break into
2416 the kernel debugger.
2417
2418 kgdboc= [KGDB,HW] kgdb over consoles.
2419 Requires a tty driver that supports console polling,
2420 or a supported polling keyboard driver (non-usb).
2421 Serial only format: <serial_device>[,baud]
2422 keyboard only format: kbd
2423 keyboard and serial format: kbd,<serial_device>[,baud]
2424 Optional Kernel mode setting:
2425 kms, kbd format: kms,kbd
2426 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2427
2428 kgdboc_earlycon= [KGDB,HW]
2429 If the boot console provides the ability to read
2430 characters and can work in polling mode, you can use
2431 this parameter to tell kgdb to use it as a backend
2432 until the normal console is registered. Intended to
2433 be used together with the kgdboc parameter which
2434 specifies the normal console to transition to.
2435
2436 The name of the early console should be specified
2437 as the value of this parameter. Note that the name of
2438 the early console might be different than the tty
2439 name passed to kgdboc. It's OK to leave the value
2440 blank and the first boot console that implements
2441 read() will be picked.
2442
2443 kgdbwait [KGDB] Stop kernel execution and enter the
2444 kernel debugger at the earliest opportunity.
2445
2446 kmac= [MIPS] Korina ethernet MAC address.
2447 Configure the RouterBoard 532 series on-chip
2448 Ethernet adapter MAC address.
2449
2450 kmemleak= [KNL] Boot-time kmemleak enable/disable
2451 Valid arguments: on, off
2452 Default: on
2453 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2454 the default is off.
2455
2456 kprobe_event=[probe-list]
2457 [FTRACE] Add kprobe events and enable at boot time.
2458 The probe-list is a semicolon delimited list of probe
2459 definitions. Each definition is same as kprobe_events
2460 interface, but the parameters are comma delimited.
2461 For example, to add a kprobe event on vfs_read with
2462 arg1 and arg2, add to the command line;
2463
2464 kprobe_event=p,vfs_read,$arg1,$arg2
2465
2466 See also Documentation/trace/kprobetrace.rst "Kernel
2467 Boot Parameter" section.
2468
2469 kpti= [ARM64] Control page table isolation of user
2470 and kernel address spaces.
2471 Default: enabled on cores which need mitigation.
2472 0: force disabled
2473 1: force enabled
2474
2475 kunit.enable= [KUNIT] Enable executing KUnit tests. Requires
2476 CONFIG_KUNIT to be set to be fully enabled. The
2477 default value can be overridden via
2478 KUNIT_DEFAULT_ENABLED.
2479 Default is 1 (enabled)
2480
2481 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2482 Default is 0 (don't ignore, but inject #GP)
2483
2484 kvm.eager_page_split=
2485 [KVM,X86] Controls whether or not KVM will try to
2486 proactively split all huge pages during dirty logging.
2487 Eager page splitting reduces interruptions to vCPU
2488 execution by eliminating the write-protection faults
2489 and MMU lock contention that would otherwise be
2490 required to split huge pages lazily.
2491
2492 VM workloads that rarely perform writes or that write
2493 only to a small region of VM memory may benefit from
2494 disabling eager page splitting to allow huge pages to
2495 still be used for reads.
2496
2497 The behavior of eager page splitting depends on whether
2498 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If
2499 disabled, all huge pages in a memslot will be eagerly
2500 split when dirty logging is enabled on that memslot. If
2501 enabled, eager page splitting will be performed during
2502 the KVM_CLEAR_DIRTY ioctl, and only for the pages being
2503 cleared.
2504
2505 Eager page splitting is only supported when kvm.tdp_mmu=Y.
2506
2507 Default is Y (on).
2508
2509 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2510 Default is false (don't support).
2511
2512 kvm.nx_huge_pages=
2513 [KVM] Controls the software workaround for the
2514 X86_BUG_ITLB_MULTIHIT bug.
2515 force : Always deploy workaround.
2516 off : Never deploy workaround.
2517 auto : Deploy workaround based on the presence of
2518 X86_BUG_ITLB_MULTIHIT.
2519
2520 Default is 'auto'.
2521
2522 If the software workaround is enabled for the host,
2523 guests do need not to enable it for nested guests.
2524
2525 kvm.nx_huge_pages_recovery_ratio=
2526 [KVM] Controls how many 4KiB pages are periodically zapped
2527 back to huge pages. 0 disables the recovery, otherwise if
2528 the value is N KVM will zap 1/Nth of the 4KiB pages every
2529 period (see below). The default is 60.
2530
2531 kvm.nx_huge_pages_recovery_period_ms=
2532 [KVM] Controls the time period at which KVM zaps 4KiB pages
2533 back to huge pages. If the value is a non-zero N, KVM will
2534 zap a portion (see ratio above) of the pages every N msecs.
2535 If the value is 0 (the default), KVM will pick a period based
2536 on the ratio, such that a page is zapped after 1 hour on average.
2537
2538 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2539 Default is 1 (enabled)
2540
2541 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2542 for all guests.
2543 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2544
2545 kvm-arm.mode=
2546 [KVM,ARM] Select one of KVM/arm64's modes of operation.
2547
2548 none: Forcefully disable KVM.
2549
2550 nvhe: Standard nVHE-based mode, without support for
2551 protected guests.
2552
2553 protected: nVHE-based mode with support for guests whose
2554 state is kept private from the host.
2555
2556 Defaults to VHE/nVHE based on hardware support. Setting
2557 mode to "protected" will disable kexec and hibernation
2558 for the host.
2559
2560 kvm-arm.vgic_v3_group0_trap=
2561 [KVM,ARM] Trap guest accesses to GICv3 group-0
2562 system registers
2563
2564 kvm-arm.vgic_v3_group1_trap=
2565 [KVM,ARM] Trap guest accesses to GICv3 group-1
2566 system registers
2567
2568 kvm-arm.vgic_v3_common_trap=
2569 [KVM,ARM] Trap guest accesses to GICv3 common
2570 system registers
2571
2572 kvm-arm.vgic_v4_enable=
2573 [KVM,ARM] Allow use of GICv4 for direct injection of
2574 LPIs.
2575
2576 kvm_cma_resv_ratio=n [PPC]
2577 Reserves given percentage from system memory area for
2578 contiguous memory allocation for KVM hash pagetable
2579 allocation.
2580 By default it reserves 5% of total system memory.
2581 Format: <integer>
2582 Default: 5
2583
2584 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2585 (virtualized MMU) support on capable Intel chips.
2586 Default is 1 (enabled)
2587
2588 kvm-intel.emulate_invalid_guest_state=
2589 [KVM,Intel] Disable emulation of invalid guest state.
2590 Ignored if kvm-intel.enable_unrestricted_guest=1, as
2591 guest state is never invalid for unrestricted guests.
2592 This param doesn't apply to nested guests (L2), as KVM
2593 never emulates invalid L2 guest state.
2594 Default is 1 (enabled)
2595
2596 kvm-intel.flexpriority=
2597 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2598 Default is 1 (enabled)
2599
2600 kvm-intel.nested=
2601 [KVM,Intel] Enable VMX nesting (nVMX).
2602 Default is 0 (disabled)
2603
2604 kvm-intel.unrestricted_guest=
2605 [KVM,Intel] Disable unrestricted guest feature
2606 (virtualized real and unpaged mode) on capable
2607 Intel chips. Default is 1 (enabled)
2608
2609 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2610 CVE-2018-3620.
2611
2612 Valid arguments: never, cond, always
2613
2614 always: L1D cache flush on every VMENTER.
2615 cond: Flush L1D on VMENTER only when the code between
2616 VMEXIT and VMENTER can leak host memory.
2617 never: Disables the mitigation
2618
2619 Default is cond (do L1 cache flush in specific instances)
2620
2621 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2622 feature (tagged TLBs) on capable Intel chips.
2623 Default is 1 (enabled)
2624
2625 l1d_flush= [X86,INTEL]
2626 Control mitigation for L1D based snooping vulnerability.
2627
2628 Certain CPUs are vulnerable to an exploit against CPU
2629 internal buffers which can forward information to a
2630 disclosure gadget under certain conditions.
2631
2632 In vulnerable processors, the speculatively
2633 forwarded data can be used in a cache side channel
2634 attack, to access data to which the attacker does
2635 not have direct access.
2636
2637 This parameter controls the mitigation. The
2638 options are:
2639
2640 on - enable the interface for the mitigation
2641
2642 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2643 affected CPUs
2644
2645 The kernel PTE inversion protection is unconditionally
2646 enabled and cannot be disabled.
2647
2648 full
2649 Provides all available mitigations for the
2650 L1TF vulnerability. Disables SMT and
2651 enables all mitigations in the
2652 hypervisors, i.e. unconditional L1D flush.
2653
2654 SMT control and L1D flush control via the
2655 sysfs interface is still possible after
2656 boot. Hypervisors will issue a warning
2657 when the first VM is started in a
2658 potentially insecure configuration,
2659 i.e. SMT enabled or L1D flush disabled.
2660
2661 full,force
2662 Same as 'full', but disables SMT and L1D
2663 flush runtime control. Implies the
2664 'nosmt=force' command line option.
2665 (i.e. sysfs control of SMT is disabled.)
2666
2667 flush
2668 Leaves SMT enabled and enables the default
2669 hypervisor mitigation, i.e. conditional
2670 L1D flush.
2671
2672 SMT control and L1D flush control via the
2673 sysfs interface is still possible after
2674 boot. Hypervisors will issue a warning
2675 when the first VM is started in a
2676 potentially insecure configuration,
2677 i.e. SMT enabled or L1D flush disabled.
2678
2679 flush,nosmt
2680
2681 Disables SMT and enables the default
2682 hypervisor mitigation.
2683
2684 SMT control and L1D flush control via the
2685 sysfs interface is still possible after
2686 boot. Hypervisors will issue a warning
2687 when the first VM is started in a
2688 potentially insecure configuration,
2689 i.e. SMT enabled or L1D flush disabled.
2690
2691 flush,nowarn
2692 Same as 'flush', but hypervisors will not
2693 warn when a VM is started in a potentially
2694 insecure configuration.
2695
2696 off
2697 Disables hypervisor mitigations and doesn't
2698 emit any warnings.
2699 It also drops the swap size and available
2700 RAM limit restriction on both hypervisor and
2701 bare metal.
2702
2703 Default is 'flush'.
2704
2705 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2706
2707 l2cr= [PPC]
2708
2709 l3cr= [PPC]
2710
2711 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2712 disabled it.
2713
2714 lapic= [X86,APIC] Do not use TSC deadline
2715 value for LAPIC timer one-shot implementation. Default
2716 back to the programmable timer unit in the LAPIC.
2717 Format: notscdeadline
2718
2719 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2720 in C2 power state.
2721
2722 libata.dma= [LIBATA] DMA control
2723 libata.dma=0 Disable all PATA and SATA DMA
2724 libata.dma=1 PATA and SATA Disk DMA only
2725 libata.dma=2 ATAPI (CDROM) DMA only
2726 libata.dma=4 Compact Flash DMA only
2727 Combinations also work, so libata.dma=3 enables DMA
2728 for disks and CDROMs, but not CFs.
2729
2730 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2731 libata.ignore_hpa=0 keep BIOS limits (default)
2732 libata.ignore_hpa=1 ignore limits, using full disk
2733
2734 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2735 when set.
2736 Format: <int>
2737
2738 libata.force= [LIBATA] Force configurations. The format is a comma-
2739 separated list of "[ID:]VAL" where ID is PORT[.DEVICE].
2740 PORT and DEVICE are decimal numbers matching port, link
2741 or device. Basically, it matches the ATA ID string
2742 printed on console by libata. If the whole ID part is
2743 omitted, the last PORT and DEVICE values are used. If
2744 ID hasn't been specified yet, the configuration applies
2745 to all ports, links and devices.
2746
2747 If only DEVICE is omitted, the parameter applies to
2748 the port and all links and devices behind it. DEVICE
2749 number of 0 either selects the first device or the
2750 first fan-out link behind PMP device. It does not
2751 select the host link. DEVICE number of 15 selects the
2752 host link and device attached to it.
2753
2754 The VAL specifies the configuration to force. As long
2755 as there is no ambiguity, shortcut notation is allowed.
2756 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2757 The following configurations can be forced.
2758
2759 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2760 Any ID with matching PORT is used.
2761
2762 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2763
2764 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2765 udma[/][16,25,33,44,66,100,133] notation is also
2766 allowed.
2767
2768 * nohrst, nosrst, norst: suppress hard, soft and both
2769 resets.
2770
2771 * rstonce: only attempt one reset during hot-unplug
2772 link recovery.
2773
2774 * [no]dbdelay: Enable or disable the extra 200ms delay
2775 before debouncing a link PHY and device presence
2776 detection.
2777
2778 * [no]ncq: Turn on or off NCQ.
2779
2780 * [no]ncqtrim: Enable or disable queued DSM TRIM.
2781
2782 * [no]ncqati: Enable or disable NCQ trim on ATI chipset.
2783
2784 * [no]trim: Enable or disable (unqueued) TRIM.
2785
2786 * trim_zero: Indicate that TRIM command zeroes data.
2787
2788 * max_trim_128m: Set 128M maximum trim size limit.
2789
2790 * [no]dma: Turn on or off DMA transfers.
2791
2792 * atapi_dmadir: Enable ATAPI DMADIR bridge support.
2793
2794 * atapi_mod16_dma: Enable the use of ATAPI DMA for
2795 commands that are not a multiple of 16 bytes.
2796
2797 * [no]dmalog: Enable or disable the use of the
2798 READ LOG DMA EXT command to access logs.
2799
2800 * [no]iddevlog: Enable or disable access to the
2801 identify device data log.
2802
2803 * [no]logdir: Enable or disable access to the general
2804 purpose log directory.
2805
2806 * max_sec_128: Set transfer size limit to 128 sectors.
2807
2808 * max_sec_1024: Set or clear transfer size limit to
2809 1024 sectors.
2810
2811 * max_sec_lba48: Set or clear transfer size limit to
2812 65535 sectors.
2813
2814 * [no]lpm: Enable or disable link power management.
2815
2816 * [no]setxfer: Indicate if transfer speed mode setting
2817 should be skipped.
2818
2819 * dump_id: Dump IDENTIFY data.
2820
2821 * disable: Disable this device.
2822
2823 If there are multiple matching configurations changing
2824 the same attribute, the last one is used.
2825
2826 load_ramdisk= [RAM] [Deprecated]
2827
2828 lockd.nlm_grace_period=P [NFS] Assign grace period.
2829 Format: <integer>
2830
2831 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2832 Format: <integer>
2833
2834 lockd.nlm_timeout=T [NFS] Assign timeout value.
2835 Format: <integer>
2836
2837 lockd.nlm_udpport=M [NFS] Assign UDP port.
2838 Format: <integer>
2839
2840 lockdown= [SECURITY]
2841 { integrity | confidentiality }
2842 Enable the kernel lockdown feature. If set to
2843 integrity, kernel features that allow userland to
2844 modify the running kernel are disabled. If set to
2845 confidentiality, kernel features that allow userland
2846 to extract confidential information from the kernel
2847 are also disabled.
2848
2849 locktorture.nreaders_stress= [KNL]
2850 Set the number of locking read-acquisition kthreads.
2851 Defaults to being automatically set based on the
2852 number of online CPUs.
2853
2854 locktorture.nwriters_stress= [KNL]
2855 Set the number of locking write-acquisition kthreads.
2856
2857 locktorture.onoff_holdoff= [KNL]
2858 Set time (s) after boot for CPU-hotplug testing.
2859
2860 locktorture.onoff_interval= [KNL]
2861 Set time (s) between CPU-hotplug operations, or
2862 zero to disable CPU-hotplug testing.
2863
2864 locktorture.shuffle_interval= [KNL]
2865 Set task-shuffle interval (jiffies). Shuffling
2866 tasks allows some CPUs to go into dyntick-idle
2867 mode during the locktorture test.
2868
2869 locktorture.shutdown_secs= [KNL]
2870 Set time (s) after boot system shutdown. This
2871 is useful for hands-off automated testing.
2872
2873 locktorture.stat_interval= [KNL]
2874 Time (s) between statistics printk()s.
2875
2876 locktorture.stutter= [KNL]
2877 Time (s) to stutter testing, for example,
2878 specifying five seconds causes the test to run for
2879 five seconds, wait for five seconds, and so on.
2880 This tests the locking primitive's ability to
2881 transition abruptly to and from idle.
2882
2883 locktorture.torture_type= [KNL]
2884 Specify the locking implementation to test.
2885
2886 locktorture.verbose= [KNL]
2887 Enable additional printk() statements.
2888
2889 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2890 Format: <irq>
2891
2892 loglevel= All Kernel Messages with a loglevel smaller than the
2893 console loglevel will be printed to the console. It can
2894 also be changed with klogd or other programs. The
2895 loglevels are defined as follows:
2896
2897 0 (KERN_EMERG) system is unusable
2898 1 (KERN_ALERT) action must be taken immediately
2899 2 (KERN_CRIT) critical conditions
2900 3 (KERN_ERR) error conditions
2901 4 (KERN_WARNING) warning conditions
2902 5 (KERN_NOTICE) normal but significant condition
2903 6 (KERN_INFO) informational
2904 7 (KERN_DEBUG) debug-level messages
2905
2906 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2907 in bytes. n must be a power of two and greater
2908 than the minimal size. The minimal size is defined
2909 by LOG_BUF_SHIFT kernel config parameter. There is
2910 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2911 that allows to increase the default size depending on
2912 the number of CPUs. See init/Kconfig for more details.
2913
2914 logo.nologo [FB] Disables display of the built-in Linux logo.
2915 This may be used to provide more screen space for
2916 kernel log messages and is useful when debugging
2917 kernel boot problems.
2918
2919 lp=0 [LP] Specify parallel ports to use, e.g,
2920 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2921 lp=reset first parallel port). 'lp=0' disables the
2922 lp=auto printer driver. 'lp=reset' (which can be
2923 specified in addition to the ports) causes
2924 attached printers to be reset. Using
2925 lp=port1,port2,... specifies the parallel ports
2926 to associate lp devices with, starting with
2927 lp0. A port specification may be 'none' to skip
2928 that lp device, or a parport name such as
2929 'parport0'. Specifying 'lp=auto' instead of a
2930 port specification list means that device IDs
2931 from each port should be examined, to see if
2932 an IEEE 1284-compliant printer is attached; if
2933 so, the driver will manage that printer.
2934 See also header of drivers/char/lp.c.
2935
2936 lpj=n [KNL]
2937 Sets loops_per_jiffy to given constant, thus avoiding
2938 time-consuming boot-time autodetection (up to 250 ms per
2939 CPU). 0 enables autodetection (default). To determine
2940 the correct value for your kernel, boot with normal
2941 autodetection and see what value is printed. Note that
2942 on SMP systems the preset will be applied to all CPUs,
2943 which is likely to cause problems if your CPUs need
2944 significantly divergent settings. An incorrect value
2945 will cause delays in the kernel to be wrong, leading to
2946 unpredictable I/O errors and other breakage. Although
2947 unlikely, in the extreme case this might damage your
2948 hardware.
2949
2950 ltpc= [NET]
2951 Format: <io>,<irq>,<dma>
2952
2953 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2954
2955 lsm=lsm1,...,lsmN
2956 [SECURITY] Choose order of LSM initialization. This
2957 overrides CONFIG_LSM, and the "security=" parameter.
2958
2959 machvec= [IA-64] Force the use of a particular machine-vector
2960 (machvec) in a generic kernel.
2961 Example: machvec=hpzx1
2962
2963 machtype= [Loongson] Share the same kernel image file between
2964 different yeeloong laptops.
2965 Example: machtype=lemote-yeeloong-2f-7inch
2966
2967 max_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory greater
2968 than or equal to this physical address is ignored.
2969
2970 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2971 will bring up during bootup. maxcpus=n : n >= 0 limits
2972 the kernel to bring up 'n' processors. Surely after
2973 bootup you can bring up the other plugged cpu by executing
2974 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2975 only takes effect during system bootup.
2976 While n=0 is a special case, it is equivalent to "nosmp",
2977 which also disables the IO APIC.
2978
2979 max_loop= [LOOP] The number of loop block devices that get
2980 (loop.max_loop) unconditionally pre-created at init time. The default
2981 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2982 of statically allocating a predefined number, loop
2983 devices can be requested on-demand with the
2984 /dev/loop-control interface.
2985
2986 mce [X86-32] Machine Check Exception
2987
2988 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2989
2990 md= [HW] RAID subsystems devices and level
2991 See Documentation/admin-guide/md.rst.
2992
2993 mdacon= [MDA]
2994 Format: <first>,<last>
2995 Specifies range of consoles to be captured by the MDA.
2996
2997 mds= [X86,INTEL]
2998 Control mitigation for the Micro-architectural Data
2999 Sampling (MDS) vulnerability.
3000
3001 Certain CPUs are vulnerable to an exploit against CPU
3002 internal buffers which can forward information to a
3003 disclosure gadget under certain conditions.
3004
3005 In vulnerable processors, the speculatively
3006 forwarded data can be used in a cache side channel
3007 attack, to access data to which the attacker does
3008 not have direct access.
3009
3010 This parameter controls the MDS mitigation. The
3011 options are:
3012
3013 full - Enable MDS mitigation on vulnerable CPUs
3014 full,nosmt - Enable MDS mitigation and disable
3015 SMT on vulnerable CPUs
3016 off - Unconditionally disable MDS mitigation
3017
3018 On TAA-affected machines, mds=off can be prevented by
3019 an active TAA mitigation as both vulnerabilities are
3020 mitigated with the same mechanism so in order to disable
3021 this mitigation, you need to specify tsx_async_abort=off
3022 too.
3023
3024 Not specifying this option is equivalent to
3025 mds=full.
3026
3027 For details see: Documentation/admin-guide/hw-vuln/mds.rst
3028
3029 mem=nn[KMG] [HEXAGON] Set the memory size.
3030 Must be specified, otherwise memory size will be 0.
3031
3032 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
3033 Amount of memory to be used in cases as follows:
3034
3035 1 for test;
3036 2 when the kernel is not able to see the whole system memory;
3037 3 memory that lies after 'mem=' boundary is excluded from
3038 the hypervisor, then assigned to KVM guests.
3039 4 to limit the memory available for kdump kernel.
3040
3041 [ARC,MICROBLAZE] - the limit applies only to low memory,
3042 high memory is not affected.
3043
3044 [ARM64] - only limits memory covered by the linear
3045 mapping. The NOMAP regions are not affected.
3046
3047 [X86] Work as limiting max address. Use together
3048 with memmap= to avoid physical address space collisions.
3049 Without memmap= PCI devices could be placed at addresses
3050 belonging to unused RAM.
3051
3052 Note that this only takes effects during boot time since
3053 in above case 3, memory may need be hot added after boot
3054 if system memory of hypervisor is not sufficient.
3055
3056 mem=nn[KMG]@ss[KMG]
3057 [ARM,MIPS] - override the memory layout reported by
3058 firmware.
3059 Define a memory region of size nn[KMG] starting at
3060 ss[KMG].
3061 Multiple different regions can be specified with
3062 multiple mem= parameters on the command line.
3063
3064 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
3065 memory.
3066
3067 memblock=debug [KNL] Enable memblock debug messages.
3068
3069 memchunk=nn[KMG]
3070 [KNL,SH] Allow user to override the default size for
3071 per-device physically contiguous DMA buffers.
3072
3073 memhp_default_state=online/offline
3074 [KNL] Set the initial state for the memory hotplug
3075 onlining policy. If not specified, the default value is
3076 set according to the
3077 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
3078 option.
3079 See Documentation/admin-guide/mm/memory-hotplug.rst.
3080
3081 memmap=exactmap [KNL,X86] Enable setting of an exact
3082 E820 memory map, as specified by the user.
3083 Such memmap=exactmap lines can be constructed based on
3084 BIOS output or other requirements. See the memmap=nn@ss
3085 option description.
3086
3087 memmap=nn[KMG]@ss[KMG]
3088 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory.
3089 Region of memory to be used is from ss to ss+nn.
3090 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
3091 which limits max address to nn[KMG].
3092 Multiple different regions can be specified,
3093 comma delimited.
3094 Example:
3095 memmap=100M@2G,100M#3G,1G!1024G
3096
3097 memmap=nn[KMG]#ss[KMG]
3098 [KNL,ACPI] Mark specific memory as ACPI data.
3099 Region of memory to be marked is from ss to ss+nn.
3100
3101 memmap=nn[KMG]$ss[KMG]
3102 [KNL,ACPI] Mark specific memory as reserved.
3103 Region of memory to be reserved is from ss to ss+nn.
3104 Example: Exclude memory from 0x18690000-0x1869ffff
3105 memmap=64K$0x18690000
3106 or
3107 memmap=0x10000$0x18690000
3108 Some bootloaders may need an escape character before '$',
3109 like Grub2, otherwise '$' and the following number
3110 will be eaten.
3111
3112 memmap=nn[KMG]!ss[KMG]
3113 [KNL,X86] Mark specific memory as protected.
3114 Region of memory to be used, from ss to ss+nn.
3115 The memory region may be marked as e820 type 12 (0xc)
3116 and is NVDIMM or ADR memory.
3117
3118 memmap=<size>%<offset>-<oldtype>+<newtype>
3119 [KNL,ACPI] Convert memory within the specified region
3120 from <oldtype> to <newtype>. If "-<oldtype>" is left
3121 out, the whole region will be marked as <newtype>,
3122 even if previously unavailable. If "+<newtype>" is left
3123 out, matching memory will be removed. Types are
3124 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
3125 3 = ACPI, 12 = PRAM.
3126
3127 memory_corruption_check=0/1 [X86]
3128 Some BIOSes seem to corrupt the first 64k of
3129 memory when doing things like suspend/resume.
3130 Setting this option will scan the memory
3131 looking for corruption. Enabling this will
3132 both detect corruption and prevent the kernel
3133 from using the memory being corrupted.
3134 However, its intended as a diagnostic tool; if
3135 repeatable BIOS-originated corruption always
3136 affects the same memory, you can use memmap=
3137 to prevent the kernel from using that memory.
3138
3139 memory_corruption_check_size=size [X86]
3140 By default it checks for corruption in the low
3141 64k, making this memory unavailable for normal
3142 use. Use this parameter to scan for
3143 corruption in more or less memory.
3144
3145 memory_corruption_check_period=seconds [X86]
3146 By default it checks for corruption every 60
3147 seconds. Use this parameter to check at some
3148 other rate. 0 disables periodic checking.
3149
3150 memory_hotplug.memmap_on_memory
3151 [KNL,X86,ARM] Boolean flag to enable this feature.
3152 Format: {on | off (default)}
3153 When enabled, runtime hotplugged memory will
3154 allocate its internal metadata (struct pages,
3155 those vmemmap pages cannot be optimized even
3156 if hugetlb_free_vmemmap is enabled) from the
3157 hotadded memory which will allow to hotadd a
3158 lot of memory without requiring additional
3159 memory to do so.
3160 This feature is disabled by default because it
3161 has some implication on large (e.g. GB)
3162 allocations in some configurations (e.g. small
3163 memory blocks).
3164 The state of the flag can be read in
3165 /sys/module/memory_hotplug/parameters/memmap_on_memory.
3166 Note that even when enabled, there are a few cases where
3167 the feature is not effective.
3168
3169 memtest= [KNL,X86,ARM,M68K,PPC,RISCV] Enable memtest
3170 Format: <integer>
3171 default : 0 <disable>
3172 Specifies the number of memtest passes to be
3173 performed. Each pass selects another test
3174 pattern from a given set of patterns. Memtest
3175 fills the memory with this pattern, validates
3176 memory contents and reserves bad memory
3177 regions that are detected.
3178
3179 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
3180 Valid arguments: on, off
3181 Default (depends on kernel configuration option):
3182 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
3183 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
3184 mem_encrypt=on: Activate SME
3185 mem_encrypt=off: Do not activate SME
3186
3187 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst
3188 for details on when memory encryption can be activated.
3189
3190 mem_sleep_default= [SUSPEND] Default system suspend mode:
3191 s2idle - Suspend-To-Idle
3192 shallow - Power-On Suspend or equivalent (if supported)
3193 deep - Suspend-To-RAM or equivalent (if supported)
3194 See Documentation/admin-guide/pm/sleep-states.rst.
3195
3196 meye.*= [HW] Set MotionEye Camera parameters
3197 See Documentation/admin-guide/media/meye.rst.
3198
3199 mfgpt_irq= [IA-32] Specify the IRQ to use for the
3200 Multi-Function General Purpose Timers on AMD Geode
3201 platforms.
3202
3203 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
3204 the BIOS has incorrectly applied a workaround. TinyBIOS
3205 version 0.98 is known to be affected, 0.99 fixes the
3206 problem by letting the user disable the workaround.
3207
3208 mga= [HW,DRM]
3209
3210 min_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory below this
3211 physical address is ignored.
3212
3213 mini2440= [ARM,HW,KNL]
3214 Format:[0..2][b][c][t]
3215 Default: "0tb"
3216 MINI2440 configuration specification:
3217 0 - The attached screen is the 3.5" TFT
3218 1 - The attached screen is the 7" TFT
3219 2 - The VGA Shield is attached (1024x768)
3220 Leaving out the screen size parameter will not load
3221 the TFT driver, and the framebuffer will be left
3222 unconfigured.
3223 b - Enable backlight. The TFT backlight pin will be
3224 linked to the kernel VESA blanking code and a GPIO
3225 LED. This parameter is not necessary when using the
3226 VGA shield.
3227 c - Enable the s3c camera interface.
3228 t - Reserved for enabling touchscreen support. The
3229 touchscreen support is not enabled in the mainstream
3230 kernel as of 2.6.30, a preliminary port can be found
3231 in the "bleeding edge" mini2440 support kernel at
3232 https://repo.or.cz/w/linux-2.6/mini2440.git
3233
3234 mitigations=
3235 [X86,PPC,S390,ARM64] Control optional mitigations for
3236 CPU vulnerabilities. This is a set of curated,
3237 arch-independent options, each of which is an
3238 aggregation of existing arch-specific options.
3239
3240 off
3241 Disable all optional CPU mitigations. This
3242 improves system performance, but it may also
3243 expose users to several CPU vulnerabilities.
3244 Equivalent to: nopti [X86,PPC]
3245 if nokaslr then kpti=0 [ARM64]
3246 nospectre_v1 [X86,PPC]
3247 nobp=0 [S390]
3248 nospectre_v2 [X86,PPC,S390,ARM64]
3249 spectre_v2_user=off [X86]
3250 spec_store_bypass_disable=off [X86,PPC]
3251 ssbd=force-off [ARM64]
3252 nospectre_bhb [ARM64]
3253 l1tf=off [X86]
3254 mds=off [X86]
3255 tsx_async_abort=off [X86]
3256 kvm.nx_huge_pages=off [X86]
3257 srbds=off [X86,INTEL]
3258 no_entry_flush [PPC]
3259 no_uaccess_flush [PPC]
3260 mmio_stale_data=off [X86]
3261 retbleed=off [X86]
3262
3263 Exceptions:
3264 This does not have any effect on
3265 kvm.nx_huge_pages when
3266 kvm.nx_huge_pages=force.
3267
3268 auto (default)
3269 Mitigate all CPU vulnerabilities, but leave SMT
3270 enabled, even if it's vulnerable. This is for
3271 users who don't want to be surprised by SMT
3272 getting disabled across kernel upgrades, or who
3273 have other ways of avoiding SMT-based attacks.
3274 Equivalent to: (default behavior)
3275
3276 auto,nosmt
3277 Mitigate all CPU vulnerabilities, disabling SMT
3278 if needed. This is for users who always want to
3279 be fully mitigated, even if it means losing SMT.
3280 Equivalent to: l1tf=flush,nosmt [X86]
3281 mds=full,nosmt [X86]
3282 tsx_async_abort=full,nosmt [X86]
3283 mmio_stale_data=full,nosmt [X86]
3284 retbleed=auto,nosmt [X86]
3285
3286 mminit_loglevel=
3287 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
3288 parameter allows control of the logging verbosity for
3289 the additional memory initialisation checks. A value
3290 of 0 disables mminit logging and a level of 4 will
3291 log everything. Information is printed at KERN_DEBUG
3292 so loglevel=8 may also need to be specified.
3293
3294 mmio_stale_data=
3295 [X86,INTEL] Control mitigation for the Processor
3296 MMIO Stale Data vulnerabilities.
3297
3298 Processor MMIO Stale Data is a class of
3299 vulnerabilities that may expose data after an MMIO
3300 operation. Exposed data could originate or end in
3301 the same CPU buffers as affected by MDS and TAA.
3302 Therefore, similar to MDS and TAA, the mitigation
3303 is to clear the affected CPU buffers.
3304
3305 This parameter controls the mitigation. The
3306 options are:
3307
3308 full - Enable mitigation on vulnerable CPUs
3309
3310 full,nosmt - Enable mitigation and disable SMT on
3311 vulnerable CPUs.
3312
3313 off - Unconditionally disable mitigation
3314
3315 On MDS or TAA affected machines,
3316 mmio_stale_data=off can be prevented by an active
3317 MDS or TAA mitigation as these vulnerabilities are
3318 mitigated with the same mechanism so in order to
3319 disable this mitigation, you need to specify
3320 mds=off and tsx_async_abort=off too.
3321
3322 Not specifying this option is equivalent to
3323 mmio_stale_data=full.
3324
3325 For details see:
3326 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst
3327
3328 module.async_probe=<bool>
3329 [KNL] When set to true, modules will use async probing
3330 by default. To enable/disable async probing for a
3331 specific module, use the module specific control that
3332 is documented under <module>.async_probe. When both
3333 module.async_probe and <module>.async_probe are
3334 specified, <module>.async_probe takes precedence for
3335 the specific module.
3336
3337 module.sig_enforce
3338 [KNL] When CONFIG_MODULE_SIG is set, this means that
3339 modules without (valid) signatures will fail to load.
3340 Note that if CONFIG_MODULE_SIG_FORCE is set, that
3341 is always true, so this option does nothing.
3342
3343 module_blacklist= [KNL] Do not load a comma-separated list of
3344 modules. Useful for debugging problem modules.
3345
3346 mousedev.tap_time=
3347 [MOUSE] Maximum time between finger touching and
3348 leaving touchpad surface for touch to be considered
3349 a tap and be reported as a left button click (for
3350 touchpads working in absolute mode only).
3351 Format: <msecs>
3352 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
3353 reporting absolute coordinates, such as tablets
3354 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
3355 reporting absolute coordinates, such as tablets
3356
3357 movablecore= [KNL,X86,IA-64,PPC]
3358 Format: nn[KMGTPE] | nn%
3359 This parameter is the complement to kernelcore=, it
3360 specifies the amount of memory used for migratable
3361 allocations. If both kernelcore and movablecore is
3362 specified, then kernelcore will be at *least* the
3363 specified value but may be more. If movablecore on its
3364 own is specified, the administrator must be careful
3365 that the amount of memory usable for all allocations
3366 is not too small.
3367
3368 movable_node [KNL] Boot-time switch to make hotplugable memory
3369 NUMA nodes to be movable. This means that the memory
3370 of such nodes will be usable only for movable
3371 allocations which rules out almost all kernel
3372 allocations. Use with caution!
3373
3374 MTD_Partition= [MTD]
3375 Format: <name>,<region-number>,<size>,<offset>
3376
3377 MTD_Region= [MTD] Format:
3378 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
3379
3380 mtdparts= [MTD]
3381 See drivers/mtd/parsers/cmdlinepart.c
3382
3383 mtdset= [ARM]
3384 ARM/S3C2412 JIVE boot control
3385
3386 See arch/arm/mach-s3c/mach-jive.c
3387
3388 mtouchusb.raw_coordinates=
3389 [HW] Make the MicroTouch USB driver use raw coordinates
3390 ('y', default) or cooked coordinates ('n')
3391
3392 mtrr_chunk_size=nn[KMG] [X86]
3393 used for mtrr cleanup. It is largest continuous chunk
3394 that could hold holes aka. UC entries.
3395
3396 mtrr_gran_size=nn[KMG] [X86]
3397 Used for mtrr cleanup. It is granularity of mtrr block.
3398 Default is 1.
3399 Large value could prevent small alignment from
3400 using up MTRRs.
3401
3402 mtrr_spare_reg_nr=n [X86]
3403 Format: <integer>
3404 Range: 0,7 : spare reg number
3405 Default : 1
3406 Used for mtrr cleanup. It is spare mtrr entries number.
3407 Set to 2 or more if your graphical card needs more.
3408
3409 multitce=off [PPC] This parameter disables the use of the pSeries
3410 firmware feature for updating multiple TCE entries
3411 at a time.
3412
3413 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
3414
3415 netdev= [NET] Network devices parameters
3416 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
3417 Note that mem_start is often overloaded to mean
3418 something different and driver-specific.
3419 This usage is only documented in each driver source
3420 file if at all.
3421
3422 netpoll.carrier_timeout=
3423 [NET] Specifies amount of time (in seconds) that
3424 netpoll should wait for a carrier. By default netpoll
3425 waits 4 seconds.
3426
3427 nf_conntrack.acct=
3428 [NETFILTER] Enable connection tracking flow accounting
3429 0 to disable accounting
3430 1 to enable accounting
3431 Default value is 0.
3432
3433 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3434 See Documentation/admin-guide/nfs/nfsroot.rst.
3435
3436 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3437 See Documentation/admin-guide/nfs/nfsroot.rst.
3438
3439 nfsrootdebug [NFS] enable nfsroot debugging messages.
3440 See Documentation/admin-guide/nfs/nfsroot.rst.
3441
3442 nfs.callback_nr_threads=
3443 [NFSv4] set the total number of threads that the
3444 NFS client will assign to service NFSv4 callback
3445 requests.
3446
3447 nfs.callback_tcpport=
3448 [NFS] set the TCP port on which the NFSv4 callback
3449 channel should listen.
3450
3451 nfs.cache_getent=
3452 [NFS] sets the pathname to the program which is used
3453 to update the NFS client cache entries.
3454
3455 nfs.cache_getent_timeout=
3456 [NFS] sets the timeout after which an attempt to
3457 update a cache entry is deemed to have failed.
3458
3459 nfs.idmap_cache_timeout=
3460 [NFS] set the maximum lifetime for idmapper cache
3461 entries.
3462
3463 nfs.enable_ino64=
3464 [NFS] enable 64-bit inode numbers.
3465 If zero, the NFS client will fake up a 32-bit inode
3466 number for the readdir() and stat() syscalls instead
3467 of returning the full 64-bit number.
3468 The default is to return 64-bit inode numbers.
3469
3470 nfs.max_session_cb_slots=
3471 [NFSv4.1] Sets the maximum number of session
3472 slots the client will assign to the callback
3473 channel. This determines the maximum number of
3474 callbacks the client will process in parallel for
3475 a particular server.
3476
3477 nfs.max_session_slots=
3478 [NFSv4.1] Sets the maximum number of session slots
3479 the client will attempt to negotiate with the server.
3480 This limits the number of simultaneous RPC requests
3481 that the client can send to the NFSv4.1 server.
3482 Note that there is little point in setting this
3483 value higher than the max_tcp_slot_table_limit.
3484
3485 nfs.nfs4_disable_idmapping=
3486 [NFSv4] When set to the default of '1', this option
3487 ensures that both the RPC level authentication
3488 scheme and the NFS level operations agree to use
3489 numeric uids/gids if the mount is using the
3490 'sec=sys' security flavour. In effect it is
3491 disabling idmapping, which can make migration from
3492 legacy NFSv2/v3 systems to NFSv4 easier.
3493 Servers that do not support this mode of operation
3494 will be autodetected by the client, and it will fall
3495 back to using the idmapper.
3496 To turn off this behaviour, set the value to '0'.
3497 nfs.nfs4_unique_id=
3498 [NFS4] Specify an additional fixed unique ident-
3499 ification string that NFSv4 clients can insert into
3500 their nfs_client_id4 string. This is typically a
3501 UUID that is generated at system install time.
3502
3503 nfs.send_implementation_id =
3504 [NFSv4.1] Send client implementation identification
3505 information in exchange_id requests.
3506 If zero, no implementation identification information
3507 will be sent.
3508 The default is to send the implementation identification
3509 information.
3510
3511 nfs.recover_lost_locks =
3512 [NFSv4] Attempt to recover locks that were lost due
3513 to a lease timeout on the server. Please note that
3514 doing this risks data corruption, since there are
3515 no guarantees that the file will remain unchanged
3516 after the locks are lost.
3517 If you want to enable the kernel legacy behaviour of
3518 attempting to recover these locks, then set this
3519 parameter to '1'.
3520 The default parameter value of '0' causes the kernel
3521 not to attempt recovery of lost locks.
3522
3523 nfs4.layoutstats_timer =
3524 [NFSv4.2] Change the rate at which the kernel sends
3525 layoutstats to the pNFS metadata server.
3526
3527 Setting this to value to 0 causes the kernel to use
3528 whatever value is the default set by the layout
3529 driver. A non-zero value sets the minimum interval
3530 in seconds between layoutstats transmissions.
3531
3532 nfsd.inter_copy_offload_enable =
3533 [NFSv4.2] When set to 1, the server will support
3534 server-to-server copies for which this server is
3535 the destination of the copy.
3536
3537 nfsd.nfsd4_ssc_umount_timeout =
3538 [NFSv4.2] When used as the destination of a
3539 server-to-server copy, knfsd temporarily mounts
3540 the source server. It caches the mount in case
3541 it will be needed again, and discards it if not
3542 used for the number of milliseconds specified by
3543 this parameter.
3544
3545 nfsd.nfs4_disable_idmapping=
3546 [NFSv4] When set to the default of '1', the NFSv4
3547 server will return only numeric uids and gids to
3548 clients using auth_sys, and will accept numeric uids
3549 and gids from such clients. This is intended to ease
3550 migration from NFSv2/v3.
3551
3552
3553 nmi_backtrace.backtrace_idle [KNL]
3554 Dump stacks even of idle CPUs in response to an
3555 NMI stack-backtrace request.
3556
3557 nmi_debug= [KNL,SH] Specify one or more actions to take
3558 when a NMI is triggered.
3559 Format: [state][,regs][,debounce][,die]
3560
3561 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3562 Format: [panic,][nopanic,][num]
3563 Valid num: 0 or 1
3564 0 - turn hardlockup detector in nmi_watchdog off
3565 1 - turn hardlockup detector in nmi_watchdog on
3566 When panic is specified, panic when an NMI watchdog
3567 timeout occurs (or 'nopanic' to not panic on an NMI
3568 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3569 To disable both hard and soft lockup detectors,
3570 please see 'nowatchdog'.
3571 This is useful when you use a panic=... timeout and
3572 need the box quickly up again.
3573
3574 These settings can be accessed at runtime via
3575 the nmi_watchdog and hardlockup_panic sysctls.
3576
3577 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3578 emulation library even if a 387 maths coprocessor
3579 is present.
3580
3581 no5lvl [X86-64] Disable 5-level paging mode. Forces
3582 kernel to use 4-level paging instead.
3583
3584 nofsgsbase [X86] Disables FSGSBASE instructions.
3585
3586 no_console_suspend
3587 [HW] Never suspend the console
3588 Disable suspending of consoles during suspend and
3589 hibernate operations. Once disabled, debugging
3590 messages can reach various consoles while the rest
3591 of the system is being put to sleep (ie, while
3592 debugging driver suspend/resume hooks). This may
3593 not work reliably with all consoles, but is known
3594 to work with serial and VGA consoles.
3595 To facilitate more flexible debugging, we also add
3596 console_suspend, a printk module parameter to control
3597 it. Users could use console_suspend (usually
3598 /sys/module/printk/parameters/console_suspend) to
3599 turn on/off it dynamically.
3600
3601 novmcoredd [KNL,KDUMP]
3602 Disable device dump. Device dump allows drivers to
3603 append dump data to vmcore so you can collect driver
3604 specified debug info. Drivers can append the data
3605 without any limit and this data is stored in memory,
3606 so this may cause significant memory stress. Disabling
3607 device dump can help save memory but the driver debug
3608 data will be no longer available. This parameter
3609 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3610 is set.
3611
3612 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3613 caches in the slab allocator. Saves per-node memory,
3614 but will impact performance.
3615
3616 noalign [KNL,ARM]
3617
3618 noaltinstr [S390] Disables alternative instructions patching
3619 (CPU alternatives feature).
3620
3621 noapic [SMP,APIC] Tells the kernel to not make use of any
3622 IOAPICs that may be present in the system.
3623
3624 noautogroup Disable scheduler automatic task group creation.
3625
3626 nocache [ARM]
3627
3628 nodsp [SH] Disable hardware DSP at boot time.
3629
3630 noefi Disable EFI runtime services support.
3631
3632 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3633
3634 noexec [IA-64]
3635
3636 nosmap [PPC]
3637 Disable SMAP (Supervisor Mode Access Prevention)
3638 even if it is supported by processor.
3639
3640 nosmep [PPC64s]
3641 Disable SMEP (Supervisor Mode Execution Prevention)
3642 even if it is supported by processor.
3643
3644 noexec32 [X86-64]
3645 This affects only 32-bit executables.
3646 noexec32=on: enable non-executable mappings (default)
3647 read doesn't imply executable mappings
3648 noexec32=off: disable non-executable mappings
3649 read implies executable mappings
3650
3651 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3652
3653 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3654 register save and restore. The kernel will only save
3655 legacy floating-point registers on task switch.
3656
3657 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings.
3658
3659 nohugevmalloc [KNL,X86,PPC,ARM64] Disable kernel huge vmalloc mappings.
3660
3661 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3662 Equivalent to smt=1.
3663
3664 [KNL,X86] Disable symmetric multithreading (SMT).
3665 nosmt=force: Force disable SMT, cannot be undone
3666 via the sysfs control file.
3667
3668 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3669 (bounds check bypass). With this option data leaks are
3670 possible in the system.
3671
3672 nospectre_v2 [X86,PPC_E500,ARM64] Disable all mitigations for
3673 the Spectre variant 2 (indirect branch prediction)
3674 vulnerability. System may allow data leaks with this
3675 option.
3676
3677 nospectre_bhb [ARM64] Disable all mitigations for Spectre-BHB (branch
3678 history injection) vulnerability. System may allow data leaks
3679 with this option.
3680
3681 nospec_store_bypass_disable
3682 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3683
3684 no_uaccess_flush
3685 [PPC] Don't flush the L1-D cache after accessing user data.
3686
3687 noxsave [BUGS=X86] Disables x86 extended register state save
3688 and restore using xsave. The kernel will fallback to
3689 enabling legacy floating-point and sse state.
3690
3691 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3692 register states. The kernel will fall back to use
3693 xsave to save the states. By using this parameter,
3694 performance of saving the states is degraded because
3695 xsave doesn't support modified optimization while
3696 xsaveopt supports it on xsaveopt enabled systems.
3697
3698 noxsaves [X86] Disables xsaves and xrstors used in saving and
3699 restoring x86 extended register state in compacted
3700 form of xsave area. The kernel will fall back to use
3701 xsaveopt and xrstor to save and restore the states
3702 in standard form of xsave area. By using this
3703 parameter, xsave area per process might occupy more
3704 memory on xsaves enabled systems.
3705
3706 nohlt [ARM,ARM64,MICROBLAZE,SH] Forces the kernel to busy wait
3707 in do_idle() and not use the arch_cpu_idle()
3708 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3709 to be effective. This is useful on platforms where the
3710 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3711 correctly or when doing power measurements to evalute
3712 the impact of the sleep instructions. This is also
3713 useful when using JTAG debugger.
3714
3715 no_file_caps Tells the kernel not to honor file capabilities. The
3716 only way then for a file to be executed with privilege
3717 is to be setuid root or executed by root.
3718
3719 nohalt [IA-64] Tells the kernel not to use the power saving
3720 function PAL_HALT_LIGHT when idle. This increases
3721 power-consumption. On the positive side, it reduces
3722 interrupt wake-up latency, which may improve performance
3723 in certain environments such as networked servers or
3724 real-time systems.
3725
3726 no_hash_pointers
3727 Force pointers printed to the console or buffers to be
3728 unhashed. By default, when a pointer is printed via %p
3729 format string, that pointer is "hashed", i.e. obscured
3730 by hashing the pointer value. This is a security feature
3731 that hides actual kernel addresses from unprivileged
3732 users, but it also makes debugging the kernel more
3733 difficult since unequal pointers can no longer be
3734 compared. However, if this command-line option is
3735 specified, then all normal pointers will have their true
3736 value printed. This option should only be specified when
3737 debugging the kernel. Please do not use on production
3738 kernels.
3739
3740 nohibernate [HIBERNATION] Disable hibernation and resume.
3741
3742 nohz= [KNL] Boottime enable/disable dynamic ticks
3743 Valid arguments: on, off
3744 Default: on
3745
3746 nohz_full= [KNL,BOOT,SMP,ISOL]
3747 The argument is a cpu list, as described above.
3748 In kernels built with CONFIG_NO_HZ_FULL=y, set
3749 the specified list of CPUs whose tick will be stopped
3750 whenever possible. The boot CPU will be forced outside
3751 the range to maintain the timekeeping. Any CPUs
3752 in this list will have their RCU callbacks offloaded,
3753 just as if they had also been called out in the
3754 rcu_nocbs= boot parameter.
3755
3756 Note that this argument takes precedence over
3757 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
3758
3759 noiotrap [SH] Disables trapped I/O port accesses.
3760
3761 noirqdebug [X86-32] Disables the code which attempts to detect and
3762 disable unhandled interrupt sources.
3763
3764 no_timer_check [X86,APIC] Disables the code which tests for
3765 broken timer IRQ sources.
3766
3767 noisapnp [ISAPNP] Disables ISA PnP code.
3768
3769 noinitrd [RAM] Tells the kernel not to load any configured
3770 initial RAM disk.
3771
3772 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3773 remapping.
3774 [Deprecated - use intremap=off]
3775
3776 nointroute [IA-64]
3777
3778 noinvpcid [X86] Disable the INVPCID cpu feature.
3779
3780 nojitter [IA-64] Disables jitter checking for ITC timers.
3781
3782 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3783
3784 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3785 fault handling.
3786
3787 no-vmw-sched-clock
3788 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3789 clock and use the default one.
3790
3791 no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES] Disable paravirtualized
3792 steal time accounting. steal time is computed, but
3793 won't influence scheduler behaviour
3794
3795 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3796
3797 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3798
3799 nomca [IA-64] Disable machine check abort handling
3800
3801 nomce [X86-32] Disable Machine Check Exception
3802
3803 nomfgpt [X86-32] Disable Multi-Function General Purpose
3804 Timer usage (for AMD Geode machines).
3805
3806 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3807 shutdown the other cpus. Instead use the REBOOT_VECTOR
3808 irq.
3809
3810 nomodeset Disable kernel modesetting. Most systems' firmware
3811 sets up a display mode and provides framebuffer memory
3812 for output. With nomodeset, DRM and fbdev drivers will
3813 not load if they could possibly displace the pre-
3814 initialized output. Only the system framebuffer will
3815 be available for use. The respective drivers will not
3816 perform display-mode changes or accelerated rendering.
3817
3818 Useful as error fallback, or for testing and debugging.
3819
3820 nomodule Disable module load
3821
3822 nopat [X86] Disable PAT (page attribute table extension of
3823 pagetables) support.
3824
3825 nopcid [X86-64] Disable the PCID cpu feature.
3826
3827 norandmaps Don't use address space randomization. Equivalent to
3828 echo 0 > /proc/sys/kernel/randomize_va_space
3829
3830 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3831 with UP alternatives
3832
3833 noresume [SWSUSP] Disables resume and restores original swap
3834 space.
3835
3836 no-scroll [VGA] Disables scrollback.
3837 This is required for the Braillex ib80-piezo Braille
3838 reader made by F.H. Papenmeier (Germany).
3839
3840 nosbagart [IA-64]
3841
3842 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3843
3844 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3845 and disable the IO APIC. legacy for "maxcpus=0".
3846
3847 nosoftlockup [KNL] Disable the soft-lockup detector.
3848
3849 nosync [HW,M68K] Disables sync negotiation for all devices.
3850
3851 nowatchdog [KNL] Disable both lockup detectors, i.e.
3852 soft-lockup and NMI watchdog (hard-lockup).
3853
3854 nowb [ARM]
3855
3856 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3857
3858 NOTE: this parameter will be ignored on systems with the
3859 LEGACY_XAPIC_DISABLED bit set in the
3860 IA32_XAPIC_DISABLE_STATUS MSR.
3861
3862 nps_mtm_hs_ctr= [KNL,ARC]
3863 This parameter sets the maximum duration, in
3864 cycles, each HW thread of the CTOP can run
3865 without interruptions, before HW switches it.
3866 The actual maximum duration is 16 times this
3867 parameter's value.
3868 Format: integer between 1 and 255
3869 Default: 255
3870
3871 nptcg= [IA-64] Override max number of concurrent global TLB
3872 purges which is reported from either PAL_VM_SUMMARY or
3873 SAL PALO.
3874
3875 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3876 could support. nr_cpus=n : n >= 1 limits the kernel to
3877 support 'n' processors. It could be larger than the
3878 number of already plugged CPU during bootup, later in
3879 runtime you can physically add extra cpu until it reaches
3880 n. So during boot up some boot time memory for per-cpu
3881 variables need be pre-allocated for later physical cpu
3882 hot plugging.
3883
3884 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3885
3886 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86] Disable NUMA, Only
3887 set up a single NUMA node spanning all memory.
3888
3889 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
3890 NUMA balancing.
3891 Allowed values are enable and disable
3892
3893 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3894 'node', 'default' can be specified
3895 This can be set from sysctl after boot.
3896 See Documentation/admin-guide/sysctl/vm.rst for details.
3897
3898 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3899 See Documentation/core-api/debugging-via-ohci1394.rst for more
3900 info.
3901
3902 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3903 Rather than timing out after 20 ms if an EC
3904 command is not properly ACKed, override the length
3905 of the timeout. We have interrupts disabled while
3906 waiting for the ACK, so if this is set too high
3907 interrupts *may* be lost!
3908
3909 omap_mux= [OMAP] Override bootloader pin multiplexing.
3910 Format: <mux_mode0.mode_name=value>...
3911 For example, to override I2C bus2:
3912 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3913
3914 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
3915
3916 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
3917
3918 boundary - index of last SLC block on Flex-OneNAND.
3919 The remaining blocks are configured as MLC blocks.
3920 lock - Configure if Flex-OneNAND boundary should be locked.
3921 Once locked, the boundary cannot be changed.
3922 1 indicates lock status, 0 indicates unlock status.
3923
3924 oops=panic Always panic on oopses. Default is to just kill the
3925 process, but there is a small probability of
3926 deadlocking the machine.
3927 This will also cause panics on machine check exceptions.
3928 Useful together with panic=30 to trigger a reboot.
3929
3930 page_alloc.shuffle=
3931 [KNL] Boolean flag to control whether the page allocator
3932 should randomize its free lists. The randomization may
3933 be automatically enabled if the kernel detects it is
3934 running on a platform with a direct-mapped memory-side
3935 cache, and this parameter can be used to
3936 override/disable that behavior. The state of the flag
3937 can be read from sysfs at:
3938 /sys/module/page_alloc/parameters/shuffle.
3939
3940 page_owner= [KNL] Boot-time page_owner enabling option.
3941 Storage of the information about who allocated
3942 each page is disabled in default. With this switch,
3943 we can turn it on.
3944 on: enable the feature
3945
3946 page_poison= [KNL] Boot-time parameter changing the state of
3947 poisoning on the buddy allocator, available with
3948 CONFIG_PAGE_POISONING=y.
3949 off: turn off poisoning (default)
3950 on: turn on poisoning
3951
3952 page_reporting.page_reporting_order=
3953 [KNL] Minimal page reporting order
3954 Format: <integer>
3955 Adjust the minimal page reporting order. The page
3956 reporting is disabled when it exceeds (MAX_ORDER-1).
3957
3958 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3959 timeout > 0: seconds before rebooting
3960 timeout = 0: wait forever
3961 timeout < 0: reboot immediately
3962 Format: <timeout>
3963
3964 panic_print= Bitmask for printing system info when panic happens.
3965 User can chose combination of the following bits:
3966 bit 0: print all tasks info
3967 bit 1: print system memory info
3968 bit 2: print timer info
3969 bit 3: print locks info if CONFIG_LOCKDEP is on
3970 bit 4: print ftrace buffer
3971 bit 5: print all printk messages in buffer
3972 bit 6: print all CPUs backtrace (if available in the arch)
3973 *Be aware* that this option may print a _lot_ of lines,
3974 so there are risks of losing older messages in the log.
3975 Use this option carefully, maybe worth to setup a
3976 bigger log buffer with "log_buf_len" along with this.
3977
3978 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3979 Format: <hex>[,nousertaint]
3980 Hexadecimal bitmask representing the set of TAINT flags
3981 that will cause the kernel to panic when add_taint() is
3982 called with any of the flags in this set.
3983 The optional switch "nousertaint" can be utilized to
3984 prevent userspace forced crashes by writing to sysctl
3985 /proc/sys/kernel/tainted any flagset matching with the
3986 bitmask set on panic_on_taint.
3987 See Documentation/admin-guide/tainted-kernels.rst for
3988 extra details on the taint flags that users can pick
3989 to compose the bitmask to assign to panic_on_taint.
3990
3991 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3992 on a WARN().
3993
3994 parkbd.port= [HW] Parallel port number the keyboard adapter is
3995 connected to, default is 0.
3996 Format: <parport#>
3997 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3998 0 for XT, 1 for AT (default is AT).
3999 Format: <mode>
4000
4001 parport= [HW,PPT] Specify parallel ports. 0 disables.
4002 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
4003 Use 'auto' to force the driver to use any
4004 IRQ/DMA settings detected (the default is to
4005 ignore detected IRQ/DMA settings because of
4006 possible conflicts). You can specify the base
4007 address, IRQ, and DMA settings; IRQ and DMA
4008 should be numbers, or 'auto' (for using detected
4009 settings on that particular port), or 'nofifo'
4010 (to avoid using a FIFO even if it is detected).
4011 Parallel ports are assigned in the order they
4012 are specified on the command line, starting
4013 with parport0.
4014
4015 parport_init_mode= [HW,PPT]
4016 Configure VIA parallel port to operate in
4017 a specific mode. This is necessary on Pegasos
4018 computer where firmware has no options for setting
4019 up parallel port mode and sets it to spp.
4020 Currently this function knows 686a and 8231 chips.
4021 Format: [spp|ps2|epp|ecp|ecpepp]
4022
4023 pata_legacy.all= [HW,LIBATA]
4024 Format: <int>
4025 Set to non-zero to probe primary and secondary ISA
4026 port ranges on PCI systems where no PCI PATA device
4027 has been found at either range. Disabled by default.
4028
4029 pata_legacy.autospeed= [HW,LIBATA]
4030 Format: <int>
4031 Set to non-zero if a chip is present that snoops speed
4032 changes. Disabled by default.
4033
4034 pata_legacy.ht6560a= [HW,LIBATA]
4035 Format: <int>
4036 Set to 1, 2, or 3 for HT 6560A on the primary channel,
4037 the secondary channel, or both channels respectively.
4038 Disabled by default.
4039
4040 pata_legacy.ht6560b= [HW,LIBATA]
4041 Format: <int>
4042 Set to 1, 2, or 3 for HT 6560B on the primary channel,
4043 the secondary channel, or both channels respectively.
4044 Disabled by default.
4045
4046 pata_legacy.iordy_mask= [HW,LIBATA]
4047 Format: <int>
4048 IORDY enable mask. Set individual bits to allow IORDY
4049 for the respective channel. Bit 0 is for the first
4050 legacy channel handled by this driver, bit 1 is for
4051 the second channel, and so on. The sequence will often
4052 correspond to the primary legacy channel, the secondary
4053 legacy channel, and so on, but the handling of a PCI
4054 bus and the use of other driver options may interfere
4055 with the sequence. By default IORDY is allowed across
4056 all channels.
4057
4058 pata_legacy.opti82c46x= [HW,LIBATA]
4059 Format: <int>
4060 Set to 1, 2, or 3 for Opti 82c611A on the primary
4061 channel, the secondary channel, or both channels
4062 respectively. Disabled by default.
4063
4064 pata_legacy.opti82c611a= [HW,LIBATA]
4065 Format: <int>
4066 Set to 1, 2, or 3 for Opti 82c465MV on the primary
4067 channel, the secondary channel, or both channels
4068 respectively. Disabled by default.
4069
4070 pata_legacy.pio_mask= [HW,LIBATA]
4071 Format: <int>
4072 PIO mode mask for autospeed devices. Set individual
4073 bits to allow the use of the respective PIO modes.
4074 Bit 0 is for mode 0, bit 1 is for mode 1, and so on.
4075 All modes allowed by default.
4076
4077 pata_legacy.probe_all= [HW,LIBATA]
4078 Format: <int>
4079 Set to non-zero to probe tertiary and further ISA
4080 port ranges on PCI systems. Disabled by default.
4081
4082 pata_legacy.probe_mask= [HW,LIBATA]
4083 Format: <int>
4084 Probe mask for legacy ISA PATA ports. Depending on
4085 platform configuration and the use of other driver
4086 options up to 6 legacy ports are supported: 0x1f0,
4087 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing
4088 of individual ports can be disabled by setting the
4089 corresponding bits in the mask to 1. Bit 0 is for
4090 the first port in the list above (0x1f0), and so on.
4091 By default all supported ports are probed.
4092
4093 pata_legacy.qdi= [HW,LIBATA]
4094 Format: <int>
4095 Set to non-zero to probe QDI controllers. By default
4096 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
4097
4098 pata_legacy.winbond= [HW,LIBATA]
4099 Format: <int>
4100 Set to non-zero to probe Winbond controllers. Use
4101 the standard I/O port (0x130) if 1, otherwise the
4102 value given is the I/O port to use (typically 0x1b0).
4103 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
4104 0 otherwise.
4105
4106 pata_platform.pio_mask= [HW,LIBATA]
4107 Format: <int>
4108 Supported PIO mode mask. Set individual bits to allow
4109 the use of the respective PIO modes. Bit 0 is for
4110 mode 0, bit 1 is for mode 1, and so on. Mode 0 only
4111 allowed by default.
4112
4113 pause_on_oops=
4114 Halt all CPUs after the first oops has been printed for
4115 the specified number of seconds. This is to be used if
4116 your oopses keep scrolling off the screen.
4117
4118 pcbit= [HW,ISDN]
4119
4120 pcd. [PARIDE]
4121 See header of drivers/block/paride/pcd.c.
4122 See also Documentation/admin-guide/blockdev/paride.rst.
4123
4124 pci=option[,option...] [PCI] various PCI subsystem options.
4125
4126 Some options herein operate on a specific device
4127 or a set of devices (<pci_dev>). These are
4128 specified in one of the following formats:
4129
4130 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
4131 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
4132
4133 Note: the first format specifies a PCI
4134 bus/device/function address which may change
4135 if new hardware is inserted, if motherboard
4136 firmware changes, or due to changes caused
4137 by other kernel parameters. If the
4138 domain is left unspecified, it is
4139 taken to be zero. Optionally, a path
4140 to a device through multiple device/function
4141 addresses can be specified after the base
4142 address (this is more robust against
4143 renumbering issues). The second format
4144 selects devices using IDs from the
4145 configuration space which may match multiple
4146 devices in the system.
4147
4148 earlydump dump PCI config space before the kernel
4149 changes anything
4150 off [X86] don't probe for the PCI bus
4151 bios [X86-32] force use of PCI BIOS, don't access
4152 the hardware directly. Use this if your machine
4153 has a non-standard PCI host bridge.
4154 nobios [X86-32] disallow use of PCI BIOS, only direct
4155 hardware access methods are allowed. Use this
4156 if you experience crashes upon bootup and you
4157 suspect they are caused by the BIOS.
4158 conf1 [X86] Force use of PCI Configuration Access
4159 Mechanism 1 (config address in IO port 0xCF8,
4160 data in IO port 0xCFC, both 32-bit).
4161 conf2 [X86] Force use of PCI Configuration Access
4162 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
4163 the function, IO port 0xCFA, also 8-bit, sets
4164 bus number. The config space is then accessed
4165 through ports 0xC000-0xCFFF).
4166 See http://wiki.osdev.org/PCI for more info
4167 on the configuration access mechanisms.
4168 noaer [PCIE] If the PCIEAER kernel config parameter is
4169 enabled, this kernel boot option can be used to
4170 disable the use of PCIE advanced error reporting.
4171 nodomains [PCI] Disable support for multiple PCI
4172 root domains (aka PCI segments, in ACPI-speak).
4173 nommconf [X86] Disable use of MMCONFIG for PCI
4174 Configuration
4175 check_enable_amd_mmconf [X86] check for and enable
4176 properly configured MMIO access to PCI
4177 config space on AMD family 10h CPU
4178 nomsi [MSI] If the PCI_MSI kernel config parameter is
4179 enabled, this kernel boot option can be used to
4180 disable the use of MSI interrupts system-wide.
4181 noioapicquirk [APIC] Disable all boot interrupt quirks.
4182 Safety option to keep boot IRQs enabled. This
4183 should never be necessary.
4184 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
4185 primary IO-APIC for bridges that cannot disable
4186 boot IRQs. This fixes a source of spurious IRQs
4187 when the system masks IRQs.
4188 noioapicreroute [APIC] Disable workaround that uses the
4189 boot IRQ equivalent of an IRQ that connects to
4190 a chipset where boot IRQs cannot be disabled.
4191 The opposite of ioapicreroute.
4192 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
4193 routing table. These calls are known to be buggy
4194 on several machines and they hang the machine
4195 when used, but on other computers it's the only
4196 way to get the interrupt routing table. Try
4197 this option if the kernel is unable to allocate
4198 IRQs or discover secondary PCI buses on your
4199 motherboard.
4200 rom [X86] Assign address space to expansion ROMs.
4201 Use with caution as certain devices share
4202 address decoders between ROMs and other
4203 resources.
4204 norom [X86] Do not assign address space to
4205 expansion ROMs that do not already have
4206 BIOS assigned address ranges.
4207 nobar [X86] Do not assign address space to the
4208 BARs that weren't assigned by the BIOS.
4209 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
4210 assigned automatically to PCI devices. You can
4211 make the kernel exclude IRQs of your ISA cards
4212 this way.
4213 pirqaddr=0xAAAAA [X86] Specify the physical address
4214 of the PIRQ table (normally generated
4215 by the BIOS) if it is outside the
4216 F0000h-100000h range.
4217 lastbus=N [X86] Scan all buses thru bus #N. Can be
4218 useful if the kernel is unable to find your
4219 secondary buses and you want to tell it
4220 explicitly which ones they are.
4221 assign-busses [X86] Always assign all PCI bus
4222 numbers ourselves, overriding
4223 whatever the firmware may have done.
4224 usepirqmask [X86] Honor the possible IRQ mask stored
4225 in the BIOS $PIR table. This is needed on
4226 some systems with broken BIOSes, notably
4227 some HP Pavilion N5400 and Omnibook XE3
4228 notebooks. This will have no effect if ACPI
4229 IRQ routing is enabled.
4230 noacpi [X86] Do not use ACPI for IRQ routing
4231 or for PCI scanning.
4232 use_crs [X86] Use PCI host bridge window information
4233 from ACPI. On BIOSes from 2008 or later, this
4234 is enabled by default. If you need to use this,
4235 please report a bug.
4236 nocrs [X86] Ignore PCI host bridge windows from ACPI.
4237 If you need to use this, please report a bug.
4238 use_e820 [X86] Use E820 reservations to exclude parts of
4239 PCI host bridge windows. This is a workaround
4240 for BIOS defects in host bridge _CRS methods.
4241 If you need to use this, please report a bug to
4242 <linux-pci@vger.kernel.org>.
4243 no_e820 [X86] Ignore E820 reservations for PCI host
4244 bridge windows. This is the default on modern
4245 hardware. If you need to use this, please report
4246 a bug to <linux-pci@vger.kernel.org>.
4247 routeirq Do IRQ routing for all PCI devices.
4248 This is normally done in pci_enable_device(),
4249 so this option is a temporary workaround
4250 for broken drivers that don't call it.
4251 skip_isa_align [X86] do not align io start addr, so can
4252 handle more pci cards
4253 noearly [X86] Don't do any early type 1 scanning.
4254 This might help on some broken boards which
4255 machine check when some devices' config space
4256 is read. But various workarounds are disabled
4257 and some IOMMU drivers will not work.
4258 bfsort Sort PCI devices into breadth-first order.
4259 This sorting is done to get a device
4260 order compatible with older (<= 2.4) kernels.
4261 nobfsort Don't sort PCI devices into breadth-first order.
4262 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
4263 tuning and use the BIOS-configured MPS defaults.
4264 pcie_bus_safe Set every device's MPS to the largest value
4265 supported by all devices below the root complex.
4266 pcie_bus_perf Set device MPS to the largest allowable MPS
4267 based on its parent bus. Also set MRRS (Max
4268 Read Request Size) to the largest supported
4269 value (no larger than the MPS that the device
4270 or bus can support) for best performance.
4271 pcie_bus_peer2peer Set every device's MPS to 128B, which
4272 every device is guaranteed to support. This
4273 configuration allows peer-to-peer DMA between
4274 any pair of devices, possibly at the cost of
4275 reduced performance. This also guarantees
4276 that hot-added devices will work.
4277 cbiosize=nn[KMG] The fixed amount of bus space which is
4278 reserved for the CardBus bridge's IO window.
4279 The default value is 256 bytes.
4280 cbmemsize=nn[KMG] The fixed amount of bus space which is
4281 reserved for the CardBus bridge's memory
4282 window. The default value is 64 megabytes.
4283 resource_alignment=
4284 Format:
4285 [<order of align>@]<pci_dev>[; ...]
4286 Specifies alignment and device to reassign
4287 aligned memory resources. How to
4288 specify the device is described above.
4289 If <order of align> is not specified,
4290 PAGE_SIZE is used as alignment.
4291 A PCI-PCI bridge can be specified if resource
4292 windows need to be expanded.
4293 To specify the alignment for several
4294 instances of a device, the PCI vendor,
4295 device, subvendor, and subdevice may be
4296 specified, e.g., 12@pci:8086:9c22:103c:198f
4297 for 4096-byte alignment.
4298 ecrc= Enable/disable PCIe ECRC (transaction layer
4299 end-to-end CRC checking).
4300 bios: Use BIOS/firmware settings. This is the
4301 the default.
4302 off: Turn ECRC off
4303 on: Turn ECRC on.
4304 hpiosize=nn[KMG] The fixed amount of bus space which is
4305 reserved for hotplug bridge's IO window.
4306 Default size is 256 bytes.
4307 hpmmiosize=nn[KMG] The fixed amount of bus space which is
4308 reserved for hotplug bridge's MMIO window.
4309 Default size is 2 megabytes.
4310 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
4311 reserved for hotplug bridge's MMIO_PREF window.
4312 Default size is 2 megabytes.
4313 hpmemsize=nn[KMG] The fixed amount of bus space which is
4314 reserved for hotplug bridge's MMIO and
4315 MMIO_PREF window.
4316 Default size is 2 megabytes.
4317 hpbussize=nn The minimum amount of additional bus numbers
4318 reserved for buses below a hotplug bridge.
4319 Default is 1.
4320 realloc= Enable/disable reallocating PCI bridge resources
4321 if allocations done by BIOS are too small to
4322 accommodate resources required by all child
4323 devices.
4324 off: Turn realloc off
4325 on: Turn realloc on
4326 realloc same as realloc=on
4327 noari do not use PCIe ARI.
4328 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
4329 do not use PCIe ATS (and IOMMU device IOTLB).
4330 pcie_scan_all Scan all possible PCIe devices. Otherwise we
4331 only look for one device below a PCIe downstream
4332 port.
4333 big_root_window Try to add a big 64bit memory window to the PCIe
4334 root complex on AMD CPUs. Some GFX hardware
4335 can resize a BAR to allow access to all VRAM.
4336 Adding the window is slightly risky (it may
4337 conflict with unreported devices), so this
4338 taints the kernel.
4339 disable_acs_redir=<pci_dev>[; ...]
4340 Specify one or more PCI devices (in the format
4341 specified above) separated by semicolons.
4342 Each device specified will have the PCI ACS
4343 redirect capabilities forced off which will
4344 allow P2P traffic between devices through
4345 bridges without forcing it upstream. Note:
4346 this removes isolation between devices and
4347 may put more devices in an IOMMU group.
4348 force_floating [S390] Force usage of floating interrupts.
4349 nomio [S390] Do not use MIO instructions.
4350 norid [S390] ignore the RID field and force use of
4351 one PCI domain per PCI function
4352
4353 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
4354 Management.
4355 off Disable ASPM.
4356 force Enable ASPM even on devices that claim not to support it.
4357 WARNING: Forcing ASPM on may cause system lockups.
4358
4359 pcie_ports= [PCIE] PCIe port services handling:
4360 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
4361 even if the platform doesn't give the OS permission to
4362 use them. This may cause conflicts if the platform
4363 also tries to use these services.
4364 dpc-native Use native PCIe service for DPC only. May
4365 cause conflicts if firmware uses AER or DPC.
4366 compat Disable native PCIe services (PME, AER, DPC, PCIe
4367 hotplug).
4368
4369 pcie_port_pm= [PCIE] PCIe port power management handling:
4370 off Disable power management of all PCIe ports
4371 force Forcibly enable power management of all PCIe ports
4372
4373 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
4374 nomsi Do not use MSI for native PCIe PME signaling (this makes
4375 all PCIe root ports use INTx for all services).
4376
4377 pcmv= [HW,PCMCIA] BadgePAD 4
4378
4379 pd_ignore_unused
4380 [PM]
4381 Keep all power-domains already enabled by bootloader on,
4382 even if no driver has claimed them. This is useful
4383 for debug and development, but should not be
4384 needed on a platform with proper driver support.
4385
4386 pd. [PARIDE]
4387 See Documentation/admin-guide/blockdev/paride.rst.
4388
4389 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
4390 boot time.
4391 Format: { 0 | 1 }
4392 See arch/parisc/kernel/pdc_chassis.c
4393
4394 percpu_alloc= Select which percpu first chunk allocator to use.
4395 Currently supported values are "embed" and "page".
4396 Archs may support subset or none of the selections.
4397 See comments in mm/percpu.c for details on each
4398 allocator. This parameter is primarily for debugging
4399 and performance comparison.
4400
4401 pf. [PARIDE]
4402 See Documentation/admin-guide/blockdev/paride.rst.
4403
4404 pg. [PARIDE]
4405 See Documentation/admin-guide/blockdev/paride.rst.
4406
4407 pirq= [SMP,APIC] Manual mp-table setup
4408 See Documentation/x86/i386/IO-APIC.rst.
4409
4410 plip= [PPT,NET] Parallel port network link
4411 Format: { parport<nr> | timid | 0 }
4412 See also Documentation/admin-guide/parport.rst.
4413
4414 pmtmr= [X86] Manual setup of pmtmr I/O Port.
4415 Override pmtimer IOPort with a hex value.
4416 e.g. pmtmr=0x508
4417
4418 pmu_override= [PPC] Override the PMU.
4419 This option takes over the PMU facility, so it is no
4420 longer usable by perf. Setting this option starts the
4421 PMU counters by setting MMCR0 to 0 (the FC bit is
4422 cleared). If a number is given, then MMCR1 is set to
4423 that number, otherwise (e.g., 'pmu_override=on'), MMCR1
4424 remains 0.
4425
4426 pm_debug_messages [SUSPEND,KNL]
4427 Enable suspend/resume debug messages during boot up.
4428
4429 pnp.debug=1 [PNP]
4430 Enable PNP debug messages (depends on the
4431 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4432 via /sys/module/pnp/parameters/debug. We always show
4433 current resource usage; turning this on also shows
4434 possible settings and some assignment information.
4435
4436 pnpacpi= [ACPI]
4437 { off }
4438
4439 pnpbios= [ISAPNP]
4440 { on | off | curr | res | no-curr | no-res }
4441
4442 pnp_reserve_irq=
4443 [ISAPNP] Exclude IRQs for the autoconfiguration
4444
4445 pnp_reserve_dma=
4446 [ISAPNP] Exclude DMAs for the autoconfiguration
4447
4448 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
4449 Ranges are in pairs (I/O port base and size).
4450
4451 pnp_reserve_mem=
4452 [ISAPNP] Exclude memory regions for the
4453 autoconfiguration.
4454 Ranges are in pairs (memory base and size).
4455
4456 ports= [IP_VS_FTP] IPVS ftp helper module
4457 Default is 21.
4458 Up to 8 (IP_VS_APP_MAX_PORTS) ports
4459 may be specified.
4460 Format: <port>,<port>....
4461
4462 powersave=off [PPC] This option disables power saving features.
4463 It specifically disables cpuidle and sets the
4464 platform machine description specific power_save
4465 function to NULL. On Idle the CPU just reduces
4466 execution priority.
4467
4468 ppc_strict_facility_enable
4469 [PPC] This option catches any kernel floating point,
4470 Altivec, VSX and SPE outside of regions specifically
4471 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
4472 There is some performance impact when enabling this.
4473
4474 ppc_tm= [PPC]
4475 Format: {"off"}
4476 Disable Hardware Transactional Memory
4477
4478 preempt= [KNL]
4479 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
4480 none - Limited to cond_resched() calls
4481 voluntary - Limited to cond_resched() and might_sleep() calls
4482 full - Any section that isn't explicitly preempt disabled
4483 can be preempted anytime.
4484
4485 print-fatal-signals=
4486 [KNL] debug: print fatal signals
4487
4488 If enabled, warn about various signal handling
4489 related application anomalies: too many signals,
4490 too many POSIX.1 timers, fatal signals causing a
4491 coredump - etc.
4492
4493 If you hit the warning due to signal overflow,
4494 you might want to try "ulimit -i unlimited".
4495
4496 default: off.
4497
4498 printk.always_kmsg_dump=
4499 Trigger kmsg_dump for cases other than kernel oops or
4500 panics
4501 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4502 default: disabled
4503
4504 printk.console_no_auto_verbose=
4505 Disable console loglevel raise on oops, panic
4506 or lockdep-detected issues (only if lock debug is on).
4507 With an exception to setups with low baudrate on
4508 serial console, keeping this 0 is a good choice
4509 in order to provide more debug information.
4510 Format: <bool>
4511 default: 0 (auto_verbose is enabled)
4512
4513 printk.devkmsg={on,off,ratelimit}
4514 Control writing to /dev/kmsg.
4515 on - unlimited logging to /dev/kmsg from userspace
4516 off - logging to /dev/kmsg disabled
4517 ratelimit - ratelimit the logging
4518 Default: ratelimit
4519
4520 printk.time= Show timing data prefixed to each printk message line
4521 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4522
4523 processor.max_cstate= [HW,ACPI]
4524 Limit processor to maximum C-state
4525 max_cstate=9 overrides any DMI blacklist limit.
4526
4527 processor.nocst [HW,ACPI]
4528 Ignore the _CST method to determine C-states,
4529 instead using the legacy FADT method
4530
4531 profile= [KNL] Enable kernel profiling via /proc/profile
4532 Format: [<profiletype>,]<number>
4533 Param: <profiletype>: "schedule", "sleep", or "kvm"
4534 [defaults to kernel profiling]
4535 Param: "schedule" - profile schedule points.
4536 Param: "sleep" - profile D-state sleeping (millisecs).
4537 Requires CONFIG_SCHEDSTATS
4538 Param: "kvm" - profile VM exits.
4539 Param: <number> - step/bucket size as a power of 2 for
4540 statistical time based profiling.
4541
4542 prompt_ramdisk= [RAM] [Deprecated]
4543
4544 prot_virt= [S390] enable hosting protected virtual machines
4545 isolated from the hypervisor (if hardware supports
4546 that).
4547 Format: <bool>
4548
4549 psi= [KNL] Enable or disable pressure stall information
4550 tracking.
4551 Format: <bool>
4552
4553 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
4554 probe for; one of (bare|imps|exps|lifebook|any).
4555 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4556 per second.
4557 psmouse.resetafter= [HW,MOUSE]
4558 Try to reset the device after so many bad packets
4559 (0 = never).
4560 psmouse.resolution=
4561 [HW,MOUSE] Set desired mouse resolution, in dpi.
4562 psmouse.smartscroll=
4563 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4564 0 = disabled, 1 = enabled (default).
4565
4566 pstore.backend= Specify the name of the pstore backend to use
4567
4568 pt. [PARIDE]
4569 See Documentation/admin-guide/blockdev/paride.rst.
4570
4571 pti= [X86-64] Control Page Table Isolation of user and
4572 kernel address spaces. Disabling this feature
4573 removes hardening, but improves performance of
4574 system calls and interrupts.
4575
4576 on - unconditionally enable
4577 off - unconditionally disable
4578 auto - kernel detects whether your CPU model is
4579 vulnerable to issues that PTI mitigates
4580
4581 Not specifying this option is equivalent to pti=auto.
4582
4583 nopti [X86-64]
4584 Equivalent to pti=off
4585
4586 pty.legacy_count=
4587 [KNL] Number of legacy pty's. Overwrites compiled-in
4588 default number.
4589
4590 quiet [KNL] Disable most log messages
4591
4592 r128= [HW,DRM]
4593
4594 raid= [HW,RAID]
4595 See Documentation/admin-guide/md.rst.
4596
4597 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4598 See Documentation/admin-guide/blockdev/ramdisk.rst.
4599
4600 ramdisk_start= [RAM] RAM disk image start address
4601
4602 random.trust_cpu=off
4603 [KNL] Disable trusting the use of the CPU's
4604 random number generator (if available) to
4605 initialize the kernel's RNG.
4606
4607 random.trust_bootloader=off
4608 [KNL] Disable trusting the use of the a seed
4609 passed by the bootloader (if available) to
4610 initialize the kernel's RNG.
4611
4612 randomize_kstack_offset=
4613 [KNL] Enable or disable kernel stack offset
4614 randomization, which provides roughly 5 bits of
4615 entropy, frustrating memory corruption attacks
4616 that depend on stack address determinism or
4617 cross-syscall address exposures. This is only
4618 available on architectures that have defined
4619 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
4620 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4621 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
4622
4623 ras=option[,option,...] [KNL] RAS-specific options
4624
4625 cec_disable [X86]
4626 Disable the Correctable Errors Collector,
4627 see CONFIG_RAS_CEC help text.
4628
4629 rcu_nocbs[=cpu-list]
4630 [KNL] The optional argument is a cpu list,
4631 as described above.
4632
4633 In kernels built with CONFIG_RCU_NOCB_CPU=y,
4634 enable the no-callback CPU mode, which prevents
4635 such CPUs' callbacks from being invoked in
4636 softirq context. Invocation of such CPUs' RCU
4637 callbacks will instead be offloaded to "rcuox/N"
4638 kthreads created for that purpose, where "x" is
4639 "p" for RCU-preempt, "s" for RCU-sched, and "g"
4640 for the kthreads that mediate grace periods; and
4641 "N" is the CPU number. This reduces OS jitter on
4642 the offloaded CPUs, which can be useful for HPC
4643 and real-time workloads. It can also improve
4644 energy efficiency for asymmetric multiprocessors.
4645
4646 If a cpulist is passed as an argument, the specified
4647 list of CPUs is set to no-callback mode from boot.
4648
4649 Otherwise, if the '=' sign and the cpulist
4650 arguments are omitted, no CPU will be set to
4651 no-callback mode from boot but the mode may be
4652 toggled at runtime via cpusets.
4653
4654 Note that this argument takes precedence over
4655 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
4656
4657 rcu_nocb_poll [KNL]
4658 Rather than requiring that offloaded CPUs
4659 (specified by rcu_nocbs= above) explicitly
4660 awaken the corresponding "rcuoN" kthreads,
4661 make these kthreads poll for callbacks.
4662 This improves the real-time response for the
4663 offloaded CPUs by relieving them of the need to
4664 wake up the corresponding kthread, but degrades
4665 energy efficiency by requiring that the kthreads
4666 periodically wake up to do the polling.
4667
4668 rcutree.blimit= [KNL]
4669 Set maximum number of finished RCU callbacks to
4670 process in one batch.
4671
4672 rcutree.dump_tree= [KNL]
4673 Dump the structure of the rcu_node combining tree
4674 out at early boot. This is used for diagnostic
4675 purposes, to verify correct tree setup.
4676
4677 rcutree.gp_cleanup_delay= [KNL]
4678 Set the number of jiffies to delay each step of
4679 RCU grace-period cleanup.
4680
4681 rcutree.gp_init_delay= [KNL]
4682 Set the number of jiffies to delay each step of
4683 RCU grace-period initialization.
4684
4685 rcutree.gp_preinit_delay= [KNL]
4686 Set the number of jiffies to delay each step of
4687 RCU grace-period pre-initialization, that is,
4688 the propagation of recent CPU-hotplug changes up
4689 the rcu_node combining tree.
4690
4691 rcutree.use_softirq= [KNL]
4692 If set to zero, move all RCU_SOFTIRQ processing to
4693 per-CPU rcuc kthreads. Defaults to a non-zero
4694 value, meaning that RCU_SOFTIRQ is used by default.
4695 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4696
4697 But note that CONFIG_PREEMPT_RT=y kernels disable
4698 this kernel boot parameter, forcibly setting it
4699 to zero.
4700
4701 rcutree.rcu_fanout_exact= [KNL]
4702 Disable autobalancing of the rcu_node combining
4703 tree. This is used by rcutorture, and might
4704 possibly be useful for architectures having high
4705 cache-to-cache transfer latencies.
4706
4707 rcutree.rcu_fanout_leaf= [KNL]
4708 Change the number of CPUs assigned to each
4709 leaf rcu_node structure. Useful for very
4710 large systems, which will choose the value 64,
4711 and for NUMA systems with large remote-access
4712 latencies, which will choose a value aligned
4713 with the appropriate hardware boundaries.
4714
4715 rcutree.rcu_min_cached_objs= [KNL]
4716 Minimum number of objects which are cached and
4717 maintained per one CPU. Object size is equal
4718 to PAGE_SIZE. The cache allows to reduce the
4719 pressure to page allocator, also it makes the
4720 whole algorithm to behave better in low memory
4721 condition.
4722
4723 rcutree.rcu_delay_page_cache_fill_msec= [KNL]
4724 Set the page-cache refill delay (in milliseconds)
4725 in response to low-memory conditions. The range
4726 of permitted values is in the range 0:100000.
4727
4728 rcutree.jiffies_till_first_fqs= [KNL]
4729 Set delay from grace-period initialization to
4730 first attempt to force quiescent states.
4731 Units are jiffies, minimum value is zero,
4732 and maximum value is HZ.
4733
4734 rcutree.jiffies_till_next_fqs= [KNL]
4735 Set delay between subsequent attempts to force
4736 quiescent states. Units are jiffies, minimum
4737 value is one, and maximum value is HZ.
4738
4739 rcutree.jiffies_till_sched_qs= [KNL]
4740 Set required age in jiffies for a
4741 given grace period before RCU starts
4742 soliciting quiescent-state help from
4743 rcu_note_context_switch() and cond_resched().
4744 If not specified, the kernel will calculate
4745 a value based on the most recent settings
4746 of rcutree.jiffies_till_first_fqs
4747 and rcutree.jiffies_till_next_fqs.
4748 This calculated value may be viewed in
4749 rcutree.jiffies_to_sched_qs. Any attempt to set
4750 rcutree.jiffies_to_sched_qs will be cheerfully
4751 overwritten.
4752
4753 rcutree.kthread_prio= [KNL,BOOT]
4754 Set the SCHED_FIFO priority of the RCU per-CPU
4755 kthreads (rcuc/N). This value is also used for
4756 the priority of the RCU boost threads (rcub/N)
4757 and for the RCU grace-period kthreads (rcu_bh,
4758 rcu_preempt, and rcu_sched). If RCU_BOOST is
4759 set, valid values are 1-99 and the default is 1
4760 (the least-favored priority). Otherwise, when
4761 RCU_BOOST is not set, valid values are 0-99 and
4762 the default is zero (non-realtime operation).
4763 When RCU_NOCB_CPU is set, also adjust the
4764 priority of NOCB callback kthreads.
4765
4766 rcutree.rcu_divisor= [KNL]
4767 Set the shift-right count to use to compute
4768 the callback-invocation batch limit bl from
4769 the number of callbacks queued on this CPU.
4770 The result will be bounded below by the value of
4771 the rcutree.blimit kernel parameter. Every bl
4772 callbacks, the softirq handler will exit in
4773 order to allow the CPU to do other work.
4774
4775 Please note that this callback-invocation batch
4776 limit applies only to non-offloaded callback
4777 invocation. Offloaded callbacks are instead
4778 invoked in the context of an rcuoc kthread, which
4779 scheduler will preempt as it does any other task.
4780
4781 rcutree.nocb_nobypass_lim_per_jiffy= [KNL]
4782 On callback-offloaded (rcu_nocbs) CPUs,
4783 RCU reduces the lock contention that would
4784 otherwise be caused by callback floods through
4785 use of the ->nocb_bypass list. However, in the
4786 common non-flooded case, RCU queues directly to
4787 the main ->cblist in order to avoid the extra
4788 overhead of the ->nocb_bypass list and its lock.
4789 But if there are too many callbacks queued during
4790 a single jiffy, RCU pre-queues the callbacks into
4791 the ->nocb_bypass queue. The definition of "too
4792 many" is supplied by this kernel boot parameter.
4793
4794 rcutree.rcu_nocb_gp_stride= [KNL]
4795 Set the number of NOCB callback kthreads in
4796 each group, which defaults to the square root
4797 of the number of CPUs. Larger numbers reduce
4798 the wakeup overhead on the global grace-period
4799 kthread, but increases that same overhead on
4800 each group's NOCB grace-period kthread.
4801
4802 rcutree.qhimark= [KNL]
4803 Set threshold of queued RCU callbacks beyond which
4804 batch limiting is disabled.
4805
4806 rcutree.qlowmark= [KNL]
4807 Set threshold of queued RCU callbacks below which
4808 batch limiting is re-enabled.
4809
4810 rcutree.qovld= [KNL]
4811 Set threshold of queued RCU callbacks beyond which
4812 RCU's force-quiescent-state scan will aggressively
4813 enlist help from cond_resched() and sched IPIs to
4814 help CPUs more quickly reach quiescent states.
4815 Set to less than zero to make this be set based
4816 on rcutree.qhimark at boot time and to zero to
4817 disable more aggressive help enlistment.
4818
4819 rcutree.rcu_kick_kthreads= [KNL]
4820 Cause the grace-period kthread to get an extra
4821 wake_up() if it sleeps three times longer than
4822 it should at force-quiescent-state time.
4823 This wake_up() will be accompanied by a
4824 WARN_ONCE() splat and an ftrace_dump().
4825
4826 rcutree.rcu_unlock_delay= [KNL]
4827 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4828 this specifies an rcu_read_unlock()-time delay
4829 in microseconds. This defaults to zero.
4830 Larger delays increase the probability of
4831 catching RCU pointer leaks, that is, buggy use
4832 of RCU-protected pointers after the relevant
4833 rcu_read_unlock() has completed.
4834
4835 rcutree.sysrq_rcu= [KNL]
4836 Commandeer a sysrq key to dump out Tree RCU's
4837 rcu_node tree with an eye towards determining
4838 why a new grace period has not yet started.
4839
4840 rcuscale.gp_async= [KNL]
4841 Measure performance of asynchronous
4842 grace-period primitives such as call_rcu().
4843
4844 rcuscale.gp_async_max= [KNL]
4845 Specify the maximum number of outstanding
4846 callbacks per writer thread. When a writer
4847 thread exceeds this limit, it invokes the
4848 corresponding flavor of rcu_barrier() to allow
4849 previously posted callbacks to drain.
4850
4851 rcuscale.gp_exp= [KNL]
4852 Measure performance of expedited synchronous
4853 grace-period primitives.
4854
4855 rcuscale.holdoff= [KNL]
4856 Set test-start holdoff period. The purpose of
4857 this parameter is to delay the start of the
4858 test until boot completes in order to avoid
4859 interference.
4860
4861 rcuscale.kfree_rcu_test= [KNL]
4862 Set to measure performance of kfree_rcu() flooding.
4863
4864 rcuscale.kfree_rcu_test_double= [KNL]
4865 Test the double-argument variant of kfree_rcu().
4866 If this parameter has the same value as
4867 rcuscale.kfree_rcu_test_single, both the single-
4868 and double-argument variants are tested.
4869
4870 rcuscale.kfree_rcu_test_single= [KNL]
4871 Test the single-argument variant of kfree_rcu().
4872 If this parameter has the same value as
4873 rcuscale.kfree_rcu_test_double, both the single-
4874 and double-argument variants are tested.
4875
4876 rcuscale.kfree_nthreads= [KNL]
4877 The number of threads running loops of kfree_rcu().
4878
4879 rcuscale.kfree_alloc_num= [KNL]
4880 Number of allocations and frees done in an iteration.
4881
4882 rcuscale.kfree_loops= [KNL]
4883 Number of loops doing rcuscale.kfree_alloc_num number
4884 of allocations and frees.
4885
4886 rcuscale.nreaders= [KNL]
4887 Set number of RCU readers. The value -1 selects
4888 N, where N is the number of CPUs. A value
4889 "n" less than -1 selects N-n+1, where N is again
4890 the number of CPUs. For example, -2 selects N
4891 (the number of CPUs), -3 selects N+1, and so on.
4892 A value of "n" less than or equal to -N selects
4893 a single reader.
4894
4895 rcuscale.nwriters= [KNL]
4896 Set number of RCU writers. The values operate
4897 the same as for rcuscale.nreaders.
4898 N, where N is the number of CPUs
4899
4900 rcuscale.perf_type= [KNL]
4901 Specify the RCU implementation to test.
4902
4903 rcuscale.shutdown= [KNL]
4904 Shut the system down after performance tests
4905 complete. This is useful for hands-off automated
4906 testing.
4907
4908 rcuscale.verbose= [KNL]
4909 Enable additional printk() statements.
4910
4911 rcuscale.writer_holdoff= [KNL]
4912 Write-side holdoff between grace periods,
4913 in microseconds. The default of zero says
4914 no holdoff.
4915
4916 rcutorture.fqs_duration= [KNL]
4917 Set duration of force_quiescent_state bursts
4918 in microseconds.
4919
4920 rcutorture.fqs_holdoff= [KNL]
4921 Set holdoff time within force_quiescent_state bursts
4922 in microseconds.
4923
4924 rcutorture.fqs_stutter= [KNL]
4925 Set wait time between force_quiescent_state bursts
4926 in seconds.
4927
4928 rcutorture.fwd_progress= [KNL]
4929 Specifies the number of kthreads to be used
4930 for RCU grace-period forward-progress testing
4931 for the types of RCU supporting this notion.
4932 Defaults to 1 kthread, values less than zero or
4933 greater than the number of CPUs cause the number
4934 of CPUs to be used.
4935
4936 rcutorture.fwd_progress_div= [KNL]
4937 Specify the fraction of a CPU-stall-warning
4938 period to do tight-loop forward-progress testing.
4939
4940 rcutorture.fwd_progress_holdoff= [KNL]
4941 Number of seconds to wait between successive
4942 forward-progress tests.
4943
4944 rcutorture.fwd_progress_need_resched= [KNL]
4945 Enclose cond_resched() calls within checks for
4946 need_resched() during tight-loop forward-progress
4947 testing.
4948
4949 rcutorture.gp_cond= [KNL]
4950 Use conditional/asynchronous update-side
4951 primitives, if available.
4952
4953 rcutorture.gp_exp= [KNL]
4954 Use expedited update-side primitives, if available.
4955
4956 rcutorture.gp_normal= [KNL]
4957 Use normal (non-expedited) asynchronous
4958 update-side primitives, if available.
4959
4960 rcutorture.gp_sync= [KNL]
4961 Use normal (non-expedited) synchronous
4962 update-side primitives, if available. If all
4963 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4964 rcutorture.gp_normal=, and rcutorture.gp_sync=
4965 are zero, rcutorture acts as if is interpreted
4966 they are all non-zero.
4967
4968 rcutorture.irqreader= [KNL]
4969 Run RCU readers from irq handlers, or, more
4970 accurately, from a timer handler. Not all RCU
4971 flavors take kindly to this sort of thing.
4972
4973 rcutorture.leakpointer= [KNL]
4974 Leak an RCU-protected pointer out of the reader.
4975 This can of course result in splats, and is
4976 intended to test the ability of things like
4977 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
4978 such leaks.
4979
4980 rcutorture.n_barrier_cbs= [KNL]
4981 Set callbacks/threads for rcu_barrier() testing.
4982
4983 rcutorture.nfakewriters= [KNL]
4984 Set number of concurrent RCU writers. These just
4985 stress RCU, they don't participate in the actual
4986 test, hence the "fake".
4987
4988 rcutorture.nocbs_nthreads= [KNL]
4989 Set number of RCU callback-offload togglers.
4990 Zero (the default) disables toggling.
4991
4992 rcutorture.nocbs_toggle= [KNL]
4993 Set the delay in milliseconds between successive
4994 callback-offload toggling attempts.
4995
4996 rcutorture.nreaders= [KNL]
4997 Set number of RCU readers. The value -1 selects
4998 N-1, where N is the number of CPUs. A value
4999 "n" less than -1 selects N-n-2, where N is again
5000 the number of CPUs. For example, -2 selects N
5001 (the number of CPUs), -3 selects N+1, and so on.
5002
5003 rcutorture.object_debug= [KNL]
5004 Enable debug-object double-call_rcu() testing.
5005
5006 rcutorture.onoff_holdoff= [KNL]
5007 Set time (s) after boot for CPU-hotplug testing.
5008
5009 rcutorture.onoff_interval= [KNL]
5010 Set time (jiffies) between CPU-hotplug operations,
5011 or zero to disable CPU-hotplug testing.
5012
5013 rcutorture.read_exit= [KNL]
5014 Set the number of read-then-exit kthreads used
5015 to test the interaction of RCU updaters and
5016 task-exit processing.
5017
5018 rcutorture.read_exit_burst= [KNL]
5019 The number of times in a given read-then-exit
5020 episode that a set of read-then-exit kthreads
5021 is spawned.
5022
5023 rcutorture.read_exit_delay= [KNL]
5024 The delay, in seconds, between successive
5025 read-then-exit testing episodes.
5026
5027 rcutorture.shuffle_interval= [KNL]
5028 Set task-shuffle interval (s). Shuffling tasks
5029 allows some CPUs to go into dyntick-idle mode
5030 during the rcutorture test.
5031
5032 rcutorture.shutdown_secs= [KNL]
5033 Set time (s) after boot system shutdown. This
5034 is useful for hands-off automated testing.
5035
5036 rcutorture.stall_cpu= [KNL]
5037 Duration of CPU stall (s) to test RCU CPU stall
5038 warnings, zero to disable.
5039
5040 rcutorture.stall_cpu_block= [KNL]
5041 Sleep while stalling if set. This will result
5042 in warnings from preemptible RCU in addition
5043 to any other stall-related activity.
5044
5045 rcutorture.stall_cpu_holdoff= [KNL]
5046 Time to wait (s) after boot before inducing stall.
5047
5048 rcutorture.stall_cpu_irqsoff= [KNL]
5049 Disable interrupts while stalling if set.
5050
5051 rcutorture.stall_gp_kthread= [KNL]
5052 Duration (s) of forced sleep within RCU
5053 grace-period kthread to test RCU CPU stall
5054 warnings, zero to disable. If both stall_cpu
5055 and stall_gp_kthread are specified, the
5056 kthread is starved first, then the CPU.
5057
5058 rcutorture.stat_interval= [KNL]
5059 Time (s) between statistics printk()s.
5060
5061 rcutorture.stutter= [KNL]
5062 Time (s) to stutter testing, for example, specifying
5063 five seconds causes the test to run for five seconds,
5064 wait for five seconds, and so on. This tests RCU's
5065 ability to transition abruptly to and from idle.
5066
5067 rcutorture.test_boost= [KNL]
5068 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
5069 "Maybe" means test if the RCU implementation
5070 under test support RCU priority boosting.
5071
5072 rcutorture.test_boost_duration= [KNL]
5073 Duration (s) of each individual boost test.
5074
5075 rcutorture.test_boost_interval= [KNL]
5076 Interval (s) between each boost test.
5077
5078 rcutorture.test_no_idle_hz= [KNL]
5079 Test RCU's dyntick-idle handling. See also the
5080 rcutorture.shuffle_interval parameter.
5081
5082 rcutorture.torture_type= [KNL]
5083 Specify the RCU implementation to test.
5084
5085 rcutorture.verbose= [KNL]
5086 Enable additional printk() statements.
5087
5088 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
5089 Dump ftrace buffer after reporting RCU CPU
5090 stall warning.
5091
5092 rcupdate.rcu_cpu_stall_suppress= [KNL]
5093 Suppress RCU CPU stall warning messages.
5094
5095 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
5096 Suppress RCU CPU stall warning messages and
5097 rcutorture writer stall warnings that occur
5098 during early boot, that is, during the time
5099 before the init task is spawned.
5100
5101 rcupdate.rcu_cpu_stall_timeout= [KNL]
5102 Set timeout for RCU CPU stall warning messages.
5103 The value is in seconds and the maximum allowed
5104 value is 300 seconds.
5105
5106 rcupdate.rcu_exp_cpu_stall_timeout= [KNL]
5107 Set timeout for expedited RCU CPU stall warning
5108 messages. The value is in milliseconds
5109 and the maximum allowed value is 21000
5110 milliseconds. Please note that this value is
5111 adjusted to an arch timer tick resolution.
5112 Setting this to zero causes the value from
5113 rcupdate.rcu_cpu_stall_timeout to be used (after
5114 conversion from seconds to milliseconds).
5115
5116 rcupdate.rcu_expedited= [KNL]
5117 Use expedited grace-period primitives, for
5118 example, synchronize_rcu_expedited() instead
5119 of synchronize_rcu(). This reduces latency,
5120 but can increase CPU utilization, degrade
5121 real-time latency, and degrade energy efficiency.
5122 No effect on CONFIG_TINY_RCU kernels.
5123
5124 rcupdate.rcu_normal= [KNL]
5125 Use only normal grace-period primitives,
5126 for example, synchronize_rcu() instead of
5127 synchronize_rcu_expedited(). This improves
5128 real-time latency, CPU utilization, and
5129 energy efficiency, but can expose users to
5130 increased grace-period latency. This parameter
5131 overrides rcupdate.rcu_expedited. No effect on
5132 CONFIG_TINY_RCU kernels.
5133
5134 rcupdate.rcu_normal_after_boot= [KNL]
5135 Once boot has completed (that is, after
5136 rcu_end_inkernel_boot() has been invoked), use
5137 only normal grace-period primitives. No effect
5138 on CONFIG_TINY_RCU kernels.
5139
5140 But note that CONFIG_PREEMPT_RT=y kernels enables
5141 this kernel boot parameter, forcibly setting
5142 it to the value one, that is, converting any
5143 post-boot attempt at an expedited RCU grace
5144 period to instead use normal non-expedited
5145 grace-period processing.
5146
5147 rcupdate.rcu_task_collapse_lim= [KNL]
5148 Set the maximum number of callbacks present
5149 at the beginning of a grace period that allows
5150 the RCU Tasks flavors to collapse back to using
5151 a single callback queue. This switching only
5152 occurs when rcupdate.rcu_task_enqueue_lim is
5153 set to the default value of -1.
5154
5155 rcupdate.rcu_task_contend_lim= [KNL]
5156 Set the minimum number of callback-queuing-time
5157 lock-contention events per jiffy required to
5158 cause the RCU Tasks flavors to switch to per-CPU
5159 callback queuing. This switching only occurs
5160 when rcupdate.rcu_task_enqueue_lim is set to
5161 the default value of -1.
5162
5163 rcupdate.rcu_task_enqueue_lim= [KNL]
5164 Set the number of callback queues to use for the
5165 RCU Tasks family of RCU flavors. The default
5166 of -1 allows this to be automatically (and
5167 dynamically) adjusted. This parameter is intended
5168 for use in testing.
5169
5170 rcupdate.rcu_task_ipi_delay= [KNL]
5171 Set time in jiffies during which RCU tasks will
5172 avoid sending IPIs, starting with the beginning
5173 of a given grace period. Setting a large
5174 number avoids disturbing real-time workloads,
5175 but lengthens grace periods.
5176
5177 rcupdate.rcu_task_stall_info= [KNL]
5178 Set initial timeout in jiffies for RCU task stall
5179 informational messages, which give some indication
5180 of the problem for those not patient enough to
5181 wait for ten minutes. Informational messages are
5182 only printed prior to the stall-warning message
5183 for a given grace period. Disable with a value
5184 less than or equal to zero. Defaults to ten
5185 seconds. A change in value does not take effect
5186 until the beginning of the next grace period.
5187
5188 rcupdate.rcu_task_stall_info_mult= [KNL]
5189 Multiplier for time interval between successive
5190 RCU task stall informational messages for a given
5191 RCU tasks grace period. This value is clamped
5192 to one through ten, inclusive. It defaults to
5193 the value three, so that the first informational
5194 message is printed 10 seconds into the grace
5195 period, the second at 40 seconds, the third at
5196 160 seconds, and then the stall warning at 600
5197 seconds would prevent a fourth at 640 seconds.
5198
5199 rcupdate.rcu_task_stall_timeout= [KNL]
5200 Set timeout in jiffies for RCU task stall
5201 warning messages. Disable with a value less
5202 than or equal to zero. Defaults to ten minutes.
5203 A change in value does not take effect until
5204 the beginning of the next grace period.
5205
5206 rcupdate.rcu_self_test= [KNL]
5207 Run the RCU early boot self tests
5208
5209 rdinit= [KNL]
5210 Format: <full_path>
5211 Run specified binary instead of /init from the ramdisk,
5212 used for early userspace startup. See initrd.
5213
5214 rdrand= [X86]
5215 force - Override the decision by the kernel to hide the
5216 advertisement of RDRAND support (this affects
5217 certain AMD processors because of buggy BIOS
5218 support, specifically around the suspend/resume
5219 path).
5220
5221 rdt= [HW,X86,RDT]
5222 Turn on/off individual RDT features. List is:
5223 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
5224 mba.
5225 E.g. to turn on cmt and turn off mba use:
5226 rdt=cmt,!mba
5227
5228 reboot= [KNL]
5229 Format (x86 or x86_64):
5230 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \
5231 [[,]s[mp]#### \
5232 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
5233 [[,]f[orce]
5234 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
5235 (prefix with 'panic_' to set mode for panic
5236 reboot only),
5237 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
5238 reboot_force is either force or not specified,
5239 reboot_cpu is s[mp]#### with #### being the processor
5240 to be used for rebooting.
5241
5242 refscale.holdoff= [KNL]
5243 Set test-start holdoff period. The purpose of
5244 this parameter is to delay the start of the
5245 test until boot completes in order to avoid
5246 interference.
5247
5248 refscale.loops= [KNL]
5249 Set the number of loops over the synchronization
5250 primitive under test. Increasing this number
5251 reduces noise due to loop start/end overhead,
5252 but the default has already reduced the per-pass
5253 noise to a handful of picoseconds on ca. 2020
5254 x86 laptops.
5255
5256 refscale.nreaders= [KNL]
5257 Set number of readers. The default value of -1
5258 selects N, where N is roughly 75% of the number
5259 of CPUs. A value of zero is an interesting choice.
5260
5261 refscale.nruns= [KNL]
5262 Set number of runs, each of which is dumped onto
5263 the console log.
5264
5265 refscale.readdelay= [KNL]
5266 Set the read-side critical-section duration,
5267 measured in microseconds.
5268
5269 refscale.scale_type= [KNL]
5270 Specify the read-protection implementation to test.
5271
5272 refscale.shutdown= [KNL]
5273 Shut down the system at the end of the performance
5274 test. This defaults to 1 (shut it down) when
5275 refscale is built into the kernel and to 0 (leave
5276 it running) when refscale is built as a module.
5277
5278 refscale.verbose= [KNL]
5279 Enable additional printk() statements.
5280
5281 refscale.verbose_batched= [KNL]
5282 Batch the additional printk() statements. If zero
5283 (the default) or negative, print everything. Otherwise,
5284 print every Nth verbose statement, where N is the value
5285 specified.
5286
5287 relax_domain_level=
5288 [KNL, SMP] Set scheduler's default relax_domain_level.
5289 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
5290
5291 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
5292 Format: <base1>,<size1>[,<base2>,<size2>,...]
5293 Reserve I/O ports or memory so the kernel won't use
5294 them. If <base> is less than 0x10000, the region
5295 is assumed to be I/O ports; otherwise it is memory.
5296
5297 reservetop= [X86-32]
5298 Format: nn[KMG]
5299 Reserves a hole at the top of the kernel virtual
5300 address space.
5301
5302 reset_devices [KNL] Force drivers to reset the underlying device
5303 during initialization.
5304
5305 resume= [SWSUSP]
5306 Specify the partition device for software suspend
5307 Format:
5308 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
5309
5310 resume_offset= [SWSUSP]
5311 Specify the offset from the beginning of the partition
5312 given by "resume=" at which the swap header is located,
5313 in <PAGE_SIZE> units (needed only for swap files).
5314 See Documentation/power/swsusp-and-swap-files.rst
5315
5316 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
5317 read the resume files
5318
5319 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
5320 Useful for devices that are detected asynchronously
5321 (e.g. USB and MMC devices).
5322
5323 retain_initrd [RAM] Keep initrd memory after extraction
5324
5325 retbleed= [X86] Control mitigation of RETBleed (Arbitrary
5326 Speculative Code Execution with Return Instructions)
5327 vulnerability.
5328
5329 AMD-based UNRET and IBPB mitigations alone do not stop
5330 sibling threads from influencing the predictions of other
5331 sibling threads. For that reason, STIBP is used on pro-
5332 cessors that support it, and mitigate SMT on processors
5333 that don't.
5334
5335 off - no mitigation
5336 auto - automatically select a migitation
5337 auto,nosmt - automatically select a mitigation,
5338 disabling SMT if necessary for
5339 the full mitigation (only on Zen1
5340 and older without STIBP).
5341 ibpb - On AMD, mitigate short speculation
5342 windows on basic block boundaries too.
5343 Safe, highest perf impact. It also
5344 enables STIBP if present. Not suitable
5345 on Intel.
5346 ibpb,nosmt - Like "ibpb" above but will disable SMT
5347 when STIBP is not available. This is
5348 the alternative for systems which do not
5349 have STIBP.
5350 unret - Force enable untrained return thunks,
5351 only effective on AMD f15h-f17h based
5352 systems.
5353 unret,nosmt - Like unret, but will disable SMT when STIBP
5354 is not available. This is the alternative for
5355 systems which do not have STIBP.
5356
5357 Selecting 'auto' will choose a mitigation method at run
5358 time according to the CPU.
5359
5360 Not specifying this option is equivalent to retbleed=auto.
5361
5362 rfkill.default_state=
5363 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
5364 etc. communication is blocked by default.
5365 1 Unblocked.
5366
5367 rfkill.master_switch_mode=
5368 0 The "airplane mode" button does nothing.
5369 1 The "airplane mode" button toggles between everything
5370 blocked and the previous configuration.
5371 2 The "airplane mode" button toggles between everything
5372 blocked and everything unblocked.
5373
5374 rhash_entries= [KNL,NET]
5375 Set number of hash buckets for route cache
5376
5377 ring3mwait=disable
5378 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
5379 CPUs.
5380
5381 ro [KNL] Mount root device read-only on boot
5382
5383 rodata= [KNL]
5384 on Mark read-only kernel memory as read-only (default).
5385 off Leave read-only kernel memory writable for debugging.
5386 full Mark read-only kernel memory and aliases as read-only
5387 [arm64]
5388
5389 rockchip.usb_uart
5390 Enable the uart passthrough on the designated usb port
5391 on Rockchip SoCs. When active, the signals of the
5392 debug-uart get routed to the D+ and D- pins of the usb
5393 port and the regular usb controller gets disabled.
5394
5395 root= [KNL] Root filesystem
5396 See name_to_dev_t comment in init/do_mounts.c.
5397
5398 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
5399 mount the root filesystem
5400
5401 rootflags= [KNL] Set root filesystem mount option string
5402
5403 rootfstype= [KNL] Set root filesystem type
5404
5405 rootwait [KNL] Wait (indefinitely) for root device to show up.
5406 Useful for devices that are detected asynchronously
5407 (e.g. USB and MMC devices).
5408
5409 rproc_mem=nn[KMG][@address]
5410 [KNL,ARM,CMA] Remoteproc physical memory block.
5411 Memory area to be used by remote processor image,
5412 managed by CMA.
5413
5414 rw [KNL] Mount root device read-write on boot
5415
5416 S [KNL] Run init in single mode
5417
5418 s390_iommu= [HW,S390]
5419 Set s390 IOTLB flushing mode
5420 strict
5421 With strict flushing every unmap operation will result in
5422 an IOTLB flush. Default is lazy flushing before reuse,
5423 which is faster.
5424
5425 s390_iommu_aperture= [KNL,S390]
5426 Specifies the size of the per device DMA address space
5427 accessible through the DMA and IOMMU APIs as a decimal
5428 factor of the size of main memory.
5429 The default is 1 meaning that one can concurrently use
5430 as many DMA addresses as physical memory is installed,
5431 if supported by hardware, and thus map all of memory
5432 once. With a value of 2 one can map all of memory twice
5433 and so on. As a special case a factor of 0 imposes no
5434 restrictions other than those given by hardware at the
5435 cost of significant additional memory use for tables.
5436
5437 sa1100ir [NET]
5438 See drivers/net/irda/sa1100_ir.c.
5439
5440 sched_verbose [KNL] Enables verbose scheduler debug messages.
5441
5442 schedstats= [KNL,X86] Enable or disable scheduled statistics.
5443 Allowed values are enable and disable. This feature
5444 incurs a small amount of overhead in the scheduler
5445 but is useful for debugging and performance tuning.
5446
5447 sched_thermal_decay_shift=
5448 [KNL, SMP] Set a decay shift for scheduler thermal
5449 pressure signal. Thermal pressure signal follows the
5450 default decay period of other scheduler pelt
5451 signals(usually 32 ms but configurable). Setting
5452 sched_thermal_decay_shift will left shift the decay
5453 period for the thermal pressure signal by the shift
5454 value.
5455 i.e. with the default pelt decay period of 32 ms
5456 sched_thermal_decay_shift thermal pressure decay pr
5457 1 64 ms
5458 2 128 ms
5459 and so on.
5460 Format: integer between 0 and 10
5461 Default is 0.
5462
5463 scftorture.holdoff= [KNL]
5464 Number of seconds to hold off before starting
5465 test. Defaults to zero for module insertion and
5466 to 10 seconds for built-in smp_call_function()
5467 tests.
5468
5469 scftorture.longwait= [KNL]
5470 Request ridiculously long waits randomly selected
5471 up to the chosen limit in seconds. Zero (the
5472 default) disables this feature. Please note
5473 that requesting even small non-zero numbers of
5474 seconds can result in RCU CPU stall warnings,
5475 softlockup complaints, and so on.
5476
5477 scftorture.nthreads= [KNL]
5478 Number of kthreads to spawn to invoke the
5479 smp_call_function() family of functions.
5480 The default of -1 specifies a number of kthreads
5481 equal to the number of CPUs.
5482
5483 scftorture.onoff_holdoff= [KNL]
5484 Number seconds to wait after the start of the
5485 test before initiating CPU-hotplug operations.
5486
5487 scftorture.onoff_interval= [KNL]
5488 Number seconds to wait between successive
5489 CPU-hotplug operations. Specifying zero (which
5490 is the default) disables CPU-hotplug operations.
5491
5492 scftorture.shutdown_secs= [KNL]
5493 The number of seconds following the start of the
5494 test after which to shut down the system. The
5495 default of zero avoids shutting down the system.
5496 Non-zero values are useful for automated tests.
5497
5498 scftorture.stat_interval= [KNL]
5499 The number of seconds between outputting the
5500 current test statistics to the console. A value
5501 of zero disables statistics output.
5502
5503 scftorture.stutter_cpus= [KNL]
5504 The number of jiffies to wait between each change
5505 to the set of CPUs under test.
5506
5507 scftorture.use_cpus_read_lock= [KNL]
5508 Use use_cpus_read_lock() instead of the default
5509 preempt_disable() to disable CPU hotplug
5510 while invoking one of the smp_call_function*()
5511 functions.
5512
5513 scftorture.verbose= [KNL]
5514 Enable additional printk() statements.
5515
5516 scftorture.weight_single= [KNL]
5517 The probability weighting to use for the
5518 smp_call_function_single() function with a zero
5519 "wait" parameter. A value of -1 selects the
5520 default if all other weights are -1. However,
5521 if at least one weight has some other value, a
5522 value of -1 will instead select a weight of zero.
5523
5524 scftorture.weight_single_wait= [KNL]
5525 The probability weighting to use for the
5526 smp_call_function_single() function with a
5527 non-zero "wait" parameter. See weight_single.
5528
5529 scftorture.weight_many= [KNL]
5530 The probability weighting to use for the
5531 smp_call_function_many() function with a zero
5532 "wait" parameter. See weight_single.
5533 Note well that setting a high probability for
5534 this weighting can place serious IPI load
5535 on the system.
5536
5537 scftorture.weight_many_wait= [KNL]
5538 The probability weighting to use for the
5539 smp_call_function_many() function with a
5540 non-zero "wait" parameter. See weight_single
5541 and weight_many.
5542
5543 scftorture.weight_all= [KNL]
5544 The probability weighting to use for the
5545 smp_call_function_all() function with a zero
5546 "wait" parameter. See weight_single and
5547 weight_many.
5548
5549 scftorture.weight_all_wait= [KNL]
5550 The probability weighting to use for the
5551 smp_call_function_all() function with a
5552 non-zero "wait" parameter. See weight_single
5553 and weight_many.
5554
5555 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
5556 xtime_lock contention on larger systems, and/or RCU lock
5557 contention on all systems with CONFIG_MAXSMP set.
5558 Format: { "0" | "1" }
5559 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
5560 1 -- enable.
5561 Note: increases power consumption, thus should only be
5562 enabled if running jitter sensitive (HPC/RT) workloads.
5563
5564 security= [SECURITY] Choose a legacy "major" security module to
5565 enable at boot. This has been deprecated by the
5566 "lsm=" parameter.
5567
5568 selinux= [SELINUX] Disable or enable SELinux at boot time.
5569 Format: { "0" | "1" }
5570 See security/selinux/Kconfig help text.
5571 0 -- disable.
5572 1 -- enable.
5573 Default value is 1.
5574
5575 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
5576 Format: { "0" | "1" }
5577 See security/apparmor/Kconfig help text
5578 0 -- disable.
5579 1 -- enable.
5580 Default value is set via kernel config option.
5581
5582 serialnumber [BUGS=X86-32]
5583
5584 sev=option[,option...] [X86-64] See Documentation/x86/x86_64/boot-options.rst
5585
5586 shapers= [NET]
5587 Maximal number of shapers.
5588
5589 simeth= [IA-64]
5590 simscsi=
5591
5592 slram= [HW,MTD]
5593
5594 slab_merge [MM]
5595 Enable merging of slabs with similar size when the
5596 kernel is built without CONFIG_SLAB_MERGE_DEFAULT.
5597
5598 slab_nomerge [MM]
5599 Disable merging of slabs with similar size. May be
5600 necessary if there is some reason to distinguish
5601 allocs to different slabs, especially in hardened
5602 environments where the risk of heap overflows and
5603 layout control by attackers can usually be
5604 frustrated by disabling merging. This will reduce
5605 most of the exposure of a heap attack to a single
5606 cache (risks via metadata attacks are mostly
5607 unchanged). Debug options disable merging on their
5608 own.
5609 For more information see Documentation/mm/slub.rst.
5610
5611 slab_max_order= [MM, SLAB]
5612 Determines the maximum allowed order for slabs.
5613 A high setting may cause OOMs due to memory
5614 fragmentation. Defaults to 1 for systems with
5615 more than 32MB of RAM, 0 otherwise.
5616
5617 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
5618 Enabling slub_debug allows one to determine the
5619 culprit if slab objects become corrupted. Enabling
5620 slub_debug can create guard zones around objects and
5621 may poison objects when not in use. Also tracks the
5622 last alloc / free. For more information see
5623 Documentation/mm/slub.rst.
5624
5625 slub_max_order= [MM, SLUB]
5626 Determines the maximum allowed order for slabs.
5627 A high setting may cause OOMs due to memory
5628 fragmentation. For more information see
5629 Documentation/mm/slub.rst.
5630
5631 slub_min_objects= [MM, SLUB]
5632 The minimum number of objects per slab. SLUB will
5633 increase the slab order up to slub_max_order to
5634 generate a sufficiently large slab able to contain
5635 the number of objects indicated. The higher the number
5636 of objects the smaller the overhead of tracking slabs
5637 and the less frequently locks need to be acquired.
5638 For more information see Documentation/mm/slub.rst.
5639
5640 slub_min_order= [MM, SLUB]
5641 Determines the minimum page order for slabs. Must be
5642 lower than slub_max_order.
5643 For more information see Documentation/mm/slub.rst.
5644
5645 slub_merge [MM, SLUB]
5646 Same with slab_merge.
5647
5648 slub_nomerge [MM, SLUB]
5649 Same with slab_nomerge. This is supported for legacy.
5650 See slab_nomerge for more information.
5651
5652 smart2= [HW]
5653 Format: <io1>[,<io2>[,...,<io8>]]
5654
5655 smp.csd_lock_timeout= [KNL]
5656 Specify the period of time in milliseconds
5657 that smp_call_function() and friends will wait
5658 for a CPU to release the CSD lock. This is
5659 useful when diagnosing bugs involving CPUs
5660 disabling interrupts for extended periods
5661 of time. Defaults to 5,000 milliseconds, and
5662 setting a value of zero disables this feature.
5663 This feature may be more efficiently disabled
5664 using the csdlock_debug- kernel parameter.
5665
5666 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
5667 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
5668 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
5669 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
5670 smsc-ircc2.ircc_irq= [HW] IRQ line
5671 smsc-ircc2.ircc_dma= [HW] DMA channel
5672 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
5673 0: Toshiba Satellite 1800 (GP data pin select)
5674 1: Fast pin select (default)
5675 2: ATC IRMode
5676
5677 smt= [KNL,S390] Set the maximum number of threads (logical
5678 CPUs) to use per physical CPU on systems capable of
5679 symmetric multithreading (SMT). Will be capped to the
5680 actual hardware limit.
5681 Format: <integer>
5682 Default: -1 (no limit)
5683
5684 softlockup_panic=
5685 [KNL] Should the soft-lockup detector generate panics.
5686 Format: 0 | 1
5687
5688 A value of 1 instructs the soft-lockup detector
5689 to panic the machine when a soft-lockup occurs. It is
5690 also controlled by the kernel.softlockup_panic sysctl
5691 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
5692 respective build-time switch to that functionality.
5693
5694 softlockup_all_cpu_backtrace=
5695 [KNL] Should the soft-lockup detector generate
5696 backtraces on all cpus.
5697 Format: 0 | 1
5698
5699 sonypi.*= [HW] Sony Programmable I/O Control Device driver
5700 See Documentation/admin-guide/laptops/sonypi.rst
5701
5702 spectre_v2= [X86] Control mitigation of Spectre variant 2
5703 (indirect branch speculation) vulnerability.
5704 The default operation protects the kernel from
5705 user space attacks.
5706
5707 on - unconditionally enable, implies
5708 spectre_v2_user=on
5709 off - unconditionally disable, implies
5710 spectre_v2_user=off
5711 auto - kernel detects whether your CPU model is
5712 vulnerable
5713
5714 Selecting 'on' will, and 'auto' may, choose a
5715 mitigation method at run time according to the
5716 CPU, the available microcode, the setting of the
5717 CONFIG_RETPOLINE configuration option, and the
5718 compiler with which the kernel was built.
5719
5720 Selecting 'on' will also enable the mitigation
5721 against user space to user space task attacks.
5722
5723 Selecting 'off' will disable both the kernel and
5724 the user space protections.
5725
5726 Specific mitigations can also be selected manually:
5727
5728 retpoline - replace indirect branches
5729 retpoline,generic - Retpolines
5730 retpoline,lfence - LFENCE; indirect branch
5731 retpoline,amd - alias for retpoline,lfence
5732 eibrs - enhanced IBRS
5733 eibrs,retpoline - enhanced IBRS + Retpolines
5734 eibrs,lfence - enhanced IBRS + LFENCE
5735 ibrs - use IBRS to protect kernel
5736
5737 Not specifying this option is equivalent to
5738 spectre_v2=auto.
5739
5740 spectre_v2_user=
5741 [X86] Control mitigation of Spectre variant 2
5742 (indirect branch speculation) vulnerability between
5743 user space tasks
5744
5745 on - Unconditionally enable mitigations. Is
5746 enforced by spectre_v2=on
5747
5748 off - Unconditionally disable mitigations. Is
5749 enforced by spectre_v2=off
5750
5751 prctl - Indirect branch speculation is enabled,
5752 but mitigation can be enabled via prctl
5753 per thread. The mitigation control state
5754 is inherited on fork.
5755
5756 prctl,ibpb
5757 - Like "prctl" above, but only STIBP is
5758 controlled per thread. IBPB is issued
5759 always when switching between different user
5760 space processes.
5761
5762 seccomp
5763 - Same as "prctl" above, but all seccomp
5764 threads will enable the mitigation unless
5765 they explicitly opt out.
5766
5767 seccomp,ibpb
5768 - Like "seccomp" above, but only STIBP is
5769 controlled per thread. IBPB is issued
5770 always when switching between different
5771 user space processes.
5772
5773 auto - Kernel selects the mitigation depending on
5774 the available CPU features and vulnerability.
5775
5776 Default mitigation: "prctl"
5777
5778 Not specifying this option is equivalent to
5779 spectre_v2_user=auto.
5780
5781 spec_store_bypass_disable=
5782 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
5783 (Speculative Store Bypass vulnerability)
5784
5785 Certain CPUs are vulnerable to an exploit against a
5786 a common industry wide performance optimization known
5787 as "Speculative Store Bypass" in which recent stores
5788 to the same memory location may not be observed by
5789 later loads during speculative execution. The idea
5790 is that such stores are unlikely and that they can
5791 be detected prior to instruction retirement at the
5792 end of a particular speculation execution window.
5793
5794 In vulnerable processors, the speculatively forwarded
5795 store can be used in a cache side channel attack, for
5796 example to read memory to which the attacker does not
5797 directly have access (e.g. inside sandboxed code).
5798
5799 This parameter controls whether the Speculative Store
5800 Bypass optimization is used.
5801
5802 On x86 the options are:
5803
5804 on - Unconditionally disable Speculative Store Bypass
5805 off - Unconditionally enable Speculative Store Bypass
5806 auto - Kernel detects whether the CPU model contains an
5807 implementation of Speculative Store Bypass and
5808 picks the most appropriate mitigation. If the
5809 CPU is not vulnerable, "off" is selected. If the
5810 CPU is vulnerable the default mitigation is
5811 architecture and Kconfig dependent. See below.
5812 prctl - Control Speculative Store Bypass per thread
5813 via prctl. Speculative Store Bypass is enabled
5814 for a process by default. The state of the control
5815 is inherited on fork.
5816 seccomp - Same as "prctl" above, but all seccomp threads
5817 will disable SSB unless they explicitly opt out.
5818
5819 Default mitigations:
5820 X86: "prctl"
5821
5822 On powerpc the options are:
5823
5824 on,auto - On Power8 and Power9 insert a store-forwarding
5825 barrier on kernel entry and exit. On Power7
5826 perform a software flush on kernel entry and
5827 exit.
5828 off - No action.
5829
5830 Not specifying this option is equivalent to
5831 spec_store_bypass_disable=auto.
5832
5833 spia_io_base= [HW,MTD]
5834 spia_fio_base=
5835 spia_pedr=
5836 spia_peddr=
5837
5838 split_lock_detect=
5839 [X86] Enable split lock detection or bus lock detection
5840
5841 When enabled (and if hardware support is present), atomic
5842 instructions that access data across cache line
5843 boundaries will result in an alignment check exception
5844 for split lock detection or a debug exception for
5845 bus lock detection.
5846
5847 off - not enabled
5848
5849 warn - the kernel will emit rate-limited warnings
5850 about applications triggering the #AC
5851 exception or the #DB exception. This mode is
5852 the default on CPUs that support split lock
5853 detection or bus lock detection. Default
5854 behavior is by #AC if both features are
5855 enabled in hardware.
5856
5857 fatal - the kernel will send SIGBUS to applications
5858 that trigger the #AC exception or the #DB
5859 exception. Default behavior is by #AC if
5860 both features are enabled in hardware.
5861
5862 ratelimit:N -
5863 Set system wide rate limit to N bus locks
5864 per second for bus lock detection.
5865 0 < N <= 1000.
5866
5867 N/A for split lock detection.
5868
5869
5870 If an #AC exception is hit in the kernel or in
5871 firmware (i.e. not while executing in user mode)
5872 the kernel will oops in either "warn" or "fatal"
5873 mode.
5874
5875 #DB exception for bus lock is triggered only when
5876 CPL > 0.
5877
5878 srbds= [X86,INTEL]
5879 Control the Special Register Buffer Data Sampling
5880 (SRBDS) mitigation.
5881
5882 Certain CPUs are vulnerable to an MDS-like
5883 exploit which can leak bits from the random
5884 number generator.
5885
5886 By default, this issue is mitigated by
5887 microcode. However, the microcode fix can cause
5888 the RDRAND and RDSEED instructions to become
5889 much slower. Among other effects, this will
5890 result in reduced throughput from /dev/urandom.
5891
5892 The microcode mitigation can be disabled with
5893 the following option:
5894
5895 off: Disable mitigation and remove
5896 performance impact to RDRAND and RDSEED
5897
5898 srcutree.big_cpu_lim [KNL]
5899 Specifies the number of CPUs constituting a
5900 large system, such that srcu_struct structures
5901 should immediately allocate an srcu_node array.
5902 This kernel-boot parameter defaults to 128,
5903 but takes effect only when the low-order four
5904 bits of srcutree.convert_to_big is equal to 3
5905 (decide at boot).
5906
5907 srcutree.convert_to_big [KNL]
5908 Specifies under what conditions an SRCU tree
5909 srcu_struct structure will be converted to big
5910 form, that is, with an rcu_node tree:
5911
5912 0: Never.
5913 1: At init_srcu_struct() time.
5914 2: When rcutorture decides to.
5915 3: Decide at boot time (default).
5916 0x1X: Above plus if high contention.
5917
5918 Either way, the srcu_node tree will be sized based
5919 on the actual runtime number of CPUs (nr_cpu_ids)
5920 instead of the compile-time CONFIG_NR_CPUS.
5921
5922 srcutree.counter_wrap_check [KNL]
5923 Specifies how frequently to check for
5924 grace-period sequence counter wrap for the
5925 srcu_data structure's ->srcu_gp_seq_needed field.
5926 The greater the number of bits set in this kernel
5927 parameter, the less frequently counter wrap will
5928 be checked for. Note that the bottom two bits
5929 are ignored.
5930
5931 srcutree.exp_holdoff [KNL]
5932 Specifies how many nanoseconds must elapse
5933 since the end of the last SRCU grace period for
5934 a given srcu_struct until the next normal SRCU
5935 grace period will be considered for automatic
5936 expediting. Set to zero to disable automatic
5937 expediting.
5938
5939 srcutree.srcu_max_nodelay [KNL]
5940 Specifies the number of no-delay instances
5941 per jiffy for which the SRCU grace period
5942 worker thread will be rescheduled with zero
5943 delay. Beyond this limit, worker thread will
5944 be rescheduled with a sleep delay of one jiffy.
5945
5946 srcutree.srcu_max_nodelay_phase [KNL]
5947 Specifies the per-grace-period phase, number of
5948 non-sleeping polls of readers. Beyond this limit,
5949 grace period worker thread will be rescheduled
5950 with a sleep delay of one jiffy, between each
5951 rescan of the readers, for a grace period phase.
5952
5953 srcutree.srcu_retry_check_delay [KNL]
5954 Specifies number of microseconds of non-sleeping
5955 delay between each non-sleeping poll of readers.
5956
5957 srcutree.small_contention_lim [KNL]
5958 Specifies the number of update-side contention
5959 events per jiffy will be tolerated before
5960 initiating a conversion of an srcu_struct
5961 structure to big form. Note that the value of
5962 srcutree.convert_to_big must have the 0x10 bit
5963 set for contention-based conversions to occur.
5964
5965 ssbd= [ARM64,HW]
5966 Speculative Store Bypass Disable control
5967
5968 On CPUs that are vulnerable to the Speculative
5969 Store Bypass vulnerability and offer a
5970 firmware based mitigation, this parameter
5971 indicates how the mitigation should be used:
5972
5973 force-on: Unconditionally enable mitigation for
5974 for both kernel and userspace
5975 force-off: Unconditionally disable mitigation for
5976 for both kernel and userspace
5977 kernel: Always enable mitigation in the
5978 kernel, and offer a prctl interface
5979 to allow userspace to register its
5980 interest in being mitigated too.
5981
5982 stack_guard_gap= [MM]
5983 override the default stack gap protection. The value
5984 is in page units and it defines how many pages prior
5985 to (for stacks growing down) resp. after (for stacks
5986 growing up) the main stack are reserved for no other
5987 mapping. Default value is 256 pages.
5988
5989 stack_depot_disable= [KNL]
5990 Setting this to true through kernel command line will
5991 disable the stack depot thereby saving the static memory
5992 consumed by the stack hash table. By default this is set
5993 to false.
5994
5995 stacktrace [FTRACE]
5996 Enabled the stack tracer on boot up.
5997
5998 stacktrace_filter=[function-list]
5999 [FTRACE] Limit the functions that the stack tracer
6000 will trace at boot up. function-list is a comma-separated
6001 list of functions. This list can be changed at run
6002 time by the stack_trace_filter file in the debugfs
6003 tracing directory. Note, this enables stack tracing
6004 and the stacktrace above is not needed.
6005
6006 sti= [PARISC,HW]
6007 Format: <num>
6008 Set the STI (builtin display/keyboard on the HP-PARISC
6009 machines) console (graphic card) which should be used
6010 as the initial boot-console.
6011 See also comment in drivers/video/console/sticore.c.
6012
6013 sti_font= [HW]
6014 See comment in drivers/video/console/sticore.c.
6015
6016 stifb= [HW]
6017 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
6018
6019 strict_sas_size=
6020 [X86]
6021 Format: <bool>
6022 Enable or disable strict sigaltstack size checks
6023 against the required signal frame size which
6024 depends on the supported FPU features. This can
6025 be used to filter out binaries which have
6026 not yet been made aware of AT_MINSIGSTKSZ.
6027
6028 sunrpc.min_resvport=
6029 sunrpc.max_resvport=
6030 [NFS,SUNRPC]
6031 SunRPC servers often require that client requests
6032 originate from a privileged port (i.e. a port in the
6033 range 0 < portnr < 1024).
6034 An administrator who wishes to reserve some of these
6035 ports for other uses may adjust the range that the
6036 kernel's sunrpc client considers to be privileged
6037 using these two parameters to set the minimum and
6038 maximum port values.
6039
6040 sunrpc.svc_rpc_per_connection_limit=
6041 [NFS,SUNRPC]
6042 Limit the number of requests that the server will
6043 process in parallel from a single connection.
6044 The default value is 0 (no limit).
6045
6046 sunrpc.pool_mode=
6047 [NFS]
6048 Control how the NFS server code allocates CPUs to
6049 service thread pools. Depending on how many NICs
6050 you have and where their interrupts are bound, this
6051 option will affect which CPUs will do NFS serving.
6052 Note: this parameter cannot be changed while the
6053 NFS server is running.
6054
6055 auto the server chooses an appropriate mode
6056 automatically using heuristics
6057 global a single global pool contains all CPUs
6058 percpu one pool for each CPU
6059 pernode one pool for each NUMA node (equivalent
6060 to global on non-NUMA machines)
6061
6062 sunrpc.tcp_slot_table_entries=
6063 sunrpc.udp_slot_table_entries=
6064 [NFS,SUNRPC]
6065 Sets the upper limit on the number of simultaneous
6066 RPC calls that can be sent from the client to a
6067 server. Increasing these values may allow you to
6068 improve throughput, but will also increase the
6069 amount of memory reserved for use by the client.
6070
6071 suspend.pm_test_delay=
6072 [SUSPEND]
6073 Sets the number of seconds to remain in a suspend test
6074 mode before resuming the system (see
6075 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
6076 is set. Default value is 5.
6077
6078 svm= [PPC]
6079 Format: { on | off | y | n | 1 | 0 }
6080 This parameter controls use of the Protected
6081 Execution Facility on pSeries.
6082
6083 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
6084 Format: { <int> [,<int>] | force | noforce }
6085 <int> -- Number of I/O TLB slabs
6086 <int> -- Second integer after comma. Number of swiotlb
6087 areas with their own lock. Will be rounded up
6088 to a power of 2.
6089 force -- force using of bounce buffers even if they
6090 wouldn't be automatically used by the kernel
6091 noforce -- Never use bounce buffers (for debugging)
6092
6093 switches= [HW,M68k]
6094
6095 sysctl.*= [KNL]
6096 Set a sysctl parameter, right before loading the init
6097 process, as if the value was written to the respective
6098 /proc/sys/... file. Both '.' and '/' are recognized as
6099 separators. Unrecognized parameters and invalid values
6100 are reported in the kernel log. Sysctls registered
6101 later by a loaded module cannot be set this way.
6102 Example: sysctl.vm.swappiness=40
6103
6104 sysfs.deprecated=0|1 [KNL]
6105 Enable/disable old style sysfs layout for old udev
6106 on older distributions. When this option is enabled
6107 very new udev will not work anymore. When this option
6108 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
6109 in older udev will not work anymore.
6110 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
6111 the kernel configuration.
6112
6113 sysrq_always_enabled
6114 [KNL]
6115 Ignore sysrq setting - this boot parameter will
6116 neutralize any effect of /proc/sys/kernel/sysrq.
6117 Useful for debugging.
6118
6119 tcpmhash_entries= [KNL,NET]
6120 Set the number of tcp_metrics_hash slots.
6121 Default value is 8192 or 16384 depending on total
6122 ram pages. This is used to specify the TCP metrics
6123 cache size. See Documentation/networking/ip-sysctl.rst
6124 "tcp_no_metrics_save" section for more details.
6125
6126 tdfx= [HW,DRM]
6127
6128 test_suspend= [SUSPEND]
6129 Format: { "mem" | "standby" | "freeze" }[,N]
6130 Specify "mem" (for Suspend-to-RAM) or "standby" (for
6131 standby suspend) or "freeze" (for suspend type freeze)
6132 as the system sleep state during system startup with
6133 the optional capability to repeat N number of times.
6134 The system is woken from this state using a
6135 wakeup-capable RTC alarm.
6136
6137 thash_entries= [KNL,NET]
6138 Set number of hash buckets for TCP connection
6139
6140 thermal.act= [HW,ACPI]
6141 -1: disable all active trip points in all thermal zones
6142 <degrees C>: override all lowest active trip points
6143
6144 thermal.crt= [HW,ACPI]
6145 -1: disable all critical trip points in all thermal zones
6146 <degrees C>: override all critical trip points
6147
6148 thermal.nocrt= [HW,ACPI]
6149 Set to disable actions on ACPI thermal zone
6150 critical and hot trip points.
6151
6152 thermal.off= [HW,ACPI]
6153 1: disable ACPI thermal control
6154
6155 thermal.psv= [HW,ACPI]
6156 -1: disable all passive trip points
6157 <degrees C>: override all passive trip points to this
6158 value
6159
6160 thermal.tzp= [HW,ACPI]
6161 Specify global default ACPI thermal zone polling rate
6162 <deci-seconds>: poll all this frequency
6163 0: no polling (default)
6164
6165 threadirqs [KNL]
6166 Force threading of all interrupt handlers except those
6167 marked explicitly IRQF_NO_THREAD.
6168
6169 topology= [S390]
6170 Format: {off | on}
6171 Specify if the kernel should make use of the cpu
6172 topology information if the hardware supports this.
6173 The scheduler will make use of this information and
6174 e.g. base its process migration decisions on it.
6175 Default is on.
6176
6177 topology_updates= [KNL, PPC, NUMA]
6178 Format: {off}
6179 Specify if the kernel should ignore (off)
6180 topology updates sent by the hypervisor to this
6181 LPAR.
6182
6183 torture.disable_onoff_at_boot= [KNL]
6184 Prevent the CPU-hotplug component of torturing
6185 until after init has spawned.
6186
6187 torture.ftrace_dump_at_shutdown= [KNL]
6188 Dump the ftrace buffer at torture-test shutdown,
6189 even if there were no errors. This can be a
6190 very costly operation when many torture tests
6191 are running concurrently, especially on systems
6192 with rotating-rust storage.
6193
6194 torture.verbose_sleep_frequency= [KNL]
6195 Specifies how many verbose printk()s should be
6196 emitted between each sleep. The default of zero
6197 disables verbose-printk() sleeping.
6198
6199 torture.verbose_sleep_duration= [KNL]
6200 Duration of each verbose-printk() sleep in jiffies.
6201
6202 tp720= [HW,PS2]
6203
6204 tpm_suspend_pcr=[HW,TPM]
6205 Format: integer pcr id
6206 Specify that at suspend time, the tpm driver
6207 should extend the specified pcr with zeros,
6208 as a workaround for some chips which fail to
6209 flush the last written pcr on TPM_SaveState.
6210 This will guarantee that all the other pcrs
6211 are saved.
6212
6213 tp_printk [FTRACE]
6214 Have the tracepoints sent to printk as well as the
6215 tracing ring buffer. This is useful for early boot up
6216 where the system hangs or reboots and does not give the
6217 option for reading the tracing buffer or performing a
6218 ftrace_dump_on_oops.
6219
6220 To turn off having tracepoints sent to printk,
6221 echo 0 > /proc/sys/kernel/tracepoint_printk
6222 Note, echoing 1 into this file without the
6223 tracepoint_printk kernel cmdline option has no effect.
6224
6225 The tp_printk_stop_on_boot (see below) can also be used
6226 to stop the printing of events to console at
6227 late_initcall_sync.
6228
6229 ** CAUTION **
6230
6231 Having tracepoints sent to printk() and activating high
6232 frequency tracepoints such as irq or sched, can cause
6233 the system to live lock.
6234
6235 tp_printk_stop_on_boot [FTRACE]
6236 When tp_printk (above) is set, it can cause a lot of noise
6237 on the console. It may be useful to only include the
6238 printing of events during boot up, as user space may
6239 make the system inoperable.
6240
6241 This command line option will stop the printing of events
6242 to console at the late_initcall_sync() time frame.
6243
6244 trace_buf_size=nn[KMG]
6245 [FTRACE] will set tracing buffer size on each cpu.
6246
6247 trace_clock= [FTRACE] Set the clock used for tracing events
6248 at boot up.
6249 local - Use the per CPU time stamp counter
6250 (converted into nanoseconds). Fast, but
6251 depending on the architecture, may not be
6252 in sync between CPUs.
6253 global - Event time stamps are synchronize across
6254 CPUs. May be slower than the local clock,
6255 but better for some race conditions.
6256 counter - Simple counting of events (1, 2, ..)
6257 note, some counts may be skipped due to the
6258 infrastructure grabbing the clock more than
6259 once per event.
6260 uptime - Use jiffies as the time stamp.
6261 perf - Use the same clock that perf uses.
6262 mono - Use ktime_get_mono_fast_ns() for time stamps.
6263 mono_raw - Use ktime_get_raw_fast_ns() for time
6264 stamps.
6265 boot - Use ktime_get_boot_fast_ns() for time stamps.
6266 Architectures may add more clocks. See
6267 Documentation/trace/ftrace.rst for more details.
6268
6269 trace_event=[event-list]
6270 [FTRACE] Set and start specified trace events in order
6271 to facilitate early boot debugging. The event-list is a
6272 comma-separated list of trace events to enable. See
6273 also Documentation/trace/events.rst
6274
6275 trace_options=[option-list]
6276 [FTRACE] Enable or disable tracer options at boot.
6277 The option-list is a comma delimited list of options
6278 that can be enabled or disabled just as if you were
6279 to echo the option name into
6280
6281 /sys/kernel/debug/tracing/trace_options
6282
6283 For example, to enable stacktrace option (to dump the
6284 stack trace of each event), add to the command line:
6285
6286 trace_options=stacktrace
6287
6288 See also Documentation/trace/ftrace.rst "trace options"
6289 section.
6290
6291 trace_trigger=[trigger-list]
6292 [FTRACE] Add a event trigger on specific events.
6293 Set a trigger on top of a specific event, with an optional
6294 filter.
6295
6296 The format is is "trace_trigger=<event>.<trigger>[ if <filter>],..."
6297 Where more than one trigger may be specified that are comma deliminated.
6298
6299 For example:
6300
6301 trace_trigger="sched_switch.stacktrace if prev_state == 2"
6302
6303 The above will enable the "stacktrace" trigger on the "sched_switch"
6304 event but only trigger it if the "prev_state" of the "sched_switch"
6305 event is "2" (TASK_UNINTERUPTIBLE).
6306
6307 See also "Event triggers" in Documentation/trace/events.rst
6308
6309
6310 traceoff_on_warning
6311 [FTRACE] enable this option to disable tracing when a
6312 warning is hit. This turns off "tracing_on". Tracing can
6313 be enabled again by echoing '1' into the "tracing_on"
6314 file located in /sys/kernel/debug/tracing/
6315
6316 This option is useful, as it disables the trace before
6317 the WARNING dump is called, which prevents the trace to
6318 be filled with content caused by the warning output.
6319
6320 This option can also be set at run time via the sysctl
6321 option: kernel/traceoff_on_warning
6322
6323 transparent_hugepage=
6324 [KNL]
6325 Format: [always|madvise|never]
6326 Can be used to control the default behavior of the system
6327 with respect to transparent hugepages.
6328 See Documentation/admin-guide/mm/transhuge.rst
6329 for more details.
6330
6331 trusted.source= [KEYS]
6332 Format: <string>
6333 This parameter identifies the trust source as a backend
6334 for trusted keys implementation. Supported trust
6335 sources:
6336 - "tpm"
6337 - "tee"
6338 - "caam"
6339 If not specified then it defaults to iterating through
6340 the trust source list starting with TPM and assigns the
6341 first trust source as a backend which is initialized
6342 successfully during iteration.
6343
6344 trusted.rng= [KEYS]
6345 Format: <string>
6346 The RNG used to generate key material for trusted keys.
6347 Can be one of:
6348 - "kernel"
6349 - the same value as trusted.source: "tpm" or "tee"
6350 - "default"
6351 If not specified, "default" is used. In this case,
6352 the RNG's choice is left to each individual trust source.
6353
6354 tsc= Disable clocksource stability checks for TSC.
6355 Format: <string>
6356 [x86] reliable: mark tsc clocksource as reliable, this
6357 disables clocksource verification at runtime, as well
6358 as the stability checks done at bootup. Used to enable
6359 high-resolution timer mode on older hardware, and in
6360 virtualized environment.
6361 [x86] noirqtime: Do not use TSC to do irq accounting.
6362 Used to run time disable IRQ_TIME_ACCOUNTING on any
6363 platforms where RDTSC is slow and this accounting
6364 can add overhead.
6365 [x86] unstable: mark the TSC clocksource as unstable, this
6366 marks the TSC unconditionally unstable at bootup and
6367 avoids any further wobbles once the TSC watchdog notices.
6368 [x86] nowatchdog: disable clocksource watchdog. Used
6369 in situations with strict latency requirements (where
6370 interruptions from clocksource watchdog are not
6371 acceptable).
6372
6373 tsc_early_khz= [X86] Skip early TSC calibration and use the given
6374 value instead. Useful when the early TSC frequency discovery
6375 procedure is not reliable, such as on overclocked systems
6376 with CPUID.16h support and partial CPUID.15h support.
6377 Format: <unsigned int>
6378
6379 tsx= [X86] Control Transactional Synchronization
6380 Extensions (TSX) feature in Intel processors that
6381 support TSX control.
6382
6383 This parameter controls the TSX feature. The options are:
6384
6385 on - Enable TSX on the system. Although there are
6386 mitigations for all known security vulnerabilities,
6387 TSX has been known to be an accelerator for
6388 several previous speculation-related CVEs, and
6389 so there may be unknown security risks associated
6390 with leaving it enabled.
6391
6392 off - Disable TSX on the system. (Note that this
6393 option takes effect only on newer CPUs which are
6394 not vulnerable to MDS, i.e., have
6395 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
6396 the new IA32_TSX_CTRL MSR through a microcode
6397 update. This new MSR allows for the reliable
6398 deactivation of the TSX functionality.)
6399
6400 auto - Disable TSX if X86_BUG_TAA is present,
6401 otherwise enable TSX on the system.
6402
6403 Not specifying this option is equivalent to tsx=off.
6404
6405 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6406 for more details.
6407
6408 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
6409 Abort (TAA) vulnerability.
6410
6411 Similar to Micro-architectural Data Sampling (MDS)
6412 certain CPUs that support Transactional
6413 Synchronization Extensions (TSX) are vulnerable to an
6414 exploit against CPU internal buffers which can forward
6415 information to a disclosure gadget under certain
6416 conditions.
6417
6418 In vulnerable processors, the speculatively forwarded
6419 data can be used in a cache side channel attack, to
6420 access data to which the attacker does not have direct
6421 access.
6422
6423 This parameter controls the TAA mitigation. The
6424 options are:
6425
6426 full - Enable TAA mitigation on vulnerable CPUs
6427 if TSX is enabled.
6428
6429 full,nosmt - Enable TAA mitigation and disable SMT on
6430 vulnerable CPUs. If TSX is disabled, SMT
6431 is not disabled because CPU is not
6432 vulnerable to cross-thread TAA attacks.
6433 off - Unconditionally disable TAA mitigation
6434
6435 On MDS-affected machines, tsx_async_abort=off can be
6436 prevented by an active MDS mitigation as both vulnerabilities
6437 are mitigated with the same mechanism so in order to disable
6438 this mitigation, you need to specify mds=off too.
6439
6440 Not specifying this option is equivalent to
6441 tsx_async_abort=full. On CPUs which are MDS affected
6442 and deploy MDS mitigation, TAA mitigation is not
6443 required and doesn't provide any additional
6444 mitigation.
6445
6446 For details see:
6447 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6448
6449 turbografx.map[2|3]= [HW,JOY]
6450 TurboGraFX parallel port interface
6451 Format:
6452 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
6453 See also Documentation/input/devices/joystick-parport.rst
6454
6455 udbg-immortal [PPC] When debugging early kernel crashes that
6456 happen after console_init() and before a proper
6457 console driver takes over, this boot options might
6458 help "seeing" what's going on.
6459
6460 uhash_entries= [KNL,NET]
6461 Set number of hash buckets for UDP/UDP-Lite connections
6462
6463 uhci-hcd.ignore_oc=
6464 [USB] Ignore overcurrent events (default N).
6465 Some badly-designed motherboards generate lots of
6466 bogus events, for ports that aren't wired to
6467 anything. Set this parameter to avoid log spamming.
6468 Note that genuine overcurrent events won't be
6469 reported either.
6470
6471 unknown_nmi_panic
6472 [X86] Cause panic on unknown NMI.
6473
6474 usbcore.authorized_default=
6475 [USB] Default USB device authorization:
6476 (default -1 = authorized except for wireless USB,
6477 0 = not authorized, 1 = authorized, 2 = authorized
6478 if device connected to internal port)
6479
6480 usbcore.autosuspend=
6481 [USB] The autosuspend time delay (in seconds) used
6482 for newly-detected USB devices (default 2). This
6483 is the time required before an idle device will be
6484 autosuspended. Devices for which the delay is set
6485 to a negative value won't be autosuspended at all.
6486
6487 usbcore.usbfs_snoop=
6488 [USB] Set to log all usbfs traffic (default 0 = off).
6489
6490 usbcore.usbfs_snoop_max=
6491 [USB] Maximum number of bytes to snoop in each URB
6492 (default = 65536).
6493
6494 usbcore.blinkenlights=
6495 [USB] Set to cycle leds on hubs (default 0 = off).
6496
6497 usbcore.old_scheme_first=
6498 [USB] Start with the old device initialization
6499 scheme (default 0 = off).
6500
6501 usbcore.usbfs_memory_mb=
6502 [USB] Memory limit (in MB) for buffers allocated by
6503 usbfs (default = 16, 0 = max = 2047).
6504
6505 usbcore.use_both_schemes=
6506 [USB] Try the other device initialization scheme
6507 if the first one fails (default 1 = enabled).
6508
6509 usbcore.initial_descriptor_timeout=
6510 [USB] Specifies timeout for the initial 64-byte
6511 USB_REQ_GET_DESCRIPTOR request in milliseconds
6512 (default 5000 = 5.0 seconds).
6513
6514 usbcore.nousb [USB] Disable the USB subsystem
6515
6516 usbcore.quirks=
6517 [USB] A list of quirk entries to augment the built-in
6518 usb core quirk list. List entries are separated by
6519 commas. Each entry has the form
6520 VendorID:ProductID:Flags. The IDs are 4-digit hex
6521 numbers and Flags is a set of letters. Each letter
6522 will change the built-in quirk; setting it if it is
6523 clear and clearing it if it is set. The letters have
6524 the following meanings:
6525 a = USB_QUIRK_STRING_FETCH_255 (string
6526 descriptors must not be fetched using
6527 a 255-byte read);
6528 b = USB_QUIRK_RESET_RESUME (device can't resume
6529 correctly so reset it instead);
6530 c = USB_QUIRK_NO_SET_INTF (device can't handle
6531 Set-Interface requests);
6532 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
6533 handle its Configuration or Interface
6534 strings);
6535 e = USB_QUIRK_RESET (device can't be reset
6536 (e.g morph devices), don't use reset);
6537 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
6538 more interface descriptions than the
6539 bNumInterfaces count, and can't handle
6540 talking to these interfaces);
6541 g = USB_QUIRK_DELAY_INIT (device needs a pause
6542 during initialization, after we read
6543 the device descriptor);
6544 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
6545 high speed and super speed interrupt
6546 endpoints, the USB 2.0 and USB 3.0 spec
6547 require the interval in microframes (1
6548 microframe = 125 microseconds) to be
6549 calculated as interval = 2 ^
6550 (bInterval-1).
6551 Devices with this quirk report their
6552 bInterval as the result of this
6553 calculation instead of the exponent
6554 variable used in the calculation);
6555 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
6556 handle device_qualifier descriptor
6557 requests);
6558 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
6559 generates spurious wakeup, ignore
6560 remote wakeup capability);
6561 k = USB_QUIRK_NO_LPM (device can't handle Link
6562 Power Management);
6563 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
6564 (Device reports its bInterval as linear
6565 frames instead of the USB 2.0
6566 calculation);
6567 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
6568 to be disconnected before suspend to
6569 prevent spurious wakeup);
6570 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
6571 pause after every control message);
6572 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
6573 delay after resetting its port);
6574 Example: quirks=0781:5580:bk,0a5c:5834:gij
6575
6576 usbhid.mousepoll=
6577 [USBHID] The interval which mice are to be polled at.
6578
6579 usbhid.jspoll=
6580 [USBHID] The interval which joysticks are to be polled at.
6581
6582 usbhid.kbpoll=
6583 [USBHID] The interval which keyboards are to be polled at.
6584
6585 usb-storage.delay_use=
6586 [UMS] The delay in seconds before a new device is
6587 scanned for Logical Units (default 1).
6588
6589 usb-storage.quirks=
6590 [UMS] A list of quirks entries to supplement or
6591 override the built-in unusual_devs list. List
6592 entries are separated by commas. Each entry has
6593 the form VID:PID:Flags where VID and PID are Vendor
6594 and Product ID values (4-digit hex numbers) and
6595 Flags is a set of characters, each corresponding
6596 to a common usb-storage quirk flag as follows:
6597 a = SANE_SENSE (collect more than 18 bytes
6598 of sense data, not on uas);
6599 b = BAD_SENSE (don't collect more than 18
6600 bytes of sense data, not on uas);
6601 c = FIX_CAPACITY (decrease the reported
6602 device capacity by one sector);
6603 d = NO_READ_DISC_INFO (don't use
6604 READ_DISC_INFO command, not on uas);
6605 e = NO_READ_CAPACITY_16 (don't use
6606 READ_CAPACITY_16 command);
6607 f = NO_REPORT_OPCODES (don't use report opcodes
6608 command, uas only);
6609 g = MAX_SECTORS_240 (don't transfer more than
6610 240 sectors at a time, uas only);
6611 h = CAPACITY_HEURISTICS (decrease the
6612 reported device capacity by one
6613 sector if the number is odd);
6614 i = IGNORE_DEVICE (don't bind to this
6615 device);
6616 j = NO_REPORT_LUNS (don't use report luns
6617 command, uas only);
6618 k = NO_SAME (do not use WRITE_SAME, uas only)
6619 l = NOT_LOCKABLE (don't try to lock and
6620 unlock ejectable media, not on uas);
6621 m = MAX_SECTORS_64 (don't transfer more
6622 than 64 sectors = 32 KB at a time,
6623 not on uas);
6624 n = INITIAL_READ10 (force a retry of the
6625 initial READ(10) command, not on uas);
6626 o = CAPACITY_OK (accept the capacity
6627 reported by the device, not on uas);
6628 p = WRITE_CACHE (the device cache is ON
6629 by default, not on uas);
6630 r = IGNORE_RESIDUE (the device reports
6631 bogus residue values, not on uas);
6632 s = SINGLE_LUN (the device has only one
6633 Logical Unit);
6634 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
6635 commands, uas only);
6636 u = IGNORE_UAS (don't bind to the uas driver);
6637 w = NO_WP_DETECT (don't test whether the
6638 medium is write-protected).
6639 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
6640 even if the device claims no cache,
6641 not on uas)
6642 Example: quirks=0419:aaf5:rl,0421:0433:rc
6643
6644 user_debug= [KNL,ARM]
6645 Format: <int>
6646 See arch/arm/Kconfig.debug help text.
6647 1 - undefined instruction events
6648 2 - system calls
6649 4 - invalid data aborts
6650 8 - SIGSEGV faults
6651 16 - SIGBUS faults
6652 Example: user_debug=31
6653
6654 userpte=
6655 [X86] Flags controlling user PTE allocations.
6656
6657 nohigh = do not allocate PTE pages in
6658 HIGHMEM regardless of setting
6659 of CONFIG_HIGHPTE.
6660
6661 vdso= [X86,SH,SPARC]
6662 On X86_32, this is an alias for vdso32=. Otherwise:
6663
6664 vdso=1: enable VDSO (the default)
6665 vdso=0: disable VDSO mapping
6666
6667 vdso32= [X86] Control the 32-bit vDSO
6668 vdso32=1: enable 32-bit VDSO
6669 vdso32=0 or vdso32=2: disable 32-bit VDSO
6670
6671 See the help text for CONFIG_COMPAT_VDSO for more
6672 details. If CONFIG_COMPAT_VDSO is set, the default is
6673 vdso32=0; otherwise, the default is vdso32=1.
6674
6675 For compatibility with older kernels, vdso32=2 is an
6676 alias for vdso32=0.
6677
6678 Try vdso32=0 if you encounter an error that says:
6679 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
6680
6681 vector= [IA-64,SMP]
6682 vector=percpu: enable percpu vector domain
6683
6684 video= [FB] Frame buffer configuration
6685 See Documentation/fb/modedb.rst.
6686
6687 video.brightness_switch_enabled= [ACPI]
6688 Format: [0|1]
6689 If set to 1, on receiving an ACPI notify event
6690 generated by hotkey, video driver will adjust brightness
6691 level and then send out the event to user space through
6692 the allocated input device. If set to 0, video driver
6693 will only send out the event without touching backlight
6694 brightness level.
6695 default: 1
6696
6697 virtio_mmio.device=
6698 [VMMIO] Memory mapped virtio (platform) device.
6699
6700 <size>@<baseaddr>:<irq>[:<id>]
6701 where:
6702 <size> := size (can use standard suffixes
6703 like K, M and G)
6704 <baseaddr> := physical base address
6705 <irq> := interrupt number (as passed to
6706 request_irq())
6707 <id> := (optional) platform device id
6708 example:
6709 virtio_mmio.device=1K@0x100b0000:48:7
6710
6711 Can be used multiple times for multiple devices.
6712
6713 vga= [BOOT,X86-32] Select a particular video mode
6714 See Documentation/x86/boot.rst and
6715 Documentation/admin-guide/svga.rst.
6716 Use vga=ask for menu.
6717 This is actually a boot loader parameter; the value is
6718 passed to the kernel using a special protocol.
6719
6720 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
6721 May slow down system boot speed, especially when
6722 enabled on systems with a large amount of memory.
6723 All options are enabled by default, and this
6724 interface is meant to allow for selectively
6725 enabling or disabling specific virtual memory
6726 debugging features.
6727
6728 Available options are:
6729 P Enable page structure init time poisoning
6730 - Disable all of the above options
6731
6732 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
6733 size of <nn>. This can be used to increase the
6734 minimum size (128MB on x86). It can also be used to
6735 decrease the size and leave more room for directly
6736 mapped kernel RAM.
6737
6738 vmcp_cma=nn[MG] [KNL,S390]
6739 Sets the memory size reserved for contiguous memory
6740 allocations for the vmcp device driver.
6741
6742 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
6743 Format: <command>
6744
6745 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
6746 Format: <command>
6747
6748 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
6749 Format: <command>
6750
6751 vsyscall= [X86-64]
6752 Controls the behavior of vsyscalls (i.e. calls to
6753 fixed addresses of 0xffffffffff600x00 from legacy
6754 code). Most statically-linked binaries and older
6755 versions of glibc use these calls. Because these
6756 functions are at fixed addresses, they make nice
6757 targets for exploits that can control RIP.
6758
6759 emulate [default] Vsyscalls turn into traps and are
6760 emulated reasonably safely. The vsyscall
6761 page is readable.
6762
6763 xonly Vsyscalls turn into traps and are
6764 emulated reasonably safely. The vsyscall
6765 page is not readable.
6766
6767 none Vsyscalls don't work at all. This makes
6768 them quite hard to use for exploits but
6769 might break your system.
6770
6771 vt.color= [VT] Default text color.
6772 Format: 0xYX, X = foreground, Y = background.
6773 Default: 0x07 = light gray on black.
6774
6775 vt.cur_default= [VT] Default cursor shape.
6776 Format: 0xCCBBAA, where AA, BB, and CC are the same as
6777 the parameters of the <Esc>[?A;B;Cc escape sequence;
6778 see VGA-softcursor.txt. Default: 2 = underline.
6779
6780 vt.default_blu= [VT]
6781 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
6782 Change the default blue palette of the console.
6783 This is a 16-member array composed of values
6784 ranging from 0-255.
6785
6786 vt.default_grn= [VT]
6787 Format: <green0>,<green1>,<green2>,...,<green15>
6788 Change the default green palette of the console.
6789 This is a 16-member array composed of values
6790 ranging from 0-255.
6791
6792 vt.default_red= [VT]
6793 Format: <red0>,<red1>,<red2>,...,<red15>
6794 Change the default red palette of the console.
6795 This is a 16-member array composed of values
6796 ranging from 0-255.
6797
6798 vt.default_utf8=
6799 [VT]
6800 Format=<0|1>
6801 Set system-wide default UTF-8 mode for all tty's.
6802 Default is 1, i.e. UTF-8 mode is enabled for all
6803 newly opened terminals.
6804
6805 vt.global_cursor_default=
6806 [VT]
6807 Format=<-1|0|1>
6808 Set system-wide default for whether a cursor
6809 is shown on new VTs. Default is -1,
6810 i.e. cursors will be created by default unless
6811 overridden by individual drivers. 0 will hide
6812 cursors, 1 will display them.
6813
6814 vt.italic= [VT] Default color for italic text; 0-15.
6815 Default: 2 = green.
6816
6817 vt.underline= [VT] Default color for underlined text; 0-15.
6818 Default: 3 = cyan.
6819
6820 watchdog timers [HW,WDT] For information on watchdog timers,
6821 see Documentation/watchdog/watchdog-parameters.rst
6822 or other driver-specific files in the
6823 Documentation/watchdog/ directory.
6824
6825 watchdog_thresh=
6826 [KNL]
6827 Set the hard lockup detector stall duration
6828 threshold in seconds. The soft lockup detector
6829 threshold is set to twice the value. A value of 0
6830 disables both lockup detectors. Default is 10
6831 seconds.
6832
6833 workqueue.watchdog_thresh=
6834 If CONFIG_WQ_WATCHDOG is configured, workqueue can
6835 warn stall conditions and dump internal state to
6836 help debugging. 0 disables workqueue stall
6837 detection; otherwise, it's the stall threshold
6838 duration in seconds. The default value is 30 and
6839 it can be updated at runtime by writing to the
6840 corresponding sysfs file.
6841
6842 workqueue.disable_numa
6843 By default, all work items queued to unbound
6844 workqueues are affine to the NUMA nodes they're
6845 issued on, which results in better behavior in
6846 general. If NUMA affinity needs to be disabled for
6847 whatever reason, this option can be used. Note
6848 that this also can be controlled per-workqueue for
6849 workqueues visible under /sys/bus/workqueue/.
6850
6851 workqueue.power_efficient
6852 Per-cpu workqueues are generally preferred because
6853 they show better performance thanks to cache
6854 locality; unfortunately, per-cpu workqueues tend to
6855 be more power hungry than unbound workqueues.
6856
6857 Enabling this makes the per-cpu workqueues which
6858 were observed to contribute significantly to power
6859 consumption unbound, leading to measurably lower
6860 power usage at the cost of small performance
6861 overhead.
6862
6863 The default value of this parameter is determined by
6864 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
6865
6866 workqueue.debug_force_rr_cpu
6867 Workqueue used to implicitly guarantee that work
6868 items queued without explicit CPU specified are put
6869 on the local CPU. This guarantee is no longer true
6870 and while local CPU is still preferred work items
6871 may be put on foreign CPUs. This debug option
6872 forces round-robin CPU selection to flush out
6873 usages which depend on the now broken guarantee.
6874 When enabled, memory and cache locality will be
6875 impacted.
6876
6877 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
6878 default x2apic cluster mode on platforms
6879 supporting x2apic.
6880
6881 xen_512gb_limit [KNL,X86-64,XEN]
6882 Restricts the kernel running paravirtualized under Xen
6883 to use only up to 512 GB of RAM. The reason to do so is
6884 crash analysis tools and Xen tools for doing domain
6885 save/restore/migration must be enabled to handle larger
6886 domains.
6887
6888 xen_emul_unplug= [HW,X86,XEN]
6889 Unplug Xen emulated devices
6890 Format: [unplug0,][unplug1]
6891 ide-disks -- unplug primary master IDE devices
6892 aux-ide-disks -- unplug non-primary-master IDE devices
6893 nics -- unplug network devices
6894 all -- unplug all emulated devices (NICs and IDE disks)
6895 unnecessary -- unplugging emulated devices is
6896 unnecessary even if the host did not respond to
6897 the unplug protocol
6898 never -- do not unplug even if version check succeeds
6899
6900 xen_legacy_crash [X86,XEN]
6901 Crash from Xen panic notifier, without executing late
6902 panic() code such as dumping handler.
6903
6904 xen_msr_safe= [X86,XEN]
6905 Format: <bool>
6906 Select whether to always use non-faulting (safe) MSR
6907 access functions when running as Xen PV guest. The
6908 default value is controlled by CONFIG_XEN_PV_MSR_SAFE.
6909
6910 xen_nopvspin [X86,XEN]
6911 Disables the qspinlock slowpath using Xen PV optimizations.
6912 This parameter is obsoleted by "nopvspin" parameter, which
6913 has equivalent effect for XEN platform.
6914
6915 xen_nopv [X86]
6916 Disables the PV optimizations forcing the HVM guest to
6917 run as generic HVM guest with no PV drivers.
6918 This option is obsoleted by the "nopv" option, which
6919 has equivalent effect for XEN platform.
6920
6921 xen_no_vector_callback
6922 [KNL,X86,XEN] Disable the vector callback for Xen
6923 event channel interrupts.
6924
6925 xen_scrub_pages= [XEN]
6926 Boolean option to control scrubbing pages before giving them back
6927 to Xen, for use by other domains. Can be also changed at runtime
6928 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
6929 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
6930
6931 xen_timer_slop= [X86-64,XEN]
6932 Set the timer slop (in nanoseconds) for the virtual Xen
6933 timers (default is 100000). This adjusts the minimum
6934 delta of virtualized Xen timers, where lower values
6935 improve timer resolution at the expense of processing
6936 more timer interrupts.
6937
6938 xen.balloon_boot_timeout= [XEN]
6939 The time (in seconds) to wait before giving up to boot
6940 in case initial ballooning fails to free enough memory.
6941 Applies only when running as HVM or PVH guest and
6942 started with less memory configured than allowed at
6943 max. Default is 180.
6944
6945 xen.event_eoi_delay= [XEN]
6946 How long to delay EOI handling in case of event
6947 storms (jiffies). Default is 10.
6948
6949 xen.event_loop_timeout= [XEN]
6950 After which time (jiffies) the event handling loop
6951 should start to delay EOI handling. Default is 2.
6952
6953 xen.fifo_events= [XEN]
6954 Boolean parameter to disable using fifo event handling
6955 even if available. Normally fifo event handling is
6956 preferred over the 2-level event handling, as it is
6957 fairer and the number of possible event channels is
6958 much higher. Default is on (use fifo events).
6959
6960 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
6961 Disables the PV optimizations forcing the guest to run
6962 as generic guest with no PV drivers. Currently support
6963 XEN HVM, KVM, HYPER_V and VMWARE guest.
6964
6965 nopvspin [X86,XEN,KVM]
6966 Disables the qspinlock slow path using PV optimizations
6967 which allow the hypervisor to 'idle' the guest on lock
6968 contention.
6969
6970 xirc2ps_cs= [NET,PCMCIA]
6971 Format:
6972 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
6973
6974 xive= [PPC]
6975 By default on POWER9 and above, the kernel will
6976 natively use the XIVE interrupt controller. This option
6977 allows the fallback firmware mode to be used:
6978
6979 off Fallback to firmware control of XIVE interrupt
6980 controller on both pseries and powernv
6981 platforms. Only useful on POWER9 and above.
6982
6983 xive.store-eoi=off [PPC]
6984 By default on POWER10 and above, the kernel will use
6985 stores for EOI handling when the XIVE interrupt mode
6986 is active. This option allows the XIVE driver to use
6987 loads instead, as on POWER9.
6988
6989 xhci-hcd.quirks [USB,KNL]
6990 A hex value specifying bitmask with supplemental xhci
6991 host controller quirks. Meaning of each bit can be
6992 consulted in header drivers/usb/host/xhci.h.
6993
6994 xmon [PPC]
6995 Format: { early | on | rw | ro | off }
6996 Controls if xmon debugger is enabled. Default is off.
6997 Passing only "xmon" is equivalent to "xmon=early".
6998 early Call xmon as early as possible on boot; xmon
6999 debugger is called from setup_arch().
7000 on xmon debugger hooks will be installed so xmon
7001 is only called on a kernel crash. Default mode,
7002 i.e. either "ro" or "rw" mode, is controlled
7003 with CONFIG_XMON_DEFAULT_RO_MODE.
7004 rw xmon debugger hooks will be installed so xmon
7005 is called only on a kernel crash, mode is write,
7006 meaning SPR registers, memory and, other data
7007 can be written using xmon commands.
7008 ro same as "rw" option above but SPR registers,
7009 memory, and other data can't be written using
7010 xmon commands.
7011 off xmon is disabled.
7012
7013 amd_pstate= [X86]
7014 disable
7015 Do not enable amd_pstate as the default
7016 scaling driver for the supported processors
7017 passive
7018 Use amd_pstate as a scaling driver, driver requests a
7019 desired performance on this abstract scale and the power
7020 management firmware translates the requests into actual
7021 hardware states (core frequency, data fabric and memory
7022 clocks etc.)